From owner-chemistry@ccl.net Fri Dec 8 04:43:00 2023
From: "Grigoriy Zhurko reg_zhurko%%chemcraftprog.com" <owner-chemistry*_*server.ccl.net>
To: CCL
Subject: CCL: Calculation of Gibbs energies of molecules with small frequencies
Message-Id: <-55053-231208032901-24889-TRiUE5FHM0T1ABrHYi2fSA*_*server.ccl.net>
X-Original-From: Grigoriy Zhurko <reg_zhurko###chemcraftprog.com>
Content-Type: multipart/mixed;
boundary="----------0690421002E5522D1"
Date: Fri, 8 Dec 2023 11:28:34 +0300
MIME-Version: 1.0
Sent to CCL by: Grigoriy Zhurko [reg_zhurko/./chemcraftprog.com]
------------0690421002E5522D1
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I have one more question concerning the msRRHO. If this approach implies the attached formula, it works improperly when the computation has negative frequencies. Such frequencies are sometimes produced as a result of numerical noise (I got a lot of them for a system with PCM water solvation model and explicit water molecules). Did anyone suggest to make the negative frequencies zero or e.g 100 cm-1 in such cases? According to my experience, it is difficult and CPU-consuming to get rid of the negative frequencies in such cases, and possibly not necessary, if we simply make small frequencies bigger.
Grigoriy Zhurko.
------------0690421002E5522D1
Content-Type: image/png;
name="form_msrrho_prakht.png"
Content-transfer-encoding: base64
Content-Disposition: attachment;
filename="form_msrrho_prakht.png"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------------0690421002E5522D1--
From owner-chemistry@ccl.net Fri Dec 8 06:51:01 2023
From: "Philipp Pracht pp555 a cam.ac.uk" <owner-chemistry=server.ccl.net>
To: CCL
Subject: CCL: Calculation of Gibbs energies of molecules with small frequen
Message-Id: <-55054-231208064914-8365-dpn+ykSwSpWmomqzxdPTCw=server.ccl.net>
X-Original-From: "Philipp Pracht" <pp555[-]cam.ac.uk>
Date: Fri, 8 Dec 2023 06:49:11 -0500
Sent to CCL by: "Philipp Pracht" [pp555===cam.ac.uk]
Dear Grigoriy,
I don't know whether a proper citation for this exists, but what one can invert all negative frequencies within a certain cut-off and treat them as positive (i.e. interpolate or set to 100 cm^-1). In our codes we us -50cm^-1 as the threshold for inversion. This works well for the numerical issues you describe. For larger imaginary modes one has to be cautious, these indicate transition states or saddle points on the energy surface.
best wishes,
Philipp
On 08.12.23 08:28, Grigoriy Zhurko reg_zhurko%%chemcraftprog.com wrote:
> I have one more question concerning the msRRHO. If this approach implies the attached formula, it works improperly when the computation has negative frequencies. Such frequencies are sometimes produced as a result of numerical noise (I got a lot of them for a system with PCM water solvation model and explicit water molecules). Did anyone suggest to make the negative frequencies zero or e.g 100 cm-1 in such cases? According to my experience, it is difficult and CPU-consuming to get rid of the negative frequencies in such cases, and possibly not necessary, if we simply make small frequencies bigger.
> Grigoriy Zhurko.
--
Dr. Philipp Pracht (pp555]~[cam.ac.uk)
Yusuf Hamied Department of Chemistry
University of Cambridge
Lensfield Road, Cambridge CB2 1EW
United Kingdom
From owner-chemistry@ccl.net Fri Dec 8 08:08:01 2023
From: "Stefan Grimme grimme * thch.uni-bonn.de" <owner-chemistry#,#server.ccl.net>
To: CCL
Subject: CCL: Calculation of Gibbs energies of molecules with small frequencies
Message-Id: <-55055-231208080547-6616-bWGleKKGhXoRgQY2yaBbuQ#,#server.ccl.net>
X-Original-From: "Stefan Grimme" <grimme[]thch.uni-bonn.de>
Date: Fri, 8 Dec 2023 08:05:43 -0500
Sent to CCL by: "Stefan Grimme" [grimme~!~thch.uni-bonn.de]
Dear Grigoriy,
our practice is to make (small) imaginary frequency values
real and include them in mRRHO. Imaginary modes caused by numerical issues (often methyl rotations) usually correspond to small valued real ones.
