Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
RadiationPressureonaDiffractiveSailcraft GROVERA.SWARTZLANDER,JR. CenterforImagingScience,RochesterInstituteofTechnology,Rochester,NY14623 Correspondingauthor:[email protected] Advanceddiffractivefilmsmayaffordadvantagesoverpassivereflectivesurfacesforavarietyspacemissions thatusesolarorlaserin-spacepropulsion.Threecasesarecompared:Sun-facingdiffractivesails,Littrow diffraction configurations, and conventional reflective sails. A simple Earth-to-Mars orbit transfer at a constantattitudewithrespecttothesun-linefindsnopenaltyfortransparentdiffractivesails.Advantagesof thelatterapproachincludeactivelycontrolledmetasailsandthereuseofphotons. 1.INTRODUCTION Radiation pressure is a process whereby optical momentum is transferredtomatterbymeansofreflection,diffraction,orabsorption. Thiseffect,firstpredictedbyMaxwell[1,2],isstrongenoughtoimpart forceandtorqueonsatellitesowingtothepressurefromsunlight[3: Mariner10,1974],aswellasonmicroscopic-scalebodiesonEarthby use of a concentrated beam of light. In the latter case the object is generallytrappedintwoorthreespatialdimensionsowingtotheuse of a focused beam. Pure optical levitation may be achieved when a uniform beam of light is used, thereby affording free movement, as occurswithin-spacesolarradiationpressure.Arichbodyofscientific literature spanning more than nine decades [4,5] describes various missionswherebythesunoralasersourceisusedtodriveasailcraft. These range from Earth-orbiting satellites, station-keeping at a Lagrange point, and fly-by missions to inner and outer planets, asteroids, comets, and distant stars [6,7]. International and citizen space agencies have recently prepared early-stage demonstration missions such as IKAROS (JAXA), NanoSail D (NASA), Lightsail (PlanetarySociety),markinganageakintotheearlydaysofairflight. Juxtaposed with this interest is a proliferation of research on advanceddiffractivematerials.Theseincludepolarizationdiffraction gratingsandothermetamaterialshavingengineeredopticalproperties [8-23]. Films of these materials may, in principle, be thin (a few micrometersthick)withalowarealdensity,andproducedacrosslarge areas, thereby affording passive or active control of a sailcraft. We propose that diffractive materials may be used in place of reflective sails to afford non-mechanical navigation (e.g., without varying the sailcraft attitude), photon re-cycling, and higher efficiencies. For example,binaryswitchingbetweenthe+1and-1orderofanelectrooptically controlled diffractive film may provide continuous navigational control authority. This report advances that vision by comparingtheradiationpressureonreflectiveanddiffractivesurfaces, and applying the force models to a two-dimensional orbit-raising scenario. The paramount finding is that no penalty is found for the areal density and momentum transfer. A complete optimization analysisbasedonmission-specificcriteria,notpresentedhere,isleft for further work. Polarization, dispersion, space-qualified materials, andotherpracticalconcernsarealsoleftforfuturedevelopment. Radiation pressure is described in Section 2 for flat films. Four specialcasesareexaminedforlightgovernedbythelawofreflection anddiffraction,includingbothreflectionandtransmissioncasesinthe Littrowconfiguration,andagratingatnormalincidence.Asinglehighefficiencydiffractionorderisassumed.Atwo-bodypointmassmodel comprisedofthesailandtheSuninatwo-dimensionalorbitalplaneis describedinSection3.TheforcemodelisappliedinSection3(A)to raise the sailcraft from an Earth orbit to that of Mars. Section 3(B) outlines advanced diffractive sail control schemes. Concluding remarksareofferedinSection4. 