ActaAstronautica](]]]])]]]–]]] ContentslistsavailableatSciVerseScienceDirect Acta Astronautica journal homepage: www.elsevier.com/locate/actaastro Experimental analysis of laser ablated plumes for asteroid deflection and exploitation Alison Gibbingsa,b,n, Massimiliano Vasilea, Ian Watsonb, John-Mark Hopkinsc, David Burnsc aAdvancedSpaceConceptsLaboratory,DepartmentofMechanical&AerospaceEngineering,UniversityofStrathclyde,75MontroseStreet,Glasgow,UK bSystems,PowerandEnergyResearchDivision,SchoolofEngineering,UniversityofGlasgow,JamesWatt(South)BuildingUniversityAvenue,Glasgow,UK cInstituteofPhotonics,UniversityofStrathclyde,WolfsonCentre,106Rottenrow,Glasgow,UK a r t i c l e i n f o a b s t r a c t Articlehistory: Ithasbeentheoreticallydemonstratedthatlaserablationiseffectiveinthepotential Received19February2012 deflection and mitigation of asteroids. However, there have been few experimental Receivedinrevisedform studiestosupportthisclaim.Thetheoreticalmodelsarecurrentlybasedonassump- 12May2012 tionsregardingthelaserbeamdiameter,thepowerrequirement,theformationofthe Accepted7July2012 ejectaplume,andthepotentialforejectatocontaminateandotherwisedegradeany exposed surface. Recent proposals suggesting the use of a solar pumped laser, in Keywords: particular, can be deeply affected by the re-condensation of the ejecta. To either Laserablation validate,amendand/oreliminatetheseassumptionsaseriesoflaserablationexperi- Asteroidexploitation mentshavebeenperformed.Usinga90W,continuous-wavelaseroperatingat808nm, Asteroiddeflection a rocky magnesium iron silica based material – olivine – has been ablated. These experiments were used to examine the validity of the theoretical model and the experiencedlevelsofcontamination.Itwillbeshownthatthecurrentmodelcorrectly predicts the ablated mass flow rate for rocky based asteroids, but overestimates the contaminationrateandthedegradationoftheoptics. &2012ElsevierLtd.Allrightsreserved. 1. Introduction Earthonceevery10,000years.Thiscancauselocaldamage, earthquakesandtsunamis.AsteroidsthatimpacttheEarth NearEarthAsteroids(NEAs)representbothanopportu- withadiameterlargerthan1kmareconsideredtobeglobal nity and a risk. Their pristine environment captures the killers. Such an impact event is considered to catastrophi- early formation of the solar system; while their impact callyannihilate90%ofalllife,resultinginanuclearwinter, potential could result in the mass extinction of life. The withlittlechanceofrecoverywithinthenearterm[1].This Earthhasbeen,andwillcontinuetobe,thesubjectofmany isthoughttohavehappened,oncebefore,approximately65 othergroundandairimpactingevents.Amidtheobserved million years ago, with the impact of a 10km diameter population, there are at least between 2000 and 200,000 asteroidat12km/s[23]. objects that could impact the Earth [1]. On average, an Thereforepotentialmethodsofasteroidmitigationand asteroid with a diameter greater than 100m impacts the deflection have been addressed by numerous authors [2–4]. Amongst the many possibilities to deflect NEAs, ablation has been shown to be theoretically one of the nCorresponding author at: University of Strathclyde Department of most effective methods [5]. Work conducted in 2009 by Mechanical&AerospaceEngineeringAdvancedSpaceConceptsLabora- Sanchez et al. [5] compared the effectiveness of six tory75MontroseStreet,GlasgowG11XJ,UnitedKingdom. Tel.:þ4407795060527. different asteroid deflection techniques. Through a E-mailaddresses:[email protected], multi-criteria,quantitativecomparisonthenuclearinter- [email protected](A.Gibbings), ceptor, kinetic impactor, mass driver, low thrust tug, [email protected](M.Vasile), ablation and the gravity tractor were assessed. Assess- [email protected](I.Watson),[email protected] (J.-M.Hopkins),[email protected](D.Burns). mentwasmaderelativetotheachievablemissdistanceat 0094-5765/$-seefrontmatter&2012ElsevierLtd.Allrightsreserved. http://dx.doi.org/10.1016/j.actaastro.2012.07.008 Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 2 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] Nomenclature R Distance from the Sun measured in Astro- nomicalUnits (Dm/A) Ejecta mass per unit area deposited on r Radiusfromthespotlocation SLIDES themicroscopeslidesduringablation SEM ScanningElectronMicroscope A Areaofthemicroscopeslide t Sublimationduration a Absorbance T Transmittance b ASPOT Areaofthelaser’ssurfacespot Tamb Ambienttemperature cA Heatcapacityoftheasteroid T0 Temperatureatthecentreoftheasteroid Cm Momentumcoupling TSUB Sublimation temperature of the asteroid in C Concentrationratio vacuumconditions R d Diameterofthesurfacespot v Averagevelocityoftheejectaplume SPOT dh/dt Surfacelayergrowth e Blackbodyemissivityoftheasteroid dm/dt Massflowrateduringsublimation Z Absorptivity E Incomingenergyduringtheablationprocess ZAB Efficiencyoftheablationprocess E Sublimationenthalpyoftheasteroid y Elevationangle,fromthesurfacenormal v FSUB Forceactingontheasteroid yMAX Limitedexpansionangle h Height of the deposited ejecta from the l Scatterfactor EXP experiment r Densityoftheejectaplume k Adiabaticindex,fordiatomicmolecules rn Densityatthenozzle kb Boltzman’sconstant rA Densityoftheasteroid Kp Jetconstant rl Layerdensity kt Thermalconductivityoftheasteroid sSB Stefan–Boltzmannconstant M Molarmassofthetargetmaterial t Degradationfactor a NEAs NearEarthAsteroids cvf Viewangle P1AU Solarpowerat1AstronomicalUnit DhEXP Measuredthicknessofthedepositedmaterial P Absorptionofthelaserbeam Dm Masslossduringablation IN Qn Energy required to ablated each kilogram of Dt Ablationduration material Zsys Overallconversionefficiencyfromsolarinput