Meson Decay Constants from the Valence Approximation to Lattice QCD
Masses and decay constants of B_q mesons in the QCD string approach

a rXiv:h ep-ph/61193v116Oct26Masses and decay constants of B q mesons in the QCD string approach A.M.Badalian ∗and Yu.A.Simonov †Institute of Theoretical and Experimental Physics,Moscow,Russia B.L.G.Bakker ‡Vrije Universiteit,Amsterdam,The Netherlands February 2,2008Abstract The relativistic string Hamiltonian is used to calculate the masses and decay constants of B q mesons:they appear to be expressed through onlythree fundamental values:the string tension σ,αs ,and the quark pole masses.The values f B =186MeV,f B s =222MeV are calculated while f B c depends on the c -quark pole mass used,namely f B c =440(424)MeV for m c =1.40(1.35)GeV.For the 1P states we predict the spin-averaged masses:¯M (B J )=5730MeV and ¯M (B sJ )=5830MeV which are in good agreement with the recent data of the D0and CDF Collaborations,at the same time owningto the string correction being by ∼50MeV smaller than in other calculations.1IntroductionThe decay constants of pseudoscalar (P)mesons f P can be directly mea-sured in P →µνdecays [1]and therefore they can be used as an importantcriterium to compare different theoretical approaches and estimate their ac-curacy.Although during the last decade f P were calculated many times:in potential models[2,3,4],the QCD sum rule method[5],and in lattice QCD [6,7],here we again address the properties of the B,B s,B c mesons for several reasons.First,we use here the relativistic string Hamiltonian(RSH)[8],which is derived from the QCD Lagrangian with the use of thefield correlator method (FCM)[9]and successfully applied to light mesons and heavy quarkonia [10,11].Here we show that the meson Green’s function and decay constants can also be derived with the use of FCM.Second,the remarkable feature of the RSH H R and also the correlator of the currents G(x)is that they are fully determined by a minimal number of fundamental parameters:the string tensionσ,ΛMS=250(5)MeV;(1) and the pole masses taken arem u(d)=0;m s=170(10)MeV;m c=1.40GeV;m b=4.84GeV.(2) Third,recently new data on the masses of B c and the P-wave mesons: B1,B2,and B s2have been reported by the D0and CDF Collaborations [12,13],which give additional information on the B q-meson spectra.Here we calculate the spin-averaged masses of the P-wave states B and B s.We would like to emphasize here that in our relativistic calculations no constituent masses are used.In the meson mass formula an overall(fitting) constant,characteristic for potential models,is absent and the whole scheme appears to be rigid.Nevertheless,we take into account an important nonperturbative(NP) self-energy contribution to the quark mass,∆SE(q)(see below eq.(18)).For the heavy b quark∆SE(b)=0and for the c quark∆SE(c)≃−20MeV[10], which is also small.For any kind of mesons we use a universal static potential with pure scalarconfining term,V0(r)=σr−4r,(3)2where the couplingαB(r)possesses the asymptotic freedom property and saturates at large distances withαcrit(n f=4)=0.52[14].The coupling can be expressed throughαB(q)in momentum space,αB(r)=2qαB(q),(4)whereαB(q)=4πβ20ln t BΛ2B.Here the QCD constantΛB,is expressed as[15]ΛB(n f)=Λ2β0· 319n f (6)and M B(σ,ΛB)=(1.00±0.05)GeV is the so called background mass[14]. For heavy-light mesons withΛ2+m2i+p2In (8)m 1(m 2)is the pole (current)mass of a quark (antiquark).The variable ωi is defined from extremum condition,which is taken either from(1)The exact condition:∂H 0p 2+m 2i .(10)ThenH 0ϕn = p 2+m 22+V 0(r ) ϕn =M n ϕn(11)reduces to the Salpeter equation,which just defines ωi (n )=∂˜ωi =0(the so-called einbein approxima-tion).As shown in [9]the difference between ωi and ˜ωi is <∼5%.For the RSH (7)the spin-averaged massM (nL )=ω12+m 212ωb +E n (µ)−2σηf p 2+m 2i nL ;µ=ω1ωbπωf ;(14)with ηf =0.9for a u (d )quark,ηf ∼=0.7for an s quark,ηf =0.4for a c quark,and ηb =0.Therefore,for a b quark ∆SE (b )=0.The mass formula(12)does not contain any overall constant C .Note that the presence of C violates linear behavior of Regge trajectories.The calculated masses of the low-lying states of B ,B s ,and B c mesons are given in Table 2,as well as their values taken from [2,3,6,7].It is of interest to notice that in our calculations the masses of the P -wave states appear to be by 30-70MeV lower than in [2]due to taking into account a string correction [11].4Table1:Masses of the low-lying B q mesons in the QCD String Approach B5280(5)a5279.0(5)5310252753B1(1P)¯M=5730a5721(8)D05734(5)CDFB s5369a5369.6(24)5390253623B s2¯M=58305839(3)D058802B∗c6330(5)a633826321(20)63Current CorrelatorThe FCM can be also used to define the correlator GΓ(x)of the currents jΓ(x),jΓ(x)=¯ψ1(x)Γψ2(x),(15) for S,P,V,and A channels(here the operatorΓ=t a⊗(1,γ5,γµ,iγµγ5)).The correlator,GΓ(x)≡ jΓ(x)jΓ(0) vac,(16) with the use of spectral decomposition of the currents jΓand the definition, vac|¯ψ1γ0γ5ψ2|P n(k=0) =f P n M n,(A,P)vac|¯ψ1γµψ2|V n(k,ε) =f V n M nεµ,(V)(17) can be presented as[3]GΓ(x)d x= n M n0|YΓe−H0T|0ω1ω2N c YΓ=p2 .(20)3Then from Eqs.(18)and(19)one obtains the following analytical expression for the decay constants(for a given state labelled n):f P(V)n 2=2N c M n|ϕn(0)|2.(21)This very transparent formula contains only well defined factors:ω1andωb, the meson mass M n,andϕn the eigenvector ofˆH0.Then in the P channelf P n 2=6(m1m2+ω1ω2− p2 )Table2:Pseudoscalar constants of B q mesons(in MeV)f B189216(34)186(5)f B s218249(42)222(2)f B swhere the w.f.at the origin,ϕn(0),is a relativistic one.In the nonrelativistic limitωi→m i,ϕn(0)→ϕNR n(0)and one comes to the standard expression:f P n(0) 2→12•In our analytic approach with minimal input of fundamental parame-ters(σ,αs,m i)the calculated decay constants are f B=186MeV,f B s=222MeV,f B s/f B=1.19.•For B c the decay constant is very sensitive to m c(pole):f B c=440MeV(m c=1.40GeV)and f B c=425MeV(m c=1.35GeV)References[1]D.Silverman and H.Yao,Phys.Rev.D38,214(1988).[2]S.Godfrey and N.Isgur,Phys.Rev.D32,189(1985);S.Godfrey,Phys.Rev.D70,054017(2004).[3]D.Ebert,R.N.Faustov,and V.O.Galkin,hep-ph/0602110;Mod.Phys.Lett.A17,803(2002),and references therein.[4]G.Cvetic,C.S.Kim,G.L.Wang,and W.K.Namgung,Phys.Lett.B596,84(2004).[5]M.Jamin,nge,Phys.Rev.D65,056005(2002)and referencestherein.[6]A.Ali Khan et al.,Phys.Rev.D70,114501(2004),ibid.64,054504(2004);C.T.H.Davies et al.,Phys.Rev.Lett.92,022001(2004).[7]A.S.Kronfeld,hep-lat/0607011and references therein;I.F.Allison etal.,Phys Rev.Lett.94172001(2005);A.Gray et al.,Phys.Rev.Lett.95,212001(2005).[8]A.Yu.Dubin,A.B.Kaidalov,and Yu.A.Simonov,Phys.Lett.B323,41(1994);Phys.Atom Nucl.56,1745(1993);E.L.Gubankova and A.Yu.Dubin,Phys.Lett.B334,180(1994).[9]H.G.Dosch and Yu.A.Simonov,Phys.Lett.B205,339(1988);Yu.A.Simonov,Z.Phys.C53,419(1992);Yu.S.Kalashnikova,A.V.Nefediev,and Yu.A.Simonov,Phys.Rev.D64,014037(2001);Yu.A.Simonov,Phys.Atom.Nucl.67,553(2004).[10]A.M.Badalian,A.I.Veselov,and B.L.G.Bakker,Phys.Rev.D70,016007(2004);Phys.Atom.Nucl.67,1367(2004).8[11]A.M.Badalian and B.L.G.Bakker,Phys.Rev.D66,034025(2002);A.M.Badalian,B.L.G.Bakker,and Yu.A.Simonov,Phys.Rev.D66,034025(2002).[12]P.Catastini(for the D0and CDF Collab.),hep-ex/0605051;M.D.Cor-coran,hep-ex/0506061.[13]D.Acosta et al.(CDF Collab.),Phys.Rev.Lett.96,202001(2006);hep-ex/0508022.[14]A.M.Badalian and D.S.Kuzmenko,Phys.Rev.D65,016004(2002);A.M.Badalian and Yu.A.Simonov,Phys.Atom.Nucl.60,636(1997).[15]M.Peter,Phys.Rev.Lett.76,602(1997);Y.Schr¨o der,Phys.Lett.B447,321(1999).[16]Yu.A.Simonov,Phys.Lett.B515,137(2001).[17]Particle Data Group,S.Eidelman,et al.,Phys.Lett.B592,1(2004).[18]A.M.Badalian and Yu.A.Simonov(in preparation).9。
Meson decays from string splitting

at which the string intersects the second brane is:
cos2 θ
=
eB(rq
(ρ′ (rq ))2 )−A(rq) + (ρ′
(rq
))2
.
(2)
The
crucial
point
now
is
that
when
θ
=
π 2
there
is
a
net
transversal
a LPTHE, Universit´es Paris VI and VII, 4 place Jussieu; 75005, Paris, France, and Physique Th´eorique et Math´ematique and International Solvay Institutes, Universit´e
e-mails: bigazzi@lpthe.jussieu.fr, cotrone@ecm.ub.es, luca.martucci@fys.kuleuven.be, walter.troost@fys.kuleuven.be
1 Introduction
The String/Gauge theory correspondence states the equivalence of a string theory (or Mtheory) on a non-trivial background and a gauge theory. Once a supergravity background dual to some gauge theory with adjoint fields is given, the easiest way to add fundamental flavors is by placing Nf “flavor branes” in the background [1]. In the probe approximation Nf ≪ Nc (Nc being the number of colors) we can just ignore the backreaction of the flavor branes. In this way it is quite easy to study meson spectra: small brane fluctuations correspond to small spin mesons, while macroscopic spinning strings attached to the branes describe large spin mesons.
