From the October 1993 issue of R & D Innovator.
Discovering a Material That's Harder Than Diamond
by Robert H. Wentorf, Jr., Ph.D.
Dr. Wentorf is Distinguished Research Professor of Chemical Engineering at Rensselaer Polytechnic Institute. He is a member of the National Academy of Engineering, has been granted 44 U.S. patents, and has received the Ipatieff Prize from the American Chemical Society.
The scene is General Electric Research and Development Center. The year is 1957, about two years after we had learned to make diamonds from graphite at high pressures (50,0000 atm.) and temperatures (about 1,500 C) using molten transition metals as catalysts. We had this marvelous high-pressure equipment, which gave us unique opportunities to explore conditions heretofore unavailable in laboratories. Furthermore, our success with diamonds had given us a breathing space from funding problems, as well as improved equipment. So we were unusually free to explore.
My boss, Tony Nerad, was a remarkable leader. One of his maxims was that if you weren't having a good time, it was your own fault. (Only later did I begin to understand the profundity of this simple statement.)
He also felt that working scientists should be shielded as much as possible from bureaucratic concerns. Tony visited each of his more than 20 scientists once a week and spent at least a half an hour chatting about their work. He often boned up on a particular subject before these visits so he could contribute to the discussion—or could find ammunition that would shock us into seeing things in a new way. His insights were often remarkable, and he was a reliable source of encouragement in the face of difficulties. When you dealt with Tony, you were dealing with intuition in big batches.
Tony and I started talking about the analogies between carbon and boron nitride, and whether a diamond-like form of boron nitride would form at extreme pressures and temperatures. We got a pound of fairly pure boron nitride powder, but nothing I tried seemed to do anything to it.
When one is on new ground, the only way to discover the ground rules is to try many things. Of course, one is guided by basic principles, but the main idea is to make mistakes as fast as possible, and never to repeat a mistake. I find that a brief (15-30 minutes) meditation can clear my mind and give me the freshness I need to choose and hold several things in awareness simultaneously and compare them. This solitude also gives me the sensitivity to become aware of and pull in the new considerations which float around the periphery of consciousness, and connect them to the problem at hand. When I am tired, these happy things simply do not happen. I might add that Tony recognized the importance of a calm, clear mind, and did his part in helping us attain it.
Up to that point, I had been trying to change the boron nitride by the seemingly logical approach of trying materials which were known to have catalytic powers for something or other. Such materials would include many of the transition metals, but would not include magnesium, which was regarded as a simple salt-former and which had no effects to speak of in diamond synthesis.
"Logic" having failed, I became more open-minded, and so one day, as part of the "make the maximum number of mistakes" approach, I put a bit of magnesium wire in the soft boron nitride and gave it the high-pressure, high-temperature treatment. While dissecting the specimen under the microscope, I saw some dark particles adhering to the remains of the magnesium wire. These particles could scratch a polished block of sintered boron carbide, something that only diamond was known to do. Clearly, this was an interesting synthesis.
I smelled ammonia, which implied that magnesium nitride was present and reacting with moisture in the air. When I put the specimen in dilute hydrochloric acid as the first cleanup step, I smelled traces of boron hydrides, suggesting the presence of magnesium borides. So evidently the magnesium had at least reacted with boron nitride, which was encouraging.
When we used freshly made magnesium nitride instead of the magnesium wire, the hard particles were white or colorless and had a cubic crystal structure. Soon I found that these cubic boron nitride particles were relatively fireproof and could scratch diamond.
Within a few months, we developed fairly effective procedures for synthesizing cubic boron nitride, which we named "Borazon." The effective catalysts were, generally speaking, the alkali and alkaline earth nitrides which, when molten at high pressure, are powerful solvents for boron nitride. How obvious it seems in retrospect: "like dissolves like"—one nitride dissolves another! The department which synthesized commercial diamonds made some test batches, and tried it as an abrasive. Borazon wasn't cost-effective for many applications. For a few years, it was an invention without an application.
Then we found that cubic boron nitride was excellent for grinding hard steels and nickel-based alloys, for which diamond isn't cost effective. The reason lay in its relative chemical inertness towards iron, nickel and cobalt, even at the high temperatures found at the interface between the abrasive and the metal. This has led to a multi-million dollar industrial technology, using machines especially built for cubic boron nitride, to manufacture high-precision steel and other alloy parts and to sharpen high-speed tools.
My attention then turned back to diamond for a few years, as we sporadically sought methods for growing large, high quality crystals ("gems"). Eventually, in 1979, we succeeded. The essentials of this method—namely to allow diamond to dissolve in a hot portion of a liquid metal melt at high pressures and grow slowly on a cooler seed—were discovered while trying to answer certain puzzling questions about diamond growth. Then, as a by-product of those studies, we unexpectedly found effective ways to sinter diamond powder into extremely hard, tough cutting tools and wire-drawing dies. They have been an outstanding industrial success.
This success with diamonds suggested that sintered cubic boron nitride tools might be useful. As in the original boron nitride synthesis, the experience with diamond was only slightly helpful—the chemistry of boron nitride was different. I had a hunch that a thin film of boron oxide on the crystals interferes with their bonding together, and I knew that aluminum was potent enough to reduce boron oxide—but how to add just the right amount of aluminum in the proper form took more than a year to work out. This search was greatly accelerated by colleagues who devised rapid, significant tests to evaluate our experimental products. It was a happy time.
It takes time for me to grow used to new concepts and be able to work with them. It takes days for a solution to a problem to form in my mind and be recognized. Once recognized, it seems almost obvious. And then it takes days for me to realize that my first solution might not have been the best one. There seem to be no rules or algorithms for solving many scientific or technical problems—the solution instead embodies entirely new concepts or views which seem to be hidden until discovered, so that a direct frontal attack on the problem is a waste of time.
After years of trying to solve problems, it becomes habit-forming, and I make up problems of various kinds to solve for the fun of it. I suppose many artists do the same thing; I have always felt a certain mutual bond when I talk with artists, most of whom are continually trying to solve a problem for the first time.