<%BANNER%>

Response of Liquid Xenon to Low-Energy Ionizing Radiation and its use in the Xenon10 Dark Matter Search

Permanent Link: http://ufdc.ufl.edu/UFE0041109/00001

Material Information

Title: Response of Liquid Xenon to Low-Energy Ionizing Radiation and its use in the Xenon10 Dark Matter Search
Physical Description: 1 online resource (159 p.)
Language: english
Creator: Manalaysay, Aaron
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: darkmatter, particle, xenon
Physics -- Dissertations, Academic -- UF
Genre: Physics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This dissertation focuses on developments aimed at improving the effectiveness and understanding of liquid xenon particle detectors in their use in the field of dark matter direct detection. Chapter \ref{chapter:xe10} covers the XENON10 experiment, which searches for evidence of direct interactions between Weakly Interacting Massive Particles (WIMPs) and Xe nuclei. The 3-D position sensitive liquid xenon time projection chamber acquired 58.6 live days of WIMP search data from October, 2006 through February, 2007. The results of these data set new limits on both spin-independent and spin-dependent interactions. The spin-independent WIMP-nucleon cross section is constrained to be less than 4.5E-44 cm^2 for WIMPs of mass 30 GeV/c^2 and less than 8.8E-44 cm^2 for WIMPs of mass 100 GeV/c^2 at the 90% confidence level. The spin-dependent WIMP-neutron and WIMP-proton cross sections are constrained to be less than 1E-39 cm^2 and 1E-36 cm^2, respectively. Finally, the mass of the heavy Majorana neutrino, in the context of a dark matter candidate, is excluded for masses in the range 10 GeV/c^2 to 2.2 TeV/c^2. Chapter 4 discusses the study of the relative scintillation efficiency of nuclear recoils in liquid xenon. The two existing measurements of the relative scintillation efficiency of nuclear recoils below 20 keV lead to inconsistent extrapolations at lower energies. This results in a different energy scale and thus sensitivity reach of liquid xenon dark matter detectors. A new measurement of the relative scintillation efficiency below 10 keV, performed with a liquid xenon scintillation detector and optimized for maximum light collection is discussed. Greater than 95% of the interior surface of this detector was instrumented with photomultiplier tubes, giving a scintillation yield of 19.6 photoelectrons/keV electron equivalent for 122 keV gamma rays. The relative scintillation efficiency for nuclear recoils of 5 keV is found to be 0.14, staying constant around this value up to 10 keV. For higher energy recoils we measure a value of 0.21, consistent with previously reported data. In light of this new measurement, the XENON10 experiment's upper limits on spin-independent WIMP-nucleon cross section, which were calculated assuming a constant 0.19 relative scintillation efficiency, change from 8.8E-44 cm^2 to 9.9E-44 cm^2 for WIMPs of mass 100 GeV/c^2, and from 4.5E-44 cm^2 to 5.6E-44 cm^2 for WIMPs of mass 30 GeV/c^2. In Chapter 6, I highlight the fact that a difficult task with many particle detectors focusing on interactions below ~100 keV is to perform a calibration in the appropriate energy range that adequately probes all regions of the detector. Because detector response can vary greatly in various locations within the device, a spatially uniform calibration is important. A new method for calibration of liquid xenon (LXe) detectors is presented, using the short-lived Kr-83m. This source has transitions at 9.4 and 32.1 keV, and as a noble gas like Xe, it disperses uniformly in all regions of the detector. Even for low source activities, the existence of the two transitions provides a method of identifying the decays that is free of background. At decreasing energies, the LXe light yield increases, while the amount of electric field quenching is diminished. Additionally, if any long-lived radioactive backgrounds are introduced by this method, it is shown that they will present less than 6.7E-5 events/kg/day/keV of background in the next generation of LXe dark matter direct detection searches.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Aaron Manalaysay.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Yelton, John M.
Local: Co-adviser: Baudis, Laura.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041109:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041109/00001

Material Information

Title: Response of Liquid Xenon to Low-Energy Ionizing Radiation and its use in the Xenon10 Dark Matter Search
Physical Description: 1 online resource (159 p.)
Language: english
Creator: Manalaysay, Aaron
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: darkmatter, particle, xenon
Physics -- Dissertations, Academic -- UF
Genre: Physics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: This dissertation focuses on developments aimed at improving the effectiveness and understanding of liquid xenon particle detectors in their use in the field of dark matter direct detection. Chapter \ref{chapter:xe10} covers the XENON10 experiment, which searches for evidence of direct interactions between Weakly Interacting Massive Particles (WIMPs) and Xe nuclei. The 3-D position sensitive liquid xenon time projection chamber acquired 58.6 live days of WIMP search data from October, 2006 through February, 2007. The results of these data set new limits on both spin-independent and spin-dependent interactions. The spin-independent WIMP-nucleon cross section is constrained to be less than 4.5E-44 cm^2 for WIMPs of mass 30 GeV/c^2 and less than 8.8E-44 cm^2 for WIMPs of mass 100 GeV/c^2 at the 90% confidence level. The spin-dependent WIMP-neutron and WIMP-proton cross sections are constrained to be less than 1E-39 cm^2 and 1E-36 cm^2, respectively. Finally, the mass of the heavy Majorana neutrino, in the context of a dark matter candidate, is excluded for masses in the range 10 GeV/c^2 to 2.2 TeV/c^2. Chapter 4 discusses the study of the relative scintillation efficiency of nuclear recoils in liquid xenon. The two existing measurements of the relative scintillation efficiency of nuclear recoils below 20 keV lead to inconsistent extrapolations at lower energies. This results in a different energy scale and thus sensitivity reach of liquid xenon dark matter detectors. A new measurement of the relative scintillation efficiency below 10 keV, performed with a liquid xenon scintillation detector and optimized for maximum light collection is discussed. Greater than 95% of the interior surface of this detector was instrumented with photomultiplier tubes, giving a scintillation yield of 19.6 photoelectrons/keV electron equivalent for 122 keV gamma rays. The relative scintillation efficiency for nuclear recoils of 5 keV is found to be 0.14, staying constant around this value up to 10 keV. For higher energy recoils we measure a value of 0.21, consistent with previously reported data. In light of this new measurement, the XENON10 experiment's upper limits on spin-independent WIMP-nucleon cross section, which were calculated assuming a constant 0.19 relative scintillation efficiency, change from 8.8E-44 cm^2 to 9.9E-44 cm^2 for WIMPs of mass 100 GeV/c^2, and from 4.5E-44 cm^2 to 5.6E-44 cm^2 for WIMPs of mass 30 GeV/c^2. In Chapter 6, I highlight the fact that a difficult task with many particle detectors focusing on interactions below ~100 keV is to perform a calibration in the appropriate energy range that adequately probes all regions of the detector. Because detector response can vary greatly in various locations within the device, a spatially uniform calibration is important. A new method for calibration of liquid xenon (LXe) detectors is presented, using the short-lived Kr-83m. This source has transitions at 9.4 and 32.1 keV, and as a noble gas like Xe, it disperses uniformly in all regions of the detector. Even for low source activities, the existence of the two transitions provides a method of identifying the decays that is free of background. At decreasing energies, the LXe light yield increases, while the amount of electric field quenching is diminished. Additionally, if any long-lived radioactive backgrounds are introduced by this method, it is shown that they will present less than 6.7E-5 events/kg/day/keV of background in the next generation of LXe dark matter direct detection searches.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Aaron Manalaysay.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Yelton, John M.
Local: Co-adviser: Baudis, Laura.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041109:00001


This item has the following downloads:


Full Text

PAGE 1

1

PAGE 2

2

PAGE 3

3

PAGE 4

Mytimeasagraduatestudenthasbeenabitatypical,spanningsixcitiesinfourcountriesontwocontinents,beginningandendinginTheSwamp.ItisthereforenotsurprisingthatIhavebenetedfrominteractionswithalargenumberofpeople.IthankmystudentcolleaguesatUFwithwhomIwentthroughthegradphysicscoursesandteaching.Guneeta,Shawn,Jesse,Tony,Corey,Dan,Alix,Dana,andLarry.Gettingthroughthoserstcoupleyearswouldhavebeenunbearablewithoutyourfriendshipandkindness.IthankZsoltMarcetforrunningaroundcampustosubmittherstdraftofthisdocumentbythedeadlinewhileIwasstillinZurich.Thephysicsdepartmentmachineshop,BillMalphursandMarcLinkinparticular,havebeenamazingintheirskillandprofessionalism.Withouttheirhardwork,muchoftheresultsreportedinthisdissertationcouldnothavebeencompleted.Youaretrulymasterartists.DarleneLatimerroutinelyroseaboveandfarbeyondherdutiesinordertohelpme.Inparticular,mytransitionfromcontinenttocontinentwouldhavefailedifnotforherassistanceanddedication.Youhavebeeninvaluabletomeinmytimeasagradstudent,andindeedtotheDepartmentyouareirreplaceable.IthankDavidHansenforhelpingmetonallyandforevershedmydependenceonWindows,andassistingmewiththemanycomputerproblemsIhadalongtheway.MyfellowXENON10gradstudentsandpost-docswithwhomIworkedinGranSassomademytimethereveryenjoyableandproductive.IthankKaixuanfortheRedstar,John,AngelandEricforthebilliardinomatches.Luiz,EricandPetercreatedawonderfulatmosphereintheapartmentandinHalldiMontaggio.Gomatlab!Peter,havingyounexttomegoingthroughthesame\patience-improvement"programwasalifesaver.Eric,IthinkIacquiredmuchofmyknowledgeofLXephysicstheoryfromconversationswithyou;it'sashameyou'velefttheeld,butgoodluckwithyourbubbles.JoergOrboeckwasapleasuretoworkwith,bothinFloridaandItaly.AlfredoFerella,the 4

PAGE 5

5

PAGE 6

6

PAGE 7

page ACKNOWLEDGMENTS ................................. 4 LISTOFTABLES ..................................... 10 LISTOFFIGURES .................................... 11 ABSTRACT ........................................ 15 CHAPTER 1INTRODUCTION .................................. 17 1.1Introduction ................................... 17 1.2EvidenceforDarkMatter ........................... 20 1.2.1GalacticScale-RotationCurves .................... 20 1.2.2ClusterScale-ClusterRedshiftSurveys ................ 21 1.2.3ClusterScale-GravitationalLensingandIntraclusterPlasma ... 23 1.2.4ClusterScale-ClustersMergers .................... 25 1.2.5CosmologicalScales ........................... 26 1.3DarkMatterCandidates ............................ 28 1.3.1Neutrinos ................................ 28 1.3.2Axions .................................. 29 1.3.3ThermalFreeze-OutandtheWeaklyInteractingMassiveParticle .. 31 2DIRECTDETECTIONANDLIQUIDXENON .................. 35 2.1TheLocalDarkMatterEnvironment ..................... 35 2.2WIMPInteractionRates ............................ 36 2.2.1Spin-IndependentInteractions ..................... 37 2.2.2Spin-DependentInteractions ...................... 38 2.3DirectDetectionStrategies ........................... 40 2.3.1ExamplesofDirectDetectionExperiments .............. 40 2.3.2WhyLiquidXenon? ........................... 43 2.4LiquidXenonInteractionPhysics ....................... 44 2.4.1MicroscopicProcessesinaParticleTrack ............... 44 2.4.2LindhardQuenching .......................... 46 2.4.3PuttingitAllTogether:Le 47 3THEXENON10EXPERIMENT .......................... 49 3.1TheXENON10DetectorandUndergroundFacility ............. 49 3.1.1DetectorDescription .......................... 49 3.1.2LaboratoriNazionalidelGranSasso .................. 51 3.1.3NuclearRecoilDiscrimination ..................... 52 3.2ElectronicRecoilBandShape ......................... 58 7

PAGE 8

.......... 59 3.2.2MonteCarloConstruction ....................... 61 3.2.3MonteCarloResults .......................... 63 3.2.4EnergyDependence ........................... 66 3.2.5Discussion ................................ 68 3.3WIMPSearch .................................. 69 3.3.1Spin-IndependentInteraction ...................... 70 3.3.2Spin-DependentInteractions ...................... 73 3.3.3ProspectsfortheHeavyMajoranaNeutrino ............. 77 4MEASUREMENTOFLEFFWITHTHEXECUBEDETECTOR ........ 81 4.1LeandtheNeedforitsFurtherStudy .................... 81 4.2MethodsforMeasuringLe 82 4.2.1MeasurementTechniqueandFacility ................. 82 4.2.2TheXecubeDetector .......................... 84 4.2.3DataAcquisition ............................ 86 4.3AnalysisandResults .............................. 88 4.3.1Calibrations ............................... 88 4.3.2EventSelection,Backgrounds,andResults .............. 89 4.4IndirectMethod ................................. 94 4.5Discussion .................................... 98 5THEXURICHDETECTOR ............................ 101 5.1TPCDesign ................................... 101 5.2AuxiliarySystems ................................ 104 5.2.1Cryostat ................................. 104 5.2.2GasSystem ............................... 105 5.3PhotomultiplierTubes ............................. 107 5.4DataAcquisitionandSignalProcessing .................... 108 5.4.1Hardware ................................. 108 5.4.2Software ................................. 110 5.4.2.1Preliminarydatamanipulation ............... 111 5.4.2.2S2nding ........................... 111 5.4.2.3S1nding ........................... 113 5.5LiquidLevel ................................... 113 5.6LXePurityandElectronLifetime ....................... 116 6LIQUIDXENONCALIBRATIONWITH83RB .................. 120 6.1TheNeedforaNewCalibrationSource .................... 120 6.2The83mKrSource ................................ 121 6.3AnalysisandResults .............................. 123 6.4Discussion .................................... 130 6.4.1LightYieldandFieldQuenching .................... 131 6.4.2RadioactiveBackgroundContamination ................ 132 8

PAGE 9

.......................... 133 6.5ExcitontoIonRatio .............................. 134 7PMTSTATISTICS .................................. 139 7.1AnalyticApproachtotheSinglePhotoelectronSpectrum .......... 140 7.2PMTMonteCarloandFunctionTest ..................... 143 7.3TheIndirectGainEstimationMethod .................... 147 8CONCLUDINGREMARKS ............................. 150 REFERENCES ....................................... 152 BIOGRAPHICALSKETCH ................................ 158 9

PAGE 10

Table page 3-1Nuclearrecoildiscriminationparametersandbackgroundestimates. ....... 57 3-2Thespinexpectationvaluesforprotonandneutrongroupsbasedonthreenuclearshellmodels. ..................................... 74 3-3ThepolynomialcoecientsofattothequasiparticleTamm-Dancospinstructurefunctions. ....................................... 75 4-1TheLeresultsfromtheneutronbeammeasurements. .............. 94 6-1Measuredlightyieldandelddependenceparameters. .............. 126 7-1MonteCarlodynodecongurationsandtfunctionperformance. ........ 146 10

PAGE 11

Figure page 1-1TherotationcurveforgalaxyNGC6503 ...................... 21 1-2Themeasuredmass-to-lightratios,insolarunits(M=L),foracollectionofgalaxyclusters,asafunctionoftheirvelocitydispersion. ............. 22 1-3Anexampleofstronggravitationallensing. ..................... 23 1-4Acompilationofthegasfractionofsixrichgalaxyclusters,asafunctionofredshift. ........................................ 24 1-5Examplesofcollidinggalaxyclusters. ........................ 26 1-6ThepredictedrelativeabundancesoflightelementsfromBigBangnucleosynthesis. 27 1-7Theaxion-photoncouplingversusaxionmassparameterspace. ......... 30 1-8Asurveyoftheinteractioncrosssectionversusparticlemassforvariousparticledarkmattercandidates. ............................... 32 2-1ThedierentialrecoilspectraofWIMPsinvariousdetectormaterials. ...... 39 2-2IonizationyieldversusenergyintheCDMS-IIexperiment. ............ 41 2-3Distributionofthediscriminationparameter,MTintheKIMSexperiment. .. 42 2-4Examplesofthreeclassesofbubble-creatinginteractionsintheCOUPPbubblechamber. ........................................ 43 2-5Potentialenergycurvesofground-stateargoninproximitytoexcitedorionizedargonatoms. ..................................... 45 3-1SchematicoftheXENON10detector. ........................ 50 3-2Thedriftvelocityofelectronsinxenonasafunctionofappliedelectriceld. .. 51 3-3LayoutoftheLaboratoriNazionalidelGranSasso ................ 52 3-4XENON10detectorandshielding. .......................... 53 3-5ElectronicandnuclearrecoilbandsintheXENON10detector. .......... 54 3-6TheattenedelectronicandnuclearrecoilbandsinXENON10. ......... 55 3-7Distributionsoflog10(S2=S1)fornuclearandelectronicrecoils. ......... 56 3-8TheelectronicrecoilrejectioninXENON10. .................... 57 3-9Decompositionoftheelectronicrecoilbandvariance. ............... 58 11

PAGE 12

.................. 60 3-11Spectrumofrecombinationuctuationsfrom131mXe. ............... 61 3-12Thephotonfractionfromlow-energy137CsComptonscattersandthecomparisonofdatatoMCinlog10(S2=S1)versusS1. ...................... 62 3-13Low-statisticcomparisonofdatatoMCoflog10(S2=S1). ............ 63 3-14High-statisticcomparisonofdatatoMCoflog10(S2=S1). ............ 64 3-15GaussianrejectionversusMCrejection. ....................... 65 3-16MC-basedcorrectionsto1RerandNleak. .................... 65 3-17Mappingofasymmetricintervalinphoton-fractionintoanasymmetricintervalinlog10(S2=S1)space. ................................ 66 3-18LinesofconstantS1andtheirspaninCES. .................... 67 3-19High-statisticcomparisonofdatatoMCoflog10(S2=S1)forvariousenergyranges. ......................................... 68 3-20CorrectionstoNleakgivenbytheMCappliedtoallenergies. ........... 69 3-21EvolutionofthelivetimeoftheXENON10blinddata. .............. 70 3-22XENON10WIMPsearchdata. ........................... 71 3-23XENON10exclusioncurveonthespin-independentWIMP-nucleoncrosssection. 72 3-24ThequasiparticleTamm-Dancospinstructurefunctionsandpolynomialts. 75 3-25PureprotonandpureneutronXENON10spin-dependentexclusionlimits. ... 76 3-26TheWIMP-neutronexclusionlimitcalculatedforfourdierentcombinationsof129Xeand131Xeshellmodels. ............................ 77 3-27XENON10C0asafunctionofmassfortheheavyMajorananeutrino. ...... 79 4-1AsurveyofLemeasurementsintheliteraturepriorto2009,alongwiththeenergyrangesrelevantfortheXENON10andZeplin-IIexperiments. ....... 81 4-2TheXENON10spin-independentWIMP-nucleoncrosssectionwithitsuncertaintyduetoLe. ...................................... 82 4-3Schematicdiagramoftheneutronbeamexperiment. ............... 83 4-4SchematicdiagramoftheXecubedetector. ..................... 85 4-5SchematicdiagramofthedataacquisitionsystemusedwiththeXecubedetector. 86 12

PAGE 13

........................ 87 4-7Spectrumfrom57CoinXecube. ........................... 89 4-8Distributionofeventsinpulseshapeparameterversustimeofight. ...... 90 4-9SelectedresultsofMonteCarlosimulationsoftheneutronbeammeasurements. 91 4-10Spectraofeventsfromtheneutronbeammeasurements. ............. 93 4-11MeasuredLevaluesasafunctionofXenuclearrecoilenergy. .......... 95 4-12RealandsimulatedspectraofelasticneutronscattersfromAmBeintheXecubedetector. ........................................ 97 4-13TheXENON10spin-independentWIMP-nucleoncrosssectionexclusionlimitusingLefromthisstudy. .............................. 99 5-1DiagramoftheXurichdual-phasetimeprojectionchamber. ........... 102 5-2Spectrumfrom57CoatzeroeldintheXurichdetector. ............. 103 5-3ThecryostatusedfortheXurichdetector. ..................... 104 5-4Cryostatperformanceoverroughlyonemonth. ................... 105 5-5ThegassysteminchargeofXelling,purication,recovery,andstorage. .... 106 5-6OneofthephotomultipliertubesusedintheXurichdetector. .......... 107 5-7SinglephotoelectronspectrafromXurich'sphotomultipliertubesatvaryingappliedcathodevoltages. ................................... 108 5-8Schematicofthedataacquisitionsystem. ...................... 109 5-9MeasuredandsimulatedtriggereciencyoftheXurichdetector. ........ 110 5-10AnexamplerawPMToutputtracefromaneventindual-phasemode. ..... 112 5-11ThecalculatedS2gainasafunctionofgasgap. .................. 114 5-12ThespectraofS2atvariousazimuthalpositionsbeforelevelingthedetector. .. 115 5-13ThespectraofS2atvariousazimuthalpositionsafterlevelingthedetector. ... 115 5-14TherateconstantforattachmentofelectronsonO2,N2O,andSF6inLXeasafunctionofappliedeld. ............................... 117 5-15S2versusdrifttimebeforeandafterpurication. ................. 118 5-16Evolutionoftheelectronlifetimeoverthecourseofoneweekofpurication. .. 118 6-1Thedecayschemeandbranchingratiosof83mKr. ................. 122 13

PAGE 14

..................................... 124 6-3DistributionofdelaytimesbetweenrstandsecondS1pulses. .......... 125 6-4Fieldquenchingofthreespectrallinesinliquidxenon. .............. 127 6-5Spectraforthelineat41.5keVinS1,S2,andcombinedenergy. ......... 129 6-6Rateof83mKrdecaysasafunctionofz-position. .................. 130 6-7ConstraintsonNex=Nionand. ........................... 135 6-8Theinversechargecollectionversustheinverseappliedelectriceldofthe41.5keVline. .......................................... 137 6-9S1versusS2shownforthe41.5keVlinetakenatvariousappliedelds,showingtheanticorrelationofthetwosignals. ........................ 138 7-1Schematicdiagramofaphotomultipliertube. ................... 139 7-2Analyticprobabilitydistributionofaphotomultipliertubeoutput. ........ 142 7-3AnexampleofrealPMTsinglephotoelectronspectra. .............. 143 7-4SampleofsimulatedMonteCarloSPEspectra. .................. 145 7-5Distributionofgainestimators. ........................... 147 7-6SpectraofPMToutputfromvaryingtheLEDintensity. ............. 148 7-7VarianceversusmeanfromvariedLEDilluminations. ............... 149 14

PAGE 15

3 coverstheXENON10experiment,whichsearchesforevidenceofdirectinteractionsbetweenWeaklyInteractingMassiveParticles(WIMPs)andXenuclei.The3-Dpositionsensitiveliquidxenontimeprojectionchamberacquired58.6livedaysofWIMPsearchdatafromOctober,2006throughFebruary,2007.Theresultsofthesedatasetnewlimitsonbothspin-independentandspin-dependentinteractions.Thespin-independentWIMP-nucleoncrosssectionisconstrainedtobelessthan4:51044cm2forWIMPsofmass30GeV/c2andlessthan8:81044cm2forWIMPsofmass100GeV/c2atthe90%condencelevel.Thespin-dependentWIMP-neutronandWIMP-protoncrosssectionsareconstrainedtobelessthan1039cm2and1036cm2,respectively.Finally,themassoftheheavyMajorananeutrino,inthecontextofadarkmattercandidate,isexcludedformassesintherange10GeV/c2to2.2TeV/c2.Chapter 4 discussesthestudyoftherelativescintillationeciencyofnuclearrecoilsinliquidxenon.Thetwoexistingmeasurementsoftherelativescintillationeciencyofnuclearrecoilsbelow20keVleadtoinconsistentextrapolationsatlowerenergies.Thisresultsinadierentenergyscaleandthussensitivityreachofliquidxenondarkmatterdetectors.Anewmeasurementoftherelativescintillationeciency 15

