Thursday, June 5, 2008Volume 8, Issue 2
Gerald G. Johnson, Jr. - Pioneer of Scientific Computing
In at the beginning
The age of modern computing began as World War II was drawing to a close. Eniac, the first electronic computer, had been developed for the Army to replace human calculators in ballistics (with contributions in symbolic logic by Penn State’s Haskell Curry). With more than 17,000 vacuum tubes, Eniac took up 3,000 cubic feet of space and weighed 30 tons. A thousand times faster than any machine of its day, it could perform 5,000 addition calculations in one second, but only 38 divisions. Eniac was soon replaced by Univac, smaller and faster, but still massive, and in the mid-1950s IBM’s first general purpose computers appeared.
When Johnson arrived at Penn State in 1962, to enter the recently created graduate program in Solid State Science (now Materials), scientific computing was still in its infancy. Although Penn State had been one of the first universities to build and operate its own computer (PENNSTAC, the Penn State Automatic Computer, built in 1956), the computer resources for researchers were limited. A large computer for the entire campus, an IBM 7074 with 10K of memory, was housed in Boucke Building with a similar set-up in the basement of Old Main that was off limits to students. The computer Johnson used for his research was an IBM 1620 Cadet, considered a small scientific computer, with a paper tape and keypunch. The Cadet, housed in the College of Earth and Mineral Sciences, was a mainframe computer about the size of a washer/dryer. It ran on the Fortran computer language and used addition tables in memory rather than calculation programs, hence the play on its Cadet nickname - Can’t Add, Doesn’t Even Try. Difficulties with programming and frequent troubleshooting made for long evenings in the computer lab.
As a graduate research assistant for Prof. Vladimer Vand, Johnson did X-ray diffraction of bedrock from meteor impact craters, using a “very ancient” XRD-5. This was the only time he ever did real experimental work at Penn State, he says. “I’m a physicist; both of my thumbs are on the same side.” Instead he learned to analyze the x-ray patterns created by others.
The X-ray files
When X-rays scatter off the atoms of a material, they form a unique pattern based on the material’s crystal structure. Today, the patterns are digitally stored in a searchable database called the Powder Diffraction File, maintained by the International Center for Diffraction Data (ICDD). Until the advent of computers, the process of archiving the large number of patterns was far more laborious. Patterns were collected from around the world, sent in by scientists to the ASTM (American Society of Testing and Materials), and published in an annual book.
At the time Johnson began working with the Powder Diffraction File in 1962, scientists worldwide were collecting and publishing only 500 – 700 X-ray patterns a year. These patterns were printed on strips of paper which were placed under glass boards and photographed. For the next year’s publication, the strips could be re-ordered with new patterns, placed under another glass board and re-photographed. The preparation for the annual publications took place at Penn State, which was well known for X-ray diffraction. The first editor of the Powder Diffraction File was Wheeler P. Davey, a Penn State faculty member for whom Davey Lab was named.
“I said, of course I can do that on a computer,” Johnson recalls of his introduction to the laborious process of typing and photographing the files. He began, in 1963, to write photo composition codes in the Fortran programming language, something that had not yet been tried.
“Today, we can calculate and analyze a new material’s pattern in six seconds. There are over 40,000 new patterns a year,” he says. The ICDD Powder Diffraction File continues to grow, with over a half million entries of searchable patterns. The PDF is used to identify substances, predict properties of new materials, and even, by forensic scientists, to solve crimes.
In the late sixties, Johnson joined the fledgling Computer Science Department in a half-time position, again without ever having taken a single computer class. In the field of X-ray diffraction, he continued his major research in conjunction with Deane Smith, a well-known Penn State geologist, studying clays and zeolites. Their work was of interest to industry, and industry turned out to be Johnson’s major support throughout his research career. “Our work was in modeling materials for catalysts. I still consult for many oil companies doing the same analytical work I’ve been doing for thirty years. They just ask more difficult questions now,” he remarks.
Distinguished Fellow of the ICDD
In 2008, Dr. Johnson was awarded the ICDD’s Distinguished Fellowship Award, joining Deane K. Smith and Johnson’s friend and mentor J. Donald Hanawalt, a retired Dow Chemical scientist, among a select group of ICDD fellows who have provided long and meritorious service to the international organization. Among his contributions to the ICDD and to the many hundreds of scientists worldwide who use the Powder Diffraction File, the following are only a few examples: Jerry Johnson introduced computer typesetting, computer internal and external checking of data, and new computer based products for the customers of the PDF; he developed the ICDD’s Search/Match program; he produced the first Max-D index; he authored ~200 papers and delivered ~400 invited talks; and he consulted for over four dozen international corporations on XRD, SEM, and computers over a 45-year career. Johnsons served as ICDD Treasurer, 1988-92; Chairman, 1992-96; and consulting editor of the Powder Diffraction File after retiring from Penn State.
In the acceptance speech for his Distinguished Fellowship Award, Johnson paid touching tribute to his mentors, Don Hanawalt and Bill Fink, founders of the ICDD, who encouraged him to dedicate his career to powder diffraction and scientific computing. “To have the same title as these two leaders is a wonderful accomplishment. To be honored by an organization that believes so strongly in something that you believe in is a great honor.”
The Materials Research Institute would like to join the ICDD in thanking Jerry Johnson for his numerous contributions to the materials community.
Contact Jerry Johnson: Validate to view address - Send Email via form
