Ten Years of Fundamental Physics inESA’s Space Science Programme

R. ReinhardSpace Science Department, ESA Directorate for Scientific Programmes, ESTEC, Noordwijk, The Netherlands

Early daysIn the beginning, ESA’s Space ScienceProgramme (actually ESRO’s in those days)consisted only of small magnetosphericsatellites. ESRO-II, ESRO-IA, HEOS-1, ESRO-IB, HEOS-2 and ESRO-IV were launchedbetween 1968 and 1972, addressing primarilyproblems in plasma physics. Throughout the1970s, Solar System exploration continued tobe limited to magnetospheric research, with thelaunches of GEOS-1 and -2 and ISEE-2. Thefirst astronomy satellites (TD-1, COS-B andIUE) were launched between 1972 and 1978,making observations at UV, X- and γ-raywavelengths.

There was also interest in fundamental-physicsmissions in those early days. COPERS (theCommission Préparatoire Européenne deRecherche Spatiale, which existed from

December 1960 until March 1964) plannedeight ad-hoc groups, representing the mainfields in space science at that time, amongthem a group on Geodesy, Relativity andGravitation; this group was, however, notrealised when ESRO came into existence in1964. A far-sighted Italian proposal in 1964suggested that ESLAR (the predecessor ofESRIN) should study drag-free satellites, a keytechnology required for fundamental-physicsmissions. This early phase culminated in thePhase-A study of the SOREL (Solar Relativity)mission in 1970-71. The goal of the SORELmission was to measure:– the gravitational redshift to 3 parts in 10

6

– the time delay and deflection of laser lightpassing close to the Sun

– the solar quadrupole moment J2.

This required a spacecraft that would be drag-free to the level of 10

-12m/s

2, and in the one-

way laser option a high-precision on-boardclock (caesium or hydrogen-maser clock) andin the two-way laser option an on-board laser.None of these technologies existed at that timeand would have had to be developed andspace-qualified at considerable cost. It wasmostly for that reason that the plans for theSOREL mission were not further pursued.

The existence of three distinctly different fieldsin space science in the 1970s was alsoreflected in ESA’s scientific advisory structure atthat time. It consisted of the LaunchingProgrammes Advisory Committee (LPAC, thepredecessor of today’s SSAC) and the– Solar System Working Group (SSWG)– Astrophysics Working Group (AWG)– Fundamental Physics Panel (FPP).

The FPP had ten members, among them aformer Director General of ESRO (H. Bondi)and a later Director General of ESA (R. Lüst).However, the founding fathers of ‘FundamentalPhysics in Space’ soon realised that thetechnology needed to carry out high-precisionexperiments in space did not yet exist and thePanel became inactive after 1979. The SSWGand the AWG still exist today.

On 15 June 1989, ESA’s Science Directorate issued a Call for MissionProposals which specifically also asked for proposals in fundamentalphysics. In response to this Call the scientific community submittedfive proposals aimed at detecting gravitational waves, testing theEquivalence Principle, measuring the gravitational constant with highprecision, and searching for the elusive Fifth Force and for DarkMatter in the Universe. Today, 10 years later, the new field offundamental physics in space has matured with one mission (GP-B)about to be launched by NASA, two industrial studies (LISA, STEP) inprogress this year in ESA at Phase-A level, and several otherexperiments either flying or being readied for flight, includingexperiments on the International Space Station. There is clearly now acommunity of fundamental physicists in Europe in need of space flightopportunities, just as there are communities of astronomers and SolarSystem scientists. The only difference is that the technologiesenabling meaningful fundamental-physics missions have only becameavailable 30 years later.

ten years of fundamental physics

Composition of ESA’s FundamentalPhysics Panel (1971-1979)

H. Bondi (Chairman) R. LüstJ. Blamont G. OcchialiniG. Cocconi I.W. RoxburghG. Colombo E. SchatzmanB. Laurent D. Sciama

In 1972, the Italian Physical Society organiseda summer school in Varenna devoted toExperimental Gravitation. The Proceedings,published in 1974, give an excellent account ofthe state of the art at that time and includepapers not only on the SOREL mission, butalso on redshift experiments using high-precision clocks such as Gravity Probe A, thegyroscope experiment (later named GravityProbe B), the STEP mission, several methodsto detect gravitational radiation and thetechnique of drag-free satellites.

With the launch of Giotto in 1985, ESA’s SolarSystem exploration began to include theexploration of the solid bodies — first cometHalley, and later, with the launch of theHuygens probe in 1997, the Saturnian moonTitan. Missions to the Moon, Mars and Mercuryare now under study. Solar physics was addedwith the launch of Soho in 1995.

Astronomy saw the addition of the new field ofastrometry with the launch of Hipparcos in1989, and ISO, launched in 1993, opened upthe possibility of observations in the infrared.

Fundamental physics as a science discipline inits own right was also included in ESA’s Long-Term Space Science Programme ‘Horizon2000’ (ESA SP-1070) with a chapter on SpaceExperiments in Relativity and Gravitation by I.W.Roxburgh. The paper does not describe aspecific mission in fundamental physics, butlists several possibilities for includingfundamental physics experiments on suitableSolar System missions, such as MercuryOrbiter, Solar Probe or a mission to the outerplanets. Concerning gravitational wavesRoxburgh wrote (in 1984):

‘At present we are unable to predict with anyconfidence the strengths of any such wavespassing through the Solar System, but thesuccessful discovery of such waves could be ofimmense significance for gravitational physicsand for astrophysics’.