PAGE 16

6 ,Ihighlightthefactthatadiculttaskwithmanyparticledetectorsfocusingoninteractionsbelow100keVistoperformacalibrationintheappropriateenergyrangethatadequatelyprobesallregionsofthedetector.Becausedetectorresponsecanvarygreatlyinvariouslocationswithinthedevice,aspatiallyuniformcalibrationisimportant.Anewmethodforcalibrationofliquidxenon(LXe)detectorsispresented,usingtheshort-lived83mKr.Thissourcehastransitionsat9.4and32.1keV,andasanoblegaslikeXe,itdispersesuniformlyinallregionsofthedetector.Evenforlowsourceactivities,theexistenceofthetwotransitionsprovidesamethodofidentifyingthedecaysthatisfreeofbackground.Atdecreasingenergies,theLXelightyieldincreases,whiletheamountofelectriceldquenchingisdiminished.Additionally,ifanylong-livedradioactivebackgroundsareintroducedbythismethod,itisshownthattheywillpresentlessthan67106eventskg1day1keV1ofbackgroundinthenextgenerationofLXedarkmatterdirectdetectionsearches. 16

PAGE 17

1 ].Newton'slawofgravitationnallyclosedforeveranypossibilityofanadherencetoageocentriccosmology,andindoingso,expandedthescaleoftheobservableuniverse 17

PAGE 18

2gR=8GN 1{1 ).Thespace-timeindices,and,runfrom0to3,andhencethetensorsinequation 1{2 containsixteenelements.Althoughtheseelementsarenotallindependentduetothesymmetryofg,wearestillleftwithtenindependent,nonlinear,second-order,coupleddierentialequations.Exactsolutionstoequation 1{2 arerare,andcanonlybemadeinsystemsthatexhibithighdegreesofgeometricandtemporalsymmetry.CosmologistsexploittheCopernicanprincipleandthefactthattheUniverseappearstobehomogeneousandisotropiconlargescales(&100Mpc).Withthesesymmetries,thesolutiontoequation 1{2 isgivenbytheFriedmann-Lema^tre-Robertson-Walker(FLRW)metric,whencetheinvariantlineelementinsphericalcoordinatesis, 18

PAGE 19

a;(1{4)withthedotdenotingthederivativewithrespecttocoordinatetime.InsertingtheFLRWmetricbackintoequation 1{2 ,andtakingthe00componentgivestheFriedmannequation, a2=8GN ii 19

PAGE 20

2 ],andlaterbyVoldersofM33in1959[ 3 ]andbyRubinofM31in1970[ 4 ],haveevolvedintoscoresofevidencethatallpointtothefactthatroughly98%ofthematterintheuniverseisnon-stellar,androughly85%isnonbaryonic[ 5 ].Idiscusssomeofthisevidence,fromgalacticscalestocosmologicalscales. 6 ].Thecloudofneutralhydrogentypicallyextendsfarbeyondthevisiblediskofstars,andhencecanprobemoreofthegalaxythanthestarsthemselves.Measurementsofthistypethenallowonetoconstructarotationcurveofthegalaxy,whichissimplyaplotoftherotationalvelocityofthegalacticmaterialasafunctionofthedistance,r,fromthegalacticcenter.Newtoniandynamicspredictstherotationcurvebasedonthetotalmass,M(r)locatedinsider, 1-1 shows 20

PAGE 21

TherotationcurveforgalaxyNGC6503,showingtheexpectedcontributionfromthediskandgas.Themeasuredvalues(datapoints)requireanadditionalcontributionfromanon-luminescenthalo.Figuretakenfrom[ 6 ]. therotationcurveofgalaxyNGC6503.Themattercontentofthediskandgascanbemeasured,andtheirexpectedcontributionstotherotationalvelocitypredictedusingequation 1{8 .Themeasuredvalues(datapoints)requiretheadditionofanadditionalhaloofmaterialnotvisiblewithtelescopes. 7 ].Theline-of-sightvelocityofthesegalaxiesisobtainedbymeasuringtheirredshifts.Usingthesemeasuredvelocities,Zwickythencalculatedthetotalgravitationalpotentialusingthevirialtheorem, 2hTi=hVtoti;(1{9)whereTistotalkineticenergy,Vtotisthetotalgravitationalpotentialenergy,andtheanglebracketsdenotetheaverageovertime. 21

PAGE 22

8 ].Thisresult,thatY1,impliesthepresenceoflargequantitiesofadditional,invisiblemass. Figure1-2. Themeasuredmass-to-lightratios,insolarunits(M=L),foracollectionofgalaxyclusters,asafunctionoftheirvelocitydispersion.Theextremedeviationofthesevaluesfromunityisaclearindicationthatmorematterexistsintheclustersthansimplythestarsandgasobservablebytelescopes.Plottakenfrom[ 9 ]. TheanomalousvalueofYindicatedaboveisnotlimitedtotheComacluster.Infact,suchalargediscrepancyisseenineverygalaxyclusterinwhichitismeasured.AsurveyovermanygalaxyclusterhasfoundanaverageclustervalueofY=24050[ 9 10 ].Theseresultsimplythatcluster=0:190:07[ 10 ]. 22

PAGE 23

1-3 (left)showsadramaticexampleofgravitationallensing.Insituationslikethese,thelightemittedfromadistantgalaxy(the\lensedobject")isbentbythegravitationalwellofagalaxycluster(the\lensobject")lyingdirectlybetweentheEarthandthedistantgalaxy.Thespaceinthevicinityofthelensingclusteriscurvedinsuchawaythatasthelightfromthedistantgalaxyfollowsgeodesics,deviatesfromastraightlineandthenreachestheEarthfrommultiplepointsinthesky,producingaseriesofwarpedimages. Figure1-3. (Left)AstunningexampleofgravitationallensingvisibleintheAbell370galaxycluster,locatedinthenorthernconstellationCetus.Thebrightyellowgalaxiesvisiblethroughouttheeldaremembersofthelensingcluster,producingthemultiple,distortedimagesofthered-bluebackgroundgalaxy.(Right)AreconstructionofthemassproleinthegalaxyCL0024+1654basedonstronggravitationallensing.Thisclusterliesroughly5billionlight-yearsawayintheconstellationPisces.Thespikesinthemassprolemarktheindividualgalaxies,however,itisclearthatanadditionalcollectionofmassliesbetweenthegalaxies.Figuretakenfrom[ 11 ]. ObservationsofgravitationallensingareaconrmationofEinstein'stheoryofgeneralrelativity.Butmorethanthat,theycanbeusedtoprobethedistributionofmasswithinthelensingcluster[ 12 ].TheclusterCL0024+1654actsasalensofasinglebackgroundgalaxy,locatedroughly10billionlight-yearsaway.Usingthemultipleimagesofthisbackgroundgalaxy,itispossibletocalculatethemass-to-lightratio,Yofthelensing 23

PAGE 24

13 ],inaccordwiththevelocitydispersionmeasurementsofthelastsection.WhilethemeasurementofYisindicativethatadditionalmatterexistsintheclustersthanjustthestars,itdoesnotruleoutthepossibilitythattheextramassisintheformofsomeother,non-optically-luminous,butbaryonic,component.Indeed,galaxyclusterscontainlargequantitiesofhot,x-rayemittingplasma.Whilethedensityofthisintraclusterplasmaisverylow,ontheorderof1026gcm3,itisnotboundtotheindividualgalaxiesandinsteadsmoothlypervadesthewholecluster.Therefore,thetotalplasmamasscanbequitelarge,andinfactexceedsthemassofluminousmaterialbyafactorof6[ 14 ]. Figure1-4. Acompilationofthegasfractionofsixrichgalaxyclusters,asafunctionofredshift.Themassfractionisdenedasthefractionthatintraclusterplasmacontributestothetotalmassofthecluster.Theintraclusterplasmaconstitutesthemajorityofbaryonicmatterinagalaxycluster,andhenceanadditional,nonbaryoniccomponentisneededtoaccountforthefactthatfgas<1.Figuretakenfrom[ 15 ]. Theluminosityoftheplasmainx-raysisproportionaltothesquareofthedensity,andthereforetheplasmamassofaclustercanbedeterminedfromobservationswithx-raytelescopes.Whenthesemeasurementsarecombinedwithmeasurementsofthetotalclustermassfromgravitationallensing,adeterminationcanbemadeofthegasfraction, 24

PAGE 25

1-4 asafunctionofredshift.Theweightedaverageoftheseresultsgivefgas=0:1130:005[ 15 ]. 1-3 (right),thedominantcomponentofagalaxycluster'smassisthedarkmatter.Additionally,thereexistvastcloudsofhot,x-rayemittingintraclusterplasma.Whileneitherofthesetwocomponentsarevisibleinopticalwavelengths,theymakeupthebulkoftheclustermass.Themaindierencebetweenthedarkmatterandtheintraclustergasisintheirinteractionstrengths:gasiscollisional,darkmatterisnot.Therefore,whentwoclustersofgalaxiescollide,theconglomerationsofdarkmatterwillpassrightthroughoneanother,astheyexperiencemainlygravitationalinteractions.Theintraclusterplasmaclouds,however,willinteractelectromagnetically,andhencewillexhibitverydierentdynamicsduringthecollisionthanthedarkmatter.Fortunately,duetotheirdierentqualities,thedierentcomponentscanbestudiedseparately.Thedensityanddistributionoftheplasmacanbestudiedbyobservingthex-rayemission[ 15 ].Incontrast,thedominantmassoftheclusterscanbestudiedbygravitationallensing,asdiscussedinsection 1.2.3 .TheresultsfromfourexamplesofclustercollisionsareshowninFigure 1-5 .Theseexamplesarefrom(clockwisefromtop-right)theBulletcluster[ 16 ],MACSJ0025.4-1222[ 17 ],MACSJ0717.5+3745[ 18 ],andAbell520[ 19 ].Ineachexample,theextentofclusterplasma(determinedfromx-rayemission)isindicatedinpink,whilethedistributionofmass(determinedfromgravitationallensing)isindicatedinblue.MostvisibleintheBulletclusterandinMACSJ0025.4-1222isthatthecloudsofhotgashavebeenstrippedawayfromtheirparentclusters.Inallotherdarkmatterevidence, 25

PAGE 26

Examplesofcollidinggalaxyclusters.Clockwisefromtop-leftareshowntheBulletcluster[ 16 ],MACSJ0025.4-1222[ 17 ],MACSJ0717.5+3745[ 18 ],andAbell520[ 19 ].Ineachcasetheintraclusterplasmaisshowninpink(whichconstitutesthemajorityofthebaryonicmass),whilethedominantclustermassisshowninblue.Thedisplacementofonefromtheothercanonlybeconsistentlyexplainedbydarkmatter. oneisconsideringdiscrepanciesinthestrengthofthegravitationalforce.However,heretheevidenceismuchmoreclear:thedominantmassislaterallydisplacedfromthebaryonicmatter.Furthermore,theresultisconsistentwiththeexpectationsofcollisionlessdarkmatter. 26

PAGE 27

Figure1-6. Thepredictedrelativeabundancesofthelightelements,dependingonasingleparameter,thebaryon-to-photonratio,.Measurementsoftheactualabundancesareindicatedbytheboxes:yellowboxesrepresent2statisticaluncertaintyonthemeasurements,larger,dashedboxesrepresentthe2statisticalandsystematicuncertaintyofthesamemeasurements.The95%condenceboundsofbh2fromBBNaremarkedbytheverticaltanlines,andthemeasurementofthesameparameterfromtheCosmicMicrowaveBackgroundisshownbythetext`CMB'.Figurefrom[ 20 ]. TheprocessesbywhichthelightelementsareproducedintheBigBanginvolvesfairlywell-studiedparticleandnuclearphysics.Thetheoryveryuniquelypredictstherelativeabundancesofthelightelements(3;4He,2H,and7Li)andischaracterizedbyasingleparameter,,thebaryon-to-photonnumberratio(seeFigure 1-6 ).Thepredictionsofthemodelareremarkablyconsistentwiththemeasurementsoftherelativeabundances,andgiveavalueof=(5:6+0:80:7)1010[ 21 ].CombiningthiswiththeknowndensityofphotonsintheUniversefromthecosmicmicrowavebackground(CMB)givesthe 27

PAGE 28

bh2=0:020+0:0030:002;(1{10)wherethesubscriptbdenotesbaryonsandhistheHubbleparameter(h=H0/100kms1Mpc1,H0=71:9+2:62:7kms1Mpc1isthepresentvalueoftheHubbleexpansionrate[ 22 ]).TheCMB,thesmoothT=2:726KblackbodyradiationleftoverfromtheBigBanghastemperatureuctuationsatthe10Klevel.Theseanisotropiesareadirectresultoftemperatureuctuationsatthetimewhenelectronsandnucleirstcombinedtoformneutralatoms.Thenatureoftheseuctuationsinturnisverysensitivetothecontentsoftheuniverse.The5-yeardataoftheWilkinsonMicrowaveAnisotropyProbe(WMAP)haverecentlybeenreleased,placingtightconstraintsonazooofcosmologicalparameters.Ofrelevancetothepresentdiscussionarethedensityofbaryonsandtotalmatter,givenas[ 22 ], bh2=0:022730:00062;mh2=0:13260:0063;(1{11)showingclearagreementwiththeresultsofBBNonb.Thevaluemisthedensityofallmatter. 1.3.1NeutrinosWiththeknowledgethatthedarkmatterisnonbaryonic,electricallyneutral,andstable,itisnaturaltorstlookattheStandardModel(SM)forapotentialculprit.TheonlySMparticlesthatmeetthesecriteriaaretheneutrinos.Neutrinoswereactiveintheearlyuniverseandplayedaroleintheformationoflightnuclei.Theirrelicabundanceisgivenby[ 7 23 ], h2=Xigimi 28

PAGE 29

24 25 ].Thisimpliesthat h2&0:0006:(1{13)Theabundanceofrelicneutrinoscanbeprobedfromcosmologicalmeasurements.Priortotheirthermaldecoupling,largeamountsofprimordialneutrinoswouldacttodecreasedampingofoscillationsintheearlyphoton-baryonplasma,whichwouldincreasethestrengthofthepeaksintheCMBanisotropies.Additionally,theexpansionrateoftheuniversewouldbealtered,therebyshiftingthepositionoftheacousticpeaks.Primordialneutrinos,decouplinghot,wouldsmoothoutstructureonsmallscales(.40Mpc).Thiswouldimplythatlargescalestructureformed\top-down",meaninglargescalesformedrst,withsmallscalestructureforminglater.Thisscenarioisunlikely,astheMilkyWayappearstobemucholderthanthelocalgroup.ThesecosmologicalconstraintsimplythatPm<0:61eV(95%C.L.).Usingthisresultwithequation 1{12 impliesthattheseresultsgiveanupperlimitonthetotalcontributionofneutrinostobe[ 22 ], h2<0:0065(95%C:L:):(1{14)Thoughitisclearthatneutrinosdocontributetothetotalenergycontentoftheuniverse,theycannotaccountforthedarkmatter. 26 ]after'tHooftshowedthatstronginteractionspossessanunboundedparameterallowingCP-violation.Axionscouldhavebeenproducedincosmologically-interestingamountsintheearlyuniversebyavarietyofmechanisms,thefavoredbeingvacuummisalignment[ 27 ].This 29

PAGE 30

28 ], a5eV 1{15 ,becauseagrowsasmadecreases.TheupperboundfromSn1987acomesfromthefactthatiftheaxion'scouplings(proportionaltoma)aretoogreat,itwouldallowsignicantcoolingduringthesupernovaandwouldbeobservable. Figure1-7. Theparameterspacetypicallyusedforaxionsearches,axion-photoncouplingversusaxionmass.Themassrangeallowedforinterestingcosmologicalconsequencesis106eV.ma.103eV.Axiondarkmattersearchesarethoselabel\microwavecavity".Figuretakenfrom[ 28 ]. 30

PAGE 31

27 ].Theexperimentsutilizeradioresonancecavitieswithtunableresonancefrequencies,withappliedstaticmagneticelds.Thestandardaxionparameterspace,alongwiththeresultsofrecentsearches,isshowninFigure 1-7 29 ]: Xh2=mXnX 30 ],whichalreadyrulesoutmostoftheparticlesintheStandardModel.Asdiscussedwithrelicneutrinosinsection 1.3.1 ,a\hot"(i.e.relativistic)darkmattercandidatewoulddestroytheformationoflarge-scalestructure.Whilethedensity 31

PAGE 32

23 29 ].Thisplacesalowerlimitonthemassat10keV.However,thefactthatsuchaparticlehasnotbeenseenincolliderslikeLEPincreasesthelowerlimitto30GeV[ 29 ].Giventheseproperties,thatsuchaparticlemusthaveaweakcrosssectionandlargemass,thistypeofdarkmattercandidateistypicallycalledaWeaklyInteractingMassiveParticle,orWIMP.Inadditiontoalowerlimitonthemass,anupperlimitcanbeinferred.Basedontheso-calledunitaritybound,whichimpliesarelationshipbetweenaparticle'smassandit'smaximumpossibleannihilationcrosssection,cosmologicalmeasurementsconstrainthatmXislessthan34TeV[ 7 ]. Figure1-8. Asurveyoftheversusmassparameterspaceforvariousparticledarkmattercandidates.ThesolidblacklineistheupperlimitfromtheXENON10WIMPsearch[ 31 ].Figureadaptedfrom[ 32 ]. Thepowerofthethermalfreeze-outmechanismisthatitismodelindependent.Itrequiresonlythatnaturesimplyallowsfortheexistenceofaparticlewiththoseproperties;thespecicsofthemodeldonotcomeintothecalculationofequation 1{17 .ItisalsoafactthatanynewphysicsbeyondtheStandardModel(BSM)almostgenericallyproducesaparticlewiththeseproperties.ExistingresultsofcolliderexperimentscanbeusedtomakeanindirectestimateoftheHiggsbosonmass,atmH=129+7449GeV/c2[ 20 ]. 32

PAGE 33

33 ],whereistheweakcouplingconstant.Additionally,newparticlesattheTeVscalewouldsignicantlyaltertheresultsofprecisionstudiesofelectroweakphysics,andthusaconservationlawmustbeinvokedthatallowsonlyevennumbersofBSMparticlesatinteractionvertices,forBSMparticlesupto5-7TeV[ 34 ].SuchaconservationlawwouldforcethelightestofsuchBSMparticlestobestable.Figure 1-8 showsvariousparticledarkmattercandidates,alongwiththeXENON10WIMPsearchexclusionlimit[ 31 ].PopularWIMPcandidatesaretheneutralinofromsupersymmetry,theLKPfromuniversalextradimensions,andthelittleHiggsmodel.IntheMinimalSupersymmetricStandardModel(MSSM),thesuperpartnersofthestandardmodelgaugebosonsmixintotwochargedmasseigenstatescalledcharginos,~1;2,andfourneutraleigenstatescalledneutralinos,~01;2;3;4[ 7 ].Inmanyscenarios,thelightestsupersymmetricparticleis~01.VarioustheoreticalargumentssuggestthatthereisanadditionalsymmetrycalledR-parity,theleadstotheconservationofR(1)2s+3B+L,wheresisthespin,Bisthebaryonnumber,andListheleptonnumber.ThereforeStandardModelparticleshaveR=1andallsuperpartnershaveR=1;thelightestsupersymmetricparticlewouldthenbestable.Theoriesthatexplorethepossibilityoftheexistenceofmorethan3+1dimensionsarecalledKaluza-Kleintheories.Theextradimensionsmustinsomewaybecompactied,meaningtheyarewrappeduponsomesmallsize,explainingwhywedonotexperiencethem.Themomentumofeldspropagatingintheseextracompactieddimensionsthusbecomesquantized.AllStandardModelparticlesexhibitthelowestmomentummodeintheextradimensions,andexcitationshaveanincreasedmassaccordingtomn/n=R,wherenistheexcitedmode(StandardModelparticleshaven=0),andRisthesizecharacterizingthescaleofcompactication.Conservationofmomentumin 33

PAGE 34

34 ].LittleHiggsmodelsgenericallycontainnewparticlesattheTeVscalewhichcouldaccountforthedarkmatter. 34

PAGE 35

1.2.1 ).Whilemanydarkmattercandidateshavebeenproposed,withvaryingdegreesofjustication,noneoftheexistingobservationscantellusmuchmoreaboutdarkmatter'sidentity.Inordertolearnmore,thedarkmattermustbeunambiguouslydetected,ortheproductsofitsdecay,annihilation,orco-annihilationmustbedetected.Thelatter,knownasindirectdetection,hasbeenoered,forexample,asapossibleexplanationfortheexcessof511keV-rayscomingfromthecenteroftheMilkyWay[ 35 ].Theformer,knownasdirectdetection,aimstoobserveWIMPsinteractingwithnormalmatter.Thebasicsofdirectdetection,themostsensitiveexamplesofexistingexperimentaleorts,thebenetsofliquidxenon(LXe),andnally,thephysicsofparticleinteractionsinLXearediscussedinthepresentchapter. 7 ].Thesetechniquesestimatethelocaldarkmatterdensitytoliesomewhereintherange0:2.0.0:6GeVc2cm3,withthepreferredvaluebeing0=0:3GeVc2cm3,acharacteristicaveragevelocityofv=230kms1andescapevelocityof600kms1[ 7 29 ]. 35

PAGE 36

36 ].Thoughfeeble,WIMPspassingthroughtheEarthshouldoccasionallyinteractwithnormalmatter.Theenergytransferedintheseinteractionsisexpectedtobesmall,butnonethelessdetectablegivenaparticledetectorwiththeappropriateproperties. mmN;(2{1)where0isthemassdensityofWIMPs,istheelasticscatteringcrosssection,andmandmNarethemassesoftheWIMPandnucleus,respectively.Thispictureis,however,toosimpletobeusefulinthisform,fortworeasons.First,thoughthepicturepaintedintheprevioussectionisthatofanEarthyingthroughagasofWIMPsat244kms1,thevelocityoftheWIMPsthemselvesisfarfromuniform,andhencetheWIMPvelocitydispersionmustbetakenintoaccount.Thisisdonebyreplacingthevelocity,v,byakinematicformfactor,T(Q),whereQistheenergytransfer,thatisweightedaccordingtotheallowedvelocities.Second,theelasticscatteringcrosssection,,isnotuniformwithenergy,andmustinsteadbereplacedby!0F2(Q),where0isthecrosssectioninthelimitofzeroenergy-transfer,andF2(Q)isthenuclearformfactor,characterizinghowthecrosssectionevolveswithenergy.Combiningthesemodicationsgivesthetotal 36

PAGE 37

29 ], dQ=00 36 ], 2{2 unaddressed:0andF2(Q).ThesedependonthetypeofinteractionthatgovernstheWIMP-nucleusscatter,andcannotbesolvedinthegeneralsense.AtthelowvaluesofQthattypicallycharacterizedirectsearches,thetwotypesofinteractionsofimportancearescalar(spin-independent)andaxial-vector(spin-dependent)[ 7 ]. 29 ],andisgivenby, 37

PAGE 38

2-1 .TheadvantagegiventoXe(A=131)bythescalingof0withA2isevidentatlowenergies,demonstratingoneadvantagethatthisnucleushasfordirectdetection. 2{2 .However,becausetheWIMPscoupletothenuclearspin,thedetailedstructureofthenucleusmustbeconsidered.Additionally,unliketheSIcasewhereitwasreasonabletotaketheprotonandneutroncouplingsasbeingidentical,herewecannotmakesuchanassumption.TheSDcrosssectionatzero-momentumtransferisgivenas[ 29 ], 38

PAGE 39

Theexpectedspin-independentdierentialrecoilspectrainAr,Ge,andXeofaWIMPofmass100GeVc2,andacrosssectionwithnucleonsof11044cm2 3.3.2 .Becausetheprotonandneutroncouplingsareunequal,thecrosssectioncannotbenormalizedtotheWIMP-nucleoncrosssection.Instead,thenormalizationisperformedbyconsideringiftheWIMPsweretoonlycoupletoprotons(i.e.an=0)andthennormalizingtotheWIMP-protoncrosssection,as 3mr 39

