ebook img

Temperature dependence of thermal conductivities of coupled rotator lattice and the momentum diffusion in standard map PDF

0.22 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Temperature dependence of thermal conductivities of coupled rotator lattice and the momentum diffusion in standard map

Temperature dependence of thermal conductivities of coupled rotator lattice and the momentum diffusion in standard map Yunyun Li,1 Nianbei Li,1,∗ and Baowen Li1,2,3,† 1Center for Phononics and Thermal Energy Science and School of Physics Science and Engineering, Tongji University, 200092 Shanghai, People’s Republic of China 2Department of Physics and Centre for Computational Science and Engineering, National University of Singapore, Singapore 117546, Republic of Singapore 5 3NUS Graduate School for Integrative Sciences and Engineering, Singapore 117456, Republic of Singapore 1 0 Incontrarytoother1Dmomentum-conservinglatticessuchastheFermi-Pasta-Ulamβ (FPU-β) 2 lattice,the1Dcoupledrotatorlatticeisanotableexceptionwhichconservestotalmomentumwhile n exhibitsnormalheatconductionbehavior. Thetemperaturebehaviorofthethermalconductivities a of 1D coupled rotator lattice had been studied in previous works trying to reveal the underlying J physical mechanism for normal heat conduction. However, two different temperature behaviors of thermal conductivities have been claimed for the same coupled rotator lattice. These different 9 2 temperature behaviors also intrigue the debate whether there is a phase transition of thermal con- ductivities as thefunction of temperature. In this work, we will revisit the temperature dependent ] thermalconductivitiesforthe1Dcoupledrotatorlattice. Wefindthatthetemperaturedependence h follows a power law behavior which is different with the previously found temperature behaviors. c Our results also support the claim that there is no phase transition for 1D coupled rotator lattice. e We also give some discussion about the similarity of diffusion behaviors between the 1D coupled m rotator lattice and the single kicked rotator also called theChirikov standard map. - t a PACSnumbers: 05.60.-k,44.10.+i,05.45.-a t s . t I. INTRODUCTION pendence of thermal conductivity of κ(T) ∝ e−T/A was a m claimed where A is a positive constant. There it was also argued that a possible phase transition at temper- - The exploration of underlying mechanism for anoma- d ature around T ∼ 0.2 − 0.3 where heat conduction is lousandnormalheatconductioninlowdimensionalsys- n normal above this temperature while anomalous below tems represents a huge challenge in the area of statisti- o this temperature exists [62]. Although this phase transi- c cal physics [1–4]. After enormous efforts for more than tion statement was challenged as a finite size effect [63], [ onedecadefromnumericalsimulations[5–44],theoretical a later work supported this phase transition conjecture predictions[45–57]andexperimentalobservations[58–60], 1 after deriving a similar temperature dependent thermal v there is still no consensus for the exact physical mecha- conductivity as κ(T)∝e−T/A [50]. 