# Geometrical Optimization not converging, where am I going wrong?

I am trying to obtain the optical properties of ZnSe Quantum Dots, the problem is that my geometrical optimization is not converging. I used VESTA to create the structure(there is a photo in the attached file), and now I am using Quantum Espresso along with its GUI BURAI. I think there is something wrong with the input file, I have also tried to increase the cut off WF and cutoff energy but it still does not converge. And I also tried switching between damped-md and bfgs but still doesn't help.
For the input and convergence graph(the output file is super big and I don't think it will be that useful)along with the structure, everything is given below,
Input file -

&CONTROL
calculation = "relax"
forc_conv_thr = 1.00000e-02
max_seconds = 1.72800e+05
nstep = 100
pseudo_dir = "C:\Users\aman\.burai\.pseudopot"
/
&SYSTEM
a = 2.50000e+01
degauss = 1.00000e-02
ecutrho = 2.25000e+02
ecutwfc = 2.50000e+01
ibrav = 1
nat = 41
ntyp = 2
occupations = "fixed"
smearing = "gaussian"
/
&ELECTRONS
conv_thr = 1.00000e-04
electron_maxstep = 200
mixing_beta = 7.00000e-01
startingpot = "atomic"
startingwfc = "atomic+random"
/

&IONS
ion_dynamics = "damp"
/
&CELL
/
K_POINTS {gamma}
ATOMIC_SPECIES
Zn 65.39000 Zn.pbe-van.UPF
Se 78.96000 Se.pbe-n-rrkjus_psl.1.0.0.UPF
ATOMIC_POSITIONS {angstrom}
Zn 9.907807 12.921427 10.192708
Zn 9.907807 10.050257 13.063878
Zn 12.778977 10.050257 10.192708
Zn 15.650147 12.921427 10.192708
Zn 9.907807 15.792597 13.063878
Zn 15.650147 10.050257 13.063878
Zn 15.650147 12.921427 15.935048
Zn 15.650147 15.792597 13.063878
Zn 12.778977 15.792597 15.935048
Zn 12.778977 12.921427 13.063878
Zn 12.778977 10.050257 15.935048
Zn 12.778977 15.792597 10.192708
Zn 9.907807 12.921427 15.935048
Se 14.214562 11.485842 8.757123
Se 8.472222 14.357012 11.628293
Se 11.343392 8.614672 14.499463
Se 17.085732 11.485842 11.628293
Se 11.343392 14.357012 8.757123
Se 14.214562 14.357012 17.370633
Se 14.214562 11.485842 14.499463
Se 11.343392 14.357012 14.499463
Se 14.214562 14.357012 11.628293
Se 17.085732 14.357012 14.499463
Se 14.214562 17.228182 14.499463
Se 11.343392 11.485842 17.370633
Se 11.343392 17.228182 11.628293
Se 8.472222 11.485842 8.757123
Se 17.085732 11.485842 17.370633
Se 17.085732 14.357012 8.757123
Se 8.472222 14.357012 17.370633
Se 11.343392 11.485842 11.628293
Se 14.214562 8.614672 11.628293
Se 8.472222 11.485842 14.499463
Se 11.343392 8.614672 8.757123
Se 11.343392 17.228182 17.370633
Se 14.214562 8.614672 17.370633
Se 14.214562 17.228182 8.757123
Se 17.085732 8.614672 14.499463
Se 8.472222 17.228182 14.499463
Se 17.085732 17.228182 11.628293
Se 8.472222 8.614672 11.628293


Convergence graph -

Structure -

• Since the output and these images are small, I think it would be better to include them directly in the post to avoid the possibility of link rot. Also, I don't know what the output for Quantum Espresso includes, but it may also help to include a plot of the mean/max force during the optimization similar to what you have for the energy.
– Tyberius
Jun 8, 2021 at 14:38
• I made the required changes but I do not think I can add Force as it is only calculate when you tell the program to calculate it and I simply did not mainly because of the computation cost. I do not think it will matter much because the optimization algorithm only uses Energy(I think). Thanks for providing your feedback Jun 8, 2021 at 17:19
• And if anyone for anyone answering the question, please also give some feedback on the structure of the quantum dot, I carved it out myself from 8 ZnSe unit cells, using VESTA. Thanks Jun 8, 2021 at 17:20
• You should check out published work on such nano dots, e.g. [Simon Liebing et al 2021 J. Phys.: Condens. Matter 33 125002]. Their ZnSe quantum dot looks quite different from yours. Jun 8, 2021 at 17:32
• The force calculation is turned on by default when you do a relax calculation in QE. Moreover you could cut the quantum dot in such a way that it is neutral. Or try to use the assume_isolated tag Jun 9, 2021 at 6:15