Dear Editor,

In a 2011 Resource Letter, Cameron Reed provided sources on the physics and history of the Manhattan Project, including general works, technical works, biographical and autobiographical works, foreign wartime programs and allied intelligence, technical papers of historical interest, and postwar policy and technical developments.1 

Among the sources included was a paper titled “Enrico Fermi and the Physics and Engineering of a Nuclear Pile: The Retrieval of Novel Documents,” by two Italian physicists from the University of Naples “Federico II,” S. Esposito and O. Pisanti, submitted March 7, 2008.2 Incidentally, Federico II, founded in 1224, is the oldest lay university in Europe.

Reed's “Resource Letter MP-2” said of the above, “The authors described a number of previously uncovered papers and patents that trace the evolution of the work of Fermi and his collaborators in developing reactors.”1 Indeed, the two physicists had discovered at the U.S. Patent Office the paperwork for 15 Fermi patents, filed up to 1952, “the vast majority ranging from 1944 to 1946; All but two dealt directly with nuclear reactors,” noting: “In practice, all these patents were issued many years after their application to the competent office, some of them being even posthumous, and were never published…”1 

The result of a Google Patent search revealed that Enrico Fermi was granted 18 patents through 1952, three from patent offices outside of the United States. The earliest, “Process for the Production of Radioactive Substances,” was filed on October 3, 1934. The most recent, “Method of Operating a Neutronic Reactor,” was filed on December 1, 1952. Fermi's 18 patents, however, are but a tiny percentage of all patents granted to Manhattan Project scientists.

But what is meant by a “published” patent application? Starting in November 2000, the U.S. Patent Office (USPTO) implemented the publication of patent applications 18 months after their earliest filing date (now defined as the earliest priority date). At some point in the patent process, the USPTO creates a full official copy of the application, complete with a unique publication number. As of the publication date, the patent application can be found by anyone who searches for it (e.g., using Google Patents or Public PAIR). PAIR refers to Patent Application Information Retrieval (at

At this point, it is important to clarify the difference between a patent application and a granted patent. Patent applications, once transformed into granted patents, are published by the USPTO. This has been the case since the 19th Century in the United States. An application, however, can be kept secret under a variety of laws passed in the 20th Century. Manhattan Project patent applications, for example, remained classified until after the war, and in many cases (like Fermi's reactor patent applications) were kept classified until the push for declassification came in the 1950–1960s as part of an attempt to create a domestic private nuclear industry. Their declassification in many cases led to their being granted as a patent.

During the war, Vannevar Bush used the Office of Scientific Research and Development (OSRD) in an attempt to acquire a monopoly on the patent rights for inventions used in the production of nuclear weapons and nuclear energy as a means of maintaining secrecy. Reference 3 contains a fascinating examination of this history.3 

Since it is impracticable to locate all patents granted to Manhattan Project scientists, I was fortunate to obtain from the Department of Energy (DOE), in 2017, via the Freedom of Information Act (FOIA: HQ-2017–00351-F), a list of approximately 1500 patent applications related to the Manhattan Project.

The first column of the above report is labeled “S-Numbers” (numbered 000001–004432), If each S-Number represents a single patent application, gaps in that numbered list would suggest that many more patents were applied for than were eventually granted. With few exceptions, these patent applications were submitted between 1942 and 1946. Most patents were later granted in the 1950s, some as late as the late-1950s and early-1960s; and were assigned to the Atomic Energy Commission (AEC).

In addition to Enrico Fermi's 18 patents, others were issued to renowned Manhattan Project scientists including J. Robert Oppenheimer, Leo Szilard, and Harold C. Urey. Patents were also issued to relative unknowns such as James I. Hoffman (S-Number 000001), an analytical chemist who was involved in the Manhattan Project from the beginning.4 

According to an undated AEC report on the purification of uranium oxide (declassified in 1947), “In the early summer of 1941, Leo Szilard (a member of Power Production Subsection of the Uranium Committee) gave a sample of impure uranyl nitrate to the author [James I. Hoffman] and requested the uranium be separated from everything else.”5 Another version of this story appeared in H.D. Smyth's Atomic Energy for Military Purposes (1945), which read, “The use of this method removed the great bulk of the difficulties in securing pure [uranium] oxide and pure materials for the production of [uranium] metal.” Hoffman, who filed his patent application on June 17, 1942, was granted a patent for the process on September 28, 1954 (U.S. Patent No. 2,690,376).

Some of the S-Numbers and their associated inventions remain exempt from disclosure in accordance with DOE Exemption (b)(3), information “specifically exempted from disclosure by statute.” In the case of nuclear patents, the statute was the Atomic Energy Act of 1954, which includes Restricted Data (RD) or Formerly Restricted Data of three types: (1) the design, manufacture, or use of atomic weapons; (2) the production of special nuclear material; and (3) the use of special nuclear material in the production of energy. Whether it is possible to imagine that the mere “title” of a patent application from the 1940s could possibly contain “Restricted Data,” as the DOE censor apparently concluded, I leave to the reader to contemplate.

Later in 2017, that list of Manhattan Project patents was posted to the Wilson Center's “International History Declassified” Digital Archive.6 Their publication will permit future historians and researchers to obtain formerly classified patent files related to the Manhattan Project, and to dig deeper into the patents of Enrico Fermi and many others, some of which undoubtedly have yet to come to light.

B. Cameron
, “
Resource letter MP-2: The Manhattan project and related nuclear research
Am. J. Phys.
, “
Enrico Fermi and the physics and engineering of a nuclear pile: The retrieval of novel documents
,” Report Number: DSF-1/2008; e-print arXiv:0803.1145 (physics.hist-ph).
, “
Patenting the bomb: Nuclear weapons, intellectual property, and technological control
“U.S. National Bureau of Standards,” Nature
“Purification of Uranium Oxide” (MDDC-777), U.S. Atomic Energy Commission, Technical Information Division, Oak Ridge Directed Operations, Oak Ridge, Tenn., Date of Manuscript Unknown (Declassified on March 18, 1947).
See <> for the Department of Energy’s “List of U.S. patents related to the Manhattan Project.”