Because mRRHO effectively damps their effect, this is much less a problem than in RRHO where such modes (even if they are just neglected) can cause
large errors. However, modes with values e.g. larger than i50-100 cm-1 should be explicitly treated. For a reference see
Angew. Chem. Int. Ed., (2022), 61, e202205735. DOI: 10.1002/anie.202205735
Best
Stefan
From owner-chemistry@ccl.net Fri Dec 8 19:49:01 2023
From: "Prasanta Bandyopadhyay mailprasanta88~~gmail.com" <owner-chemistry:_:server.ccl.net>
To: CCL
Subject: CCL:G: Start post-Hartree Fock calculation from Localized orbitals in PySCF
Message-Id: <-55056-231208190550-18524-eDAy5arUVRUdMo0s1jdFiw:_:server.ccl.net>
X-Original-From: "Prasanta Bandyopadhyay" <mailprasanta88||gmail.com>
Date: Fri, 8 Dec 2023 19:05:49 -0500
Sent to CCL by: "Prasanta Bandyopadhyay" [mailprasanta88=gmail.com]
Hello everyone. This is my first message to the community.
It is known that the Localized Molecular Orbitals can do CCSD calculations in less time. After a canonical HF calculation, I am trying to do Boys Localization and feed the localized orbitals to start the CCSD calculation in PySCF. I did not find any such references or examples. So, my input may also be erroneous. If my inputs are correct, then there may be some problems with CCSD convergence.
I have run a sample code in ipython notebook and sharing the input/output.
Please help.
Python 3.10.11 | packaged by conda-forge | (main, May 10 2023, 18:58:44) [GCC 11.3.0]
Type 'copyright', 'credits' or 'license' for more information
IPython 8.15.0 -- An enhanced Interactive Python. Type '?' for help.
...: atom = [
...: [ 'H' , 0.000000 , 0.000000 , 0.100000],
...: [ 'H1' , 0.000000 , 0.000000 , 0.860000],
...: [ 'H' , 2.000000 , 0.000000 , 0.100000],
...: [ 'H1' , 2.000000 , 0.000000 , 0.860000]],
...: basis = {'H': gto.parse('''
...: H S
...: 0.9000000000D+00 0.1000000000D+01
...: H S
...: 0.3000000000D+00 0.1000000000D+01
...: '''),'H1': gto.parse('''
...: H S
...: 0.9900000000D+00 0.1000000000D+01
...: H S
...: 0.3900000000D+00 0.1000000000D+01
...: ''')
...: },
...: cart=True
...: )
...: mf=scf.RHF(mol)
...: mf.kernel()
...: # Canonical CCSD calculation
...: ccsd_can = cc.CCSD(mf)
...: ccsd_can.kernel()
System: uname_result(system='Linux', node='prasanta', release='6.2.0-37-generic', version='#38~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov 2 18:01:13 UTC 2', machine='x86_64') Threads 6
Python 3.10.11 | packaged by conda-forge | (main, May 10 2023, 18:58:44) [GCC 11.3.0]
numpy 1.24.3 scipy 1.10.1
Date: Fri Dec 8 23:45:32 2023
PySCF version 2.3.0
PySCF path /home/pro/psi4conda/lib/python3.10/site-packages/pyscf
[CONFIG] conf_file None
[INPUT] verbose = 4
[INPUT] num. atoms = 4
[INPUT] num. electrons = 4
[INPUT] charge = 0
[INPUT] spin (= nelec alpha-beta = 2S) = 0
[INPUT] symmetry False subgroup None
[INPUT] Mole.unit = angstrom
[INPUT] Cartesian GTO integrals (6d 10f)
[INPUT] Symbol X Y Z unit X Y Z unit Magmom
