Computational materials discovery has grown in utility over the past decade due to advances in computing power and crystal structure prediction algorithms(CSPA).However,the computational cost of the ab initio calculat...Computational materials discovery has grown in utility over the past decade due to advances in computing power and crystal structure prediction algorithms(CSPA).However,the computational cost of the ab initio calculations required by CSPA limits its utility to small unit cells,reducing the compositional and structural space the algorithms can explore.Past studies have bypassed unneeded ab initio calculations by utilizing machine learning to predict the stability of a material.Specifically,graph neural networks trained on large datasets of relaxed structures display high fidelity in predicting formation energy.Unfortunately,the geometries of structures produced by CSPA deviate from the relaxed state,which leads to poor predictions,hindering the model’s ability to filter unstable material.To remedy this behavior,we propose a simple,physically motivated,computationally efficient perturbation technique that augments training data,improving predictions on unrelaxed structures by 66%.Finally,we show how this error reduction can accelerate CSPA.展开更多
The use of carbon nanotubes in composite hard armor is discussed in this study.The processing techniques to make various armor composite panels consisting of Kevlar■29 woven fabric in an epoxy matrix and the subseque...The use of carbon nanotubes in composite hard armor is discussed in this study.The processing techniques to make various armor composite panels consisting of Kevlar■29 woven fabric in an epoxy matrix and the subsequent V50 test results for both 44 caliber soft-point rounds and 30 caliber FSP(fragment simulated projectile)threats are presented.A 6.5%improvement in the V50 test results was found for a combination of 1.65 wt%loading of carbon nanotubes and 1.65 wt%loading of milled fibers.The failure mechanism of carbon nanotubes during the ballistic event is discussed through scanning electron microscope images of the panels after the failure.Raman Spectroscopy was also utilized to evaluate the residual strain in the Kevlar■29 fibers post shoot.The Raman Spectroscopy shows a Raman shift of 25 cm^(−1) for the Kevlar■29 fiber utilized in the composite panel that had an enhancement in the V50 performance by using milled fiber and multi-walled carbon nanotubes.Evaluating both scenarios where an improvement was made and other panels without any improvement allows for understanding of how loading levels and synergistic effects between carbon nanotubes and milled fibers can further enhance ballistic performance.展开更多
基金This work was supported by the National Science Foundation under grants Nos.PHY-1549132the Center for Bright Beams,and the software fellowship awarded to J.B.G.by the Molecular Sciences Software Institute funded by the National Science Foundation(Grant No.ACI-1547580).
文摘Computational materials discovery has grown in utility over the past decade due to advances in computing power and crystal structure prediction algorithms(CSPA).However,the computational cost of the ab initio calculations required by CSPA limits its utility to small unit cells,reducing the compositional and structural space the algorithms can explore.Past studies have bypassed unneeded ab initio calculations by utilizing machine learning to predict the stability of a material.Specifically,graph neural networks trained on large datasets of relaxed structures display high fidelity in predicting formation energy.Unfortunately,the geometries of structures produced by CSPA deviate from the relaxed state,which leads to poor predictions,hindering the model’s ability to filter unstable material.To remedy this behavior,we propose a simple,physically motivated,computationally efficient perturbation technique that augments training data,improving predictions on unrelaxed structures by 66%.Finally,we show how this error reduction can accelerate CSPA.
基金supported by Florida Space Grant Consortium(FSGC)under grant number NASA NNX10AM01H.
文摘The use of carbon nanotubes in composite hard armor is discussed in this study.The processing techniques to make various armor composite panels consisting of Kevlar■29 woven fabric in an epoxy matrix and the subsequent V50 test results for both 44 caliber soft-point rounds and 30 caliber FSP(fragment simulated projectile)threats are presented.A 6.5%improvement in the V50 test results was found for a combination of 1.65 wt%loading of carbon nanotubes and 1.65 wt%loading of milled fibers.The failure mechanism of carbon nanotubes during the ballistic event is discussed through scanning electron microscope images of the panels after the failure.Raman Spectroscopy was also utilized to evaluate the residual strain in the Kevlar■29 fibers post shoot.The Raman Spectroscopy shows a Raman shift of 25 cm^(−1) for the Kevlar■29 fiber utilized in the composite panel that had an enhancement in the V50 performance by using milled fiber and multi-walled carbon nanotubes.Evaluating both scenarios where an improvement was made and other panels without any improvement allows for understanding of how loading levels and synergistic effects between carbon nanotubes and milled fibers can further enhance ballistic performance.
