1: %mp_arc cond-mat/
2: %**start of header
3: \newcount\mgnf %ingrandimento
4: \mgnf=1
5:
6: \ifnum\mgnf=0
7: \magnification=1000
8: \hsize=15truecm\vsize=21truecm%\voffset2.truecm\hoffset.5truecm
9: \parindent=0.3cm\baselineskip=0.45cm\fi
10: \ifnum\mgnf=1
11: \magnification=\magstephalf
12: %\voffset=.5truecm
13: % \hoffset=0.truecm
14: % \hsize=15truecm\vsize=20.2truecm
15: \baselineskip=18truept plus0.1pt minus0.1pt \parindent=0.9truecm
16: % \lineskip=0.5truecm\lineskiplimit=0.1pt \parskip=0.1pt plus1pt
17: \fi
18:
19: %\ifnum\mgnf=2%
20: % \magnification=1200%
21: % \hsize=15truecm\vsize=20.2truecm%
22: % \baselineskip=18truept plus0.1pt minus0.1pt \parindent=0.9truecm%
23: % \lineskip=0.5truecm\lineskiplimit=0.1pt \parskip=0.1pt plus1pt%
24: %\fi
25: \ifnum\mgnf=2\magnification=1200\fi
26:
27: \ifnum\mgnf=0
28: \def\openone{\leavevmode\hbox{\ninerm 1\kern-3.3pt\tenrm1}}%
29: \def\*{\vskip.7truemm}\fi
30: \ifnum\mgnf=1
31: \def\openone{\leavevmode\hbox{\ninerm 1\kern-3.63pt\tenrm1}}%
32: \def\*{\vglue0.3truecm}\fi
33: \ifnum\mgnf=2\def\*{\vglue0.7truecm}\fi
34: \openout15=\jobname.aux
35:
36:
37: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38: %%%%%%%%%%%%%%%%%%%%%%%%% DEFINIZIONI DI FONT %%%%%%%%%%%%%%%%%%%%%%%%%%%
39: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
40: \font\titolo=cmbx12\font\titolone=cmbx10 scaled\magstep 2\font \titolino=cmbx10%
41: \font\cs=cmcsc10\font\sc=cmcsc10\font\css=cmcsc8%
42: \font\ss=cmss10\font\sss=cmss8%
43: \font\crs=cmbx8%
44: \font\indbf=cmbx10 scaled\magstep2
45: \font\type=cmtt10%
46: \font\ottorm=cmr8\font\ninerm=cmr9%
47: \font\msxtw=msbm9 scaled\magstep1%
48: \font\msytw=msbm9 scaled\magstep1%
49: \font\msytww=msbm7 scaled\magstep1%
50: \font\msytwww=msbm5 scaled\magstep1%
51: %\font\msytwwww=msbm4 scaled\magstep1%
52: \font\euftw=eufm9 scaled\magstep1%
53: \font\euftww=eufm7 scaled\magstep1%
54: \font\euftwww=eufm5 scaled\magstep1%
55: \def\st{\scriptstyle}%
56: \def\dt{\displaystyle}%
57:
58: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
59: %%%%%%%%%%%%%%%%% LETTERE GRECHE E LATINE IN NERETTO %%%%%%%%%%%%%%%%%
60: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
61:
62: % lettere greche e latine in neretto italico - pag.430 del manuale
63: \font\tenmib=cmmib10 \font\eightmib=cmmib8
64: \font\sevenmib=cmmib7\font\fivemib=cmmib5
65: \font\ottoit=cmti8
66: \font\fiveit=cmti5\font\sixit=cmti6%%
67: %!!!@@@\font\fiveit=cmti7\font\sixit=cmti7%%
68: \font\fivei=cmmi5\font\sixi=cmmi6\font\ottoi=cmmi8
69: \font\ottorm=cmr8\font\fiverm=cmr5\font\sixrm=cmr6
70: \font\ottosy=cmsy8\font\sixsy=cmsy6\font\fivesy=cmsy5%%
71: \font\ottobf=cmbx8\font\sixbf=cmbx6\font\fivebf=cmbx5%
72: \font\ottott=cmtt8%
73: \font\ottocss=cmcsc8%
74: \font\ottosl=cmsl8%
75:
76:
77: \def\ottopunti{\def\rm{\fam0\ottorm}\def\it{\fam6\ottoit}%
78: \def\bf{\fam7\ottobf}%
79: \textfont1=\ottoi\scriptfont1=\sixi\scriptscriptfont1=\fivei%
80: \textfont2=\ottosy\scriptfont2=\sixsy\scriptscriptfont2=\fivesy%
81: %\textfont3=\tenex\scriptfont3=\tenex\scriptscriptfont3=\tenex%
82: \textfont4=\ottocss\scriptfont4=\sc\scriptscriptfont4=\sc%
83: %\scriptfont4=\ottocss\scriptscriptfont4=\ottocss%
84: \textfont5=\eightmib\scriptfont5=\sevenmib\scriptscriptfont5=\fivemib%
85: \textfont6=\ottoit\scriptfont6=\sixit\scriptscriptfont6=\fiveit%
86: \textfont7=\ottobf\scriptfont7=\sixbf\scriptscriptfont7=\fivebf%
87: %\textfont\bffam=\eightmib\scriptfont\bffam=\sevenmib%
88: %\scriptscriptfont\bffam=\fivemib%
89: \setbox\strutbox=\hbox{\vrule height7pt depth2pt width0pt}%
90: \normalbaselineskip=9pt\rm}
91: \let\nota=\ottopunti%
92:
93:
94: \textfont5=\tenmib\scriptfont5=\sevenmib\scriptscriptfont5=\fivemib
95: \mathchardef\Ba = "050B %alfa
96: \mathchardef\Bb = "050C %beta
97: \mathchardef\Bg = "050D %gamma
98: \mathchardef\Bd = "050E %delta
99: \mathchardef\Be = "0522 %varepsilon
100: \mathchardef\Bee = "050F %epsilon
101: \mathchardef\Bz = "0510 %zeta
102: \mathchardef\Bh = "0511 %eta
103: \mathchardef\Bthh = "0512 %teta
104: \mathchardef\Bth = "0523 %varteta
105: \mathchardef\Bi = "0513 %iota
106: \mathchardef\Bk = "0514 %kappa
107: \mathchardef\Bl = "0515 %lambda
108: \mathchardef\Bm = "0516 %mu
109: \mathchardef\Bn = "0517 %nu
110: \mathchardef\Bx = "0518 %xi
111: \mathchardef\Bom = "0530 %omi
112: \mathchardef\Bp = "0519 %pi
113: \mathchardef\Br = "0525 %ro
114: \mathchardef\Bro = "051A %varrho
115: \mathchardef\Bs = "051B %sigma
116: \mathchardef\Bsi = "0526 %varsigma
117: \mathchardef\Bt = "051C %tau
118: \mathchardef\Bu = "051D %upsilon
119: \mathchardef\Bf = "0527 %phi
120: \mathchardef\Bff = "051E %varphi
121: \mathchardef\Bch = "051F %chi
122: \mathchardef\Bps = "0520 %psi
123: \mathchardef\Bo = "0521 %omega
124: \mathchardef\Bome = "0524 %varomega
125: \mathchardef\BG = "0500 %Gamma
126: \mathchardef\BD = "0501 %Delta
127: \mathchardef\BTh = "0502 %Theta
128: \mathchardef\BL = "0503 %Lambda
129: \mathchardef\BX = "0504 %Xi
130: \mathchardef\BP = "0505 %Pi
131: \mathchardef\BS = "0506 %Sigma
132: \mathchardef\BU = "0507 %Upsilon
133: \mathchardef\BF = "0508 %Fi
134: \mathchardef\BPs = "0509 %Psi
135: \mathchardef\BO = "050A %Omega
136: \mathchardef\BDpr = "0540 %Dpr
137: \mathchardef\Bstl = "053F %*
138: \def\BK{\bf K}
139: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
140: %%%%%%%%% RIFERIMENTI SIMBOLICI A FORMULE, PARAGRAFI E FIGURE %%%%%%%%%%
141: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
142: %
143: % Ogni paragrafo deve iniziare con il comando \section(#1,#2), dove #1
144: % e' il simbolo associato al paragrafo e #2 e' il titolo. Per le
145: % appendici bisogna pero' usare \appendix(#1,#2).
146: %
147: % Se nel titolo compaiono riferimenti ad altri simboli, questi vanno
148: % racchiusi fra parentesi graffe, per es. {\equ(1.2)}; in caso contrario
149: % si provoca un errore.
150: %
151: % Ogni sottoparagrafo deve iniziare con il comando \sub(#1) o \asub(#1),
152: % nelle appendici.
153: %
154: % I riferimenti a paragrafi e sottoparagrafi si realizzano con il comando
155: % \sec(#1), che produce il numero effettivo preceduto dal simbolo di
156: % paragrafo, o \secc(#1), che produce solo il numero (serve nel caso si
157: % faccia riferimento ad un sottoparagrafo, che e' un Lemma, un Teorema o
158: % altro oggetto suscettibile di una denominazione speciale).
159: %
160: % Le formule sono contrassegnate con \Eq(#1), eccetto che all'interno
161: % del comando \eqalignno, dove si deve usare \eq(#1). Nelle appendici
162: % i comandi corrispondenti sono \Eqa(#1) e \eqa(#1).
163: % I riferimenti alle formule si realizzano con \equ(#1).
164: %
165: % La numerazione delle figure utilizza il comando \eqg(#1), per
166: % contrassegnarle, e \graf(#1) per citarle.
