Computational Chemistry

Computational Chemistry

  • Bradley D Rose added an answer:
    Which is better on MP2, Dunning's triple zeta vs Pople's 6-311 G ?

    Hello there. 

    I am a student. 

    I am working with a system which involves mapping PES with a system that has multiple rings,  with MP2 and MO6-2X.

    Does Dunning triple zeta better than the conventional Pople's triple split basis? 

    Thank you in advance for your ideas. 

    Bradley D Rose

    Dunnings basis sets were parametrized with correlation and should be preferred for post-HF calculations, MP2, CCSD, and the like. Pople basis sets were parametrized for Hartree-Fock and are preferred for this, and they have the benefit of being a bit smaller than the Dunning basis sets. DFT tends to show little difference between the two if not a slight preference for the Pople basis sets, and they are faster. The Ahlrichs def2 basis sets are my preferred 'go to' since they are a little more generally applicable than the other two in my experience. Take a look at Christopher Cramers book "Essentials of Computational Chemistry" for a good introduction.

  • Krzysztof Beć added an answer:
    How can I obtain the UV-VIS Spectrum using Gaussian 09?

    I'm quite new to this software and can only design quite basic molecules, not run calculations just yet. I have already tried: Job Type (Energy) - Method (TD-SCF, N=6 states) - Solvation (Default, Default) from other guides, but the UV-VIS result is not an available option to select in the Results tab once run. Are there any key steps that I have missed? An example for a simple molecule would be very much appreciated. 

  • Artem A Babaryk added an answer:
    ¿Does anyone Knows about any software that could compare the same asymmetric crystallographic unit and unit cells at different temperatures?

    ¿Does anyone Knows about any software that could compare the same asymmetric crystallographic unit and unit cells at different temperatures?

    Artem A Babaryk

    Hola Roberto,

    The choice depends on that what it is planned to examine as a trend. If experimental conditions are supposedly the same as Kenneth suggests, more effective analysis can be carried out using the parametric refinement. Please, refer to:

    G. W. Stinton & J. S. O. Evans, Parametric Rietveld refinement, J. Appl. Cryst. (2007). 40, 87-95.

    For sure TOPAS does this due to the macro-language, and GSAS-II is now able to perform similar tasks.



  • Mohamed Safy added an answer:
    How can i using gaussian 09 to optimize a molecule on different concentration?

    I wanted to study the effect of solvent concentration in some properties of molecule? 

    Mohamed Safy

    These are very useful  answers thank  you Prof.karaman,  Mr hosseini and very  big thanks  prof. Ivanova

  • Haamid Rasool Bhat added an answer:
    How can I studies the reaction path of a complex by Gaussian 09 ?
    1. The scheme of reaction is like 

    M-H2O + Im ------> M-Im + H2

    where M = metal complex, Im = imidazole

    Haamid Rasool Bhat

    Dear Siddhartha,

    There are three popular methods to model a reaction:

    1. QST2 

    2. QST3

    3. Initial guess (with or without potential energy scanning). 

    Mind it, Reaction modelling is a very tricky art and please start from very simple systems.

    Please go through the pages i have attached ,,hope it will help

    Good Luck!!!!!!!!!!!!!!!!

    with regards Haamid R Bhat

    + 2 more attachments

  • Sergey Gusarov added an answer:
    I want to know, How can i calculate the mixture of solvent in gaussian 09?. How will i specified the mixture of solvents in gaussiaan inputs?

    I want to know, How can i calculate the mixture of solvent in gaussian 09?. How will i specified the mixture of solvents in gaussiaan inputs? Please anyone can help.....Thanks in advance

    Sergey Gusarov

    You can model thermodynamics/properties of mixtures by :

    1. MD Explicit solvent - any MD package (Amber, LAmmps, etc)

    2. COSMO-RS (COSMO for real solvents e.g. ADF, Turmomol

    3. Three dimensional reference interaction site model (3D-RISM) - ADF, Amber (Ambertools). See e.g. J. Phys. Chem. A, 2006, 110 (18), pp 6083–6090

  • Rafik Karaman added an answer:
    How do I format the PES scan of a proton-transfer reaction in G09?

    Attached is an image of two molecules for which I would like to carry PES calculations- as a proton is migrated from one molecule to the other. I've been trying the attached input in Gaussian09 but I don't get the desired scan output files that I want, it only calculates the SP energy of the input structure. I have been trying to read up on this issue, but most examples are for a relaxed dihedral angle scan. I know that there are differences in input between rigid and relaxed and I seem to be getting them confused.

