arXiv:cond-mat/0201310v1 [cond-mat.stat-mech] 17 Jan 2002Equation of state for systems with Goldstone bosons Massimo CampostriniaMartin HasenbuschbAndrea Pelissettoc Paolo RossiaEttore Vicaria aDipartimento di Fisica, Universit`a di Pisa, and INFN, Sez. di Pisa, I-56126 Pisa, Italy bNIC/DESY Zeuthen, Platanenallee 6, D-51738 Zeuthen, Germany cDipartimento di Fisica, Universit`a di Roma La Sapienza, and INFN, Sez. di Roma I, I-00185 Roma, Italy Abstract We discuss some recent determinations of the equation of state for the XY and the Heisenberg universality class. Key words:Heisenberg model, XY model, equation of state, critical behavior PACS:67.40.-w, 64.60.Fr, 11.15.Me, 05.70.Jk In the last few years there has been a significant progress in the determination of the critical properties ofO(N) models; see, e.g., Ref. [1] for a comprehen- sive review. First of all, high-precision estimates of critical exponents and of several high-temperature universal ratios have been obtained by usingim- provedHamiltonians. Improved models are such that the leading nonanalytic correction is absent in the expansion of any thermodynamic quantity near the critical point. The idea is quite old [2–4]. However, the early attempts that used high-temperature techniques were not able to determine improved models with high accuracy, so that final results did not significantly improve the estimates of standard analyses. Recently [5–12], it has been realized that Monte Carlo simulations using finite-size scaling techniques are very effective for this purpose, obtaining accurate determinations of several improved mod- els in the Ising, XY, and O(3) universality class. Once an improved models is accurately determined, one can use standard high-temperature techniques in order to obtain very precise determinations of the critical exponents. For instance, for the experimentally relevant cases, we obtained [13,14,10,12]: γ= 1.2373(2),ν= 0.63012(16),N= 1, Preprint submitted to Elsevier Science1 November 2018
γ= 1.3177(5),ν= 0.67155(27),N= 2, γ= 1.3960(9),ν= 0.7112(5),N= 3. Beside the critical exponents, experiments may determine several other univer- sal properties. We consider here the equation of state that relates the magnetic field~H, the magnetization~M, and the reduced temperaturet≡(T−Tc)/Tc. In a neighborhood of the critical pointt= 0,~H= 0, it can be written in the scaling form ~H= (Bc)−δ~M Mδ−1f(x),x≡t(M/B)−1/β,(1) whereBcandBare the amplitudes of the magnetization on the critical isotherm and on the coexistence curve, M=BcH1/δt= 0.(2) M=B(−t)βH= 0, t <0.(3) With these choices, the coexistence line corresponds tox=−1, andf(−1) = 0,f(0) = 1. Alternatively, one can write ~H=k1 ~M M|t|βδF±(|z|),z≡k2Mt−β,(4) whereF+(z) applies fort >0 andF−(|z|) fort <0. The constantsk1andk2 are fixed by requiring F+(z) =z+ 1 6z3+∑ n=3 r2n (2n−1)!z2n−1(5) forz→0 in the high-temperature phase. The behavior of the functionsf(x) andF−(|z|) at the coexistence curve depends crucially onN. ForN= 1 they vanish linearly. On the other hand, forN≥2, the presence of the Goldstone modes implies in three dimensions [15–19]: f(x)≈cf(1 +x)2.(6) The nature of the corrections to this behavior is not clear [17–20]. In particular, logarithmic terms are expected [20]. In order to obtain approximations of the equation of state, we parametrize the thermodynamic variables in terms of two parametersθandR: M=m0Rβm(θ),t=R(1−θ2),H=h0Rβδh(θ).(7) 2
-1.0-0.50.00.51.0 x 0 1 2 f(x) n=1 (A) n=1 (B) MC -1.0-0.50.00.51.0 x 0 1 2 f(x) n=0 n=1 (A) n=1 (B) Fig. 1. Graph of the functionf(x) forN= 2 (left) andN= 3 (right). ForN= 2 we also report the Monte Carlo results of Ref. [27]. Here,m0andh0are nonuniversal constants,m(θ) andh(θ) are odd functions ofθ, normalized so thatm(θ) =θ+O(θ3) andh(θ) =θ+O(θ3). The variable Ris nonnegative and measures the distance from the critical point in the (t, H) plane, while the variableθparametrizes the displacement along the lines of constantR. In particular,θ= 0 corresponds to the high-temperature linet >0,H= 0,θ= 1 to the critical isothermt= 0, andθ=θ0, where θ0is the smallest positive zero ofh(θ)—it must satisfy of courseθ0>1—to the coexistence line. Such a mapping has been extensively used in the Ising case and provides accurate approximations if one uses low-order polynomials form(θ) andh(θ) [21–24,13,25]. In systems with Goldstone bosons we must additionally ensure the condition (6). For this purpose, it is enough to require [26]h(θ)∼(θ0−θ)2forθ→θ0. In Refs. [26,10,12] we obtained the equation of state in the scaling limit by using two different approximation schemes for the functionsm(θ) andh(θ): scheme(A) :m(θ) =θ ( 1 + n∑ i=1 ciθ2i ) , h(θ) =θ( 1−θ2/θ2 0 )2 ,(8) scheme(B) :m(θ) =θ, h(θ) =θ( 1−θ2/θ2 0 )2 ( 1 + n∑ i=1 ciθ2i ) .(9) The constantsciandθ0were fixed by requiringF+(z) to have the expan- sion (5), with the coefficients determined by high-temperature expansion tech- niques. Since we were able to compute accurately onlyr6andr8, we used the two schemes forn= 0 andn= 1. The results, especially those forN= 3, see Fig. 1, are quite independent of the scheme used, indicating the good convergence of the method. 3
