/usr/share/doc/psi4/html/dfmp2.html is in psi4-data 1:0.3-5.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 | <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<title>DF-MP2: Density-Fitted Second-Order Møller-Plesset Perturbation Theory — Psi4 [] Docs</title>
<link rel="stylesheet" href="_static/psi4.css" type="text/css" />
<link rel="stylesheet" href="_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="./" type="text/css" />
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT: './',
VERSION: '',
COLLAPSE_INDEX: false,
FILE_SUFFIX: '.html',
HAS_SOURCE: true
};
</script>
<script type="text/javascript" src="_static/jquery.js"></script>
<script type="text/javascript" src="_static/underscore.js"></script>
<script type="text/javascript" src="_static/doctools.js"></script>
<script type="text/javascript" src="_static/jquery.cookie.js"></script>
<script type="text/javascript" src="_static/toggle_sections.js"></script>
<script type="text/javascript" src="_static/toggle_sidebar.js"></script>
<script type="text/javascript" src="_static/toggle_codeprompt.js"></script>
<link rel="shortcut icon" href="_static/favicon-psi4.ico"/>
<link rel="top" title="Psi4 [] Docs" href="index.html" />
<link rel="up" title="Theoretical Methods: SCF to FCI" href="methods.html" />
<link rel="next" title="CC: Coupled Cluster Methods" href="cc.html" />
<link rel="prev" title="DCFT: Density Cumulant Functional Theory" href="dcft.html" />
</head>
<body role="document">
<div class="relbar-top">
<div class="related" role="navigation" aria-label="related navigation">
<h3>Navigation</h3>
<ul>
<li class="right" style="margin-right: 10px">
<a href="genindex.html" title="General Index"
accesskey="I">index</a></li>
<li class="right" >
<a href="py-modindex.html" title="Python Module Index"
>modules</a> </li>
<li class="right" >
<a href="contents.html" title="Table Of Contents"
accesskey="C">toc</a> </li>
<li class="right" >
<a href="cc.html" title="CC: Coupled Cluster Methods"
accesskey="N">next</a> </li>
<li class="right" >
<a href="dcft.html" title="DCFT: Density Cumulant Functional Theory"
accesskey="P">previous</a> </li>
<li><a href="index.html">Psi4 []</a> » </li>
<li class="nav-item nav-item-1"><a href="methods.html" accesskey="U">Theoretical Methods: SCF to FCI</a> »</li>
</ul>
</div>
</div>
<div class="document">
<div class="documentwrapper">
<div class="bodywrapper">
<div class="body" role="main">
<a class="reference internal image-reference" href="_images/psi4banner.png"><img alt="Psi4 Project Logo" src="_images/psi4banner.png" style="width: 100%;" /></a>
<div class="section" id="df-mp2-density-fitted-second-order-mo-slashller-plesset-perturbation-theory">
<span id="sec-dfmp2"></span><span id="index-0"></span><h1>DF-MP2: Density-Fitted Second-Order Møller-Plesset Perturbation Theory<a class="headerlink" href="#df-mp2-density-fitted-second-order-mo-slashller-plesset-perturbation-theory" title="Permalink to this headline">¶</a></h1>
<p><em>Code author: Robert M. Parrish</em></p>
<p><em>Section author: Robert M. Parrish</em></p>
<p><em>Module:</em> <a class="reference internal" href="autodir_options_c/module__dfmp2.html#apdx-dfmp2"><span>Keywords</span></a>, <a class="reference internal" href="autodir_psivariables/module__dfmp2.html#apdx-dfmp2-psivar"><span>PSI Variables</span></a>, <a class="reference external" href="https://github.com/psi4/psi4public/blob/master/src/bin/dfmp2">DFMP2</a></p>
<div class="section" id="introduction">
<h2>Introduction<a class="headerlink" href="#introduction" title="Permalink to this headline">¶</a></h2>
<p>Second-order Møller-Plesset Perturbation Theory (MP2) occupies a unique role