167: %
168:
169: \global\newcount\numsec\global\newcount\numapp
170: \global\newcount\numfor\global\newcount\numfig
171: \global\newcount\numsub
172: \numsec=0\numapp=0\numfig=0
173: \def\veroparagrafo{\number\numsec}\def\veraformula{\number\numfor}
174: \def\veraappendice{\number\numapp}\def\verasub{\number\numsub}
175: \def\verafigura{\number\numfig}
176:
177: \def\Section(#1,#2){\advance\numsec by 1\numfor=1\numsub=1\numfig=1%
178: \SIA p,#1,{\veroparagrafo} %
179: \write15{\string\Fp (#1){\secc(#1)}}%
180: \write16{ sec. #1 ==> \secc(#1) }%
181: \0\hbox%to \hsize
182: {\titolo\hfill
183: \number\numsec. #2\hfill%
184: \expandafter{\hglue-1truecm\alato(sec. #1)}}}
185:
186: \def\appendix(#1,#2){\advance\numapp by 1\numfor=1\numsub=1\numfig=1%
187: \SIA p,#1,{A\veraappendice} %
188: \write15{\string\Fp (#1){\secc(#1)}}%
189: \write16{ app. #1 ==> \secc(#1) }%
190: \hbox to \hsize{\titolo Appendix A\number\numapp. #2\hfill%
191: \expandafter{\alato(app. #1)}}\*%
192: }
193:
194: \def\senondefinito#1{\expandafter\ifx\csname#1\endcsname\relax}
195:
196: \def\SIA #1,#2,#3 {\senondefinito{#1#2}%
197: \expandafter\xdef\csname #1#2\endcsname{#3}\else
198: \write16{???? ma #1#2 e' gia' stato definito !!!!} \fi}
199:
200: \def \Fe(#1)#2{\SIA fe,#1,#2 }
201: \def \Fp(#1)#2{\SIA fp,#1,#2 }
202: \def \Fg(#1)#2{\SIA fg,#1,#2 }
203:
204: \def\etichetta(#1){(\veroparagrafo.\veraformula)%
205: \SIA e,#1,(\veroparagrafo.\veraformula) %
206: \global\advance\numfor by 1%
207: \write15{\string\Fe (#1){\equ(#1)}}%
208: \write16{ EQ #1 ==> \equ(#1) }}
209:
210: \def\etichettaa(#1){(A\veraappendice.\veraformula)%
211: \SIA e,#1,(A\veraappendice.\veraformula) %
212: \global\advance\numfor by 1%
213: \write15{\string\Fe (#1){\equ(#1)}}%
214: \write16{ EQ #1 ==> \equ(#1) }}
215:
216: %\def\getichetta(#1){Fig. \verafigura%
217: \def\getichetta(#1){\veroparagrafo.\verafigura%
218: \SIA g,#1,{\veroparagrafo.\verafigura} %
219: \global\advance\numfig by 1%
220: \write15{\string\Fg (#1){\graf(#1)}}%
221: \write16{ Fig. #1 ==> \graf(#1) }}
222:
223: \def\etichettap(#1){\veroparagrafo.\verasub%
224: \SIA p,#1,{\veroparagrafo.\verasub} %
225: \global\advance\numsub by 1%
226: \write15{\string\Fp (#1){\secc(#1)}}%
227: \write16{ par #1 ==> \secc(#1) }}
228:
229:
230: \def\Eq(#1){\eqno{\etichetta(#1)\alato(#1)}}
231: \def\eq(#1){\etichetta(#1)\alato(#1)}
232: \def\Eqa(#1){\eqno{\etichettaa(#1)\alato(#1)}}
233: \def\eqa(#1){\etichettaa(#1)\alato(#1)}
234: \def\eqg(#1){\getichetta(#1)\alato(fig. #1)}
235: \def\sub(#1){\0\palato(p. #1){\bf \etichettap(#1).}}
236: \def\asub(#1){\0\palato(p. #1){\bf \etichettapa(#1).}}
237: \def\apprif(#1){\senondefinito{e#1}%
238: \eqv(#1)\else\csname e#1\endcsname\fi}
239:
240: \def\equv(#1){\senondefinito{fe#1}$\clubsuit$#1%
241: \write16{eq. #1 non e' (ancora) definita}%
242: \else\csname fe#1\endcsname\fi}
243: \def\grafv(#1){\senondefinito{fg#1}$\clubsuit$#1%
244: \write16{fig. #1 non e' (ancora) definito}%
245: \else\csname fg#1\endcsname\fi}
246: \def\secv(#1){\senondefinito{fp#1}$\clubsuit$#1%
247: \write16{par. #1 non e' (ancora) definito}%
248: \else\csname fp#1\endcsname\fi}
249:
250: \def\eqo{{\global\advance\numfor by 1}}
251: \def\equ(#1){\senondefinito{e#1}\equv(#1)\else\csname e#1\endcsname\fi}
252: \def\graf(#1){\senondefinito{g#1}\grafv(#1)\else\csname g#1\endcsname\fi}
253: \def\figura(#1){{\css Figura} \getichetta(#1)}
254: %\def\fig(#1){\0\veroparagrafo.\getichetta(#1)}
255: \def\secc(#1){\senondefinito{p#1}\secv(#1)\else\csname p#1\endcsname\fi}
256: %\def\sec(#1){{\S\secc(#1)}}
257: \def\sec(#1){{\secc(#1)}}
258: \def\refe(#1){{[\secc(#1)]}}
259:
260: \def\BOZZA{%\bz=1
261: \def\alato(##1){\rlap{\kern-\hsize\kern-.5truecm{$\scriptstyle##1$}}}
262: \def\palato(##1){\rlap{\kern-.5truecm{$\scriptstyle##1$}}}
263: }
264:
265: \def\alato(#1){}
266: \def\galato(#1){}
267: \def\palato(#1){}
268:
269: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
270: %%%%%%%%%%%%%%%%%%%% DATA E PIE' DI PAGINA %%%%%%%%%%%%%%%%%%%%%%
271: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
272:
273: {\count255=\time\divide\count255 by 60 \xdef\hourmin{\number\count255}
274: \multiply\count255 by-60\advance\count255 by\time
275: \xdef\hourmin{\hourmin:\ifnum\count255<10 0\fi\the\count255}}
276:
277: \def\oramin{\hourmin }
278:
279: \def\data{\number\day/\ifcase\month\or gennaio \or febbraio \or marzo \or
280: aprile \or maggio \or giugno \or luglio \or agosto \or settembre
281: \or ottobre \or novembre \or dicembre \fi/\number\year;\ \oramin}
282: \setbox200\hbox{$\scriptscriptstyle \data $}
283:
284: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
285: %%%%%%%%%%%%%%% INSERIMENTO FIGURE ( se si usa DVIPS ) %%%%%%%%%%%%%%
286: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
287: \newdimen\xshift \newdimen\xwidth \newdimen\yshift \newdimen\ywidth
288:
289: \def\ins#1#2#3{\vbox to0pt{\kern-#2\hbox{\kern#1 #3}\vss}\nointerlineskip}
290:
291: \def\eqfig#1#2#3#4#5{
292: \par\xwidth=#1 \xshift=\hsize \advance\xshift
293: by-\xwidth \divide\xshift by 2
294: \yshift=#2 \divide\yshift by 2%
295: %\line
296: {\hglue\xshift \vbox to #2{\vfil
297: #3 \special{psfile=#4.ps}
298: }\hfill\raise\yshift\hbox{#5}}}
299:
300: \def\8{\write12}
301:
302: \def\figini#1{
303: \catcode`\%=12\catcode`\{=12\catcode`\}=12
304: \catcode`\<=1\catcode`\>=2
305: \openout12=#1.ps}
306:
307: \def\figfin{
308: \closeout12
309: \catcode`\%=14\catcode`\{=1%
310: \catcode`\}=2\catcode`\<=12\catcode`\>=12}
311:
312: \openin13=#1.aux \ifeof13 \relax \else
313: \input #1.aux \closein13\fi
314: \openin14=\jobname.aux \ifeof14 \relax \else
315: \input \jobname.aux \closein14 \fi
316: \immediate\openout15=\jobname.aux
317:
318: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
319: %%%%%%%%%%%%%%%%%%%%%% SIMBOLI VARI %%%%%%%%%%%%%%%%%%%%%%%
320: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
321:
322: \let\a=\alpha \let\b=\beta \let\g=\gamma \let\d=\delta \let\e=\varepsilon
323: \let\z=\zeta \let\h=\eta \let\th=\theta \let\k=\kappa \let\l=\lambda
324: \let\m=\mu \let\n=\nu \let\x=\xi \let\p=\pi \let\r=\rho
325: \let\s=\sigma \let\t=\tau \let\f=\varphi \let\ph=\varphi\let\c=\chi
326: \let\ch=\chi \let\ps=\psi \let\y=\upsilon \let\o=\omega\let\si=\varsigma
327: \let\G=\Gamma \let\D=\Delta \let\Th=\Theta\let\L=\Lambda \let\X=\Xi
328: \let\P=\Pi \let\Si=\Sigma \let\F=\Phi \let\Ps=\Psi
329: \let\O=\Omega \let\Y=\Upsilon
330:
331: \def\\{\hfill\break} \let\==\equiv
332: \let\txt=\textstyle\let\dis=\displaystyle
333:
334: \let\io=\infty
335: \def\Dpr{\BDpr\,}%\def\Dpr{\V\dpr\,}
336: \def\aps{{\it a posteriori\ }}\def\ap{{\it a priori\ }}
337: \let\0=\noindent\def\pagina{{\vfill\eject}}
338: \def\bra#1{{\langle#1|}}\def\ket#1{{|#1\rangle}}
339: \def\media#1{{\langle#1\rangle}}
340: \def\ie{\hbox{\it i.e.\ }}\def\eg{\hbox{\it e.g.\ }}
341: %\def\ie{{i.e. }}\def\eg{{e.g. }}
342: \let\dpr=\partial \def\der{{\rm d}} \let\circa=\cong
343: \def\arccot{{\rm arccot}}
344:
345: \def\tende#1{\,\vtop{\ialign{##\crcr\rightarrowfill\crcr
346: \noalign{\kern-1pt\nointerlineskip} \hskip3.pt${\scriptstyle
347: #1}$\hskip3.pt\crcr}}\,}
348: \def\circage{\lower2pt\hbox{$\,\buildrel > \over {\scriptstyle \sim}\,$}}
349: \def\otto{\,{\kern-1.truept\leftarrow\kern-5.truept\to\kern-1.truept}\,}
350: \def\fra#1#2{{#1\over#2}}
351:
352: \def\PP{{\cal P}}\def\EE{{\cal E}} \def\MM{{\cal M}} \def\VV{{\cal V}}
353: \def\CC{{\cal C}}\def\FF{{\cal F}} \def\HH{{\cal H}} \def\WW{{\cal W}}
354: \def\TT{{\cal T}}\def\NN{{\cal N}} \def\BB{{\cal B}} \def\II{{\cal I}}
355: \def\RR{{\cal R}}\def\LL{{\cal L}} \def\JJ{{\cal J}} \def\OO{{\cal O}}
356: \def\DD{{\cal D}}\def\AA{{\cal A}} \def\GG{{\cal G}} \def\SS{{\cal S}}
357: \def\KK{{\cal K}}\def\UU{{\cal U}} \def\QQ{{\cal Q}} \def\XX{{\cal X}}
358:
359: \def\T#1{{#1_{\kern-3pt\lower7pt\hbox{$\widetilde{}$}}\kern3pt}}
360: \def\VVV#1{{\underline #1}_{\kern-3pt
361: \lower7pt\hbox{$\widetilde{}$}}\kern3pt\,}
362: \def\W#1{#1_{\kern-3pt\lower7.5pt\hbox{$\widetilde{}$}}\kern2pt\,}