2.RADIATIONPRESSUREONAFLATSAIL A uniform plane wave propagating along the incident unit vector direction k̂i withirradiance I i isassumed.Theincidentforceona flatfilmofsurfaceareaAscaleswiththecollectedattitude-dependent incidentbeampower Pi (θ i ) : ! Fi = Pi (θi )k̂i / c = (I i A / c)k̂i cosθi (1) wherecisthespeedoflight, θ i istheanglesubtendingtheincident wavevectorandtheoutwardsurfacenormalofthefrontface n̂ f (see Fig.1).Forsailcraftmovingmuchslowerthanthespeedoflightthe Dopplershiftofreflectedandtransmittedwavesmaybeignored,and thusthemagnitudesofthewavevectorsareequal: ki = kr = kt .Fora non-absorbingfilmofreflectanceR,transmittanceT,withR+T=1,the netforcemaybeexpressed: ! ! ! ! Frp = Fi + Fr + Ft = (I i A / c)cosθi k̂i − Rk̂r − Tk̂t ( ) (2) CorrectionstoEq.(2)arerequiredifasignificantfractionofthelightis diffusely reflected or transmitted, or if heat is emitted owing to absorption. Here we seek to investigate ideal cases, leaving higher ordermaterial-specificcomplicationsforlaterexplorations. 1 ! ηrr,t = ηrpr,t ⋅ r̂ = cosθi (1+ cos(θ r,t − θi )) ! ηφr,t = ηrpr,t ⋅ φ̂ = −cosθi sin(θ r,t − θi ) (6) C.SpecialCases Severalcasesofspecialinterestmaybedescribed,includingfilmsthat obey the Reflection Law, diffraction from a grating at the Littrow conditionforbothreflectionandtransmission,and,diffractionfroma gratingforthecaseofnormalincidence. i)ReflectionLaw(RL). θ r = −θ i , θ r ≤ π / 2 ! η RL = 2cos3 θi n̂ = 2cos 2 θi cosθi r̂ − sin θiφ̂ ( ) (7) Fig.1.Planeofincidenceforraysataflatplanarfilm.Filmcoordinates (n,p)andsunlinecoordinates(r,φ)areshown. ii)LittrowReflection(LR)Grating. θ r = θ i , θ r ≤ π / 2 A.Forcecomponentsinthefilmcoordinatesystem(n,p) If the reflected and transmitted rays are constrained to the plane of incidence,theforcemaybeprojectedalongtwounitvectors:normal and parallel to the film with respective unit vectors n̂, p̂ . The ray directionsmaythenbeexpressed(seeFig.1): k̂i = cosθi n̂ − sin θi p̂ , k̂r,t = −cosθ r,t n̂ + sin θ r,t p̂ (3) whereweassign n̂ = n̂b = −n̂ f .Thethreeradiationpressureforces maythenbeexpressed: ! Fi = F0 cosθi k̂i = F0 cos 2 θi n̂ − cosθi sin θi p̂ ! Fr = −F0 Rcosθi k̂r = F0 Rcosθi (cosθ r n̂ − sin θ r p̂ ) ! Ft = −F0T cosθi k̂t = F0T cosθi (cosθ t n̂ − sin θ t p̂ ) ( ) ! η LR = 2cosθi (cosθi n̂ − sin θi p̂) = 2cosθi r̂ (8) iii)LittrowTransmission(LT)Grating. θ t = π − θ i , θ t ≥ π / 2 ! η LT = −2cosθi sin θi p̂ ( = 2cosθi sin θi sin θi r̂ − cosθiφ̂ ) (9) iv)Gratingatnormalincidence: θ i = 0, n̂ = x̂, p̂ = ŷ ! η NI = (1+ cosθ r,t )r̂ − sin θ r,tφ̂ ( ) (10) Theforceisdirectedalongthesurfacenormalofthebackfacefora reflective film, but the force is parallel to the film for a transmission grating at the Littrow condition. A reflection grating at the Littrow condition is the vector sum of these two values, with a magnitude equalto 2F0 cosθ i . (4) wherethescalingforce F0 = (I i A / c) isdefinedforconvenience.For example, at 1 AU the solar irradiance is 1.37 [kW/m2] and thus F0 = 4.57 [µ N] for a square meter sail. Below we shall consider Thecomponentsofforceinboththe(n,p)and(r,φ)basisareplottedfor thesethreecasesinFig.2(a,b)asafunctionoftheangleofincidence. WeseefromEq.