In 1987, the Austrian Space Agency togetherwith ESA organised a summer school inAlpbach devoted to ‘Space Science andFundamental Physics’. The Proceedings of thatsummer school (ESA SP-283) include severalpapers describing future missions infundamental physics, such as a test of theEquivalence Principle and the detection ofgravitational waves with long-baselineinterferometers in space.

Restarting fundamental physics in spaceWith the discontinuation of the FPP in ESA’sscience advisory structure in 1979, thepossibilities of realising a fundamental-physicsmission in ESA essentially disappeared andinterest within the Agency in fundamentalphysics almost completely vanished. Thischanged in the late 1980s with the enunciationof the ‘fifth force’ hypothesis by E. Fischbachand co-workers, triggering a suite of highlypublicised experiments worldwide and callingattention to the fact that gravity, itself the‘oldest’ of the known forces, was in some waysalso the least understood. A ‘fifth force’,coexisting with conventional gravity, would leadto a net interaction between macroscopicbodies, which would show small deviationsfrom the behaviour expected from the classicalNewtonian inverse-square law of gravity. If sucha force were to exist, it would be very small andpossibly easier to detect in a gravitationally lessdisturbed experiment in space.

r bulletin 98 — june 1999 bull

Fundamental Physics in Space

The field of ‘Fundamental Physics in Space’ includes those research activities in gravitationaland particle physics aimed at finding new, more comprehensive concepts and laws, the testingof existing ones, and the resolution of some very basic inconsistencies. This includes:

• the direct detection and detailed analysis of gravitational waves• the investigation of possible violations of the Equivalence Principle• the search for new hypothetical long-range forces• the testing of General Relativity and its alternative theories• the unification of the fundamental interactions of nature• particle physics, in particular the search for antimatter in space• the development and fundamental application of space-based

ultrahigh-precision atomic and other clocks.

The technologies used in fundamental-physics experiments (e.g. high-precisionaccelerometers), the requirements on spacecraft, and the high degree of spacecraft/experimentinterrelationship are distinctly different from missions in Solar System exploration and astronomy.For example, fundamental-physics spacecraft typically have to be in purely gravitational orbits,i.e. they have to be drag-free.

The STEP missionSTEP was proposed to ESA in November 1989by a team of scientists from Europe andStanford University. Work at Stanford on thedesign of the STEP experiment had alreadybegun much earlier, in 1971. The originalproposal comprised three accelerometersystems accommodated in a cryogenic dewarto test the Equivalence Principle to a precisionof 1 part in 10

17, along with a limited geodesy

co-experiment and an aeronomy co-experiment. The boil-off helium from the dewarwould be used to feed small proportionalthrusters to compensate for the drag of theresidual atmosphere at orbital (~ 500 km)altitude.

This proposal was studied as an ESA/NASAcollaborative project in 1991 at Assessmentlevel and in 1992 at Phase-A level. In the end itwas not selected as the M2 project and wassubsequently re-proposed in May 1993 as acandidate for the third medium-size project(M3). During the M3 cycle, STEP was studiedagain at Assessment and Phase-A levels, thistime as a European-only project. Again, it wasnot selected as a flight project because at thetime of the selection (April 1996) there wereother proposals in the scientific communityaimed at achieving STEP’s main scientificobjective, a test of the Equivalence Principle, atmuch lower cost and even higher precision. A20 M€ contribution by ESA to a NASA-ledSTEP project with a launch in 2004 is nowincluded in ESA’s Space Science Programme.ESA’s contribution to the project is envisaged tobe the Service Module and possibly a few otherelements. The Service Module design ispresently the subject of a four-month industrialstudy.

The Equivalence Principle postulates theequivalence between inertial and gravitationalmass or, stated differently, that bodies ofdifferent mass and/or composition fall with thesame acceleration in a gravitational field. Thiscontention cannot be proven, it can only betested to higher and higher precision. The mostprecise ground-based tests today haveachieved a precision of 1 part in 10

12.

Experiments on the ground are limited at thislevel because of unshieldable seismic noiseand the weak driving acceleration. In space,this test could be done a factor 10

6more

precisely.

Einstein generalised the Equivalence Principleand made it the foundation of his theory ofGeneral Relativity. A violation of the EquivalencePrinciple at some level would either require amodification of Einstein’s theory or constitutethe discovery of a new force. There are, in fact,

ESA’s Call for Mission Proposals for the secondmedium-size project (M2) within the frame-work of ESA’s Long-Term Space ScienceProgramme ‘Horizon 2000’ was issued on 15 June 1989. It specifically also asked forproposals in fundamental physics. In responseto this Call, five proposals were submitted inthe field of fundamental physics (out of a totalof 22):– a mission to detect and observe gravitational

waves– a Satellite Test of the Equivalence Principle

(STEP)– Newton, a man-made ‘planetary system’ in

space to measure the Constant of Gravity G– GRAVCON, an experiment to measure the

Constant of Gravity G– APPLE, a proposal for the determination of

the Fifth Force and the detection of DarkMatter in the Universe.

Of these five proposals, STEP and Newtonwere given the highest priority by the ‘Ad-hocWorking Group on Fundamental Physics’.However, only STEP was selected by the SSACon 9 February 1990 for study at AssessmentLevel. Later, in mid-1990, the STEP payloadwas enlarged by incorporating the scientificobjectives of Newton and GRAVCON.

