Item # 24698 INSTRUCTOR’S MANUAL CONTAINING SOLUTIONS TO. OVER 200 PROBLEMS SELECTED FROM STATISTICAL MECHANICS (SECOND EDITION) BY R.K. PATHRIAPREFACE ‘The idea of producing this manual first came from my friend and colleague Wing-Ki Liu several years ago when I had just embarked on the task of preparing the second edition of my book on Statistical ‘Mechanics. Though the merits of this idea were clear to me right away, I somehow shirked the effort involved in pursuing it and quietly chose to let it slide by. Later on, when, soon after the appearance of the new edition, my publisher reiterated the same suggestion, I decided to consult a few people who were known users of my book to find out their reaction to this suggestion. Almost unanimously, they endorsed the idea, after which I had no choice but to go ahead and do it, The result is a manual containing solutions to some 220 problems selected from the text, ‘The usefulness of this manual to someone who teaches Statistical Mechanics from my book can be manifold, First of all, there is the obvious advantage of saving time that one would otherwise spend on solving these problems oneself. Secondly, before one selects problems either for homework or for an exam, ‘one can now consult the manual to determine the level of difficulty of the various problems and make one’s selection accordingly. Thirdly, one may even use some of these solved problems, especially the ones appearing in later chapters, as “lecture material” supplementing the text, I trust that the availability of the ‘manual wil enhance the usefulness of the text — both for the instructors and (indirectly) for the students. For the production of a camera-ready copy of the manuscript for this manual, I am once again indebted to Mrs. Debbie Guenther whose skillful typing and careful proof-reading have gone a long way in ‘making my task a truly enjoyable one, RKP, Waterloo, Ontario, CanadaCHAPTER 1 AL, (@ We expand the quantity £n Q (E,) as a Taylor series in the variable (E, - Ei) and get $n (E,) = tnQ,(E,) + &nQ,(E,) (2=8°-8) ={énQ, (E:)+ (nQ,(E2)} + dtnQ(E) , 9 tn Q, (Ey) dE, | “ki io a zB gen E) , 22) 2 OE, 1 oF; ‘The first term of this expansion is a constant, the second term vanishes as a result of equilibrium (8, = B,), ‘while the third term may be written as 1 Pca ied ia ae = Ce ees men ; with 7; =7,. Ignoring the subsequent terms (which is justified if the systems involved are large) and taking the exponentials, we readily see that the function °(E,) is a Gaussian in the variable (E,-E), with variance KT (Cy), (Cy), 1{(Cy), + (Cy)2}- Note that if (C,), >> (C,), — corresponding to system 1 being in thermal ‘contact with a very large reservoir — then the variance becomes simply kT?(C,,),, regardless of the nature of the reservoir, cf, eqn, (3.6.3). (b) If the systems involved are ideal classical gases, then (C,); =3NE and (Cy); HE Nik the variance then becomes fer N,N, /(N,+N,). Again, it N, >>N,, we obtain the simplified expression Suen: cf. Problem 3.18. 12, Since $ is additive and © multiplicative, the function /(S2) must satisfy the condition F(2,2,)= f(Q,)+ FQ,) - oO Differentiating (1) with respect to ©, (and with respect to Q,), we get Q,F'(Q,2,)= FQ) and 2,f(Q,Q,)= f(Qz) + so that Q,F'Q,)= 2, f(Q,) @ iteSince the left-hand side of (2) is independent of €2, and the right-hand side is independent of @,, each side must be equal to a constant, k, independent of both ©, and ©,. It follows that f“(Q)= k/Q and hence f(Q)=k én Q + const. ® Substituting (3) into (1), we find that the constant of integration is zero, 14, Instead of eqn. (1.4.1), we now have Qe V(V-v,)(V-2v,)...(V-N=1 vy), so that E &nQ=C+lnV4+ln (V=v,)+ tn (V—2vq)t.