9 nism causing anomalous heat conduction. It is believed 1 that momentum conservation plays an important role Most recently, the 1D coupled rotator lattice re- 4 in determining the actual heat conduction behavior. In attracts much attention due to the new finding of simul- 7 general,1D non-integrablelattices with momentum con- taneouslyexistingnormaldiffusionofmomentumaswell 0 servingpropertyshouldhaveanomalousheatconduction as heat energy [64]. It is argued that the normal be- 1. where the thermal conductivity κ diverges with the lat- haviormight be due to the reducednumber ofconserved 0 tice size N as κ ∝ Nα where 0< α < 1 [1–4]. However, quantitieswhichisuniqueforperiodicinteractionpoten- 5 the 1Dcoupledrotatorlattice is a wellknownexception. tials [65, 66]. But the new debate is that whether the 1 It exhibits normal heat conduction behavior despite its stretch is conservedor not in 1D coupled rotator lattice. : v momentum conserving nature [61, 62]. The investigation of 1D coupled rotator lattice will be i the key to unravel the true mechanism behind the con- X The normal heat conduction in 1D coupled rotator nection between momentum conservation and normal or r lattice was discovered via numerical simulations by two anomalous heat conduction. Therefore, it is the right a groups independently [61, 62]. In order to understand time to revisit the temperature dependence of thermal the underlying mechanism, the temperature dependence conductivity of 1D coupled rotatorlattice as a first step. of thermal conductivity has been studied in detail in bothworks. InRef. [61], the temperature dependence of In this work, we will revisit the temperature depen- thermal conductivity of coupled rotator lattice has been dence of thermal conductivities for the 1D coupled rota- found to be like κ(T) ∝ e∆V/T where ∆V is a posi- tor lattice. We find that the temperature dependence is tive constant. However, in Ref. [62], a temperature de- neither κ(T) ∝ e∆V/T nor κ(T) ∝ e−T/A as previously claimed [61, 62]. The actual temperature dependence is a power-law dependence as κ(T) ∝ T−3.2. The possible connectionwiththemomentumdiffusionofsinglekicked ∗Electronicaddress: [email protected] rotator or the Chirikov standard map has also been dis- †Electronicaddress: [email protected] cussed. In order to determine whether there is a phase 2 transition, we also present the temperature dependent thermal conductance at different system sizes. All the 0.11 thermal conductances for different sizes collapse to the T= 0.1 same value at low temperatures while approach to the Ti power-law behavior as κ(T) ∝ T−3.3 at high tempera- 0.10 tures. However, the crossover temperature decreases as thesystemsizeincreases. Thisfactindicatesthatthereis (a) no phase transitionbetween normaland anomalousheat 0.09 0.0 0.5 1.0 conduction. In thermodynamical limit, the heat conduc- i/ N 0.9 tionisnormalforalltemperaturesexceptthetrivialzero N=50 T= 0.8 N=100 temperature point. In Sec. II the lattice model will be N=200 introduced. Numerical results and discussions will be Ti N=400 0.8 N=800 presented in Sec. III and the results will be summarized N=1600 in Sec. IV. (b) 0.7 0.0 0.5 1.0 II. MODEL i/N FIG. 1: (color online). Temperature profiles for the 1D cou- pled rotator lattice at (a) T = 0.1 and (b) T = 0.8. The The Hamiltonian for the 1D coupled rotator lattice is lattice sizes are N = 50,100,200,400,800 and 1600 and the defined as the following [61, 62]: color ruleof thelines arethesamefor both (a) and(b). The firstandlast atom areputintocontactwith aLangevinheat N p2 bathwithtemperaturesetasT =T(1±∆)where∆=0.1 H = i +K[1−cos(q −q )] (1) L/R (cid:20) 2 i+1 i (cid:21) here. Xi=1 whereq andp