363: \def\Re{{\rm Re}\,}\def\Im{{\rm Im}\,}
364: \def\lis{\overline}\def\tto{\Rightarrow}
365: \def\etc{{\it etc}} \def\acapo{\hfill\break}
366: \def\mod{{\rm mod}\,} \def\per{{\rm per}\,} \def\sign{{\rm sign}\,}
367: \def\indica{\leaders \hbox to 0.5cm{\hss.\hss}\hfill}
368: \def\guida{\leaders\hbox to 1em{\hss.\hss}\hfill}
369:
370:
371: \let\ciao=\bye
372: \def\qed{\raise1pt\hbox{\vrule height5pt width5pt depth0pt}}
373: \def\hf#1{{\hat \f_{#1}}}
374: \def\barf#1{{\tilde \f_{#1}}} \def\tg#1{{\tilde g_{#1}}}
375: \def\bq{{\bar q}}
376: \def\Val{{\rm Val}}
377: \def\indic{\hbox{\raise-2pt \hbox{\indbf 1}}}
378:
379: \def\RRR{\hbox{\msytw R}} \def\rrrr{\hbox{\msytww R}}
380: \def\rrr{\hbox{\msytwww R}} \def\CCC{\hbox{\msytw C}}
381: \def\cccc{\hbox{\msytww C}} \def\ccc{\hbox{\msytwww C}}
382: \def\NNN{\hbox{\msytw N}} \def\nnnn{\hbox{\msytww N}}
383: \def\nnn{\hbox{\msytwww N}} \def\ZZZ{\hbox{\msytw Z}}
384: \def\zzzz{\hbox{\msytww Z}} \def\zzz{\hbox{\msytwww Z}}
385: \def\TTT{\hbox{\msytw T}} \def\tttt{\hbox{\msytww T}}
386: \def\ttt{\hbox{\msytwww T}}
387: \def\QQQ{\hbox{\msytw Q}} \def\qqqq{\hbox{\msytww Q}}
388: \def\qqq{\hbox{\msytwww Q}}
389:
390: \def\vvv{\hbox{\euftw v}} \def\vvvv{\hbox{\euftww v}}
391: \def\vvvvv{\hbox{\euftwww v}}\def\www{\hbox{\euftw w}}
392: \def\wwww{\hbox{\euftww w}} \def\wwwww{\hbox{\euftwww w}}
393: \def\vvr{\hbox{\euftw r}} \def\vvvr{\hbox{\euftww r}}
394:
395: \def\Sqrt#1{{\sqrt{#1}}}
396: %\def\Sqrt#1{{{#1}^{\fra{1}{2}}}}
397:
398: \def\defi{\,{\buildrel def\over=}\,}
399: \def\lhs{{\it l.h.s.}\ }
400: \def\rhs{{\it r.h.s.}\ }
401: \def\cfr{{\it c.f.\ }}
402:
403: \def\sqr#1#2{{\vcenter{\vbox{\hrule height.#2pt%
404: \hbox{\vrule width.#2pt height#1pt \kern#1pt%
405: \vrule width.#2pt}%
406: \hrule height.#2pt}}}}
407: \def\dalam{{\,\mathchoice\sqr66\sqr55\sqr44\sqr33\,}}
408: \def\square{\mathchoice\sqr34\sqr34\sqr{2.1}3\sqr{1.5}3}
409: \def\didascalia#1#2{\0Fig.#1{ {\nota\it#2}}}
410: \def\ig{\int}
411:
412: \footline={\rlap{\hbox{\copy200}}\tenrm\hss \number\pageno\hss}
413: \def\V#1{{\bf#1}}
414: \let\wt=\widetilde
415: \def\fk{\f^{(\le k)}}
416:
417: %\usepackage{eqalignno}
418: \def\fiat{}
419:
420: %BOZZA
421: \def\asint#1{\,\vtop{\ialign{##\crcr\hss $\simeq$\hss\crcr
422: \noalign{\kern-1pt\nointerlineskip} \hskip3.pt${\scriptstyle
423: #1}$\hskip3.pt\crcr}}\,}
424:
425: \def\Tr#1{\,{\rm Tr}\,#1\,}
426: % \def++{(+,+)}\def\pm{(+,-)}
427:
428: \fiat
429: %**end of header
430:
431: %\vglue1.truecm
432: \centerline{\titolone Stationary nonequilibrium statistical mechanics}
433: \*
434:
435: \centerline
436: {\it Giovanni Gallavotti}
437:
438: \centerline
439: {\it I.N.F.N. Roma 1, Fisica Roma1}
440:
441: \*
442: \Section(1, Nonequilibrium.)
443: \*
444:
445: Systems in stationary nonequilibrium are mechanical systems subject to
446: nonconservative external forces and to thermostat forces which forbid
447: indefinite increase of the energy and allow reaching statistically
448: stationary states. A system $\Si$ is described by the positions and
449: velocities of its $n$ particles $\V X,\dot{\V X}$, with the particles
450: positions confined to a finite volume container $C_0$.
451:
452: If $\V X=(\V x_1,\ldots$, $\V x_n)$ are the particles positions in a
453: Cartesian inertial system of coordinates, the equations of motion are
454: determined by their masses $m_i>0,\, i=1,\ldots,n$, by the
455: {\it potential energy} of interaction $V(\V x_1,\ldots,\V x_n)\=V(\V
456: X)$, by the external nonconservative forces $\V F_i(\V X,\BF)$ and by the {\it
457: thermostat} forces $-\Bth_i$ as
458:
459: $$m_i {\ddot{\V x}}_i=-\dpr_{\V x_i} V(\V X)+ \V F_i(\V X; \BF)
460: -\Bth_i,\qquad i=1,\ldots,n\Eq(1.1)$$
461: %
462: where $\BF=(\f_1,\ldots,\f_q)$ are strength parameters on which the
463: external forces depend. All forces and potentials will be supposed
464: smooth, \ie analytic, in their variables aside from {\it possible}
465: impulsive elastic forces describing shocks, and with the property: $\V
466: F(\V X;\V 0)=\V0$. The impulsive forces are allowed here to model
467: possible shocks with the walls of the container $C_0$ or between hard
468: core particles.
469:
470: A kind of thermostats are {\it reservoirs} which may consist of one or
471: more {\it infinite} systems which are asymptotically in thermal
472: equilibrium and are separated by boundary surfaces from each other aas
473: well as from the system: with the latter they
474: interact through short range conservative forces, see Fig.1.
475: %\input fig1%
476: \eqfig{380pt}{100pt}
477: {\ins{110pt}{100pt}{\vbox{\hsize=250pt%
478: \0The reservoirs occupy infinite regions of the space outside $C_0$,
479: \eg sectors $C_a\subset \RRR^3$, $a=1,2\ldots$, in space and their
480: particles are in a configuration which is typical of an equilibrium
481: state at temperature $T_a$. This means that the {\it empirical}
482: probability of configurations in each $C_a$ is Gibbsian with some
483: temperature $T_a$. In other words the frequency with which a
484: configuration $(\dot {\V Y},\V Y+\V r)$ occurs in}}
485: \ins{60pt}{30pt}{Fig.1}
486: }{fig1}{}
487:
488: \vskip4pt
489: \0a region $\L+\V
490: r\subset C_a$ and a configuration $(\dot{\V W},\V W+\V r)$ occurs
491: outside $\L+\V r$ (with $\V Y\subset \L,\V W\cap\L$\\$=\emptyset$)
492: {\it averaged over the translations $\L+\V r$ of $\L$ by $\V r$} (with
493: the restriction that $\L+\V r\subset C_a$) is
494:
495: $${\rm average}\kern-30pt\lower8pt\hbox{$\st \V r+\L\subset C_a$}
496: (f_{\L+\V r}[(\dot{\V Y},\V Y+\V r); \dot{\V W},\V W+\V r])=
497: \fra{e^{-{\b_a}\big(\fra1{2m_a} | \dot{\V Y}|^2+ V_a(\V Y|\V
498: W)\big)}}{\rm normalization}\Eq(1.2)$$
499: %
500: here $m_a$ is the mass of the particles in the $a$-th reservoir and
501: $V_a(\V Y|\V W)$ is the energy of the short range potential between
502: pairs of particles in $\V Y\subset C_a$ or with one point in $\V Y$
503: and one in $\V W$. Since the configurations in the system and in the
504: thermostats are not random the \equ(1.2) should be considered as an
505: ``empirical'' probability in the sense that it is the frequency
506: density of the events $\{(\dot{\V Y},\V Y+\V r); \V W+\V r\}$: in
507: other words the configurations $\Bo_a$ in the reservoirs should be
508: ``typical'' in the sense of probability theory of distributions which
509: are asymptotically Gibbsian.
510:
511: The property of being ``thermostats'' means that
512: \equ(1.2) remains true for all times, if initially satisfied.
513:
514: Mathematically there is a problem at this point: the latter property
515: is either true or false, but a proof of its validity seems out of
516: reach of the present techniques except in very simple cases. Therefore
517: here we follow an intuitive approach and assume that such thermostats
518: exist and, actually, that any configuration which is typical of a
519: stationary state of an infinite size system of interacting particles
520: in the $C_a$'s, with physically reasonable microscopic interactions,
521: satisfies the property \equ(1.2).
522:
523: The above thermostats are examples of ``deterministic thermostats''
524: because, together with the system they form a deterministic dynamical
525: system. They are called ``Hamiltonian thermostats'' and are often
526: considered as the most appropriate models of ``physical thermostats''.
527:
528: A closely related thermostat model is obtained by assuming that the
529: particles outside the system are not in a given configuration but they
530: have a probability distribution whose conditional distributions
531: satisfy \equ(1.2) initially. Also in this case it is necessary to
532: assume that \equ(1.2) remains true for all times, if initially
533: satisfied. Such thermostats are examples of ``stochastic
534: thermostats'' because their action on the system depends on random
535: variables $\Bo_a$ which are the {\it initial} configurations of the
536: particles belonging to the thermostats.
