q-bio0601034/nma2.tex
1: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
2: \documentclass[showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
3: %\documentclass[11pt,aps,preprint]{revtex4}
4: 
5: \usepackage{graphicx}
6: 
7: \date{\today}
8: 
9: \begin{document}
10: 
11: \title{Time-dependent perturbation theory 
12: for vibrational energy relaxation and dephasing in peptides and proteins}
13: 
14: \author{Hiroshi Fujisaki}\email{fujisaki@bu.edu} 
15: \author{Yong Zhang}\email{zhangy@bu.edu} 
16: \author{John E. Straub}\email{straub@bu.edu} 
17: \affiliation{
18: Department of chemistry,
19: Boston University, 590 Commonwealth Ave.,
20: Boston, 02215, Massachusetts, USA
21: }
22: 
23: \begin{abstract}
24: 
25: Without invoking the Markov approximation, we derive formulas 
26: for vibrational energy relaxation (VER) and dephasing for an anharmonic 
27: system oscillator using a time-dependent perturbation theory. 
28: The system-bath Hamiltonian 
29: contains more than the third order coupling terms
30: since we take a normal mode picture as a zeroth order approximation. 
31: When we invoke the Markov approximation, 
32: our theory reduces to the Maradudin-Fein formula which is used to 
33: describe VER properties of glass and proteins. 
34: When the system anharmonicity and the 
35: renormalization effect due to the environment vanishes, 
36: our formulas reduce to 
37: those derived by Mikami and Okazaki invoking the 
38: path-integral influence functional method [J.~Chem.~Phys.~{\bf 121}, 10052 (2004)].
39: We apply our formulas to VER of the amide I mode of a small amino-acide like molecule, 
40: $N$-methylacetamide, in heavy water.
41: 
42: 
43: 
44: \end{abstract}
45: 
46: \pacs{33.80.Be,05.45.Mt,03.65.Ud,03.67.-a}
47: % PACS, the Physics and Astronomy Classification Scheme.
48: % 33.80.Be  Level crossing in molecules 
49: % 05.45.Mt  Semiclassical chaos (Quantum chaos)
50: % 03.65.Ud  Entanglement and quantum nonlocality (e.g. EPR paradox,
51: %           Bell's inequalities, GHZ states, etc.)
52: % 03.67.-a  Quantum information
53: 
54: \maketitle
55: 
56: \section{Introduction}
57: 
58: %\section{Introduction}
59: 
60: %We can do numerical simulations of protein dynamics 
61: %with sophisticated algorithms and powerful computers!
62: %But how to analyze the huge numerical results of them?
63: 
64: 
65: %\subsection{Why new formulas?}
66: 
67: 
68: Vibrational energy relaxation (VER) and dephasing are fundamental 
69: properties of molecular dynamics, energy transfer, and reactivity.
70: Many experimental and theoretical studies have explored these fundamental
71: processes in gas phase, the liquid state, and in glasses and biomolecular systems \cite{FS05}.
72: Though our methodology can be applied to any molecular system, 
73: we are primarily interested in addressing VER and dephasing in peptides or proteins.
74: While recent advanced experimental techniques using absorption spectra or
75: time-resolved spectra can deduce  
76: the structure and dynamics of such a peptide or protein system, 
77: theoretical approaches are needed to clarify 
78: the mechanisms of VER and dephasing underlying the experimental data.
79: 
80: The most standard approach to this problem is through the 
81: perturbation theory of quantum mechanics 
82: as initiated by Oxtoby \cite{Oxtoby}.
83: Recently Hynes's group \cite{Hynes} and Skinner's group \cite{Skinner} 
84: thoroughly studied the VER and dephasing properties of water 
85: (their target mode was the OH bond of HOD in heavy water)
86: using this strategy.
87: This approach is applicable to 
88: peptides or proteins as was first illustrated by Straub and coworkers \cite{Straub}.
89: Derived from this strategy is the use of the 
90: Maradudin-Fein formula (or its equivalent),
91: which was pursued by Leitner \cite{Leitner05} 
92: and Straub and coworkers \cite{FBS05a}.
93: This formula requires the normal modes of the system and 
94: the cubic anharmonic coefficients between the normal modes.
95: This methodology can provide a reasonable account of VER
96: properties of peptides or proteins, 
97: but there are several deficiencies:
98: the most serious one is that it assumes 
99: the Markov properties of the system, so 
100: it cannot describe the short time dynamics \cite{FBS05b}.
101: Another problem is the determination of 
102: the ``lifetime'' width parameter \cite{FS05,FBS05a,FBS05b}.
103: We also want to describe the dephasing properties 
104: of the system, crucial to the interpretation of the 
105: experimental results; it cannot be directly 
106: described by the MF formula (but see \cite{Leitner02}). 
107: 
108: 
109: To meet these goals, we derive the formulas for VER 
110: and dephasing without assuming the Markov properties, i.e., 
111: without taking an infinite-time limit.
112: As a result, we can avoid the annoying ``width parameter'' problem
113: inherent to the MF approach. 
114: In this sense, Mikami and Okazaki \cite{MO04} took 
115: a similar path using the path-integral 
116: influence functional theory.
117: %, and derived formulas for VER and 
118: %dephasing without assuming the Markov properties. 
119: We use a simple 
120: time-dependent perturbation theory of quantum mechanics, 
121: and derive the VER and dephasing formulas more easily. 
122: We find there is a difference between our formulas and theirs 
123: in terms of renormalization of the system Hamiltonian.
124: Another difference is that our system oscillator is taken to be 
125: a cubic anharmonic oscillator, whereas their mode is a harmonic
126: oscillator. This can affect the result when the formulas are applied 
127: to real systems with strong anharmonicity.
128: 
129: This paper is organized as follows:
130: In Sec.~\ref{sec:derivation},
131: we derive the VER and dephasing formulas 
132: for an anharmonic oscillator (mode)
133: without assuming the Markov properties. 
134: In Sec.~\ref{sec:NMA}, we apply our formulas 
135: to the amide I mode of $N$-methylacetamide in
136: heavy water, and discuss the numerical results and the 
137: limitations of our strategy.
138: In Sec.~\ref{sec:summary}, we summarize the paper.
139: Several system parameters and coefficients in our formulas 
140: are defined in the Appendix.
141: 
142: 
143: 
144: 
145: 
146: \section{Derivation of the formulas for VER and dephasing}
147: \label{sec:derivation}
148: 
149: \subsection{System, Bath, and Coupling}
150: 
151: We take our Hamiltonian of a solvated 
152: peptide or protein to be
153: \begin{eqnarray}
154: {\cal H} 
155: &=& {\cal H}_0 + {\cal V} 
156: = {\cal H}_S +{\cal H}_B+ {\cal V} 
157: = {\cal H}_S^{(0)} +{\cal H}_f +{\cal H}_B+ {\cal V},
158: %\nonumber
159: %\\
160: %&=& 
161: \\
162: {\cal V}
163: &=&
164: -q_S ({\cal F} -\langle {\cal F} \rangle)+ q_S^2 
165: ({\cal G}-\langle {\cal G} \rangle)
166: =
167: -q_S \delta{\cal F}+ q_S^2 \delta{\cal G},
168: \\
169: {\cal H}_S^{(0)}
170: &=& 
171: \frac{p_S^2}{2} + \frac{\omega_S^2}{2} q_S^2 
172: -q_S \langle {\cal F} \rangle + q_S^2 \langle {\cal G} \rangle
173: =
174: \frac{p_S^2}{2} + \frac{\bar{\omega}_S^2}{2} \bar{q}_S^2 
175: -\frac{\langle {\cal F} \rangle^2}{2 \bar{\omega}_S^2},
176: \\
177: {\cal H}_f
178: &=& \frac{f}{6} \bar{q}_S^3,
179: \\
180: {\cal H}_B
181: &=&
182: \sum_{\alpha} 
183: \left(
184: \frac{p_{\alpha}^2}{2}+\frac{\omega_{\alpha}^2}{2} q_{\alpha}^2
185: \right),
186: \end{eqnarray}
187: where
188: \begin{eqnarray}
189: \bar{\omega}_S
190: &=& \omega_S \sqrt{1+\frac{2 \langle {\cal G} \rangle}{\omega_S^2}}, 
191: \label{eq:omegas}
192: \\
193: \bar{q}_S
194: &=&
195: q_S-\frac{\langle {\cal F} \rangle}{\bar{\omega}_S^2}=q_S-b.
196: \label{eq:qs}
197: \end{eqnarray}
198: ${\cal H}_S={\cal H}_S^{(0)}+{\cal H}_f$ 
199: is the renormalized system Hamiltonian
200: representing a vibrational mode $q_S$ with cubic anharmonicity $f$,
201: ${\cal H}_B$ the bath Hamiltonian representing 
202: solvent or environmental degrees of freedom with harmonic 
203: frequencies $\omega_{\alpha}$,
204: and ${\cal V}$ the interaction Hamiltonian describing the 
205: coupling between the system and the bath. 
206: We have assumed that the interation can be 
207: Taylor expanded, and we have only included up to the 
208: second order in $q_S$.
209: Note that we need to renormalize the system 
210: to assure that $\langle {\cal V} \rangle =0$ where 
211: the bracket denotes the bath average throughout this paper.
212: (For the definition of $\delta {\cal F}$ and $\delta {\cal G}$, 
213: see Appendix \ref{app:system}.)
214: This is automatically satisfied in the case of 
215: bilinear coupling like the Caldeira-Leggett-Zwanzig model \cite{Weiss},
216: but this is not usually the case. 
217: The system variable becomes ${\bar q}_S$ instead of $q_S$, 
218: and the system frequency does ${\bar \omega}_S$ instead of $\omega_S$.
219: This is similar to previous treatments of the system-bath interaction
220: in the literature \cite{Skinner,XYK02}.
221: 
222: 
223: 
224: 
225: 
226: \subsection{Perturbation theory for VER and dephasing}
227: 
228: %and
229: Starting from the interaction picture of the von Neumann equation, 
230: we can expand the density operator for the full system as 
231: %If we {\it assume} that ${\cal V}$ is small in some sense, 
232: %we can carry out the perturbation expansion for ${\cal V}$ as
233: \begin{eqnarray}
234: \tilde{\rho}(t)
235: &=& \rho(0)+ \frac{1}{i \hbar} \int_0^t dt' 
236: [\tilde{\cal V}(t'), \tilde{\rho}(t')]
237: \nonumber
238: \\
239: &=&
240: \rho(0)+ \frac{1}{i \hbar} \int_0^t dt' 
241: [\tilde{\cal V}(t'), \rho(0)]
242: \nonumber
243: \\
244: &&
245: +\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
246: [\tilde{\cal V}(t'), [\tilde{\cal V}(t''),\rho(0)]]
247: + \cdots
248: \end{eqnarray}
249: where 
250: \begin{equation}
251: \tilde{\rho}(t) \equiv 
252: e^{i {\cal H}_0 t/\hbar} \rho(t) e^{-i {\cal H}_0 t/\hbar},
253: \hspace{1cm}
254: \tilde{\cal V}(t) \equiv 
255: e^{i {\cal H}_0 t/\hbar} {\cal V} e^{-i {\cal H}_0 t/\hbar}.
256: \end{equation}
257: 
258: The {\it reduced} density matrix for the system oscillator 
259: is introduced as 
260: \begin{eqnarray}
261: (\rho_S)_{mn}(t) 
262: &\equiv& {\rm Tr} \{ P_{mn}  \rho(t) \}
263: = {\rm Tr} \{ P_{mn}
264: e^{-i {\cal H}_0 t/\hbar} \tilde{\rho}(t) e^{i {\cal H}_0 t/\hbar}  \},
265: \\
266: P_{mn} &\equiv& |n \rangle \langle m| \otimes 1_B,
267: \\
268: \rho(0) &=& 
269: \rho_S \otimes \rho_B
270: =
271: \sum_{k,l} (\rho_S)_{kl} |k \rangle \langle l| 
272: \otimes e^{-\beta {\cal H}_B}/Z_B,
273: \\
274: Z_B &=& {\rm Tr}_B \{ e^{-\beta {\cal H}_B} \},
275: \end{eqnarray}
276: where the initial state is assumed to be a direct product 
277: state of $\rho_S$ and 
278: $\rho_B = e^{-\beta {\cal H}_B}/Z_B$, i.e.,
279: we have assumed that the bath is in thermal equilibrium.
280: Here $|k \rangle$ 
281: is the vibrational eigenstate for the system Hamiltonian ${\cal H}_S$,
282: i.e., ${\cal H}_S | k \rangle =E_k | k \rangle$.
283: If we assume that ${\cal H}_f$ is small, we can calculate $|k \rangle$
284: and $E_k$ using the time-independent perturbation theory as shown 
285: in Appendix \ref{app:system}.
286: 
287: %In the above case, ${\cal H}_S$ is just a harmonic oscillator,
288: %so we can easily calculate $E_k$ and $|k \rangle$.
289: 
290: 
291: We note that
292: \begin{equation}
293: (\rho_S)_{mn}(t)={\rm Tr} \{ P_{mn}
294: e^{-i {\cal H}_0 t/\hbar} \tilde{\rho}(t) e^{i {\cal H}_0 t/\hbar}  \}
295: =
296: {\rm Tr}_B \{ \tilde{\rho}_{mn}(t)  \} e^{-i \omega_{mn}t}.
297: \end{equation}
298: %as $\rho_v$ commutes with ${\cal H}_0$.
299: %If we assume that $v \neq v_0$, then $\rho_v \rho(0)=0$.
300: %Using this fact,
301: %we obtain the lowest (second) order result  
302: The lowest (second) order result for the density matrix is 
303: \begin{eqnarray}
304: (\rho_S)_{mn}(t)
305: &\simeq&
306: (\rho_S)^{(0)}_{mn}(t)
307: +(\rho_S)^{(1)}_{mn}(t)
308: +(\rho_S)^{(2)}_{mn}(t)
309: +\cdots,
310: \label{eq:density}
311: \\
312: (\rho_S)^{(0)}_{mn}(t)
313: &=&
314: {\rm Tr}_B \{ \tilde{\rho}_{mn}(0)  \} e^{-i \omega_{mn}t}
315: =(\rho_S)_{mn}e^{-i \omega_{mn}t},
316: \\
317: (\rho_S)^{(1)}_{mn}(t)
318: &=&
319: \frac{1}{i \hbar}
320: \int^{t}_0 dt' 
321: {\rm Tr}_B \{ \langle m| [\tilde{\cal V}(t'),\rho(0)] | n \rangle 
322: \} 
323: e^{-i \omega_{mn}t}
324: \nonumber
325: \\
326: &=&
327: \frac{1}{i \hbar}
328: \int^{t}_0 dt'
329: \sum_k
330: \left \{ 
331: \langle \tilde{\cal V}_{mk}(t') \rangle e^{i \omega_{mk}t'} (\rho_S)_{kn}
332: -
333: \langle \tilde{\cal V}_{kn}(t') \rangle e^{i \omega_{kn}t'} (\rho_S)_{mk}
334: \right \}
335: e^{-i \omega_{mn}t},
336: %\nonumber
337: %\\
338: %&=& 0,
339: \\
340: (\rho_S)^{(2)}_{mn}(t)
341: &=&
342: \frac{1}{(i \hbar)^2}
343: \int^{t}_0 dt' 
344: \int^{t'}_0 dt'' 
345: {\rm Tr}_B \{ \langle m| [\tilde{\cal V}(t'),[\tilde{\cal V}(t''),\rho(0)]] | n \rangle 
346: \} 
347: e^{-i \omega_{mn}t}
348: \nonumber
349: \\
350: &=&
351: \frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
352: \sum_{k,l}
353: \left \{
354: \langle \tilde{\cal V}_{mk}(t') \tilde{\cal V}_{kl}(t'') \rangle 
355: (\rho_S)_{ln} e^{i (\omega_{mk}t'+\omega_{kl}t'')} 
356: \right \}
357: e^{-i \omega_{mn}t}
358: \nonumber
359: \\
360: &+&
361: \frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
362: \sum_{k,l}
363: \left \{
364: \langle \tilde{\cal V}_{kl}(t'') \tilde{\cal V}_{ln}(t') \rangle 
365: (\rho_S)_{mk} e^{i (\omega_{kl}t''+\omega_{ln}t')} 
366: \right \}
367: e^{-i \omega_{mn}t}
368: \nonumber
369: \\
370: &-&
371: \frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
372: \sum_{k,l}
373: \left \{
374: \langle \tilde{\cal V}_{ln}(t'') \tilde{\cal V}_{mk}(t') \rangle 
375: (\rho_S)_{kl} e^{i (\omega_{mk}t'+\omega_{ln}t'')} 
376: \right \}
377: e^{-i \omega_{mn}t}
378: \nonumber
379: \\
380: &-&
381: \frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
382: \sum_{k,l}
383: \left \{
384: \langle \tilde{\cal V}_{ln}(t') \tilde{\cal V}_{mk}(t'') \rangle 
385: (\rho_S)_{kl} e^{i (\omega_{mk}t''+\omega_{ln}t')} 
386: \right \}
387: e^{-i \omega_{mn}t}
388: \end{eqnarray}
389: where 
390: \begin{eqnarray}
391: \langle \tilde{\cal V}_{kl} (t) \tilde{\cal V}_{mn}(t') \rangle
392: &\equiv&
393: {\rm Tr}_B \{ \rho_B
394: \tilde{\cal V}_{kl}(t) \tilde{\cal V}_{mn}(t') \},
395: \\
396: \tilde{\cal V}_{kl}(t)
397: &=& \langle k | \tilde{\cal V}(t)| l \rangle
398: = \langle k | e^{i {\cal H}_B t/\hbar} {\cal V} e^{-i {\cal H}_B t/\hbar} | l \rangle,
399: \\
400: \omega_{kl} &=& (E_{k}-E_{l})/\hbar.