denotethedisplacementfromequilibrium i i This is because the kinetic energy is proportional to the and momentum for i-th atom, respectively. The mass of temperatureas p2 =T andthepotentialenergyinEq. theatommandtheBoltzmannconstantk hasbeenset i B (1) is confined b(cid:10)y t(cid:11)he cosine function. intounity. TheparameterK withenergydimensionrep- Withoutgettingintonumerics,wecangetaqualitative resentsthecouplingstrengthoftheinter-atompotential. picture of the thermal conductivity of 1D coupled rota- Therefore, the system temperature T can be rescaled by tor lattice as the function of temperature. The thermal T/K and we can also set K =1 for simplicity [67]. The conductivity κ(T) will diverge in the low temperature equations of motion for i-th atom are limit approaching to the harmonic limit and will decay to zero in the high temperature limit as approaching to q˙ = p i i the unconnected N free particles. As a general picture, p˙i = Ksin(qi+1−qi)−Ksin(qi−qi−1) (2) the thermal conductivity κ(T) will decrease as the tem- perature increases. At low temperature limit, the displacements are small In non-equilibrium numerical simulations, we put the values so that |q −q | ≪ 1. The Hamiltonian can be i+1 i firstandlastatomintocontactwithLangevinheatbath. expanded into the Taylor series as: The temperatures for left and right heat bath are set as T =T(1±∆) where T denotes the averagetempera- H = N p2i + (qi+1−qi)2 tuLr/eRand ∆ = 0.1 gives rise to the temperature gradient (cid:20) 2 2 along the lattice. The fixed boundary conditions with Xi=1 q =q =0 are also been applied. (q −q )4 (q −q )6 0 N+1 i+1 i i+1 i − + −... (3) 4! 6! (cid:21) III. RESULTS AND DISCUSSIONS This Hamiltonian will approach to the integrable Har- monic lattice only at zero temperature T = 0. The first Before we discuss the results of thermal conductivi- stable nonlinear potential term will be the sextic poten- ties, we first show the temperature profiles. In Fig. 1, tial. the temperature profiles at two different temperatures On the other hand, at infinite high temperature limit T = 0.1 and T = 0.8 are plotted. The lattice sizes are T =∞,the rotatorlattice willapproachtoanotherinte- chosen as N = 50,100,200,400,800and 1600. For rela- grable system consisting of N independent and isolated tive low temperature at T =0.1, the temperature jumps free particles as at two boundaries are obvious. However, the tempera- ture jumps are reduced with the increase of lattice size N p2 H = i (4) N as can be seen in Fig. 1(a). In Fig. 1(b) where the Xi=1 2 temperature is relatively high at T = 0.8, all the tem- 3 100 N=1600 - 3.2 10 ~ T 10 - 3.2 ~ T N=50 1 N=100 1 N=200 N=400 N=800 N=1600 0.1 0.1 0.01 0.1 1 1 2 T T FIG. 2: (color online). Thermal conductivities κ(T) as FIG. 3: (color online). Thermal conductivities κ(T) as the the function of temperature for different lattice sizes N = function of temperature for lattice size N = 1600. The data 50,100,200,400,800 and 1600. The straight line is propor- aretakenfromFig. 