[gmx-developers] TIP3P and TIP4P parameters

[Erik Lindahl]

Hi Michael,

I have no idea where the differences appeared, but to make things even 
worse, these are the parameters I get from Bill Jorgensen's OPLS 
datafiles when using the factor 4.184:

  opls_111   OW     15.99940    -0.834       A    3.15061e-01  
6.36386e-01
  opls_113   OW     15.99940     0.000       A    3.15365e-01  
6.48520e-01

Yep, a third alternative - probably due to rounding in the opls 
parameter file - at least I haven't done anything except applying the 
conversion and using lots of extra digits for the accuracy.

(These are the ones I sent to Eric Sorin in the updated OPLS files a 
couple of months ago.)

I agree that the published parameters is the best source, so I'll 
change them to your values.

Cheers,

Erik

On Tuesday, June 10, 2003, at 04:56  PM, Michael Shirts wrote:

>
> So, looking over the original TIP3P and TIP4P paper (Jorgensen, 
> Chandrasekhar,
> and Madura, JCP, v. 79 n. 2, July 15 1983, p. 926), and looking over 
> the
> parameters there, and comparing them to ffoplsaanb.itp it seems as 
> though the
> LJ parameters for the waters should be:
>
> TIP3P
>  opls_111   OW     15.99940    -0.834       A    3.15061e-01  
> 6.36272e-01
> TIP4P
>  opls_113   OW     15.99940     0.000       A    3.15358e-01  
> 6.48694e-01
>
> instead of:
>
> TIP3P
>  opls_111   OW     15.99940    -0.834       A    3.15061e-01  
> 6.36812e-01
> TIP4P
>  opls_113   OW     15.99940     0.000       A    3.15365e-01  
> 6.48954e-01
>
> Yes, it is in the last digit of accuracy as pubished in the paper, but
> original model parameters should be treated as exact when a unit 
> conversion (1
> kcal/mol = 4.184 kJ/mol) is applied.  I know they aren't big 
> differences, but
> some weird non-linear effects can sometime occur in the density, etc.
>
> Cheers,
> Michael Shirts
> Pande Group
> Stanford University
>
>
> _______________________________________________
> gmx-developers mailing list
> gmx-developers at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-developers
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-developers-request at gromacs.org.