¢n V-N=1y,) , where C is independent of V. The expression on the right may be written as xa wet conten v4 5 tn(1 - a) =C4HNin vsd(- ie) C+Ninv-— Sy v v 2 Equation (1.4.2) is then replaced by Since Nv, << V , (1+Nv,/2V)" = 1-Nv,/2V . Our last result then takes the form: P(V—b) = NkT, where b= A little reflection shows that v, = (4m /3)o° , with the result that i An 1 ; pO ANT L5, This problem is essentially solved in Appendix A; all that remains to be done is to substitute from eqn. (B.12) into (B.11), to get 22 pp? #29 (eM IL) yg te LY 6° lon \L?, we obtain eqns. (1.4.15 and 16). Z(er Substituting V= P’ and $=Next, the ratio of the Hinear term to the main one is #9/4e*'*, which may be written as ¥ 9h/{8L(2me)'"}. With the given data, this ratio turns out to be about F 2.6x10™” . 117. Since the energy cigenvalues in this case are proportional to V~, the entropy of the system will be of the form S(N,V,E) =S(N,V'E); cf. eqn. (1.4.8). An isentropic process is, therefore, characterized by the fact that the quantity (VE) stays cOnstant during such a process. It follows that E =) 5 OV Js v and hence the quantity (PV“*) also stays constant during an isentropic process. This suggests that the ratio Cp, of this system may be equal to 4/3. To see this explicitly, we proceed as follows. ‘Since the entropy of any given system is an extensive quantity, we must have for the present one. S=NfVIEIN’), where f(x) is some function of the variable x. It follows that 1_( as Vi? A VPE\_N (Vez 7G), er) EG): where g(x) = x/’(x). On inversion, this gives E=NTX some function of (v'°7) (v=V/N). Now, in the case of an ideal, classical gas we cannot have a v-dependence in this relationship; accordingly, E ‘must be of the form ciV7, where c is a constant. The precise value of ¢ for this system is 3k; see Problems 2.8 and 3.15. ILis now straightforward to see that C, = cN and C, = (4/3)cN ; cf. eqns. (1.4.24 and 27). 18. Convert the total energy £ into quanta, each of energy fiv. Let R be the number of these quanta; clearly, R= E/hy, Now follow the argument developed on page 70 of the text, whereby the number {2 for this problem turns out to be (R+N-D)YRUN-D! a For NV >> 1, expression (1) gives £nQ=(R+N) fn(R+N)-RénR-N iN .The expression for T now follows straightforwardly; we get so that rota For E >> Nhv, we recover the classical result: T= E/ Nk . 12 Since the function S(N, V,£) of a given thermodynamic system is an extensive quantity, we may write sv = mi ¥)=ar(we) It follows that ee x afro ween Gr). MS) AS), ‘Adding these expressions, we obtain the desired result. LLL, Clearly, the initial temperatures and the initial particle densities of the two gases (and hence of the mixture) are the same, ‘The entropy of mixing may, therefore, be obtained from eqn. (1.5.4), with N, =4N, and N,=N,. Weget (AS) = KAN, én(5/4)+N,én5] = R[4 fn (5/4) + fn] = 2.502 R, which is equivalent to about 0.5 R per mole of the mixture, 12, (a) The expression in question is given by eqn. (1.5.3a). Without loss of generality, we may keep N:, No and V; fixed and vary only V;. ‘The first and second derivatives of this expression are then given by MMM] goa of AMM] da» V+v, Vy Wtyrespectively. Equating (1a) to zero gives the desired condition, viz. N,V, =N,V,, ie. N/V, =N,/V;, say. Expression (1b) then reduces to n n Inv, ae Clearly, (AS) q» is atts minimum when N, /V; = N, /'V;, and itis straightforward to check that the value at the Soe nea (b) The expression now in question is given by eqn. (1.5.4). With N,=aN and N, =(1-a)N, where N= N, +N, (which is fixed), the expression for (AS) */k takes the form — GN fn @-(1-@)N fn (1-ar), ‘The first and second derivatives of this expression with respect to @¢ are NN [-N én +N én(l-a)] and [-2 I : (2a.b) respectively. Equating (2a) to zero gives the condition o = 1/2, which reduces (2b) to 4. Clearly, (AS)*/k is at its maximum when N, = N, Nén2. (1/2)N, and it is straightforward to check that the value at the maximum is 1.13. Proceeding with eqn. (1.5.1), with T replaced by T;, itis straightforward to see that the extra contribution to AS, owing to the fact that T; # T,, is given by the expression Nk tn (T/T )+ 3 Nk OT, /T) where T, = (N,T, +N,T,)/(N, +N,). Itis worth checking that this expression is always greater