]物理化学名词解释
![]物理化学名词解释](https://img.taocdn.com/s3/m/9200f457804d2b160b4ec034.png)
物理化学概念及术语A B C D E F G H I J K L M N O P Q R S T U V W X Y Z概念及术语 (16)BET公式BET formula (16)DLVO理论DLVO theory (16)HLB法hydrophile-lipophile balance method (16)pVT性质pVT property (16)ζ电势zeta potential (16)阿伏加德罗常数Avogadro’number (16)阿伏加德罗定律Avogadro law (16)阿累尼乌斯电离理论Arrhenius ionization theory (16)阿累尼乌斯方程Arrhenius equation (17)阿累尼乌斯活化能Arrhenius activation energy (17)阿马格定律Amagat law (17)艾林方程Erying equation (17)爱因斯坦光化当量定律Einstein’s law of photochemical equivalence (17)爱因斯坦-斯托克斯方程Einstein-Stokes equation (17)安托万常数Antoine constant (17)安托万方程Antoine equation (17)盎萨格电导理论Onsager’s theory of conductance (17)半电池half cell (17)半衰期half time period (18)饱和液体saturated liquids (18)饱和蒸气saturated vapor (18)饱和吸附量saturated extent of adsorption (18)饱和蒸气压saturated vapor pressure (18)爆炸界限explosion limits (18)比表面功specific surface work (18)比表面吉布斯函数specific surface Gibbs function (18)比浓粘度reduced viscosity (18)标准电动势standard electromotive force (18)标准电极电势standard electrode potential (18)标准摩尔反应焓standard molar reaction enthalpy (18)标准摩尔反应吉布斯函数standard Gibbs function of molar reaction (18)标准摩尔反应熵standard molar reaction entropy (19)标准摩尔焓函数standard molar enthalpy function (19)标准摩尔吉布斯自由能函数standard molar Gibbs free energy function (19)标准摩尔燃烧焓standard molar combustion enthalpy (19)标准摩尔熵standard molar entropy (19)标准摩尔生成焓standard molar formation enthalpy (19)标准摩尔生成吉布斯函数standard molar formation Gibbs function (19)标准平衡常数standard equilibrium constant (19)标准氢电极standard hydrogen electrode (19)标准态standard state (19)标准熵standard entropy (20)标准压力standard pressure (20)标准状况standard condition (20)表观活化能apparent activation energy (20)表观摩尔质量apparent molecular weight (20)表观迁移数apparent transference number (20)表面surfaces (20)表面过程控制surface process control (20)表面吸附量surface excess (21)表面张力surface tension (21)表面质量作用定律surface mass action law (21)波义尔定律Boyle law (21)波义尔温度Boyle temperature (21)波义尔点Boyle point (21)玻尔兹曼常数Boltzmann constant (22)玻尔兹曼分布Boltzmann distribution (22)玻尔兹曼公式Boltzmann formula (22)玻尔兹曼熵定理Boltzmann entropy theorem (22)泊Poise (22)不可逆过程irreversible process (22)不可逆过程热力学thermodynamics of irreversible processes (22)不可逆相变化irreversible phase change (22)布朗运动brownian movement (22)查理定律Charle’s law (22)产率yield (23)敞开系统open system (23)超电势over potential (23)沉降sedimentation (23)沉降电势sedimentation potential (23)沉降平衡sedimentation equilibrium (23)触变thixotropy (23)粗分散系统thick disperse system (23)催化剂catalyst (23)单分子层吸附理论mono molecule layer adsorption (23)单分子反应unimolecular reaction (23)单链反应straight chain reactions (24)弹式量热计bomb calorimeter (24)道尔顿定律Dalton law (24)道尔顿分压定律Dalton partial pressure law (24)德拜和法尔肯哈根效应Debye and Falkenhagen effect (24)德拜立方公式Debye cubic formula (24)德拜-休克尔极限公式Debye-Huckel’s limiting equation (24)等焓过程isenthalpic process (24)等焓线isenthalpic line (24)等几率定理theorem of equal probability (24)等温等容位Helmholtz free energy (25)等温等压位Gibbs free energy (25)等温方程equation at constant temperature (25)低共熔点eutectic point (25)低共熔混合物eutectic mixture (25)低会溶点lower consolute point (25)低熔冰盐合晶cryohydric (26)第二类永动机perpetual machine of the second kind (26)第三定律熵Third-Law entropy (26)第一类永动机perpetual machine of the first kind (26)缔合化学吸附association chemical adsorption (26)电池常数cell constant (26)电池电动势electromotive force of cells (26)电池反应cell reaction (27)电导conductance (27)电导率conductivity (27)电动势的温度系数temperature coefficient of electromotive force (27)电动电势zeta potential (27)电功electric work (27)电化学electrochemistry (27)电化学极化electrochemical polarization (27)电极反应reactions on the electrode (27)电极种类type of electrodes (27)电解池electrolytic cell (28)电量计coulometer (28)电流效率current efficiency (28)电迁移electro migration (28)电迁移率electromobility (28)电渗electroosmosis (28)电渗析electrodialysis (28)电泳electrophoresis (28)丁达尔效应Dyndall effect (28)定容摩尔热容molar heat capacity under constant volume (28)定容温度计Constant voIume thermometer (28)定压摩尔热容molar heat capacity under constant pressure (29)定压温度计constant pressure thermometer (29)定域子系统localized particle system (29)动力学方程kinetic equations (29)动力学控制kinetics control (29)独立子系统independent particle system (29)对比摩尔体积reduced mole volume (29)对比体积reduced volume (29)对比温度reduced temperature (29)对比压力reduced pressure (29)对称数symmetry number (29)对行反应reversible reactions (29)对应状态原理principle of corresponding state (29)多方过程polytropic process (30)多分子层吸附理论adsorption theory of multi-molecular layers (30)二级反应second order reaction (30)二级相变second order phase change (30)法拉第常数faraday constant (31)法拉第定律Faraday’s law (31)反电动势back E.M.F (31)反渗透reverse osmosis (31)反应分子数molecularity (31)反应级数reaction orders (31)反应进度extent of reaction (32)反应热heat of reaction (32)反应速率rate of reaction (32)反应速率常数constant of reaction rate (32)范德华常数van der Waals constant (32)范德华方程van der Waals equation (32)范德华力van der Waals force (32)范德华气体van der Waals gases (32)范特霍夫方程van’t Hoff equation (32)范特霍夫规则van’t Hoff rule (33)范特霍夫渗透压公式van’t Hoff equation of osmotic pressure (33)非基元反应non-elementary reactions (33)非体积功non-volume work (33)非依时计量学反应time independent stoichiometric reactions (33)菲克扩散第一定律Fick’s first law of diffusion (33)沸点boiling point (33)沸点升高elevation of boiling point (33)费米-狄拉克统计Fermi-Dirac statistics (33)分布distribution (33)分布数distribution numbers (34)分解电压decomposition voltage (34)分配定律distribution law (34)分散相dispersion phase (34)分体积partial volume (34)分体积定律partial volume law (34)分压partial pressure (34)分压定律partial pressure law (34)分子反应力学mechanics of molecular reactions (34)分子间力intermolecular force (34)分子蒸馏molecular distillation (35)封闭系统closed system (35)附加压力excess pressure (35)弗罗因德利希吸附经验式Freundlich empirical formula of adsorption (35)负极negative pole (35)负吸附negative adsorption (35)复合反应composite reaction (35)盖.吕萨克定律Gay-Lussac law (35)盖斯定律Hess law (35)甘汞电极calomel electrode (35)感胶离子序lyotropic series (35)杠杆规则lever rule (35)高分子溶液macromolecular solution (36)高会溶点upper consolute point (36)隔离法the isolation method (36)格罗塞斯-德雷珀定律Grotthus-Draoer’s law (36)隔离系统isolated system (37)根均方速率root-mean-square speed (37)功work (37)功函work content (37)共轭溶液conjugate solution (37)共沸温度azeotropic temperature (37)构型熵configurational entropy (37)孤立系统isolated system (37)固溶胶solid sol (37)固态混合物solid solution (38)固相线solid phase line (38)光反应photoreaction (38)光化学第二定律the second law of actinochemistry (38)光化学第一定律the first law of actinochemistry (38)光敏反应photosensitized reactions (38)光谱熵spectrum entropy (38)广度性质extensive property (38)广延量extensive quantity (38)广延性质extensive property (38)规定熵stipulated entropy (38)过饱和溶液oversaturated solution (38)过饱和蒸气oversaturated vapor (38)过程process (39)过渡状态理论transition state theory (39)过冷水super-cooled water (39)过冷液体overcooled liquid (39)过热液体overheated liquid (39)亥姆霍兹函数Helmholtz function (39)亥姆霍兹函数判据Helmholtz function criterion (39)亥姆霍兹自由能Helmholtz free energy (39)亥氏函数Helmholtz function (39)焓enthalpy (39)亨利常数Henry constant (39)亨利定律Henry law (39)恒沸混合物constant boiling mixture (40)恒容摩尔热容molar heat capacity at constant volume (40)恒容热heat at constant volume (40)恒外压constant external pressure (40)恒压摩尔热容molar heat capacity at constant pressure (40)恒压热heat at constant pressure (40)化学动力学chemical kinetics (40)化学反应计量式stoichiometric equation of chemical reaction (40)化学反应计量系数stoichiometric coefficient of chemical reaction (40)化学反应进度extent of chemical reaction (41)化学亲合势chemical affinity (41)化学热力学chemical thermodynamics (41)化学势chemical potential (41)化学势判据chemical potential criterion (41)化学吸附chemisorptions (41)环境environment (41)环境熵变entropy change in environment (41)挥发度volatility (41)混合熵entropy of mixing (42)混合物mixture (42)活度activity (42)活化控制activation control (42)活化络合物理论activated complex theory (42)活化能activation energy (43)霍根-华森图Hougen-Watson Chart (43)基态能级energy level at ground state (43)基希霍夫公式Kirchhoff formula (43)基元反应elementary reactions (43)积分溶解热integration heat of dissolution (43)吉布斯-杜亥姆方程Gibbs-Duhem equation (43)吉布斯-亥姆霍兹方程Gibbs-Helmhotz equation (43)吉布斯函数Gibbs function (43)吉布斯函数判据Gibbs function criterion (44)吉布斯吸附公式Gibbs adsorption formula (44)吉布斯自由能Gibbs free energy (44)吉氏函数Gibbs function (44)极化电极电势polarization potential of electrode (44)极化曲线polarization curves (44)极化作用polarization (44)极限摩尔电导率limiting molar conductivity (44)几率因子steric factor (44)计量式stoichiometric equation (44)计量系数stoichiometric coefficient (45)价数规则rule of valence (45)简并度degeneracy (45)键焓bond enthalpy (45)胶冻broth jelly (45)胶核colloidal nucleus (45)胶凝作用demulsification (45)胶束micelle (45)胶体colloid (45)胶体分散系统dispersion system of colloid (45)胶体化学collochemistry (45)胶体粒子colloidal particles (45)胶团micelle (45)焦耳Joule (45)焦耳-汤姆生实验Joule-Thomson experiment (46)焦耳-汤姆生系数Joule-Thomson coefficient (46)焦耳-汤姆生效应Joule-Thomson effect (46)焦耳定律Joule's law (46)接触电势contact potential (46)接触角contact angle (46)节流过程throttling process (46)节流膨胀throttling expansion (46)节流膨胀系数coefficient of throttling expansion (46)结线tie line (46)结晶热heat of crystallization (47)解离化学吸附dissociation chemical adsorption (47)界面interfaces (47)界面张力surface tension (47)浸湿immersion wetting (47)浸湿功immersion wetting work (47)精馏rectify (47)聚(合)电解质polyelectrolyte (47)聚沉coagulation (47)聚沉值coagulation value (47)绝对反应速率理论absolute reaction rate theory (47)绝对熵absolute entropy (47)绝对温标absolute temperature scale (48)绝热过程adiabatic process (48)绝热量热计adiabatic calorimeter (48)绝热指数adiabatic index (48)卡诺定理Carnot theorem (48)卡诺循环Carnot cycle (48)开尔文公式Kelvin formula (48)柯诺瓦洛夫-吉布斯定律Konovalov-Gibbs law (48)科尔劳施离子独立运动定律Kohlrausch’s Law of Independent Migration of Ions (48)可能的电解质potential electrolyte (49)可逆电池reversible cell (49)可逆过程reversible process (49)可逆过程方程reversible process equation (49)可逆体积功reversible volume work (49)可逆相变reversible phase change (49)克拉佩龙方程Clapeyron equation (49)克劳修斯不等式Clausius inequality (49)克劳修斯-克拉佩龙方程Clausius-Clapeyron equation (49)控制步骤control step (50)库仑计coulometer (50)扩散控制diffusion controlled (50)拉普拉斯方程Laplace’s equation (50)拉乌尔定律Raoult law (50)兰格缪尔-欣谢尔伍德机理Langmuir-Hinshelwood mechanism (50)雷利公式Rayleigh equation (50)兰格缪尔吸附等温式Langmuir adsorption isotherm formula (50)冷冻系数coefficient of refrigeration (50)冷却曲线cooling curve (51)离解热heat of dissociation (51)离解压力dissociation pressure (51)离域子系统non-localized particle systems (51)离子的标准摩尔生成焓standard molar formation of ion (51)离子的电迁移率mobility of ions (51)离子的迁移数transport number of ions (51)离子独立运动定律law of the independent migration of ions (51)离子氛ionic atmosphere (51)离子强度ionic strength (51)理想混合物perfect mixture (52)理想气体ideal gas (52)理想气体的绝热指数adiabatic index of ideal gases (52)理想气体的微观模型micro-model of ideal gas (52)理想气体反应的等温方程isothermal equation of ideal gaseous reactions (52)理想气体绝热可逆过程方程adiabatic reversible process equation of ideal gases (52)理想气体状态方程state equation of ideal gas (52)理想稀溶液ideal dilute solution (52)理想液态混合物perfect liquid mixture (52)粒子particles (52)粒子的配分函数partition function of particles (53)连串反应consecutive reactions (53)链的传递物chain carrier (53)链反应chain reactions (53)量热熵calorimetric entropy (53)量子统计quantum statistics (53)量子效率quantum yield (53)临界参数critical parameter (53)临界常数critical constant (53)临界点critical point (53)临界胶束浓度critical micelle concentration (53)临界摩尔体积critical molar volume (54)临界温度critical temperature (54)临界压力critical pressure (54)临界状态critical state (54)零级反应zero order reaction (54)流动电势streaming potential (54)流动功flow work (54)笼罩效应cage effect (54)路易斯-兰德尔逸度规则Lewis-Randall rule of fugacity (54)露点dew point (54)露点线dew point line (54)麦克斯韦关系式Maxwell relations (55)麦克斯韦速率分布Maxwell distribution of speeds (55)麦克斯韦能量分布MaxwelIdistribution of energy (55)毛细管凝结condensation in capillary (55)毛细现象capillary phenomena (55)米凯利斯常数Michaelis constant (55)摩尔电导率molar conductivity (56)摩尔反应焓molar reaction enthalpy (56)摩尔混合熵mole entropy of mixing (56)摩尔气体常数molar gas constant (56)摩尔热容molar heat capacity (56)摩尔溶解焓mole dissolution enthalpy (56)摩尔稀释焓mole dilution enthalpy (56)内扩散控制internal diffusions control (56)内能internal energy (56)内压力internal pressure (56)能级energy levels (56)能级分布energy level distribution (57)能量均分原理principle of the equipartition of energy (57)能斯特方程Nernst equation (57)能斯特热定理Nernst heat theorem (57)凝固点freezing point (57)凝固点降低lowering of freezing point (57)凝固点曲线freezing point curve (58)凝胶gelatin (58)凝聚态condensed state (58)凝聚相condensed phase (58)浓差超电势concentration over-potential (58)浓差极化concentration polarization (58)浓差电池concentration cells (58)帕斯卡pascal (58)泡点线bubble point line (58)配分函数partition function (58)配分函数的析因子性质property that partition function to be expressed as a product of the separate partition functions for each kind of state (58)碰撞截面collision cross section (59)碰撞数the number of collisions (59)偏摩尔量partial mole quantities (59)平衡常数(理想气体反应)equilibrium constants for reactions of ideal gases (59)平动配分函数partition function of translation (59)平衡分布equilibrium distribution (59)平衡态equilibrium state (60)平衡态近似法equilibrium state approximation (60)平衡状态图equilibrium state diagram (60)平均活度mean activity (60)平均活度系统mean activity coefficient (60)平均摩尔热容mean molar heat capacity (60)平均质量摩尔浓度mean mass molarity (60)平均自由程mean free path (60)平行反应parallel reactions (61)破乳demulsification (61)铺展spreading (61)普遍化范德华方程universal van der Waals equation (61)其它功the other work (61)气化热heat of vaporization (61)气溶胶aerosol (61)气体常数gas constant (61)气体分子运动论kinetic theory of gases (61)气体分子运动论的基本方程foundamental equation of kinetic theory of gases (62)气溶胶aerosol (62)气相线vapor line (62)迁移数transport number (62)潜热latent heat (62)强度量intensive quantity (62)强度性质intensive property (62)亲液溶胶hydrophilic sol (62)氢电极hydrogen electrodes (62)区域熔化zone melting (62)热heat (62)热爆炸heat explosion (62)热泵heat pump (63)热功当量mechanical equivalent of heat (63)热函heat content (63)热化学thermochemistry (63)热化学方程thermochemical equation (63)热机heat engine (63)热机效率efficiency of heat engine (63)热力学thermodynamics (63)热力学第二定律the second law of thermodynamics (63)热力学第三定律the third law of thermodynamics (63)热力学第一定律the first law of thermodynamics (63)热力学基本方程fundamental equation of thermodynamics (64)热力学几率thermodynamic probability (64)热力学能thermodynamic energy (64)热力学特性函数characteristic thermodynamic function (64)热力学温标thermodynamic scale of temperature (64)热力学温度thermodynamic temperature (64)热熵thermal entropy (64)热效应heat effect (64)熔化热heat of fusion (64)溶胶colloidal sol (65)溶解焓dissolution enthalpy (65)溶液solution (65)溶胀swelling (65)乳化剂emulsifier (65)乳状液emulsion (65)润湿wetting (65)润湿角wetting angle (65)萨克尔-泰特洛德方程Sackur-Tetrode equation (66)三相点triple point (66)三相平衡线triple-phase line (66)熵entropy (66)熵判据entropy criterion (66)熵增原理principle of entropy increase (66)渗透压osmotic pressure (66)渗析法dialytic process (67)生成反应formation reaction (67)升华热heat of sublimation (67)实际气体real gas (67)舒尔采-哈迪规则Schulze-Hardy rule (67)松驰力relaxation force (67)松驰时间time of relaxation (67)速度常数reaction rate constant (67)速率方程rate equations (67)速率控制步骤rate determining step (68)塔费尔公式Tafel equation (68)态-态反应state-state reactions (68)唐南平衡Donnan equilibrium (68)淌度mobility (68)特鲁顿规则Trouton rule (68)特性粘度intrinsic viscosity (68)体积功volume work (68)统计权重statistical weight (68)统计热力学statistic thermodynamics (68)统计熵statistic entropy (68)途径path (68)途径函数path function (69)外扩散控制external diffusion control (69)完美晶体perfect crystalline (69)完全气体perfect gas (69)微观状态microstate (69)微态microstate (69)韦斯顿标准电池Weston standard battery (69)维恩效应Wien effect (69)维里方程virial equation (69)维里系数virial coefficient (69)稳流过程steady flow process (69)稳态近似法stationary state approximation (69)无热溶液athermal solution (70)无限稀溶液solutions in the limit of extreme dilution (70)物理化学Physical Chemistry (70)物理吸附physisorptions (70)吸附adsorption (70)吸附等量线adsorption isostere (70)吸附等温线adsorption isotherm (70)吸附等压线adsorption isobar (70)吸附剂adsorbent (70)吸附热heat of adsorption (70)吸附质adsorbate (70)析出电势evolution or deposition potential (71)稀溶液的依数性colligative properties of dilute solutions (71)稀释焓dilution enthalpy (71)系统system (71)系统点system point (71)系统的环境environment of system (71)相phase (71)相变phase change (71)相变焓enthalpy of phase change (71)相变化phase change (71)相变热heat of phase change (71)相点phase point (71)相对挥发度relative volatility (72)相对粘度relative viscosity (72)相律phase rule (72)相平衡热容heat capacity in phase equilibrium (72)相图phase diagram (72)相倚子系统system of dependent particles (72)悬浮液suspension (72)循环过程cyclic process (72)压力商pressure quotient (72)压缩因子compressibility factor (73)压缩因子图diagram of compressibility factor (73)亚稳状态metastable state (73)盐桥salt bridge (73)盐析salting out (73)阳极anode (73)杨氏方程Young’s equation (73)液体接界电势liquid junction potential (73)液相线liquid phase lines (73)一级反应first order reaction (73)一级相变first order phase change (74)依时计量学反应time dependent stoichiometric reactions (74)逸度fugacity (74)逸度系数coefficient of fugacity (74)阴极cathode (75)荧光fluorescence (75)永动机perpetual motion machine (75)永久气体Permanent gas (75)有效能available energy (75)原电池primary cell (75)原盐效应salt effect (75)增比粘度specific viscosity (75)憎液溶胶lyophobic sol (75)沾湿adhesional wetting (75)沾湿功the work of adhesional wetting (75)真溶液true solution (76)真实电解质real electrolyte (76)真实气体real gas (76)真实迁移数true transference number (76)振动配分函数partition function of vibration (76)振动特征温度characteristic temperature of vibration (76)蒸气压下降depression of vapor pressure (76)正常沸点normal point (76)正吸附positive adsorption (76)支链反应branched chain reactions (76)直链反应straight chain reactions (77)指前因子pre-exponential factor (77)质量作用定律mass action law (77)制冷系数coefficient of refrigeration (77)中和热heat of neutralization (77)轴功shaft work (77)转动配分函数partition function of rotation (77)转动特征温度characteristic temperature of vibration (78)转化率convert ratio (78)转化温度conversion temperature (78)状态state (78)状态方程state equation (78)状态分布state distribution (78)状态函数state function (78)准静态过程quasi-static process (78)准一级反应pseudo first order reaction (78)自动催化作用auto-catalysis (78)自由度degree of freedom (78)自由度数number of degree of freedom (79)自由焓free enthalpy (79)自由能free energy (79)自由膨胀free expansion (79)组分数component number (79)最低恒沸点lower azeotropic point (79)最高恒沸点upper azeotropic point (79)最佳反应温度optimal reaction temperature (79)最可几分布most probable distribution (80)最可几速率most propable speed (80)概念及术语BET公式BET formula1938年布鲁瑙尔(Brunauer)、埃米特(Emmett)和特勒(Teller)三人在兰格缪尔单分子层吸附理论的基础上提出多分子层吸附理论。
Decay Constants $f_{D_s^}$ and $f_{D_s}$ from ${bar{B}}^0to D^+ l^- {bar{nu}}$ and ${bar{B}

form factor.
PACS index : 12.15.-y, 13.20.-v, 13.25.Hw, 14.40.Nd, 14.65.Fy Keywards : Factorization, Non-leptonic Decays, Decay Constant, Penguin Effects
∗ experimentally from leptonic B and Ds decays. For instance, determine fB , fBs fDs and fDs
+ the decay rate for Ds is given by [1]
+ Γ(Ds
m2 G2 2 2 l 1 − m M → ℓ ν ) = F fD D s 2 8π s ℓ MD s
1/2
(4)
.
(5)
In the zero lepton-mass limit, 0 ≤ q 2 ≤ (mB − mD )2 .
