CHAPTER-VII
A novel method to detect and estimate traces level contamination of crystalline polymorphic
form in Amorphous form by XRD
chapter Vll
Introduction:
Wilhelm Conrad RSntgen discovered 1895 the X-rays. 1901 he was honoured
by the Noble prize for physics.
1912: von Laue, Friedrich, and Knipping: "Interference Effects with Roentgen Rays"
passed x-rays through crystal of sphalerite (zinc sulfide); distinct diffraction pattern
observed.
(a) Crystals cause distinct x-ray diffraction patterns due to atoms.
(b) Crystals are composed of periodic arrays of atoms.
1914: English physicists Sir William Henry Bragg and his son Sir William Lawrence.
Bragg showed that the scattering of x-rays could be represented as a "reflection" by
successive planes of atoms within a crystal diffraction pattern can be used to
determine relative positions of atoms within a single crystal (i.e., molecular structure)
now called Bragg's law, d= nX/2 sin9. First single crystal structure: NaCl. 1915:
Braggs awarded Nobel Prize.
There are several X-ray diffraction techniques. Two of the most common are:
Single crystal X-ray diffraction: used to solve structure of crystalline materials
ranging from inorganic compounds to complex macromolecules such as proteins or
polymers. You can learn everything about a crystal structure, but requires a single
crystal. Although obtaining single crystals is difficult, single crystal X-ray
crystallography is a primary method for determining the molecular conformations of
biological interest such as DNA, RNA and proteins.
Powder X-ray diffraction: used to characterize crystallographic structure, grain size,
and preferred orientation in polycrystalline or powder solid samples. This is a
preferred method of analysis for characterization of unknown crystalline materials.
Compounds are identified by comparing diffraction data against a database of known
materials. It can be used to follow phase changes as a function of variable such as
temperature, pressure.
Uses of X-Ray Powder Diffi-action:
The most widespread use of x-ray powder diffraction, and the one we focus on
here, is for the identification of crystalline compounds in amorphous compound by
their diffraction pattern. Listed below are some specific uses:
194
CHapterVlI
Identification of single-phase materials - minerals, chemical compounds,
ceramics or other engineered materials.
Identification of multiple phases in microcrystalline mixtures (i.e., rocks).
Determination of the crystal structure of identified materials.
Identification and structural analysis of clay minerals
Recognition of amorphous materials in partially crystalline mixtures
Crystallographic structural analysis and unit-cell calculations for crystalline
materials.
Quantitative determination of amounts of different phases in multi-phase
mixtures by peak-ratio calculations.
Quantitative determination of phases by whole-pattern refinement.
Determination of crystallite size fi-om analysis of peak broadening.
Determine of crystallite shape from study of peak symmetry.
Study of thermal expansion in crystal structures using in-situ heating stage
equipment.
Basic overview of solid forms
Organic molecular solid Molecular weight
I I Amorphous Crystalline
^ — ^ X Polymorphs Hydrates
.—^-. r^—. Monotropic Enantiotrqpic Reversible Non>reverstt>le .4 p.
Ail can have numercnis habits
Figure 1: Basic overview of solid forms.
From the above figure 1 we can say that a solid can exist in two forms viz.
either amorphous or crystalline. In crystalline form a solid can exist as polymorph,
hydrate, solvate or co-crystal.
195
CHapterVlI
Polymorphs arise when molecules of a compound attain different
conformations with differing energies of stabilization. Although their chemical
properties may be same but they differ in their structural orientation. Generally a less
stable polymorph converts into its stable polymorph. But the conversion might take
years or few seconds to occur as it depends on the activation energy supplied to the
polymorph and free energy of that solid. Lowest energy form is more stable. At
transition temperature when one form (polymorph) is more stable than the other forms
then they are called as enantiotropic polymorphs and when the all forms are stable at
transition temperature, then they are called monotropic polymorphs. When the
crystal's repeating arrangement unit has a molecule of solvent attached to it then it is
called as a solvate. When the solvate molecule is replaced by water molecule then it is
called as hydrate. A co crystal is a distinct solid-state material with unique,
unpredictable structure and physical property profile '̂̂ . Different polymorphs give a
drastic effect on the dissolution rate which is in turn dependent on the solubility '^l
Amorphous state means lack of form. It has no long range order of molecular packing.
They generally possess higher internal energy, have greater thermodynamic properties
and have increased mobility due to greater intermolecular distances. We can say that
an amorphous system is actually fluid but appears to be a solid in the time scale
observation. This statement can be explained by the following figure 2:
Til T „ T. ,
T«nip*r«tui«
RGURg .Sch«matteftpf««»fttttioftof4KWh»lpy(ofwhim<^v«t»u»t«iiy«wtuwdH»^^ for a Nqijld caiMbI* of oryttatl l^ and forming ̂ fforant glassM at dHforant cooling ratat (Tic Kauzmann tamparatura. T^ glass transition tamparatura. T«: fosien tamparatura).
Figure 2 [3]
196
CHapterVlI
Cooling from melt, there is a chance that it may convert to a crystalline form
at temperature Tm. This is due to an exothermic crystallization which occurs due to
decrease in volume and enthalpy. Secondly, if we don't give enough time for the
molecule to crystallize then we don't see discontinuity at Tm and the system enters
into supercooled state. Now on further cooling the viscosity of the compound
decreases and it appears to be in a rubbery state. When we further cool it we observe
that at some point we don't see the decrease in viscosity as rubbery state is converted
to glassy state. The temperature at which this phenomenon occurs is called as glass
transition temperature (Tg). We can say that Tg is a thermodynamic requirement
because otherwise the material may get lower enthalpy than crystalline material or
even get negative enthalpy Tk. This is a significant difference between amorphous
and crystalline materials. Amorphous materials are easy to formulate than crystalline
materials but they degrade faster than the latter (generally).
