Scribd Logo
Cargar

¡Te damos la bienvenida a Scribd!

  • Capitulo 4 Brejan PDF

    Cargado por

    Gustavo Alves
    18 vistas18 páginas
    Capitula 4 Brejan

    Título original

    Capitulo 4 Brejan.pdf

    Derechos de autor

    © © All Rights Reserved

    Formatos disponibles

    PDF o lea en línea desde Scribd

    ¿Le pareció útil este documento?

    ¿Este contenido es inapropiado?

    Denunciar este documento
    Capitula 4 Brejan

    Copyright:

    © All Rights Reserved
    Descargue como PDF o lea en línea desde Scribd
    Marcar por contenido inapropiado
    18 vistas18 páginas

    Capitulo 4 Brejan PDF

    Cargado por

    Gustavo Alves
    Capitula 4 Brejan

    Copyright:

    © All Rights Reserved
    Descargue como PDF o lea en línea desde Scribd
    Marcar por contenido inapropiado
    de 18
    Chapter 4 SINGLE-PHASE SYSTEMS Problem 4.1 Starting with the Euler equation and fundamental relation in entropy representation by eye sn S=pUtEV-EN (4.30) ds 1 Py tg dU +p dV—FdN (4.18) we differentiate eq. (4.30") dU +Ud @ P val +Eaveva(E Ean-na(E and subtract eq. (a) from eq. (4.18): =ud 2 -Na{#) o= U(x) +velF)-Na{F) ) “The alternate approach is to start with the Gibbs-Duhem relation SdT - VdP +Ndp=0 (4.32) replace S using eq. (4.30"), pU+R- dT -VaP + Nau =0 © divide eq, (c) by T, aT, y PaT=TaP , y Téw-paT _ Cie ee @ and recognize the emergerice of the differentials that appear on the right-hand-side of eq. (b). Problem 4.2 We have a system of four equations U(S, V, N)= (a) TS, V, n=(¥ hs (b) WS. VN) =(F) © : Ris, (SS, S F,(S, V,N)=U-TS-HN=Up (¥,) ex (1 7 ne) @ To obtain Fy as a function of (T, V, 1) we eliminate U, S and N between eqs. (b) - (d). First, dividing (b) and (c) yields E?5SS © Second, combining (4) and (¢) we obtain the wanted result Sen 4 FAT VW =Uo(y) ee(-ar © Problem 4. In order to derive F(T, V, N) we begin with three equations (review Table 4.3): : : U(S,V,N)= RI oxy eS) @ n (9 TS,V.N= ©) &,V.N) (Sly, x FS, V.N=U-TS=Uo(q] The objective is to find F = F(T, V, N). First, we combine (b) and (c) to write F =TNe, {1- F=TNe, (! ne) @ Second, we eliminate S between (d) and (b), and obtain the wanted result: F=(T,)N, V)=TNe, © ‘The second part of the problem concerns the derivation of G(T, P, N). We have a system of four equations, namely (a) and (b) as well as ® GS, V,N)=U_ TS +PV=U9/ (@ First we eliminate and V between (b), (£) and (g). Dividing (b) by (£) yields 42 TL PTNR ® Next, we divide (g) by (b) @ Using (h) and (i) to eliminate V and S from (g) yields, finally, G=TN (é,+R)-SoT-TN which is a function of T, P and N. Problem 4.4 The (n+2)th Legendre transform is noday-¥ on The fundamental relation has (n+2) arguments U= UG, V,Nive- Nay oF Y= Y¥ (Kr, X2, XB, +» X42) which means that the (n+2)'® transform of U is its "total" Legendre transform aff?) =U -TS +PV-wiN-...-HnNn However, according to eq. (4.30") the right-hand-side is identically equal to zero, hence ng=0 Problem 4.5 We must prove that ap) Sp _/aP (= (35), oO Invoking eqs. (4.58) and (4.60) we work successively on the left-hand-side (LHS): =(22) 7(2) 1 (20) us =(5)7($4), + (3s), ®) Next, we eliminate (4P/2v)p using the cyclical relation between P, v and T, 43 ($3), (3), (p),--1 © such that eg. (b) becomes (OT) av 3) @ Putting (d) and (a) together, ~(35), 8), GF, @ we recognize the cyclical relaiton for P, v and s, LG, ° The fact that (f) is an identity proves the validity of equation (a). Problem 4.6 (3) = ($F), using eq. (4.561) _ _ @vlT)p i : - GaP using the cyclical relation for P, v and T 1 (av +t), 5 = oy ov\eas -+(5l, ay yar) ; ) a ~($), using eq. (4.563) 1 ~vfy(Sp),=-8 (©) First, a method involving the use of Jacobians: using eq. (4.58) using eq. (4.86) using eq. (4.87) using eq. (4.562) = cp- EBT ‘An alternate proof, one that does not involve the use of Jacobians, begins with recognizing that the objective is to expres the difference (cp - cy) as a function of P, v, T and their derivatives, uf ° We go back to the entropic definitions of ep and cy and write using eq. (4.56g) = pont (SF) a using eq. (4.561) Eliminating ds between the two expressions developed above, we solve for dT and find ($i), pee (Ge, & 45 in other words, ara lorl,=(GF) ae (5),-(, Equations (b) and (c) are the same equation namely 7(64), Ge, With an eye on