# Computational Chemistry

5
How do I calculate hydrogen bond energies?

We investigate a dimer of an organic molecule. It is linked together by two hydrogen bridges.

a) What is the best way to calculate the bonding energy of the dimer?

One possible approach is to optimize the geometries of the dimer and the monomer with classical DFT methods, let’s say B3LYP/6-311+G(d,p). The energy of the monomer is corrected with BSSE correction. The difference between (2x)monomer and dimer energy gives the bond energy of the hydrogen bond.

b) Is that correct?

c) How to do this type of calculation with modern dispersion-corrected DFT methods?

d) Which method / basis set combination would you propose?

Usually people optimize the structure at B3LYP and then use MP2 or MP3 level of theory for calculation of interaction energy.

45
How to obtain QST2 Calculation to find the transition state?
I have a problem doing the qst2 calculation. My purpose is to find the transition state of the decomposition mechanism process. I have already used the scan process to find the maximum energy of the molecule and use the opt=(ts,noeigen) to optimize the transition state. But I can't find any imaginary frequency. Then I tried use qst2 calculation to find the transition state. The results from qst2 are error in internal coordinate system and error termination via Lnk1e in/pkg/chem/gaussian/g09c01/g09/l103.exe. Is there any ways to find out the transition state or some suggestion to do this calculation?

Mrs/Miss Wibowo,

Please pay attention to Gaussian technical notes available at:

http://www.gaussian.com/g_whitepap/qst2.htm

1
Why does water dissociate in AIMD simulation?

I am trying to simulate water by using AIMD simulation with VASP.

The LDA is used and about 70 water molecules are placed in a box.  The following parameters are my setting in INCAR. However, the results seem abnormal. A few water molecules prefer to dissociate into H and OH after 5ps. Does anyone know how to solve this problem? Thanks in advance!!!

PREC = low
ENCUT = 300.000

IBRION = 0
NSW = 6000
POTIM = 0.5
SMASS = 2
TEBEG = 300
TEEND = 300

ISIF = 2
ISYM = 0
ALGO = V
ISMEAR = -1
SIGMA = 0.0258
LREAL= Auto
NELECT=1392 #1393
NELMIN=6

Most likely there is a problem with the energies of the system. Have you double checked your molecule's topologies to make sure they're correct?

3
Gaussian the error in link 716?

While I put a job for Scan 2 molecules in gaussian the error 716 appeared. Could you please suggest how I can resolve it?
Regards,

dehdab

Explanation of Error
This means that the Gaussian job terminated abnormally because an angle x in the z-matrix optimized to become outside the allowed range of 0 < x < 180.

Fixing the Error
This can happen if there are large geometry changes in a molecule, especially one composed of interacting fragments, and one must either re-define the z-matrix, or use a different coordinate system.

8
How can I model Fe3O4 (111) surface slab with a ligand (like benzene) on top of the slab's surface?

I want to model the slab surface of Fe3O4 (111) with a ligand (non-attached) to the slab's surface for DFT studies. Which software and detailed tutorials are available for this molecular modelling?

Sure, no problems!

Best regards,

Aleksey.

8
Why the adiabatic Born–Oppenheimer approximation (BOA) is broken down in graphene?

I want to know is the broken down of BOA in graphene has been solved or not yet,

http://www.nature.com/nmat/journal/v6/n3/full/nmat1846.html

what are the other cases where the BOA is also broken down in them and Why

The Born-Oppenheimer Approximation is mostly a wonderful approach in molecular physics, but it often breaks down, cf. the Jahn-Teller Mechanism, conical intersections and in condensed matter problems like superconductivity and strongly correlated systems, in addition to graphene.

In such systems one usually have two subsystems, (I) the fermionic system of light carriers and (II) the nuclear skeleton.  One can then use the mirror theorem for mappings between (I) and (II), see e.g. the discussion as well as references in the enclosed article below.

4
What does a TDDFT Oscillator Strength greater than experimental value indicate?

I have synthesized a solvatochromic molecule whose oscillator strength indicates smaller value for the highest absorbed peak in experimental absorption spectra while vice versa for the other peak. Can anyone please suggest me where I have gone wrong?

