1: % ****** Start of file apssamp.tex ******
2: %
3: % This file is part of the APS files in the REVTeX 4 distribution.
4: % Version 4.0 of REVTeX, August 2001
5: %
6: % Copyright (c) 2001 The American Physical Society.
7: %
8: % See the REVTeX 4 README file for restrictions and more information.
9: %
10: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
11: % as well as the rest of the prerequisites for REVTeX 4.0
12: %
13: % See the REVTeX 4 README file
14: % It also requires running BibTeX. The commands are as follows:
15: %
16: % 1) latex apssamp.tex
17: % 2) bibtex apssamp
18: % 3) latex apssamp.tex
19: % 4) latex apssamp.tex
20: %
21: %\documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
22: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
23:
24: % Some other (several out of many) possibilities
25: %\documentclass[aps,prl,preprint,groupedaddress]{revtex4}
26: \documentclass[11pt,aps,superscriptaddress]{revtex4}
27: %\documentclass[preprint,aps,draft]{revtex4}
28: %\documentclass[prb]{revtex4}% Physical Review B
29:
30: \usepackage{graphicx}% Include figure files
31: \usepackage{dcolumn}% Align table columns on decimal point
32: \usepackage{bm}% bold math
33:
34: %\nofiles
35:
36: \begin{document} % INITIALIZE - DONT CHANGE
37: %
38: %
39: %
40: \title{Quantal Density Functional Theory of the Hydrogen Molecule }
41:
42:
43: \author{Xiao-Yin Pan and Viraht Sahni}
44: \affiliation{Department of Physics, Brooklyn College and The
45: Graduate School of the City University of New York,365 Fifth Avenue, New York, New
46: York 10016. }
47: %address. (Remove the left % marks)
48: %
49:
50: \begin{abstract} % DON'T CHANGE THIS LINE
51: In this paper we perform a Quantal Density Functional Theory (Q-DFT) study of the Hydrogen molecule
52: in its ground state. In common with traditional Kohn-Sham density functional theory (KS-DFT),
53: Q-DFT transforms the interacting system as described by Schrodinger theory, to one of noninteracting
54: fermions -- the S system -- such that the equivalent density, total energy, and ionization potential are obtained.
55: The Q-DFT description of the S system is in terms of `classical' fields and their quantal sources
56: that are quantum-mechanical expectations of Hermitian operators taken with respect to the
57: interacting and S system wave functions.. The sources, and hence the fields, are separately
58: representative of all the many-body effects the S system must account for, viz.
59: electron correlations due to the Pauli exclusion principle, Coulomb repulsion and
60: Correlation-Kinetic effects. The local electron-interaction potential energy of each model
61: fermion is the work done to move it in the force of a conservative effective field that is the
62: sum of the individual fields. The Hartree, Pauli, Coulomb, and Correlation-Kinetic energy
63: components of the total energy are also expressed in virial form in terms of the corresponding fields.
64: The highest occupied eigenvalue of the S system is the negative of the ionization potential energy.
65: The Q-DFT analysis of the Hydrogen molecule is performed employing the highly accurate correlated wave
66: function of Kolos and Roothaan. The structure of the sources -- the density, Fermi-Coulomb,
67: Fermi, and Coulomb holes -- as a function of the electron position are obtained, and
68: from them the corresponding fields. (To our knowledge, these are the first accurate graphs of the
69: Fermi-Coulomb and Coulomb holes as a function of electron position for the Hydrogen molecule.) As a consequence of the symmetry of the molecule,
70: the individual fields -- Hartree, Pauli, Coulomb, Correlation-Kinetic -- are all \emph{antisymmetric}
71: about the center of the nuclear bond.
72: Thus, the electron-interaction potential energy, and its Hartree, Pauli,
73: Coulomb, and Correlation-Kinetic components are each \emph{symmetric} about this center.
74: The Coulomb correlation and Correlation-Kinetic fields, and hence their contributions to the
75: potential and total energy are an order of magnitude smaller than those due to the Hartree and Pauli terms.
76: However, the Correlation-Kinetic contribution is more significant than that due to Coulomb correlations.
77: This new fact is important to the construction of approximate KS-DFT `correlation' energy functionals for
78: molecules. Finally, there is a striking similarity in the structure of the
79: various sources, fields, and potential energies of the Hydrogen molecule for
80: electron positions in the positive half-space encompassing one nucleus, and those of the Helium atom.
81:
82: % Insert your abstract here. No {\it REVTEX} limit to number of lines.
83: \end{abstract}
84: %
85: \maketitle
86: \newpage
87: \section{ Introduction}
88: In this paper we analyze the Hydrogen molecule ($H_{2}$) in its ground-state electronic
89: configuration $(\sigma_{g}1s)^{2}$ from the perspective of time-independent Quantal density
90: functional theory (Q-DFT) \cite{1,2,3,4,5,6,7,8}.
91: The \textit{in principle exact} framework of Q-DFT for ground and
92: excited states, both nondegenerate and degenerate, has been
93: demonstrated by application to exactly solvable model atomic
94: systems \cite{3, 5,6,7,8} as well as by the use of essentially
95: exact atomic correlated wave functions \cite{2,4,9}.
96: In its approximate form, Q-DFT has been applied to atoms, atomic ions, atoms
97: in excited states,
98: and positron binding, as well as to the many-electron inhomogeneity at metallic
99: surfaces and metallic clusters.
100: We refer the reader to the review articles of Refs \cite{2,10} for further
101: references on these applications.
102: This paper constitutes a first step in the application of Q-DFT to molecules.
103: Here we present the essentially exact analysis of the $H_{2}$ molecule via Q-DFT by
104: employing the highly
105: accurate correlated wave function of Kolos and Roothaan \cite{11}.
106: Beyond the understandings achieved, a principal attribute of the
107: calculation is the knowledge that the structure of the
108: corresponding Q-DFT properties for other diatomic molecules will
109: then be qualitatively similar. Furthermore, these essentially
110: exact properties can be used as the basis for comparison and
111: testing of various approximations within Q-DFT prior to their
112: application to
113: more complex molecules.\\
114:
115:
116:
117:
118: Q-DFT, in common with traditional Kohn-Sham density functional theory (KS-DFT)\cite{12},
119: maps a system of electrons in an external field ${\bf F}^{ext} = -\nabla v({\bf r})$ in their
120: \textit{ground} state to one of \textit{noninteracting} fermions in their \textit{ground} state with
121: equivalent density. The equivalent ground-state energy and ionization potential are thereby also obtained.
122: The model system of noninteracting fermions is referred to as the S system, S being a mnemonic for a single
123: Slater determinant. (Within the framework of Q-DFT, it is also possible \textit{in principle} to map
124: into an S system in which the noninteracting fermions are in an \textit{excited} state.) The local
125: (multiplicative) effective potential energy $v_{s}({\bf r})$ of the model fermions is the sum of the
126: external $v({\bf r})$ and an electron-interaction $v_{ee}({\bf r})$ potential energy, the latter
127: being representative of all the electron correlations the S system must account for.
128: These correlations are those due to the Pauli exclusion principle, Coulomb repulsion,
129: and Correlation-Kinetic effects. The Correlation-Kinetic contribution is a consequence of
130: the difference in the kinetic energies of the interacting and noninteracting systems with equivalent density.
131: In Q-DFT, the potential energy $v_{ee}({\bf r})$ is defined as the work done to move a model fermion
132: in the force of a conservative `classical' field. The components of this field each separately represent
133: a different electron correlation. The sources of these component fields are quantal in that they are
134: expectations of Hermitian operators taken with respect to the Schr{\" o}dinger and S system wave functions.
135: The Pauli (exchange), Coulomb (correlation), and Correlation-Kinetic components of the total energy are
136: also separately expressed in integral virial form in terms of the fields representative of these correlations.
137: The highest occupied eigenvalue of the S system differential equation is the negative of the ionization
138: potential\cite{13}.