Q Heatlossesduetoconduction tolaseroutput COND QRAD Heatlossesduetoradiation Zabs Absorptivityoftheilluminatedasteroid Earth,thewarningtime,thetotalmassintoorbitandthe infrastructure and resources. Therefore an alternative currenttechnologyreadinesslevel.Withbotharelatively option was to mount a mega-watt laser onto a large short warning time and a low mass into space, ablation single spacecraft. The laser would be powered by a can provide significantly higher and more controllable nuclear reactor [20]. However, manoeuvring and operat- ratesofdeflection.Thetechniqueisalsoadvantageousas ing such a large structure, at close proximity to the itavoidsthecatastrophicfragmentationoftheasteroid.It asteroid, under an irregular gravity field was considered alsoeliminatestheneedofhavingtophysicallylandand/ tobeverydifficult.Thisisfurthercoupledwithdevelop- orattachasystemontothesurfaceoftheasteroid[5]. inganuclearreactorforspace-basedapplications,andthe Ablationisachievedbyirradiatingtheasteroidwitha associatedpoliticalramifications.Thereforeanalternative light source. This can either be collected and focussed conceptwasproposed.Insteadofalargesinglestructure, solar radiation or with a laser light source. Within the a swarm of small spacecraft, each equipped with an illuminatedfocalpoint,theabsorbedenergyincreasesthe identical kilo-watt solar-pumped laser could be used temperatureoftheasteroid,enablingittosublimate.This [10]. This provides a much lighter and more adaptable transformstheexposedmaterialdirectlyfromasolidtoa concept.Bysuperimposingeachlaserbeam,thecumula- gas.Theablatedmaterialthenexpandstoformanejecta tive surface power density would be used to initiate the plume. Over an extended period of time, the resultant ablationprocess[10].Singularormultipleablationspots thrust, induced by the ejecta plume and acting on the canalsobeused.Thisincreasestheflexibilityandoverall asteroid can be used to push the asteroid away from its redundancy of the deflection mission. As required, more original threatening trajectory [2]. This increases the spacecraft can be added or removed from the existing minimum orbit interception distance between the Earth configuration,eliminatingtheneedtodevelopanddesign andtheasteroid,otherwisepreventingtheEarthimpact- new spacecraft(s) [6,7,9]. The potential for deflection is ingevent[5–8]. thereforedependentonthenumberofspacecraftlocated Previous proposals for the initiation of laser ablation within the vicinity of the asteroid, their combined laser considered using either a ground-based or space-based powerandthematerialpropertiesoftheasteroid. facility [1,19]. For a ground-based facility an average Within this configuration, each laser would be powered power level of several giga-watts would be required to bytheSun,eitherdirectlyorindirectly.Fordirectpumping deflect a small, 40–80m in diameter asteroid [19]. This thesolarradiationiscollectedandconcentrateddirectlyonto was considered to be a substantial investment in thelasermedium.Forindirectpumping,theincomingsolar Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] 3 radiationisfirstfocussedontoasetofhighlyefficientsolar If the mass flow is negative then there is not enough cells. This immediate step is used to convert the incoming energytoinitiatetheablationprocess. solar radiation into electrical energy that is then used to Theheatlossthroughradiationwasassumedtoactas powerthelaser.Forbothcases–directandindirectpumping ablack-bodyandisthereforedefinedas –alight,deployableprimarymirrorandasmallersecondary Q ¼s eA ðT4 (cid:2)T4 Þ ð2Þ (bothknownasthesolarconcentrators)areneededtocollect RAD SB SPOT SUB amb the freely available solar radiation of the Sun. A steering where s is the Stefan–Boltzmann constant (5.6704(cid:3) SB mirror is also required to target the laser beam onto the 10(cid:2)8W/m2K4), e is the black body emissivity of the surfaceoftheasteroid.Largeradiatorsareneededtokeepthe illuminatedmaterial,A is theareaof thespot illumi- SPOT laserwithinitsoperationallimitandtocoolthespacecraft. natedbythelaser,T isthesublimationtemperatureof SUB However, any exposed surface within the vicinity of theilluminatedmaterial(invacuumconditions)andT amb theejectavolume,includingthesteeringmirrorandsolar istheambienttemperaturethatinspaceis4Kandinthe concentrators, would be subjected to the continual con- laboratoryenvironmentis298K. taminatingeffectsofthecondensingejecta.Itiscurrently Theheatlossduetoconductionisdeterminedfrom assumed that once the ejecta plume comes into contact rfficffiffiffiffirffiffiffiffiffiffikffiffiffiffi withanygivensurfacethattheparticleswillimmediately Q ¼ðT (cid:2)T ÞA A A t ð3Þ COND SUB 0 SPOT pt re-condenseandstick.Thecontinuedaccumulationofthe ejecta will decrease the transmittance and increase the wherec ,r andk aretheheatcapacity,densityandthermal A A t absorbance of the exposed surface. The degradation conductivityof the asteroid respectively,t is the equivalent caused by the ejecta is considered to follow the Beer– timetobringalayer(o1mmthick)ofthetargetmaterialto Lambert–Bouguerlawandisdependentontheabsorptiv- sublimation temperature (this depends on the diffusion in ity of the condensed material. The laser beam is also thetargetmaterialandthespeedofrecessionoftheablated- expectedtobeattenuatedbytheablatedplumeofejecta. ejectalayer),andT isthetemperatureatthecentreofthe 0 This paper will show that the effect of the ejecta con- asteroid.Thisisassumedtobe298K,whichcoincideswith tamination, according to the current model, has a major thelocaltemperatureofthelaboratory.Forthesimulations