    Currently, my input section is as follows

    # scan hf/3-21g geom=connectivity
    [internal coordinates/connectivity]
    B31 10 0.34 !H atom B31 with 10 optimization steps 0.34units apart!

    1. How can I first carry out this calculation using the RIGID molecules in the input?
    2. How can I carry the RELAXED scan of the same input structure.

    For the purpose of these calculations I want to keep all atoms on the molecule fixed in one spot while the H-atom is being translated towards the other molecule. My ultimate goal is to scan the potential energy surface of this proton-transfer reaction.

    Help is appreciated,


    + 1 more attachment

    Rafik Karaman

    Dear Patrick,

    Attached is the requested file.


  • Hamed Haghshenas added an answer:
    Can someone advise on the Error l914 in TD-OPT calculation with gaussian?
    I do TD-opt calculations for geometry optimization in excited state, and I get this error: You need to solve for more vectors in order to follow this state.
    Error termination via Lnk1e in /..../.../g09/l914.exe.
    How can I fix this error?
    Hamed Haghshenas

    Dear samaneh,

    Error termination in l914.exe followed by "NIJ> Max2 in MMCore" failure message is usually related to the memory allocation bug for large systems in gaussian revision B and earlier. Fortunately this problem has been solved in gaussian revision C.

    All the Best


  • Aleksey E. Kuznetsov added an answer:
    Issue with the frequency calculations: 36676847 words are not enough for AlAXAO?

    Dear All,

    while running the frequency calculation for the system with 2 Pd atoms and 96 lighter atoms, B3LYP functional used, 6-31G* basis set for lighter atoms and Lanl2dz for Pd, SCRF (PCM model, DMFA as a solvent) used, the calculation stopped with the following error message:

    AlAXAO: NMat= 225 NPMax= 16 NPMax1= 0 MaxMat= 72 Max3X=T.
    36676847 words are not enough for AlAXAO.
    Error termination via Lnk1e in /opt/apps/gaussian/09/g09/l1002.exe

    Unfortunately, Google search did not give me any hints how to solve this problem. Did anybody encounter with such issue? Does it mean that I should assign read-write files for this calculations?

    Thank you in advance!

    Best regards,


    Aleksey E. Kuznetsov

    Dear Massimiliano and Bartosz,

    thank you for your advices! Actually, I always do exactly as you described: first, optimization and freqs in the gas phase, then, on top of it, optimization followed by freqs in the solvent phase. And I was able to overcome this issue with "NaN" using larger amount of memory in frequencies calculations.

    Best regards,


  • Vaibhav Kaware added an answer:
    How to define a crystal structure in quantum-espresso (pwscf)?

    In the tutorial of quantum-espresso (pwscf), the tutor demonstrated how to set up a unit cell using the parameters list below.




    Only 1 atom is enough to define this face-centered cubic. As far as I understand from this parameter, the software will know where to place 14 atoms in one cubic (8 corners and 6 face centers).

    However, I was confused by the second example. The tutor only used 2 atoms to define a diamond structure (Fig 1).





    Si 0.0 0.0 0.0

    Si 0.25 0.25 0.25

    If the first Si was considered as the atoms whom should be placed in the 8 corners and 6 face centers in one cubic, how does the software know where to place the second Si? How does the software distinguish the diamond structure from fluorite structure (Fig2)?

    + 1 more attachment

    Vaibhav Kaware

    Dear Sinkander,

      I typically use these options. Check if this helps.

    cif2cell -f nis2-1991.cif -p quantum-espresso -o --cartesian --coordinate-tolerance=0.01 --setup-all

    Please also consider the small but finite possibility that  " case.cif "  may not comply with the cif format and/or is erroneous.

  • Md Junaid added an answer:
    What are the steps to run the Gaussian 09 for Ruthenium complexes?
    I tried to do the optimization job for entire ruthenium complex by using the method contained Ground state, DFT, Default spin, B3LYP, with Basis sets LanL2DZ. But it terminated due convergence criterion not met with in 128 and hence I obtained as follows

    >>>>>>>>>> Convergence criterion not met.
    SCF Done: E(RB3LYP) = -5139.07405391 A.U. after 129 cycles
    Convg = 0.7104D-04 -V/T = 2.0298
    Convergence failure -- run terminated.
    Error termination via Lnk1e in C:\G09W\l502.exe at Sat Mar 22 20:44:28 2014.

