|
08/01/2011
Why the Secrecy
CHAPTER 12 - A Device Spurs Collaboration with Physicists
I'm starting this column on July 15 after watching Tom Watson's hole-in-one at the British Open. This memorable event reminds me of a day in 1980 when I attended the U.S. Open at nearby Baltusrol. After an hour or so in the stands watching the players come through on Baltusrol's signature par-3 4th hole, I left to follow the action elsewhere on the course. Just a few minutes after abandoning the hole, I hear a huge roar from the crowd there. Tom Watson had a hole-in-one and I missed it! Longtime readers will, of course, anticipate that I will shamelessly take this occasion to point out that I have also had a hole-in-one. In contrast, full disclosure requires me to note that my most humiliating golfing experience was taking a 13 on that very 4th hole! A neighbor who was a member of Baltusrol invited me to join him that day; he never invited me again!
I'm back a week later and the temperatures here in New Jersey, and over a major portion of the USA, are 100 degrees Fahrenheit or higher. Nearby Newark would record an all-time high of 108 degrees. As one who is convinced of the reality of global warming, I'm tempted to point to this extreme heat wave as convincing evidence that climate scientists are correct in their dire warnings about civilization's future if we don't take remedial action soon. However, I realize that one heat wave isn't sufficient to confirm that global warming is "real". Nevertheless, when I look at what's been going on worldwide in the way of glaciers melting, record heat waves, widespread wildfires, torrential rains, unusual windstorms and tornadoes, etc., I can't help thinking that we've already blown it and that the "tipping point" sealing our climatic fate has passed.
Appropriately, StocksandNews editor Brian Trumbore gave me an article by Al Gore titled "Climate of Denial" clipped from the July 7-21 issue of Rolling Stone magazine. I don't imagine many of my aged generation peruse this magazine and was surprised that Gore chose this magazine for such a thorough, penetrating analysis of global warming and its politicization as an issue along party lines. Frankly, I am disgusted and dismayed by the fact that the scientific issues such as climate change and evolution have become political issues in the minds of too many people. After reading Gore's article, I regret even more my vote against him in that contentious election and wonder where we would be if he had been president. (In another case of full disclosure, I am a registered Republican and have voted in 15 presidential elections, voting 9 times for Republican candidates and 6 times for Democrat candidates. I've only voted for two presidents twice, Eisenhower and Reagan.)
Well, enough about politics. Let's get back to the scientific side of my memoirs. In my last column, I had settled down at Bell Labs and had found a niche working on impurities in germanium and silicon. In that column, I failed to mention that there was a Japanese fellow, Leo Esaki, who played an important role in my career in the late 1950s. Esaki, in 1958, published his work on a new device, the Esaki diode, also known as the tunnel diode. In a normal diode, as you increase the voltage the current goes up. However, in the Esaki diode the current goes up, peaks and then falls and, as the voltage is increased further, goes up again. This unusual behavior is due to a weird quantum effect known as "tunneling" and in 1973 Leo Esaki would share the Nobel Prize in Physics for his work on this tunneling in his diode. I won't attempt to go into the electrical properties of the device except to say that it was a really hot device, arousing great interest at the time.
For me, the important point was that device consisted of p-n junctions in "heavily doped" semiconductors. In my case, this meant germanium or silicon containing relatively large amounts of impurities. In my work I had amassed a collection of crystals containing just the amounts of impurities such as arsenic that our device people needed. I became an instantaneous expert on heavily doped materials and the resultant interactions as a maker and supplier of materials for devices would become a theme for the rest of my career at Bell Labs working on LEDs and lithium batteries.
One of the biggest joys of working at Bell Labs was the opportunity to collaborate with virtually anyone sharing a mutual interest. Physicist Ralph Logan and his assistant, Harry White, were two of the "customers" for my heavily doped crystals. They, and a fellow named Emil Dickten, would make the diodes and study their electrical properties. Ralph had a dry sense of humor and we traded good natured insults about the quality of the materials I supplied. I thoroughly enjoyed the banter we exchanged in our interactions, which would continue in my later light-emitting work. Another physicist with whom I interacted was William Spitzer, who was interested in the optical (infrared) characteristics of the heavily doped materials.
Of particular interest was arsenic-doped germanium. You may recall that in an earlier column I described how work on the vapor pressures of arsenic over germanium-arsenic alloys saved me from being fired from Bell Labs! Now I had to supply more of the arsenic-doped germanium than I had in my desk drawer. I was having my assistant, Ed Porbansky, pull crystals from heavily doped melts of germanium and arsenic using our conventional crystal pulling machine. We were having trouble pulling good quality crystals thanks to arsenic piling up in the melt at the growing interface of the crystal and the melt. The result was crystal that contained occlusions, clumps of arsenic and arsenic-doped germanium that disrupted the single crystal structure.