I again optimised the structure and tried getting its spectral data using the same basis set def2-TZVP B3LYP. I got the required results with oscillator strength as pert he experimental values. Thank you for all the guidance V.N. Ravi kishore V. and Bojidarka B. Ivanova

9
Can you suggest a suitable level of theory and basis set for the determination of dipole bound state of an anion?
Optimization of a cytidine anion molecule using HF/6-31+G(d) accuracy level results in a valence bound state (i.e., the extra electron is located on the cytosine nucleobase) whereas at MP2/6-31+G(d) level, the excess electron is residing in a diffuse orbital outside the molecular framework. Should I decrease the exponent in the basis function? If so, how can we determine the exponent range? I can give more information if required?

Thank you Jamshidi...

15
How can I correct/improve Gaussian Frequency Calculations?

Hi all,

For a while now I have been building a small application to generate AMBER forcefield compatible force constants (stretch and angle). I use the full Force Constant matrix for a Gaussian OPT->FREQ calculation. I am not working on normal modes, I am using the vibrations projected on the bond through a displaced of one atom from the other:

# B3LYP/<VARIES> OPT=(calcfc,verytight,noraman) int=ultrafine FREQ=noraman geom=connectivity

Testing on Methane (CH4) over a number levels of theory I sometime get unequal force constants for the C-H bonds, but only in pairs, e.g., C-H1 and C-H3 are always equal, and C-H2 and C-H4 are always equal, e.g.,

B3LYP/6-311++g**
101.12 101.12 101.12 101.12

B3LYP/6-311++g(2d,2p)

98.76 98.76 207.12 207.12

After looking at the two different frequency results I see that one is very slightly different to the other - see attached. Observing the actual vibrations in motion and I see whilst one calculation (where the force constants are all equal) demonstrates a simultaneous stretch vibration, the other (where the force constants are not equal) pins two of the hydrogens in place, keeping them still whilst the remaining two hydrogens vibrate.

How can I address this obvious error in the frequency calculations? I have checked that the wave function is stable, it is.

See attached for a better description.

+ 2 more attachments

Many thanks Eugene.

3
How can I predict the reaction mechanism in which both reactants and products are known and to find which enzyme residues are important?

If we know the reactants molecules, let us assume X and Y, and final product (z), and the enzyme involved and possible residues (of enzyme) important for reaction to proceed; then can we predict the reaction mechanism?

Is there any computational tool to predict the possible reaction mechanism?

As far as I know, there is no computational tool to predict the reaction mechanism,

To find which enzyme residues are important, usually we can see by blast your protein sequence, and align your enzymes sequence with enzymes related,
For example, your enzyme is shikimate kinase from A organism, so you can check about shikimate kinase from B organism in other paper.

after that you can see the conserved residue, usually conserved residue are important residue in the enzymes. Check about conserved residues role in other paper.
You will see if there is a paper related with your enzymes.

4
How can I get help for supercell of MgB2?

Am working on MgB2 supercell on quantum espresso -5.0 .But when i run scf file on pw.x its always given me error 1 atom overlap 4 atom .Sir i need your assistance on it

Attach with the scf file of MgB2

Thanks

Indeed I forgot that celldm(3) is c/a ratio when structure is set to hexagonal (ibrav=4).

In this way, as Roberto pointed, you should leave celldm(3)=1.155 as before to maintain proper c/a ratio.

4
Which software is best for design of experiment (DOE) in chemistry?

design of experiment in chemistry is important and caused  saving time and material.

many software like spss, mini tab, Design expert 9 and etc used for this Purposes

I like this site"s advice, very much so:

https://learnandteachstatistics.wordpress.com/2013/02/11/excel-spss-minitab-or-r/

4
How do I parallelise the for loop in square matrix multiplication ?

actually i need to reduce time in multiplication of two square matrices of high dimension, let say 2500/2500. presently i am using three nested for loops, it is taking 5 mints time. please any body suggest if there is an alternative way to reduce time.

Try:

square matrix multiplication parallel

6
How can I eliminate the spin contamination in unrestricted states?

I am optimizing some structures at the doublet potential energy surface. I know that the eigenvalue of operator S2 should be near to 0.75 on the doublet potential energy surface (S(S+1)=(0.5(0.5+1)=0.75). But, my structures show 1.45 for the eigenvalue of operator S2.  How can I eliminate these spin contaminations? Does anyone have experience regarding the elimination of the spin contamination?

9
CCSD(T) theory for SPE calculation is taking much time, what should I do?

Dear all

I have to calculate SPEs for some (25-30 molecules of max 15 atoms of 60-65 electrons in one molecule) molecules and applied CCSD(T)/6-311+g(3df, 2p) theory to do that but it is taking a lot of time and above that, for reading & writing file (rwf) it is consuming a lot of disk space also.