[INPUT] 1 H 0.000000000000 0.000000000000 0.100000000000 AA 0.000000000000 0.000000000000 0.188972612457 Bohr 0.0
[INPUT] 2 H1 0.000000000000 0.000000000000 0.860000000000 AA 0.000000000000 0.000000000000 1.625164467126 Bohr 0.0
[INPUT] 3 H 2.000000000000 0.000000000000 0.100000000000 AA 3.779452249130 0.000000000000 0.188972612457 Bohr 0.0
[INPUT] 4 H1 2.000000000000 0.000000000000 0.860000000000 AA 3.779452249130 0.000000000000 1.625164467126 Bohr 0.0
nuclear repulsion = 2.41641498659376
number of shells = 8
number of NR pGTOs = 8
number of NR cGTOs = 8
basis = {'H': [[0, [0.9, 1.0]], [0, [0.3, 1.0]]], 'H1': [[0, [0.99, 1.0]], [0, [0.39, 1.0]]]}
ecp = {}
CPU time: 0.76
******** <class 'pyscf.scf.hf.RHF'> ********
method = RHF
initial guess = minao
damping factor = 0
level_shift factor = 0
DIIS = <class 'pyscf.scf.diis.CDIIS'>
diis_start_cycle = 1
diis_space = 8
SCF conv_tol = 1e-09
SCF conv_tol_grad = None
SCF max_cycles = 50
direct_scf = True
direct_scf_tol = 1e-13
chkfile to save SCF result = /home/pro/tmpho51b2t8
max_memory 4000 MB (current use 133 MB)
Set gradient conv threshold to 3.16228e-05
Initial guess from minao.
init E= -1.39106822755651
HOMO = -0.433617041428962 LUMO = 0.282308074774963
cycle= 1 E= -2.06295523982167 delta_E= -0.672 |g|= 0.0727 |ddm|= 0.799
HOMO = -0.473556403806346 LUMO = 0.412321896221053
cycle= 2 E= -2.06454743844608 delta_E= -0.00159 |g|= 0.0118 |ddm|= 0.0747
HOMO = -0.469582406635784 LUMO = 0.416563705787143
cycle= 3 E= -2.06459105895304 delta_E= -4.36e-05 |g|= 0.000111 |ddm|= 0.0144
HOMO = -0.469553180529867 LUMO = 0.416580776242043
cycle= 4 E= -2.06459106222831 delta_E= -3.28e-09 |g|= 6.29e-06 |ddm|= 0.000152
HOMO = -0.469554412248582 LUMO = 0.4165811859113
cycle= 5 E= -2.06459106224563 delta_E= -1.73e-11 |g|= 3.17e-07 |ddm|= 1.39e-05
HOMO = -0.469554206106277 LUMO = 0.416581169256107
Extra cycle E= -2.06459106224566 delta_E= -3.73e-14 |g|= 5.96e-08 |ddm|= 3.61e-07
converged SCF energy = -2.06459106224566
******** <class 'pyscf.cc.ccsd.CCSD'> ********
CC2 = 0
CCSD nocc = 2, nmo = 8
max_cycle = 50
direct = 0
conv_tol = 1e-07
conv_tol_normt = 1e-05
diis_space = 6
diis_start_cycle = 0
diis_start_energy_diff = 1e+09
max_memory 4000 MB (current use 141 MB)
Init t2, MP2 energy = -2.09735912778898 E_corr(MP2) -0.0327680655433131
Init E_corr(CCSD) = -0.0327680655433144
cycle = 1 E_corr(CCSD) = -0.0433288723433739 dE = -0.0105608068 norm(t1,t2) = 0.0338082
cycle = 2 E_corr(CCSD) = -0.0469749314685062 dE = -0.00364605913 norm(t1,t2) = 0.0129309
cycle = 3 E_corr(CCSD) = -0.0491424777892665 dE = -0.00216754632 norm(t1,t2) = 0.00506568
cycle = 4 E_corr(CCSD) = -0.0491155494004375 dE = 2.69283888e-05 norm(t1,t2) = 0.000214851
cycle = 5 E_corr(CCSD) = -0.0491166716803206 dE = -1.12227988e-06 norm(t1,t2) = 2.73751e-05
cycle = 6 E_corr(CCSD) = -0.0491164004888309 dE = 2.7119149e-07 norm(t1,t2) = 7.07441e-06
