1: % For submission to J. Chem. Phys.
2: % (Last update: 02 24 04)
3: %
4: % Title:
5: % Heavy atom tunneling in chemical reactions:
6: % study of $\mbox{H}+\mbox{LiF}$ collisions
7: %
8: % Authors:
9: % P. F. Weck and N. Balakrishnan
10: %
11: %
12: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb,aps]{revtex4}
13: \documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb,aps]{revtex4}
14:
15: % Some other (several out of many) possibilities
16: %\documentclass[preprint,aps]{revtex4}
17: %\documentclass[preprint,aps,draft]{revtex4}
18: %\documentclass[prb]{revtex4}% Physical Review B
19:
20: \usepackage{graphicx}% Include figure files
21: \usepackage{dcolumn}% Align table columns on decimal point
22: \usepackage{bm}% bold math
23: \usepackage{epsfig}
24:
25: %\newcommand{\ro}{\bm{\rho}}
26:
27: %\nofiles
28:
29: %===============
30: \begin{document}
31: %===============
32:
33:
34: %\preprint{APS/123-XXX}
35:
36: \title{Heavy atom tunneling in chemical reactions: study of $\mbox{H}+\mbox{LiF}$ collisions}
37:
38: \author{P. F. Weck}
39: \email{weckp@unlv.nevada.edu}
40: \author{N. Balakrishnan}
41: \email{naduvala@unlv.nevada.edu}
42:
43: \affiliation{Department of Chemistry, University of Nevada Las Vegas,
44: 4505 Maryland Parkway, Las Vegas, NV 89154, USA}
45:
46: \date{\today}
47:
48: %===============
49: \begin{abstract}
50:
51: The $\mbox{H}+\mbox{LiF}(X ^1\Sigma^+,v=0-2,j=0)\rightarrow\mbox{HF}(X ^1\Sigma^+,v',j')+\mbox{Li}(^2S)$
52: bimolecular process is investigated by means of quantum scattering calculations on the chemically accurate
53: $X~^2A'$ LiHF potential energy surface of \citeauthor{agua03} [J. Chem. Phys. \textbf{119}, 10088 (2003)].
54: Calculations have been performed for zero total angular momentum for translational energies from $10^{-7}$
55: to $10^{-1}~\mbox{eV}$.
56: Initial-state selected reaction probabilities and cross sections are characterized
57: by resonances originating from the decay of metastable states of the
58: $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ and $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ van der Waals complexes.
59: Extensive assignment of the resonances has been carried out by performing quasibound states calculations
60: in the entrance and exit channel wells. Chemical reactivity is found to be significantly enhanced by
61: vibrational excitation at low temperatures, although reactivity appears much less favorable
62: than non-reactive processes due to the inefficient tunneling of the relatively heavy fluorine atom
63: strongly bound in van der Waals complexes.
64:
65: \end{abstract}
66: %=============
67:
68: \pacs{33.70.-w}% PACS, the Physics and Astronomy
69: % Classification Scheme.
70: \maketitle
71:
72: %=======================================
73: \section{Introduction \label{sec:intro}}
74: %=======================================
75:
76: Quantum mechanical tunneling through an energy barrier has long been recognized as an
77: important process in a variety of chemical reactions \cite{bell80}.
78: Recently, evidences for tunneling in reactions have been reported, for instance, in
79: enzymology \cite{gao02}, organic \cite{zuev03}, atmospheric \cite{sulta04},
80: and interstellar chemistry \cite{hira01}.
81: This mechanism is also being investigated at cold and ultracold temperatures
82: in the context of chemical reactivity involving Bose-Einstein condensates (BEC)
83: of diatomic molecules \cite{joc03,gre03,zwi03} where tunneling is thought to be the
84: dominant reactive process \cite{bala01,bodo02,bala03,weck04}.
85:
86: Analysis of subthreshold resonances
87: in the energy dependence of reaction cross sections can provide important
88: insight into the reaction mechanism as illustrated recently for FH$_2$ \cite{taka98,taka99},
89: HOCl \cite{zou01,xie02}, HCl$_2$\cite{bowman05} and LiHF \cite{pania99,wei03,weck04b}.
90: For these systems, resonances were attributed to the decay of quasibound van der Waals
91: complexes in the entrance and/or exit channels of the reaction.
92: %
93: The LiHF system is especially interesting since at
94: low temperatures where tunneling is preponderant, the reaction pathway
95: between the $\mbox{H}+\mbox{LiF}$ and $\mbox{Li}+\mbox{HF}$ arrangements involves
96: the transfer of the relatively heavy F atom.
97: The $\mbox{LiH}+\mbox{F}$ product channel is highly endoergic and is not accessible
98: at low energies.
99: %
100: The LiHF system has been extensively studied experimentally and theoretically.
101: Indeed, after the original crossed beam measurement of \citet{tayl55}, the
102: $\mbox{Li}+\mbox{HF}$ reaction became a prototype for experimental
103: investigation of the ``harpoon'' mechanism \cite{hers66} and many measurements of the
104: integral and differential reactive cross sections have been
105: reported \cite{beck80,loes93,baer94,aoiz99,aoiz00,casa00,hobe01,aoiz01,hobe04}.
106: %
107: Moreover, LiHF is particularly amenable to {\em ab initio} calculations of chemical accuracy,
108: therefore explaining the variety of potential energy surfaces (PESs) available for the $X^2A'$ symmetry
109: electronic ground state \cite{zeir78,chen80,cart80,laga84,garc84,
110: pani89,palm89,suar94,agua95,agua97a,agua97b,laga98a,laga98b,burc00,jasp01,jasp02,
111: burc02,agua03}. A large number of quantum mechanical
112: \cite{baer94,park95,gogt96,agua97a,agua97b,lara98,pani98,aoiz99,aoiz00,lara00,
113: aoiz01,wei03,laga04,weck04b} and classical trajectory \cite{aoiz00,aoiz00b} scattering
114: calculations have been performed on these PESs for the $\mbox{Li}+\mbox{HF}$ collision.
115: In contrast, the $\mbox{H}+\mbox{LiF}$ collision has received no theoretical
116: and experimental attention, to the best of our knowledge. The lack of experimental results
117: may be explained by the difficulty in working with metastable hydrogen atom sources in
118: crossed beams experiments.
119:
120: In this work, we present quantum scattering calculations
121: on the $X~^2A'$ LiHF electronic ground state for the
122: $\mbox{H}+\mbox{LiF}(X ^1\Sigma^+,v=0-2,j=0)\rightarrow\mbox{HF}(X ^1\Sigma^+,v',j')+\mbox{Li}(^2S)$
123: reaction.
