1: %% "Quantum Force due to Distinct Boundary Conditions
2: %% T. Fulop, H. Miyazaki and I. Tsutsui
3:
4: %%% 1 September 2003
5: %%% Plain TeX
6:
7: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
8: %%%%%%%% Collection of macro's for use %%%%%%%%%
9: %%%%%%%% with plain tex %%%%%%%%%
10: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
11:
12: %%% First some fonts %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
13: \font\bigbold=cmbx12
14: \font\ninerm=cmr9 \font\eightrm=cmr8 \font\sixrm=cmr6
15: \font\fiverm=cmr5
16: \font\ninebf=cmbx9 \font\eightbf=cmbx8 \font\sixbf=cmbx6
17: \font\fivebf=cmbx5
18: \font\ninei=cmmi9 \skewchar\ninei='177 \font\eighti=cmmi8
19: \skewchar\eighti='177 \font\sixi=cmmi6 \skewchar\sixi='177
20: \font\fivei=cmmi5
21: \font\ninesy=cmsy9 \skewchar\ninesy='60 \font\eightsy=cmsy8
22: \skewchar\eightsy='60 \font\sixsy=cmsy6 \skewchar\sixsy='60
23: \font\fivesy=cmsy5 \font\nineit=cmti9 \font\eightit=cmti8
24: \font\ninesl=cmsl9 \font\eightsl=cmsl8
25: \font\ninett=cmtt9 \font\eighttt=cmtt8
26: \font\tenfrak=eufm10 \font\ninefrak=eufm9 \font\eightfrak=eufm8
27: \font\sevenfrak=eufm7 \font\fivefrak=eufm5
28: \font\tenbb=msbm10 \font\ninebb=msbm9 \font\eightbb=msbm8
29: \font\sevenbb=msbm7 \font\fivebb=msbm5
30: \font\tenssf=cmss10 \font\ninessf=cmss9 \font\eightssf=cmss8
31: \font\tensmc=cmcsc10
32: \font\mfd=msbm10
33:
34: %%% bold font for math vectors %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35: \font\tensvec=cmmib10
36: \font\sevensvec=cmmib10 at 7pt
37: \font\fivesvec=cmmib10 at 5 pt
38: \def\svec#1{{\mathchoice%
39: {\hbox{$\displaystyle\textfont1=\tensvec%
40: \scriptfont1=\sevensvec%
41: \scriptscriptfont1=\fivesvec#1$}}%
42: {\hbox{$\textstyle\textfont1=\tensvec%
43: \scriptfont1=\sevensvec%
44: \scriptscriptfont1=\fivesvec#1$}}%
45: {\hbox{$\scriptstyle\scriptfont1=\sevensvec%
46: \scriptscriptfont1=\fivesvec#1$}}%
47: {\hbox{$\scriptscriptstyle\scriptscriptfont1=\fivesvec#1$}}%
48: }}
49:
50: %%% Some Families %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
51: \newfam\bbfam \textfont\bbfam=\tenbb \scriptfont\bbfam=\sevenbb
52: \scriptscriptfont\bbfam=\fivebb \def\Bbb{\fam\bbfam}
53: \newfam\frakfam \textfont\frakfam=\tenfrak \scriptfont\frakfam=%
54: \sevenfrak \scriptscriptfont\frakfam=\fivefrak \def\frak{\fam\frakfam}
55: \newfam\ssffam \textfont\ssffam=\tenssf \scriptfont\ssffam=\ninessf
56: \scriptscriptfont\ssffam=\eightssf \def\ssf{\fam\ssffam}
57: \def\smc{\tensmc}
58:
59: %%% Definition of 8 point %%%%%%%%%%%%%%%%%%%%%%%%%%%%
60: \def\eightpoint{\textfont0=\eightrm \scriptfont0=\sixrm
61: \scriptscriptfont0=\fiverm \def\rm{\fam0\eightrm}%
62: \textfont1=\eighti \scriptfont1=\sixi \scriptscriptfont1=\fivei
63: \def\oldstyle{\fam1\eighti}\textfont2=\eightsy
64: \scriptfont2=\sixsy \scriptscriptfont2=\fivesy
65: \textfont\itfam=\eightit \def\it{\fam\itfam\eightit}%
66: \textfont\slfam=\eightsl \def\sl{\fam\slfam\eightsl}%
67: \textfont\ttfam=\eighttt \def\tt{\fam\ttfam\eighttt}%
68: \textfont\frakfam=\eightfrak \def\frak{\fam\frakfam\eightfrak}%
69: \textfont\bbfam=\eightbb \def\Bbb{\fam\bbfam\eightbb}%
70: \textfont\bffam=\eightbf \scriptfont\bffam=\sixbf
71: \scriptscriptfont\bffam=\fivebf \def\bf{\fam\bffam\eightbf}%
72: \abovedisplayskip=9pt plus 2pt minus 6pt \belowdisplayskip=%
73: \abovedisplayskip \abovedisplayshortskip=0pt plus 2pt
74: \belowdisplayshortskip=5pt plus2pt minus 3pt \smallskipamount=%
75: 2pt plus 1pt minus 1pt \medskipamount=4pt plus 2pt minus 2pt
76: \bigskipamount=9pt plus4pt minus 4pt \setbox\strutbox=%
77: \hbox{\vrule height 7pt depth 2pt width 0pt}%
78: \normalbaselineskip=9pt \normalbaselines \rm}
79:
80: %%% Definition of 9 point %%%%%%%%%%%%%%%%%%%%%%%%%%%%
81: \def\ninepoint{\textfont0=\ninerm \scriptfont0=\sixrm
82: \scriptscriptfont0=\fiverm \def\rm{\fam0\ninerm}\textfont1=\ninei
83: \scriptfont1=\sixi \scriptscriptfont1=\fivei \def\oldstyle%
84: {\fam1\ninei}\textfont2=\ninesy \scriptfont2=\sixsy
85: \scriptscriptfont2=\fivesy
86: \textfont\itfam=\nineit \def\it{\fam\itfam\nineit}%
87: \textfont\slfam=\ninesl \def\sl{\fam\slfam\ninesl}%
88: \textfont\ttfam=\ninett \def\tt{\fam\ttfam\ninett}%
89: \textfont\frakfam=\ninefrak \def\frak{\fam\frakfam\ninefrak}%
90: \textfont\bbfam=\ninebb \def\Bbb{\fam\bbfam\ninebb}%
91: \textfont\bffam=\ninebf \scriptfont\bffam=\sixbf
92: \scriptscriptfont\bffam=\fivebf \def\bf{\fam\bffam\ninebf}%
93: \abovedisplayskip=10pt plus 2pt minus 6pt \belowdisplayskip=%
94: \abovedisplayskip \abovedisplayshortskip=0pt plus 2pt
95: \belowdisplayshortskip=5pt plus2pt minus 3pt \smallskipamount=%
96: 2pt plus 1pt minus 1pt \medskipamount=4pt plus 2pt minus 2pt
97: \bigskipamount=10pt plus4pt minus 4pt \setbox\strutbox=%
98: \hbox{\vrule height 7pt depth 2pt width 0pt}%
99: \normalbaselineskip=10pt \normalbaselines \rm}
100:
101: %%% Macro to generate the equation #'s automatically.
