Release log for Q-Chem 4.4



Download Q-Chem 4.4



4.4.2 Release

[New improvements]

- Iterative EOM solvers with improved stability and performance

- Improvements to the RASCI method: RASCI(2), RAS-UNO, new analysis tools

- TDDFT/TDA with functionals containing VV10

- Improvements in default DFT integration grids

- Karlsruhe def2 basis sets added to the basis library

- Analytic gradient for RI-SCF with RI-J and RI-K

- Correction for occ-RI-K virtual energy

[Bug fixes]

- Fix for finite-difference and geometry optimization with RI-SCF

- Beta density missing in unrestricted TDDFT/PCM

- Fix for inconsistency of CAM-B3LYP CIS derivative coupling results in serial-run and OpenMP

- Fix for PCM hessian MPI job

- Fix for NTOs and Detach-Attach densities when using a denser grid

- Enabled user-specified guess to work with FNO


4.4.1 Release

[New improvements]

- New density functionals: BEEF-vdW and mBEEF

- Improved shared-memory parallel scaling of CIS and TD-DFT calculations

- Improved Davidson solvers for equation-of-motion coupled-cluster (EOM-CC) methods

- GPLMR solver is temporarily disabled

- Extended user-specified guess to handle multiple states

- Evaluation of <S2> for EOM excited states

- Fragment-style input for freezing string method (FSM) jobs

[Bug fixes]

- Resolved an issue with SF-DFT analytical gradient

- Resolved an issue with saving the full Hessian matrix to disk

- Resolved an issue causing crash in QM/MM calculation when the number of MM atoms is greater than 10000

- Fixed crashes in EOM-IP/EA/SF gradient calculations

- Enabled user-specified guess to work with FNO


4.4 Release

- OCC-RI-K algorithm for the evaluation of exact exchange in energy and force calculations

- Combinatorially-optimized exchange-correlation functionals:

- wB97M-V (range-separated hybrid, meta-GGA functional with VV10 nonlocal correlation)

- B97M-V (meta-GGA functional with VV10 nonlocal correlation)

- wB97X-V (range-separated hybrid functional with VV10 nonlocal correlation)

- Implementation of new exchange-correlation functionals from the literature, including:


- N12, N12-SX, GAM, MN12-L, MN12-SX, MN15-L, dlDF

- VV10, LC-VV10

- B97-K, B97-D3(0), B97-3, tau-HCTH, tau-HCTHh

- SRC1-R1, SRC1-R2, SRC2-R1, SRC2-R2


- Hessian-free minimum point verification

- Exciton-based excited-state models:

- Ab initio Frenkel-Davydov model for coupled excitations in multi-chromophore systems.

- TDDFT for molecular interactions [TDDFT(MI)], a set of local excitation approximations for efficient TDDFT calculations in multi-chromophore systems and for single chromophores in the presence of explicit solvent molecules.

- Improvements to many-body and XSAPT methods

- MPI-parallelized many-body expansion with analytic gradient.

- Efficient atomic orbital implementation of XSAPT for both closed- and open-shell systems.

- Thermostats for ab initio molecular dynamics.

- Analytic energy gradient for the Ewald summation in QM/MM calculations.

- Zeolite QM/MM methods.

- EOM-MP2 methods for excitation, ionization and electron attachment energies.

- Evaluation of polarizabilities using CCSD and EOM-CCSD wave functions.

- Distributed-memory parallel implementation of CC and EOM-CC methods and performance improvements in disk-based algorithms.

- Improvements to the maximum overlap method (MOM) for SCF calculations.

- Non-equilibrium PCM method to describe solvent effects in ADC excited-state calculations.

- Spin-flip ADC method.