than or equal to zero, the equality holding if and only if 7, =, . Furthermore, the result quoted here does not depend on whether the two gases were different or identical. Li By eqn. (1.5.1a), given on page 24 of the text, we get (45), =3ako(r, 7) Now, since PV = NKT, the same equation may also be written as Saf of) . It follows that(AS), =e, 17)=3(A5), : A numerical verification of this result is straightforward. It should be noted that quite generally, (AS) T(@®S/0T), _C, (9, T@s/an, ~ Cy which, in the present case, happens to be 5/3 115. Foran ideal gas, C, —C, = nR, where nis the number of moles of the gas. With C, / Cy =7, one gets Cp=ynR/(y=1) and For a mixture of two ideal gases, mRI(y-1) nk Jo +n )R. n Equating this to the conventional expression (1, +7,)R/(y—1), we get the desired result. G 1.16. In view of eqn. (1.3.15), E~7S+ PV = MN. It follows that dE~TdS - Sd? + PaV + VaP = pid + Naps Combining this with eqn. (1.3.4), we get “Sd + VaP = Nat, i.e. dP =(N/V)du+(S/V)dT. Clearly, then, (@P/Out)p=NIV and (@P/ dT), =S/V. ‘Now, for the ideal gas 1 PHN nd p=Kr ing X(—* j Vv V ( QamkT see eqn. (1.5.7), Eliminating (V/V), we get 2 mo sr{ 2008) en, cs which is the desired expression. It follows quite readily now that for this system ee ou), kT which is indeed equal to N/V, whereas7 oa[ 3m nl which, by eqn. (1.5.1a), is precisely equal to $/ V Problem 2.3CHAPTER 2 23, The rotator in this problem may be regarded as confined to the (z = 0)-plane and time 1 may be denoted by the azimuthal angle g. The conjugate variable p, is then mp* position at where the various symbols have their usual meanings. The energy of rotation is given by 1 E=> N, we obtain the asymptotic result. QE) = R*"1(N-1)!, where R= El ho B.8.25a) ‘The corresponding expression for P(E;A) would be (Ethos A EMA (N=)! ha (N=1)i(hoy” ‘The “volume” of the relevant region of the phase space may be derived from the integral J Tlténde, with 3 (Zeat +4 of) T(E;A) = a10 ‘This is equal to, see en. (7a) of Appendix C, ay eileen WW 2) Om" = (=) ae k NI o) NI where @ = yk /m, The “volume” of the shell in question is then given by (2) NE* (2) ry -A=(2= @ o N! @/) (N-1)! Dividing (2) by (1), we see that the conversion factor a, is precisely h ty 28. We write V,, = AR, so that dVjy =A-3NR™ "dR. Atthe same time, we have a ‘The integral on the left may be written as Jean" dvy = e*(4n)\ A-3NR"'dR =(42)" A-3N TN) - Q 3 Equating (1) and (2), we get: A= (81)" / (3NV)!,, which yields the desited result for Vjy ‘The “volume” of the relevant region of the phase space is given by {Tae wy Tl@mdp) =V" Gr EB /c*)"/GN)!, ‘so that LON, V,E) = V" (8 Ef h’c?)" 1(3N)!, which is a function of N and VE" . An isentropic process then implies that VE? = const. ‘The temperature of the system is given by = OME ig E=3NEr. E hw ‘The equation for the isentropic process then becomes VT? =const., ic. To V~"; this implies that ¥ =4/3, The rest of the thermodynamics follows straightforwardly. See also Problems 1.7 and 3.15.CHAPTER 3 34, For the first part, we use eqn. (3.2.31) with all @, =1. We get £ oar =k Se jo. which is indeed equal to -(A/T)+(U/T) =. For the second part, we use eqn. (3.2.5), with the result that k ae hea a” w{or}=#[renn—Sor inn] (nt) Ber}. = inh rar ‘Substituting for 1; from eqn. (3.2.10), we get etn W{n} = KB(E,) + ken which is precisely the result obtained in the first part, 3.5, Since the function A(N, V,7) of a given thermodynamic system is an extensive quantity, we may write A(N,V,T) = Nf(v.T) (v=V/N) It follows that 8), on) Adding these expressions, we obtain the desired result, 4 - 36. Let's go to part (c) right away. Our problem here is to maximize the expression S/k=—P.,fnP.,, subject to the constraints }) 2, =1, )\E,P,,=E and) N,P.,=N. Varying P's and using the method of ‘Lagrange’s undetermined multipliers, we are led to the condition YH1+ mP,,)-7-BE,- on, } 6P,, 0 In view of the arbitrariness of the SP"s in this expression, we require that —(I+ tnP,,)-7—BE,-ON, for