[gmx-users] Lennard Jones Parameters for Ammonia in ffoplsaa

[Darrell Koskinen]

Hi Mark,
By "missing", I meant zero by design. I found the initial opls paper in 
which the H atom in ammonia is assigned a zero LJ parameter and the 
entire LJ parameter for ammonia is assigned to the N atom.

I was just wondering if assigning the entire LJ parameter to the N atom 
might affect the adsorption of the polar ammonia molecule within the MD 
model since instead of there being four potential energy wells in the 
ammonia molecule, there would only be one potential energy well and this 
single well may not interact as significantly with the multiple 
potential energy wells present at the edge of the hydrogen terminated 
graphene sheet. Just looking for your comments and thoughts.

Thanks.

Darrell
> Date: Sun, 01 Nov 2009 18:20:26 +1100
> From: Mark Abraham <Mark.Abraham at anu.edu.au>
> Subject: Re: [gmx-users] Lennard Jones Parameters for Ammonia in
> 	ffoplsaa
> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> Message-ID: <4AED36BA.1040908 at anu.edu.au>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
> Darrell Koskinen wrote:
>   
>> Hi Mark,
>> I thought opls_127 and opls_128 were the representative atoms to use for 
>> ammonia. Please correct me if I am wrong.
>>     
>
> I don't know anything more than what's in ffoplsa* files or the papers.
>
>   
>> Would the adsorption of the polar ammonia molecule onto the polar 
>> (hydrogen-terminated) edges of the graphene sheet be affected by the 
>> missing LJ parameter on the H atoms?
>>     
>
> Define "missing". If it is zero by design, that's because zero served 
> the purpose during the parameterization, and the documentation as quoted 
> by you is perhaps erroneous with respect to small molecules. If it 
> oplsa_128 should have non-zero LJ then you'll need to find more evidence 
> of that in the OPLSAA papers, or by consulting the authors. Note that 
> most models of water do not have LJ on the H atoms, and so not having LJ 
> on H atoms of solvated ammonia is probably quite consistent.
>
> Doesn't this bring us back to the point of quite a while ago, that an 
> ammonia model that was probably not parameterized for the gas phase 
> might not be useful in the gas phase?
>
> Mark
>
>   
>>> Date: Mon, 26 Oct 2009 08:47:14 +1100
>>> From: Mark Abraham <Mark.Abraham at anu.edu.au>
>>> Subject: Re: [gmx-users] Lennard Jones Parameters for Ammonia in
>>>     ffoplsaa
>>> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>>> Message-ID: <4AE4C762.2060609 at anu.edu.au>
>>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>>
>>> Justin A. Lemkul wrote:
>>>  
>>>       
>>>> Darrell Koskinen wrote:
>>>>    
>>>>         
>>>>> I looked through the ffoplsaanb.itp file and see that all the H 
>>>>> atoms except for the following four have zero LJ parameters:
>>>>>
>>>>> HC OPLS_140 alkane H
>>>>> HA OPLS_146 Benzene H
>>>>> H4 OPLS_345 Cytosine H-C6
>>>>> H5 OPLS_355 Adenine & Guanine H-C2
>>>>>
>>>>> However, in the paper "Development and Testing of the OPLS All-Atom 
>>>>> Force Field on Conformational Energetics and Properties of Organic 
>>>>> Liquids", it states: "The original OPLS (optimized potentials for 
>>>>> liquid simulations) potential functions used a partially united-atom 
>>>>> (UA) model; sites for nonbonded interactions are placed on all 
>>>>> nonhydrogen atoms and on hydrogens attached to heteroatoms or 
>>>>> carbons in aromatic rings". Later, the paper indicates that the 
>>>>> paramters were adopted as much as possible from the OPLS-UA force 
>>>>> field. Thus, since "sites for nonbonded interactions are placed on 
>>>>> hydrogens attached to heteroatoms ", it appears to me that a site 
>>>>> for a non-bonded interaction should exist on the H atoms within 
>>>>> ammonia. Is my interpretation incorrect?
>>>>>
>>>>>       
>>>>>           
>>>> I don't think so.  Each of the above examples you've cited above is 
>>>> for a C-H bond in an aromatic ring.  In OPLS, all N-H bonds involved 
>>>> hydrogen atoms with zero LJ parameters.  See, for example, any 
>>>> backbone NH, amide (ASN/GLN), or amine (LYS/LYSH), etc and you will 
>>>> see the atom types used.
>>>>     
>>>>         
>>> I think you're both partly wrong. Type 140 is not for C-H in an 
>>> aromatic ring, and if you grep 1.008 ffoplsaanb.itp you get many more 
>>> H atom types with LJ. Certainly Justin's right inasmuch as amide N-H 
>>> type 241 has no H LJ parameters. Perhaps there's other text that 
>>> Darrell hasn't found yet, or backbone nitrogen is not considered a 
>>> heteroatom :-).
>>>
>>> Mark
>>>
>>>  
>>>       
>>>>> Darrell
>>>>>      
>>>>>           
>>>>>> Date: Mon, 19 Oct 2009 19:07:22 +1100
>>>>>> From: Mark Abraham <Mark.Abraham at anu.edu.au>
>>>>>> Subject: Re: [gmx-users] Lennard Jones Parameters for Ammonia in
>>>>>>     ffoplsaa
>>>>>> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>>>>>> Message-ID: <4ADC1E3A.8010502 at anu.edu.au>
>>>>>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>>>>>>
>>>>>> Darrell Koskinen wrote:
>>>>>>  
>>>>>>        
>>>>>>             
>>>>>>> Hi,
>>>>>>> I see that the Lennard Jones parameters for the N & H atoms in 
>>>>>>> ammonia, represented by opls_127 and opls_128 in the file 
>>>>>>> ffoplsaanb.itp are:
>>>>>>>
>>>>>>> opls_127   NT  7     14.00670    -1.020       A    3.42000e-01  
>>>>>>> 7.11280e-01
>>>>>>> opls_128   H   1       1.00800     0.340       A    0.00000e+00  
>>>>>>> 0.00000e+00
>>>>>>>
>>>>>>> Why are sigma and epsilon both zero for H? Are the LJ parameters 
>>>>>>> for H embedded in the parameter for NT? If so, how were these 
>>>>>>> parameters combined?
>>>>>>>               
>>>>>>>               
>>>>>> What do other H have? What do the OPLS-AA paper(s) have to say 
>>>>>> about such H atoms?
>>>>>>
>>>>>> Mark
>>>>>>