139: \\
140:
141:
142:
143: The traditional KS-DFT description of the S system differs from
144: that of Q-DFT in the following manner. Traditional theory is in
145: terms of the ground state energy functional $E[\rho]$ of the
146: density $\rho({\bf r})$, and of its functional derivative. In
147: KS-DFT, the many-body correlations noted above are embedded in the
148: KS
149: electron-interaction energy functional $E^{KS}_{ee}[\rho]$. The corresponding
150: electron-interaction potential energy of the noninteracting fermions is defined as the
151: functional derivative of this functional taken at the true ground state density value:
152: $v_{ee}({\bf r}) = \delta E^{KS}_{ee}[\rho]/\delta \rho({\bf r})$. Within KS-DFT, it is common practice to
153: subtract the known Hartree or Coulomb self energy functional $E_{H}[\rho]$ from $E_{ee}[\rho]$ , thereby defining the
154: KS `exchange-correlation' energy functional $E^{KS}_{xc}[\rho]$ and its functional derivative
155: $v_{xc}({\bf r}) = \delta E^{KS}_{xc}[\rho]/\delta \rho({\bf r})$. The functionals
156: ($E^{KS}_{ee}[\rho], E^{KS}_{xc}[\rho]$) and their respective derivatives
157: ($v_{ee}({\bf r}), v_{xc}({\bf r})$) are therefore also representative of the Pauli
158: and Coulomb correlations and Correlation-Kinetic effects. KS-DFT, however does not describe how
159: the different electron correlations are incorporated in the functionals ($E^{KS}_{ee}[\rho], E^{KS}_{xc}[\rho]$)
160: and hence how they are represented in their functional derivatives.
161: Furthermore, the functionals ($E^{KS}_{ee}[\rho], E^{KS}_{xc}[\rho]$) are themselves unknown.
162: As such, even if the exact wave function of an interacting system were known, it is not possible to
163: construct the corresponding S system directly by following the prescription of KS-DFT.
164: Hence, the potential energy $v_{xc}({\bf r})$ is usually constructed indirectly via
165: density-based methods [14-17] that employ knowledge of the `exact' density as determined
166: from \textit{ab initio} calculations.\\
167:
168: In Section II we give a brief description of ground state Q-DFT. Section III is a
169: description of the various quantal sources, fields, energies and potential
170: energies pertaining to the S system as determined via Q-DFT employing the
171: $51$-parameter correlated wave function of Kolos-Roothaan. Concluding remarks
172: are made in Section IV. \\
173:
174:
175:
176: \section{ Q-DFT OF A NONDEGENERATE GROUND STATE}
177: The Schr{\"o}dinger equation for a system of $N$ electrons in an
178: external field $\textbf{F}^{ext}({\bf r}) = -\bm{\nabla} v({\bf
179: r})$, and in a
180: nondegenerate ground
181: state, is
182: \begin{equation}
183: [\hat{T}+\hat{V}+\hat{U}]\Psi({\bf X})=E \Psi ({\bf X}),
184: \end{equation}
185: where $\hat{T} = -\frac{1}{2} \sum_{i} \nabla_{i}^{2},\;\;\;
186: \hat{V}= \sum_{i} v({\bf r}_{i})$, and $\hat{U} = \frac{1}{2}
187: \sum^{\prime}_{i,j} \frac{1}{|{\bf r}_{i}-{\bf r}_{j}|}$ are the
188: kinetic energy, local external potential energy, and
189: electron-interaction potential energy operators, $\Psi({\bf X})$
190: and $E$ are the ground state wave function and energy, with $
191: {\bf X}={\bf x}_{1}, {\bf x}_{2}, \dots,{\bf x}_{N} $, ${\bf
192: x}={\bf r}\sigma$, and ${\bf r}$ and $\sigma$ the spatial and
193: spin coordinates. The ground state electronic density
194: is the expectation
195: \begin{equation}
196: \rho({\bf r}) = \left\langle \Psi|\hat{\rho}|\Psi\right\rangle,
197: \end{equation}
198: where $\hat{\rho} = \sum_{i}\delta(\bf{r}-{\bf r}_{i})$ is the
199: Hermitian density operator. The corresponding spinless single
200: particle density matrix is the expectation
201: \begin{equation}
202: \gamma({\bf r}{\bf r}^{\prime})= \left\langle \Psi|\hat{ \gamma
203: }|\Psi\right\rangle,
204: \end{equation}
205: where the Hermitian operator $\hat{\gamma} = \hat{A}+i\hat{B}$,
206: $\hat{A}= \frac{1}{2}\sum_{j} [\delta({\bf r}_{j}-{\bf
207: r})T_{j}({\bf a})+ \delta({\bf r}_{j}-{\bf r}^{\prime})
208: T_{j}(-{\bf a})]$, $\hat{B}=-\frac{i}{2}\sum_{j} [\delta({\bf
209: r}_{j}-{\bf r})T_{j}({\bf a})- \delta({\bf r}_{j}-{\bf
210: r}^{\prime}) T_{j}(-{\bf a})]$, $T_{j}({\bf a})$ is a translation
211: operator, and ${\bf a}={\bf r}^{\prime}- {\bf r}$. The diagonal
212: matrix element of $\gamma({\bf r}{\bf r}^{\prime})$ is the
213: density: $\gamma({\bf r}{\bf r})=\rho({\bf r})$. The ground state
214: energy is the expectation
215: \begin{equation}
216: E=\left\langle \Psi|\hat{ T }+\hat{V}+\hat{U}|\Psi\right\rangle =T+E_{ext}+E_{ee},
217: \end{equation}
218: with the kinetic energy $T=\left\langle
219: \Psi|\hat{T}|\Psi\right\rangle$, the external potential energy
220: $E_{ext}=\left\langle \Psi|\hat{V}|\Psi\right\rangle=\int
221: \rho({\bf r})v({\bf r}) d {\bf r}$, and the electron-interaction
222: energy $E_{ee}=\left\langle \Psi|\hat{U}|\Psi\right\rangle$. The
223: ionization potential is $I=E^{ion}-E$, where $E^{ion}$ is the
224: energy of the resulting ion when the
225: least bound electron is removed to infinity.\\
226:
227: The differential equation for the S-system in its ground state
228: that leads to the same density $\rho({\bf r})$ as that of the
229: electrons is
230: \begin{equation}
231: [-\frac{1}{2}\nabla^{2}+v({\bf r}) +
232: v_{ee}({\bf r})]\phi_{i}({\bf x})=\epsilon_{i}\phi_{i}
233: ({\bf x}); \;\;\;\; i=1,\dots,N,
234: \end{equation}
235: where $v_{ee}({\bf r})$ is the electron-interaction potential
236: energy of the noninteracting fermions. The S system wave function
237: is the Slater determinant $\Phi\{\phi_{i}({\bf x})\}$ of the
238: orbitals $\phi_{i}({\bf x})$, so that the density is the
239: expectation
240: \begin{equation}
241: \rho({\bf r})=\left\langle \Phi \{\phi_{i}\}|
242: \hat{\rho}|\Phi\{\phi_{i}\} \right\rangle =\sum_{i} \sum_{\sigma}
243: |\phi_{i}({\bf r}\sigma)|^{2},
244: \end{equation}
245: and the corresponding spinless Dirac density matrix
246: is the expectation
247: \begin{equation}
248: \gamma_{s}({\bf r}{\bf r}^{\prime})= \left\langle \Phi
249: \{\phi_{i}\}| \hat{{\bf \gamma}}|\Phi\{\phi_{i}\} \right\rangle
250: =\sum_{i} \sum_{\sigma} \phi_{i}^{*}({\bf r}\sigma)\phi_{i}({\bf
251: r}^{\prime}\sigma).
252: \end{equation}
253: \\
254:
255: The potential energy $v_{ee}({\bf r})$ is the work done to move
256: the model fermion from a reference point at infinity to its
257: position at ${\bf r}$ in the force of a \textit{conservative}
258: effective field $\bm{\mathcal{F}}^{e\!f\!f}({\bf r})$:
259: \begin{equation}
260: v_{ee}({\bf r}) = - \int_{\infty}^{{\bf r}}
261: \bm{\mathcal{F}}_{k}^{e\!f\!f}({\bf r}^{\prime})\cdot d{\bf l}^{\prime}.
262: \end{equation}
263: The field $\bm{\mathcal{F}}_{k}^{e\!f\!f}({\bf r})$ is the sum
264: of an electron-interaction field $\bm{\mathcal{E}}_{ee}({\bf r})$
265: representative of Pauli and Coulomb correlations, and a
266: Correlation-Kinetic field $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$ that is representative of the
267: correlation contribution to the kinetic energy:
268: \begin{equation}
269: \bm{\mathcal{F}}^{eff}({\bf r})=
270: \bm{\mathcal{E}}_{ee}({\bf r})+\bm{\mathcal{Z}}_{t_{c}}({\bf r}).