all r and s. It follows that “ue12 P., «exp(-BE, -aN,) ‘The parameters @ and f are to be determined by the given values of N and E . In the absence of the constraint imposed by 1V , the parameter o¢ does not even figure in the calculation, ‘and we obtain P, e> 1, wy A mals va l(a) } ] AT inZ , whence sa son ef 522) jog foma {22 (22) arfe*™(ULKT*)E%G T yy - fewa%g T Kinv—— z Now while kinZ=ktn{Ver™} ‘Substituting these results into the above expression for S, we obtain the desired result for S. In passing, we note tat (1) = A4TS = 3 NAT +T. For the second part of the question, we write U(q) =) u(x), so that =TT(+4). Tey fey and follow Problems 3.23 and 1.4. ‘The quantity V then appears to be in the nature of a “free volume” for the molecules of the system. ‘The partition function of the mixture at any stage of the reaction is given by 1 WW Na Naw Vie Ei Sa” fia » where the symbols N,N, and N,, represent the number of atoms A, atoms B and molecules AB, respectively; this leads to the free energy A=-KT inQ=-KTY(N,fn f,-N,£nN,+N,) (j= A,B, AB). @ ‘To determine the equilibrium distribution of these numbers, we minimize (1) under the obvious constraints Ny+Ngp=const. and Ny +N, =const. @4 Varying N; to N; +8N,, we get GA = fn f, 1 N, JON, + En(f,/ Ny ON, + El Fn / Nan ON @ subject to the constraints ON, +8V,,=0 and SN, +5N,,=0 @ ‘Combining (3) and (4), we obtain the condition for equilibrium, viz, En f, 1N4)—€n( fn! Np) + &n fan! Nga, which may be written as Nyy /NyNp= fap! fufar its yp /4My = Vian ! f,fa In view of the fact that each of the f's, by virtue of the translational degrees of freedom, is proportional to V, the final result is a function of T oily. 3.18. Here, O,(V,T)=(1/ N1)Q*(V,T), while Toto V-4mp%dp _8nV_1 =fet ous QV.) Ie i PPE’ which yields the desired result for Qy. The thermodynamics of the system now follows straightforwardly. As regards the density of states, the expression eae BBO for a single particle, while the expression for Q,,(V,7) leads to 1 (8nv y EG ee Ge TGM) for the N-particle system; ef, the expression for 2(E) derived in Problem 2.8. AZ, Differentiate the stated result with respect to to get au o mu -Hhe Mdo=0 Hep ‘This means that,15 ‘which amounts to the desired result: (H®) —(H)’ =—(aU/ aB) B18, We start with eqn. (3.6.2), viz, a8, ae ae, ec Bp Ye™ : ‘and differentiate it with respect to B, keeping the F, fixed, We get au Sar =(E)-(F 2 e)-(\o+0 ‘Substituting for (AU / 98) from eqn. (1), we get Seo (p) Xe usa" ' which is precisely equal to ((E-U)’). As for 2°U / dB*, we note that, since (5), "Ge (#) =-kT" Zura], er[are,+r(%) J. Hence the desired result, “IPC, For the ideal classical gas, U 3 ner ana ©, =k, which ready yield te stated resus 3.1% Since G=Pg,p;, G= > (4.p; +4:2,). Averaging over a time interval 7, we get LFS Gp. +ap a2 f Gar = SEHD=GO = Jae. +4.p)dt a Gat = c T qa For a finite V and finite E, the quantity G is bounded; therefore, in the limit t—> ce, the right-hand side of (1) vanishes. The left-hand side then gives, (Sten+es))=0. which leads to the desired result,16 3.20, The virial of the noninteracting system, by eqn. (3.7.12), is -3PV. The contribution from interparticle interactions, by eqn. (3.7.15), is given by the “expectation value of the sum of the quantity —r(du/ 2r) over all f u(r) is a homogeneous function (of degree n) of the particle coordinates, this contribution will be —nU, where U is the mean potential energy (not the internal energy) of the system, The total pairs of particles in the system”. vVirial is then given by Y=-3PV—nU. ‘The relation K = ——¥ still holds, and the rest of the results follow straightforwardly. 2.21, All systems considered here are localized. ‘The pressure term, therefore, drops out, and we are left with the result Le 2 n+ Example (a) pertains to n= 2, while examples (b) and (c) pertain to n=~I. In the former case, K = U=-E ‘The next problem pertains to n= 4 3.22, Note that a force proportional to q” implies a potential energy proportional to g‘. ‘Thus oer es (c>0) 2m It follows that Je ?"