PAGE 40

2{6 ),demanddetectormaterialswithlargenuclei.Thelowexpectedeventrate(Figure 2-1 )requiresadetectorwithalargeoveralltargetmass.Severaldirectdetectionexperimentsfollowingtheserequirements,whichhavethecurrentbestsensitivities,arediscussedinthissection,nishingwithadiscussionofwhatLXehastooertheeld. 31 ](thefocusofChapter 3 )andCDMS-II[ 37 ].Thephysicsunderlyingthetechniquesthatbothdetectorsuseinordertodistinguishelectronicfromnuclearrecoils,nuclearrecoildiscrimination,aresimilar.Thatis,bothexperimentsmakeuseofparametersrelatedtotheionizationyield,ortheamountofionizationcollectedforagivenenergy.TheprocessofnuclearrecoildiscriminationintheXENON10experiment 40

PAGE 41

3.1.3 .Theionizationyieldisemployedbecausenuclearrecoilsproduceahigherionizationdensitythanelectronicrecoils,andhenceleadtostrongerelectron-ionrecombinationfollowinganinteraction.TheCDMS-IIexperimentusesanarrayofgermaniumandsilicondetectorscooledtotensofmK.Thedetectorsmeasureenergydepositionsimultaneouslyintheformofathermalballisticphononsandionization.Theionizationyieldoftheinteractionistakenfromtheratioofthetwosignals,showninFigure 2-2 .Asexpected,thestrongerelectron-ionrecombinationofnuclearrecoilsresultsinasuppressedionizationyieldcomparedtoelectronicrecoils. Figure2-2. IonizationyieldversusenergyintheCDMS-IIexperimentfromcalibrationsources.Electronicrecoilsareinblue,nuclearrecoilsingreen.Solidanddashedlinescorrespondtothe2boundsoftheelectronicandnuclearrecoilbands,respectively.Figuretakenfrom[ 38 ]. Thecurrenttwostrongestlimitsonpure-protonSDinteractionscomefromtheKIMS[ 39 ]andCOUPP[ 40 ].Thetwoexperimentsusevastlydierenttechniques,butsharetheirexceptionalsensitivitytopure-protonSDinteractionsduetothehighprotoncontentoftheirnuclearspins. 41

PAGE 42

39 ]usesanarrayofCsI(Tl)scintillatingcrystals,heldatT=0C.Thisexperimentmakesuseofthefactthatthescintillationemissiontimescaleforelectronicandnuclearrecoilsisstatisticallydierent,whichresultsfromthedieringlinearenergytransfer(LET)ofthetwospecies.Adistributionfunctioncharacterizingthearrivaltimeofphotoelectronsfromthephotomultipliertubes,f(t),isconstructed,andisthenusedtondthemeantime(MT)ofaneventfromMT=Rtf(t)dt=Rf(t)dt.ThedistributionsofMTforelectronicandnuclearrecoilcalibrationdataareshowninFigure 2-3 .Giventherelativeoverlapofthetwosignals,nuclearrecoildiscriminationmustbeperformedonastatisticalbases,ratherthananevent-by-eventbasisasintheCDMS-IIandXENON10experiments. Figure2-3. Distributionsofthediscriminationparameter,meantime,fromonecrystalusedintheKIMSexperiment.Opensquaresarefromnuclearrecoilcalibrationdata,opencirclesfromelectronicrecoilcalibration,andclosedtrianglesfromWIMPsearchdata.Plottakenfrom[ 39 ]. TheCOUPPexperiment[ 40 ]usesasuperheatedCF3Iliquidbubblechamberheldatclosetoroomtemperature,andimagestheliquidwithhigh-speedcamerassearchingforthecreationofbubbles.Bubblesnucleatefromregionsofionizedliquidandgrowtomacroscopicsizes.Thepowerofthistechniqueisthatbytuningthepressureandtemperature,thethresholdforbubblenucleationcanbeadjusted.Thesethermodynamicparametersaresetsothattherelativelylowionizationdensityofelectronicrecoils 42

PAGE 43

2-4 .Theactualtotalenergyofaninteractioncannotbedetermined,however,andinsteadanintegratedrateisobserved.Thespectrumofrecoilenergiesisprobedbycollectingdatasetswithvariedenergythresholds. Figure2-4. ExamplesofthreeclassesofparticleinteractionsintheCOUPPbubblechamber.Photographscorrespondto(A)cosmicrays,(B)neutrons,and(C)WIMP-likeinteractions.Figuretakenfrom[ 40 ]. 2{6 ).Additionally,nearlyhalfofitsnaturallyoccurringisotopescarryspin,presentingsensitivitytoSDinteractions.DespitetheabilitytorejectelectronicrecoilsinLXe(seeSection 3.1.3 ),backgroundeventsmustnonethelessbeminimized.LXeoersseveralfeaturesthathelptofacilitatethiseort.First,thereexistnolong-livednaturallyoccurringxenonradioisotopes(unlikeArwhichsuersfrom39Ar),andhencetherearenointrinsicbackgroundsourcesattheinteriorofaLXedetector.Additionally,xenonisaformidableabsorberof-raysduetoitshighZ.Givenasucientlylargedetectorvolume,theouterregionsofthedetector 43

PAGE 44

2.4.1MicroscopicProcessesinaParticleTrackArecoilingparticleinLXeleavesbehindatrackofelectricallyneutral,excitedxenonatoms(`excitons')andpositively-chargedionizedxenonatoms(`ions').Theprocessesoccurringaftertheseionsandexcitonsarecreatedarewhatleadtothescintillationandionizationsignalsthatareusedforparticledetection.Figure 2-5 showsthepotentialenergyofelectronicallyexcitedAratomsinthevicinityofground-stateAratoms,asafunctionofseparation.Thoughtwoground-stateargonatomsarestronglyrepulsiveatshortdistances,Ar+ArandAr++Arseepotentialwellsthatformboundstates,calledself-trappedexcitonsandions,respectively.Thisenergyschemeischaracteristicofraregases,includingxenon.Anionizedxenonatomcangothroughaprocessofdimerformationandelectronrecombinationthatleadstoasingly 44

PAGE 45

Potentialenergycurvesofground-stateargoninproximitytoexcitedorionizedargonatoms.Themaincomponentofthescintillationspectrumcomesfromthetransitionlabeled`II'.Figuretakenfrom[ 41 ]. excitedxenonatom[ 42 ]:Xe++Xe!Xe+2Xe+2+e!Xe+XeXe!Xe+heat; 2{11 .Thiscanhappeneitherbecausetheelectronhasbeencarriedawaybythermalmotion,orbecauseithasmigratedawayfromthetrackundertheinuenceofanappliedelectriceld.Thelattercaseleadstoacloudofdriftingelectronsthatcanbereadoutasanionizationsignal.Thepositiveionsthatresultfrom 45

PAGE 46

43 ].ThenalstateXeatomisinthesamestateasthoseneutralatomsinthetrackthatexperienceonlyelectronicexcitation.Theseexcitonsrelaxtoground-stateatomsinasimilarprocess:Xe+Xe!Xe2Xe2!2Xe+h 2{12 correspondstothethetransitioninFigure 2-5 labeled`II',releasing7.0eV.Thiscorrespondstothepeakinthescintillationspectrumof178nm,withawidthof13nm[ 44 ].Forregionsofhighexcitationdensity,itcanbepossiblefortwoexcitonstointeractdirectly,beforebecomingself-trapped,intheprocess, Xe+Xe!Xe+Xe++e:(2{13)Thisprocessconvertstwoexcitonsintoaneutralandsingly-ionizedatom,andactstoquenchtheoverallparticlesignal:thetwoexcitonswhichmightnormallyeachproduceascintillationphotonarenowreplacedbyasingleion,capableofyieldingatmostonephoton.Asthisprocessrequiresexcitonsinteractingdirectly,itisexpectedtoplaysignicantrolesinonlythoserecoiltrackswiththehighestexcitonicdensities:nuclearrecoils,alphaparticles,andssionfragments[ 45 ]. 46

PAGE 47

36 ].InthecaseofWIMPscatters,therelevantinteractionsareXerecoilsinXe;whenthetargetandtheprojectileareidentical,asinthiscase,fnisgivenby[ 46 ], 1+kg(");(2{14)wherekadimensionlessconstantcharacterizingthenuclearsizeandcharge,"isafunctionoftherecoilenergyandZ,andg(")isanalgebraicfunctionof".Thesethreequantitiesaredeterminedempiricallyandcanbefoundin[ 36 ]. 2.4.1 and 2.4.2 ,itisclearthattheconnectionbetweenthetotalenergyofaprojectile,andthenumberofcollectedscintillationphotons,isnotsodirect.Energycanbelostviaelectronsescapingrecombination(Equation 2{11 ),biexcitonicquenching(Equation 2{13 ),andLindhardquenching(Equation 2{14 ).Furthermore,theamountofenergythatislostineachofthesethreeeectsdependsontheidentityoftherecoilingparticle.Becauseofthis,theunitofenergyassignedtoaneventcarriesasuxthatdesignatesthetypeofrecoilingparticle.Theunit`keVee'standsfor`keVelectron-equivalent',meaningthenumberofscintillationphotonsacquiredisequivalenttothenumberthatwouldbeemittedbyarecoilingelectronofthatenergy.Theunit`keVr'indicates`keVnuclearrecoilequivalent',andsimilarlyindicatestheamountofcollectedscintillationlightisequivalenttowhatwouldbeemittedfromarecoilingXenucleusofthatenergy. 47

PAGE 48

45 ],andexperimentallybyvariousgroups[ 47 { 53 ],andisthefocusofChapter 4 48

PAGE 49

3.1.1DetectorDescriptionTheXENON10detectorisa3-Dpositionsensitivedualphase(liquid-gas)xenonTimeProjectionChamber(TPC),seeninFigure 3-1 .Theactivevolumeis20cmindiameterand15cminheight,denedontheperimeterbyahollowpolytetrauoroethylene(PTFE)cylinderandontopandbottombymeshelectrodes.Thecathodemeshelectrodeatthebottomandthegatemeshatthetopdeneadownwardelectriceld,Ed,of0.73kVcm1;5mmabovethegatemeshistheanodemesh,withtheliquidlevellyingbetweenthegateandanode.Afourthmesh,5mmabovetheanode,isheldatthesamepotentialasthegate,andservestopreventanyextractedelectronsfromescapingtheanode.Afterducialcuts,themassofLXeusedfortheWIMPsearchis5.4kg.Thetemperatureiskeptconstantat180K,coolingprovidedbyapulsetuberefrigerator(PTR).Anarrayof47photomultipliertubes(PMTs)viewthevolumefromthetop,inthegas.Asecondarrayof41PMTsviewstheactivevolumefrombelow,lyingbelowthecathodemesh.Followingaparticleinteraction,theexcitonsandrecombiningelectronsproducescintillationlightwithintensofnanoseconds.TheelectronsthatescaperecombinationaredrifteduptotheliquidsurfacebyEdandintothegas,wheretheyproducesecondaryscintillationlightastheycollidewithgaseousxenonatomsduringtheirtransittowardstheanode.Inthiswaybothpromptscintillation(S1)andionization(S2)signalscanbemeasuredsimultaneouslywiththePMTs.ThepositionofaneventisdeterminedbythecharacteristicsoftheS2signal.SeeninFigure 3-2 ,thedriftvelocityofelectronsinliquidxenoniswellknownasafunction 49

PAGE 50

SchematicoftheXENON10detector.TheactiveLXedetectorisdenedontopandbottombymeshelectrodes,andontheperimeterbyaPTFEcylinder.CoolingisprovidedbythePTRatthetop-left. ofappliedeld.ThedelaytimebetweenS1andS2thusgivesthetransittimeoftheelectrons,andthereforethez-positionoftheeventwith1mmresolution.BecausetheS2scintillationlightisemittedinthegasgap,1cmbelowthetopPMTarray,thissignalwillbehighlylocalizedinthePMTsliedirectlyabovetheinteractionsite.Inordertoobtainaprecisereconstructionofthex-yposition,aMonteCarlo(MC)simulationisusedtoestimatetheexpectedPMThitpatterngivenaS2position.Anevenlyspacedgridofpointsinx-yisselected,andforeachpointinthegridthePMThitpatternestimated.ThisMChitmapisthenusedtotrainaneuralnetworkinreconstructingthex-ypositionfromameasuredsignalpatternonthetopPMTarray,withprecisiontowithinafewmillimeters. 50

PAGE 51

Thedriftvelocityofelectronsinxenonasafunctionofappliedelectriceld.Figuretakenfrom[ 54 ]. CalibrationofXENON10isdonewithavarietyofsources,eachforadierentpurpose.TheS1-basedenergyscale,determinedby122keV-raysfrom57Co,isfoundtogiveavolume-averagedlightyieldof3.00.1(sys)0.1(stat)p.e./keVee.Theresponseofthedetector,likealldetectors,variesdependingonthelocationoftheeventvertex.Inordertomeasurethesevariations,131mXewasintroduced,providingaspatially-uniformsourceof164keV-raysandconversionelectrons.Thelow-energyresponsetoelectronicrecoilswasmeasuredwith662keVgammaraysfrom137CsundergoingComptonscatterswithintheactiveLXevolume.Similarly,thelow-energynuclearrecoilresponsewasmeasuredwithmulti-MeVneutronsfromaAmBesource.ThesetwocalibrationsarediscussedinSection 3.1.3 55 ].Anexisting10kmundergroundhighwaytunnelprovidesaccesstothelaboratory,which 51

PAGE 52

3-3 Figure3-3. ThelayoutoftheundergroundLNGSfacility. Thedetectorandcryostatwerelocatedinsideaspeciallydesignedshield,providing20cmofleadand20cmofhigh-densitypolyethylene(HDPE).Theleadshieldactstoattenuateexternalelectromagneticbackgrounds,whiletheHDPEslowsneutrons.ThedetectorcanbeseenintheopenedshieldinFigure 3-4 .Thevisiblepartoftheshieldisthedoor;undernormaloperationthedoorisclosedandthecavityhousingthedetectorisushedwithboil-onitrogengasinordertopurgethevolumeofradon. 52

PAGE 53

TheXENON10detectorseeninsidetheleadandHDPEshield. ionizationdensitythanelectronicrecoils[ 56 ],fewerelectronsescaperecombinationfromrecoilingnucleithanelectronsforagivenenergyandEd.Electronsthatrecombinecontributetothepromptscintillationsignal(S1),whilethoseescapingrecombinationaredriftedtotheanodeinthegasandproducetheproportionalsignal(S2).Therefore,therelativestrengthofrecombinationforagiveneventcanbemeasuredbytheratioS2/S1,andhencethisparametercanbeusedtodiscriminatebetweenrecoilingspecies.Figure 3-5 showsthebehavioroflog10(S2=S1)asafunctionofenergyforpopulationsofbothrecoiltypes,calledtheelectronicandnuclearrecoilbands,orERandNRbands,respectively.Themainpurposeofsuchcalibrationsistoidentifyaregioninlog10(S2=S1){S1space,calledtheWIMPacceptancewindow,whichshouldbenearlyfreeofEReventswhilecoveringasignicantportionoftheNRband.Thelowerboundofthiswindowalongthehorizontalaxisisdeterminedbythedetector'sS1threshold.ThecorrespondingupperboundischosentomaximizethepotentialintegratedWIMPratewhileminimizingtheeectsof`gamma-x'eventswhich 53

PAGE 54

Figure3-5. Theelectronicandnuclearrecoilbandsshowninlog10(S2=S1)versusS1space. ThecalibrationoftheERbandisperformedusinga2Ci137Cssourcethatemitsa662keV-ray,placedoutsidethecryostatandPTFEshieldbutinsidetheleadshield.Theattenuationlengthof662keV-raysinLXeisroughly4.5cm,whichmeansthesephotonsareabletoreachallregionsofther=10cmdetectorgivensucientexposure.DataweretakenwiththissourcethroughoutmostofNovember2006,andintermittentlyfromDecember1throughFebruary142007,accumulatingatotalof2100events(afterqualityandducialcuts)intheWIMPacceptancewindow'sS1range,4.4p.e.S126.4p.e.Fluctuationsinlog10(S2=S1)overmostofthisrangearedominatedbyrecombinationuctuations,untilthelowestenergieswhereuncorrelatedstatisticaluctuationstakeover.ThewidthoftheERbandisveryimportantinregardstonuclearrecoildiscriminationbecauseitpartiallyoverlapswiththeNRband.Duemainlytothenon-uniformS1responseatdierentlocationswithintheactiveregion,performingspatially-dependentcorrectionstoS1basedonthe131mXecalibrationimprovestheoverallS1resolutionandthushelpstoreducethevarianceofthebands(thesuperscript`m'followingtheatomic 54

PAGE 55

3-6 showsthebandsinlog10(S2=S1)space. Figure3-6. ThebandsinFigure 3-5 havebeentransformedtoshowthedistanceinlog10(S2=S1)spacefromtheERbandcenter,givingthenewdiscriminationparameter,log10(S2=S1).TheverticallinesindicatetheWIMPregionofinterest(ROI). AlthoughtheenergydependenceoftheERbandcentroidhasbeenremoved,theNRbandcentroidandwidthstillexhibitenergydependence.Theattenedbandsare 55

PAGE 56

3-7 .TheWIMPacceptancewindowisdenedtolieintherange(3)
PAGE 57

Thenuclearrecoilacceptance,Anr,andtheelectronrecoilrejectioneciency,Rer,foreachofthesevenenergybins.Thepredictednumberofleakageevents,Nleak,isbasedonRerandthenumberofbackgroundevents,Nevt,ineachenergybin,forthe58.6live-daysWIMP-searchdata.ErrorsarethestatisticaluncertaintyfromtheGaussiantsontheelectronrecoillog10(S2=S1)distribution. 0.446 0.8+0:70:4 0.2+0:20:16.7-9.0 0.458 1.7+1:60:9 0.3+0:30:29.0-11.2 0.457 1.1+0:90:5 0.2+0:20:111.2-13.4 0.442 4.1+3:62:0 0.8+0:70:413.4-17.9 0.493 4.2+1:81:3 1.4+0:60:417.9-22.4 0.466 4.3+1:71:2 1.4+0:50:422.4-26.9 0.446 7.2+2:41:9 2.7+0:90:7 1815 7.0+1:41:0 3-1 andFigure 3-8 .Additionally,theexpectednumberofbackgroundeventsintheWIMPacceptancewindow,Nleak,areshown,whicharecalculatedbasedonthepredictedrejectionandbackgroundrateinthe58.6live-daysexposure. Figure3-8. TheERrejectionasafunctionofS1forlog10(S2=S1)<.Therejectionimprovesatlowerenergies,tobetterthan99.9%intherange2{3keVee. TheobservedtrendoftheERrejectionpowerwithenergyisunexpected.Ifrecombinationuctuationswereatatallenergies,orifthebandwidthsweredominatedbybinomialuctuationsfromlightcollection,photoelectronemission,etc.,onewould 57

PAGE 58

3-9 showsadecompositionoftheERbandvarianceintostatisticalandanticorrelatedrecombinationuctuations.Itisquiteevidentthatuncorrelatedstatisticaluctuationscannotaloneaccountfortheobserveddegreeofvariance.Unfortunately,amodeldoesnotyetexistthatsuccessfullypredictsrecombinationuctuationsinliquidnoblegases,andhencemorecannotbesaidonthesubjectbesidesemphasisontheneedforitsfurtherstudy. Figure3-9. DecompositionoftheERbandvariance.Theanticorrelatedrecombinationuctuationsareinferredbycomparingtheexpectedstatisticaluctuationstothefullobservedbandvariance. 3-1 ,aresensitivetothepredictedlevelofelectronicrecoilrejection,Rer.Thequantitativeperformanceofthisrejection,showninthesametable,isinturnbasedupontheassumptionthatthelog(S2=S1)spectrumforelectronicrecoilsisGaussiandistributed.Thisassumptionseemsreasonable,butisdiculttojustifygiventherelativelylowstatisticsofthe137Cscalibration(Figure 3-6 ).Whatisknownisthatthewidthofthelog(S2=S1)bandisdominatedby 58

PAGE 59

3-9 ).Ifthedistributionoftheserecombinationuctuationsareknown,itispossibletosimulatetheshapeofthelog(S2=S1)spectrum.Thefollowingsectionillustratesameasurementofthisdistribution,whichwillthenbeusedastheinputtoaMonteCarlo(MC)simulationtodeterminetheshapeofthelog(S2=S1)band. 57 ].Thelifetimeofthe39.6keVstate,roughly1ns,istooshorttoallowseparateidenticationofthetwotransitions,andtheobservedsignalisinsteadthatofasingle236.2keVevent.Thepreparationofthissourceisdescribedindetailin[ 58 ].Oneadvantageoftheseisomericxenoncalibrationsourcesisthattheydiuseuniformlythroughoutthedetector,andallowacharacterizationofthedetector'sresponseasafunctionofposition.Additionally,themeasurementwithactivatedxenonallowsacalibrationofXENON10'scombinedenergyscale(CES).Thisenergyscale,describedlaterinSection 6.3 ,countsthetotalnumberofquanta,n+ne,andisinsensitivetorecombinationuctuations.TheserecombinationuctuationsleadtoananticorrelationbetweentheS1andS2signals,seeninFigure 3-10 (left).ThecalibrationofS1innumberoftotalphotons(n)andS2innumberofelectrons(ne)isdonebyadjustingtheS1andS2scalinguntilthemajoraxisofthe131mXeellipsehasaslopeof-1.ThisprocedureleavesS1andS2inastatesuchthattheirsumisproportionaltothetotalquanta,andtheabsolutescalingisthendeterminedbyn+ne=E=W,whereEisthedepositedenergy(163.9keV)and 59

PAGE 60

(Left)Theactivatedxenondata,131mXeand129mXewithde-excitationlinesat164keVand236keV,respectively,showninS1versusS2.Thedistinctanticorrelationisduetouctuationsinthefractionofrecombiningelectron-ionpairs.Theblackdashedlineindicatesthemajoraxisofthe131mXeellipse,andhasaslopeof-1.(Right)SpectrafromtheactivatedxenoninS1,S2,andcombinedenergyscale(CES);theimprovementinenergyresolutionofthecombinedscaleisdueitsinsensitivitytorecombinationuctuations. 59 ]istheaverageenergyrequiredtoproduceasinglequantum.ThespectrameasuredfromtheactivatedxenoninS1,S2,andCESareshowninFigure 3-10 (right).TheimprovementinenergyresolutiongainedbytheCESisimmediatelyapparent.Inordertodeterminetheshapeoftherecombinationuctuations,eventsresultingfrom131mXedecaysareexamined.DataareselectedbasedonFigure 3-10 (right),fromanarrow,1 2intervalaroundaboutthe164keVpeak.Suchanarrowrangeischosensothatn+neisaconstantvalue,andthusnenisaveryaccuraterepresentationoftherecombinationuctuations,seeninFigure 3-11 .TheagreementbetweenthehistogramandtheGausstevenouttomanyisconsistentwiththehypothesisthattherecombinationuctuationsareGaussian-distributed.ThisassumptionwillbeusedasaninputtotheMCdescribedinthenextsection. 60

PAGE 61

Thespectrumofrecombinationuctuations,innen,alongwithaGausst,from164keVdecaysof131mXe.Datawereselectedfromanarrow1 2bandaboutthemeanofthepeakintheCESspectrum. 3-12 (left)showsthisband,alongwiththebandt.ThebandtisdonebybreakingupdataintoCESslices,andttingthen=(ne+n)spectrumofeachslicewithaGaussian.Despitetherelativelylowstatisticsfrom137Csdata,theassumptionthatthisbandisGaussian-distributedisjustiedduetotheobservedGaussianityofthe164keVrecombinationuctuations.OncethebandtparametersareobtainedfromFigure 3-12 (left),aGaussian-randomnumbergeneratorisusedtocreateasetofphotonfractionvalueswhosemeanandstandarddeviationmatch 61