2
For the q 2 dependence of the form factors, Wirbel et al. [8] assumed a simple pole formula for both F1 (q 2 ) and F0 (q 2 ) (we designate this scenario ’pole/pole’): q2 F1 (q ) = F1 (0) /(1 − 2 ), mF1
∗ amount to about 11 % for B → DDs and 5 % for B → DDs , which have been mentioned in
CP violation in the radiative dileptonic B-meson decays

Current address: Kernfysisch Versneller Institute, Zernikelaan 25, 9747 AA Groningen, The Netherlands E-mail address: erkol@kvi.nl ‡ E-mail address: gsevgur@.tr
†
in the SM [14] and beyond [15]-[19]. So, we think that it would be interesting and complementary to consider the remaining exclusive mode Bd → γ ℓ+ ℓ− . In this paper, we would like to study the CP violation in the exclusive Bd → γ ℓ+ ℓ− decay in the context of the SM. Bd → γ ℓ+ ℓ− decay is induced by the pure-leptonic decay Bd → ℓ+ ℓ− , which is well known to have helicity suppression for light lepton modes, having branching ratios (BR) of the order of 10−15 for ℓ = e and 10−10 for ℓ = µ channels [1]. However, when a photon line is attached to any of the charged lines in Bd → ℓ+ ℓ− process, it changes into the corresponding radiative ones, Bd → γ ℓ+ ℓ− , so helicity suppression is overcome and larger branching ratios are expected. In [2]( [3]), it was found that in the SM, BR(Bd → ℓ+ ℓ− γ ) = (1.5(1.5) , 1.2(1.8) , − (6.2)) × 10−10 for ℓ = e, µ, τ , respectively. Although these BR’s are quite low, in models beyond the SM they can be enhanced by two (one) orders, as shown e.g. in [20]([21]) for Bs(d) → γ ℓ+ ℓ− decay, so investigation of this process may also be interesting from the point of view of the new physics effects. In Bd → γ ℓ+ ℓ− decays, depending on whether the photon is released from the initial quark or final lepton lines, there exist two different types of contributions, namely the so-called ”structure dependent” (SD) and the ”internal Bremsstrahlung” (IB) respectively, while contributions coming from the release of the free photon from any charged internal line will be suppressed by a factor 2 of m2 b /MW . The SD contribution is governed by the vector and axial vector form factors and it is free from the helicity suppression. Therefore, it could enhance the decay rates of the radiative processes Bd → ℓ+ ℓ− γ in comparison to the decay rates of the pure leptonic ones Bd → ℓ+ ℓ− . As for the IB part of the contribution, it is proportional to the ratio mℓ /mB and therefore it is still helicity suppressed for the light charged lepton modes while it is expected to enhance the amplitude considerably for ℓ = τ mode. However, we note that IB part of the amplitude does not contribute to CP violating asymmetry ACP and the forward-backward asymmetry AF B (see section 2). We organized the paper as follows: In section 2, first the effective Hamiltonian is presented and the form factors are defined. Then, the basic formulas of the differential branching ratio dBR/dx, ACP , AF B and CP violating asymmetry in forward-backward asymmetry ACP (AF B ) for Bd → γ ℓ+ ℓ− decay are introduced. Section 3 is devoted to the numerical analysis and discussion.
Heavy-light decay constants with three dynamical flavors

light quark masses, the chiral extrapolation was
a major source of systematic error. Here, chang-
ing from linear to quadratic chiral fits of decay constants changes the results by <∼2%.
quette following Refs. [3,11], and then define the 1-loop coefficient ζA by ZAtad = 1 + αPs ζA.
(2) Fix the scale from mρ on each set, with valence quarks extrapolated to physical values.
∞/∞
8.40 0.09 417 200
0.031/0.031 7.18 0.09 162 40
0.0124/0.031 7.11 0.09 25 –
tors for 5 light and 5 heavy masses. This gives
good control over both chiral and heavy quark
We present preliminary results for the heavy-light leptonic decay constants in the presence of three light dynamical flavors. We generate dynamical configurations with improved staggered and gauge actions and analyze them for heavy-light physics with tadpole improved clover valence quarks. When the scale is set by mρ, we find an increase of ≈ 23% in fB with three dynamical flavors over the quenched case. Discretization errors appear to be small (<∼3%) in the quenched case but have not yet been measured in the dynamical case.
The Delta(1232) at RHIC

a r X i v :n u c l -t h /0409026v 3 26 J a n 2005The ∆(1232)at RHICHendrik van Hees †and Ralf RappCyclotron Institute and Physics Department,Texas A&M University,CollegeStation,Texas 77843-3366E-mail:hees@,rapp@Abstract.We investigate properties of the ∆(1232)and nucleon spectral functions at finite temperature and baryon density within a hadronic model.The medium modifications of the ∆consist of a renormalization of its pion-nucleon cloud and resonant π∆scattering.Underlying coupling constants and form factors are determined by the elastic πN scattering phase shift in the isobar channel,as well as empirical partial decay widths of excited baryon resonances.For cold nuclear matter the model provides reasonable agreement with photoabsorption data on nuclei in the ∆-resonance region.In hot hadronic matter typical for late stages of central Au -Au collisions at RHIC we find the ∆-spectral function to be broadened by ∼65MeV together with a slight upward mass shift of 5-10MeV,in qualitative agreement with preliminary data from the STAR collaboration.PACS numbers:25.75.-q,21.65.+f,12.40.-y 1.Introduction At low energies,the main features of Quantum Chromodynamics (QCD)are confinement and the spontaneous breaking of chiral symmetry.The former implies that we onlyobserve hadrons (rather than quarks and gluons),while the latter is believed to govern the (low-lying)hadron-mass ttice-QCD calculations predict a phase transition from nuclear/hadronic matter to a deconfined,chirally symmetric state [1]at temperatures T ≃150-200MeV,dictating a major reshaping of the hadronic spectrum in terms of degenerate chiral partners.The observation of such medium modifications is therefore an important objective in relativistic heavy-ion collision rge theoretical efforts have been devoted to evaluate in-medium properties of vector mesons which are accessible experimentally through dilepton invariant-mass spectra [2].In most of these studies,baryon-driven effects are essential to account for the dilepton enhancement observed in P b -Au collisions at the SPS below the free ρmass [3,4].Thus,changes in the baryon properties themselves deserve further investigation.In addition,recent measurements of πN invariant-mass spectra in nuclear collisions [5,6,7]may open a more direct window on modifications of the ∆(1232).†presenting authorTo date,in-medium properties of the∆have mostly been assessed in cold nuclear matter[8,9,10,11,12,13],with few exceptions[14,15].In this article we will discuss properties of the nucleon and the∆(1232)in a hot and dense medium[16],employing a finite-temperaturefield theory framework based on hadronic interactions.Both direct interactions of the∆with thermal pions as well as modifications of its freeπN self-energy(incuding vertex corrections)will be accounted for.The article is organised as follows.In Sec.2we introduce the hadronic Lagrangian and outline how its parameters are determined using scattering and decay data in vacuum.In Sec.3we compute in-medium self-energies for nucleon and∆.In Sec.4we first check our model against photoabsorption cross sections on the nucleon and nuclei, followed by a discussion of the spectral functions under conditions expected to occur in high-energy heavy-ion collisions.We close with a summary and outlook.2.Hadronic Interaction Lagrangians in VacuumThe basic element of our analysis are3-point interaction vertices involving a pion and two baryons,πB1B2.Baryonfields are treated using relativistic kinematics, E2B(p)=m2B+p2,but neglecting anti-particle contributions and restricting Rarita-Schwinger spinors to their non-relativistic spin-3/2components.Pions are treated fully relativisticly(ω2π(k)=m2π+k2).The resulting interaction Lagrangians are thus of the usual nonrelativistic form involving(iso-)spin-1/2Pauli matrices,1/2to3/2transition operators,as well as spin-3/2matrices[9,17,18,19,20],see Ref.[16]for explicit expressions.To simulatefinite-size effects we employed hadronic form factors with auniform cutoffparameterΛπB1B2=500MeV(except forπNN andπN∆vertices).The imaginary part of the vacuum self-energy for the∆→Nπdecay takes the formImΣ(Nπ)∆(M)=−f2πN∆MF2(k cm)Θ(M−m N−mπ)(1)with k cm the center-of-mass decay momentum(an extra factor m N/E N(k cm)has been introduced in Eq.(1)to restore Lorentz-invariance),and the real part is determined via a dispersion relation.With a bare mass of m(0)∆=1302MeV,a form-factor cutoffΛπN∆=290MeV and a coupling constant fπN∆=3.2we obtain a satisfactoryfit to the experimentalδ33-phase shift[21,15,22].To account for resonant interactions of the∆with pions we identified the relevant excited baryons via their decay branchings B→π∆.The pertinent coupling constants have been determined assuming the lowest partial wave to be dominant(unless otherwise specified)[23],using pole masses and(total)widths of the resonances.The same procedure has been adopted for resonantπN interactions(which are used to evaluate finite-temperature effects on the nucleon).The resonances included are N(1440), N(1520),N(1535),∆(1600),∆(1620)and∆(1700).The total widthsfiguring into the resonance propagators have been obtained by scaling up the partialπN andπ∆channels,and vacuum renormalizations of the masses have been neglected.Figure 1.Diagrammatic representation ofπN∆vertex corrections(dashed lines:pion,solid lines:nucleon,double solid lines:∆(1232));a bubble with labelαcorresponds to a Lindhard functionΠα(α∈{1,2})attached to baryon lines withpertinent Migdal parameters,i.e.,g′12forα=1and g′22forα=2.Finally,the evaluation of the photoabsorption cross section requires aγN∆vertex for which we employ the magnetic coupling[10]LγN∆=−fγN∆3m2π d4lE N(l)k2F2π(|k|)(3)×{[Θ(k0)+σ(k0)fπ(|k0|)]Aπ(k)G N(l)−f N(l0)A N(l)Gπ(k)}, where k=p−l is the pion4-momentum.The thermal distributions are defined by f N(l0)=f fermi(l0−µN,T)and fπ(|k0|)=f bose(|k0|,T)exp(µπ/T),with f fermi and f bose the Fermi and Bose functions,respectively.For simplicity,finite pion-chemical potentials,µπ>0,are treated in the Boltzmann limit to avoid Bose singularities in the presence of broad pion spectral functions(a more detailed discussion of this point will be given elsewhere).In Eq.(3)positive energies k0>0correspond to outgoing pions,i.e.,∆→πN decays,while k0<0accounts for scattering with(incoming)pions from the heat bath.The key quantities in Eq.(3)are the in-medium pion and nucleon propagators, Gπand G N,and pertinent spectral functions A N=−2Im G N and Aπ=−2Im Gπ.The modifications of the pion propagator are implemetend via a self-energy,arising from two parts:(i)interactions with thermal pions modeled by a four-point interaction in second order(“sunset diagram”)[24],with a coupling constant adjusted to qualitatively reproduce the results of more elaborateππinteractions in s,p,and d-wave[25]; (ii)interactions with baryons via p-wave nucleon-and∆-hole excitations atfinite temperature,described by standard Lindhard functions,supplemented by short-range correlations encoded in Migdal parameters[26](our default values are g′NN=0.8, g′N∆=g′∆∆=0.33).These excitations induce a softening of the pion-dispersion relation which can even lead to a(near)vanishing of the pion group velocity atfinite momentum, inducing an artificial threshold enhancement in the∆self-energy[14].This feature isk γ[GeV]σγ/Ak γ[GeV]σγ/A [µb ]Figure 2.Photoabsorption cross sections on nucleons (left panel,data from [29])andnuclei (right panel;data from [30,31,32,33,34,35]).remedied by accounting for appropriate vertex corrections,which in the case of ρ→ππdecays are required to maintain a conserved vector current in the medium [27,28].Here we apply the same technique to the πN ∆vertex,cf.Fig.1.The nucleon self-energy is calculated in terms of resonant interactions with thermal pions,at the same level of approximation as the pion Lindhard functions (i.e.,neglecting offenergy-shell dependencies in the spectral functions of the excited baryons).The second contribution to the in-medium ∆self-energy consists of resonant π∆→B interactions,corresponding to the finite-temperature part of πB loops.The resulting self-energy expressions are similar to Eq.