271: \end{equation}
272: The field $\bm{\mathcal{E}}_{ee}({\bf r})$ is obtained via
273: Coulomb's law from its quantal source $g({\bf r}{\bf r}^{\prime})$, the pair-correlation density. Thus,
274: \begin{equation}
275: \bm{\mathcal{E}}_{ee}({\bf r})=\int
276: \frac{g({\bf r}{\bf r}^{\prime})
277: ({\bf r}-{\bf r}^{\prime})}{|{\bf r}-{\bf r}^{\prime}|^{3}}
278: d{\bf r}^{\prime},
279: \end{equation}
280: where $g({\bf r}{\bf r}^{\prime}) = <\Psi|\hat{P}({\bf r}{\bf
281: r}^{\prime})|\Psi>/\rho({\bf r})$, and $\hat{P}({\bf r}{\bf
282: r}^{\prime})=\sum_{i,j}^{\prime} \delta({\bf r}_{i}-{\bf
283: r})\delta({\bf r}_{j}-{\bf r}^{\prime})$
284: the Hermitian pair-correlation operator. The
285: pair-correlation density may be further separated into
286: its local(static) and nonlocal (dynamic) components as
287: \begin{eqnarray}
288: g({\bf r}{\bf r}^{\prime})&=& \rho({\bf r}^{\prime})+
289: \rho_{xc}({\bf r}{\bf r}^{\prime})\\
290: &=&\rho({\bf r}^{\prime})+\rho_{x}({\bf r}{\bf r}^{\prime})+
291: \rho_{c}({\bf r}{\bf r}^{\prime}),
292: \end{eqnarray}
293: where the sources $ \rho_{xc}({\bf r}{\bf r}^{\prime})$,
294: $\rho_{x}({\bf r}{\bf r}^{\prime})$, and
295: $\rho_{c}({\bf r}{\bf r}^{\prime})$ are the Fermi-Coulomb,
296: Fermi, and Coulomb hole charge distributions. The Fermi hole
297: is defined as $\rho_{x}({\bf r}{\bf r}^{\prime})=
298: -|\gamma_{s}({\bf r}{\bf r}^{\prime})|^2/2\rho({\bf r})$,
299: and the Coulomb hole is defined via Eqs.(11) and (12).
300: The sum rules satisfied by these charge distributions
301: are $\int \rho_{xc}({\bf r}{\bf r}^{\prime})
302: d{\bf r}^{\prime}=-1$; $\int \rho_{x}({\bf r}
303: {\bf r}^{\prime}) d{\bf r}^{\prime}=-1$;
304: $\rho_{x}({\bf r}{\bf r}^{\prime})\leq 0$; $\rho_{x}({\bf r}{\bf r})=
305: -\rho({\bf r})/2$, and $\int \rho_{c}({\bf r}{\bf r}^{\prime})
306: d{\bf r}^{\prime}=0$. With the above separation,
307: the electron-interaction field may
308: then be written in terms of its components as
309: \begin{eqnarray}
310: \bm{\mathcal{E}}_{ee}({\bf r})&=& \bm{\mathcal{E}}_{H}({\bf r})
311: + \bm{\mathcal{E}}_{xc}({\bf r})\\
312: &=& \bm{\mathcal{E}}_{H}({\bf r})+\bm{\mathcal{E}}_{x}({\bf r})
313: + \bm{\mathcal{E}}_{c}({\bf r}),
314: \end{eqnarray}
315: where the Hartree $\bm{\mathcal{E}}_{H}({\bf r})$, Pauli-Coulomb
316: $\bm{\mathcal{E}}_{xc}({\bf r})$, Pauli
317: $\bm{\mathcal{E}}_{x}({\bf r})$, and Coulomb
318: $\bm{\mathcal{E}}_{c}({\bf r})$ fields are due to
319: their respective quantal sources $\rho({\bf r})$,
320: $ \rho_{xc}({\bf r}{\bf r}^{\prime})$, $\rho_{x}({\bf r}{\bf r}^{\prime})$,
321: and $\rho_{c}({\bf r}{\bf r}^{\prime})$.\\
322:
323: The Correlation-Kinetic field $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$
324: is the difference of the kinetic fields $\bm{\mathcal{Z}}({\bf r})$
325: and $\bm{\mathcal{Z}}_{s}({\bf r})$ of the interacting and noninteracting systems, respectively:
326: \begin{equation}
327: \bm{\mathcal{Z}}_{t_{c}}({\bf r})=
328: \bm{\mathcal{Z}}_{s}({\bf r})-\bm{\mathcal{Z}}({\bf r}),
329: \end{equation}
330: where $\bm{\mathcal{Z}}({\bf r})={\bf z}({\bf r};
331: [\gamma])/\rho({\bf r})$
332: and $\bm{\mathcal{Z}}_{s}({\bf r})={\bf z}_{s}({\bf r};
333: [\gamma_{s}])/\rho({\bf r})$. The quantal sources
334: of the fields $\bm{\mathcal{Z}}({\bf r})$ and
335: $\bm{\mathcal{Z}}_{s}({\bf r})$
336: are the single particle and Dirac density matrices.
337: The kinetic `force' ${\bf z}({\bf r};[\gamma])$ is defined in terms
338: of its components as $z_{\alpha}({\bf r};
339: [\gamma])=2\sum_{\beta}\partial t_{\alpha\beta}({\bf r};
340: [\gamma])/\partial r_{\beta}$, where $t_{\alpha\beta}({\bf r};
341: [\gamma])=\frac{1}{4}[\partial^{2}/\partial r_{\alpha}^{\prime}
342: \partial r_{\beta}^{\prime\prime}+\partial^{2}/\partial r_{\beta}^
343: {\prime}
344: \partial r_{\alpha}^{\prime\prime}]\gamma
345: ({\bf r}^{\prime}{\bf r}^{\prime\prime})|_{{\bf r}^{\prime}=
346: {\bf r}^{\prime\prime}
347: ={\bf r}}$ is the kinetic energy tensor. The field ${\bf z}_{s}({\bf r};
348: [\gamma_{s}])$ is similarly defined in terms of the S-system tensor
349: $t_{\alpha\beta,s}({\bf r};[\gamma_{s}])$.\\
350:
351: The Hartree field $\bm{\mathcal{E}}_{H}({\bf r})$ is conservative,
352: and $\nabla \times \bm{\mathcal{E}}_{H}({\bf r})=0$. This is because
353: its source $\rho({\bf r})$ is a static charge, and the field may
354: consequently be written as $\bm{\mathcal{E}}_{H}({\bf r})=-\nabla W_{H}({\bf r})$,
355: where $W_{H}({\bf r})=\int d{\bf r'} \rho({\bf r'})/|{\bf r}-{\bf r'}|$. The
356: fields $\bm{\mathcal{E}}_{xc}({\bf r})$, $\bm{\mathcal{E}}_{x}({\bf r})$, and
357: $\bm{\mathcal{E}}_{c}({\bf r})$
358: are in general not conservative as their sources
359: are nonlocal.