401: \end{eqnarray}
402: Note that, in the above formulas, the time dependence 
403: is only induced by the bath Hamiltonian ${\cal H}_B$.
404: 
405: 
406: For the matrix elements of the interaction Hamiltonian ${\cal V}$,
407: we have 
408: \begin{eqnarray}
409: \langle \tilde{\cal V}_{kl}(t) \rangle
410: &=&
411: -(q_S)_{kl} \langle \delta{\cal F}(t) \rangle
412: +(q_S^2)_{kl} \langle \delta{\cal G}(t) \rangle,
413: %=0
414: \\
415: \langle \tilde{\cal V}_{kl}(t) \tilde{\cal V}_{mn}(t') 
416: \rangle
417: %&=&
418: %(q_S)_{ab} (q_S)_{mn} 
419: %\langle \delta{\cal F}(t) \delta{\cal F}(t') \rangle
420: %+(q_S^2)_{ab} (q_S^2)_{mn} 
421: %\langle \delta{\cal G}(t) \delta{\cal G}(t') \rangle
422: %\nonumber
423: %\\
424: %&&
425: %-(q_S)_{ab} (q_S^2)_{mn} 
426: %\langle \delta{\cal F}(t) \delta{\cal G}(t') \rangle
427: %-(q_S^2)_{ab} (q_S)_{mn} 
428: %\langle \delta{\cal G}(t) \delta{\cal F}(t') \rangle
429: %\nonumber
430: %\\
431: &=&
432: (q_S)_{kl} (q_S)_{mn} 
433: \langle \delta{\cal F}(t) \delta{\cal F}(t') \rangle
434: +(q_S^2)_{kl} (q_S^2)_{mn} 
435: \langle \delta{\cal G}(t) \delta{\cal G}(t') \rangle
436: \nonumber
437: \\
438: &&
439: -[(q_S)_{kl} (q_S^2)_{mn}+(q_S^2)_{kl} (q_S)_{mn}] 
440: \langle \delta{\cal F}(t) \delta{\cal G}(t') \rangle
441: \label{eq:matrix}
442: \end{eqnarray}
443: where the value of $(q_S)_{kl}$ and $(q_S^2)_{kl}$ are 
444: given in Eqs.~(\ref{eq:qs1})-(\ref{eq:qs2}) for the case 
445: of a cubic oscillator.
446: Since $\langle \delta {\cal F} \rangle=0$ and 
447: $\langle \delta {\cal G} \rangle=0$,
448: we have $\langle \tilde{\cal V}_{kl}(t) \rangle=0$ and 
449: $(\rho_S)^{(1)}_{mn}(t)=0$.
450: 
451: 
452: 
453: \subsection{VER formula}
454: 
455: We first calculate the diagonal elements of the density matrix 
456: $(\rho_S)_{ii}(t)$ ($i=0,1$) by assuming that the initial state is 
457: the first {\it vibrationally} excited state: 
458: $\rho_S= |1 \rangle \langle 1|$. This is a typical situation 
459: for VER though VER from highly excited states can be 
460: considered \cite{VFVAMJ04}.
461: The density matrix $(\rho_S)_{00}(t)$
462: is written as 
463: \begin{eqnarray}
464: (\rho_S)_{00}(t)
465: &\simeq& 
466: \frac{2}{\hbar^2}
467: \int_0^t dt' \int_0^{t'} dt''  
468: {\rm Re} \left \{
469: \langle \tilde{\cal V}_{10}(t') \tilde{\cal V}_{01}(t'') \rangle 
470: e^{i \tilde{\omega}_S (t'-t'')} 
471: \right \}
472: \end{eqnarray}
473: where $\tilde{\omega}_S$ is the anharmonicity-corrected 
474: system frequency given by Eq.~(\ref{eq:freq}).
475: 
476: 
477: From Eq.~(\ref{eq:matrix}), we have
478: \begin{eqnarray}
479: (\rho_S)_{00}(t)
480: &\simeq& 
481: \frac{2}{\hbar^2}
482: (q_S)^2_{10}
483: \int_0^t dt' \int_0^{t'} dt''  
484: {\rm Re} \left \{
485: \langle \delta {\cal F}(t'-t'') \delta {\cal F}(0) \rangle 
486: e^{i \tilde{\omega}_S (t'-t'')} 
487: \right \}
488: \nonumber
489: \\
490: &&
491: +
492: \frac{2}{\hbar^2}
493: (q^2_S)^2_{10}
494: \int_0^t dt' \int_0^{t'} dt''  
495: {\rm Re} \left \{
496: \langle \delta {\cal G}(t'-t'') \delta {\cal G}(0) \rangle 
497: e^{i \tilde{\omega}_S (t'-t'')} 
498: \right \}
499: \nonumber
500: \\
501: &&
502: -\frac{4}{\hbar^2}
503: (q_S)_{10}(q^2_S)_{10}
504: \int_0^t dt' \int_0^{t'} dt''  
505: {\rm Re} \left \{
506: \langle \delta {\cal F}(t'-t'') \delta {\cal G}(0) \rangle 
507: e^{i \tilde{\omega}_S (t'-t'')} 
508: \right \}.
509: \label{eq:VER0}
510: \end{eqnarray}
511: Using the explicit expressions for the correlation functions \cite{FBS05a},
512: %(see appendix \ref{app:coef}),
513: the final VER formula is obtained as
514: \begin{eqnarray}
515: %e^{i \tilde{\omega}_S (t'-t'')}
516: %\\
517: (\rho_S)_{00}(t)
518: &\simeq&
519: %\left[
520: \frac{2}{\hbar^2}
521: \sum_{\alpha,\beta} 
522: %(
523: %C_{FF} A^{(--)FF}_{\alpha \beta}  
524: %+C_{FG} A^{(--)FG}_{\alpha \beta}  
525: %+C_{GG} A^{(--)GG}_{\alpha \beta}  
526: %)
527: \left[
528: C_{--}^{\alpha \beta}  
529: %\nonumber
530: %\\
531: %&&
532: %\times
533: u_t(\tilde{\omega}_S-\omega_{\alpha}-\omega_{\beta})
534: %\right]
535: \nonumber
536: %\\
537: %&&
538: %+\frac{2}{\hbar^2}
539: %\left[
540: %\sum_{\alpha,\beta} 
541: %(
542: %C_{FF} A^{(++)FF}_{\alpha \beta}  
543: %+C_{FG} A^{(++)FG}_{\alpha \beta}  
544: %+C_{GG} A^{(++)GG}_{\alpha \beta}  
545: %)
546: +C_{++}^{\alpha \beta}  
547: %\nonumber
548: %\\
549: %&&
550: %\times
551: u_t(\tilde{\omega}_S+\omega_{\alpha}+\omega_{\beta})
552: %\nonumber
553: %\\
554: %&&
555: %+\frac{2}{\hbar^2}
556: %\left[
557: %\sum_{\alpha,\beta} 
558: %(
559: %C_{FF} A^{(+-)FF}_{\alpha \beta}  
560: %+C_{FG} A^{(+-)FG}_{\alpha \beta}  
561: %+C_{GG} A^{(+-)GG}_{\alpha \beta}  
562: %)
563: +C_{+-}^{\alpha \beta}  
564: %\nonumber
565: %\\
566: %&&
567: %\times
568: u_t(\tilde{\omega}_S-\omega_{\alpha}+\omega_{\beta})
569: \right]
570: \nonumber
571: \\
572: &&
573: +\frac{2}{\hbar^2}
574: \sum_{\alpha} 
575: \left[ 
576: C^{\alpha}_{-} u_t(\tilde{\omega}_S-\omega_{\alpha})
577: +
578: C^{\alpha}_{+} u_t(\tilde{\omega}_S+\omega_{\alpha})
579: \right]
580: \label{eq:VER}
581: \end{eqnarray}
582: where $u_t(\Omega)$ is defined as 
583: \begin{eqnarray}
584: u_t(\Omega)
585: &=&
586: \int_0^t dt' \int_0^{t'} dt''  
587: \cos \Omega (t'-t'') 
588: = \frac{1-\cos \Omega t}{\Omega^2},
589: \\
590: v_t(\Omega)
591: &=&
592: \int_0^t dt' \int_0^{t'} dt''  
593: \sin \Omega (t'-t'') 
594: = 
595: \frac{\Omega t-\sin \Omega t}{\Omega^2},
596: \end{eqnarray}
597: and $v_t(\Omega)$ is defined for later use.
598: The coefficients are defined in Appendix \ref{app:coef}.
599: Equation (\ref{eq:VER}) is our final formula for VER.
600: 
601: If we take the long time limit of this formula (which is equivalent to 
602: the Markov approximation), we obtain a formula for the VER rate
603: \begin{eqnarray}
604: k_{0 \leftarrow 1} 
605: &\equiv& 
606: \left. \frac{d}{dt}(\rho_S)_{00}(t) \right|_{t \rightarrow \infty}
607: \nonumber
608: \\
609: &=&
610: \frac{2 \pi}{\hbar^2}
611: \sum_{\alpha,\beta} 
612: \left[
613: %(
614: %C_{FF} A^{(--)FF}_{\alpha \beta}  
615: %+C_{FG} A^{(--)FG}_{\alpha \beta}  
616: %+C_{GG} A^{(--)GG}_{\alpha \beta}  
617: %)
618: C_{--}^{\alpha \beta}  
619: %\nonumber
620: %\\
621: %&&
622: %\times
623: \delta(\tilde{\omega}_S-\omega_{\alpha}-\omega_{\beta})
624: %\right]
625: %\nonumber
626: %\\
627: %&&
628: +
629: %\frac{2 \pi}{\hbar^2}
630: %\left[
631: %\sum_{\alpha,\beta} 
632: %(
633: %C_{FF} A^{(++)FF}_{\alpha \beta}  
634: %+C_{FG} A^{(++)FG}_{\alpha \beta}  
635: %+C_{GG} A^{(++)GG}_{\alpha \beta}  
636: %)
637: C_{++}^{\alpha \beta}  
638: %\nonumber
639: %\\
640: %&&
641: %\times
642: \delta(\tilde{\omega}_S+\omega_{\alpha}+\omega_{\beta})
643: %\nonumber
644: %\\
645: %&&
646: +
647: %\frac{2 \pi}{\hbar^2}
648: %\left[
649: %\sum_{\alpha,\beta} 
650: %(
651: %C_{FF} A^{(+-)FF}_{\alpha \beta}  
652: %+C_{FG} A^{(+-)FG}_{\alpha \beta}  
653: %+C_{GG} A^{(+-)GG}_{\alpha \beta}  
654: %)
655: C_{+-}^{\alpha \beta}  
656: %\nonumber
657: %\\
658: %&&
659: %\times
660: \delta(\tilde{\omega}_S-\omega_{\alpha}+\omega_{\beta})
661: \right]
662: \nonumber
663: \\
664: &&
665: +\frac{2 \pi}{\hbar^2}
666: \sum_{\alpha} 
667: \left[ 
668: C^{\alpha}_{-} \delta(\tilde{\omega}_S-\omega_{\alpha})
669: +
670: C^{\alpha}_{+} \delta(\tilde{\omega}_S+\omega_{\alpha})
671: \right]
672: \label{eq:MF}
673: \end{eqnarray}
674: where we have used 
675: \begin{equation}
676: \left. \frac{d}{dt} u_t(\Omega) \right|_{t \rightarrow \infty} 
677: =\left. \frac{\sin \Omega t}{\Omega} \right|_{t \rightarrow \infty} 
678: = \pi \delta(\Omega).
679: \end{equation}
680: If $\bar{q}_S=q_S$ and $\tilde{\omega}_S=\omega_S$, i.e., 
681: $\langle {\cal F} \rangle = \langle {\cal G} \rangle = 0$ 
682: {\it and} $f=0$,
683: we recover the Maradudin-Fein formula \cite{FBS05a} from 
684: Eq.~(\ref{eq:MF}). It follows that Eq.~(\ref{eq:VER}) is a generalization of 
685: the Maradudin-Fein formula, which can describe the time development 
686: of a density matrix.
687: % which assumes the Markov property 
688: %of the system.
689: 
690: 
691: \subsection{Dephasing formula}
692: 
693: We calculate the off diagonal elements of the density matrix 
694: $(\rho_S)_{10}(t)$ by assuming
695: that the initial state is the superposition state 
696: between $|0 \rangle$ and  $|1 \rangle$: 
697: $\rho_S=(1/2)(
698: |0 \rangle \langle 0|+
699: |0 \rangle \langle 1|+
700: |1 \rangle \langle 0|+
701: |1 \rangle \langle 1|)
702: $ \cite{MO04}. 
703: That is, $(\rho_S)_{kl}=1/2$ for all $k$ and $l$.
704: This is a simplified situation to consider dephasing in a two-level system.
705: We have 
706: \begin{eqnarray}
707: (\rho_S)_{10}(t)
708: &=& (\rho_S)^{(0)}_{10}(t)+(\rho_S)^{(1)}_{10}(t)+(\rho_S)^{(2)}_{10}(t) + \cdots 
709: \nonumber
710: \\
711: &=&
712: \frac{1}{2}e^{-i \tilde{\omega}_S t}(1+ r^{(1)}(t)+r^{(2)}(t)+ \cdots).
713: \label{eq:dephasing}
714: %\nonumber
715: %\\
716: %&\simeq& \frac{1}{2}e^{-i \tilde{\omega}_S t} \exp \{r^{(1)}(t)+r^{(2)}(t) -[r^{(1)}(t)]^2/2 + \cdots \}
717: \end{eqnarray}
718: %Note that 
719: %\begin{eqnarray}
720: %r^{(1)}(t)
721: %&=&
722: %\frac{1}{i \hbar} 
723: %\int^{t}_0 dt'
724: %\sum_k
725: %\left \{ 
726: %\langle \tilde{\cal V}_{1k}(t') \rangle e^{i \omega_{1k}t'} 
727: %-
728: %\langle \tilde{\cal V}_{k0}(t') \rangle e^{i \omega_{k0}t'} 
729: %\right \}
730: %\nonumber
731: %\\
732: %&=&
733: %\frac{1}{i \hbar} 
734: %\int^{t}_0 dt'
735: %\left \{ 
736: %\langle \tilde{\cal V}_{11}(t') \rangle 
737: %-
738: %\langle \tilde{\cal V}_{00}(t') \rangle 
739: %\right \}
740: %%=-i \frac{\langle {\cal G}(0) \rangle}{\tilde{\omega}_S} t
741: %=0
742: %\end{eqnarray}
743: By the definition of the interaction Hamiltonian (Appendix \ref{app:system}), 
744: we have $r^{(1)}(t)=0$.
