La lecture à portée de main
Description
Sujets
Informations
Publié par | universitat_stuttgart |
Publié le | 01 janvier 2011 |
Nombre de lectures | 31 |
Langue | Deutsch |
Poids de l'ouvrage | 1 Mo |
Extrait
IUnnsitviteurstitfa¨u¨trSTthuettogreatritschePhysikI
7P0f5a50enSwtaultdtrginargt57
QuantumThermodynamicsunder
Observation:
TheInuenceofQuantumMeasurements
VonderFakulta¨tMathematikundPhysikderUniversita¨tStuttgartzur
ErlangungderWu¨rdeeinesDoktorsderNaturwissenschaften(Dr.rer.nat.)
genehmigteAbhandlung
Hauptberichter:
Mitberichter:
Vorgelegtvon
ThomasJahnke
ausLudwigsburg
Prof.Dr.Gu¨nterMahler
Prof.Dr.HansPeterBu¨chler
Tagdermu¨ndlichenPru¨fung:28.Februar2011
Institutfu¨rTheoretischePhysikderUniversita¨tStuttgart
1102
D(
issertation
U
n
iversita¨t
S
ut
ttgart)
Danksagung
Ichmo¨chtemichbeieinigenPersonenbedanken,diemichbeidieserDisserta-
tionaufvielfa¨ltigeWeiseunterstu¨tzthaben.
ZuallererstbedankeichmichganzherzlichbeiProf.Dr.Gu¨nterMahlerdafu¨r,
dassermirdieForschunginseinerArbeitsgruppeermo¨glichthat.Seinesehr
guteBetreuungundseinInteresseanmeinerArbeit,welcheszuzahlreichen
anregendenDiskussionenfu¨hrte,habenmaßgeblichzudieserDissertationbei-
getragen.
Prof.Dr.HansPeterBu¨chlerdankeichherzlichfu¨rdieU¨bernahmedesMit-
berichts.
EbensodankeichProf.Dr.ClemensBechingerfu¨rdieU¨bernahmedesPru¨-
fungsvorsitzes.
Bedankenmo¨chteichmichauchbeimeinenehemaligenKollegenSuzanne
Lane´ry,KilianRambach,FlorianRempp,HeikoSchro¨der,JensTeifel,Pedro
Vidal,GeraldWaldherrundHendrikWeimerfu¨rvieleinteressanteDiskussio-
nen.InsbesonderedankeichHeiko,JensundHendrikfu¨rdievielenunterhalt-
samenStunden,diewirzusammenverbrachthaben.
EinbesondererDankgiltmeinenElternManfredundPetraJahnkesowie
meinerSchwesterJasminfu¨rihrevielfa¨ltigeUnterstu¨tzungwa¨hrendmeines
gesamtenStudiumsundderPromotion.
iii
Contents
1.Introduction
I.Quantumthermodynamicsundertheinuenceofperi-
odicmeasurements
1
5
2.Thequantumthermodynamicalmodel7
2.1.Generalpropertiesofquantumthermodynamicalmodels....7
2.2.Theconcretemodel.........................8
2.2.1.Spectrumofthemodularenvironment..........8
2.2.2.Interactionbetweensystemandenvironment......9
2.3.Numericalillustrationofthethermalization...........10
2.4.Quantumthermodynamicsandobservation...........13
3.Quantummeasurementtheory15
3.1.Themeasurementpostulateforprojectivemeasurements....15
3.1.1.Measurementsonbipartitesystems:Theco-jump....16
3.2.POVMmeasurement........................17
3.3.Concretemeasurementmodels..................18
4.Inuenceofperiodicmeasurementsoftheenvironment21
4.1.Eectsofthemeasurementoftheenvironmentalenergy....21
4.2.Short-timedynamics........................23
4.2.1.DiagonalelementsoftheTLSstate............25
4.2.2.O-diagonalelementsoftheTLSstate..........31
4.2.3.Probabilitiesforthemeasurementresults........34
4.3.Normalizationoftheinteraction.................36
4.4.Trajectoriesduetoperiodicmeasurements............38
4.5.Analyticalcalculationoftheensembleaverage..........40
4.5.1.O-diagonalelements...................40
4.5.2.Diagonalelements:Generalproperties..........43
4.5.3.Attractorstatefortheresonantcase...........44
4.5.4.Attractorstatefortheo-resonantcase.........46
v
iv
Contents
4.5.5.Testoftheanalyticalresults...............49
4.5.6.Long-timeaverageandergodicity.............49
4.6.Entropyandlackofknowledge..................52
4.6.1.Informationinthermodynamics..............52
4.6.2.Measurementlogic.....................53
5.Measurementsonasmallspin-environment59
5.1.Dynamicswithoutexternaldisturbance.............59
5.2.Dynamicswithmeasurementsoftheenvironment........60
5.3.DynamicswithdirectmeasurementsoftheTLS.........68
5.4.ExplicitmeasurementmodelbasedonaCNOT-gate......71
5.4.1.HamiltonianfortheCNOTgate.............72
5.4.2.DynamicsoftheTLSunderperiodicapplicationofthe
CNOT-operation......................74
6.ComparisonwiththemodelofKurizkietal.