Paracetamol:
NHCOCH,
o Paracetamol
H
Figure 3: The Structure of paracetamol [4]
Paracetamol is used as an antipyretic and analgesic. Commercially it is
available as tablets, capsules, solutions, suspensions etc. Paracetamol has three known
polymorphic form viz. Monoclinic (Form I), Orthorhombic (Form II) and a third
instable form III. Efforts were made in characterising form III with use of rapid
heating DSC and HPMC (crystal growth modifier) ^̂ l Thermal analysis of the
polymorphs was carried out using FT-IR spectroscopies "̂l Form II is the
commercially available form of paracetamol. However form I has a stififer
arrangements in the crystal lattice compared to the orthorhombic metastable form II.
Hence form I shows poor compression ability *̂̂ . Although form II shows better
compression ability compared to form I but still form I is preferred as API. This is due
to the stability issues related with form II. The stability of the form II varies with
alteration in temperature and humidity. Hence, the stability issue prompts the
197
CHapterVlI
manufacturer to liken form I. Efforts should be made to make form II more stable
under ambient temperature and humidity.
Itraconazole;
A I H v_y \ /
Ct
Figure 4: The Structure of Itraconazole '̂ ^
Itraconazole (Figure 4) is an antimicrobial agent. It is commercially used for
treatment of onychomychosis. The main disadvantage for this drug is its solubility. It
is a poorly soluble drug. Hence attempts were made to form a co crystal of the drug
which will increase the solubility of the drug. Amorphous state of the drug has a
greater solubility than the crystalline form but the amorphous form is less stable and
degrades quickly in presence of humidity and temperature. Hence in one such attempt
a CO crystal of itraconazole was prepared using 1,4-dicarboxylic acids ^^\ A trimeric
CO crystal was prepared with itraconazole atoms at the two ends of the dicarboxylic
acid (Succinic acid).
anWtmgal agent
«»i r"^
Co-crystallization The dissolution proflie of co<ry$tal$ with L-malic acid matches that of amorphous cis-itraconazol*
1. Sporjrwx* 2. >-malic JCi<) co-crystar 3. MMtMtc aetd c<xrystai 4. Succtnsc aetd co-crystals 9. c^-itraconazol«
mmtwMH JAmCmm&m 2603, J2S,*««-««?
Figure 5: Dissolution profile of itraconazole
From the figure 5 we can conclude that the co crystal with malic acid matches
the solubility that of the amorphous itraconazole ^^\ Thus this study suggests that it is
198
CHapterVlI
possible to have a form which matches the bioavailability of the amorphous form but
has the physical and chemical stability of the crystalline form.
Basic overview of Amorphous Forms:
Celecoxib:
Celecoxib is a nonsteroidal anti-inflammatory drug indicated for Rheumatic
arthritis. It acts as a COX2 inhibitor. It is commercially marketed as Celebrex and is
supplied in the form of capsules.
Figure 6: The structure of Celecoxib *̂̂
The crystalline form of the drug is marketed. However it was found that the
higher bioavailability was shown by the amorphous state. The downfall of the
amorphous state was its stability. This was due to the structural relaxation which
causes devitrification of the celecoxib if stored at room temperature ^'l This was
enhanced by mixing it with polymers like PVP, PVP-Meglumine which helped in
stabilising the amorphous system 1'°""]. A new solid state form was developed by
Pharmacia Corporation and patented ^̂ l The polymer used form a composite with the
amorphous form and prevents its conversion to the crystalline form. Crystalline form
of the Celecoxib is less pharmaceutically active. The amorphous lactose was prepared
from the crystalline one by spray drying techniques with isopropanol. Studies were
conducted to check the stability and solubility of the drug-polymer amorphous system
'^l The measurement of the crystalline content, glass transition temperatures and
enthalpy relaxation were characterised using DSC. Dissolution and solubility studies
were carried out to check the release mechanisms. Hence by these techniques the new
form of Celecoxib was discovered and an improved formulation was developed.
199
Simvastatin:
CHaperVlI
H.C
Figure 7: The structure of Simvastatin ''^'
Simvastatin is a cholesterol reducing agent belonging to the statin family. It
belongs to class II (low solubility but high permeability). It has two known
polymorphs, however the second polymorph requires extreme conditions to form and
form I is the most known stable form available. Simvastatin is suitable candidate for
amorphous form. In one such study the physicochemical properties of two differently
prepared amorphous forms were prepared and characterized '̂̂ l The two methods
were cryo-milling and quench cooling. In quench cooling the compound was melted
in oven and then quench cooled by putting liquid nitrogen on the plate carrying the
compound. In cryo milling the compound was ball-milled and the mill was
surrounded liquid nitrogen. The amorphous material obtained by two different
methods were further characterized by using FT-IR, DSC, High Speed DSC, TGA,
XRPD, PLS and SEM.
Figure 8: SEM images of simvastatin (a) crystalline, (b) cryo-milled, (c) 4°C milled, (d) RT milled, (e) recrystallized cryo-milled, and (f) quench cooled''^^.