537:
538: Other kinds of stochastic thermostats are {\it collision rules} with
539: the container boundary $\dpr C_0$ of $\Si$: every time a
540: particle collides with $\dpr C_0$ it is reflected with a momentum
541: $\V p$ in $d^3\V p$ that has a probability distribution proportional
542: to $e^{-\b_a\fra1{2m}\V p^2}d^3\V p$ where $\b_a$, $a=1,2,\ldots$
543: depend on which boundary portion (labeled by $a=1,2\ldots$ and, if
544: $k_B$ is Boltzmann's constant, at ``temperature'' $T_a=(k_B
545: \b_a)^{-1}$) the collision takes place. Which $\V p$ is actually
546: chosen after each collision is determined by a random variable
547: $\Bo=(\Bo_1,\Bo_2,\ldots)$.
548:
549: It is also possible, and convenient, to consider deterministic
550: thermostats which are {\it finite}. In the latter case
551: $\Bth$ is a force {\it only} depending upon the configuration of the
552: $n$ particles in their finite container $C_0$. The distinction between
553: stochastic and deterministic thermostats ultimately rests on what we
554: call ``system''. If reservoirs or the randomness generators are
555: included in the system then the system becomes deterministic (possibly
556: infinite); and finite deterministic thermostats can also regarded as
557: simplified models for infinite reservoirs, see Sect.\sec(4).
558:
559: Examples of finite deterministic reservoirs are forces obtained by
560: imposing a nonholonomic constraint via some {\it ad hoc} principle
561: like the Gauss' principle. For instance if a system of particles
562: driven by a force $\V G_i\defi-\Dpr_{\V x_i}V(\V X)+\V F_i(\V X)$ is
563: enclosed in a box $C_0$ and $\Bth$ is a thermostat enforcing an
564: anholonomic constraint $\ps(\dot{\V X},\V X)\=0$ via Gauss' principle
565: then
566:
567: $$\Bth_i(\dot{\V X},\V X)=
568: \Big[\fra{\sum_j \dot{\V x}_j\cdot\Dpr_{\V
569: x_j}\ps(\dot{\V X},\V X)+
570: \fra1m \V G_j\cdot \Dpr_{\dot{\V x}_j}\ps(\dot{\V X},\V X)}
571: {\sum_j \fra1m (\Dpr_{\dot{\V
572: x}_j}\ps(\dot{\V X},\V X))^2}\Big]\, \Dpr_{\dot{\V
573: x}_i}\ps(\dot{\V X},\V X)
574: \Eq(1.3)$$
575: %
576: Gauss' principle is remarkable as it says that the force which needs
577: to be added to the other forces $\V G_i$ acting on the system minimizes $\sum_i
578: \fra{(\V G_i- m_i\V a_i)^2}{m_i}$, given $\dot{\V X},\V X$,
579: among all accelerations $\V a_i$ which are compatible with the
580: constraint $\ps$.
581:
582: {\it For simplicity stochastic or infinite thermostats will not be
583: considered} here. It should be kept in mind that the only known
584: examples of mathematically treatable thermostats modeled by infinite
585: reservoirs are cases in which the thermostats particles are either
586: noninteracting particles or linear (\ie noninteracting) oscillators.
587:
588: In general in order that a force $\Bth$ can be considered a
589: deterministic ``thermostat force'' a further property is necessary:
590: namely that the system evolves according to \equ(1.1) towards a {\it
591: stationary state}. This means that for all initial particles
592: configurations $(\dot{\V X},\V X)$, {\it except possibly for a set of
593: zero phase space volume}, {\it any} smooth function $f(\dot{\V X},\V
594: X)$ evolves in time so that, if $S_t(\dot{\V X},\V X)$ denotes the
595: configuration into which the initial data evolve in time $t$ according
596: to \equ(1.1), then the limit
597:
598: $$\lim_{T\to\io} \fra1T \ig_0^T f(S_t(\dot{\V X},\V X))\,dt
599: =\ig f(z)\m(dz)\Eq(1.4)$$
600: %
601: exists and is independent of $(\dot{\V X},\V X)$. The probability
602: distribution $\m$ is then called the {\it SRB distribution} for the
603: system. The maps $S_t$ will have the group property $S_t\cdot
604: S_{t'}=S_{t+t'}$ and the SRB distribution $\m$ will be invariant under
605: time evolution.
606:
607: It is important to stress that the requirement that the exceptional
608: configurations form just a set of zero phase volume (rather than a set
609: of zero probability with respect to another distribution, singular
610: with respect to the phase volume) is a strong assumption and {\it it
611: should be considered an axiom of the theory}: it corresponds to the
612: assumption that the initial configuration is prepared as a typical
613: configuration of an equilibrium state, which by the classical
614: equidistribution axiom of equilibrium statistical mechanics is a
615: typical configuration with respect to the phase volume .
616:
617: For this reason the SRB distribution is said to describe a stationary
618: state of the system. The SRB distribution depends on the parameters on
619: which the forces acting on the system depend, \eg $|C_0|$ (volume),
620: $\BF$ (strength of the forcings), $\{\b_a^{-1}\}$ (temperatures)
621: \etc. The collection of SRB distributions obtained by letting the
622: parameters vary defines a {\it nonequilibrium ensemble}.
623:
624: In the stochastic case the distribution $\m$ is required to be
625: invariant in the sense that it can be regarded as a marginal
626: distribution of an invariant distribution for the larger
627: (deterministic) system formed by the thermostats and the system
628: itself.
629:
630: \*
631: \0{\it References}: [EM90], [Ru97a], [EPR99].
632: \*
633: %\pagina
634:
635: \Section(2, Nonequilibrium thermodynamics)
636: \*
637:
638: The key problem of nonequilibrium statistical mechanics is to derive a
639: macroscopic ``nonequilibrium thermodynamics'' in a way similar to the
640: derivation of equilibrium thermodynamics from equilibrium statistical
641: mechanics.
642:
643: The first difficulty is that nonequilibrium thermodynamics is not well
644: understood. For instance there is no (agreed upon) definition of {\it
645: entropy}, while it should be kept in mind that the effort to find the
646: microscopic interpretation of equilibrium entropy, as defined by
647: Clausius, was a driving factor in the foundations of equilibrium
648: statistical mechanics.
649:
650: The importance of entropy in classical equilibrium thermodynamics
651: rests on the implication of universal, parameter free, relations which
652: follow from its existence ({\it e.g.} $\dpr_V \fra1T\=\dpr_U\fra{p}T$ if
653: $U$ is the internal energy, $T$ the absolute temperature and $p$ the
654: pressure of a simple homogeneous material).
655:
656: Are there universal relations among averages of observables with
657: respect to SRB distributions?
658:
659: The question has to be posed for systems ``really'' out of
660: equilibrium, \ie for $\BF\ne\V0$ (see \equ(1.1)): in fact there is a
661: well developed theory of the derivatives with respect to $\BF$ of
662: averages of observables evaluated at $\BF=\V 0$. The latter theory is
663: often called, and here we shall do so as well, {\it classical
664: nonequilibrium thermodynamics} or {\it near equilibrium
665: thermodynamics} and it has been quite successfully developed on the
666: basis of the notions of equilibrium thermodynamics, paying particular
667: attention on the macroscopic evolution of systems described by
668: macroscopic continuum equations of motion.
669:
670: ``Stationary nonequilibrium statistical mechanics'' will indicate a
671: theory of the relations between averages of observables with respect
672: to SRB distributions. Systems so large that their volume elements can
673: be regarded as being in locally stationary nonequilibrium states could
674: also be considered. This would extend the familiar ``local equilibrium
675: states'' of classical nonequilibrium thermodynamics: however they are
676: not considered here. This means that we shall not attempt at finding
677: the macroscopic equations regulating the time evolution of continua
678: locally in nonequilibrium stationary states but we shall only try to
679: determine the properties of their ``volume elements'' assuming that
680: the time scale for the evolution of large assemblies of volume
681: elements is slow compared to the time scales necessary to reach local
682: stationarity.
683: \*
684: {\it References}: [DGM84], [Le93], [Ru97a], [Ru99a], [Ga98], [GL03], [Ga04].
685: \*
686:
687: \Section(3, Chaotic hypothesis)
688: \*
689:
690: In equilibrium statistical mechanics the {\it ergodic hypothesis}
691: plays an important conceptual role as it implies that the motions of
692: ergodic systems have a SRB statistics and that the latter coincides
693: with the Liouville distribution on the energy surface.
694:
695: An analogous role has been proposed for the {\it chaotic hypothesis}:
696: which states that the {\it motion of a chaotic system, developing on
697: its attracting set, can be regarded as an Anosov system}. This means
698: that the attracting sets of chaotic systems, physically defined as systems
699: with at least one positive Lyapunov exponent, can be regarded as
700: smooth surfaces on which motion is highly unstable:
701: \*
702:
703: \0(i) around every point a curvilinear coordinate system can be
704: established which has three planes, varying continuously with $x$,
705: which are {\it covariant} (\ie are coordinate planes at a point $x$
706: which are mapped, by the evolution $S_t$, into the corresponding
707: coordinate planes around $S_t x$) and
708: \\
709: (ii) the planes are of three types, {\it stable, unstable and
710: marginal}, with respective positive dimensions $d_s,d_u$ and $1$:
711: lengths on the stable surface and on the unstable surface of any point
712: contract at exponential rate as time proceeds towards the future or
713: towards the past. The length along the marginal direction neither
714: contracts nor expands (\ie it varies around the initial value staying
715: bounded away from $0$ and $\io$): its tangent vector is parallel to
716: the flow. In cases in which time evolution is discrete, and determined
717: by a map $S$, the marginal direction is missing.
718: \\
719: (iii) the contraction over a time $t$, positive for lines on the
720: stable plane and negative for those on the unstable plane, is
721: exponential, \ie lengths are contracted by a factor uniformly
722: bounded by $C e^{-\k |t|}$ with $C,\k>0$.
723: \\
724: (iv) there is a dense trajectory.
