research article derivation of field equations in space...
TRANSCRIPT
Research ArticleDerivation of Field Equations in Space with the GeometricStructure Generated by Metric and Torsion
Nikolay Yaremenko
Yu A Mitropolsky Department International Mathematical Center JO Mitropolsky Kiev 01601 Ukraine
Correspondence should be addressed to Nikolay Yaremenko mathkievgmailcom
Received 21 August 2014 Revised 29 October 2014 Accepted 29 October 2014 Published 11 December 2014
Academic Editor Kazuharu Bamba
Copyright copy 2014 Nikolay YaremenkoThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
This paper is devoted to the derivation of field equations in space with the geometric structure generated by metric and torsiontensors We also study the geometry of the space generated jointly and agreed on by the metric tensor and the torsion tensorWe showed that in such space the structure of the curvature tensor has special features and for this tensor we obtained analogRicci-Jacobi identity and evaluated the gap that occurs at the transition from the original to the image and vice versa in the case ofinfinitely small contoursWe have researched the geodesic lines equationWe introduce the tensor120587
120572120573which is similar to the second
fundamental tensor of hypersurfaces 119884119899minus1 but the structure of this tensor is substantially different from the case of Riemannianspaces with zero torsion Then we obtained formulas which characterize the change of vectors in accompanying basis relative tothis basis itself Taking into considerations our results about the structure of such space we derived from the variation principle thegeneral field equations (electromagnetic and gravitational)
1 Introduction
In this paper we study the properties of the geometry of thespace generated jointly and agreed on by the metric tensorand the torsion tensor so we investigate the spaces withconnection in the presence of the metric tensor We obtainedsome results on the structure of the curvature tensor andconsidered the construction of geodesic lines and an estimateof the gap that occurs when traversing the contour of aparallelogram in these spaces
The investigation of properties of metric spaces and affineconnection spaces began approximately at the beginning ofthe 20th century [1 2] and continued to develop so far [2ndash16]
The importance of this kind of research is on the onehand due to the internal logic of mathematical science basesitself [1 2 9 13] and on the other hand to applications toproblems in physics analytical and theoretical mechanics[3 12] the theory of relativity [7 14ndash16] and continuummechanics and cosmology [10] Fairly well studied Riemannspaces [9] because of the wealth of geometric propertiesand less explored space with affine connection [5] are notsufficiently considered the most interesting geometry which
is obtained by combining geometry and affine connectiongenerated by the metric tensor and this is the subject ofthis work From the theory of spaces with affine connectionit is known that parallel displacement vector depends onpathways that is if the vector is parallel transported at thegiven contour with it return to the starting point we obtainthe other vector than the original (the gap appears) In thespaces which are studied in this work not only do they holda similar statement but also there are new properties of thisgap
The principle of least action (more correctly the principleof stationary action) is the basic variational principle ofparticle and continuum systems Let the starting point be theaction denoted as 119878 of a physical system It is defined as theintegral of the Lagrangian 119871 between two instants of time afunctional of the 119899 generalized coordinates 119902which define theconfiguration of the system
119878 (119902 (119905)) = int
1199052
1199051
119871 (119902 (119905) 119902 (119905)) 119889119905 (1)
where the dot denotes the time derivative and 119905 is timeMath-ematically the principle is 120575119878 = 0 where 120575means a variation
Hindawi Publishing CorporationJournal of GravityVolume 2014 Article ID 420123 13 pageshttpdxdoiorg1011552014420123
2 Journal of Gravity
In applications the statement and definition of action aretaken together
120575int
1199052
1199051
119871 (119902 (119905) 119902 (119905)) 119889119905 = 0 (2)
The action and Lagrangian both contain the dynamics ofthe system for all times The term ldquopathrdquo simply refers to acurve traced out by the system in terms of the coordinates inthe configuration space that is the curve 119902(119905) parameterizedby time On the other hand a Finsler manifold is a differ-entiable manifold together with the structure of an intrinsicquasimetric space in which the length of any rectifiable curve120591 [119886 119887] rarr 119872 is given by the length functional
119878 (120591 (119905)) = int
1199052
1199051
119865 (120591 (119905) 120591 (119905)) 119889119905 (3)
where 119865(119909 sdot) is a Minkowski norm on each tangent space Itis obvious from these definitions that there is a connectionbetween these two concepts which can be realized byHamiltonian formalism Thus any Riemannian space can beregarded as Finsler manifold with the length functional 1198652 =119892119894119895(119909)119889119909
119894
119889119909119895 and so the geodesics of a Finsler manifold are
geodesics of Riemannian spaceThe geodesics of the space that are being studied in
our work (with the geometric structure generated by metric119892119894119895(119909) and torsion 119878119896
119894119895(119909) tensors) are different from geodesics
of corresponding Riemann space (with 119892119894119895(119909)) and so of
geodesics Finsler manifold (with 1198652 = 119892119894119895(119909)119889119909
119894
119889119909119895)
These spaces retain all the properties of geometry ofan affine space but important features associated with thepresence of the metric appear the structure of the curvaturetensor has a specific characteristics as well as an opportunityto assess the gap that occurs at the transportation from theoriginal to the image and conversely in the case of infinitelysmall contours
The main objective of this work-the study of the geomet-ric properties of the space that is generated by metric 119892
119894119895(119909)
and torsion 119878119896
119894119895(119909) tensors obtain the field equations from
variation principle in such spacesWe remind that according to Albert Einstein proposal
the free falling gravitating massive bodies follow geodesic lineIf we postulate this proposal we can obtain some results ofNewton theory as a consequenceWe have another importantassumption of Albert Einstein that the geodesic equation ofmotion can be derived from the field equations for emptyspace
Since we believe that gravitational and electromagneticfields are determined geometric structure of empty space(torsion and curvature) it is interesting to have exampleabout geometric sense of torsion
Now we discuss one example about geometric sense oftorsionWe consider the surface 119878 At point119860 on 119878 construct atangent plane 119875 We choose an arbitrary infinitesimal square119860119861119862119863 in the plane 119875 with vertex 119860 From point 119860 on thesurface 119878 will draw the geodesic in the direction of 119860119861 Wepass along the distance corresponding parameter equal to the
length of119860119861 and get to point 1198611015840 Similarly from119860 on 119878 drawgeodesic towards 119860119863 and get into 1198631015840 We perform a paralleltransportation of vector 119860119863 to point 1198611015840 along the geodesic1198601198611015840 and draw geodesic 1198611015840 along the transportation of this
vector we reach the point 1198621015840 Similarly the vector 119860119861 willmove parallel along the1198601198631015840 and along the transported vectorfrom 119863
1015840 draw a geodesic to get to 11986210158401015840 If torsion is zero then1198621015840
= 11986210158401015840 and geodesic square up to small higher order will
be closed otherwise not In our case due to the presence of themetric can calculate the length of the gap more precisely we canestimate the length of this gap
These considerations are true only up to the second orderrelative to the length of square side If we want more strictresult we must consider the component of curvature tensorNext this example is true only when length of square sidetends to zero that is remains very small in other words ingeneral it is a local property
We go to the discussion of physical interpretation of thisexampleThe physical properties of the space-time (more justspace) are defined by the presence of matter (electromagneticfields and mass) in this space and from the viewpoint ofmathematics are described by the geometrical structure ofspace (torsion andmetric tensors)The empty space (withoutmatter) corresponds to the geometric structure of Euclideanspace (torsion tensor and curvature tensor are identicallyequal to zero) Similarly gravitation (mass and without elec-tromagnetic fields) corresponds to the geometric structure ofRiemannian space (torsion tensor is identically equal to zero)And similarly the electromagnetism (electromagnetic fieldsand without mass) corresponds to the geometric structure ofaffine connection space (curvature tensor is identically equalto zero) In the last two cases the result is conditional (notstrict) because the matter division by the mass and field isconditional
To describe the internal geometry of the subspace issufficient to define two tensors the metric tensor 119892
119894119896and the
torsion tensor 119878119898119894119896 using these tensors one can build connec-
tion of the space and all the geometry of the spaceIf we consider the space of hypersurface 119884119899minus1 as a sub-
space then for a complete description of the geometry itis necessary and sufficient to define the tensors 119886
120572120573and
119879120574
120572120573 If we consider the hypersurface 119884
119899minus1 as a subspacethat embedded in a space 119884
119899 then it is not sufficient toknow tensors 119886
120572120573and 119879
120574
120572120573for identifying the space 119884119899 (for
obtaining the metric tensor 119892119894119896and the torsion tensor 119878119898
119894119896)
to describe the geometry of the ambient space it is necessaryand sufficient to specify exactly how the hypersurface 119884119899minus1 isembedded in it that is specify the embedment it can be donein various ways We assume that Jacobi matrix is known sothe way of embedment hypersurface119884119899minus1 in119884119899 is determined(also the space can be recovered by determination of normalfor the hypersurface 119884119899minus1)
We consider the followingmethod of constructing space-time the space is constructed on the basis of manifoldsby determining these manifolds metric tensor and torsiontensorMetric and torsion tensors are calculated from the dif-ferential equations of the field Hence torsion as the curvature
Journal of Gravity 3
arises from the physical features of the distribution of matterin space-time Roughly the same way as the masses leads tocurvature space-time electromagnetism leads to appearanceof torsion But on the other hand from the mathematicalpoint of view if we assume that the space-time embeddedin Euclidean space of higher dimension then the appearanceof torsion can be explained by violation of the smoothnessembedding Therefore we can conditionally determine thetorsion and curvature by violation of smoothness regardlessof the dimension and embedment
The aim of our paper is to study the property of metricspace with torsion and obtain analog of Einstein-Hilbertequation at such space
This paper is organized as follows In Section 2 somegeneral properties of structure of a metric space with torsionare discussed In Section 3 we present the study of geodesicsin the space with torsion These results can be used inldquogeodesic principlerdquo In Section 4 the theory of hypersurfacesin the space with torsion is dedicated In Section 5 weobtained some interesting relationships which is used inSection 6 In Section 6 we derive analog of Einstein-Hilbert(for electromagnetic and gravitational fields) in case of ametric space with torsion
The main natural assumption that is used below that ascalar product of two any vectors that is transporting parallelalong an arbitrary path does not change of the 20th century[2 8 11ndash13] and continues to develop so far [3ndash7 14ndash22]
2 Structure of a Metric Space with Torsion
There are many ways to represent the physical four-dimensional space where events of our reality are occurringFrom a mathematical point of view there are two possibleconceptions of space geometry which might be identifiedwith the physical space
The first scheme is a generalization of Euclidean geom-etry the geometry of the Riemannian metric that is 119899-dimensional manifold equipped with a field twice covariantsymmetric metric tensor which is non-degenerate 119892
119894119896(119872)
where Det|119892119894119896| = 0 and 119892
119894119896= 119892119896119894 Note that the metric tensor
is chosen arbitrarily but in addition to conditions laid abovewe demand that the manifold was sufficiently smooth
This definition can be rewritten as follows invariantdifferential quadratic form 119892
119894119896119889119909119894
119889119909119896 determined on the
manifold and satisfying the conditions Det|119892119894119896| = 0 119892
119894119896= 119892119896119894
defines the geometry of RiemannAs a consequence of the invariance of the form
1198891199042
= 119892119894119896119889119909119894
119889119909119896
(4)
we find that the coefficients 119892119894119896are forming a tensor field
In this model for the arc length of the curve 119909119894 = 119909119894
(119905)119905 isin [119886 119887] sub 119877 is given by integral
119904 = int
119887
119886
radic119892119894119896119889119909119894119889119909119896
= int
119887
119886
radic119892119894119896(1199091(119905) 119909
119899(119905))
119889119909119894
(119905)
119889119905
119889119909119896
(119905)
119889119905
119889119905
(5)
The second scheme is a generalization of affine geometrythe geometry of affine connection Γ119894
119895119896(119872) that is based on 119899-
dimensional manifoldThe connection Γ
119894
119895119896(119872) is a geometric object on a man-
ifold and is subjected to the law of the transformation fromone coordinate system 119909
119894 to another 1199091198941015840
in the form of
Γ1198941015840
11989510158401198961015840 = Γ119894
119895119896
1205971199091198941015840
120597119909119894
120597119909119895
1205971199091198951015840
120597119909119896
1205971199091198961015840+
1205972
119909119894
1205971199091198951015840
1205971199091198961015840
1205971199091198941015840
120597119909119894 (6)
where the functions Γ119894119895119896are sufficiently smooth
Let along the curve 119909119894 = 119909119894
(119905) 119905 isin [119886 119887] sub 119877 is giventensor field 119860119894 = 119860
119894
(119905) if for each infinitesimal displacementtensor 119889119860119894(119905) coordinates is changing the law
119889119860119894
= minusΓ119894
119895119896119860119895
119889119909119896
(7)
then we say that the tensor 119860119894 is transported parallel to thecurve 119905
Depending on the physical investigated problem one oranother geometricmodel is choosing but as the internal logicand common sense requires that in the physical world thesetwo models coexist together and complement each otherThere is well-known result that in an arbitrary Riemannianspace can always construct a connection Γ
119894
119895119896(119872) An inter-
esting question is the uniqueness of such a construction Ingeneral such a construction Γ119894
119895119896is not unique but completely
natural (in terms of mathematics and physics to a greaterextent) there is the requirement that whenever two vectors119860119894and 119861119894 simultaneous parallel transported along a path (due tothe presence of the connection the transportation is defined)their scalar product does not change (the scalar product isdefined by metric) Mathematically this can be written as thevanishing differential
119889 (119892119894119896119860119894
119861119896
) = 0 (8)
If the requested coefficients Γ119894119895119896
are symmetric namelyΓ119894
119895119896= Γ119894
119896119895 then the connectivity is uniquely defined using a
metricAlways below we would not require the symmetry of
connection And so if the metric 119892119894119896
is defined then ageometric object Γ119894
119895119896subject to certain requirements but still
there is some arbitrariness in the choice of connectedness ofthe space namely we need to define a torsion tensor
119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895 (9)
then the geometric object Γ119894119895119896that generated the connection
is uniquely determined
Theorem 1 Suppose that a Riemannian space with the metric119892119894119896and in this space is given torsion tensor 119878119894
119895119896-antisymmetric
If demand 119889(119892119894119896119860119894
119861119896
) = 0 for arbitrary 119860119894 and 119861119896 then the
connection (geometric object that defines it) Γ119894119895119896
is uniquelydefined
4 Journal of Gravity
Proof It is pretty easy to see the truth of this statement itselfbut for further more importantly those symbols and valuesthat relate to the nature of the entered values