2andcanbebestfittedtobeapower-law tional to T−3.2 which describes the temperature behavior of dependenceas κ(T)∝T−3.2. κ(T) at high temperatureregion very well. more than two orders of magnitudes for the κ(T) value peratureprofilescollapseto the samestraightline asthe as shown in Fig. 3. This also explains the poor fitting temperature jumps are very small for all lattice sizes. of the κ(T) ∝ e∆V/T dependence in Ref. [61] and the Inordertoobtainthetemperaturedependenceofther- κ(T)∝e−T/A dependence in Ref. [62]. malconductivities,weneedfirstdefinethewayhowκ(T) Thispower-lawdependenceofκ(T)∝T−3.2cannotbe can be calculated numerically. We notice that the tem- explained by the effective phonon theory which is able perature profiles all collapse to the same straight line if to predict the temperature dependent thermal conduc- the temperature jumps can be ignored at high tempera- tivities for other typical 1D nonlinear lattices such as tures or long lattice sizes. It is appropriate to define the FPU-β lattice and generalized nonlinear Klein-Gordon thermal conductivity κ(T) as: lattices [68–75]. According to the effective phonon the- ory,thethermalconductivityforlowtemperaturerotator JN lattice with Hamiltonian of Eq. (3) can be derived as κ(T)= (5) T −T L R 1 κ(T)∝ ∝T−2 (6) whereJ =hJiiistheaverageheatfluxalongthelatticein ǫ the stationary state and the local heat flux J is defined i where ǫ is the nonlinearity strength with the following as J = −p Ksin(q −q ) via the energy continuity i i i+1 i temperature dependence equation. InFig. 2, the thermalconductivities κ(T)asthe func- (q −q )6 tion of temperature are plotted for different lattice sizes ǫ∝ i+1 i ∝T2 (7) (cid:10)h(q −q )2i(cid:11) N =50,100,200,400,800and1600. Athightemperature i+1 i region, all the thermal conductivities κ(T) for different at low temperature region. Here we consider the sextic lattice sizes collapse together indicating the saturation potential term as the lowest nonlinear term because the of thermal conductivities as the increase of lattice size. dynamicsgovernedbythenegativequarticpotentialterm This is characteristic for lattices with normal heat con- is unstable. Therefore, the actual temperature behavior duction. As the temperature decreases,the thermalcon- ofκ(T)∝T−3.2 is steeper thanthe prediction ofκ(T)∝ ductivities first increases and then becomes flat. This is T−2 from effective phonon theory. because the phonon mean free paths are getting longer Although it is difficult to unravel the exact physical as the temperature reduces and the ballistic regime will mechanism behind the power-law dependence of ther- be approached if the phonon mean free paths are longer mal conductivities for coupled rotator lattice, it is very thanthelattice size. Theboundaryjumps willdominate helpful tolook intothe transportpropertiesofthe single the temperature profiles as in Fig. 1(a) and the defini- kicked rotator (the Chirikov standard map) [76]. As the tion of κ(T) of Eq. (5) will no longer reflect the actual name indicates, the coupled rotator lattice is a kind of thermal conductivities. connected kicked rotators. The equations of motion for At high temperatures, it is clearly seen that the ther- the single kicked rotator are mal conductivities follows a power-law dependence as κ(T)∝T−3.2. For the longestsize we consideredhere as pn+1−pn = Ksin(qn) N =1600,this power-lawbehaviorcanbe best fitted for q −q = p (8) n+1 n n+1 4 -1 10 10 -2 DP 1 - 3.2 10 ~ T N=50 -3 N=100 10 N=200 - 3.2 0.1 N=400 ~ T N=800 -4 N=1600 10 0.01 1 2 3 0.01 0.1 1 T T FIG. 4: (color online). Momentum diffusion constant DP as FIG. 5: (color online). Heat conductance σ as the the function of temperature for 1D coupled rotator lattice. function of temperature T for different lattice size N = ThenumericaldataareobtainedviaequilibriumMDsimula- 