4.3.2 Release

[New Features and Improvements]

- More effective parallel implementation of VV10 XC calculations using OpenMP

- Implemented QCISD(T) method in CCMAN2 module

- Improved stability of GVB and PP calculations

[Bug fixes]

- Resolved an integer overflow issue in file IO causing Q-Chem to crash when reading and writing very large files (file size exceeding 2GB)

- Fixed an error in analytical frequency calculations involving GGA functionals with Gab crossing term, e.g., P86, PW91 and PBE functionals

- Resolved issues with basis set superposition error calculations involving Grimme's D3 correction

- Fixed a bug which leads to erroreous MOS-MP2 energy

- Resolved an issue causing Q-Chem to crash upon startup on certain Linux distributions

- Fixed a memory leak in the SCF procedure leading to increased memory consumption in large optimization and finite difference calculations

- Fixes in the wavefunction analysis module



4.3.1 Release

[New Features and Improvements]

- Streamlined implementation of density functional derivative computations

- New implementation of Dyson orbital computation with real and complex CC/EOM methods

- Properties calculations and density plotting with complex CC/EOM methods

- Parallelization of computations using omegaB97X-D3, omegaM05-D, omegaM06-D3, AK13, LFAs, TAO-DFT

- Enabled gradient computation with direct TDDFT/RPA bypassing TDDFT/TDA

- Several improvements in TAO-DFT

[Bug fixes]

- Resolved issues with reported CC/EOM dipole moment for calculations with SYM_IGNORE and with ECPs

- Resolved errors in computing CC/EOM analytic gradients with frozen core approximation

- Resolved issues that disabled the use of NBO analysis with Q-Chem

- Corrected omegaB97X-D3 second derivatives computation

- Resolved issues with the Nose-Hoover thermostat in AIMD

- Resolved issues with restarting SCF calculations that use fractional occupation numbers



4.3 Release

[New Features]

- Analytic derivative couplings (i.e. nonadiabatic couplings) between electronic states computed at the CIS, spin-flip CIS, TDDFT, and spin-flip TDDFT levels

- A third-generation +D3 dispersion potential for XSAPT

- Non-equilibrium PCM for computing vertical excitation energies and ionization potentials in solution

- Spin-orbit couplings between electronic states for CC and EOM-CC wavefunctions

- The PARI-K method for evaluation of exact exchange yields dramatic speedups for TZ and greater basis set hybrid DFT calculations

- Transition moments and cross sections for two-photon absorption using EOM-CC wave functions

- New excited-state analysis for ADC and CC/EOM-CC methods

- New Dyson orbital code for EOM-IP-CCSD and EOM-EA-CCSD

- Thermally-Assisted-Occupation Density Functional Theory (TAO-DFT)

- MP2[V], a dual basis method that approximates the MP2 energy

- LFAs asymptotic correction scheme for semilocal exchange-correlation functionals

- wM05-D, wM06-D3, and wB97X-D3 long-range corrected hybrid functionals with dispersion corrections

- PBE0-2 and PBE0-DH parameter-free double-hybrid functionals

- Derivative discontinuity restoration scheme for energy gap correction


- Shared-memory parallelization of TDDFT energy and gradient calculations

Dec 03 2014 (4.2.2 Release)

[New features]

- Extended complex-absorbing-potential (CAP) calculation to use unrestricted wavefunctions (CAP-UHF and CAP-UCCSD)

- Enabled CAP calculations based on EOMIP-CCSD and EOMSF-CCSD methods


- Enabled direct RPA and TDDFT calculations (RPA=2) without going through CIS or TDDFT/TDA calculations first

- Added OpenMP implementation for 1-E integral computing and improved hybrid MPI/OpenMP performance

- Added license scheme checking IP address

- Updated libefp module to allow Cartesian coordinates (xyz format) specification for fragments consisting of less than 3 atoms

- Multiple improvements in complex-scaled (CS) and complex-absorbing-potential (CAP) coupled-cluster and equation-of-motion methods

- Enabled printing orbitals in MOLDEN format at each step of AIMD

- Fixed an input issue of sos_factor for O2 MP2

- Fixed a printing format issue for Raman intensity, allowing two more digits after the decimal point

- Allow ranges to be used in "$occupied" and "$qm_atom" input sections

- Enabled the use of LANL2DZ-SV basis set in conjunction with LANL2DZ effective core potentials.