The renewed interest in fundamental physicsmay have been triggered in ESA by thecuriosity about the ‘fifth force’, but what wasnot perhaps fully realised at that time was that,unlike in the 1970s, in the early 1990s thetechnologies for carrying out meaningfulfundamental-physics missions had in themeantime become available. Key technologies,such as high-precision accelerometers, drag-free control using He-proportional thrusters orsmall ion thrusters, ultra-stable lasers in space,He-dewars, high-precision displacementsensors (SQUIDs), magnetic spectrometers,small lightweight H-maser clocks and atomicclocks using laser-cooled caesium atoms hadbeen developed and were either already space-qualified or about to be space-qualified.Already in 1976, NASA sent an H-maser clockto an altitude of 18 000 km on a suborbitalrocket flight (Gravity Probe A) to test Einstein’s‘clock gravitational frequency shift’ formula (theclock on the rocket appears to run faster thanthe one on the ground). NASA’s Gravity ProbeB (GP-B), to be launched in October 2000 intoa polar orbit at 650 km altitude, will test twopredictions of Einstein’s theory of GeneralRelativity — geodetic precession and frame-dragging precession — to 1 part in 10

5and

400, respectively. For GP-B several keytechnologies had to be developed which arenow available for other fundamental-physicsmissions.

ten years of fundamental physics

Figure 1. Relative motion oftwo concentric cylindricaltest masses in the case of

an Equivalence Principle(EP) violation. The masses

are constrained to one-dimensional motion. In this

example the inner testmass falls faster towards

the Earth. The EP violationsignal is periodic at orbital

frequency

good reasons to believe that General Relativityis not the ultimate theory of gravity. Gravitation,electromagnetism and the weak and stronginteractions are the four known fundamentalforces of Nature. Einstein’s theory of gravity,General Relativity, provides the basis for ourdescription of the Big Bang, the cosmologicalexpansion, gravitational collapse, neutronstars, black holes and gravitational waves. It isa ‘classical’, non-quantum field theory ofcurved space-time, constituting an as yetunchallenged description of gravitationalinteractions at macroscopic scales. The otherthree interactions are dealt with by a quantumfield theory called the ‘Standard Model’ ofparticle physics, which accurately describesphysics at short distances where quantumeffects play a crucial role. But, at present, norealistic theory of quantum gravity exists. Thisfact is the most fundamental motivation forpursuing our quest into the nature of gravity.

The Standard Model successfully accounts forall existing non-gravitational particle data.However, just as in the case of GeneralRelativity, it is not a fully satisfactory theory. Itscomplicated structure lacks an underlyingrationale. Even worse, it suffers fromunresolved problems concerning the violationof the charge conjugation parity symmetrybetween matter and antimatter and the variousunexplained mass scales. Purported solutionsof these shortcomings typically involve newinteractions that could manifest themselves asapparent violations of the EquivalencePrinciple.

The truly outstanding problem remains theconstruction of a consistent quantum theory ofgravity, a necessary ingredient for a completeand unified description of all particleinteractions. Super-string theories — in whichelementary particles would no longer be point-like — are the only known candidates for sucha grand construction. They systematicallyrequire the existence of spinless partners of thegraviton: dilatons and axion-like particles. Thedilaton, in particular, could remain almostmassless and induce violations of theEquivalence Principle at a level that — albeittiny — may well be within STEP’s reach.

The simplest way of testing the EquivalencePrinciple would be to throw two masses (e.g.spheres) of different composition from a hightower and measure any difference in the arrivaltime on the ground (taking into account theeffect of air resistance). Galileo was, for a longtime, reputed to have performed such anexperiment from the Leaning Tower of Pisaalthough, as we know today, he never actuallydid so himself.

The STEP Project is the modern version of thisexperiment. The test masses are placed insidea satellite in low-Earth orbit, where they ‘fallaround the Earth’ (Fig. 1). In this way, the testmasses never strike the ground, and anydifference in the rate of fall can build up over along time period. In Earth orbit the signal isperiodic and the experiment can be repeatedseveral thousand times during the missionlifetime. However, even at 500 km altitude, thedensity of the Earth’s atmosphere is sufficient tobrake the satellite and to disturb theexperiment. The satellite must thereforecompensate for the braking by firing acombination of proportional thrusters so thatthe satellite is ‘drag-free’ and the test massesinside are free-floating and follow a purelygravitational orbit.

The test masses are in the form of hollowcylinders whose axes are centred on eachother to eliminate any disturbances from theEarth’s gravity gradient. Any differential motionof these two test masses is sensed by coilscoupled to SQUID (Superconducting QuantumInterference Device) magnetometers, forming asuper-conducting differential accelerometer.The SQUIDs can detect any differential motionof the two typically 500 g test masses with asensitivity of 10

-15m, the diameter of the

nucleus of an atom. To sample a variety of test-mass materials, STEP carries four suchdifferential accelerometers.

The payload chamber is accommodated insidea superfluid helium dewar, which cools the

r bulletin 98 — june 1999 bull

Figure 2. The STEPspacecraft in low-Earth,Sun-synchronous orbit at500 km altitude

payload to 1.8 K. The cryogenic dewar ismounted below a three-axis-stabilised, drag-free spacecraft (Fig. 2) which uses the heliumboil-off from the dewar to feed a number ofproportional thrusters to compensate for theresidual air drag at orbital altitude. STEP’s orbitis circular and at low altitude (~500 km). A Sun-synchronous (i.e. almost polar) orbit waschosen to avoid eclipses, thus providing ahighly stable thermal environment throughoutthe mission lifetime of 6-8 months. During flight,the spacecraft rotates about its long axis at asmall multiple of the orbital frequency, in orderto spectrally shift the science signal from orbit-fixed systematic error sources.

The LISA missionThe primary objective of the LISA (LaserInterferometer Space Antenna) mission is thedetection and observation of gravitationalwaves from massive black holes and galacticbinaries in the frequency range 10

-4– 10

-1Hz.

This low-frequency range is inaccessible toground-based interferometers because of theunshieldable background of local gravitationalnoise and because ground-based inter-ferometers are limited in length to a fewkilometres.