124: Calculations have been performed for zero total angular momentum using the recent
125: high accuracy global PES of \citet{agua03}.
126: In the context of our study of the reverse reaction \cite{weck04b},
127: particular effort is made here to assign resonances due to the decay
128: of metastable states in van der Waals wells of the entrance and exit channels.
129: The LiHF electronic ground state is typical of alkali metal
130: halides with a saddle point resulting from the crossing of
131: the $\mbox{Li}^++\mbox{HF}^-$ ionic state and a covalent configuration
132: correlating to $\mbox{Li}(^2S)+\mbox{HF}(X ^1\Sigma^+)$. The strong
133: dipole electric field of the products is at the origin of an exceptionally deep
134: van der Waals minimum of about $0.24~\mbox{eV}$ for the
135: $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ complex, while the
136: $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ complex in the entrance channel is about
137: $0.07~\mbox{eV}$ deep.
138: These van der Waals wells give rise to long-lived collision complexes and
139: narrow scattering resonances in the energy dependence of reaction probabilities.
140: Since $\mbox{H}+\mbox{LiF}\rightarrow\mbox{HF}+\mbox{Li}$ involves the transfer of
141: the relatively heavy F atom, it will be of particular interest to see
142: whether the reaction will occur with significant rate coefficient at low
143: energies and how resonances due to tunneling affect reaction rates.
144:
145: In Sec. II, we give a summary of the principal features of the PES
146: of \citet{agua03}, followed by a concise review of quantum scattering theory
147: and convergence tests assessing the validity of our numerical results.
148: Initial-state-selected probabilities, cross sections, cumulative reaction probabilities
149: as well as rate coefficients for both reactive and non-reactive open channels of the
150: collision are presented in Sec. III. We also provide a detailed analysis of the resonances present
151: in the energy depencence of cross sections by means of quasibound state calculations
152: of the van der Waals complexes formed in the reagent and product channels.
153: The effect of vibrational excitation on chemical reactivity at low temperatures and the
154: contribution of resonances in reaction rates are discussed. Finally, a
155: summary of our findings is given in Sec. IV.
156:
157:
158: %=====================
159: \section{Calculations} \label{sec:1}
160: %=====================
161:
162: %---------------------------------------------------------
163: \subsection{$X^2A'$ LiHF potential energy surface }
164: %---------------------------------------------------------
165:
166: The $X^2A'$ LiHF ground state PES of \citet{agua03} was used for the
167: present scattering calculations.
168: {\em Ab initio} calculations were performed for 6000 nuclear geometries at the
169: multireference configuration interaction (MRCI) level of theory, using internally
170: contracted wave functions including all single and double excitations and the Davidson
171: correction (+Q).
172: The $\mbox{Li}^++\mbox{HF}^-$ and $\mbox{Li}^++\mbox{H}^-\mbox{F}$ ionic configurations,
173: responsible for the curve crossing leading to the LiF products in the
174: adiabatic $X^2A'$ state, were adequately described using
175: a large atomic basis set. The crossing between the $\mbox{Li}^++\mbox{HF}^-$ ionic state
176: and a covalent configuration correlating to $\mbox{Li}(^2S)+\mbox{HF}(X ^1\Sigma^+)$ was
177: found to produce a pronounced saddle point in the Born-Oppenheimer electronic ground
178: state. The resulting analytic PES constructed using the modified many-body expansion
179: of \citet{agua92} exhibits deep van der Waals wells of
180: $0.2407~\mbox{eV}~(5.551~\mbox{kcal/mol})$
181: and $0.0686~\mbox{eV}~(1.582~\mbox{kcal/mol})$ corresponding to the
182: $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ and $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ complexes,
183: respectively. The saddle point and the $\mbox{Li}(^2S)+\mbox{HF}$ products asymptote
184: are at $+0.0642~\mbox{eV}$ and $-0.1861~\mbox{eV}$, respectively, relative to the
185: $\mbox{H}(^2S)+\mbox{LiF}(X ^1\Sigma^+)$ asymptote with $E=0$ corresponding
186: to the bottom of the LiF potential. The $\mbox{LiH}(X ^1\Sigma^+)+\mbox{F}(^2P)$ products
187: channel lies at $+3.3839~\mbox{eV}$ relative to the energy origin, thus this reaction
188: channel is closed for the energy range considered in this study.
189: The PES of \citet{agua03} significantly improves on the widely used PES of \citet*{park95}
190: constructed from a limited set of {\em ab initio} data. It is also markedly different
191: from the recent {\em ab initio} PES of \citet{jasp01}, which exhibits a $+0.076~\mbox{eV}$
192: higher saddle point separating the $\mbox{H}\cdots\mbox{FLi}$ and
193: $\mbox{Li}\cdots\mbox{FH}$ complexes and an additional saddle
194: point at $+0.014~\mbox{eV}$ in the reagent valley.
195: Using the PES of \citet{agua03}, we found that, when the zero-point energies of
196: reactants and products are included, the
197: $\mbox{H}+\mbox{LiF}(v=0,j=0)\rightarrow\mbox{Li}+\mbox{HF}(v'=0,j'=0)$ reaction has
198: an endoergicity of $0.0112~\mbox{eV}~(0.2588~\mbox{kcal/mol})$, while for
199: vibrationally excited $\mbox{LiF}(v=1,j=0)$ and $\mbox{LiF}(v=2,j=0)$ reagents the
200: reaction producing $\mbox{HF}(v'=0,j'=0)$ becomes exoergic by
201: $0.0998~\mbox{eV}~(2.3011~\mbox{kcal/mol})$ and $0.2089~\mbox{eV}~(4.8170~\mbox{kcal/mol})$,
202: respectively.
203:
204:
205: %-------------------------------------------
206: \subsection{Quantum scattering calculations}
207: %-------------------------------------------
208:
209: Quantum reactive scattering calculations were carried out using the coupled-channel
210: hyperspherical coordinate method as implemented in the ABC program \cite{sko00}.
211: In this method, the Schr\"{o}dinger equation for the motion of three nuclei on the parametric
212: representation of a single Born-Oppenheimer PES is solved in Delves hyperspherical
213: coordinates with exact reactive scattering boundary conditions.