We rewrite 119889(119892119894119896119860119894
119861119896
) = 0 as (119892119894119896119897
minus 119892119898119896Γ119898
119894119897minus
119892119894119898Γ119898
119896119897)119860119894
119861119896
119889119909119897
= 0 due to the fact that 119889119860119894 = minusΓ119894
119901119897119860119901
119889119909119897
where Γ119898119894119897 unknown coefficients of connection are a geomet-
ric object and 119889119909119897 are the differentials of coordinates of a
point under infinitesimal displacement along the path 119892119894119896119897
equiv
(120597120597119909119897
)119892119894119896 Since119860119894 119861119896 119889119909119897 are arbitrary the equalities must
be identity relative to 119860119894 119861119896 119889119909119897 By circular permutation weobtain the system of equations
119892119894119896119897
= 119892119898119896Γ119898
119894119897+ 119892119894119898Γ119898
119896119897
119892119897119894119896
= 119892119898119894Γ119898
119897119896+ 119892119897119898Γ119898
119894119896
119892119896119897119894
= 119892119898119897Γ119898
119896119894+ 119892119896119898Γ119898
119897119894
(10)
Since the technique is similar to the classical one then wegive formulas without justification
119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
= 119892119898119896119878119898
119894119897+ 119892119898119897119878119898
119894119896+ 119892119894119898Γ119898
119896119897+ 119892119898119894Γ119898
119897119896 (11)
where 119878119898119894119897= Γ119898
119894119897minus Γ119898
119897119894is torsion tensor and we have
119892119894119898(Γ119898
119896119897+ Γ119898
119897119896)
= 119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894
Γ119901
119896119897+ Γ119901
119897119896
= 119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(12)
and complementing the obvious equation-definition Γ119901
119896119897minus
Γ119901
119897119896= 119878119901
119896119897 then we obtain
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(13)
Thenwe introduce the notation and from the last formulawe see that
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (14)
is geometric object and
119871119901
119896119897equiv
1
2
119878119901
119896119897+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (15)
is tensorThe geometric object Γ
119901
119896119897 which generate connection
space is completely determined by the tensors 119892119894119896and 119878
119898
119894119896
Therefore the connection Γ119901119896119897is the sum of a geometric object
119875119901
119896119897which is composed of derivatives of the metric tensor 119892
119894119896
and tensor 119871119901119896119897is compiled of 119892
119894119896and the tensor 119878119898
119896119897 namely
Γ119901
119896119897= 119875119901
119896119897+ 119871119901
119896119897 (16)
Remark 2 Tensor 119871119901
119896119897represents the sum of two tensors
symmetric (12)119892119901119894(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) and torsion (12)119878119901
119896119897
Remark 3 It is not difficult to prove the relation
Γ119901
119901119897=
1
2
119892119894119901119897119892119894119901
=
1
radic119892
120597radic119892
120597119909119897
where 119892 = det 1003816100381610038161003816119892119894119896
1003816100381610038161003816 (17)
The next step in building a geometric theory is the con-sideration of the parallel transport tensor-vector 119860119894 which isgiven by
119889119860119894
= minusΓ119894
119901119897119860119901
119889119909119897
(18)
where the coefficients Γ119894119901119897are the connection of space
As further arguments are similar to the classic and oftenrepeated them the presentation of intermediate results willwear schematic character
By a covariant derivative of 119906119894with respect to 119897 we mean
119906119894119897equiv 119906119894119897minus Γ119896
119894119897119906119896
119906119894
119897equiv 119906119894
119897+ Γ119894
119896119897119906119896
(19)
Then we consider the difference
119906119894119897minus 119906119897119894= 119906119894119897minus 119906119897119894minus 119878119896
119894119897119906119896 (20)
During the transition along a parallelogram in the imageto the original polygon the gap 119885
119896 is formed (breaking thecircuit) which can be estimated as follows up to the 2ndorder of smallness relative to sides of a parallelogram
119885119896
= 119878119896
119894119895119860119894
119861119895
1205912
(21)
It is the result of coagulation at the torsion tensor with vectorparties 119860119894120591 and 119861
119895
120591 that express geodetic displacement Thebasic geometric meaning of the torsion tensor is an estimateof the gap (up to 2nd order) at which the open loop shrinks ifthe torsion is zero then the gap will not infinitesimal secondbut higher order smallness
Next we consider the difference of the second orderderivatives
119906119894119897119896
minus 119906119894119896119897
= 119877119901
119896119897119894119906119901+ 119878119902
119896119897119906119894119902 (22)
where we identified
119877119901
119896119897119894equiv Γ119901
119894119896119897minus Γ119901
119894119897119896+ Γ119901
119902119897Γ119902
119894119896minus Γ119901
119902119896Γ119902
119894119897 (23)
119877119901
119896119897119894is curvature tensorSimilarly we have
119906119894
119897119896minus 119906119894
119896119897= minus119877119894
119896119897119901119906119901
+ 119878119902
119896119897119906119894
119902 (24)
Remark 4 Also possible and slightly different way of defini-tion a covariant derivative namely the absolute differential119863119860119894 is first determined using the formula
119863119860119894
equivsim 119889119860119894
+ Γ119894
119895119896119860119895
119889119909119896
= (119860119894
119896+ Γ119894
119895119896119860119895
) 119889119909119896
(25)
Journal of Gravity 5
Absolute differential can be defined as derivative coeffi-cients all the results obtained with this approach to the con-struction is identical with the analysis which has been madeabove Then more clearly we can assert that for any space topossess absolute parallelism it is a necessary and sufficientcondition that curvature tensor be identity vanishing (recallthat the space is called with absolute parallelism if the resultof the parallel transport of an arbitrary tensor-vector doesnot depend on the choice of path for all points of space) Theproof of this theorem is generally known we note only that itfollows from the formula
119863119863119860119894
minus 119863119863119860119894
= minus119877119894
119896119897119901119860119901 119889119909119896
119889119909119897
(26)
which we could get by folding (24) with 119889119909119896
119889119909119897
All the constructions outlined above are general in naturewithout specifying space further we will investigate thestructure of the tensor 119877119901
119894119896119897 So by definition of (23) we have
119877119901
119894119896119897= Γ119901
119897119894119896minus Γ119901
119897119896119894+ Γ119901
119902119896Γ119902
119897119894minus Γ119901
119902119894Γ119902
119897119896 (27)
Then we use (16) and obtain
119877119901
119894119896119897= 119875119901
119897119894119896+ 119871119901
119897119894119896minus 119875119901
119897119896119894+ 119871119901
119897119896119894+ (119875119901
119902119896+ 119871119901
119902119896) (119875119901
119897119894+ 119871119902
119897119894)
minus (119875119901
119902119894+ 119871119901
119902119894) (119875119901
119897119896+ 119871119902
119897119896)
= 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894minus 119875119901
119902119894119875119902
119897119896+ 119871119901
119897119894119896minus 119871119901
119897119896119894
+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
+ 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896
(28)
Next we introduce the notation
119875119901
119894119896119897equiv 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894+ 119875119901
119902119894119875119902
119897119896(29)
is a tensor like the Riemann curvature tensor composed ofthe metric tensor and its derivatives Consider the following
119885119901
119894119896119897equiv 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896(30)
is tensor and
119879119901
119894119896119897equiv 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894(31)
is tensorIf we take into account that119877119901
119894119896119897is tensor the last assertion
is obvious It is interesting to obtain this important result inanother way namely
119879119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894Γ119901
119902119896+ 119871119901
119902119894Γ119902
119897119896+ 119871119901
119897119902Γ119902
119894119896+ 119871119902
119897119896Γ119901
119902119894
minus 119871119901
119902119896Γ119902
119897119894minus 119871119901
119897119902Γ119902
119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119902
119902119896
minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894119871119901
119902119896+ 119871119901
119902119894119871119902
119897119896+ 119871119902
119897119896119871119901
119902119894
minus 119871119901
119902119896119871119902
119897119894+ 119871119901
119897119902119878119902
119894119896
(32)
is obviously the tensor because the absolute derivatives havetensor character
We introduce the notation
119872119901
119894119896119897equiv 119879119901
119894119896119897+ 119885119901
119894119896119897 (33)
then we obtain
119872119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119871119901
119897119902119878119902
119894119896+ 119871119901
119902119894119871119902
119897119896minus 119871119901
119902119896119871119902
119897119894 (34)
and in the new notation
119877119901
119894119896119897= 119875119901
119894119896119897+119872119901
119894119896119897 (35)
Formula (35) shows that the curvature tensor in generalcases can be represented as the sum of two tensors (suchrepresentation is not accidental it is associatedwith a physicaldescription of the field roughly speaking in the absence ofgravitational fields tensor119872119901
119894119896119897is not equal to zero) Although
the formula (34) gives a qualitative representation of thegeometric structure it is a little convenient since it reentersthe values of Γ119894
119895119896
Further we establish the equation which is similar toequation of Ricci-Jacobi
119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ Γ119901
119902119896119878119902
119897119894+ Γ119901
119902119896119878119902
119894119896+ Γ119901
119902119894119878119902
119896119897
= 119878119901
119894119896119897+ Γ119902
119894119897119878119901
119902119896+ Γ119902
119896119897119878119901
119894119902+ 119878119901
119896119897119894+ Γ119902
119896119894119878119901
119902119897+ Γ119902
119897119894119878119901
119896119902
+ 119878119901
119897119894119896+ Γ119902
119897119896119878119901
119902119894+ Γ119902
119894119896119878119901
119897119902
= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896
+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897
(36)
It is easy to prove the equations
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895= 0
119878119894
119894119901119878119901
119895119896= 0
(37)
and as a consequence we obtain the equation
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895+ 119878119894
119894119901119878119901
119895119896= 0 (38)
3 The Study of Geodesics inthe Space with Torsion
Formula (13) consists of three summand of various kindsThe sum (14) is a geometric object second valence its
components are converted by a formula similar to the one thattakes place for the connection
1198751199011015840
11989610158401198971015840= 119875119901
119896119897
1205971199091199011015840
120597119909119901
120597119909119896
1205971199091198961015840
120597119909119897
1205971199091198971015840+
1205972
119909119901
1205971199091198961015840
1205971199091198971015840
1205971199091199011015840
120597119909119901 (39)
but note that in this space this ldquoconnectionrdquo does not satisfythe ldquotorsion conditionrdquo (119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895)
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
2 Journal of Gravity
In applications the statement and definition of action aretaken together
120575int
1199052
1199051
119871 (119902 (119905) 119902 (119905)) 119889119905 = 0 (2)
The action and Lagrangian both contain the dynamics ofthe system for all times The term ldquopathrdquo simply refers to acurve traced out by the system in terms of the coordinates inthe configuration space that is the curve 119902(119905) parameterizedby time On the other hand a Finsler manifold is a differ-entiable manifold together with the structure of an intrinsicquasimetric space in which the length of any rectifiable curve120591 [119886 119887] rarr 119872 is given by the length functional
119878 (120591 (119905)) = int
1199052
1199051
119865 (120591 (119905) 120591 (119905)) 119889119905 (3)
where 119865(119909 sdot) is a Minkowski norm on each tangent space Itis obvious from these definitions that there is a connectionbetween these two concepts which can be realized byHamiltonian formalism Thus any Riemannian space can beregarded as Finsler manifold with the length functional 1198652 =119892119894119895(119909)119889119909
119894
119889119909119895 and so the geodesics of a Finsler manifold are
geodesics of Riemannian spaceThe geodesics of the space that are being studied in
our work (with the geometric structure generated by metric119892119894119895(119909) and torsion 119878119896
119894119895(119909) tensors) are different from geodesics
of corresponding Riemann space (with 119892119894119895(119909)) and so of
geodesics Finsler manifold (with 1198652 = 119892119894119895(119909)119889119909
119894
119889119909119895)
These spaces retain all the properties of geometry ofan affine space but important features associated with thepresence of the metric appear the structure of the curvaturetensor has a specific characteristics as well as an opportunityto assess the gap that occurs at the transportation from theoriginal to the image and conversely in the case of infinitelysmall contours
The main objective of this work-the study of the geomet-ric properties of the space that is generated by metric 119892
119894119895(119909)
and torsion 119878119896
119894119895(119909) tensors obtain the field equations from
variation principle in such spacesWe remind that according to Albert Einstein proposal
the free falling gravitating massive bodies follow geodesic lineIf we postulate this proposal we can obtain some results ofNewton theory as a consequenceWe have another importantassumption of Albert Einstein that the geodesic equation ofmotion can be derived from the field equations for emptyspace
Since we believe that gravitational and electromagneticfields are determined geometric structure of empty space(torsion and curvature) it is interesting to have exampleabout geometric sense of torsion
Now we discuss one example about geometric sense oftorsionWe consider the surface 119878 At point119860 on 119878 construct atangent plane 119875 We choose an arbitrary infinitesimal square119860119861119862119863 in the plane 119875 with vertex 119860 From point 119860 on thesurface 119878 will draw the geodesic in the direction of 119860119861 Wepass along the distance corresponding parameter equal to the
length of119860119861 and get to point 1198611015840 Similarly from119860 on 119878 drawgeodesic towards 119860119863 and get into 1198631015840 We perform a paralleltransportation of vector 119860119863 to point 1198611015840 along the geodesic1198601198611015840 and draw geodesic 1198611015840 along the transportation of this
vector we reach the point 1198621015840 Similarly the vector 119860119861 willmove parallel along the1198601198631015840 and along the transported vectorfrom 119863
1015840 draw a geodesic to get to 11986210158401015840 If torsion is zero then1198621015840
= 11986210158401015840 and geodesic square up to small higher order will
be closed otherwise not In our case due to the presence of themetric can calculate the length of the gap more precisely we canestimate the length of this gap
These considerations are true only up to the second orderrelative to the length of square side If we want more strictresult we must consider the component of curvature tensorNext this example is true only when length of square sidetends to zero that is remains very small in other words ingeneral it is a local property
We go to the discussion of physical interpretation of thisexampleThe physical properties of the space-time (more justspace) are defined by the presence of matter (electromagneticfields and mass) in this space and from the viewpoint ofmathematics are described by the geometrical structure ofspace (torsion andmetric tensors)The empty space (withoutmatter) corresponds to the geometric structure of Euclideanspace (torsion tensor and curvature tensor are identicallyequal to zero) Similarly gravitation (mass and without elec-tromagnetic fields) corresponds to the geometric structure ofRiemannian space (torsion tensor is identically equal to zero)And similarly the electromagnetism (electromagnetic fieldsand without mass) corresponds to the geometric structure ofaffine connection space (curvature tensor is identically equalto zero) In the last two cases the result is conditional (notstrict) because the matter division by the mass and field isconditional