(3)but with only the scattering part (k 0<0)retained (note that this is consistent with our description of the ∆(1232)in vacuum where πB loops are not included).4.In-medium Spectral Properties of the ∆4.1.Photoabsorption on Nucleons and NucleiValuable constraints on the ∆spectral function in cold nuclear matter can be obtained from photoabsorption cross sections on nuclei.To leading order in αem ,the latter can be related to the photon self-energy (electromagnetic current correlator),Πγ,by [18]σabs γA k 12ImΠγ(k 0=k ),Πγ=1M N [GeV]-I m G N [G e V -1]k 0 [GeV]-I m G π [G e V -2]Figure 3.Left panel:nucleon spectral function at RHIC (solid line T =180MeV,̺N =0.68̺0;dashed line:T =100MeV,̺N =0.12̺0).Right panel:pion spectralfunction for cold nuclear matter (dashed line:T =0,̺N =0.68̺0)and at RHIC (solidline:T =180MeV,̺N =0.68̺0);the dash-dotted line corresponds to switching offbaryonic effects leaving only the 4-point interactions with thermal pions.Migdal parameters or the nuclear density is very moderate.Given our rather simple approach for the cross section,the agreement with data is fair.The discrepancies at low energy (which seem to be present already for the nucleon)could be due to interference with the background,collective effects involving direct NN −1-excitations,or transverse contributions with in-medium ρmesons in the vertex corrections of the ∆decay.At higher energies,further resonances in the photon self-energy need to be included.4.2.Hot Hadronic MatterLet us finally turn to the results for hot hadronic matter.In heavy-ion collisions one expects a hierarchy of chemical freeze-out (determining the ratios of stable hadrons)and thermal freeze-out (where elastic rescattering ceases).The former is characterized by a temperature T chem and a common baryon chemical potential µB .Thermal freezeout occurs at lower T fo ≃100MeV,which requires the build-up of additional chemical potentials for pions,kaons,etc.[36],to conserve the observed hadron ratios,including relative chemical equilibrium for elastic processes,e.g.πN ↔∆implying µ∆=µN +µπ.Under RHIC conditions the nucleon spectral function exhibits an appreciable broadening and a moderate downward mass shift (left panel of Fig.3)due to resonant scattering offthermal pions.The pion spectral function (right panel of Fig.3)is strongly broadened mostly due to scattering offbaryons,with little mass shift.Thermal motion completely washes out the multi-level structure visible at zero temperature (dashed line).Also for the ∆spectral function (left panel in Fig.4)the main effect is a broadening with a slight repulsive mass shift.Half of the increase of the in-medium width is due to baryon-resonance excitations (slightly enhanced due to in-medium pion propagators),adding to the contribution of the πN loop.In the real part,however,the predominantly repulsive contributions from baryon resonances are counterbalanced by net attraction in the πN loop (mostly due to the pion-Bose factor).At thermal freeze-out we find1 1.1 1.2 1.3 1.4 1.5 1.6M ∆ [GeV]0510152025-I m G ∆ [G e V -1]vacuum T=100 MeV T=180 MeV 1 1.1 1.2 1.3 1.4 1.5 1.6M ∆ [GeV]0510152025-I m G ∆ [G e V -1]vacuum T=70 MeV T=160 MeV Figure 4.In-medium ∆(1232)spectral functions in heavy-ion collisions compared tofree space (dash-dotted lines);left panel:RHIC;dashed line:T =100MeV,̺N =0.12̺0(µN =531MeV),µπ=96MeV;solid line:T =180MeV,̺N =0.68̺0(µN =333MeV),µπ=0.Right:future GSI facility;dashed line:T =70MeV,̺N =0.19̺0(µN =727MeV),µπ=105MeV;solid line:T =160MeV,̺N =1.80̺0(µN =593MeV),µπ=0.a peak position at about M ≃1.226GeV and a width Γ≃177MeV,to be compared to the corresponding vacuum values of M ≃1.219GeV and Γ≃110MeV,in qualitative agreement with preliminary data from STAR [7].For more conclusive comparison a detailed treatment of the freeze-out dynamics is mandatory.In the vicinity of T c ,the ∆width increases substantially.We expect this trend to be further magnified when including transverse parts in the vertex corrections,especially in combination with in-medium ρ-mesons [2].In the right panel of Fig.4we show the ∆-spectral function in a net-baryon rich medium,representative for the future GSI facility.Whereas in dilute matter the line shape is only little affected,the resonance structure has essentially melted close to T c ,mostly due to a strong renormalization of the pion propagator at high density.5.Conclusions and outlookBased on hadronic interaction Lagrangians employed within a finite-temperature many-body approach we have evaluated medium effects on pions,nucleons and deltas.The resulting ∆-spectral functions in cold nuclear matter provide fair agreement with photoabsorption data on nuclei.In hot hadronic matter,we found a significant broadening and a slight upward peak shift of the ∆resonance,qualitatively in line with preliminary measurements of πN invariant-mass spectra at RHIC.Future improvements of the πN ∆system in vacuum include u -channel exchange diagrams as well as spin-3/2-∆∗excitations which we expect to increase the rather low form-factor cut-offused so far.We further plan to implement in-medium baryon propagators into the description of axial-/vector mesons within a chiral framework to arrive at a more consistent picture of the equation of state of hadronic matter under extreme conditions [37]and the chiralphase transition.Another interesting ramification[38]concerns the role of the medium-modified∆spectral functions in the soft photon enhancement as recently observed at the SPS[39].AcknowledgmentsOne of us(HvH)acknowledges support from the Alexander-von-Humboldt Foundation as a Feodor-Lynen Fellow.References[1]Karsch F2002Lect.Notes Phys.583209[2]Rapp R and Wambach J2000Adv.Nucl.Phys.251[3]Agakishiev G et al.(CERES/NA45)1998Phys.Lett.B422405[4]Adamova D et al.(CERES/NA45)2003Phys.Rev.Lett.91042301[5]Hjort E L et al.1997Phys.Rev.Lett.794345[6]Pelte D et al.(FOPI)1997Z.Phys.A35955[7]Fachini P2004J.Phys.G30S735[8]Horikawa Y,Thies M and Lenz F1980Nucl.Phys.A345386[9]Oset E and Salcedo L L1987Nucl.Phys.A468631[10]Ericson T and Weise W1988Pions and Nuclei(Clarendon Press,Oxford)[11]Migdal A B,Saperstein E E,Troitsky M A et al.1990Phys.Rept.192179[12]Xia L H,Siemens P J and Soyeur M1994Nucl.Phys.A578493[13]Korpa C L and Lutz M F M2004Nucl.Phys.A742305[14]Ko C M,Xia L H and Siemens P J1989Phys.Lett.B23116[15]Korpa C L and Malfliet R1995Phys.Rev.C522756[16]van Hees H and Rapp R2004Preprint nucl-th/0407050[17]Cubero M1990Ph.D.thesis TH Darmstadt[18]Rapp R,Urban M,Buballa M et al.1998Phys.Lett.B4171[19]Urban M,Buballa M,Rapp R et al.2000Nucl.Phys.A673357[20]Nacher J C,Oset E,Vicente M J et al.2001Nucl.Phys.A695295[21]Moniz E J1981Nucl.Phys.A354535c[22]Weinhold W,Friman B and N¨o renberg W1998Phys.Lett.B433236[23]Hagiwara K et al.2002Phys.Rev.D6601001[24]van Hees H and Knoll J2002Phys.Rev.D6*******[25]Rapp R and Wambach J1995Phys.Lett.B35150[26]Migdal A B1978Rev.Mod.Phys.50107[27]Chanfray G and Schuck P1993Nucl.Phys.A555329[28]Herrmann M,Friman B L and N¨o renberg W1993Nucl.Phys.A560411[29]Lepretre A et al.1978Phys.Lett.B7943[30]Ahrens J et al.1984Phys.Lett.B146303[31]Ahrens J1985Nucl.Phys.A446229c[32]Frommhold T et al.1992Phys.Lett.B29528[33]Frommhold T et al.1994Z.Phys.A350249[34]Bianchi N et al.1993Phys.Lett.B299219[35]Bianchi N et al.1996Phys.Rev.C541688[36]Rapp R2002Phys.Rev.C66017901[37]Voskresensky D N2004Preprint hep-ph/0402020[38]Rapp R2004Mod.Phys.Lett.A191717[39]Aggarwal M M et al.(WA98)2004Phys.Rev.Lett.93022301。
Bs Mixing and decays at the Tevatron

a r X i v :0707.1007v 2 [h e p -e x ] 11 J u l 2007B 0s mixing and decays at the TevatronMossadek Talby (On behalf of the D0and CDF Collaborations)CPPM,IN2P3-CNRS,Universit ´e de la M´editerran ´ee,Marseille,France This short review reports on recent results from CDF and DØexperiments at the Tevatron collider on B 0smixing and the lifetimes of B 0s and Λb .1.IntroductionDue to the large b ¯b cross section at 1.96TeV p ¯p col-lisions,the Tevatron collider at Fermilalb is currentlythe largest source of b -hadrons and provides a very rich environment for the study of b -hadrons.It is also the unique place to study high mass b -hadrons suchas B 0s ,B c ,b -baryons and excited b -hadrons states.CDF and DØare both symmetric multipurpose de-tectors [1,2].They are essentially similar and consist of vertex detectors,high resolution tracking cham-bers in a magnetic field,finely segmented hermitic calorimeters and muons momentum spectrometers,both providing a good lepton identification.They have fast data acquisition systems with several levels of online triggers and filters and are able to trigger at the hardware level on large track impact parameters,enhancing the potential of their physics programs.2.B 0s mixingThe B 0-¯B0mixing is a well established phenomenon in particle physics.It proceeds via a flavor changing weak interaction in which the flavor eigenstates B 0and ¯B0are quantum superpositions of the two mass eigenstates B H and B L .The probability for a B 0me-son produced at time t =0to decay as B 0or ¯B0at proper time t >0is an oscillatory function with a frequency ∆m ,the difference in mass between B Hand B L .Oscillation in the B 0dsystem is well estab-lished experimentally with a precisely measured os-cillation frequency ∆m d .The world average value is∆m d =0.507±0.005ps −1[3].In the B 0s system,the expected oscillation frequency value within the standard model (SM)is approximately 35times faster than ∆m d .In the SM,the oscillation frequencies ∆m d and ∆m s are proportional to the fundamental CKM matrix elements |V td |and |V ts |respectively,and can be used to determine their values.This determina-tion,however,has large theoretical uncertainties,but the combination of the ∆m s measurement with the precisely measured ∆m d allows the determination of the ratio |V td |/|V ts |with a significantly smaller theo-retical uncertainty.Both DØand CDF have performed B 0s -¯B 0s mixinganalysis using 1fb −1of data [4,5,6].The strategies used by the two experiments to measure ∆m s are very similar.They schematically proceed as follows:theB 0s decay is reconstructed in one side of the event and its flavor at decay time is determined from its decayproducts.The B 0s proper decay time is measured fromthe the difference between the B 0s vertex and the pri-mary vertex of the event.The B 0s flavor at production time is determined from information in the opposite and/or the same-side of the event.finally,∆m s is ex-tracted from an unbinned maximum likelihood fit of mixed and unmixed events,which combines,among other information,the decay time,the decay time res-olution and b -hadron flavor tagging.In the following only the latest CDF result is presented.2.1.B 0ssignal yields The CDF experiment has reconstructed B 0sevents in both semileptonic B 0s →D −(⋆)s ℓ+νℓX (ℓ=e orµ)and hadronic B 0s→D −s (π+π−)π+decays.In both cases the D −s is reconstructed in the channels D −s →φπ−,D −s →K ⋆0K −and D −s →π−π+π−with φ→K +K −and K ⋆0→K +π−.Additional par-tially reconstructed hadronic decays,B 0s →D ⋆−sπ+and B 0s →D −s ρ+with unreconstructed γand π0in D ⋆−s→D −s (φπ−)γ/π0and ρ+→π+π0decay modes,have also been used.The signal yields are 61,500semileptonic decays,5,600fully reconstructed and 3,100partially reconstructed hadronic decays.This correponds to an effective statistical increase in the number of reconstructed events of 2.5compared to the first CDF published analysis [5].This improve-ment was obtained mainly by using particle identifi-cation in the event selection,by using the artificial neural network (ANN)selection for hadronic modes and by loosening the kinematical selection.Figure 1shows the distributions of the invariant masses of the D +s (φπ+)ℓ−pairs m D s ℓand of the ¯B 0s →D +s (φπ+)π−decays including the contributions from the partially reconstructed hadronic decays.2.2.B 0s proper decay time reconstructionThe proper decay time of the reconstructed B 0s events is determined from the transverse decay length L xy which corresponds to the distance between theprimary vertex and the reconstructed B 0s vertex pro-jected onto the transverse plane to the beam axis.Forfpcp07Figure1:The invariant mass distributions for the D+s(φπ+)ℓ−pairs(upper plot)and for the¯B0s→D+s(φπ+)π−decays(bottom plot)including the contri-butions from the partially reconstructed hadronic decays. the fully reconstructed B0s decay channels the proper decay time is well defined and is given by:t=L xyM(B0s)P T(D sℓ(π))×K,K=P T(D sℓ(π))1Neutrino,π0andγ.fpcp07an efficiencyǫ,the fraction of signal candidates with aflavor tag,and a dilution D=1−2ω,whereωis the probablity of mistagging.The taggers used in the opposite-side of the event are the charge of the lepton(e andµ),the jet charge and the charge of identified kaons.The information from these taggers are combined in an ANN.The use of an ANN improves the combined opposite-side tag effectiveness by20%(Q=1.8±0.1%)compared to the previous analysis[5].The dilution is measured in data using large samples of kinematically similar B0d and B+decays.The same-sideflavor tags rely on the identification of the charge of the kaon produced from the left over ¯s in the process of B0s fragmentation.Any nearby charged particle to the reconstructed B0s,identified as a kaon,is expected to be correlated to the B0sflavor, with a K+correlated to a B0s and K−correlated to¯B s .An ANN is used to combine particle-identificationlikelihood based on information from the dE/dx and from the Time-of-Flight system,with kinematic quan-tities of the kaon candidate into a single variable.The dilution of the same side tag is estimated using Monte Carlo simulated data samples.The predicted effec-tiveness of the same-sideflavor tag is Q=3.7%(4.8%) in the hadronic(semileptonic)decay sample.The use of ANN increased the Q value by10%compared to the previous analysis[5].If both a same-side tag and an opposite-side tag are present,the information from both tags are combined assuming they are independent.2.4.