in quantum chemistry due to its small-prefactor <img class="math" src="_images/math/ba3bb4e90152921f5602f9ee346b254f2d1ccabf.png" alt="{\cal O}(N^5)" style="vertical-align: -4px"/> treatment of
dynamic electron correlation. This unusually cheap
<em>ab initio</em> treatment of electron correlation may be made even more
efficient by means of the Density-Fitting (DF) approximation (also known as
Resolution-of-the-Identity or RI), wherein the quadratic <img class="math" src="_images/math/de177e872a957abf603c2a0d589c8933f8b4c280.png" alt="ov" style="vertical-align: 0px"/> products in the
bra- and ket- of the <img class="math" src="_images/math/374638e4af7ba1754e92c03bf8a9aa644bcda0f9.png" alt="(ov|ov)" style="vertical-align: -5px"/>-type Electron Repulsion Integrals (ERIs)
appearing in MP2 are cast onto a linear-scaling auxiliary basis by least-squares
fitting. Substitution of the DF factorization into the MP2 equations results in
a formal scaling and prefactor reduction of MP2, and further speed gains are
possible due to heavy utilization of matrix-multiplication kernels and minimal
storage requirements in a DF approach. The method has been found to be quite
robust and accurate, and it should be preferred unless extreme accuracy is required
or a fitting basis is not defined for the primary basis and atom type
encountered. In particular, we have found excellent efficiency and tractability
gains when using DF-MP2 in concert with a DF-SCF reference. An efficient,
threaded, disk-based DF-MP2 code is available in <span class="sc">Psi4</span> for all single
reference types available in the SCF module.</p>
<p>An example utilization of the code is:</p>
<div class="highlight-python"><div class="highlight"><pre>molecule h2o {
0 1
O
H 1 1.0
H 1 1.0 2 104.5
}
set basis cc-pvdz
set scf_type df
set freeze_core True
energy('df-mp2')
</pre></div>
</div>
<p>The <code class="docutils literal"><span class="pre">energy('df-mp2')</span></code> call to <a class="reference internal" href="energy.html#driver.energy" title="driver.energy"><code class="xref py py-func docutils literal"><span class="pre">energy()</span></code></a> executes
the predefined DF-MP2 procedure, first calling
the SCF module with a default RHF reference and DF algorithm for the
two-electron integrals. When the orbitals are converged, the DF-MP2 module is
launched, which forms the density-fitted <img class="math" src="_images/math/451fb39270282db655a2a19099925f8cb363d74c.png" alt="(Q|ov)" style="vertical-align: -5px"/> integrals and then builds the
full <img class="math" src="_images/math/374638e4af7ba1754e92c03bf8a9aa644bcda0f9.png" alt="(ov|ov)" style="vertical-align: -5px"/> tensor in blocks, evaluating the contributions to the MP2 energy
as it goes. A RHF-MP2 wavefunction is selected automatically due to the RHF
reference. In this example, we freeze the core, both for efficiency and
because split-valence bases like cc-pVDZ do not contain core correlation
functions. The result looks something like:</p>
<div class="highlight-python"><div class="highlight"><pre>----------------------------------------------------------
====================> MP2 Energies <====================
----------------------------------------------------------
Reference Energy = -76.0213974789664633 [H]
Singles Energy = -0.0000000000000001 [H]
Same-Spin Energy = -0.0512503261762665 [H]
Opposite-Spin Energy = -0.1534098129352447 [H]
Correlation Energy = -0.2046601391115113 [H]
Total Energy = -76.2260576180779736 [H]
----------------------------------------------------------
==================> SCS-MP2 Energies <==================
----------------------------------------------------------