impactontheabilityofthelaserablationprocesstoproduce withinthispaper,thedensityoftheasteroidanaloguetarget a significant deflection action. Therefore to examine the materialwasmeasured.Allothervalueswereassumedbased actual operational and environmental constraints of laser onthephysicalpropertiesofolivine.ThisisgiveninTable1. ablation, a series of experiments have been performed. The enthalpy of sublimation, as given in Table 1, combines Within a vacuum chamber, and using a 90W continuous- the enthalpy of vaporisation and the heat fusion. It is wave laser, the ablation response of a magnesium iron considered to be the complete enthalpy of vaporisation, silicate(olivine)samplehasbeenassessed.Thisassessment ratherthantheincidentofvaporisation. includedthedevelopmentoftheejecta–coneanglediver- Note that, the model assumes that the internal tem- gence,massflowrate,andplumedensity–andtheaffectsof perature of the asteroid is unchanged and therefore thecondensingejecta.Allresultshavebeencomparedtothe remainsat298Kthroughouttheablationprocess.Further- numericalmodel.Thispaperthereforeinvestigatestheeffect more,theasteroidisnotrotating.Theinternaltemperature of laser irradiation on a rocky based asteroid simulate. correspondstotheconstantandcontrolledroomtempera- Providing experimental data and model validation is an tureofthelaboratory.Thelatterassumptionisintroduced important step towards the realisation of a laser asteroid for consistency with the experiments that will be con- deflectionsystem.Thispaperdetailsthecurrentablationand ductedonanon-rotatingtarget. asteroid deflection models; highlighting the conditions The average velocity of the ejecta plume v can be placed upon the assumed physical parameters and the calculatedassumingMaxwell’sdistributionofanidealgas response of the ablation process. The laser ablation experi- [5,6].ThisisdefinedbythesublimationtemperatureT , SUB mentisthenpresented.Theresultsfromtheexperimentare themolarmassoftheablatedgaseousphaseofthetarget then given and placed within the context of a revised material(thisistakentobe50g/moleandaccountsfora deflection simulation. Conclusions have been drawn and diatomicmoleculeofforsteritewithinthevapourregime) areasoffutureworkaddressed. M ,andBoltzman’sconstantk .Thisisgivenby a b sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 8k T v¼ b SUB ð4Þ 2. Currentablationandasteroiddeflectionmodels pMa Thecurrentablationmodelisbasedontheenergybalance Table1 ofsublimation[6,11].Thiscombinestheabsorbedlaserbeam Assumedparametersfortheasteroidanaloguetargetmaterial. powerP ,thesublimationenthalpyoftheilluminedmaterial IN bythelaserE ,andtheheatlossesthroughconductionQ Parameter Value v COND andradiationQ respectively.Thereforethemassflowrate m_ of the abRAlaDted material during sublimation can be Density(kg/m3) 3500 exepxrpessedas BSulabclkimbaotdioynemenitshsaivlpityy,,Eev[(6J]/kg)[26] 10.49.55n10 dm 1 Temperatureatthecentreoftheasteroid,T0(K) 298 dt ¼m_exp¼ EvðPIN(cid:2)QRAD(cid:2)QCONDÞ ð1Þ THheeartmcaaplaccointyd,uccAti(vJi/tkyg,Kk)A(W/mK)[25] 143.7661 Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 4 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] The force acting on the asteroid F is given by the SUB product of the ejecta velocity and the mass flow rate of theablatedmaterial.Thisisgivenas F ¼lvm_ ð5Þ SUB exp A constant scatter factor l is used to account for the uniformexpansionoftheejectaoveragivenhalfsphere. This is considered to be a worst case, conservative assumption.Thevelocityoftheejectaisalsoassumedto beconstant,independentofthelocalelevationangle,and therefore is only a function of the sublimation tempera- Fig.1. Geometryoftheplume—localreferenceframe. ture. This is given in Eq. (4). The force acting on the asteroidisconsequentlyonlydependentonthemassflow theBeer–Lambert–Bouguerlaw,andastickycoefficientof rate.Ahighermassflowrateresultsinahigherthrust. oneisusedthroughout[11,6]. The input power P is computed assuming that the Bytakingthefirstassumptionthedensityoftheejecta IN beam is generated by pumping a fibre laser. The electric plumeatanygivendistancerfromthespotlocation,and powerisgeneratedbyasolarconcentratorfocussingthe elevationangle y fromthe surface normalcanbedefined sunlightontoasolararray.Itisassumedthatonly80%of [11].ThisisillustratedinFig.1andcanbeexpressedas the incoming laser light is absorbed and that the rest is d2 (cid:5) (cid:3) py (cid:4)(cid:6)2=k(cid:2)1 reflected.Theabsorbedpoweristherefore rðr,yÞ¼rnK SPOT cos ð7Þ Pð2rþdSPOTÞ2 2yMAX P A PIN¼tZabsZsysCR 1AUR2SPOT ð6Þ A compressible, friction free gas is typically classified with a constant adiabatic index k, where for diatomic where t is the degradation factor due to contamination molecules k¼1.44 [11]. In Eq. (7) the jet constant K for p (see below), Zabs is the absorptivity of the illuminated diatomic molecules is defined as 0.345 with the limited asteroid,Zsysistheoverallconversionefficiencyfromthe expansionangleyMAXas130.451[24].Byconsideringthe solar input to the laser output, CR is the concentration standardrocketequation,thedensityatthenozzle,rnis ratio or ratio between the area of the solar concentrator alsogivenas and the area of the spot, P ¼1378W/m2 is the solar 1AU m_ powerat1AstronomicalUnit,A istheareaofspoton rn¼ exp ð8Þ SPOT A v thesurfaceoftheasteroid,andRisthedistancefromthe SPOT SunmeasuredinAstronomicalUnits.Theoverallconver- Kahle et al.,2006 [11]alsoconsideredtherate ofejecta sion efficiency is assumed to be 19%. This accounts for contaminating and degrading any exposed surface. The solar cells with an efficiency of 40%, a laser system with variation of