200
chapter VII
Figure 8 depict the difference in the crystal structure of the amorphous
material prepared by different techniques. Accelerated stability studies were carried
out on the amorphous material prepared by two techniques. The cryo-milled
amorphous product showed less stability compared to quench cooled product. The
logic behind making amorphous form is that it shows more solubility than the
crystalline form. Amorphous form has more bioavailability compared to the
crystalline form. The main disadvantage of this form is that it has excess of
thermodynamic properties hence it tends to recrystallize into its crystalline form. If
we are able to keep the amorphous form stable enough till the expiry of the medicine
then we can achieve a novel and alternative formulation.
Summary of literature survey:
The crystalline and amorphous form of the drug have same molecular
configuration but different physical and chemical properties. The crystalline form of
the drug is not easy to formulate but the formulator is sure that the formulation will be
stable enough till its expiry date. But it is essential to identify the most stable
polymorph early in the development. This is of utter importance as the goal is to avoid
the scenario of ritonavir where the stable polymorph occurred after manufacturing
240 batches which had impaired bioavailability '̂̂ l Compounds used as active
pharmaceutical ingredients (APIs) must be screened for polymorphism according to
ICH guidelines. Whether to use a hydrate form or a solvate form will depend on the
ability of the compound to be stable over varied temperature and humidity. Co
crystals offer a new dimension to the formulator as it has the solubility as that of the
amorphous material but also retains the physical and chemical properties of the
crystalline form. However to choose an ideal co crystal former is of the utmost
importance. The amorphous form of the drug has increased solubility than its
crystalline counterpart and hence has increased bioavailability. The amorphous
material is easy to formulate than the crystalline material due to its high internal
energy and increased mobility. But amorphous state is not a state where the molecules
love to be and hence the molecules try to give out energy and recrystallize to a more
stable crystalline form. This is the major issue with amorphous materials. Little
change in temperature or humidity may largely affect the stability of the dosage form.
Various pharmaceutical processes affect the final form of the drug. Processes like
spray- drying, lyophilisation may cause formation of amorphous form. Amorphous
201
material is formed during mechanical shearing processes (Ball milling, wet
granulation, compaction). To conclude one must do extended preformulation studies
and then formulate the best possible stable drug available.
Polymorphism specifying the diversity of nature is widely observed in
pharmaceutical compounds. Differences in their physico-chemical and mechanical
properties led to the emergence of characterization based stringent quality control
measures of these altered solid-state forms in active pharmaceutical ingredients
(APIs) and drug products '̂'*l This complexity generally observed due to the
differences inflow properties, compatibility, hygroscopisity, crystal morphology and
hence process ability of different forms ''^l The sudden appearance or disappearance
of a crystalline form can threaten process development, and can lead to serious
pharmaceutical consequences if the transformation occurs during the manufacturing
or storage of the dosage form. Therefore, qualitative and quantitative analyses of
polymorphs must be incorporated early on in the drug development stage, both in the
API manufacturing and the formulation stage ̂ '*l
Pharmaceutical companies engaged in generic products aim at developing a
formulation that is qualitatively and quantitatively closest to the innovator product in
terms of excipients and the API polymorphic form. However, USFDA allows the use
of alternate polymorphic forms as long as the criteria of pharmaceutical equivalence
and bioequivalence are met ^"l
Amorphous materials can be naturally occurring and offer several advantages,
such as insulin (where as Insulin Zn complex is highly crystalline), large molecule
excipients (mostly polymers), and some small drug molecules that are difficult to
crystallize. The pharmaceutical industry often engages in costly high energy
processes, such as lypholization, high energy milling, melt extrusion and
coprecipitation with polymers to obtain an amorphous material. The desired benefits
of amorphous drugs include higher rate of solution and higher kinetic solubility.
Improved bioavailability of poorly soluble drugs is an especially desired consequence
of higher kinetic solubility. Several additional excellent references exist which discuss
the use and characteristics of amorphous materials in pharmaceuticals t'*'̂ °J.
The use of amorphous materials in any field is associated with some challenges.
202
CHapterVlI
The transformation from amorphous to a crystalline phase is dependent on diffusion
properties of the amorphous material, as well as the kinetics of nucleation and growth
of the crystalline phase ^ '̂"^^1. Additionally, the identity of the resultant crystalline
phase(s) cannot be absolutely predicted. First, though the identity of the most stable
crystalline form of a drug is sought, it is possible that the most stable form is not yet
identified. Second, there is no guarantee that the amorphous phase will transform to
the most stable crystalline phase '̂*l
Pure amorphous materials are often produced in the pharmaceutical industry.
Materials in the amorphous phase are thermodynamically less stable than any
crystalline form, leading to a tendency for the amorphous materials to transform to a
known or potentially unknown crystalline phase '̂*l The time scale for any
transformation is also unknown and can, in part, be evaluated via stability studies. The
detection of crystalline phases in mixed systems is often performed by powder X-ray
diffraction.
In this work, a limit of detection method was established to detect low levels
of crystalline material in an amorphous material ['*"''].
Materials and Methods:
Instrumentation:
XRD patterns of pure amorphous, pure crystalline and amorphous/crystalline
mixes were collected at room temperature on Bruker D8 Advance diffractometer
(Kralschue, West Germany) in which used the Bragg-Brentano configuration, a
copper X-ray tube (Cu Ka=1.5406 A) and Scintillation detector.