I gave 25-30 SPE simulations to one server at a time and it took all the disk space of around 660 GBs in rwf files and gave "link died" for all the simulations.

1) What should I do?

2) Which another level of theory should I apply for SPE (which can give appx. results) and which is balanced b/w time and accuracy?

7
Mean square displacement is constant during the simulations (Gromacs)?

I am simulating the formic acid and formate ion seperately in 1000 spc/e water molecules with Gromacs package.

after 5ns run in nvt ensemble I have calculated MSD from g_msd command for water it is increasing with time and it is giving reasonable diffusion constant value.

but when I calculate for formic acid or formate ion which are single molecule in 1000 water msd is not changing and remains constant and diffusion coefficient is very different.

can anyone suggest me where I am doing wrong and how to correct it.

12 ns is still a very short time. The diffusion of one molecule is going to be extremely noisy, and largely dependent upon the water molecules around it. You may have to simulate for a much longer time (> 50 ns or longer) to get anything sensible here.

3
How do I calculate the adiabatic electron affinity of H3PO at MP2/6-311G++(dp) level?

Hello,

I'm using Gaussian 09 to try and calculate the adiabatic electron affinity (AEA) of H3PO4. I've done this by running a geometry optimizations and frequency calculations of both H3PO4 (neutral) and H3PO4- (anion), after which I try to calculate the AEA by

AEA = E_neutral - E_anion.

where E is the electronic energy + non-electrostatic effects. Regardless of whether I include zero point corrections etc, I get on the order of 7eV.

However, in the paper "A model of single-electron transport. Calculation of the thermodynamic parameters for electron capture by the bound proton of oxyacids" they claim to calculate an AEA of 1.7eV, which is off by quite a lot compared to mine.

Am I doing the calculation of EA wrong in some way? The paper states that they calculate EA using "calculated as the sum of the total energy of interaction of the electrons and nuclei, energy of formation of the solvation cavity (for the aquatic environment), and zero-point energy"; still, doing that I consistently get on the order of 7eV, regardless of molecule conformation.

I know that in the paper they use Gaussion 03, whereas I use 09, but that shouldn't make that much of a difference. My input routes look like this:

#P UMP2/6-311++G(d,p) SCF=(XQC,MaxCycle=512,NoVarAcc)
OPT=(MaxCycles=999)
IOp(1/6=999,1/152=999) Test
Int(SuperFineGrid) CPHF=SuperFineGrid
SCRF(SMD,Solvent=Water)
Freq FormCheck

Can anyone help me figure out how there's such a large discrepancy between my results and the ones in the paper?

Dear Mathias,

For electron affinity the equation to be used is:

AEA = E_neutral - E_anion radical and NOT AEA = E_neutral - E_anion.

This makes a lot of difference.

The G03 and G09 are comparable (no effect is expected).

The basis set will not make this large difference.

In addition, you might consider RHF and UHF (closed and open shell calculations).

Rafik

14
How to calculate all the force constants in Gaussian09?
When use "opt freq" to calculate water, I only get three frc consts, but the literature had four data. Thanks very much.

i am using siesta code to create force constant matrix but i still have problem with big matrix, if any one have some explain can help me

5
Is there any CHARMM force field parameters for heme b and c type for both reduced and oxidized state of heme?

From what I read so far, I see that there are some heme a/a3 force filed parameters for both reduced and oxidized state of heme. Is there similar kind of parameters and also topology files to be used in NAMD for both reduced and oxidized form of heme b and heme c. I know there is reduced form parameters of heme b for CHARMM.

Atanu

Thanks Daniel. Those were really helpful.

5
Can the HOMO-LUMO gap be correlated with electrical conductivity?
I am working with a 13-atom cluster. I can measure its homo-lumo gap. So, from homo-lumo gap, can I get the idea of electrical conductivity. Let, Cluster A has its homo-lumo gap 0.5 eV and cluster B has its homo lumo gap 0.2 eV. In that case, can I say like, cluster B is more electrically conductive than cluster A?

Other thing is: as optical band gap and electrical band gap is quite analogous, can I say like low homo-lumo gap means high optical conductivity?

Dr  Shaikat

HOMO-LUMO gap is defined as E(LUMO) - E(HOMO). Always E(LUMO)  in more positive than E(HOMO), therefor the gap is positive. It also is zero for metallic compounds. You can also find the relation between electrical conductivity and the gap in one of my articles (attached here).

