1.3.2 New Features in Q-Chem 6.0

Changes to default behavior:

• –

  Tightened default integral threshold (THRESH) to SCF_CONVERGENCE + 4 and used same threshold for DIIS and GDM

• –

  Set default of FD_MAT_VEC_PROD to FALSE for VV10 functional (Yuezhi Mao)

• –

  Turned off automatic evaluation of electrostatic potentils on a grid (Felix Plasser)

• –

  Set finite difference as default for energy derivatives in electric field (Yuezhi Mao)

General features and improvements:

• –

  Next-generation interface of Q-Chem with external tools (generation of archive files in the HDF5 format)

• –

  Implemented the nuclear-electronic orbital CCSD (NEO-CCSD) method (Fabijan Pavosevic, Sharon Hammes-Schiffer)

• –

  Implemented NEO-TDDFT analytical gradient and Hessian (Zhen (Coraline) Tao, Patrick E. Schneider, Sharon Hammes-Schiffer)

• –

  Enabled subset selection of atoms in NMR J-coupling calculations (JOBTYPE = ISSC) via spin input section

• –

  Disabled steepest descent in geometry optimization with fixed atoms

• –

  Added delocalized natural internal coordinate optimization in new optimizer

• –

  Updated geometry in the MOLDEN file for each step in finite difference optimizations (John Herbert)

• –

  Stabilized density fitting for JK and MP2

• –

  Set new optimizer as default for unconstrained optimization (GEOM_OPT_DRIVER = 2022)

• –

  Added the minimal-augmented and heavy-augmented versions of the Karlsruhe basis sets (John Herbert)

• –

  Removed MPI support

• –

  Resolved issues with:

  • *

    incorrect Hirshfeld charges based on molecule input orders (Abdulrahman Aldossary)

  • *

    not-a-number (NAN) errors in SOC calculations

  • *

    missing nuclear repulsion energies in Fock projection (BASIS2) calculations

  • *

    removed restriction on number of atoms (MAX_ATOM) that can be included in random search and basin hopping

  • *

    ordering of localized MOs in formatted checkpoint files (Abdulrahman Aldossary)

  • *

    missing ECP for the def2-SVPD basis set

  • *

    failure to compute NMR properties with linearly dependent basis sets

  • *

    parsing input files with 100k+ lines

  • *

    character table of C3 point group

Density functional theory and self-consistent field:

• –

  Accelerated convergence of the SCF algorithm ADIIS and add a new combined algorithm option ADIIS_DIIS. (Yuezhi Mao)

• –

  Enabled gauge-independent atomic orbitals (GIAOs) in SCF calculations using GEN_SCFMAN (Brad Ganoe, Tim Neudecker, Joonho Lee, Adam Rettig, Jonathan Wong)

• –

  Disabled user setting of coefficients (via HFK_LR_COEF/HFK_SR_COEF) if using built-in range-separated functionals

• –

  Implemented frequency calculation and analytic Hessian for the VV10 functional (Jiashu Liang)

• –

  Enabled generation of formatted checkpoint files in CIS/TDDFT calculations with frozen occupied/virtual orbitals via GUI = 2 (Yuezhi Mao)

• –

  Enabled STATE_ANALYSIS for the new plot section (PLOT = 1) (Yuezhi Mao)

• –

  Performed consistency check on TDKS Fock matrices based on the SCF convergence threshold (SCF_CONVERGENCE) instead of the field amplitude (John Herbert)

• –

  Added new energy density functionals: revSCAN, regSCAN, r++SCAN, r2SCAN, r4SCAN, TASK, mTASK, regTM, rregTM, revTM

• –

  Enabled computing spin-orbit couplings (SOC) (1-electron and 2-electron mean-field) with TDDFT (both restricted and unrestricted) and spin-flip TDDFT (SF-TDDFT) (Saikiran Kotaru, Ana Krylov)

• –

  Implemented analytic gradient for density-corrected DFT (DC-DFT) for self-interaction correction (Marc Coons, Bhaskar Rana, John Herbert)

• –

  Resolved issues with:

  • *

    incorrect results of fractional electron SCF calculations using GEN_SCFMAN (Yuezhi Mao)

  • *

    hanging qints (USE_LIBQINTS = true) jobs with large number of OpenMP threads

  • *

    non-variational initial SCF guess for ADIIS (Yuezhi Mao)