Experimental conditions/Method:
The Bruker D8 Advance configuration included power of 40 kV x 40 mA, Cu
Ka radiation passing through nickel filter, 0.3 mm divergence slit, 0.3 mm anti-
scattering slit, 0.1 mm detector slit and 0.6 mm receiving slit. The scan 20 ranges 3.0
to 60°, step size 0.03° and time per step O.Ssec. The sample holder was rotated in
plane parallel to their surface at 30 rpm during the measurement. Obtained
diffractograms were analysed with DIFFRAC'"* diffraction software.
203
CHapter'UlI
Samples:
This report will mainly focus on two material Atorvastatin Calcium
Amorphous (figure 9), Atorvastatin Calcium crystalline known form (figure 10) and
Fluvastatin Sodium Amorphous, Fluvastatin Sodium crystalline known form. To
determine the limit of detection in each case, blends using the amorphous material and
its related most stable known crystalline API phase, were prepared directly on the
sample holder using the manual blend method. All standard mixes are expressed as
%crystalline phase/total API. For example, a 2.0% crystalline standard contains 2.0%
crystalline per total API. The Si Zero background sample holders (Supplied by
Bruker). All sample holders were prepared with flat 511 Si inserts to assure sample
depth of 0.5 mm.
iUi '̂ M'̂ ^*.l»uw^<..l,A,̂ >w. spsa
•KSr.rarr*
Figure 9 Figure 10
METHOD DEVELOPMENT, EXPERIMENTAL DESIGN AND STUDY:
Several data analysis methods were evaluated to obtain a reproducible analysis
technique to observe the limit of detection (LOD) of a crystalline Form 1 in
amorphous material A. With the noise generated by the diffuse disorder scattering
fi-om the amorphous material, reproducible interpretation of a peak is difficult, but
necessary for a validated method. The FDA (food and drug administration) published
guidelines for LOD methods, which suggested LOD can be determined by visual
evaluation, signal-to-noise evaluation and Standard Deviation of the Response and
Slope. This last method is not possible for this case because the signal is not linear
with the degree of crystallinity. Also, as pointed out previously, the method must be
general enough to detect any peak, not just the peaks fi"om the most stable known
form. This means that peak detection at known 2 D values was not appropriate, though
204
chapter'V'lI
this is the most common method of analysis and was certainly used for method
development and identifying limit of detection standards. Often, a sufficiently trained
scientist can recognize a peak, which would make the visual evaluation possible.
However, validation of that visual recognition is close to impossible. Therefore, a
numeric approach to analysis must be taken. Jade, 10a program which reads several
PXRD file formats, was used for the analyses. Smoothing algorithms, background
subtraction and peak finding routines were evaluated. The Fast Fourier Transform
(FFT) smoothing algorithms work well with crystalline materials but tend to introduce
low intensity peaks for an amorphous material, which lead to problems distinguishing
real peaks from FFT-generated peaks. The traditional smoothing methods, including
the Sovitsky-Golay least squares smoothing algorithm, tended to reduce the
detectability of low intensity peaks. Background subtraction techniques often did not
adequately describe the background of a halo. This version of Jade allows the scientist
to select the background more accurately, but again, that lead to validation and
repeatability concerns. The analysis that appeared most reproducible was a simple
peak finds routine, with a parabolic filter, variable filter length, summit peak location,
0.1% intensity cutoff, 1.0 20 range to find background, and 7 points for the
background average. Normally, a threshold of 3a is used for peak detection, but 5a
was required due to the occasional interference of noise from the amorphous halo.
The limit of detection of the known most stable crystalline form for each
system was determined by making mixes of crystalline compound in the amorphous
compound.
Method development:
The calculated penetration depth information was used as a guide, but was not
expected to be accurate since the sample materials were imagined has a not pure
crystalline materials, nor isotropic but homogeneous. Tests were performed to
determine the effect of depth of penetration so that the appropriate sample depth,
especially for standard preparation, could be designed. Instead of thoroughly mixing
the 5.0% crystalline form with the amorphous blend, 5.0% crystalline form was
layered in the sample holder at 0.5 mm depth and at the top surface. Figure 11, which
demonstrates the effect of depth of penetration, shows that crystalline form can be
detected more precisely when layered on top surface, but not at 0.5 mm depth. The
powder pattern for 5.0% crystalline form on the top surface shown similar pattern
205
CHapterVlI
with that of crystalline form. Figure 11 is the diffractogram of crystalline form layered
on upper surface and lower surface of the sample holder, overlay. But in case of
pharmaceutical industry if there is a contamination or conversion of amorphous to
more stable crystalline form then it will be for entire bulk. So uniform mixing of
crystalline form with that of the amorphous method is adapted for the study rather
than layering on the top surface. In-order to suffice the intended purpose.
*»»k*M*rtnkH*W
•Uli'l^lllVHtl i
Figure 11 Figure 12
It begins with defining the requirements and the information desired from the
technique. The purpose of the method discussed here is to demonstrate that the test
samples contain amorphous material without evidence of crystalline materials, within
the limits of detection. A "low" limit of detection is desired, with the eventual desire
to relate degree of crystallmity to product performance, such as dissolution rate. The
development of a technique to understand the structure (truly amorphous and/or
crystallites too small to be detected by X-ray) of the amorphous material is not a goal
of this method, and would require more advanced instrumentation. The method,
including interpretation of the powder patterns, must be able to be run and analyzed
reproducibly by several scientists, and possibly may need to be transferred to a
production site.
