<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
At least two of you have approached me about optimization problems
(Gaussian taking a LONG time to optimize a structure). Here is a
strategy that I generally follow to speed up this process. Since
most of you are optimizing normal molecules (not trying to do
bizarre things like force atoms to be close to each other) then I
recommend the following.<br>
<br>
<ol>
<li>Build the basic molecule backbone and then add the functional
groups. Make sure that no atoms are too close to each other.
You can easily do this now by chaning the angles and dihedrals
for individual atoms.</li>
<li>Optimize using a Semiempirical technique (I usually selecte
PM3). In most case this will put the molecule in a confirmation
that is near the minimum.</li>
<li>With the PM3 optimized structure start doing your "real"
optimization using either HF/DFT or MP2. As many of you saw in
P-Chem -- you are going to need to put polarization functions on
those basis sets. In other words, do you optimizations with at
least 3-21G* and, it you have time, 6-31G(d,p)</li>
</ol>
If you are still having issues getting molecules to optimize, send
me the .gjf file and I'll see what I can do.<br>
<br>
<br>
<pre class="moz-signature" cols="72">--
Andrew J. Pounds, Ph.D. (<a class="moz-txt-link-abbreviated" href="mailto:pounds@theochem.mercer.edu">pounds@theochem.mercer.edu</a>)
Associate Professor of Chemistry and Computer Science
Mercer University, Macon, GA 31207 (478) 301-5627
</pre>
</body>
</html>