360: The sum of the fields
361: $\bm{\mathcal{E}}_{xc}({\bf r})+\bm{\mathcal{Z}}_{t_{c}}({\bf
362: r})$ and $
363: \bm{\mathcal{E}}_{x}({\bf r})+\bm{\mathcal{E}}_{c}({\bf r})+\bm{\mathcal{Z}}_{t_{c}}({\bf r})$ are always conservative.\\
364:
365: For systems of symmetry such that the component fields
366: $\bm{\mathcal{E}}_{ee}({\bf r})$ and
367: $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$ are separately conservative,
368: the potential energy $v_{ee}({\bf r})$ may be expressed as
369: the sum of the separate work done in these fields. Thus
370: \begin{eqnarray}
371: v_{ee}({\bf r})&=& W_{ee}({\bf r})+W_{t_c}({\bf r})\\
372: &=&W_{H}({\bf r})+W_{xc}({\bf r})+W_{t_{c}}({\bf r})\\
373: &=& W_{H}({\bf r})+W_{x}({\bf r})+W_{c}({\bf r})+W_{t_{c}}({\bf r}),
374: \end{eqnarray}
375: where $W_{ee}({\bf r})$, $W_{H}({\bf r})$, $W_{xc}({\bf r})$,
376: $W_{x}({\bf r})$, $W_{c}({\bf r})$, and $W_{t_{c}}({\bf r})$
377: are respectively the work done in the fields
378: $\bm{\mathcal{E}}_{ee}({\bf r})$, $\bm{\mathcal{E}}_{H}({\bf r})$,
379: $\bm{\mathcal{E}}_{xc}({\bf r})$, $\bm{\mathcal{E}}_{x}({\bf r})$,
380: $\bm{\mathcal{E}}_{c}({\bf r})$, and
381: $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$.\\
382:
383: The ground state energy is
384: \begin{equation}
385: E=T_{s}+\int \rho({\bf r})v({\bf r})d{\bf r}+E_{ee}+T_{c},
386: \end{equation}
387: where $T_{s} = \left\langle \Phi\{\phi_{i}\}|\hat{T}|\Phi\{\phi_{i}\}
388: \right\rangle$ is the kinetic energy of the noninteracting Fermions.
389: The electron-interaction $E_{ee}$ and Correlation-Kinetic $T_{c}$
390: energies are expressed in terms of the fields
391: $\bm{\mathcal{E}}_{ee}({\bf r})$
392: and $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$, respectively,
393: in integral virial form as
394: \begin{equation}
395: E_{ee}=\int d{\bf r}\rho({\bf r})
396: {\bf r}\cdot\bm{\mathcal{E}}_{ee}({\bf r})
397: \;\;\;\; {\textstyle and}\;\; T_{c} =\frac{1}{2}\int d{\bf r}
398: \rho({\bf r}){\bf r}\cdot \bm{\mathcal{Z}}_{t_{c}}({\bf r}).
399: \end{equation}
400: These expressions for the energy are valid whether the fields
401: $\bm{\mathcal{E}}_{ee}({\bf r})$ and
402: $\bm{\mathcal{Z}}_{t_{c}}({\bf r})$ are separately conservative or not. Employing Eq.(13)
403: and (14) in Eq.(20), the energy $E_{ee}$ may be written as a sum of the Hartree
404: $E_{H}$ and Pauli-Coulomb $E_{xc}$ (or Pauli $E_{x}$ plus Coulomb $E_{c}$) energies with each component term expressed in integral virial form.\\
405:
406: Finally, the highest occupied eigenvalue of the S system differential equation
407: is the negative of the ionization potential: $\epsilon_{m}= -I$.\\
408:
409:
410: \section{ APPLICATION TO THE HYDROGEN MOLECULE}
411:
412: \textbf{A. Wave functions, Orbitals, and Density}\\
413:
414: The purely electronic part of the Hamiltonian for $H_{2}$ in atomic units ($e=m=\hbar=1$) is
415: \begin{equation}
416: \hat{H}= -\frac{1}{2} \nabla_{1}^{2}- \frac{1}{2}
417: \nabla_{2}^{2}-\frac {1} {r_{1a}}- \frac {1}{r_{2a}}-\frac {1}{r_{1b}}-\frac {1}{r_{2b}}+\frac {1}
418: {r_{12}}
419: \end{equation}
420: where $1$ and $2$ are the electrons, and $a$ and $b$ are the
421: nuclei.
422: As the wave function of the molecule in its ground state is unknown, we employ
423: the essentially exact $51$-parameter correlated wave function of Kolos-Roothaan \cite{11}
424: in our calculations. The symmetric spatial part of the wave
425: function is
426: \begin{equation}
427: \Psi({\bf r}_{1} {\bf r}_{2})= exp[-\delta (\xi_{1} +\xi_{2})]
428: \sum_{m n j k l} C_{mnjkp} \;[\xi_{1}^{m} \xi_{2}^{n} \eta_{1}^{j}
429: \eta_{2}^{k} + \xi_{1}^{n} \xi_{2}^{m} \eta_{1}^{k} \eta_{2}^{j}]
430: r_{12}^{p}
431: \end{equation}
432: with
433: \begin{equation}
434: \xi_{1}=(r_{1a}+r_{1b})/R; \;\; \xi_{2}=(r_{2a}+r_{2b})/R ;
435: \end{equation}
436: \begin{equation}
437: \eta_{1}=(r_{1a}-r_{1b})/R; \;\; \eta_{2}=(r_{2a}-r_{2b})/R,
438: \end{equation}
439: where the variational parameters are $\delta$ and the coefficients
440: $C_{mnjkp}$, $r_{12}=|{\bf r}_{1}-{\bf r}_{2}|$, and $R=2a$ is
441: the internucleus separation. The values of the variational
442: parameters are given in the Appendix. The total energy
443: (inclusive of the internuclear potential energy
444: $V_{nn}=1/R$) is $E_{tot}(H_{2})=-1.174448$ (a.u.) at $a=0.7005$
445: (a.u.). The kinetic energy $T=-E_{tot}$, and the total potential
446: energy $E_{ext}+E_{ee}+V_{nn}=-2.348851 $(a.u.). The virial
447: theorem ratio, which is the ratio of the total potential energy to
448: twice the total energy, is $0.999981$. The electron interaction
449: energy component $E_{ee}=0.58737$(a.u.), and the external energy
450: $E_{ext}=-3.65005$(a.u.). The total energy \cite{18} of the
451: Hydrogen molecular ion $H_{2}^{+}$ at the equilibrium internuclear
452: separation of the Hydrogen molecule is
453: $E_{tot}(H_{2}^{+})|_{a=0.7005}=-0.56998$ (a.u.). Thus, the
454: ionization potential
455: of the $H_{2}$ molecule is $I=E_{tot}(H_{2}^{+})|_{a=0.7005}-E_{tot}(H_{2})=0.60447$ (a.u.).\\
456:
457: For two electron systems such as the Hooke's atom\cite{19}, Helium atom, or the Hydrogen molecule,
458: the orbitals of the S system in its ground (singlet) state that lead to the interacting system
459: density are known. These orbitals are $\phi_{i}({\bf r})=\sqrt{\rho({\bf r})/2}, i=1,2$, and are therefore
460: known to the same accuracy as the wave function or density.\\
461:
462: The density $\rho(0,z)$ along the nuclear bond z-axis is plotted in Fig.1. The density is extremely accurate
463: throughout space except at and very near each nucleus. Thus, although on the scale of this figure, it appears that
464: the density satisfies the electron-nucleus cusp
465: condition\cite{20} exactly, in fact it does not.\\
466:
467: \textbf{B. Fermi-Coulomb, Fermi, and Coulomb Holes }\\
468:
469:
470: For the $H_{2}$ molecule in its singlet ground state, there are no correlations due to the Pauli
471: exclusion principle as the two electrons have opposite spin. However, within the S system framework,
472: it is customary in local effective potential energy theories to define a Fermi
473: hole as $\rho_{x}({\bf r} {\bf r'}) = -\rho({\bf r'})/2$.
474: (This is because the pair-correlation density as determined from the corresponding S system wave
475: function is $g({\bf r} {\bf r'}) = \rho({\bf r'}) /2$.) \\
476:
477: In Fig.2 we plot cross-sections through the Fermi-Coulomb
478: $\rho_{xc}({\bf r} {\bf r'})$, Fermi $\rho_{x}({\bf r} {\bf r'})$,
479: and Coulomb $\rho_{c}({\bf r} {\bf r'})$ hole sources as a
480: function of ${\bf r'} = (0,z')$ for an electron at the origin
481: ${\bf r} = (0,0)$ at the center of the nuclear bond. (Because of
482: the cylindrical symmetry of the molecule, cylindrical coordinates
483: are employed throughout.) The electron position is indicated by
484: the arrow. The three charge distributions, of course, have
485: cylindrical symmetry about the bond axis. More significantly,
486: they are symmetrical about the
487: electron along the $z'$ axis. Observe that at the electron position, both the Fermi-Coulomb and
488: Coulomb holes exhibit a cusp corresponding to
489: the electron-electron cusp condition \cite{20}. (Based on the work of Ref. \cite{21} it is
490: known that the wave function does not satisfy this cusp condition exactly. It obviously satisfies it
491: to a good degree as evidenced by the figure.) As expected, at the electron position, the Fermi-Coulomb hole is more negative than the Fermi hole. Thus, in the region about the electron, the Coulomb hole is negative. (This is also the case for all the other electron positions considered.) As both the Fermi-Coulomb and Fermi holes satisfy the same charge conservation sum rule, there must then be regions where the former lies above the latter. This is clearly evident in the figure. Hence, in the outer and classically forbidden regions of the molecule, the Coulomb hole is positive. (The positive part of the Coulomb hole is more clearly evident in the figures that follow.) The Coulomb hole is both positive and negative as its total charge is zero. The positive part of the Coulomb hole is an indication that the other electron is equally likely to be in the classically forbidden region on either side of each nucleus. \\
492:
493:
494:
495: As the Fermi hole is independent of electron position, we now focus on the Fermi-Coulomb and
496: Coulomb holes. In Figs.3-5, we plot the cross-sections of these holes for electron positions
497: at ${\bf r} = (0, a/3), {\bf r} = (0, 2a/3), {\bf r} = (0, a)$. Again, observe the cusp at
498: the electron position for both the Fermi-Coulomb and Coulomb holes of each figure.