∆m s measurementAn unbinned maximum likelihoodfit is used to search for B0s oscillations in the reconstructed B0s de-cays samples.The likelihood combines masses,decay time,decay-time rsolution,andflavor tagging infor-mation for each reconstructed B0s candidate,and in-cludes terms for signal and each type of background. The technique used to extract∆m s from the un-binned maximum likelihoodfit,is the amplitude scan method[7]which consists of multiplying the oscilla-tion term of the signal probablity density function in the likelihood by an amplitude A,andfit its value for different∆m s values.The oscillation amplitude is expected to be consistent with A=1when the probe value is the true oscillation frequency,and con-sitent with A=0when the probe value is far fromthe true oscillation frequency.Figure3shows thefit-ted value of the amplitude as function of the oscil-lation frequency for the combination of semileptonic and hadronic B0s candidates.The sensitivity is31.3ps−1for the combination2∆m d m B0s for the hadronic decays alone.fpcp07Figure4:The logarithm of the ratio of likelihoodsΛ≡log L A=0/L A=1(∆m s) ,versus the oscillation frequency. The horizontal line indicates the valueΛ=−15that cor-responds to a probability of5.7×10−7(5σ)in the case of randomly tagged data.3.b-hadrons lifetime measurements atthe Tevatron RunIILifetime measurements of b-hadrons provide impor-tant information on the interactions between heavy and light quarks.These interactions are responsible for lifetime hierarchy among b-hadrons observed ex-perimentally:τ(B+)≥τ(B0d)≃τ(B0s)>τ(Λb)≫τ(B c) Currently most of the theoretical calculations of the light quark effects on b hadrons lifetimes are per-formed in the framework oftheHeavy QuarkEx-pansion(HQE)[10]in which the decay rate of heavy hadron to an inclusivefinal state f is expressed as an expansion inΛQCD/m b.At leading order of the expansion,light quarks are considered as spectators and all b hadrons have the same lifetime.Differ-ences between meson and baryon lifetimes arise at O(Λ2QCD/m2b)and splitting of the meson lifetimes ap-pears at O(Λ3QCD/m3b).Both CDF and DØhave performed a number of b-hadrons lifetimes measurements for all b-hadrons species.Most of these measurements are already included in the world averages and are summarised in[11].In this note focus will be on the latest results on B0s andΛb measurements from CDF and DØ. 3.1.B0s lifetime measurementsIn the standard model the light B L and the heavy B H mass eigenstates of the mixed B0s system are ex-pected to have a sizebale decay width difference of order∆Γs=ΓL−ΓH=0.096±0.039ps−1[12].If CP violation is neglected,the two B0s mass eigenstates are expected to be CP eigenstates,with B L being the CP even state and B H being the CP odd state.Various B0s decay channels have a different propor-tion of B L and B H eigenstates:•Flavor specific decays,such as B0s→D+sℓ−¯νℓand B0s→D+sπ−have equal fractions of B L andB H at t=0.Thefit to the proper decay lengthdistributions of these decays with a single signalexponential lead to aflavor specific lifetime:τB s(fs)=12Γs 22Γs 2,Γs=ΓL+ΓH3123.1.2.B0s lifetime measurements in B0s→J/ψφDØexperiment has performed a new B0s mean life-time measurement in B0s→J/ψφdecay mode.The analysis uses a data set of1.1fb−1and extracts three parameters,the average B0s lifetime¯τ(B0s)=1/Γs, the width difference between the B0s mass eigenstates ∆Γs and the CP-violating phaseφs,through a study of time-dependent angular distribution of the decay products of the J/ψandφmesons.Figure6shows the distribution of the proper decay length andfits to the B0s candidates.From afit to the CP-conserving time-dependent angular distributions of untagged de-cay B0s→J/ψφ,the measured values of the average lifetime of the B0s system and the width difference be-tween the two B0s mass eigenstates are[16]:τDØB s =1.52±0.08(stat)+0.01−0.03(sys)ps∆Γs=0.12+0.08−0.10(stat)±0.02(sys)ps−1 Allowing for CP-violation in B0s mixing,DØprovidesFigure6:The proper decay length,ct,of the B0s candi-dates in the signal mass region.The curves show:the signal contribution,dashed(red);the CP-even(dotted) and CP-odd(dashed-dotted)contributions of the signal, the background,light solid(green);and total,solid(blue).thefirst direct constraint on the CP-violating phase,φs=−0.79±0.56(stat)+0.14−0.01(sys),value compatiblewith the standard model expectations.3.2.Λb lifetime measurementsBoth CDF and DØhave measured theΛb lifetime in the golden decay modeΛb→J/ψΛ.Similar analysis procedure have been used by the two experiments,on respectively1and1.2fb−1of data..TheΛb lifetime was extracted from an unbinned simultaneous likeli-hoodfit to the mass and proper decay lenghts distribu-tions.To cross check the validity of the method simi-lar analysis were performed on the kinematically sim-ilar decay B0→J/ψK s.Figure7shows the proper decay time distributions of the J/ψΛpair samples from CDF and DØ.TheΛb lifetime values extracted from the maximum likelihoodfit to these distributions are[17,18]:τCDFΛb=1.580±0.077(stat)±0.012(sys)ps andτDØΛb=1.218+0.130−0.115(stat)±0.042(sys)ps.Figure7:Proper decay length distribution of theΛb candi-dates from CDF(upper plot)and DØ(bottom plot),with thefit result superimposed.The shaded regions represent the signal.The CDF measured value is the single most pre-cise measurement of theΛb lifetime but is 3.2σhigher than the current world average[3](τW.A.Λb= 1.230±0.074ps).The DØresult however is con-sistent with the world average value.The CDF and DØB0→J/ψK s measured lifetimes are:τCDFB0=1.551±0.019(stat)±0.011(sys)ps andτDØB0=1.501+0.078−0.074(stat)±0.05(sys)ps.Both are compatiblewith the world average value[11](τW.A.B0=1.527±0.008ps).One needs more experimental input to conclude about the difference between the CDF and the DØ/world averageΛb lifetime values.One of the Λb decay modes that can be exploited is the fullyfpcp07hadronicΛb→Λ+cπ−,withΛ+c→pK−π+.CDF has in this decay mode about3000reconstructed events which is5.6more than inΛb→J/ψΛ.Recently,the DØexperiment has performed a new measurement of theΛb lieftime in the semileptonic de-cay channelΛb→Λ+cµ−¯νµX,withΛ+c→K s p[19]. This measurement is based on1.2fb−1of data.As this is a partially reconstructed decay mode the proper decay time is corrected by a kinematical factor K= P T(Λ+cµ−)/P T(Λb),estimated from Monte Carlo sim-ulation.TheΛb lifetime is not determined from the usually performed unbinned maximum likelihoodfit, but is extracted from the number of K s pµ−events in bins of their visible proper decay length(VPDL). Figure8shows the distribution of the number of Λ+cµ−as function of the VPDL with the result of thefit superimposed.ThefittedΛb lifetime value is τ(Λb)=1.28±0.12(stat)±0.09(sys)ps.This results is compatible with the lifetime value fromΛb→J/ψΛand the world average.Figure8:Measured yields in the VPDL bins and the result of the lifetimefit.The dashed line shows the c¯c contribu-tion.AcknowledgmentsI would like to thank the local organizer committee for the wonderful and very successful FPCP07confer-ence.References[1]R.Blair et al.(DØCollaboration),FERMILAB-PUB-96/390-E(1996).[2]V.M.Abazov et al.(DØCollaboration),Nucl.In-strum.Methods A565,243(2006).[3]W.-M.Yao el al.J.Phys.G33,1(2006)[4]V.M.Abazov et al.(DØCollaboration),Phys.Rev.Lett.97,021802(2006).[5]A.Abulancia et al.(CDF Collaboration),Phys.Rev.Lett.97,062003(2006).[6]A.Abulancia et al.(CDF Collaboration),Phys.Rev.Lett.97,242003(2006).[7]H.G.Moser and A.Roussarie,Nucl.Instrum.Methods Phys.Res.,Sect.A384,491(1997). [8]D.Acosta et al.(CDF Collaboration)Phys.Rev.Lett.96,202001(2006).[9]M.Okamoto,T2005(2005)013hep-lat/0510113].[[10]J.Chay,H.Georgi and B.Grinstein,Phys.Lett.B247,399(1990);C.Tarantino,hep-ph/0310241;E.Franco,V.Lubicz,F.Mescia,C.Trantino,Nucl.Phys.B633,212,hep-ph/0203089;F.Gabbiani, A.I.Onishchenko, A.A.Petrov,Phys.Rev.D70,094031,hep-ph/0407004;M.Beneke,G.Buchalla,C.Greub,A.Lenz,U.Nierste,Nucl.Phys.B639,389,hep-ph/0202106.[11]Heavy Flavor Averaging Group(HFAG),“Aver-ages of b-hadron Properties at the End of2006”, hep-ex/07043575(2007).[12]A.Lenz and U.Nierste,hep-ph/0612167,TTP06-31,December2006;M.Beneke et al.,Phys.Lett.B459,631(1999).[13]V.M.Abazov et al.(DØCollaboration),Phys.Rev.Lett.97,241801(2006).[14]CDF Collaboration,CDF note7757,13August2005.[15]CDF Collaboration,CDF note7386,23March2005.[16]V.M.Abazov et al.(DØCollaboration),Phys.Rev.Lett.98,121801(2007).[17]V.M.Abazov et al.(DØCollaboration),hep-ex/07043909,FERMILAB-PUB-07/094-E;Submitted to PRL.[18]CDF Collaboration,CDF note8524,30Novem-ber2006.[19]V.M.Abazov et al.(DØCollaboration),hep-ex/07062358,FERMILAB-PUB-07/196-E;Submitted to PRL.fpcp07。
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
a rXiv:h ep-la t/9319v17Oct1993Meson Decay Constants from the Valence Approximation to Lattice QCD F.Butler,H.Chen,J.Sexton 1,A.Vaccarino,and D.Weingarten IBM Research P.O.Box 218,Yorktown Heights,NY 10598ABSTRACT We evaluate f π/m ρ,f K /m ρ,1/f ρ,and m φ/(f φm ρ),extrapolated to physical quark mass,zero lattice spacing and infinite volume,for lattice QCD with Wilson quarks in the valence (quenched)approximation.The predicted ratios differ from experiment by amounts ranging from 12%to 17%equivalent to between 0.9and 2.8times the corresponding statistical uncertainties.1INTRODUCTIONIn a recent paper[1]we presented lattice QCD predictions for the masses of eight low-lying hardrons,extrapolated to physical quark mass,zero lattice spacing, and infinite volume,using Wilson quarks in the valence(quenched)approximation. The masses we found were within6%of experiment,and all differences between pre-diction and experiment were consistent with the calculation’s statistical uncertainty. We argued that this result could be interpreted as quantitative confirmation of the low-lying mass predictions both of QCD and of the valence approximation.It ap-peared to us unlikely that eight different valence approximation masses would agree with experiment yet differ significantly from QCD’s predictions including the full effect of quark-antiquark vacuum polarization.We have now evaluated the infinite volume,zero lattice spacing,physical quark mass limit of fπ/mρ,f K/mρ,1/fρ,and mφ/(fφmρ).To our knowledge there have been no previous systematic calculations of this physical limit of lattice meson decay constants.A review of earlier work is given in Ref.