1997 TMS Annual Meeting: Wednesday Session

APPLICATIONS OF SENSORS AND MODELING TO MATERIALS PROCESSING: Session VI

Session Chairs: L.L. Shaw, Dept. of Metallurgy & Materials Science, Univ. of Connecticut, Storrs, CT 06269-3136; R. Abbaschian, P.O. Box 116400, 132 Rhines Hall, Univ. of Florida, Gainesville, FL 32611-6400

2:00 pm

SPECTROSCOPIC METHODS FOR CONTROL OF THIN FILM GROWTH: A.G. Jackson, S.J.P. Laube, J. Jones, TMC, Inc., Materials Directorate, Wright Laboratory, WPAFB, OH 45433

Control of thin films during the growth process is dependent on several difficult-to-monitor parameters, including flux of film material, pressure, substrate temperature, composition, and thickness. One means for easing this difficulty is to take advantage of emission spectra associated with the process and the film. For pulsed laser deposition (PLD) the ion cloud generated by the laser has characteristic emission spectra that can be used to control the deposition. Raman spectroscopy affords the possibility of real-time control by sensing characteristic peaks associated with film composition and thickness. These spectroscopic methods offer a real-time alternative to process control that is very attractive because of the ability to accurately control films to achieve the engineered structures and properties sought. Examples are presented to illustrate the capabilities of these methods for this film preparation.

2:25 pm

RAMAN SPECTROSCOPY FOR DETERMINING YBCO THIN FILM PARAMETERS IN SITU: David P. Lubbers, Univ. of Cincinnati, Cincinnati, OH; John D. Busbee, A.G. Jackson, TMC, Inc., Materials Directorate, Wright Laboratory, WPAFB, OH 45433; Rand R. Biggers, David C. Liptak, SOCHE, Dayton, OH

This paper describes the application of Raman spectroscopy for characterizing superconducting YBCO thin film parameters. Attenuation of the substrate Raman spectrum as a function of material deposited is established. Also, a correlation between film quality and Raman spectrum is explored. The identification and discrimination of superconducting and nonsuperconducting phases of YBCO is presented. The critical temperature (Tc) of a film as a function of its oxygen content is also established using Raman peak ratios. These results provide significant implications toward the use of Raman spectroscopy for in situ monitoring and control of the PLD process. Film quality can be controlled via optimization of film oxygen content and reduction of improper YBCO phases. Film thickness can be controlled by monitoring the response of the substrate material. Also, the PLD process can be studied and modeled using this powerful observation tool.

2:50 pm

INTELLIGENT PROCESSING OF MATERIALS: AN APPLICATION TO POLYMER COMPOSITES: J.F. Maguire, M.A. Miller, Materials Development Dept., Southwest Research Institute, San Antonio, TX

The process control of polymeric components is a difficult problem in modern process control. These materials consist of a chemically reactive resin which impregnates a reinforcing fiber. The resin is subject to aging phenomena and is found that frequent adjustments of the process are needed in order to assure quality of the product. In order to increase quality and reduce manufacturing delays the manufacturing process was analyzed from a physico-chemical standpoint and intelligent control system was developed and implemented. It was determined that integration of a novel advanced sensor technology based on in situ Raman and Rayleigh light scattering with a state-of-the-art chemical kinetic and polymer transport model of the polymerization process would provide real (sensor) and virtual (model) information on which material process control decisions could be based. The sensor information provides chemical and physical analytic data in real-time. This data is analyzed and compared with the predictions of the model and adaptive control decisions are implemented. The control system was interfaced with customized processing hardware via a programmable logic controller (PLC) employing conventional ladder logic. The system has been tested extensively and has been able to process adaptively advanced composite components fabricated from quartz reinforced polyimides. Finally, the process logic control algorithms, which rely on the physical and chemical state of the material based on sensor information, are generic to any chemically reactive resin with minor modifications to initial parameter values.

3:15 pm BREAK

3:25 pm

ALUMINUM SENSOR FOR STEEL GALVANIZING BATHS: R.Sridhar, J.M. Toguri, Dept. of Metallurgy and Materials Science, University of Toronto, Toronto, Ontario M4S 3E4, Canada

The aluminum content of zinc galvanizing baths is an important parameter for producing good quality steel sheets for automotive applications. The present method of bath sampling and chemical analysis measures the total aluminum content of the bath which includes dissolved aluminum and aluminum in entrained dross. Also, this method does not provide quick analysis for process control purposes. It has been reported that the measurement of dissolved aluminum is essential to obtain a good zinc coating on steel. Laboratory experiments have shown that an electrochemical cell consisting of an aluminum electrode, an ionic fused salt and an electrode like Mo (inert to Zn) dipping in the zinc bath has good potential for quick dissolved aluminum measurements. Such a sensor can be ready adopted for process control. These results will be discussed.

3:50 pm

SOLID-STATE ALUMINUM SENSOR FOR USE IN MOLTEN ZINC: J.W. Fergus, S. Hui, Materials Research and Education Center, 201 Ross Hall, Auburn University, AL 36849

Aluminum is an important alloying addition to the zinc used in the hot-dip galvanization of sheet steel to control the properties and appearance of the resulting coating. Optimization of the galvanization process requires control of the alloy concentration, which can be improved through the use of chemical sensors. Aluminum sensors, based on molten electrolytes, are commercially available but are not widely used because of their high cost. One approach to reducing the cost is to use a solid electrolyte, which can simplify the sensor design and, thus, reduce the cost of fabrication. In this paper, the development of an aluminum sensor based on a solid fluoride electrolyte, specifically magnesium fluoride, will be described.

4:15 pm

APPLICATIONS OF Al SENSORS IN CONTINUOUS GALVANIZING: N. Qiang, N.-Y. Tang, G.R. Adams, Cominco Ltd., Product Technology Centre, Sheridan Science and Technology Park, Mississauga, Ontario L5K 1B4 Canada

Aluminum is the most important alloy addition to Zn baths for controlling the reaction kinetics between the steel substrate and molten Zn in continuous galvanizing operations. The control of effective bath Al content is particularly critical for galvannealing where a high bath Al content may lead to incomplete alloying during postannealing, and a low bath Al content may result in other problems, such as an excessive bottom dross accumulation and difficulties in coating thickness control. The amount of Al in a galvanized coating has a strong influence on coating properties and the overall product quality. Yet, due to the complexity of the Zn-Fe-Al ternary system, the measurement and accurate of Al sensors has made the real-time measurements of bath Al contents possible. To facilitate their application, a computer interface and a palm-size data logger have been developed. Plant applications indicated that the sensor possessed sufficient accuracy for process monitoring and control. Potential benefits of Al sensor applications include improvements in process control and product quality, reductions in product transition periods for dual product lines, and a reduction in the overall production cost.