A full understanding of the material to be tested is also important. The test
samples are defined as containing API that is amorphous by X-ray. There is a
possibility that some of the samples will also contain API that has transformed to a
crystalline phase. Thus, the expected PXRD pattern will be an amorphous halo,
sometimes with peaks due to crystalline excipients. The limit of detection was
determined using the most stable crystalline phase. One assumption is that the
206
CHapterVlI
amorphous-to-crystalline transition can be simulated with the physical mixture added
to the amorphous material. This assumes that any preferred orientation in the physical
mixture is similar to the naturally occurring crystallization in amorphous material.
Since there is no guarantee that the most stable known phase will be the result
of the transition from the amorphous phase, the method must be general enough to
pick up any new crystalline peaks. The kinetics of the amorphous-to-crystalline
transition is unknown and likely to depend on temperature, humidity and physical
treatment such as grinding. The standard samples are defined as amorphous by X-ray
and/or the related crystalline form. Preparing standard samples to determine the limit
of detection introduced a sample uniformity challenge that needed to be handled. To
reduce errors from sample loss, standards were prepared by adding amorphous
material directly on the sample holder. The instrument and sample holder
configuration can also be selected to reduce sample uniformity issues. The first
selection is for an instrument with a wide beam so that the largest sample surface can
be analyzed ''*l
The scan range, scan step size, time per step and the analysis method were
evaluated simultaneously. The scan ranges were selected based on strong peak
positions and limiting interference, which lead to scan ranges set to 3° to 60° 29. This
allowed the appearance of multiple peaks from the most likely crystal form and a
large enough range to allow detection of a form other than the most stable known
form. The scan step size and time per step were set to optimize signal/noise, with the
optimum appearing to be 0.03° 26 step size and 0.5 seconds per step. Several data
analysis methods were evaluated to obtain a reproducible analysis technique to
observe the limit of detection (LOD) of a crystalline form in amorphous material.
With the noise generated by the diffuse disorder scattering from the amorphous
material, reproducible interpretation of a peak is difficult, but necessary for a
validated method. The FDA published guidelines for LOD methods, which suggested
LOD can be determined by visual evaluation, signal-to-noise evaluation and Standard
Deviation of the Response and Slope.
This last method is not possible for this case because the signal is not linear
with the degree of crystallinity. Also, as pointed out previously, the method must be
general enough to detect any peak, not just the peaks fi-om the most stable known
207
CftapterVlI
form. This means that peak detection at known 2Q values was not appropriate, though
this is the most common method of analysis and was certainly used for method
development and identifying limit of detection standards. Often, a sufficiently trained
scientist can recognize a peak, which would make the visual evaluation possible.
However, validation of that visual recognition is close to impossible. Therefore, a
numeric approach to analysis must be taken, a program Eva which reads PXRD file
formats, was used for the analyses. Smoothing algorithms, background subtraction
and peak finding routines were evaluated. The Fast Fourier Transform (FFT)
smoothing algorithms work well with crystalline materials but tend to introduce low
intensity peaks for an amorphous material, which lead to problems distinguishing real
peaks from FFT-generated peaks. The traditional smoothing methods, including the
Sovitsky-Golay least squares smoothing algorithm, tended to reduce the detectability
of low intensity peaks. Background subtraction techniques often did not adequately
describe the background of a halo. This version of Eva allows the scientist to select
the background more accurately, but again, that lead to validation and repeatability
concerns. The analysis that appeared most reproducible was a simple peak find
routine, normally, a threshold of 1 is used for peak detection I'*"'*!.
The limit of detection of the known most stable crystalline form for system
was determined by making mixes of crystalline in the amorphous material. The limit
of detection was determined to be 3% (refer figure 17) by observing that 3.0%
crystalline form was repeatedly detected, but that lower amounts were not
reproducibly detected. To assure that 3% crystalline form could always be detected,
the source intensity was varied by decreasing the source energy to 35 mA, 30 mA and
25 mA. With 25 mA source energy crystalline peaks were found to be undetected. It
gives an inference that source energy to be monitored using external standard
corundum (AI2O3), intensity counts were recorded for every by day (Diffractogram
were not presented).
Experimental design and Study:
Talcum (figure 12) and NaCl (figure 13) are selected as an external standard
and diffractogram of both are recorded. NaCl is selected as an amorphous blank as it
does not show interference in 20 region were major peaks of Atorvastatin calcium
crystalline form appear and shows straight pattern, mean no halo pattern where as
208
CHapUrVlI
amorphous pattern shows. Refer figure 14 for overlay of Talcum and Atorvastatin
calcium crystalline form and for overlay of NaCl and Atorvastatin calcium crystalline
form refer figure 15.
It has been observed major peaks of Atorvastatin calciimi crystalline form, are at 20
9.13 ±0.2° and 21.56±0.2°. Limit of detection is considered based on the detection
of these two peaks of Atorvastatin calcium crystalline form.
Jii.iMiyM.ww " A M T &«K-.jiCACr.<m.liM •
iUMMlfM^Mik
Figure 13 Figure 14
Figure 15 Figure 16
Study details for finding limit of detection by considering NaCl as an External
amorphous standard:
Detection of Atorvastatin calcium crystalline form with NaCl.
a. 1.0% w/w of Atorvastatin calcium crystalline form in NaCl - diifi-actogram
shows absence of peak at 28 9.l3±0.2°and 21.56±0.2°.
b. 2.0% w/w of Atorvastatin calcium crystalline form in NaCl - diifractogram
shows absence of peak at 28 9.13±0.2° and 21.56±0.2°.
209
CfiapterVlI = ^
c. 3.0% w/w of Atorvastatin calcium crystalline form in NaCI - diifractogram
shows presence of peak at 29 9.13±0.2° and 21.56±0.2°; refer figure 16.