102: %%% To use start each new section (eg 3) with the commands
103: %%% \secno=3 \meqno=1 :this will start the equations with (3.1)
104: %%% Then in place of \eqno(3.1) type \eqn\descriptivename . To refer
105: %%% back to the equation simply type (\descritivename)
106: %%% For the appendixset \secno=0, \appno=1\meqno=1 etc
107: %%%
108: \global\newcount\secno \global\secno=0 \global\newcount\meqno
109: \global\meqno=1 \global\newcount\appno \global\appno=0
110: \newwrite\eqmac \def\romappno{\ifcase\appno\or A\or B\or C\or D\or
111: E\or F\or G\or H\or I\or J\or K\or L\or M\or N\or O\or P\or Q\or
112: R\or S\or T\or U\or V\or W\or X\or Y\or Z\fi}
113: \def\eqn#1{ \ifnum\secno>0 \eqno(\the\secno.\the\meqno)
114: \xdef#1{\the\secno.\the\meqno} \else\ifnum\appno>0
115: \eqno({\rm\romappno}.\the\meqno)\xdef#1{{\rm\romappno}.\the\meqno}
116: \else \eqno(\the\meqno)\xdef#1{\the\meqno} \fi \fi
117: \global\advance\meqno by1 }
118:
119: %%% Macro to do the refs %%%%%%%%%%%%%%%%%%%%%%%%%%%%%
120: \global\newcount\refno \global\refno=1 \newwrite\reffile
121: \newwrite\refmac \newlinechar=`\^^J \def\ref#1#2%
122: {\the\refno\nref#1{#2}} \def\nref#1#2{\xdef#1{\the\refno}
123: \ifnum\refno=1\immediate\openout\reffile=refs.tmp\fi
124: \immediate\write\reffile{\noexpand\item{[\noexpand#1]\ }#2\noexpand%
125: \nobreak.} \immediate\write\refmac{\def\noexpand#1{\the\refno}}
126: \global\advance\refno by1} \def\semi{;\hfil\noexpand\break ^^J}
127: \def\nl{\hfil\noexpand\break ^^J} \def\refn#1#2{\nref#1{#2}}
128: %\def\listrefs{\vfill\eject\immediate\closeout\reffile%\parindent=20pt
129: \def\listrefs{\yex{7}\immediate\closeout\reffile%\parindent=20pt
130: \centerline{{\bf References}}\bigskip\frenchspacing%
131: \input refs.tmp\vfill\eject\nonfrenchspacing}
132:
133: \def\ann#1#2#3{{\it Ann.\ Phys.}\ {\bf {#1}} ({#2}) #3}
134: \def\cmp#1#2#3{{\it Commun.\ Math.\ Phys.}\ {\bf {#1}} ({#2}) #3}
135: \def\jgp#1#2#3{{\it J.\ Geom.\ Phys.}\ {\bf {#1}} ({#2}) #3}
136: \def\jmp#1#2#3{{\it J.\ Math.\ Phys.}\ {\bf {#1}} ({#2}) #3}
137: \def\jpA#1#2#3{{\it J.\ Phys.}\ {\bf A{#1}} ({#2}) #3}
138: \def\ijmp#1#2#3{{\it Int.\ J.\ Mod.\ Phys.}\ {\bf A{#1}} ({#2}) #3}
139: \def\ijtp#1#2#3{{\it Int.\ J.\ Theor.\ Phys.}\ {\bf {#1}} ({#2}) #3}
140: \def\mplA#1#2#3{{\it Mod.\ Phys.\ Lett.}\ {\bf A{#1}} ({#2}) #3}
141: \def\nc#1#2#3{{\it Nuovo Cimento} {\bf {#1}D} ({#2}) #3}
142: \def\np#1#2#3{{\it Nucl.\ Phys.}\ {\bf B{#1}} ({#2}) #3}
143: \def\pl#1#2#3{{\it Phys.\ Lett.}\ {\bf B{#1}} ({#2}) #3}
144: \def\plA#1#2#3{{\it Phys.\ Lett.}\ {\bf {#1}A} ({#2}) #3}
145: \def\pr#1#2#3{{\it Phys.\ Rev.}\ {\bf {#1}} ({#2}) #3}
146: \def\prA#1#2#3{{\it Phys.\ Rev.}\ {\bf A{#1}} ({#2}) #3}
147: \def\Prb#1#2#3{{\it Phys.\ Rev.}\ {\bf B{#1}} ({#2}) #3}
148: \def\prD#1#2#3{{\it Phys.\ Rev.}\ {\bf D{#1}} ({#2}) #3}
149: \def\prl#1#2#3{{\it Phys.\ Rev.\ Lett.}\ {\bf #1} ({#2}) #3}
150: \def\ptp#1#2#3{{\it Prog.\ Theor.\ Phys.}\ {\bf {#1}} ({#2}) #3}
151: \def\rmp#1#2#3{{\it Rev.\ Mod.\ Phys.}\ {\bf {#1}} ({#2}) #3}
152: \def\prp#1#2#3{{\it Phys.\ Rep.}\ {\bf {#1}C} ({#2}) #3}
153: \def\zpc#1#2#3{{\it Z.\ Phys.}\ {\bf C{#1}} ({#2}) #3}
154:
155: %%% Numbering does not start on title page %%%%%%%%%%%
156: \newif\iftitlepage \titlepagetrue \newtoks\titlepagefoot
157: \titlepagefoot={\hfil} \newtoks\otherpagesfoot \otherpagesfoot=%
158: {\hfil\tenrm\folio\hfil} \footline={\iftitlepage\the\titlepagefoot%
159: \global\titlepagefalse \else\the\otherpagesfoot\fi}
160:
161: %%% Abstract %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
162: \def\abstract#1{{\parindent=30pt\narrower\noindent\ninepoint\openup
163: 2pt #1\par}}
164:
165: %%% A nicer footnote (\note) %%%%%%%%%%%%%%%%%%%%%%%%%
166: \newcount\notenumber\notenumber=1 \def\note#1
167: {\unskip\footnote{$^{\the\notenumber}$} {\eightpoint\openup 1pt #1}
168: \global\advance\notenumber by 1}
169:
170: %%% Date %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
171: \def\today{\ifcase\month\or January\or February\or March\or
172: April\or May\or June\or July\or August\or September\or October\or
173: November\or December\fi \space\number\day, \number\year}
174:
175: %%% More general stuff %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
176: \def\pagewidth#1{\hsize= #1} \def\pageheight#1{\vsize= #1}
177: \def\hcorrection#1{\advance\hoffset by #1}
178: \def\vcorrection#1{\advance\voffset by #1}
179:
180: %%% Output layout of the text %%%%%%%%%%%%%%%%%%%%%%%%
181: \pageheight{23cm}
182: \pagewidth{15.7cm}
183: \hcorrection{-1mm}
184: \magnification= \magstep1
185: \parskip=5pt plus 1pt minus 1pt
186: \tolerance 8000
187: \def\bsk{\baselineskip= 14.5pt plus 1pt minus 1pt}
188: \bsk
189:
190: %%% Definition of extra symbols R, C, Z, N %%%%%%%%%%%
191: \font\extra=cmss10 scaled \magstep0 \setbox1 = \hbox{{{\extra R}}}
192: \setbox2 = \hbox{{{\extra I}}} \setbox3 = \hbox{{{\extra C}}}
193: \setbox4 = \hbox{{{\extra Z}}} \setbox5 = \hbox{{{\extra N}}}