One day, in disgust, I told Ed to forget about trying to pull any more that day and just leave the seed crystal rotating in the melt overnight, thinking we could start up the next morning with everything ready to go. Well, the next morning we came in and surprise. There was a beautiful, large single crystal that had grown overnight! What had happened was that the volatile arsenic had slowly evaporated, causing the germanium to "precipitate" out of the arsenic-germanium melt as a single crystal. The arsenic was acting in essence as a solvent and the single crystal was free of the occlusions in our other pulled crystals. This "solvent evaporation" technique worked for other heavily-doped crystals and resulted in me giving my first invited paper, at a meeting of the American Institute of Mining, Metallurgical and Petroleum Engineers in Boston in 1960. We did file a patent application on the process but, in my opinion, the examiner brought up an objection that was irrelevant and we let the application die.
Now for a little chemistry as background for what I think may be the incident of most interest in this column. As with every impurity in germanium or silicon, the maximum solubility of arsenic in germanium depends on temperature. Typically the solubility is zero at the melting point of germanium (no arsenic added), increases as arsenic is added below the melting point until, at a certain temperature, begins to decrease at lower temperatures. Plotting a curve with temperature on the y-axis and solubility on the x-axis, the curve resembles Dolly Parton's bosomy profile. This solubility behavior is known in the trade as "retrograde" solubility.
This retrograde solubility of arsenic led to a weird and wondrous encounter with the U.S. military involving my first patent. Bill Spitzer, Ralph Logan and I had filed a patent application on a method of increasing the concentration of electrically active arsenic in heavily doped germanium used in Esaki diodes. What's the problem? Suppose we've grown a crystal at a high temperature at which a lot of arsenic is soluble in the germanium. We analyze the crystal and find half a percent of arsenic in the crystal. But when we measure the electrical properties we only find that a quarter of a percent of the arsenic have given up their electrons. What has happened is that on cooling the crystal, the solubility of arsenic is lower and the arsenic atoms want to get together and precipitate out of solution. What we found was that by simply heating the germanium sample back up to near or at the growth temperature and then quenching the germanium quickly, we could "freeze" in the arsenic atoms and they didn't have time to move around and get together to precipitate out.
This was a fairly simple patent and, as was generally the custom at Bell Labs, once the application was filed we were free to talk about and/or publish the work. So, Bill Spitzer presented our work at a Physical Society meeting and we submitted a paper to the Journal of Applied Physics. The paper was accepted and all seemed in order. Then one day we were all summoned to the inner sanctum of the Bell Labs patent department and were informed that one of the branches of the U. S. military, my recollection is that it was the Air Force, had classified the work as secret and that we were not allowed to talk about or publish it! Whoa! We already talked about it and the paper was already in press! None of us could think of any possible reason the military would consider our work of such importance that it should be classified. Then our patent attorneys started discussing the possibility that we would have to impound the entire run of the Journal of Applied Physics issue containing our paper, bringing down on us the wrath of the entire worldwide physics community! Fortunately, sanity prevailed. The patent was declassified and was issued. We never did find out the reason it was classified in the first place.
In addition to working with Logan, Spitzer and others at Bell Labs in Murray Hill, there was a fellow, Charlie Burrus, in Bell Labs at Holmdel, NJ who was also interested in Esaki diodes and I supplied him with some of my heavily doped materials. Charlie also wondered if I could supply him with some other materials for his studies. In particular, he was interested in what are known as III-V compounds. These are compounds of elements from group III and group V of the periodic table. The group III elements of interest to Burrus included gallium (Ga) and indium (In) alloyed with the group V element antimony (Sb). I obliged and supplied Burrus with some GaInSb crystals containing various amounts of gallium and indium pulled by my assistants Paul Freeland and Ed Porbansky.
Little did I know that this initial dipping of my toes into the growth of III-V compounds was only the beginning of years of work on another III-V compound, gallium phosphide (GaP). Somewhere along the line, I had transferred from the semiconductor research area to the metallurgy department at Bell Labs. I took the opportunity to transfer because I felt that in my work on germanium and silicon I had become closely identified with Carl Thurmond, my mentor so to speak, and I wanted to strike out more on my own. Morry Tanenbaum, whom I've mentioned recently as a golfing buddy and the guy who made the first silicon transistor, was my boss in metallurgy area. It was Morry who suggested that perhaps I should spread my wings and get into something other than germanium and silicon.
I agreed and there followed a period of a year or so in which I dabbled in growing crystals of such things as garnets, tungstates and various rare earth compounds of possible use in lasers. It was a time of intense interest in the new field of lasers but none of the stuff I was involved in lased. Not a single bit of publishable material came of these efforts. Morry, meanwhile had begun his rapid climb into the management field which was to end up with him becoming chief Financial officer of AT&T. My new boss was Robert Laudise. Bob had made his mark with the growth of quartz crystals for use in the Bell system and it would turn out to be interesting in that I was soon, although in his department, to be working more closely than ever with my old area associated with Carl Thurmond.
Stay tuned. Next column , hopefully, will be posted on or about September 1.
Allen F. Bortrum
|