ejecta thickness – known as the surface layer an efficiency of 60% and a nominal reflectivity of the growth–onanyexposedsurfacecanbeexpressedas[6,11] concentratorsof90%. dh 2vr The sublimation process is also considered to be dt ¼ r coscvf ð9Þ governed by three fundamental assumptions. The first l considersthattheformationoftheejectaplumeissimilar wherec istheviewanglei.e.theanglebetweenthenormal vf to,althoughnotidenticalto,thedevelopmentofarocket tothesurfaceandthesurface-to-spotvector.Toaccountfor exhaust in standard methods of rocket propulsion. The the expansion of gas into a vacuum the average velocity is same approach is used to model cometary sublimation multipliedbyafactoroftwo.Thedenominatorr isthelayer l [12]. This assumes that the ablated ejecta is a compres- density. Based on current literature, for a magnesium iron sible,frictionfreegas(withnosolidparticles),whichhas silicamaterial,thisisassumedtobe1g/cm3[11].Allofthe a constant velocity and temperature, and occurs below ablated particles of ejecta coming from the asteroid are the limit of ionisation of plasma formation [11,6]. It assumedtosticktoeveryexposedsurfaceonthespacecraft. therefore assumes a vapour-only flow regime. At Theincreasingthicknessofthecontaminantswillultimately increased intensities a strong laser generated plasma reducethereflectivityoftheconcentratorsandthereforethe can form. Plasma, through the inverse Bremsstrahlung laser output power. The consequence is a reduction of the effect, is known to absorb the incoming laser beam and thrust imparted onto the asteroid until the sublimation therefore shield the target from its affects. This would ceases completely and the thrust associated with it. The dramaticallyreducethecouplingbetweenthelaserbeam reductionofthereflectivityoftheconcentrator,knownasthe and the asteroid that otherwise provides the low thrust degradation factor, t can be computed from the Beer– requiredforthedeflectiontechniquetobecomeeffective. Lambert–Bouguerlaw[11].Thisisgivenby Thesecondassumptionisthattheasteroidisaspherical, t¼e(cid:2)2Zh ð10Þ dense, non-porous homogenous body. The asteroid is considered to have an infinite heat sink, where during where Z is the absorptivity (absorbance per unit length) of the ablation process the asteroid maintains a constant the accumulated ejecta. For olivine, at 808nm (the wave- internaltemperature.Thethirdassumptionisthatallthe lengthofthelaser)thisisassumedtobeapproximately104/ ejected particles will immediately re-condense and stick cm[11,10].Afactorof2withintheexpressionaccountsfor toanyexposuresurface.Degradationisassumedtofollow thedoublepassingofthesurfacelayer[11]. Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] 5 Gauss planetary equations are used to compute how the sublimation process will affect the orbit of the asteroid. Itis assumedthat the forcein Eq.(5) isalways aligned to the velocity vector of the asteroid [18]. The Gaussplanetaryequationsarepropagatedforwardintime fromthemomentthesublimationstarts totheexpected timeoftheimpactoftheasteroidwiththeEarth[18].At the expected time of impact with the Earth, this can be usedtocomputethepositionandvelocityofthedeflected asteroid. From the position and velocity, the impact parameter at the expected time of impact can be computed. For example, for a relatively small NEA, with an assumed diameter of 250m and a mass of 2.7(cid:3)1010kg, the miss distance is illustrated in Fig. 2. For this case, multiple spacecraft are flying in formation with the Fig. 3. Miss distance of a 250m diameter, 2.7(cid:3)1010kg (based on asteroid. Each spacecraft carries a laser (with an output Apophis)asteroidasafunctionofthewarningtimeandthenumberof power of 22kW) and a 10m diameter primary mirror satellites:withcontamination. (solar concentrator). Together they concurrently beam their lasers onto the same spot on the surface of the asteroid. It is assumed that the temperature of the spot details[18]).Thisisthereasonfortheperiodictrendwith remainsconstantandisequaltotheassumedsublimation thewarningtimethatcanbeseeninFig.3. temperature of Forsterite, which is 1800K. Fig. 2 shows Asillustrated inFig. 3, accountingforejectadegrada- the miss distance as a function of the number of space- tion and contamination, the maximum achievable miss craftandthewarningtime.Themissdistance,asgivenin distance reduces to 4500m. This results in a significant Fig 2, is shown in kilometers. The warning time is the reductionofperformanceof85%.Contaminationreduces time from the beginning of the deflection action to the theachievablesurfacepowerdensityofthelaser,andthe expectedimpactoftheasteroidwiththeEarth.Theresult associatedthrustimpartedontotheasteroid.Contamina- in Fig. 2 does not include the effects of contamination. tionofanyopticalsurfacewillhaveamajorimpactonthe With deflection operations of between 1 and 9 years overallsuccessofanyasteroiddeflectionmission. (fromtheexpectedimpactwiththeEarth)Fig.2demon- The model developed in this section also allows for the stratesanachievablemissdistanceofbetween5000and estimationofthecouplingbetweentheoutputpowerfrom 30,000kmrespectively. thelaserandthechangeoflinearmomentumoftheasteroid. Fig. 3 shows the result for the same mission scenario This is known as the momentum coupling coefficient, C m but this time including the contamination of the solar [13–16].Forcontinuouslaseroperationsitisdefinedasthe concentrator.Thisisinaccordancewiththemodeldevel- induced force F relative to the incoming absorbed laser SUB oped by Kahle et al., 2006 [11]. The contamination powerthatoccursatthespotlocation.Thiscanbewrittenas: induces a