745: The remaining term $r^{(2)}(t)$ is decomposed as
746: %\begin{eqnarray}
747: %r^{(2)}(t)
748: %&=&
749: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
750: %\sum_{k,l}
751: %\left \{
752: %\langle \tilde{\cal V}_{1k}(t') \tilde{\cal V}_{kl}(t'') \rangle 
753: %e^{i (\omega_{1k}t'+\omega_{kl}t'')} 
754: %\right \}
755: %\nonumber
756: %\\
757: %&+&
758: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
759: %\sum_{k,l}
760: %\left \{
761: %\langle \tilde{\cal V}_{kl}(t'') \tilde{\cal V}_{l0}(t') \rangle 
762: %e^{i (\omega_{kl}t''+\omega_{l0}t')} 
763: %\right \}
764: %\nonumber
765: %\\
766: %&-&
767: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
768: %\sum_{k,l}
769: %\left \{
770: %\langle \tilde{\cal V}_{l0}(t'') \tilde{\cal V}_{1k}(t') \rangle 
771: %e^{i (\omega_{1k}t'+\omega_{l0}t'')} 
772: %\right \}
773: %\nonumber
774: %\\
775: %&-&
776: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
777: %\sum_{k,l}
778: %\left \{
779: %\langle \tilde{\cal V}_{l0}(t') \tilde{\cal V}_{1k}(t'') \rangle 
780: %e^{i (\omega_{1k}t''+\omega_{l0}t')} 
781: %\right \}
782: %\nonumber
783: %\\
784: %&=&
785: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
786: %[\langle {\cal V}_{10}(t'){\cal V}_{01}(t'') \rangle 
787: %+
788: %\langle {\cal V}_{01}(t''){\cal V}_{10}(t') \rangle ] e^{i \tilde{\omega}_S(t'-t'')}
789: %\nonumber
790: %\\
791: %&+&
792: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
793: %[\langle {\cal V}_{10}(t'){\cal V}_{00}(t'') \rangle 
794: %+
795: %\langle {\cal V}_{11}(t''){\cal V}_{10}(t') \rangle ] e^{i \tilde{\omega}_S t'}
796: %\nonumber
797: %\\
798: %&+&
799: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
800: %[\langle {\cal V}_{11}(t'){\cal V}_{10}(t'') \rangle 
801: %+
802: %\langle {\cal V}_{10}(t''){\cal V}_{00}(t') \rangle ] e^{i \tilde{\omega}_S t''}
803: %\nonumber
804: %\\
805: %&+&
806: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
807: %[\langle {\cal V}_{11}(t'){\cal V}_{11}(t'') \rangle 
808: %+
809: %\langle {\cal V}_{00}(t''){\cal V}_{00}(t') \rangle ] 
810: %\nonumber
811: %\\
812: %&-&
813: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
814: %[\langle {\cal V}_{10}(t'){\cal V}_{10}(t'') \rangle 
815: %+
816: %\langle {\cal V}_{10}(t''){\cal V}_{10}(t') \rangle ] e^{i \tilde{\omega}_S(t'+t'')}
817: %\nonumber
818: %\\
819: %&-&
820: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
821: %[\langle {\cal V}_{10}(t'){\cal V}_{11}(t'') \rangle 
822: %+
823: %\langle {\cal V}_{00}(t''){\cal V}_{10}(t') \rangle ] e^{i \tilde{\omega}_S t'}
824: %\nonumber
825: %\\
826: %&-&
827: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
828: %[\langle {\cal V}_{00}(t'){\cal V}_{10}(t'') \rangle 
829: %+
830: %\langle {\cal V}_{10}(t''){\cal V}_{11}(t') \rangle ] e^{i \tilde{\omega}_S t''}
831: %\nonumber
832: %\\
833: %&-&
834: %\frac{1}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
835: %[\langle {\cal V}_{00}(t'){\cal V}_{11}(t'') \rangle 
836: %+
837: %\langle {\cal V}_{00}(t''){\cal V}_{11}(t') \rangle ] 
838: %\nonumber
839: %\\
840: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
841: %&=&
842: %\frac{1}{2 \hbar \tilde{\omega}_S} \int_0^t dt' \int_0^{t'} dt''  
843: %[\langle {\cal F}(t'){\cal F}(t'') \rangle 
844: %+
845: %\langle {\cal F}(t''){\cal F}(t') \rangle ] [e^{i \tilde{\omega}_S(t'+t'')}-e^{i \tilde{\omega}_S(t'-t'')}]
846: %\nonumber
847: %\\
848: %&-&
849: %\frac{1}{2 \tilde{\omega}_S^2} \int_0^t dt' \int_0^{t'} dt''  
850: %[3 \langle {\cal G}(t'){\cal G}(t'') \rangle 
851: %-
852: %\langle {\cal G}(t''){\cal G}(t') \rangle ] 
853: %\nonumber
854: %\\
855: %&-&
856: %\frac{1}{\hbar \tilde{\omega}_S} \sqrt{\frac{\hbar}{2 \tilde{\omega}_S}}
857: % \int_0^t dt' \int_0^{t'} dt''  
858: %[\langle {\cal F}(t'){\cal G}(t'') \rangle 
859: %-
860: %\langle {\cal F}(t''){\cal G}(t') \rangle ] (e^{i \tilde{\omega}_S t'}-e^{i \tilde{\omega}_S t''})
861: %\end{eqnarray}
862: %This $r^{(2)}(t)$ can be rewritten as 
863: \begin{eqnarray}
864: r^{(2)}(t)
865: &=& -r^{(2)}_{FF}(t)-r^{(2)}_{GG}(t)-r^{(2)}_{FG}(t),
866: \\
867: r^{(2)}_{FF}(t)
868: &=&
869: \frac{2}{\hbar^2}
870:  \int_0^t dt' \int_0^{t'} dt''  
871:  {\rm Re} \{ \langle {\cal V}_{10}(t'){\cal V}_{01}(t'') \rangle \}
872: [e^{i \tilde{\omega}_S(t'-t'')}-e^{i \tilde{\omega}_S(t'+t'')}]
873: \nonumber
874: \\
875: &=&
876: \frac{2}{\hbar^2} 
877: (q_S)_{10}^2
878:  \int_0^t dt' \int_0^{t'} dt''  
879:  {\rm Re} \{ \langle \delta {\cal F}(t') \delta {\cal F}(t'') \rangle \}
880: [e^{i \tilde{\omega}_S(t'-t'')}-e^{i \tilde{\omega}_S(t'+t'')}]
881: \nonumber
882: \\
883: &+&
884: \frac{2}{\hbar^2} 
885: (q^2_S)_{10}^2
886:  \int_0^t dt' \int_0^{t'} dt''  
887:  {\rm Re} \{ \langle \delta {\cal G}(t') \delta {\cal G}(t'') \rangle \}
888: [e^{i \tilde{\omega}_S(t'-t'')}-e^{i \tilde{\omega}_S(t'+t'')}]
889: \nonumber
890: \\
891: &-&
892: \frac{4}{\hbar^2} 
893: (q_S)_{10} (q^2_S)_{10}
894:  \int_0^t dt' \int_0^{t'} dt''  
895:  {\rm Re} \{ \langle \delta {\cal F}(t') \delta {\cal G}(t'') \rangle \}
896: [e^{i \tilde{\omega}_S(t'-t'')}-e^{i \tilde{\omega}_S(t'+t'')}],
897: \\
898: r^{(2)}_{GG}(t)
899: &=&
900: \frac{1}{\hbar^2} \int_0^t dt' \int_0^{t'} dt''  
901: \{ 
902: \langle 
903: [{\cal V}_{11}(t')- {\cal V}_{00}(t')] {\cal V}_{11}(t'') 
904: \rangle 
905: %\nonumber
906: %\\
907: %&&
908: +
909: \langle
910: {\cal V}_{00}(t'') [{\cal V}_{00}(t')- {\cal V}_{11}(t')]
911: \rangle  
912: \}
913: \nonumber
914: \\
915: &=&
916: \frac{1}{\hbar^2} [(q_S)_{11}-(q_S)_{00}]^2
917: \int_0^t dt' \int_0^{t'} dt''  
918: {\rm Re}
919: \langle 
920: \delta {\cal F}(t') \delta {\cal F}(t'') 
921: \rangle 
922: \nonumber
923: \\
924: &+&
925: \frac{1}{\hbar^2} [(q^2_S)_{11}-(q^2_S)_{00}]^2
926: \int_0^t dt' \int_0^{t'} dt''  
927: {\rm Re}
928: \langle 
929: \delta {\cal G}(t') \delta {\cal G}(t'') 
930: \rangle 
931: \nonumber
932: \\
933: &-&
934: \frac{2}{\hbar^2} [(q_S)_{11}-(q_S)_{00}] [(q^2_S)_{11}-(q^2_S)_{00}]
935: \int_0^t dt' \int_0^{t'} dt''  
936: {\rm Re}
937: \langle 
938: \delta {\cal F}(t') \delta {\cal G}(t'') 
939: \rangle 
940: \nonumber
941: \\
942: &+&
943: \frac{i}{\hbar^2} [(q_S)^2_{11}-(q_S)^2_{00}]
944: \int_0^t dt' \int_0^{t'} dt''  
945: {\rm Im}
946: \langle 
947: \delta {\cal F}(t') \delta {\cal F}(t'') 
948: \rangle 
949: \nonumber
950: \\
951: &+&
952: \frac{i}{\hbar^2} [(q^2_S)^2_{11}-(q^2_S)^2_{00}]
953: \int_0^t dt' \int_0^{t'} dt''  
954: {\rm Im}
955: \langle 
956: \delta {\cal G}(t') \delta {\cal G}(t'') 
957: \rangle 
958: \nonumber
959: \\
960: &-&
961: \frac{2 i}{\hbar^2} [(q_S)_{11} (q^2_S)_{11} - (q^2_S)_{00}(q^2_S)_{00}]
962: \int_0^t dt' \int_0^{t'} dt''  
963: {\rm Im}
964: \langle 
965: \delta {\cal F}(t') \delta {\cal G}(t'') 
966: \rangle, 
967: \label{eq:gg}
968: \\
969: r^{(2)}_{FG}(t)
970: &=&
971: \frac{2i}{(i \hbar)^2} \int_0^t dt' \int_0^{t'} dt''  
972: {\rm Im} 
973: \{ 
974: \langle [{\cal V}_{11}(t')-{\cal V}_{00}(t')] {\cal V}_{10}(t'') \rangle 
975: \}
976: [e^{i \tilde{\omega}_S t'}-e^{i \tilde{\omega}_S t''}]
977: \nonumber
978: \\
979: &=&
980: \frac{2i}{(i \hbar)^2} 
981: [(q_S)_{11}-(q_S)_{00}](q_S)_{10}
982: \int_0^t dt' \int_0^{t'} dt''  
983: {\rm Im} 
984: \{ 
985: \langle \delta {\cal F}(t') \delta {\cal F}(t'') \rangle 
986: \}
987: [e^{i \tilde{\omega}_S t'}-e^{i \tilde{\omega}_S t''}]
988: \nonumber
989: \\
990: &+&
991: \frac{2i}{(i \hbar)^2} 
992: [(q_S^2)_{11}-(q_S^2)_{00}] (q_S^2)_{10}
993: \int_0^t dt' \int_0^{t'} dt''  
994: {\rm Im} 
995: \{ 
996: \langle \delta {\cal G}(t') \delta {\cal G}(t'') \rangle 
997: \}
998: [e^{i \tilde{\omega}_S t'}-e^{i \tilde{\omega}_S t''}]
999: \nonumber
1000: \\
1001: &-&
1002: \frac{2i}{(i \hbar)^2} 
1003: \{ 
1004: [(q_S)_{11}-(q_S)_{00}]
1005: (q_S^2)_{10}
1006: +
1007: [(q_S^2)_{11}-(q_S^2)_{00}]
1008: (q_S)_{10}
1009: \}
1010: \nonumber
1011: \\
1012: &&
1013: \times
1014: \int_0^t dt' \int_0^{t'} dt''  
1015: {\rm Im} 
1016: \{ 
1017: \langle \delta {\cal F}(t') \delta {\cal G}(t'') \rangle 
1018: \}
1019: [e^{i \tilde{\omega}_S t'}-e^{i \tilde{\omega}_S t''}]
1020: \end{eqnarray}
1021: where the subscript denotes that, e.g., 
1022: for $r^{(2)}_{FF}(t)$, the dominant contribution comes 
1023: from $\langle \delta {\cal F}(t) \delta {\cal F}(0) \rangle$ 
1024: when 
1025: the effects of the bath and the system anharmonicity are both weak. 
1026: 
1027: 
1028: After similar calculations as done for the VER formula above, 
1029: we obtain
1030: \begin{eqnarray}
1031: r^{(2)}_{FF}(t)
1032: &=&
1033: \frac{2}{\hbar^2}
1034: %\left[
1035: \sum_{\alpha,\beta} 
1036: %(
1037: %C_{FF} A^{(--)FF}_{\alpha \beta}  
1038: %+C_{FG} A^{(--)FG}_{\alpha \beta}  
1039: %+C_{GG} A^{(--)GG}_{\alpha \beta}  
1040: %)
1041: \left[
1042: (C_{--}^{\alpha \beta}  +C_{++}^{\alpha \beta})
1043: %\nonumber
1044: %\\
1045: %&&
1046: %\times
1047: f_t (\omega_{\alpha}+\omega_{\beta}) 
1048: %\right]
1049: %\nonumber
1050: %\\
1051: %&&
1052: %-\frac{2}{\hbar^2}
1053: %\left[
1054: %\sum_{\alpha,\beta} 
1055: %(
1056: %C_{FF} A^{(++)FF}_{\alpha \beta}  
1057: %+C_{FG} A^{(++)FG}_{\alpha \beta}  
1058: %+C_{GG} A^{(++)GG}_{\alpha \beta}  
1059: %)
1060: %\nonumber
1061: %\\
1062: %&&
1063: %\times
1064: %f_t(\omega_{\alpha}+\omega_{\beta}) 
1065: %\nonumber
1066: %\\
1067: %&&
1068: %+\frac{2}{\hbar^2}
1069: %\left[
1070: %\sum_{\alpha,\beta} 
1071: %(
1072: %C_{FF} A^{(+-)FF}_{\alpha \beta}  
1073: %+C_{FG} A^{(+-)FG}_{\alpha \beta}  
1074: %+C_{GG} A^{(+-)GG}_{\alpha \beta}  
1075: %)
1076: +C_{+-}^{\alpha \beta}  
1077: %\nonumber
1078: %\\
1079: %&&
1080: %\times
1081: f_t(\omega_{\alpha}-\omega_{\beta}) 
1082: %\right]
1083: %\nonumber
1084: %\\
1085: %&&
1086: %+\frac{2}{\hbar^2}
1087: %\sum_{\alpha} 
1088: +
1089: (C^{\alpha}_{-} + C^{\alpha}_{+}) f_t(\omega_{\alpha})
1090: \right],
1091: \label{eq:FF}
1092: \\
1093: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1094: r^{(2)}_{GG}(t)
1095: &=&
1096: \frac{1}{\hbar^2}
1097: \sum_{\alpha,\beta}
1098: %\left(
1099: %D_{FF}^{(R)} A_{\alpha \beta}^{(--)FF} 
1100: %+
1101: %D_{GG}^{(R)} A_{\alpha \beta}^{(--)GG} 
1102: %+
1103: %D_{FG}^{(R)} A_{\alpha \beta}^{(--)FG} 
1104: %\right)
1105: \left[
1106: (D_{--}^{R\alpha \beta}+D_{++}^{R\alpha \beta})
1107: u_t(\omega_{\alpha}+\omega_{\beta})
1108: %\nonumber
1109: %\\
1110: %&&
1111: %+\frac{1}{\hbar^2}
1112: %\sum_{\alpha,\beta}
1113: %\left(
1114: %D_{FF}^{(R)} A_{\alpha \beta}^{(++)FF} 
1115: %+
1116: %D_{GG}^{(R)} A_{\alpha \beta}^{(++)GG} 
1117: %+
1118: %D_{FG}^{(R)} A_{\alpha \beta}^{(++)FG} 
1119: %\right)
1120: %u_t(\omega_{\alpha}+\omega_{\beta})
1121: %\nonumber
1122: %\\
1123: %&&
1124: +
1125: %\frac{1}{\hbar^2}
1126: %\sum_{\alpha,\beta}
1127: %\left(
1128: %D_{FF}^{(R)} A_{\alpha \beta}^{(+-)FF} 
1129: %+
1130: %D_{GG}^{(R)} A_{\alpha \beta}^{(+-)GG} 
1131: %+
1132: %D_{FG}^{(R)} A_{\alpha \beta}^{(+-)FG} 
1133: %\right)
1134: D_{+-}^{R\alpha \beta}
1135: u_t(\omega_{\alpha}-\omega_{\beta})
1136: %\nonumber
1137: %\\
1138: %&&
1139: %+\frac{1}{\hbar^2}
1140: %\sum_{\alpha} 
1141: +
1142: (D_{-}^{R \alpha}+D_{+}^{R\alpha})
1143: u_t(\omega_{\alpha})
1144: \right]
1145: \nonumber
1146: \\
1147: &&
1148: - \frac{i}{\hbar^2}
1149: \sum_{\alpha,\beta}
1150: %\left(
1151: %D_{FF}^{(I)} A_{\alpha \beta}^{(--)FF} 
1152: %+
1153: %D_{GG}^{(I)} A_{\alpha \beta}^{(--)GG} 
1154: %+
1155: %D_{FG}^{(I)} A_{\alpha \beta}^{(--)FG} 
1156: %\right)
1157: \left[
1158: (D_{--}^{I\alpha \beta}-D_{++}^{I\alpha \beta})
1159: v_t(\omega_{\alpha}+\omega_{\beta})
1160: %\nonumber
1161: %\\
1162: %&&
1163: %- \frac{i}{\hbar^2}
1164: %\sum_{\alpha,\beta}
1165: %\left(
1166: %D_{FF}^{(I)} A_{\alpha \beta}^{(++)FF} 
1167: %+
1168: %D_{GG}^{(I)} A_{\alpha \beta}^{(++)GG} 
1169: %+
1170: %D_{FG}^{(I)} A_{\alpha \beta}^{(++)FG} 
1171: %\right)
1172: %v_t(\omega_{\alpha}+\omega_{\beta})
1173: %\nonumber
1174: %\\
1175: %&&
1176: %- \frac{i}{\hbar^2}
1177: %\sum_{\alpha,\beta}
1178: %\left(
1179: %D_{FF}^{(I)} A_{\alpha \beta}^{(+-)FF} 
1180: %+
1181: %D_{GG}^{(I)} A_{\alpha \beta}^{(+-)GG} 
1182: %+
1183: %D_{FG}^{(I)} A_{\alpha \beta}^{(+-)FG} 
1184: %\right)
1185: +
1186: D_{+-}^{I\alpha \beta}
1187: v_t(\omega_{\alpha}-\omega_{\beta})
1188: %\nonumber
1189: %\\
1190: %&&
1191: %- \frac{i}{\hbar^2} \sum_{\alpha} 
1192: +(D^{I \alpha}_{-}-D^{I \alpha}_{+})
1193: v_t(\omega_{\alpha})
