石墨烯
GRAPHENEINTEGERQUANTUMHALLEFFECTINTHE¼PHYSICALREVIEWB74,075422 2006
ThoulesstransitioninthequantumHalleffect.Furthertheo-reticalexplorationofthisphenomenoninthepresentcontextwouldcertainlybeinteresting.
2.Fillingfactor =4
Similarlytowhatwefoundat =0,thegroundstateat =4dependsonthestrengthofZeemancouplingandtheon-siteUrelativetothatoftheinteractionsv1andu2.Theformertermsagainfavoraspin-polarized avor-singletstate,whilethelatterinteractionsnowfavoranin-plane avor-polarizedspin-singletstate.Evaluatingthevariationalener-giesofthesestates,we ndthatthespin-polarizedstatehasalowerenergyifg BB+u0 0 u1 0+u2 0/4.Thisinequal-ityissatis edevenwithavanishingon-siteUandagfactorassmallasunity.Thuswewillassumeherethat =4groundstateisspinpolarized:
=4 = c↑R,mc↑L,m vac .
m
62
gestthatthisisindeedthecase.However,themeasuredac-tivationenergyat =±4wasfoundtobedominatedbythe
single-particleZeemansplitting,suggestingthatherethesys-temisperhapsmoreappropriatelycharacterizedasapara-magnet.Atpresentitisunclearwhetherthesameistruefortheplateausat =0or =±1.Measurementsoftheactiva-tionenergiestherewilllikelyprovidesomehintsastotheirorigin.
Inthefollowingsection,wewillexplorethepossibilitythatthesystemisindeedparamagneticatthesenewlyob-servedquantizedHallplateausandthattheiroriginisduetoexplicitsymmetry-breakingtermsintheHamiltonian ratherthanexchange asappearstobethecaseat =±4.Inanattempttoaccessthisregime,wewillassumethattheeffectsofexchangeareeffectively“canceled”bydisorderanduti-lizeHartreetheorytoexaminethespinand avorsplittingofthen=0andn=1Landaulevelsbysuchterms.Thisanalysiswillhopefullyprovideusefulinputfordeterminingthetruenatureofthe =0and =±1states.
Wenote,however,thatthegapsforspinwavesoutofthisstatearepositiveonlyifg BB+2u0 0 2u1 0+u2 0/2,whichprovidesaslightlymorerestrictiveconditionforhav-ingafullyspin-polarizedgroundstate.Withg=1.4,thelatterinequalityissatis edprovidedU 2eV.
3.Fillingfactor =5
V.PARAMAGNETICREGIME
Finally,thesituationat =5isanalogoustothatat =3.Thegroundstatewillhaveallspin-upstatesinthen=1LandauleveloccupiedtosatisfytheZeemancoupling.Duetothesublatticerepulsionv1,theremainingspin-downelec-tronsinthegroundstateareexpectedtobein-plane avorpolarized:
i
=5 c↓R,m+ec↓L,m =4 .
m
63
Itwillbeinstructivetobeginouranalysisoftheparamag-neticregimebyapplyingHartreetheoryinitiallyinaclean
system.Disordereffectswillbediscussedattheendofthissection.Sinceinourtreatmentoftheferromagneticregimewefoundthattheu2interactiondroppedoutatthen=0Landaulevelandwaseffectiveonlythroughexchangeatn=1,wewillignorethistermhere.Doingsowillgreatlystreamlineourdiscussion,andourresultsdonotdependonthissimpli cation.Weretaintheremainingsymmetry-breakingtermsinH1andconsiderthefollowinginteractingHamiltonian:
Thisstateexhibitsthesametypeoflattice-scaleorderasthe =3groundstateandalsospontaneouslybreaksU 1 avorsymmetrysothataKosterlitz-Thoulesstransitionisexpectedhereaswell.
C.Experimentalrelevance
=H
d2x i v xDx+ yDy g BB·Stot
64
2
+u0 ↑,tot ↓,tot+ ↑,stag ↓,stag u1 stag ,
InitialsightingsofthegraphenequantumHalleffectob-servedintegerquantumHallstatesonlyat llingfactors
=±2,±6,±10,etc.,2,3correspondingto llednearlyfourfold-degenerateLandaulevels.QuantumHallferromagnetismwasthusentirelyabsentintheseexperiments.ThisisalmostcertainlyduetoacollapseoftheexchangegapresponsibleforquantumHallferromagnetismbystrongdisorder,aphe-nomenonthathasbeenestablishedboththeoretically15,32,33andexperimentally.34,35
AsdiscussedintheIntroduction,morerecentexperimentsutilizinghighermagnetic eldsobservedadditionalquan-tizedHallplateausat =0,±1,and±4.5Anaturalquestiontoaskiswhethertheappearanceoftheseplateausisamanifes-tationofquantumHallferromagnetism.NomuraandMac-Donald,whoveryrecentlyderivedanexperimentalcriterionforrealizingquantumHallferromagnetismingraphene,sug-where ,totand ,stagarethetotalandstaggereddensities
zz
forspin e.g., ↑,stag= ↑ ↑ .Wehaveignoredthe niterangeofthesublatticerepulsionv1,asalocalformwillnowbeadequate.Furthermore,wehavedispensedwiththemani-festlySU 2 spin-invariantformoftheon-siteUterminfavorofaformwhichwillbemoreconvenienttoworkwithhere.Notealsothatwehavedroppedpiecesoftheon-siteUin-teractioninvolvingJ±,sinceagainthesewillbeunimportant.OuraimwillbetoemployHartreetheorytoobtaineffec-tivesingle-particleenergylevelsfortheinteractingsystemandinparticulartoextracttheresultingparticle-holeexcita-tionenergieswhichwillbeofpotentialrelevanceforexperi-mentallymeasuredactivationenergies.Toproceedwede-coupletheinteractionsbylinearizingin uctuationsofthetotalandstaggereddensitiesabouttheirmeanvalues,yield-inganeffectivemean- eldHamiltonian
C. Meyer等人对他们观察到的这一现象提出了理论模型:石墨烯并不是绝对 的平面,而是存在一定的小山丘似的起伏。[7]随后,Meyer等人又研究了单层石 墨烯和双层石墨...
热门文档