    What are steps to follow the optimization of ruthenium complexes?
    Md Junaid

    Dear the easier method you use is that Ground state, DFT, Default spin, B3LYP, with Basis sets SDD.Remind that when you use metal basis set SDD must needed.

    Thnk u

  • Srimanta Pakhira added an answer:
    How can I resolve the 'convergence failure' error in Gaussian 09?

    Hello everyone,

    I deal with the calculations of QM charges for protein fragments. Each fragment consists of aproximately 250 non-hydrogens atoms.

    Currently I started these calculation using B3LYP theory level, NPA distribution scheme and 6-31G* basis set.

    I ran these calculations for 41 fragments. However, only 20 of them ended with the 'Normal termination'. The remaining 20 fragments ended with this error message:

    >>>>>>>>>> Convergence criterion not met.
    SCF Done: E(RB3LYP) = -14532.8135584 A.U. after 129 cycles
    NFock=128 Conv=0.48D-03 -V/T= 2.0071
    Convergence failure -- run terminated.
    Error termination via Lnk1e in /afs/ at Sat Mar 28 13:01:44 2015.

    Do you have any idea how to solve this problem?

    Srimanta Pakhira

    You can try scf=(maxcycle=3000) scf=(conver=4). It may helpful for you.

  • Ashar J Malik added an answer:
    Can anyone explain the algorithm for the calculation of change in energy by Ewald sum method?

    I am having problem with ewald sum energy calculation part.

    I have written the code and validated it with NIST data for total coulomb potential, it is matching well. But when i am trying to calculate  change in coulomb potential (new configuration - old configuration), i am confused with this part.

    Ashar J Malik


    It is not clear from what you have written - to completely grasp what the problem is. Could you please add some more information?

    What is the system you are studying - values will depend on that.

    When you say you wrote some code - does that mean you wrote it yourself - or put together libraries? What language are you using?

    What exactly is confusing you? Are the values not what you are expecting?

    If you add the above information you will be better able to get a coherent answer.


  • Xiao-yong Zhang added an answer:
    Is it convergence problem with my Transition State Calculation?

    Hello, dear friends.  

    I have encountered serious convergence problems in the process of finding right TSs. I first conducted the calculation with maxstep=20, then decreased the step to 10, finally to 1, but as the attached figure showed, the energy was in shape oscillation and it seems much difficulty to reach the convergence criteria of displacement. 

    The keywords were set as "# opt(ts,readfc,modredundant) b3lyp/6-31g(d) empirical dispersion=gd3 iop(2/15=1,5/13=1,1/11=1,1/8=1)  optcyc=50" 

    Any suggestions are welcome. 

    Best Regards

    Xiao-yong Zhang

    Thanks for your answer. The system contains up to 180 atoms with significant π-π interactions, it is till of doubt wether the computer facility could afford such large system using QST2 to locate TS. I am sorry I could not provide the detailed molecular structure, since those are related to unpublished work.

    The molecular model competes in triplet and quintet state, and fortunately we have got the TSs on the quintet PES, while failed on some cases of the triplet PES. For those faied cases, the first two convergency criteria can  be easily reached and the eigenvector is as expected. 

    So I thought the problem maybe related to the opt algorithnm of Gaussian software, or something beyond my knowledge. 

  • Edwin A. Murillo added an answer:
    Can I run this calculation without NETZSCH ThermoKinetics software?

    how can I make calculations of activation energy of a polymer by TGA analysis I do not have the NETZSCH ThermoKinetics software ¿can I make this calculation without employed this software?

    Edwin A. Murillo

    Thank you very much for all.

    Best regards

  • Maximo Ramirez added an answer:
    Can anyone tell that the total electron density is the representation of total electrons or valence electrons?

    When we plot the CHGCAR in VESTA the representation shows the electron density. My doubt is that is it the representation of total electrons or valence electron?

    Maximo Ramirez

    About of CHGCAR,

    Can I get the valence state or oxidation number from one element of compound?

    For example if you have AB2

    and you can get the valence of A or B element in the compound.

  • Asma Iqbal Khan added an answer:
    Can someone help me with the pervaporation Using Aspen Plus?

     I am working on the sepration processes for azeotropic mixtures, one of them is pervaporation.
    I have been trying to model and then simulate the pervaporation process in either Aspen Plus with a user defined model or with Aspen custom modeler. I have been trying to see how I can modify the Excel template supplied by Aspen plus for filtration to suit pervaporation but I haven't been successful to find parameters and equations for pervaporation process.