1194: \right],
1195: \label{eq:GG}
1196: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1197: \\
1198: r^{(2)}_{FG}(t)
1199: &=&
1200: \frac{2}{\hbar^2}
1201: \sum_{\alpha,\beta}
1202: %\left(
1203: %E_{FF} A_{\alpha \beta}^{(--)FF} 
1204: %+
1205: %E_{GG} A_{\alpha \beta}^{(--)GG} 
1206: %+
1207: %E_{FG} A_{\alpha \beta}^{(--)FG} 
1208: %\right)
1209: \left[
1210: (E_{--}^{\alpha \beta}-E_{++}^{\alpha \beta})
1211: g_t(\omega_{\alpha}+\omega_{\beta})
1212: %\nonumber
1213: %\\
1214: %&&
1215: %+\frac{2}{\hbar^2}
1216: %\sum_{\alpha,\beta}
1217: %\left(
1218: %E_{FF} A_{\alpha \beta}^{(++)FF} 
1219: %+
1220: %E_{GG} A_{\alpha \beta}^{(++)GG} 
1221: %+
1222: %E_{FG} A_{\alpha \beta}^{(++)FG} 
1223: %\right)
1224: %g_t(\omega_{\alpha}+\omega_{\beta})
1225: %\nonumber
1226: %\\
1227: %&&
1228: %+\frac{2}{\hbar^2}
1229: %\sum_{\alpha,\beta}
1230: %\left(
1231: %E_{FF} A_{\alpha \beta}^{(+-)FF} 
1232: %+
1233: %E_{GG} A_{\alpha \beta}^{(+-)GG} 
1234: %+
1235: %E_{FG} A_{\alpha \beta}^{(+-)FG} 
1236: %\right)
1237: +E_{+-}^{\alpha \beta}
1238: g_t(\omega_{\alpha}-\omega_{\beta})
1239: %\nonumber
1240: %\\
1241: %&&
1242: %+\frac{2}{\hbar^2} \sum_{\alpha} 
1243: +(E^{\alpha}_{-}-E^{\alpha}_{+})
1244: g_t(\omega_{\alpha})
1245: \right]
1246: \label{eq:FG}
1247: \end{eqnarray}
1248: where 
1249: \begin{eqnarray}
1250: f_t(\Omega)
1251: &=&
1252: \int_0^t dt' \int_0^{t'} dt''  
1253: \cos \Omega (t'-t'') 
1254: [e^{i \tilde{\omega}_S(t'-t'')} -e^{i \tilde{\omega}_S(t'+t'')}]
1255: \nonumber
1256: \\
1257: &=& 
1258: \frac{1}{2}
1259: \left[
1260: \frac{1}{\tilde{\omega}_S-\Omega}+\frac{1}{\tilde{\omega}_S+\Omega}
1261: \right]
1262: \left\{
1263: \frac{1-e^{i(\tilde{\omega}_S+\Omega)t}}{\tilde{\omega}_S+\Omega}
1264: +\frac{1-e^{i(\tilde{\omega}_S-\Omega)t}}{\tilde{\omega}_S-\Omega}
1265: +\frac{2i \tilde{\omega}_S t -1+e^{2 i\tilde{\omega}_S t}}{2\tilde{\omega}_S}
1266: \right\},
1267: %\nonumber
1268: %\\
1269: %&=&
1270: %\frac{1-e^{i(\tilde{\omega}_S-\Omega)t}}{2(\tilde{\omega}_S-\Omega)^2}
1271: %+\frac{1-e^{i(\tilde{\omega}_S+\Omega)t}}{2(\tilde{\omega}_S+\Omega)^2}
1272: %\nonumber
1273: %\\
1274: %&&
1275: %+\frac{1-e^{-i \tilde{\omega}_S t} \cos \Omega t}{\Omega^2-\tilde{\omega}_S^2}
1276: %+
1277: %\frac{1-2 e^{-i \tilde{\omega}_S t} \cos \Omega t+ e^{2 i\tilde{\omega}_S t} + 2i \tilde{\omega}_S t}
1278: %{2 (\tilde{\omega}_S^2-\Omega^2)}
1279: %\\
1280: %f_t(\tilde{\omega}_S)
1281: %&=&
1282: %\frac{t^2}{4}+\frac{1-(1-2 i \tilde{\omega}_S t)e^{2 i \tilde{\omega}_S t}}{8 \tilde{\omega}_S^2}
1283: \\
1284: g_t(\Omega)
1285: &=&
1286: i \int_0^t dt' \int_0^{t'} dt''  
1287: \sin \Omega (t'-t'') 
1288: [e^{i \tilde{\omega}_S t'} -e^{i \tilde{\omega}_S t''}]
1289: \nonumber
1290: \\
1291: %&=& 
1292: %\frac{e^{i \tilde{\omega}_S t}-1}{\tilde{\omega}_S}
1293: %\left[
1294: %\frac{1}{\Omega}-\frac{1}{2(\Omega-\tilde{\omega}_S)}
1295: %-\frac{1}{2(\Omega+\tilde{\omega}_S)}
1296: %\right]
1297: %\nonumber
1298: %\\
1299: %&&
1300: %+\frac{e^{i \Omega t}+e^{-i(\Omega-\tilde{\omega}_S)t}-2}{2 \Omega(\Omega-\tilde{\omega}_S)}
1301: %-\frac{e^{-i \Omega t}+e^{i(\Omega+\tilde{\omega}_S)t}-2}{2 \Omega(\Omega+\tilde{\omega}_S)}
1302: %\nonumber
1303: %\\
1304: &=&
1305: %\frac{\tilde{\omega}_S(1-e^{i \tilde{\omega}_S t})}{\Omega( \Omega^2-\tilde{\omega}_S^2)}
1306: \frac{
1307: \tilde{\omega}_S (1+e^{i \tilde{\omega}_S t}) (1-\cos \Omega t )
1308: -i \Omega (1-e^{i \tilde{\omega}_S t}) \sin \Omega t } 
1309: {\Omega (\tilde{\omega}_S^2-\Omega^2)},
1310: %\\
1311: %g_t(\tilde{\omega}_S)
1312: %&=&
1313: %\frac{
1314: %-\tilde{\omega}_S t(1+e^{i\tilde{\omega}_S t}) \sin \tilde{\omega}_S t
1315: %+ i (1-e^{i\tilde{\omega}_S t}) \sin \tilde{\omega}_S t
1316: %+ i \tilde{\omega}_S t(1-e^{i\tilde{\omega}_S t}) \cos \tilde{\omega}_S t
1317: %}
1318: %{2 \tilde{\omega}_S^2}
1319: %\nonumber
1320: %\\
1321: %&=&
1322: %\frac{
1323: %i(e^{i\tilde{\omega}_S t}-1) (\tilde{\omega}_S t - \sin \tilde{\omega}_S t)
1324: %}
1325: %{2 \tilde{\omega}_S^2}
1326: %\\
1327: %g_t(0)
1328: %&=&
1329: \end{eqnarray}
1330: and the coefficients are defined in Appendix \ref{app:coef}.
1331: Equation (\ref{eq:dephasing}) and Eqs.~(\ref{eq:FF})-(\ref{eq:FG}) 
1332: constitute the dephasing formula.
1333: %We notice that if the resonant condition is important 
1334: %\begin{equation}
1335: %{\rm Re} \{ r^{(2)}_{FF}(t) \} 
1336: %\simeq 
1337: %\frac{1}{2} (\rho_S)_{00}(t)
1338: %\end{equation}
1339: 
1340: Dephasing properties are characterized by the decaying behavior of 
1341: this off diagonal density matrix.
1342: Incidentally, as an alternative approach, 
1343: one might use the von Neuman entropy or linear entropy for the 
1344: reduced system as an indicator of dephasing \cite{FMT03}.
1345: 
1346: 
1347: 
1348: \subsection{Frequency autocorrelation function}
1349: 
1350: Using the time-independent perturbation theory 
1351: for the interaction, we obtain the fluctuation of the 
1352: system frequency as 
1353: \begin{equation}
1354: \delta \omega(t)= \frac{{\cal V}_{11}(t)-{\cal V}_{00}(t)}{\hbar}
1355: =-\frac{(q_S)_{11}-(q_S)_{00}}{\hbar} \delta {\cal F}(t)
1356: +\frac{(q_S^2)_{11}-(q_S^2)_{00}}{\hbar} \delta {\cal G}(t).
1357: \end{equation}
1358: Hence we have 
1359: \begin{eqnarray}
1360: {\rm Re}\langle \delta \omega(t')\delta \omega(t'') \rangle
1361: &=& 
1362: \frac{1}{\hbar^2}
1363: \left(
1364: [(q_S)_{11}-(q_S)_{00}]^2
1365: {\rm Re} \langle \delta {\cal F}(t')\delta {\cal F}(t'') \rangle
1366: +[(q_S^2)_{11}-(q_S^2)_{00}]^2
1367: {\rm Re} \langle \delta {\cal G}(t')\delta {\cal G}(t'') \rangle
1368: \right.
1369: \nonumber
1370: \\
1371: &&
1372: \left.
1373: -2 [(q_S)_{11}-(q_S)_{00}]
1374: [(q_S^2)_{11}-(q_S^2)_{00}]
1375: {\rm Re} \langle \delta {\cal F}(t')\delta {\cal G}(t'') \rangle
1376: \right).
1377: %=\frac{1}{\hbar^2} {\rm Re}I_{GG} = C(t'-t'')
1378: \end{eqnarray}
1379: This turns out to be the second derivative of 
1380: ${\rm Re} \{ r^{(2)}_{GG}(t) \}$, i.e.,
1381: \begin{equation}
1382: C(t) \equiv {\rm Re}\langle \delta \omega(t)\delta \omega(0) \rangle
1383: =
1384: \frac{d^2}{dt^2} {\rm Re} \{ r^{(2)}_{GG}(t) \}.
1385: \end{equation}
1386: From this correlation function, 
1387: we can define a {\it pure dephasing time} $T_2^*$ as
1388: \begin{eqnarray}
1389: %\delta \omega^2
1390: %&=& C(0)
1391: %\\
1392: \frac{1}{T_2^*}
1393: =
1394: \int_0^{\infty} C(t) dt. 
1395: %\\
1396: %\tau_c
1397: %&=&
1398: %\frac{1}{\delta \omega^2} \int_0^{\infty} C(t) dt =
1399: %\frac{1}{\delta \omega^2} \frac{1}{T_2^*}
1400: \end{eqnarray}
1401: Note that this is different from a correlation time defined by
1402: \begin{eqnarray}
1403: %\delta \omega^2
1404: %&=& C(0)
1405: %\\
1406: \tau_c
1407: =
1408: \frac{1}{C(0)} \int_0^{\infty} C(t) dt 
1409: %\\
1410: %\tau_c
1411: %&=&
1412: %\frac{1}{\delta \omega^2} \int_0^{\infty} C(t) dt =
1413: %\frac{1}{\delta \omega^2} \frac{1}{T_2^*}
1414: \end{eqnarray}
1415: which leads to the relation $1/T_2^*=C(0) \tau_c$.
1416: Note that $T_2^*$ and $\tau_c$ are inversely related.
1417: 
1418: %where $T_2^*$ is a pure dephasing time, and $\tau_c$ 
1419: %is a correlation time for the bath.
1420: %Note that the latter two equations only hold when the 
1421: %Markov approximation is valid, i.e., when $C(t)$ decays quickly.
1422: 
1423: 
1424: 
1425: 
1426: \section{Application: NMA in heavy water}
1427: \label{sec:NMA}
1428: 
1429: \subsection{NMA-D in heavy water}
1430: 
1431: 
1432: We now apply our formulas to VER and dephasing problems 
1433: of $N$-methylacetamide (NMA) in heavy water.
1434: %Note that we can calculate a VER time as well as a dephasing time,
1435: %whereas in the previous studies a VER time is usually just borrowed 
1436: %from experiment.
1437: In many theoretical and experimental studies,
1438: this molecule CH$_3$-NH-CO-CH$_3$ is taken to be 
1439: a model ``minimal'' peptide system 
1440: because it contains a peptide bond (-NH-CO-).
1441: For example, Gerber and coworkers calculated the 
1442: vibrational frequencies for this molecule
1443: using the vibrational self-consistent field (VSCF) method \cite{GCG02}. 
1444: Nguyen and Stock worked to characterize VER in
1445: this molecule using a quasi-classical method \cite{NS03}.
1446: Skinner and coworkers investigated the dephasing properties of 
1447: the amide I mode using their correlation method combined with ab initio (DFT) 
1448: calculations \cite{SCS04}.
1449: Employing 2D-IR spectroscopy, Zanni {\it et al}.~\cite{ZAH01}
1450: measured $T_1$ and $T_2^*$ for the amide I mode in this molecule,
1451: which were reported to be $T_1 \simeq 0.45$ ps and $T_2^* \simeq 1.12$ ps,
1452: whereas Woutersen {\it et al}. \cite{WPHMKS02} obtained  $T_2^* \simeq 0.8$ ps.
1453: 
1454: In our study, we deuterate the system to NMAD/D$_2$O so that 
1455: the amide I mode,
1456: localized around the CO bond,
1457: can be clearly recognized as a single peak in the spectrum. 
1458: In the following numerical calculations 
1459: based on the CHARMM force field \cite{CHARMM}, 
1460: its frequency is $\sim$ 1690 cm$^{-1}$ which fluctuates depending on the structure 
1461: (see Fig.~\ref{fig:shift}),
1462: whereas the experimental and DFT values are 1717 cm$^{-1}$ and 
1463: 1738 cm$^{-1}$, respectively \cite{SCS04}.
1464: 
1465: 
1466: 
1467: \subsection{Procedure}
1468: 
1469: We have applied the following general procedure.
1470: %\begin{itemize}
1471: %\item
1472: (1) Run an equilibrium simulation. % , say, up to 100ps. 
1473: %(One can take any convenient ensemble like NEV, NPT etc.)
1474: %The total number of atoms is 1665.
1475: %\item
1476: (2) Sample several trajectories during the run.
1477: %, say, $N=$100 trajectories during the run.
1478: %\item
1479: (3) Delete atoms of each configuration except the ``active'' region 
1480: around the system oscillator.
1481: We introduce a cut off radius $R_c$ around a certain 
1482: atom within the active site.
1483: %for VER.
1484: %(This strategy is similar to that of QM/MM methods.) See the right of 
1485: %Fig.~\ref{fig:NMA}. When the cut off length is 9\AA, the 
1486: %number of the degrees of freedom approximately reduces to 400 atoms.
1487: %\item
1488: (4) Calculate instantaneous normal modes (INMs) \cite{Keyes} for each ``reduced'' configuration,
1489: ignoring all the imaginary frequencies \cite{MO04}.
1490: %\item
1491: (5) Calculate anharmonic coupling elements with the finite difference method \cite{FBS05a} 
1492: using the obtained INMs.
1493: %\item
1494: (6) Insert the results in the VER formula [Eq.~(\ref{eq:VER})] and dephasing formula 
1495: [Eq.~(\ref{eq:dephasing}) with Eqs.~(\ref{eq:FF})-(\ref{eq:FG})]. 
1496: %Eqs.~(\ref{eq:MO1}) and (\ref{eq:MO2}), for a density matrix. 
1497: %See Fig.~\ref{fig:VER1}.
1498: %\item
1499: (7) Ensemble average the resultant density matrix, and 
1500: estimate the VER and dephasing rates (times), if possible. 
1501: %\begin{equation}
1502: %\bar{\rho}_{mn}(t) \equiv
1503: %\frac{1}{N} \sum_{i=1}^N 
1504: %\rho^{(i)}_{mn}(t)
1505: %\end{equation}
1506: %\end{itemize}
1507: 
1508: This procedure seems to be straighforward.
1509: However, when applied to real systems like peptides or proteins,
1510: we need to carefully treat the the effects of the bath.
1511: %First we see the effects of the bath and the system anharmonicity.
1512: In Fig.~\ref{fig:shift}, we plot the system frequency ${\omega}_S$ 
1513: %$\bar{\omega}_S/\omega_S=\sqrt{1+\frac{2 \langle {\cal G} \rangle}{\omega_S^2}}$, 
1514: %and 
1515: %$b/a= 
1516: %\frac{\langle {\cal F} \rangle}{\bar{\omega}_S^2}/\sqrt{\frac{\hbar}{2 \bar{\omega}_S}$ 
1517: %($a=\sqrt{\frac{\hbar}{2 \bar{\omega}_S}}$)
1518: %and $\varepsilon = \frac{1}{\hbar \bar{\omega}_S} \frac{f}{6} 
1519: %\left(\frac{\hbar}{2 \bar{\omega}_S}\right)^{3/2}$ 
1520: for 100 sample trajectories from an equilibrium run at 300K.
1521: We see that the amide I mode frequency changes depending on 
1522: the structure; the deviation can amount to 1\%. %(Fig.~\ref{fig:NMA2}, Left).
1523: %On the other hand, the renormalized frequency also changes 
1524: %but its deviation can be much larger $\sim$ 10 \% (Fig.~\ref{fig:NMA2}, Middle).
1525: %The shift $b/a$ can be large at the same structure when 
1526: %the renormalized frequency deviates from the ``original''
1527: %frequency $\omega_S$ (Fig.~\ref{fig:NMA2}, Right).