fast degradation of the collected power and a F rapidhaltofthesublimationprocess.Dependingonwhen Cm¼PSUB ð11Þ the sublimation starts and ends the miss distance can INEXP increase or decrease (see Colombo et al., 2008 for more During the experiment the force was not directly measured, but was calculated from Eq. (5) combining themeasuredablated massflowratewiththemeasured ejecta gas velocity. The absorbed power P was taken INEXP from the known power illumination of the spot and combined with the assumed absorption of the target materialandthelaserbeamwithintheplume.Theenergy required to ablate each kilogram of asteroid material Qn couldalsobedetermined.ThisismeasuredinJ/kgandis expressedas E Qn¼ ð12Þ Dm The energy E is the incoming laser energy during the ablationprocess.Itisthereforeaproductoftheincoming absorbed laser power at the spot location multiplied by theablationtime.Dmisthemassoftheablatedmaterial in a given time period. From this and the momentum coupling, theefficiency of the ablation process Z could Fig. 2. Miss distance of a 250m diameter, 2.7(cid:3)1010kg (based on AB be determined. This defines the efficiency at which the Apophis)asteroidasafunctionofthewarningtimeandthenumberof satellites:withoutcontamination. laserenergyisconvertedintokineticenergy,carriedaway Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 6 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] bytheablatingejecta.Thisisgivenas[14,21] Within the chamber, the asteroid analogue target material was mounted on a raised pedestal, at a pre- 1 ZAB¼2C2mQn: ð13Þ determinedlocation.Thiswasrelativetotheknownfocal point of the laser, which had an approximate spot size Note that ZAB does not include the effect of any diameter of (cid:4)1.8mm. This corresponded with a surface absorption losses within the ejecta plume. It is based power density(accountingfor optical losses throughthe purely on the incoming laser energy and power at the system) of (cid:4)2.44-1.69kW/cm2. For each experiment spotlocation. reported herein, ablation occurred for 10min. To collect and assess the ablated ejecta, highly cleanedmicroscope 3. Theexperiment slideswerepositionedwithintheablationvolume.Before andaftereachablationevent,massmeasurementsofthe A 90W continuous-wave laser, operating at 808nm, microscope slides enabled the mass of the deposited was used to initiate all ablation events. This was below ejecta at different points within the plume to be deter- the threshold of plasma formation. The formation of mined.Bymeasuringthemassperunitareadepositedon plasma was not accounted for within the model, and eachmicroscopeslide(Dm/A) ,whereAistheareaof SLIDES would otherwise have a degrading effect on the quality the microscope slide, and by measuring the thickness of andefficiencyoftheablationprocess[14,15,6].Forinitial thedepositedmaterialDh ,thedensityofthedeposited EXP calibration, all ablation events occurred within a sealed materialcanbecomputed.Thisisachievedby and self contained glass test chamber. The chamber was (cid:7) (cid:8) Dmðr,yÞ purged with nitrogen to reduce the occurrence of atmo- r ðr,yÞ¼ A SLIDES ð14Þ spheric combustion to negligible levels. The ejecta mass EXP Dh EXP flow rate, plume density, plume divergence and the From the model, the expected collection rate of the influenceofthere-condensingejectawerethenassessed ejectaoneachmicroscopeslidecanalsobederived.This [22].Theexperimentwasthenrepeatedwithinavacuum isgivenas chamber. A pump down pressure of 2(cid:3)10(cid:2)5mbar was achieved. This removed the atmospheric particle drag 1dm ¼2rðr,yÞv ð15Þ disturbance;allowingforthemaximumandfreespread- A dt ing of the plume [17]. This was considered to provide a Eq.(15)assumesthatthevelocityoftheexpandedgas realisticsimulationofthelaser-to-asteroidinteraction. isabout2v(toaccountforfullexpansionintoavacuum) Olivine was used to represent a dense, rocky, s-type and that all the particles are sticking onto the surface of asteroid. It had a density of 3500kg/m3 and was shaped the microscope slides. The effect of the deposited ejecta into a cube by cutting a larger rock sample with a masswasassessedbymeasuringthelighttransmittance diamond blade.This enabled allablation events to occur acrosseachmicroscopeslide.Therelativereductioninthe ontoaflatface.Aflatfacewasusedtoavoid,orseverely transmittanceTandtheincreaseinabsorbancea caused b limit,theablationprocessbeingsubjectedtoanyirregu- bythedepositedejectaweremeasured.Thiswasachieved larityofthesurfacematerial(includingsurfacecurvature) byeitherusingapowermeterorphotographicanalysisof andanyinhomogeneitiesofthetargetmaterial’scompo- the actual microscope slides. By considering only a one sition. The aim was to provide a tightly focussed laser way transmittanceof themicroscope slides,the value of beam onto thesurface of thetargetmaterial. It was also the associated absorbance can be calculated from the consideredtobearealisticanalogueofthein-spaceevent, Beer–Lambert–Bouguer law for optical absorption. This where thespot sizeof thelaser beamwouldbesmallin isgivenas comparisontothesize,andmajorfeaturesonthesurface oftheasteroid.Massmeasurementsofthetargetmaterial a ¼log1 ð16Þ before and after each ablation event enabled the rate of b T sublimationtobedetermined. Given the absorbance a , the associated absorptivity b Thetestchamberwassurroundedbytwocameras,and Z (absorbanceperunitlength)giveninEq.