PAGE 62

3-12 (left),whenappliedtothepreviously-generatedCESenergyvalues,describedabove. Figure3-12. (Left)Thephotonfraction,whichmeasurestheelectron-ionrecombination,asafunctionofenergy,fromreal137Csdata.Themean(solidredline)andwidth(dashedredlines)areusedasinputtotheMC.(Right)ThecomparisonoftheMClog10(S2=S1)versusS1bandtorealdata. ThephotonfractionforeachenergyvaluethengivesS1andS2innandne,respectively,assumingperfectanticorrelation.Thesevaluesarethenbothconvertedtophotoelectrons,withbinomialuctuationsappliedtosimulatetherealisticlightcollection,quantumeciencyofthephotocathodes,andcollectioneciencies.TheendresultofthisprocessisshowninFigure 3-12 (right),wherethelog10(S2=S1)dataandMCbandsarecompared,basedonequalstatistics.ItisworthemphasizingthatthegoalofthisMCisnotsomuchtoreproducetheenergydependenceoftheband,buttoaccuratelyreproduce,andstudy,theshapeofthelog10(S2=S1)spectrum.TheMCbandfromFigure 3-12 (right)istakenand\attened"bythesametechniqueastherealdatadescribedinSection 3.1.3 ,toproducethequantityofinterest,log10(S2=S1).The10keVeeS112keVeebinischosentocompareMCagainstdata. 3-13 ,thelog10(S2=S1)spectrumisshownfortheactual 2.4.3 62

PAGE 63

Figure3-13. Theequal-statisticspectrumoflog10(S2=S1)forthe10-12keVeebin,forbothrealdataandtheMC.Alsoshownarethemeanand3levels(greenlines). 3-1 )isbaseduponGausststothehistogramsofFigure 3-13 ,thegoalofthisMCistocomparethepredictedrejectionbaseduponaGausst(fromhereonreferredtoasRG)towhattheactualrejectionis,aspredictedbytheMC(fromherereferredtoasRMC).ThebluehistograminFigure 3-14 representsthesamespectrumsimulatedinFigure 3-13 ,butwith107events.ThemagentacurveisaGaussttothebluehistogram,andhighlightsitsdeparturefromGaussianity.OfinterestisthedepartureofthebluehistograminFigure 3-14 fromthemagentacurve,onthelowend(becausethisiswheretheNRbandappears).Interestingly,this 63

PAGE 64

Thehigh-statisticMCspectrumoflog10(S2=S1)forS1intherange10-12keVee.ThedepartureofthebluehistogramfromGaussianity,representedbythemagentacurve,becomesreadilyapparentoutsideof3(reddashedlines). iswherethereexiststhelargestdegreeofdiscrepancy.BystudyingthecurvesinFigure 3-14 ,aconversionisconstructedbetweenRG(magentacurve)andRMC(bluehistogram).Figure 3-15 showstherelationbetweenRGandRMC,coveringtendecadesinRG.ForreferencetheregionofinterestinFigure 3-15 istherange103(1RG)102.Asmentionedbefore,themainpurposeofthisstudyisnottoreproducetheenergydependenceoftheERband,butinmodelingtheshapeofthelog10(S2=S1).TheconversioncurveofFigure 3-15 isappliedtoallenergiesintheWIMPsearchregionofinterest(2-12keVee).Figure 3-16 (left)showsthepreviously-reportedvaluesoftherejectioninblue,andinredarethecorrectedvaluesbasedonthisMC.TheuncertaintiesarebasedontheoriginalGausststotherealdata,andtheseremainthedominantuncertaintyfollowingthecorrections.Thesecorrectionsscaledirectlytothetotalamountofpredictedleakage 64

PAGE 65

(1RMC)versus(1RG)forS1intherange10-12keVee.Atallvalues,(1RMC)>(1RG),withequalrejectionindicatedbytheredline.ThetypeofrejectioncharacterizingtheXENON10WIMPsearchwindowisforrejectionintherange103(1RG)102. Figure3-16. (Left)Theoriginal(blue)andMC-corrected(red)electronicrecoilrejectionintheWIMP-searchenergyrange.(Right)ThecorrectedpredictionsonthenumberofbackgroundelectronicrecoilsleakingintotheWIMPsearchwindow,fromTable 3-1 (Nleak),seeninFigure 3-16 (right).Thepreviously-reportedvalueofNleak(original)=7:0+2:11:0shiftsuptoNleak(corrected)=11:5+2:41:6. 65

PAGE 66

3-15 hasbeentakentoapplyatallenergieswithintheWIMPsearchregion.However,itisnotguaranteedthatthisassumptionisvalid.Forexample,becauseithasbeenassumedthattherecombinationuctuationsareGaussian-distributed,thisalsoentailstheimplicitassumptionthatthedistributionissymmetric.Thequestionthenbecomes,howdosymmetricintervalsinphoton-fractionspacetransformontolog10(S2=S1)space?InFigure 3-17 ,asetof Figure3-17. Themappingofa0.2-widesymmetricintervalfromphoton-fractionspaceontolog10(S2=S1)space.Aclearasymmetryarises,whosepolarityipsoneithersideofthe50%photonfraction,orn=(ne+n)=0:5. symmetricintervalsinphoton-fractionspace,withcentersrangingfrom0.1to0.9withafullwidthof0.2(i.e.0:1),isconsidered.Theblackcurveisthemappingoftheintervals'centersontolog10(S2=S1)space,theblueandredcurvesarethemappingoftheintervals'lowerandupperbounds,respectively.Clearly,symmetricintervalsinphoton-fractiondonotretaintheirsymmetrywhenmappedontolog10(S2=S1).ItisalreadyclearfromanexaminationofFigure 3-14 thatthelog10(S2=S1)spectrumisasymmetric,butwhatisnowevidentisthatsignofthelog10(S2=S1)skewmightnotalwayscomeoutthesameway.Figure 3-17 showsthattheskewofthelog10(S2=S1)intervalchangesoneitherside 66

PAGE 67

3-12 (left)shows,thephotonfractionoftheERbandcrossesthe50%markatroughly6keVee.AnadditionalcomplicationisthatthesevenbinsinS1(fromTable 3-1 )haveboundsatconstantS1,notatconstantenergy.TheMCisconstructedsothatuctuationsinphotonfractionareGaussiandistributedatagivenenergy.ButeachS1binspansarangeofenergy,whilethecurvesinFigure 3-17 applytoonlyasingleenergy.Figure 3-18 Figure3-18. ThesamedataasinFigure 3-12 (left),withlinesofconstantS1overlaid.EachcurvecorrespondstotheboundofanintervalofTable 3-1 showsthephotonfractionwiththeboundsofthesevenWIMPsearchbinssuperimposed.Clearly,theshapeofthebandinlog10(S2=S1)spaceforbinsofconstantS1dependsquitestronglyonthewaythesebinsintersectthebandinFigure 3-18 .Inordertoaddressthisissue,additionalMCsimulationsareconstructedtocovertheelectronicrecoilbandoverthefullrangeoftheWIMPsearchwindow.TheresultsareshowninFigure 3-19 ,forotherWIMPsearchS1energybins.Notshownarethetwolowestbins,2{3keVeeand3{4keVeeastheseexhibitthesamequalitativebehaviorasthespectrainthe4{5keVeeand5{6keVeebins. 67

PAGE 68

Thesamelog10(S2=S1)spectraasinFigure 3-14 ,butforvariousenergyranges.Althoughtheskewofthespectradoesnotremainthesame,thenon-Gaussiantailsonthelowendareaconsistentfeature. ThoughthespectrainFigure 3-19 donotmaintainthesamesymmetry,allhistogramsexhibitthesamecharacteristicnon-Gaussiantailsatlowvaluesoflog10(S2=S1).EachofthesesimulationscanagainbeusedtoformulateaconversionbetweentheoriginalGaussian-predictedrejectionfactors,RG,andthosegivenbytheMC.TheseupdatedrejectionsareturnedintoMC-correctedbackgroundestimates,showninFigure 3-20 .ThenewbackgroundestimatebecomesNleak(corrected)=10:2+2:11:5. 68

PAGE 69

CorrectionstothevaluesofNleakreportedinTable 3-1 basedontheenergy-dependentMCresultsshowninFigure 3-19 unlikelythatanymajorchangesoccurbecausethephotonfractionspectrumofFigure 3-12 (left)appearstobeGaussiandistributedaswell.ThoughtheresultsoftheMCsimulationindicatethattheactualelectronicrecoilrejectionpowerisworsethantheestimatesgiveninTable 3-1 ,theconsequencesareencouraging.WhatthismeansisthatthebackgroundpredictionsbasedontheGaussianrejection,RG,actuallyunderestimatethetruebackground.Thus,anyresultsthatusebackgroundsubtractionbasedonNleakareactuallyconservativeresults. 69

PAGE 70

ThelivetimeofXENON10duringthedurationoffall2006throughwinter2007.Blueandgreenpointsindicatecalibrationdatathathasbeenscaledtoequivalentbackgroundlivedaysbasedonthenumberofacquiredtriggers.FigureprovidedbyL.deViveiros. onDecember1stforanAmBecalibrationtodenethenuclearrecoilband(describedinSection 3.1.3 ).TheprogressionofdatacollectionthroughoutthistimeperiodisshowninFigure 3-21 .Thedetailsofthecutsandunblindingprocedurescanbefoundin[ 31 60 ].Followingunblinding,theS1andS2valuesareusedtoconstructthelog10(S2=S1)bandasinSection 3.1.3 .SeeninFigure 3-22 ,thisprocedureresultsinteneventswithintheWIMPsearchacceptancewindow.Thoughnoneoftheseeventsarelikelytoresultfromnuclearrecoilsscatters,theyareallconsideredindeterminingtheexperimentalupperlimitsdescribedinthefollowingsections.Foradiscussionofthelikelyoriginofeachoftheseevents,see[ 60 ]. 61 ].Thismethodisadvantageousinthatitallowsalimittobesetinthepresenceofbothknownandunknownbackgrounds.Theunknownbackgroundishandledbycomparingnotonlythemeasurednumberofeventstotheexpectednumber,butalsocomparingpredictedandexpecteddistributions.The\gap"betweentwoevents 70

PAGE 71

ThedistributionofblindWIMPsearchdatainlog10(S2=S1)versusS1.Thesignalacceptanceregionisboundedhorizontallybythebluelinesandverticallybythebrownlines.Theteneventsremainingintheacceptancewindowaftercutsareindicatedbytheredcircles. adjacentinenergy,x1;2,isdenedby, dQdQ;(3{1)whereQ1istheenergyofthersteventandQ2istheenergyofthesecond.Thedierentialrate,dR=dQ,includesexpectedsignal(Equation 2{2 )andknownbackground(ifany).Astatisticalparameter,C0,iscalculatedwhichrepresentstheprobabilitythatthemaximumgapfromarandomsamplingofdR=dQwouldbesmallerthantheobservedmaximumgap,andisgivenby, 3{2 reducestoC0(;)=1e,equivalenttotheone-sidedPoissonnullresult. 71

PAGE 72

2{2 ,isafunctionofcrosssectionandWIMPmass.ForagivenWIMPmass,theWIMP-nucleoncrosssectionisvarieduntilC0=0:9,representingthe90%CondenceLevel(C.L.)exclusionlimit.ThedatashowninFigure 3-22 areconvertedtonuclearrecoilequivalentenergybyassumingLe=0:19atallenergies. Figure3-23. XENON1058.6livedaySIWIMP-nucleonexclusionlimits,inred.Thedashedlineiswithbackgroundsubtraction,solidlineiswithout.ResultsfromacombinationofCDMS-II2008datawithare-analysisofCDMS-II2004-2005areshowninblue[ 37 ].TheshadedregionsarefavoredbytwostudiesofMSSMmodels,dark[ 62 ]andlight[ 63 ]. AsdiscussedinSection 3.1.3 ,theexpectednumberofbackgroundeventscanbeestimatedundertheassumptionthattheeventsintheERbandareGaussiandistributedinlog10(S2=S1)space.Thesepredictedbackgrounds,perenergybin,areshowninTable 3-1 .Theresulting90%C.L.exclusioncurves,withandwithoutbackgroundsubtraction,areshowninFigure 3-23 ,alongwiththetheoretically-favoredregionsoftwo 72

PAGE 73

62 63 ].Resultsfromcurrentbestcompetingexperiment,CDMS-II[ 37 ],arealsoshown.AstudyoftheactualdeviationoftheERbandfromGaussianityispresentedinSection 3.2 .TheuncertaintyintroducedbydeparturesofLefrom0.19isdiscussedinChapter 4 .Figure 3-23 canbecomparedwithFigure 1-8 .XENON10,andmorerecentlyCDMS-II,arenowbeginningtoprobetheinterestingregionsoftheMSSMparameterspacerelevanttotheneutralino.TheshadedregionsofFigure 3-23 coverthe95%C.L.regionallowedbytheanalyses;themostfavoredregionsarestilloutsidethesensitivityreachofexistingsearches. 2.2.2 .Inordertoapplytheseformulaetoanactualdetector,vepiecesofinformationmustbeknown.Thersttwoarethespincontentofthenucleus,hSpiandhSni.Theremainingthreeunknownsarethespinstructurefunctions,Sij(Equation 2{10 ).WhereastheSIinteractionstreateveryxenonnucleusintheducialregionasasensitivetarget,SDinteractionscoupleonlywiththosenucleihavingnon-zerospin.Thetwonaturallyoccurringxenonisotopeswithspinare129Xe(J=1 2)and131Xe(J=3 2),existingwithnaturalabundancesof26.44%and21.18%,respectively.Thenuclearstructuresof129Xeand131Xecannotbeconsideredidentical,andthereforemustbetreatedseparately.For129Xe,thereexistintheliteratureaccuratecalculationsbasedontwodierenteectivenucleon-nucleonpotentials,BonnA[ 64 ]andNijmegenII[ 65 ].Theaccuracyofthemodelsisquantiedbytheagreementbetweenpredictedandmeasurednuclearmagneticmoment.ThismetricischosenbecausethematrixelementforWIMP-nucleusscatteringisverysimilartothatofthenuclearmagneticmoment.Thesetwomodelshavealsobeenappliedto131Xe,givingsimilaraccuraciesasinthecaseof129Xe.However,athirdmodelexistsfor131Xebasedonthequasiparticle 73

PAGE 74

Thespinexpectationvaluesforprotonandneutrongroupsbasedonthethreenuclearshellmodelsdiscussedinthetext.Alsoshownarethedeviationofthemodels'predictionsofthenuclearmagneticmomentfromthemeasuredvalue,`{acc.'.ValuesaretakenfromTableIIof[ 67 ]. Nucleus Model hSni BonnA 0.028 0.359 19% NijmegenII 0.0128 0.300 51% BonnA -0.009 -0.227 8% NijmegenII -0.012 -0.217 50% QTDA -0.041 -0.236 1% Tamm-Dancoapproximation(QTDA)[ 66 ].Thismodelyieldsamagneticmomentaccuracytowithin1%ofthemeasuredvalue,andisrecommendedforuseovertheBonnAandNijmegenIImodelsbytheauthorsof[ 67 ].TheresultsofhSpiandhSpicalculationsbasedonthemodelsdescribedherearetabulatedinTable 3-2 ,alongwiththeiraccuraciesintermsofnuclearmagneticmoment.Inordertocapturethelevelofuncertaintyintroducedbythevariousnuclearshellmodels,limitsarecalculatedaccordingtoa`main'model(BonnAfor129Xe,QTDAfor131Xe)andan`alternate'model(NijmegenIIfor129XeandQTDAfor131Xe).QTDAisusedfor131Xeinbothcasesduetoitshighdegreeofaccuracyinthemagneticmoment.ThespinstructurefunctionsarepresentedbyRessellandDean[ 67 ]decomposedasanexponentialmultipliedbyapolynomial,givenby, 67 ]forBonnAandNijmegenII.Aparameterizationofthe131XespinstructurefunctionsbasedontheQTDAmodelisnotfoundin[ 67 ]orevenintheoriginalpaper[ 66 ].BednyakovandSimkovic[ 68 ]haveattemptedtoextractasetofSijvaluesfromFigure3of[ 66 ],however,thesevaluesprovideonlycoarsecoverageintheregionofinterestforXENON10'sdarkmattersearch. 74

PAGE 75

ThepolynomialcoecientsofattotheQTDAspinstructurefunctionsshowninFigure3of[ 66 ].Thefunctionsareparameterizedasafunctionofq2asSij=P5k=0ckq2,withqinunitsofGeVc1. mode 3.4638 -8.9189 1.1992 -8.7745 0.0375 1.6063 -4.4275 0.6139 -4.4670 0.0175 -2.7828 8.7248 -1.3769 11.2749 -0.0498 TheQTDAspinstructurefunctionsareheredeterminedinawaythatismoreappropriateforXENON10.TheSijcurvesinFigure3of[ 66 ]arecopiedintotheGraphClicksoftware[ 69 ],whereasetofpointsisextractedfromeachcurve.Thesepointsarethentwithafth-orderpolynomialinthelowenergyregion,showninFigure 3-24 ,withpolynomialcoecientstabulatedinTable 3-3 .Thetsarevalidforvaluesofq2.0:015GeV2c2. Figure3-24. TheQTDAspinstructurefunctionsfor131Xe.OpencirclesaretakenfromFigure3of[ 66 ],solidlinesarethepolynomialtsshowninTable 3-3 .TheenergyrangeusedfortheWIMPsearchofXENON10isindicatedbytheshadedyellowregion. TheexclusionlimitsforSDcouplinghavebeennormalizedtopureprotonandpureneutroncouplingsasdenedinEquation 2{9 .ResultsareshowninFigure 3-25 forthe 75

PAGE 76

3{2 ).Asbothofthexenonisotopesdiscussedherehaveanunpairedneutron,mostofthenuclearspiniscarriedbytheneutrongroup(Table 3-2 ).Asaresult,theXENON10exclusionlimitonpureneutroncouplingissignicantlymoreconstrainingthanthatforpureprotoncoupling. Figure3-25. TheXENON10SDexclusionlimitsnormalizedtopureproton(left)andpureneutron(right)formain(solidred)andalternate(dashedred)spinformfactors.Theresultsthebestcompetingdirectdetectionexperimentsineachcategoryareshownforcomparison:COUPP{darkblue[ 40 ];KIMS{black[ 39 ],CDMS-II{lightblue[ 37 ].Theshadedareaisthetheoretical95%probabilityregionfromoneanalysisofCMSSM[ 63 ]. Thedecisiontoholdthe131Xemodelxedforbothmainandalternateshellmodelsisjustiedbecausethevariationintheexclusionlimitisdominatedbythe129Xemodel.TheWIMP-neutronexclusionlimitisshowninFigure 3-26 (left)forfoursetsofshellmodelchoices.Itisclearthatthechosen131Xeshellmodelhasonlyaverysmalleectontheresultingexclusionlimit.AnalternativewayofinterpretingtheXENON10resultsistoconstraintheSDWIMP-nucleoncouplingsthemselves,apandan.FromEquation 2{8 ,itisclearthatdR=dQ/[aphSpi+anhSni]2.Therefore,foragivennucleus,anypairofvaluesofapandanthatliealongtheline, hSpi;(3{4) 76

PAGE 77

3-26 (right)foraWIMPmassof50GeVc2,avalueofC0(Equation 3{2 )isassignedtoeachpointintheparameterspace.The90%C.L.exclusionisthengivenbythecontourdenedbyC0=0:9.Alsoshownarethe129Xeand131Xeaxesofnullcrosssection(Equation 3{4 ). Figure3-26. (Left)TheWIMP-neutronexclusionlimitcalculatedforfourdierentcombinationsof129Xeand131Xeshellmodels.Itisclearthattheshellmodelsfor129Xeproducethegreatestvariationintheresultingexclusionlimit.(Right)TheC0mapfor50GeVc2WIMPs,alongwiththecorrespondingcontourthatexcludestheexteriorparameterspaceatthe90%condencelevel.Thedashedlinesindicatethe129Xeand131Xeaxesofnullcrosssection. 1.3.1 coveredthetopicofthecosmologicalabundanceofrelicneutrinos.Ifoneignoresargumentsrelatedtotheformationoflargescalestructure,Equations 1{11 and 1{12 alonerequirethattheheaviestneutrinospeciesmusthaveamasslessthan10eVcm2sothattheirdensitydoesnotconictwithmeasurementsofm.ItisalreadyknownthatnoneoftheStandardModelneutrinosevencomeclosetoexceedingthismass,butitcouldbepossiblethatmoreneutrinosexist,possiblypartofafourthgenerationoffermions.Equation 1{12 appliesonlytoneutrinosthatfreeze-outrelativistically,but 77

PAGE 78

70 ].Therelicdensityofaheavyneutrinowouldbetoosmalltoaccountforthedarkmatterunderthestandardfreeze-outscenario[ 71 ].However,givenadynamicallyevolvingdarkenergydensitypriortoBBN,itcouldbepossibleforheavyneutrinostobeproducedwithanabundancelargeenoughtoaccountform[ 72 ].AheavyDiracneutrinowouldinteractwithnormalmatterviaSIinteractions,buthaslongsincebeenruledoutasapossibledarkmattercandidatebypreviousdirectdetectionexperiments[ 73 ].Incontrast,aheavyMajorananeutrinointeractsonlyviaSDinteractions,anditselasticscattercrosssectionwithnucleiisgivenby[ 74 75 ], 2{8 exceptfortheprefactor.SuchaheavyMajorananeutrinowithmassintherange100{500GeVc2hasbeenpredictedinminimal[ 72 ]andwalking[ 76 ]technicolortheories.ThesemodelsprovideamechanismforelectroweaksymmetrybreakingthatisalternativetotheHiggsmechanism,andposittheexistenceofnewgaugeinteractionswithnonStandard-Modelfermions.UnliketheSDWIMP-nucleuscrosssection,hereapandanaregivenfromparticlephysicsexperiments.Thecouplingsaretakentobeap=0:68andan=0:58[ 36 77 78 ]. 3.3.2 ,thecrosssectioninEquation 3{5 dependsonlyontheneutrino-nucleusreducedmass,mr.ThiscrosssectionisthenusedtondC0asa 36 ]diersbyafactoroffour,butthisappearstobeamistake.3 36 ],basedon[ 77 ],aregivenonlyasa2p;nandhencedonotpreservethesignofthecoupling.Theauthorsprovide[ 78 ]asawebsupplementwiththefullvalues. 78

PAGE 79

3-27 .TheC0versusmass Figure3-27. ThecondencelevelofexclusionoftheheavyMajorananeutrino,givenbytheMaximumGapparameter,C0.TheMajorananeutrinoisthereforeexcludedatthe90%C.L.wherethecurveisgreaterthanC0=0:9,indicatedbythathorizontalblackdottedline.MajorananeutrinoswithmasslessthanhalftheZbosonhavebeenexcludedbytheLargeElectron-Positroncollider[ 79 ],indicatedbytheverticaldashedline. curvecrossesC0=0:9at9.4GeVc2and2.2TeVc2usingthemainshellmodel,and9.6GeVc2and1.8TeVc2usingthealternateshellmodel.Heavyfourth-generationneutrinoswithamasslessthanhalftheZbosonmass(45.6GeVc2)havealreadybeenexcludedattheLargeElectron-Positroncollider(LEP)[ 79 ],indicatedbytheverticaldashedlineinFigure 3-27 .AlthoughMajorananeutrinoswithM;Maj>2:2TeVarenotexcludedbycosmologicalconstraints[ 71 ],technicolorheavyneutrinosareunlikelytohaveamassgreaterthan500GeVc2[ 72 ],andthereforethelowerlimitonM;MajgivenbyXENON10andLEPeectivelyrulesouttheseparticlesasasignicantcontributortom.ItisworthemphasizingthattheconstraintsshownhereapplyonlytotheheavyMajorananeutrinoasadarkmattercandidate.Ifnoneofthespecialpre-BBNdynamical 79

