Evaluation of optical glass composition by optimization methods

Optical glass comprises SiO(2) and various other oxides that create the basic glass structure network. The Huggins-Sun-Davis (HSD) model, later modified by several authors, explains the influence of glass composition on glass properties, such as refractive index and density. A new technique for calculating the composition of a given glass whose Buchdahl or Schott dispersion coefficients and density are known is described. The well-known damped-least-squares method implementing Lagrange multipliers for boundary constraints on the composition parameters is used to provide a powerful iteration scheme with a high rate of convergence. The method based on the modified HSD model has been tested on several commercial glasses and is found to converge to very realistic composition values. The method can be easily programmed and provides a good tool in graded-index profile computations and in forming new optical glasses.