[Bug fixes]

- Fixed a random error in Davidson diagonalization in ADC calculation

- Updated libefp module to handle newline on mac and windows correctly and to fix a rare bug in OpenMP parallel code

- Fixed incorrect polarization correction to EOM-XX/EFP excited states (only in EFPMAN2+CCMAN2 combination)

- Fixed a bug causing QM/EFP jobs with pure hgfd functions to crash

- Fixed the crash in ChemSol calculation by doubling the maximum number of Langevian dipoles.

- Fixed an input parser error with calculations of single-atom fragments

- Fixed an issue causing RI-MP2 optimization jobs to crash when checkpoint file output is enabled

- Fixed a bug in RPath code in MP2 calculations

- Fix a bug in RIMP2 calculations using large auxiliary basis



4.2.1 Release

[New features]

- added Linearized CCD (LCCD) method

- added ADC(3) method

- added feature to request a PCM calculation using a single, spherical cavity


- added COSMO radii for many transition metal elements

- PCM solvation is now supported for systems involving effective core potentials

- Density embedding now calculates 1-in-2 + 2-in-1 in a single job

- improvements in the wB97 family density functionals

- Interface with NBO5 improved for orbital and density visualization

- cleaned-up and unified input keywords for implicit solvation models (also see Q-Chem manual)

- bug fix for many-body BSSE correction for MP2

- bug fix for energy gradient calculations with f and higher angular momentum basis functions on systems involving effective core potentials

- bug fix for Raman intensity calculations with MP2

- fixed a bug that disables CCSD(dT) and CCSD(fT) energy computations



4.2.0 Release

- enhanced NMR capability of indirect spin-spin coupling

- QM/MM interface for modeling Zeolites

- SM12 solvation model

- NBO6 Integration

- performance improvement in the CCSD(T)


- improvement for computing higher EOM-CC eigenvalues

- Fermi broadening for the SCF occupations to treat metallic systems with DFT.

- Density embedding techniques to run high-level-embedded-in--low-level calculations

- unified set of REM keywords for PCMs, SMx, and Chipman's SS(V)PE

- added fragment-based initial guess

- QM/MM Ewald single point energies and gradients

- Many-Body methods, MBE, EE-MBE, EE-MBE (SCF charges) for binding energies, and VMFC for BSSE corrections

- Spin-opposite scaling variant of ADC(2) and ADC(2)-x, including optimized set of parameters.

- State-to-state properties for ADC excited states

- ROKS excited states

- spin-flip TDDFT MECP optimizer

- RI-ADC code upto ADC(2)-x

- Basin hopping and numerical PH

- Added regularized OOMP2

- bug fix to limit maximum amount of scratch memory used in openmp integral computing

- bug fix in state following, gsm code

- bug fix for mecp optimization

- bug fix in XSAPT code

- bug fixes in SCF driver routine to fix some crashing jobs



4.1.2 Release

- added OpenMP implementation for DFT with pure and range-separated functionals

- added Coulomb screening and other performance enchancements for OpenMP code

- bug fix for OpenMP HF/DFT gradient calculation with g basis function

- bug fix for XSAPT and MBE

- bug fix for M08s/M11s/SOGGA11s TDDFT when beta electrons=0

- added XSAPT+D

- enabled TDDFT calculations with SOGGA11 and SOGGA11X

- added wB97X-V functional

- enabled gaussian blur with ECP

- enabled PCM using ECP

- bug fix for CISD calculation crash

- bug fix for random sporatic error in ccman2 OpenMP gradient

- enabled ccman2 EOM CCSD(T)

- bugfix for ECP RIMP2 energy and ECP with general aux basis

- bug fix for default purecart flag on ghost atoms

- added minimum energy crossing point (MECP) calculation with spin-flip TDDFT

- bug fix for local XYGJ-OS functional

- bug fix for RI-CIS(D) crash without specifying AO2MO_DISK value

- full implementation of QACF with PBE potential correlation functional

- bug fix for DFT-D3 geometry optimization

- added CIS-to-CIS derivative coupling

- fixed the crash for HF frequency calculation in certain cases

- bug fix for incorrect active space setup in Coupled Cluster calculation in cases that involve high spin open shell system with frozen core and ECP