214: %
215: For all arrangements of the collision products, parity-adapted $S-$matrix
216: elements, $S^{J,P}_{v'j'k',vjk}$, are calculated for each $(J,P,p)$ triple. Here,
217: $J$ is the total angular momentum quantum number, $P$ and $p$ are the triatomic and diatomic
218: parity eigenvalues, respectively; $v$ and $j$ are the usual diatomic vibrational
219: and rotational quantum numbers and $k$ is the helicity quantum number for
220: the reactants, their primed counterparts referring to the products. After
221: transformation of the parity-adapted $S-$matrix elements into their standard
222: helicity representation, $S^{J}_{v'j'k',vjk}$, initial state selected cross sections
223: are computed as a function of the kinetic energy, $E_{kin}$, according to
224: %
225: \begin{equation}\label{stscs}
226: \sigma_{vj}(E_{kin}) = \frac{\pi}{k^2_{vj}(2j+1)}\sum^{J_{max}}_{J=0}
227: (2J+1) \sum_{v'j'k'k} \vert S^{J}_{v'j'k',vjk}(E_{kin})\vert ^2,
228: \end{equation}
229: %
230: where $k_{vj}$ is the incident channel wave vector and the helicity quantum numbers
231: $k$ and $k'$ are restricted to the ranges $0\leqslant k \leqslant \mbox{min}(J,j)$ and
232: $0\leqslant k' \leqslant \mbox{min}(J,j')$. For zero total molecular
233: angular momentum and $s-$wave scattering in the incident channel, Eq. (\ref{stscs}) merely
234: reduces to a summation over the quantum number $v'$ and $j'$.
235: The cumulative reaction reaction probability (CRP) is also computed
236: using the formula
237: %
238: \begin{equation}\label{crp}
239: N_{J=0}(E) = \sum_{v'j'k',vjk} \vert S^{J=0}_{v'j'k',vjk}(E)\vert ^2,
240: \end{equation}
241: %
242: where $E$ is the total energy and the summation is over all open channels of the reactant
243: and product molecules.
244: For $J=0$, reaction rate coefficients are further evaluated as the product of the cross section
245: and the relative velocity, as a function of the translational temperature, $T=E_{kin}/k_B$,
246: where $k_B$ is the Boltzmann constant.
247:
248: %-----------------------------
249: %\subsection{Convergence tests}
250: %-----------------------------
251:
252: Owing to the predominance of quantum tunneling in reactions involving energy
253: barriers at low temperatures, reaction probabilities are generally small for
254: translationally cold and ultracold collisions.
255: Therefore, extensive convergence tests have been carried out for
256: the initial-state-selected and state-to-state reaction probabilities of
257: the exothermic $\mbox{H}+\mbox{LiF}(v=1,2,j=0)\rightarrow\mbox{Li}+\mbox{HF}(v',j')$
258: reactions. Convergence of our scattering calculations was checked with respect to
259: the maximum rotational quantum number, $j_{max}$, the cut-off energy, $E_{max}$
260: that control the basis set size, the maximum value of the hyperradius,
261: $\rho_{max}$, and the size of the log derivative propagation sectors, $\Delta{\rho}$.
262:
263: Converged reaction probability for LiF formation in $\mbox{H}+\mbox{LiF}(v=1,2,j=0)$
264: collisions were obtained over the range $10^{-5}-10^{-3}~\mbox{eV}$ using the
265: values $\rho_{max}=25.0~\mbox{a.u.}$ and $\Delta{\rho}=0.01~\mbox{a.u.}$
266: The accuracy of our numerical results over this energy range was better than $10^{-10}$
267: using a large basis set size constrained with the values of $j_{max}=25$ and
268: $E_{max}=2.5~\mbox{eV}$. Using the same values of $\rho_{max}$ and $\Delta{\rho}$ as above,
269: the size of the basis set was further optimized by adjusting the value of the cut-off
270: energy.
271: In general, the reaction probabilities are more sensitive to the
272: basis set parameters at low energies. We performed the basis set optimization
273: at an incident kinetic energy of $10^{-5}~\mbox{eV}$. We found that
274: the state-to-state reactions
275: $\mbox{Li}+\mbox{HF}(v=1,2,j=0)\rightarrow\mbox{H}+\mbox{LiF}(v'=0,j')$
276: with $E_{max}=2.5~\mbox{eV}$ and $E_{max}=2.0~\mbox{eV}$ resulted in
277: nearly identical values of the reaction probabilities.
278: The basis set corresponding to $E_{max}=2.0~\mbox{eV}$
279: was composed of 634 basis functions.
280: Based on these convergence tests, values of $\rho_{max}=25.0~\mbox{a.u.}$,
281: $\Delta{\rho}=0.01~\mbox{a.u.}$, $j_{max}=25$ and $E_{max}=2.0~\mbox{eV}$
282: were adopted in the calculations reported hereafter.
283:
284:
285: %===============================
286: \section{Results and discussion} \label{sec:2}
287: %===============================
288:
289: Fig. \ref{fig1}~ shows the initial state-selected reaction probability for
290: HF formation in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions as a function of
291: the translational energy. For LiF reagent initially in the rovibrational
292: ground state, the reaction leading to HF formation has an endoergicity of
293: $0.0112~\mbox{eV}~(0.2588~\mbox{kcal/mol})$, thus explaining the presence of
294: the kinetic energy threshold in the lower panel of Fig. \ref{fig1}.
295: Probability for nonreactive collisions also exhibits a threshold near
296: $3\times 10^{-4}~\mbox{eV}~(0.007~\mbox{kcal/mol})$ associated with rotational
297: excitation to $\mbox{LiF}(v=0,j=1)$.
298: At low energies, the reaction probabilities are small, because quantum tunneling
299: of the fluorine atom is the main reaction pathway. Vibrational excitation of LiF
300: significantly enhances the reaction probability in the ultracold and cold regimes,
301: for instance, an energy increase of $0.109~\mbox{eV}$ from LiF$(v=1,j=0)$ to
302: LiF$(v=2,j=0)$ translates in a 350\% increase of the reaction probability in the
303: zero temperature limit.
304: Our results also show a dense resonance structure in the energy dependence of
305: the reaction probability at low energies associated with quasibound states of the
306: $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ and $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ van
307: der Waals complexes. Assignment of these resonances will
308: be presented further in this study.
309:
310: Cross sections for $s-$wave elastic scattering in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$
311: collisions are displayed in Fig. \ref{fig2} as a function of the incident kinetic energy.
312: For low translational energies, only $s-$wave scattering is expected to play a significant
313: role \cite{kaj03}. For collisions involving excited LiF reagents, elastic cross sections
314: are found to have similar values for kinetic energies below $5\times10^{-3}~\mbox{eV}$.
315: The nearly identical values of the limiting elastic cross sections for $v=1$ and 2 suggest
316: that vibrational excitation does not affect the isotropic part of the interaction potential.