To describe the internal geometry of the subspace issufficient to define two tensors the metric tensor 119892
119894119896and the
torsion tensor 119878119898119894119896 using these tensors one can build connec-
tion of the space and all the geometry of the spaceIf we consider the space of hypersurface 119884119899minus1 as a sub-
space then for a complete description of the geometry itis necessary and sufficient to define the tensors 119886
120572120573and
119879120574
120572120573 If we consider the hypersurface 119884
119899minus1 as a subspacethat embedded in a space 119884
119899 then it is not sufficient toknow tensors 119886
120572120573and 119879
120574
120572120573for identifying the space 119884119899 (for
obtaining the metric tensor 119892119894119896and the torsion tensor 119878119898
119894119896)
to describe the geometry of the ambient space it is necessaryand sufficient to specify exactly how the hypersurface 119884119899minus1 isembedded in it that is specify the embedment it can be donein various ways We assume that Jacobi matrix is known sothe way of embedment hypersurface119884119899minus1 in119884119899 is determined(also the space can be recovered by determination of normalfor the hypersurface 119884119899minus1)
We consider the followingmethod of constructing space-time the space is constructed on the basis of manifoldsby determining these manifolds metric tensor and torsiontensorMetric and torsion tensors are calculated from the dif-ferential equations of the field Hence torsion as the curvature
Journal of Gravity 3
arises from the physical features of the distribution of matterin space-time Roughly the same way as the masses leads tocurvature space-time electromagnetism leads to appearanceof torsion But on the other hand from the mathematicalpoint of view if we assume that the space-time embeddedin Euclidean space of higher dimension then the appearanceof torsion can be explained by violation of the smoothnessembedding Therefore we can conditionally determine thetorsion and curvature by violation of smoothness regardlessof the dimension and embedment
The aim of our paper is to study the property of metricspace with torsion and obtain analog of Einstein-Hilbertequation at such space
This paper is organized as follows In Section 2 somegeneral properties of structure of a metric space with torsionare discussed In Section 3 we present the study of geodesicsin the space with torsion These results can be used inldquogeodesic principlerdquo In Section 4 the theory of hypersurfacesin the space with torsion is dedicated In Section 5 weobtained some interesting relationships which is used inSection 6 In Section 6 we derive analog of Einstein-Hilbert(for electromagnetic and gravitational fields) in case of ametric space with torsion
The main natural assumption that is used below that ascalar product of two any vectors that is transporting parallelalong an arbitrary path does not change of the 20th century[2 8 11ndash13] and continues to develop so far [3ndash7 14ndash22]
2 Structure of a Metric Space with Torsion
There are many ways to represent the physical four-dimensional space where events of our reality are occurringFrom a mathematical point of view there are two possibleconceptions of space geometry which might be identifiedwith the physical space
The first scheme is a generalization of Euclidean geom-etry the geometry of the Riemannian metric that is 119899-dimensional manifold equipped with a field twice covariantsymmetric metric tensor which is non-degenerate 119892
119894119896(119872)
where Det|119892119894119896| = 0 and 119892
119894119896= 119892119896119894 Note that the metric tensor
is chosen arbitrarily but in addition to conditions laid abovewe demand that the manifold was sufficiently smooth
This definition can be rewritten as follows invariantdifferential quadratic form 119892
119894119896119889119909119894
119889119909119896 determined on the
manifold and satisfying the conditions Det|119892119894119896| = 0 119892
119894119896= 119892119896119894
defines the geometry of RiemannAs a consequence of the invariance of the form
1198891199042
= 119892119894119896119889119909119894
119889119909119896
(4)
we find that the coefficients 119892119894119896are forming a tensor field
In this model for the arc length of the curve 119909119894 = 119909119894
(119905)119905 isin [119886 119887] sub 119877 is given by integral
119904 = int
119887
119886
radic119892119894119896119889119909119894119889119909119896
= int
119887
119886
radic119892119894119896(1199091(119905) 119909
119899(119905))
119889119909119894
(119905)
119889119905
119889119909119896
(119905)
119889119905
119889119905
(5)
The second scheme is a generalization of affine geometrythe geometry of affine connection Γ119894
119895119896(119872) that is based on 119899-
dimensional manifoldThe connection Γ
119894
119895119896(119872) is a geometric object on a man-
ifold and is subjected to the law of the transformation fromone coordinate system 119909
119894 to another 1199091198941015840
in the form of
Γ1198941015840
11989510158401198961015840 = Γ119894
119895119896
1205971199091198941015840
120597119909119894
120597119909119895
1205971199091198951015840
120597119909119896
1205971199091198961015840+
1205972
119909119894
1205971199091198951015840
1205971199091198961015840
1205971199091198941015840
120597119909119894 (6)
where the functions Γ119894119895119896are sufficiently smooth
Let along the curve 119909119894 = 119909119894
(119905) 119905 isin [119886 119887] sub 119877 is giventensor field 119860119894 = 119860
119894
(119905) if for each infinitesimal displacementtensor 119889119860119894(119905) coordinates is changing the law
119889119860119894
= minusΓ119894
119895119896119860119895
119889119909119896
(7)
then we say that the tensor 119860119894 is transported parallel to thecurve 119905
Depending on the physical investigated problem one oranother geometricmodel is choosing but as the internal logicand common sense requires that in the physical world thesetwo models coexist together and complement each otherThere is well-known result that in an arbitrary Riemannianspace can always construct a connection Γ
119894
119895119896(119872) An inter-
esting question is the uniqueness of such a construction Ingeneral such a construction Γ119894
119895119896is not unique but completely
natural (in terms of mathematics and physics to a greaterextent) there is the requirement that whenever two vectors119860119894and 119861119894 simultaneous parallel transported along a path (due tothe presence of the connection the transportation is defined)their scalar product does not change (the scalar product isdefined by metric) Mathematically this can be written as thevanishing differential
119889 (119892119894119896119860119894
119861119896
) = 0 (8)
If the requested coefficients Γ119894119895119896
are symmetric namelyΓ119894
119895119896= Γ119894
119896119895 then the connectivity is uniquely defined using a
metricAlways below we would not require the symmetry of
connection And so if the metric 119892119894119896
is defined then ageometric object Γ119894
119895119896subject to certain requirements but still
there is some arbitrariness in the choice of connectedness ofthe space namely we need to define a torsion tensor
119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895 (9)
then the geometric object Γ119894119895119896that generated the connection
is uniquely determined
Theorem 1 Suppose that a Riemannian space with the metric119892119894119896and in this space is given torsion tensor 119878119894
119895119896-antisymmetric
If demand 119889(119892119894119896119860119894
119861119896
) = 0 for arbitrary 119860119894 and 119861119896 then the
connection (geometric object that defines it) Γ119894119895119896
is uniquelydefined
4 Journal of Gravity
Proof It is pretty easy to see the truth of this statement itselfbut for further more importantly those symbols and valuesthat relate to the nature of the entered values
We rewrite 119889(119892119894119896119860119894
119861119896
) = 0 as (119892119894119896119897
minus 119892119898119896Γ119898
119894119897minus
119892119894119898Γ119898
119896119897)119860119894
119861119896
119889119909119897
= 0 due to the fact that 119889119860119894 = minusΓ119894
119901119897119860119901
119889119909119897
where Γ119898119894119897 unknown coefficients of connection are a geomet-
ric object and 119889119909119897 are the differentials of coordinates of a
point under infinitesimal displacement along the path 119892119894119896119897
equiv
(120597120597119909119897
)119892119894119896 Since119860119894 119861119896 119889119909119897 are arbitrary the equalities must
be identity relative to 119860119894 119861119896 119889119909119897 By circular permutation weobtain the system of equations
119892119894119896119897
= 119892119898119896Γ119898
119894119897+ 119892119894119898Γ119898
119896119897
119892119897119894119896
= 119892119898119894Γ119898
119897119896+ 119892119897119898Γ119898
119894119896
119892119896119897119894
= 119892119898119897Γ119898
119896119894+ 119892119896119898Γ119898
119897119894
(10)
Since the technique is similar to the classical one then wegive formulas without justification
119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
= 119892119898119896119878119898
119894119897+ 119892119898119897119878119898
119894119896+ 119892119894119898Γ119898
119896119897+ 119892119898119894Γ119898
119897119896 (11)
where 119878119898119894119897= Γ119898
119894119897minus Γ119898
119897119894is torsion tensor and we have
119892119894119898(Γ119898
119896119897+ Γ119898
119897119896)
= 119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894
Γ119901
119896119897+ Γ119901
119897119896
= 119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(12)
and complementing the obvious equation-definition Γ119901
119896119897minus
Γ119901
119897119896= 119878119901
119896119897 then we obtain
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(13)
Thenwe introduce the notation and from the last formulawe see that
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (14)
is geometric object and
119871119901
119896119897equiv
1
2
119878119901
119896119897+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (15)
is tensorThe geometric object Γ
119901
119896119897 which generate connection
space is completely determined by the tensors 119892119894119896and 119878
119898
119894119896
Therefore the connection Γ119901119896119897is the sum of a geometric object
119875119901
119896119897which is composed of derivatives of the metric tensor 119892
119894119896
and tensor 119871119901119896119897is compiled of 119892
119894119896and the tensor 119878119898
119896119897 namely
Γ119901
119896119897= 119875119901
119896119897+ 119871119901
119896119897 (16)
Remark 2 Tensor 119871119901
119896119897represents the sum of two tensors
symmetric (12)119892119901119894(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) and torsion (12)119878119901
119896119897
Remark 3 It is not difficult to prove the relation
Γ119901
119901119897=
1
2
119892119894119901119897119892119894119901
=
1
radic119892
120597radic119892
120597119909119897
where 119892 = det 1003816100381610038161003816119892119894119896
1003816100381610038161003816 (17)
The next step in building a geometric theory is the con-sideration of the parallel transport tensor-vector 119860119894 which isgiven by
119889119860119894
= minusΓ119894
119901119897119860119901
119889119909119897
(18)
where the coefficients Γ119894119901119897are the connection of space
As further arguments are similar to the classic and oftenrepeated them the presentation of intermediate results willwear schematic character
By a covariant derivative of 119906119894with respect to 119897 we mean
119906119894119897equiv 119906119894119897minus Γ119896
119894119897119906119896
119906119894
119897equiv 119906119894
119897+ Γ119894
119896119897119906119896
(19)
Then we consider the difference
119906119894119897minus 119906119897119894= 119906119894119897minus 119906119897119894minus 119878119896
119894119897119906119896 (20)
During the transition along a parallelogram in the imageto the original polygon the gap 119885
119896 is formed (breaking thecircuit) which can be estimated as follows up to the 2ndorder of smallness relative to sides of a parallelogram
119885119896
= 119878119896
119894119895119860119894
119861119895
1205912
(21)
It is the result of coagulation at the torsion tensor with vectorparties 119860119894120591 and 119861
119895
120591 that express geodetic displacement Thebasic geometric meaning of the torsion tensor is an estimateof the gap (up to 2nd order) at which the open loop shrinks ifthe torsion is zero then the gap will not infinitesimal secondbut higher order smallness
Next we consider the difference of the second orderderivatives
119906119894119897119896
minus 119906119894119896119897
= 119877119901
119896119897119894119906119901+ 119878119902
119896119897119906119894119902 (22)
where we identified
119877119901
119896119897119894equiv Γ119901
119894119896119897minus Γ119901
119894119897119896+ Γ119901
119902119897Γ119902
119894119896minus Γ119901
119902119896Γ119902
119894119897 (23)
119877119901
119896119897119894is curvature tensorSimilarly we have
119906119894
119897119896minus 119906119894
119896119897= minus119877119894
119896119897119901119906119901
+ 119878119902
119896119897119906119894
119902 (24)
Remark 4 Also possible and slightly different way of defini-tion a covariant derivative namely the absolute differential119863119860119894 is first determined using the formula
119863119860119894
equivsim 119889119860119894
+ Γ119894
119895119896119860119895
119889119909119896
= (119860119894
119896+ Γ119894
119895119896119860119895
) 119889119909119896
(25)
Journal of Gravity 5
Absolute differential can be defined as derivative coeffi-cients all the results obtained with this approach to the con-struction is identical with the analysis which has been madeabove Then more clearly we can assert that for any space topossess absolute parallelism it is a necessary and sufficientcondition that curvature tensor be identity vanishing (recallthat the space is called with absolute parallelism if the resultof the parallel transport of an arbitrary tensor-vector doesnot depend on the choice of path for all points of space) Theproof of this theorem is generally known we note only that itfollows from the formula
119863119863119860119894
minus 119863119863119860119894
= minus119877119894
119896119897119901119860119901 119889119909119896
119889119909119897
(26)
which we could get by folding (24) with 119889119909119896
119889119909119897
All the constructions outlined above are general in naturewithout specifying space further we will investigate thestructure of the tensor 119877119901
119894119896119897 So by definition of (23) we have
119877119901
119894119896119897= Γ119901
119897119894119896minus Γ119901
119897119896119894+ Γ119901
119902119896Γ119902
119897119894minus Γ119901
119902119894Γ119902
119897119896 (27)
Then we use (16) and obtain
119877119901
119894119896119897= 119875119901
119897119894119896+ 119871119901
119897119894119896minus 119875119901
119897119896119894+ 119871119901
119897119896119894+ (119875119901
119902119896+ 119871119901
119902119896) (119875119901
119897119894+ 119871119902
119897119894)
minus (119875119901
119902119894+ 119871119901
119902119894) (119875119901
119897119896+ 119871119902
119897119896)
= 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894minus 119875119901
119902119894119875119902
119897119896+ 119871119901
119897119894119896minus 119871119901
119897119896119894
+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
+ 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896
(28)
Next we introduce the notation
119875119901
119894119896119897equiv 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894+ 119875119901
119902119894119875119902
119897119896(29)
is a tensor like the Riemann curvature tensor composed ofthe metric tensor and its derivatives Consider the following
119885119901
119894119896119897equiv 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896(30)
is tensor and
119879119901
119894119896119897equiv 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894(31)
is tensorIf we take into account that119877119901
119894119896119897is tensor the last assertion
is obvious It is interesting to obtain this important result inanother way namely
119879119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894Γ119901
119902119896+ 119871119901
119902119894Γ119902
119897119896+ 119871119901
119897119902Γ119902
119894119896+ 119871119902
119897119896Γ119901
119902119894
minus 119871119901
119902119896Γ119902
119897119894minus 119871119901
119897119902Γ119902
119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119902
119902119896
minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894119871119901
119902119896+ 119871119901
119902119894119871119902
119897119896+ 119871119902
119897119896119871119901
119902119894
minus 119871119901
119902119896119871119902
119897119894+ 119871119901
119897119902119878119902
119894119896
(32)
is obviously the tensor because the absolute derivatives havetensor character
We introduce the notation
119872119901
119894119896119897equiv 119879119901
119894119896119897+ 119885119901
119894119896119897 (33)
then we obtain