(p? /2m)dp Gen") ieee for the values of these integrals, see eqns. (13a) of Appendix B. Next, Jem (ca*)dq @ 7 th) ; (ca")= [ma7 where 1() denotes the integral in the denominator, It is straightforward to see that 1() is proportional to BY, whence (cq*) = 1/ 48, which proves the desired result. 3.23. _(@) In this problem we are concerned with the integral I= [genre Integrating by parts, we pet mc Kes A Je +¢)Br'dr . An arbitrary constant c has been introduced here to secure “proper behavior” at r=09, Since exp(—Bu) 1 as re, we choose c=~1. The integrated part then vanishes [assuming that u(r) — 0 faster than 1/r?}, and ‘we are left with the result I alee 2 ‘This reduces eqn. (3.7.17) to the desired form, (b) In the case of hard-sphere potential, the function f(r) = PV _ |, 2mo* ( a) + 5 1 for rSo and 0 for r> 0. We then get ais oe MKT 3 For no” << 1, we may write this result in the approximate form 2nno* my - ] = MEI. Comparison with Problem 1.4 shows that the parameter # of that problem is equal to (277/3)No*, which is indeed four times the actual space occupied by the particles. 3.24, By eqn. (3.7.5), we have, for a single particle, (Sra)=sxr. @ ‘at ‘The left-hand side of (1) is the expectation value of the quantity p-u, ic, pu which, for a relativistic particle, is equal to myu®(1—u? / c*)”. The desired result follows readily18 In the non-relativistic limit (w< Fota(ore) which is essentially the same as eqn, (3.10.9). S=kinQ= For the temperature of the system, we get i () k { ==(2) --m, T \@)y 2 which agrees with eqn, (3.10.8).25 3.43, The partition function of this system is given by the usual expression (3.5.5), except for the fact that the Hamiltonian of the system is now a function of the quantities p; +(e; /c)A(x)), and not of the p, as such, However, on integration over any component of p;, from —s= to +e, we obtain the same standard factor 2rankT — regardless of the value of the corresponding component of A. The partition function is, therefore, independent of the applied field and hence the net magnetization of the system is zero.CHAPTER 4 4. By eqns, (4.1.9), (4.3.10) and (4.1.8), we get =(uN-U-PV)/kT a Since uN = G =U + PV TS, the right-hand side of (1) equals —S/. Hence the result. 4.2, According to the grand canonical ensemble theory, pv=sred3ct@,cran}. a * Now, the largest term in the sum pertains to the value N°, of N,, which is determined by the condition A nsnz-in0,,] By Sec. 3.3, this is equivalent to the statement: z=exp(u" /KT) , where 1" is the chemical potential of the given system in a canonical ensemble (with N = N"), If we replace the sum in (1) by its largest term, we would get PV=N'p’-A’=P'V, where P” is the pressure of the system in the canonical ensemble (with N ="), How different would P be from P* depends essentially on how different the particle density 7 is from n” — a question thoroughly discussed in Sec. 4.5. 43, The probability distribution in question is the binomial distribution no! Nt Vv =. PMO ee (e=fea=t po, We note that ae DPW,V)= (q+) m For part (), wehave i N= YNPW,V)=Np(gt p)*"'* =Np, while a -26-21 yo NW=D= NW PON, V) = NON — yp? p)"? = NWN = 1p? Pt It follows that, =NW=D+N=(Np) -N%p? +p, whence (ANY = N?-N? = N p(t =p) , ete For part (ii), we shift the origin to N = N©'p, write N=Np+x, N° -N=Nq—x sand examine the function tnP(x) = fn NO! — tn(N p+ x)! —En(Nq—x)! +(N p+x) inp+(N%q—x) ing. Since N p and Ng are both >> 1, we apply Stirling's formula, fnv!~vfnv— v, and get (after some reduction) x o) x én P(x)= re sy tei)-(0 ra -ziy) For x<>1 and NJ? << V, we may write > 3 2u0v.=v[iew sf |e vila a | . Z(V,T)=V" + rn Diy It follows that tn Q,(V,T)=-NinN+ nano. Me Jen ee) 733 ING AN TA ea ee ie Vamp y gy where _v=V/N; cf. eqns. (7.1.13) and (8.1.17). 