PAGE 80

80

PAGE 81

2 .BecausetheWIMPrecoilspectruminLXeisexpectedtobesteeplyfallingwithenergy,ascomparedwithlightertargetnuclei(seeChapter 2 ),theunderstandingofthenuclearrecoilenergyscalestronglyaectstheconclusionsdrawnfromdarkmattersearchesthatuseLXe. Figure4-1. AsurveyofLemeasurementsintheliteraturepriorto2009.Bluetrianglesarefrom[ 49 ],greensquaresfrom[ 48 ],withtheremainingmeasurementscitedlaterinthischapter.ThepurpleandredverticallinescorrespondtotheenergyrangesusedbyZeplin-II[ 80 ]andXENON10[ 31 ],respectively.ThebeigeshadedareaisusedasanestimateoftheuncertaintyofLeinXENON10'sresults. Figure 4-1 showsthemeasurementsofLeintheliteraturepriorto2009.TheZeplin-IIdarkmattersearch[ 80 ]operatedinanenergyregimethathasbeenwell-studied, 81

PAGE 82

49 ]andChepel[ 48 ]appeartoindicateopposingtrendswithdecreasingenergy.WhilethechoiceinXENON10touseaatLe=0.19isfairlywelljustied,giventheexistinghigh-energymeasurements,thereisclearlyalargeuncertaintyintroduced,indicatedbythebeigeshadedareainFigure 4-1 .Theuncertaintyinthelow-energybehaviorofLecanbepropagatedthroughtothenalresultsofXENON10,indicatedbythebeigeregioninFigure 4-2 ,andrepresentsXENON10'slargestsystematicuncertainty.ItbecomesclearthatanimprovedunderstandingofLe'slow-energybehaviorisnecessary,requiringnewmeasurements. Figure4-2. TheXENON10upperlimitonthespin-independentWIMP-nucleoncrosssection.Thebeigeareaindicatesthelimit'suncertaintycorrespondingtothebeigeregioninFigure 4-1 82

PAGE 83

Schematicdiagramoftheexperimentalsetup.Incoming1MeVneutronsscatterintheLXeandaretaggedbytheEJ301organicscintillatoratanglesof48,62,70.5,and109.5.TheparanandleadareusedtoshieldtheEJ301fromdirectneutronsandgammarays. monoenergeticneutronsareincidentuponaLXetarget,someofwhichscatterunderanangleandarecollectedwithanEJ301organicscintillator(seeFig. 4-3 ),capableofdistinguishingelectronic(gammarays)fromnuclear(neutron)recoilsviaPulseShapeDiscrimination(PSD)[ 81 82 ].EJ301isaproprietaryname;thescintillatormaterialisalsoknownbytheproprietarynamesBC501AandNE213.Inthisway,theenergyoftherecoilingxenonnucleusisknownkinematically,andisgivenbytherelation 49 ].Inthepresentwork,1.9MeVprotonsareincidentonatritiumtarget,yielding1MeVneutronsintheT(p,n)3Hereaction.Thisreactionproduces 83

PAGE 84

83 ].Theterminalvoltageoftheprotonaccelerator(andhencetheincidentprotonenergy,Ep)isknowntowithin0.1%.Thesetwosystematicuncertainties,comingfromtheangulardependenceofEnandtheuncertaintyinEp,areconsiderednegligibleandarenotincludedinthecalculationsofsection 4.3.2 .ThedominantspreadintheincidentneutronenergycomesfromthethicknessoftheTiT2target,inwhichtheprotonscanloseupto260keVbeforeproducingneutrons[ 84 ];thistranslatestoa1-spreadof7.8%inEn.AlsoseeninFigure 4-3 ,a30cm-thickparanblockisplacedalongthelineofsightbetweenthetritiumtargetandtheEJ301scintillator,inordertoblockneutronsfromdirectlyinteractingintheEJ301.Inadditiontotheparanblock,5cmofPbshieldtheEJ301fromgammasproducedintheT(p,n)3Hereaction. 4-4 .TheLXevolumeisviewedbysix1in2HamamatsumetalchannelR8520-06-Alphotomultipliertubes(PMTs),fourofwhichuseanewbialkaliphotocathodethatyieldsquantumecienciesto178nmlightaround40%atroomtemperature[ 85 ].ThePMTs,heldtogetherwithapolytetrauoroethylene(PTFE)frame,formacubesuchthateachPMTwindowcoversafaceofthecube.BoththephotocathodeandmetalbodyofthePMTsareheldatgroundpotential,withpositivehighvoltageappliedtotheanodes.ThiscongurationguaranteesthatnoresidualelectriceldsexistedintheLXe,whosescintillationyieldcanbestronglydependentontheappliedeld[ 56 ](bydenition,Leistherelativelightyieldatzeroeld).Thexenoniscooledandliqueedbyacopperringcoldngerwhichisthermallycoupledtoaliquidnitrogenbath,andthexenonliquidlevelkeptabovethetopPMT. 84

PAGE 85

SchematicdiagramoftheLXechamberusedfortheLemeasurement.VisiblearefourofthesixPMTsusedtoviewthe1in3LXevolume.Coolingisachievedbythecoppercoldngerabove;temperatureandpressureareregulatedbyheaters(notshown)placedonthestainlesssteelvessel. Thetemperatureisheldconstantat180K(sameasinXENON10[ 31 ]),withuctuationsvaryingbylessthan0.03%.Theentiredetectorassemblyiscontainedinastainlesssteelvacuumvessel,surroundedbyberglassforthermalinsulationfromtheoutsideworld.Followingassemblyandxenonliquefaction,thedetectorismovedintothebeamroom.TheEJ301scintillatoriscontainedinanaluminumcylinder3"indiameterand3"tall,heldatroomtemperature.TheliquidisviewedbyasinglePhotonisXP4312BPMTandreadoutwiththesameelectronicsasthePMTsintheLXechamber. 85

PAGE 86

86 ],withneutronscatteringcrosssectionstakenfromtheJEFF-3.1databases.Theresultsofthesimulationsarediscussedfurtherinsection 4.3.2 SchematicdiagramofthedataacquisitionsystemusedwiththeXecubedetector.ThesixchannelsfromtheLXeareaddedintothreechannelsoftwoforthetriggeringsystem,requiringcoincidencebetweenthesethreechannelsandtheneutronscintillator(EJ301). AschematicdiagramofthetriggeringanddataacquisitionsystemisseeninFig. 4-5 .TheanalogPMTsignalsarefedintoaPhillips776amplierwithagainof10,withtwoidenticaloutputsperchannel.OneoutputisdigitizedbyaCAEN8-channelV1724100MHzashADC,whiletheotheroutputisfedtothetriggeringsystem.FortheLXetrigger,thesixLXePMTchannelsarecombinedinpairswithFANmodules,toproducethreetriggeringchannels,connectedtodiscriminatorssettotriggeratthesinglephotoelectron(p.e.)level.Thelogicaloutputsofthethreediscriminator 86

PAGE 87

TheeciencyoftheLXetriggercondition,basedonaMonteCarlosimulation.Thetriggerrequiresthatallthreepairsoftriggerchannelsreceiveatleastonephotoelectroneach.ThoughthePMTssignalsarecombinedintopairsforthetrigger,thechannelsaredigitizedindividually. channelsarepassedtoanN=3coincidenceunit.Thus,theLXetriggerconditionissimilartoasimpleN3p.e.requirement,butwiththeaddedstipulationthattheNp.e.mustbedistributedtocertainPMTs(i.e.ifasinglep.e.isreceivedinPMTs1,4,and6,thiscanproduceatrigger;asinglep.e.receivedinPMTs1,2,and6cannotbecause1and2arecombinedinthesametriggerchannel).TheeciencyofthistriggerconditionisdeterminedbytheMonteCarlomethod.SeeninFig. 4-6 ,itindicates100%eciencyat1keVee,slowlyrollingoto90%at0.5keVee.TheEJ301triggeristakensimplyastheoutputofthediscriminator.Forthemeasurementoftheneutrons'TimeofFlight(ToF)theLXetriggerisfeddirectlytothe\start"inputofaTime-to-AmplitudeConverter(TAC),Ortec556,whiletheEJ301triggerprovidesthe\stop"afterappropriatedelay.TheoutputoftheTACisdigitizedbythesameCAENunit.CalibrationoftheToFsignalisdiscussedfurtherinsection 4.3.1 .TheshapeofthesignalintheEJ301dependsontheincomingparticlespecies,andcanbeusedtodistinguishneutronsfromgammarayssincethecharacteristicscintillationdecaytimeisdierentfortheseparticles.Thiscanbeexplainedbythepresenceof 87

PAGE 88

81 ].InEJ301,the\slow"componentistwoordersofmagnitudelongerthanthe\fast"component,reportedtobe3.2ns[ 87 ].APSDparameterisconstructedbydividingtheareaofthepulse'stailbythetotalareaofthepulse,withthetaildenedasthepartofthetracestarting30nsafterthepeakuntilthetracereaches5%ofthepeakvalue. 4.3.1CalibrationsThePMTsarecalibratedinsituwithapulsedblueLED,inordertomeasurethegain.ThelightfromtheLEDproducesasinglep.e.spectrum,whosemeandeterminesthegainofthemultiplierchain.WithacompletesetofsuchLEDcalibrationmeasurements,thesignalsobtainedforallacquisitionscanbeconvertedtoavalueinnumberofp.e.Therelationshipbetweenthenumberofcollectedp.e.andthetotalnumberofemittedphotonsdependsonthegeometricallightcollectioneciency,thequantumeciencyofthephotocathodes,andthecollectioneciencybetweenthephotocathodeandtherstdynode.Althoughthesevaluesarenotknowntohighprecision,theyrepresentcompletelylinearprocessesandhenceleadtoalinearrelationshipbetweenthetotalnumberofscintillationphotonsandthemeasurednumberofp.e.Comparingthep.e.yieldsofvarioussourcesthusgivesameasureoftheirrelativescintillationyields.AsLeisdenedagainstthescintillationyieldof122keVgammarays,datafroma100Ci57Cosourcearetakenperiodicallyduringtheexperiment.Fig. 4-7 showsthespectrumfromonesuchcalibration.The57Coyieldismeasuredtobe19.640.07(stat)0.11(sys)p.e./keVee,wherethestatisticaluncertaintyisthecombinationofthe 88

PAGE 89

Thescintillationlightspectrumof122keVgammaraysfrom57Co,usedtocalibratetheelectronicrecoilenergyscale.Thiscalibrationgivesascintillationyieldof19.64p.e./keV. parameteruncertaintiesofthetsfromthevariouscalibrationdata,andthesystematicuncertaintyistakenfromthevariationinthisyieldoverthetwo-daydurationoftheexperiment.OnesetofPMTgainvaluesisappliedtoalldata,andthusthesystematicuncertaintyinthe57CoyieldquotedaboveaccountsforbothvariationsinyieldandPMTgain.Inadditionto57Co,datawerealsocollectedfroma22Nasource.Thissourceemitsa+thatpromptlylosesenergyintheNaandannihilates,producingtwo511keVgammaraysemittedsimultaneouslyinoppositedirections.WiththesourceplacedbetweentheLXedetectorandtheEJ301detector,thetwogammarayswillinteractatessentiallythesametimeinthetwodetectors.Inthisway,22NaprovidesabaselineToF=0which,whenusedinconjunctionwithavariabledelaygenerator,isusedtocalibratetheToFmeasurementsystem. 89

PAGE 90

4-8 showsthedistributionofeventsinPSDparameterandToF.Clearlyvisiblearethenuclearrecoilandelectronicrecoilbands,inadditiontothepeaksfrombothgammaandneutronscatters.ThePSDcutischosentoacceptamajorityofthenuclearrecoilbandwhilerejectingelectronicrecoils.ThewidthoftheToFcutis10ns,whichistheexpectationbasedonthespreadinEnandthenitesizeofthedetectors.ThetailoftheToFpeakisduemainlytoeventswheretheneutronscatteredinoneofthedetectormaterialsinadditiontotheLXe,beforeinteractingintheEJ301scintillator.MultiplescattersintheLXealsoaddtothetail,althoughM.C.simulationsindicatethattheiroverallcontributionislessthan2%. Figure4-8. ThedistributionoftriggeredeventsinPSDvs.ToFspacefromthedatasetat70.5.An\upper"bandand\lower"bandarereadilyidentiableinthedata,andcorrespondtonuclearrecoilsandelectronicrecoils,respectively.ThepeakatthelowerleftnearToF=0,duetogammaraysthatComptonscatterintheLXebeforestrikingtheEJ301,iseasilyvetoedbythePSDcut.Apopulationofaccidentaltriggers(seetext)havingaatToFspectrumisvisibleinbothbandsandcontributesbackgroundeventswithintheneutronpeak.TheLXespectraofeventswithintheleftboxareusedastheexpectationofthisbackground.Thewidthoftherightbox|10ns|ischosentoacceptneutronsthatinteractinanyregionofthenitely-sizeddetectors. 90

PAGE 91

SelectedresultsoftheMonteCarlosimulations,whichdonotincludetheaccidentalsbackground.(a){Thespectrumofeventstaggedat70.5,scaledwiththemeasuredvalueofLegivingtheelectron-equivalentenergy(keVee),convolutedassumingPoissonstatisticsforthenumberofp.e.andmultipliedbythesimulatedtriggereciencycurve.Thegreenhistogramisthetotalspectrum,andtheblackcirclesindicatethetruematerialsbackground.Thereddashedlineisanexponentialttothehigh-energyregionofthegreenhistogram;itsagreementwiththetruematerialsbackgroundconrmsthevalidityofthistechnique'suseintherealdata.Theshadedblueareashowsthespectrumoftrueelasticsinglescatteredneutrons.(b){Thespectrumofeventstaggedat109.5.Thedataareshownintheoriginal,recoilequivalentenergyscale(keVr)withoutPoissonconvolution.Thematerialsbackgroundinthisregiondepartsfromtheexponentialbehaviorseenatlowerenergies,anddistortsthepositionofthepeakfromtruesinglescatters,at20keV.Thereddashedlinesaretheresultofanexponential+Gaussiant.TheGaussiancomponent,centeredat22:944:34keV,isusedasthe`true'energyoftheGaussiancomponentintherealspectrum. TwobackgroundscontributetotheLXespectrumwhichcannotbevetoedwiththecutsdescribedabove,andmustinsteadbesubtracted.ItisclearfromFig. 4-8 thatbeneaththeneutronpeakliesapopulationofeventswhichhaveaatToFspectrum.TheseareidentiedasneutronsthataccidentallyinteractintheEJ301incoincidencewithanunrelatedeventintheLXe,andarereferredtoasaccidentals.AstheseeventsareuniforminToFspace,accidentalsoutsideoftheToFpeakshouldhavethesameenergyspectrumasthosewithinthepeak.TheLXespectrumoftheeventsinsidetheboxof 91

PAGE 92

4-8 labeled\accidentals"isusedastheexpectationoftheaccidentalsbackground.Theregiontotheleftofthepeakischosenbecausethepeak'sextendedtailcontaminatestheaccidentalsspectrumtotherightofthe\neutron"peak.ThesecondbackgroundthatcannotbevetoedcomesfromneutronsthatscatterinvariousdetectormaterialsinadditiontotheLXe,beforeinteractingintheEJ301.Herereferredtoasmaterialsbackground,MCsimulationsshowthatthespectrumoftheseeventsintheLXefollowsanapproximatelyexponentialdistributionintheregionofthepeak.Fig. 4-9 (a)displaystheresultsoftheMCsimulationofthedatasetat70.5,indicatingthecontributionfromthematerialsbackground.Inordertoestimatethespectrumoftheseeventsintherealdata,adecayingexponentialwasttothehighenergyportionofthedistributionsaftersubtractingtheaccidentalsbackground.Afterapplyingcuts(PSDandToF)andsubtractingbackgrounds(accidentalsandmaterials),aspectrumresultsinwhichthepeakfromsingle-scatterneutronscanbereadilyidentied,seenasthesolidcirclesinFig. 4-10 .Thehorizontalscaleofthesespectraisgivenas\keVee"meaning\keVelectron-equivalent",indicatingitistheenergyscalederivedfromthe57Cocalibration.Leisfoundfromthefollowingrelation: 4-1 .TheuncertaintiesinLearecalculatedbyconsideringthespreadinErmentionedabove,statisticalerrorsintheGaussiants,thevariation 92

PAGE 93

ThespectraofeventsintheLXeforthefouranglesusedinthisstudy:(a)-48;(b)-62;(c)-70.5;(d)-109.5.Inallfourplots,theblackdot-dashedlineistheoriginalspectrum,blackdashedlineisthespectrumofaccidentals(seeFig. 4-8 ),greenlineisthespectrumaftersubtractingtheaccidentalsbackground,shaded-grayregionistheexponentialttothetailofthegreenspectrumandusedastheexpectationofthematerialsbackground,andthebluedotsarethespectraaftersubtractingbothbackgrounds.Errorbarsonthebluedotsarethecombinederrorsoftheoriginal,accidentals,andmaterialsbackground(thegrayareacoversthe1-regionofthetparameters),andareincludedintheGaussianttothebluedots,indicatedbythesolidbluecurves. in57Colightyield,theuncertaintyinthebackgroundestimations,andtheeectofthetriggerthresholdroll-o.Thislastuncertaintywascalculatedbyndingthepeakpositionsbeforeandafterdividingthespectrabythetriggereciencydiscussedinsection 4.2.3 .However,onlythelowestangle(48)isaectedbythistriggerroll-o.Theasymmetricerrorbarofthe5keVdatapointisduetoboththetriggerrolloandthe 93

PAGE 94

Table4-1. ThevaluesofLeobtainedatthefouranglesusedinthisstudy.ErrorbarsontherecoilenergiesarethespreadofEnasmentionedinsection 4.2.1 combinedwiththegeometricaluncertainties.TheuncertaintiesinLearethecombinationofallstatisticalandsystematicerrorsmentionedinthetext. ThoughthepurposeofthisstudyistoinvestigatethebehaviorofLebelow10keV,itisnecessarytocollectdatafromhigher-energyrecoilsinordertoestablishaconnectionwithpreviousstudies.Forthis,theEJ301isplacedatascatteringangleof109.5,correspondingto20.0keVrecoils.However,thisangleisclosetotheminimuminthedierentialscatteringcrosssectionof1MeVneutronsinXe[ 88 ],andsothesignalfrom\true"singlescattersiswellbelowthebackground.Additionally,thematerialsbackgroundinthisenergyrangedepartsfromadecayingexponential.AscanbeseenintheMCdataofFig. 4-9 (b),theactual\bump"inthespectrum,comingprimarilyfromneutronswhichhavealsoscatteredinthePTFE,isactuallyslightlyhigherthan20keV.Inordertondthetrueenergyofthepeakposition,thesameprocedureusedinexaminingtherealdatawasappliedheretotheMCdata,givingarecoilenergyof22:944:34keV.ThespreadinEristakenasthewidthoftheGaussiancomponentintheMCspectrum.ThevaluesobtainedforLe[ 47 ]arelistedinTable 4-1 ,andadditionallyshowninFig. 4-11 alongwiththeresultsofpreviousstudies[ 48 { 52 ].Shownaswellisthe1-allowedregionofthebest-tproceduredescribedinsection 4.4 94

PAGE 95

MeasuredLevaluesasafunctionofXenuclearrecoilenergy.Symbolscorrespondto( 47 ];( 48 ];( 49 ];( 50 ];(){Bernabeietal.[ 51 ];( 52 ].Thesolidgraycurveistheresultfromabest-tanalysisofXENON10AmBesourcedataandMC[ 53 ].AlsoshownisthetheoreticalpredictionofHitachi(dashedline)[ 45 ].Theshaded-blueregionistheresultoftheXecubebesttbetweenAmBesourceandMonteCarlo. emitsneutronsviathe(,n)reaction.TheAmBebranchingratioforneutronemissionis6105[ 89 ],giving4106neutrons/s.TheGeant4MonteCarlo(M.C.)packageprovidesonlyenergydepositionandparticletrackinginformation,anddoesnotsimulatescintillationmechanismsindetectormaterials.Hence,oneextractstheabsoluteenergiesfromparticlehits,regardlessoftheinteractiontype(i.e.electronicversusnuclearrecoils).Therefore,inordertocompareaspectrumfromanAmBesimulationtothatfromrealdata,thesimulatedhitenergiesmustbescaledrstbyLe.If,however,Leisconsideredaparameter,itispossibletoestimate 95

PAGE 96

53 ],theresultsofwhichareincludedinFigure 4-11 .ArigorousmeasurementofLewiththistechnique(asdonebySorensen)isdicultandextremelytimeconsuming,fortworeasons.First,oneshouldworkwithasucientnumberofsplineknots,coveringtheentireenergyrangeofthespectrum,soastocaptureallofLe'senergy-dependentfeatures.However,thetaskofextractingabest-tincreasesincomplexitydramaticallyasthenumberfreeparametersareincreased;themulti-parameter2spacecontainsmanylocalminimaandhencethethetisverysensitivetotheparameterstartingpointsthatarechosen.Calculationtimeforamany-parametergradientdescentcanalsobenon-trivial.Thesecondproblem,andperhapsthemosttimeconsuming,istoconstructanaccurateestimateofthesystematicuncertainties.ThistaskinvolvestrackingdownthemeasurementsusedfortheGeant4Xe(n,n)Xecrosssectiondatabasestondthetotaluncertaintiesinthosestudies.OnemustthenvarytheXe(n,n)Xecrosssectiondatabasevaluesaccordingtothoseuncertaintiesmanytimes,eachtimere-runningtheM.C.simulationandperformingadditionalbest-ts.AdditionalsystematicuncertaintiesariseduetodiscrepanciesbetweentherealdetectorgeometryandthatwhichhasbeencodedintoGeant4. 96

PAGE 97

RealandsimulatedspectraofelasticneutronscattersfromAmBeintheXecubedetector.Thedashedlinesarethefullspectra,whilethesolidlinesindicatethepartofthespectrausedinthettingprocedure.(Top)ThetwospectraaftervaryingtheLesplineknotstoformabest-t(2/ndf=1.1).(Bottom)Thespectrashownafterscalingthesimulateddatabyanenergy-independentLe=0.19(2/ndf=26.3),asusedinXENON10. Despitethesediculties,abest-tresultwithouttherigordescribedinthepreviousparagraphcanstillbeusefulasaconsistencycheckofthecoincidentbeamdata.Figure 4-12 comparestherealandsimulatedAmBespectra,usingtheatLe=0.19asinXENON10[ 31 ](giving2/ndf=26.3),andaswellthespectraaftervaryingtheLesplinetoobtainabest-t(giving2/ndf=1.1).Thefourxedsplineknotsarelocatedat4,10,15,and22keVr,andthetisperformedbycomparingthehistogramsinFigure 4-12 intherange1-8keVee.AfterscalingtherawM.C.databyLe,thespectrumisthenconvolutedwiththedetector'senergyresolution,andnormalizedtomatchthetotalnumberofeventsastherealdatainthetenergyrange.Theelectron-equivalentenergyresolution(=)isassumedtobeproportionalto1=E2,withtheconstantofproportionalitytakenfrom57Co's122keVline(9.0%).Ateachiteration,2iscomputed, 97

PAGE 98

4-11 48 ],withaconsiderableimprovementinprecision.Thecentralvalueat10keVisconsistentwiththelowest-energydatapointofAprileetal.[ 49 ],enforcingtheaccuracyofthismeasurement.Unfortunately,thetheoreticalmodelsofneitherLindhard[ 46 ]norHitachi[ 45 ]canshedanylightonthebehaviorofLeinthisenergyrange.Hitachi'smodel,whichattemptstotakeintoaccountincompletechargerecombinationandadditionalelectronicquenching,isbasedonLindhardquenchingaswellastheThomas-Fermiapproximation;forXenuclearrecoils,bothbreakdownbelow10keV[ 90 91 ].Asmentionedintheintroduction,theuncertaintyinLeatlowrecoilenergiespresentsthelargestsystematicuncertaintyintheresultsoftheXENON10darkmatterexperiment,whereitwaschosentouseaatLe=0.19asacompromisebetweentheseeminglyopposingtrendsobservedbyChepelandAprile.Underthisassumption,theWIMP-nucleonspin-independentcrosssectionforWIMPsofmass100GeV/c2and30GeV/c2wasconstrainedtobelessthan8:81044cm2and4:51044cm2,respectively,indicatedbythesolidcurveinFig 4-13 .AllowingforLescenariosbelow20keVthatcoverthevaluesallowedbybothChepelandAprilegivesupperlimitsthatvaryby40%at30GeV/c2and18%at100GeV/c2,withvariationsbecominglessseverewithincreasingWIMPmass.WithanLemodelthatfollowsthenewdatapointsofthisstudy,theresultingupperlimitisshowninFig 4-13 asthedashedcurve.Thelimitis 98

