The recent obituary of my longtime friend and colleague Sheldon Datz (Physics Today, November 2002, page 88) contains several inaccuracies. As a result, the description of the work with molecular beams presents a confusing chronology.
The steps that led to the first successful crossed-beam experiment on a chemical reaction were as follows. In 1945, I joined the chemistry division of Clinton Laboratories (now Oak Ridge National Laboratory); by 1949, I was working toward a crossed-beam study of a chemical reaction, that of hydrogen bromide with potassium. The extent of reaction was to be determined by collecting the resulting KBr on a cooled foil and measuring the amount by neutron activation of the bromine. The surface ionization gauge with a tungsten filament could not be used to measure the resulting KBr separately, because it detected both K and KBr with high efficiency. It was to be used to measure the total number of collisions, both reactive and elastic.
When Sheldon accepted a position at the laboratory in 1951, we began designing and building an apparatus specifically for the study of such chemical reactions. We pushed ahead with adapting neutron activation, but we both hoped that some variant of the much more convenient surface ionization gauge would be found that could distinguish between K and KBr. One came by accident. Like many other researchers, we found the large and noisy background signal from tungsten to be a problem. When we heard that a filament of platinum—which, like tungsten, is a metal with a high work function—tended to show less positive-ion noise, Sheldon obtained a sample. He measured the relative efficiencies for K and for KBr, quantities necessary for analyzing the planned experiments, and found that platinum detected K with an efficiency around 60 to 65% but KBr with only about 0.1%. That was the variant we sought.
Thus, the two-filament gauge was the fortunate first step in the birth of experimental chemical dynamics, not the logical result of our two later papers on surface ionization. Nor were those papers the first quantitative studies of surface ionization detection, a subject going back to Irving Langmuir in 1913. 1
For the record, Sheldon needed no guidance for his PhD thesis: It was his project from conception to interpretation. His thesis adviser was W. T. Smith, not Richard D. Present. In addition, Sheldon was not assistant director of the lab’s chemistry division; he was associate director along with Ralph Livingston for the period mentioned in the obituary.
It was a privilege and a pleasure to be closely associated with Sheldon for so long. We were friends as well as colleagues from a few days after he joined my group on the Manhattan Project.