In similar manner Atorvastatin calcium crystalline form with Atorvastatin
Calcium amorphous
a. 1.0% w/w of Atorvastatin calcium crystalline form in Atorvastatin
Amorphous - diifi-actogram shows absence of peak at 20 9.13±0.2°and
21.56±0.2°.
b. 2.0% w/w of Atorvastatin calcium crystalline form in Atorvastatin
Amorphous - diifractogram shows absence of peak at 29 9.13±0.2° and
21.56±0.2°.
c. 3.0% w/w of Atorvastatin calcium crystalline form in Atorvastatin
Amorphous - diifractogram shows presence of peak at 29 9.13±0.2° and
21.56±0.2°, refer figure 17.
m —,
ao -ao — m -__ an — 290 -
** ̂ 230 -
HD -
m '
UiJjCt^ 1T0
MO —
ISO —
MO -
I M -
130 -
110 —
M»
10 -
«e —
70 ^
M —
90
« 10 —
at — M —
,
nts)
CMS7B12
1
A
k l/\ ' I uJ V AM/T^TM*u^^JM iJ^ • Yyyjvfr Ti|^v«nrT"
w
^ A T V C a - F M aWctyMillMtnwnorphMO
* > 4 1 W 1
fcAPi .
Mwi* ft ("7 jy ' " H N L
ir^Hf
zo
i^lMm. • M I F M V W I I h A J ^ 1 * 1 u i> II . . u
^^^"^'>''*'wM^wHfV*A<^^
M M
2-Theta - Scale
rf# jj-y^. iij.a ii 1 . . • ' ^ * ^ ' > * * M A A / f k / W v v - « f « i y H r ^ ^
90 00
Figure 17
Demonstrating that the method will work reproducibly is an important step to
completing method validation. In the pharmaceutical industry, the ICH has prepared
guidelines for method validation ^^^\ Most of the guidelines were initially based on
chromatography method development, but many of the concepts, such as specificity,
linearity, precision (refer figure 18), accuracy, sensitivity and robustness, can be used
for other methods, including PXRD methods.
Specificity is the ability to measure a specific analyte free fi-om interference of
other components. In this case, the method was developed to test the 20 range where
210
CfiapterVlI s = s
the amorphous halo or excipients provide the least interference while crystalline form
presence is obvious.
The analysis method was not designed to be specific, in that it is designed to
pick up any peak associated with a known or unknown crystal form. Linearity is
defined as showing direct and proportional response to changes in concentration. The
interferences by diffuse disorder scattering changes with the relative amounts of
amorphous/crystalline content, which interferes with linearity. In practice, we found
that the response is not linear near the limit of detection, which is an expected result.
Therefore, linearity is not a parameter that is tested to validate this limit of detection
PXRD method. Accuracy is also a concept that is not applicable to this method. Since
there is no linearity, we are not using this method to predict the amount of crystalline
Form present; we are simply determining that the amount of crystalline phases present
are either greater than or less than the limit of detection. Precision, which is the
closeness of agreement among a series of measurements, is a parameter that should be
evaluated for this technique. In this case, standards at the limit of detection (LOD) are
run repeatedly and are demonstrated to consistently detect peaks of crystalline form.
Precision:
Atorvastatin calcium crystalline form 9.0% w/w in Atorvastatin calcium
amorphous - diffractogram shows presence of peak at 29 9.13±0.2° and 21.56±0.2°
the six replicate sample preparation of same sample shows almost same intensity at
26 9.13±0.2° and 21.56±0.2°, refer figure 18.
211
Cfiapterl'n
[A
fJ>»*̂ «rvj*Wv-̂ w.»w»̂ '*
.•«V^w...W^>V*MA>"*v^„,^, / ^ - ~ , wVt.-.v-v--'»«™^yA~''*->-<^(—\-- ••'-^^^,-wU>.~
•"Atorvastatin Ca 9%crystaUine preparation 1 '^Atorvastatin Ca9%crystalline preparation 2 "'Atorvastatin Ca9%crvstalline preparation 3 ^Atorvastatin Ca9%crvstaliine preparation 4
2-Theta - Scale ^AtOTvastatinCa 9%crystalline preparations ^Atorvastatin Ca9%crystaUine preparation 6
Figure 18
In a similar manner, study has been conducted for Fluvastatin Na based on the
study it has been observed, form of NaCl does not have any peak which corresponds
to the major peaks of Fluvastatin crystallhne form, refer figure 19.
_J liL k m fkt\ fi ft l-'-ra -SyK
•]pC£,i t»ttSM*. T\)aS;ee V i ^ *«|iakiiM*NM«Mjblfl rti>ii a | l I i i^^i i • » JL
%l3<-sa> K i s^a^lM'
Figure 19 Figure 20
It has been observed major peaks of Fluvastatin crystallhne form, are at 28
12.21±0.2° and 12.97±0.2°. Limit of detection is considered based on the detection of
these two peaks of Fluvastatin crystallline form.
Detection of.Eluvastatin Crystallline form in NaCl,
a) 1.0% w/w of Fluvastatin crystallline form in NaCl - diifractogram shows
absence of peak at 29 12.21±0.2° and 12.97±0.2°.
212
CHapterl^I --—
b) 2.0% w/w of Fluvastatin crystallline form in NaCl- diifractogram shows
presence of peak at 29 12.21±0.2° (12.130°) and 12.97±0.2°. (12.910°), refer
figure 20.