SCS Same-Spin Scale = 0.3333333333333333 [-]
SCS Opposite-Spin Scale = 1.2000000000000000 [-]
SCS Same-Spin Energy = -0.0170834420587555 [H]
SCS Opposite-Spin Energy = -0.1840917755222936 [H]
SCS Correlation Energy = -0.2011752175810492 [H]
SCS Total Energy = -76.2225726965475161 [H]
----------------------------------------------------------
</pre></div>
</div>
<p>The theory, breakdown of results, and common keywords used in DF-MP2 are presented below.</p>
</div>
<div class="section" id="theory">
<span id="index-1"></span><h2>Theory<a class="headerlink" href="#theory" title="Permalink to this headline">¶</a></h2>
<p>Møller-Plesset Theory (MPn) or Many-Body Perturbation Theory
(MBPT) through second order has the spin-orbital formula:</p>
<div class="math" id="equation-MP2">
<p><span class="eqno">(1)</span><img src="_images/math/176c01c041d9055cc82769102e4a124f722ec160.png" alt="E_{\mathrm{total}}^{(2)} = E_{\mathrm{Reference}} -
\frac{f_{ia} f_{ia}}{\epsilon_a - \epsilon_i} -
\frac{1}{4} \frac{\langle ij||ab\rangle^2}{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}"/></p>
</div><p>Here <img class="math" src="_images/math/4cc7324e0d6c8c591e4d865a21144bda81fd3011.png" alt="i" style="vertical-align: 0px"/> and <img class="math" src="_images/math/c18271eddf460603079ed91e4dc4af329a59eab2.png" alt="j" style="vertical-align: -4px"/> are occupied spin orbitals, <img class="math" src="_images/math/d7bd06a860633f7d8ead0d84fbb30ac7e88b558b.png" alt="a" style="vertical-align: 0px"/> and
<img class="math" src="_images/math/fdcf4bd764a50e69250d8e0a9e2212d63e8ecfe0.png" alt="b" style="vertical-align: 0px"/> are virtual spin orbitals, <img class="math" src="_images/math/db20b3a8d35fb65e5e099984ac052cf105c92e49.png" alt="f_{ia}" style="vertical-align: -4px"/> are the <img class="math" src="_images/math/de177e872a957abf603c2a0d589c8933f8b4c280.png" alt="ov" style="vertical-align: 0px"/>
Fock Matrix elements, <img class="math" src="_images/math/1c67e848e51a0f11ff730889cbd43af9034630a0.png" alt="\epsilon" style="vertical-align: 0px"/> are the orbital eigenvalues, and
<img class="math" src="_images/math/3da940473420e618cac0091dd3ffd080d839ab6c.png" alt="\langle ij||ab\rangle" style="vertical-align: -5px"/> are the antisymmetrized physicist’s ERIs.
For converged RHF and UHF references, the singles correction,</p>
<div class="math">
<p><img src="_images/math/c6131c65d2db04986f489df147750c3fe98c93d1.png" alt="E_{\mathrm{MBPT}}^{(1)} = - \frac{f_{ia} f_{ia}}{\epsilon_a - \epsilon_i},"/></p>
</div><p>is zero due to the Brillioun Condition, and the first contribution to the
perturbation series is at the second order:</p>
<div class="math">
<p><img src="_images/math/cf2adf852b6e3d29622206c3a713fc78e2f27d1b.png" alt="E_{\mathrm{MBPT}}^{(2)} = - \frac{1}{4} \frac{\langle ij|ab\rangle^2}{\epsilon_a +
\epsilon_b - \epsilon_i - \epsilon_j}."/></p>
</div><p>In the DFMP2 module, the first-order contribution, or “singles energy” is
always evaluated. This term is a significant contributor to the total
second-order energy if a ROHF reference is used. In this case, we have chosen
to use the ROHF-MBPT(2) ansatz, in which the ROHF orbitals are
semicanonicalized, the resultant nonzero Fock matrix elements <img class="math" src="_images/math/db20b3a8d35fb65e5e099984ac052cf105c92e49.png" alt="f_{ia}" style="vertical-align: -4px"/> are used
to form the singles amplitudes, and then the second-order amplitudes are formed
with the semicanonical spin orbitals via the same machinery as a UHF-MP2. Note
that the singles energy should be very close to zero for RHF and UHF references;
if it is not, there is a good chance your orbitals are not well converged.