194: \def\RRR{{{\extra R}}\hskip-\wd1\hskip2.0true pt{{\extra I}}
195: \hskip-\wd2 \hskip-2.0 true pt\hskip\wd1}
196: \def\Real{\hbox{{\extra\RRR}}} % Actual special symbol: R
197: \def\CCC{{{\extra C}}\hskip-\wd3\hskip 2.5 true pt{{\extra I}}
198: \hskip-\wd2\hskip-2.5 true pt\hskip\wd3}
199: \def\Complex{\hbox{{\extra\CCC}}\!\!} % Actual special symbol: C
200: \def\ZZZ{{{\extra Z}}\hskip-\wd4\hskip 2.5 true pt{{\extra Z}}}
201: \def\Zed{\hbox{{\extra\ZZZ}}} % Actual special symbol: Z
202: \def\NNN{{{\extra N}}\hskip-\wd5\hskip -1.5 true pt{{\extra I}}}
203: \def\En{\hbox{{\extra\NNN}}} % Actual special symbol: N
204:
205: %%% Some useful macros %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
206: \def\frac#1#2{{#1\over#2}}
207: \def\dfrac#1#2{{\displaystyle{#1\over#2}}}
208: \def\tfrac#1#2{{\textstyle{#1\over#2}}}
209: \def\ket#1{|#1\rangle}
210: \def\bra#1{\langle#1|}
211: \def\pmb#1{\setbox0=\hbox{$#1$} \kern-.025em\copy0\kern-\wd0
212: \kern.05em\copy0\kern-\wd0 \kern-.025em\raise.0433em\box0 }
213: \def\noblackboxes{\overfullrule 0pt}
214: \def\ve{\vfill\eject}
215: \def\pa{\partial}
216: \def\tr{\hbox{tr}}
217: \def\la{\langle}
218: \def\ra{\rangle}
219:
220: %%% special macros for the paper:
221: %% Some useful macros %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
222: \def\frac#1#2{{#1\over#2}}
223: \def\dfrac#1#2{{\displaystyle{#1\over#2}}}
224: \def\tfrac#1#2{{\textstyle{#1\over#2}}}
225: \def\ket#1{|#1\rangle}
226: \def\bra#1{\langle#1|}
227: \def\pmb#1{\setbox0=\hbox{$#1$} \kern-.025em\copy0\kern-\wd0
228: \kern.05em\copy0\kern-\wd0 \kern-.025em\raise.0433em\box0 }
229: \def\noblackboxes{\overfullrule 0pt}
230: \def\ve{\vfill\eject}
231: \def\pa{\partial}
232: \def\tr{\hbox{tr}}
233: \def\Z{{\Zed}}
234: \def\R{{\Real}}
235: \def\C{{\Complex}}
236: \def\({\left(}
237: \def\){\right)}
238: \def\la{\langle}
239: \def\ra{\rangle}
240:
241:
242: \def\omit#1{} \def\vomit#1{#1} %% The difference is only one letter
243: \def\HB{\hbox} \def\SB{\setbox1\HB} %% For definitions that use boxes
244: \def\HC{\hbox} \def\SC{\setbox2\HC}
245: \def\XB{\wd1} \def\YB{\ht1} \def\CB{\copy1} \def\RB#1{\raise#1\CB}
246: \def\XC{\wd2} \def\YC{\ht2} \def\CC{\copy2} \def\RC#1{\raise#1\CC}
247: \def\ph{\phantom} %% Kiprobalt tomor valtozat
248: \def\xph{\hphantom} \def\xex#1{\x#1ex} \def\yex#1{\y#1ex}
249: \def\yph{\vphantom} \def\xpt#1{\x#1truept} \def\ypt#1{\y#1truept}
250: \def\x{\hskip} \def\xcm#1{\x#1truecm} \def\ycm#1{\y#1truecm}
251: \def\y{\vskip} \def\xmm#1{\x#1truemm} \def\ymm#1{\y#1truemm}
252: \def\bi{} \def\bitt{\xex{.3}} \def\biTTT{\xex{-.45}}
253: \def\bit{\xex{.15}} \def\biTT{\xex{-.3}} \def\bitttt{\xex{.6}}
254: \def\biT{\xex{-.15}} \def\bittt{\xex{.45}} \def\biTTTT{\xex{-.6}}
255: \def\m#1{$#1$} \def\mmm#1{$\,\,#1\,\,$}
256: \def\mm#1{$\,#1\,$} \def\mmmm#1{$\,\,\,#1\,\,\,$}
257: \def\hf{\hfill} \def\newline{\hfill\break}
258:
259: \def\smile{{\tt :-)}} \def\smiley{\smile} \def\mosoly{\smile}
260: \def\m#1{$#1$} \def\mm#1{$\,#1\,$} \def\mmm#1{$\,\,#1\,\,$}
261: \def\mmmm#1{$\,\,\,#1\,\,\,$}
262: \def\E#1#2{$$ #2 \eqn{#1} $$}
263: \def\f#1#2{{#1\over#2}} %% Smaller and smaller fractions
264: \def\ff#1#2{\raise.5pt\hbox{\eightpoint${\displaystyle\f{#1}{#2}}$}}
265: \def\fff#1#2{{\textstyle \f{#1}{#2}}}
266: \def\sectit#1{\bigskip\noindent{\bf #1}\smallskip}
267: \def\quotes#1{`#1'} \def\idez#1{\quotes{#1}}
268: \def\quotess#1{``#1''} \def\idezz#1{\x0ex\sB{"}\rB{-1.35ex}#1"}
269: \def\emp#1{{\sl #1}} \def\hangs{\emp}
270: \def\rem#1{({\it #1})} \def\megj{\rem}
271: \def\QED{{\it Q.E.D.}}
272: \def\eg{{\it e.g.,\ }} %% pl.-t
273: \def\ie{{\it i.e.,\ }}
274: \def\etc{{\it etc.}} %% stb-t
275: \def\etal{{\it et al.}}
276: \def\tbw{\rem{tbw}}
277: \def\cf{cf.\ } %% ld.-t
278: \def\lhs{l.h.s.\ } \def\rhs{r.h.s.\ } \def\wrt{w.r.t.\ }
279: \def\kb{kb.\ } %% De \approx -t is \kb -vel roviditettem mar...
280: \def\ii{\' \i}
281: %% \F: For fractions P/Q. Easy to switch to and from \f
282: %% When switching from \f, some "wrapping" may be needed:
283: %% \f{P}{Q} R --> \G{P}{Q} R : P/Q R is ambigous: (P/Q)R or P/(QR)?
284: %% \f{A+B}{C} --> \Fa{A+B}{C} : To give (A+B)/C instead of A + B/C
285: %% abc, xyz: inner wrappings
286: %% \G versions: Add outer wrapping as well (the appropriate one)
287: \def\F#1#2{#1/#2} \def\G#1#2{(#1/#2)} %% P/Q
288: \def\Fa#1#2{(#1)/#2} \def\Ga#1#2{[(#1)/#2]} %% (P)/Q
289: \def\Fb#1#2{[#1]/#2} \def\Gb#1#2{\{[#1]/#2\}} %% [P]/Q
290: \def\Fc#1#2{\{#1\}/#2} \def\Gc#1#2{(\{#1\}/#2)} %% {P}/Q
291: \def\Fx#1#2{#1/(#2)} \def\Gx#1#2{[#1/(#2)]} %% P/(Q)
292: \def\Fy#1#2{#1/[#2]} \def\Gy#1#2{\{#1/[#2]\}} %% P/[Q]
293: \def\Fz#1#2{#1/\{#2\}} \def\Gz#1#2{(#1/\{#2\})} %% P/{Q}
294: \def\Fax#1#2{(#1)/(#2)} \def\Gax#1#2{[(#1)/(#2)]} %% (P)/(Q)
295: \def\Fay#1#2{(#1)/[#2]} \def\Gay#1#2{\{(#1)/[#2]\}} %% (P)/[Q]
296: \def\Faz#1#2{(#1)/\{#2\}} \def\Gaz#1#2{[(#1)/\{#2\}]} %% (P)/{Q}
