Neural Network Prediction of Interfacial Tension at Crystal/Solution Interface

Interfacial tension at the crystal/liquid interface is a crucial and important parameter in crystal growth kinetics. The objective of the present study is to develop a neural network that is simple to use for predicting this important parameter using only from the information of solubility, molecular weight, and density of the studied systems. A three-layer feed-forward neural network was constructed and tested to predict the interfacial tension at the crystal/solution interface. The concentration of solute in liquid phase, concentration of solute in solid phase, temperature, density and molecular weight of crystal were used as inputs to predict the interfacial tension at the crystal/liquid interface (σSL). The network was trained using the solubility information for 28 systems to predict the σSL value and was validated with 29 new systems. Despite the limited number of data used for training, the neural network was capable of predicting σSL successfully for the new inputs, which are kept unaware during the training process. The σSL value that is predicted by the artificial neural network during the training and testing process was compared with σSL predicted from the widely used empirical expression. For most of the systems, ANN better predicts σSL, when compared to empirical correlation.

晶体/液体界面处的界面张力在晶体生长动力学中是一项至关重要的参数。本研究的目的是开发一个神经网络，该网络易于使用，仅通过溶解度、分子量和研究系统的密度信息来预测这一重要参数。构建并测试了一个三层前馈神经网络，以预测晶体/溶液界面处的界面张力。将液相中溶质的浓度、固相中溶质的浓度、温度、晶体密度和分子量作为输入，用于预测晶体/液体界面（σSL）的界面张力。该网络使用28个系统的溶解度信息进行训练，以预测σSL值，并使用29个新的系统进行验证。尽管用于训练的数据量有限，但神经网络能够成功预测未经训练过程知晓的新输入的σSL值。在训练和测试过程中，由人工神经网络预测的σSL值与从广泛使用的经验公式预测的σSL值进行了比较。与经验相关性相比，对于大多数系统，ANN在预测σSL方面表现更佳。