eq. (a), we invoke the cyclical relation cp ey a a a (ar, (a (Ge), and eliminate (@P/2T)y between (e) and (d): = -1(35),¢" = CBT+Dv (we i, (fy 46 (by © (d) (©) @ using dh = Tds + vaP using eq. (4.562) using the cyclical relation for T, P and h using the answer to part (d) above using ds = z aT+ using eq. (4.568) © Gl aia Problem 4.7 We must prove that sa Since du = Tds - Pd, we recognize that Hu _ at @) as? av? as dv. (8) ant P= G8) ‘The left-hand-side of eq. (a) becomes -B@) Las = - (2) and, using Table 4.6, LHS =— 5 rh fam) Gr (shy petcal! wee For the right-hand-side of eq. (a) we note that df = -sdT - Pdv, hence (5) Ov) al (P)r rus =-(54), Or (ov), In other words, we conclude that LHS = RHS. Problem 4.8 ds= ds = cpdT + (-BT + 1) vdP (Se), et+ (SB), T aT Sp Far —pv ar (Se), at (3), = av +(2) dv c 48 using du = Tas - Pdv using dh = Tds + vdP using eq. (4.56) using eq. (4.561) using the cyclical relation Problem 4.9 (a) From Problem 4.6g we learn that Gl, Integrating at constant temperature we obtain av SY) =-2vg AT (a }, i) cp (T, P)- cp (T, 0) = - 2voATP hence ep = cg- D(T - Tp) - 2avoTP. (b) From Table 4.7 we have = gr ds =< dT Bv dP where We integrate 7 : ds= (2 ao ZavgP) AT — Zip (T-Ty + aP) P from the reference state (T = To, P = 0) to an arbitrary state (T, P), by choosing the path (Tp, 0) > (T, 0) followed by (T, 0) > (T, P). The result is 49 S(T, P) - S(T, 0) = co + bTo) In i -b(T - To) - 2AvoP (T - To) ~ Avo aP2 0 (©) For the Gibbs free energy g(T, P) we start with (see Table 4.3) dg =-sdT + vdP where we know both s(T, P) and v(T, P). We integrate from (To, 0) to any state (T, P) along the path (Tp, 0) — (T, 0) followed by (T, 0) + (T, P), t P A(T, P)= (Ty, 0) ~~ 1,0) a+ [ ver, a b a Setting both g(To, 0) and s(To, 0) equal to zero we obtain g(T,P) = —(cg + bTyJT In b +(Co# bTo)(T- To) +3 (TT)? +y0(P-2P?} + AvgP (T= To}? +aP(T Ty) + PP (d) The incompressible liquid model with constant specific heat means that 2 = 0, Ko = 0 and b=0. In this case the g(T, P) expression reduces to a(T, P)=-coT In zt + c(T - To) + voP. 0 Problem 4.10 This process is a zero-work process, therefore the first law requires 8Q=dU (constant V) If the process represents "heating", then (dU)y >0 () Constant-volume pressurization means that (dP)y >0 @ Dividing (1) and (2), we find that the condition for the described process to be a heating process is (pl>° 4-10 or ve(B>9 ° Finally, invoking the cyclical relation (By (Fh (or! the inequality (3) reduces to hence cy BPO Problem 4.11 oe le) = +97($) (SE). using the cyclical relation for P, v, and s = (34), 50), (35), (FF) HELE a) ~ HG), using the cyclical relation for P, v and T all P'!, therefore, (b) According to Table 4.5, in the ideal gas limit we have vep Ploy cj P ay ART. = Pv Alternatively, according to Table, 4.7, for an ideal gas we can write ep. ov ds= of dves oP and, in conclusion, Problem 4.12 (i) Regarding u as a function of T and v, we have ave(B oto =eyot+{r($), ~Plav seat 4 [T (F), -Plav and, after evaluating (@P/@T)y for a van der Waals gas, we obtain Integrating from (To, vo) to (T, v), we obtain finally = ug = ey (T- To +a(ye y (ii) The same procedure can be used for deriving the function s(T,v), 4-12 yielding finally vet sespsevin 5 tRin be (iii) Using the ds expression obtained above, we write dh= Teds + vdP R-ar-—RT v4 28 24 gy (v-b} RTb_, 2a] gy cy +R valet “ale ‘The difference (h - ho) is independent of the path linking the end states (To vo) and (T, v). Therefore, we can integrate dh by first proceeding isothermally from (To, vo) to (To, ¥), and later by keeping the volume constant from (To, v) until (T, v): Fare ya}+y because du = Tds - Pdv because cy = 7( $8) 5 (Dy because dh = Tds + vdP because cp = 1), 413 (-P), because df=-sdT —Pdv | because x= - {2% : v\ oP Problem 4.14 The “system” is the total volume V, which contains two compartments, A and B, V=Va+ Vp a) or Va=xV and Vp=(1-x)V 2) where the volume fraction x is a number between 0 and 1 The initial equilibrium state of the system is specified. Chamber A contains the gas mass ma at initial pressure Pas and temperature T,;, from which follows the ideal gas volume maRTa xV= (3) Pal Chamber B contains the gas mass mp at given initial pressure Pa) and temperature Tp), therefore mpRT 31 (l-x)V= (4) Par Dividing eqs. (3) and (4) we