297: \def\Fbx#1#2{[#1]/(#2)} \def\Gbx#1#2{\{[#1]/(#2)\}} %% [P]/(Q)
298: \def\Fby#1#2{[#1]/[#2]} \def\Gby#1#2{\{[#1]/[#2]\}} %% [P]/[Q]
299: \def\Fbz#1#2{[#1]/\{#2\}} \def\Gbz#1#2{([#1]/\{#2\})} %% [P]/{Q}
300: \def\Fcx#1#2{\{#1\}/(#2)} \def\Gcx#1#2{[\{#1\}/(#2)]} %% {P}/(Q)
301: \def\Fcy#1#2{\{#1\}/[#2]} \def\Gcy#1#2{(\{#1\}/[#2])} %% {P}/[Q]
302: \def\Fcz#1#2{\{#1\}/\{#2\}} \def\Gcz#1#2{(\{#1\}/\{#2\})} %% {P}/{Q}
303: \def\co{\, ,} \def\pe{\, .} \def\sc{\, ;}
304: \def\mdef#1{\left\{ %% For defining functions piecewise, separ-
305: \matrix{#1}\right.} %% ating the entries by & and \cr as usual
306: \def\binom#1#2{{\pmatrix{ #1 \cr #2 }}}
307: \def\restr#1#2{\left. #1 \vph{\f{}{}} \right|_#2} %% Restriction; value
308: \def\megsz#1#2{\left. #1 \vph{\f{}{}} \right|_#2} %% taken at #2
309: \def\arrow#1{\buildrel #1 \over \longrightarrow}
310: \def\idx#1{{\rm #1}}
311: \def\scal#1#2{{\langle #1, #2 \rangle}}
312: \def\abs#1{{\left|#1\right|}} \def\absz#1{\abs{#1}}
313: \def\norm#1{{\left\|#1\right\|}} \def\norma{\norm}
314: \def\bra#1{\langle#1|} \def\ket#1{|#1\rangle}
315: \def\ordo#1{ {\cal O} \( #1 \) }
316: \def\s#1{\sqrt{#1}}
317: \def\lt{\left} \def\({\lt(} \def\[{\lt[} \def\<{\lt\langle}
318: \def\rt{\right} \def\){\rt)} \def\]{\rt]} \def\>{\rt\rangle}
319: \def\_{^{}_}
320: \def\Dot{^{\Bf{\cdot}}} \def\pont{\point}
321: \def\PL{^{(+)}} \def\PM{^{(\pm)}} \def\ONE{^{(1)}}
322: \def\MI{^{(-)}} \def\MP{^{(\mp)}} \def\TWO{^{(2)}}
323: \def\fel{\raisebox{.5pt}{$\scriptscriptstyle{1\over2}$}}
324: \def\negyed{\raisebox{.5pt}{$\scriptscriptstyle{1\over4}$}}
325: \def\haromnegyed{\raisebox{.5pt}{$\scriptscriptstyle{3\over4}$}}
326: \def\fel{$\textstyle{1\over2}$}
327: \def\negyed{$\textstyle{1\over4}$}
328: \def\haromnegyed{$\textstyle{3\over4}$}
329: \def\p{\partial}
330: \def\impl{\Rightarrow} \def\Impl{\Longrightarrow}
331: \def\intl{\int\limits} \def\suml{\sum\limits}
332: \def\homeo{\cong}
333: \def\Bf#1{\HB{$\SB{$#1$} %% Poor man's boldface in formulas
334: \def\P{\CB\x-\XB\RB{.05ex}}\P\x-\XB\x.07ex\P$}}
335: \def\cA{{\cal A}} \def\cH{{\cal H}} %% The whole \cal ABC
336: \def\cB{{\cal B}} \def\cI{{\cal I}} \def\cO{{\cal O}} \def\cU{{\cal U}}
337: \def\cC{{\cal C}} \def\cJ{{\cal J}} \def\cP{{\cal P}} \def\cV{{\cal V}}
338: \def\cD{{\cal D}} \def\cK{{\cal K}} \def\cQ{{\cal Q}} \def\cW{{\cal W}}
339: \def\cE{{\cal E}} \def\cL{{\cal L}} \def\cR{{\cal R}} \def\cX{{\cal X}}
340: \def\cF{{\cal F}} \def\cM{{\cal M}} \def\cS{{\cal S}} \def\cY{{\cal Y}}
341: \def\cG{{\cal G}} \def\cN{{\cal N}} \def\cT{{\cal T}} \def\cZ{{\cal Z}}
342: \def\eps{\varepsilon} \def\phy{\varphi} \def\thet{\vartheta}
343: \def\chy{\chi^{}}
344: \def\text#1{{ \rm #1 }} \def\ttexttt#1{{ \rm \, #1 \ }}
345: \def\textt#1{{ \rm #1 \, }} \def\tttext#1{{ \rm \ #1 }}
346: \def\texttt#1{{ \rm #1 \ }} \def\tttextt#1{{ \rm \ #1 \, }}
347: \def\ttext#1{{ \rm \, #1 }} \def\tttexttt#1{{ \rm \ #1 \ }}
348: \def\ttextt#1{{ \rm \, #1 \, }} %% Use \tt for Latex's old \texttt
349: \def\t#1{{\rm #1}} %% VIGYAZZ, Latex eqnarray-jaban masra hasznalom \t-t!
350: \def\text#1{\t{#1}} \def\ttext#1{\,\t{#1}} \def\tttext#1{\ \t{#1}}
351: \def\textt#1{\t{#1}\,} \def\ttextt#1{\,\t{#1}\,} \def\tttextt#1{\ \t{#1}\,}
352: \def\texttt#1{\t{#1}\ }\def\ttexttt#1{\,\t{#1}\ }\def\tttexttt#1{\ \t{#1}\ }
353: \def\d{\text{d}}
354: \def\ee#1{{\text{e}^{#1}}}
355: \def\Dom{\texttt{Dom}}
356: \def\mod{\texttt{mod}}
357: \def\arccot{\textt{arccot}}
358: \def\Re{\text{Re}} \def\Im{\text{Im}} \def\cc{\ttext{c.}\ttext{c.}}
359: \def\const{\textt{const.}}
360: \def\R{{\rm I \hskip-0.47ex R}}
361: \def\szor{\biT \cdot \biT}
362: \def\h{\hbar}
363: \def\fok{$\vph{.}^{\circ}$} %% Legyen \deg , es csak utana a magyar
364: \def\aR{\alpha_{\rm R}} \def\aI{\alpha_{\rm I}}
365: \def\bR{\beta_{ \rm R}} \def\bI{ \beta_{\rm I}}
366: \def\etaR{\eta_{\rm R}} \def\etaI{\eta_{\rm I}}
367: \def\alphaR{\aR} \def\alphaI{\aI} \def\betaR{\bR} \def\betaI{\bI}
368:
369:
370: %% Abbreviations for this note
371: \def\qf{y}
372: \def\qg{g}
373: \def\qk{k}
374: \def\ql{l}
375: \def\qm{m}
376: \def\qq{q}
377: \def\qs{\tau}
378: \def\qt{t}
379: \def\qy{y}
380: \def\Qy{y}
381: \def\qA{A}
382: \def\qF{F}
383: \def\qN{N}
384: \def\qsigma{\sigma}
385: \def\qeta{\eta}
386: \def\qEpsilon{\cE}
387: \def\Dir{^-}
388: \def\Neu{^+}
389:
390: \def\figone{qfllow5.eps}
391: \def\figtwo{qflmed7.eps}
392: \def\figthree{qflhigh6.eps}
393:
394: \input epsf
395:
396: %\let\omitpictures=Y
397: \let\omitpictures=N
398:
399: %% References for the paper:
400:
401: {
402:
403: %% 1
404: \refn\CFT
405: {T. Cheon, T. F\"{u}l\"{o}p and I. Tsutsui,
406: {\it Ann.\ Phys.} {\bf 294} (2001) 1}
407:
408: %% 2
409: \refn\Warburton
410: {R.J. Warburton, C. Sch\"aflein, D. Haft, F. Bickel, A. Lorke, K. Karrai, J.M.
411: Garcia, W. Scoenfeld and P.M. Petroff, {\it Nature} {\bf 405} (2000) 926}
412:
413: %% 3
414: \refn\Lorke
415: {A. Lorke, R.J. Luyken, A.O. Govorov, J.P. Kotthaus, J.M. Garcia and P.M.