PAGE 99

Figure4-13. TheupperlimitontheWIMP-nucleonspin-independentcrosssectionbasedonthe58.6livedaysofXENON10'sWIMPsearch,shownwithaatLe=0:19(solid).AnLefunctionconsistentwiththeresultsofthisstudy,appliedtothesameXENON10dataisshownaswell(dashed). IthasbecomeclearfromXENON10thatfuturedarkmattersearchesusingLXemusthavesensitivitytonuclearrecoilsbelow10keVinordertobecompetitive.TheimprovedunderstandingofLe'sbehaviorpresentedinthisstudynotonlypermitsamorepreciseinterpretationofXENON10'sresults,butbenetsfuturedarkmattersearchesalsousingLXe.SeveralnextgenerationLXedarkmattersearchesarecurrentlyinoperationorunderconstruction,suchasXENON100[ 92 ],LUX[ 93 ]andXMASS[ 94 ].Theseexperimentswillbegintoprobeforthersttimethoseregionsofparameterspacemostfavoredbymanytheoreticalmodels,andwillconsequentlyrelyquiteheavilyona 99

PAGE 100

100

PAGE 101

5-1 .Thestainlesssteel(SS)vesselishousedwithinavacuumcryostatwithcoolingprovidedviaacoppercoldngerimmersedinliquidnitrogen(seeSection 5.2.2 ).Thetemperatureandpressureareheldconstantat175Kand1.8bar(absolute),respectively,andthedetectoroperatedstablyforseveralmonthsatatime.Xurich'scylindricalactiveregion,3.5cmindiameterand3cminheight(80.8gofLXe),isdenedbyapolytetrauoroethylene(PTFE)cylinderontheperimeterandgridelectrodesabove(gate)andbelow(cathode).Athirdgridelectrode(anode)islocatedabovethegategrid,withtheliquidlevellyingbetweenthegateandanodegrids.TwoHamamatsuR9869[ 85 ]photomultipliertubes(PMTs)viewtheactivevolume,onefrombelowandonefromabove.Atotalof1.76kgisusedtollthestainlesssteelvessel.APTFEspill-overcupsurroundstheTPCstructure,whichxestheheightoftheliquid.TheLXeremovedforrecirculationistakenfromthiscup,andthereforetheliquidlevelintheTPCcannotexceedtheheightofthecup.Thecathodeandgategridsapplyanelectriceldoftypically1kVcm1whichisusedtodriftelectronsawayfromaninteractionsitetowardsthegategrid.Oncetheelectronspassthroughthegategrid,theyarriveattheliquidsurfaceandareextractedtothegasbyanelectriceldof10kVcm1thatthenacceleratestheelectronsthroughthe 101

PAGE 102

Photographandschematicdiagramofthedual-phaseXurichdetector.ThePTFEstructureholdsthePMTsandgridelectrodes(seetext),deninganactiveregion3.5cmindiameterand3cmhigh.ThephotographontopisaviewupthroughtheanodegridtothetopPMT,whilethebottomphotographisaside-viewoftheassembledTPC.DiagrampreparedbyTeresaMarrodanUndagoitia. gasuntiltheycollectontheanodegrid.ThehighvoltageappliedtothegridsissuppliedbyaCAENmodelA1526module.Duringtheirtransitthroughthegas,theelectronswillcollidewithXeatomswithsucientenergytoproducescintillationlight.Therefore,thetypicalresultofaparticleinteractionisapromptscintillationsignal(S1)emittedfromtheinteractionsiteitself,followedbyadelayedscintillationsignal(S2)producedastheelectronstravelthroughthegasundertheinuenceoftheextractioneld.Inthisway,boththescintillationandionizationsignalsaremeasuredbythePMTs.Thistechniqueisusedforchargereadoutbecauseitprovidessuperioramplicationovermoretraditionalmethods[ 95 96 ].Additionally,thez-positionoftheeventcanbeinferredfromthedelaytimebetweentheS1andS2signalssincetheelectrondriftvelocityiswellknownasafunctionoftheappliedeld[ 54 ]. 102

PAGE 103

Spectrumobtainedfrom57CoatzeroeldwithGaussiant.Therun-averagedlightyieldis6.38p.e./keV. Thelightyieldinthiscaseisdenedasthenumberofphotoelectrons(p.e.)emittedfromthePMTphotocathodesperunitenergy,andiscustomarilyquotedbasedontheprimaryemissionof57Co.WhenXurichisoperatedinsingle-phasemodewiththeliquidlevelabovethetopPMT,the57Cosourceproduces10p.e./keV.Inthedual-phasemode,wheretheliquidlevelliesbelowthetopPMT(betweengateandanodegrids),scintillationlightthatreachestheliquidlevelfrombelowisreectedorrefractedduetothedieringindicesofrefractionbetweenliquidandgasxenon[ 97 ].ThoughsomeoftherefractedphotonsmaybedetectedbythetopPMT,andsomeofthereectedphotonsdetectedbythebottomPMT,roughly35%arelostoverall.TheresultisasignicantlylargerS1signalinthebottomPMTcomparedtothetop(70%onbottom,30%ontop),andanoverallreducedlightyieldascomparedwiththevaluetakeninsingle-phasemode.Thedual-phase57Cozero-eldlightyieldismeasuredtobe6:380:05(stat)0:36(sys)p.e./keV,with11.5%resolution(=).Thesystematicuncertaintyistakenfromthelevelofuctuationsinthislightyieldovertime,andthestatisticaluncertaintyisthecombinationoftuncertaintiesfromeach57Cozero-elddataset.Thespectrumobtainedfromone57CocalibrationisshowninFigure 5-2 103

PAGE 104

5-3 ,constructedattheUniversityofFlorida.Acoppercoldnger,alsovacuuminsulated,isimmersedinaliquidnitrogen(LN)bath,at77K.ThecoldngerattachestothebottomofanaluminumcanthatinturnattachesatthetoptothestainlesssteelvesselcontainingtheTPC.ThepathofheatowisthusfromtheSSvesseltothealuminumradiationshield,fromtheretothecoppercoldngerandnallytotheLNbath. Figure5-3. Aphotographandcross-sectionalschematicofthecryostatthathousestheXurichdetector.Coolingisprovidedbyavacuum-insulatedcoppercoldngerimmersedinliquidnitrogen. 104

PAGE 105

98 ].Thecryostatinthismannerprovidesexceptionallystablecoolingpower,withuctuationsatthelevelof0.01{0.1Kovermonthsofcontinuousoperation.Thoughthenormaloperatingtemperatureis180K,thecryostatiscapableofreachingroughly140Kwhennoheatloadisapplied.Twotemperature Figure5-4. Thecryostatperformanceoverroughlyonemonth.Inthisplot,theinitialliquidnitrogenllisdoneatt1:5days,andproceedsforanother1.5days.Theabruptriseintemperatureatt3dayscorrespondstotheheatersbeingturnedon. sensorsarenormallyreadout,onelocatedonthetopoftheradiationcan,whiletheotherislocatedintheLXe.Thevacuuminthecryostatspaceiskeptbelow105mbarbyaVarianturbomolecularpump. 4inSwagelokconnections[ 99 ].Apictureofthegassystem,andaschematicdiagram,areshowninFigure 5-5 .Whennotinuse,theXegasisstoredinCylinder1.Priortocoolingthecryostat,theinnerLXespaceofthedetectorisevacuatedandthenlledwith2bar(absolute)of 105

PAGE 106

ThegassysteminchargeofXelling,purication,recovery,andstorage.ThearrowsindicatethepathoftheXegasduringrecirculation.DiagrampreparedbyTeresaMarrodanUndagoitia. Xegasatroomtemperature,thatactsasathermaltransfergasduringcooldown.Onceoperatingtemperature(175K)hasbeenreached,XegasistransferedfromCylinder1viathepressureregulator,throughthegetterandowmeterandintotheLXeinnerchamberwhereitiscondensed.Cylinder2storesexcessXeandalsoactsasanemergencyrecoveryvolumeincaseofanyproblemsduringlling.ALNdewarisconnectedthroughanelectroniccryogenvalvetoacopperloopsurroundingtheradiationcan.Ifthepressureintheinnerchamberexceeds3bar,thevalveopensautomatically,providingadditionalcoolingpower.Thegetterusesaheatedmetalthatabsorbselectronegativeimpurities.OncetheXellingiscomplete,therecirculationpump(labeled\Rec.pump"inFigure 5-5 )isturnedonandtheXeisdirectedalongthepathindicatedbythearrows.Theowrateiskeptat8.5SLMandiscontrolledbyameteringvalve,indicatedonthediagramasthevalveiconwithadiagonalarrowthroughit,locatedbeforethebuervolume.ThemeasurementandevolutionoftheLXepurityisdiscussedinSection 5.6 106

PAGE 107

5-6 .Withtheexceptionofthephotocathodes,thetwoPMTsareidenticalindesign.Themultipliersectionconsistsoftwelvestagesofametalchanneldynodestructure.ThePMTthatisplacedonthetopofthedetector,inthegas,hasanewtypeofphotocathodedesignedtohaveaquantumeciencyof&35%[ 85 ],whilethebottomPMThasamorestandardphotocathodewithquantumeciency25%.VoltagesaredistributedtothecathodeanddynodechainbyavoltagedividerbuiltontoaPTFEdiscsubstrate. Figure5-6. OneofthephotomultipliertubesusedintheXurichdetector.Thephotocathodeisfacingdown,andvisibleistheinitialtestvoltagedivider. ThegainofthePMTsiscalibratedwithapulsedbluelightemittingdiode,inaprocessexplainedinChapter 7 .Thesinglep.e.spectraobtainedfromthesePMTsatvaryingappliedvoltagesisdisplayedinFigure 5-7 .Alsoshownisthebehaviorofthe 107

PAGE 108

SinglephotoelectronspectrafromXurich'sphotomultipliertubesatvaryingappliedcathodevoltages.Ineachpanel,theredhistogramisfromthetopPMT,whilethebluehistogramisfromthebottom.Theextractedgainandsinglephotoelectronresolutionisalsoshown. gainandsinglephotoelectronresolutionasafunctionofappliedvoltage.TheoperatingvoltagesusedforthetwoPMTs|900VforPMT1and850VforPMT2|aresuppliedbyaNHQ225MNIMmoduleandarechosentominimizetheresolutionswhilenearlyequatingthegains. 5.4.1HardwareTherawPMTsignalsarefedtoanexternalfastvoltageamplier(Phillips777),orwhennoexternalgainisneeded,toalinearfan-out(CAENN454).Bothunitshavetwooutputs;oneoutputisconnecteddirectlytotheanalog-to-digitalconverter(ADC),AcqirismodelDC436100MS/s,whiletheotheroutputisfedtothetriggeringsystem.ACAENN840leadingedgediscriminatorprovidesachannel-by-channeltriggerwhosethresholdissetat1.5p.e..TheselogicsignalsarethentimedbyaN93Btimingunitsothateachpulselasts10s.ThetimedsignalsareconnectedtoaN455coincidenceunitsetto`AND'(requiringcoincidenceinthetwoPMTchannels),andthissignalthenfunctionsasthetriggerfortheADC.ThetriggersetupisshownschematicallyinFigure 5-8 .TheADCsareoutttedwithinternalbandwidthltersthatsuppresssignalcomponentswithfrequencylargerthan50MHz,toavoidNyquistaliasing. 108

PAGE 109

Schematicofthedataacquisitionsystem Theeciencyofthetriggerisstudiedbytwomethods.Therstmethodinvolvesuseofa137Cswhichgivesa662keV-ray.ThissourceisusedbecauseatthelowenergiesitsComptonspectrumisfeaturelessandat;indeed,itisthesamesourceusedtocalibratetheERbandintheXENON10experiment,describedinSection 3.1.3 .ThesecondmethodusedtostudytheeciencyisbyconstructingaMonteCarlosimulation(MC).TheMCbeginsbysimulatingrealisticPMTresponse,describedlaterinSection 7.2 ,inordertodeterminetheeciencytocatchNp.e.givenatriggerthresholdof1.5p.e..Next,thecombinationofgeometricallightcollectioneciency,quantumeciency,and1stdynodecollectioneciency,tot,isestimatedfrom, 100 ]istheenergyrequiredforarecoilingelectrontoproduceasinglescintillationphotoninLXeatzeroappliedeld,andLp:e:=6:74p.e./keVisthemeasuredlightyieldofXurichat9.4keV(seeTable 6-1 ).FromthedetectedsignalreachingthePMTs,30%isdetectedintherstPMT(top),while70%isdetectedinthesecondPMT(bottom).TheindividualPMTecienciesfordetectinganinitialscintillationphoton,(1;2),aretherefore(1)=0:3totand(2)=0:7tot.TheMCstarts 109

PAGE 110

Figure5-9. StudyofthetriggereciencyoftheXurichdetector,givenindividualPMTtriggerthresholdsof1.5p.e..Thespectrumof137Csisinblue,whiletheeciencyfromtheMonteCarlosimulationisgiveninred. triggereciencyisshowninFigure 5-9 inred,alongwiththereal137Csdata.Theresultsshow95%eciencyat20p.e.,rollingdownto70%at10p.e.and10%at5p.e.. 101 ]programconstructedspecicallyfortheXurichAcqirisADCs,isrunonaPCandcommunicateswiththeADCsviaaCompactPCI(cPCI)connection.TheverticalresolutionoftheAcqirisis12-bitandthesamplesarestoredasshortintegersandtransfereddirectlytodiskwith1000eventsperle.Theprocessingprocedureoccursinthreesteps:preliminarydatamanipulation,S2nding,andS1nding. 110

PAGE 111

5-10 (top).Additionally,somesmallS2eventsarenotacharacteristic`pulse',butinsteadaseriesofsmallpulsesspreadoutover1s,asseenbythesignalintheinsetplot.TheS2ndingalgorithmtakesadvantageofthefactthat,thoughoddlyshaped,S2pulsesalwaysoccuroveraspanof1s,whileS1pulsesarenowiderthanhundredsofns.First,a`S1-like'boxarea,A(1)i,iscomputed, 111

PAGE 112

(Top)AnexamplerawPMToutputtracefromaneventindual-phasemode.TheinsetboxshowsazoomedviewofthesmallS2pulseenclosedbytheblackdashedbox.(Bottom)Thesametrace,withazoomedverticalaxis,andtheresultoftheS2lterinred.ThelterhasmanagedtorespondtothelegitimateS2pulses,whileremainingunaectedbytheS1. identicallyzero.Next,alteredsignaliscreatedbycalculatinga`S2-like'boxlter,similartoA(1)i,andsubtractingthevalueofthelargestA(1)ithatlieswithintheS2box: 5-10 inred,superimposedovertherealtrace.Oncethislteredsignalhasbeencomputed,freeofanyS1contribution,itisusedtondthepositionandwidthsoftheS2pulses.Thepulsendingalgorithmhereisquitesimple;ittakesthemaximumvalueofA(2),andstepsiterativelytotheleftuntilA(2)reacheszero.Theextentofthepulsetotherightislikewisefound.Theareainthiswindowisthencomputedfromthesumoftheoriginal(unattened)traces.Thevalues 112

PAGE 113

113

PAGE 114

95 ], 5-11 showsthegasgainasafunctionofgasgapforaxedsetofp,Vgate,andd. Figure5-11. TheS2gain,calculatedafter[ 95 ]asafunctionofgasgap,foraxedVgate=3kV,p=1:8bar,andgate-anodespacingd=5cm. Ifanoveralltiltinthedeviceexists,thismeanstheS2gainfromonelocationofthedetectorwillbedierentthaninotherregions,andwilldegradetheresolutionoftheionizationsignal.Itisthereforenecessarytoensuretheliquidsurfaceisascloseaspossibletobeingparallelwithgateandanodegrids.Inordertotesttheliquidlevel,thelocalizedenergydepositionof-raysfrom57Coisemployed.Dataaretakenwiththesourceplacedonthecryostatbodyatfourdierentazimuthal()positions,allatthesameheightz.Assumingthedrifteld,Ed,is 114

PAGE 115

5-12 showstheS2spectraandpeakpositionsfordatatakenatfoursourcepositions. Figure5-12. (Left)TheS2spectrafrom57Cotakenatvariousazimuthalpositionsbeforelevelingthedetector.(Right)ThepositionoftheseS2peaksasafunctionofsourceazimuthalcoordinate. Figure5-13. TheS2peakpositionsafterperformingleveling,conrmingconsistency.Thepeakpositionat270,determinedbyaGaussiant,isdisplacedfromthepositionofthemaximumhistogrambinduetoanon-negligibleskewinthespectrum(inset).Thepositionofthemaximumbinisindicatedbythered 115

PAGE 116

5-13 showstheS2peakpositionsafterleveling.Withtheexceptionofthepointat270,thepositionsarestatisticallyconsistent.TheS2spectrumtakenat270(Figure 5-13 ,inset),unliketheotherthreepositions,showsapronouncedskew,andthepeakpositionoftheGaussiantisconsiderablydisplacedfromthepeakbininthehistogram.Thered` 5-14 [ 102 ].AlthoughSF6clearlyshowsthestrongesteect,themostimportofthesethreeimpuritiestoXurichisO2,asthesystemisrstexposedtoroomair.ThemetalgetterisparticularlygoodatremovingO2,however,andtheeectivenessofrecirculationcanbereadilyseenbymeasuringthepurityovertime.Thelevelofpurityisdeterminedbymonitoringtheparameterknownastheelectronlifetime,.Givenaknown(oratleastuniform)amountofchargeemittedfromaninteraction,theamountofchargereachingthegasgap,Q(t)givenaninitialamountQ0followsanexponentialdecayasafunctionofthedrifttime(thetimebetweenS1andS2),as, 116

PAGE 117

TherateconstantforattachmentofelectronsforthreedierentimpuritiesinLXeasafunctionofappliedeld.Figuretakenfrom[ 102 ]. ThemeasurementofisaccomplishedbymeasuringtheS2peakinthe57Cospectrumasafunctionofdrifttime.Figure 5-15 showsaplotofS2versusdrifttimeatthebeginningoftherun(left)andagainafterapproximatelyoneweekofrecirculation(right).TheverticalaxisisgivenasthenaturallogarithmoftheS2size,andhencetheslopeofthebandgives.TheprogressionofthemeasuredelectronlifetimeasafunctionofdateisshowninFigure 5-16 .Althoughavalueisreportedforthelaterdatasets,thedatafromFigure 5-15 (right)arestatisticallyconsistentwithzeroslope,andhencethereportedlifetimeisalowerlimit.Themaximumdrifttimeis15-20s(dependingonthedrifteld),andthusacharacteristiclifetimeof&300sensureslessthan5%chargelossfromeventsoccurringatthebottomofthedetector.Theelectronlifetime,,canbeusedtondtheconcentrationofimpurities,typicallygivenin`O2equivalent'.Theconcentrationoffreeelectrons,Ce,followstherelation, 117

PAGE 118

S2versusdrifttimefrom57Cotakenatthebeginningoftherun(left),andafterapproximatelyoneweekofpurication(right). Figure5-16. Themeasuredelectronlifetimeoverthecourseofoneweekofxenonrecirculation. whereCXe+istheconcentrationofXeions,CO2istheconcentrationofdissolvedO2,andkn;r;O2istherateconstantforneutralization,recombination,andattachmenttoO2,respectively.Theattachmenttermdominatesbyseveralordersofmagnitudeovertheneutralizationandrecombinationterms,andcanbeneglectedfordriftingelectrons[ 102 ].Theconcentrationofelectronsisthengivenby, 118

PAGE 119

5{6 and 5{8 gives, 5-14 ,CO2isinunitsofmol/m3(molar).UsingkO271010M1s1,theconcentrationofO2atthebeginningandendofpurication(fromFigure 5-16 )is182pptand0.794ppt(g/g),respectively. 119

PAGE 120

31 103 ]. 120

PAGE 121

104 ].ThissourcehasbeenusedforcalibrationsofdetectorsintheLargeElectron-PositronCollider[ 105 106 ],aswellasintheKATRINexperimentwhichattemptstomeasuretheelectronneutrinoabsolutemass[ 107 ].83mKrshoulddiuseuniformlyinaLXedetector,addressingtheissueofspatialuniformity.Additionally,itstwode-excitationlinesat9.4and32.1keVlieintheenergyrangeofinterestfordarkmatterdirectsearches,anditshalf-lifeofonly1.8hoursallowsforashortturnaroundtimefollowingmeasurement.ThischapterpresentsasuccessfulimplementationofthiscalibrationsourceintheXurichdetector.Furthermore,resultsofmeasurementsoftheLXeenergyscalelinearity,evolutionofenergyresolutionwithenergy,eectsofLXeresponseunderappliedelectriceldsareshown,andlimitsontheleveloflong-livedradiocontaminantsintroducedbythismethodareset. 57 ].Thedecayschemeof83mKrisshowninFigure 6-1 ,indicatingthatmostofthereleasedenergyiscarriedbyinternalconversionandAugerelectrons[ 106 ].The6kBq83RbsourceusedinthisstudywasproducedattheNuclearPhysicsInstitute,Rez(CzechRepublic).Thisinstitutealsoprovides83RbfortheKATRINexperiment[ 107 ].Theparent83Rbis 121

PAGE 122

Thedecayschemeandbranchingratiosof83mKr.Thedecayalwayspassesthroughtwotransitions,givingmostlyinternalconversion(IC)andAuger(A)electrons.Asmallamountoftheenergyiscarriedbygamma-()andX-rays(X)[ 106 ](thedistinctionbetween-andX-raysisintheirsource:-raysarephotonsemittedbynuclei,X-raysarephotonsemittedbyelectrons). producedintheU-120McyclotronfromthereactionnatKr(p,xn)83Rbbyirradiatingamedium-pressuregaseouskryptontargetwith27MeVprotons.Theproduct,depositedonthetargetchamberwalls,isthenwashedintoseveraltensofmillilitersofhighpuritywater(<0.07S/cm).Anappropriateamountofthetargeteluateisthenabsorbedinzeolitebeads(2mmdiameter,Merck),whichactsasamolecularsieve.Zeolitewaschosenduetoitsabilitytoallowforecientemanationof83mKrinvacuum,whileexhibitinghighretentionofthemother83Rbinitsporousstructure.Thedetailsofthesourceproductionprocessaredescribedmorethoroughlyin[ 108 ].Inadditionto83Rb,84Rb(t1=2=38days)and86Rb(t1=2=19days)arealsoproduced,however,theydecaytostableKrisotopesandhenceintroducenoradioactivebackgrounds.Since83Rbdecayswithahalf-lifeof86.2days,thesourcestrengthdecreasedto3kBqbytheendofthesemeasurements.83mKrisintroducedintotheowoftheclosedrecirculationcircuitbymeansofasingleportwithavalve.Thezeolitebeadscontainingthe83Rbresideinasmallchamberlledwiththesamexenongasthatowsinthegassystem.Gaseous83mKremanatingfromthe83Rbdecaymaythendiuseintotherecirculationcircuit,itsintroductionbeingeasilycontrolledbyeitheropeningorclosingthevalveattheport,denotedastheRbvalve.Duetotheratherlonghalf-lifeof83Rb(86.2d),itisimperativethatnotraceofthis 122

PAGE 123

6.3 and 6.4 6-2 (top).TheeventswithsuchadoubleS1structureareshownfromonedatasetinFigure 6-2 (bottom),withtheareaoftherstpulseplottedversustheareaofthesecondpulse.Inthisspace,itisevidentthatthe83mKrdecaysformapopulationofeventsthatisclearlyseparatedfrombackground.TheboxindicatestheenergycutsforrstandsecondS1pulsesusedtoidentify83mKrdecays;beforeopeningtheRbvalve,backgrounddatashownoeventswithinthisbox.AftertheRbvalvehasbeenopened,therateof83mKrdecaysinthetotalLXevolumeincreasestothe20Bqlevelinroughly10h.Inordertofurthercheckthattheseareindeed83mKrdecays,thedistributionofS1delaytimes(i.e.thetimebetweentherstandsecondS1pulses),tS1,ofeventswithintheboxofFigure 123