317: In the zero-temperature limit, the scattering length can be expressed as a complex number,
318: $a_{vj}=\alpha_{vj}-i\beta_{vj}$, where the real and imaginary parts are related to
319: the elastic component of the $S-$matrix, $S^{el}_{vj}$, according to
320: %
321: \begin{equation}\label{rscatl}
322: \alpha_{vj} = -\lim_{k_{vj}\to 0}\frac{\mbox{Im}(S^{el}_{vj})}{2k_{vj}},
323: \end{equation}
324: %
325: \begin{equation}\label{rscat2}
326: \beta_{vj} = \lim_{k_{vj}\to 0}\frac{1-\mbox{Re}(S^{el}_{vj})}{2k_{vj}},
327: \end{equation}
328: %
329: where $k_{vj}$ is the wave vector in the incident channel. We found $\alpha_{10}=+9.100~\mbox{\AA}$
330: and $\alpha_{20}=+9.159~\mbox{\AA}$ for $v=1$ and $v=2$, respectively.
331: However, for numerical reasons, the Wigner's threshold law for inelastic reactive and nonreactive
332: collisions, i. e.
333: %
334: \begin{equation}\label{sigin}
335: \sigma^{in}_{vj} = \frac{4\pi\beta_{vj}}{k_{vj}},
336: \end{equation}
337: %
338: is preferred over Eq. (\ref{rscat2}) to evaluate the imaginary part of the scattering length.
339: In Eq. (\ref{sigin}), $\sigma^{in}_{vj}$ is the sum of the nonreactive and reactive cross sections.
340: Using Eq. (\ref{sigin}), we found $\beta_{10}=2.365\times 10^{-4}~\mbox{\AA}$ and
341: $\beta_{20}=3.478\times 10^{-3}~\mbox{\AA}$ for $v=1$ and $v=2$, respectively.
342: The total elastic cross section in the zero energy limit is given by
343: %
344: \begin{equation}\label{sigel}
345: \sigma^{el}_{vj} = 4\pi \left| a_{vj}\right|^2,
346: \end{equation}
347: %
348: which yields $\sigma^{el}_{20}=1054.05\times 10^{-16}~\mbox{cm}^2$ and
349: $\sigma^{el}_{10}=1040.58\times 10^{-16}~\mbox{cm}^2$.
350:
351: Initial-state-selected cross sections for HF formation and for
352: nonreactive scattering in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions are displayed
353: in Fig. \ref{fig3} as a function of the incident translational energy.
354: For $v=1$ and $v=2$, reaction cross sections reach the Wigner regime \cite{wig48}
355: for kinetic energies below $10^{-5}~\mbox{eV}$ in accordance with Eq. (\ref{sigin}).
356: Nonreactive scattering dominates
357: over the complete kinetic energy range
358: investigated, with a $\mbox{LiF}/\mbox{HF}$ product branching ratio
359: of 4 for $v=1$ and 15 for $v=2$ in the Wigner regime.
360: For translational energies beyond $10^{-4}~\mbox{eV}$, reaction cross sections
361: are characterized by resonant spikes due to metastable states of the
362: $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ and $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ van der Waals
363: complexes. These resonances tend to be washed out for reactions with vibrational excited
364: reactants.
365:
366: Assignments of the resonances in the energy dependence of the reactive
367: cross sections in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions
368: are presented in Table \ref{tab1}. Bound- and quasi-bound states calculations have
369: been performed for the $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ van der Waals potentials
370: correlating with the $\mbox{LiF}(v=0-2)$ manifold.
371: The adiabatic potentials are obtained by constructing the matrix elements of the
372: interaction potential in a basis set of the rovibrational levels of the LiF molecule
373: and diagonalizing the resulting diabatic potentials as a function of the
374: atom-molecule separation, $R$. The resonance energies and the corresponding
375: wave functions are computed using the Fourier grid Hamiltonian (FGH) method \cite{mars89,bali92}.
376: For constructing the adiabatic potentials, we used a 20-term Legendre expansion
377: of the interaction potential, 25 angular orientations to project out the expansion
378: coefficients, 17 Gauss-Hermite quadrature points for the vibrational wave functions
379: and a grid of 2000 points in the atom-molecule separation ranging from $R=0.1-100~a_0$.
380: Similar calculations for the $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ exit channel complexes have
381: been carried out previously \cite{weck04b} and results for the van der Waals potential correlating
382: with the $\mbox{HF}(v=0)$ manifold were reported. Total energy values given in
383: Table \ref{tab1} are relative to separated $\mbox{H}+\mbox{LiF}$ system with energy
384: origin at the bottom of the LiF potential.
385: Eigenenergies of the reactant and product diatoms calculated using the FGH and ABC codes
386: were found to be consistent within an error margin of $10^{-6}~\mbox{eV}$ for both
387: $\mbox{H}+\mbox{LiF}$ and $\mbox{Li}+\mbox{HF}$ asymptotes.
388: The resonances in the $\mbox{H}\cdots\mbox{LiF}$ entrance channel
389: van der Waals well are characterized by quantum numbers $(v,j,t)$, where
390: $v$ and $j$ are the LiF vibrational and rotational quantum numbers, respectively, and
391: $t$ refers to the $\mbox{H}-\mbox{LiF}(v,j)$ van der Waals stretching
392: vibration; their primed counterparts refer to resonances originating from the
393: $\mbox{Li}\cdots\mbox{HF}$ exit channel well. As shown in Table \ref{tab1}, the agreement between
394: the FGH energy eigenvalues and the peak positions from our scattering calculations is
395: excellent. For $v=0$, resonances are mostly due to the decay of $(v'=0,j',t')$ metastable
396: states of the $\mbox{Li}\cdots\mbox{HF}$ exit channel van der Waals complex, with $j'$ and $t'$ ranging from
397: relatively high $t'$-values for $j'=1$ to the lowest-lying stretching vibrational states of $j'=5$.
398: Resonances attributed to the $\mbox{H}\cdots\mbox{LiF}(v=0,j=10-15,t=1-2)$ states also appear
399: for reactants in this vibrational state. For $v=1$ and $v=2$, this tendency is reverted with
400: a dominant imprint of resonances corresponding to low$-t$ quasibound states of the
401: adiabatic potentials correlating with $j=2-12$ of the entrance channel
402: $\mbox{H}\cdots\mbox{LiF}$ complex. For these
403: vibrationally excited states of the reagents, only few high-lying stretching
404: vibrational states of the $\mbox{Li}\cdots\mbox{HF}$ exit channel van der Waals complex give rise
405: to resonances: $j'=6$ for $v=1$ and $j'=9$ for $v=2$. Although the exit channel van der Waals well is much deeper than
406: the entrance channel well, the $\mbox{Li}\cdots\mbox{HF}$ exit channel complex dominates only for
407: $v=0$, while for higher vibrational levels of the reagents, the entrance channel complex plays the
408: major role and the signature of the exit channel quasi-bound states tends to disappear.