(8)andfromFig.2(b)thatastheattitudeofaLittrow reflection grating changes, the force remains directed only along the sunline.This,andtherelativelylargerforce,mayprovideadynamic advantageinsomecases(forexample,anoscillatingattitudedoesnot produce an undulating trajectory). In contrast, an orbit-raising eitheraperfectingreflectingorperfectlytransmittingfilm(notatedby r or t). In such cases the normalized force may be expressed as an ! ! efficiencyvector η = Frp / F0 : maneuvermayrequireasignificanttransverseliftforcewith Fφ ≥ Fr . ! η r,t = cosθi ((cosθi + cosθ r,t )n̂ − (sin θi + sin θ r,t ) p̂) constitutetheoptimalliftforce,owingtheweakeningfactorof cosθ i (5) B.Forcecomponentsintheorbitalcoordinatesystem(r,φ ) Thesunline k̂i isassumedtoradiatefromapoint-likesun,parallel totheorbitalradialunitvector r̂ inatwodimensionalplane: r̂ = k̂i = cosθi n̂ − sin θi p̂ and φ̂ = sin θi n̂ + cosθi p̂ .Theangleof In those cases a reflective mirror or a transmissive grating at the Littrowconditionmaybedesired.However,thosetwocasesdonot in both Eq. (7) and (9). A diffractive film illuminated at normal incidenceovercomesthisdisadvantage.Inthiscasethemagnitudeof 3/2 force for a normally incident grating is 2 F0 cos(θ r,t / 2) , providing themaximumamountofforceatanydiffractionangle.Whatismore, NI theliftforceefficiency, ηφ maybeaslargeat100%. incidence θ i describestheattitudeofthefilmwithrespecttothe sunline.FromFig.1wesee x̂ = k̂i = cosθ i n̂ − sin θ i p̂ and ŷ = sin θi n̂ + cosθi p̂ andthusweexpressthecomponentsofEq.s(5) 2 ! ! ! ! ! mM F = FG + FRP = −G 2 r̂ + A(I E / c)(RE / r) 2 η = ma r (11) whereIE=1.37[kW/m2]istheso-calledsolarirradianceconstantatr= ! ! RE=1[AU],Gistheuniversalgravitationalconstant,and η = Frp / F0 is the radiation pressure efficiency vector. The acceleration may be expressedincircularcoordinates: ! ! d 2r a = 2 = !! r − rφ! 2 r̂ + 2r!φ! + rφ!! φ̂ dt ( ) ( ) (12) Expressingtheareamassdensity σ = m / A ,Eq.(11)maybewritten: ! F = −maM (RE / r) 2 (1− ηrσ cr / 2σ )r̂ − (ηφσ cr / 2σ )φ̂ ( ) (13) 2 2 where σ cr = 2RE I E / GMc = 1.54g / m is a characteristic mass Fig.2.Radiationpressureefficiencycomponentsprojectedalongthe body coordinates (n,p) and sunline coordinates (r, φ)., plotted as a function of the angle of incidence, θi. (RLaw = Reflection Law, LR = LittrowReflection,LT=LittrowTransmission) Parametric force lines, shown in Fig. 3, provide a convenient way to representtheforcecomponentsforthefourcasesexaminedabove.A largemagnitudeofforceandliftcomponentareclearlyaffordedbythe normallyincidentgratingcase. density and aM = −GMRE−2 = 5.931 [mm/s 2 ] is the gravitational acceleration of the sun at the radial orbit of the Earth. The ratio σ * = σ cr / σ iscalledthelightnessnumber.Notethatthesailcraftis * neutrallybuoyantwithrespecttosolargravitywhen σ = 1 ,assuming ηr = 2 (asun-facingmirrororretro-reflectinggrating).Thevaluesof the radiation pressure terms in Eq. (13) tend toward infinity as the areal density of the sailcraft vanishes (σ → 0) . With