1528: 
1529: 
1530: 
1531: %Since such a large deviation in the system frequency implies that the 
1532: %perturbative treatment is inaccurate, 
1533: %we have a criterion to {\it reject} a trajectory (configuration):
1534: %if $|\bar{\omega}_S/\omega_S-1|$ is more than 2\%,
1535: %we do not include such a trajectory in the ensemble average.
1536: %This might be a reasonable one because if there is such a 
1537: %deviation, the system frequency shifts about 40 cm$^{-1}$ 
1538: %($\omega_S \sim 2000$ cm$^{-1}$), and the structure of an absorption 
1539: %spectrum completely changes.
1540: 
1541: \begin{figure}[htbp]
1542: \hfill
1543: \begin{center}
1544: %\begin{minipage}{.42\linewidth}
1545: \includegraphics[scale=1.2]{shift4.eps}
1546: %\end{minipage}
1547: %\hspace{1cm}
1548: %\begin{minipage}{.42\linewidth}
1549: %\includegraphics[scale=1.2]{VER_all.eps}
1550: %\end{minipage}
1551: %\vspace{1cm}
1552: %\begin{minipage}{.32\linewidth}
1553: %\includegraphics[scale=0.8]{shift1.eps}
1554: %\end{minipage}
1555: %\begin{minipage}{.43\linewidth}
1556: %\includegraphics[scale=1.0]{shift3.eps}
1557: %\end{minipage}
1558: %\vspace{1cm}
1559: \end{center}
1560: \caption{
1561: %{\sf 
1562: %Left: 
1563: Instantaneous normal mode frequency of the system
1564: ${\omega}_S$ %, Middle: $\bar{\omega}_S/\omega_S$, Right: $b/a$ 
1565: for 100 different sample trajectories at 300K where
1566: the cut off radius is $R_c= 10$ \AA.
1567: %}
1568: }
1569: \label{fig:shift}
1570: \end{figure}
1571: 
1572: {\it Furthermore} the frequency is renormalized according to Eqs.~(\ref{eq:omegas}) and (\ref{eq:freq}),
1573: and such an effect can be anomalously large if we include all the contribution from 
1574: low frequency components. Hence we need to introduce a cutoff frequency $\omega_c$,
1575: below which the contribution is neglected.
1576: This is physically sound, because we are dealing with time-dependent phenomena,
1577: and such low frequency components correspond to longer time behavior. However,
1578: we are now interested in rather short time dynamics, so such contributions 
1579: should not play a role. In fact, the final result of VER does not depend 
1580: much on the choice of $\omega_c$, whereas that of dephasing does.
1581: We need to admit that for now this is just a remedy.
1582: We discuss how to improve this situation later.
1583: 
1584: 
1585: 
1586: 
1587: %\subsection{Numerical results}
1588: 
1589: \subsection{VER properties of NMA-D}
1590: 
1591: First we consider the VER properties of the amide I mode as shown in
1592: Fig.~\ref{fig:VER1}.
1593: We use the following relation 
1594: \begin{equation}
1595: \rho_{11}(t)=1 - \rho_{00}(t)=\exp[-s(t)]
1596: \end{equation}
1597: and hypothesize that $s(t) \simeq \rho_{00}(t)$, which is definitely 
1598: true  when $\rho_{00}(t) \ll 1$, and might be justified using 
1599: the cumulant expansion technique \cite{Breuer,IMM03}.
1600: 
1601: We calculated the density matrix for the following three cases: 
1602: (a) NMA-D in heavy water with CHARMM force field at 300K,
1603: (b) NMA-D in vacuum with CHARMM force field at 0K,
1604: and 
1605: (c) NMA-D in vacuum with DFT force field at 0K.
1606: Here we have used $R_c=10$ \AA \, and $\omega_c=10$ cm$^{-1}$ for case (a).
1607: The result for VER does not depend sensitively on these parameters.
1608: For cases (b) and (c), we must take a special care:
1609: It is known that the low frequency components cause serious 
1610: problems for vibrational frequency calculations \cite{Yagi},
1611: so we need to eliminate the low frequency components.
1612: In this work, we exclude several normal modes if their frequency 
1613: is less than 300 cm$^{-1}$. See Table \ref{table:NMAfreq}.
1614: 
1615: 
1616: \begin{table}[htbp]
1617: \caption{
1618: Normal mode frequencies (in cm$^{-1}$) 
1619: for ab initio (left) and CHARMM (right) NMA.
1620: The level of the ab initio calculation is B3LYP/6-31+G(d).
1621: }
1622: \hfill
1623: \begin{center}
1624: \begin{tabular}{c|c|c}
1625: \hline
1626: Mode index $\alpha$   & $\omega_{\alpha}$ (ab initio) & $\omega_{\alpha}$ (CHARMM) \\
1627: \hline \hline
1628: 1 &      31.5 	       &   64.1    \\ \hline
1629: 2 &      71.6	       &   88.9    \\ \hline
1630: 3 &     170.2	       &  192.3    \\ \hline
1631: 4 &     259.6	       &  269.5    \\ \hline
1632: 5 &     282.3	       &  426.7    \\ \hline
1633: 6 &     421.9	       &  536.3    \\ \hline
1634: 7 &     619.1	       &  575.9    \\ \hline
1635: 8 &     619.9	       &  741.9    \\ \hline
1636: 9 &     868.7	       &  757.0    \\ \hline
1637: 10 &    946.1	       &  854.4    \\ \hline
1638: 11 &   1012.4	       &  964.3    \\ \hline
1639: 12 &   1066.1	       & 1055.9    \\ \hline
1640: 13 &   1144.5	       & 1075.7    \\ \hline
1641: 14 &   1158.9	       & 1088.5    \\ \hline
1642: 15 &   1207.6	       & 1123.5    \\ \hline
1643: \end{tabular}
1644: \hspace{1cm}
1645: \begin{tabular}{c|c|c}
1646: \hline
1647: Mode index $\alpha$   & $\omega_{\alpha}$ (ab initio) & $\omega_{\alpha}$ (CHARMM) \\
1648: \hline \hline
1649: 16 &   1417.9	       &     1380.2    \\ \hline
1650: 17 &   1436.1	       &     1408.7    \\ \hline
1651: 18 &   1483.6	       &     1415.4    \\ \hline
1652: 19 &   1495.1	       &     1418.5    \\ \hline
1653: 20 &   1499.4	       &     1425.7    \\ \hline
1654: 21 &   1516.7	       &     1444.7    \\ \hline
1655: 22 &   1535.9	       &     1563.1    \\ \hline
1656: 23 &   1745.9	       &     1678.1    \\ \hline
1657: 24 &   2671.1	       &     2445.0    \\ \hline
1658: 25 &   3058.5	       &     2852.8    \\ \hline
1659: 26 &   3058.9	       &     2914.3    \\ \hline
1660: 27 &   3116.5	       &     2914.8    \\ \hline
1661: 28 &   3130.9	       &     2917.2    \\ \hline
1662: 29 &   3135.9	       &     2975.3    \\ \hline
1663: 30 &   3148.9	       &     2975.5    \\ \hline
1664: \end{tabular}
1665: \end{center}
1666: \label{table:NMAfreq}
1667: \end{table}
1668: 
1669: 
1670: 
1671: By using a fitting form $s(t)=t/T_1$,
1672: where $T_1$ is the VER time,
1673: we estimate that 
1674: $T_1 \simeq 0.5$ ps at 300K and 0.6 ps at 0K from the initial decay. 
1675: The former estimate is rather similar to the experimental value $T_1 \simeq 0.45$ ps \cite{ZAH01},
1676: %However, we should not trust this result completely, because $\rho_{00}(t)$ should not 
1677: %exceed one ($0 \leq \rho_{00}(t) \leq 1$).
1678: whereas Nguyen-Stock's quasi-classical estimate is $T_1 \simeq 1.5$ ps \cite{NS03}.
1679: Considering that the estimate at 300K is rather close to that at 0K,
1680: we can conclude that quantum effects are important to describe VER for 
1681: the amide I mode of NMA-D in heavy water.
1682: However, the decay at the later stage becomes very slow at 0K as expected 
1683: because there is no environment. 
1684: (In the vacuum cases, we only use one minimized structure, 
1685: thus there is no ensemble average, and the oscillatory behavior remains.)
1686: 
1687: The results are similar for 
1688: NMA-D in vacuum with different force fields.
1689: It is known that NMA with CHARMM force field is not well
1690: characterized around the methyl groups \cite{GK92}, but
1691: this fact does not affect the VER properties of the amide I mode.
1692: 
1693: We have analyzed the mechanism of VER in terms of the VER pathway.
1694: In Table \ref{table:pathway}, we show 
1695: several mode combinations that contribute most to $s(t)$ for NMA-D 
1696: in heavy water at 300K.
1697: These eigenvectors (normal modes) are well localized 
1698: around NMA (Fig.~\ref{fig:norm}), especially on the CO bond (Table \ref{table:norm}).
1699: There is very little contribution from the surrounding water.
1700: (This is expected from the previous result of Kidera and coworkers \cite{Kidera}.)
1701: Similar ``resonant'' mode combinations can be found in the 
1702: isolated NMA-D cases. See Tables \ref{table:pathway2} and \ref{table:pathway3}.
1703: This means that the initial stage of VER of NMA-D in heavy water 
1704: is dominated by intramolecular vibrational redistribution (IVR) 
1705: localized near the peptide bond.
1706: This result might explain why the amide I mode, 
1707: in many peptide systems with differing environments, 
1708: appears to have similar VER times \cite{MKFKAZ04}.
1709: Note that this is the case for a {\it localized mode} such as the 
1710: deuterated amide I mode. 
1711: A {\it collective mode} can decay 
1712: with a different VER pathway,
1713: as shown by Austin's group \cite{XMHA00}.
1714: 
1715: 
1716: \begin{figure}[htbp]
1717: \hfill
1718: \begin{center}
1719: %\begin{minipage}{.42\linewidth}
1720: %\includegraphics[scale=1.2]{shift4.eps}
1721: %\end{minipage}
1722: %\hspace{1cm}
1723: %\begin{minipage}{.42\linewidth}
1724: \includegraphics[scale=1.2]{VER_all.eps}
1725: %\end{minipage}
1726: %\vspace{1cm}
1727: %\begin{minipage}{.32\linewidth}
1728: %\includegraphics[scale=0.8]{shift1.eps}
1729: %\end{minipage}
1730: %\begin{minipage}{.43\linewidth}
1731: %\includegraphics[scale=1.0]{shift3.eps}
1732: %\end{minipage}
1733: %\vspace{1cm}
1734: \end{center}
1735: \caption{
1736: %{\sf 
1737: %Left: 
1738: Time evolution of the excited density matrix 
1739: for NMA-D in vacuum (DFT and CHARMM) at 0K, and for 
1740: NMA-D in heavy water (CHARMM) at 300K.
1741: The level of DFT is B3LYP/6-31+G(d).
1742: %}
1743: }
1744: \label{fig:VER1}
1745: \end{figure}
1746: 
1747: 
1748: \begin{table}[htbp]
1749: \caption{
1750: %{\sf 
1751: The most dominant VER pathways for the amide I mode of NMA-D in heavy water.
1752: $\Delta \omega$ is defined by $|\omega_S-\omega_{\alpha}-\omega_{\beta}|$.
1753: %}
1754: }
1755: \hfill
1756: \begin{center}
1757: \begin{tabular}{c|c|c|c}
1758: \hline \hline
1759: Mode combination  ($\alpha, \beta$) &  frequency (cm$^{-1}$) &  Contribution to $s(t)$ & 
1760:             $\Delta \omega$ (cm$^{-1}$) \\ \hline \hline           
1761:        1143   +     1143 & 778.6   +    778.6  & 0.04 &       125.7    \\ \hline
1762:        1147   +     1134 & 1085.3  +    570.0  & 0.04 &       27.6    \\ \hline
1763:        1147   +     1135 & 1085.3  +    570.9  & 0.01 &       26.6      \\ \hline
1764:        1147   +     1136 & 1085.3  +    578.8  & 0.02 &       18.8      \\ \hline
1765:        1147   +     1137 & 1085.3  +    581.3  & 0.03 &       16.2      \\ \hline
1766:        1147   +     1140 & 1085.3  +    612.1  & 0.46 &       14.5      \\ \hline
1767:        1148   +     1132 & 1127.8  +    558.5  & 0.01 &       3.4      \\ \hline
1768:        1148   +     1134 & 1127.8  +    570.0  & 0.11 &       14.9      \\ \hline
1769:        1148   +     1135 & 1127.8  +    570.9  & 0.03 &       15.9      \\ \hline
1770: \end{tabular}
1771: \end{center}
1772: \label{table:pathway}
1773: \end{table}
1774: 
1775: \begin{figure}[htbp]
1776: \hfill
1777: \begin{center}
1778: %\begin{minipage}{.42\linewidth}
1779: %\includegraphics[scale=1.2]{norm2.eps}
1780: %\end{minipage}
1781: %\hspace{1cm}
1782: %\begin{minipage}{.42\linewidth}
1783: \includegraphics[scale=1.2]{norm3.eps}
1784: %\end{minipage}
1785: %\begin{minipage}{.32\linewidth}
1786: %\includegraphics[scale=0.8]{corr_NMA.eps}
1787: %\end{minipage}
1788: %\vspace{1cm}
1789: \end{center}
1790: \caption{
1791: %{\sf 
1792: Norm of the eigenvectors (normal modes)
1793: with the exception of the contribution from NMA-D, 
1794: which is defined by
1795: $\sum_{i \in {\rm Water}} (x_i^2 +y_i^2+z_i^2)$,
1796: where $i$ comes from water degrees of freedom alone.
1797: %}
1798: }
1799: \label{fig:norm}
1800: \end{figure}
1801: 
1802: \begin{table}[htbp]
1803: \caption{
1804: %{\sf 
1805: The most localized modes around the CO bond in NMA-D.
1806: The norm is defined by $\sum_{i \in {\rm CO bond}} (x_i^2 +y_i^2+z_i^2)$.
1807: The 1345th mode is the amide I mode.
1808: %}
1809: }
1810: \hfill
1811: \begin{center}
1812: \begin{tabular}{c|c|c}
1813: \hline \hline
1814: Mode index $\alpha$ &  frequency (cm$^{-1}$) & Contribution to norm  \\ \hline \hline           
1815:         1134 & 570.0 & 0.14    \\ \hline
1816:         1140 & 612.1 & 0.26    \\ \hline
1817:         1142 & 771.2 & 0.35    \\ \hline
1818:         1143 & 778.6 & 0.41    \\ \hline
1819:         1146 & 1013.6 & 0.13    \\ \hline
1820:         1147 & 1085.3 & 0.26    \\ \hline
1821:         1148 & 1127.8 & 0.17    \\ \hline
1822:         1340 & 1452.1 & 0.36    \\ \hline
1823:         1345 & 1689.6 & 0.92    \\ \hline
1824: \end{tabular}
1825: \end{center}
1826: \label{table:norm}
1827: \end{table}
1828: 
1829: 
1830: \begin{table}[htbp]
1831: \caption{
1832: The most dominant VER pathways for 
1833: NMA-D in vacuum with the ab initio potential (B3LYP/6-31+G(d)).
1834: }
1835: \hfill
1836: \begin{center}
1837: \begin{tabular}{c|c|c|c}
1838: \hline
1839: Mode combination  ($\alpha, \beta$) &  frequency (cm$^{-1}$) &  Contribution to $s(t)$ &  
1840: $\Delta \omega$ (cm$^{-1}$) \\ \hline \hline           
1841:            9 +      9 & 868.7 + 868.7 & 0.13         &     2.7  \\ \hline
1842:           13 +      8 & 1144.5+ 620.0 & 0.02         &     29.8 \\ \hline
1843: \end{tabular}
1844: \end{center}
1845: \label{table:pathway2}
1846: \end{table}
1847: 
1848: \begin{table}[htbp]
1849: \caption{
1850: The most dominant VER pathways for 
1851: NMA-D in vacuum with the CHARMM force field.
1852: }
1853: \hfill
1854: \begin{center}
1855: \begin{tabular}{c|c|c|c}
1856: \hline
1857: Mode combination  ($\alpha, \beta$) &  frequency (cm$^{-1}$) & Contribution to $s(t)$ &  
1858:  $\Delta \omega$ (cm$^{-1}$) \\ \hline \hline           
1859:            8   +    8 & 741.9 +741.9 & 0.02  &  194.1   \\ \hline
1860:           14   +    6 & 1088.5+536.3 & 0.08  &  53.1   \\ \hline
1861:           14   +    8 & 1088.5+741.9 & 0.02  &  152.5   \\ \hline
1862:           15   +    7 & 1123.5+575.9 & 0.08  &  21.6   \\ \hline
1863: \end{tabular}
1864: \end{center}
1865: \label{table:pathway3}
1866: \end{table}
1867: 
1868: 
1869: 
1870: 
1871: 
1872: %\subsubsection{Comparison with ab initio result at 0K without water}
1873: 
1874: 
1875: \subsection{Dephasing properties of NMA-D}
1876: 
1877: We now consider the dephasing properties of the amide I mode.