499: Note also how the positive part of the Coulomb hole becomes more pronounced relative to
500: the negative part as the electron is moved away from the center of the nuclear bond towards
501: one nucleus. The positive part of the Coulomb hole is also largest about the other nucleus,
502: thereby indicating that the second electron is about this nucleus. \\
503:
504:
505: In Figs.6-8, we plot the Fermi-Coulomb and Coulomb hole
506: cross-sections for an electron in the classically forbidden region
507: at ${\bf r} = (0, 2a), {\bf r} = (0, 4a)$, and ${\bf r} = (0,
508: 6a)$. The positive part of the Coulomb hole continues to increase
509: about the left nucleus at the expense of the negative part as the
510: electron is moved further from the molecule. Thus, even for the
511: asymptotic
512: position of an electron at ${\bf r} = (0, 6a)$, the other electron is still mainly about the left nucleus.
513: For all electron positions, the Fermi-Coulomb hole $\rho_{xc}({\bf r} {\bf r'})$ is negative.\\
514:
515: (We note that the same cross-sections of the Fermi-Coulomb, Fermi, and Coulomb holes for an
516: electron position $0.3$ (a.u.) to the left of the right nucleus, which corresponds
517: approximately to our Fig.4, has been plotted by Baerends et al \cite{22} in their study
518: of the dissociation of the
519: molecule. However, in their figure, the electron-electron cusp in the
520: Fermi-Coulomb and Coulomb holes is not present because the wave function employed by these
521: authors is a configuration-interaction type wave function.)\\
522:
523:
524:
525: %section III .B ends
526:
527: \textbf{C. Fields, Potential Energies, and Energies.} \\
528:
529: The electron-interaction field ${\mathcal E}_{ee}({\bf r})$, and
530: its Hartree ${\mathcal E}_{H}({\bf r})$ and Pauli-Coulomb
531: ${\mathcal E}_{xc}({\bf r})$ components along the nuclear bond
532: axis are plotted in Fig.9.
533: Observe that these fields all vanish at the center of the bond axis or origin. This is because their
534: corresponding sources -- the pair-correlation density $g({\bf r} {\bf r'})$,
535: the density $\rho({\bf r})$, and the Fermi-Coulomb hole
536: charge $\rho_{xc}({\bf r} {\bf r'} )$ --- are symmetrical
537: about the center of the nuclear bond for this electron
538: position (see Figs.1 and 2). The existence (non-zero value) of these
539: fields for all other electron positions is a consequence of the fact
540: that their sources are \textit{not symmetrical} about the
541: electron (see Figs.1 and 3-8). The fields are
542: also all \textit{antisymmetric} about the
543: center of the nuclear bond.
544: (This is a reflection of the symmetry about
545: the x-y plane at the center of the nuclear bond.
546: As such the potential energies obtained from these
547: fields will be \textit{symmetric} about this point.)
548: In the positive half-space, there is a maximum in the
549: electron-interaction and Hartree fields, and a minimum in
550: the Pauli-Coulomb field. The Hartree and Pauli-Coulomb fields
551: are of the same order of magnitude and opposite in sign.
552: This is because their sources, $\rho({\bf r})$ and $\rho_{xc}({\bf r} {\bf r'})$
553: respectively, are of the same order of magnitude and opposite in sign.
554: Asymptotically, in the z direction these fields decay
555: as ${\mathcal E}_{ee}({\bf r}) \sim 1/z^{2} $,
556: ${\mathcal E}_{H}({\bf r}) \sim 2/z^{2} $, and
557: ${\mathcal E}_{xc}({\bf r}) \sim -1/z^{2}$ as they must \cite{3,9}. (It is interesting to note that with a slight translation to the right, the plots of the fields in the positive half-space, are strikingly similar to those of the Helium atom \cite{2,9}.)\\
558:
559:
560:
561: The Pauli ${\mathcal E}_{x}({\bf r})$ and Coulomb ${\mathcal
562: E}_{c}({\bf r})$ field components of the Pauli-Coulomb field
563: ${\mathcal E}_{xc}({\bf r})$ along the nuclear bond axis are
564: plotted in Fig.10. Again, these fields vanish at the origin and
565: are \textit{antisymmetric} about it. Hence, the corresponding
566: potential energies obtained from these fields will be symmetric.
567: In the positive half space, the Pauli field ${\mathcal E}_{x}({\bf
568: r})$ is negative as its source
569: is a negative charge. The Coulomb field ${\mathcal E}_{c}({\bf r})$, on the other hand, is positive in the
570: inter-nuclear region and negative throughout the region beyond the right nucleus. This structure
571: is attributable to the fact that the Coulomb hole has both a positive and negative component.
572: Asymptotically, the Pauli field decays as ${\mathcal E}_{x}({\bf r}) \sim -1/z^{2} $,
573: whereas the Coulomb field ${\mathcal E}_{c}({\bf r})$ has essentially vanished by about $z = 5$
574: (a.u.).
575: (Once again in the positive half-space, the structure of these fields when translated slightly to the
576: right, is similar to those of the Helium atom. In particular, we note that the structure of the
577: Coulomb holes of the Hydrogen molecule for electron positions $z > a$ (see Figs. 6-8) is very
578: similar to those of the Helium atom for electron positions away from its nucleus (see Figs.3,4 of \cite{9}).)
579: As is the case for atoms, it turns out that the asymptotic structure along the nuclear
580: bond axis of $\{{\mathcal F}^{eff}({\bf r}) - {\mathcal E}_{H}({\bf r})\} \sim {\mathcal E}_{x}({\bf r})
581: \sim -1/z^{2}$ . Thus, the asymptotic structure of the electron-interaction potential
582: energy $v_{ee}({\bf r})$ minus the Hartree potential energy $W_{H}({\bf r})$ is again due
583: to Pauli correlations: $\{v_{ee}({\bf r}) - W_{H}({\bf r})\} \sim W_{x}({\bf r}) \sim - 1/z $ as shown
584: in Fig. 11. \\
585:
586:
587:
588: Since in the S system description of two electron systems ${\mathcal E}_{x}({\bf r}) = - {\mathcal E}_{H}({\bf r})/2$,
589: the curl of the Fermi field along the nuclear bond z axis
590: direction vanishes: $\nabla \times {\mathcal E}_{x}({\bf r})|_{z}= 0$, as it does in all directions.
591: Hence, the work done $W_{x}(0,z)$ plotted in Fig. 11 is path independent.
592: Along the nuclear bond axis, however, the $\nabla \times {\mathcal E}_{c}({\bf r})|_{z}\neq 0$
593: and $\nabla \times {\mathcal Z}_{t_{c}}({\bf r})|_{z}\neq 0$ . But in this and all directions,
594: the curl of the sum of the fields ${\mathcal E}_{c}({\bf r})$ and ${\mathcal Z}_{t_{c}}({\bf r})$
595: vanishes: $\nabla \times [{\mathcal E}_{c}({\bf r}) + {\mathcal Z}_{t_{c}}({\bf r})])|_{z} = 0$.