2
How to Visualize Natural Atomic Orbitals of systems wtih 286+ basis functions using JANPA software?

My question concerns visualizing the natural orbitals from the NPA occupation analysis in JANPA for systems with a large number of basis functions. I am currently looking at excited states of TiCl4 and this system has 286 basis functions. I attempted to plot the natural orbitals of TiCl4 using the -NAO_Molden_File runtime flag. It begins to build the molden file but then I get the following error:

Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: 286

at onpa.printout.Export_Orbitals(printout.java:90)

at onpa.Main.main(Main.java:330)

My interpretation of this error is that there is a particular array in the code that has a set limit, and 286 is past that limit. Is there a way to change this so I can view the natural orbitals, or is there a way to selectively plot specific NAOs, I really only need to visualize a few of them. I am currently only using the JANPA executable and have not downloaded the source code.

• Source
##### Article: JANPA: An open source cross-platform implementation of the Natural Population Analysis on the Java platform
[Hide abstract]
ABSTRACT: The algorithms for constructing natural atomic orbitals (NAOs) from a reduced one-particle density matrix as well as for natural population analysis (NPA) are reviewed and their novel independent cross-platform open-source implementation on Java platform, a JANPA package, is presented. The package consists of three programs: two of them generate MOLDEN files from the output of popular quantum chemistry packages (PSI4, ORCA, NwChem, Gaussian, GAMESS-US etc.), and one program performs NAO/NPA transformations taking these data as an input. The developed programs can be freely obtained at http://janpa.sourceforge.net
Theoretical and Computational Chemistry 10/2014; 1050. DOI:10.1016/j.comptc.2014.10.002

Hi, Wallace, how are you?

Is there a way to render the NBOs from a JANPA output?

10
Any literature reference for the kinetics of reversible bimolecular association?

Imagine that you that mix two chemical or biochemical reagents of any kind (enzyme and inhibitor; protein and DNA; pharmacological receptor and a ligand).  These two molecules reversibly form a non-covalent complex.  Necessarily it takes a certain amount of time for the reversible equilibrium to be fully established. The question is: exactly how much time is needed given some information about the "on" and "off" rate constants.

A question like this can be answered by using an integrated rate equation, showing the evolution of all reagent concentrations over time.  My question is, has anyone seen an equation like that, specifically for the reversible bimolecular case. I looked at a good number of textbooks already, and many journal articles, but I simply could not find the equation anywhere, so I had to derive it from scratch - please see the attached graphics.

If, for some reason, the relatively simple equation I just came up with has not been published yet (despite the fact that literally thousands of people must have thought about this simple two-component problem over many decades), then perhaps the attached algebraic formula is even "publishable" in some suitable research journal. At the same time I find it hard to believe that that there isn't a textbook or a monograph showing the same thing.

So: Calling all experts and students of bio/chemical kinetics -- can you please point me to a published integral rate equation that describes the kinetics of reversible bimolecular association?  Thanks in advance for your time...

P.S.: In the attached graphics, E0 and I0 are the initial  (total, analytic) concentrations of the enzyme and the inhibitor, respectively; ka and kd are the association and dissociation rate constants; and C(t) is the concentration of the non-covalent complex "C" at time t.

Gert, thanks very much for the offer to copy pages from the book, but in the meantime I got a copy of it from a used-book vendor.   The bottom line is the Nelson book does not show the equation I had to derive, either.  However, I did find something reasonably close to "my" equation:

• Pladziewicz, J. R., Lesniak, J. S., and Abraham, A. J., Treatment of kinetic data for opposing second-order and mixed first- and second-order reactions, J. Chem. Educ. 63 (1986) 850-851.

It's not the same thing as I have, but it's a good jumping off point for a possible paper presenting the newly derived formula.

Thanks again.

3
How do I calculate band gap by using HSE06 hybrid functional?

computational chemistry

i am also  want to calculate same?

5
Why do I find an unusual conformation of glutamic acid found during MD simulation?

I have performed MD simulation of protein-peptide complex for 300ns time interval using GROMACS-5.0.4 on cluster and write frames at every1ns. I ran the simulation in triplicates. When I was extracting pdb files at different time interval after performing MD. At few instances, COOH group of GLUTAMIC ACID is showing unusual conformation. These unusual conformation are found in all triplicates but at different time frames.
There were clashing at some point of time such as at 50ns the COOH group is showing clash but it is normal at 51ns.
Such type of abnormalities, I have found in many places in protein simulation.
Please let me know the exact reason of this abnormality.

I am not able to understand the reason.