416: Petroff, \prl{84}{2000}{2223}}
417:
418: %% 4
419: \refn\Nature
420: {A. Fuhrer, S. L{\"u}sher, T. Ihn, T. Heinzel, K. Ensslin, W. Wegscheider
421: and M. Bichler, {\it Nature} {\bf 413} (2001) 822}
422:
423: %% 5
424: \refn\FTC
425: {T. F\"{u}l\"{o}p, I. Tsutsui and T. Cheon,
426: {\sl Spectral Properties on a Circle with a Singularity},
427: {\it J.\ Phys.\ Soc.\ Japan}, to appear; quant-ph/0307002}
428:
429: %% 6
430: \refn\elsewhere
431: {to be reported elsewhere}
432:
433: }
434:
435: %%% frontpage
436:
437: %%% Output of frontpage %%%%%%%%%%%%%%%%%%%%
438:
439: \pageheight{23cm}
440: \pagewidth{15.7cm}
441: \hcorrection{0mm}
442: \magnification= \magstep1
443: \def\bsk{%
444: \baselineskip= 16.8pt plus 1pt minus 1pt}
445: \parskip=5pt plus 1pt minus 1pt
446: \tolerance 6000
447:
448: %\vsize 19.2cm \voffset -1.2cm %% For temporary purposes !!!
449:
450: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
451:
452: %%% %%% %%%
453: % FrontPage
454: %%% %%% %%%
455:
456: \hfill
457: %\phantom{KEK Preprint 2002-107}
458: \vskip -4pt
459: \hfill
460: \phantom{quant-ph/0207xxx}
461:
462: \vskip 42pt
463:
464: %%% Setting of the baselineskip for frontpage
465: {\baselineskip=18pt
466: %%%
467:
468: \centerline{\bigbold
469: Quantum Force due to Distinct Boundary Conditions}
470: %Quantum Force Emerging from Distinct Boundary Conditions}
471:
472: \vskip 30pt
473:
474: \centerline{\smc
475: Tam\'{a}s F\"{u}l\"{o}p\footnote{${}^{*}$}
476: {\eightpoint email:\ fulopt@post.kek.jp},
477: %\quad
478: %{\rm and}
479: \quad
480: Hitoshi Miyazaki
481: \quad
482: {\rm and}
483: \quad
484: Izumi Tsutsui\footnote{${}^\dagger$}
485: {\eightpoint email:\ izumi.tsutsui@kek.jp}
486: }
487:
488:
489:
490: \vskip 15pt
491:
492: {
493: \baselineskip=13pt
494: \centerline{\it
495: Institute of Particle and Nuclear Studies}
496: \centerline{\it
497: High Energy Accelerator Research Organization (KEK)}
498: \centerline{\it Tsukuba 305-0801}
499: \centerline{\it Japan}
500: }
501:
502:
503:
504: \vskip 100pt
505:
506: \abstract{%
507: {\bf Abstract.}\quad
508: We calculate the quantum statistical force acting on a partition wall that divides a one
509: dimensional box into two halves. The two half boxes contain the same (fixed) number of
510: noninteracting bosons, are kept at the same temperature, and admit the same boundary
511: conditions at the outer walls; the only difference is the distinct
512: boundary conditions imposed at the two sides of the partition wall.
513: The net force acting on the partition wall is nonzero at zero temperature and remains almost
514: constant for low temperatures. As the temperature increases, the force starts to decrease
515: considerably, but after reaching a minimum it starts to increase, and tends to infinity
516: with a square-root-of-temperature asympotics. This example demonstrates clearly that distinct
517: boundary conditions cause remarkable physical effects for quantum systems.}
518:
519: \vskip 10pt
520: %{\baselineskip=10pt
521: %{\ninepoint
522: %\indent{PACS codes: 3.65.-w, 2.20.-a, 73.20.Dx\hfill\break}
523: %\indent{Keywords: Singular interaction, Inequivalent Quantizations,
524: % Caustics}
525: %}
526: %}
527: %
528: %\vskip 10pt
529: %
530: %\noindent
531: %Number of manuscript pages: xx, \quad figures: x, \quad tables: x
532: %
533: %\centerline{
534: %(Running title: {\sl M{\" o}bius Structure of the Spectral Space
535: %under Point Interaction})
536: %}
537: %
538: %%% End setting of the baselineskip for frontpage
539: }
540: %%%
541: %\bigskip
542: \ve
543:
544: %%% Output layout of the text %%%%%%%%%%%%%%
545:
546: \pageheight{23cm}
547: \pagewidth{15.7cm}
548: \hcorrection{-1mm}
549: \magnification= \magstep1
550: \def\bsk{%
551: \baselineskip= 15.2pt plus 1pt minus 1pt}
552: \parskip=5pt plus 1pt minus 1pt
553: \tolerance 8000
554: \bsk
555:
556: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
557:
558:
559: \secno=1 \meqno=1
560:
561: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
562: \bigskip
563: \noindent{\bf 1. Introduction}
564: \medskip
565: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
566:
567: Quantum systems in less than three dimensions enjoy an increasing popularity
568: and importance in physics. These seemingly simple systems (rings, boxes, ``dots", etc.)
569: exhibit unexpectedly interesting properties [\CFT], many of which have originally been
570: found in connection with quantum field theories. The recent developments in nanotechnology
571: now make it possible to manufacture such quantum mechanical devices, allowing us to
572: study these various features in laboratories
573: % [\Warburton, \Lorke, \Nature].
574: [\Warburton--\Nature].
575: With tunable control parameters --- such as varying the magnetic flux that is driven through
576: a ring, or influencing the parameters that characterize the boundary conditions at the walls
577: of a box --- these properties become governable, giving us thus controllable quantum devices
578: that may be useful for many technological applications.
579:
580: In fact, the dependence of the physical properties on the control parameters is quite
581: strong. For example, the energy spectrum of a box or a ring with junctions changes
582: considerably with different boundary/fitting conditions applied at the walls/junctions
583: [\FTC]. The aim of the present paper is to demonstrate a physical consequence of this
584: effect in the context of quantum statistical mechanics. Namely, we show by the example of the
585: box systems how different quantum statistical behavior emerges from distinct boundary conditions
586: and the corresponding distinct spectra. One of the remarkable results is that not only the
587: low-temperature behavior is sensitive to the difference in spectrum but also the
588: high-temperature properties depend clearly on it.
589:
590: The setting we consider is a one-dimensional quantum well/box, divided into two halves
591: with an internal partition wall. Both halves contain the same number of
592: noninteracting particles. We impose the same (Dirichlet) boundary conditions at the two ends
593: of the box, while the boundary conditions at the two sides of the partition wall are
594: chosen to be different (Dirichlet from the left, Neumann from the right). {}From the different
595: emerging spectra in the two half regions, we calculate the quantum statistical force/pressure
596: acting on the internal wall from the left and from the right, and the corresponding net force
597: acting on it, as the function of temperature. We present both numerical results (which are
598: obtained by appropriate truncations of the arising infinite sums) and analytical
599: approximations, the latter ones aiming at understanding the low and the high temperature
600: regimes.
601:
602: The number of particles on both sides, \m{N}, is arbitrary, and is not necessarily
603: macroscopically large. Our numerical results are presented for \mm{N = 100}, which is a
604: realistic population number in nanoscale quantum experiments [\Nature]. The particles are
605: considered as bosons in this paper, however, we mention that the results prove to be
606: qualitatively (and partly quantitatively) similar for fermionic particles, too [\elsewhere].
607:
608:
609: \secno=2 \meqno=1
610:
611: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
612: \bigskip
613: \noindent{\bf 2. Low temperature regime}
614: \medskip
615: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
616:
617: The system we consider is formulated as a one-dimensional quantum well of width $2l$ given
618: by the interval $[-l, l]$ with a partition placed at the centre $x = 0$. At the
619: end walls of the well the states are supposed to obey the
620: Dirichlet boundary condition $\psi(\pm l) = 0$. At the centre, we assume that the
621: partition imposes distinct boundary conditions on the left and the right, our choice is
622: to impose the Dirichlet one for $ x = -0$ and the Neumann one for $x = +0$:
623: $$
624: \psi(-0) = 0, \qquad
625: \psi'(+0) = 0.