PAGE 124

(Top)PMToutputfroma83mKrdecay.Inthisdoublepulseofprimaryscintillationlight(S1),therstpulsecorrespondstothe32.1keVtransitionwiththesecondpulseresultingfromthe9.4keVtransition.(Bottom)TheareaoftherstS1pulseversustheareaofthesecond,foreventsshowingthischaracteristictwo-pulsestructure.ShownaredistributionstakenbeforeRbexposure(`Background')andduringRbexposure(`83mKr'),demonstratingthatthepopulationof83mKrdecaysisclearlyseparatedfrombackgroundevents.Theboxrepresentstheenergycutsusedasthe83mKracceptancewindow. 6-2 (bottom)istwithadecayingexponential.Theresultofthet,showninFigure 6-3 (top),givest1=2=1565ns,consistentwiththepublishedvalueof154ns[ 57 ].Thisexcellentagreementvalidatestheclaimthattheseeventsareindeedcausedby83mKrdecays.DuetotheshapingofthePMTsignalsbythevariousDAQcomponents,multipleS1pulsesthataredelayedbylessthan100nscannotbeseparatelyresolved.Additionally,thesignalisrequiredtobe`clean'(i.e.atbaseline)twosamplesbeforeandafterthe 124

PAGE 125

(Top)ThedistributionofdelaytimesbetweenrstandsecondS1pulsesforeventsinthe83mKracceptancewindow.Anexponentialttothedistributiongivesahalf-lifeof1565ns,consistentwiththepublishedvalueof154ns.(Bottom)Spectrafromthetwo83mKrtransitions,summedoverallrunstakenatzeroeld. pulse,inordertoregisterasapositiveS1identicationduringtheoineprocessingofthedata.ThismakestheeciencyfordetectingmultipleS1pulseslessthanunityfortS1<250ns,asisobviousfromFigure 6-3 (top).Therefore,thedoubleS1selectioncutdetects83mKrdecayswithaneciencyofapproximately32%undertheseconditions.Thespectra,inp.e.,obtainedatzeroeldfromthetwotransitionsof83mKraredisplayedinFigure 6-3 (bottom).AGaussianfunctionisttoeachspectrumthatisusedtodeterminethelightyieldandenergyresolution,showninTable 6-1 .Asmentionedinsection 6.1 ,57Coemitsprimarily122keV-rays.However,thereisasmalladditionalcontributionfrom136keV.Thetwolines,however,arenotdistinguishablefromoneanotherduetothedetector'senergyresolutionandinsteadgiveasinglepeak,whose 125

PAGE 126

Themeasuredzero-eldlightyield(L.Y.)andpeakresolution(Res.),andelddependencetparameters,ai.Therowfollowing41.5keVgivesthechargecollectionofthesummedsignal.Uncertaintiesshowninlightyieldarestatisticalonly;becausethesetwopeaksaretakenfromidenticalevents,theirsystematicuncertaintiesarehighlycorrelated,andhencedonotaectthesignicanceoftherelativeriseinlightyield. L.Y.(p.e./keV) Res.(=) 6.740.06 20.0% -0.340.06 63 132.1 6.430.04 14.4% -0.550.03 8.31.5 141.5 | | 0.390.01 132 0.100.01123.6 6.380.05 11.5% -0.6710.003 14.00.2 1 averageenergyis123.6keV.Themeasurementssuggestariseinthelightyieldatlowerenergies,consistentwithbehaviorpreviouslyobservedinLXe[ 109 ]andalsointheXENON10detector[ 60 ].Thepeakresolutions(=)arealsoshownatzeroeld.BecauseLXedetectorstypicallyuseanappliedelectriceldinordertoextractanionizationsignal,itisinterestingtoconsiderwhathappenstothedetectorresponseundersuchanappliedeld.Astheappliedeldisincreased,moreandmoreelectronsleavetheinteractionsite,suppressingtherecombinationprocessthatcontributesphotonstothescintillationsignal.Theresultisthatboththescintillationandionizationresponsesvarystronglywithappliedeld,withthetwosignalsexhibitinganti-correlation.Itisthencrucialthattheeldquenchingbehaviorforanycalibrationsourcesbeknownquantitatively.Figure 6-4 showsthelightyieldasafunctionofappliedeld,normalizedtothezeroeldvalue,ofthetwo83mKrtransitionsandthe57Coline.Thetimescaleoftheionizationsignal,1-2s,doesnotpermitthetwo83mKrtransitionstoberesolvedseparately,andinsteadtheS2signalcontainsthecombinationofchargeemittedfrombothdecays.This41.5keVsummed-signalionizationyieldisalsoshowninFigure 6-4 normalizedtoQ0,thetheoreticaltotalamountofinitialchargeproducedpriortoelectron-ionrecombination.ThisvalueisdeterminedbyplottingtheS1peakpositionsversustheS2peakpositionsfromdatatakenatvariousappliedelds.AsS1andS2areanti-correlated,thesedataliealongalinehavingnegativeslope,withtheline'sintercepts 126

PAGE 127

Fieldquenching,denedasthelightyieldofaspectrallinedividedbythelightyieldobtainedatzeroeld,orS(E)=S(0).Thelevelofeldquenchingdecreasesatlowerenergies,indicatingstrongerelectron-ionrecombinationalongtherecoiltrack.Datacollectedfrom57Coareconsistentwiththosepreviouslyreportedintheliterature[ 56 ].Dashedlinescorrespondtoatparameterizationdescribedinthetext.Alsoshownistheeld-dependentchargecollectionofthecombinationofboth83mKrtransitions,Q(E)=Q0;thetwotransitionsoccurtoocloseintimefortheirionizationsignalstobeindividuallyresolved. representingthetotalnumberofquanta,ionsplusexcitons(Nion+Nex).Forelectronicrecoils,theratioofexcitonstoions,Nex=Nion,istakentobe0.06[ 110 ],andhenceQ0is94.3%thevalueoftheS2intercept.Thedataaretwithathree-parameterfunctionbasedontheThomas-Imelboxmodelforelectron-ionrecombination[ 43 ],givenby 6-1 .Becausethescintillationyieldsarenormalizedtothevalueatzeroeld,a3isunityandthefunction 127

PAGE 128

59 ],andIS1(S2)istheS1(S2)interceptinunitsofp.e..TheCEShastheadvantagethatitisnotaectedbycorrelatedrecombinationuctuationswhichdominatetheS1resolutionovermostenergies[ 43 ],andhencegivesanenergyestimatewithbetterresolutionthanS1orS2alone.Forexample,theS1,S2,andCESspectraofthe41.5keVpeaktakenat500V/cmareshowninFigure 6-5 .TheS1-onlyandS2-onlypeakresolutionsare14.2%and20.1%,respectively.TheresolutionoftheCESpeakatthiseldis10.0%.ThedelaytimebetweenS1andS2givesthedrifttimeoftheelectrons,andhencethez-positionoftheinteraction.OneimportantmotivationforusingthissourceisthatitshoulddisperseuniformlyintheactiveLXevolume,providingaspatially-uniformcalibration.Thesummedz-positiondistributionof83mKreventstakenatdrifteldsfrom100-1000V/cmisshowninFigure 6-6 (top).Theobservedz-dependentrateisatwithvariationsconsistentwithstatisticaluctuationsoneachbin.Withthisuniformcalibration,theposition-dependenceofthedetector'sresponsecanbemeasuredandcorrectedfor.MostoftheS1signalisdetectedbythebottomPMT,andthereforeoneexpectstoseealightyieldthatisamonotonicallydecreasingfunctionofz-position(i.e.morelightiscollectedfromeventsoccurringclosetothebottomPMTthanforeventsclosetothetop).Figure 6-6 (bottom)showsthelightyieldofthe83mKrdecaysatall 128

PAGE 129

Spectrafor41.5keVat500V/cm.TheS1,S2,andcombinedenergyresolutionsare14.2%,20.1%and10.0%,respectively. positionsalongthez-axisbetweenthecathodeandgategridsforthedatarunat1kV/cm;solidlinesarethebandcentroids,shadedbandscover1.Inbothtransitions,thelightyieldatthecathode(bottomofactiveregion)isafactorof1.3higherthanthelightyieldatthegategrid(topofactiveregion).Althoughthe83mKrdecaysawayinamatterofhours,the83Rbwilllivefornearly1.5yrbeforedecayingbelow1%oftheinitialactivity.Ifthistechniqueistobeusedinlow-backgroundexperiments,itisthenimperativethatno83Rbatomsenterthesystem,andinsteadmustremaintrappedwithinthezeoliteorthelter.Inordertotestthis,thevalvetothe83Rbchamberwasclosed.Therateof83mKrdecaysisexpectedtodecreaseexponentiallytozeroduringthefollowingday;however,if83Rbhasenteredthesystem,theratevs.timewillbehaveasanexponentialdecaywithaverticaloset.Nosuchosetwasobservedinthe83mKrratefollowingtheclosingoftheRbvalve.Indeed,2.5hofdatacollectedonedayafterclosingtheRbvalveresultedinzeroobservedevents.Werethevalvetobeleftopen,approximately3000eventswouldbeseeninthistimeperiod. 129

PAGE 130

(Top)Rateof83mKrdecaysasafunctionofz-position,indicatingauniformconcentration.(Bottom)Measuredz-dependenceonthelightyieldfrom83mKr'stwotransitionstakenat1kV/cm.Thesolidlinesindicatedthebandcenters,with1coveredbytheshadedareas.Bothlinesshowalightyieldatthecathodethatisafactorof1.3largerthanatthegategrid. Anullobservationcorrespondstoaone-sided90%condencePoissonupperlimitoflog(10:9)=2:3events.Therefore,therateof83mKrdecayscanbeconstrainedtobelessthan, 2:3events 2:5h32%=800Bq(90%C:L:);(6{2)intheactiveregion. 130

PAGE 131

111 ]. 6-1 ,thelightyieldincreasesatlowenergies.Althoughanaccuratequantitativeunderstandingofthisprocessisincomplete,theobservedbehaviorcanbeunderstoodqualitativelyinthefollowingmanner.TheelectronicstoppingpowerofelectronsinLXeincreasesatdecreasingenergies[ 112 ],andthustheionizationdensityproducedbyarecoilingelectronincreasesalongthetrack,withthehighestdensitiesconcentratedatthetrack'send.Becauseofthis,theoverallionizationdensitycausedbyalowenergyelectronwillbegreaterthanforanelectronofhigherenergy.Theelectronsandionsproducedalongthetrackwillrapidlyrecombineandproducescintillationphotonsastheelectronsfalltotheirgroundstates.Thestrengthofrecombinationiscorrelatedwiththeionizationdensity,becausethecharacteristicelectron-iondistanceisshorterforhigherionizationdensities.Evenatzeroappliedelectriceld,notalloftheelectron-ionpairsproducedwillrecombinetogivescintillationphotons[ 100 ].Itisthenexpectedthatthezeroeldrecombinationisstrongeratlowerenergies(higherdE=dx),givingahigheroveralllightyield.Thispictureisalsoconsistentwiththemeasurementsofthescintillationeldquenching,showninFigure 6-4 131

PAGE 132

113 114 ].Anexposureto83Rbof10hwouldbesucient,undertheseconditions,toprovideadequatestatisticsforsuchacalibration(1000Krevents/kg).Ourupperlimitof83Rbcontaminationtranslatestoaresidualrateof<0.46decays/kg/dayinthis300kg.Evenifthisamountof83Rbwaspresentinthesystem,thevastmajorityofdecayswouldnotintroducedangerousbackgrounds.Inorderforabackgroundeventtobe`dangerous'(i.e.appearintheWIMPsignalacceptancewindow),itmusthavetwofeatures:(1)itmustproduceasinglescatterevent;(2)theeventmustdepositasmallamountofenergythatiswithintheWIMPsearchenergywindow.Anadditionalfeaturethatdual-phase 132

PAGE 133

3-8 ).However,statisticaluctuationscancauseasmallfractionofelectronicrecoileventstoyieldaS2/S1ratiosimilartovaluescharacteristicofanuclearrecoilfromWIMPs,andthustheoverallbackgroundlevelmustbeminimizedasmuchaspossible.Any83mKrdecaysintheactivevolumewouldnotpresentaproblembecausetheywouldeitherhaveadoubleS1structure(andcouldbevetoedonthatbasis),orwouldgive41.5keV,outsideoftheWIMPsearchregion.Theonlypossibilityforadangerousbackgroundisfromoneofthe-raysproducedastheinitialexcited83Krdecaystothemetastablestate.These-raysaremostlyemittedintherangeof500-600keV;again,tobedangeroustheyarerequiredtosingle-scatterintheducialregion,whichishighlyunlikelygiventheir3-4cmattenuationlength.With83Rbcontaminationatthelevelofourupperlimit,MonteCarlosimulationsindicatethat0.46decays/kg/daywouldcontributelessthan67DRUofsinglescattersintheWIMPsearchenergyregion(1DRU1eventkg1day1keV1).Theprojectedbackgroundratein[ 113 ]and[ 114 ]duetonaturalradioactivityinthedetectormaterialsaloneisroughly1mDRU,fullyfteentimesgreaterthanourupperlimitonthe83Rbbackground. 115 ]andthereforewillnotimpedelightcollection. 133

PAGE 134

6.3 todetermineQ0ischosenonlysoastofacilitateawaytoseta\standard"chargescale.ThevalueofNex=Nion=0:06isthetheoreticalvaluebasedonabsorptionspectraofsolidxenon.However,eortstoactuallymeasurethisquantityhavenotconrmedthisresult.In[ 100 ],theauthorsmeasured1MeVconversionelectronsinLXeanddeterminedNex=Nion=0:20.InordertoobtainNex=Nion=0:20,theauthorsof[ 100 ]usedtwomeasurements.Therstmeasurementisofthezero-eldreductionfactor,,denedasthescintillationeciencyrelativetothatofrelativisticheavyions.Itassumesthereductioninscintillationyieldisdueentirelytoescapingelectrons,andisrelatedtoNex=Nionby, 134

PAGE 135

Figure6-7. ConstraintsonNex=Nionand(Nion0=Nion)basedonthedatapresentedin[ 100 ].Shadedareasrepresenttheallowedregionsatthe1-level.Thegreen Therearetwoproblemswiththemethodsused.First,althoughthevalueofisdenedrelativetothescintillationyieldofrelativisticheavyions,theauthorshaveneglectedtheuncertaintyinthescintillationyieldoftheseheavyions.Includingthisadditionaluncertaintychangestheoverallerrorbarby30%,sothatthetruemeasurement 135

PAGE 136

6{4 )arestrongerthanthosebasedon(Equation 6{3 ),becausetherearemultipledatapointsthatcanbeconsidered.Oddly,theauthorsdiscardmorethanhalfofthedatapointswithnojusticationfordoingso.Theyspeculatethattheneglecteddatapointsareinconsistentwiththeresultbecauseofpossibleampliernon-linearity.Thisiscurious,becausemostofthedatadonotexhibitanysuchnonlinearity.Moreimportantly,neglectingthepointsastheyhavedoneistheonlywaythatthismethodcanbemadeconsistentwiththeconstraintsbasedon.IfinsteadoneconsidersalltheirdatathatshowlinearityinS1versusS2,theconstraintsonandNex=Nionareinfactinconsistentwiththemeasurement.TheseconstraintsareshowninFigure 6-7 .Theshadedareasrepresentthe1-allowedregionbybothmethods.Theauthorsdonotreporttheuncertaintiesintheeld-dependencemeasurements;theseuncertaintieshavebeenestimatedbaseduponthelevelofuctuationsinthedatapoints,andhencetheshaded-blueregionmaybeaninaccuraterepresentationofthe1-error.Thebesttheregives=0:2240:027andNex=Nion=0:2950:021.AverysimilarapproachcanbemadefromtheXurichdataofthe41.5keVline.Inthiscase,insteadofusingagivenvalueofNex=NiontodetermineQ0,thevalueofQ0ismeasuredandusedtodetermineNex=Nion.Q0istakentobeQmax=limE!1Q(E).Thecommonapproachtothisproblem(andiswhatwasusedin[ 100 ])isperformedbymakingtheinverseofFigure 6-4 ,thatis,plottingQ1versusE1andextrapolatingtotheverticalintercept,seeninFigure 6-8 .TheredlineisthesametthatwasdeterminedinTable 6-1 .ThemodelitselfgivesthemaximumchargecollectionasQmax=a1+a3.Theassumption,then,isthatQmax=Q0(arisingfromNion),andthatanyscintillationlightremainingafterthisvalueistheresultofNex.AlsoshowninFigure 6-8 istheexpectedQmaxifNex=Nionis0.06and0.20. 136

PAGE 137

Theinversechargecollectionversustheinverseappliedelectriceldofthe41.5keVline.Theplotisusedtoestimatethemaximumchargecollection(atE1=0).TheredlineisthesametgiveninTable 6-1 ,butscalingoutQ0.ShownarewhattheverticalinterceptshouldbeifNex=Nionis0.06(brownline)and0.20(light-blueline). Figure 6-9 showsS1versusS2,bothscaledtonumberofquanta,foreldsrangingfrom0.1{1.0kVcm1.Theverticalintercept,IS2,isassumedtobegivenbyNex+Nion.Thentheratioofexcitonstoionsisgivenby, 116 ],anewmodelforelectron-ionrecombinationstudiedinconjunctionwithdatatakenfromasimilarLXeprototypedetectorindicatedvaluesofNex=Nion0:90fornuclearrecoils.TheXurichresultisnotsensitivetobecauseS1andS2havenotbeenscaledtoS(0)andQ0asdonein[ 100 ].ThevalueofNex=NionfrombothXurichand[ 100 ]depend 137

PAGE 138

S1scaledtophotons,versusS2scaledtoelectrons,foreldsrangingfrom0.1{1.0kVcm1.TheverticalinterceptistakenasNex+Nion,whiletheextrapolationofQmaxisindicatedbytheblack-dashedline. stronglyonhowtheextrapolationtoQmaxisdone.ThemodelfromEquation 6{1 mightnotholdtoveryhighvaluesoftheappliedeld,andhenceQmaxcoulddeviatefroma1+a3.Additionally,itisassumedthattheeciencyforexcitonstoyieldscintillationphotonsisthesameasthatforrecombiningions.If,forexample,thescintillationeciencyforrecombiningionsislessthanthatforexcitons,thenthehorizontalandverticalinterceptsofFigure 6-9 donotrepresentNex+Nion,andthevalueofNex=Nionextractedhereisarticiallytoohigh.ThediscrepancybetweentheoryandmeasurementofNex=Nioncanthereforenotberesolvedwithoutrst,amorerobustmethodofextrapolatingQmax,andsecond,ameasurementoftheeciencyforrecombiningionstogivescintillationphotons. 138

PAGE 139

7-1 .Thecomponentsresideinsideavacuumchambertoallowthefree Figure7-1. Schematicdiagramofaphotomultipliertube.FigurereproducedwithpermissionfromHamamatsuCorporationfrom[ 117 ]. transitofelectrons.Photonsenterthechamberthroughatransparentwindowandareincidentuponasemi-transparentphotocathodewheretheyemitelectronsthroughthephotoelectriceect.Thesephotoelectronsareacceleratedbyanelectriceldontotherstofaseriesofdynodes.Astheelectronscollidewiththerstdynode,secondaryelectronsareemittedviatheAugereectandareinturndirectedtotheseconddynodeandthethird,eachtimemultiplyinginnumberuntilreachingtheanodewheretheyarereadoutbychargesensitiveelectronics.The\gain"ofaPMTistheaveragetotalamplicationoftheentiredynodechain,andcanrangeanywherefrom105to108[ 81 ]. 139

PAGE 140

118 119 ].Thereasonforthenon-consensusisthatdeterminationofanexplicitexpressionforthePMToutputprobabilitydistributionappearstobeintractable[ 120 ].Thischapterapproachestheproblemfromananalyticperspective,followedbyaquantitativetestofseveralapproximationstotheoutputprobabilitydistribution,andnallyanevaluationofanindependentgaindeterminationmethod. 119 ].Thatis,thenumberofelectrons,t,leavingeachdynodeisarandomnumberfollowingaPoissondistributionwithmeanequaltothenumberofincidentelectronsmultipliedbytheamplicationfactorofthedynode,Pn(t),whereisthedynode'samplicationfactor(typicallyaround3to5[ 81 ])andnisthenumberofincidentelectrons.AlthoughsomeattentionhasbeenfocusedondeparturesofsecondaryemissionfromPoissonianity,thereisnoclearevidenceforthis[ 121 ].IlabeltheprobabilityofreceivingtelectronsfromtheNthdynodeasPN(t).BecauseIamconsideringthebehavioroftheSPEspectrum,thenumberofelectronsincidentupontherstdynodeisunity,andthereforetheprobabilityofobtainingtelectronsfromtherstdynode,P1(t),isgivensimplybyaPoissondistributionwithmeanof, 140

PAGE 141

7-2 showstheresultofEquation 7{6 with=4andN=1;2;3;4.Theresolutionofeachdistribution,shownas=where2isthevarianceandisthemean, 141

PAGE 142

AnalyticprobabilitydistributionofaphotomultipliertubeoutputafterNdynodes(Equation 7{6 ),eachwithanamplicationfactorof4.Theverticaldashedlinesarelocatedat(4N+1),where4Nisthemeanofeachdistribution.ThehorizontalaxisisgivenasnN+1sothatnN=0canbeshownonthislog-logplot. increasesateachstep.However,theamountofincreasediminishes;thisisreectiveofthefactthattheresolutionofthenalsignalisexpectedtoberoughlyproportionaltoageometricseriesin1[ 81 ], /1 119 ].BecauseanoutputofnN=0atanydynodeisequivalenttonosignalatall,theimpulsedensityistypicallyignored.ThoughEquation 7{6 iscompact,itisnotatalluseful.AtypicalPMThasnofewerthantendynodes,inwhichcaseusingEquation 7{6 tocalculateevenasinglevalueofnNbecomescomputationallyprohibitive.Additionally,evenatN=3itisclearthattypical 142

PAGE 143

7{6 canbeadequatelyapproximatedasbeingcontinuous. 5.3 .Twospectra,onefromeachPMT,areshownagainhereinFigure 7-3 .Inorder Figure7-3. AnexampleofrealPMTsinglephotoelectronspectra,alsoshowninFigure 5-7 toobtainsuchaspectrum,thePMTisilluminatedbyapulsed,bluelightemittingdiode(LED),withapulsedurationof4sandarepetitionrateof1kHz.Withineachpulse,thecentral1sisintegrated.TheintensityoftheLEDisadjustedsothatroughly95%oftheLEDpulsesgivenoPMTsignal.Withthissmallprobabilityofsuccess,thenumberofphotoelectronsfallingwithinthe1ssignalwindowisPoissondistributedwithanaverageofln(0:95)=0:0513photoelectrons.Suchalowintensityischoseninordertominimizethecontributionfromdoubleandtriplephotoelectrons.Withthisaveragenumber,thefrequencyofdoublephotoelectronsrelativetosinglephotoelectronsisln(0:95)=2=0:026,andhencetheresultingspectrumhasanegligiblecontaminationfrommultiplephotoelectronemission.Thelargepeaknearzero,calledthepedestal,isduetotheintegrationofbaselinenoiseandistreatedasbeingGaussiandistributed;hereitisclearwhytheimpulsedensity 143