Detection of FluvastatinCrystallline form in Fluvastatin amorphous.
a) Diffractogram of Fluvastatin amorphous refer figure 21.
b) 2.0% w/w of Fluvastatin crystallline form in Fluvastatin amorphous -
diifractogram shows presence of peak at 20 12.21±0.2° (12.279°) and 12.97±0.2°
(12.951°) refer figure 22.
c) Precision: Fluvastatin crystallline form 6.0% w/w in Fluvastatin amorphous -
diifractogram shows presence of peak at 29 12.21±0.2° and 12.97±0.2°, all the six
replicate sample preparation shows same intensity at 29 12.21±0.2° and
12.97±0.2° refer figure 23.
Figure 21 Figure 22
213
CfuifterVn = =
/v̂ fr̂ %
'/ww^v»«Jl^^•^u^
-v.V*,JV,t. ^1^,, . , • 'V •H-''t",-^fi,M-niy!y^i_^ ^.^
•"Fluvastatin Na 6% crystalline preparation I "•Fluvastatin Na 6% crystalline preparation 2 ™Fluvastatin Na 6% crystalline preparation 3 'Fluvastatin Na 6% crystalline preparation 4
2-Theta - Scale
'^Fluvastatin Na 6% crystalline preparation 5 ^'Fluvastatin Na 6% crystalline preparation 6
Figure 23
Demonstrating that the method will work reproducibly is an important step to
completing method validation. In the pharmaceutical industry, the FDA has prepared
guidelines for method validation. Most of the guidelines were initially based on
chromatography method development, but many of the concepts, such as specificity,
linearity, precision, accuracy, sensitivity and robustness, can be used for other
methods, including PXRD methods.
Specificity is the ability to measure a specific analyte free from interference of
other components. In this case, the method was developed to test the 29 range where
the amorphous halo or excipients provide the least interference while crystalline form
presence is obvious. The analysis method was not designed to be specific, in that it is
designed to pick up any peak associated with a known or unknown crystal form
Linearity is defined as showing direct and proportional response to changes in
concentration. The interferences by diffuse disorder scattering changes with the
relative amounts of amorphous/crystalline content, which interferes with linearity. In
214
CHapterVlI
practice, we found that the response is not linear near the limit of detection, which is
an expected result. Therefore, linearity is not a parameter that is tested to validate this
limit of detection PXRD method. Accuracy is also a concept that is not applicable to
this method. Since there is no linearity, we are not using this method to predict the
amount of crystalline present; we are simply determining that the amount of
crystalline phases present are either greater than or less than the limit of detection.
Precision, which is the closeness of agreement among a series of
measurements, is a parameter that should be evaluated for this technique. In this case,
standards at the limit of detection (LOD) are run repeatedly and are demonstrated to
consistently detect peaks of crystalline. The related phenomenon, robustness, is often
defined as the capacity of a method to remain unaffected by small, deliberate
variations in the method. In regards to robustness, we demonstrated that the
amorphous material A was not induced to form crystalline API by grinding or holding
the sample for several days beyond the normal analysis time. We also demonstrate
that additional grinding of a standard LOD mix results in a sample from which
crystalline content can still be detected.
Limitations of the Procedure,
(i) The demarcation of the crystalline (Ic) and amorphous (la) intensities from the
experimentally obtained X-ray diffraction pattern is done arbitrarily and is subject
to individual judgment. The procedure may thus lead to high variability in the
results, especially when the degree of crystallinity is low.
(ii) Samples may show preferred orientation effects, which may cause significant
errors in peak intensity measurement.
An effective way to minimize preferred orientation is to decrease the particle size
of the sample by grinding. However, the grinding process itself can cause
undesired changes in the solid state of the material.
(iii) It is necessary to generate a standard curve using samples with different degrees
of crystallinity. A common procedure is to mix crystalline and amorphous
standard phases in appropriate proportions. Non-homogenous mixing becomes a
potentially important issue, especially when preparing mixtures of exfreme
compositions.
215
CHapter'Un
Our investigation had two objectives: (i) to develop an X-ray diffractometric
technique, based on synchrotron X-rays and a 2-D area detector, having increased
sensitivity and rapid data acquisition capability compared to conventional instruments
and (ii) to develop an objective method for calculating the degree of crystallinity
based on an algorithm. The key analytical challenge, from a pharmaceutical
perspective, has been to develop a technique capable of discerning subtle changes in
lattice order in substantially amorphous materials. Our focus was therefore on the
detection and quantification of the first evidence of crystallization in an amorphous
material. An in situ crystallization approach was used to overcome the problem of in
homogeneity in mixing, which is a particularly serious issue at extreme mixture
compositions.
Conclusions:
The method described in this report was developed to determine if amorphous
material has crystalline character that is detectable by PXRD. Limit of detection of
Atorvastatin Ca crystalline from I in Atorvastatin amorphous is 3.0% and method is
found to be precise at concentration 3 times of its limit of detection.(i.e9.0%) and
Limit of detection of Fluvastatin crystalline from B in Fluvastatin amorphous is 2.0%
and method is found to be precise at concentration 3 times of its limit of detection.(i.e
6,0%).The most stable crystalline known form, was used to prepare standards to
determine its limit of detection, though the method was also designed to detect peaks
from other crystal forms. This methods, and similar methods for other amorphous
materials, have been used for formulation feasibility and stability studies. In the case
of amorphous materials, stability studies are especially important because, besides the
possibility of crystallization, the amorphous materials are also more hygroscopic and
often less chemically stable than related crystalline forms.