The ground-based interferometers LIGO,VIRGO, TAMA 300 and GEO 600, withbaselines from 0.3 to 4 km and the LISAinterferometer in space, with a baseline of 5 million km, complement each other in anessential way. Just as it is important tocomplement the optical and radio observationsfrom the ground with observations from spaceat submillimetre, infrared, ultraviolet, X-ray andgamma-ray wavelengths, so too is it importantto complement the gravitational-waveobservations made by the ground-basedinterferometers in the high-frequency regime(10 to 10

3Hz) with observations in space in the

low-frequency regime.

Ground-based interferometers can observe thebursts of gravitational radiation emitted bygalactic binaries during the final stages(minutes and seconds) of coalescence, whenthe frequencies are high and both theamplitudes and frequencies increase quicklywith time. At low frequencies, which are onlyobservable in space, the orbital radii of thebinary systems are larger and the frequenciesare stable over millions of years. Coalescencesof massive black holes are only observablefrom space. Both ground- and space-baseddetectors will also search for a cosmologicalbackground of gravitational waves. Since bothkinds of detectors have similar energysensitivities, their different observingfrequencies are ideally complementary:

ten years of fundamental physics

observations can provide crucial spectralinformation.

In Newton’s theory, the gravitational interactionbetween two bodies is instantaneous, butaccording to Einstein’s theory of gravity thisshould be impossible because the speed of lightrepresents the limiting speed for all interactions.If a body changes its shape, the resultingchange in the force field will make its wayoutward at the speed of light. In Einstein’s theoryof gravity massive bodies produce ‘indentations’in the ‘fabric’ of space-time and other bodiesmove in this curved space-time taking theshortest path. If a mass distribution moves in aspherically asymmetric way, then theindentations travel outwards as ripples in space-time called ‘gravitational waves’.

Gravitational waves are fundamentally differentfrom the familiar electromagnetic waves. Whilethe latter, created by the acceleration of electriccharges, propagate in the framework of spaceand time, gravitational waves, created by theacceleration of masses, are waves of the space-time fabric itself. Unlike charge, which exists intwo polarities, mass always comes with thesame sign. This is why the lowest-orderasymmetry producing electromagnetic radiationis the dipole moment of the charge distribution,whereas for gravitational waves it is a change inthe quadrupole moment of the massdistribution. Hence those gravitational effectsthat are spherically symmetric will not give rise togravitational radiation. A perfectly symmetric

Figure 3. Schematic of theLISA configuration (not to

scale). Three distantsatellites linked by infrared

laser beams form a giant 5 million km triangular

interferometer, which issensitive to fluctuations in

the separations betweenthe satellites caused by

gravitational waves. Theplane of the triangle istilted by 60º out of the

ecliptic

collapse of a supernova will produce no waves,whilst a non-spherical one will emit gravitationalradiation. A binary system will always radiate.

Gravitational waves are a direct consequenceof Einstein’s theory of General Relativity. If thattheory is correct, gravitational waves mustexist, but up to now they have not beendetected. There is, however, strong indirectevidence for the existence of gravitationalwaves: the binary pulsar PSR 1913+16 losesenergy exactly at the rate predicted by GeneralRelativity through the emission of gravitationalradiation.

Gravitational waves distort space-time: in otherwords, they change the distances between freemacroscopic bodies. A gravitational wavepassing through the Solar System creates atime-varying strain in space that periodicallychanges the distances between all bodies inthe Solar System in a direction perpendicular tothat of wave propagation. These could be thedistances between spacecraft and the Earth,as in the case of Ulysses or Cassini (attemptshave been and will be made to measure thesedistance fluctuations) or the distances betweenshielded proof masses inside widely separatedspacecraft, as in the case of LISA. The mainproblem is that the relative length change dueto the passage of a gravitational wave isexceedingly small. For example, the periodicchange in distance between two proof masses,separated by a sufficiently large distance, dueto a typical white-dwarf binary at a distance of50 pc, is only 10

-10m. This does not mean that

gravitational waves are weak in the sense thatthey carry little energy. On the contrary, asupernova in a not too distant galaxy willdrench every square metre here on Earth withkilowatts of gravitational radiation intensity. The

resulting length changes, though, are very smallbecause space-time is an extremely stiff elasticmedium, so that extremely large energies areneeded to produce even minute distortions.

It is because of the extremely small distancechanges that gravitational waves have not yetbeen detected. However, with the LISA spaceinterferometer, orbiting the Sun at 1 AU, millionsof sources will be detected in one year ofobservation with a signal-to-noise ratio of 5 orbetter. The LISA mission comprises threeidentical spacecraft located 5x10

6km apart

forming an equilateral triangle (Fig. 3). Thedistance between the spacecraft — theinterferometer arm length — determines thefrequency range in which LISA can makeobservations; it was carefully chosen to allowfor the observation of most of the interestingsources of gravitational radiation. The centre ofthe triangular formation is in the ecliptic plane, 1 AU from the Sun and 20º behind the Earth.The plane of the triangle is inclined at 60º withrespect to the ecliptic. These particularheliocentric orbits for the three spacecraft werechosen such that the triangular formation ismaintained throughout the year, with thetriangle appearing to rotate about the centre ofthe formation once per year.

