Release Log for Q-Chem 5.3
Q-Chem 5.3.2 Release
December 10, 2020
New features
- Dual Hessian for anharmonic frequency calculations (Magnus Hanson-Heine)
- State-targeted energy projection with restricted, unrestricted, and restricted open-shell orbitals (Kevin Carter-Fenk)
- Real-valued and mixed real/complex absorbing potentials for HF/CC/EOM calculations (Florian Hampe)
- EOM-DEA-MP2 and EOM-DIP-MP2 methods, EOM-DEA for open-shell systems (Sahil Gulania, Anna Krylov)
- Diabatization of electronic states computed with RAS-CI using Boys, Edmiston-Ruedenberg, and dipole-quadrupole schemes (Abel Carreras)
- Spin-orbit couplings and SOC-NTOs using RAS-CI wavefunctions (Abel Carreras)
- Extended QM/MM interface for Gromacs to support surface hopping (Vale Cofer-Shabica, Joe Subotnik)
Bugfixes and Improvements
DFT and SCF improvements:
- Extended support for density functionals in GPU-accelerated DFT calculations using BrianQC: B97-1, B97-2, B97-3, B97M-V, ωB97, ωB97X-V, revPBE, revPBE0
- Improved parallel performance of force and vibrational frequency jobs with effective core potentials
- Improved parallel performance of Fock derivative contributions to analytic Hessian
- Resolved issues with:
- missing spin density plots with PLOT_SPIN_DENSITY=TRUE (Yuezhi Mao)
- memory exhaustion in long DFT optimization and dynamics jobs
- memory errors and incorrect frequencies in large jobs with ECPs
- missing output files in molecular junction calculations
- printout of partial Hessian frequency analysis
- crashing linear response property calculations via RESPONSE=TRUE
Correlated methods improvements:
- Improved integral screening for CC/EOM two-electron nuclear gradient
- Improved performance of EOM-CC computations (Anna Krylov)
- Resolved issues with:
- computation of transition density matrices for EOM-EE-CCSD, CVS-EOM-EE-CCSD, their RI and Cholesky counterparts (Sahil Gulania, Anna Krylov)
- failure to remove large temporary files after performing CC/EOM jobs on Windows
- insufficient memory error when evaluating spin-orbit coupling integrals in EOM when highly contracted basis sets are in use
- fragment excitation difference (FED) computations with RAS-CI (Aaditya Manjanath, Cherri Hsu)
Solvation and QM/MM improvements:
- Improved shared memory parallel performance in SCF jobs with effective fragment potentials
- Added capability to specify fractional nuclear charges for QM/MM (John Herbert)
- Resolved issues with:
- crashing jobs that use SMD on Windows
- geometry optimizations using ROKS and IEF-PCM (Yuezhi Mao)
Q-Chem 5.3.1 Release
August 31, 2020, updated on September 3, 2020
Changes in Default Settings
- Dispersion energy is now automatically produced in ALMO-EDA calculations with ECPs
New Features and Improvements
DFT and SCF improvements:
- Enabled RCA-DIIS algorithm in jobs that request FRAGMO SCF guess (Yuezhi Mao)
- Enabled RCA-DIIS algorithm in DFT jobs with non-local correlation (Yuezhi Mao)
- Enabled damping in conjunction with the level-shifting DIIS algorithm
- Extended availability of density functionals to include range-separated functionals when using BrianQC
- Resolved issues with:
- performance regression in incremental Fock builds
- logic error that caused SCF jobs using the DIIS-DM algorithm to crash (Yuezhi Mao)
- performance regressions in the evaluation of Fock matrix nuclear derivatives
- exchange-correlation nuclear gradients to be reported as NaN for some large molecules
- BASIS=MIXED in some FRAGMO and EDA calculations (Yuezhi Mao)
- calculations invoking the HFPC/DFPC (John Herbert)
- using AUTOSAD and FRAGMO SCF guesses for post-SCF calculations (Yuezhi Mao)
- interperting input EXCHANGE=Becke (Yuezhi Mao)
- crashing static hyperpolarizability calculations with MOPROP=103
- geometry optimization using ROKS (Yuezhi Mao)