409: It is also interesting to note that the resonances appear mostly in the reactive cross sections
410: despite the fact that for $v=1$ and 2 it is the entrance channel van der Waals potential that supports
411: the resonances. Thus, it appears that the long lived resonances decay mostly by vibrational pre-reaction of
412: the quasibound states by tunneling of the F atom.
413:
414: The cumulative reaction probability for HF formation in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$
415: collisions is depicted in Fig. \ref{fig4} as a function of the total energy. CRP is
416: represented over an energy range of $0.1~\mbox{eV}$ starting from the total energy values
417: of the $\mbox{LiF}(v,j=0)$ reagents, thus explaining the presence of thresholds at
418: $0.0675~\mbox{eV}$, $0.1673~\mbox{eV}$ and $0.2764~\mbox{eV}$, respectively, for $v=0-2$.
419: The resonance structure still subsists in the CRP, but becomes less sharper with increase
420: in energy and vibrational excitation.
421:
422: The $J=0$ contribution to the reaction rate coefficients for HF formation in
423: $\mbox{H}+\mbox{LiF}(v=1,2,j=0)$ collisions is shown in Fig. \ref{fig5}.
424: For $v=1$, the reaction rate coefficient reaches the Wigner regime for temperatures below
425: $0.1~\mbox{K}$, with a finite value of
426: $3.86\times 10^{-15}~\mbox{cm}^3~\mbox{s}^{-1}$ in the zero-temperature limit.
427: Vibrational excitation of HF to the $v=2$ level enhances reactivity by about a factor of 4
428: in the cold and ultracold regimes.
429: Although rate coefficients appear to be small at low temperatures, the values are
430: not negligible compared to those for reactions involving tunneling
431: of atoms lighter than fluorine. This suggests that heavy atom tunneling may play a major
432: role at low energies, as observed recently by \citet{zuev03} for carbon tunneling
433: in ring expansion reactions of 1-methylcyclobutylhalocarbenes at 8 to 25 K.
434: In Fig. \ref{fig5}, for $v=1$ and $v=2$, the reactivity rapidly increases beyond $3~\mbox{K}$,
435: however, accurate prediction of rate coefficients for higher temperatures requires calculations
436: for $J>0$ which is beyond the scope of this work.
437:
438:
439: %===================
440: \section{Conclusion} \label{sec:conclu}
441: %===================
442:
443: We have investigated the scattering dynamics of the
444: $\mbox{H}+\mbox{LiF}(X ^1\Sigma^+,v=0-2,j=0)\rightarrow\mbox{HF}(X ^1\Sigma^+,v',j')+\mbox{Li}(^2S)$
445: reaction for zero total angular momentum. Using the coupled-channel hyperspherical coordinate method,
446: probabilities, cross sections, and rate coefficients have been
447: computed from the zero-temperature limit up to a translational energy of $0.1~\mbox{eV}$ using
448: the most recent PES for the LiHF electronic ground state.
449: Extensive calculations have been carried out for quasibound states in the entrance and exit channel van
450: der Waals wells to characterize the rich resonance structure observed in our scattering calculations.
451: We found that the resonances arise from the decay of metastable states of both
452: $\mbox{H}\cdots\mbox{F}-\mbox{Li}$ and $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ van der Waals complexes.
453: In particular, the quasibound states of the deep $\mbox{Li}\cdots\mbox{F}-\mbox{H}$ van der Waals
454: well were found to strongly imprint the resonance structure for reactions involving
455: $\mbox{LiF}(v=0,j=0)$. For vibrationally excited reagents, however, resonances are mainly
456: originating from tunneling of metastable states of $\mbox{H}\cdots\mbox{F}-\mbox{Li}$~ into
457: the $\mbox{HF}(v'=0)$ product manifold.
458: Consistent with our recent findings for the reverse reaction \cite{weck04b}, chemical reactivity for
459: $\mbox{H}+\mbox{LiF}\rightarrow\mbox{HF}+\mbox{Li}$ is enhanced by vibrational excitation of the
460: reagents at low and ultralow temperatures. In contrast to $\mbox{Li}+\mbox{HF}(v=1,j=0)$,
461: nonreactive processes dominate over HF formation in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions.
462: However, quasibound states of the entrance channel van der Waals potential do decay
463: through vibrational pre-reaction by tunneling of the fluorine atom.
464: The results suggest that heavy atom tunneling at low temperatures may be more important
465: than generally recognized, as observed recently by \citet{zuev03} for carbon tunneling
466: in ring expansion reactions.
467:
468: %======================
469: \begin{acknowledgments}
470: %======================
471:
472: This work was supported by NSF grant PHYS-0245019, the Research Corporation and by the
473: United States-Israel Binational Science Foundation.
474:
475: %====================
476: \end{acknowledgments}
477: %====================
478:
479: %=============
480: % Bibliography
481: %=============
482:
483: \begin{thebibliography}{}
484:
485: \bibitem{bell80}
486: R. P. Bell,
487: The Tunnel Effect in Chemistry (Chapman and Hall, New York) (1980).
488: \bibitem{gao02}
489: J. Gao and D. G. Truhlar,
490: Ann. Rev. Phys. Chem. \textbf{53}, 457 (2002).
491: \bibitem[ Zuev et al.(2003)]{zuev03}
492: P. S. Zuev, R. S. Sheridan, T. V. Albu, D. G. Truhlar, D. A. Hrovat,
493: and W. T. Borden,
494: Science \textbf{299}, 867 (2003).
495: \bibitem{sulta04}
496: R. A. Sultanov and N. Balakrishnan,
497: J. Chem. Phys. \textbf{121}, 11038 (2004).
498: \bibitem{hira01}
499: K. Hiraoka, T. Sato, and T. Takayama,
500: Science \textbf{292}, 869 (2001).
501: \bibitem{joc03}
502: S. Jochim {\em et al},
503: Science \textbf{302}, 2101 (2003).
504: \bibitem{gre03}
505: M. Greiner, C. A. Regal, and D. S. Jin,
506: Nature (London) \textbf{426}, 537 (2003).
507: \bibitem{zwi03}
508: M. W. Zwierlein, C. A. Stan, C. H. Schunk, S. M. Raupach, S. Gupta,
509: Z. Hafzibabic, and W. Ketterle,
510: Phys. Rev. Lett. \textbf{91}, 250401 (2003).
511: \bibitem{bala01}
512: N. Balakrishnan and A. Dalgarno,
513: Chem. Phys. Lett. \textbf{341}, 652 (2001).