While LISA is basically a giant Michelsoninterferometer in space, the actualimplementation in space is very different from alaser interferometer on the ground and is muchmore reminiscent of a ‘spacecraft tracking’technique, but then realised with infrared laserlight instead of radio waves. The laser lightgoing out from the centre spacecraft to theother corners is not directly reflected backbecause very little light intensity would be leftover that way. Instead, analogous to an RF

r bulletin 98 — june 1999 bull

The spacecraft mainly serve to shield the proofmasses from the adverse effects due to thesolar radiation pressure, and the spacecraftposition does not directly enter into themeasurement. It is nevertheless necessary tokeep all spacecraft moderately accurately (10

-8mHz

-1/2in the measurement band) centred

on their respective proof masses to reducespurious local noise forces. This is achieved bya ‘drag-free’ control system, consisting of anaccelerometer (or inertial sensor) and a systemof ion thrusters. Capacitive sensing is used tomonitor the relative motion between eachspacecraft and its test masses. These positionsignals are used in a feedback loop tocommand micro-Newton ion-emittingproportional thrusters, to enable the spacecraftto follow its test masses precisely and withoutintroducing disturbances in the bandwidth ofinterest. The same thrusters are used forprecision attitude control relative to theincoming optical wave fronts.

Each of the three LISA spacecraft has a launchmass of about 460 kg (incl. margin). Ion drivesare used for the transfer from the Earth orbit tothe final position in interplanetary orbit. All threespacecraft can be launched by a single Delta II7925H.

LISA was proposed to ESA in May 1993 inresponse to ESA’s Call for Mission Proposalsfor the third Medium-Size Project (M3). Theproposal was submitted by a team of Americanand European scientists who envisaged LISA

ten years of fundamental physics

transponder scheme, the laser on the distantspacecraft is phase-locked to the incominglight providing a return beam with full intensityagain. After being transponded back from thefar spacecraft to the centre spacecraft, the lightis superposed with the on-board laser lightserving as a local oscillator in a heterodynedetection.

Each spacecraft contains two opticalassemblies (Fig. 4). The two assemblies on onespacecraft each point towards an identicalassembly on each of the other two spacecraft.A 1 W infrared laser beam is transmitted to thecorresponding remote spacecraft via a 30-cmaperture f/1 Cassegrain telescope. The sametelescope is used to focus the very weak beam(a few pW) coming from the distant spacecraftand to direct the light to a sensitivephotodetector, where it is superimposed with afraction of the original local light. At the heart ofeach assembly is a vacuum enclosurecontaining a free-flying polished platinum-goldcube, 4 cm in size, referred to as the ‘proofmass’, which serves as an optical reference(‘mirror’) for the light beams. A passinggravitational wave will change the length of theoptical path between the proof masses of onearm of the interferometer relative to the otherarm. The distance fluctuations are measured tosub-Angstrom precision which, whencombined with the large separation betweenthe spacecraft, allows LISA to detectgravitational-wave strains down to a level oforder ∆l / l =10

-23in one year of observation.

Figure 4. Left: Cut-awayview of one of the threeidentical LISA spacecraft.The main structure is a ringwith a diameter of 1.8 mand a height of 0.48 m,made from graphite-epoxyfor low thermal expansion.A lid on top of thespacecraft is removed toallow a view of the Y-shapedpayload.Right: Detail of the payloadon each Y-shaped LISAspacecraft, consisting oftwo identical telescopesand two optical bencheseach housing a drag-freetest mass (the yellowcubes in the centres)

as an ESA/NASA collaborative project. Themission was conceived as comprising fourspacecraft in a heliocentric orbit forming aninterferometer with a baseline of 5x10

6km.

LISA was selected for study as an ESA-onlyproject, but it became clear quite early in theAssessment Phase that it was not likely to be asuccessful candidate for M3 because the costfor an ESA-only LISA considerably exceededthe M3 limit of 350 M€. In December 1993,LISA was therefore re-proposed to ESA, thistime as a Cornerstone project for ‘Horizon2000 Plus’, involving six spacecraft in aheliocentric orbit with a pair of spacecraft ateach vertex of an equilateral triangle. Both theFundamental Physics Topical Team and theSurvey Committee realised the enormousdiscovery potential and timeliness of the LISAProject and recommended it as the ThirdCornerstone for ‘Horizon 2000 Plus’. However,the Survey Committee also noted that theinclusion of LISA as a Cornerstone ‘will requirea modest increase in the funding of the ESAScientific Programme beginning in 2001’.

Being a Cornerstone in ESA’s Space ScienceProgramme implies that, in principle, themission is approved and that funding forindustrial studies and for technologydevelopment is provided right away. The launchyear, however, is dictated by scientific prioritiesand the availability of funding. Consideringrealistic funding scenarios for ESA’s SpaceScience Programme, LISA could probably onlybe launched after the other two Cornerstonesof Horizon 2000 Plus, namely Mercury Orbiterand Interferometry (either GAIA or IRSI), i.e.after 2017. Because of the large inequalitybetween the ESA and NASA science budgets,it must then be expected that even the mostoptimistic opportunity for ESA to launch theLISA Cornerstone will be pre-empted by anearlier NASA mission. For this and several otherreasons, it was decided in January 1997 to putLISA on an equal footing with the other twoCornerstones in Horizon 2000 Plus, with apossible launch as early as 2009. A launcharound 2009 would be ideal as it is around thattime that the first detection of gravitationalwaves by the ground-based interferometers inthe high-frequency regime can be expected.There still remained, however, the problem ofLISA’s cost exceeding the Cornerstone limit,with the cost of the six-spacecraft projectinitially estimated to be about 800 M€.

In 1996 and 1997, the LISA team made severalproposals as to how the cost might bedrastically reduced without compromising thescience, the most important being thereduction in the number of spacecraft from six

to three. This was achieved by replacing eachpair of spacecraft at the vertices of thetriangular configuration by a single spacecraftcarrying essentially two identical instruments ina Y-shaped configuration. With these and a fewother measures, the total launch mass could bereduced from 6.8 to 1.4 t and the total costcould be reduced accordingly (to $330 M,excluding the payload, according to a recentJPL Team-X cost estimate, a figure not yetconfirmed by ESA).

