This PDF file contains the front matter associated with SPIE Proceedings Volume 12806, including the Title Page, Member Companies, BACUS Dedication, Table of Contents, Author Biographies, and Welcome Note.

Basically, I would like to put some general concepts of the industry in perspective. The industry has a right to be very proud of itself, but I think it is necessary, at times, to step back and see how we fit in the world picture. One fairly simple way to measure this impact is in dollars. Bob Kopp, who is vice president of our firm, has prepared some material to help visualize some of the concepts I will be presenting. Then, we'll take a look at what we are going to say this year. Hopefully, we'll be able to come back next year and say that we have improved.

This is a Panel Discussion from the Bay Area Chrome Users Society Symposium 1982.

In the early days when masks changed from emulsion to hard surface materials, only green glass was being used. Now we have also white crown and borosilicate glasses (from 3 different vendors) and in addition quartz. Because of this increased complexity, we will try with this talk to make you better aquainted with the different substrate materials and some methods of substrate identification.

The subject of chromium on glass substrates used in the production of photomasks was addressed last year in Interface '80, with an analysis of currently available materials. The data demonstrated high degrees of variance between all materials. Evidence of improvement was noted within lots and batches of some of the vendors' products. This investigation has been extended to cover the material performance as a function of use. In a typical wafer patterning process, a photomask is used for several contact prints and then returned for cleaning and defect inspection. Very often, the material performance of masks limits their use to only one or a few such cycles. They are then rejected for all types of defect failures which would result in unacceptable wafer yields. Factors which play a dominant role in selecting useable materials are durability, resistance to scratches or abrasion, and variations in reflectivity caused by mechanical and chemical stress. Experiments were performed to simulate these stress situations in order to follow the evolution of so called hard defects, reflectivity, optical density and critical dimensions. The products of seven vendors, four domestic and three foreign, were tested. Vendor selection was done with respect to the different chrome deposition processes used. RF - or DC - sputtered and vacuum evaporated chrome blanks are considered. The names of the vendors are coded to protect the innocent.

The integration of IC's is increasing at about a two fold speed from year to year. Of late, 64K RAM's are being manufactured on a mass-production scale and 256K RAM's might be made on a production scale by the end of this year. These highly integrated IC's require high quality masks and high quality blanks. I will discuss in this paper the characteristics of chromium blanks being produced topically.

The subject of photomask inspection has received increasing attention over the past few years. Mask makers have come to rely on automatic optical inspection for two reasons. First, it is a necessary means to provide quality assurance to their fabline customers. The fablines will specify a maximum defect dimension and a maximum defect density. These specifications can be guaranteed. Second, photomask inspection is a vital link in control of the mask making process itself. The feedback provides excellent information on raw material inputs (i.e., chrome, glass, and resist), on the lithography process, and on handling and cleanliness in the mask shop.

I'd like to begin this presentation by bringing you in focus with our improvements in the area of manufacturing E-beam photomask blanks.

I will emphasize in this presentation the pellicle protection problem, for I think that's a relevant subject for the audience today. This is a 2.85 micron pellicle. Perhaps it's had as much impact on yield improvement with Perkin Elmer Aligners as any other single innovation in the lithography tooling industry in the last ten years. This pellicle was made out of nitrous cellulose which has a high degree of uniformity and a low haze as we'll discuss later.

A year ago my colleague, Val Sneberg, bravely stood here and described to you a conclusive experiment he had conducted to begin to unravel the relationship between the consistency of substrate reflectivity and critical dimension control. All of those experiments were conducted by proxy, by gracious users, and were outside of Val's control. They were difficult to repeat and difficult to correlate from one user's process to another.

There have been many discussions on different aspects of photolithography and mask making over the years, but very little has been devoted to hard surface chemistry in general.

I must explain a misunderstanding over the title of this paper. Last year when I was invited to this meeting, the topics for the papers were assigned by the BACUS Committee. The subject I was given was "Functional Control of C.D. Variance through Resist Process and Etch Rates", and I thought that trying to explain C.D. Variance to this group was a pretty tough assignment. This year, when I was informed that I could choose my own topic, I was delighted. I decided I would talk about photoresist thickness measurement. But you will notice that the title of this paper, as listed in the program, has me back at C.D. control again. I wonder if the learned members of BACUS are trying to tell me something.

Full field 1:1 projection printing, the backbone of today's IC manufacturing process, was first made practical by Perkin-Elmer. Continuing the evolution of this critical technology, R the company is now in full production of another new generation of Micralign printers - the Model 500 , Perkin-Elmer's latest entry in the evolution of full field 1:1 projection lithography. We believe that the Model 500 will significantly influence the future direction of mask technology in the 1980’s, as did the introduction of the Micralign Model 100 in the 1970's. Initial customer response to equipment in the field strongly suggests that the Model 500 will play a key role in the manufacturing process for future generations of IC devices, and that major demands will be placed on all segments of the mask-making industry to provide the mask quality necessary to exploit the full potential of the Model 500 in the years to come.

The initial concern of everyone trying to make high quality masks is pinholes on CR blanks. However, there seems to be no common standard in the evaluation of pinholes. Therefore, we made a study of the following areas: A) The comparison between automatic and visual inspection methods. B) The increase of pinholes after cleaning. C) The increase of pinholes after mask processing.

I'd like to welcome all of you to the final panel of the 1982 BACUS Symposium.

Let's stop for a few moments and see where we are and what we have learned today.