514: \bibitem{bodo02}
515: E. Bodo, F. A. Gianturco, and A. Dalgarno,
516: J. Chem. Phys. \textbf{116}, 9222 (2002).
517: \bibitem{bala03}
518: N. Balakrishnan and A. Dalgarno,
519: J. Phys. Chem. A \textbf{107}, 7101 (2003).
520: \bibitem[Weck and Balakrishnan(2004)]{weck04}
521: P. F. Weck and N. Balakrishnan,
522: Euro. Phys. J. D \textbf{31}, 417 (2004).
523: \bibitem{taka98}
524: T. Takayanagi and Y. Kurosaki,
525: J. Chem. Phys. \textbf{109}, 8929 (1998).
526: \bibitem{taka99}
527: T. Takayanagi and Y. Kurosaki,
528: Phys. Chem. Chem. Phys. \textbf{1}, 1099 (1999).
529: \bibitem{zou01}
530: S. Zou, S. Skokov, and J. M. Bowman,
531: Chem. Phys. Lett. \textbf{339}, 290 (2001).
532: \bibitem{xie02}
533: T. Xie, D. Wang, J. M. Bowman, and D. E. Manolopoulos,
534: J. Chem. Phys. \textbf{116}, 7461 (2002).
535: \bibitem{bowman05} J. M. Bowman,
536: Chem. Phys. \textbf{308}, 255 (2005).
537: \bibitem{pania99}
538: M. Paniagua, A. Aguado, M. Lara, and O. Roncero,
539: J. Chem. Phys. \textbf{111}, 6712 (1999).
540: \bibitem[Wei et al.(2003)Wei, Jasper, and Truhlar]{wei03}
541: L. Wei, A. W. Jasper, and D. G. Truhlar.
542: J. Phys. Chem. A \textbf{107}, 7236 (2003).
543: \bibitem[Weck and Balakrishnan(2004b)]{weck04b}
544: P. F. Weck and N. Balakrishnan,
545: J. Chem. Phys. \textbf{117}, in press;
546: physics/0412105 (2005).
547: \bibitem[Taylor and Datz(1955)]{tayl55}
548: E. H. Taylor and S. Datz,
549: J. Chem. Phys. \textbf{23}, 1711 (1955).
550: \bibitem{hers66}
551: D. R. Herschbach,
552: Adv. Chem. Phys. \textbf{10}, 319 (1966).
553: \bibitem{beck80}
554: C. H. Becker, P. Casavecchia, P. W. Tiedermann, J. J. Valentini,
555: and Y. T. Lee,
556: J. Chem. Phys. \textbf{73}, 2833 (1980).
557: \bibitem[Loesch and Stienkemeier(1993)]{loes93}
558: H. J. Loesch and F. Stienkemeier,
559: J. Chem. Phys. \textbf{99}, 9598 (1993).
560: \bibitem{baer94}
561: M. Baer, I. Last, and H.-J. Loesch,
562: J. Chem. Phys. \textbf{101}, 9648 (1994).
563: \bibitem{aoiz99}
564: F. J. Aoiz, M. T. Martinez, M. Men\'{e}ndez, V. R\'{a}banos, and E. Verdasco,
565: Chem. Phys. Lett. \textbf{299}, 25 (1999).
566: \bibitem{aoiz00}
567: F. J. Aoiz, E. Verdasco, V. S. R\'{a}banos, H.-J. Loesch, M. Men\'{e}ndez,
568: and F. Stienkemeier,
569: Phys. Chem. Chem. Phys. \textbf{2}, 541 (2000).
570: \bibitem{aoiz01}
571: F. J. Aoiz, M. T. Martinez, and V. S. R\'{a}banos,
572: J. Chem. Phys. \textbf{114}, 8880 (2001).
573: \bibitem{casa00}
574: P. Casavecchia,
575: Rep. Prog. Phys. \textbf{63}, 355 (2000).
576: \bibitem{hobe01}
577: O. Hobel, M. Men\'{e}ndez, and H.-J. Loesch,
578: Phys. Chem. Chem. Phys. \textbf{3}, 3633 (2001).
579: \bibitem{hobe04}
580: O. Hobel, R. Bobbenkamp, A. Paladini, A. Russo, and H.-J. Loesch,
581: Phys. Chem. Chem. Phys. \textbf{6}, 2198 (2004).
582: \bibitem{agua97a}
583: A. Aguado, M. Paniagua, M. Lara, and O. Roncero,
584: J. Chem. Phys. \textbf{106}, 1013 (1997).
585: \bibitem[Aguado et al.(1997)]{agua97b}
586: A. Aguado, M. Paniagua, M. Lara, and O. Roncero,
587: J. Chem. Phys. \textbf{107}, 10085 (1997).
588: \bibitem{zeir78}
589: Y. Zeiri and M. Shapiro,
590: Chem. Phys. \textbf{31}, 217 (1978).
591: \bibitem{chen80}
592: M. M. Chen and H. F. Schaefer, III,
593: J. Chem. Phys. \textbf{72}, 4376 (1980).
594: \bibitem{cart80}
595: S. Carter and J. N. Murrell,
596: Mol. Phys. \textbf{41}, 567 (1980).
597: \bibitem{laga84}
598: A. Lagan\`{a} and E. Garcia,
599: Theochem \textbf{16}, 91 (1984).
600: \bibitem{garc84}
601: E. Garcia and A. Lagan\`{a},
602: Mol. Phys. \textbf{52}, 1115 (1984).
603: \bibitem{pani89}
604: M. Paniagua and A. Aguado,
605: Chem. Phys. \textbf{134}, 287 (1989).
606: \bibitem{palm89}
607: P. Palmeiri and A. Lagan\`{a},
608: J. Chem. Phys. \textbf{91}, 7303 (1989).
609: \bibitem{suar94}
610: C. Suarez, A. Aguado, C. Tablero and M. Paniagua,
611: Int. J. Quantum Chem. \textbf{52}, 935 (1994).
612: \bibitem{agua95}
613: A. Aguado, C. Suarez, and M. Paniagua,
614: Chem. Phys. \textbf{201}, 107 (1995).
615: \bibitem{laga98a}
616: A. Lagan\`{a}, O. Gervasi, and E. Garcia,
617: Chem. Phys. Lett. \textbf{143}, 174 (1998).
618: \bibitem{laga98b}
619: A. Lagan\`{a}, G. Ochoa de Aspuru, and E. Garcia,
620: J. Chem. Phys. \textbf{168}, 3886 (1998).