(10)canbe EXP a spectrometer. Viewing access was granted through determined. This is achieved by dividing the experimen- optical windows. Each window had a transmittance of tally determined absorbance a by the experimentally b approximately 94%. Two high resolution, digital cameras measuredheightofthedepositedejectah .Theheight EXP (Panasonic HDC), mounted perpendicular to each other, of the collected ejecta was measured with a Nikon recordedeachablationevent. Duringeachablationevent Nomarskimicroscope.Therefore the cameras were used to measure the divergence and a formationoftheejectaplume.Aspectrometerwasusedto Z ¼ b ð17Þ EXP h measuretheinferredtemperatureoftheablatedgas.This EXP was determined from the Wien’s Displacement law, and In the following section the absorptivity of the con- wasachievedbymeasuringtheintensityandwavelength taminationlayerestimatedfromtheexperimentswillbe ofthe emittedspectra.Thevelocityofthe gaseousejecta usedtore-assesstheeffectivenessofthedeflectionaction. was then calculated, assuming Maxwell’s distribution of The degradation is a function of both the absorptivity of an ideal gas. All experiments were repeated three times. the condensed layer and the thickness, which in turns Thisaimedtoprovidemoreviableandwellcalibrateddata dependsonthelayergrowthrate.Bymeasuringh over EXP points. timeonecanalsoderiveacorrectionfactortoEq.(10)and Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] 7 compute a degradation factor that is consistent withthe compositionoftherimandthatofthetargetmaterialalso experimentalresults. varied. Confirmed by a Scanning Electron Microscope (SEM) the surface material was initially composted of (Mg,Fe) SiO . However during the ablation event the 2 4 4. Results subsurface inclusions of aluminium and calcium were brought to the surface (i.e. from the bottom of the Laserablationresultedintheformationofasmalland ablation hole and/or from the material matrix of the extended rocket plume. This was dominated by the sample). These inclusions represent impurities within gaseous exhaust of material, and is illustrated in Fig. 4. thetargetmaterial’sinhomogeneousstructure[30].Again Fromtheinitialilluminationofthelaserbeam,inaddition confirmedbytheSEMtheilluminatedrimwascomposed tothegaseousexhaust,theablationprocessalsoresulted of (Ca,Al)SiO . The sublimation temperature at 1 atmo- 4 in the formation of small solid particles. However, sphereofolivinewithcalciumdepositsis3800Kandfor throughout all the experiments this only occurred for a Mg SiO is 3350K [31]. This variation will result in the 2 4 very small portion of time; almost instantaneously once incongruentsublimationofthetargetmaterial[31].This the laser beam had initially illuminated the olivine will occur over a range of temperature limits and can sample. The solid particles were also distributed over include the full, partial and un-vaporised sections of the 1801. This is also illustrated in Fig. 4, and is portrayed illuminatedmaterial[25,31].Theexposureofnewmate- by the creation of white streaks. The generation of solid rial may therefore increase the sublimation temperature ejectaisnotcurrentlyaccountedforwithinthenumerical of the illuminated material. The laser beam will have to model, and would otherwise enhance the momentum heattherocktoahighertemperature,whilealsoexperi- exchangebetweenthelaserbeamandtargetmaterial. encing the effects of a slightly defocused laser beam. Underhighvacuum,basedonthemodelwithaninput Lower pockets of sublimation may also be encountered. illuminationpower of either 62 or 43W, the initial mass This can be caused by the transparency of the pure flow of the ablated ejecta was 2.10-7 kg/s. At the end of mineralscontainedwithinthetargetmaterial.Theplume theablationperiod(10min)theexperimentalmassflow of gaseous ejecta will also increase the local pressure rateoftheablatedmaterialhadreducedto2.5(cid:5)10-8kg/s. surrounding the ablation zone. This will assist in The rapid reduction in the mass flow rate is expected to be due, in part, due to the defocusing of the laser beam and the additional thermal effects that are not currently accounted for within the numerical model. Critically the modeldoesnotaccountforthethreedimensionalthermal diffusivity.Itislimitedtoamono-dimensionaltransferof heat.Italsoassumedaconstantvalueofemissivity,heat capacity, density and thermal conductivity. In reality these parameters have a temperature dependence on the optical and thermal properties of the target material [29]. In practice the ablation response resulted in the widening and deepening of the surface hole. This was causedbythetunnellingawayofthesurfaceandsubsur- face material. A small, narrow hole was formed that extended into the target material. Similar to a rocket exhaust,asafunctionoftime,thiswouldhaveassistedin constraining the formation of the gaseous ejecta plume. The surface hole diameter was also widened slightly. An initial surface spot radius of (cid:4)0.9 mm increased to (cid:4)0.95 mm. The rim was also partially evaporated, Fig. 5. Mass deposition per unit area: experimental result vs. model which suggests that the rim was also illuminated. The predictionat3cmfromthespot. Fig.4. Ablationresponseoftheolivinesample(Left,withejecta;Right,minirocketplume). Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 8 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] r=3cm 10−4 10−5 m] h [ Δ 10−6 Model @ 2.59E−08 kg/s Test 1 @ 2.40E−08 kg/s Test 2 @ 2.03E−07 kg/s Model @ 3.84E−08 kg/s Test 3 @ 4.63E−08 kg/s 10−7 −60 −40 −20 0 20 40 60 80 θ [deg] Fig.6. MassdepositionperUnitArea:ExperimentalResultvs.Model Fig.8. Thicknessofthematerialdepositedontheslidesat3cmfrom Predictionat7cmfromtheSpot. thespot.Comparisonbetweenexperimentalmeasurementsandsimula- tionresults. r=10cm 10−2 r=7cm Model @ 2.82E−08 kg/s Test 1 @ 2.12E−08 kg/s 10−4 Model @ 2.59E−08 kg/s Test 2 @ 4.43E−08 kg/s Test 1 @ 3.90E−08 kg/s Model @ 4.07E−08 kg/s Test 2 @ 3.07E−08 kg/s Test 3 @ 3.28E−08 kg/s Model @ 3.84E−08 kg/s 2m] 10−3 Test 3 @ 5.65E−08 kg/s g/ 10−5 k A [ Δ m] Δ m/ 10−4 Δ h [ 10−6 10−5 −80 −60 −40 −20 0 20 40 60 80 θ [deg] 10−7 −80 −60 −40 −20 0 20 40 60 80 Fig. 7. Mass deposition per unit area: experimental result vs. model θ [deg] predictionat10cmfromtheSpot. Fig.9. Thicknessofthematerialdepositedontheslidesat7cmfrom thespot.Comparisonbetweenexperimentalmeasurementsandsimula- tionresults. increasingthetemperatureoftheplumeand inpartially meltingthetargetmaterialaroundtheablationrim[32]. Theself-cleaningactionofthelaserbeam,asdiscussed olivine at the critical frequency of the visible range and later, would also contribute to the rise in the ablated gas 1.2isthepeakincrementofreflectivelyatthefrequency temperature. The laser is considered to either: (1) re-heat of the laser (808nm). It was also assumed that the gas and re-sublimate the deposited ejecta [32] and/or (2) re- within the ejecta plume absorbed about 10% of the excite and re-direct any ejecta that enters the laser beam, incominglaserbeam.Thereforethepowervalueasgiven preventing it from reaching or depositing onto the micro- in Eq. (6) is the power absorbed at the spot. Z in abs scopeslides.Howevereachhypothesisneedstobevalidated Eq.(6)accountsfortheabsorptionofthelaserbeaminto furtherwithmoreexperimentsandSEManalysis.Afterthe the target material and the absorption of gas within initialablationevent,theablatedparticleswillalsorecom- the plume of ejecta. It is therefore expressed as bineandseparateintomuchsimplermoleculesandatoms. Z =0.9(1(cid:2)(0.3(cid:5)1.2)). abs This will release more energy into the flow and assist in These cumulative effects would explain the tempera- increasingthetemperatureoftheplume[27]. ture difference between the expected spot temperature The absorption of the laser radiation into the bulk of (3100–3800K) and the result measured by the spectro- thetargetmaterialalsohastobeconsidered.Tointroduce meter(4285–4747K).Theadditionalheatabsorbedinthe theactualabsorptionofthetargetmaterialtheworstcase Knudsenlayerisequivalenttoincreasingtheenthalpyof scenario was taken. This was considered to be sublimation by approximately 1–2n106 J/kg [28] and 1(cid:2)(0.3(cid:5)1.2). i.e. 0.3 is the worst-case reflectivity of heating up thegasfrom 3100 to 4747Kwouldconsume Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] 9 r=10cm r=7cm 10−5 1 0.9 0.8 0.7 Δ h [m]10−6 T 0.6 Model @ 2.59E−08 kg/s 0.5 Test 1 @ 3.90E−08 kg/s Test 2 @ 3.07E−08 kg/s Test 2 @ 4.43E−08 kg/s 0.4 Model @ 3.84E−08 kg/s Model @ 3.84E−08 kg/s Test 3 @ 5.65E−08 kg/s Test 3 @ 3.28E−08 kg/s 10−7 −80 −60 −40 −20 0 20 40 60 80 −60 −40 −20 0 20 40 60 θ [deg] θ [deg] Fig.12. DegradationFactor:comparisonbetweenExperimentalresults Fig.10. Thicknessofthematerialdepositedontheslidesat10cmfrom andmodelpredictionat7cmfromthespot. thespot.Comparisonbetweenexperimentalmeasurementsandsimula- tionresults. r=10cm 1 r=3cm 1 0.95 0.9 Model @ 2.59E−08 kg/s 0.8 Test 1 @ 2.40E−08 kg/s 0.9 Test 2 @ 2.03E−07 kg/s 0.7 Model @ 3.84E−08 kg/s 0.85 Test 3 @ 4.63E−08 kg/s 0.6 T 0.8 T 0.5 0.75 0.4 0.3 0.7 Test 2 @ 4.43E−08 kg/s Model @ 3.84E−08 kg/s 0.2 0.65 Test 3 @ 3.28E−08 kg/s 0.1 0 −60 −40 −20 0 20 40 60 −60 −40 −20 0 20 40 60 80 θ [deg] θ [deg] Fig.13. Degradationfactor:comparisonbetweenexperimentalresults Fig.11. Degradationfactor:comparisonbetweenexperimentalresults andmodelpredictionat10cmfromthespot. andmodelpredictionat3cmfromthespot. 6.15(cid:3)10(cid:2)7 and 4.10(cid:3)10(cid:2)6 N/W and an energy usage of (assuming a specific heat of 1361J/kgK) approximately between1.76(cid:3)108and1.17(cid:3)109J/kg.Therangeofdatais 2MW/m2ofenergy. caused by the variation in the mass flow rate of the Each experimental test reported in the figures is the ablated material. All values given are, in part, calculated averagemassflowrateoftheablatedtargetmaterial.The from the average ablated mass flow rate (over the abla- model presented in this paper, predicted a much higher tion period) gained from the experiment. The reported ablatedmassflowrate.Theinclusionoftheformationof momentum coupling coefficient was considered to be a theKnudsenlayer[30,33,34]andtheabsorptionofenergy more important parameter than the energy usage and in the plume provided some corrective factors that efficiency of the ablation process. An increase in the allowed for a more accurate prediction of the mass flow momentum coupling coefficient implies an increase in rate.Howeverthecurrentnumericalmodelisstilllimited. the ablated mass flow rate. Also unlike conventional It does not account for the three dimensional thermal propulsion-based ablation, the deflection of asteroids diffusioneffects.Thiscouldpotentiallyplayanimportant through laser ablation is not fuel (i.e. mass) limited. If roleinthereductionofthemassflowwithtime.Further required the entire asteroid could be ablated. This could analyses,therefore,needstobeperformedtoupdatethe provideapotentiallyendlesssupplyofpropellantforthe simulations with a more accurate model of the energy ablationprocess.Howeverthecurrentshowstopperisin drainingeffects. the degrading and contaminating effects of the ablated The efficiency of the ablation process also varied ejecta.Thisiscurrentlybeingaddressed. between 0.022% and 0.148%. This corresponded with a The velocity of the gaseous ejecta plume v was cal- measured momentum coupling coefficient of between culatedtobe(cid:4)1130.8m/s.Thiswasinferredbyusingthe Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008 