596: Therefore, the work done in the sum of these fields in all directions, and hence along the nuclear
597: bond axis $v_{c}(0,z) = W_{c}(0,z) + W_{t_{c}}(0,z)$ is path independent.
598: The calculation of the potential energy $v_{c}(0,z)$ is straightforward. However,
599: our use of the Kolos-Roothaan wave function, in spite of its accuracy, leads to $v_{c}(0,z)$ being
600: singular at the nucleus. This occurs due to the component $W_{t_{c}}(0,z)$ that requires a
601: cancellation of the kinetic fields of the interacting and noninteracting systems.
602: The underlying reason for the singularity, however, is that the wave function does not
603: satisfy the \textit{electron-nucleus} cusp condition exactly. In a recent paper \cite{23},
604: we have proved by employing the integral form of the electron-nucleus cusp condition \cite{24}, that
605: in local effective potential energy theories and for arbitrary symmetry,
606: the potential energy $v_{ee}({\bf r})$ is finite at the nucleus. Furthermore,
607: it is shown that this finiteness is a direct consequence of the satisfaction of the
608: electron-nucleus cusp condition by the Schrodinger wave
609: function. (As a consequence, for example, this potential
610: energy is singular at each nucleus when determined either
611: from Gaussian geminal \cite{23} or configuration interaction
612: \cite{25} wave functions.) Hence, in order to obtain $v_{c}(0,z)$, we have employed
613: our calculated results in regions other than near the
614: nucleus, and smoothed the curve through each nucleus.( A comparison of our results with
615: the work of Gritsenko \emph{et al}\cite{26} who in their self-consistent calculations assumed
616: $v_{ee}({\bf r})$ to be finite at the nucleus show the two curves to be indistinguishable throughout space.)
617: The potential
618: energy $v_{c}(0,z)$ is plotted in Fig.12. Observe that $v_{c}(0,z)$, and thus the sum of
619: the Coulomb and Correlation-Kinetic potential energies is an order of magnitude
620: smaller than $W_{x}(0,z)$, the Pauli contribution. The potential energy $v_{c}(0,z)$ has
621: considerable structure, is symmetric about the origin, and is mainly positive,
622: indicating thereby that its principal contribution is Correlation-Kinetic.
623: (Recall that the Coulomb field is principally negative in the right-half space
624: (see Fig. 10) so that the Coulomb potential energy $W_{c}({\bf r})$ is negative.) The plot of
625: $v_{c}(0,z)$ translated to the right nucleus is very similar in shape and magnitude to the
626: corresponding potential energy $v_{c}({\bf r})$ of the Helium atom (see Fig. 4 of \cite{2}). \\
627:
628:
629: To obtain a quantitative sense of the separate Coulomb and Correlation-Kinetic
630: contributions to $v_{c}(0,z)$, we plot in both Figs.11 and 12 the work done
631: $W_{c}(0,z)$ along the path of the nuclear bond in the force of the Coulomb
632: field ${\mathcal E}_{c}({\bf r})$. From $v_{c}(0,z)$ and $W_{c}(0,z)$ we
633: obtain $W_{t_{c}}(0,z)$ which is also plotted in Fig. 12. The corresponding
634: Correlation-Kinetic field $Z_{t_{c}}(0,z)$ is shown in Fig.13. Note that
635: this field too is antisymmetric about the origin. Once again, there is a striking similarity
636: between the plots of $W_{c}(0,z)$, $Z_{t_{c}}(0,z)$, and $W_{t_{c}}(0,z)$ when translated to
637: the right nucleus to those of the corresponding properties of the Helium atom \cite{2}.
638: The Coulomb correlation part $W_{c}(0,z)$ is negative throughout space and vanishes by about
639: $z = 5 $ (a.u.). The Correlation-Kinetic piece $W_{t_{c}}(0,z)$ is throughout
640: positive and asymptotically decays more slowly. The field $Z_{t_{c}}(0,z)$ is principally
641: positive throughout space. Thus, the Correlation-Kinetic energy $T_{c}$ is
642: positive: $T = 1.1745$ (a.u.), $T_{s} = 1.1414$ (a.u.), $T_{c} = 0.0331$ (a.u.)
643: (The corresponding value of $T_{c}$ for the Helium atom is $0.0365$ (a.u.)\cite{9}).
644: (We note that the $W_{c}(0,z)$ and $W_{t_{c}}(0,z)$ do not each separately represent a potential energy.
645: Their sum which is $v_{c}(0,z)$ does.)\\
646:
647:
648: \section { \textbf{CONCLUDING REMARKS}}
649:
650: This paper is the first application of the Q-DFT quantal source and field perspective
651: to a molecule, and much has been learned as explained in the previous section.
652: The symmetry of the $H_{2}$ molecule dictates that the individual fields ${\mathcal E}_{x} ({\bf r})$,
653: ${\mathcal E}_{H} ({\bf r})$, ${\mathcal E}_{c} ({\bf r})$, ${\mathcal Z}_{t_{c}}(r)$ representative of the
654: Pauli and Coulomb correlations, and Correlation-Kinetic effects respectively, must each be antisymmetric
655: about the center of the nuclear bond. The corresponding electron-interaction potential
656: energy $v_{ee}({\bf r})$ representative of these correlations as determined by the work done
657: in the force of these fields is then symmetric about this point as also dictated by the
658: symmetry of the molecule. The potential energy $v_{ee}({\bf r})$ is also finite at each nucleus,
659: as must be the case \cite{23}. The Hartree ${\mathcal E}_{H} ({\bf r})$ and Pauli ${\mathcal E}_{x} ({\bf r})$
660: fields are the largest in magnitude and opposite in sign, the former being positive
661: and twice as large as the latter. As such the principal contributions to the electron-interaction
662: energy $E_{ee}$ and potential energy $v_{ee}({\bf r})$ are due
663: to the Hartree and the Pauli correlation terms. The Coulomb ${\mathcal E}_{c} ({\bf r})$ and
664: Correlation-Kinetic ${\mathcal Z}_{t_{c}}(r)$ fields tend to cancel each other,
665: so that the contribution of their sum to the potential energy $v_{ee}({\bf r})$ is an
666: order of magnitude smaller. However, as the potential energy component $v_{c}({\bf r})$ representing
667: the sum of these correlations is principally positive (see Fig. 12), it is evident that the
668: Correlation-Kinetic effects are more significant. They are also more significant asymptotically,
669: where the Coulomb correlation contributions to the potential energy vanish. Thus,
670: Correlation-Kinetic effects play an important role in local effective potential energy
671: theories of the $H_{2}$ molecule.
672: We further note that in the construction of approximate KS-DFT `exchange-correlation' and correlation energy
673: functionals $E^{KS}_{xc}[{\bf r}]$ and $E^{KS}_{c}[{\bf r}]$ for molecules, Correlation-Kinetic
674: effects must be
675: incorporated if an accurate S system representation of molecules is to be obtained.\\
676:
677: On the basis of the Q-DFT results determined from the
678: $H_{2}$ molecule, it is evident that the qualitative
679: features of the quantal sources, fields, and potential
680: energies for other diatomic molecules will be similar.
681: However, the fields and hence the potential energies of
682: these diatomics will have more structure as a
683: consequence of the additional molecular subshells. We
684: expect this added structure to be similar to that
685: observed in atoms as the number of shells is increased.
686: Finally, we note that the accuracy of approximation
687: methods within Q-DFT \cite{27} and KS-DFT can be tested
688: by comparison with these essentially exact results.\\
689:
690: We conclude by reiterating the striking similarity between the Q-DFT properties of
691: the Hydrogen molecule and the Helium atom for electron positions in the positive half space.
692: It is interesting that in spite of the presence of a second nucleus, and therefore of a different
693: symmetry, the quantal sources and fields representative of the various electron correlations
694: in the Hydrogen molecule are so similar to those of the Helium atom. This
695: speaks to the commonality of properties of these distinct quantum systems as exhibited within the framework of Q-DFT. \\
696:
697:
698: \begin{acknowledgements}
699:
700: We thank Prof. E. J. Baerends and Dr. Myrta Gr{\"u}ning for
701: sending us the results of Ref. \cite{26}. This work was supported
702: in part by the Research Foundation of the City University of New
703: York.