Perhaps most importantly, the LISA StudyTeam and ESA’s Fundamental Physics AdvisoryGroup (FPAG) proposed in February 1997 thatthe LISA mission be carried out in collaborationwith NASA. This makes sense not only from acost-saving point of view, but also because theLISA team is an international one and the LISAmission-definition work was carried out jointlybetween Europe and the USA. The FPAGrecommended limiting the cost to ESA to 150 M€ in this collaboration, as was done on asmaller scale with a cost cap of 20 M€ forSTEP.

Presently, ESA is carrying out a six-monthindustrial system-level study, with support fromthe LISA Science Team of about 30 scientists.On the US side, a LISA Mission Definition Teamconsisting also of about 30 scientists has beenformed. Both teams have partial teammembership overlap to ensure that both teamswork towards defining the same mission. As afirst activity, in February 1999 the US teamagreed on a Technology Plan with a totalbudget of $33 M to be implemented by theLISA Pre-Project Office at JPL in the nextcouple of years.

The need to demonstrate key LISAtechnologies in spaceWhen LISA was presented together with theother three Cornerstone missions in ‘Horizon2000 Plus’ to the scientific communities in theESA Member States in 1995, prior to theMinisterial Conference in Toulouse, severalDelegations expressed concern about thetestability of key LISA technologies on theground, remarking: ‘It is very risky to launch amission costing nearly 800 M€ without havinga high degree of confidence that the keytechnologies will work; however, testing these technologies on the ground under 1 gconditions is not possible’.

Similar concern was also raised by ESA’sScience Programme Committee (SPC) in May1996: ‘The technology required for asuccessful LISA mission is extremelydemanding and, furthermore, some keysubsystem elements (e.g. drag-free control,

r bulletin 98 — june 1999 bull

Figure 5. Cut-away view of the proposed LISAtechnology-demonstrationspacecraft, showing twotest masses (4 cm Pt-Aucubes, highlighted inyellow) with lasermetrology

On the European side, a technology-demonstration mission addressing thetechnology for both the LISA and a multi-satellite infrared interferometry mission isforeseen for launch in 2005 as the second in aseries of Small Missions for AdvancedResearch in Technology (SMART-2). ESA wouldnevertheless be interested in exploring thepossibility of carrying out this mission, at leastin part, in cooperation with NASA at an earliertime if ESA’s technology needs could berealised in a more cost-effective manner.

Next-generation fundamental-physicsmissionsIn response to the Call for Mission Ideas forHorizon 2000 Plus in 1993, ESA received 28proposals for fundamental-physics missions,which were subsequently evaluated by TopicalTeam 5 (TT-5). This constitutes the mostcomplete survey of the possibilities of space forfundamental physics so far. Based on theirscientific objectives, the proposals werecategorised into six groups. One group wasaimed at testing General Relativity and itsalternative theories of gravity by measuring theso-called ‘PPN parameters’ (explained below),while another group of proposals searched fornew particles in space. One in each group wasselected and briefly described below; apartfrom STEP and LISA, these two proposalsconsistently received the highest rankings intwo evaluations (by the FPAG for M3 and by TT-5 for Horizon 2000 Plus).

Testing theories of gravity (and GeneralRelativity in particular) has received renewedattention especially for the cosmologicalconsequences of possible violations. A largeclass of alternative scalar-tensor theories

zero-g accelerometry) can only be tested inspace. The overall technical feasibility of LISAstill needs to be demonstrated.’

In February 1997, during a meeting at JPL, theLISA team came to the conclusion that atechnology-demonstration mission is anecessary precursor to the LISA project. Bythat time it had become clear that the cost forsuch a mission could be kept relatively low. TheLISA team, together with their US colleagues,therefore submitted a detailed proposal to ESAin May 1998 for a technology-demonstrationmission (Fig. 5) that would test:

– the inertial sensor performance to within anorder of magnitude of the LISA requirements

– the low-frequency laser interferometrybetween two inertial sensors

– drag-free satellite operations using fieldemission ion thrusters.

In December 1998, an updated proposal wassubmitted which was broader in scope, alsoaddressing key IRSI and XEUS technology-demonstration needs. This revised proposalsuggested a collaboration with NASA on ST-5(previously called DS-5), with a launch in 2003.For the ST-5 slot of $28 M, there are currentlythree candidates: solar-sail technology, acluster of nano-satellites, and a DisturbanceReduction System. The latter would include thetesting of key technologies for LISA and wouldalso involve some aspects of spaceinterferometry for imaging interferometryprojects. One of these three candidate projectswill be selected by NASA in late July 1999,following parallel six-month studies at Phase-Alevel. ESA was recently invited by NASA toconsider a possible collaboration on theDisturbance Reduction System. In the mostlikely collaborative scenario, ESA wouldcontribute all or part of a small drag-freesatellite (with a cost cap of 10 M€) which wouldcarry the relevant technology items from bothESA and NASA, with NASA providing thelaunch and mission operations.