10 A.Gibbingsetal./ActaAstronautica](]]]])]]]–]]] temperatureofsublimationofthetargetmaterial(3100– oftheexposedsurface.Theprofileofthedegradationfactor 3800K) and assuming Maxwell’s distribution of an ideal followscloselythedistributionofthethickness. gas.Itshouldbenoticedthatthetemperaturerecordedin The model predicted the ejecta to result in a signifi- theexperimentwasmorethantheminimumsublimation cantlygreaterdegradationthanwasotherwiseobserved. temperature assumed for the deflection simulations in Thisdegradationisexpectedtobehigheratlowereleva- Section2. tion angles,wheretheplume densityislarge. The thick- Figs. 5–7 shows the accumulated mass per unit area nessinsteadismuchhigherthanexpectedwithanequal for the different experiments (i.e. tests) and at different massperunitarea.Thedensityofthematerialdeposited distancesfromthespot.Thevaluesgivencorrespondtoa on the microscope slides is therefore lower than the fewdiscrete,butrepresentativesamplesthatweretaken 1000kg/m3 assumed in the simulation model. At 7 and alongeachmicroscopeslide.Eachtestproducedadiffer- 10cm the average density is about 250kg/m3. At 3cm entablatedmassflowrate,whichisalsoreportedinthe this is much higher with an average value over the figures. It should be noted that the reported mass flow centralslideofabout700kg/m3.Theaverageabsorptivity rateasgivenineachfigureistheaveragevaluethatwas at 7 and 10cm from the spot location is around experienced over a sublimation period of 10 mins. From 5(cid:3)104m(cid:2)1. However at 3cm from the spot location, on inspection it can be observed that there is a direct the central slide the average absorptivity increases to correlation between the amount of deposited material about 105m(cid:2)1. The absorptivity values then drops off andtheejectedmassflowrate.Theinputpowerfortest1 rapidly below 104m(cid:2)1 over the two microscope slides was43Wwhilefortest2and3itwas62W.Forallthree positionedat 7451. distances–either3,7or10cmawayfromthespot–the Itisthereforereasonabletoassumethatat3cmfrom depositedmasspredictedbythemodelisverysimilarto thespotlocationthattheplumeisveryfocussedandthat theexperiment. the material is mainly distributed over the central slide. Figs.8–10showsthethicknessofaccumulatedmaterial. At7and10cmfromthespotlocationtheplumeismore Theexperimentalresultsshowamuchhigherthicknessthan expandedandthereforeleadstoamoredistributedlayer predicted by the model, although with a similar variation ofmaterial.Inallcasesitseemsthatthemodelassumes withtheelevationangle.Anexceptiontothisisrepresented an incorrect growth of the deposited material, an incor- by the central point at elevation of zero degrees. This rectdensityandanincorrectabsorptivityvalue.Italsohas demonstrates that the laser beam has a cleaning effect on to be noted that the deposited material is not bonded the microscope slides. This is partially due to the fact that with the slides, i.e. by vibrating the slides most of the the particles in the laser beam tend to be excited while accumulatedmaterialcanberemoved. travelling towards the microscope slides and partially becausetheyarere-evaporatedoncedeposited.Figs.11–13 4.1. Reviseddeflectionsimulations showsthemeasureddegradationofthelightintensitygoing through the microscope slides. Within these figures the Data gained from the experiment campaign demon- experimentally derived degradation factor, as given in stratedagoodrepresentationtotheexpected,theoretically Eq(10)isdenotedT.Degradationiscausedbytheaccumula- predicted mass flow rate. However, the ejecta plume tionofthedepositedejecta.Thisreducesthetransmittance appearstobemorefocussedthanwasotherwisepredicted bythemodel.Thisfactorwouldincreasetheperformance ofthe deflection method butisdependent onthe typeof material and the composition of the given asteroid. For a Mirror diameter d=10m, Cr=5000, laser efficiency=0.6, rocky material, composed of olivine it is reasonable to cell efficeincy=0.4 considerthemodelasbeingconservative.However,further 10 10000 analysiswithdifferentmaterialsisstillrequiredtoprovide Number of Spacecraft 345678985 613613114111411669121976691272514219712197272516632539378127430923225531973253272534738019483732589353536574433803798124831794370953652724538352786994593642317538365158493309642148536974396585080597436774216949747788005533395899064175538830169646799484287095307935150361 3456789000000000000000000000 ftmbiiTmbmdntuhheeucloaaedaerlrstdimooiecetnsneeerharglvesmiflssaae,taeproailpssbacowbtiieacutntisltgrlahtslearatmyepiawrtitzoouitetpuesienwhrdlaoc.eldethweiiIinnforottauelohcfmnadrirtrrsrtteeoudeghhiannfatseeeesetslctrnresrih.aristodseebiyTias.talohisylaswnT1aneethtsi.ilti8seeoydlhitlsrrnhhimeeiaieanissnmgsmbuctmcpthirlgeern,oaeeoegtmtgttaredaehi.soripsseetTesnethleodhsahrafptiaratbitshertgsthyoehuasawhacuesrietomheenpmrfhsitcealoshtteorces.hhfeestlsNeaeattnahnhsfltoaceihbeetonrbeielaenwagldsltaaattps,tthttisooahheeateantnddoeesrt. 613 1669 23722553 2000 process extends beyond the area initially illuminated by 2 1141 21196769 1000 thelaser. 1 85 613 1141 Theexperimentalanalysishasalsodemonstratedthat 1 2 3 4 5 6 7 8 9 the contamination due to the re-condensation of the Warning Time [year] sublimatedmaterialisfarlowerthanforecastedinKahle et al., 2006 [11]. One reason seems to be that the Fig. 14. Revised deflection simulations with the experimentally mea- suredZ. condensed material has a much lower density than Please cite this article as: A. Gibbings, et al., Experimental analysis of laser ablated plumes for asteroid deflection andexploitation,ActaAstronautica(2012),http://dx.doi.org/10.1016/j.actaastro.2012.07.008