119872119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119871119901
119897119902119878119902
119894119896+ 119871119901
119902119894119871119902
119897119896minus 119871119901
119902119896119871119902
119897119894 (34)
and in the new notation
119877119901
119894119896119897= 119875119901
119894119896119897+119872119901
119894119896119897 (35)
Formula (35) shows that the curvature tensor in generalcases can be represented as the sum of two tensors (suchrepresentation is not accidental it is associatedwith a physicaldescription of the field roughly speaking in the absence ofgravitational fields tensor119872119901
119894119896119897is not equal to zero) Although
the formula (34) gives a qualitative representation of thegeometric structure it is a little convenient since it reentersthe values of Γ119894
119895119896
Further we establish the equation which is similar toequation of Ricci-Jacobi
119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ Γ119901
119902119896119878119902
119897119894+ Γ119901
119902119896119878119902
119894119896+ Γ119901
119902119894119878119902
119896119897
= 119878119901
119894119896119897+ Γ119902
119894119897119878119901
119902119896+ Γ119902
119896119897119878119901
119894119902+ 119878119901
119896119897119894+ Γ119902
119896119894119878119901
119902119897+ Γ119902
119897119894119878119901
119896119902
+ 119878119901
119897119894119896+ Γ119902
119897119896119878119901
119902119894+ Γ119902
119894119896119878119901
119897119902
= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896
+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897
(36)
It is easy to prove the equations
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895= 0
119878119894
119894119901119878119901
119895119896= 0
(37)
and as a consequence we obtain the equation
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895+ 119878119894
119894119901119878119901
119895119896= 0 (38)
3 The Study of Geodesics inthe Space with Torsion
Formula (13) consists of three summand of various kindsThe sum (14) is a geometric object second valence its
components are converted by a formula similar to the one thattakes place for the connection
1198751199011015840
11989610158401198971015840= 119875119901
119896119897
1205971199091199011015840
120597119909119901
120597119909119896
1205971199091198961015840
120597119909119897
1205971199091198971015840+
1205972
119909119901
1205971199091198961015840
1205971199091198971015840
1205971199091199011015840
120597119909119901 (39)
but note that in this space this ldquoconnectionrdquo does not satisfythe ldquotorsion conditionrdquo (119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895)
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 3
arises from the physical features of the distribution of matterin space-time Roughly the same way as the masses leads tocurvature space-time electromagnetism leads to appearanceof torsion But on the other hand from the mathematicalpoint of view if we assume that the space-time embeddedin Euclidean space of higher dimension then the appearanceof torsion can be explained by violation of the smoothnessembedding Therefore we can conditionally determine thetorsion and curvature by violation of smoothness regardlessof the dimension and embedment
The aim of our paper is to study the property of metricspace with torsion and obtain analog of Einstein-Hilbertequation at such space
This paper is organized as follows In Section 2 somegeneral properties of structure of a metric space with torsionare discussed In Section 3 we present the study of geodesicsin the space with torsion These results can be used inldquogeodesic principlerdquo In Section 4 the theory of hypersurfacesin the space with torsion is dedicated In Section 5 weobtained some interesting relationships which is used inSection 6 In Section 6 we derive analog of Einstein-Hilbert(for electromagnetic and gravitational fields) in case of ametric space with torsion
The main natural assumption that is used below that ascalar product of two any vectors that is transporting parallelalong an arbitrary path does not change of the 20th century[2 8 11ndash13] and continues to develop so far [3ndash7 14ndash22]
2 Structure of a Metric Space with Torsion
There are many ways to represent the physical four-dimensional space where events of our reality are occurringFrom a mathematical point of view there are two possibleconceptions of space geometry which might be identifiedwith the physical space
The first scheme is a generalization of Euclidean geom-etry the geometry of the Riemannian metric that is 119899-dimensional manifold equipped with a field twice covariantsymmetric metric tensor which is non-degenerate 119892
119894119896(119872)
where Det|119892119894119896| = 0 and 119892
119894119896= 119892119896119894 Note that the metric tensor
is chosen arbitrarily but in addition to conditions laid abovewe demand that the manifold was sufficiently smooth
This definition can be rewritten as follows invariantdifferential quadratic form 119892
119894119896119889119909119894
119889119909119896 determined on the
manifold and satisfying the conditions Det|119892119894119896| = 0 119892
119894119896= 119892119896119894
defines the geometry of RiemannAs a consequence of the invariance of the form
1198891199042
= 119892119894119896119889119909119894
119889119909119896
(4)
we find that the coefficients 119892119894119896are forming a tensor field
In this model for the arc length of the curve 119909119894 = 119909119894
(119905)119905 isin [119886 119887] sub 119877 is given by integral
119904 = int
119887
119886
radic119892119894119896119889119909119894119889119909119896
= int
119887
119886
radic119892119894119896(1199091(119905) 119909
119899(119905))
119889119909119894
(119905)
119889119905
119889119909119896
(119905)
119889119905
119889119905
(5)
The second scheme is a generalization of affine geometrythe geometry of affine connection Γ119894
119895119896(119872) that is based on 119899-
dimensional manifoldThe connection Γ
119894
119895119896(119872) is a geometric object on a man-
ifold and is subjected to the law of the transformation fromone coordinate system 119909
119894 to another 1199091198941015840
in the form of
Γ1198941015840
11989510158401198961015840 = Γ119894
119895119896
1205971199091198941015840
120597119909119894
120597119909119895
1205971199091198951015840
120597119909119896
1205971199091198961015840+
1205972
119909119894
1205971199091198951015840
1205971199091198961015840
1205971199091198941015840
120597119909119894 (6)
where the functions Γ119894119895119896are sufficiently smooth
Let along the curve 119909119894 = 119909119894
(119905) 119905 isin [119886 119887] sub 119877 is giventensor field 119860119894 = 119860
119894
(119905) if for each infinitesimal displacementtensor 119889119860119894(119905) coordinates is changing the law
119889119860119894
= minusΓ119894
119895119896119860119895
119889119909119896
(7)
then we say that the tensor 119860119894 is transported parallel to thecurve 119905
Depending on the physical investigated problem one oranother geometricmodel is choosing but as the internal logicand common sense requires that in the physical world thesetwo models coexist together and complement each otherThere is well-known result that in an arbitrary Riemannianspace can always construct a connection Γ
119894
119895119896(119872) An inter-
esting question is the uniqueness of such a construction Ingeneral such a construction Γ119894
119895119896is not unique but completely
natural (in terms of mathematics and physics to a greaterextent) there is the requirement that whenever two vectors119860119894and 119861119894 simultaneous parallel transported along a path (due tothe presence of the connection the transportation is defined)their scalar product does not change (the scalar product isdefined by metric) Mathematically this can be written as thevanishing differential
119889 (119892119894119896119860119894
119861119896
) = 0 (8)
If the requested coefficients Γ119894119895119896
are symmetric namelyΓ119894
119895119896= Γ119894
119896119895 then the connectivity is uniquely defined using a
metricAlways below we would not require the symmetry of
connection And so if the metric 119892119894119896
is defined then ageometric object Γ119894
119895119896subject to certain requirements but still
there is some arbitrariness in the choice of connectedness ofthe space namely we need to define a torsion tensor
119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895 (9)
then the geometric object Γ119894119895119896that generated the connection
is uniquely determined
Theorem 1 Suppose that a Riemannian space with the metric119892119894119896and in this space is given torsion tensor 119878119894
119895119896-antisymmetric
If demand 119889(119892119894119896119860119894
119861119896
) = 0 for arbitrary 119860119894 and 119861119896 then the
connection (geometric object that defines it) Γ119894119895119896
is uniquelydefined
4 Journal of Gravity
Proof It is pretty easy to see the truth of this statement itselfbut for further more importantly those symbols and valuesthat relate to the nature of the entered values
We rewrite 119889(119892119894119896119860119894
119861119896
) = 0 as (119892119894119896119897
minus 119892119898119896Γ119898
119894119897minus
119892119894119898Γ119898
119896119897)119860119894
119861119896
119889119909119897
= 0 due to the fact that 119889119860119894 = minusΓ119894
119901119897119860119901
119889119909119897
where Γ119898119894119897 unknown coefficients of connection are a geomet-
ric object and 119889119909119897 are the differentials of coordinates of a
point under infinitesimal displacement along the path 119892119894119896119897
equiv
(120597120597119909119897
)119892119894119896 Since119860119894 119861119896 119889119909119897 are arbitrary the equalities must
be identity relative to 119860119894 119861119896 119889119909119897 By circular permutation weobtain the system of equations
119892119894119896119897
= 119892119898119896Γ119898
119894119897+ 119892119894119898Γ119898
119896119897
119892119897119894119896
= 119892119898119894Γ119898
119897119896+ 119892119897119898Γ119898
119894119896
119892119896119897119894
= 119892119898119897Γ119898
119896119894+ 119892119896119898Γ119898
119897119894
(10)
Since the technique is similar to the classical one then wegive formulas without justification
119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
= 119892119898119896119878119898
119894119897+ 119892119898119897119878119898
119894119896+ 119892119894119898Γ119898
119896119897+ 119892119898119894Γ119898
119897119896 (11)
where 119878119898119894119897= Γ119898
119894119897minus Γ119898
119897119894is torsion tensor and we have
119892119894119898(Γ119898
119896119897+ Γ119898
119897119896)
= 119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894
Γ119901
119896119897+ Γ119901
119897119896
= 119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(12)
and complementing the obvious equation-definition Γ119901
119896119897minus
Γ119901
119897119896= 119878119901
119896119897 then we obtain
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(13)
Thenwe introduce the notation and from the last formulawe see that
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (14)
is geometric object and
119871119901
119896119897equiv
1
2
119878119901
119896119897+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (15)
is tensorThe geometric object Γ
119901
119896119897 which generate connection
space is completely determined by the tensors 119892119894119896and 119878
119898
119894119896
Therefore the connection Γ119901119896119897is the sum of a geometric object
119875119901
119896119897which is composed of derivatives of the metric tensor 119892
119894119896
and tensor 119871119901119896119897is compiled of 119892
119894119896and the tensor 119878119898
119896119897 namely
Γ119901
119896119897= 119875119901
119896119897+ 119871119901
119896119897 (16)
Remark 2 Tensor 119871119901
119896119897represents the sum of two tensors
symmetric (12)119892119901119894(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) and torsion (12)119878119901
119896119897
Remark 3 It is not difficult to prove the relation
Γ119901
119901119897=
1
2
119892119894119901119897119892119894119901
=
1
radic119892
120597radic119892
120597119909119897
where 119892 = det 1003816100381610038161003816119892119894119896
1003816100381610038161003816 (17)
The next step in building a geometric theory is the con-sideration of the parallel transport tensor-vector 119860119894 which isgiven by
119889119860119894
= minusΓ119894
119901119897119860119901
119889119909119897
(18)
where the coefficients Γ119894119901119897are the connection of space
As further arguments are similar to the classic and oftenrepeated them the presentation of intermediate results willwear schematic character
By a covariant derivative of 119906119894with respect to 119897 we mean
119906119894119897equiv 119906119894119897minus Γ119896
119894119897119906119896
119906119894
119897equiv 119906119894
119897+ Γ119894
119896119897119906119896
(19)
Then we consider the difference
119906119894119897minus 119906119897119894= 119906119894119897minus 119906119897119894minus 119878119896
119894119897119906119896 (20)
During the transition along a parallelogram in the imageto the original polygon the gap 119885
119896 is formed (breaking thecircuit) which can be estimated as follows up to the 2ndorder of smallness relative to sides of a parallelogram
119885119896
= 119878119896
119894119895119860119894
119861119895
1205912
(21)
It is the result of coagulation at the torsion tensor with vectorparties 119860119894120591 and 119861
119895
120591 that express geodetic displacement Thebasic geometric meaning of the torsion tensor is an estimateof the gap (up to 2nd order) at which the open loop shrinks ifthe torsion is zero then the gap will not infinitesimal secondbut higher order smallness
Next we consider the difference of the second orderderivatives
119906119894119897119896
minus 119906119894119896119897
= 119877119901
119896119897119894119906119901+ 119878119902
119896119897119906119894119902 (22)
where we identified
119877119901
119896119897119894equiv Γ119901
119894119896119897minus Γ119901
119894119897119896+ Γ119901
119902119897Γ119902
119894119896minus Γ119901
119902119896Γ119902
119894119897 (23)
119877119901
119896119897119894is curvature tensorSimilarly we have
119906119894
119897119896minus 119906119894
119896119897= minus119877119894
119896119897119901119906119901
+ 119878119902
119896119897119906119894
119902 (24)
Remark 4 Also possible and slightly different way of defini-tion a covariant derivative namely the absolute differential119863119860119894 is first determined using the formula
119863119860119894
equivsim 119889119860119894
+ Γ119894
119895119896119860119895
119889119909119896
= (119860119894
119896+ Γ119894
119895119896119860119895
) 119889119909119896
(25)
Journal of Gravity 5
Absolute differential can be defined as derivative coeffi-cients all the results obtained with this approach to the con-struction is identical with the analysis which has been madeabove Then more clearly we can assert that for any space topossess absolute parallelism it is a necessary and sufficientcondition that curvature tensor be identity vanishing (recallthat the space is called with absolute parallelism if the resultof the parallel transport of an arbitrary tensor-vector doesnot depend on the choice of path for all points of space) Theproof of this theorem is generally known we note only that itfollows from the formula
119863119863119860119894
minus 119863119863119860119894
= minus119877119894
119896119897119901119860119901 119889119909119896
119889119909119897
(26)
which we could get by folding (24) with 119889119909119896
119889119909119897
All the constructions outlined above are general in naturewithout specifying space further we will investigate thestructure of the tensor 119877119901
119894119896119897 So by definition of (23) we have
119877119901
119894119896119897= Γ119901
119897119894119896minus Γ119901
119897119896119894+ Γ119901
119902119896Γ119902
119897119894minus Γ119901
119902119894Γ119902
119897119896 (27)
Then we use (16) and obtain
119877119901
119894119896119897= 119875119901
119897119894119896+ 119871119901
119897119894119896minus 119875119901
119897119896119894+ 119871119901
119897119896119894+ (119875119901
119902119896+ 119871119901
119902119896) (119875119901
119897119894+ 119871119902
119897119894)
minus (119875119901
119902119894+ 119871119901
119902119894) (119875119901
119897119896+ 119871119902
119897119896)
= 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894minus 119875119901
119902119894119875119902
119897119896+ 119871119901
119897119894119896minus 119871119901
119897119896119894
+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
+ 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896
(28)
Next we introduce the notation
119875119901
119894119896119897equiv 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894+ 119875119901
119902119894119875119902
119897119896(29)
is a tensor like the Riemann curvature tensor composed ofthe metric tensor and its derivatives Consider the following
119885119901
119894119896119897equiv 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896(30)
is tensor and
119879119901