By using the equation of state, one can determine several amplitude ratios. We mention here the experimentally relevant U0=A+ A−,Rχ=C+Bδ−1 Bδ c ,(10) whereC+andA±are related to the critical behavior of the susceptibilityχ and of the specific heatCforH= 0: χ=C+t−γ,t >0, C=A±(±t)−α+B±t >0. Using the approximate equation of state we obtain [14,10,12]:U0= 1.062(4), Rχ= 1.35(7) forN= 2 andU0= 1.57(4),Rχ= 1.33(8) forN= 3. ForN= 3 the approximate equation of state can be compared with experi- ments. We observe good qualitative and quantitative agreement. ForN= 2 we can use our results to predict critical properties of theλ-transition in4He. In this case, the equation of state does not have a direct physical meaning, but we can still compare the predictions for the singular specific-heat ratio U0. A precise determination of the exponentαand ofU0was done recently by means of a calorimetric experiment in microgravity [28] (see also reference 4 in Ref. [10]) obtainingα=−0.01056(38) andU0≈1.0442. The result for U0is lower than the theoretical one. This is strictly related to the disagree- ment in the value ofα(see also Ref. [29]). Indeed, using hyperscaling we find α=−0.0146(8), that significantly differs from the experimental estimate. The origin of this discrepancy is unclear and further theoretical and experimental investigations are needed. A new generation of experiments in microgravity environment that is currently in preparation should clarify the issue on the experimental side [30]. References [1]A. Pelissetto and E. Vicari, cond-mat/0012164. [2]J.-H. Chen, M.E. Fisher, and B.G. Nickel, Phys. Rev. Lett.48(1982) 630. [3]M.J. George and J.J. Rehr, Phys. Rev. Lett.53(1984) 2063. [4]M.E. Fisher and J.H. Chen, J. Physique46(1985) 1645. [5]H.G. Ballesteros, L.A. Fern´andez, V. Mart´ın-Mayor, and A. Mu˜noz Sudupe, Phys. Lett. B441(1998) 330. [6]M. Hasenbusch, K. Pinn, and S. Vinti, Phys. Rev. B59(1999) 11471. 4
[7]H.G.Ballesteros,L.A.Fern´andez,V.Mart´ın-Mayor,A.Mu˜nozSudupe, G. Parisi, and J.J. Ruiz-Lorenzo, J. Phys. A32(1999) 1. [8]M. Hasenbusch, J. Phys. A32(1999) 4851. [9]M. Hasenbusch and T. T¨or¨ok, J. Phys. A32(1999) 6361. [10] M. Campostrini, M. Hasenbusch, A. Pelissetto, P. Rossi, and E. Vicari, Phys. Rev. B63(2001) 214503. [11] M. Hasenbusch, J. Phys. A34(2001) 8221. [12] M. Campostrini, M. Hasenbusch, A. Pelissetto, P. Rossi, and E. Vicari, cond- mat/0110336. [13] M. Campostrini, A. Pelissetto, P. Rossi, and E. Vicari, Phys. Rev. E60(1999) 3526; cond-mat/0201180. [14] M. Campostrini, A. Pelissetto, P. Rossi, and E. Vicari, Phys. Rev. B61(2000) 5905. [15] E. Br´ezin and D.J. Wallace, Phys. Rev. B7(1973) 1967. [16] E. Br´ezin and J. Zinn-Justin, Phys. Rev. B14(1976) 3110. [17] D.J. Wallace and R.P.K. Zia, Phys. Rev. B12(1975) 5340. [18] L. Sch¨afer and H. Horner, Z. Phys. B29(1978) 251. [19] I.D. Lawrie, J. Phys. A14(1981) 2489. [20] A. Pelissetto and E. Vicari, Nucl. Phys. B540(1999) 639. [21] P. Schofield, Phys. Rev. Lett.22(1969) 606. [22] P. Schofield, J.D. Lister, and J.T. Ho, Phys. Rev. Lett.23(1969) 1098. [23] B.D. Josephson, J. Phys. C: Solid State Phys.2(1969) 1113. [24] R. Guida and J. Zinn-Justin, Nucl. Phys. B489(1997) 626. [25] M. Caselle, M. Hasenbusch, A. Pelissetto, and E. Vicari, J. Phys. A34(2001) 2923. [26] M. Campostrini, A. Pelissetto, P. Rossi, and E. Vicari, Phys. Rev. B62(2000) 5843. [27] J. Engels, S. Holtmann, T. Mendes, and T. Schulze, Phys. Lett. B492(2000) 219. [28] J.A. Lipaet al., Phys. Rev. Lett.76(1996) 944;84(2000) 4894. [29] S. Holtmann, J. Engels, and T. Schulze, hep-lat/0109013. [30] J.A. Nissen, D.R. Swanson, Z.K. Geng, V. Dohm, U.E. Israelsson, M.J. DiPirro, and J.A. Lipa, Low Temp. Phys.24(1998) 86. 5
100%
Scan to connect with one of our mobile apps
Coinbase Wallet app
Connect with your self-custody wallet
Coinbase app
Connect with your Coinbase account
Open Coinbase Wallet app
Tap Scan
Or try the Coinbase Wallet browser extension
Connect with dapps with just one click on your desktop browser
Add an additional layer of security by using a supported Ledger hardware wallet