- evaluation of transition dipole moments using delta-SCF and ROKS (Diptarka Hait)
- out of memory errors in large DFT vibrational frequency calculations
- numerical errors in gradient and hessian of Grimme's empirical dispersion corrections
CIS and TDDFT improvements:
- Enabled wavefunction analysis between SF-TDDFT states, resolved a number of issues with SF-TDDFT state analysis (Felix Plasser)
- Added sample inputs demonstrating XAS and XES TDDFT capabilities (John Herbert)
- Resolved issues with:
- erroneously reported zero spin-orbit couplings in some large TDDFT calculations
- including the empirical dispersion contribution in excited states gradients (Yuezhi Mao)
Correlated methods improvements:
- Added capability to evaluate g-tensor using CCSD (Sven Kaehler)
- Resolved issues with:
- crash in some RIXS EOM-CC calculations (Kaushik Nanda)
- out of memory error during computation of RI-CASSCF energies with large RI basis sets
- CCVB calculations for large systems (Alan Rask)
Solvation and QM/MM improvements:
- Restored capabilities to have fractional nuclear charges in QM/MM calculations (John Herbert)
- Resolved an issue causing MM optimizations with over 10,000 atoms to crash (John Herbert, Sahil Gulania)
- Added capability to use Poisson boundary conditions in systems with ECPs (John Herbert)
Energy decomposition analysis improvements:
- Enabled visualization of ALMOs in standalone SCF-MI calculations (Yuezhi Mao)
- Added plots for NOCV pair contribution to difference density in EDA2 (Yuezhi Mao)
- Resolved an issue with EDA_PRINT_COVP=AUTOMATED in ALMO-EDA jobs (Yuezhi Mao)
Q-Chem 5.3.0 Release
June 16, 2020, updated on June 22, 2020
Changes in Default Settings
- Renamed rem variable ADIABATIC_CTA to VFB_CTA
- Changed ROHF_DIAG_SPEC default from 0 to 2 for ROHF and set GEN_SCFMAN as default ROSCF engine
New Features and Methods
General improvements:
- Added support for the jun-cc-pvdz basis set (K. Carter-Fenk)
New features and improvements to the DFT capabilities:
- TD-DFT analytic force and frequencies for meta-GGA density functionals
- Level shifting in DIIS for better SCF convergence in difficult cases
- M06-SX density functional (P. Morgante, R. Peverati)
- HF-3c method (B. Rana, J. Herbert)
New features and improvements in the CC/EOM-CC package:
- Calculation of RIXS and orbital analysis of RIXS transition moments (K. Nanda, A.I. Krylov)
- New features in the CVS-EOM-CC suite (M. Vidal, S. Coriani)
- Energies and properties for EOM-DEA-CCSD (S. Gulania, M. Ivanov, A.I. Krylov)
- Transition properties and S2 for EOM-DIP-CCSD (S. Gulania, W. Skomorowski, A.I. Krylov)
- New NLO properties (hyperpolarizabilities) in EOM-CC (K. Nanda, A.I. Krylov)
- New tools for strongly correlated and magnetic systems: Extention of FNO to open-shell systems (P. Pokhilko, A.I. Krylov)
- Construction of effective Hamiltonians from EOM-CC wavefunctions (P. Pokhilko, A.I. Krylov)
- NTO analysis of spin-forbidden transitions (P. Pokhilko, A.I. Krylov)
- Search for special points of complex PES (minima, MECP, and exceptional points) within CAP-EOM-CCSD (Z. Benda, T.-C. Jagau)
- Voronoi CAP and projected CAP methods (J. Gayvert, K. Bravaya)
- New tools for computing Auger decay rates and resonance lifetimes by the Feshbach-Dyson method (W. Skomorowski, A.I. Krylov)
- Stability improvements in EOM-CC (P. Pokhilko, A.I. Krylov)
New features and improvements in MP2 methods:
- Geometry optimization with regularized orbital-optimized second-order Moller-Plesset perturbation theory (kappa-OOMP2) (Joonho Lee, M. Head-Gordon)
New capabilities for intermolecular interactions:
- Implementation of the XSAPT+MBD method (K. Carter-Fenk, J. Herbert)
QM/MM improvements:
- L-BFGS algorithm for geometry optimization (B. Rana and J. Herbert)
- Harmonic confining potentials (S. Dasgupta and J. Herbert)