725:
726: \kern2pt
727: It has to be stressed that the chaotic hypothesis concerns physical
728: systems: mathematically {\it it is very easy to find dynamical systems
729: for which it does not hold}. As it is easy (actually even easier) to
730: find systems in which the ergodic hypothesis does not hold (\eg
731: harmonic lattices or black body radiation). However, if suitably
732: interpreted, the ergodic hypothesis leads even for these systems to
733: physically correct results (the specific heats at high temperature,
734: the Raileigh-Jeans distribution at low frequencies). Moreover the
735: failures of the ergodic hypothesis in physically important systems
736: have led to new scientific paradigms (like quantum mechanics from the
737: specific heats at low temperature and Planck's law).
738:
739: Since physical systems are almost always not Anosov systems it is very
740: likely that probing motions in extreme regimes will make visible the
741: features that distinguish Anosov systems from non Anosov systems: much
742: as it happens with the ergodic hypothesis.
743:
744: The interest of the hypothesis is to provide a framework in which
745: properties like the existence of an SRB distribution is \ap
746: guaranteed: the role of Anosov systems in chaotic dynamics is similar
747: to the role of harmonic oscillators in the theory of regular motions.
748: They are the paradigm of chaotic systems as the harmonic oscillators
749: are the paradigm of order. Of course the hypothesis is only a
750: beginning and one has to learn how to extract information from it, as
751: it was the case with the use of the Liouville distribution once the
752: ergodic hypothesis guaranteed that it was the appropriate distribution
753: for the study of the statistics of motions in equilibrium situations.
754: \*
755: \0{\it References}: [Ru76], [GC95], [Ru97a], [Ga98], [GBG04].
756: \*
757: %\pagina
758:
759: \Section(4, Heat, temperature and entropy production)
760: \*
761:
762: The amount of heat $\dot Q$ that a system produces while in a stationary state
763: is naturally identified with the work that the thermostat forces
764: $\Bth$ perform per unit time
765:
766: $$\dot Q=\sum_i \Bth_i\cdot\dot{\V x}_i\Eq(4.1)$$
767: %
768: A system may be in contact with several reservoirs:
769: in models this will be reflected by a decomposition
770:
771: $$\Bth=\sum_{a=1}^m \Bth^{(a)}(\dot{\V X},\V X)\Eq(4.2)$$
772: %
773: where $\Bth^{(a)}$ is the force due to the $a$-th thermostat and depends
774: on the coordinates of the particles which are in a region
775: $\L_a\subseteq C_0$ of a decomposition $\cup_{a=1}^m \L_a= C_0$ of the
776: container $C_0$ occupied by the system ($\L_a\cap\L_{a'}=\emptyset$ if
777: $a\ne a'$).
778:
779: From several studies based on simulations of thermostated systems of
780: particles arose the proposal to consider the average of the phase space
781: contraction $\s^{(a)}(\dot{\V X},\V X)$ due to the $a$-th thermostat
782:
783: $$\s^{(a)}(\dot{\V X},\V X)\defi \sum_j
784: \dpr_{\dot{\V x}_j}\cdot \Bth^{(a)}_j (\dot{\V X},\V X)\Eq(4.3)$$
785: %
786: and to identify it with the {\it rate of entropy creation} in the $a$-th
787: thermostat.
788:
789: Another key notion in thermodynamics is the {\it temperature} of a
790: reservoir; in the infinite deterministic thermostats case, of Sect.\sec(1),
791: it is defined as $(k_B
792: \b_a)^{-1}$ but in the finite determinstic thermostats
793: considered here it needs to be defined. If there are $m$ reservoirs
794: with which the system is in contact one sets
795:
796: $$\eqalign{
797: \s^{(a)}_+&\defi
798: \media{\s^{(a)}(\dot{\V X},\V X)}\=\ig \s^{(a)}(\dot{\V X},\V X)
799: \,\m(d \dot{\V X}\,d\V X)\cr
800: \dot Q_a&\defi\sum_i \Bth^{(a)}_i\cdot\dot{\V x}_i\cr}\Eq(4.4)$$
801: %
802: where $\m$ is the SRB distribution describing the stationary state. It
803: is natural to define, if $k_B$ is Boltzmann's constant,
804: the {\it absolute temperature} of the $a$-th
805: thermostat to be
806:
807: $$T_a=\fra{\media{\dot Q_a}}{k_B\s^{(a)}_+}.\Eq(4.5)$$
808: %
809:
810: Although it is known that $\fra{\sum_a \media{\dot Q_a}}{\sum_a
811: \s^{(a)}_+}\ge0$ it is not clear that
812: $T_a>0$: this happens in a rather general class of models and it would
813: be desirable, for the interpretation that is proposed here, that it
814: could be considered a property to be added to the requirements that
815: the forces $\Bth^{a}$ be thermostats models.
816:
817: An important class of thermostats for which the property $T_a>0$ holds
818: can be described as follows. Imagine $N$ particles in a container
819: $C_0$ interacting via a potential $V_0=\sum_{i<j} \f (\V q_i-\V
820: q_j)+\sum_j V'(\V q_j)$ (where $V'$ models external conservative
821: forces like obstacles, walls, gravity, $\ldots$) and, furthermore,
822: interacting with $M$ other systems $\Si_a$, of $N_a$ particles of mass
823: $m_a$, in containers $C_a$ contiguous to $C_0$. The latter will model
824: $M$ parts of the system in contact with thermostats at temperatures
825: $T_a$, $a=1,\ldots,M$.
826:
827: The coordinates of the particles in the $a$-th system $\Si_a$ will be
828: denoted $\V x^a_j, \, j=1,\ldots,N_a$, and they will interact with
829: each other via a potential $V_a=\sum_{i,j}^{N_a}
830: \f_a(\V x^a_i-\V x^a_j)$. Furthermore there will be an interaction
831: between the particles of each thermostat and those of the system via
832: potentials $W_a=\sum_{i=1}^N\sum_{j=1}^{N_a} w_a(\V q_i-\V
833: x^a_j)$, $a=1,\ldots,M$.
834:
835: The potentials will be assumed to be either hard core or non singular
836: potentials and the external $V'$ is supposed to be at least such that
837: it forbids existence of obvious constants of motion.
838:
839: The temperature of each $\Si_a$ will be defined by the total kinetic
840: energy of its particles, \ie by $K_a=\sum_{j=1}^{N_a} \fra12 m_a (\dot{\V
841: x}^a_j)^2\defi \fra32 N_a k_B T_a$: the particles of the $a$-th
842: thermostat will be kept at constant temperature by further forces
843: $\Bth^a_j$. The latter are defined by imposing via a Gaussian
844: constraint that $K_a$ is a constant of motion (see \equ(1.3) with
845: $\ps\= K_a$) . This means that the equations of motion are
846:
847: $$\eqalign{m \,\ddot{\V q}_j&= -\dpr_{\V q_j} \big(V_0(\V Q) +\sum_{a=1}^{N_a}
848: W_a(\V Q,\V x^a)\big)\cr
849: m_a\, \ddot{\V x}^a_j&= -\dpr_{\V x^a_j} \big( V_a(\V x^a)+W_a(\V Q,\V
850: x^a)\big)-
851: \Bth^a_j\cr}
852: \Eq(4.6)$$
853: %
854: and an application of Gauss' principle yields $\Bth_j^a=
855: \fra{L_a-\dot V_a} {3N_a k_B T_a}\,\,\dot{\V x}^a_j\defi \a^a \,\dot{\V
856: x}^a_j $ where $L_a$ is the work per unit time done by the particles
857: in $C_0$ on the particles of $\Si_a$ and $V_a$ is their potential
858: energy.
859:
860: In this case the partial divergence $\s^a=3 N_a \a^a=
861: \fra{L_a}{k_B T_a}-
862: \fra{\dot V_a}{k_B T_a}$ will make \equ(4.5) identically satisfied
863: with $T_a>0$ because $L_a$ can be naturally interpreted as {\it heat}
864: $Q_a$ ceded, per unit time, by the particles in $C_0$ to the subsystem
865: $\Si_a$ (hence to the $a$-th thermostat because the temperature of
866: $\Si_a$ is constant), while the derivative of $V_a$ {\it will not
867: contribute to the value of} $\s^a_+$. The phase space contraction
868: rate is, {\it neglecting the total derivative terms},
869: %
870: $$\s_{true}(\dot{\V X},\V X) = \sum_{a=1}^{N_a}
871: \fra{\dot Q_a}{k_B T_a}\,.\Eq(4.7)$$
872: %
873: where the subscript ``true'' is to remind that an additive total
874: derivative term distinguishes it from the complete phase space
875: contraction.
876:
877: \0{\it Remarks:} (1)
878: The above formula provides the motivation of the name ``entropy
879: creation rate'' attributed to the phase space contraction $\s$. Note
880: that in this way the definition of entropy creation is ``reduced'' to
881: the equilibrium notion because what is being defined is the entropy
882: increase of the thermostats which have to be considered in
883: equilibrium. No attempt is made here to define the entropy of the
884: stationary state. Nor any attempt is made to define the notion of
885: temperature of the nonequilibrium system in $C_0$ (the $T_a$ are temperatures of
886: the $\Si_a$, not of the particles in $C_0$). This is an important
887: point as it leaves open the possibility of envisaging the notion of
888: ``local equilibrium'' which becomes necessary in the approximation
889: (not considered here) in which the system is regarded as a continuum.
890:
891: \0(2) In the above model another viewpoint is possible: \ie to consider the
892: system to consist of only the $N$ particles in $C_0$ and the $M$
893: systems $\Si_a$ to be thermostats. From this point of view the above
894: can be considered a model of a system subject to thermostats. The
895: Gibbs distribution characterizing the infinite thermostats of
896: Sect.\sec(1) becomes in this case the constraint that the kinetic
897: energies $K_a$ are constants, enforced by the Gaussian forces. This
898: shows that the phase space contraction can be an appropriate
899: definition of entropy creation rate only if the system is subject to
900: finite deterministic thermostats. In the new viewpoint the appropriate
901: definition should be simply the \rhs of \equ(4.7), \ie {\it the work per
902: unit time done by the forces of the system on the thermostats divided
903: by the temperature of the thermostats}. This suggests a more general
904: definition of entropy creation rate, applying also to thermostats that
905: are often considered ``more physical'' and that needs to be further
906: investigated.
907:
908:
909: \*
910: \0{\it References:} [EM90], [GC95], [Ru96], [Ru97b], [Ga04].
911:
912: \*
913: \Section(5, Thermodynamic fluxes and forces)
914: \*
915:
916: Nonequilibrium stationary states depend upon external parameters
917: $\f_j$ like the temperatures $T_a$ of the thermostats or the size of
918: the force parameters $\BF=(\f_1,\ldots,\f_q)$, see\ \equ(1.1).