1878: The off-diagonal density matrix is written as 
1879: \begin{equation}
1880: \rho_{10}(t)
1881: =\frac{1}{2} e^{- i \tilde{\omega}_S t} 
1882: [1 - r^{(2)}_{FF}(t)-r^{(2)}_{GG}(t)-r^{(2)}_{FG}(t)]
1883: \simeq 
1884: \frac{1}{2} e^{- i \tilde{\omega}_S t - r^{(2)}_{FF}(t)-r^{(2)}_{GG}(t)-r^{(2)}_{FG}(t)}
1885: \end{equation}
1886: and we analyze each contribution to the density matrix seperately.
1887: %Here we do not consider the effect of the fluctuations of $\tilde{\omega}_S$,
1888: %(This might affect the result. I will consider this problem later.)
1889: %ensemble averages for $r^{(2)}_{FF}(t), r^{(2)}_{GG}(t), r^{(2)}_{FG}(t)$.
1890: 
1891: In Fig.~\ref{fig:deph}, we show the result with $R_c=10$ \AA \, and $\omega_c=10$ cm$^{-1}$.
1892: We can see that the following 
1893: relation holds
1894: \begin{equation}
1895: {\rm Re} \{ r^{(2)}_{FF}(t)\} \simeq s(t)/2 \simeq t/(2 T_1).
1896: \end{equation}
1897: If we {\it further} assume that ${\rm Re} \{ r^{(2)}_{GG}(t) \}\simeq t/T_2^*$ and 
1898: ${\rm Re} \{ r^{(2)}_{FG}(t) \} \simeq 0$,
1899: we have 
1900: \begin{equation}
1901: \frac{1}{T_2}=\frac{1}{2T_1}+\frac{1}{T_2^*}.
1902: \label{eq:T2}
1903: \end{equation}
1904: This is a standard expression connecting $T_1$ and $T_2$ \cite{Mchale}, 
1905: and holds under the Markov approximation. 
1906: We can see that ${\rm Re} \{ r^{(2)}_{FG}(t) \}\simeq 0$ holds,
1907: but it is difficult to judge whether ${\rm Re} \{ r^{(2)}_{GG}(t) \} \simeq t/T_2^*$ holds or not.
1908: 
1909: 
1910: 
1911: %This indicates the difficulty to calculate dephasing properties 
1912: %using our strategy: we employed a cut off radius to reduce the system
1913: %size, but the dephasing properties are determined by the 
1914: %delocalized modes, so this implies that the convergence problem
1915: %appears.
1916: %We have checked that the result is converged for the VER calculations,
1917: %but it is rather hard to do that for the dephasing calculations.
1918: %We believe that our calculations with $R_c=10$ \AA \, should have 
1919: %some implications.
1920: %It is a future task to overcome this problem and 
1921: %to connect with the other theoretical methods to calculate dephasing \cite{SCS04}.
1922: 
1923: 
1924: \begin{figure}[htbp]
1925: \hfill
1926: \begin{center}
1927: %\begin{minipage}{.42\linewidth}
1928: \includegraphics[scale=1.2]{DEPH_fq10.eps}
1929: %\end{minipage}
1930: %\hspace{1cm}
1931: %\begin{minipage}{.42\linewidth}
1932: %\includegraphics[scale=1.2]{corr_comp.eps}
1933: %\end{minipage}
1934: %\begin{minipage}{.32\linewidth}
1935: %\includegraphics[scale=0.8]{corr_NMA.eps}
1936: %\end{minipage}
1937: %\vspace{1cm}
1938: \end{center}
1939: \caption{
1940: %{\sf 
1941: Dephasing properties of the amide I mode of NMA-D in heavy water.
1942: %}
1943: }
1944: \label{fig:deph}
1945: \end{figure}
1946: 
1947: 
1948: 
1949: 
1950: 
1951: 
1952: \begin{figure}[htbp]
1953: \hfill
1954: \begin{center}
1955: \begin{minipage}{.42\linewidth}
1956: \includegraphics[scale=1.0]{VER_comp.eps}
1957: \end{minipage}
1958: \hspace{1cm}
1959: \begin{minipage}{.42\linewidth}
1960: \includegraphics[scale=1.0]{corr_comp.eps}
1961: \end{minipage}
1962: %\begin{minipage}{.32\linewidth}
1963: %\includegraphics[scale=0.8]{corr_NMA.eps}
1964: %\end{minipage}
1965: %\vspace{1cm}
1966: \end{center}
1967: \caption{
1968: %{\sf 
1969: Left: VER (left) and frequency auto correlation 
1970: calculations (right) at 300K with different 
1971: cut-off frequencies $\omega_c$.
1972: %}
1973: }
1974: \label{fig:comp}
1975: \end{figure}
1976: 
1977: 
1978: 
1979: There are more serious problems:
1980: as mentioned by Mikami and Okazaki \cite{MO04},
1981: the diagonal terms contribute most for dephasing, i.e.,
1982: the second term in Eq.~(\ref{eq:GG}) is a dominant contribution for dephasing.
1983: Furthermore, the coefficients are 
1984: dominant factors, so this means that the low frequency (and thus delocalized) 
1985: modes contribute most. 
1986: In this paper, we have employed two cut-off parameters: $R_c$ and $\omega_c$.
1987: If $R_c$ is large enough, it is fine, but the choice of $\omega_c$ can be 
1988: arbitrary. Figure \ref{fig:comp} shows the dependence of the results 
1989: on $\omega_c$. The VER results do not depend on the choice of $\omega_c$ 
1990: because there is a resonant condition which should be met, but 
1991: the dephasing results do.
1992: We need to be cautious in the interpretation of our results for dephasing.
1993: One way to get rid of this problem is to go back to 
1994: the original expression Eq.~(\ref{eq:gg}) 
1995: using the force and force-constant autocorrelation functions
1996: $\langle \delta {\cal F}(t) \delta {\cal F}(0) \rangle$ and 
1997: $\langle \delta {\cal G}(t) \delta {\cal G}(0) \rangle$.
1998: Here $r^{(2)}_{GG}(t)$ is calculated as time integral of 
1999: these correlation functions, which can be calculated using 
2000: classical mechanics. This is in the same spirit as the 
2001: quantum correction factor method \cite{SP01}, which is 
2002: an approximation to quantum effects.
2003: In this case, we only need to consider the 
2004: zero frequency component, so the classical mechanics should work well
2005: and quantum effects should be less important.
2006: %This algorithm will be implemented elsewhere.
2007: 
2008: 
2009: 
2010: \subsection{Discussions}
2011: 
2012: %Using our formulas [Eq.~(\ref{eq:VER}) and Eq.~(\ref{eq:dephasing}) 
2013: %with Eqs.~(\ref{eq:FF})-(\ref{eq:FG})], 
2014: %we can analyze the machanisms of VER and dephasing in a molecule.
2015: %For VER problems,
2016: %as Shiga and Okazaki did, we can specify the most resonant 
2017: %two modes with the amide I mode.
2018: %It turned out that such resonant modes are strongly localized around 
2019: %peptide bond (-NH-CO-), and little contribution from the around water molecules.
2020: %This means that the VER of the amide I mode is determined from 
2021: %the intramolecular VER, and not from the intermolecular VER.
2022: %This might explain that the VER time of  most experimental data 
2023: %for a deuterated amide I mode is about 0.5 ps, i.e., 
2024: %the VER mechanism is mainly determined by the local interaction 
2025: %within the peptide bond.
2026: %Note that this is the case for a {\it localized mode} such as the 
2027: %deuterated amide I mode, and a {\it collective mode} can decay 
2028: %with a different VER pathway as shown by Austin's group \cite{XMHA00}.
2029: 
2030: 
2031: 
2032: 
2033: We found that there are several resonant modes in 
2034: NMA-D, which form the main VER pathways {\it within} the molecule.
2035: Gerber and coworkers reported that the amide I mode in NMA 
2036: is very weakly coupled to other modes \cite{GCG02}. 
2037: We expect that this discrepancy results from 
2038: (a) the use of only pair interactions between normal modes to reduce the computational cost,
2039: (b) the level of the ab initio method: they
2040: used MP2/DZP whereas we used B3LYP/6-31+G(d),
2041: and 
2042: (c) the criterion of the mode-mode coupling: 
2043: their criterion is not directly related to VER.
2044: 
2045: It is important and interesting to clarify the nature of 
2046: VER in the amide I mode in more detail. 
2047: Note the importance of the system anharmonicity.
2048: The effect of the system anharmonicity defined in Eq.~(\ref{eq:anhar}) 
2049: is very weak for the CHARMM case:
2050: $\varepsilon=10^{-5}$, but it is not for the ab initio case: 
2051: $\varepsilon=10^{-2}$. According to Eq.~(\ref{eq:freq}), 
2052: this anharmonicity shifts the system frequency by 0.6\%, which amounts 
2053: to 10 cm$^{-1}$. The resonant condition changes compared 
2054: to the case without anharmonicity. 
2055: Of course, dephasing is also affected by this amount of anharmonicity. 
2056: To address these issues, we must develop QM/MM type methods, which will be described elsewhere.
2057: Another interesting system to investigate anharmonicity is 
2058: a highly excited bond such as 
2059: the highly excited CO bond in myoglobin \cite{VFVAMJ04}. 
2060:  
2061: %other molecules like HOD in heavy water \cite{Skinner,Hynes} 
2062: %or highly excited vibrational states of CO in myoglobin \cite{VFVAMJ04}.  
2063: %It is also interesting to examine the VER and dephasing properties 
2064: %of delocalized modes like delocalized amide I mode in an alpha helix \cite{XMHA00}.
2065: 
2066: It is important to assess our strategy: 
2067: our perturbative expansion and cut-off strategy are 
2068: approximate. It would be profitable and interesting to compare this strategy 
2069: with others, including the time-dependent vibrational self-consistent field methods \cite{JG99,FYZSH},
2070: the semiclassical method \cite{Geva}, and the path integral method \cite{Krilov}.
2071: The application of our strategy to protein systems, including cytochrome c, 
2072: will be described elsewhere \cite{Matt}.
2073: 
2074: 
2075: \section{Summary}
2076: \label{sec:summary}
2077: 
2078: 
2079: In this paper,
2080: we have derived formulas for VER and dephasing for an 
2081: anharmonic (cubic) oscillator coupled to a harmonic bath 
2082: through 3rd and 4th order coupling elements.
2083: We employed time-dependent perturbation theory and did not
2084: take the infinite time limit as is done in the derivation of the Maradudin-Fein formula.
2085: Hence our formulas do not assume the Markov properties 
2086: of the system, and can describe short time behavior 
2087: that can be important for VER and dephasing properties 
2088: of localized modes in peptides or proteins.
2089: %We are also free from the problem due to the ad hoc 
2090: %``lifetime'' width parameter.
2091: Our final results are the VER formula 
2092: [Eq.~(\ref{eq:VER})] and dephasing formula 
2093: [Eq.~(\ref{eq:dephasing}) with Eqs.~(\ref{eq:FF})-(\ref{eq:FG})]. 
2094: As a test case, we have studied the amide I mode of 
2095: $N$-methylacetamide in heavy water.
2096: We found that the VER time is 0.5 ps at 300K,
2097: which is in good accord with the experimental value,
2098: and clarified that the VER mechanism is mainly 
2099: localized around the peptide bond in NMA-D;
2100: VER is dominated by IVR within the molecule.
2101: We also investigated the dephasing properties of the 
2102: amide I mode, and met with some problems.
2103: We proposed a new method to overcome these problems 
2104: using classical correlation function calculations.
2105: 
2106: 
2107: 
2108: \acknowledgments
2109: 
2110: We thank S.~Okazaki, T.~Mikami, K.~Yagi, T.~Miyadera, A.~Szabo, E.~Geva, G.~Krilov,
2111: H.-P.~Breuer, S.~Maniscalco, 
2112: F.~Romesberg and M.~Cremeens for useful discussions. We also thank the National Science 
2113: Foundation (CHE-036551) and Boston University's Center for 
2114: Computer Science for generous support to
2115: our research.
2116: 
2117: 
2118: \begin{appendix}
2119: 
2120: \section{System parameters for a cubic oscillator}
2121: \label{app:system}
2122: 
2123: We assume that the system-bath interaction can be 
2124: Taylor expanded using the bath coordinate $q_{\alpha}$,
2125: and that the fluctuating force and 
2126: the fluctuating force constant can be expressed as 
2127: \begin{eqnarray}
2128: \delta{\cal F} 
2129: &=&
2130: \sum_{\alpha,\beta}C_{S \alpha \beta} 
2131: (q_{\alpha} q_{\beta} 
2132: -\langle q_{\alpha} q_{\beta} \rangle),
2133: \\
2134: \delta{\cal G}
2135: &=&
2136: \sum_{\alpha,\beta}C_{SS \alpha \beta} 
2137: (q_{\alpha} q_{\beta}
2138: -\langle q_{\alpha} q_{\beta} \rangle)
2139: +\sum_{\alpha}C_{SS \alpha} q_{\alpha}. 
2140: %=\omega_S \Delta \omega_S(t) 
2141: \end{eqnarray}
2142: 
2143: 
2144: In real molecular systems such as peptides or proteins, the coefficients 
2145: in ${\cal V}$ and the anharmonicity parameter in ${\cal H}_f$ 
2146: are calculated as  
2147: \begin{eqnarray}
2148: C_{S \alpha \beta}
2149:  &=& 
2150: -\frac{1}{2} \frac{\partial^3 V}{\partial q_S \partial q_{\alpha} \partial q_{\beta}},
2151: \\
2152: C_{SS \alpha}
2153:  &=& 
2154: \frac{1}{2} \frac{\partial^3 V}{\partial q_S^2 \partial q_{\alpha}},
2155: \\
2156: C_{SS \alpha \beta}
2157:  &=& 
2158: \frac{1}{4} \frac{\partial^4 V}{\partial q_S^2 \partial q_{\alpha} \partial q_{\beta}},
2159: \\
2160: f
2161: &=&
2162: \frac{\partial^3 V}{\partial q_S^3}
2163: \end{eqnarray}
2164: where $V$ represents a potential function for the system considered.
2165: This potential function can be an empirical force field (CHARMM, Amber) or 
2166: an ab initio potential calculated by any level of theory.
2167: 
2168: 
2169: 
2170: Assuming that the cubic anhamonicity $f$ in the system 
2171: is small, we use the time-independent perturbation 
2172: theory to calculate the eigen energies and vectors.
2173: We quote from J.J.~Sakurai \cite{JJ}:
2174: \begin{eqnarray}
2175: E_n
2176: &=& 
2177: E^{(0)}_n+ V_{nn}+ \sum_{k \neq n} \frac{|V_{nk}|^2}{E^{(0)}_n-E^{(0)}_k},
2178: \\
2179: |n \rangle 
2180: &=& |n^{(0)} \rangle 
2181: +
2182: \sum_{k \neq n} |k^{(0)} \rangle  
2183: \frac{V_{kn}}{E^{(0)}_n-E^{(0)}_k}
2184: \nonumber
2185: \\
2186: &&
2187: +
2188: \left(
2189: \sum_{k \neq n} 
2190: \sum_{l \neq n} 
2191: |k^{(0)} \rangle  
2192: \frac{V_{kl}V_{ln}}
2193: {(E^{(0)}_n-E^{(0)}_k)(E^{(0)}_n-E^{(0)}_l)}
2194: -\sum_{k \neq n} 
2195: |k^{(0)} \rangle  
2196: \frac{V_{nn}V_{kn}}
2197: {(E^{(0)}_n-E^{(0)}_k)^2}
2198: \right)
2199: \end{eqnarray}
2200: where $E^{(0)}_n=\hbar \bar{\omega}_S (n+1/2)$,
2201: $|k^{(0)} \rangle$ is the
2202: $k$-th eigenfunction of the harmonic oscillator,
2203: and 
2204: \begin{eqnarray}
2205: V_{kn} 
2206: &=& \frac{f}{6} 
2207: \langle k^{(0)}| \bar{q}_S^3 |n^{(0)} \rangle
2208: \nonumber
2209: %\\
2210: %= \frac{f}{6} 
2211: %\left( \sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2212: %\right)^3
2213: %\langle k^{(0)}| (a+a^{\dagger})^3 |n^{(0)} \rangle
2214: %\nonumber
2215: \\
2216: &=& \hbar \bar{\omega}_S \varepsilon
2217: %\frac{f}{6} 
2218: %\left( \sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2219: %\right)^3
2220: \left[
2221: \sqrt{n(n-1)(n-2)} \delta_{k,n-3}
2222: +3n \sqrt{n} \delta_{k,n-1}
2223: \right.
2224: \nonumber
2225: \\
2226: &&
2227: \left.
2228: +3(n+1) \sqrt{n+1} \delta_{k,n+1}
2229: +\sqrt{(n+1)(n+2)(n+3)} \delta_{k,n+3}
2230: \right]
2231: \end{eqnarray}
2232: where 
2233: \begin{equation}
2234: \varepsilon=\frac{1}{\hbar \bar{\omega}_S} 
2235: \frac{f}{6} \left( \frac{\hbar}{2 \bar{\omega}_S} \right)^{3/2} 
2236: \label{eq:anhar}
2237: \end{equation}
2238: is a dimensionless paramater representing the strength of the anharmonicity of the system.
2239: Note that $V_{kn}$ becomes nonzero only when $|k-n|=1$ or $|k-n|=3$.
