Editing Leo Szilard (section)

===Columbia University===
Szilard visited Béla and Rose and her husband Roland (Lorand) Detre, in Switzerland in September 1937. After a rainstorm, he and his siblings spent an afternoon in an unsuccessful attempt to build a prototype collapsible umbrella. One reason for the visit was that he had decided to emigrate to the United States, as he believed that another war in Europe was inevitable and imminent. He reached New York on the liner {{RMS|Franconia|1922|6}} on January 2, 1938.{{sfn|Lanouette|Silard|1992|pp=166–167}} Over the next few months, he moved from place to place, conducting research with [[Maurice Goldhaber]] at the [[University of Illinois at Urbana–Champaign]], and then the [[University of Chicago]], [[University of Michigan]] and the [[University of Rochester]], where he undertook experiments with [[indium]] but again failed to initiate a chain reaction.{{sfn|Lanouette|Silard|1992|pp=172–173}}

[[File:Fermi and Szilard.gif|thumb|left|upright|Army Intelligence report on Enrico Fermi and Leo Szilard]]
In early 1938, Szilard would settle down at "what would become a haven for much of the rest of his life" when he took a room at the King's Crown Hotel in New York City, being near [[Columbia University]] where he now conducted research without a formal title or position.<ref name="Lanouette 18">{{Cite journal |last=Lanouette |first=William |date=December 1992 |title=Ideas by Szilard, Physics by Fermi |url=https://www.tandfonline.com/doi/abs/10.1080/00963402.1992.11460136 |journal=Bulletin of the Atomic Scientists|volume=48 |issue=10 |pages=18 |doi=10.1080/00963402.1992.11460136 |bibcode=1992BuAtS..48j..16L |via=Taylor & Francis}}</ref> He encountered [[John R. Dunning]], who invited him to speak about his research at an afternoon seminar in January 1939.{{sfn|Lanouette|Silard|1992|pp=172–173}} That same month, [[Niels Bohr]] brought news with him to New York that nuclear fission had accidentally been observed by chemists [[Otto Hahn]] and [[Fritz Strassmann]] at the Kaiser Wilhelm Institute for Chemistry in Berlin on December 19, 1938. Hahn and Strassman's misunderstanding of their observation would be corrected theoretically and thus explained by [[Lise Meitner]] and [[Otto Frisch]], as Meitner had known of Szilard's theory back in 1933 and after re-conducting the experiment themselves, confirmed Szilard's theory to have been correct all along.<ref name="Lanouette 18"/> When Szilard found out about it on a visit to Wigner at [[Princeton University]], he immediately realized that [[uranium]] might be the element capable of sustaining a chain reaction.{{sfn|Lanouette|Silard|1992|pp=178–179}}

Unable to convince Fermi that this was the case, Szilard set out on his own. He obtained permission from the head of the physics department at Columbia, [[George B. Pegram]], to use a laboratory for three months. To fund his experiment, he borrowed $2,000 from a fellow inventor, Benjamin Liebowitz. He wired [[Frederick Lindemann]] at Oxford and asked him to send a beryllium cylinder. He persuaded [[Walter Zinn]] to become his collaborator and hired Semyon Krewer to investigate processes for manufacturing pure uranium and [[graphite]].{{sfn|Lanouette|Silard|1992|pp=182–183}}

Szilard and Zinn conducted a simple experiment on the seventh floor of Pupin Hall at Columbia, using a radium–beryllium source to bombard uranium with neutrons. Initially nothing registered on the [[oscilloscope]], but then Zinn realized that it was not plugged in. On doing so, they discovered significant neutron multiplication in natural uranium, proving that a chain reaction might be possible.{{sfn|Lanouette|Silard|1992|pp=186–187}} Szilard later described the event: "We turned the switch and saw the flashes. We watched them for a little while and then we switched everything off and went home."{{sfn|Rhodes|1986|p=291}} He understood the implications and consequences of this discovery, though. "That night, there was very little doubt in my mind that the world was headed for grief".{{sfn|Rhodes|1986|p=292}}