PAGE 144

7{6 isnotusefulisthatarealPMTspectrumwillhaveinstrumentalnoiseuctuationsappliedinadditiontothetrueuctuationsalreadyresultingfromtheamplicationprocess.AttothespectrumofFigure 7-3 canbemadewithseveraldierentfunctions[ 118 119 ],andthreeareinvestigatedhere:Gaussian,truncatedGaussian,andcontinuousPoisson.ForsomePMTs,thepeakvalueoftheSPEresponseissignicantlyseparatedfromthepedestalthattheSPEspectrumcanbeapproximatedbyathreeparameterGaussianfunction: 7-3 ,thenon-physicalnegativeportionoftheGaussianfunctionmustsuppressed,or`truncated'.TheresulthasthesameparameterizationastheGaussianfunction,butisdenedtobezerofornegativevaluesofx: 7{1 ,isconvertedtoacontinuousfunctionbytheintroductionofanormalizationparameter,A,acontinuousindependentvariable,t!x,agammafunction,k!!(x+1),andascaleparameter,B: B+1)(ContinuousPoisson):(7{10) 144

PAGE 145

7-1 ,chosentoproduceSPEspectrathatarecharacteristicallysimilartothoseseeninFigure 7-3 .TheresultsfromonesimulationareshowninFigure 7-4 Figure7-4. Anexampleofoneofthe1000setsofsimulatedspectrageneratedbytheMonteCarlosimulation.Colorsrepresentthepedestal(blue),singlephotoelectrons(red),doublephotoelectrons(green),sum(black),andthetruegain(cyan). Eachsimulationbeginsbypickingarandomnumberfromabinomialdistributionwith105trialsand95%probabilityofsuccess;thisnumber,Np,representsthenumberofeventsinthepedestal.Thenumberofsingle,Ns,anddouble,Nd,photoelectroneventsaresimilarlychosen.Thesesingleanddoubleeventsareusedasinputtothedynodesimulation,whichtakestheinputnumberofphotoelectronsasincidentontherstdynode,andchosesanumberfromaPoissondistributionwithmeanofmultipliedbytheinputnumber.Thisresultingnumberisthentreatedasinputtotheseconddynode, 145

PAGE 146

ThedynodeamplicationfactorsusedinthesixcongurationssimulatedbytheMonteCarlo.Anarrowindicatesthatthesamevalueisusedinallsubsequentdynodes.ThebarchartsontherightshowtheperformanceoftheGaussian(purple),truncatedGaussian(blue),andcontinuousPoisson(green)ttingfunctions.Performanceisquantiedbytherelativebias,b=twherebistheestimatorbiasofthegainandtisthetruegain,andtherelativestandarddeviation,=twhereistheestimatorstandarddeviation.ThesevaluesaredeterminedfromthehistogramsinFigure 7-5 .Bybothmeasures,thetruncatedGaussianconsistentlyoutperformstheothertwofunctions. Cong Dyn1 Dyn2 Dyn3 Dyn4 Dyn5-12 3.4! 3.3! 3.2! 3.1! 3.0! 2.02.53.03.03.7! 7-5 foreachofthesixdynodecongurations.Thesespectraarethenusedtodeterminetheestimatorbiasandestimatorvariance.UponvisualinspectionofFigure 7-5 ,thetruncatedGaussianandcontinuousPoissonfunctionsappeartohaveequivalentestimatorvariance,whilethetruncatedGaussianshowsconsistentlysmallerestimatorbias.TheactualbiasandvarianceareshowninTable 7-1 ,andbearoutthisqualitativeassessment. 146

PAGE 147

DistributionsofthegainestimatorsofthethreeSPEtfunctionsdescribedinthetext.ColorsrepresentGaussian(purple),truncatedGaussian(blue),andcontinuousPoisson(green).Ineachframe,theverticalblacklinerepresentsthetruegain. 122 ].Incidentally,thePMTsusedin[ 122 ](HamamatsuR6041Q)areverysimilartothoseusedintheXurichdetector.Insteadofalow-intensityLEDintendedtoproducesinglephotoelectrons,Baldinietal.useaLEDofvaryingintensitiesandtakeadvantageofthefactthattheuctuationsinthenumberofphotoelectronsiscoupledtotheirabsolutenumber.Fromcountingstatistics,therelationbetweenthesignalvariance,thegain,andthechargeoutput,is, 147

PAGE 148

7{12 reducestoEquation 7{11 .However,theSPEspectraofFigure 7-3 haver-valuesof0.8(left)and0.6(right),andthereforeuseofthismethodtodeterminegwouldresultinanerrorof64%and36%,respectively.Theplotof2versusqisstilllinear,butmeasurementofitsslopeoersnowaytoseparatelydeterminegandr. Figure7-6. SpectraofPMToutputfromvaryingtheLEDintensity. However,thistechniquecanprovideacheckoftheparametersobtainedintheSPEt.Figure 7-6 showsthespectrafromZB2183illuminatedatseveraldierentintensities.ThemeanversusvarianceofthesepeaksareshowninFigure 7-7 .Theredlineisattoallvedatapoints.Thislinehasaslopethatisroughly35%higherthanwhatisexpectedfromtheSPEspectrumofthissamePMT.ItispossiblethatthePMTsuersfromnonlinearityatthehighestilluminations;indeed,ifthisisthecasethenthevariancewould 148

PAGE 149

VarianceversusmeanforPMToutputinresponsetovariousLEDilluminations.Theinsertaxesareazoomofthethreelowestdatapoints,whichwereusedforthegreen-linet. beunchanged,whilethemeanwouldbelowerthanexpected,givingahigherslopethanthatpredictedfromlow-illuminationmeasurements.When,instead,onlythethreelowestdatapointsareusedinthet(greenline),theslopeiswithin2%ofthevaluederivedfromtheSPEspectrum. 149

PAGE 150

3 representanimportantresultfromseveraldierentperspectives.First,atthetimeofitsrelease,theresultsforSIinteractionsrepresentedthemostsensitivesearchever,andindeedremainsthemostsensitivemeasurementforWIMPmassesbelow40{50GeVc2(Figure 3-23 ).XENON10'sexclusionlimitsonthepure-neutronSDcrosssectionarethemostsensitiveforallWIMPmasses.Second,XENON10achievedthesesensitivitiesintherstresults.XENON10wasconstructedasaproofofprincipledetector,andinitsrstrunwasabletosurpassthebestresultsofothersearchesusingtechnologiesfarmoremature.IfLXecanyieldsuchimpressiveresultsinitsproofofprincipleapplication,thenfuturesearchesusinglarger,moresophisticatedLXedetectorsaresuretodominatetheeld.WhilenotyetsensitivetothevaluesoftheSIWIMP-nucleoncrosssectionmostfavoredbytheneutralino,XENON10hasbeenabletoexcludeasignicantportionoftheparameterspacedeemedtoliewithinthe95%probabilitycontourforforSUSYmodels.SDsensitivitytorelevantneutralinointeractionsremainsweak,however,thisstudyhasmanagedtoexcludeforthersttimeheavyMajorananeutrinoswithmassesfavoredbyparticletheories.IncombinationwithresultsfromLEP,thelimitonthemassoftheheavyMajorananeutrinoisexcludedbelow2.2TeVc2.TheXENON10experimenthasalsopushedthelimitsofenergythresholdlowerthanotherLXedetectors.Indoingso,anenergyrangeinwhichLewaspoorlyunderstoodsuddenlybecameimportantinunderstandingXENON10'ssensitivity.TheuncertaintyinXENON10'sresultscomingfromthelackofLeunderstandingwasdiscussedatthebeginningofChapter 4 ,andpresentedastrongmotivationforfurtherstudyofthis 150

PAGE 151

6 presentedanewtechniqueforLXeenergycalibration:83mKr.Useofthissourceisnon-trivial,butwasdemonstratedwithremarkablesuccessintheXurichdetector,whosedevelopmentwaspresentedinChapter 5 .83mKrwasnotonlyshowntohavetheadvantageofoeringabackground-freemethodofmeasurementatlowenergies,themethodofintroductionintothedetectorwasshowntobefreeofanyradioactivecontaminantscapableofhinderingalow-backgroundWIMPsearch.TheXurichdetectornotonlyfacilitatedasuccessfulimplementationofthisnewcalibrationsource,butwasusedtostudysomepropertiesofLXeattheselowenergies,relevantfordarkmattersearches.Thelightyield,atzeroappliedeld,wasobservedtoshownonlinearitiesatthelevelof6%between122keVand9.4keV.Liquidxenonisexpectedtoplayanimportantroleinthefutureofdarkmatterdirectdetection,andcouldverywellbethersttechnologytoprobetheregionsofparameterspacemostinterestingforSUSY,atthesametimethatSUSYisbeingprobedwithproton-protoncollisionsattheLHC.ThestudiespresentedinthisdissertationprovideimportantdevelopmentsintheunderstandingofLXeinthecontextofdarkmattersearches,inadditiontothedevelopmentoftechniquesthatwillproveusefulforfutureexperiments. 151

PAGE 152

[1] T.Ferris,ComingofAgeintheMilkyWay(WilliamMorowandCompany,Inc.,NewYork,NY,1988). [2] F.Zwicky,Helv.Phys.Acta6,110(1933). [3] L.Volders,Bull.Astron.Inst.Neth.1,323(1959). [4] V.C.RubinandJ.Ford,W.Kent,Astrophys.J.159,379(1970). [5] M.S.Turner,(1998),astro-ph/9904051. [6] K.G.Begeman,A.H.Broeils,andR.H.Sanders,Mon.Not.Roy.Astron.Soc.249,523(1991). [7] G.Bertone,D.Hooper,andJ.Silk,Phys.Rept.405,279(2005),hep-ph/0404175. [8] Z.Jun-Liang,H.Song-nian,P.Rong-shi,andH.Yan-ping,Chin.AstronomyandAstrophysics15,260(1991). [9] R.G.Carlbergetal.,Astrophys.J.462,32(1996). [10] R.G.Carlberg,H.K.C.Yee,andE.Ellingson,Astrophys.J.478,462(1997). [11] B.Laboratories,http://www.bell-labs.com/org/physicalsciences/projects/-darkmatter/darkmatter.html. [12] J.Wambsganss,LivingRev.Rel.1,12(1998),astro-ph/9812021. [13] J.A.Tyson,G.P.Kochanski,andI.P.Dell'Antonio,Astrophys.J.498,L107(1998),astro-ph/9801193. [14] M.Fukugita,C.J.Hogan,andP.J.E.Peebles,Astrophys.J.503,518(1998),astro-ph/9712020. [15] S.W.Allen,R.W.Schmidt,andA.C.Fabian,Mon.Not.Roy.Astron.Soc.334,L11(2002),astro-ph/0205007. [16] D.Cloweetal.,Astrophys.J.648,L109(2006),astro-ph/0608407. [17] M.Bradacetal.,Astrophys.J.687,959(2008). [18] C.-J.Ma,H.Ebeling,andE.Barrett,Astrophys.J.Lett.693,L56(2009),0901.4783. [19] A.Mahdavietal.,Astrophys.J.668,806(2007). [20] ParticleDataGroup,C.Amsleretal.,Phys.Lett.B667,1(2008). [21] K.A.Olive,(2003),astro-ph/0301505. 152

PAGE 153

WMAPCollaboration,J.Dunkleyetal.,Astrophys.J.Suppl.180,306(2009),0803.0586. [23] R.J.Gaitskell,Ann.Rev.Nucl.Part.Sci.54,315(2004). [24] Super-KamiokandeCollaboration,Y.Fukudaetal.,Phys.Rev.Lett.81,1562(1998),hep-ex/9807003. [25] SNOCollaboration,Q.R.Ahmadetal.,Phys.Rev.Lett.89,011301(2002),nucl-ex/0204008. [26] R.D.PecceiandH.R.Quinn,Phys.Rev.Lett.38,1440(1977). [27] L.J.RosenbergandK.A.vanBibber,Phys.Rept.325,1(2000). [28] L.J.Rosenberg,2009,http://taup2009.lngs.infn.it/slides/jul2/rosenberg.pdf. [29] G.Jungman,M.Kamionkowski,andK.Griest,Phys.Rept.267,195(1996),hep-ph/9506380. [30] L.Baudis,Int.J.Mod.Phys.A21,1925(2006),astro-ph/0511805. [31] XENON10Collaboration,J.Angleetal.,Phys.Rev.Lett.100,021303(2008),0706.0039. [32] L.Roszkowski,Pramana62,389(2004),hep-ph/0404052. [33] N.Weiner,Exoticdarkmattercandidates,2007,Presentedatthe2007SLACSummerInstitute,http://www-conf.slac.stanford.edu/ssi/2007/lateReg/-program.htm. [34] H.-C.ChengandI.Low,JHEP09,051(2003),hep-ph/0308199. [35] J.Knodlsederetal.,Astron.Astrophys.441,513(2005),astro-ph/0506026. [36] J.D.LewinandP.F.Smith,AstroparticlePhysics6,87(1996). [37] CDMSCollaboration,Z.Ahmedetal.,Phys.Rev.Lett.102,011301(2009),0802.3530. [38] CDMSCollaboration,D.S.Akeribetal.,Phys.Rev.Lett.93,211301(2004),astro-ph/0405033. [39] KIMSCollaboration,H.S.Lee.etal.,Phys.Rev.Lett.99,091301(2007),0704.0423. [40] COUPPCollaboration,E.Behnkeetal.,Science319,933(2008),0804.2886. [41] A.M.Bochenkoetal.,QuantumElectronics23,3(1993). [42] M.Tanakaetal.,Nucl.Instrum.Meth.A457,454(2001). 153

PAGE 154

J.ThomasandD.A.Imel,Phys.Rev.A36,614(1987). [44] J.Jortneretal.,J.Chem.Phys.42,4250(1965). [45] A.Hitachi,Astropart.Phys.24,247(2005). [46] J.Lindhardetal.,Mat.Fys.Medd.Dan.Vid.Selsk.33,10(1963). [47] E.Aprileetal.,(2008),0810.0274. [48] V.Chepeletal.,Astropart.Phys.26,58(2006). [49] E.Aprileetal.,Phys.Rev.D72,072006(2005),astro-ph/0503621. [50] D.Akimovetal.,Phys.Lett.B524,245(2002),hep-ex/0106042. [51] R.Bernabeietal.,EPJdirectC11,1(2001). [52] F.Arneodoetal.,Nucl.Instrum.Meth.A449,147(2000). [53] P.Sorensenetal.,Nucl.Instrum.Meth.A601,339(2009),0807.0459. [54] L.S.Miller,S.Howe,andW.E.Spear,Phys.Rev.166,871(1968). [55] D.-M.MeiandA.Hime,Phys.Rev.D73,053004(2006),astro-ph/0512125. [56] E.Aprileetal.,Phys.Rev.Lett.97,081302(2006). [57] S.Chu,L.Ekstrom,andR.Firestone,WWWTableofRadioactiveIsotopes,http://ie.lbl.gov/toi/. [58] K.Nietal.,Nucl.Instrum.Meth.A582,569(2007),0708.1976. [59] T.Shutt,C.E.Dahl,J.Kwong,A.Bolozdynya,andP.Brusov,Nucl.Instrum.Meth.A579,451(2007),astro-ph/0608137. [60] P.Sorensen,APosition-SensitiveLiquidXenonTime-ProjectionChamberforDirectDetectionofDarkMatter:TheXENON10Experiment,PhDthesis,BrownUniversity,Providence,RhodeIsland,2008. [61] S.Yellin,Phys.Rev.D66,032005(2002),physics/0203002. [62] J.R.Ellis,K.A.Olive,Y.Santoso,andV.C.Spanos,Phys.Rev.D71,095007(2005),hep-ph/0502001. [63] L.Roszkowski,R.RuizdeAustri,andR.Trotta,JHEP07,075(2007),0705.2012. [64] M.Hjorth-Jensen,T.T.S.Kuo,andE.Osnes,Phys.Rept.261,125(1995). [65] V.G.J.Stoks,R.A.M.Klomp,C.P.F.Terheggen,andJ.J.deSwart,Phys.Rev.C49,2950(1994),nucl-th/9406039. 154

PAGE 155

J.Engel,Phys.Lett.B264,114(1991). [67] M.T.RessellandD.J.Dean,Phys.Rev.C56,535(1997),hep-ph/9702290. [68] V.A.BednyakovandF.Simkovic,Phys.Part.Nucl.37,S106(2006),hep-ph/0608097. [69] GraphClickv3.0,http://www.arizona-software.ch/graphclick/. [70] B.W.LeeandS.Weinberg,Phys.Rev.Lett.39,165(1977). [71] K.Enqvist,K.Kainulainen,andJ.Maalampi,Nucl.Phys.B317,647(1989). [72] Kainulainen,K.andTuominen,K.andVirkajarvi,J.,Phys.Rev.D75,085003(2007),hep-ph/0612247. [73] H.V.Klapdor-KleingrothausandK.Zuber,ParticleAstrophysics.Studiesinhighenergyphysics,cosmology,andgravitation(InstituteofPhysicsPub.,Bristol,UK,2000). [74] J.R.Primack,D.Seckel,andB.Sadoulet,Ann.Rev.Nucl.Part.Sci.38,751(1988). [75] T.VarleyandL.Roszkowski,2008,Privatecommunication. [76] C.Kouvaris,Phys.Rev.D76,015011(2007),hep-ph/0703266. [77] J.R.EllisandM.Karliner,Phys.Lett.B341,397(1995),hep-ph/9407287. [78] Lewin,J.D.andSmith,P.F.,http://hepwww.rl.ac.uk/UKDMC/pub/papers/-journal/ApP6 add.pdf. [79] ParticleDataGroup,S.Eidelmanetal.,Phys.Lett.B592,1(2004). [80] UKDarkMatterCollaboration,G.J.Alneretal.,Astropart.Phys.28,287(2007),astro-ph/0701858. [81] G.F.Knoll,RadiationDetectionandMeasurement,3rded.(JohnWileySons;NewYork,USA,2000). [82] S.Marroneetal.,Nucl.Instr.andMeth.A490,299(2002). [83] H.LiskienandA.Paulsen,Nucl.DataTables11,569(1973). [84] S.Marino,2007,Privatecommunictaion. [85] H.Photonics,http://www.hamamatsu.com/. [86] GEANT4Collaboration,S.Agostinellietal.,Nucl.Instrum.Meth.A506,250(2003). [87] E.Technology,http://www.eljentechnology.com. 155

PAGE 156

M.B.Chadwicketal.,Nucl.DataSheets107,2931(2006),http://www.nndc.bnl.gov/exfor/endf00.htm. [89] ParticleDataGroup,W.M.Yaoetal.,J.Phys.G33,1(2006). [90] A.Hitachi,J.Phys.Conf.Ser.65,012013(2007). [91] A.Mangiarottietal.,Nucl.Instrum.Meth.A580,114(2007),physics/0610286. [92] E.Aprile,PoS(idm2008),018(2008),http://pos.sissa.it/POSreaders.html. [93] R.Gaitskell,PoS(idm2008),015(2008),http://pos.sissa.it/POSreaders.html. [94] Y.Suzuki,PoS(idm2008),001(2008),http://pos.sissa.it/POSreaders.html. [95] A.I.Bolozdynya,Nucl.Instrum.andMeth.A422,314(1999). [96] M.Yamashitaetal.,Astropart.Phys.20,79(2003). [97] V.N.Solovov,V.Chepel,M.I.Lopes,andA.Hitachi,Nucl.Instrum.Meth.A516,462(2004),physics/0307044. [98] CryogenicControlSystemsInc.,http://www.cryocon.com/. [99] SwagelokFluidSystemTechnologies,http://www.swagelok.com/. [100] T.Dokeetal.,Jpn.J.Appl.Phys.41,1538(2002). [101] NationalInstruments,http://www.ni.com/labview/. [102] G.Bakale,U.Sowada,andW.F.Schmidt,J.Chem.Phys.80,2556(1976). [103] XENON10Collaboration,J.Angleetal.,Phys.Rev.Lett.101,091301(2008),0805.2939. [104] D.McKinsey,2008,presentedattheApril2008DUSELS4/ISEWorkshop,http://neutrino.lbl.gov/dusel/HomestakeWorkshopApril2008/Physics/. [105] ALEPHCollaboration,D.Decampetal.,Nucl.Instrum.Meth.A294,121(1990). [106] DELPHICollaboration,A.Chanetal.,IEEETrans.Nucl.Sci.42,491(1995). [107] KATRINCollaboration,J.Angriketal.,KATRINdesignreport2004,FZKA-7090. [108] D.Venos,A.Spalek,O.Lebeda,andM.Fiser,AppliedRadiationandIsotopes63,323(2005). [109] M.Yamashitaetal.,Nucl.Instrum.Meth.A535,692(2004). [110] T.Takahashietal.,Phys.Rev.A12,1771(1975). 156

PAGE 157

L.W.Kastens,S.B.Cahn,A.Manzur,andD.N.McKinsey,arXiv(2009),0905.1766. [112] ESTAR:StoppingPowersandRangesforElectrons,http://physics.nist.gov/-PhysRefData/Star/Text/ESTAR.html. [113] E.Aprile,ProceedingsofIdenticationofDarkMatter2008PoS(idm2008),018(http://pos.sissa.it/). [114] R.Gaitskell,ProceedingsofIdenticationofDarkMatter2008PoS(idm2008),015(http://pos.sissa.it/). [115] G.Baldini,Phys.Rev.128,1562(1962). [116] C.Dahl,ThePhysicsofBackgroundDiscriminationinLiquidXenon,andFirstResultsfromXENON10intheHuntforWIMPDarkMatter,PhDthesis,PrincetonUniversity,Princeton,NewJersey,2009. [117] HamamatsuPhotonicsK.K.EditorialCommittee,PhotomultiplierTubes:BasicsandApplications,3aed.(HamamatsuPhotonicsK.K.ElectronTubeDivision,2007),http://sales.hamamatsu.com/assets/applications/ETD/-pmt handbook complete.pdf. [118] R.Dossi,A.Ianni,G.Ranucci,andO.Y.Smirnov,Nucl.Instrum.Meth.A451,623(2000). [119] R.J.StokeyandP.J.Lee,TDAProgressReports42-73,36(1983). [120] H.H.Tan,TDAProgressReports42-68,55(1982). [121] J.Prescott,NuclearInstrumentsandMethods39,173(1966). [122] A.Baldinietal.,Nucl.Instrum.Meth.A545,753(2005),physics/0407033. 157

PAGE 158

AaronGostaManalaysaywasborninBethesda,Maryland,spendingmostofhisyouthintheWashingtonD.C.area.Ingradeschoolhebeganplayingthesaxophone,laterbecominginvolvedinvariousfunk,rock,andjazzbands.Hismusicactivities,alwaysacompetitionforhistimewithacademics,nallycametoanendwhenhestartedgraduateschool.AaronenrolledatCaseWesternReserveUniversity(CWRU),inCleveland,Ohio.Followingasetofpositiveexperiencesinhisfreshmanphysicscourses,hedecidedtomajorinphysicsandstayintheeldaslongasitheldhisinterest.InhisjunioryearatCWRUhesawatalkgivenbyProfessorDanAkeribontheeldofdarkmatterdirectdetection.Immediatelyfollowingthetalk,heapproachedDanandaskedtodohisbachelor'sthesisinthatgroup.Hisbachelor'sthesis,entitledSimulatingtheneutronbackgroundintheCDMS-IIexperiment,focusedontheprospectofusingproportionalcountergastubesinordertovetofastneutronsresultingfromhadroniccascadesinducedbycosmic-raymuonstravelingthroughtherocksurroundingtheSoudanminewheretheCDMS-IIexperimentwaslocated.Followinggraduation,AaronworkedforayearinAkerib'sCDMSgroup,workingasalabtechnician.Duringthistime,hedecidedtogototheUniversityofFlorida(UF)forgraduateschool,andlatermetLauraBaudis(thenapost-docinCDMS,andlater,coincidentally,takingafacultypositionatUF)andlearnedaboutthethenproposedXENONdarkmattersearch.WhileinhisrstyearasagraduatestudentatUF,AarondecidedtoapproachLauraaboutdoinghisdissertationinhergroup.ThischoiceledhimtoeventuallyleaveFloridaafterhisthirdyearandtraveltoItaly,Germany,NewYork,andnallySwitzerland,wherehenishedhisdissertation.Intheraremomentsthathehasfreetime,Aaronenjoysrockclimbing,hiking,mountainbikingandsnowboarding.HeplanstostayinZurichfollowinghisPh.D.while 158

PAGE 159

159