704: %\end{}
705: \end{acknowledgements}
706: \newpage
707:
708:
709: %\section{\textbf{APPENDIX}}
710:
711:
712: \appendix
713: \textbf{APPENDIX: Wave function parameters}\\
714: The values of the parameter $\delta$ and the coefficients
715: $c_{mnjkp}$ for the wave function of Eq.(21) are listed in the
716: table I.
717:
718: %\begin{minipage} [t]{8cm}
719: \begin{table}
720: \caption{\label{tab:table1} Variational parameters in the
721: normalized 51-parameter correlated wave function for the ground
722: state of $H_{2}$\cite{11}.}
723: \renewcommand{\arraystretch}{0.4}
724: %\begin{ruledtabular}
725: \begin{tabular}{ll@{\hspace{2mm }}ll@{\hspace{2mm }}l@{\hspace{1mm }}|r}
726: \hline \hline
727: \multicolumn{5}{c}{No. of terms} & $50 $\\
728: %\cline{2-3}
729:
730: \multicolumn{5}{r} {$\delta$ =} & $0.995 $\\ \hline
731: $\xi_{1}$ & $\eta_{1}$ & $\xi_{2}$ & $\eta_{2}$ & $ r_{12}$ & \multicolumn{1}{c} {Coefficients} \\ \hline
732: 0 &0 & 0 & 0 & 0 & $2.065908$ \\
733: 0 &0 & 0 & 2 & 0 & $1.282032$ \\
734: 0 &0 & 1 & 0 & 0 & $0.144619$ \\
735: 0 &1 & 0 & 1 & 0 & $-0.430253$ \\
736: 0 &0 & 0 & 0 & 1 & $0.787198$ \\
737:
738: 1 &1 & 0 & 1 & 0 & $-0.235454$ \\
739: 1 &0 & 0 & 2 & 0 & $0.148273$ \\
740: 0 &0 & 2 & 0 & 0 & $0.109859$ \\
741: 0 &0 & 0 & 0 & 2 & $-0.212159$ \\
742: 1 &0 & 1 & 0 & 0 & $-0.081387$ \\
743:
744: 0 &2 & 0 & 2 & 0 & $0.182892$ \\
745: 0 &0 & 0 & 2 & 1 & $0.198555$ \\
746: 0 &0 & 1 & 0 & 1 & $0.324658$ \\
747: 1 &1 & 1 & 1 & 0 & $-0.010794$ \\
748: 0 &0 & 1 & 0 & 2 & $0.077830$ \\
749:
750: 1 &0 & 2 & 0 & 0 & $-0.055114$ \\
751: 0 &1 & 0 & 1 & 1 & $0.130714$ \\
752: 0 &1 & 0 & 1 & 2 & $-0.050854$ \\
753: 1 &0 & 2 & 0 & 1 & $0.014963$ \\
754: 0 &0 & 2 & 0 & 1 & $-0.132980$ \\
755:
756: 1 &1 & 1 & 1 & 2 & $0.000362$ \\
757: 0 &0 & 2 & 0 & 2 & $0.006992$ \\
758: 1 &0 & 0 & 2 & 1 & $-0.050940$ \\
759: 1 &1 & 1 & 1 & 1 & $0.018027$ \\
760: 1 &0 & 1 & 0 & 1 & $0.017554$ \\
761:
762: 0 &0 & 0 & 2 & 2 & $-0.014601$ \\
763: 1 &0 & 1 & 0 & 2 & $-0.015172$ \\
764: 1 &0 & 0 & 2 & 2 & $0.012656$ \\
765: 1 &2 & 3 & 0 & 0 & $-0.000202$ \\
766: 2 &0 & 3 & 0 & 0 & $-0.000856$ \\
767:
768:
769: 0 &0 & 1 & 2 & 0 & $-0.009469$ \\
770: 0 &0 & 3 & 0 & 0 & $0.036963$ \\
771: 1 &0 & 1 & 2 & 0 & $-0.022325$ \\
772: 0 &1 & 2 & 1 & 0 & $0.053233$ \\
773: 1 &0 & 3 & 0 & 0 & $0.004690$ \\
774:
775: 1 &2 & 1 & 2 & 0 & $0.004707$ \\
776: 1 &1 & 2 & 1 & 0 & $-0.017531$ \\
777: 0 &2 & 3 & 0 & 0 & $0.017270$ \\
778: 3 &0 & 3 & 0 & 0 & $0.000082$ \\
779: 2 &1 & 2 & 1 & 0 & $0.000031$ \\
780:
781: 0 &0 & 1 & 2 & 1 & $0.094436$ \\
782: 0 &0 & 3 & 0 & 1 & $0.001789$ \\
783: 0 &0 & 3 & 0 & 2 & $-0.000394$ \\
784: 0 &0 & 1 & 2 & 2 & $-0.004475$ \\
785: 2 &0 & 3 & 0 & 1 & $-0.000121$ \\
786:
787: 1 &0 & 1 & 2& 1 & $-0.014893$ \\
788: 2 &0 & 3 & 0 & 2 & $0.000011$ \\
789: 1 &0 & 1 & 2& 2 & $0.001016$ \\
790: 0 &2 & 3 & 0 & 1 & $-0.003443$ \\
791: 0 &2 & 3 & 0 & 2 & $0.000225$ \\ \hline \hline
792:
793: \end{tabular}
794: %\end{ruledtabular}
795: \end{table}
796: %\end{minipage}
797: %\end{appendxi}
798:
799:
800: \begin{references}
801: \bibitem{1}V. Sahni, Phys. Rev A {\bf 55}, 1846(1997).
802: \bibitem{2}V. Sahni, Top. Curr. Chem. {\bf 182}, 1(1996).
803: \bibitem{3} Z. Qian and V. Sahni, Phys. Rev. A {\bf 57}, 2527(1998).
804: \bibitem{4} Z. Qian and V. Sahni, Phys. Rev. B {\bf 62}, 16364(2000).
805: \bibitem{5} V. Sahni, L. Massa, R. Singh and M. Slamet, Phys. Rev. Lett.
806: {\bf 87} 113002(2001).
807: \bibitem{6} M. Slamet and V. Sahni, Int. J. Quantum Chem.
808: {\bf 85} 436(2001).
809: \bibitem{7} V. Sahni and X.-Y. Pan, Phys. Rev. Lett. \textbf{90}, 123001 (2003).
810: \bibitem{8} M. Slamet, R. Singh, L. Massa, and V. Sahni, (submitted to Phys. Rev. A).
811: \bibitem{9} M. Slamet and V. Sahni, Phys. Rev. A \textbf{51}, 2815 (1995).
812:
813: \bibitem{10} \textit{Atoms and atomic ions}: V. Sahni in \textit{Structure and Dynamics of Atoms and Molecules: Conceptual Trends}, edited by J. L. Calais and E. S. Kryacho, (Kluwer Academic Publishers, 1995); V. Sahni, Y. Li, and M. K. Harbola, Phys. Rev. A \textbf{45}, 1434 (1992). \textit{Atomic excited states}: R. Singh and B. M. Deb, Phys. Rep.\textbf{ 311}, 47 (1999). \textit{Positron binding}: R. R. Zope, Phys. Rev. A \textbf{60}, 218 (1999)\textit{. Metallic surfaces}: V. Sahni and A. Solomatin, Adv. Quantum Chem. \textbf{33}, 241 (1999). \textit{Metallic clusters}: M. K. Harbola, J. Chem. Phys. \textbf{97}, 2578 (1992).
814: \bibitem{11} W. Kolos and C. C. J. Roothaan, Rev. Mod. Phys. \textbf{32}, 219 (1960).
815: \bibitem{12} W. Kohn and L. J. Sham, Phys. Rev. \textbf{140}, A1133 (1965).
816: \bibitem{13} J. P. Perdew, R. G. Parr, M. Levy, and J. L. Balduz, Phys. Rev. Lett. \textbf{49}, 1691(1982); M. Levy, J. P. Perdew, and V. Sahni, Phys. Rev. A \textbf{30}, 2745 (1984); C.-O. Almbladh and U. von Barth, Phys. Rev. B \textbf{31}, 3231 (1985).
817: \bibitem{14} A. Gorling, Phys. Rev. A \textbf{46}, 3753 (1992).
818: \bibitem{15} Y. Wang and R. G. Parr, Phys. Rev. A \textbf{47}, R1591 (1993).
