MOFSimplify: Machine Learning Models with Extracted Stability Data of Three Thousand Metal-Organic Frameworks

We report a workflow and the output of a natural language processing (NLP)-based procedure to mine the extant metal-organic framework (MOF) literature describing structurally characterized MOFs and their solvent removal and thermal stabilities. We obtain over 2,000 solvent removal stability measures from text mining and 3,000 thermal decomposition temperatures from thermogravimetric analysis data. We assess the validity of our NLP methods and the accuracy of our extracted data by comparing to a hand-labeled subset. Machine learning (ML, i.e. artificial neural network) models trained on this data using graph- and pore-geometry-based representations enable prediction of stability on new MOFs with quantified uncertainty. Our web interface, MOFSimplify, provides users access to our curated data and enables them to harness that data for predictions on new MOFs. MOFSimplify also encourages community feedback on existing data and on ML model predictions for community-based active learning for improved MOF stability models.

NMR of $^{31}$P Nuclear Spin Singlet States in Organic Diphosphates

$^{31}$P NMR and MRI are commonly used to study organophosphates that are central to cellular energy metabolism. In some molecules of interest, such as adenosine diphosphate (ADP) and nicotinamide adenine dinucleotide (NAD), pairs of coupled $^{31}$P nuclei in the diphosphate moiety should enable the creation of nuclear spin singlet states, which may be long-lived and can be selectively detected via quantum filters. Here, we show that $^{31}$P singlet states can be created on ADP and NAD, but their lifetimes are shorter than T$_{1}$ and are strongly sensitive to pH. Nevertheless, the singlet states were used with a quantum filter to successfully isolate the $^{31}$P NMR spectra of those molecules from the adenosine triphosphate (ATP) background signal.

Second-order perturbative correlation energy functional in the ensemble density-functional theory

We derive the second-order approximation (PT2) to the ensemble correlation energy functional by applying the G{o}rling-Levy perturbation theory on the ensemble density-functional theory (EDFT). Its performance is checked by calculating excitation energies with the direct ensemble correction method in 1D model systems and 3D atoms using numerically exact Kohn-Sham orbitals and potentials. Comparing with the exchange-only approximation, the inclusion of the ensemble PT2 correlation improves the excitation energies in 1D model systems in most cases, including double excitations and charge-transfer excitations. However, the excitation energies for atoms are generally worse with PT2. We find that the failure of PT2 in atoms is due to the two contributions of an orbital-dependent functional to excitation energies being inconsistent in the calculations. We also analyze the convergence of PT2 excitation energies with respect to the number of unoccupied orbitals.

Explaining classifiers to understand coarse-grained models

Bottom-up coarse-grained molecular dynamics models are parameterized using complex effective Hamiltonians. These models are typically optimized to approximate high dimensional data from atomistic simulations. In contrast, human validation of these models is often limited to low dimensional statistics that do not necessarily differentiate between the CG model and said atomistic simulations. We propose that explainable machine learning can directly convey high-dimensional error to scientists and use Shapley additive explanations do so in two coarse-grained protein models.

Federated Learning of Molecular Properties in a Heterogeneous Setting

Chemistry research has both high material and computational costs to conduct experiments. Institutions thus consider chemical data to be valuable and there have been few efforts to construct large public datasets for machine learning. Another challenge is that different intuitions are interested in different classes of molecules, creating heterogeneous data that cannot be easily joined by conventional distributed training. In this work, we introduce federated heterogeneous molecular learning to address these challenges. Federated learning allows end-users to build a global model collaboratively while preserving the training data distributed over isolated clients. Due to the lack of related research, we first simulate a federated heterogeneous benchmark called FedChem. FedChem is constructed by jointly performing scaffold splitting and Latent Dirichlet Allocation on existing datasets. Our results on FedChem show that significant learning challenges arise when working with heterogeneous molecules. We then propose a method to alleviate the problem, namely Federated Learning by Instance reweighTing (FLIT). FLIT can align the local training across heterogeneous clients by improving the performance for uncertain samples. Comprehensive experiments conducted on our new benchmark FedChem validate the advantages of this method over other federated learning schemes. FedChem should enable a new type of collaboration for improving AI in chemistry that mitigates concerns about valuable chemical data.

Circumneutral concentrated ammonium acetate solution as water-in-salt electrolyte

The exponentially growing market of electrochemical energy storage devices requires substitution of flammable, volatile, and toxic electrolytes. The use of Water in salt solutions (WiSE) regarded as green electrolyte might be of interest thanks to an association of key features such as high safety, low cost, wide electrochemical stability, and high ionic conductivity. Here, we report comprehensive chemical-physical study of circumneutral WiSE based on ammonium acetate so as to investigate application in electrochemical energy storage systems, with focus on the effect of pH, density, viscosity, conductivity, and the ESW with salt concentration ranging from 1 to 30 mol/kg . Data are reported and discussed with respect to the structure of the solutions investigated by complemental IR and molecular dynamic study. The study is addressed through the showcase of an asymmetric supercapacitor based on Argan shell-derived carbon electrodes tested at temperatures ranging from -10 to 80 {deg}C.

Quantum Krylov subspace algorithms for ground and excited state energy estimation

Quantum Krylov subspace diagonalization (QKSD) algorithms provide a low-cost alternative to the conventional quantum phase estimation algorithm for estimating the ground and excited-state energies of a quantum many-body system. While QKSD algorithms have typically relied on using the Hadamard test for estimating Krylov subspace matrix elements of the form, $langle phi_i|e^{-ihat{H}tau}|phi_j rangle$, the associated quantum circuits require an ancilla qubit with controlled multi-qubit gates that can be quite costly for near-term quantum hardware. In this work, we show that a wide class of Hamiltonians relevant to condensed matter physics and quantum chemistry contain symmetries that can be exploited to avoid the use of the Hadamard test. We propose a multi-fidelity estimation protocol that can be used to compute such quantities showing that our approach, when combined with efficient single-fidelity estimation protocols, provides a substantial reduction in circuit depth. In addition, we develop a unified theory of quantum Krylov subspace algorithms and present three new quantum-classical algorithms for the ground and excited-state energy estimation problem, where each new algorithm provides various advantages and disadvantages in terms of total number of calls to the quantum computer, gate depth, classical complexity, and stability of the generalized eigenvalue problem within the Krylov subspace.

Differentiable Physics: A Position Piece

Differentiable physics provides a new approach for modeling and understanding the physical systems by pairing the new technology of differentiable programming with classical numerical methods for physical simulation. We survey the rapidly growing literature of differentiable physics techniques and highlight methods for parameter estimation, learning representations, solving differential equations, and developing what we call scientific foundation models using data and inductive priors. We argue that differentiable physics offers a new paradigm for modeling physical phenomena by combining classical analytic solutions with numerical methodology using the bridge of differentiable programming.

A bosonic perspective on the classical mapping of fermionic quantum dynamics

We consider the application of the original Meyer-Miller (MM) Hamiltonian to mapping fermionic quantum dynamics to classical equations of motion. Non-interacting fermionic and bosonic systems share the same one-body density dynamics when evolving from the same initial many-body state. The MM classical mapping is exact for non-interacting bosons, and therefore it yields the exact time-dependent one-body density for non-interacting fermions as well. Starting from this observation, the MM mapping is compared to different mappings specific for fermionic systems, namely the spin mapping (SM) with and without including a Jordan-Wigner transformation, and the Li-Miller mapping (LMM). For non-interacting systems, the inclusion of fermionic anti-symmetry through the Jordan-Wigner transform does not lead to any improvement in the performance of the mappings and instead it worsens the classical description. For an interacting impurity model and for models of excitonic energy transfer, the MM and LMM mappings perform similarly, and in some cases the former outperforms the latter when compared to a full quantum description. The classical mappings are able to capture interference effects, both constructive and destructive, that originate from equivalent energy transfer pathways in the models.

Subsurface Carbon-Induced Local Charge of Copper for On-Surface Displacement Reaction

Transition metal carbides have sparked unprecedented enthusiasm as high-performance catalysts in recent years. Still, the catalytic properties of copper (Cu) carbide remain unexplored. By introducing subsurface carbon (C) to Cu(111), displacement reaction of proton in carboxyl acid group with single Cu atom is demonstrated at the atomic scale and room temperature. Its occurrence is attributed to the C-doping induced local charge of surface Cu atoms (up to +0.30 e/atom), which accelerates the rate of on-surface deprotonation via reduction of the corresponding energy barrier, thus enabling the instant displacement of a proton with a Cu atom when the molecules land on the surface. Such well-defined and robust Cu$^{delta +}$ surface based on the subsurface C doping offers a novel catalytic platform for on-surface synthesis.