919: Nonequilibrium thermodynamics is well developed at ``low forcing'':
920: strictly speaking this means that it is widely believed that we
921: understand properties of the derivatives of the averages of
922: observables with respect to the external parameters {\it if evaluated
923: at $\f_j=0$}. Important notions are the notions of {\it thermodynamic
924: fluxes} $J_i$ and of {\it thermodynamic forces} $\f_i$; hence it seems
925: important to extend such notions to nonequilibrium systems (\ie
926: $\BF\ne\V0$).
927:
928: A possible extension could be to define the thermodynamic flux $J_i$
929: associated with a force $\f_i$ as $J_i=\media{\dpr_{\f_i} \s}_{SRB}$
930: where $\s(\V X,\dot{\V X}; \BF)$ is the volume contraction per unit
931: time. This definition seems appropriate in several concrete cases that
932: have been studied and it is appealing for its generality.
933:
934: An interesting example is provided by the model of thermostated
935: system in \equ(4.6): if the container of the system is a box with
936: periodic boundary conditions one can imagine to add an extra constant
937: force $\V E$ acting on the particles in the container. Imagining the
938: particles to be charged by a charge $e$ and regarding such force as an
939: electric field the first equation in \equ(4.6) is modified by the
940: addition of a term $e \V E$.
941:
942: The constraints on the thermostats temperatures imply that
943: $\s$ depends also on $\V E$: in fact, if $\V J=e\sum_j \dot{\V q}_j$
944: is the electric current, energy balance implies $\dot U_{tot}=\V
945: E\cdot\V J-\sum_a (L_a-\dot V_a)$ if $U_{tot}$ is the sum of {\it all}
946: kinetic and potential energies. Then the phase space contraction
947: $\sum_a \fra{L_a-\dot V_a}{T_a}$ can be written, to first order in the
948: temperature variations $\d T_a$ with respect to a common value
949: $T_a=T$, as $ -\sum_a \fra{L_a-\dot V_a}T
950: \fra{\d T_a}{T}+\fra{\V
951: E\cdot\V J-\dot U_{tot}}{T}$ hence $\s_{true}$, see \equ(4.7), is
952:
953: $$\s_{true}=\fra{\V E\cdot \V J}{k_BT}-\sum_a
954: \fra{\dot Q_a}{k_B T}\fra{\d T_a}{T}
955: %- \fra{\dot U_{tot}}{k_B T}
956: \Eq(5.1)$$
957: %
958:
959: The definition and extension of the conjugacy between thermodynamic
960: forces and fluxes is compatible with the key results of classical
961: nonequilibrium thermodynamics, at least as far as
962: Onsager reciprocity and Green-Kubo's formulae are concerned. It can be
963: checked that if the equilibrium system is {\it reversible}, \ie if
964: there is an isometry $I$ on phase space which anticommutes with the
965: evolution ($I S_t=S_{-t} I$ in the case of continuous time dynamics
966: $t\to S_t$ or $IS=S^{-1}I$ in the case of discrete time dynamics $S$)
967: then, shortening $(\dot{\V X},\V X)$ into $x$,
968:
969: $$\eqalign{
970: L_{ij}\defi&\dpr_{\F_i} J_j|_{\BF=\V0}=
971: \dpr_{\F_i} \media{\dpr_{\F_j} \s(x;\BF)}_{SRB}|_{\BF=\V0}
972: =\dpr_{\F_j} J_i|_{\BF=\V0}=\cr
973: &=L_{ji}=\fra12\ig_{-\io}^\io
974: \media{\dpr_{\F_j} \s(S_t x; \BF)\,\dpr_{\F_i} \s(x;
975: \BF)}_{SRB}\Big|_{\BF=\V0}\, dt\cr}
976: \Eq(5.2)$$
977: %
978: The $\s( x; \BF)$ plays the role of ``Lagrangian'' generating the
979: duality between forces and fluxes. The extension of the duality just
980: considered might be of interest in situations in which $\BF\ne\V0$.
981: \*
982: \0{\it References:} [DGM84],[Ga96],[GR97].
983: \*
984:
985: \Section(6, Fluctuations)
986: \*
987:
988: As in equilibrium, large statistical fluctuations of observables are of great
989: interest and already there is, at the moment, a rather large set of
990: experiments dedicated to the analysis of large fluctuations in
991: stationary states out of equilibrium.
992:
993: If one defines the dimensionless phase space contraction
994:
995: $$p(x)=\fra1\t \ig_0^\t \fra{\s(S_t x) }{\s_+}\, dt\Eq(6.1)$$
996: %
997: (see also \equ(4.7)) then there exists $p^*\ge1$ such that the
998: probability $P_\t$ of the event $p\in [a,b]$ with $[a,b]\subset
999: (-p^*,p^*)$ has the form
1000:
1001: $$P_\t(p\in[a,b])\,=\,const\, e^{\t \max_{p\in [a,b]} \z(p)
1002: +O(1)}\Eq(6.2)$$
1003: %
1004: with $\z(p)$ analytic in $(-p^*,p^*)$. The function $\z(p)$ can be
1005: conveniently normalized to have value $0$ at $p=1$ (\ie at the average
1006: value of $p$).
1007:
1008: Then, {\it in Anosov systems which are reversible and dissipative}
1009: (see Sect.\sec(5)) a general symmetry property, called the {\it
1010: fluctuation theorem} and reflecting the reversibility symmetry, yields
1011: the {\it parameterless} relation
1012:
1013: $$\z(-p)=\z(p)-p\s_+ \qquad p\in(-p^*,p^*)\Eq(6.3)$$
1014: %
1015: This relation is interesting because it has no free parameters, in
1016: other words it is {\it universal} for reversible dissipative Anosov
1017: systems. In connection with the duality fluxes-forces in Sec.5, it can
1018: be checked to reduce to the Green--Kubo formula and to Onsager
1019: reciprocity, see \equ(5.2), in the case in which the evolution depends
1020: on several fields $\BF$ and $\BF\to\V0$ (of course the relation
1021: becomes trivial as $\BF\to\V0$ because $\s_+\to0$ and to obtain the
1022: result one has first to divide both sides by suitable powers of the fields
1023: $\BF$).
1024:
1025: A more informal (but imprecise) way of writing \equ(6.2),\equ(6.3) is
1026:
1027: $$\fra{P_\t(p)}{P_\t(-p)}=e^{\t p \s_++ O(1)}, \qquad \ {\rm for\ all} \
1028: p\in(-p^*,p^*)
1029: \Eq(6.4)$$
1030: %
1031: where $P_\t(p)$ is the probability density of $p$. An interesting
1032: consequence
1033: of \equ(6.4) is $\media{e^{-\t\,p\,\s_+}}_{SRB}=1$ in the sense that
1034: $\fra1\t \log \media{e^{-\t\,p\,\s_+}}_{SRB}\tende{\t\to\io}0$.
1035:
1036: Occasionally systems with singularities have to be considered: in such
1037: cases the relation \equ(6.3) may change in the sense that the function
1038: $\z(p)$ may be not analytic: in such cases one expects that the
1039: relation holds in the largest analyticity interval symmetric around
1040: the origin. In various cases considered in the literature such
1041: interval appears to contain the interval $(-1,1)$ and sometimes this
1042: can be proved rigorously. For instance in simple, although admittedly
1043: special, examples of systems close to equilibrium.
1044:
1045: It is important to remark that the above fluctuation relation is the
1046: first representative of remarkable consequences of the reversibility
1047: and chaotic hypotheses. For instance given $F_1,\ldots,F_n$ {\it
1048: arbitrary} observables which are (say) odd under time reversal $I$ (\ie
1049: $F(I x)=-F(x)$) and given $n$ functions $t\in[-\fra\t2,\fra\t2]\to
1050: \f_j(t)$, $j=1,\ldots,n$ one can ask which is the probability that
1051: $F_j(S_tx)$ ``closely follows'' the {\it pattern} $\f_j(t)$ and at the
1052: same time $\fra1\t\ig_0^\t
1053: \fra{\s(S_\th x)}{\s_+}\,d\th$ has value $p$. Then calling
1054: $P_\t(F_1\sim\f_1,\ldots,F_n\sim\f_n,p)$ the probability of this event,
1055: which we write in the imprecise form corresponding to \equ(6.4) for
1056: simplicity, and defining $I\f_j(t)\defi-\f_j(-t)$ it is
1057:
1058: $$\fra{P_\t(F_1\sim\f_1,\ldots,F_n\sim\f_n,p)}
1059: {P_\t(F_1\sim I\f_1,\ldots,F_n\sim I\f_n,-p)}
1060: =e^{\t \s_+ p}, \qquad p\in (-p^*,p^*)\Eq(6.5)$$
1061: %
1062: which is remarkable because it is parameterless and at the same time
1063: surprisingly independent of the choice of the observables $F_j$. The
1064: relation \equ(6.5) has far reaching consequences: for instance if
1065: $n=1$ and $F_1=\dpr_{\F_i}\s(x;\BF)$ the relation \equ(6.5) has been
1066: used to derive the mentioned Onsager reciprocity and Green--Kubo's
1067: formulae at $\BF=\V0$.
1068:
1069: Eq. \equ(6.5) can be read as follows: the probability that the
1070: observables $F_j$ follow given {evolution patterns} $\f_j$ conditioned
1071: to entropy creation rate $p\s_+$ is {\it the same} that they follow
1072: the time reversed patterns if conditioned to entropy creation rate
1073: $-p\s_+$. In other words to change the sign of time it is {\it just}
1074: sufficient to reverse the sign of entropy creation rate, no ``extra
1075: effort'' is needed.
1076: \*
1077: \0{\it References:} [Si72],[Si94],[ECM93],%
1078: [GC85],[Ga96],[GR97],[Ga99],[GBG04],[BGGZ05].
1079: \*
1080:
1081: \Section(7, Fractal attractors. Pairing. Time reversal.)
1082: \*
1083:
1084: Attracting sets (\ie sets which are the closure of attractors) are
1085: fractal in most dissipative systems. However the chaotic hypothesis
1086: assumes that fractality can be neglected. Aside from the very
1087: interesting cases in which systems are close to equilibrium, in which
1088: the closure of an attractor is the whole phase space (under the
1089: chaotic hypothesis, \ie if the system is Anosov) there are, however,
1090: serious problems in preserving validity of the fluctuation theorem.