50,100,200,400,800 and 1600. All the other parameters are tions asin Ref. [64]. Thesolid line of T−3.2 is guided for the thesame as in Fig. 2. Theleft-pointing arrow representsthe eye. trend of decreasing crossover temperature as the lattice size increases. where q and p denote the coordinate and momentum n n after n-th kick. As a consistence check, we plot the momentum diffu- The variation of momentum p is unbounded and can sion constant D for the coupled rotator lattice as the P be characterized by a normal diffusion [76, 77] functionoftemperatureinFig. 4. Thesametemperature ∆p2(t) ∼Dt (9) dependence of DP ∝T−3.2 has been obtained. Therefore, the prediction of κ(T) ∝ T−3 with anal- (cid:10) (cid:11) where D is the diffusion constant. This is similar to the ogy to the single kicked rotator at high temperature or normal momentum diffusion for 1D coupled rotator lat- low K region is close to the numerical observation of tice. Accordingto Ref. [76,77],the diffusionconstantD κ(T)∝T−3.2. Thisindicatestheremightbesomedeeper depends on the coupling strength K as connectionbetweenthediffusionbehaviorofsinglekicked rotator and 1D coupled rotator lattice. In addition, the D ∝ (K−1.2)3, 1.2<K <4 (10) analogy analysis also predicts a κ(T)∝T−2 behavior at D ∝ K2, K >4 (11) low temperature region which is the same as the predic- tionfromtheeffectivephonontheoryofEq. (6). Therea- where 1.2 is the chaos threshold for the kicked rotator. sonwhythistemperaturebehaviorcannotbeobservedin Ifoneassumethetransportpropertiesofcoupledrota- numerical simulations might be due to the severe finite tor lattice are the same as that of single kicked rotator, size effect at low temperature region as can be seen in one would expect that the energy diffusion constant D E Fig. 1(a). as well as the momentum diffusion constant D of cou- P pled rotator lattice should also follow the same depen- In the final part we will discuss the issue about the dence as that of single kicked rotator as in Eq. (10). To possiblephasetransitionbetweennormalandanomalous translatetheK dependenceintoT dependence,wenotice heat conduction for 1D coupled rotator lattice. As we that the parameter K in single kicked rotator of Eq. (8) haveshowninFig. 3,theactualtemperaturedependence plays the same role as that in the coupled rotatorlattice ofκ(T)isapower-lawdependenceasκ(T)∝T−3.2. This asin Eq. (2). As wehavediscussedabove,the dynamics indicatesthe analyticderivationofκ(T)∝e−T/A inRef. ofcoupledrotatorlatticecanberescaledwithT/K. The [50] originally for a lattice with pinned on-site potential K dependence should be inversely proportionalto the T can not be extended to 1D coupled rotator lattice. dependence. LowK valueregioncorrespondstothehigh On the other hand, the claim that the transition tem- T regionandviceverse. Andfornormalheatconduction, perature is around(0.2−0.3)in Ref. [62] has been chal- the thermal conductivity κ is proportionalto the energy lenged in Ref. [63] with numerical simulations of longer diffusion constant D . This will finally give rise to the sizes. The possible phase transitioncouldbe a finite size E prediction oftemperature dependent thermal conductiv- effect. Here we reconfirm the finite size effect by giving ity κ(T)ofcoupledrotatorlattice fromthe analogywith a more illustrative picture in Fig. 5. The heat conduc- single kicked rotator: tance σ ≡ κ(T)/N as the function of temperature