Last December our group held our first ForteCon, a three-day small conference/workshop on coding in Forte. You can see a summary of the topics covered in ForteCon2018 on our public GitHub resources folder.
A 2007 review of coupled cluster theory by Bartlett and Musial (Bartlett, R. J.; Musiał, M. Coupled-Cluster Theory in Quantum Chemistry. Rev. Mod. Phys. 2007, 79, 291–352) opens with an intriguing quotation of a 1989 article by physics Nobel prize winner Kenneth Wilson:
Ab initio quantum chemistry is an emerging computational area that is fifty years ahead of lattice gauge theory…and a rich source of new ideas and new approaches to the computation of many fermion systems.
I recently read this paper (Wilson, K. G. Ab Initio Quantum Chemistry: a Source of Ideas for Lattice Gauge Theorists. Nuclear Physics B – Proc. Suppl. 1990, 17, 82–92) and found it extremely interesting. The article begins by saying that Wilsons had switched his field of research from lattice gauge theory to computational quantum chemistry. One of the reasons for this change, Wilson says, is ancestral, adding an interesting explanatory note in which he acknowledges discussions with his father, E. B. Wilson. The second reason he reports is the lack of sources of inspiration in his field.
In his paper Wilson discusses the benefits of using Gaussian basis functions compared to numerical grids, the emphasis of quantum chemistry on analytic approaches rather than stochastic methods, and “the status of my efforts to ease the programming burden using the C++ programming language”.
In just a few pages, Wilson summarizes with extreme clarity the state-of-the-art in quantum chemistry (in 1989). He provides a very clear description of the major challenges in this field, and succinctly analyzes purpose and limitations of some of the most important methods used in quantum chemistry.
To my surprise, Wilson was at the time also concerned with the software productivity of scientists, an issue that is still very relevant today. I was quite excited to see that Wilson had already recognized the potential of C++ to simply the organization of large computer codes. In Wilsons words:
The ability to attach variable names and subroutines and functions to a common block is, I find, a powerful organizing
tool for large programs.
I think this article is a little gem and recommend it especially to those interested in the history of physics and quantum chemistry. I hope you will enjoy reading it!
A photo of the group in front of the Center for Computational Quantum Chemistry at the University Georgia. We were there to attend the 2016 Psi4 developers workshop. Kevin, Sam, Wallace, and Francesco gave talks on parallel CASSCF, new data structures for sparse CI, valence virtual orbitals, and the Wick&d project.
I recently watched an interesting interview of Freeman Dyson in which he recalls a meeting with Enrico Fermi. During the meeting they discussed calculations done by Dyson, but Fermi was skeptical of the model used to produce the data. Hence, Fermi quoted von Neumann: “With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.” Today I found this article, which actually does fit the shape of an elephant with four complex parameters.
Our new espresso machine (SAECO Via Venezia) and coffee grinder (Cuisinart Supreme Grind™ Automatic Burr Mill) have arrived! We’ll be happy to serve you an espresso or a cappuccino when you visit us.
Here is a photo of Jeff receiving the prize for his poster presented at SETCA 2015, University of Central Florida, Orlando.
- Tensor product methods and entanglement optimization for ab initio quantum chemistry. This review of tensor product approximations in quantum chemistry by Szalay et al. was just posted on arXiv. At times a bit formal, this is a nice introduction to the topic.
- In Search of a Rational Dressing of Intermediate Effective Hamiltonians. This is a paper on intermediate Hamiltonian theory, an approach to deal with the intruder state problem in effective Hamiltonian theories. From the abstract: The intermediate effective Hamiltonians are designed to provide M exact energies and the components of the corresponding eigenvectors in the N-dimensional model space, with N > M. The effective Hamiltonian is not entirely defined by these N × M conditions, and several dressings of the Hamiltonian matrix in the model space are possible.
- Appointing silver and bronze standards for noncovalent interactions: A comparison of spin-component-scaled (SCS), explicitly correlated (F12), and specialized wavefunction approaches. A systematic investigation of the accuracy of non-covalent interaction predicted with various quantum chemical methods by Burns and co-workers. From the abstract: After examination of both accuracy and performance for 394 model chemistries, SCS(MI)-MP2/cc-pVQZ can be recommended for general use, having good accuracy at low cost and no ill-effects such as imbalance between hydrogen-bonding and dispersion-dominated systems or non-parallelity across dissociation curves. Moreover, when benchmarking accuracy is desirable but gold-standard computations are unaffordable, this work recommends silver-standard [DW-CCSD(T**)-F12/aug-cc-pVDZ] and bronze-standard [MP2C-F12/aug-cc-pVDZ] model chemistries, which support accuracies of 0.05 and 0.16 kcal/mol and efficiencies of 97.3 and 5.5 h for adenine·thymine, respectively.
I recently purchased a couple dozen books for the group and out of curiosity decided to add this book to the order. Although I don’t use integrals much in my research (we mostly deal with linear algebra and tensor algebra) I found this book to be extremely interesting. It’s so interesting that I cannot stop reading it and put it down.
Inside Interesting Integrals is a great book if you want to revisit your integration skills and learn all sorts of tricks to compute definite integrals.
In the past two weeks I read two papers from arXiv that present some very interesting new ideas. The first paper, Extended Møller-Plesset perturbation theory for dynamical and static correlations, by Takashi Tsuchimochi and Troy Van Voorhis, deals with symmetry restoration in second-order perturbation theory. The extended MP2 method presented by the authors looks promising and I found it very interesting because it can automatically adapt to single- and multireference problems. At the very end of the paper the authors even show that their technique can be used to compute excited states.
The second one,Compact wavefunctions from compressed imaginary time evolution, is from Jarrod R. McClean and Alán Aspuru-Guzik. This paper mixes compression techniques with a propagation of the Schrödinger equation in imaginary time to compute the ground state energy.