New methods and capabilities:
- Nuclear electronic orbital DFT and TD-DFT methods (F. Pavosevic, Zhen Tao, S. Hammes-Schiffer)
- New module for RAS-SF methods (S. Houck, N. Mayhall)
- A family of configuration-interaction methods: non-orthogonal configuration interaction singles (NOCIS), static exchange (STEX), and one-center NOCIS (K. Oosterbaan, M. Head-Gordon)
- Integral screening and resolution-of-the-identity capabilities for complex basis functions (T.-C. Jagau)
- RI-MP2 method for complex basis functions (M. Hernandez Vera, T.-C. Jagau)
- New method (concentric localization) for truncating the virtual space in projector-based embedding theory (Yuezhi Mao)
- Square gradient minimization for excited state orbital optimization (D. Hait, M. Head-Gordon)
- Resonance Raman spectroscopy simulation (S. Dasgupta, B. Rana, J. Herbert)
- Population analysis of antibonding orbitals (A. Aldossary)
- Fragment-based diabatization schemes (Yuezhi Mao)
- Enabled ghost atoms without basis functions (B. Alam and J. Herbert)
- Electron localization function (A. Bushra, J. Herbert)
- New input options for wavefunction analysis (F. Plasser)
New features of the BrianQC GPU module:
- Extended support for GPU accelerated DFT exchange-correlation with support for LDA, GGA, and meta-GGA functionals
- Partially GPU accelerated DFT frequency calculations
Other Fixes and Improvements
- Added NTO/NBO plots for multiple CIS/TDDFT states with the new format of \$plots input section
- Added capability to produce AIMPAC WFN files in GEN_SCFMAN (J. Herbert)
- Improved readability of SAPT output (J. Herbert)
- Added spin density plots capability using ALMO-EDA (Yuezhi Mao)
- Added visualization of fragment orbitals using ALMO-EDA (Yuezhi Mao)
- Resolved a numerical problem with zero initial gradient in the GDM algorithm for SCF (Yuezhi Mao)
- Resolved an issue with SCF convergence during geometry optimization with SM8 (Yuezhi Mao)
- Resolved an issue with frozen active-space ROHF calculation giving inconsistent results (Yuezhi Mao)
- Resolved an issue causing not-a-number errors when evaluating empirical D3 corrections
- Resolved multiple issues with the Poisson-Boltzmann solver (C. Stein)
- Resolved an issue with the normalization of vectors in CVS-EOMIP-CCSD (P. Pokhilko and A.I. Krylov)
- Resolved an issue preventing the computation of CCSD to EOM-EE transition properties (P. Pokhilko)
- Resolved an issue with unrestricted EOM-DIP transition properties (P. Pokhilko)
- Resolved EOM convergence issues due to violation of permutational symmetry (P. Pokhilko)
- Resolved an issue with EOM two-photon absorption due to an incorrect prefactor in the response equation (K. Nanda)
- Resolved an issue with evaluating Hirshfeld charges for open-shell systems with basis sets containing high angular momentum functions (Yuezhi Mao)
- Resolved an issue that caused SOS-ADC(2) computations to crash
- Resolved issues with using user-specified MM force field parameter file
- Resolved an issue preventing BSSE computations from properly executing (Yuezhi Mao)
- Resolved an issue causing incorrect SM12/CM5 solvation energies (Yuezhi Mao; note: only development versions of Q-Chem were affected by this issue)
- Resolved an issue with ghost atoms in PCM calculations (Yuezhi Mao)
- Resolved an issue with wavefunction analysis in QM/MM (J. Herbert)
- Resolved an issue with missing MM energies in QM/MM calculations (Yuezhi Mao)
- Resolved multiple issues in ALMO-EDA (Yuezhi Mao)
- Resolved an issue causing ALMO-EDA computations to terminate abnormally on Windows
- Restored the behavior of EDA_ALIGN_FRGM_SPIN as documented (Yuezhi Mao)
Other Release Logs:
- Release Log for Q-Chem 6.0
- Release Log for Q-Chem 5.4
- Release Log for Q-Chem 5.2
- Release Log for Q-Chem 5.1
- Release Log for Q-Chem 5.0