Hydrogen bond life time?

How can I calculate Hydrogen bond life time in Gromacs between two groups like water-water and water-Ammonium group?

gmx hbond can provide lifetime of a specific h-bond. But I am not sure, it is the best approach for a liquid mixture (like your case).
Best regards

I don’t remember, but probably VMD has that option?

Dear alevilla

Thank you for your reply. Its true but I have heard the life time from gmx hbond is not very exact.

Thank you

Dear jakubh

in VMD we can just see hydrogen bonds.

Thank you

There are different approaches to evaluate h-bond life time (not use if one can say which one is more exact of another). The lifetime of the H-bonds in gmx hbond is calculated from the average over all autocorrelation functions of the existence functions (either 0 or 1) of all H-bonds:
C(τ) = ⟨si(t) si(t+τ)⟩ with si(t)={0,1}
for H-bond i at time t. The integral of C(τ) gives a rough estimate of the average H-bond lifetime τHB. Note that
both the integral and the complete autocorrelation function C(τ) will be output, so that more sophisticated analysis ( e.g. using multi-exponential fits) can be used to get better estimates for τHB.
For more detail and reference you can look at

The references (in the page above) may help you to find the best approach for your analysis.

Best regards

Dear Allevila

Thank you very much for sharing your valuable information. It helped me alot.


very simple, please check terminal after use " gmx hbond" module, gromacs automatically report number and life time of hbond in screen terminal

Dear Hamid
Thank you very much for your reply.


Dear hamid,


I calculated H-bond life time and gromacs reported number of lines on the terminal screen like these:

Type Rate (1/ps) Time (ps) DG (kJ/mol) Chi^2
Forward 0.007 138.003 16.742 0.000466026
Backward 0.000 3999.828 25.088
One-way 0.014 71.387 15.108
Integral 0.005 204.424 17.716
Relaxation 0.015 67.350 14.964
HB lifetime=0.85ps

which line is related to H-bond exactly? I think HB lifetime=0.85ps is true. But, what are others?

Thank you , in advance


Hi Mohammad,
people from my department calculated in this way 12 years ago. In gromacs you use autocorellation function for hbonds, then you must do Multiexponential Nonlinear Fits to Time Correlation Functions Table 5 https://pubs.acs.org/doi/10.1021/jp8048235 to get A and T - A will give you information how many bonds break (%) in T time. For example A1 = 20,6 % T1 = 4,1 ps - so 20,6 % hbonds breaking in 4,1 ps

Dear jakubH

Thank you very much for you reply. Your method provide better results compared to results provided by Gromacs.
I try to do your method.

Thanks again