cycle = 7 E_corr(CCSD) = -0.0491163959830784 dE = 4.50575245e-09 norm(t1,t2) = 2.03493e-06
CCSD converged
E(CCSD) = -2.113707458228743 E_corr = -0.04911639598307844
Out[1]:
(-0.04911639598307844,
array([[ 9.66395727e-04, -7.15861468e-17, 3.73379105e-03,
2.44065462e-18, -1.57904882e-03, -1.32952145e-16],
[ 2.00856366e-16, -1.07204164e-03, 2.38841448e-16,
3.31332992e-03, 8.65144493e-17, -1.87163682e-03]]),
array([[[[-5.54907257e-02, -6.98304215e-17, 4.96365459e-03,
6.15454933e-17, 1.62904246e-02, 2.35860458e-16],
[-6.98304215e-17, -4.32600566e-02, -1.46096206e-17,
-6.18073805e-03, 2.82455911e-16, -1.38761644e-02],
[ 4.96365459e-03, -1.46096206e-17, -1.76383333e-02,
3.47288702e-17, -7.08184505e-04, -3.53184896e-18],
[ 6.15454933e-17, -6.18073805e-03, 3.47288702e-17,
-1.34470086e-02, 2.46462720e-17, -7.67761587e-04],
[ 1.62904246e-02, 2.82455911e-16, -7.08184505e-04,
2.46462720e-17, -1.08204284e-02, -1.05483455e-17],
[ 2.35860458e-16, -1.38761644e-02, -3.53184896e-18,
-7.67761587e-04, -1.05483455e-17, -9.06019104e-03]],
[[-2.71717382e-16, 5.62101570e-02, 8.10011333e-17,
7.13198951e-03, -2.29155940e-16, 1.58849468e-02],
[ 4.83892195e-02, 2.06293400e-16, -5.44502289e-03,
-2.97797424e-17, -1.81692514e-02, -2.42204538e-16],
[ 5.75513408e-17, -5.73018600e-03, 4.25602864e-17,
-1.63495118e-02, -1.40076323e-18, 5.15636384e-04],
[ 6.22957127e-03, -3.83416194e-17, -1.65269939e-02,
-6.88327791e-18, -1.97040906e-03, -2.58176270e-17],
[-2.00659956e-16, -1.75761121e-02, 4.95767349e-18,
-2.03951127e-03, 3.50784025e-16, -1.06212681e-02],
[ 1.48299958e-02, -2.38310658e-16, 4.72373240e-04,
-2.78193986e-17, -1.06327588e-02, -3.40485550e-16]]],
[[[-2.71717382e-16, 4.83892195e-02, 5.75513408e-17,
6.22957127e-03, -2.00659956e-16, 1.48299958e-02],
[ 5.62101570e-02, 2.06293400e-16, -5.73018600e-03,
-3.83416194e-17, -1.75761121e-02, -2.38310658e-16],
[ 8.10011333e-17, -5.44502289e-03, 4.25602864e-17,
-1.65269939e-02, 4.95767349e-18, 4.72373240e-04],
[ 7.13198951e-03, -2.97797424e-17, -1.63495118e-02,
-6.88327791e-18, -2.03951127e-03, -2.78193986e-17],
[-2.29155940e-16, -1.81692514e-02, -1.40076323e-18,
-1.97040906e-03, 3.50784025e-16, -1.06327588e-02],
[ 1.58849468e-02, -2.42204538e-16, 5.15636384e-04,
-2.58176270e-17, -1.06212681e-02, -3.40485550e-16]],
[[-5.46358905e-02, 1.30287310e-16, 5.13339295e-03,
8.49872253e-17, 1.95471507e-02, 3.66679032e-16],
[ 1.30287310e-16, -5.74430970e-02, -5.56956115e-17,
-7.49341692e-03, 2.84447125e-16, -1.68101818e-02],
[ 5.13339295e-03, -5.56956115e-17, -2.14792869e-02,
-1.04757120e-17, -6.44811716e-04, -1.83577455e-18],
[ 8.49872253e-17, -7.49341692e-03, -1.04757120e-17,
-1.65366923e-02, 1.87452778e-17, -7.38305129e-04],
[ 1.95471507e-02, 2.84447125e-16, -6.44811716e-04,
1.87452778e-17, -1.26396619e-02, -4.00584700e-17],
[ 3.66679032e-16, -1.68101818e-02, -1.83577455e-18,
-7.38305129e-04, -4.00584700e-17, -1.06101681e-02]]]]))
In [2]:
In [2]: localization_method = 'boys' # You can choose 'boys', 'iao', 'pipek', e
...: tc.