621: \bibitem{burc00}
622: R. Burcl, P. Piecuch, V. \u{S}pirko and O. Bludsk\'{y},
623: Int. J. Quantum Chem. \textbf{80}, 916 (2000).
624: \bibitem[Jasper et al.(2001)]{jasp01}
625: A. W. Jasper, M. D. Hack, A. Chakraborty, D. G. Truhlar,
626: and P. J. Piecuch,
627: J. Chem. Phys. \textbf{115}, 7945 (2001).
628: \bibitem[Jasper et al.(2002)]{jasp02}
629: A. W. Jasper, M. D. Hack, D. G. Truhlar, and P. J. Piecuch,
630: J. Chem. Phys. \textbf{116}, 8353 (2002).
631: \bibitem{burc02}
632: R. Burcl, P. Piecuch, V. \u{S}pirko and O. Bludsk\'{y},
633: Theochem \textbf{591}, 151 (2002).
634: \bibitem[Aguado et al.(2003)]{agua03}
635: A. Aguado, M. Paniagua, C. Sanz, and O. Roncero,
636: J. Chem. Phys. \textbf{119}, 10088 (2003).
637: \bibitem[Parker et al.(1995)]{park95}
638: G. A. Parker, A. Lagan\`{a}, S. Crocchianti, and R. T. Pack,
639: J. Chem. Phys. \textbf{102}, 1238 (1995).
640: \bibitem{gogt96}
641: F. G\"{o}gtas, G. G. Balint-Kurti, and A. R. Offer,
642: J. Chem. Phys. \textbf{104}, 7927 (1996).
643: \bibitem{lara98}
644: M. Lara, A. Aguado, O. Roncero, and M. Paniagua,
645: J. Chem. Phys. \textbf{109}, 9391 (1998).
646: \bibitem{pani98}
647: M. Paniagua, A. Aguado, M. Lara, and O. Roncero,
648: J. Chem. Phys. \textbf{109}, 2971 (1998).
649: \bibitem{lara00}
650: M. Lara, A. Aguado, M. Paniagua, and O. Roncero,
651: J. Chem. Phys. \textbf{113}, 1781 (2000).
652: \bibitem[Lagan\`{a} et al.(2004)Lagan\`{a}, Crocchianti, and Piermarini]{laga04}
653: A. Lagan\`{a}, S. Crocchianti and V. Piermarini,
654: Lect. Notes Comp. Science \textbf{3044}, 422 (2004).
655: \bibitem{aoiz00b}
656: F. J. Aoiz, M. T. Martinez, M. Men\'{e}ndez, V. S. R\'{a}banos,
657: and E. Verdasco,
658: Chem. Phys. Lett. \textbf{299}, 25 (2000).
659: \bibitem[Aguado and Paniagua(1992)]{agua92}
660: A. Aguado and M. Paniagua,
661: J. Chem. Phys. \textbf{96}, 1265 (1992).
662: \bibitem[Skouteris, Castillo and Manolopoulos(2000)]{sko00}
663: D. Skouteris, J. F. Castillo and D. E. Manolopoulos,
664: Comput. Phys. Commun. \textbf{133}, (2000) 128.
665: \bibitem{kaj03}
666: M. Kajita,
667: Eur. Phys. J. D \textbf{23}, 337 (2003).
668: \bibitem{wig48}
669: E. P. Wigner,
670: Phys. Rev. \textbf{73}, 1002 (1948).
671: \bibitem[Marston and Balint-Kurti(1989)]{mars89}
672: C. C. Marston and G. G. Balint-Kurti,
673: J. Chem. Phys. \textbf{91}, 3571 (1989).
674: \bibitem[Balint-Kurti, Dixon, and Marston(1992)]{bali92}
675: G. G. Balint-Kurti, R. N. Dixon, and C. C. Marston,
676: Internat. Rev. Phys. Chem. \textbf{11}, 317 (1992).
677:
678: %====================
679: \end{thebibliography}
680: %====================
681:
682: \clearpage
683:
684: %======
685: %Tables
686: %======
687:
688: \begingroup
689: \squeezetable
690: \begin{table}
691: \caption{\label{tab1} Assignment of the resonances in the energy dependence of the
692: cross sections for HF formation in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions
693: (energies in eV).}
694: \begin{ruledtabular}
695: \begin{tabular}{cccccccccccc}
696: \multicolumn{1}{c}{}
697: & \multicolumn{2}{c}{Position\footnotemark[2]}
698: & \multicolumn{1}{c}{}
699: & \multicolumn{3}{c}{$\mbox{H}\cdots\mbox{LiF}(v,j,t)$}
700: & \multicolumn{1}{c}{}
701: & \multicolumn{4}{c}{$\mbox{Li}\cdots\mbox{HF}(v',j',t')$}
702: \\
703: \cline{2-3} \cline{5-7} \cline{9-12}
704: \multicolumn{1}{c}{$v$\footnotemark[1]}
705: & \multicolumn{1}{c}{$E_{kin}$}
706: & \multicolumn{1}{c}{$E_{tot}$}
707: & \multicolumn{1}{c}{}
708: & \multicolumn{1}{c}{$E_{v,j,t}$\footnotemark[3] }
709: & \multicolumn{1}{c}{$j$\footnotemark[4]}
710: & \multicolumn{1}{c}{$t$\footnotemark[5]}
711: & \multicolumn{1}{c}{}
712: & \multicolumn{1}{c}{$E_{v',j',t'}$\footnotemark[3] }
713: & \multicolumn{1}{c}{$v'$\footnotemark[6]}
714: & \multicolumn{1}{c}{$j'$\footnotemark[7]}
715: & \multicolumn{1}{c}{$t'$\footnotemark[8]}
716: \\
717: \hline
718: %$\mbox{H}+\mbox{LiF}(v=0,j=0)$ && & & & & & & & & \\
719: %\hline
720: %
721: %0& &0.0680 && & & && X & x & x & x \\
722: 0&0.0127 &0.0690 && & & && 0.0689 & 0 & 2 & 6 \\
723: 0&0.0132 &0.0695 && 0.0692 & 10 & 1 && 0.0694 & 0 & 1 & 11 \\
724: 0&0.0146 &0.0709 && & & && 0.0709 & 0 & 3 & 3 \\
725: 0&0.0157 &0.0720 && & & && 0.0718 & 0 & 1 & 12 \\
726: 0&0.0165 &0.0728 && & & && 0.0725 & 0 & 4 & 1 \\
727: 0&0.0175 &0.0738 && & & && 0.0737 & 0 & 2 & 7 \\
728: 0&0.0222 &0.0785 && 0.0787 & 12 & 1 && 0.0786 & 0 & 3 & 4 \\
729: 0&0.0244 &0.0807 && 0.0809 & 13 & 1 && 0.0804 & 0 & 2 & 9 \\
730: 0&0.0264 &0.0827 && & & && 0.0826 & 0 & 2 & 10 \\
731: 0&0.0272 &0.0835 && & & && 0.0836 & 0 & 4 & 1 \\
732: 0&0.0280 &0.0843 && & & && 0.0843 & 0 & 3 & 5 \\
733: 0&0.0298 &0.0861 && 0.0868 & 14 & 1 && & & & \\
