qmcommits

[qmcommits]

Augustin Bussy has pushed RI Hartree-Fock Exchange Forces for CP2K.

[qmcommits]

After the MPQC3 development has slowed down to basically a stand still without a stable version ever released and several papers alluding to a MPQC4 project in development, the mpqc Github project was updated to the MPQC4 codebase yesterday. The difference between the v3 and v4 branches is substantial:

6997 files changed, 134919 insertions(+), 1834751 deletions(-)

Documentation is rather light at this point, so the exact amount of added/removed features compared to MPQC3 is not yet clear at this point, but based on the validation tests inputs, the following methods might be available:

HF

CIS

MP2, MP2-F12

CCSD, CCSD-F12,

PNO-CCSD

EOM-CCSD, EOM-EA-CCSD, EOM-IP-CCSD, PNO-EOM-CCSD

CC3

CCSD(T)

GF2F12

Many of the methods can apparently be used together with density fitting. For the pair natural orbitals CCSD (PNO-CCSD), a recent ArXiv preprint by the MPQC4 authors is available. The explicitly-correlated second order single particle Green’s function method (GF2F12) has been published recently as well.

[qmcommits]

Sergey Chulkov has implemented the Non-Equilibrium Green's Functions (NEGF) method in CP2K. The implementation is based on this paper by Papior et al.

[qmcommits]

Vladimir Rybkin has added AXK-RPA (Approximate Exchange Kernel-Random Phase Approximation) to CP2K. For further details, see the original paper by Bates and Furche.

[qmcommits]

Ashutosh Kumar has finished implementing (over a series of commits) analytic gradients for the Second-Order Approximate Coupled-Cluster (CC2) method in Psi4.

[qmcommits]

Nico Holmberg has implemented CDFT (Constrained DFT). For details of the implementation in CP2K, see his paper with Kari Laasonen.

[qmcommits]

Jae Woo Park has implemented Nonadiabatic Coupling Matrix Elements (NACME) in Bagel. For the CASSCF and CASPT2 methods, the Nonadiabatic Coupling is evaluated analytically (there is also an accompanying paper by Park and Shiozaki for the multistate CASPT2 case), for other methods it is evaluated via finite differences. Additionally, conical intersection optimizations (which require NACMEs) are now possible for CASSCF and CASPT2 methods via the gradient projection method.

[qmcommits]

Rustam Khaliullin has extended the ALMO-SCF (Absolutely Localized Molecular Orbitals SCF) code in CP2K to allow for (variational only) molecular dynamics.

[qmcommits]

In February, analytical multi-state (MS) and extended multi-state (XMS) gradients for the CASPT2 (complete active space second order perturbation theory) method were implemented and published by Bess Vlaisavljevich and Toru Shiozaki. Those were not completely debugged, however, but now apparently they are production ready and a test case has been added.

[qmcommits]

While Psi4 (and Psi3 before) had analytical CCSD(T) gradients for UHF reference wavefunctions for a long time, Ashutosh Kumar has now finally implemented analytic gradients for RHF reference wavefunctions as well.

[qmcommits]

Justin Turney has added the original Psi 1.0 release from the late 1980s to Gtihub; it makes for some interesting quantum chemistry archeology, the license is unclear however.

[qmcommits]

A pull request by Jerome Gonthier has been merged into Psi4, adding unrestricted SAPT0 (USAPT0).

[qmcommits]

Toru Shiozaki has completed implementing a two-step second-order orbital optimization algorithm for CASSCF.

[qmcommits]

A pull request by Andy Simmonett implementing analytical second derivatives for regular and  density-fitted restricted Hartree-Fock has been merged, adding Hessians for the first ab initio method in Psi4.

[qmcommits]

Bess Vlaisavljevich: “Eref updated. XMS-CASPT2 now working” This adds extended multi-state CASPT2 (XMS-CASPT2) to BAGEL, allowing for more accurate potential energy surfaces at near-degenerate energies than conventional MS-CASPT2 . See the original paper on XMCQDPT theory by Granovsky or the JCP communication on XMS-CASPT2 by Shiozaki et al. for details on this.

[qmcommits]

Matthias Krack has added the self-consistent continuum solvation (SCCS) model to CP2K. For a description of this model, see the paper by Oliviero Andreussi et al.

[qmcommits]

Justin Turney has integrated density matrix renormalization group (DMRG) as implemented in the CheMPS2  project into Psi4.