119894119896119897equiv 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894(31)
is tensorIf we take into account that119877119901
119894119896119897is tensor the last assertion
is obvious It is interesting to obtain this important result inanother way namely
119879119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894Γ119901
119902119896+ 119871119901
119902119894Γ119902
119897119896+ 119871119901
119897119902Γ119902
119894119896+ 119871119902
119897119896Γ119901
119902119894
minus 119871119901
119902119896Γ119902
119897119894minus 119871119901
119897119902Γ119902
119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119902
119902119896
minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894119871119901
119902119896+ 119871119901
119902119894119871119902
119897119896+ 119871119902
119897119896119871119901
119902119894
minus 119871119901
119902119896119871119902
119897119894+ 119871119901
119897119902119878119902
119894119896
(32)
is obviously the tensor because the absolute derivatives havetensor character
We introduce the notation
119872119901
119894119896119897equiv 119879119901
119894119896119897+ 119885119901
119894119896119897 (33)
then we obtain
119872119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119871119901
119897119902119878119902
119894119896+ 119871119901
119902119894119871119902
119897119896minus 119871119901
119902119896119871119902
119897119894 (34)
and in the new notation
119877119901
119894119896119897= 119875119901
119894119896119897+119872119901
119894119896119897 (35)
Formula (35) shows that the curvature tensor in generalcases can be represented as the sum of two tensors (suchrepresentation is not accidental it is associatedwith a physicaldescription of the field roughly speaking in the absence ofgravitational fields tensor119872119901
119894119896119897is not equal to zero) Although
the formula (34) gives a qualitative representation of thegeometric structure it is a little convenient since it reentersthe values of Γ119894
119895119896
Further we establish the equation which is similar toequation of Ricci-Jacobi
119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ Γ119901
119902119896119878119902
119897119894+ Γ119901
119902119896119878119902
119894119896+ Γ119901
119902119894119878119902
119896119897
= 119878119901
119894119896119897+ Γ119902
119894119897119878119901
119902119896+ Γ119902
119896119897119878119901
119894119902+ 119878119901
119896119897119894+ Γ119902
119896119894119878119901
119902119897+ Γ119902
119897119894119878119901
119896119902
+ 119878119901
119897119894119896+ Γ119902
119897119896119878119901
119902119894+ Γ119902
119894119896119878119901
119897119902
= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896
+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897
(36)
It is easy to prove the equations
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895= 0
119878119894
119894119901119878119901
119895119896= 0
(37)
and as a consequence we obtain the equation
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895+ 119878119894
119894119901119878119901
119895119896= 0 (38)
3 The Study of Geodesics inthe Space with Torsion
Formula (13) consists of three summand of various kindsThe sum (14) is a geometric object second valence its
components are converted by a formula similar to the one thattakes place for the connection
1198751199011015840
11989610158401198971015840= 119875119901
119896119897
1205971199091199011015840
120597119909119901
120597119909119896
1205971199091198961015840
120597119909119897
1205971199091198971015840+
1205972
119909119901
1205971199091198961015840
1205971199091198971015840
1205971199091199011015840
120597119909119901 (39)
but note that in this space this ldquoconnectionrdquo does not satisfythe ldquotorsion conditionrdquo (119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895)
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
4 Journal of Gravity
Proof It is pretty easy to see the truth of this statement itselfbut for further more importantly those symbols and valuesthat relate to the nature of the entered values
We rewrite 119889(119892119894119896119860119894
119861119896
) = 0 as (119892119894119896119897
minus 119892119898119896Γ119898
119894119897minus
119892119894119898Γ119898
119896119897)119860119894
119861119896
119889119909119897
= 0 due to the fact that 119889119860119894 = minusΓ119894
119901119897119860119901
119889119909119897
where Γ119898119894119897 unknown coefficients of connection are a geomet-
ric object and 119889119909119897 are the differentials of coordinates of a
point under infinitesimal displacement along the path 119892119894119896119897
equiv
(120597120597119909119897
)119892119894119896 Since119860119894 119861119896 119889119909119897 are arbitrary the equalities must
be identity relative to 119860119894 119861119896 119889119909119897 By circular permutation weobtain the system of equations
119892119894119896119897
= 119892119898119896Γ119898
119894119897+ 119892119894119898Γ119898
119896119897
119892119897119894119896
= 119892119898119894Γ119898
119897119896+ 119892119897119898Γ119898
119894119896
119892119896119897119894
= 119892119898119897Γ119898
119896119894+ 119892119896119898Γ119898
119897119894
(10)
Since the technique is similar to the classical one then wegive formulas without justification
119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
= 119892119898119896119878119898
119894119897+ 119892119898119897119878119898
119894119896+ 119892119894119898Γ119898
119896119897+ 119892119898119894Γ119898
119897119896 (11)
where 119878119898119894119897= Γ119898
119894119897minus Γ119898
119897119894is torsion tensor and we have
119892119894119898(Γ119898
119896119897+ Γ119898
119897119896)
= 119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894
Γ119901
119896119897+ Γ119901
119897119896
= 119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(12)
and complementing the obvious equation-definition Γ119901
119896119897minus
Γ119901
119897119896= 119878119901
119896119897 then we obtain
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(13)
Thenwe introduce the notation and from the last formulawe see that
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (14)
is geometric object and
119871119901
119896119897equiv
1
2
119878119901
119896119897+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (15)
is tensorThe geometric object Γ
119901
119896119897 which generate connection
space is completely determined by the tensors 119892119894119896and 119878
119898
119894119896
Therefore the connection Γ119901119896119897is the sum of a geometric object
119875119901
119896119897which is composed of derivatives of the metric tensor 119892
119894119896
and tensor 119871119901119896119897is compiled of 119892
119894119896and the tensor 119878119898
119896119897 namely
Γ119901
119896119897= 119875119901
119896119897+ 119871119901
119896119897 (16)
Remark 2 Tensor 119871119901
119896119897represents the sum of two tensors
symmetric (12)119892119901119894(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) and torsion (12)119878119901
119896119897
Remark 3 It is not difficult to prove the relation
Γ119901
119901119897=
1
2
119892119894119901119897119892119894119901
=
1
radic119892
120597radic119892
120597119909119897
where 119892 = det 1003816100381610038161003816119892119894119896
1003816100381610038161003816 (17)
The next step in building a geometric theory is the con-sideration of the parallel transport tensor-vector 119860119894 which isgiven by
119889119860119894
= minusΓ119894
119901119897119860119901
119889119909119897
(18)
where the coefficients Γ119894119901119897are the connection of space
As further arguments are similar to the classic and oftenrepeated them the presentation of intermediate results willwear schematic character
By a covariant derivative of 119906119894with respect to 119897 we mean
119906119894119897equiv 119906119894119897minus Γ119896
119894119897119906119896
119906119894
119897equiv 119906119894
119897+ Γ119894
119896119897119906119896
(19)
Then we consider the difference
119906119894119897minus 119906119897119894= 119906119894119897minus 119906119897119894minus 119878119896
119894119897119906119896 (20)
During the transition along a parallelogram in the imageto the original polygon the gap 119885
119896 is formed (breaking thecircuit) which can be estimated as follows up to the 2ndorder of smallness relative to sides of a parallelogram
119885119896
= 119878119896
119894119895119860119894
119861119895
1205912
(21)
It is the result of coagulation at the torsion tensor with vectorparties 119860119894120591 and 119861
119895
120591 that express geodetic displacement Thebasic geometric meaning of the torsion tensor is an estimateof the gap (up to 2nd order) at which the open loop shrinks ifthe torsion is zero then the gap will not infinitesimal secondbut higher order smallness
Next we consider the difference of the second orderderivatives
119906119894119897119896
minus 119906119894119896119897
= 119877119901
119896119897119894119906119901+ 119878119902
119896119897119906119894119902 (22)
where we identified
119877119901
119896119897119894equiv Γ119901
119894119896119897minus Γ119901
119894119897119896+ Γ119901
119902119897Γ119902
119894119896minus Γ119901
119902119896Γ119902
119894119897 (23)
119877119901
119896119897119894is curvature tensorSimilarly we have
119906119894
119897119896minus 119906119894
119896119897= minus119877119894
119896119897119901119906119901
+ 119878119902
119896119897119906119894
119902 (24)
Remark 4 Also possible and slightly different way of defini-tion a covariant derivative namely the absolute differential119863119860119894 is first determined using the formula
119863119860119894
equivsim 119889119860119894
+ Γ119894
119895119896119860119895
119889119909119896
= (119860119894
119896+ Γ119894
119895119896119860119895
) 119889119909119896
(25)
Journal of Gravity 5
Absolute differential can be defined as derivative coeffi-cients all the results obtained with this approach to the con-struction is identical with the analysis which has been madeabove Then more clearly we can assert that for any space topossess absolute parallelism it is a necessary and sufficientcondition that curvature tensor be identity vanishing (recallthat the space is called with absolute parallelism if the resultof the parallel transport of an arbitrary tensor-vector doesnot depend on the choice of path for all points of space) Theproof of this theorem is generally known we note only that itfollows from the formula
119863119863119860119894
minus 119863119863119860119894
= minus119877119894
119896119897119901119860119901 119889119909119896
119889119909119897
(26)
which we could get by folding (24) with 119889119909119896
119889119909119897
All the constructions outlined above are general in naturewithout specifying space further we will investigate thestructure of the tensor 119877119901
119894119896119897 So by definition of (23) we have
119877119901
119894119896119897= Γ119901
119897119894119896minus Γ119901
119897119896119894+ Γ119901
119902119896Γ119902
119897119894minus Γ119901
119902119894Γ119902
119897119896 (27)
Then we use (16) and obtain
119877119901
119894119896119897= 119875119901
119897119894119896+ 119871119901
119897119894119896minus 119875119901
119897119896119894+ 119871119901
119897119896119894+ (119875119901
119902119896+ 119871119901
119902119896) (119875119901
119897119894+ 119871119902
119897119894)
minus (119875119901
119902119894+ 119871119901
119902119894) (119875119901
119897119896+ 119871119902
119897119896)
= 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894minus 119875119901
119902119894119875119902
119897119896+ 119871119901
119897119894119896minus 119871119901
119897119896119894
+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
+ 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896
(28)
Next we introduce the notation
119875119901
119894119896119897equiv 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894+ 119875119901
119902119894119875119902
119897119896(29)
is a tensor like the Riemann curvature tensor composed ofthe metric tensor and its derivatives Consider the following
119885119901
119894119896119897equiv 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896(30)
is tensor and
119879119901
119894119896119897equiv 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894(31)
is tensorIf we take into account that119877119901
119894119896119897is tensor the last assertion
is obvious It is interesting to obtain this important result inanother way namely
119879119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894Γ119901
119902119896+ 119871119901
119902119894Γ119902
119897119896+ 119871119901
119897119902Γ119902
119894119896+ 119871119902
119897119896Γ119901
119902119894
minus 119871119901
119902119896Γ119902
119897119894minus 119871119901
119897119902Γ119902
119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119902
119902119896
minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894119871119901
119902119896+ 119871119901
119902119894119871119902
119897119896+ 119871119902
119897119896119871119901
119902119894
minus 119871119901
119902119896119871119902
119897119894+ 119871119901
119897119902119878119902
119894119896
(32)
is obviously the tensor because the absolute derivatives havetensor character
We introduce the notation
119872119901
119894119896119897equiv 119879119901
119894119896119897+ 119885119901
119894119896119897 (33)
then we obtain
119872119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119871119901
119897119902119878119902
119894119896+ 119871119901
119902119894119871119902
119897119896minus 119871119901
119902119896119871119902
119897119894 (34)
and in the new notation
119877119901
119894119896119897= 119875119901
119894119896119897+119872119901
119894119896119897 (35)
Formula (35) shows that the curvature tensor in generalcases can be represented as the sum of two tensors (suchrepresentation is not accidental it is associatedwith a physicaldescription of the field roughly speaking in the absence ofgravitational fields tensor119872119901
119894119896119897is not equal to zero) Although
the formula (34) gives a qualitative representation of thegeometric structure it is a little convenient since it reentersthe values of Γ119894
119895119896
Further we establish the equation which is similar toequation of Ricci-Jacobi
119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ Γ119901
119902119896119878119902
119897119894+ Γ119901
119902119896119878119902
119894119896+ Γ119901
119902119894119878119902
119896119897
= 119878119901
119894119896119897+ Γ119902
119894119897119878119901
119902119896+ Γ119902
119896119897119878119901
119894119902+ 119878119901
119896119897119894+ Γ119902
119896119894119878119901
119902119897+ Γ119902
119897119894119878119901
119896119902
+ 119878119901
119897119894119896+ Γ119902
119897119896119878119901
119902119894+ Γ119902
119894119896119878119901
119897119902
= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896
+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897
(36)
It is easy to prove the equations
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895= 0
119878119894
119894119901119878119901
119895119896= 0
(37)
and as a consequence we obtain the equation
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895+ 119878119894
119894119901119878119901
119895119896= 0 (38)
3 The Study of Geodesics inthe Space with Torsion
Formula (13) consists of three summand of various kindsThe sum (14) is a geometric object second valence its
components are converted by a formula similar to the one thattakes place for the connection
1198751199011015840
11989610158401198971015840= 119875119901
119896119897
1205971199091199011015840
120597119909119901
120597119909119896
1205971199091198961015840
120597119909119897
1205971199091198971015840+
1205972
119909119901
1205971199091198961015840
1205971199091198971015840
1205971199091199011015840
120597119909119901 (39)
but note that in this space this ldquoconnectionrdquo does not satisfythe ldquotorsion conditionrdquo (119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895)
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 5
Absolute differential can be defined as derivative coeffi-cients all the results obtained with this approach to the con-struction is identical with the analysis which has been madeabove Then more clearly we can assert that for any space topossess absolute parallelism it is a necessary and sufficientcondition that curvature tensor be identity vanishing (recallthat the space is called with absolute parallelism if the resultof the parallel transport of an arbitrary tensor-vector doesnot depend on the choice of path for all points of space) Theproof of this theorem is generally known we note only that itfollows from the formula
119863119863119860119894
minus 119863119863119860119894
= minus119877119894
119896119897119901119860119901 119889119909119896
119889119909119897
(26)
which we could get by folding (24) with 119889119909119896
119889119909119897
All the constructions outlined above are general in naturewithout specifying space further we will investigate thestructure of the tensor 119877119901
119894119896119897 So by definition of (23) we have
119877119901
119894119896119897= Γ119901
119897119894119896minus Γ119901
119897119896119894+ Γ119901
119902119896Γ119902
119897119894minus Γ119901
119902119894Γ119902
119897119896 (27)
Then we use (16) and obtain