current * technology a modest value of σ = 0.1 is within reach for CubeSat compatiblesailcraftforexample. A.SynchronousTransferOrbitswithConstantSolarAttitude Todemonstratethatthereisnoobviousadvantageofreflectiveover diffractive sailcraft, let us examine a simplified Earth to Mars rendezvousmission.Thetrajectorywillbeasmoothoutwardspiral,as showninFig.3.Forconvenience,weassumetheplanetshavecircular co-planarorbitsofrespectiveradiiRE=1.0[AU]andRM=1.5[AU].The initial (t = 0) and final (t = T) conditions for the rendezvous are ! ! r(t = 0) = RE , v(t = 0) = v Eφ̂ , r(t = T ) = RM , and v(t = T ) = v M φ̂ , where v E = GM / RE and v M = GM / RM .Thedesiredvalueof Fig. 3. Parametric force efficiency lines for reflective and diffractive films. 3. SOLAR PRESSURE ON AN ORBITING DIFFRACTIVE ORREFLECTIVESAIL The solar propulsion on a sailcraft of area A may be estimated by assuming point-like masses for the sailcraft, m, and sun, M, and ignoringothergravitationalbodies.Furthersimplicityisachievedby assumingatwodimensionalsailcrafttrajectoryinthex,y-plane(r,φplane), with the sun at the origin. The net force from gravity and radiationpressuremaybeexpressed[6] theradialcomponentofvelocityatthetwotimepointsiszero.Lacking a general analytic solution, this non-central potential type problem maybeconvenientlysolvedbynumericalmeans(e.g.,4thorderRungeKutta). As a further simplification, let us assume a fixed sailcraft attitudewithrespecttothesunlineandafixeddiffractionorreflection anglethroughouttheorbit.Thesailmaybejettisonedorstowedonce the desired orbit is reached. The numerical challenge then is to determinethediffractionorreflectionanglethatsatisfiestheboundary conditions(towithinasmallerror)intheshortesttime,T.Inprinciple, an exact matching of the boundary conditions may not exist for this type of orbit. Nevertheless, they may be satisfied to within a given degreeoferror.Inthecaseofafixedsolarattitude,weachievedquasisynchronous transfers with errors for terminal radius, energy, and azimuthalvelocitybelow0.01%oftheexpectedvaluesforMars.The gravitationalsphereofinfluenceofMarsis170radii,or0.25%ofthe orbitalradius,hence,anerror<0.01%isakintoabull’seye.Theerror for the radial component of velocity, however, was typically several tenthsofapercent.Thissuggeststhatafixedattitudesailcraftcannot make a perfectly synchronous transfer unless either the initial 3 condition is changed by the boost-phase rocket from Earth, or if the components of radiation pressure are actively controlled during the mission[24,25]. timebranchremainsrelativelyconstant,asfoundinCase(i),whereas * thelongertimebranchincreaseslinearlywith σ .Thecut-offat Fig.4.TypicalsynchronousEarth-Martransferorbitforafixedsolar attitude(lightnessnumberσ*=0.10).Sailisjettisoneduponreaching thedesiredorbit. 1.IdealMirror(Casei) * At a lightness value of σ = 0.1 we find a numerical solution for a matched Earth to Mars orbit occurs when θ i = 50° , resulting in a transfertimeofT=1.58years.Asthelightnessincreasesvalueof θ i mustalsochangetosatisfytheorbitalboundaryconditions,asshowin Fig.X(a).Reducingthearealdensityofthesailcraft,however,doesnot significantlychangetotimetoreachthematchedtheorbit.Asshown in Fig. X(b), the azimuthal acceleration term in Eq. (13), ηφσ cr / 2σ , * remains relatively