[2].The predicted mφ/(fφmρ) lies above its observed value by about its statistical uncertainty of approximately 15%.The predicted fπ/mρand1/fρare close to12%below experiment,equivalent to about0.9times the corresponding statistical uncertainties,and f K/mρis below experiment by17%,equal to2.8times its statistical uncertainty.Although overall the predicted values could be considered in fair agreement with experiment,the result that three of four decay constants range from0.9to2.8 standard deviations below experiment suggests the possiblity that the true,underlying values,if determined without statistical uncertainties,would fall somewhat below experiment.One possible source of such disagreement is that we have assumed exact isospin symmetry and ignored electromagnetic effects.In particular,although our prediction for fπlies below measured values determined from charged pion decays, reported numbers for the neutral pion decay constant[3]are quite close to our result. The significance of the closer agreement between our predicted fπand the neutral pion decay constant is unclear to us,however,as a consequence of the systematic uncertainties arising in the experimental determination of the neutral pion decay1constant.Another possible source of disagreement between our numbers and experiment may be the valence approximation itself.A simple physical argument tends to support of this alternative[4].The valence approximation may be viewed as replacing the momentum and frequency dependent color dielectric constant arising from quark-antiquark vacuum polarization with its low momentum limit[5].At low momentum, then,the effective quark charge appearing in the valence approximation will agree with the low momentum effective charge of the full theory.The valence approximation might thus be expected to be fairly reliable for low-lying baryon and meson masses, which are determined largely by the long distance behavior of the chromoelectric field.The valence approximation’s effective quark charge at higher momentum can be obtained from the low momentum charge by the Callan-Symanzik equation.As a consequence of the absence of dynamical quark-antiquark vacuum polarization,the quark charge in the valence approximation will fall faster with momentum than it does in the full theory,at short distance the attractive quark-antiquark potential in the valence approximation will be weaker than in the full theory,and meson wave functions in the valence approximation will be pulled into the origin less than in the full theory.Since decay constants are proportional to the square of wave functions at the origin,decay constants in the valence approximation could be expected to be smaller than in the full theory.The calculations described here were done on the GF11parallel computer at IBM Research[6]and use the same collection of gauge configurations and quark propagators generated for the mass calculations of Ref.[1].The full set of mass and decay constant calculations took approximately one year to complete.GF11was used in configurations ranging from384to480processors,with sustained speeds ranging from5Gflops to7Gflops.With the present set of improved algorithms and480 processors,these calculations could be repeated in less than four months.22DEFINITIONSThe normalization we adopt for pseudoscalar and vector decay constants in continuum QCD is<0|Jµj(0)|V(p,ǫ,j)>=ǫµm j F j,<0|J5µj(0)|P(p,j)>=pµf j,for vector and pseudoscalar states,|V(p,ǫ,j)>and|P(p,j)>,respectively,normal-ized by<p|q>=(2π)3p0δ( p− q).Here j is aflavor-SU(3)octet index and vector and axial vectorflavor-SU(3)currents Jµj(x)and J5µj(x)are related to cannonical continuum quark and antiquarkfields,ψc(x)γµλjψc(x),J5µj(x)=.2Assuming exact isospin symmetry,we havef i=fπi=1,...3,f i=f K i=4,...7,F i=mρ√therefore,also been adopted for the present calculation.The smeared field φr ( x ,t )is related by the lattice antiquark field ψℓ( x ,t )byφr ( x ,t )= y G r ( x − y )ψℓ( y ,t ),G r ( z )=(√r 2).The fieldψℓ( x ,t ).We take the smeared fields φ0(x )andψ(x ),respectively.From these fields,define smearedcurrents by J 5jr (x )=φr (x )γµλj φr (x ),J 5µjr (x )=C P jr ′r (t )→Z A jr ′r ′′Measured in units of the lattice spacing a the decay constants f j a and F j a are then given,for any choice of smearing size r,by2(z A j Z AP j0r)2(f j a)2=.(2.9)m j aZ V jrrThe coefficients z A j and z V j arefinite renormalizations chosen so that the lattice cur-rents z A j a3J5µj0and z V j a3Jµj0approach the continuum currents J5µj and Jµj,respectively, as the lattice spacing approaches zero.These constants are often given the“naive”values=2k P j,z ANjz V N=2k V j.(2.10)jwhere k P j and k V j are the hopping constants corresponding to the mass of the quark and antiquark for a pseudoscalar or vector meson,respectively,withflavor j,assumed here to have m q=m,4k c3k V jz V0j=1−and z V1j z V0wherez A1=1−0.31α(1/a).(2.12)msDecay constants for mesons with m q=mFor the lattice83×32atβof5.70we calculated propagators only for source r and sink r′of size0.In all other cases we calculated propagators only for source size r of2.To determine decay constants,according to Eqs.(2.8)and(2.9),fits for the lattice83×32to propagators for the single sink of size0are sufficient,while for all other latticesfits are needed for sink sizes of both0and2.To determine the range of time separations to be used infitting for eachβand k value,we evaluated effective masses m P(t),m V(t)and m P′(t)byfitting C P r′r(t),C V r′r(t),and C A r′r(t),respectively, to Eqs.(2.4)-(2.6)at time separations t and t+1.The largest interval at large t showing an approximate plateau in an effective mass graph we chose as the initial trial range on which tofit each propatator to the corresponding asymptotic form of Eqs.(2.4)-(2.6).Similarly,the largest interval at large t showing an approximate plateau in a graph of C A r′r′′(t)/C P r′r(t)we chose as the initial trial range for afit to Eq.(2.7).Figures(1)-(12)show the plateaus at large t in the effective masses m P(t), m P′(t),the plateau at large t in the ratio C A02(t)/C P02(t),and the effective mass m V(t) for propagators with source r of2and sink r′of0.Figures(1)-(4)show results for the lattice163×32atβof5.70and the largest corresponding k value,0.1675. Figures(5)-(8)show results for the lattice243×36atβof5.93and the largest corresponding k,0.1581.Figures(9)-(12)show results for30×322×40atβof6.17 and the largest corresponding k,0.1532.Fits to data for a range of t were done by minimizing thefit’sχ2determined from the full correlation matrix for the data beingfit.An automaticfitting program repeatedly carried outfits on every connected interval of four or more points within the initial trialfitting range.Forfits to Eq.(2.7),which require only a singlefitting parameter,we looked at intervals of three or more points.Thefinalfitting range was chosen by the program to be the interval with the smallest value ofχ2per degree of freedom.Altough a variety of other criteria could be used to determine thefinal fitting range,an advantage of the method we adopted is that it can be implemented automatically thereby reducing the potential for biases.The reliability of our in our final extrapolated results depends to some degree on adopting a consistent choice of fitting ranges at different parameter values.The horizontal lines in Figures(1)-(12)show thefitted values of masses and7Z A02/Z P02,and the pair of vertical lines in eachfigure indicates the interval of t values inthefinalfitting range chosen.It is perhaps useful to mention that since the effectivemass at each t depends on data both at t and t+1,the effective mass shown at thehighest t within eachfinalfitting range depends on data outside thefitting range.Thus thefitted lines tend to approximate the average of the effective masses withinthefitting range but with the effective mass at highest t omitted.In all but onecase,the data shows clear plateaus at large t extending over more than four timevalues.These plateaus appear to befitted fairly reliably.For the rho propagator,C V02(t),on the lattice243×36,the plateau is more ambiguous.Thefit is made over the four t values from9to12,for which the three corresponding effective masses fallslowly.The effective mases at t of12to15then fall about one standard deviationbelow thefit.At t of16and17,the effective masses then return to the originalfittedvalue.This behavior is consistent with a statisticalfluctuation although it makes theidentication of the plateau at which tofit more parable ambiguitiesdo not occur elsewhere in our data.The interval chosen is approximately a rescalingof the rhofitting intervals of5to8,atβof5.70and13to16,atβof6.17.Also,therho decay constant obtained from thisfit in Section4is interpolates smoothly valuesobtained from less ambiguousfits at other values of quark mass and lattice spacing.If this point were simply elimated from our extrapolations of fρin quark mass andin lattice spacing,ourfinal continuum predictions would be nearly unchanged.Value ofχ2for thefits in Figures(1)-(12)are shown in Table2.Ourfitsfor sinks with size r′of2,fits at smaller k,andfits on the lattice243×32are of comparable quality to those shown and give comparableχ2.Statistical uncertainties of parameters obtained fromfits and of any function ofthese parameters were determined by the bootstrap method[7].From each ensembleof N gauge configurations,100bootstrap ensembles were generated.Each bootstrapensemble consists of a set of N gauge configurations randomly selected from theunderlying N member ensemble allowing repeats.For each bootstrap ensemble theentirefit was repeated,including a possibly new choice of thefinalfitting interval.The collection of100bootstrap ensembles thus yields a collection of100values of anyfitted parameter or any function of anyfitted parameter.The statistical uncertainty8of any parameter is taken to be half the difference between a value which is higher than all but15.9%of the bootstrap values and a value which is lower than all but 15.9%of the bootstrap values.In the limit of large N the collection of bootstrap values of a parameter p approaches a gaussian distribution and the definition we use for statistical uncertainty approaches the dispersion,d,given by√ms(1/a).For k c we used the values determined in Ref.[1].These are listed in Table(3).To determineαms(π/a)using the mean-field improved perturbation theory relation[8,4]1MS (π/a)=<T rU/3>ms(π/a)given by4π/g2ms(1/a).The corresponding values of z A1and z V1are shown in Table(4).9Values of the decay constant in lattice units fπa for the various lattices,βandk shown in Table(1)are listed in Tables(5)-(9).As a measure of the lattice spacingin each case,Table(10)gives the rho mass in lattice units mρ(m n)a,extrapolated tothe“normal”quark mass m n which produces the physical value of mπ(m n)/mρ(m n)[1].We list also values of m n itself.These parameters are not given for the lattice83×32since in this case we were not able to calculate propagators at small enough quark mass to perform the required extrapolation reliably.Thefinite renormalizationsfor the decay constants in Tables(5)-(9)all include both the leading term z A0ofEqs.(2.11)and thefirst order mean-field improved perturbative correction z A1ofEqs.(2.12).The second column in each table gives values of fπa found from Z A r′rdetermined from a directfit of C A r′r(t)to Eq.(2.6).The third column gives fπa foundfrom Z A r′r determined byfitting the ratio C A r′r(t)/C P r′r(t)to Eq.(2.7)and then usingthe value of Z P r′r found from afit of C P r′r(t)to Eq.(2.4).The two sets of data inall cases are statistically consistent and in all cases,except for the lattice83×32 for k below0.1500,fπdetermined from ratiofits has a smaller statistical error fπdetermined from directfits.The ratio method tends to give less statistical noise,in effect,because it uses C P r′r(t),which is relatively less noisey,to determine m P and then extracts only Z A r′r from the more noisey propagator C A r′r(t).The direct method extracts both Z A r′r and m P from C A r′r(t)yielding an m P with greater noise,which is then multiplied by a possibly large t and exponentiated,feeding additional noise back into the value of Z A r′r.In the remainder of this paper we use only values of fπdetermined from ratiofits.Values of the decay constant in lattice units Fρa for the lattices shown in Table(1)are listed in the fourth column of Tables(5)-(9).Thefinite renormalizationsfor Fρshown in these tables all include both the leading term z V0of Eqs.(2.11)andthefirst order mean-field improved perturbative correction z V1of Eqs.(2.12).Thevalues of Z V r′r used to determine Fρwere all extracted from directfits of C V r′r(t)to Eq.