2240: 
2241: We explicitly have 
2242: \begin{eqnarray}
2243: E_0
2244: &=&
2245: \frac{\hbar \bar{\omega}_S}{2}- 
2246: \frac{|V_{01}|^2}{\hbar \bar{\omega}_S}
2247: -\frac{|V_{03}|^2}{3 \hbar \bar{\omega}_S}
2248: %\nonumber
2249: %\\
2250: %&=&
2251: =
2252: \frac{\hbar \bar{\omega}_S}{2}(1-22 \varepsilon^2),
2253: %\frac{\hbar \bar{\omega}_S}{2}- 
2254: %\frac{1}{\hbar \bar{\omega}_S}
2255: %\frac{f^2}{4}
2256: %\left(
2257: %\frac{\hbar}{2 \bar{\omega}_S}
2258: %\right)^3
2259: %-\frac{1}{3 \hbar \bar{\omega}_S}
2260: %\frac{f^2}{6}
2261: %\left(
2262: %\frac{\hbar}{2 \bar{\omega}_S}
2263: %\right)^3
2264: %=
2265: %\frac{\hbar \bar{\omega}_S}{2}- 
2266: %\frac{11 f^2}{36 \hbar \bar{\omega}_S}
2267: %\left(
2268: %\frac{\hbar}{2 \bar{\omega}_S}
2269: %\right)^3
2270: \\
2271: E_1
2272: &=&
2273: \frac{3 \hbar \bar{\omega}_S}{2}+ 
2274: \frac{|V_{10}|^2}{\hbar \bar{\omega}_S}
2275: -\frac{|V_{12}|^2}{\hbar \bar{\omega}_S}
2276: -\frac{|V_{14}|^2}{3 \hbar \bar{\omega}_S}
2277: =
2278: \frac{\hbar \bar{\omega}_S}{2}(3-142 \varepsilon^2).
2279: %\nonumber
2280: %\\
2281: %&=&
2282: %\frac{3 \hbar \bar{\omega}_S}{2}+ 
2283: %\frac{1}{\hbar \bar{\omega}_S}
2284: %\frac{f^2}{4}
2285: %\left(
2286: %\frac{\hbar}{2 \bar{\omega}_S}
2287: %\right)^3
2288: %-\frac{1}{\hbar \bar{\omega}_S}
2289: %2 f^2
2290: %\left(
2291: %\frac{\hbar}{2 \bar{\omega}_S}
2292: %\right)^3
2293: %-\frac{1}{3 \hbar \bar{\omega}_S}
2294: %\frac{2 f^2}{3}
2295: %\left(
2296: %\frac{\hbar}{2 \bar{\omega}_S}
2297: %\right)^3
2298: %\nonumber
2299: %\\
2300: %&=&
2301: %\frac{3 \hbar \bar{\omega}_S}{2}
2302: %- 
2303: %\frac{71 f^2}{36 \hbar \bar{\omega}_S}
2304: %\left(
2305: %\frac{\hbar}{2 \bar{\omega}_S}
2306: %\right)^3
2307: \end{eqnarray}
2308: The anharmonicity-corrected frequency is 
2309: \begin{equation}
2310: \tilde{\omega}_S=\frac{E_1-E_0}{\hbar}
2311: = 
2312: \bar{\omega}_S (1-60 \varepsilon^2).
2313: %- \frac{5 f^2 \hbar}{24 \bar{\omega}_S^4}
2314: \label{eq:freq}
2315: \end{equation}
2316: 
2317: Next we calculate the matrix elements for ${q}_S$ and ${q}_S^2$.
2318: We write the eigenfunctions:
2319: \begin{eqnarray}
2320: |0 \rangle 
2321: &=& |0^{(0)} \rangle 
2322: +
2323: \sum_{k=1,3} |k^{(0)} \rangle  
2324: \frac{V_{k0}}{E^{(0)}_0-E^{(0)}_k}
2325: %\nonumber
2326: %\\
2327: %&&
2328: +
2329: %\left(
2330: \sum_{k,l \in S_0} 
2331: |k^{(0)} \rangle  
2332: \frac{V_{kl}V_{l0}}
2333: {(E^{(0)}_0-E^{(0)}_k)(E^{(0)}_0-E^{(0)}_l)},
2334: %\right)
2335: \\
2336: |1 \rangle 
2337: &=& |1^{(0)} \rangle 
2338: +
2339: \sum_{k=0,2,4} |k^{(0)} \rangle  
2340: \frac{V_{k1}}{E^{(0)}_1-E^{(0)}_k}
2341: %\nonumber
2342: %\\
2343: %&&
2344: +
2345: %\left(
2346: \sum_{k,l \in S_1} 
2347: |k^{(0)} \rangle  
2348: \frac{V_{kl}V_{l1}}
2349: {(E^{(0)}_1-E^{(0)}_k)(E^{(0)}_1-E^{(0)}_l)}
2350: %\right)
2351: \end{eqnarray}
2352: where 
2353: $S_0$ represents $(l=1, k=2)$ or $(l=1, k=4)$ or 
2354: $(l=3, k=2)$ or $(l=3, k=4)$ or $(l=3, k=6)$, and 
2355: $S_1$ does $(l=0, k=3)$ or $(l=2, k=3)$ or 
2356: $(l=2, k=5)$ or $(l=4, k=3)$ or $(l=4, k=5)$, or $(l=4, k=7)$.
2357: Note that these eigenvectors are not normalized, so 
2358: we need to renormalize them before or after calculations.
2359: 
2360: 
2361: After some lengthy but straighforward calculations, 
2362: we have 
2363: \begin{eqnarray}
2364: (q_S)_{10}
2365: &=&
2366: \langle 1| \bar{q}_S+b |0 \rangle
2367: %=\sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2368: %\left[1+f^2 p^{(1)}_{10}\right]
2369: =(q_S)_{01}
2370: %\equiv c_{10}^{(0)}+c_{10}^{(2)} f^2
2371: %=a+a p^{(1)}_{10} f^2
2372: =a(1+22 \varepsilon^2),
2373: \label{eq:qs1}
2374: \\
2375: (q_S)_{00}
2376: &=&
2377: \langle 0| \bar{q}_S+b |0 \rangle
2378: %=b + 
2379: %\sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2380: %f p^{(1)}_{00}
2381: %\equiv b+c_{00}^{(1)} f
2382: %=b+a p^{(1)}_{00} f
2383: =b-6 a \varepsilon,
2384: \\
2385: (q_S)_{11}
2386: &=&
2387: \langle 1| \bar{q}_S+b |1 \rangle
2388: %=b+
2389: %\sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2390: %f p^{(1)}_{11}
2391: %\equiv b+c_{11}^{(1)} f
2392: %=b+a p^{(1)}_{11} f
2393: =b- 18 a \varepsilon,
2394: \\
2395: (q^2_S)_{10}
2396: &=&
2397: \langle 1| (\bar{q}_S+b)^2 |0 \rangle
2398: %=
2399: %2 b \sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2400: %\left[1+f^2 p^{(1)}_{10}\right]
2401: %+
2402: %\frac{\hbar}{2 \bar{\omega}_S}
2403: %f p^{(2)}_{10}
2404: =(q^2_S)_{01}
2405: %\nonumber
2406: %\\
2407: %\equiv d_{10}^{(0)}+d_{10}^{(1)} f+d_{10}^{(2)} f^2
2408: %=2ab +a^2 p^{(2)}_{10} f + 2 ab p^{(1)}_{10} f^2
2409: =2 ab-20 a^2 \varepsilon +44 a b \varepsilon^2,
2410: \\
2411: (q^2_S)_{00}
2412: &=&
2413: \langle 0| (\bar{q}_S+b)^2 |0 \rangle
2414: %= \frac{\hbar}{2 \bar{\omega}_S}\left[1+f^2 p^{(2)}_{00}\right]
2415: %+2b \sqrt{\frac{\hbar}{2 \bar{\omega}_S}} f p^{(1)}_{00}
2416: %+b^2
2417: %\nonumber
2418: %\\
2419: %&\equiv& 
2420: %=d_{00}^{(0)}+d_{00}^{(1)} f+d_{00}^{(2)} f^2
2421: =a^2+b^2 -12 a b \varepsilon
2422:  + 88 a^2 \varepsilon^2,
2423: \\
2424: (q^2_S)_{11}
2425: &=&
2426: \langle 1| (\bar{q}_S+b)^2 |1 \rangle
2427: %= \frac{\hbar}{2 \bar{\omega}_S}\left[3+f^2 p^{(2)}_{11}\right]
2428: %+2b \sqrt{\frac{\hbar}{2 \bar{\omega}_S}} f p^{(1)}_{11}
2429: %+b^2
2430: %\nonumber
2431: %\\
2432: %&\equiv& 
2433: %d_{11}^{(0)}+d_{11}^{(1)} f+d_{11}^{(2)} f^2
2434: =3 a^2+b^2 
2435: -36 ab \varepsilon
2436: +568 a^2 \varepsilon^2
2437: \label{eq:qs2}
2438: \end{eqnarray}
2439: where 
2440: \begin{eqnarray}
2441: a
2442: =
2443: \sqrt{\frac{\hbar}{2 \bar{\omega}_S}}
2444: \end{eqnarray}
2445: is the fundamental length charactering the system oscillator.
2446: 
2447: 
2448: \section{The coefficients used in the formulas}
2449: \label{app:coef}
2450: 
2451: Using the expression derived previously for the 
2452: force-force correlation function \cite{FBS05a}, 
2453: the coefficients in our VER and dephasing formulas 
2454: are expressed as 
2455: \begin{eqnarray}
2456: \mathbf{C}^{\alpha \beta}  
2457: &=&
2458: \left(
2459: \begin{array}{cc}
2460: C^{\alpha \beta}_{--} & C^{\alpha \beta}_{+-} \\
2461: C^{\alpha \beta}_{+-} & C^{\alpha \beta}_{++} \\
2462: \end{array}
2463: \right)
2464: =
2465: \left \{ 
2466: (q_S)_{10} C_{S \alpha \beta} -(q^2_S)_{10} C_{SS \alpha \beta}
2467: \right \}^2 
2468: \mathbf{S}^{\alpha \beta},
2469: \\
2470: \mathbf{D}^{R \alpha \beta}  
2471: &=&
2472: \left(
2473: \begin{array}{cc}
2474: D^{R \alpha \beta}_{--} & D^{R\alpha \beta}_{+-} \\
2475: D^{R \alpha \beta}_{+-} & D^{R \alpha \beta}_{++} \\
2476: \end{array}
2477: \right)
2478: \nonumber
2479: \\
2480: &=&
2481: \left \{
2482: [(q_S)_{11}-(q_S)_{00}] C_{S \alpha \beta} -[(q^2_S)_{11}-(q^2_S)_{00}] C_{SS \alpha \beta}
2483: \right \}^2 
2484: \mathbf{S}^{\alpha \beta},
2485: \\
2486: \mathbf{D}^{I \alpha \beta}  
2487: &=&
2488: \left(
2489: \begin{array}{cc}
2490: D^{I \alpha \beta}_{--} & D^{I \alpha \beta}_{+-} \\
2491: D^{I \alpha \beta}_{+-} & D^{I \alpha \beta}_{++} \\
2492: \end{array}
2493: \right)
2494: \nonumber
2495: \\
2496: &=&
2497: \left \{
2498: [(q_S)_{11}-(q_S)_{00}] C_{S \alpha \beta} -[(q^2_S)_{11}-(q^2_S)_{00}] C_{SS \alpha \beta}
2499: \right \} 
2500: \nonumber
2501: \\
2502: &&
2503: \times
2504: \left \{
2505: [(q_S)_{11}+(q_S)_{00}] C_{S \alpha \beta} -[(q^2_S)_{11}+(q^2_S)_{00}] C_{SS \alpha \beta}
2506: \right \} 
2507: \mathbf{S}^{\alpha \beta},
2508: \\
2509: \mathbf{E}^{\alpha \beta}  
2510: &=&
2511: \left(
2512: \begin{array}{cc}
2513: E^{\alpha \beta}_{--} & E^{\alpha \beta}_{+-} \\
2514: E^{\alpha \beta}_{+-} & E^{\alpha \beta}_{++} \\
2515: \end{array}
2516: \right)
2517: \nonumber
2518: \\
2519: &=&
2520: \left \{
2521: [(q_S)_{11}-(q_S)_{00}] C_{S \alpha \beta} -[(q^2_S)_{11}-(q^2_S)_{00}] C_{SS \alpha \beta}
2522: \right \} 
2523: \nonumber
2524: \\
2525: &&
2526: \times
2527: \left \{
2528: (q_S)_{10} C_{S \alpha \beta} -(q^2_S)_{10} C_{SS \alpha \beta}
2529: \right \} 
2530: \mathbf{S}^{\alpha \beta},
2531: \\
2532: \mathbf{S}^{\alpha \beta}
2533: &=&
2534: \frac{\hbar^2}{2 \omega_{\alpha} \omega_{\beta}}
2535: \left(
2536: \begin{array}{cc}
2537: (1+n_{\alpha})(1+n_{\beta}) & 2 (1+n_{\alpha})n_{\beta} \\ 
2538: 2 (1+n_{\alpha}) n_{\beta}  &  n_{\alpha} n_{\beta} \\
2539: \end{array}
2540: \right),
2541: \\
2542: \mathbf{C}^{\alpha}
2543: &=&
2544: \left(
2545: \begin{array}{c}
2546: C^{\alpha}_{-} \\
2547: C^{\alpha}_{+} \\
2548: \end{array}
2549: \right)
2550: =
2551: (q_S^2)_{10}^2 C_{SS \alpha}^2 
2552: \mathbf{R}^{\alpha},
2553: \\
2554: \mathbf{D}^{R\alpha}
2555: &=&
2556: \left(
2557: \begin{array}{c}
2558: D^{R\alpha}_{-} \\
2559: D^{R\alpha}_{+} \\
2560: \end{array}
2561: \right)
2562: =
2563: [(q_S^2)_{11} - (q_S^2)_{00}]^2 
2564: C_{SS \alpha}^2 
2565: \mathbf{R}^{\alpha},
2566: \\
2567: \mathbf{D}^{I\alpha}
2568: &=&
2569: \left(
2570: \begin{array}{c}
2571: D^{I\alpha}_{-} \\
2572: D^{I\alpha}_{+} \\
2573: \end{array}
2574: \right)
2575: =
2576: [(q_S^2)_{11} - (q_S^2)_{00}][(q_S^2)_{11} + (q_S^2)_{00}]
2577: C_{SS \alpha}^2 
2578: \mathbf{R}^{\alpha},
2579: \\
2580: \mathbf{E}^{\alpha}
2581: &=&
2582: \left(
2583: \begin{array}{c}
2584: E^{\alpha}_{-} \\
2585: E^{\alpha}_{+} \\
2586: \end{array}
2587: \right)
2588: =
2589: [(q_S^2)_{11} - (q_S^2)_{00}](q_S^2)_{10}
2590: C_{SS \alpha}^2 
2591: \mathbf{R}^{\alpha},
2592: \\
2593: \mathbf{R}^{\alpha}
2594: &=&
2595: \frac{\hbar}{2 \omega_{\alpha}}
2596: \left(
2597: \begin{array}{c}
2598: 1 + n_{\alpha} \\
2599:  n_{\alpha} \\
2600: \end{array}
2601: \right)
2602: \end{eqnarray}
2603: where $n_{\alpha}=1/(e^{\beta \hbar \omega_{\alpha}}-1)$ is 
2604: the thermal phonon number.
2605: 
2606: To calculate $\bar{\omega}_S$ and $b$ in Eqs.~(\ref{eq:omegas}) and (\ref{eq:qs}),
2607: we use the following
2608: \begin{eqnarray}
2609: \langle {\cal F}(t) \rangle
2610: &=& \langle {\cal F}(0) \rangle
2611: = \frac{\hbar}{2} \sum_{\alpha} 
2612: \frac{C_{S \alpha \alpha}}{\omega_{\alpha}}
2613: (1+ 2 n_{\alpha}), 
2614: \\
2615: \langle {\cal G}(t) \rangle
2616: &=& \langle {\cal G}(0) \rangle
2617: = \frac{\hbar}{2} \sum_{\alpha} 
2618: \frac{C_{S S \alpha \alpha}}{\omega_{\alpha}}
2619: (1+ 2 n_{\alpha}). 
2620: \end{eqnarray}
2621: 
2622: 
2623: %\section{Vibrational frequencies of NMA in vacuum}
2624: 
2625: \end{appendix}
2626: 
2627: 
2628: \begin{thebibliography}{99}
2629: 
2630: \bibitem{FS05}
2631: See, e.g.,
2632: H.~Fujisaki and J.E.~Straub,
2633: %``Vibrational energy relaxation in proteins,''
2634: Proc.~Natl.~Acad.~Sci.~USA {\bf 102}, %6726-6731 
2635: 6726 (2005) and references therein; e-print q-bio.BM/0412048.
2636: 
2637: \bibitem{Oxtoby}
2638: D.W.~Oxtoby, 
2639: Adv.~Chem.~Phys.~{\bf 40}, 1 (1979);
2640: ibid.~{\bf 47}, 487 (1981).
2641: 
2642: \bibitem{Hynes}
2643: R.~Rey, K.B.~Moller, and J.T.~Hynes,
2644: Chem.~Rev.~{\bf 104}, 1915 (2004);
2645: K.B.~Moller, R.~Rey, and J.T.~Hynes,
2646: J.~Phys.~Chem.~A {\bf 108}, 1275 (2004).