constant as σ is varied, while the radial value, ηrσ cr / 2σ , decreases as the sailcraft become more buoyant against the sun. We note that a matched orbit cannot be achieved unless σ * > ~ 0.08 and θi ≥ 45° .Asverifiedbythecasesbelow,thecut-off * conditionfor σ occurswhen ηr = ηφ orequivalently.Thereforewe seethattheliftforcemustbegreaterthanorequaltotheradialforce fromradiationpressure.Therelativeinvarianceofthetimeofarrival offersnocompellingneedtoreducethearealdensityofthesailcraft below 10σ cr , assuming the mission flight path allows an incidence angleofroughly50°.Ofcourse,othermissionfactorsnotconsidered heremaybenefitfromlowerarealdensities. Fig.5.Quasi-synchronoustransferorbitforamirror-basedsailcraft. Optimized attitude, θi, as a function of lightness number σ* . Also plottedarethecorrespondingtransittime(a),andaccelerationfactors (b). * the longer time branch increases linearly with σ . The cut-off at σ * = 0.08 and θi = 45° is seen again, with the lower short-time branch(upperlong-timebranch)correspondingto ηr < ηφ (ηr > ηφ ) , asevidentinFig.X(b).Asforthecaseofareflectivesail,asmallerradial componentofradiationpressureforce,comparedtotheazimuthallift componentisfavorableforashortarrivaltime.Whatismore,froma materials fabrication point of view it may be desirable to require a smallLittrowangle,whichinprincipleiseasiertoachieve.Thislatter point also benefits from a larger lightness value, since θ i becomes * vanishingly small as σ approaches unity. As shown in Fig. X the azimuthal values ηφσ cr / 2σ are found to be nearly invariant with * lightness. For example, a value of θ i = 9.4° is matched to σ = 0.2 , providingatransittimeofT=1.42years. 2.LittrowReflectionGrating(Caseii) Lacking azimuthal acceleration, there are no solutions for a Littrow reflectiongrating. Atransmissivesailcraftpropelledbyasun-facing( θ i = 0 )diffractive 3.LittrowTransmissionGrating(Caseiii) (θ tʹ = 180° − θ t = 39°) and σ * = 0.1 . A reflective branch was also 4.Sun-FacingGrating(Caseiv) sail provides a transit time of T=1.44 years if θ t = 141° condition is reached in T=1.44 years when θ i = 21.5° . A second found,butthetransittimewasgreaterthantwoyearsacrossalarge rangeoflightnessvalues,soitisnotanalyzedhere.Thetransfertime forthetransmissivebranchisrelativelyinvarianttolightness,asseen solution is also predicted: T=1.52 years when θ i = 50.7° . If the incasesabove.Therequiredvalueof θ tʹ decreaseswithincreasing lightnessvalueincreases,asshowninFig.x(a),wefindthattheshorter timebranchremainsrelativelyconstant,asfoundinCase(i),whereas * lightness.Forexample,at σ = 0.2 avalueof θ tʹ = 18.6° isrequired, * ForatransmissiveLittrowgratingoflightness σ = 0.1 aMarsorbital providingajourneyofT=1.42years.Thecut-offanglecorresponding 4 to ηr = ηφ is θ t = 90° .FromFig.7(b) weagainseethat ηr < ηφ is required,and ηφσ cr / 2σ isrelativelyinvarianttolightness. Fig. 6. Quasi-synchronous transfer orbit for a Littrow transmission grating sailcraft. Optimized attitude, θi, as a function of lightness numberσ*.Alsoplottedarethecorrespondingtransittime(a),and acceleration factors (b). Red-colored branches corresponding to the shortesttransittimes. 5.Comparisons Theidealmirrorrequires10%moretimetoreachaMartianorbitwith a fixed attitude sailcraft. This difference may be expected to significantlyvaryiftheattitudeisactivelycontrolledorifadiffractive metasailincludesanelectropticcontrolmechanism.Forexample,an optimallycontrolledreflectivesailmissiontoMarsprovidedpredicted a transfer in as short as 324 days [24,25]. At a lightness number σ * = 0.1 wefoundtheLittrowtransmissioncaserequiresanattitude inclinationof θ i = 21.5° ,andthus,atotalbeamdeviationoftwicethis value. In comparison the sun-facing grating must deviate the transmitted beam by θ tʹ = 39° . These angular deviation values are comparable,andthusneitherconfigurationismostfavorablefromthe point of view of engineering the diffractive film. On the other hand, smaller deviations angle may be easier to fabricate – especially if uniformbroadbandperformanceisdesired.Inthatcase,alowerareal densityispreferred,asitlessensthedeviationangle.Finallyisshould benotedthattheEarth-Marstransfertimesinthisreportarelonger than the chemically fuel Hohmann transfer, which is roughly T=0.7 years.Itremainsanopenquestionweatheranactivelycontrolledsolar sailcanachieveshortertimes. Fig. 7. Quasi-synchronous transfer orbit for a sun-facing diffractive sailcraft.Optimizedattitude,θi,asafunctionoflightnessnumberσ*. Also plotted are the corresponding transit time (a), and acceleration factors (b). Blue-colored branches corresponding to the shortest transittimes. B.ControlledBinaryMetamaterialArrayedGrating Asanexampleofanelectro-opticallycontrolleddiffractivesailcraft,let us chose a sun-facing sail with diffractive panels that are switchable betweentwoequalbutoppositediffractionorders: ±θ1 .Notethatthe radialforceonthesailcraftdoesnotchangewhenthediffractionorder is switched, whereas the lift component changes sign. For a large numberofsuchpanelsarrayedacrossthesail,theaverageazimuthal force may be varied nearly continuously between two equal but oppositevalues.CombiningEq.s(10)and(13)wewrite ⎛ ⎞ ! σ* σ* F = −maM (RE / r) 2 ⎜⎜ (1− (1+ cosθ1 )) r̂ + g(t) sin θ1φ̂ ⎟⎟ 2 2 ⎝ ⎠ where g(t) is the control variable with g(t) ≤ 1 and defined assuming θ1 > 0 .Forexample, g(t) hasanegativevaluewhenmost of the panels diffract into the −θ1 order. For a reflective grating, 0° ≤ θ1 ≤ 90° ,whereas 90° ≤ θ1 ≤ 180° foratransmissivegrating(see Fig.1).Ifwerequire ηφ > ηr duringsomepartofthetransfer,aswas necessary for short transfer times in Section A, then a transmissive gratingisrequired.Thedeterminationofacontrolsignalthatachieves asynchronoustransferistheshortesttimeisbeyondthescopeofthis report,andwillbeexploredinfuturestudies. 5 4.CONCLUSIONS Diffractive and reflective sails having a fixed attitude with respect to thesunlinewereexamined.NearlysynchronousEarth-Marstransfers wereachieved,withshorttransfertimeswhenthetransverseliftforce exceededtheradialscatteringforcefromsolarradiationpressure.A transmissive sun-facing diffractive sail composed of an array of switchable diffractive elements is of particular interest, affording a variableliftcomponentofforcewhenthesignofthediffractiveorderis switched.Futureworkisneededtofindanoptimizedcontrolscheme forsuchasailcraft. FundingInformation.NationalScienceFoundation(NSF):ECCS1309517. References 1. J.C.Maxwell,Atreatiseonelectricityandmagnetism,Vol.2(Macmillan andCo.,London,1873). 2. P.Mulser,“Radiationpressureonmacroscopicbodies,”J.Opt.Soc.Am.B 2,1814-1829(1985). 