has been plotted for different lattice sizes. At high temper- κ(T) ∝ T−2, low T (12) ature regions, all the heat conductances follow the same κ(T) ∝ T−3, high T (13) temperature behavior of σ ∝ T−3.2. As the tempera- 5 ture decreases,the phononmeanfree pathwillovercome connectionwiththesinglekickedrotatororthe Chirikov the lattice size and the heat conductance will saturate standard map has been discussed where a κ(T) ∝ T−3 to a value determined by the harmonic limit of coupled dependence can be implied. Our results also recon- rotator lattice. As can be seen from Fig. 5, the heat firmthatthepreviouslyclaimedpossiblephasetransition conductance for short lattice will first bend and become should be a finite size effect. flat as the temperature decreases. This crossover tem- perature decreases as the lattice size increases smoothly and no sign of phase transition can be observed. It is V. ACKNOWLEDGMENTS also noticed that the crossover temperature happens to be around(0.2−0.3)for lattice sizesof afew thousands. The numericalcalculations were carriedout at Shang- hai Supercomputer Center. This work has been sup- portedbytheNSFChinawithgrantNo. 11334007(Y.L., IV. SUMMARY N.L., B.L.), the NSF China with grant No. 11347216 (Y.L), Tongji University under Grant No. 2013KJ025 We have systematically investigated the temperature (Y.L), the NSF China with Grant No. 11205114(N.L.), dependence of thermal conductivities of 1D coupled ro- the Program for New Century Excellent Talents of the tator lattice. The actual temperature dependence is a Ministry of Education of China with Grant No. NCET- power-lawdependence ofκ(T)∝T−3.2 whichisdifferent 12-0409 (N.L.) and the Shanghai Rising-Star Program with the observations of previous studies. The possible with grant No. 13QA1403600(N.L.). [1] S.Lepri,R.Livi,andA.Politi,Phys.Rev.Lett.78,1896 [27] H. Zhao, Phys. Rev.Lett. 96, 140602 (2006) (1997) [28] D. He, S. Buyukdagli, and B. Hu, Phys. Rev. E 78, [2] S.Lepri,R.Livi,andA.Politi,Phys.Rep.377,1(2003) 061103 (2008) [3] A.Dhar, Adv.Phys.57, 457 (2008) [29] Z.-G.Shao,L.Yang,W.-R.Zhong,D.-H.He,andB.Hu, [4] S. Liu, X. F. Xu, R. G. Xie, G. Zhang and B. Li, Eur. Phys. Rev.E 78, 061130 (2008) Phys.J. B 85, 337 (2013) [30] Y. Dubi and M. Di Ventra, Phys. Rev. E 79, 042101 [5] B.Hu,B.Li,andH.Zhao,Phys.Rev.E57,2992(1998) (2009) [6] P.Tong,B.Li,andB.Hu,Phys.Rev.B59,8639(1999) [31] A. Henry and G. Chen, Phys.Rev. B 79, 144305 (2009) [7] T. Hatano, Phys. Rev.E 59, R1 (1999) [32] K. Saito and A. Dhar, Phys. Rev. Lett. 104, 040601 [8] G. P. Tsironis, A. R. Bishop, A. V. Savin, and A. V. (2010) Zolotaryuk, Phys.Rev.E 60, 6610 (1999) [33] L.Wang,D.He,andB.Hu,Phys.Rev.Lett.105,160601 [9] A. Sarmiento, R. Reigada, A. H. Romero, and K. Lin- (2010) denberg,Phys. Rev.E 60, 5317 (1999) [34] N.Yang,G. Zhang,and B. Li,NanoToday 5, 85 (2010) [10] A.Dhar and D.Dhar, Phys. Rev.Lett. 82, 480 (1999) [35] L. Wang and T. Wang, EPL 93, 54002 (2011) [11] D. Alonso, R. Artuso, G. Casati, and I. Guarneri, Phys. [36] L. Wang, B. Hu, and B. Li, Phys. Rev. E 86, 040101 Rev.Lett. 82, 1859 (1999) (2012) [12] B.Hu,B.Li,andH.Zhao,Phys.Rev.E61,3828(2000) [37] D. Xiong, J. Wang, Y. Zhang, and H. Zhao, Phys. Rev. [13] K.AokiandD.Kusnezov,Phys.Lett.A265,250(2000) E 85, 020102(R) (2012) [14] B.Li,H.Zhao,andB.Hu,Phys.Rev.Lett.86,63(2001) [38] G. T. Landi and M. J. de Oliveira, Phys. Rev. E 87, [15] A.Dhar, Phys.Rev. Lett.86, 5882 (2001) 052126 (2013) [16] K. Aoki and D. Kusnezov, Phys. Rev. Lett. 86, 4029 [39] D. Xiong, Y. Zhang, and H. Zhao, Phys. Rev. E 90, (2001) 022117 (2014) [17] Y.Zhang and H. Zhao, Phys. Rev.E 66, 026106 (2002) [40] S.G.Das,A.Dhar,K.Saito,C.B.Mendl,andH.Spohn, [18] B. Li, L.Wang, and B. Hu,Phys.Rev.Lett. 88, 223901 Phys. Rev.E 90, 012124 (2014) (2002) [41] C. B. Mendl and H. Spohn, Phys. Rev. E 90, 012147 [19] K.Saito, EPL 61, 34 (2003) (2014) [20] A. V. Savin and O. V. Gendelman, Phys. Rev. E 67, [42] A.V.SavinandY.A.Kosevich,Phys.Rev.E89,032102 041205 (2003) (2014) [21] S.Lepri,R.Livi,andA.Politi, Phys.Rev.E68,067102 [43] S. Liu, J. Liu, P. Hanggi, C. Wu, and B. Li, Phys. Rev. (2003) B 90, 174304 (2014) [22] D.Segal,A.Nitzan,andP.H¨anggi,J.Chem.Phys.119, [44] L. Wang, L. Xu, and H. Zhao, Phys. Rev. E 91, 012110 6840 (2003) (2015) [23] O.V.Gendelman and A.V.Savin,Phys.Rev.Lett. 92, [45] S. Lepri, R. Livi, and A. Politi, Europhy. Lett. 43, 271 074301 (2004) (1998) [24] B. Li, J. Wang, L. Wang, and G. Zhang, Chaos 15, [46] S. Lepri, Phys. Rev.E 58, 7165 (1998) 015121 (2005) [47] O. Narayan and S. Ramaswamy, Phys. Rev. Lett. 89, [25] G.ZhangandB. Li,J. Chem.Phys.123, 114714 (2005) 200601 (2002) [26] H. Zhao, Z. Wen, Y. Zhang, and D. Zheng, Phys. Rev. [48] J.-S.WangandB.Li,Phys.Rev.Lett.92,074302(2004) Lett.94, 025507 (2005) [49] P. Cipriani, S. Denisov, and A. Politi, Phys. Rev. Lett. 6 94, 244301 (2005) Rev. Lett.84, 2144 (2000). [50] E. Pereira and R. Falcao, Phys. Rev. Lett. 96, 100601 [62] O. V.Gendelman and A. V.Savin,Phys.Rev.Lett.84, (2006) 2381 (2000). [51] L. Delfini, S. Lepri, R. Livi, and A. Politi, Phys. Rev. E [63] L. Yangand B. Hu,Phys.Rev.Lett. 94, 219404 (2005). 73, 060201 (2006) [64] Y.Li,S.Liu,N.Li,P.Hanggi,andB.Li,arXiv:1407.1161 [52] G. Basile, C. Bernardin, and S. Olla, Phys. Rev. Lett. [65] S. G. Das and A. Dhar, arXiv:1411.5247v2 96, 204303 (2006) [66] H. Spohn,arXiv:1411.3907 [53] H.van Beijeren, Phys. Rev.Lett. 108, 180601 (2012) [67] N.Li,J.Ren,L.Wang,G.Zhang,P.Hanggi, andB.Li, [54] C.B.MendlandH.Spohn,Phys.Rev.Lett.111,230601 Rev. Mod. Phys. 84, 1045 (2012). (2013) [68] N.Li,P.Tong, andB.Li, Europhys.Lett.7549(2006). [55] E. Pereira, R. Falcao, and H.C.F. Lemos, Phys. Rev. E [69] N. Li and B. Li, Europhys.Lett. 78, 34001 (2007). 87, 032158, (2013) [70] N. Li and B. Li, Phys.Rev.E 76, 011108 (2007). [56] H.Spohn,J. Stat.Phys. 154, 1191 (2014) [71] N. Li and B. Li, J. Phys. Soc. Japan 78, 044001 (2009). [57] S. Liu, P. H¨anggi, N. Li, J. Ren, and B. Li, Phys. Rev. [72] N.Li, B. Li, andS.Flach, Phys.Rev.Lett.105, 054102 Lett.112, 040601 (2014) (2010). [58] C. W. Chang, D. Okawa, H. Garcia, A. Majumdar, and [73] N. Li and B. Li, AIPAdvances2, 041408 (2012). A.Zettl, Phys. Rev.Lett. 101, 075903 (2008) [74] N. Li and B. Li, Phys.Rev.E 87, 042125 (2013). [59] X. Xu, L. F. C. Pereira, Y. Wang, J. Wu, K. Zhang, X. [75] L. Yang, N. Li, and B. Li, Phys. Rev. E 90, 062122 Zhao, S. Bae, C. T. Bui, R. Xie, J. T. L. Thong, B. H. (2014). Hong, K. P. Loh, D. Donadio, B. Li, and B. Ozyilmaz, [76] B. V.Chirikov, Phys. Rep.52, 263 (1979). Nat.Commun. 5, 3689 (2014) [77] R. S. MacKay, J. D. Meiss, I. C. Percival, Phys. D 13, [60] T.Meier, F.Menges, P.Nirmalraj, H.H¨olscher,H.Riel, 55 (1984). and B. Gotsmann, Phys.Rev.Lett. 113, 060801 (2014) [61] C. Giardina, R. Livi, A. Politi, and M. Vassalli, Phys.

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.