819: \bibitem{16} R. Van Leeuwen and E. J. Baerends, Phys. Rev. A \textbf{49}, 2421 (1994).
820: \bibitem{17} Q. Zhao, R. C. Morrison, and R. G. Parr, Phys. Rev. A \textbf{50}, 2138 (1994).
821: \bibitem{18} H. Wind, J. Chem. Phys. \textbf{42}, 2371 (1965); D. R. Bates, K. Ledsham, and A. L. Stewart, Phil. Trans. Roy. Soc. A \textbf{246}, 215 (1953).
822: \bibitem{19} M. Taut, Phys. Rev. A \textbf{48}, 3561 (1993).
823: \bibitem{20} T. Kato, Commun. Pure. Appl. Math. \textbf{10}, 151 (1957); E. Steiner, J. Chem. Phys. \textbf{39}, 2365 (1963).
824: \bibitem{21} W. Kolos and C. C. J. Roothaan, Rev. Mod. Phys.\textbf{32}, 205 (1960).
825: \bibitem{22} E. J. Baerends, Phys. Rev. Lett. \textbf{87}, 133004 (2001); E. J. Baerends and O. V. Gritsenko, J. Phys. Chem. A \textbf{101}, 5383 (1997); R. van Leeuwen, O. V. Gritsenko, and E. J. Baerends, Top. Curr. Chem. \textbf{180}, 107 (1996).
826: \bibitem{23} X.-Y. Pan and V. Sahni, Phys. Rev. A \textbf{67}, 012501 (2003).
827: \bibitem{24} W. A. Bingel, Z. Naturforsch. A \textbf{18a}, 1249 (1963); Theor. Chim. Acta (Berl) \textbf{8}, 54 (1967); R. T. Pack and W. Byers Brown, J. Chem. Phys. \textbf{45}, 556 (1966).
828: \bibitem{25} M. E. Mura, P. J. Knowles, and C. A. Reynolds, J.
829: Chem. Phys. \textbf{106}, 9659 (1997).
830: \bibitem{26} O. V. Gritsenko, R. van Leeuwen, and E. J. Baerends, Phys. Rev. A \textbf{52}, 1870 (1995).
831: \bibitem{27} R. Singh , L. Massa, and V. Sahni, Phys. Rev. A
832: \textbf{60}, 4135(1999).
833:
834:
835:
836:
837: \end{references}
838:
839:
840:
841: %\begin{figure}
842: %\begin{center}
843: %\includegraphics[bb=60 202 791 534, angle=0.5, scale=0.7]{fig_2.eps}
844: %\caption{A schematic representation of the Hohenberg-Kohn theorem and its corollary.\label{}}
845: %\end{center}
846: %\end{figure}
847:
848:
849: %
850: \begin{figure}
851: \includegraphics[bb= 1 2 500 658, angle=90,scale=0.8]{F_1.eps}
852: \caption{ The electron density $\rho(0,z)$ of the hydrogen
853: molecule along the nuclear bond axis in atomic units (a.u.). The
854: nuclei are on the axis at $a = ± 0.7005 $ (a.u.) indicated by the
855: two dots.\label{}}
856: \end{figure}
857:
858: \begin{figure}
859: \includegraphics[bb=1 1 504 627, angle=90.8,scale=0.8]{F_2.eps}
860: \caption{ Cross-sections of the Fermi-Coulomb $\rho_{xc}({\bf r} {\bf r'})$, Fermi $\rho_{x}({\bf r} {\bf r'})$, and Coulomb $\rho_{c}({\bf r} {\bf r'})$ holes along the nuclear bond axis for an electron at the center ${\bf r} = (0,0)$ of the bond. The electron position is indicated by the arrow.\label{}}
861: \end{figure}
862:
863: \begin{figure}
864: \includegraphics[bb=1 1 504 679,angle=90.5,scale=0.75]{F_3.eps}
865: \caption{ Cross-sections of the Fermi-Coulomb $\rho_{xc}({\bf r} {\bf r'})$ and Coulomb $\rho_{c}({\bf r} {\bf r'})$ holes along the nuclear bond axis for an electron at the center ${\bf r} = (0,a/3)$ of the bond with the electron position indicated by the arrow.\label{}}
866: \end{figure}
867:
868: \begin{figure}
869: \includegraphics[bb=1 1 502 646,angle=90,scale=0.45]{F_4.eps}
870: \caption{ The same as in Fig.3, but with the electron at ${\bf r}
871: = (0, 2a/3)$.\label{}}
872: \end{figure}
873:
874:
875: \begin{figure}
876: \includegraphics[bb=0 0 50 676,angle=90,scale=0.45]{F_5.eps}
877: \caption{ The same as in Fig.3, but with the electron at ${\bf r}
878: = (0, a)$.\label{}}
879: \end{figure}
880:
881: \begin{figure}
882: \includegraphics[bb=1 1 502 630,angle=90,scale=0.8]{F_6.eps}
883: \caption{ The same as in Fig.3, but with the electron at ${\bf r}
884: = (0, 2a)$. \label{}}
885: \end{figure}
886:
887: \begin{figure}
888: \includegraphics[bb=1 1 491 633,angle=90,scale=0.8]{F_7.eps}
889: \caption{ The same as in Fig.3, but with the electron at ${\bf r}
890: = (0, 4a)$. \label{}}
891: \end{figure}
892:
893: \begin{figure}
894: \includegraphics[bb=1 1 504 628,angle=90.4,scale=0.8]{F_8.eps}
895: \caption{ The same as in Fig.3, but with the electron at ${\bf r}
896: = (0, 6a)$. \label{}}
897: \end{figure}
898:
899: \begin{figure}
900: \includegraphics[bb=1 1 495 617,angle=90,scale=0.8]{F_9.eps}
901: \caption{ The electron-interaction ${\mathcal E}_{ee}(0,z)$ field, and its Hartree ${\mathcal E}_{H}(0,z)$
902: and Pauli-Coulomb ${\mathcal E}_{xc}(0,z)$ components along the nuclear bond axis. \label{}}
903: \end{figure}
904:
905: \begin{figure}
906: \includegraphics[bb=1 1 495 620,angle=90,scale=0.8]{F_10.eps}
907: \caption{ The Pauli ${\mathcal E}_{x}(0,z)$ and Coulomb ${\mathcal E}_{c}(0,z)$ fields
908: along the nuclear bond axis. The function $- 1/z^{2}$ is also plotted. \label{}}
909: \end{figure}
910:
911: \begin{figure}
912: \includegraphics[bb=1 1 495 610,angle=90,scale=0.8]{F_11.eps}
913: \caption{ The Pauli potential energy $W_{x}(0,z)$ along the nuclear bond axis.
914: The work done $W_{c}(0,z)$ in this direction in the force of the Coulomb field ${\mathcal E}_{c}(0,z)$,
915: and the function $- 1/z$, are also plotted.\label{}}
916: \end{figure}
917:
918:
919: \begin{figure}
920: \includegraphics[bb=1 1 497 608,angle=90,scale=0.8]{F_12.eps}
921: \caption{ The potential energy $v_{c}(0,z)$, the sum of the
922: Coulomb and Correlation-Kinetic potential energies, along the
923: nuclear bond axis. The work done $W_{c}(0,z)$ of Fig. 11, and the
924: work done $W_{t_{c}}(0,z)$ in the force of the Correlation-Kinetic
925: field ${\mathcal Z}_{t_{c}}(0,z)$ of Fig.13, are also
926: plotted.\label{}}
927: \end{figure}
928:
929: \begin{figure}
930: \includegraphics[bb=1 1 498 611,angle=89.5,scale=0.8]{F_13.eps}
931: \caption{ The Correlation-Kinetic field ${\mathcal Z}_{t_{c}}(0,z)$ along the nuclear bond axis. \label{}}
932: \end{figure}
933: %\begin{figure}
934: %\includegraphics[scale=1.0,angle=0]{Fig2.eps}% Here is how to import EPS art
935:
936: %\caption{\label{fig:fig2} The radial probability densities
937: %$r^{2}\rho_{1s^{2}}( r)$
938: %and $r^{2}\rho_{1s}( r)$ and $r^{2}\rho_{2s}( r)$}
939: %\end{figure}
940:
941:
942: \end{document}
943: %
944: % ****** End of file apssamp.tex ******
945: