Editing Calutron (section)

== Research ==
While the process had been demonstrated to work, considerable effort was still required before a prototype could be tested in the field. Lawrence assembled a team of physicists to tackle the problems, including [[David Bohm]],{{sfn|Peat|1997|pp=64–65}} [[Edward Condon]], [[Donald Cooksey]],{{sfn|Smyth|1945|p=190}} A. Theodore Forrester,<ref>{{cite news |newspaper=[[Los Angeles Times]] |title=A. Theodore Forrester; UCLA Professor, Acclaimed Inventor |date=31 March 1987 |url=https://www.latimes.com/archives/la-xpm-1987-03-31-me-1315-story.html |access-date=1 September 2015 }}</ref> [[Irving Langmuir]], [[Kenneth Ross MacKenzie]], [[Frank Oppenheimer]], [[J. Robert Oppenheimer]], William E. Parkins, Bernard Peters and [[Joseph Slepian]].{{sfn|Smyth|1945|p=190}} In November 1943 they were joined by a [[British contribution to the Manhattan Project|British Mission]] headed by Oliphant that included fellow Australian physicists [[Harrie Massey]] and [[Eric Burhop]], and British physicists such as [[Joan Curran]] and [[Thomas Allibone]].{{sfn|Gowing|1964|pp=256–260}}{{sfn|Jones|1985|p=124}}

[[File:XAX development unit.JPG|thumb|left|alt=Very large building with a strange contraption. Stairs lead up to parts of it.|The XAX development unit at Oak Ridge was used for research, development and training.]]
Lawrence had a large cyclotron under construction at Berkeley, one with a {{convert|184|in|cm|adj=on}} magnet.{{sfn|Smyth|1945|p=192}} This was converted into a calutron that was switched on for the first time on 26 May 1942.{{sfn|Manhattan District|1947b|p=1.8}} Like the 37-inch version, it looked like a giant C when viewed from above. The operator sat in the open end, whence the temperature could be regulated, the position of the electrodes adjusted, and even components replaced through an airlock while it was running. The new, more powerful calutron was not used to produce enriched uranium, but for experiments with multiple ion sources. This meant having more collectors, but it multiplied the throughput.{{sfn|Parkins|2005|p=48}}{{sfn|Hewlett|Anderson|1962|pp=92–93}}

The problem was that the beams interfered with each other, producing a series of oscillations called hash. An arrangement was devised that minimized the interference, resulting in reasonably good beams being produced, in September 1942. Robert Oppenheimer and [[Stan Frankel]] invented the [[shim (magnetism)|magnetic shim]], a device used to adjust the homogeneity of a magnetic field.<ref name="Magnetic shims">{{cite patent|country=US|number=2719924|pubdate=1955-10-04|title=Magnetic shims|assign1=[[United States Atomic Energy Commission]]|inventor1-last=J. Robert|inventor1-first=Oppenheimer|inventorlink1=J._Robert_Oppenheimer|inventor2-last=Frankel|inventor2-first=Stanley Phillips|inventorlink2=Stan_Frankel|inventor3-last=Carlyle|inventor3-first=Nelson Eldred}}</ref> These were sheets of iron about {{convert|3|ft|m|0}} in width that were bolted to the top and bottom of the vacuum tank. The effect of the shims was to slightly increase the magnetic field in such a way as to help focus the ion beam. Work would continue on the shims through 1943.{{sfn|Parkins|2005|p=48}}{{sfn|Hewlett|Anderson|1962|pp=92–93}} The main calutron patents were ''Methods of and apparatus for separating materials'' (Lawrence),<ref>{{cite patent|title=Methods of and apparatus for separating materials|country=US|number=2709222|pubdate=1955-05-24|assign1=[[United States Atomic Energy Commission]]|inventor1-last=Lawrence|inventor1-first=Ernest O.|inventorlink1=Ernest_Lawrence}}</ref> ''Magnetic shims'' (Oppenheimer and Frankel),<ref name="Magnetic shims"/> and ''Calutron system'' (Lawrence).<ref>{{cite patent|||title=Calutron system|country=US|number=2847576|pubdate=1958-08-12|assign1=[[United States Atomic Energy Commission]]|inventor1-last=Lawrence|inventor1-first=Ernest O.|inventorlink1=Ernest_Lawrence}}</ref>

Burhop and Bohm later studied the characteristics of electric discharges in magnetic fields, today known as [[Bohm diffusion]]. Their papers on the properties of plasmas under magnetic containment would find usage in the post-war world in research into controlled [[nuclear fusion]].<ref>{{cite journal |title=Eric Henry Stoneley Burhop 31 January 1911 – 22 January 1980 |first1=Harrie |last1=Massey |author-link=Harrie Massey |first2=D. H. |last2=Davis |s2cid=123018692 |journal=Biographical Memoirs of Fellows of the Royal Society |volume=27 |pages=131–152 |date=November 1981 |jstor=769868|doi=10.1098/rsbm.1981.0006}}</ref> Other technical problems were more mundane but no less important. Although the beams had low intensity, they could, over many hours of operation, still melt the collectors. A water cooling system was therefore added to the collectors and the tank liner. Procedures were developed for cleaning the "gunk" that condensed inside the vacuum tank. A particular problem was blockage of the slits by "crud", which caused the ion beams to lose focus, or stop entirely.<ref name="Lawrence and his Laboratory">{{cite web |url=http://www.lbl.gov/Science-Articles/Research-Review/Magazine/1981/81fepi2.html |title=Lawrence and his Laboratory |access-date=3 September 2007 |year=1981 |work=LBL Newsmagazine |publisher=Lawrence Berkeley Lab |archive-url=https://web.archive.org/web/20150208134741/http://www2.lbl.gov/Science-Articles/Research-Review/Magazine/1981/81fepi2.html |archive-date=8 February 2015 }}</ref>

The chemists had to find a way of producing quantities of uranium tetrachloride ({{chem|U|Cl|4}}) from [[uranium oxide]].{{sfn|Larson|2003|p=102}} (Nier had used uranium bromide.){{sfn|Smyth|1945|p=188}} Initially, they produced it by using hydrogen to reduce [[uranium trioxide]] ({{chem|U|O|3}}) to [[uranium dioxide]] ({{chem|U|O|2}}), which was then reacted with [[carbon tetrachloride]] ({{chem|C|Cl|4}}) to produce uranium tetrachloride. [[Charles A. Kraus]] proposed a better method for large-scale production that involved reacting the uranium oxide with carbon tetrachloride at high temperature and pressure. This produced [[uranium pentachloride]] ({{chem|U|Cl|5}}) and [[phosgene]] ({{chem|C|O|Cl|2}}). While nowhere near as nasty as the [[uranium hexafluoride]] used by the gaseous diffusion process, uranium tetrachloride is [[hygroscopic]], so work with it had to be undertaken in [[glovebox]]es that were kept dry with [[phosphorus pentoxide]] ({{chem|P|4|O|10}}). The presence of phosgene, a lethal gas responsible for 85,000 deaths as a [[Chemical weapons in World War I|chemical weapon]] during [[World War I]], required that the chemists wear gas masks when handling it.{{sfn|Larson|2003|p=102}}

Of the $19.6&nbsp;million spent on research and development of the electromagnetic process, $18&nbsp;million (92 percent) was spent at the Radiation Laboratory in Berkeley, and further work conducted at [[Brown University]], [[Johns Hopkins University]] and [[Purdue University]], and by the [[Tennessee Eastman]] corporation.{{sfn|Manhattan District|1947b|p=2.10}} During 1943, the emphasis shifted from research to development, engineering, and the training of workers to operate the production facilities at the [[Clinton Engineer Works]] in [[Oak Ridge, Tennessee]]. By the middle of 1944, there were nearly 1,200 people working at the Radiation Laboratory.{{sfn|Jones|1985|p=123}}