This study deals with quantification of crystalline content in amorphous form
using crystallographic (XRD) method. A simple novel XRD method has been
developed and validated for the quantification of crystalline form in amorphous form.
The two drug substances namely Atorvastatin Calcium and Fluvastatin Sodium have
been considered for the study. XRD pattern of pure amorphous, crystalline and
mixture of amorphous and crystalline of different concentration (w/w) has been
collected by using Bruker DS-Advance diffractometer, on in which the Braggs
bretano configuration, a copper X-ray tube (Cu Ka= 1.5406), power of 40kv x 40mA,
scan range 3 to 60°, step size 0.03° and time per step 0.5 sec. Limit of detection for
Atorvastatin Calcium and Fluvastatin Sodium is found to be 3% and 2% respectively.
216
copter'UII : =
References:
1. Christer B. Aakero'y et al. practical guidelines for co-crystal Synthesis.
Pharmaceutical co-crystals meeting Amsterdam, The Netherlands, 2008.
2. Tros de Ilarduya et al. Drug. Dev.Ind Pharm, 1997, 23, 1095.
3. Aditya Mohan Kaushal et.al.. amorphous drug deliver systems: molecular aspects,
design and performance. Critical reviews in therapeutic drug carrier system, 2004,
21(3), 133-193.
4. Zimmermann B et.al., thermal analysis of paracetamol polymorphs by FT-IR
spectroscopies. J. pharm Biomed Anal, 2011, 54(2), 295-302.
5. S. Gaisford, et al.. Characterisation of paracetamol form III with rapid-heating
DSC, J. Pharm. Biomed. Anal, 2010.
6. Etienne Joiris et.al. Compression behaviour of orthorhombic paracetamol.
Pharmaceutical Research, 1998, 15(7), 1122-1130.
7. Julius F. Remenar et.al.. Crystal engineering of novel cocrystals of a Triazole
drug with 1,4-Dicarboxylic acids. J. Am. Chem. Soc, 2003, 125 (28), 8456-8457.
8. Patent No. U.S. 6.964,978. Solid-State Form of Celecoxib Having Enhanced
Bioavailability. 2005.
9. K. Grzybowska et al. Molecular dynamics and physical stability of amorphous
anti-inflammatory drug: Celecoxib. J. Phys. Chem. B, 2010, 114, 12792-12801.
10. Piyush Gupta et.al., stability and solubility of Celecoxib-PVP amorphous
aispersions: A molecular perspective. Pharm Research 2004,21(10), 1762-1769.
11. Piyush Gupta et al. Molecular interactions in celecoxib-PVP-Meglumine
amorphous system. JPP 2005, 57,1-8.
12. Kirsten A. Graeser et al. physicochemical properties and stability of two
differently prepared amorphous forms of Simvastatin. Crystal growth and design
(2008), 8(1), 128-135.
13. Scott L. Childs etal.A metastable polymorph of Metformin hydrochloride:
isolation and characterization using capillary crystallization and thermal
microscopy techniques. Crystal growth and design, 2004, 4 (3), 441-449.
14. S. Yamamura and Y. Momose, Quantitative analysis of crystalline
pharmaceuticals in powders and tablets by a pattern-fitting procedure using X-ray
diffraction data, Int. J. Pharm, 2001, 212,203-212.
217
CfiapterVlI
15. M. J. Kontny and G. Zografi. Sorption of water by solids. In H. G. Brittain (ed.),
physical characterization of pharmaceutical solids. Marcel Dekker, New York,
1995, 387-418.
16. Vishal koradia, Garimachawla, Arvind k. Bansal, qualitative and quantitative
analysis of clopidogrel bisulphate polymorphs, acta pharm., 2004, 54, 193-204
17. The United States Pharmacopoeia 32, United States Pharmacopoeial Convection,
Rockville, USA.
18. B. A. Sarsfield, M. Davidovich, S. Desikan, M. Fakes, S. Futemik, J. L. Hilden,J.
S.Tan, S. Yin, G. Young, B. Vakkalagadda, and, K. Volk, Bristol Myers Squibb
Co., New Brunswick, NJPurdue University, West Lafayette, powder x-ray
diffraction detection of crystalline phases in amorphous pharmaceuticals.
International Centre for Diffraction Data, 2006, ISSN 1097-0002 , 322-327.
19. Cletus Nunes, ArumugamMahendrasingam, and Raj Suryanarayanan,
Quantification of crystallinity in substantially amorphous materials by synchrotron
x-ray powder diffractometry. Pharmaceutical Research, 2005, 22 (11).
20. R. Surana and R. Suryanarayanan. Quantitation of crystallinity in substantially
amorphous pharmaceuticals and study of crystallization kinetics by X-ray powder
diffractometry. Powder Diffr, 2000, 15, 2-6.
21. C. J. Kedward, W. MacNaughtan, and J. R. Mitchell. Isothermal and non-
isothermal crystallization in amorphous sucrose and lactose at low moisture
contents. Carbohydr. Res, 2000, 329, 423-430.
22. S. L. Shamblin, E. Y. Huang, and G. Zografi. The effects of co-lyophilized
polymeric additives on the glass transition temperature and crystallization of
amorphous sucrose. J. Therm. Anal, 1996,47, 1567-1579.
23. ICH Topic Q 2 (Rl) Validation of analytical procedures text and methodology
note for guidance on validation of analytical procedures, text and methodology
(CPMP/ICH/381/95), June 1995.
218