[ayse]

Hi all,

I am new in Gaussian and I want to optimize the icosahedral gold geometry by using Gaussian 03. My system consists of 144 atom. I've used Rhf/LANL2DZ basis set. When I try to run input file with the coordinates of 144 gold atom, it gives an error message as below;


Maximum number of bonds=100 exceeded for atom   1.
 Error termination via Lnk1e in C:\G03W\l101.exe at Mon Jan 12 16:07:18 2015.
 Job cpu time:  0 days  0 hours  0 minutes  0.0 seconds.
 File lengths (MBytes):  RWF=      7 Int=      0 D2E=      0 Chk=      1 Scr= 

How can I solve this problem?

Thanks,

Ayse.

[curiouscat]

Can you post your starting coordinates? Are they reasonable?

[Irlanur]

An honest question: How does this work? Is it actually a Calculation involving all electrons or is it some kind of force-field approach? Because I would expect HF based or DFT based methods to fail miserably in this situation.

[curiouscat]

An honest question: How does this work? Is it actually a Calculation involving all electrons or is it some kind of force-field approach? Because I would expect HF based or DFT based methods to fail miserably in this situation.

I'm no expert but my two cents: I've tried a full DFT calculation & if you relax it it will find some local energy minima. Problem is that the energy landscape with such clusters is so kinked up that it's easy to get trapped in some such local minimum.

In other words, you can use DFT to compare two known configs, but it will do miserably at doing a geometric search.

Is that what you were implying? Or did I misunderstand?