Tighten the SCF <a class="reference internal" href="autodoc_glossary_options_c.html#term-e-convergence-scf"><span class="xref std std-term">E_CONVERGENCE</span></a> and/or <a class="reference internal" href="autodoc_glossary_options_c.html#term-d-convergence-scf"><span class="xref std std-term">D_CONVERGENCE</span></a> keywords
and try again.</p>
<p>To increase the efficiency of MP2 energy evaluation, spin integration
and simplification is carried out. This also allows for the identification of
Same-Spin (SS) and Opposite-Spin (OS) terms for use in Grimme’s Spin-Component
Scaled (SCS) MP2. For RHF-MP2 (also labeled as RMP2), the spin-free equations are
(note that the integrals are now chemist’s integrals over spatial orbitals)</p>
<div class="math">
<p><img src="_images/math/5c74f8f9b9932c4692d6e07df4086b89f7893723.png" alt="E_{\mathrm{MBPT,OS}}^{(2)} =
- \frac{(ia|jb)(ia|jb)}{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}"/></p>
</div><p>and</p>
<div class="math">
<p><img src="_images/math/231060b20f6e9ea7fd9f1c1f9a3549968dac5f0c.png" alt="E_{\mathrm{MBPT,SS}}^{(2)} =
- \frac{[(ia|jb)-(ib|ja)](ia|jb)}{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}."/></p>
</div><p>For UHF-MP2 (also labeled as UMP2) and the second-order contribution to
ROHF-MBPT(2) using semicanonical orbitals, the spin-free equations are</p>
<div class="math">
<p><img src="_images/math/3e84d2743af561ab12676cb7fdff2c83d754cc74.png" alt="E_{\mathrm{MBPT,OS}}^{(2)} =
- \frac{(ia^\alpha|jb^\beta)(ia^\alpha|jb^\beta)}{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}"/></p>
</div><p>and</p>
<div class="math">
<p><img src="_images/math/524a9005c012a2b49124c8b23128c51b175bbbaf.png" alt="E_{\mathrm{MBPT,SS}}^{(2)} =
- \frac{1}{2}\frac{[(ia^\alpha|jb^\alpha)-(ib^\alpha|ja^\alpha)](ia^\alpha|jb^\alpha)}
{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}
- \frac{1}{2}\frac{[(ia^\beta|jb^\beta)-(ib^\beta|ja^\beta)](ia^\beta|jb^\beta)}
{\epsilon_a + \epsilon_b - \epsilon_i - \epsilon_j}."/></p>
</div><p>Note that the UHF-MP2 equations use three classes of integrals, while the
RHF-MP2 equations use only one class. Because of this, a UHF-MP2 or
ROHF-MBPT(2) energy should take roughly three times as long as an RHF-MP2
energy.</p>
</div>
<div class="section" id="recommendations">
<h2>Recommendations<a class="headerlink" href="#recommendations" title="Permalink to this headline">¶</a></h2>
<p>All-in-all, DFMP2 should be a simple module to use, with few keywords (fully
documented in the Appendix <a class="reference internal" href="autodir_options_c/module__dfmp2.html#apdx-dfmp2"><span>DFMP2</span></a>). Some basic recommendations are included below:</p>
<ul class="simple">
<li>DFMP2 should be run with the <img class="math" src="_images/math/de177e872a957abf603c2a0d589c8933f8b4c280.png" alt="ov" style="vertical-align: 0px"/>-type RI or MP2FIT auxiliary
basis sets, <em>not</em> the -JKFIT basis sets. The automatic basis selector
should work fine for all of the Dunning and Pople bases (provided the auxiliary
basis exists for the atom in question). If it does not, use the
<a class="reference internal" href="autodoc_glossary_options_c.html#term-df-basis-mp2-dfmp2"><span class="xref std std-term">DF_BASIS_MP2</span></a> keyword to manually specify the basis.</li>
<li>DFMP2 likes memory. At a minimum, <img class="math" src="_images/math/55e58ade139da24241a829ef5c18c6f965d4a372.png" alt="2Q^2" style="vertical-align: -4px"/> doubles are required,
where <img class="math" src="_images/math/f59dd7fde255c3f2d4d4d2f0631914aa7604d031.png" alt="Q" style="vertical-align: -4px"/> is the size of the auxiliary basis set. However, there is
one disk transpose of the <img class="math" src="_images/math/451fb39270282db655a2a19099925f8cb363d74c.png" alt="(Q|ov)" style="vertical-align: -5px"/> tensor in the RHF-MP2 algorithm
[two for UHF-MP2 and ROHF-MBPT(2)], so more memory will reduce seek times.
If you notice DFMP2 using more memory than allowed, it is possible that
the threaded three-index ERI computers are using too much overhead memory.