- bug fix for QCISD(T) calculation under low-memory condition


4.1.1 Release

- performance improvement and parallelization of Effective Fragment Potential calculation

- addded new EFP features, E.g., polarization damping, overlap-based damping

- fixed a small memory leak in geometry optimization

- fixed a memory leak in DFT calculation

- store unrelaxed and relaxed one particle density matrix for coupled-cluster and equation-of-motion coupled-cluster wavefunction analysis

- fixed SVP calculation crash

- fixed a bug in OpenMP HF gradient

- fixed a crash in potential energy surface scan

- added Truhlar's SM12 model

- enabled cutting of non C-C bonds in internal QM/MM calculations

- added excitation energy decomposition with wB97X and wB97X-D functionals

- fixed a bug of integral screening in OpenMP HF gradient calculation

- QACF from Yousung Jung's group implemented



4.1 Release

- Performance enhancement for OpenMP gradient calculations

- Fixed basis2 bug in OpenMP calculations

- Fix bug for attenuated MP2

- Added T-Chem and T-Chem samples

- Updated with parallel RIMP2 fixes

- Updated CAM-B3LYP functional

- Updated FSM

- Updated RASCI method

- Fixed bug in fEFP

- Added new OpenMP implementation for HF and DFT(GGA) energy and gradient

- Updated bond order analysis code

- Added RAS-SF code

- Added omegaB97X-V functional

- Fixed hydrogen atom issue for all Minnesota functionals after 2006.omegaB97X-V

- Added RI/Cholesky in ccman2

- Add VSA analysis

- Updated for MOM singlet states

- Updated partial hessian code

- Added PES scan code

- Fixed a bug in Q-Chem/CHARMM excited state force calculation

- Fixed a problem with SM8

- Added RASCI method

- Added analytical gradients for XYGJ-OS

- Added ECP for higher angular momentum

- Updated for truncated (subspace) TDDFT

- Fixed bugs for unrestricted TDDFT extending to full TDDFT

- Fixed a bug in OpenMP MO transformation

- Updated geometry optimization output format

- Added OpenMP MO transformation code

-Updated QCT code

-Fixed a segfault bug in FSM

- Fixed a bug in ECP frequency calculation

- Updated convergence threshold for coupled cluster jobs

- Fixed crash in finite difference gradient calculation for EOM-CC

- Performance improvement of EOM DM

- Fixed crash for cations isoelectronic with noble gases with FC

- Fixed numerical ECP gradients

- Added CPCM TDDFT hessian

- Updated XDM code with latest data for free atomic polarizabilities and volumes

- Added AOCF code

- Performance improvement of tensor library

- Added unrestricted fragmo guess

- Fixed a bug in computing SCFMI DIIS error vector RMS

- Fixed a bug in BSSE calculation with pure/cartisian functions.

- Added excited states properties for ADC calculation

- Added Hessian-free TS search

- Added ab initio dispersion code in EFP

- Fixed bugs for XPol and SAPT

- Updated EOMIP EA property calculation

- Added M08/M11 TDDFT

- Fixed an ECP bug

- Added the new LSD Ec functional PK09

- Added RI-EOM

- Fixed a bug in CCSD(T) energy calculation

- Fixed a bug related to xpol gradients utilizing Lowdin charges

- Fix the D3 part of makenn in integral computing

- Allow QM/MM JANUS with coupled cluster

- Added LRC-wPBEPBE, B97-D functionals

- Added input REM for BR89B94h and B3TLAP

- Fixed a parallel bug in COSMO calculation

- Updated beck_rsc code

- Updated CAM-B3LYP and mPW1PW implementations

- Added CAMB3LYP, mPW1PW, B5050LYP

- Fixed a bug in ssgman

- Fixed a bug in ADC calculation