Ideally, such a technology-demonstrationmission should be launched about five yearsbefore LISA. A launch much earlier would notallow full utilisation of the latest technologies tobe tested, while a launch much later would notallow full advantage to be taken of theknowhow obtained during the technology-demonstrator flight in the design phase of theLISA mission. To preserve the possibility of aaunch of the NASA/ESA collaborative LISAmission in 2009, the technology-demonstrationmission should therefore ideally be launched inthe 2003-2005 time frame.

ten years of fundamental physics

contain a cosmological attractor mechanismtoward General Relativity. Approximateestimates based upon inflationary cosmologiesindicate that in the present epoch GeneralRelativity provides indeed an excellentdescription of gravitation, accurate to a level of10

-7— 10

-5. However, due to the uncertainties

of the theoretical estimates, every experimentable to improve the present accuracies issignificant. If a discrepancy is confirmed, itwould indicate that mass does not curve spaceas predicted by Einstein’s theory. Such a resultwould be a milestone in our knowledge of thefundamental laws of the Universe. Ameasurable violation could indicate the type ofscalar-tensor theory evolution of the Universe.In the weak-field, small-velocity approximation,theories of gravity are classified using theParametrised-Post Newtonian formalism(PPN). In the simplest case, this classification isbased upon just two parameters, called β andγ. The former measures the amount ofnonlinearity in the superposition law for gravity,and the latter the space curvature produced bya unit mass.

The proposed SORT (Solar Orbit Relativity Test)mission is aimed at improving, by four orders ofmagnitude, the measurement of the PPNparameter γ by measuring the deflection andthe time delay of laser light passing close to theSun. γ is a measure of the strength of thecoupling of mass to the curvature of space; inGeneral Relativity γ =1. Present experimentallimits on |γ-1| are about 10

-3. Measurements of

γ at the 10-7

level have considerable theoreticalsignificance because generic tensor-scalartheories of gravity predict a natural weakeningduring the cosmological expansion of theobservable deviations from General Relativitydown to the level of 10

-7. SORT proposes to

combine a time-delay experiment (via lasersignals sent from the Earth and recorded byprecise clocks on board two satellites orbitingthe Sun) with a light-deflection experiment(interferometric measurement on Earth of theangle between the two light flashes emittedfrom the same satellites). SORT is the modernversion of the SOREL mission studied by ESAin 1970/71. For comparison, SOREL aimed atdetermining γ to a precision of 5x10

-4and β to

5x10-3

.

The proposed SSPIN (Satellite Search forPseudoscalar Interactions) mission is designedto search for a weak spin-dependent force atthe level of gp • gs = 6x10

-37(where gp and gs

are spin-coupling constants) at the range of 1 mm, 10 orders of magnitude better than iscurrently achievable on the ground. This ispossible by placing a highly sensitive payload atcryogenic temperature in a drag-free satellite in

a geosynchronous orbit. This experiment waspreviously included in the M2- and M3-STEPpayloads and studied for several years. SSPINwould be three orders of magnitude moresensitive than the MSC experiment on M3-STEP, which would put us in the realm ofactually detecting the axion, a hypothetical,weakly interacting, massive particle which hasbeen postulated to reconcile the theoreticallyallowed level of charge conjugation parity (CP)violation in the strong interactions, with thecurrent upper limit to the electric dipolemoment of the neutron. It has also beeninvoked as a possible candidate for the elusive‘Dark Matter’ in the Universe.

Both missions rely on drag-free control andhigh-precision accelerometers. In addition,SSPIN needs cryogenics and SQUID sensing.These techniques will already have beendeveloped for STEP, which is therefore an idealtestbed not only for LISA, but also forfundamental-physics missions in the moredistant future. As these missions will probablyonly fly after 2010, further improvements intechnology can be expected which will makethem even more exciting for the physicscommunity. As fundamental physics is a veryyoung field, many more competitive ideas formissions can be expected to emerge by thetime the next fundamental-physics mission willbe selected after STEP and LISA.

Fundamental physics on non-dedicatedESA missionsExisting and future ESA Solar Systemexploration and astronomy missions also offerattractive possibilities for fundamental physics.Precise, two- or three-way tracking ofinterplanetary probes, such as the Ulysses andCassini spacecraft, can set upper limits on low-frequency gravitational waves. These appear asirregularities in the time-of-communicationresiduals after the orbit of the spacecraft hasbeen fitted. The irregularities have a particularsignature. Searches for gravitational waveshave produced only upper limits so far, but thisis not surprising: their sensitivity is far short ofpredicted wave amplitudes. This technique isinexpensive and well worth pursuing, but will belimited for the foreseeable future by somecombination of measurement noise, thestability of the frequency standards, and theuncorrected parts of the fluctuations inpropagation delays due to the interplanetaryplasma and the Earth’s atmosphere.Consequently, it is unlikely that this method willrealise an r.m.s strain sensitivity much betterthan 10

-17, which is six orders of magnitude

worse than that of the space-based LISAinterferometer.

r bulletin 98 — june 1999 bull

space this test can be done three orders ofmagnitude more precisely than on the ground.

Fundamental Physics as an emergingspace-science disciplineIt is now widely understood that the scientificobjectives of fundamental-physics missions aredistinctly different — questioning the laws ofNature — from the scientific objectives ofastronomy and Solar System missions —accepting and applying the laws of Nature.Also, the technologies used in fundamental-physics missions (the spacecraft typically carryproportional thrusters and high-precisionaccelerometers to allow drag-free operation),the requirements on spacecraft design (e.g. nomoving parts, no deployable solar arrays,extremely high thermal stability) and the highdegree of spacecraft/experiment inter-relationship (e.g. the signals from the payloadare used to control the spacecraft) are distinctlydifferent from Solar System exploration andastronomy missions.

Fundamental-physics experiments on theground are limited in achievable precision.Many tests can be carried out in space withmuch higher precision (e.g. STEP: 10

6better),

whilst some observations can only be made inspace (e.g. gravitational waves at lowfrequencies). It is therefore no surprise that thenumber of proposals received in response toESA’s Calls for Mission Proposals has beensteadily growing: 5 (out of 22) proposals for M2,16 (out of 48) proposals for M3 and 28 (out of~100) proposals for Horizon 2000 Plus.Roughly a quarter of all mission proposals havebeen submitted in the past and are likely to besubmitted in the future by the fundamental-physics community.