    If anyone has done any similar work on pervaporation, its modeling and simulation with either Excel, Aspen Plus or Aspen Custom modeler or a combination of any.?I would be grateful to receive any example files or notes.

    Asma Iqbal Khan
    Thanks Martin. J.Wolf sir.
  • Radhika Geethu added an answer:
    Does docking of a telomere with a ligand is reliable using blind docking method?

    Dear Friends, I used auto dock vina to dock telomere with a ligand. As I don't know which guanine in the G-quadruplex  could be much  specific for that ligand, I have done blind docking. But the result was not the one which I expected. The intercalation has to be happen in between the  stacked bases. But in this case, interaction is something different (end stacking interaction). Whether this is because of blind docking ? Else can I consider this kind of interaction as a significant result? Herewith I have attached the telomere and the docked complex for your reference. Can I get any suggestions.  

    With regards,

    R. Radhika

    + 2 more attachments

    Radhika Geethu

    Thank you so much Dr. Ignor Dubey, I will do it using ONIOM ,which gives considerable accuracy with low computation cost. All the best for your research..

  • Christopher Xiaoquan Sun added an answer:
    What things should be take into account, when building structure for CASTEP calculations in material Studio?

    I want to Calculate band structures and phonon calculations by using CASTEP, so kindly guid that what things are important for building the crystal structures for such calcualtion in Material Studio?

    Christopher Xiaoquan Sun

    Agree with Otto, reading through those document is like studying simulation throughly.

  • Christopher Xiaoquan Sun added an answer:
    What does CPU time stand for in the OUTCAR of VASP?

    The CPU time in at the end of OUTCAR is very much different for the time it take in reality.

    I want to know from where I will got the exact time of calculation?

    As I am checking the same file on different number of nodes so I can compare the time of calculation? 

    Christopher Xiaoquan Sun

    I know the computing time in Gaussian is real time * the number of core

  • Rabi Khanal added an answer:
    How I can select different memory in vasp pbs file for calculating our result?

    Computational Chemistry

    Rabi Khanal

    If you mean in your job script - one way is to use unpacked nodes. For example you chose 12 nodes each with 16 cores and you run only in 12 cores in each node so that memory of remaining 4 cores in each node will be shared among 12 cores that you are using. You could do this by using -N (nodes)  and n (total tasks).  Here -N 12 -n 144 will use 12 core/nodes. It also depends upon the computer system you are using to do your calculations. 

    Some supercomputers have designated nodes with higher memory and the way you assigned depend on how it is designed in particular to that supercomputer. Depending upon the system you are doing calculation on using ISYM=2 tag in INCAR helps to reduce memory requirement in your calculation as well. 

    Good Luck! 

  • Vladimir Potemkin added an answer:
    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?

    Vladimir Potemkin

    You can calculate the difference between Gibbs Free Energies for dimer (many-particle system) and separate molecules using on-line "Thermodynamics: Heat Capacity, Enthaly, Entropy, Gibbs Free Energy, log(BCF)" software at

  • Vladimir Potemkin added an answer:
    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? 

    Vladimir Potemkin

    You can model the slab surface of Fe3O4 (111) with any ligand using "Electron properties calculation: Integration over atomic basins" (uses AlteQ quantum approach) or "ChemoSophia Elastic Calculations" (uses MERA non-quantum approach) for many-particle systems at

  • John Clarkson asked a question:
    How do I use VEDA to do quick PED of isotopomers?

    I have used VEDA to perform PED for my molecule. The vibrational analysis was done in Gaussian 09W. How do I get Gaussian to provide files for VEDA to do PED using the same internal coordinates for my molecules isotopomers?

    Do I have to re-calculate the frequencies in Gaussian 09W to provide VEDA with the formatted checkpoint file for every isotopomer?

  • Mansour H. Almatarneh added an answer:
    Can anyone tell me where I can find input for basis set for iodine atom, I need 6-31G(d,p) format code for Gaussian09?
    Iodine 6-31G(d,p)
    Mansour H. Almatarneh

    check this site: and you will find many basis sets that you can choose from..happy computing!

  • Rituparna Duarah added an answer:
    What factors determine the complete conversion in the synthesis of nanoparticle?

    What factors determine the 100% conversion during synthesis of nanoparticles?