  • *

    incorrect memory estimation in TDDFT/TDA calculations

  • *

    crash of TDA excited state frequency jobs

  • *

    crash of geometry optimization with fixed atoms

  • *

    frequency calculations using basis functions with g or higher angular momenta

  • *

    sign error with TDDFT spin-orbit coupling calculations (Nicole Bellonzi)

  • *

    crash of projection-based embedding calculations (Yuezhi Mao)

  • *

    incorrect result of RPA TDDFT frequency using non-Pople basis set

  • *

    insufficient memory allocation for NMR calculations with meta-GGA functionals

  • *

    erroneous results in DC-DFT calculations using hybrid functionals with larger basis sets (Marc Coons, Bhaskar Rana, John Herbert)

  • *

    crash of excited state potential energy surface scans with CIS/TDDFT (John Herbert)

Correlated methods:

• –

  Implemented EOM oscillator strengths in velocity and mixed gauges (Josefine Andersen, Sonia Coriani)

• –

  Implemented CCSD optical rotation evaluation (Josefine Andersen, Kaushik Nanda)

• –

  Implemented the fragment charge difference (FCD) scheme in RASMAN2 (Chou-Hsun (Jeff) Yang, Aaditya Manjanath, Chao-Ping (Cherri) Hsu)

• –

  Implemented complex-valued CC2, RI-CC2, and RI-CCSD (Cansu Utku, Garrette Paran, Thomas Jagau)

• –

  Implemented the complex absorption potential (CAP) method in AIMD calculations (Jerryman A. Gyamfi, Thomas Jagau)

• –

  Implemented the v2RDM-CASSCF-PDFT method using density fitted basis sets (Mohammad Mostafanejad, Run Li, A. Eugene DePrince III)

• –

  Resolved formatting error in output of SOC calculation with RAS-CI method (Abel Carreras, David Casanova)

Molecular dynamics, non-adiabatic dynamics, embedding, and solvation:

• –

  Implemented projection-based embedding with complex basis functions (Valentina Parravicini, Thomas Jagau)

• –

  Enabled user-defined permittivity grid for Poisson equation solver (PEqS) (Suranjan Paul)

• –

  Improved PCM printing (John Herbert)

• –

  Implemented CIS and TDDFT wavefunction overlaps including their spin-flip variants for (A)FSSH (Theta Chen, Junhan Chen, Zuxin Jin, Vishikh Athavale, Vale Cofer-Shabica, Joe Subotnik)

• –

  Resolved issues with QM/MM optimzation not reading previous MOs as a guess for the next cycle

Fragment and energy decomposition analysis:

• –

  Implemented pairwise fragment excitation energy decomposition analysis (EDA) in QM/EFP calculations (Lyudmila Slipchenko)

• –

  Increased the maximum angular momentum of basis functions to 5 for XSAPT calculations

• –

  Implemented SPADE- and ALMO-based partitioning schemes for electric field calculations (Yuezhi Mao)

• –

  Implemented a new MP2 EDA scheme and added a non-perturbative polarization analysis for DFT EDA (Kevin Ikeda, Hengyuan Shen)

• –

  Enabled ALMO-CIS/TDA calculations with excitation amplitudes localized on one fragment (Yuezhi Mao)

• –

  Enabled ALMO-CIS/TDA calculations with excitation from one fragment’s occupied orbitals to all virtuals in the system (Yuezhi Mao)

• –

  Enabled ALMO-CIS/TDA calculations with excitation from one fragment’s occupied orbitals to another’s virtual orbitals (Yuezhi Mao)

• –

  Enabled user-defined occupied-virtual pairs in ALMO-CIS/TDA calculations (Yuezhi Mao)

• –

  Resolved miscellaneous issues with ALMO-CIS and excited-state ALMO-EDA calculations (Yuezhi Mao)

Miscellaneous:

• –

  Printed orbital kinetic energies using SCF_PRINT = 3

• –

  Enabled EXTERNAL_CHARGES specification in an external file (Vale Cofer-Shabica, Joseph Subotnik)

• –

  Added parameter check for many-body dispersion calculations (John Herbert)

• –

  Restored finite difference for wB97M2 and the XYG series of energy functionals

• –

  Restored finite difference banner for SA-SF-RPA