...: mo_coeff_localized = lo.orth.orth_ao(mol, localization_method)
...:
...: mo_occ_localized = mf.mo_occ
/home/pro/psi4conda/lib/python3.10/site-packages/pyscf/dft/libxc.py:772: UserWarning: Since PySCF-2.3, B3LYP (and B3P86) are changed to the VWN-RPA variant, the same to the B3LYP functional in Gaussian and ORCA (issue 1480). To restore the VWN5 definition, you can put the setting "B3LYP_WITH_VWN5 = True" in pyscf_conf.py
warnings.warn('Since PySCF-2.3, B3LYP (and B3P86) are changed to the VWN-RPA variant, '
In [3]: ccsd_localized = cc.CCSD(mf)
...: ccsd_localized.mo_coeff = mo_coeff_localized
...: mo_occ_localized = mf.mo_occ
...: ccsd_localized.mo_occ = mo_occ_localized
...: ccsd_localized.kernel()
******** <class 'pyscf.cc.ccsd.CCSD'> ********
CC2 = 0
CCSD nocc = 2, nmo = 8
max_cycle = 50
direct = 0
conv_tol = 1e-07
conv_tol_normt = 1e-05
diis_space = 6
diis_start_cycle = 0
diis_start_energy_diff = 1e+09
max_memory 4000 MB (current use 150 MB)
Init t2, MP2 energy = 3.27530080154194 E_corr(MP2) -0.00569411069633339
Init E_corr(CCSD) = 0.216714827164018
cycle = 1 E_corr(CCSD) = -3.04461060576688 dE = -3.26132543 norm(t1,t2) = 36.1895
cycle = 2 E_corr(CCSD) = -19.1686545365852 dE = -16.1240439 norm(t1,t2) = 2491.19
cycle = 3 E_corr(CCSD) = 19.8047923107763 dE = 38.9734468 norm(t1,t2) = 2.68056e+06
cycle = 4 E_corr(CCSD) = 25.879418175191 dE = 6.07462586 norm(t1,t2) = 1.40707e+06
cycle = 5 E_corr(CCSD) = 27.2766254838092 dE = 1.39720731 norm(t1,t2) = 329057
cycle = 6 E_corr(CCSD) = -2.01780064492512 dE = -29.2944261 norm(t1,t2) = 364817
cycle = 7 E_corr(CCSD) = -2.97753280814283 dE = -0.959732163 norm(t1,t2) = 47972.5
cycle = 8 E_corr(CCSD) = -1817.22134315427 dE = -1814.24381 norm(t1,t2) = 109750
cycle = 9 E_corr(CCSD) = 0 dE = 1817.22134 norm(t1,t2) = 6.03432e+10
cycle = 10 E_corr(CCSD) = 0 dE = 0 norm(t1,t2) = 2.2255
cycle = 11 E_corr(CCSD) = 0 dE = 0 norm(t1,t2) = 2.2255
cycle = 12 E_corr(CCSD) = 0 dE = 0 norm(t1,t2) = 2.2255
cycle = 13 E_corr(CCSD) = 0 dE = 0 norm(t1,t2) = 2.2255
cycle = 14 E_corr(CCSD) = 0 dE = 0 norm(t1,t2) = 2.2255
cycle = 15 E_corr(CCSD) = 0.216714827164027 dE = 0.216714827 norm(t1,t2) = 2.2255
cycle = 16 E_corr(CCSD) = 0.108084643542402 dE = -0.108630184 norm(t1,t2) = 36.1895
cycle = 17 E_corr(CCSD) = -0.170149619083507 dE = -0.278234263 norm(t1,t2) = 35.9229
WARN: diis singular, eigh(h) [-3.72002754e-01 -1.08414679e-13 -2.85868422e-14 4.94132623e+00
2.12412279e+01 4.71628680e+01 2.69791993e+03]
---------------------------------------------------------------------------
LinAlgError Traceback (most recent call last)
Cell In[3], line 5
3 mo_occ_localized = mf.mo_occ
4 ccsd_localized.mo_occ = mo_occ_localized
----> 5 ccsd_localized.kernel()
File ~/psi4conda/lib/python3.10/site-packages/pyscf/cc/ccsd.py:1076, in CCSD.kernel(self, t1, t2, eris)