While they had demonstrated that the fission of uranium produced more neutrons than it consumed, this was still not a chain reaction. Szilard persuaded Fermi and [[Herbert L. Anderson]] to try a larger experiment using {{convert|500|lb}} of uranium. To maximize the chance of fission, they needed a [[neutron moderator]] to slow the neutrons down. [[Hydrogen]] was a known moderator, so they used water. The results were disappointing. It became apparent that hydrogen slowed neutrons down, but also absorbed them, leaving fewer for the chain reaction. Szilard then suggested Fermi use [[carbon]], in the form of graphite. He felt he would need about {{convert|50|t}} (50.8 metric ton) of graphite and {{convert|5|t}} of uranium. As a back-up plan, Szilard also considered where he might find a few tons of [[heavy water]]; [[deuterium]] would not absorb neutrons like ordinary hydrogen but would have the similar value as a moderator. Such quantities of material would require a lot of money.{{sfn|Lanouette|Silard|1992|pp=194–195}}

Szilard drafted a confidential letter to the President, [[Franklin D. Roosevelt]], explaining the possibility of nuclear weapons, warning of the [[German nuclear weapon project]], and encouraging the development of a program that could result in their creation. With the help of Wigner and [[Edward Teller]], he approached his old friend and collaborator Einstein in August 1939, and persuaded him to sign the letter, lending his fame to the proposal.<ref>{{cite web| last = The Atomic Heritage Foundation| title = Einstein's Letter to Franklin D. Roosevelt| url = http://www.mphpa.org/index2.php?option=com_content&do_pdf=1&id=172| access-date = May 26, 2007| archive-date = June 17, 2022| archive-url = https://web.archive.org/web/20220617102507/http://www.mphpa.org/index2.php?option=com_content| url-status = dead}}</ref> The [[Einstein–Szilárd letter]] resulted in the establishment of research into nuclear fission by the US government, and ultimately to the creation of the [[Manhattan Project]]. Roosevelt gave the letter to his aide, [[Brigadier General (United States)|Brigadier General]] [[Edwin "Pa" Watson|Edwin M. "Pa" Watson]] with the instruction: "Pa, this requires action!"<ref>{{cite web|last=The Atomic Heritage Foundation |title=Pa, this requires action! |url=http://www.mphpa.org/index2.php?option=com_content&do_pdf=1&id=173 |access-date=May 26, 2007 |url-status=dead |archive-url=https://web.archive.org/web/20121029020707/http://www.mphpa.org/index2.php?option=com_content&do_pdf=1&id=173 |archive-date=October 29, 2012 }}</ref>

An Advisory Committee on Uranium was formed under [[Lyman J. Briggs]], a scientist and the director of the [[National Bureau of Standards]]. Its first meeting on October 21, 1939, was attended by Szilard, Teller, and Wigner, who persuaded the Army and Navy to provide $6,000 for Szilard to purchase supplies for experiments—in particular, more graphite.{{sfn|Hewlett|Anderson|1962|pp=19–21}} A 1940 Army intelligence report on Fermi and Szilard, prepared when the United States had not yet entered World War II, expressed reservations about both. While it contained some errors of fact about Szilard, it correctly noted his dire prediction that Germany would win the war.{{sfn|Lanouette|Silard|1992|pp=223–224}}

Fermi and Szilard met with [[Herbert G. MacPherson]] and V. C. Hamister of the [[National Carbon Company]], who manufactured graphite, and Szilard made another important discovery. He asked about impurities in graphite and learned{{sfn|Weinberg|1994b}} from MacPherson that it usually contained [[boron]], a neutron absorber. He then had special boron-free graphite produced.{{sfn|Lanouette|Silard|1992|p=222}} Had he not done so, they might have concluded, as the German nuclear researchers did, that graphite was unsuitable for use as a neutron moderator.<ref>{{cite journal | last = Bethe | first = Hans A. | title = The German Uranium Project | journal = Physics Today | volume = 53| issue = 7| pages = 34–36| date = March 27, 2000 |doi=10.1063/1.1292473|bibcode = 2000PhT....53g..34B }}</ref> Like the German researchers, Fermi and Szilard still believed that enormous quantities of uranium would be required for an [[atomic bomb]], and therefore concentrated on producing a controlled chain reaction.{{sfn|Lanouette|Silard|1992|p=227}} Fermi determined that a fissioning uranium atom produced 1.73 neutrons on average. It was enough, but a careful design was called for to minimize losses.{{sfn|Hewlett|Anderson|1962|p=28}} Szilard worked up various designs for a [[nuclear reactor]]. "If the uranium project could have been run on ideas alone," Wigner later remarked, "no one but Leo Szilard would have been needed."{{sfn|Lanouette|Silard|1992|p=227}}