1091:
1092: The reason is very simple: if the attractor closure is smaller than
1093: phase space then it is to be expected that time reversal will change
1094: the attractor into a repeller disjoint from it. Thus even if the
1095: chaotic hypothesis is assumed, so that the attracting set $\AA$ can be
1096: considered a smooth surface, the motion on the attractor will not be
1097: time reversal symmetric (as its time reversal image will develop on
1098: the repeller): one can say that an attracting set with dimension lower
1099: than that of phase space in a time reversible system corresponds to a
1100: {\it spontaneous breakdown of time reversal symmetry}.
1101:
1102: It has been noted however that there are classes of systems, forming a
1103: large set in the space of evolutions depending on a parameter $\F$, in
1104: which geometric reasons imply that if beyond a critical value $\F_c$
1105: the attracting set becomes smaller than phase space then a map $I_P$
1106: is generated mapping the attractor $\AA$ into the repeller $\RR$, and
1107: {\it viceversa}, such that $I_P^2$ is the identity on $\AA\cup\RR$ and $I_P$
1108: {\it commutes with the evolution}: therefore the composition $I\cdot
1109: I_P$ is a time reversal symmetry (\ie it anticommutes with evolution)
1110: for the motions on the attracting set $\AA$ (as well as on the
1111: repeller $\RR$).
1112:
1113: In other words the time reversal symmetry in such systems ``cannot be
1114: broken'': if spontaneous breakdown occurs (\ie $\AA$ is not mapped into
1115: itself under time reversal $I$) a new symmetry $I_P$ is spawned and
1116: $I\cdot I_P$ is a new time reversal symmetry (an analogy with the
1117: spontaneous violation of time reversal in quantum theory, where time
1118: reversal $T$ is violated but $TCP$ is still a symmetry: so $T$ plays
1119: the role of $I$ and $CP$ that of $I_P$).
1120:
1121: Thus a fluctuation relation will hold for the phase space contraction
1122: of the motions taking place on the attracting set for the class of
1123: systems with the geometric property mentioned above (technically the
1124: latter is called {\it axiom C} property). This is interesting but it
1125: still is quite far from being checkable even in numerical experiments.
1126:
1127: There are nevertheless systems in which also a {\it pairing} property
1128: holds: this means that, considering the case of discrete time maps
1129: $S$, the Jacobian matrix $\dpr_x S(x)$ has $2N$ eigenvalues that can
1130: be labeled, in decreasing order, $\l_N(x),\ldots,\l_{\fra12 N}(x),$
1131: $\ldots,\l_1(x)$ with the remarkable property that
1132: $\fra12(\l_{N-j}(x)+\l_j(x))\defi\a(x)$ is $j$--independent. In such
1133: systems a relation can be established between phase space
1134: contractions in the full phase space and on the surface of the
1135: attracting set: the fluctuation theorem for the motion on the
1136: attracting set can therefore be related to the properties of the
1137: fluctuations of the total phase space contraction measured on the
1138: attracting set (which includes the contraction transversal to the
1139: attracting set) and if $2M$ is the attracting set dimension and $2N$
1140: is the total dimension of phase space it is, in the analyticity
1141: interval $(-p^*,p^*)$ of the function $\z(p)$,
1142:
1143: $$\z(-p)=\z(p)-p \fra{M}N \s_+\Eq(7.1)$$
1144: %
1145: which is an interesting relation. It is however very difficult to
1146: test in mechanical systems because in such systems it seems very
1147: difficult to make the field so high to see an attracting set thinner
1148: than the whole phase space and {\it still} observe large
1149: fluctuations.
1150: \*
1151:
1152: \0{\it References:} [DM96], [Ga99].
1153: \*
1154:
1155:
1156: \*
1157: \Section(8, Nonequilibrium ensembles and their equivalence)
1158: \*
1159:
1160: Given a chaotic system the collection of the SRB distributions
1161: associated with the various control parameters (volume, density,
1162: external forces, ...) forms an ``ensemble'' describing the possible
1163: stationary states of the system and their statistical properties.
1164:
1165: As in equilibrium one can imagine that the system can be described
1166: equivalently in several ways at least when the system is large (``in
1167: the thermodynamic'' or ``macroscopic limit''). In
1168: nonequilibrium equivalence can be quite different and more structured
1169: than in equilibrium because one can imagine to change not only the
1170: control parameters but also the thermostatting mechanism.
1171:
1172: It is intuitive that a system may behave in the same way under the
1173: influence of different thermostats: the important phenomenon being the
1174: extraction of heat and not the way in which it is extracted from the
1175: system. Therefore one should ask when two systems are ``physically
1176: equivalent'', \ie when the SRB distributions associated with them give
1177: the same statistical properties for the same observables, at least for
1178: the {\it very few} observables which are macroscopically relevant. The
1179: latter may be a few more than the usual ones in equilibrium
1180: (temperature, pressure, density, ...) and include currents,
1181: conducibilities, viscosities, ... but they will always be very few
1182: compared to the (infinite) number of functions on phase space.
1183:
1184: As an example consider a system of $N$ interacting particles (say hard
1185: spheres) of mass $m$ moving in a periodic box $C_0$ of side $L$
1186: containing a regular array of spherical scatterers (a basic model for
1187: electrons in a crystal) which reflect particles, elastically and are
1188: arranged so that no straight line exixts in $C_0$ which avoids the
1189: obstacles (to eliminate obvious constants of motion). An external
1190: field $E \V u$ acts also along the $\V u$-direction: hence the
1191: equations of motion are
1192:
1193: $$ m\ddot{\V x}_i=\V f_i+ E\V u-\Bth_i\Eq(8.1)$$
1194: %
1195: where $\V f_i$ are the interparticle and scatterers-particles forces
1196: and $\Bth_i$ are the thermostating forces. The following thermostats
1197: models have been considered
1198:
1199: \0(1) $\Bth_i= \n\,\dot {\V x}_i$ ({\it viscosity thermostat})
1200:
1201: \0(2) immediately after elastic collision with an obstacle the
1202: velocity is rescaled to a prefixed value $\sqrt{3 k_B Tm^{-1}}$ for some $T$
1203: ({\it Drude's thermostat})
1204:
1205: \0(3) $\Bth_i=\fra{\V E\cdot \sum \dot{\V x}_i}{\sum_i\dot{\V x}_i^2}$
1206: ({\it Gauss' thermostat})
1207:
1208: The first two are not reversible. At least not manifestly such,
1209: because the natural time reversal, \ie change of velocity sign, is not
1210: a symmetry (there might be however more hidden, hitherto unknown, symmetries
1211: which anticommute with time evolution). The third is reversible and
1212: time reversal is just the change of the velocity sign. The third
1213: thermostat model generates a time evolution in which the total kinetic
1214: energy $K$ is constant.
1215:
1216: Let $\m'_\n,\m''_T,\m'''_K$ be the SRB distributions for the system in
1217: a container $C_0$ with volume $|C_0|=L^3$ and density
1218: $\r=\fra{N}{L^3}$ fixed. Imagine to tune the values of the control
1219: parameters $\n,T,K$ in such a way that $\media{\rm kinetic\
1220: energy}_\m=\EE$, {\it with the same} $\EE$ for
1221: $\m=\m'_\n,\m''_T,\m'''_K$ and consider a {\it local} observable
1222: $F(\dot{\V X},\V X)>0$ depending only on the coordinates of the
1223: particles located in a region $\L\subset C_0$. Then a reasonable
1224: conjecture is that
1225:
1226: $$\lim_{L\to\io\atop \fra{N}{L^3}=\r}
1227: \fra{\media{F}_{\m'_\n}}{\media{F}_{\m''_T}}=
1228: \lim_{L\to\io\atop \fra{N}{L^3}=\r}
1229: \fra{\media{F}_{\m'_\n}}{\media{F}_{\m'''_T}}=1\Eq(8.2)$$
1230: %
1231: if the limits are taken at fixed $F$ (hence at fixed $\L$ while
1232: $L\to\to\io$). The conjecture is an open problem: it illustrates,
1233: however, the a kind of questions arising in nonequilibrium statistical
1234: mechanics.
1235:
1236: \*
1237: \0{\it References:} [ES93],[Ga99],[Ru99b].
1238:
1239: \*
1240: \Section(9, Outlook)
1241: \*
1242:
1243: \0The subject is (clearly) at a very early stage of development.
1244: \*
1245:
1246: (1) The theory can be extended to stochastic thermostats quite
1247: satisfactorily, at least as far as the fluctuation theorem is
1248: concerned.
1249:
1250: (2) Remarkable works have appeared on the theory of systems which are
1251: purely Hamiltonian and (therefore) with thermostats that are
1252: infinite: unfortunately the infinite thermostats can be treated,
1253: so far, only if the systems are ``free'' at infinity (either free
1254: gases or harmonic lattices).
1255:
1256: (3) The notion of entropy turns out to be extremely difficult to
1257: extend to stationary states and there are even doubts that it could
1258: be actually extended. Conceptually this is certainly a major open
1259: problem.
1260:
1261: (4) The statistical properties of stationary states out of equilibrium
1262: are still quite mysterious and surprising: recently appeared exactly
1263: solvable models as well as attempts at unveiling the deep reasons
1264: for their solubility and at deriving from them general guiding
1265: principles.
1266:
1267: (5) Numerical simulations have given a strong impulse to the subject,
1268: in fact one can even say that they created it: introducing the
1269: notion of thermostat and providing the first reliable results on
1270: the properties of systems out of equilibrium. Simulations continue
1271: to be an essential part of the effort of research on the field.
1272:
1273: (6) Approach to stationarity leads to many important questions: is
1274: there a Lyapunov function measuring the distance between an
1275: evolving state and the stationary state towards which it evolves?
1276: In other words can one define an analogous of Boltzmann's
1277: $H$-function? About this question there have been proposals
1278: and the answer seems affirmative, but it does not seem
1279: that it is possible to find a universal, system independent,
1280: such function (search for it is related to the problem of defining an
1281: entropy function for stationary states: its existence is at least
1282: controversial, see Sec.2,3).