conclude that the initial (locked) position of the adiabatic partition (x:) is known: =x myn Pat 6 a maTar Pp ‘The partition must be “locked” in this position because in general Pa, #Pp 4-14 The question is to determine the final equilibrium position of the adiabatic partition (x2) after it is allowed to slide freely. Equilibrium of this kind (mechanical) means that Pao = Poo 6) or, in order, maRTay _ mpRTpo : : ” Var Vp2 =x. _ mplp2 8) XQ mTar Equation (8) shows that the unknown x2 depends on two additional unknowns, Taz and Ter. We need two more equations, The first additional equation is the first law of thermodynamics invoked for the entire (isolated) system during process 1-2: Qe2 — Wi-2 = (U2 — Ups + (U2 - Upp 0) in which Q\_7 = Wj = 0. The result is macy(Ta2 ~ Tai) + mpey(Tp2 ~ Tai) = 0 (10) ‘The second additional equation is the subleity that apparently has eluded thermodynamicists for a hundred years. The partition is massless. ‘This means that when the locking mechanism is removed and the partition slides freely, the pressure must be the same on both sides of the partition. Which pressure will that be, Pa(t) or Pa(t)? Let us convene that subscript A means that initially chamber A is more pressurized than chamber B, Par > Pai (ily When the partition is free to move (i.e. when the force that used to be provided by the lock is no longer present), the pressure on both sides of the partition will be the ower of the two pressures, namely Pp. This means that a pressure nonuniformity is created inside chamber A, and from 41 this the opportunity for gas A to expand, The expansion of A means that B will be compressed Whereas gas A expands as a nonequilibrium (nonuniform P4) subsystem, gas B contracts quasistatically: at every point in time, the pressure on its moving wall, Pp(t), is the same as the pressure throughout the gas mass my. So, the final equation comes from the fact that the adiabatic compression of mp is quasistatic. ‘The first law for an infinitesimal change of state of subsystem B requires 8Qp — Wy = dUy (12) where 5Qp = 0, and OWg = PadVp (13) dU = mpeydTg (4) From eqs. (12) ~ (14) follows the conclusion that cydTp /Tp + RdVp / Vp = 0, or that the path of the process undergone by mp is TyV$ = constant, where a= R/c,; fom this we obtain the final equation, Oe) or = BL (16) x In summary, we have three equations [namely, eqs. (8), (10) and (16)] for three unknowns, Xs, Tar and Typ. The final state is unique, and it has just been determined. For example, by climinating T2 and Tyo from the three equations we obtain the final position of the partition, I=x =(I- 4) (7) 4-16 where B=1+ Tala (18) mpTpt The other properties that define the final state are calculated after x is determined: a Tp2 = ta : } (9) 1-x jm Ta = Th ( —*L | Mp _*2_ 20 42° Tax | ma 19 mpRTy (1- x, Bo ee et eo 21 Beer | en What role does the second law play in all this? It plays the role of test of irreversibility. The solution developed above must not violate the second law, which, for the process 1-2 executed by the entire (isolated) system, requires that 2 (22) 8-512 f 1 The right-hand side is zero because the system is isolated (6Q = 0). The second law becomes (Sz - 81), + (82 - $1), 20 (23) where the B contribution is zero +Rin2B2 |= (24) Vai (S2 - Si) =m (som Equation (24) is a reaffirmation of the fact that the process executed by the B chamber is adiabatic and quasistatic, hence isentropic. ‘This means that all the irreversibility of process 1-2 is located in the gas of chamber A. Left to prove is that 417 (Sy - Si), 20 (25) as long as the assumption of initial mechanical nonequilibrium (Pq 2 Pip) is correct. This can be shown numerically, by assigning numerical values to all the symbols, given and calculated. Here I show it analytically, after I make the “convenient” assumptions that the setup is such that ma=mp, Ta: = Tpi anda = 1, This means that B = 2 and, from eq. (5), 1% = Fas) (26) x] Pai therefore x) < i. The second law test (25) becomes S$) -§, 2 (=), nNA2 nf xa) 20 en TMyty Vat Tar Vai The question is whether the quantity in the square brackets is greater than 1 ee ee 2 (28) 1=x_ 1x) % This question can be rewritten in several ways until this form: (1-year) 20? (29) This conclusion is correct as soon as. x,

    También podría gustarte