119877119901
119894119896119897= 119875119901
119897119894119896+ 119871119901
119897119894119896minus 119875119901
119897119896119894+ 119871119901
119897119896119894+ (119875119901
119902119896+ 119871119901
119902119896) (119875119901
119897119894+ 119871119902
119897119894)
minus (119875119901
119902119894+ 119871119901
119902119894) (119875119901
119897119896+ 119871119902
119897119896)
= 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894minus 119875119901
119902119894119875119902
119897119896+ 119871119901
119897119894119896minus 119871119901
119897119896119894
+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
+ 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896
(28)
Next we introduce the notation
119875119901
119894119896119897equiv 119875119901
119897119894119896minus 119875119901
119897119896119894+ 119875119901
119902119896119875119902
119897119894+ 119875119901
119902119894119875119902
119897119896(29)
is a tensor like the Riemann curvature tensor composed ofthe metric tensor and its derivatives Consider the following
119885119901
119894119896119897equiv 119871119901
119902119896119871119902
119897119894minus 119871119901
119902119894119871119902
119897119896(30)
is tensor and
119879119901
119894119896119897equiv 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119901
119902119896minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894(31)
is tensorIf we take into account that119877119901
119894119896119897is tensor the last assertion
is obvious It is interesting to obtain this important result inanother way namely
119879119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894Γ119901
119902119896+ 119871119901
119902119894Γ119902
119897119896+ 119871119901
119897119902Γ119902
119894119896+ 119871119902
119897119896Γ119901
119902119894
minus 119871119901
119902119896Γ119902
119897119894minus 119871119901
119897119902Γ119902
119896119894+ 119875119901
119902119896119871119902
119897119894+ 119875119902
119897119894119871119902
119902119896
minus 119875119901
119902119894119871119902
119897119896minus 119875119902
119897119896119871119901
119902119894
= 119871119901
119897119894119896minus 119871119901
119897119896119894minus 119871119902
119897119894119871119901
119902119896+ 119871119901
119902119894119871119902
119897119896+ 119871119902
119897119896119871119901
119902119894
minus 119871119901
119902119896119871119902
119897119894+ 119871119901
119897119902119878119902
119894119896
(32)
is obviously the tensor because the absolute derivatives havetensor character
We introduce the notation
119872119901
119894119896119897equiv 119879119901
119894119896119897+ 119885119901
119894119896119897 (33)
then we obtain
119872119901
119894119896119897= 119871119901
119897119894119896minus 119871119901
119897119896119894+ 119871119901
119897119902119878119902
119894119896+ 119871119901
119902119894119871119902
119897119896minus 119871119901
119902119896119871119902
119897119894 (34)
and in the new notation
119877119901
119894119896119897= 119875119901
119894119896119897+119872119901
119894119896119897 (35)
Formula (35) shows that the curvature tensor in generalcases can be represented as the sum of two tensors (suchrepresentation is not accidental it is associatedwith a physicaldescription of the field roughly speaking in the absence ofgravitational fields tensor119872119901
119894119896119897is not equal to zero) Although
the formula (34) gives a qualitative representation of thegeometric structure it is a little convenient since it reentersthe values of Γ119894
119895119896
Further we establish the equation which is similar toequation of Ricci-Jacobi
119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ Γ119901
119902119896119878119902
119897119894+ Γ119901
119902119896119878119902
119894119896+ Γ119901
119902119894119878119902
119896119897
= 119878119901
119894119896119897+ Γ119902
119894119897119878119901
119902119896+ Γ119902
119896119897119878119901
119894119902+ 119878119901
119896119897119894+ Γ119902
119896119894119878119901
119902119897+ Γ119902
119897119894119878119901
119896119902
+ 119878119901
119897119894119896+ Γ119902
119897119896119878119901
119902119894+ Γ119902
119894119896119878119901
119897119902
= 119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896
+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897
(36)
It is easy to prove the equations
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895= 0
119878119894
119894119901119878119901
119895119896= 0
(37)
and as a consequence we obtain the equation
119878119894
119895119901119878119901
119896119894+ 119878119894
119896119901119878119901
119894119895+ 119878119894
119894119901119878119901
119895119896= 0 (38)
3 The Study of Geodesics inthe Space with Torsion
Formula (13) consists of three summand of various kindsThe sum (14) is a geometric object second valence its
components are converted by a formula similar to the one thattakes place for the connection
1198751199011015840
11989610158401198971015840= 119875119901
119896119897
1205971199091199011015840
120597119909119901
120597119909119896
1205971199091198961015840
120597119909119897
1205971199091198971015840+
1205972
119909119901
1205971199091198961015840
1205971199091198971015840
1205971199091199011015840
120597119909119901 (39)
but note that in this space this ldquoconnectionrdquo does not satisfythe ldquotorsion conditionrdquo (119878119894
119895119896equiv Γ119894
119895119896minus Γ119894
119896119895)
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
6 Journal of Gravity
Value (12)119878119901119896119897 this tensor can be to be used as a ldquocon-
nectionrdquo but then this ldquoconnectionrdquo is not used to satisfy thecompatibility condition
Value
119872119901
119896119897equiv
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) (40)
is a tensor that expresses the combined effect of the metricand torsion
First of all it is easy to show that the element (12)119878119901119896119897does
not affect the geodesic since by the asymmetry of the torsionit is not included in the equations of geodesic lines
1198892
119909119896
1198891205912= minusΓ119896
119894119895
119889119909119894
119889120591
119889119909119895
119889120591
(41)
that is (6) will be determined only by sum 119875119901
119896119897+119872119901
119896119897 In (6) 120591
is the canonical parameter that is the vector 119889119909119894119889120591 is paralleltransported vector For no isotropic geodesic the length of arc119904 is a canonical parameter for geodetic related to 119904 takes placein differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(42)
Definition 5 The line is called a geodesic if any tangent to thisline at some point vector remains tangent to it at the paralleltransport along it
Theorem 6 The equations of geodesic lines in space 119884119899 is
determined by the geometric object in form of the sum
119875119901
119896119897+119872119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894)
+
1
2
119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
(43)
In case classical Riemannian space geodetic lines haveknown extreme properties in this case analogical propertiesof geodesic require additional research
Thus for a no isotropic geodesic with canonical parame-ter arc length 119904 we have differential equations
1198892
119909119896
1198891199042
= minusΓ119896
119894119895
119889119909119894
119889119904
119889119909119895
119889119904
(44)
Consider the problem of calculating the variation of thearc length
Let a nonisotropic curve 119909119894
= 119909119894
(119905) 119905 isin [1199051 1199052] We
calculate the variation of length 120575119878 of the curve 119878
120575119878 = int
1199052
1199051
120575radic119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
119889119905
= int
1199052
1199051
120575 (119892119894119895(119889119909119894
119889119905) (119889119909119895
119889119905))
2radic119892119894119895(119889119909119894119889119905) (119889119909
119895119889119905)
119889119905
120575 (119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 119892119894119895119863
119889119909119894
119889119905
119889119909119895
119889119905
+ 119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
= 2119892119894119895
119889119909119894
119889119905
119863
119889119909119895
119889119905
119863
119889119909119895
119889119905
= 120575
119889119909119895
119889119905
+ Γ119895
119901119896
119889119909119901
119889119905
120575119909119896
119863
120575119909119895
119889119905
=
119889
119889119905
120575119909119895
+ Γ119895
119896119901
119889119909119896
119889119905
120575119909119901
119863
119889119909119895
119889119905
= 119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
(45)
where119863 denotes the absolute differential through the param-eter curves of the family at a constant value 119905 and119863 is absolutedifferential charge small displacement 119889119905 curve at a constantparameter of the family then
120575(119892119894119895
119889119909119894
119889119905
119889119909119895
119889119905
) = 2119892119894119895
119889119909119894
119889119905
(119863
120575119909119895
119889119905
+ 119878119895
119901119896
119889119909119901
119889119905
120575119909119896
)
120575119904 = int
1199052
1199051
119892119894119895
119889119909119894
119889119904
119863120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
= int
1199052
1199051
119863(119892119894119895
119889119909119894
119889119904
120575119909119895
)
minus int
1199052
1199051
119892119894119895119863
119889119909119894
119889119904
120575119909119895
+ int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
(46)
if the ends of the variable curve are fixed then
120575119904 = int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119905
120575119909119895
) (47)
if the considered curve has fixed length (analytically 120575119904 = 0)then we obtain
int
1199052
1199051
(119892119894119895119878119895
119901119896
119889119909119894
119889119904
119889119909119901
120575119909119896
minus 119892119894119895119863
119889119909119894
119889119904
120575119909119895
) = 0 (48)
Using the fundamental lemma calculus of variations itfollows that
119892119894119896119878119896
119901119895
119889119909119894
119889119904
119889119909119901
minus 119892119894119895119863
119889119909119894
119889119904
= 0 (49)
This equation means that the tangent vector 120585119902 of thecurve is transported according to the law119863120585
119902
119892119895119902
119892119894119896119878119896
119901119895120585119894
119889119909119901
which means that 119904 is not geodesic curve Conversely thevariation of the length of the geodesic lines is
120575119904 = int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(50)
Properties in the new geodesic geometry defined bymeans of two tensors 119892
119894119896and 119878
119895
119894119896differ greatly from similar
properties in Riemann geometry The theorem is proved
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 7
Theorem 7 In order a line in space 119884119899 (that is generated by119892119894119896and 119878
119895
119894119896) was geodetic it is necessary and sufficient that a
variation of the arc this line was equaled
int
1199052
1199051
119892119894119895119878119895
119901119896
119889119909119894
119889119905
119889119909119901
120575119909119896
(51)
Consequence In the case of spaces with affine connectionis known that there is a gap (breach) of closure during thetransition from the original to the image and vice versa inthe case of an infinitely small contour (determined up to thesecond order relative to 120591) If you specify the torsion tensor 119878119896
119894119895
at the corresponding point then if this gap is denoted by Ψ119896then Ψ
119896
= 119878119896
119894119895119860119894
119861119895
1205912 where the parallelogram 119860
119894
120591 and 119861119895
120591
shrink to a point at 120591 rarr 0 In this case we have the formulafor square of the length |Ψ|2 = 119892
119901119902119878119901
119894119895119860119894
119861119895
119878119902
119896119897119860119896
119861119897
1205914
Theorem 8 Let classical Riemannian space be given (Rie-mannian manifold with Riemannian metric tensor 119892
119894119896) with
the connection 119875119896
119894119895 connection Riemannian space Let 119884119899 be
the space generated jointly and agreed on by metric 119892119894119896and
torsion 119878119896119894119895tensors together with connection Γ119896
119894119895 To coincide the
geodesics in classical Riemannian space with the geodesics inspace 119884119899 it is necessary and sufficient that the connections 119875119896
119894119895
and Γ119896119894119895be differed by tensor
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (52)
that is
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (53)
Remark 9 The last summand in
Γ119896
119894119895=
1
2
119892119896119901
(119892119894119901119895119897
+ 119892119895119901119894
minus 119892119894119895119901
+ 119892119894119898119878119898
119895119901+ 119892119895119898119878119898
119894119901)
+
1
2
119878119896
119894119895
(54)
that is (12)119878119896119894119895 is irrelevant (it is clear)
Proof The necessity means that if the geodesics in classicalRiemannian space coincide with the geodesics in space 119884119899then
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (55)
Let the vector 119860119894 be tangent to any geodesic and 119860119894 is
parallel transported to the connection Γ119901119896119897
Γ119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894
+ 119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894)
+
1
2
119878119901
119896119897
(56)
and a vector 119861119894 is parallel transported to the connection 119875119901119896119897
119875119901
119896119897=
1
2
119892119901119894
(119892119894119896119897
+ 119892119897119894119896
minus 119892119896119897119894) (57)
It was found that the first geodesic connections coincidewith geodesic connection 119875
119901
119896119897+ 119872119901
119896119897 here 119872119896
119894119895is an arbitrary
symmetric tensor Since both vectors are tangential then
119861119894
= 119886119860119894
(58)
where the coefficient 119886 is variable and 119886 = 0 Tensor-vector119860119894 is given by
119889119860119896
= minus (119875119896
119894119895+119872119896
119894119895)119860119894
119889119909119895
119889119861119896
= minus119875119896
119894119895119861119894
119889119909119895
(59)
Then we have
119860119896
119889119886 + 119886119889119860119896
= minus119875119896
119894119895119886119860119894
119889119909119895
(60)
119860119896
119889119886
119886
= 119872119896
119894119895119860119894
119889119909119895
(61)
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 is thecanonical parameter relative to the connection 119875119901
119896119897+119872119901
119896119897
Then after dividing by 119889120591 we have
119889 ln 119886119889120591
119860119896
= 119872119896
119894119895119860119894
119860119895
(62)
Since geodesic lines can be carried out through any point andin any direction then equality must be true at any point andfor any vector119860119894 the functional dependence of the point anddirection obviously there
The last equality is multiplied by 119860119897 and alternated by 119896and 119897
119860119897
119872119896
119894119895119860119894
119860119895
minus 119860119896
119872119897
119894119895119860119894
119860119895
= 0 (63)
or
120575119897
119898119872119896
119894119895119860119894
119860119895
119860119898
minus 120575119896
119898119872119897
119894119895119860119894
119860119895
119860119898
= 0 (64)
where we denote 120575119897
119898= 119892119898119901119892119901119897 This equation must hold
identically with respect to vectors 1198601 119860119899 consequentlyafter summed similar summands all the coefficients of thecubic form must vanish We compute the total coefficient
120575119897
119898119872119896
119894119895minus 120575119896
119898119872119897
119894119895+ 120575119897
119894119872119896
119895119898minus 120575119896
119894119872119897
119895119898+ 120575119897
119895119872119896
119898119894minus 120575119896
119895119872119897
119898119894
= 0
(65)
Then we contracted tensor by indices 119897 and 119895 Since 120575119894119894= 119899 we
have
119872119896
119894119895=
1
119899 + 1
(120575119896
119894119872119897
119895119897+ 120575119896
119895119872119897
119894119897) (66)
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
8 Journal of Gravity
All calculations above are correct for any symmetricaltensor119872119896
119894119895
All calculations presented above do not take intoaccount the specificity of the tensor 119872119896
119894119895 then let 119872119901
119896119897equiv
(12)119892119901119894
(119892119896119898119878119898
119897119894+ 119892119897119898119878119898
119896119894) then119872119897
119896119897equiv (12)119878
119897
119896119897
We obtained
Γ119896
119894119895minus 119875119896
119894119895=
1
2
1
119899 + 1
(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) (67)
The necessity is proof
Remark 10 In the beginning we considered more generalsituation and then specified tensor119872119896
119894119895
We prove the sufficiency We assume that Γ119896119894119895minus 119875119896
119894119895=
(12)(1(119899 + 1))(120575119896
119894119878119897
119895119897+ 120575119896
119895119878119897
119894119897) we have to show that the