Thanks

Shipra

Shipra, If that's the case you probably dont need to regenerate. But I would suggest to manually inspect your minimized structure that you feed into the MD. If there is a problem in the minimized structure itself it would show up in the trajectory as well. All would recommend to use a smaller times scale for writing the frames to inspect what is happening to the structure in between; may be in the order of few ps.

8
Why would the bond stretch force constants for CH4 be different?

I am reconstructing the force constants from the weighted hessian matrix, as calculated using HF/6-311+G(2d,2p) in Gaussian G09.

The mathematics simple used a submatrix of the atoms of interest (A---B) from the weighted hessian matrix, and projects it onto the bond eigenvectors.

I am running a lot of test cases, one of which is CH4. I would have expected all four bond force constants to be identical. Unfortunately I get:

C1-H2 70 kcal/(mol ang^2)

C1-H3 70 kcal/(mol ang^2)

C1-H4 204 kcal/(mol ang^2)

C1-H5 204 kcal/(mol ang^2)

My maths and my code seems to be correct (from many test e.g., the bond force constant of both O-H in water are BOTH approx 144 kcal/(mol ang^2) at the same level of theory and basis set). I am guessing the results are a direct product of the frequency analysis calculations in Gaussian. Any suggestions?

Thanks

Anthony

Dear Anthony,

The answer to your question can be found in the following paragraph:

One of the cases where Hartree-Fock calculations fail is in the calculation of dissociation energies. This is because these calculations usually start with a molecule that has a closed shell of electrons and the molecule breaks into two fragments that are open shell molecules. A closed shell molecule has all its electrons paired; in an open shell molecule, there is at least one unpaired electron. The energy difference between closed and open shell molecules is very difficult to calculate using Hartree-Fock methods.

Rafik

• Nicola Porta asked a question:
Open
Does anyone have experience in using ligand-protein Interaction Fingerprints with Free Maestro (Schrödinger)?

Hello to everyone,

I am working on the modeling of a class of FMN-dependent NAD(P)H enzymes, where FMN is tightly bound to the protein. Interaction Fingerprints (IF) seems to be a powerfull tool to convert complex 3D-data (eg. a binding pocket) into more menageable, and easier to compare 1D-bit strings. Free-Maestro works well for this purpose but it seems that the IF were calculated only for the ligand-protein couple and not for FMN. In the workspace the first entry is the FMN-protein and the second is the ligand. How can I compute IF for the ligand-protein-FMN complex as if FMN were considered like a residue that belongs to the protein?

Nicola Porta

9
Can anyone suggest how to differentiate the electron transfer and energy transfer phenomenon between two neutral molecules ?

I would like do experiments in order to confirm the electron transfer/ energy transfer phenomenon between the two neutral molecules in solution state. So which experiment is preferred to differentiate the above 2 process in solution state??

14
How to calculate Bond dissociation energy (BDE)?
I want to calculate BDE for hydroxyl groups (-OH) present in my synthetic molecules. Could you please suggest software that can do these calculations. Thanks, Giri

4
Why are the frontier-orbtials' electrons in these conjugated molecular systems occupying the states in such way?

On writing down the Hamiltonian for the particle-in-a-ring problem, and solving the Schrodinger to get the quantized energy in terms of the radius (R) of the ring, for the two examples of Benzene & Coronene.

Why is it acceptable during calculating the excitation energy, like in Benzene, to deal with the model as the the energy is driving the ring from quantum state (n=1) to (n=2)? I mean why are the pairs distributed as: 1) One pair in n=0  2) Two pairs in n=1 ? And similar for Coronene, up to n=6.

Maybe i lead to similar question but now comparatively for Benzene and Coronene & β-Carotene: Why do the distribution of the pairs begin from state (n=1) to (n=11) sequentially each level is filled only by one pair, in the case of β-Carotene, and for the other two examples from (n=0) with one pair, and up to any states by two pairs.

Hopefully, it is clear enough.

Also remember that excitation energies are NOT equal to orbital energy differences. In benzene, both the HOMO and the LUMO are doubly degenerate. The HOMO has E1g symmetry and the LUMO has E2u symmetry (if my memory serves me well).  Single excitation of electrons just among those orbitals gives rise to 4 states (2x2 = 4) of symmetry B1u, B2u, and E1u (direct product E1gxE2u = B1U + B2u + E1u). Two non-degenerate states (B1u and B2u) and one doubly degenerate state (E1u). The B2u and E1u states are optically allowed in one-photon absorption.