1283:
1284: (7) How irreversible is a given {\it irreversible process} in which
1285: the initial state $\m_0$ is a stationary state and at time $t=0$
1286: and the external parameters $\BF_{0}$ start changing into functions
1287: $\BF(t)$ of $t$ and tend to a limit $\BF_{\io}$ as $t\to\io$? In
1288: this case the stationary distribution $\m_0$ starts changing and
1289: becomes a function $\m_t$ of $t$ which {\it is not stationary} but
1290: approaches another stationary distribution $\m_\io$ as
1291: $t\to\io$. The process is, in general, irreversible and the
1292: question is how to measure its {\it degree of irreversibility}. A
1293: natural quantity $\II$ associated with the evolution from an
1294: initial stationary state to a final stationary state through a
1295: change in the control parameters can be defined as
1296: follows. Consider the distribution $\m_t$ into which $\m_0$
1297: evolves in time $t$, and consider also the SRB distribution
1298: $\m_{\BF(t)}$ corresponding to the control parameters ``frozen''
1299: at the value at time $t$, \ie $\BF(t)$. Let the phase space
1300: contraction, when the forces are ``frozen'' at the
1301: value $\BF(t)$, be $\s_t(x)=\s(x;\BF(t))$. In general $\m_t\ne
1302: \m_{\BF(t)}$. Then
1303:
1304: $$\II(\{\BF(t)\},\m_{0},\m_{\io})\defi \ig_0^\io
1305: (\m_t(\s_t)-\m_{\BF(t)}(\s_t))^2\,dt\Eq(9.1)$$
1306: %
1307: can be called the {\it degree of irreversibility of the process}: it has the
1308: property that in the limit of infinitely slow evolution of $\BF(t)$,
1309: \eg if $\BF(t)=\BF_0+(1- e^{-\g \k t})\BD $ (a {\it quasi static
1310: evolution} on time scale $\g^{-1}\k^{-1}$ from $\BF_0$ to
1311: $\BF_\io=\BF_0+\BD$), the irreversibility degree
1312: $\II_\g\tende{\g\to0}0$ if (as in the case of Anosov evolutions, hence
1313: under the chaotic hypothesis) the approach to a stationary state is
1314: exponentially fast at fixed external forces $\BF$.
1315: \*
1316:
1317: The entire subject is dominated by the initial insights of Onsager on
1318: classical nonequilibrium thermodynamics: which concern the
1319: properties of the infinitesimal deviations from equilibrium (\ie
1320: averages of observables differentiated with respect to the control
1321: parameters $\BF$ and evaluated at $\BF=\V0$). The present efforts are
1322: devoted to studying properties at $\BF\ne\V0$. In this direction the
1323: classical theory provides certainly firm constraints (like Onsager
1324: reciprocity or Green-Kubo relations or fluctuation dissipation
1325: theorem) but at a technical level it gives little help to enter
1326: the {\it terra incognita} of nonequilibrium thermodynamics of
1327: stationary states.
1328:
1329: \*
1330: \halign{#&#\hfill\cr
1331: \0{\it References:}&\ [Ku98],[LS99],[Ma99]; [EPR99],[BLR00],[EY05];
1332: [DLS01];[BDGJL01]; \cr
1333: &
1334: [EM90],[ECM33]; [GL03],[Ga04],[BGGZ05].\cr}
1335: \*
1336:
1337: \0{\it Bibliography}
1338: \*
1339:
1340: \def\*{\vskip1mm}
1341:
1342: \0[BDGJL01] L. Bertini, A. De Sole, D. Gabrielli, G. Jona, C. Landim:
1343: {\it Fluctuations in Stationary Nonequilibrium States of Irreversible
1344: Processes}, Physical Review Letters, {\bf??}, , 2001.87 (2001) 040601
1345:
1346: \*\0[BLR00] F. Bonetto, J.L. Lebowitz, L. Rey-Bellet: {\it Fourier's
1347: law: a challenge to theorists}, in ``Mathematical Physics 2000'', p. 128--150,
1348: editor R. Streater, Imperial College Press, 2000.
1349:
1350: \*\0[BGGZ05] F. Bonetto, G. Gallavotti, A. Giuliani, F. Zamponi:
1351: {\it Chaotic Hypothesis, Fluctuation Theorem and Singularities},
1352: mp$\_$arc 05-257, cond-mat/0507672
1353:
1354: \*\0[DM96] C. Dettman, G.P. Morriss: {\it Proof of conjugate pairing
1355: for an isokinetic thermostat}, Physical Review {\bf 53 E}, 5545--5549,
1356: 1996.
1357:
1358: \*\0[DGM84] S. de Groot, P., Mazur: {\it Non equilibrium
1359: thermodynamics}, Dover, 1984, (reprint).
1360:
1361: \*\0[DLS01] B. Derrida, J.L. Lebowitz, E. Speer: {\it Free Energy
1362: Functional for Nonequilibrium Systems: An Exactly Solvable Case},
1363: Physical Review Letters, 2001.
1364:
1365: \*\0[ECM93] D.J. Evans, E.G.D. Cohen, G., Morriss:
1366: {\it Probability of second law violations in shearing steady flows},
1367: Physical Review Letters, {\bf 70}, 2401--2404, 1993.
1368:
1369: \*\0[EM90] D.J. Evans and G.P. Morriss,
1370: {\it Statistical Mechanics of Nonequilibrium Fluids}, Academic Press,
1371: 1990, New-York.
1372:
1373: \*\0[EPR99] J.P. Eckmann, C.A. Pillet, L. Rey Bellet:
1374: {\it Non-Equilibrium Statistical Mechanics of Anharmonic Chains Coupled to
1375: Two Heat Baths at Different Temperatures},
1376: Communications in Mathematical Physics, {\bf 201}, 657 - 697, 1999.
1377:
1378: \*\0[ES93] D.J. Evans, S. Sarman: {\it Equivalence of
1379: thermostatted nonlinear responses}, Physical Review, {\bf E 48},
1380: 65--70, 1993.
1381:
1382: \*\0[EY05] J.P. Eckmann, L.S. Young: {\it Nonequilibrium energy
1383: profiles for a class of 1D models}, Communications in Mathematical
1384: Physics, {\bf??}, ????, 2005.
1385:
1386: \*\0[Ga96] G. Gallavotti: {\it Chaotic hypothesis: Onsager reciprocity
1387: and fluctuation--dissipation theorem}, Journal of Statistical
1388: Physics, {\bf 84}, 899--926, 1996.
1389:
1390: \*\0[Ga98] G. Gallavotti,
1391: {\it Chaotic dynamics, fluctuations, non-equilibrium ensembles},
1392: Chaos, {\bf 8}, 384--392, 1998.
1393:
1394: \*\0[Ga99] G. Gallavotti: {\it Statistical Mechanics}, Springer Verlag,
1395: Berlin, 1999.
1396:
1397: \*\0[Ga04] G. Gallavotti,
1398: {\it Entropy production in nonequilibrium stationary states: a
1399: point of view}, Chaos, {\bf 14}, 680--690, 2004.
1400:
1401: \*\0[GBG04] G. Gallavotti, F. Bonetto, G. Gentile: {\it
1402: Aspects of the ergodic, qualitative and statistical theory of
1403: motion}, Springer--Verlag, p.1--434, 2004.
1404:
1405: \*\0[GC95] G. Gallavotti, E.G.D. Cohen: {\it Dynamical
1406: ensembles in non-equilibrium statistical mechanics}, Physical Review
1407: Letters, {\bf74}, 2694--2697, 1995.
1408:
1409: \*\0[GR97] G. Gallavotti, D. Ruelle: {\it SRB states and
1410: nonequilibrium statistical mechanics close to equilibrium},
1411: Communications in Mathematical Physics, {\bf190} (1997), 279--285.
1412:
1413: \*\0[GL03] S. Goldstein, J. Lebowitz,
1414: {\it On the ({B}oltzmann) entropy of Nonequilibrium systems}, Physica
1415: D, {\bf 193}, 53--66, 2004.
1416:
1417: \*\0[Ku98] J. Kurchan, Fluctuation theorem for stochastic dynamics,
1418: Journal of Physics A, 31, 3719--3729, 1998.
1419:
1420: \*\0[Le93] J.L. Lebowitz: {\it Boltzmann's entropy
1421: and time's arrow}, Physics Today, 32--38, 1993 (september).
1422:
1423: \*\0[LS99] J.L. Lebowitz, H. Spohn: {\it The Gallavotti-Cohen Fluctuation
1424: Theorem for Stochastic Dynamics}, Journal of Statistical
1425: Physics, {\bf 95}, 333--365, 1999.
1426:
1427: \*\0[Ma99] C. Maes: {\it The Fluctuation Theorem as a {G}ibbs Property},
1428: Journal of Statistical Physics, {\bf 95}, 367--392, 1999.
1429:
1430: \*\0[Ru76] D. Ruelle: {\it A measure associated with axiom $A$
1431: attractors}, American Journal of Mathematics {\bf98}, 619--654, 1976.
1432:
1433: \*\0[Ru96] Ruelle, D.: {\it Positivity of entropy production in
1434: nonequilibrium statistical mechanics}, Journal of Statistical Physics,
1435: {\bf 85}, 1--25, 1996.
1436:
1437: \*\0[Ru97a] Ruelle, D.: {\it Smooth dynamics and new theoretical
1438: ideas in non-equilibrium statistical mechanics}, Journal of
1439: Statistical Physics, {\bf 95}, 393--468, 1999.
1440:
1441: \*\0[Ru97b] D. Ruelle: {\it Entropy production in
1442: nonequilibrium statistical mechanics}, Communications in Mathematical
1443: Physics, {\bf189}, 365--371, 1997.
1444:
1445: \*\0[Ru99a] Ruelle, D.: {\it A remark on the equivalence of isokinetic
1446: and isoenergetic thermostats in the thermodynamic limit}, Journal of
1447: Statistical Physics, {\bf100}, 757--763, 2000.
1448:
1449: \*\0[Ru99b] D. Ruelle: {\it A remark on the equivalence of isokinetic
1450: and isoenergetic thermostats in the thermodynamic limit}, Journal of
1451: Statistical Physics, {\bf100}, 757--763, 2000.
1452:
1453: \*\0[Si72] Ya.G. Sinai: {\it Gibbs measures in ergodic theory},
1454: Russian Mathematical Surveys {\bf 166} (1972), 21--69.
1455:
1456: \*\0[Si94] Ya.G. Sinai: {\it Topics in ergodic theory},
1457: Princeton Mathematical Series, Vol. 44, Princeton University Press,
1458: Princeton, 1994.
1459:
1460: %\input riferimenti
1461:
1462: \end
1463: