Abstract
▶ Simulation the acoustic properties of some tellurite glasses using one of the artificial intelligence techniques (artificial neural network). ▶ The glass network is strengthened by enhancing the linkage of Te–O chains. The tellurite network will also come to homogenization, because of uniform distribution of Nb5+ ions among the Te–O chains, though some of the tellurium-oxide polyhedra still link each other in edge sharing. ▶ Excellent agreements between the measured values and the predicted values were obtained for over 50 different tellurite glass compositions. ▶ The model we designed gives a better agreement as compared with Makishima and Machenzie model.
The developments in the field of industry raise the need for simulating the acoustic properties of glass materials before melting raw material oxides. In this paper, we are trying to simulate the acoustic properties of some tellurite glasses using one of the artificial intelligence techniques (artificial neural network). The artificial neural network (ANN) technique is introduced in the current study to simulate and predict important parameters such as density, longitudinal and shear ultrasonic velocities and elastic moduli (longitudinal and shear moduli). The ANN results were found to be in successful good agreement with those experimentally measured parameters. Then the presented ANN model is used to predict the acoustic properties of some new tellurite glasses. For this purpose, four glass systems xNb2O5–(1−x)TeO2, 0.1PbO–xNb2O5–(0.9−x)TeO2, 0.2PbO–xNb2O5–(0.8−x)TeO2 and 0.05Bi2O3–xNb2O5–(0.95−x)TeO2 were prepared using melt quenching technique. The results of ultrasonic velocities and elastic moduli showed that the addition of Nb2O5 as a network modifier provides oxygen ions to change [TeO4] tbps into [TeO3] tps.