626: \eqn\no
627: $$
628: The two half wells seperated by the partition then admit
629: the energy levels $E^{\pm}_n = e^{\pm}_n {\cal E}$, $n = 1$, 2, 3 $\ldots$, with
630: $$
631: e^{+}_n = \left(n-{1\over 2}\right)^2, \qquad
632: e^{-}_n = n^2, \qquad
633: {\cal E} = {{\hbar^2}\over{2m}}\left({{\pi}\over{l}}\right)^2.
634: \eqn\unitenergy
635: $$
636:
637: Suppose that we put $N$ identical bosonic particles into each of the two half
638: wells. The particles will then distribute among the eigenstates according to the
639: Bose-Einstein statistics,
640: $$
641: N = \sum_{n} N_n^\pm,
642: \qquad
643: N_n^\pm
644: %= {1\over{e^{\alpha^\pm + \beta E_n^\pm} - 1}}
645: = {1\over{e^{\alpha^\pm + b e^\pm_n} - 1}},
646: \eqn\popl
647: $$
648: where we have
649: introduced
650: $b = \beta {\cal E}$ ($\beta = 1/k T$).
651: Note that
652: $\alpha^\pm$ are determined by
653: the particle number constraint (the first of (\popl)) and
654: are dependent on the temperature. The forces (or pressure) acting on the partition
655: from the right and the left are then given by
656: $$
657: F^\pm = - \sum_n {{\partial E_n^\pm}\over{\partial
658: l}} N_n^\pm.
659: \eqn\express
660: $$
661: For simplicity, in what follows we use the dimensionless force and temperature
662: defined by
663: $$
664: f^\pm = {{l}\over{2\cal E}} F^\pm, \qquad
665: t = {1\over{b}} = {k\over{{\cal E}}}T,
666: \eqn\dlq
667: $$
668: with which the net force on the partition becomes
669: $$
670: \Delta f = f^{-} - f^{+}, \qquad
671: f^\pm = \sum_n
672: %f^\pm_n, \qquad f^\pm_n =
673: e^\pm_n N^\pm_n.
674: \eqn\nfp
675: $$
676: Our objective is to find how the net force $\Delta f$ behaves
677: as a function of the temperature variable $t$.
678:
679:
680: \topinsert
681: \epsfxsize 7.0cm
682: \ifx\omitpictures N \centerline{\epsfbox {\figone}} \fi
683: \yex{1}
684: \abstract{{\bf Figure 1.} The net force $\Delta f(t)$ for \mm{N = 100},
685: in the temperature region \mm{t < 1 \co} obtained by a numerical computation (solid
686: line), and approximated by Eq.~(2.8) (dashed line).}
687: \bigskip
688: \endinsert
689:
690: To proceed, we first study the low temperature regime where a finite number of
691: particles remain in the ground state $n = 1$. This occurs if
692: $\alpha^\pm + b e^\pm_1 \ll 1$, and in this regime
693: a numerical computation exhibits interesting
694: behaviors of the net force $\Delta f$. Namely, as shown in Fig.1, $\Delta f$ starts
695: off by the value $\Delta f = 3N/4$ at $t = 0$ and decreases slightly but
696: basically stays there for
697: $t < 1$. Above $t \approx 1$ the net force starts to decrease almost linearly until
698: it reaches its minimum at around $t \sim N$ (see Fig.2), and from there it increases
699: to infinity. To see how these behaviors arise, let us consider
700: the case of the extremely low temperatures
701: $t < 1$, where most of the $N$ particles are in the ground state.
702: In this case, if we write $f^\pm$ in (\nfp) as
703: $$
704: f^\pm = e_1^\pm N + \sum_{n = 2}^\infty g_n^\pm,
705: \qquad g_n^\pm = \left(e_n^\pm - e_1^\pm\right) N^\pm_n,
706: \eqn\reexpress
707: $$
708: then we find that the first term gives the zero temperature value $f^\pm(0)$
709: and the rests represent the corrections from the higher energy levels at finite
710: temperatures. Since $N^\pm_n$ decreases exponentially fast for higher $n$, we may keep
711: only the first contribution $n = 2$ to get
712: $$
713: \Delta f(t) \approx {3\over 4} N + (3\, e^{-3/t} - 2\, e^{-2/t}).
714: \eqn\nfres
715: $$
716: This gives a good approximation for $t < 1$ as seen in Fig.1.
717:
718: The linear decrease of the net force observed for temperatures higher (but not much
719: higher) than $t = 1$ may be understood heuristcially as follows. First, we classify
720: the energy levels into three categories, where the first is those levels for which
721: $b(e_n^\pm - e_1^\pm) \ll 1$, the second is those for which
722: $b(e_n^\pm - e_1^\pm) \gg 1$, and the third is the rest, {\it i.e.}, those for which
723: $b(e_n^\pm - e_1^\pm)$ is not far from 1. Let
724: $m$ be the level whose $b(e_n^\pm - e_1^\pm)$ is closest to 1,
725: that is,
726: $b(e_m^\pm - e_1^\pm) \approx 1$.
727: Levels with $n \ll m$ thus belong to the first category, and for these one
728: has $\alpha^\pm + b e_n^\pm \approx b(e_n^\pm - e_1^\pm)$ and hence
729: $N_n^\pm \approx {1\over{b(e_n^\pm - e_1^\pm)}}$.
730: The corresponding $g_n^\pm$ given in (\reexpress) are then found to be
731: $g_n^\pm \approx {1\over{b}}$. Thus we have $g_n^- - g_n^+ \approx 0$ showing
732: that the levels in this category do not contribute to the net
733: force.
734: On the other hand, levels with $n \gg m$ belong to the second category, and for
735: these we have
736: $N_n^\pm
737: \approx e^{-b(e_n^\pm - e_1^\pm)}$ but
738: since
739: $b(e_n^\pm - e_1^\pm) \gg 1$, the corresponding force $F_n$ is seen to be
740: exponentially small and can be ignored.
741: Thus the contribution to the net force can comes only from the intermediate
742: levels belonging to the third category. As a representative, let us
743: choose
744: $m$ such that $b(e_m^+ - e_1^+) = 1$ holds in effect,
745: which is equivalent to
746: $m^2 - m = t$, for the right half well.
747: Since we have $N_m^{+} = 1/(e-1)$,
748: we get
749: $$
750: g_m^{+} = {1\over b}\cdot{1\over{e - 1}}.
751: \eqn\qfmr
752: $$
753: In contrast, for the level $m$ for the left half well, we have
754: $b(e_m^- - e_1^-) = {{m^2 - 1}\over t} = 1 + {1\over m}$
755: and hence $N_m^{-} = 1/(e^{1 + 1/m} - 1)$. Thus
756: the contribution to the force becomes
757: $$
758: g_m^{-} = {1\over b}\cdot
759: {{1 + {1\over m}}\over{e^{1 + 1/m} - 1}}.
760: \eqn\qfml
761: $$
762: Combining (\qfmr) and (\qfml),
763: and estimating roughly that
764: the total number of levels in the third
765: category is of the order of $m$, ranging from $m/2 < n < 3m/2$, say, we just
766: multiply the number $m$ to each of the contribution to find
767: $$
768: \Delta f(t) \approx {3\over 4} N + \sum_{n = m/2}^{3m/2} (g_n^- - g_n^+)
769: = {3\over 4} N - {t\over{(e - 1)^2}} .
770: \eqn\nflt
771: $$
772: The linear decrease of the net force is now seen in (\nflt).
773: Our argument assumes
774: $\alpha^\pm + b e_1^\pm < b(e_n^\pm - e_1^\pm)$ which gives the upper limit
775: of the temperature $t$ for which the heuristic formula (\nflt) is available.
776: The numerical result shows that the limit is around $t \approx 2N/3$.