geodesics coincideWe again use (61) we have 119860119896119889119886119886 = 119872
119896
119894119895119860119894
119889119909119895
Tangent vector119860119894 can be written119860119894 = 119889119909119894
119889120591 where 120591 isthe canonical parameter relative to the connection Γ119896
119894119895
Since along the curve 119878119897119894119897119860119894 there is a definite function of
the parameter 120591 then it will find ln 119886 after integration withup to a constant 119886 but only up to a constant factor (we canintegrate) Therefore the vector is found to be 119861119894 = 119886119860
119894 andall geodesics coincide The sufficiency is proved The theoremis proved
4 The Theory of Hypersurfaces
First we formulate a well-known classical result on thederivation formulas of hypersurface in Riemannian torsion-free space where the connection is uniquely generated by themetric
Formal statement of Gauss-Codazzi equations in spacewith symmetrical connection 119878119896
119894119895= 0 Assume that 119894 119872 sub 119875
is 119899-dimensional embedded submanifold of a Riemannianmanifold 119875 of dimension 119899 + 119901 There is a natural inclusionof the tangent bundle of119872 into that of 119875 by the pushforwardand the cokernel is the normal bundle of 119872 0 rarr 119879
119909rarr
119879119909119875|119872
rarr 119879perp
119909119872 rarr 0 The metric splits this short exact
sequence and so 119879119875|119872= 119879119872 oplus 119879
perp
119872The Levi-Civita connection nabla
1015840 of 119875 decomposes intotangential and normal components For each 119883 isin 119879119872 andvector field 119884 on 119872 nabla1015840
119883119884 = 119879(nabla
1015840
119883119884) + perp (nabla
1015840
119883119884) Let it be
nabla119883119884 = 119879(nabla
1015840
119883119884) 120572(119883 119884) = perp (nabla
119883119884) Gaussrsquo formula asserts
that nabla119883is Levi-Civita connection on119872 and 120572 is a symmetric
form with values into the normal bundle It is also referred toas the second fundamental form An immediate corollary isthe Gauss equation For119883119884 119885119882 isin 119879119872
⟨1198771015840
(119883 119884)119885119882⟩ = ⟨119877 (119883 119884)119885119882⟩
+ ⟨120572 (119883 119885) 120572 (119884119882)⟩
minus ⟨120572 (119884 119885) 120572 (119883119882)⟩
(68)
where 1198771015840 is the Riemann curvature tensor of 119875 and 119877 is thatof119872
There are thus a pair of connections nabla defined on thetangent bundle of 119872 and 119863 defined on the normal bundleof 119872 These combine to form a connection on any tensorproduct of copies of 119879119872 and 119879 perp 119872 In particular theydefined the covariant derivative of 120572
(nabla10158401015840
119883120572) (119884 119885) = 119863
119883(120572 (119884 119885)) minus 120572 (nabla
119883119884119885)
minus 120572 (119884 nabla119883119885)
(69)
Codazzi equation is
perp ⟨1198771015840
(119883 119884)119885⟩ = (nabla10158401015840
119883120572) (119884 119885) minus (nabla
10158401015840
119884120572) (119883 119885) (70)
The above formulas also hold for immersions becauseevery immersion is in particular a local embedding Theimportant assumption in the theory stated above is thatconnection is symmetrical so analog of second form ofhypersurface is symmetrical and we could obtain the deriva-tion formulas of hypersurface
In our case we do not assume any conditions aboutthe connection symmetry so below we study much morecomplicated problem
A similar manner as in the Riemannian space can bedeveloped as a theory of hypersurfaces so in themetric spacewith torsion a theory of hypersurfaces can be constructedBut due to the presence of torsion in these cases there is asignificant difference For example the derivation equations(analog PetersonCodazzi equations) take amore complicatedform in which there are new summands which are caused bythe presence of torsion in the space
We make a study of the hypersurfaces 119884119899minus1 in a metricspace with torsion119884119899 We are assuming that the hypersurfaceis defined by a system of equations
119909119894
= 119909119894
(1199101
119910119899minus1
) (71)
and the rank of the matrix [120597119909119894120597119910120572] equals 119899 minus 1 The metrictensor of hypersurface 119884119899minus1 is given by
119886120572120573
= 119892119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
(72)
Then we obtain the formula for tensor of torsion 119879120574
120572120573of
hypersurface 119884119899minus1 (assuming that functions 119909119894(1199101 119910119899minus1)are smooth enough) Let 119866120572
120573120574be the connection of 119884119899minus1 and
we assume that 119866120572120573120574
express via metric 119886120572120573
and torsion 119879120574
120572120573
similarly to as the connection Γ119896119894119895express by means of 119892
119894119895and
119878119896
119894119895 we have
119866120572
120573120574
=
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
(73)
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 9
119866120572
120573120574is geometric object and is subjected to the law of the
transformation from one coordinate system 119906120572 to another 119906120572
1015840
by the formula
119866120572
120573120574= 1198661205721015840
12057310158401205741015840
120597119906120572
1205971199061205721015840
1205971199061205731015840
120597119906120573
1205971199061205741015840
120597119906120574+
120597119906120572
1205971199061205721015840
1205972
1199061205721015840
120597119906120573120597119906120574
(74)
Then we assume that connection 119866120572
120573120574of 119884119899minus1 are associated
with connection Γ119896119894119895of 119884119899 mean of formula
119866120572
120573120574
120597119909119896
120597119906120572= Γ119896
119894119895
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(75)
We obtain
1
2
(119886120572120578
(119886120573120578120574
+ 119886120574120578120573
+ 119886120573120574120578
+ 119886120573120583119879120583
120574120578+ 119886120574120583119879120583
120573120578) + 119879120572
120574120573)
120597119909119896
120597119906120572
=
1
2
(119892119896119899
(119892119899119894119895
+ 119892119899119895119894
minus 119892119894119895119899
+ 119892119894119898119878119898
119895119899+ 119892119895119898119878119898
119894119899) + 119878119896
119894119895)
times
120597119909119894
120597119906120573
120597119909119895
120597119906120574+
1205972
119909119896
120597119906120573120597119906120574
(76)
By permuting indices we have the next formula for thetorsion tensor of hypersurface 119884119899minus1
119879120574
120572120573= 119886120574120578
119892119901119902119878119901
119894119895
120597119909119894
120597119910120572
120597119909119895
120597119910120573
120597119909119902
120597119910120578 (77)
using tensors 119886120572120573
and 119879120574
120572120573 both metric and torsion we can
explore the geometry of the space hypersurface 119884119899minus1The connection of119884119899minus1 will be determined by the formula
(73)Below we use the mixed tensors values which enumer-
ated two types of indices while Latin indices refer to thecontaining space 119884
119899 and are responsive to the coordinatetransformation 119909
119894 and Greek indices belong to the spacehypersurface 119884
119899minus1 and are responsive to the coordinatetransformation 119910120572
The index 119894 is not responsive to the coordinate 119910120572
transformation into 119884119899minus1 and the index 120572 does not respondto the coordinate 119909
119894 transformation in 119884119899 For example
the formula to calculate the covariant derivative of a mixedtensor
119860119894120572
119895120573120574= 119860119894120572
119895120573120574+ Γ119894
119901119896119860119901120572
119895120573
120597119909119896
120597119910120574minus Γ119901
119895119902119860119894120572
119901120573
120597119909119902
120597119910120574+ 119866120572
120578120574119860119894120578
119895120573
minus 119866120578
120573120574119860119894120572
119895120578
(78)
The direct calculations lead us to formulas
119906119894120572120573
minus 119906119894120573120572
= 119877119901
119896119897119894119906119901
120597119909119897
120597119910120572
120597119909119896
120597119910120573
+ 119878119902
119896119897119906119894119902
120597119909119897
120597119910120572
120597119909119896
120597119910120573
119906120574120572120573
minus 119906120574120573120572
= 119877120578
120573120572120574119906120578+ 119879120578
120573120572119906120574120578
(79)
where 119877120578
120573120572120574is curvature tensor of space 119884
119899minus1 compiled byusing the components of connection 119866120572
120573120574
A further aim of our study is to obtain some analogs ofPeterson-Kodachi equations To do this consider the systemof values
120585119894
120572=
120597119909119894
120597119910120572 (80)
At each point of the hypersurface 119884119899minus1 we can build
rapper consisting of the vectors
120585119894
1 120585
119894
119899minus1 ]119894 (81)
where 120585119894
1 120585
119894
119899minus1are linearly independent tangent vectors
and ]119894 is normal vector defined since the metric and connec-tivity agreed
Next we act formally the idea is the same as in theclassical case and we will indicate significant new momentsWe compute the derivative of the mixed tensors 120585119894
120572
120585119894
120572120574= 120585119894
120572120574+ Γ119894
119901119902120585119901
120572
120597119909119902
120597119910120574minus 119866120578
120572120574120585119894
120578 (82)
In contrast to the case of torsion-free connection we havethe equality
120585119894
120572120574minus 120585119894
120574120572= 119878119894
119901119902120585119901
120572120585119902
120574+ 119879120578
120574120572120585119894
120578 (83)
Next we permute the indices in equation
0 = 119886120572120573120574
= (119892119894119895120585119894
120572120585119895
120573)
120574
= 119892119894119895120585119894
120572120574120585119895
120573+ 119892119894119895120585119894
120572120585119895
120573120574 (84)
we obtain
119892119894119895120585119894
120572120574120585119895
120573= 0 (85)
Hence we can write a decomposition
120585119894
120573120572= 120587120572120573]119894 (86)
Remark 11 120587120572120573
is tensor which is similar to the secondfundamental tensor of hypersurfaces 119884119899minus1 but its structurein this space is substantially different from the case ofRiemannian spaces with zero torsion
Then we have obtained (by differentiating 119892119894119895]119894120585119895120572= 0 by
120574) the formula
119892119894119895]119894120574120585119894
120572= minus120587120574120572 (87)
Similarly by differentiating 119892119894119895]119894]119895 = 1 by 120574 we obtain
]119894120574= minus119886120578120583
120587120583120574120585119894
120578 (88)
Formulae (86) and (88) characterize the change of vectorsin the small accompanying frame relative to this frame itself
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
10 Journal of Gravity
Further we obtain
120585119894
120573120594120582minus 120585119894
120573120582120594= minus119877119894
119896119897119901120585119896
120582120585119897
120594120585119901
120573+ 119877120590
120582120594120573120585119894
120590+ 119879120590
120582120594120585119894
120573120590
= (120587120594120573120582
minus 120587120582120573120594
) ]119894
minus (120587120594120573120587120578120582119886120578120590
minus 120587120582120573120587120578120594119886120578120590
) 120585119894
120590
(89)
Equation (89) is multiplied by 119892119894119895120585119895
120572 we have
119877120572120582120594120573
= 119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573120585119894
120572minus (120587120594120573120587120572120582
minus 120587120582120573120587120572120594) (90)
Similarly we derive a formula
]119894120594120582
minus ]119894120582120594
= minus119877119894
119896119897119901120585119896
120582120585119897
120594]119901 + 119879
120590
120582120594]119894120590
= (120587120578120582120594
119886120578120590
minus 120587120578120594120582
119886120578120590
) 120585119894
120590
(91)
We contract (89) with 119892119894119895]119895 then
minus119877119894119896119897119901
120585119896
120582120585119897
120594120585119901
120573]119894 + 119879
120590
120582120594120587120590120573
= 120587120594120573120582
minus 120587120582120573120594
(92)
Formula (91) is multiplied by 119892119894119895120585119895
120572 we concluded that
minus119877119894119896119897119901
120585119896
120582120585119897
120594]119901120585119894120572+ 119879120590
120582120594120587120572120590
= 120587120572120582120594
minus 120587120572120594120582
(93)
Remark 12 If (91) contract with 119892119894119895]119895 then we obtain identi-
cally zero equals zeroThus we have the two types of formulas Formula (90)
does not contain the torsion tensor explicitly but it is countedin the tensor 120587
120572120573 In the formula (91) the torsion tensor
of the hypersurface is present explicitly and in the formof coefficients of 120587
120572120573and appears in the calculation of the
covariant derivative
5 Establish Some Important Relationships
We consider the equation
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
+ 119878119905
119902119901119878119894
119895119896119905
(94)
or
119878119894
119895119896119901119902minus 119878119894
119895119896119902119901minus 119878119905
119902119901119878119894
119895119896119905
= 119877119905
119902119901119895119878119894
119905119896+ 119877119905
119902119901119896119878119894
119895119905minus 119877119894
119902119901119905119878119905
119895119896
(95)
we contract these tensors by indices 119894 119902 then the left side ofthis equation can be transformed into
119878119894
119895119896119901119894minus 119878119894
119895119896119894119901minus 119878119905
119894119901119878119894
119895119896119905
= (119878119894
119895119896119901minus 119878119894
119902119901119878119902
119895119896)
119894
minus 119878119894
119895119896119894119901minus 119878119894
119901119902119894119878119902
119895119896
(96)
then we contract this equation by indices 119896 119901 and raising theindex 119895 we obtained
(119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896)119894
minus 119892119896119901
119892119895119904
119878119894
119904119896119894119901minus 119892119896119901
119892119895119904
119878119894
119901119902119894119878119902
119904119896
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905
minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(97)
we introduce the notation
119867119895119894
= 119892119896119901
119892119895119904
119878119894
119904119896119901minus 119892119896119901
119892119895119904
119878119894
119902119901119878119902
119904119896
119865119895119901
= 119892119896119901
119892119895119904
119878119894
119904119896119894
(98)
Then without any loss of generality we obtain therelations
119867119895119894
119894minus 119865119895119894
119894minus 119892119896119901
119892119895119904
119878119902
119904119896119865119901119902
= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
(99)
where 119865119901119902
= 119878119894
119901119902119894
Suppose now that the identity Ricci-Jacobi run in astandard form 119877119901
119894119896119897+ 119877119901
119896119897119894+ 119877119901
119897119894119896= 0 hence
119878119901
119894119896119897+ 119878119901
119896119897119894+ 119878119901
119897119894119896+ 119878119901
119897119902119878119902
119894119896+ 119878119901
119896119902119878119902
119897119894+ 119878119901
119894119902119878119902
119896119897= 0 (100)
We contract this equation by indices 119901 119897 and we foundidentity
119878119901
119894119896119901+ 119878119901
119896119901119894+ 119878119901
119901119894119896= 0 (101)
Next we are assuming that 119878119901119894119901
= 120593119894and taking into
account the identity 119878119901
119894119895119878119902
119901119902= 0 we obtain the following
expression
119878119901
119894119895119901= 120593119894119895minus 120593119895119894 (102)
Next if we put 119878119901119894119895119901
= 0 then it follows that 120593119894119895minus 120593119895119894
=
0 and hence the value 119878119901119894119901can be expressed in terms of the
partial derivative of the scalar 119878119901119894119901= 120593119894= (ln120595)
119894 System (99)
takes the form
119867119895119894
119894= 119892119896119901
119892119895119904
119877119905
119894119901119904119878119894
119905119896+ 119892119896119901
119892119895119904
119877119905
119894119901119896119878119894
119904119905minus 119892119896119901
119892119895119904
119877119894
119894119901119905119878119905
119904119896
119865119894119895
= 0
(103)
We consider the tensor
119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+ 119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902 (104)
obvious that tensor is antisymmetricWe have the equality
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+ 119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904
minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
(105)
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 11
By direct calculations we can conclude that
119892119896119901
119892119902119904
119878119905
119901119902119878119895
119905119904minus 119892119895119901
119892119902119904
119878119905
119901119902119878119896
119905119904
=
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
(106)
hence
119867119895119896
minus 119867119896119895
= 119862119894119896119895
119894+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902) (107)
We calculate the covariant derivative
119862119894119896119895
119894= minus119862119894119895119896
119894= minus (119862
119894119895119896
119894+ Γ119895
119901119894119862119894119901119896
+ Γ119896
119901119894119862119894119895119901
+ Γ119894
119901119894119862119901119896119895
) (108)
By virtue of the fact that tensor 119862119894119895119896
= 119892119901119895
119892119902119896
119878119894
119901119902+
119892119901119896
119892119902119894
119878119895
119901119902+ 119892119901119894
119892119902119895
119878119896
119901119902is antisymmetric we have
Γ119895
119901119894119862119894119901119896
= Γ119895
119894119901119862119894119901119896
=
1
2
(Γ119895
119894119901119862119894119901119896
+ Γ119895
119901119894119862119901119894119896
) =
1
2
119878119895
119894119901119862119901119896119894
= minus
1
2
119878119895
119901119894119862119896119901119894
(109)
similarly we obtain
Γ119896
119901119894119862119894119895119901
= Γ119896
119894119901119862119901119895119894
=
1
2
119862119895119901119894
(Γ119896
119894119901minus Γ119896
119901119894) =
1
2
119878119896
119894119901119862119895119901119894
=
1
2
119878119896
119901119902119862119895119902119901
(110)
Then we write
119862119894119896119895
119894= minus119862119894119895119896
119894
= minus119862119894119895119896
119894minus
1
2
119878119895
119901119902119862119896119901119902
+
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894+
1
2
119878119895
119901119902119862119896119901119902
minus
1
2
119878119896
119901119902119862119895119901119902
minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
+
1
2
(119862119895119901119902
119878119896
119901119902minus 119862119896119901119902
119878119895
119901119902)
119867119895119896
minus 119867119896119895
= minus119862119894119895119896
119894minus Γ119902
119901119902119862119901119896119895
+ 119865119895119896
(111)
We will compute Γ119901
119897119901 for this we recall that Γ
119901
119901119897=
(12)119892119894119901119897119892119894119901
= (1radic119892)(120597radic119892120597119909119897
) and Γ119901
119897119901= Γ119901
119901119897+ 119878119901
119897119901and
obtain
Γ119901
119901119897=
1
radicminus119892
120597radicminus119892
120597119909119897
+ (ln120595)119897= (ln (120595radicminus119892))
119897 (112)
Then we obtain
119867119895119896
minus 119867119896119895
minus 119865119895119896
= minus119862119894119895119896
119894minus (ln (120595radicminus119892))
119894119862119894119896119895
(113)
We multiply by 120595radicminus119892 and have
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus120595radicminus119892 (119862119894119895119896
119894+ (ln (120595radicminus119892))
119894119862119894119896119895
)
120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
)
= minus (120595radicminus119892119862119894119895119896
)119894
(114)
We differentiate the last equality in view of the anti-symmetry of the tensors and we obtain the next importantequality
(120595radicminus119892 (119867119895119896
minus 119867119896119895
minus 119865119895119896
))119896
= 0 (115)
6 The Field Equations
Below we consider the derivation of the field equations independing on conditions
61 The Field Equations in the Absence of Symmetry Con-ditions Let Γ119894
119895119896be constructed on the basis of connection
119899-dimensional manifold It is not assumed that the Γ119894119895119896
aresymmetric in 119895 and 119896 Regardless of connection Γ
119894
119895119896 we
introduce symmetric metric tensor 119892119894119896(4)
We will derive the field equations from the variationprinciple of least action by varying the function Γ
119894
119895119896and 119892
119894119896
independentlyUsing (23) we can write the Riemann tensor
119877119894119895= Γ119901
119894119901119895minus Γ119901
119894119895119901+ Γ119901
119902119895Γ119902
119894119901minus Γ119901
119902119901Γ119902
119894119895 (116)
We form the scalar density as (119877119894119896+119878119895
119894119897119878119897
119896119895)119892119894119896
radicminus119892 where 119878119894119895119896=
Γ119894
119895119896minus Γ119894
119896119895 and postulate that all the variations of the integral
int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881 (117)
with respect to Γ119894119895119896and 119892119894119896radicminus119892 as the independent variables
are zero (at the boundaries they do not vary)Without dwelling on the standard intermediate calcula-
tions we find that the variation with respect to 119892119894119896radicminus119892 leadsto the equation
119877119894119896+ 119878119895
119894119897119878119897
119896119895= 0 (118)
then the variation with respect to Γ119894119895119896gives the equation
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901
minus 120575119895
119896(119892119894119901
119901minus
1
2
119892119894119901
119892119898119899119892119898119899
119901+ 119892119901119902
Γ119894
119901119902) minus 119892119894119895
Γ119901
119896119901
+ 2 (119892119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(119)
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
12 Journal of Gravity
If we contract the left side of the last equality by indices119894 and 119896 then we obtain zero identically equal to zero If wecontract by indices 119895 and 119896 we obtain the equation
minus3119892119894119895
119895+
3
2
119892119894119895
119892119901119902119892119901119902
119895minus 3119892119901119902
Γ119894
119901119902+ 119892119894119901
119878119895
119901119895= 0 (120)
Therefore we have
119892119894119895
119896minus
1
2
119892119894119895
119892119901119902119892119901119902
119896+ 119892119901119895
Γ119894
119901119896+ 119892119894119901
Γ119895
119896119901minus 119892119894119895
Γ119901
119896119901
minus
1
3
120575119895
119896119892119894119901
119878119902
119901119902+ 2 (119892
119894119901
119878119895
119901119896+ 119892119901119895
119878119894
119896119901) = 0
(121)
Then we lowered upper indices by using the metric andobtain the equation
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901+ 2 (119892
119901119895119878119901
119894119896+ 119892119894119901119878119901
119896119895) = 0
(122)
We use the symmetry of tensor 119892119894119895and rearrange 119894 and 119895
and deduce the equation
5 (119892119894119901119878119901
119896119895+ 119892119895119901119878119901
119894119896) +
1
3
(119892119895119896119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0 (123)
Further we obtained
minus 119892119894119895119896
minus
1
2
119892119894119895119892119901119902
119892119901119902119896
+ 119892119894119901Γ119901
119896119895+ 119892119895119901Γ119901
119894119896minus 119892119894119895Γ119901
119896119901
minus
1
3
119892119895119896119878119901
119894119901minus
2
15
(119892119896119895119878119901
119894119901+ 119892119894119896119878119901
119895119901) = 0
(124)
Thus we obtained the field equations (122) by varyingconnection obtained equation and as a result of them weobtained (123) where there are only metric and torsion Weassuming that the electromagnetic component is absent 120593 equiv
0 then from (123) we have 119878119894
119895119896equiv 0 and connection is
symmetrical as in Riemann geometry and (124) shows theknown law of Einstein-Hilbert problem for the gravitationalfield If 120593 = 0 and the metric is flat (no gravitationalfield) then from (124) can be obtainedMaxwell equations forelectromagnetic field in vacuum
62 The Einstein-Hilbert Equation in Case of the Absenceof Symmetry Conditions When the Lagrange Function IsDepending on the Torsion Tensor Now we also start from thevariation principle of the least action in the form 120575(119882
119898+
119882119892) = 0 where 119882
119898and 119882
119892-action respectively for matter
and field values we are varying 119892119894119896
By standard calculations we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int (119877119894119896radicminus119892120575119892
119894119896
+ 119877119894119896119892119894119896
120575radicminus119892 + 119892119894119896
radicminus119892120575119877119894119896
+ 119878119895
119894119897119878119897
119896119895radicminus119892120575119892
119894119896
+ 119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892) 119889119881
119877119894119896119892119894119896
120575radicminus119892 = minus
1
2
119877119901119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(125)
Similarly we obtain
119878119895
119894119897119878119897
119896119895119892119894119896
120575radicminus119892 = minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896radicminus119892120575119892
119894119896
(126)
Nowwe compute119892119894119896radicminus119892120575119877119894119896directly by using the defini-
tion and thus obtain two types of summands the first has thestandard form 119892
119894119896
(120575Γ119897
119896119894)119897minus 119892119894119896
(120575Γ119897
119896119897)119894= (119892119894119896
120575Γ119897
119896119894minus 119892119894119897
120575Γ119901
119897119901)119897
where it is considered that 119892119894119896119897
= 0 and by Stokesrsquo theoremturns into zeros The term of the second type exists due tothe absence of symmetry connection 119892119894119896120575119877
119894119896= 119892119894119896
120575(Γ119901
119902119901Γ119902
119896119894minus
Γ119901
119902119894Γ119902119897
119896119901) Then we express the connection coefficients via the
metric and torsion and after a rather lengthy calculation weobtain
119892119894119896
120575119877119894119896= (119878119901
119894119901119878119902
119902119896minus 119878119901
119894119902119878119902
119901119896minus 119892119896119901119892119902119905
119878119898
119894119905119878119901
119902119898) 120575119892119894119896
(127)
Thus we have
120575int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(128)
Then we obtain the conclusions
120575119882119892
= 1198701int (119877119894119896+ 119878119895
119894119897119878119897
119896119895) 119892119894119896
radicminus119892119889119881
= 1198701int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896)radicminus119892120575119892
119894119896
119889119881
(129)
where a physical constant 1198701 as a rule in a classic case is
1198883
16120587119896 and 119896 is called the universal gravitational constantFor variation of the action of matter we find
120575119882119892= 1198702int119879119894119896radicminus119892120575119892
119894119896
119889119881 (130)
where 119879119894119896is energy-momentum tensor of matter 119870
2usually
take a constant equal to minus12119888Therefore by the principle of least action for 120575119882
119892+120575119882119898=
0 we find relations
int(119877119894119896minus
1
2
119892119894119896119877 + (119878
119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902)
+ 119878119895
119894119897119878119897
119896119895minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896minus 119870119879119894119896)radicminus119892120575119892
119894119896
119889119881 = 0
(131)
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Journal of Gravity 13
because of the arbitrariness 120575119892119894119896 we have
119877119894119896minus
1
2
119892119894119896119877 + 119878119901
119894119901119878119902
119902119896+ 119878119902
119894119901119878119901
119896119902+ 119892119896119901119892119902119905
119878119898
119894119905119878119901
119898119902+ 119878119895
119894119897119878119897
119896119895
minus
1
2
119878119895
119901119897119878119897
119896119902119892119901119902
119892119894119896= 119870119879119894119896
(132)
where the constant119870 can be determined by 1198701and119870
2
7 Conclusions
We have investigated the properties of the space generatedjointly and agreed on by the metric and the torsion tensorsWe have presented the structure of the curvature tensor andstudied its special features and for this tensor we obtainedanalog Ricci-Jacobi identity and also evaluated the gap thatoccurs at the transition from the original to the imageand vice versa in the case of an infinitely small contoursThe geodesic lines equation has been researched We haveshown that the structure of tensor 120587
120572120573 which is similar
to the second fundamental tensor of hypersurfaces 119884119899minus1 issubstantially different from the case of Riemannian spaceswith zero torsion Then we have obtained formulas forhypersurfaces 119884119899minus1 which characterize the change of vectorsin accompanying basis relative to this basis itself
Taking into consideration the structure of the space withmetric and torsion we have reach the aim of our paperthat is derived from the variation principle the generalfields equations (electromagnetic and gravitational) so we areobtained analog of Einstein-Hilbert equation in space 119884119899
Conflict of Interests
The author declares that there is no conflict of interestsregarding the publication of this paper
References
[1] K Bredies ldquoSymmetric tensor fields of bounded deformationrdquoAnnali di Matematica Pura ed Applicata vol 192 no 5 pp 815ndash851 2013
[2] E Cartan and J Schouten ldquoOn Riemannian geometries admit-ting an absolute parallelismrdquoNederlandse Akademie vanWeten-schappen Proceedings Series A vol 29 pp 933ndash946 1926
[3] I Agricola and T Friedrich ldquoOn the holonomy of connectionswith skew-symmetric torsionrdquo Mathematische Annalen vol328 no 4 pp 711ndash748 2004
[4] I Agricola and T Friedrich ldquoA note on flat metric connectionswith antisymmetric torsionrdquo Differential Geometry and ItsApplications vol 28 no 4 pp 480ndash487 2010
[5] B Alexandrov and S Ivanov ldquoVanishing theorems on Hermi-tian manifoldsrdquo Differential Geometry and its Applications vol14 no 3 pp 251ndash265 2001
[6] A Saa ldquoA geometrical action for dilaton gravityrdquo Classical andQuantum Gravity vol 12 no 8 pp L85ndashL88 1995
[7] G Bonneau ldquoCompact Einstein-Weyl four-dimensional mani-foldsrdquo Classical and Quantum Gravity vol 16 no 3 pp 1057ndash1068 1999
[8] E Cartan and J Schouten ldquoOn the geometry of the groupman-ifold of simple and semisim-ple groupsrdquoNederlandse Akademievan Wetenschappen Series A vol 29 pp 803ndash815 1926
[9] G R Cavalcanti ldquoReduction of metric structures on CourantalgebroidsrdquoThe Journal of Symplectic Geometry vol 4 no 3 pp317ndash343 2006
[10] T Dereli and R W Tucker ldquoAn Einstein-Hilbert action foraxi-dilaton gravity in four dimensionsrdquo Classical and QuantumGravity vol 12 no 4 pp L31ndashL36 1995
[11] A Einstein The Meaning of Relativity Princeton UniversityPress Princeton NJ USA 1921
[12] A Einstein Relativity The Special and General the Ory H Holtand Company New York NY USA 1920
[13] A Einstein ldquoTheorie unitaire de champ physiquerdquo Annales delrsquoInstitut Henri Poincare no 1 pp 1ndash24 1930
[14] S Manoff ldquoFrames of reference in spaces with affine connec-tions and metricsrdquo Classical and Quantum Gravity vol 18 no6 pp 1111ndash1125 2001
[15] R AMosna andA Saa ldquoVolume elements and torsionrdquo Journalof Mathematical Physics vol 46 no 11 Article ID 112502 2005
[16] J A Peacock Cosmological Physics Cambridge UniversityPress Cambridge UK 1999
[17] P J E Peebles Principles of Physical Cosmology PrincetonUniversity Press Princeton NJ USA 1993
[18] W Rindler Essential Relativity Special General and Cosmologi-cal Texts and Monographs in Physics Springer New York NYUSA 2nd edition 1977
[19] J Jost Riemannian Geometry and Geometric Analysis SpringerBerlin Germany 2005
[20] H Pedersen and A Swann ldquoRiemannian submersions four-manifolds and Einstein-Weyl geometryrdquo Proceedings of the Lon-don Mathematical Society Third Series vol 66 no 2 pp 381ndash399 1993
[21] S Dineen Multivariate Calculus and Geometry SpringerUndergraduate Mathematics Series Springer Berlin Germany3rd edition 2014
[22] J G VargasDifferential Geometry for Physicists andMathemati-cians Moving Frames and Differential Forms From Euclid PastRiemann World Scientific 2014
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of
Submit your manuscripts athttpwwwhindawicom
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
High Energy PhysicsAdvances in
The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
FluidsJournal of
Atomic and Molecular Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Advances in Condensed Matter Physics
OpticsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstronomyAdvances in
International Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Superconductivity
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Statistical MechanicsInternational Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
GravityJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
AstrophysicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Physics Research International
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Solid State PhysicsJournal of
Computational Methods in Physics
Journal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Soft MatterJournal of
Hindawi Publishing Corporationhttpwwwhindawicom
AerodynamicsJournal of
Volume 2014
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
PhotonicsJournal of
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
Journal of
Biophysics
Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014
ThermodynamicsJournal of