3. “SP-424TheVoyageofMariner10,”Ch.7,8.https://history.nasa.gov/SP424/ch7.htm(cited16January2017). 4. K.Tsander,“Fromascientificheritage”(1924).NASA.Technical Translationno.TTF-5411967.(http://epizodsspace.no-ip.org/bibl/inostryazyki/nasa/Tsander_From_a_Scientific_Heritage_1969.pdf(cited14 January2017). 5. K.E.Tsiolkovsky(Tsiolkovskiy),“Extensionofmanintoouterspace,” (1921);SymposiumJetPropulsion2,UnitedScientificandTechnical Presses(1936). 6. C.R.McInnes,Solarsailing:technology,dynamicsandmissionapplications, (SpringerScience&BusinessMedia,2013). 7. G.Vulpetti,L.Johnson,andG.L.Matloff,Solarsails:anovelapproachto interplanetarytravel,(Springer,2014). 8. F.T.ChenandH.G.Craighead,“Diffractivephaseelementsbasedontwodimensionalartificialdielectrics,”Opt.Lett.20,121–123(1995). 9. P.Lalanne,S.Astilean,P.Chavel,E.Cambril,andH.Launois,“Designand fabricationofblazedbinarydiffractiveelementswithsamplingperiods smallerthanthestructuralcutoff,”J.Opt.Soc.Am.A16,1143–1156 (1999). 10. J.Tervo,andJ.Turunen,“Paraxial-domaindiffractiveelementswith100% efficiencybasedonpolarizationgratings,”Opt.Lett.25,785-786(2000). 11. E.Hasman,Z.E.Bomzon,A.Nivetal.,“Polarizationbeam-splittersand opticalswitchesbasedonspace-variantcomputer-generated subwavelengthquasi-periodicstructures,”Opt.Comm.209,45-54(2002). 12. J.Tervo,V.Kettunen,M.Honkanenetal.,“Designofspace-variant diffractivepolarizationelements,”J.Opt.Soc.Am.A20,282-289(2003). 13. H.Sarkissian,S.V.Serak,N.V.Tabiryanetal.,“Polarization-controlled switchingbetweendiffractionordersintransverse-periodicallyaligned nematicliquidcrystals,”Opt.Lett.31,2248-2250(2006). 14. V.Presnyakov,K.Asatryan,T.Galstianetal.,“Opticalpolarizationgrating inducedliquidcrystalmicro-structureusingazo-dyecommandlayer,” OpticsExpress14,10558-10564(2006). 15. L.Nikolova,andP.S.Ramanujam,Polarizationholography(Cambridge UniversityPress,2009). 16. R.K.Komanduri,andM.J.Escuti,“Highefficiencyreflectiveliquidcrystal polarizationgratings,”Appl.Phys.Lett95,091106(2009). 17. E.Nicolescu,andM.J.Escuti,“Polarization-independenttunableoptical filtersusingbilayerpolarizationgratings,”AppliedOptics49,3900-3904 (2010). 18. J.Kim,C.Oh,S.Seratietal.,“Wide-angle,nonmechanicalbeamsteering withhighthroughpututilizingpolarizationgratings,”AppliedOptics50, 2636-2639(2011). 19. O.D.Lavrentovich,“Liquidcrystals,photoniccrystals,metamaterials,and transformationoptics,”ProceedingsoftheNationalAcademyofSciences 108,5143-5144(2011). 20. D.Lin,P.Fan,E.Hasman,andM.L.Brongersma,“Dielectricgradient metasurfaceopticalelements,”Science345,298–302(2014). 21. N.YuandF.Capasso,“Flatopticswithdesignermetasurfaces,”Nat. Mater.13,139–50(2014). 22. J.Kim,Y.Li,M.N.Miskiewicz,C.Oh,M.W.Kudenov,andM.J.Escuti, “Fabricationofidealgeometric-phasehologramswitharbitrary wavefronts,”Optica2,958-964(2015). 23. G.Zheng,H.Mühlenbernd,M.Kenney,G.Li,T.Zentgraf,andS.Zhang, “Metasurfacehologramsreaching80%efficiency,”Nature nanotechnology,10(4),308-312(2015). 24. A.N.ZhukovandV.N.Lebedev,“Variationalproblemoftransfer betweenheliocentriccircularorbitsbymeansofasolarsail,”Cosmic Research2,41-44(1964). 25. V.L.CoverstoneandJ.E.Prussing,“Techniqueforescapefrom geosynchronoustransferorbitusingasolarsail,”J.ofGuidance,Control, andDynamics26,628-634(2003). 6