Fundamental Physics in Space is a rapidlygrowing new field with enormous discoverypotential in physics, and a major technologydriver. ESA recognised this in early 1994 bysetting up the Topical Team on FundamentalPhysics (TT-5) and in December 1994 bysetting up the Fundamental Physics AdvisoryGroup (FPAG). NASA followed in 1997 bysetting up the Gravitational and RelativisticPhysics Panel and the Fundamental PhysicsDiscipline Working Group. In 1996, COSPARdecided to create a new Scientific Commission(SC-H) to cater for the needs of thefundamental-physics community. ESA’s Long-Term Space Science Programme was fullydescribed in 1984 in ‘Space Science: Horizon2000’ and in 1995 in ‘Horizon 2000 Plus’; bothpublications include chapters on fundamentalphysics, the latter describing a Cornerstonemission and a number of medium-size andsmall missions. However, it is also stated in the

During conjunctions, the radio signal passingclose to the Sun experiences a measurabletime delay and frequency shift, which allowsone to determine the space curvatureparameter γ. The radio-science investigation onCassini will allow us to test this prediction to aprecision of ~10

-4. The best experiment to date

was performed by the Viking Mars Landermission in 1971, which achieved a precision of2x10

-3.

A Mercury Orbiter would allow γ to bemeasured to a precision of ~ 2x10

-5and β to a

precision of 7x10-4

. On the astronomy side,Hipparcos has already achieved a precision of10

-3for γ and GAIA is expected to achieve a

two orders of magnitude improvement overthat.

Fundamental-physics experiments on theISSAlready under development is the AMS (AlphaMagnetic Spectrometer) to be flown on theInternational Space Station (ISS) in 2004/5.There are also two precursor flights on theSpace Shuttle; one has already taken placesuccessfully in 1998, the other will take place ina few years’ time. A major objective of the AMSis the search for antimatter in the Universe.According to standard theories, at the verybeginning there should be as much matter asantimatter, but the antimatter has not yet beendetected. The AMS is a US-led project withmost contributions coming from Europe.

Also under development for flight on the ISS isthe PHARAO (Project d’Horloge Atomique parRefroidissement d’Atomes en Orbite) clock,which uses laser-cooled caesium atoms toimprove the accuracy of the time frequencystandard by two orders of magnitude. The French PHARAO atomic clock iscomplemented on the ISS by a Swiss hydrogenmaser clock to provide a longer-term frequencystandard; together they form the Atomic ClockEnsemble in Space (ACES). The ultra-highprecision of PHARAO will allow ‘new records’to be set for two fundamental-physicsconstants:

– determination of the space-curvatureparameter γ to 1 part in 10

5(two

orders of magnitude improvement)– determination of the gravitational redshift to

3 parts in 106

(almost two orders ofmagnitude improvement over the GravityProbe A suborbital flight in 1976).

Proposed for flight on the ISS, but not yetaccepted, is a test of the Equivalence Principleusing protons and antiprotons — the WeakEquivalence Antiproton Experiment (WEAX). In

ten years of fundamental physics

latter publication that the inclusion of thefundamental-physics discipline will require amodest increase in the funding of ESA’s SpaceScience Programme.

NASA has recently published a ‘Roadmap forFundamental Physics in Space’, whichrepresents a long-term framework within whichto establish and advocate NASA’s futureresearch and technology developmentprogramme in fundamental physics. It identifiesthree sets of focussed scientific investigations(called ‘campaigns’), which comprise ascientifically rewarding, technologicallychallenging, flexible and exciting programme offundamental-physics research in space. Thesecampaigns are:

– Gravitational and Relativistic Physics– Laser Cooling and Atomic Physics – Low-Temperature and Condensed-Matter

Physics.

The Roadmap describes a number of missionsor experiments in each of these campaigns.The Gravitational and Relativistic PhysicsCampaign describes six missions, among themthe NASA/ESA collaborative STEP and LISAmissions.

Fundamental physics as a discipline in its ownright is also now recognised by many Europeanspace agencies. CNES defined in 1993 a newscientific theme ‘Fundamental Physics’ and setup a Fundamental Physics Working Group. ASIhas recognised the emergence of this newdiscipline through the appointment of arepresentative specifically for fundamentalphysics in its Scientific Council. PPARC hasrecently set up a science committee and isalready demonstrating its interest infundamental physics by encouragingsubmissions on gravitational-wave detectors inspace. DLR has identified fundamental physicsas a scientific discipline with a ‘high priority inthe future’ and has explicitly expressed a wishto participate in LISA and STEP. The AustrianSpace Agency (ASA) organised (together withESA) the 1997 Alpbach Summer School‘Fundamental Physics in Space’ (Proceedingsavailable as ESA SP-420).

Following a recommendation by the ESASpace Science Department (SSD) AdvisoryCommittee in 1997 (a group of outside seniorscientists who review SSD’s activities every twoyears), a Fundamental Physics Office was setup in SSD in mid-1998, initially with two staffscientists, in addition to the already existingSolar System, Astrophysics and Earth SciencesDivisions.

ConclusionConsidering that the required technologies arenow mature, that there is a sizeable communityin Europe in need of space flights (a quarter toa third of all proposals come from thatcommunity), and that fundamental physics inspace is now recognised by many spaceagencies, it is only a question of time andmoney until Europe realises its firstfundamental-physics mission. r

r bulletin 98 — june 1999 bull