2647: 
2648: \bibitem{Skinner}
2649: C.P.~Lawrence and J.L.~Skinner,
2650: J.~Chem.~Phys.~{\bf 117}, 5827 (2002);
2651: ibid.~{\bf 117}, 8847 (2002);
2652: ibid.~{\bf 118}, 264 (2003); 
2653: A.~Piryatinski, C.P.~Lawrence, and J.L.~Skinner, 
2654: ibid.~{\bf 118}, 9664 (2003);
2655: ibid.~{\bf 118}, 9672 (2003);
2656: C.P.~Lawrence and J.L.~Skinner,
2657: ibid.~{\bf 119}, 1623 (2003);
2658: ibid.~{\bf 119}, 3840 (2003).
2659: 
2660: 
2661: \bibitem{Straub}
2662: D.E.~Sagnella and J.E.~Straub,
2663: Biophys.~J.~{\bf 77}, 70 (1999); 
2664: L.~Bu and J.E.~Straub,
2665: ibid.~{\bf 85}, 1429 (2003).
2666: 
2667: \bibitem{Leitner05}
2668: D.M.~Leitner,
2669: %``Heat transport in molecules and reaction kinetics:
2670: %The role of quantum energy flow and localization,''
2671: Adv.~Chem.~Phys.~{\bf 130}B, %205-256 
2672: 205 (2005).
2673: 
2674: \bibitem{FBS05a}
2675: H.~Fujisaki, L.~Bu, and J.E.~Straub,
2676: %``Vibrational energy relaxation of a CD stretching mode in cytochrome c,''
2677: Adv.~Chem.~Phys.~{\bf 130}B, %179-203 
2678: 179 (2005); 
2679: e-print q-bio.BM/0403019. 
2680: 
2681: \bibitem{FBS05b}
2682: H.~Fujisaki, L.~Bu, and J.E.~Straub,
2683: %``Probing vibrational energy relaxation in proteins using normal modes,''
2684: in {\it Normal Mode Analysis: Theory and Applications to Biological 
2685: and Chemical Systems}, edited by Q. Cui and I. Bahar (2005);
2686: e-print q-bio.BM/0408023.
2687: 
2688: \bibitem{Leitner02}
2689: D.M.~Leitner,
2690: Chem.~Phys.~Lett.~{\bf 359}, %434-439 
2691: 434 (2002).
2692: 
2693: \bibitem{MO04}
2694: T.~Mikami and S.~Okazaki, 
2695: %``Path integral influence functional theory of dynamics of 
2696: %coherence between vibrational states of solute in condensed phase,'' 
2697: J.~Chem.~Phys.~{\bf 121}, %10052-10064 
2698: 10052 (2004). 
2699: 
2700: \bibitem{Weiss}
2701: U.~Weiss, {\it Quantum Dissipative Systems}, 
2702: 2nd Ed. World Scientific Publishing, 
2703: Singapore (1999).
2704: 
2705: \bibitem{XYK02}
2706: R.~Xu, Y.J.~Yan, and O.~K\"uhn,
2707: %Quantum chemical construction of a reduced reaction Hamiltonian 
2708: %and T-1-relaxation and pure T-2-dephasing rates for the proton transfer in 3-chlorotropolone 
2709: Europhys.~J.~D {\bf 19}, %293-305 
2710: 293 (2002).
2711: 
2712: 
2713: \bibitem{VFVAMJ04}
2714: C.~Ventalon, J.M.~Fraser, M.H.~Vos, A.~Alexandrou, J.L.~Martin, and M.~Joffre,
2715: %``Coherent vibrational climbing in carboxyhemoglobin,''
2716: Proc.~Natl.~Acad.~Sci.~USA {\bf 101}, %13216-13220 
2717: 13216 (2004);
2718: O.~K\"uhn,
2719: %``Multidimensional vibrational quantum dynamics of CO-heme compounds: 
2720: %ultrafast IVR mediated Fe-CO bond-breaking after CO excitation?''
2721: Chem.~Phys.~Lett.~{\bf 402}, %48-53 
2722: 48 (2005).
2723: 
2724: 
2725: \bibitem{FMT03}
2726: %A.~Tanaka, H.~Fujisaki, and T.~Miyadera,
2727: %Phys.~Rev.~E.~{\bf 66}, 045201(R) (2002); 
2728: H.~Fujisaki, T.~Miyadera, and A.~Tanaka,
2729: Phys.~Rev.~E.~{\bf 67}, 066201 (2003).  
2730: %H.~Fujisaki, A.~Tanaka, and T.~Miyadera,
2731: %J.~Phys.~Soc.~Jpn.~Suppl.~ C {\bf 72}, 111 (2003); 
2732: %e-print quant-ph/0302015.
2733: 
2734: 
2735: 
2736: \bibitem{GCG02}
2737: S.K.~Gregurick, G.M.~Chaban, and R.B.~Gerber, 
2738: J.~Phys.~Chem.~A {\bf 106}, 8696 (2002). 
2739: 
2740: \bibitem{NS03}
2741: P.H.~Nguyen and G.~Stock,
2742: %``Nonequilibrium molecular-dynamics study of the vibrational 
2743: %energy relaxation of peptides in water,'' 
2744: J.~Chem.~Phys.~{\bf 119}, 11350 (2003).
2745: 
2746: \bibitem{SCS04}
2747: J.R.~Schmidt, S.A.~Corcelli, and J.L.~Skinner,
2748: %``Ultrafast vibrational spectcroscopy of water and aqueous N-methylacetamide:
2749: %Comparison of different electronic structure/molecular dynamics approaches,''
2750: J.~Chem.~Phys.~{\bf 121}, %8887-8896 
2751: 8887 (2004). 
2752: 
2753: \bibitem{ZAH01}
2754: M.T.~Zanni, M.C.~Asplund, and R.M.~Hochstrasser,
2755: %``Two-dimensional heterodyned and stimulated infrared photon echoes of N-methylacetamide-D,''
2756: J.~Chem.~Phys.~{\bf 114}, %4579-4590
2757: 4579 (2001). 
2758: 
2759: 
2760: \bibitem{WPHMKS02}
2761: S.~Woutersen, R.~Pfister, P.~Hamm, 
2762: Y.~Mu, D.S.~Kosov, and G.~Stock,
2763: %``Peptide conformational heterogeneity revealed from nonlinear vibrational 
2764: %spectroscopy and molecular-dynamics simulations,''
2765: J.~Chem.~Phys.~{\bf 117}, %6833-6840 
2766: 6833 (2002). 
2767: 
2768: \bibitem{CHARMM}
2769: B.R.~Brooks, R.E.~Bruccoleri, B.D.~Olafson, D.J.~States, S.~Swaminathan, and M.~Karplus,
2770: %``CHARMM: A Program for Macromolecular Energy, 
2771: %Minimization, and Dynamics Calculations,'' 
2772: J.~Comp.~Chem.~{\bf 4}, 187 (1983);
2773: A.D.~MacKerell, Jr., B.~Brooks, C.L.~Brooks III, L.~Nilsson, B.~Roux, Y.~Won, and M.~Karplus,
2774: %``CHARMM: The Energy Function and Its Parameterization with an Overview of the Program,''
2775: in {\it The Encyclopedia of Computational Chemistry}, {\bf 1}, 271, 
2776: edited by P.v.R.~Schleyer {\it et al}., John Wiley \& Sons: Chichester (1998).
2777: 
2778: 
2779: \bibitem{Keyes}
2780: M.~Cho, G.R.~Fleming, S.~Saito, I.~Ohmine, and R.M.~Stratt,
2781: J.~Chem.~Phys.~{\bf 100}, %6672-6683 
2782: 6672 (1994); 
2783: T.~Keyes, 
2784: J.~Phys.~Chem.~A {\bf 101}, 2921 (1997).
2785: 
2786: \bibitem{Breuer}
2787: H.-P.~Breuer and F.~Petruccione,
2788: {\it The Theory of Open Quantum Systems},
2789: Oxford University Press, New York (2002);
2790: %Time-local master equations: Influence functional and cumulant expansion
2791: %H. P. Breuer, A. Ma, F. Petruccione
2792: in {\it Quantum computing and quantum bits in mesoscopic systems},
2793: edited by A.~Leggett, B.~Ruggiero and P.~Silvestrini, 
2794: 263-271, Kluwer Academic/Plenum Publishers, New York (2004).
2795: 
2796: \bibitem{IMM03}
2797: F.~Intravaia, S.~Maniscalco, and A.~Messina,
2798: Eur.~Phys.~J.~B {\bf 32}, %97-107 
2799: 97 (2003).
2800: 
2801: 
2802: \bibitem{Yagi}
2803: %\bibitem{YHTSG04}
2804: K.~Yagi, K.~Hirao, T.~Taketsugu, M.W.~Schmidt, and M.S.~Gordon,
2805: J.~Chem.~Phys.~{\bf 121}, %1383-1389 
2806: 1383 (2004).
2807: 
2808: \bibitem{GK92}
2809: H.~Guo and M.~Karplus,
2810: J.~Phys.~Chem.~{\bf 96}, %7273-7287 
2811: 7273 (1992).
2812: 
2813: %\bibitem{MNAFGMLF03}
2814: %K.A.~Merchant, W.G.~Noid, R.~Akiyama, I.J.~Finkelstein,
2815: %A.~Goun, B.L.~McClain, R.F.~Loring, and M.D.~Fayer,
2816: %``Myoglobin-CO substate structures and dynamics:
2817: %Multidimensional vibrational echoes and molecular 
2818: %dynamics simulations,''
2819: %J.~Am.~Chem.~Soc.~{\bf 125}, 13804 (2003).
2820: 
2821: %\bibitem{RH96}
2822: %R.~Rey and J.T.~Hynes,
2823: %``Vibrational energy relaxation of HOD in liquid D$_2$O,''
2824: %J.~Chem.~Phys.~{\bf 104}, 2356 (1996).
2825: 
2826: %\bibitem{WWH92}
2827: %R.E.~Whitnell, K.R.~Wilson, and J.T.~Hynes, 
2828: %``Vibrational relaxation of a dipolar molecule in water,'' 
2829: %J.~Chem.~Phys.~{\bf 96}, 5354-5369 (1992). 
2830: 
2831: 
2832: %\bibitem{SSJLA99}
2833: %D.E.~Sagnella, J.E.~Straub, T.A.~Jackson, M.~Lim, and P.A.~Anfinrud,
2834: %``Vibrational population relaxation of carbon monoxide in the 
2835: %heme pocket of photolyzed carbonmonoxy myoglobin:
2836: %Comparison of time-resolved mid-IR absorbance experiments and molecular
2837: %dynamics simulations,''
2838: %Proc.~Natl.~Acad.~Sci.~USA {\bf 96}, 14324-14329 (1999). 
2839: 
2840: %\bibitem{SS99}
2841: %D.E.~Sagnella and J.E.~Straub,
2842: %``A study of vibrational relaxation of B-state carbon monoxide in 
2843: %the heme pocket of photolyzed carboxymyglobin,''
2844: %Biophys.~J.~{\bf 77}, 70-84 (1999).
2845: 
2846: %\bibitem{ME98}
2847: %J.~Meller and R.~Elber,
2848: %``Computer simulation of carbon monoxide photodissociation 
2849: %in myoglobin: Structural interpretation of the B state,''
2850: %Biophys.~J.~{\bf 74}, 789-802 (1998).
2851: 
2852: %\bibitem{HSA04}
2853: %G.~Hummer, F.~Schotte, and P.A.~Anfinrud,
2854: %``Unveiling functional protein motions with picosecond
2855: %x-ray crystrallography and molecular dynamics simulations,''
2856: %Proc.~Natl.~Acad.~Sci.~USA {\bf 101}, 15330-15334 (2004). 
2857: 
2858: 
2859: \bibitem{Kidera}
2860: K.~Moritsugu, O.~Miyashita, and A.~Kidera,
2861: Phys.~Rev.~Lett.~{\bf 85}, 3970 (2000).
2862: 
2863: 
2864: \bibitem{MKFKAZ04}
2865: P.~Mukherjee, A.T.~Krummel, E.C.~Fulmer, I.~Kass, I.T.~Arkin, and M.T.~Zanni,
2866: %``Site-specific vibrational dynamics of the CD3 zeta membrane peptide 
2867: %using heterodyned two-dimensional infrared photon echo spectroscopy,''
2868: J.~Chem.~Phys.~{\bf 120}, 10215 (2004).  
2869: 
2870: \bibitem{XMHA00}
2871: A. Xie, L. van der Meer, W. Hoff, and R.H. Austin,
2872: %``Long-lived amide I vibrational mode in myoglobin,''
2873: Phys.~Rev.~Lett.~{\bf 84}, %5435-5438.
2874: 5435 (2000). 
2875: 
2876: \bibitem{Mchale}
2877: J.L.~McHale,
2878: {\it Molecular Spectroscopy},
2879: Prentice-Hall, Inc. (1999).
2880: 
2881: \bibitem{SP01}
2882: J.L.~Skinner and K.~Park,
2883: %``Calculation vibrational energy relaxation rates from classical molecular 
2884: %dynamics simulations: Quantum correction factors for proceeses involving
2885: %vibration-vibration energy transfer,''
2886: J.~Phys.~Chem.~B {\bf 105}, %6716-6721 
2887: 6716 (2001). 
2888: 
2889: 
2890: %\bibitem{PPBZ02}
2891: %S.K. Pal, J. Peon, B. Bagchi, and A.H. Zewail,
2892: %``Biological water: Femtosecond dynamics of macromolecular hydration,''
2893: %J. Phys. Chem. B {\bf 106} (2002) 12376-12395.
2894: 
2895: %\bibitem{SPASIBA}
2896: %P.~Lagamt. D.~Nolde, R.~Stote, G.~Vergoten, and M.~Karplus,
2897: %``Increasing normal modes analysis accuracy: The SPASIBA spectroscopic force 
2898: %field introduced into the CHARMM program,'' 
2899: %J.~Phys.~Chem.~A {\bf 108}, 4019-4029 (2004).
2900: 
2901: 
2902: %\bibitem{GKS05}
2903: %Roman D. Gorbunov, Daniil S. Kosov, and Gerhard Stock,
2904: %``Ab initio-based exciton model of amide I vibrations in peptides: 
2905: %Definition, conformational dependence, and transferability,''
2906: %J. Chem. Phys. {\bf 122}, 224904-1-12 (2005).
2907: 
2908: 
2909: %\bibitem{Yong}
2910: %If we use B3LYP/6-31+G(d) theory, 
2911: %the DFT value can be 1722 cm$^{-1}$, which 
2912: %is closer to the experimental value than B3LYP/6-311+G(d,p).
2913: %{\bf Is this right $\rightarrow$ Yong?}
2914: %We thank Yong Zhang for noting this fact.
2915: 
2916: \bibitem{JG99}
2917: P.~Jungwirth and R.B.~Gerber,
2918: Chem.~Rev.~{\bf 99}, 1583 (1999);
2919: C.~Iung, F.~Gatti, H.-D.~Meyer,
2920: J.~Chem.~Phys.~{\bf 120}, 6992 (2004);
2921: F.~Gatti and H-D.~Meyer,
2922: Chem.~Phys.~{\bf 304}, 3 (2004).
2923: 
2924: \bibitem{FYZSH}
2925: H.~Fujisaki, K.~Yagi, Y.~Zhang, J.E.~Straub, and K.~Hirao,
2926: unpublished.
2927: 
2928: 
2929: \bibitem{Geva}
2930: Q.~Shi and E.~Geva,
2931: %``Semiclassical theory of vibrational energy relaxation in the condensed phase,''
2932: J.~Phys.~Chem.~A {\bf 107}, 9059 (2003);
2933: %``Vibrational energy relaxation in liquid oxgen from a semiclassical molecular
2934: %dynamics simulation,''
2935: ibid.~{\bf 107}, 9070 (2003);
2936: %\bibitem{SG03b}
2937: %Q.~Shi and E.~Geva,
2938: %``On the calculation of vibrational energy relaxation rate constants from 
2939: %centroid molecular dynamics simulations,''
2940: J.~Chem.~Phys.~{\bf 119}, 9030 (2003).
2941: 
2942: \bibitem{Krilov}
2943: G.~Krilov, E.~Sim, and B.J.~Berne,
2944: Chem.~Phys.~{\bf 268}, %21-34 
2945: 21 (2001);
2946: E.~Rabani, D.R.~Reichman, G.~Krilov, and B.J.~Berne,
2947: Proc.~Natl.~Acad.~Sci.~USA {\bf 99}, 1129 (2002).
2948: 
2949: \bibitem{Matt}
2950: M.~Cremeens, H.~Fujisaki, Y.~Zhang, J.~Zimmermann, L.B.~Sagle,
2951: S.~Matsuda, P.E.~Dawson, J.E.~Straub, and F.E.~Romesberg,
2952: unpublished. 
2953: 
2954: \bibitem{JJ}
2955: J.J.~Sakurai, {\it Modern Quantum Mechanics}, 
2956: Benjamin/Cummings,
2957: 2nd edition (1994).
2958: 
2959: 
2960: \end{thebibliography}
2961: 
2962: 
2963: 
2964: 
2965: \end{document}
2966: 
2967: 
2968: 
2969: