Structure-Based Synthesizability Prediction of Crystals Using Partially Supervised Learning

Abstract

Predicting the synthesizability of inorganic materials is one of the major challenges in accelerated material discovery. A widely employed approximate approach is to consider the thermodynamic decomposition stability due to its simplicity of computing, but it is notorious for either producing too many candidates or missing important metastable materials. These results, however, are not unexcepted since the synthesizability is a complex phenomenon, and the thermodynamic stability is just one contributor. Here, we suggest a machine-learning model to quantify the probability of synthesis based on the partially supervised learning of materials database. We adapted the positive and unlabeled machine learning (PU learning) by implementing the graph convolutional neural network as a classifier in which the model outputs crystal-likeness scores (CLscore). The model shows 87.4% true positive (CLscore > 0.5) prediction accuracy for the test set of experimentally reported cases (9356 materials) in the Materials Project. We further validated the model by predicting the synthesizability of newly reported experimental materials in the last 5 years (2015–2019) with an 86.2% true positive rate using the model trained with the database as of the end of year 2014. Our analysis shows that our model captures the structural motif for synthesizability beyond what is possible by E hull. We find that 71 materials among the top 100 high-scoring virtual materials have indeed been previously synthesized in the literature. With the proposed data-driven metric of the crystal-likeness score, high-throughput virtual screenings and generative models can benefit significantly by effectively reducing the chemical space that needs to be explored experimentally in the future toward more rational materials design.

Shifting the scaling relations of single-atom catalysts for facile methane activation by tuning the coordination number

“… 51 Before suggesting potentially promising materials for practical applications, the synthetic possibility is important to consider. 52 Since the synthesizability is a result of many factors, including precursors, reaction conditions, thermodynamics, kinetics, etc. , here we only consider the thermodynamic stability of the SACs due to their simplicity.…”

Generative models for inverse design of inorganic solid materials

“…In a standard visualization process, the encoder encodes high-dimensional material descriptors into low-dimensional latent vectors, and then dimension reduction algorithm is applied to map these vectors to two-dimensional space for visualization. Common dimensionality reduction algorithms include principal component analysis [93] , multidimensional scaling [94] , t-distributed stochastic neighbor embedding [95] , sketch-map [96] , etc. As the input of the [81] .…”

Rational Solid-State Synthesis Routes for Inorganic Materials

“…[1][2][3][4] Formation of a new solid phase often occurs through nucleation and growth, phenomena that are driven by a complex interplay of bulk, surface and interface thermodynamics and transport, 5,6 hindering any straightforward, step-by-step reconstruction of crystals from smaller components as in organic retrosynthesis. 7 Hence, machine-learning and text-mining based approaches are being sought towards enabling predictive synthetic capability for inorganic solids, [8][9][10][11][12][13][14] accompanied by first-principles studies providing in-depth analyses of reaction mechanisms in individual systems or higher-level synthesizability trends. [15][16][17][18] Lack of a generally applicable rational synthesis planning framework is considered the missing link for realization of computer-designed functional inorganic materials.…”