5.Zand 8, For solutions to these problems, consult the references cited in Notes 10 and 11 on page 126 of the text.CHAPTER 6 6.1, Westart with eqn. (6.1.19) and write it in the form sau [rioff}o(v—BJo(i-o%] wo Now, setting all g, =1 and identifying (n} / g,) with (n,) , see eqns. (6.1.18a) and (6.2.22), we get 8=AE[-(.)nls,)+(ro)-2)ent—alo)] @ Choosing a= —I or +1, we obtain the desired results, Next we have to verify that cay {5 Dovey: oo} =#L(enp.cn). ® ‘Substituting for p,(n) from eqn. (6.3.10) into (3) leads to the desired result (2), with a = —1; substituting from eqn. (6.3.11) instead leads to the desired result (2), with a= +1, 62, Inthe BE. case, see eqn, (6.3.10), Pn) =(-n)r" [r=bay/( (n.)+1 }; 2 =012, Je It follows that (n.)=C-N3Snr =r/-1), (2) ==) Snr" = r+) (1-1), so that (n2)-(n,)? =r/ =r) =(n.)+(n,)° - @ In the F.D. case, see eqn. (6.3.11), (02)= Sonne = pe(1)=(n,) , 80 that (n2)~(n.)? =(n.)—(n,)° « ® In the M.B. case, see eqn. (6.3.12), one can readily see that (rae) = Stn tale) a(n ry ee a(n) soma (nd) (ne)? =(ne) « @ 34235 For the second part, we note, from eqn, 6.2.22, that (nyt =e! a Differentiating this result with respect to Jt, we get ® ‘Comparing (4) with our previous results (1)-(3), and with formula (6.3.9), we infer that, quite generally, (ni)-(n,)° = ef a(n.) au], . 6.3, Starting with eqn. (6.2.15), we now have ae, rn=TI] Ze" | TI a, V-T)= Dl enfi—(ee™ 1} —enft—ze™}] ; cf. eqn. (6.2.17). IL follows that For £ = 1, we obtain the Fermi-Dirac result; for £9 2 and ze" > 1 [see eqn, (6,2.16a)], we obtain the Bose-Einstein result. To determine the state of equilibrium of the given system, we minimize its free energy, U~TS, under the constraint that the total number of particles, N, is fixed. For this, we vary the particle distribution from n(r) to n(r)-+ dx(r) and require thatthe resulting variation6U-TS) = <, ve 2am A SI2 aw-(72) eee @ Bh} Bmg ‘The thermodynamic properties of the system now follow straightforwardly. In particular, U turns out to be 5 NET and hence Cy = Nk. The extra contribution comes from the potential energy of the system, which too rises with 7. Note, from eqns. (1) and (2), that the effective height of the gas molecules is (1-e"*)/ Bmg which for small heights is essentially L itself but for large heights is essentially KT’ / mg making the total potential energy of the gas equal to NET. 6.9. — Consider a layer of the gas confined between heights z and z + dz. For hydrostatic equilibrium, we must have P(z+ de) + pgdz = Plz) , where p is the mass density of the gas. In differential form, one gets dP | dz =—pg =—(mg/kT)P a (a) If Tis uniform, eqn, (1) can be readily integrated, with the result inP =—(mg/kT)z+ const. , @ Which yields the desired formula: P(z) = P(O)exp (~mgz / KT). (b) If, on the other hand, the equilibrium is attained adiabatically, then T is related to P; in fact, Tx PY" Wenow get @ This means that T now decreases essentially linearly with height. The pressure P and the density p go hand in hand with 7— varying as 77" and T”’~, respectively.39. 6.10, (a) For the given system, F(p)dp = const.e"™* (4ap*dp) = CoP mie” p2dp . ‘The normalization constant C is determined by the condition Jr@ap=cfe®o" pa Substituting = m,c sinh @, we get for the left-hand side of this equation Ce" mc? sinh® Bcosh 640 soften eed ; =Cmic' “Binet th ecosh a, + [yaar cmos = Cimjc? (Bmyc*)* K, (Bm,c?) ‘Equating this result with 1, we obtain the desired expression for C (b) Using the limiting forms (x /2x)"2e* >) ee at) We <<), ‘we obtain, rather straightforwardly, the nonrelativistic and the extreme relativistic limits of the distribution. (© Since de _ myc*d(cosh@) _ anh @ , dp mecd(sinh@) (pu) = Cf {mgc? sinh @tanh 6} cP" mic? sinh? Acosh 640 a mgc* | e-Print 640 0 Once again, integrating by parts (this time twice), we obtain (pu) = Come? -3(Bm.c*)* K, (Bc?) Substituting for C, we obtain: (pu) =3/ B — regardless of the severity of the relativistic effects and in conformity with the results of Secs. 3.7 and 6.4. GAL, Ordinarily, when a molecule is reflected from a stationary wall that is perpendicular to the 2- direction, the z-component of its velocity u