734: 0&0.0357 &0.0920 && 0.0922 & 15 & 2 && 0.0920 & 0 & 5 & 1 \\
735: 0&0.0397 &0.0960 && & & && 0.0957 & 0 & 3 & 8 \\
736: 0&0.0437 &0.1000 && & & && 0.0992 & 0 & 4 & 4 \\
737: 0&0.0477 &0.1040 && & & && 0.1043 & 0 & 5 & 2 \\
738: %
739: 1&0.0003 &0.1676 && 0.1675 & 2 & 2 && & & & \\
740: 1&0.0007 &0.1680 && 0.1679 & 6 & 2 && & & & \\
741: 1&0.0018 &0.1691 && 0.1690 & 3 & 2 && & & & \\
742: 1&0.0028 &0.1701 && & & && 0.1701 & 0 & 6 & 8 \\
743: 1&0.0033 &0.1706 && 0.1705 & 4 & 2 && & & & \\
744: 1&0.0049 &0.1722 && & & && 0.1726 & 0 & 6 & 9 \\
745: 1&0.0061 &0.1734 && 0.1733 & 8 & 1 && & & & \\
746: 1&0.0093 &0.1766 && 0.1766 & 9 & 1 && & & & \\
747: 1&0.0131 &0.1804 && 0.1802 & 10 & 1 && & & & \\
748: 1&0.0177 &0.1850 && 0.1856 & 11 & 1 && & & & \\
749: 1&0.0235 &0.1908 && 0.1896 & 12 & 1 && & & & \\
750: %
751: 2&0.0010 &0.2774 && 0.2772 & 6 & 1 && & & & \\
752: 2&0.0029 &0.2793 && 0.2794 & 7 & 1 && & & & \\
753: 2&0.0046 &0.2810 && & & && 0.2807 & 0 & 9 & 6 \\
754: 2&0.0056 &0.2820 && 0.2822 & 8 & 1 && & & & \\
755: 2&0.0095 &0.2859 && 0.2854 & 9 & 1 && & & & \\
756: 2&0.0118 &0.2882 && 0.2889 & 10 & 1 && 0.2880 & 0 & 9 & 8 \\
757: 2&0.0168 &0.2932 && 0.2936 & 11 & 1 && & & & \\
758: 2&0.0225 &0.2989 && 0.2985 & 12 & 1 && & & & \\
759: \end{tabular}
760: \end{ruledtabular}
761: %
762: \footnotetext[1]{LiF vibrational quantum number.}
763: \footnotetext[2]{Position of the observed resonances. Total energy, $E_{tot}$, is relative to
764: separated $\mbox{H}+\mbox{LiF}$ system with energy zero corresponding to the bottom of the LiF potential.}
765: \footnotetext[3]{Binding energies $E_{v,j,t}$ and $E_{v',j',t'}$ are calculated with the Fourier
766: grid Hamiltonian method in the $\mbox{H}\cdots\mbox{LiF}$ and $\mbox{Li}\cdots\mbox{HF}$
767: van der Waals wells, respectively.}
768: \footnotetext[4]{LiF rotational quantum number.}
769: \footnotetext[5]{Quantum number for the $\mbox{H}-\mbox{LiF}(v,j)$ van der Waals stretching vibration.}
770: \footnotetext[6]{HF vibrational quantum number.}
771: \footnotetext[7]{HF rotational quantum number.}
772: \footnotetext[8]{Quantum number for the $\mbox{Li}-\mbox{HF}(v',j')$ van der Waals stretching
773: vibration.}
774:
775: %
776: \end{table}
777: \endgroup
778:
779: \clearpage
780:
781:
782: %================
783: % Figure captions
784: %================
785:
786: \begin{figure}
787: \caption{\label{fig1} Initial state-selected probability for
788: HF formation (solid curve) and for non-reactive scattering (dashed curve) in
789: $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions as a function of the total energy.}
790: \end{figure}
791:
792: \begin{figure}
793: \caption{\label{fig2} Elastic cross sections for $s-$wave scattering in
794: $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions as a function of the
795: incident kinetic energy. Solid curve: $v=0$; dotted curve: $v=1$; dashed curve: $v=2$.}
796: \end{figure}
797:
798: \begin{figure}
799: \caption{\label{fig3} Cross sections for HF formation and nonreactive
800: scattering in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions as a function of the
801: incident kinetic energy. solid curve: HF product channel; dashed curve: nonreactive
802: scattering.}
803: \end{figure}
804:
805: \begin{figure}
806: \caption{\label{fig4} Cumulative reaction probability for HF formation
807: in $\mbox{H}+\mbox{LiF}(v=0-2,j=0)$ collisions as a function of the
808: total energy. Solid curve: $v=0$; dotted curve: $v=1$; dashed curve: $v=2$.}
809: \end{figure}
810:
811: \begin{figure}
812: \caption{\label{fig5} Temperature dependence of reaction rate
813: coefficients for HF formation and nonreactive vibrational quenching
814: in $\mbox{Li}+\mbox{HF}(v=1,2,j=0)$ collisions. }
815: \end{figure}
816:
817: %================
818: % Figure captions
819: %================
820:
821: \clearpage
822:
823: \begin{figure*}
824: \begin{center}
825: \includegraphics[height=14cm,width=16cm]{f1}
826: \end{center}
827: \end{figure*}
828:
829: \clearpage
830: \begin{figure*}
831: \begin{center}
832: \includegraphics[height=14cm,width=16cm]{f2}
833: \end{center}
834: \end{figure*}
835:
836: \clearpage
837: \begin{figure*}
838: \begin{center}
839: \includegraphics[height=14cm,width=16cm]{f3}
840: \end{center}
841: \end{figure*}
842:
843: \clearpage
844: \begin{figure*}
845: \begin{center}
846: \includegraphics[height=14cm,width=16cm]{f4}
847: \end{center}
848: \end{figure*}
849:
850: \clearpage
851: \begin{figure*}
852: \begin{center}
853: \includegraphics[height=14cm,width=16cm]{f5}
854: \end{center}
855: \end{figure*}
856:
857: %=============
858: \end{document}
859: %=============