777:
778: \topinsert
779: \epsfxsize 7.0cm
780: \ifx\omitpictures N \centerline{\epsfbox {\figtwo}} \fi
781: \yex{1}
782: \abstract{{\bf Figure 2.} The net force $\Delta f(t)$ for \mm{N = 100} and in the
783: temperature region \mm{ 0 < t < 160}, obtained by a numerical computation (solid line),
784: and approximated by Eq.~(2.11) (dotted line), and by Eq.~(2.16) using Eq.~(2.18)
785: (dashed line).}
786: \bigskip
787: \endinsert
788:
789: If the temperatures are even higher $t \gg 1$ but still satisfy $\alpha^\pm + b
790: e^\pm_1
791: \ll 1$ remaining in the low temperature regime, the above approximation
792: becomes worse and we need to resort to more systematic means to reproduce the
793: numerical results. One such method valid for $t \gg 1$ is to consider the infinite
794: sums as the trapezoidal approximations of corresponding integrals. Thus we can write
795: $$
796: \sum_{n=1}^{\infty} y(s_n) \approx \f{ y(s_1) }{2} +
797: \f{1}{\Delta s} \int_{s_1}^{\infty} y(s) \bit \d s
798: \eqn\apf
799: $$
800: for functions $y(s)$ vanishing at infinity, $\lim_{s \to \infty} y(s) = 0$. The
801: summation in (\apf) is taken at equidistant points,
802: $s_n$, $n = 1, 2, \ldots$, with $\Delta s = s_{n+1} - s_n =$ constant, and obviously
803: the approximation is better for smaller $\Delta s$.
804: Using this approximation, we obtain
805: $$
806: \eqalign{
807: f^\pm
808: &= \f{1}{b} \sum_{n=1}^{\infty}
809: \f{ b e_n }{ \ee{ \alpha^\pm + b e^\pm_n } - 1 }
810: = - \f{N \alpha^\pm}{b} + \f{1}{b} \sum_{n=1}^{\infty}
811: \f{ \alpha^\pm + (s^\pm_n)^2 }{ \ee{ \alpha^\pm + (s^\pm_n)^2 } - 1 } \cr
812: &\approx
813: - \f{N \alpha^\pm}{b} + \f{1}{b} \[ \f{1}{2}
814: \f{ \alpha^\pm + (s^\pm_1)^2 }{ \ee{ \alpha^\pm + (s^\pm_1)^2 } - 1 } +
815: \f{1}{\s{b}} \int_{s^\pm_1}^{\infty}
816: \f{ \alpha^\pm + s^2 }{ \ee{ \alpha^\pm + s^2 } - 1 } \d s \]
817: \co }
818: $$
819: where we have
820: introduced $s^\pm_n = \s{ b e_n^\pm }$ which are equidistant
821: on account of (\unitenergy). Note that the increments
822: \m{ \Delta s^\pm = \s{b} = 1/\s{t} } are indeed small for $t \gg 1$.
823: Since \mm{ \f{z}{\ee{z} - 1} \approx 1 } for small \m{z},
824: we can approximate the second term by
825: \m{ \f{1}{2 b} }, and similarly the integral \mm{ \int_{0}^{s^\pm_1}
826: \f{ \alpha^\pm + s^2 }{ \ee{ \alpha^\pm + s^2 } - 1 } \d s } by
827: \m{s^\pm_1}. We thus obtain
828: $$
829: f^\pm \approx \f{ - N \alpha^\pm + 1/2 - s^\pm_1 / \s{b} }{b} +
830: \f{1}{ b^{3/2} } \int_{0}^{\infty}
831: \f{ \alpha^\pm + s^2 }{ \ee{ \alpha^\pm + s^2 } - 1 } \d s \pe
832: \eqn\no
833: $$
834: Employing the formula
835: $\f{z}{\ee{z} - 1} \approx \ee{-z} \( 1 + \f{z}{2} + \f{z^2}{12} \)$
836: which is handy for evaluating the integral approximately, we find
837: $$
838: \int_{0}^{\infty} \f{ \alpha^\pm + s^2 }{ \ee{ \alpha^\pm + s^2 } - 1 }
839: \d s \approx
840: %\int_{0}^{\infty} \ee{ - ( \alpha + \qy^2 ) }
841: %\[ 1 + \f{ \alpha + \qy^2 }{2} + \f{ ( \alpha + \qy^2 )^2 }{12} \]
842: %\d \qy =
843: \f{ \s{\pi} }{96} (63 - 35 \alpha^\pm) + \ordo{\alpha^\pm{}^2},
844: \eqn\no
845: $$
846: and hence
847: $$
848: f^\pm \approx
849: %\f{ - N \alpha + 1/2 - \s{e_1} }{b} +
850: % \f{ \s{\pi} }{96} (63 - 35 \alpha) \f{1}{ b^{3/2} } =
851: \( - N \alpha^\pm + 1/2 - \s{e^\pm_1} \) \qt +
852: \f{ \s{\pi} }{96} (63 - 35 \alpha^\pm) \bitt \qt^{3/2} \pe
853: \eqn\no
854: $$
855: The net force is then
856: $$
857: \Delta f \approx \( N \qt + \f{35}{96} \s{\pi} \bittt \qt^{3/2} \)
858: \( \alpha\Neu - \alpha\Dir \)
859: + \( \s{ e_1\Neu } - \s{ e_1\Dir } \) \qt \pe
860: \eqn\qnf
861: $$
862: In passing we remark that this formula (\qnf) turns out to be valid even for
863: small temperatures. This can be seen, for example, by considering the \mm{t \to 0}
864: limit, where $\qN_1^\pm \approx N$ and hence from (\popl) we have
865: $\alpha^\pm \approx \ln( 1 + 1/N ) - b e^\pm_1$. Thus the net force
866: (\qnf) is found to be $\Delta f \approx N \( e_1\Dir - e_1\Neu \) = 3N/4$
867: for $t \ll 1$ as seen in (\nfres).
868:
869: To make use of (\qnf), we need to determine the \m{\alpha^\pm} as functions of
870: $t$ from the total number costraint in (\popl). This can be done with the help of
871: the same approximation method as used above. Namely, upon using (\apf)
872: we write down the constraint condition as
873: $$
874: \eqalign{
875: N
876: &= \sum_{n=1}^{\infty} \qN_n^\pm = \qN_1^\pm
877: + \sum_{n=2}^\infty \f{1}{ \ee{ \alpha^\pm + b e_n^\pm } - 1 }\cr
878: &\approx \f{1}{ \ee{ \alpha^\pm + b e_1^\pm } - 1 } +
879: \f{1/2}{ \ee{ \alpha^\pm + b e_2^\pm } - 1 } +
880: \f{1}{\s{b}} \int_{s_2^\pm}^{\infty}
881: \f{ \d s }{ \ee{ \alpha^\pm + s^2 } - 1 },
882: }
883: \eqn\no
884: $$
885: where we have kept \m{\qN_1^\pm} separately to achieve a better approximation (since
886: \m{\qN_n^\pm} are rapidly decreasing functions of \m{n} for small \m{n}).
887: Limiting the range of integration to \mm{ [0,\s{2}] } which provides the main
888: contribution to the integral, and using
889: \mm{ \f{1}{\ee{z} - 1} \approx \f{1}{z} - \f{1}{2}} valid on the range,
890: we find
891: $$
892: \eqalign{
893: N &\approx
894: \f{1}{ \alpha^\pm + b e^\pm_1 } + \f{1/2}{ \alpha^\pm + b e^\pm_2 } -
895: \f{3}{4} -\f{1}{ \s{b} } \f{ \s{2 - \alpha^\pm} - s^\pm_2 }{2} \cr
896: & \quad +\f{1}{ \s{b |\alpha^\pm|} }
897: \[ \qA \( \f{\s{|\alpha^\pm|}}{s^\pm_2} \)
898: - \qA \( \f{\s{|\alpha^\pm|}}{ \s{2 - \alpha^\pm} } \) \] \co }
899: \eqn\aat
900: $$
901: where the function \m{\qA} is the \mm{\arctan} function for positive
902: \m{\alpha^\pm}, and is the \mm{\hbox{arctanh}} function for negative
903: \m{\alpha^\pm}. Unfortunately, it is not easy to solve (\aat) directly for
904: \m{\alpha^\pm} even approximately to obtain an analytic formula
905: that can be used in (\qnf) to reproduce the numerical result.
906: Nevertheless, the formula (\aat) has allowed us to evaluate
907: the infinite sum to a good accuracy, and we can solve (\aat) indirectly for
908: \m{\alpha^\pm} by numerical means. The outcome of this semi-analytic analysis
909: is in good agreement with the numerical computation as shown in Fig.2.
910:
911:
912: \secno=3 \meqno=1
913:
914: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
915: \bigskip
916: \noindent{\bf 3. High temperature regime}
917: \medskip
918: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
919:
920:
921: \topinsert
922: \epsfxsize 7.0cm
923: \ifx\omitpictures N \centerline{\epsfbox {\figthree}} \fi
924: \yex{1}
925: \abstract{{\bf Figure 3.} The net force $\Delta f(t)$ for \mm{N = 100},
926: obtained by a numerical computation (solid line), and approximated for
927: high temperatures with Eq.~(3.7) (dashed line).}
928: \bigskip
929: \endinsert
930:
931: Let us now determine the high-temperature asymptotic behavior of the net force.
932: For increasing temperature, we expect \mm{N^\pm_1} to decrease so [see
933: (\popl) for \m{n = 1}] we expect \m{\alpha^\pm} to increase to higher
934: positive values. Inspired by this, let us
935: expand \m{N_n^\pm} in \mm{ \qq^\pm := \ee{- \alpha^\pm} } as
936: \E{\aad}{
937: N_n^\pm = \f{ \qq^\pm \ee{- b e_n^\pm} }{ 1 - \qq^\pm \ee{- b e_n^\pm} }
938: = \suml_{\qk = 1}^{\infty} (\qq^\pm)^\qk \ee{- \qk b e^\pm_n} \co
939: }
940: which is valid for any positive \m{\alpha^\pm}. Thus
941: \E{\aae}{
942: N = \suml_{n = 1}^{\infty} N_n^\pm = \suml_{\qk = 1}^{\infty}
943: (\qq^\pm)^\qk \suml_{n = 1}^{\infty} \ee{- \qk b e_n^\pm}
944: = \suml_{\qk = 1}^{\infty} (\qq^\pm)^\qk \[ - \f{\qsigma^\pm}{2}
945: + \f{1}{2} \suml_{n = - \infty}^{\infty} \ee{- \qk b e^\pm_n} \] ,
946: }
947: with the constants \mm{\qsigma\Neu = 0 \co} \mm{\qsigma\Dir = 1} corresponding to the
948: \m{\pm} half wells, where we have extended the meaning of the notation \m{e_n^\pm} [see
949: (\unitenergy)] to negative \m{n}s, too. Applying now the Poisson summation formula
950: \E{\aal}{
951: \suml_{n = - \infty}^{\infty} \qf(n) = \suml_{\qm = - \infty}^{\infty}
952: \intl_{-\infty}^{\infty} \d s \bitt \qf(s) \bitt \ee{ 2 \pi i \qm s } \co
953: }
954: we obtain
955: \E{\aag}{
956: N = \suml_{\qk = 1}^{\infty} (\qq^\pm)^\qk \[
957: - \f{\qsigma^\pm}{2} + \s{ \f{\pi}{ 4 \qk b } } \suml_{\qm =
958: - \infty}^{\infty} (\qs^\pm)^m \ee{ - \f{\pi^2}{ \qk b } \qm^2 } \]
959: }
960: with \mm{\qs^\pm = \mp 1 \pe}
961: Similarly, for the force \m{f^\pm}, one can find
962: \E{\aah}{
963: f^\pm = \suml_{n = 1}^{\infty} e_n^\pm N_n^\pm =
964: \suml_{\qk = 1}^{\infty} (\qq^\pm)^\qk
965: \s{ \f{\pi}{ 16 \qk^3 b^3 } } \suml_{\qm = - \infty}^{\infty}
966: (\qs^\pm)^m \( 1 - \ff{2 \pi^2}{\qk b} \qm^2 \)
967: \ee{ - \f{\pi^2}{ \qk b } \qm^2 } \pe
968: }
969:
970: For the high-temperature asymptotic behavior (\mm{\qq^\pm \to 0}), it
971: suffices to consider only the first some terms in the sums over
972: \mm{\qk} [both in (\aag) and (\aah)], and within each term to keep
973: only the \mm{\qm = 0} term in the sums over \m{\qm} (the \mm{\qm \ne 0}
974: terms being exponentially suppressed). Now, the leading, \mm{\qk = 1} term in (\aag)
975: gives that \mm{ \qq^\pm = 2 N \( \f{b}{\pi} \)^{1/2} + \ordo{b} \pe }
976: Since this leading behavior of
977: \m{\qq^\pm} is independent of \m{\qsigma^\pm}, inserting it into (\aah)
978: gives that the leading, \m{ \ordo{ b^{-1} } } term of \m{f^\pm} (coming from
979: \mm{\qk = 1}, \mm{\qm = 0}) is also \m{\qsigma}-independent. Hence,
980: this term will drop out from the net force. Therefore, to
981: have the first nonvanishing term in the net force we need
982: the first subleading term in \m{\qq}, too. Incorporating the \mm{\qk = 2} term
983: as well for \m{\qq^\pm}, we find
984: $$
985: \qq^\pm = 2 N \( \f{b}{\pi} \)^{1/2} + 2 N \[ \qsigma^\pm - \s{2} N \]
986: \f{b}{\pi} + \ordo{ b^{3/2} } \pe
987: \eqn\aai
988: $$
989: Plugging this into (\aah) and then calculating the net force yields
990: \E{\aaj}{
991: \Delta f = \f{N}{2} \( \f{\qt}{\pi} \)^{1/2} + \ordo{ \qt^{0} } \pe
992: }
993: We can see in Fig.3 how the net force actually reaches this square-root asymptotic behavior
994: at high temperatures.
995:
996: \secno=4 \meqno=1
997:
998: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
999: \bigskip
1000: \noindent{\bf 4. Discussion}
1001: \medskip
1002: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1003:
1004: We have found that the net force acting on the separating wall is nonzero at
1005: low temperatures, being practically constant for very small temperatures and starting to
1006: decrease when temperature is increased. Knowing that the energy spectrum is different on the
1007: two halves this property is not very surprising. What is surprising, however, is that this
1008: decrease stops at a certain temperature and the net force starts to increase above this
1009: value. Furthermore, a remarkable fact is that this increase does not stop
1010: nor converges to some finite high-temperature limit but increases to infinity, as the square root
1011: of the temperature. {}From the naive expectation that such quantum effects coming from the
1012: different boundary conditions should vanish at high temperatures where the classical picture
1013: would be available, this result seems quite unusual. However, this may be understood by the
1014: fact that, contrary to most quantum systems, one dimensional boxes have such energy spectra that
1015: the level spacing is not decreasing but increasing for higher energy levels (which is actually
1016: valid not only for boxes with Dirichlet and/or Neumann boundary conditions but for all other boxes
1017: as well [\FTC]). In other words, quantum boxes can be
1018: distinguished by their high-temperature behavior, too.
1019:
1020: We mention that the calculation presented here could be repeated for boxes with other
1021: boundary conditions, too. We note however that, for most box systems, the energy values
1022: are determined by a transcendental equation [\FTC], and hence a certain additional difficulty
1023: for carrying out the calculations, especially the analytical ones, will arise there.
1024:
1025:
1026:
1027: \bigskip
1028: \noindent
1029: {\bf Acknowledgement:}
1030: This work has been supported in part by
1031: the Grant-in-Aid for Scientific
1032: Research on Priority Areas (No.~13135206) by
1033: the Japanese
1034: Ministry of
1035: Education, Science, Sports and Culture.
1036:
1037:
1038: \baselineskip= 15.5pt plus 1pt minus 1pt
1039: \parskip=5pt plus 1pt minus 1pt
1040: \tolerance 8000
1041: \listrefs
1042:
1043: \bye