simply changes sign, ic. u{ =—u,. If the wall is receding at40 velocity v in the direction of its normal, the above result changes to (u/—v) (4-9), 20 tat uf =—(u, ~2v). This results in a change in the translational energy of the molecule which, for small v, is given by at ema 2 If A is the area of the wall, the net change in the energy of the gas, in time ét, is then given by, cf. eqn. (6.4.10), 86 = Abcn f i J omu,yyu,fluddu,dida, ten oo Avétn fff f2mu* cos? 9 f(a)}(u? sin @du dd do) 0 @ where E, is the total kinetic energy of the gas. Note that, since the gas continues to be in a state of (quasi- static) equilibrium, the change SE (even though it originates in the translational motion of the molecules) becomes eventually a change in the internal energy U of the gas (which may well have contributions from ‘degrees of freedom other than translational). If U = a, we may write @ @) Re-arranging (3) and integrating it, we obtain the desired result. In the extreme relativistic case, the factor 2/3 is replaced throughout by 1/3, leading to the alternate value of 7 6.12, We refer to expression (6.4.11) of the text. For part (a) of the question, we integrate only over 1 and @, to get aR, = ni 4n) 2s eas d8 => rtsind cos a841 For part (b), we integrate only over @ and @, to get aR, =nr- f(wu'du, where f(a), (25) re Eogeu})oer For part (c), we refer to expression (6.4.10) instead and get a) ek u,du,du,du, Tefollows that R5() _(1)" ogo[-E( 41 men) ni le ae ny” With J; = 300K, 7, = 310K and E =10~” J, this ratio turns out to be about 2.2, 6.13, (a) We start by calculating the kinetic energy associated with the z-component of the motion of the effused molecules. Proceeding as on page 139 of the text, we get [see eqn. (6.4.11)] J Je cos? a) Foun’ sin 8 due do aa i J (ucos 6) f(u)u? sin@ dud@ note that the averages on the right-hand side are taken over the gas inside the vessel. It is not difficult to show, see the corresponding calculation in Problem 6.6 and the formulae (B.13b), that (OEEee (i) 1, Bm” so that (ma) for the effused molecules, = 1/ 8 = kT’. The kinetic energy associated with the x- and y- 1 components of the molecular motion wil be the same as inside the vessel, viz. ET each. Tt follows that the mean energy & of an effused molecule is 2KT.42 (b) Assuming quasi-static equilibrium, the relations E=(3/2)NKT and P=NkT/V will ‘continue to hold for the gas inside the vessel. However, in view of the result obtained in part (a), we shall also have d Itfollows that, AN snshence To N'*; T 3N it further follows that Pe N*? . As for explicit variations with t, we make use of eqn. (6.4.13) and write aN y dt . ‘Combining the last two results, we get aT__la ( AT. J a =) a, T V\2am. so that T=T,(1+ct)*, where c=(a/6V)(KT, / 2mm)". The variations of N and P with 1 follow straightforwardly. 6.14, If nj, is the number of holes per unit area of the surface of the balloon (of radius 7), a the area of each hole and ¢ the duration of the leak, then the total number of molecules leaking is given by AN KT I7mn)'?] fri +My (4a? Jat The fraction of the molecules leaking is thus given by AN _ 1 (8kTY? % = [—) -n,(4ar’)at . N ayaa a(n") Since V = (4 /3)r*, we get AN oor Czy" ny, =— | N 3at Substituting the data given, we obtain: ny ~187 holes/m?,43 6.15, The rate of effusion of molecules from side A to side B, through a hole of cross-section S, is given by the expression 1 (RE, 2, =th, AS; Ret 4" am,” Flam the same from side B to side A is given by 1 [BE B Rpg = Mg, — & mn ga om, Jara kT, In the stationary state, these two expressions will be equal — which leads to the condition of dynamic ‘equilibrium: : PI Py = (mT, 1 m3)” Ite two gases are samples of the same gas, the condition simplifies to PIP, (T/T). 