MXene Material Property Prediction via Transfer Learning with Graph Neural Networks

MXenes are an important class of 2D solid materials with expected novel properties and myriad applications, including efficient energy conversion in batteries and solar cells, environmental and water treatment, supercapacitors, and catalysts. MXenes have the broadest known theoretical chemical space among 2D materials. However, MXenes are challenging to synthesize in the laboratory, and numerical models for simulating MXene properties, such as density functional theory (DFT), remain computationally expensive. We apply transfer learning to train Atomistic Line Graph Neural Network Transfer Learning (ALIGNN and ALIGNNTL)-based models to rapidly predict MXene properties, account for crystal graph structure, and leverage additional, non-MXenes data to assist with model initialization. We first train models by incorporating training data from the Computational 2D Materials Database, followed by training these models with MXenes data, to leverage these larger datasets improve model performance. Our best models obtain the following test set mean absolute errors and percent errors, respectively, for the following, individually-predicted MXene properties: Work Function <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$0.32 \pm 0.11(\text{eV})$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$7.5 \pm 2.4 \%$</tex> error, Magnetic (T/F) of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0. 0 6 7} \pm \mathbf{0. 0 2 0}$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{6. 7} \pm \mathbf{2. 0 \%}$</tex> error, Bulk Modulus <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 5. 5} \pm \mathbf{4. 2} (\mathrm{N} / \mathrm{m})$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$7.3 \pm 1.9 \%$</tex> error, Shear Modulus <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$18.5 \pm 4.0(~\mathrm{N} / \mathrm{m})$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$14.6 \pm 3.0 \%$</tex> error, and Young's Modulus <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$38.0 \pm 10.7(~\mathrm{N} / \mathrm{m})$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$12.2 \pm 3.3 \%$</tex> error. We predict Work Function, Magnetic (T/F), Bulk Modulus, Shear Modulus, and Young's modulus for each of the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{2 3, 8 5 7}$</tex> MXenes in the aNANt database and for each of the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{2, 7 8 4}$</tex> MXenes presented in Rems et al., a well known paper that claims to lay “the foundation for the materials genomics of MXenes, i.e., MXene genomics.” We use the crystal structure files calculated with DFT from each source as input to the best performing models for each property. Neither of these sources predicted these target properties for their datasets. To our knowledge, no one has predicted the specified target properties for this dataset using the crystal structure information from this dataset as input features such as we predict here. Links to our predicted values for these target properties for all MXenes in these two datasets can be found in the supplemental information.