OPTIMIZATION OF J(C) OF YBCO THIN-FILMS PREPARED BY PHOTO-ASSISTED MOCVD THROUGH STATISTICAL ROBUST DESIGN

Intrinsically, the thin-film preparation technique of metalorganic che mical vapor deposition involves a large number of control variables wh ich are especially critical in work with quaternary or higher high-T-c materials. Thus, effective methods are needed to optimize variable se ttings in the preparation of high-T-c films. A matrix experimental pla n and robust design methodology has been employed for this purpose. Th e first design phase was based on existing knowledge regarding growth temperature, pressure and annealing temperature for photo-assisted pre paration of YBCO thin films on LaAlO3 (100) substrates. A battery of o nly nine experiments plus three confirmation depositions were then use d to optimize the process control parameters of precursor oven tempera ture, carrier gas (Ar) flow rate, and the flow rates of two oxidizing gases, O-2 and N2O, with respect to critical current density (J(c)) of the films. J(c) (77 K, zero field) values for deposited YBCO films (w ith thickness in the 0.4-0.6 mu m range) were found to be improved fro m about 1 to 3 X 10(6) A/cm(2). J(c) for a thinner sample (0.15 mu m) reached 5.0 X 10(6) A/cm(2). It was found that under these processing conditions the most significant process control parameter affecting J( c) is the flow rate of O-2, or more precisely, the oxidizing agent now -rate ratio of O-2 and N2O. Analysis of the variance of the J(c) data indicated that there likely exist interactions among some of the contr ol variables, most obviously between the flow rates of O-2 and N2O.