6.17, The (un-normalized) velocity distribution for a pair of molecules is given by fed rion op, F(a,.u)P udu, ~e 4, dy. We define the relative velocity, v, and the velocity of the centre-of-mass, V,, in the usual manner, viz, 1 v=u,-u,, SMa) This results in a new distribution for the variables v and V_ FV) dvdV~ 6 dye dy It is now straightforward to show that (v) = (16/ xm)'” =/2(u), while (v*) = (6/ Bm) = 2(u*). "The latter result implies that v, =v2u, We note that, since v= 42 -2u,-u,, (v?)=2(u2), rogardtess of the law of distribution of velocities — so long as it is isouropic, making (u,-u,)=0.44 6.18, ‘The (un-normalized) joint distribution for the molecular energies €, and €, is S(€,,8,)de,de, ~ Pele? dede, To obtain the desired distribution, we set €, = E—e, and integrate over all relevant values of &, with the result that 2 P(E)dE ~ e® [fee] dede a ~e ME dE; cf. eqns, (3.4.3) and (3.5.16), with N = 2. It is now straightforward to check that 6.19, The relative fraction of the excited atoms in the given sample of the helium gas would be 3e**, where he = 38,32 . Be, iTA, ‘The desired fraction turns out to be extremely small — about 7107”, 6.20, We extend the treatment of Problem 3.14 to the reaction AB+ CD <> AD+CB and obtain, in equilibrium, Tisbea water — Sober - cry, aco Santon For the given reaction, Li Fanfoo where each fis a product of three factors — the translational, the rotational and the vibrational. K(1) Now, for a heteronuclear molecule like HD we have, at high temperatures, i =o Sl)" Bak kT a ah? BRO yp” while for a homonuclear molecule like HH we have instead45 fame ak)" Tak RT i 2nh? A Nee see Note 11 on page 156 of the text. It follows that, at high temperatures, RT Minn Assuming the internuclear distances to be the same, the /’s here will be proportional to the reduced masses of the molecules; the @’s, on the other hand, are inversely proportional to the square roots of the reduced a masses, Accordingly, Fay Way yp Hey —_ {mutmg Hy +p) a Tmuloo Gm Manton (1, ) (Ly) Cae Nie, ana (rig ay @ mygmys —(2m,)" (2my) ‘Substituting (2) and (3) into (1), we see that K(T) = 4 6.22. The potential V(r) is minimum at r=r,, which determines the equilibrium value of r. Accordingly, the quantum of the rotational motion of the molecule is A? /2/, where J = jur;. This gives for ©, the expression f? /2yir3k = h? / mrzk because the reduced mass [L in this case is equal to m/2. ‘Substituting the given data, ©, turns out to be about 75 K. This gives a fairly clear idea of the “temperature range” where the rotational motion of the hydrogen molecules begins to contribute towards the specific heat of the gas. Next we expand V(r) in the neighborhood of =r and write Vir) =-%+(V a \(r—n) + This gives an @ equal to (2¥, / a)" =(4V, /ma*)'” and hence a ©, equal to A(4V, / ma)” /k ‘Substituting the given data, ©, turns out to be about 6260 K. Again, this gives a fairly clear idea of the “temperature range” where the vibrational motion of the hydrogen molecules begins to contribute towards ‘the specific heat of the gas.46 6.23, The effective potential of a diatomic molecule (including both rotation and vibration) is given by e vin= Vo +5 p0%(r=n) + (+1) ur’ ‘The equilibrium value of r is obtained by minimizing V(r), with the result 2 2 (4-5) = uy = 4040) 4 wor, Ho’ It follows that 2 we a qiegsurn= 2) s+1). Using data from the preceding problem, we find that for a hydrogen molecule the fractional change in. r is alo"), 6.24, The occupation number N,, is proportional to (2/ +1) e™", It follows that 1 Ateoreayar Ny N, 5 é, Substituting the given data, we get XN, ss ste 0.0675, ™ Ny: N,2N, #3 0.050:0.208:0.742 . 6.28 The various contributions to the molar specific heat of the gas at 300 K are: (® translational — the amount being (3/2)R. (i) rotational — since the characteristic values of the parameter ©, in this case are of the order of 10 K, these degrees of freedom may be treated classically, which yields a contribution of (3/2)R; see eqn. (6.5.42) (ii) vibrational — here, the parameters ©, are such that the various contributions have to be calculated quantum-mechanically, using formula (6.5.44). We find that ) 9, } e. 12 = 16,00, —* = 4.56, — = 16.37, —§ = 7.80, eT T mn