97
II.Temperatureestimation:Fluctuationsarisingfromquan-
tummeasurements83
7.Temperatureestimationformodularsystems85
7.1.Repeatedtemperatureestimations:Averageanductuations.86
7.1.1.Boundsforthevalidityoftheuctuationformula....91
7.1.2.Temperatureestimationbyusingmorethanonemea-
surement..........................92
7.2.Example:Thenspinmodel....................92
7.2.1.Averagetemperatureestimate..............93
7.2.2.Fluctuationsoftheestimatedtemperature........96
7.2.3.Temperatureestimationwithseveralmeasurements..99
8.Conclusion
III.Appendices
A.Numericaltestoftheapproximations
B.Short-timeapproximation
C.Interactionenergy
301
701
901
131
171
Contents
iiv
D.Evolutionof00forperiodicmeasurementofthesameenergyband119
E.Vanishingoftheo-diagonalelements
F.POVMmeasurementsoftheTLS
G.LowerlimitfortheGaussianapproximation
121
521
721
H.Germansummary-DeutscheZusammenfassung129
H.1.DasquantenthermodynamischeModell..............130
H.2.DerEinussperiodischerMessungen...............132
H.2.1.KurzzeitdynamikundTrajektorien............132
H.2.2.Ensemblemittelung:RelaxationundAttraktorzustand.135
H.2.3.PeriodischeMessungenbeikleinenSpin-Umgebungen.136
H.2.4.VergleichmitdemModellvonKurizkietal........138
H.3.Temperaturscha¨tzungundFluktuationen............138
H.3.1.Temperaturscha¨tzungdurchEnergiemessung:Erwartungswert
undFluktuationen.....................139
H.3.2.KonkretesBeispiel:Dasn-SpinSystem.........141
H.4.Fazit.................................141
ListofSymbols
Bibliography
341
741
1.Introduction
Observationsnotonlydisturbwhathastobemeasured,theypro-
duceit.
—PascualJordan[27]
Thermodynamics[8,47]asdevelopedmainlyinthe19thcentury,isapow-
erfultheorywithalargerangeofapplications.Originallybeingdevelopedto
describeheatengines,itsconceptsareforexamplealsoapplicabletochemical
reactionsorevenincosmology.However,thistheoryispurelyphenomeno-
logical,basedontheaxiomaticlawsofthermodynamics.Thus,thequestion
arises,whetherthermodynamicscouldbederivedfromamorefundamental
theory.
Originally,classicalmechanicshasbeenconsideredassuchanunderlying
theory,whichledtothedevelopmentofstatisticalmechanics.Butsuchclassi-
calapproachesdidnotgetalongwithoutadditionalassumptionsas,e.g.,the
ergodichypothesis.
Inthe20thcenturyquantummechanics[4,44]wasdevelopedtodescribe
systemsatatomicandsubatomicscales.Itturnedoutthatclassicalbehavior
canbeobtainedfromquantummechanicsasaspeciallimitingcase,implying
thatquantumtheoryismorefundamentalthantheclassicaltheories.Con-
cerningthermodynamics,thisledtotwomainquestions:
•Isitpossibletoderivethermodynamicsdirectlyfromquantummechan-
ics,withoutanyfurtherassumptions?
•Aretheconceptsofthermodynamicsinsomesensealsoapplicablefor
quantumsystemsbeyondtheclassicallimit?
Theanswerstothesequestionsaresearchedwithintheratherneweld
ofquantumthermodynamics[15,17,41].Itturnsoutthat,indeed,thermal
propertiescanarisefromquantummechanics:Asystemtypicallyrelaxestoa
stationary,thermalstateduetothecouplingtoanappropriateenvironment.
Remarkably,thisevenholdsforverysmallquantumsystemsasforexample
asingletwo-levelsystem.
However,inquantumthermodynamicsthetotalsystem–consistingofsys-
temandenvironment–isconsideredtobeclosed.Inparticular,thereexists
1
2
1.Introduction
nointeractionwithanexternalobserver.Asweknow,quantummeasurements
leadtoadisturbanceofthemeasuredsystemaccordingtothemeasurement
postulate.Thus,thequestionarises,howanexternalobservationeectsthe
propertiesofaquantumthermodynamicalsystem.
Inthisthesis,weprovidesuchaconnectiontoanexternalobserverby
includingquantummeasurements(i.e.informationaspects)intothequan-
tumthermodynamicalmodel.Aswewillsee,variousconceptsknownfrom
statisticalmechanicsarisefromsucha“quantumthermodynamicsunderob-
servation”,althoughtheydonothaveanymeaningintheisolatedquantum
thermodynamicalsetting:
Instatisticalmechanics,thesystemsaresupposedtopasstroughatra-
jectoryinphasespace.Incontrasttoquantumthermodynamics,thermal
propertiesemergeonlybyaveragingoveranensembleofsuchsystemsoras
along-timeaverageoverasingletrajectory.Theabovementionedergodic
hypothesisthenstatesthatbothoftheseaveragesshouldbeequal.Inour
modelwewill,indeed,recoverquasi-classicaltrajectoriesandobservether-
malizationinthesenseofstatisticalmechanics.Furthermore,wewillprove
theergodicityofthesystem.
Anotherissueconcernstheinterpretationofentropyasameasureoflackof
knowledge.Aslongastheunderlyingtheoryisclassical,thislackofknowledge
canonlybesubjective,sincetheexactmicrostateatanypointintimeiswell
denedandtherefore–atleastinprinciple–canbeknown.Incontrast,the
vonNeumannentropyofasysteminquantumthermodynamicsisbasedona
fundamentallackofknowledgeduetotheentanglementwiththeenvironment.
Basedonthemeasurementlogic,wewillndapossibilitytoconnectthe
objectiveentropyofthesystemwiththesubjectivelackofknowledgeofthe
observercarryingoutthemeasurements.
Afurtherapparentcontradictionbetweenstatisticalmechanicsandquan-
tumthermodynamicsisrelatedtouctuationsofthermodynamicvariables.
Especially,therehavebeenlong-standingcontroversies[29,35]concerningthe
existenceandthemeaningoftemperatureuctuatio