1075 def kernel(self, t1=None, t2=None, eris=None):
-> 1076 return self.ccsd(t1, t2, eris)
File ~/psi4conda/lib/python3.10/site-packages/pyscf/cc/ccsd.py:1091, in CCSD.ccsd(self, t1, t2, eris)
1087 if eris is None:
1088 eris = self.ao2mo(self.mo_coeff)
1090 self.converged, self.e_corr, self.t1, self.t2 = \
-> 1091 kernel(self, eris, t1, t2, max_cycle=self.max_cycle,
1092 tol=self.conv_tol, tolnormt=self.conv_tol_normt,
1093 verbose=self.verbose, callback=self.callback)
1094 self._finalize()
1095 return self.e_corr, self.t1, self.t2
File ~/psi4conda/lib/python3.10/site-packages/pyscf/cc/ccsd.py:83, in kernel(mycc, eris, t1, t2, max_cycle, tol, tolnormt, verbose, callback)
81 t1, t2 = t1new, t2new
82 t1new = t2new = None
---> 83 t1, t2 = mycc.run_diis(t1, t2, istep, normt, eccsd-eold, adiis)
84 eold, eccsd = eccsd, mycc.energy(t1, t2, eris)
85 log.info('cycle = %d E_corr(CCSD) = %.15g dE = %.9g norm(t1,t2) = %.6g',
86 istep+1, eccsd, eccsd - eold, normt)
File ~/psi4conda/lib/python3.10/site-packages/pyscf/cc/ccsd.py:1243, in CCSD.run_diis(self, t1, t2, istep, normt, de, adiis)
1239 if (adiis and
1240 istep >= self.diis_start_cycle and
1241 abs(de) < self.diis_start_energy_diff):
1242 vec = self.amplitudes_to_vector(t1, t2)
-> 1243 t1, t2 = self.vector_to_amplitudes(adiis.update(vec))
1244 logger.debug1(self, 'DIIS for step %d', istep)
1245 return t1, t2
File ~/psi4conda/lib/python3.10/site-packages/pyscf/lib/diis.py:237, in DIIS.update(self, x, xerr)
235 else:
236 self._xprev = None # release memory first
--> 237 self._xprev = xnew = self.extrapolate(nd)
239 self._store('xprev', xnew)
240 if 'xprev' not in self._buffer: # not incore
File ~/psi4conda/lib/python3.10/site-packages/pyscf/lib/diis.py:264, in DIIS.extrapolate(self, nd)
262 except numpy.linalg.linalg.LinAlgError as e:
263 logger.warn(self, ' diis singular, eigh(h) %s', w)
--> 264 raise e
265 logger.debug1(self, 'diis-c %s', c)
267 xnew = None
File ~/psi4conda/lib/python3.10/site-packages/pyscf/lib/diis.py:261, in DIIS.extrapolate(self, nd)
259 else:
260 try:
--> 261 c = numpy.linalg.solve(h, g)
262 except numpy.linalg.linalg.LinAlgError as e:
263 logger.warn(self, ' diis singular, eigh(h) %s', w)
File <__array_function__ internals>:200, in solve(*args, **kwargs)
File ~/psi4conda/lib/python3.10/site-packages/numpy/linalg/linalg.py:386, in solve(a, b)
384 signature = 'DD->D' if isComplexType(t) else 'dd->d'
385 extobj = get_linalg_error_extobj(_raise_linalgerror_singular)
--> 386 r = gufunc(a, b, signature=signature, extobj=extobj)
388 return wrap(r.astype(result_t, copy=False))
File ~/psi4conda/lib/python3.10/site-packages/numpy/linalg/linalg.py:89, in _raise_linalgerror_singular(err, flag)
88 def _raise_linalgerror_singular(err, flag):
---> 89 raise LinAlgError("Singular matrix")
LinAlgError: Singular matrix