Set the <a class="reference internal" href="autodoc_glossary_options_c.html#term-df-ints-num-threads-dfmp2"><span class="xref std std-term">DF_INTS_NUM_THREADS</span></a> to a smaller number to prevent this
in this section (does not affect threaded efficiency in the rest of the
code).</li>
<li>DFMP2 likes disk. At a minimum, <img class="math" src="_images/math/e2d0571e37359e9f12adee406b8ed582ed1ecdb5.png" alt="2Qov" style="vertical-align: -4px"/> doubles are required for
RHF-MP2, and <img class="math" src="_images/math/7d9bde787055a2f4f8787e0d2722e65337a3162e.png" alt="4Qov" style="vertical-align: -4px"/> doubles are required for UHF-MP2.</li>
<li>DFMP2 likes threads. Some of the formation of the <img class="math" src="_images/math/451fb39270282db655a2a19099925f8cb363d74c.png" alt="(Q|ov)" style="vertical-align: -5px"/> tensor
relies on threaded BLAS (such as MKL) for efficiency. The main
<img class="math" src="_images/math/ba3bb4e90152921f5602f9ee346b254f2d1ccabf.png" alt="{\cal O}(N^5)" style="vertical-align: -4px"/> step is done via small/medium-sized DGEMMs inside of
OpenMP, so make sure to set the <span class="target" id="index-2"></span><a class="reference internal" href="external.html#envvar-OMP_NESTED"><code class="xref std std-envvar docutils literal"><span class="pre">OMP_NESTED</span></code></a> environment variable
to <code class="docutils literal"><span class="pre">FALSE</span></code> to prevent thread thrash (or just as well, do not define
<span class="target" id="index-3"></span><a class="reference internal" href="external.html#envvar-OMP_NESTED"><code class="xref std std-envvar docutils literal"><span class="pre">OMP_NESTED</span></code></a> at all).</li>
<li>Freezing core is good for both efficiency and correctness purposes.
Freezing virtuals is not recommended. The DFMP2 module will remind you how
many frozen/active orbitals it is using in a section just below the title.</li>
<li>ROHF-MBPT(2) may be preferred to UHF-MP2, as the latter can suffer from
severe spin contamination in some cases.</li>
<li>MP2 is not suitable for systems with multireference character. The
orbital energies will come together and an explosion will occur.</li>
</ul>
<style type="text/css"><!--
.green {color: red;}
.sc {font-variant: small-caps;}
--></style></div>
</div>
</div>
</div>
</div>
<div class="sphinxsidebar" role="navigation" aria-label="main navigation">
<div class="sphinxsidebarwrapper">
<h3><a href="index.html">Table Of Contents</a></h3>
<ul>
<li><a class="reference internal" href="#">DF-MP2: Density-Fitted Second-Order Møller-Plesset Perturbation Theory</a><ul>
<li><a class="reference internal" href="#introduction">Introduction</a></li>
<li><a class="reference internal" href="#theory">Theory</a></li>
<li><a class="reference internal" href="#recommendations">Recommendations</a></li>
</ul>
</li>
</ul>
<h4>Previous topic</h4>
<p class="topless"><a href="dcft.html"
title="previous chapter">DCFT: Density Cumulant Functional Theory</a></p>
<h4>Next topic</h4>
<p class="topless"><a href="cc.html"
title="next chapter">CC: Coupled Cluster Methods</a></p>
<div role="note" aria-label="source link">
<h3>This Page</h3>
<ul class="this-page-menu">
<li><a href="_sources/dfmp2.txt"
rel="nofollow">Show Source</a></li>
</ul>
</div>
<div id="searchbox" style="display: none" role="search">
<h3>Quick search</h3>
<form class="search" action="search.html" method="get">
<input type="text" name="q" />
<input type="submit" value="Go" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
<p class="searchtip" style="font-size: 90%">
Enter search terms or a module, class or function name.
</p>
</div>
<script type="text/javascript">$('#searchbox').show(0);</script>
</div>
</div>
<div class="clearer"></div>
</div>
<div class="relbar-bottom">
<div class="related" role="navigation" aria-label="related navigation">
<h3>Navigation</h3>
<ul>
<li class="right" style="margin-right: 10px">
<a href="genindex.html" title="General Index"
>index</a></li>
<li class="right" >
<a href="py-modindex.html" title="Python Module Index"
>modules</a> </li>
<li class="right" >
<a href="contents.html" title="Table Of Contents"
>toc</a> </li>
<li class="right" >
<a href="cc.html" title="CC: Coupled Cluster Methods"
>next</a> </li>
<li class="right" >
<a href="dcft.html" title="DCFT: Density Cumulant Functional Theory"
>previous</a> </li>
<li><a href="index.html">Psi4 []</a> » </li>
<li class="nav-item nav-item-1"><a href="methods.html" >Theoretical Methods: SCF to FCI</a> »</li>
</ul>
</div>
</div>
<div class="footer" role="contentinfo">
© Copyright 2015, The Psi4 Project.
Last updated on Tuesday, 12 January 2016 03:10PM.
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 1.3.3.
</div>
<!-- cloud_sptheme 1.3 -->
</body>
</html>
|