(2.5).105VOLUME DEPENDENCEPercentage changes in decay constants going from83×32and163×32to 243×32,atβof5.70,are given in Table11.These changes are the same for all choices offinite renormalization.All of the differences appear to be of marginal statistical significance and may therefore best be viewed as upper bounds on the volume dependence of our results.A variety of different arguments suggest that, for the range of k,β,and lattice volume we have examined,the errors in valence approximation decay constants due to calculation in afinite volume L3are bounded by an expression of the form Ce−L/R.A simple non-relativistic potential model gives this expression with the radius of a hadron’s wave function for R.A more elaborate field theory argument gives for R the Compton wave length of a pair of pions,which is the lightest state that can be exchanged between a pair of identical pseudoscalar or vector mesons.Atβof5.70,R is thus very likely to be between3and5lattice units. Since the changes in decay constants shown in Table11between163and243are all less than5%for k≥0.1650,it follows that the differences between these values in 243and in infinite volume should be less than1.3%.For extrapolations to physical quark masses,we use only decay constants with k≥0.1650.6QUARK MASS EXTRAPOLATIONAt the largest k on each lattice,the ratio mπ/mρis significantly larger than its experimentally observed value of0.179.Thus to produce decay constants for hadrons containing only light quarks,our data has to be extrapolated to larger k or,equivalently,to smaller quark mass.We did not calculate directly at larger k both because the algorithms we used tofind quark propagators became too slow and because the statistical errors we found in trial calculations became too large.Define the quark mass in lattice units m q a to bem q a=12k c,(6.1)where k c is the critical hopping constant at which mπbecomes zero.We found fπa11and Fρa,for all three possible choices offinite renormalization in Eqs.(2.10)-(2.12) to be nearly linear functions of m q a over the entire range of k considered on each lattice.Figure13shows fπand Fρ,withfinite renormalizations including thefirst order mean-field improved perturbative correction,as functions of m q.Data is shown from the three lattices of Table1which we use to evaluate continuum limits,163×32, 243×36and30×322×40.For convenience,decay constants at eachβare shownin units of the central value of the rho mass mρ(m n),at the sameβ,extrapolated to the the“normal”quark mass m n.Quark masses m q for eachβare shown in units of the central value of the strange quark mass m s at the sameβ.The data Figure13has been scaled by the central values of mρ(m n)and m s taken as arbitrary external parameters,and the error bars shown do not include the effect of statistical fluctuations in mρ(m n)or m s.Table(10)gives values of m s found in Ref.[1]by requiring mπ[(m n+m s)/2]/mρ(m n)to be equal to the physical value of m K/mρ. The lines in Figure13arefits of decay constants measured in lattice units,fπa and Fρa,to linear functions of the quark mass in lattice units,m q a,at the three smallest quark masses in the data set at eachβ.Thesefits were obtained by minimizingχ2 obtained form the full correlation matrix amoung the data points.The correlation matrix was calculated by taking averages of data values and products of data values over bootstrap ensembles generated as described in Section(3).Theχ2for these fits,and correspondingfits on the lattice243×32atβof5.70,are given in Table 12.Thefits in Figure13appear to be quite good and provide,we believe,a reliable method for extrapolating decay constants down to light quark masses.With naive finite renormalization,Eq.(2.10),or zero order mean-fieldfinite renormalization, Eq.(2.11),fπa and Fρafit straight lines in m q a about as well as thefirst order perturbatively renormalized data of Figure13.The linearfits of Figure13permit the determination of f K and Fφin addition to fπand Fρ.For a pion composed of a quark and antiquark with mass m q=mmqCharge conjugation invariance then givesαq=α=m n,will have the same decay constant as aqpion composed of a single type of quark and antiquark with m q=m=m s which,in the valence approximation,gives Fφ.Tables13andq14give the value of fπa,f K a,Fρa and Fφa obtained from thefits in Figure13.The statistical uncertainties in these quantities were obtained by a further application of the bootstrap method of Section(3).Bootstrap ensembles of the underlying gauge configurations were generated,and on each bootstrap ensemble the extrapolated decay constants were recalculated.The uncertainty in each decay constant was obtained from the resulting distribution.The correlation matrices used tofit bootstrap data to linear functions of m q a were taken to be the same as the correlation matrices for the full ensemble.To recalulate correlation matrices separately on each bootstrap ensemble by a further bootstrap would have been too time consuming.7CONTINUUM LIMITThe ratios fπ/mρ,f K/mρ,Fρ/mρand Fφ/mρfor physical quark masses we then extrapolated to zero lattice spacing.For Wilson fermions the leading asymp-totic lattice spacing dependence in these decay ratios is expected to be linear in a. On the other hand,as shown in Ref.[1],mρa follows the two-loop Callan-Symanzik scaling prediction inαcase is nearly the same.For lattice period L,the quantity mρL is respectively,9.08±0.13,9.24±0.19and,averaged over three directions,8.67±0.12[1].Thefits shown in Figure14were found by minimizingχ2obtained from the full correlation matrix among thefitted data.Since both the x and y coordinates of each of the threefitted points on each line have statistical uncertainties,we evaluated χ2among all six pieces of data and chose asfitting parameters the slope and inter-cept of the line along with the x coordinate of each point.The required correlation matrices were found by the bootstrap method as were the statistical uncertainties of the extrapolated predictions.The correlation matrices used infits for each bootstrap ensemble were again taken from the full ensemble and not recalculated on each boot-strap ensemble independently.Theχ2per degree of freedom for thefits in Figure14, are given in Table15.For the lattice spacing dependence of decay ratios found using zeroth order mean-fieldfinite renormalization and using naivefinite renormalization we also made fits to linear functions of mρa.The slopes with respect to mρa of the ratios fπ/mρ, f K/mρ,Fρ/mρand Fφ/mρalong withχ2for eachfit are also given in Table15.It is clear from these results,as menitioned in Section2,that naive renormalization leads to decay ratios with significantly stronger lattice spacing dependence than found for either zeroth orfirst order improved perturbative renormalization.There also appears to be some tendency forfirst order perturbative renormalization to lead to weaker lat-tice spacing dependence than zeroth order.It follows that the extrapolations we have done to zero lattice spacing are likely to be most reliable forfirst order perturbative renormalzation and least reliable for naive renormalization.8INFINITE VOLUME LIMITThe continuum ratios we found infinite volume were then corrected to infinite volume by an adaptation of the method used in Ref.[1]to correctfinite volume continuum mass ratios to infinite volume.From fπ/mρ,withfirst order perturbative renormalization,as a function of lattice spacing a and lattice period L,both measured14in physical units,define thefinite volume correction term∆(a,L)to be∆(a,L)=fπmρ(a,L).(8.1)The quantity which we would like to determine is∆(0,9/mρ).Now the ratio fπ/mρ, for L of9/mρ,undergoes a relative change of a bit less than20%as a goes from its value a5.7atβof5.70to0.Thus we would expect an error of about20%of∆(0,9/mρ) for the approximation∆(0,9mρ).(8.2)Moreover,from our earlier discussion of the exponential approach of decay ratios to their infinite volume limits,it follows that with an additional error of about20%of ∆(0,9/mρ)we have∆(a5.7,9mρ(a5.7,13.5mρ(a5.7,9mρ(0,∞)≈fπmρ)+fπmρ)−fπmρ).(8.4)The error in this approximation should be less than about40%of6%of fπ/mρ, which is2.4%of fπ/mρ.Table16lists estimates of the systematic uncertainties in equations corresponding to Eq.(8.4)for other decay constants and other choices offinite renormalization.First order perturbative renormalization consistently gives the smallest systematic error in volume correction largely because the lattice spacing dependence of these decay ratios is smallest.The ratios fπ/mρ,f K/mρ,Fρ/mρand Fφ/mρ,for all three different choices of finite renormalization,extrapolated to zero lattice spacing with mρLfixed at9,and then corrected to infinite volume are shown in Table17.Forfirst order perturbative15renormalization we also give,in Table18,finite and infinite volume values of the ratios fπ/f K,Fρ/f K and Fφ/f K.The errors shown for infinite volume ratios are statistical only and do not include the estimates we have just given for the systematic error in our procedure for making infinite volume corrections.Forfirst order perturbative finite renormalization,a comparision of Tables16and17shows,however,that the systematic errors arising from our method of obtaining infinite volume results are much smaller than the statistical errors.For the range ofβused in our extrapolation to zero lattice spacing,first order mean-field theory improved perturbation expan-sions have been shown[8]to work quite well for a wide variety of different quantities. In addition,as we mentioned earlier,the extrapolation to zero lattice spacing should be most reliable for this renormalization scheme.Thus we believe that the num-bers in Table17obtained usingfirst order mean-field perturbative renormalization are significantly more reliable than those found using the other two renormalization methods.Zeroth order perturbative renormalization has been included,however,to provided some measure of the degree to which our results may be sensitive to the choice of renormalization.Half of the difference betweenfirst order and zeroth order mean-field perturbative renormalization appears to us to be a conservative estimate of the systematic uncertainty in thefirst order results arising from the missing sec-ond and higher order perturbative renormalization contributions.In all cases this uncertainty is significantly less than the statistical errors.Predictions obtained with naive renormalization have been included in Table17largely as a curiosity.It is interesting to notice,however,that the difference between thefinal,infinite volume results found with naive renormalization and those found withfirst order perturbative renormalization is still less than1.5times the naive renormalization statistical errors.The predicted infinite volume ratios in Table17are all statistically consistent with the correspondingfinite volume ratios.The main consequence of the correction to infinite volume is an increase in the size of the statistical uncertainty in each prediction.The experimental numbers shown in Table17for fπand f K are from charged particle decays and for fρfrom neutral decays.In all cases the uncertainties in the experimental values are0.001or less.As mentioned in the introduction,an16experimental value for the neutral pion decay[3]gives fπ/mρof0.110±0.005,which is quite close to our prediction.The systematic uncertainties in this experimental number,however,are larger than those for the charged pion decay.As a result the significance of the improved agreement of our prediction with the observed neutral pion decay constant is unclear to us.We would like to thank Paul Mackenzie for discussions,and Mike Cassera, Molly Elliott,Dave George,Chi Chai Huang and Ed Nowicki for their work on GF11. We are particularly grateful to Chris Sachrajda for calling our attention to an error in an earlier version of this paper.References[1]F.Butler,H.Chen,J.Sexton,A.Vaccarino and D.Weingarten,Physical ReviewLetters70,2849(1993).[2]D.Toussaint,Nucl.Phys.B(Proc.Suppl.)26(1992)3.[3]H.-J.Behrend et al.,Z.Phys.C49(1991)401.[4]A.X.El-Khadra,G.H.Hockney,A.S.Kronfeld and P.B.Mackenzie,Phys.Rev.Letts.69(1992)729.[5]D.H.Weingarten,Phys.Lett.109B,57(1982);Nuclear Physics,B215[FS7],1(1983).[6]D.Weingarten,Nucl.Phys.B(Proc.Suppl.)17(1990)272.[7]B.Efron,The Jacknife,the Bootstrap and Other Resampling Plans,Society forIndustrial and Applied Mathematics,Philadelphia,1982.[8]G.P.Lepage and P.B.Mackenzie,to appear in Physical Review D48(1993).1783×32 5.700.1400-0.165010002439163×32 5.700.1600-0.16752000219243×32 5.700.1600-0.1675400058243×36 5.930.1543-0.158******** 30×322×40 6.170.1500-0.153********latticeβC P02C A02C A02/C P02C V02Table2:Values ofχ2per degree of freedom forfits to propagators.83×32 5.700.169012±0.000102163×32 5.700.169405±0.000052243×32 5.700.169304±0.000035243×36 5.930.158948±0.00002630×322×40 6.170.153763±0.000018。