Hello,

do you mean adding explicit solvent molecules to the main molecule? Or adding the solvent as continuous medium?

I did both in Gaussian, ie, I added solvent on polarizable continuum modal as well as added explicit solvent molecule  to the main molecule. Can both be called as they are in solvent phase? or any explanation!

NO both are different, if you run in some solvent model it will be taken as calculation in the solvent.

Hi I think in either case (explicit or implicit solvent) you can consider the molecule to be in a 'solution phase' but the two methods are very different. The implicit model provides a perturbation to your gas phase Hamiltonian to account for some important solvent interactions. Explicit solvent provides a change to the environment your main system exists in so alters the wave function you calculate.

If you are using explicit solvents, you should be careful how many solvent molecules you are adding, to few and it is not really simulating a solvent body, more the specific interactions of a main solute and a few close by solvent molecules. There are other considerations for QM/MM calculations. 

S. Xavier, you may find the following open access article of use. Apologies if it seems like a shameless plug for my own article, but I think it might actually be useful to you. You probably want to ignore section 1 and 2, but section 3 and 4 discuss at length the use of implicit and explicit solvent models.

http://pubs.rsc.org/en/Content/ArticleLanding/2015/CP/C5CP00288E#!divAbstract

As James explained, those are two different methods. In the explicit solvent addition method you can run calculation using PBC (periodic boundary condition), so you can define specific number of solvent molecules based on the density of the solvent.  This can be conjunction with MD simulation using BOMD keywords and also QM/MM.  

Thank you so much. I have not yet used running PBC, could you kindly explain me how to use it. For example if I want to add KMno4 with L-valine. The solvent is not found in the implicit method. Through this PBC can we add solvent with the main molecule, like that other solvent too? Please explain it to me

Hi Xavier, I believe you are using GaussView, so followings are good step by step tutorials for creating PBC for your system. Please try those. A free software Avogadro also capable of creating PBC, but I don't have tutorials for it.

http://www.gaussian.com/g_tech/gv5ref/pbctut1.htm

http://www.gaussian.com/g_tech/gv5ref/pbctut2.htm

http://www.gaussian.com/g_tech/gv5ref/pbctut3.htm

http://www.gaussian.com/g_tech/gv5ref/pbctut4.htm

http://www.gaussian.com/g_tech/gv5ref/pbctut5.htm

http://www.gaussian.com/g_tech/gv5ref/pbctut6.htm

Hi, 

I am quite new in using Gaussian, and I want to do optimization for a sugar system in explicit solvent, I have few questions:

1. I have 8 water molecules and I don't know how to arrange them around sugar 

2. which command I should use to consider as in solvent system 

I will appreciated if any one can help me

Thank you 

Hi Sara, are you trying to treat the solvent quantum mechanically or classically (QM/MM) ? What is your goal for the calculation what are you trying to find out?

You will find solvated systems have a large number of fairly shallow minima and some deep minima that Gaussian may optimise to. You may have some difficulty optimising  fully quantum mechanically as a result of the number of shallow minima accessible from a random starting point.

Might I suggest an alternative:

1. Run an MD simulation for a maybe 10 ps  in a box of solvent to equilibrate.

2. Take that geometry and cut out the solute and the eight closest water molecules this should give you a reasonable starting structure. 

3. Place the equilibrated solute and eight closest water molecules in Guassian and define the solute as quantum and solvent as classical using Gaussian's ONIOM scheme. (http://www.gaussian.com/g_whitepap/oniom_technote.htm)

4. Run a QM/MM optimisation in Gaussian. 

5. Try running a fully quantum calculation see if it works. Look up Gaussian optimisation in the manual for other options if it doesn't work or repost here. 

Alternatively if the solvent is not such an important consideration, maybe think about a continuum solvent model, PCM, CPCM, SMD in Gaussian.

Hope at least some of that is helpful to start with. 

Sara, you may also be best starting a new question to get the broadest possible answers as your question title might attract different people to answer your question than answered the original question. 

Hello, everyone! I'm a newbie in GaussView and Gaussian software, also.

I have an idea to simulate the model including a main molecule (an inhibitor) in Sodium Chloride (NaCl) solvent to generate theoretical paratmeters i.e. HOMO energy, LUMO energy...

I don't know how can I create a solvent environment for this inhibitor. Please kindly explain it^^