************************************************************************** ************************************************************************** ** ** ** /$$ /$$ /$$$$$$ /$$$$$$ /$$$$$$$ /$$ ** ** | $$$ /$$$/$$__ $$/$$__ $| $$__ $$ | $$ ** ** | $$$$ /$$$| $$ \ $| $$ \__| $$ \ $$/$$$$$$ /$$$$$$| $$ /$$ ** ** | $$ $$/$$ $| $$ | $| $$ | $$$$$$$|____ $$/$$_____| $$ /$$/ ** ** | $$ $$$| $| $$ | $| $$ | $$____/ /$$$$$$| $$ | $$$$$$/ ** ** | $$\ $ | $| $$/$$ $| $$ $| $$ /$$__ $| $$ | $$_ $$ ** ** | $$ \/ | $| $$$$$$| $$$$$$| $$ | $$$$$$| $$$$$$| $$ \ $$ ** ** |__/ |__/\____ $$$\______/|__/ \_______/\_______|__/ \__/ ** ** \__/ ** ** ** ** The Merced Quantum Chemistry Package ** ** (MQCPack) ** ** Development Version ** ** Based On: ** ** Development Version 0.1 ** ** ** ** ** ** Written By: ** ** Lee M. Thompson, Xianghai Sheng, Andrew D. Mahler, Dave Mullally ** ** and Hrant P. Hratchian ** ** ** ** Version 20.05.1 Completed ** ** May 22, 2020 ** ** ** ************************************************************************** **************************************************************************
The MQC package is a library containing objects and procedures to streamline develoments of quantum chemical methodologies. MQC is built using object-orientated Fortran and has three key features.
- Algebra
MQC has a number of mathematical objects that are designed to hide away and automate manipulation of arrays. Arrays can be packed for efficient memory use and used for operations completely transparently to the user. Furthermore, there is no need to type arrays, as this can be manipulated on the fly. Arrays carry their own procedures, and use underlying lapack routines for efficiency. See the documentation for more information, but as an example:
scalar = vector%pop()
removes the value from the end of a vector and returns it to scalar,
call matrix%eigensystem(metric,eigenvalues,rightvectors)
computes the generalized eigenequation of the matrix given the metric matrix and returns the eigenvalues and the right eigenvectors (you can return the left eigenvectors as well if you wish, or just the left eigenvectors, or neither set of eigenvectors). MQC can deal with integer, real and complex arrays.
- Quantum chemistry
MQC has objects that built over algebra objects designed specifically for quantum chemistry calculations. The most developed set of objects are the EST objects. This can work out whether you have a restricted, unrestricted or general spin framework and store the spin blocks as makes sense, it can automatically determine how to multiply different objects or carry out manipulations. Procedure interfaces allow you to use these objects just like normal matrices or take advantage of their higher-level organization.
- Interfacing with QM codes
A key feature of MQC is that it makes interacting with a range of computational chemistry codes entirely seamless. By having interfaces with different code APIs, MQC is able to fill the quantum chemistry objects and pass them back out in a manner that is completely transparent to the user. Thus, regardless regardles of what your favorite software is, your program will work without any modifications.
- Required Software:
- autotools (only for developers)
- BLAS
- LAPACK library
- gcc v8 or newer
- gfortran v8 or newer, or pgfortran v19 or newer
- Optional Software:
- gauopen
- Gaussian 16 or higher
To build:
- ./mqc_install
- {Answer questions}
This project is licensed under the Apache License - see the LICENSE file for details