    Rituparna Duarah

    Yes, that's one of the confusions I had whether 100% conversion is actually possible during nano particle synthesis or not.

  • Rafik Karaman added an answer:
    How do I join together the steps of a reaction coordinate's free energy with intermediates in mind?

    I have a very long pathway to map using DFT. There are approximately 15 intermediates. I am using Gaussian to capture the reactants, intermediates, transitions states and the products. 

    I have so far managed to successfully isolate my first five transition states. From going from one transition state to the other, the product of one which essentially becomes the reactant of the next, needs some changing e.g., a change in number of water molecules. Now I fin that the DeltaG (also the DeltaH, DeltaE and DeltaSCF) are way off.

    For examples:

    Re<--->(+16.54kcal/mol)TS1<--->(+16.10kcal/mol)Int1  This involved two water molecules.

    The next step only involves one water molecule on the other side of this long fragment. So I took Int1 (from above) and added a new water molecule on the other side. Note: I took out the first two water molecules, as they are not directly involved at this stage of the reaction mechanism. 


    Unfortunately, the problem occurs when I try and compare any of the energy measurements (enthalpy, gibbs free, SCF, ZPE, etc..) between In1 (product 2 water molecules) with Int2 (the reactant, only one water molecule). There is a huge +401037.58 kcal/mol difference. So in total....

    Re<--->(+16.54kcal/mol)TS1<--->(+16.10kcal/mol)Int1 <--->(+401037.58kcal/mol) Int2<--->(+42.27kcal/mol)TS2<--->(+30.29kcal/mol)Int3.

    I am now certain that it is because I took out the first two water molecules. But I thought DeltaG was a state function, so in theory, that reaction would/could happen even without those two water molecules present from the previous reaction. 

    Would really appreciate some assistance with this one. 



    Rafik Karaman

    Dear Jonathan,

    If you add the Free energy of separate water molecule you will end with incorrect value.

    The best way is to add the water molecule to the structure with less water molecule and to optimize by DFT. By this way you balance the number of atoms.

    Doing the calculation as it is shown below will lead to false result.

    G(intA) = G(Molecule+3 water)

    G(intB) = G(Molecule+1 water) + G(2 water)


  • Saeideh Hosseini added an answer:
    Vibrational frequencies - How can I study ibuprofen?

    I have the vibrational frequencies of ibuprofen(IBF) using DFT methods which performed using gaussian 03 program.I want to assign these frequencies with the suitable vibration motion seen at gauss proplem, in some frequencies there are many vibration motion in different moieties of IBF molecules and i want know which motions is corresponding to these frequency. Also I need some tutorial in vibrational assignments and i want some books to explain me the forms of vibration motions of  groups such as (Methylene, Methyl, Carboxyl, Nitro,......)

    can anyone help me in this proplem?

    Saeideh Hosseini

    you can see the attached links

    + 2 more attachments

  • Oleg Gromov added an answer:
    How do cpu cores and max memory effect DFT numerical frequencies calculations?

    I am a complete beginner to computational chemistry but have recently started using ORCA to calculate vibrational spectra of various long side chain glycolamides. However, these molecules can have 100-200 atoms and while the geometry optimisation is relatively quick the numerical frequencies calculation is very slow.

    Currently I am using serial processing on an i7 4790 using only about 4 gb of memory out of a total of 16 gb (DDR3) available.

    In order to speed up the numerical frequencies calculation should I use parallel processing on two of the other cores available (using 3 cores in total out of the 4) and split the memory or would it be better to still use a serial method and just assign more memory to the calculation (8 or 12 gb)?

    I also might have access to an older AMD Opteron 2423 HE (6-core) dual cpu system with 32 gb of memory (DDR2). Would it be faster on this system compared to my current i7 system using parallel processing and assigning 3 gb of memory per core?

    Oleg Gromov


    That's quite reasonable, as in short runs amount of time program is performing i/o operations, garbage collection, passing data from one subroutine to other (and many other operations that are not paralleled) becomes comparable to calculation itself



  • Justin Lemkul added an answer:
    How to calculate Lennard-Jones parameters for a particular atom?

    It is often necessary to know the parameters of Lennard-Jones (LJ) potential for various atoms. There may be database for LJ parameters. I am just curious to know how these parameters are obtained from experimental data or using ab initio computational methods.

    Justin Lemkul

    That is the point of the question - one cannot do the simulation without the requisite parameters.

Topic followers (6,402) See all