Development of Mass Spectrometry in the United States: The First 50 Years

Gary D. Christian of the University of Washington organized a symposium titled "Evolution of Analytical Sciences in the United States" for the Fall 2008 meeting of the American Chemical Society. In that symposium, I presented an overview of the development of mass spectrometry (MS) in the United States, covering a period of about 50 years. Half of a century is long enough to gain perspective, but not so long that the documents and artifacts from the beginning of the period have been lost or moved into museums. Some of the presentations from that symposium have been summarized on teh ACS website, but MS was not included. The main points of my presentation are condensed into the present column.

Evolution

is a particularly apt term to describe the development of analytical sciences. In reviewing the past 50 years of mass spectrometry (MS), it became clear that evolution's usual iterative progress was interrupted more than a few times by unforeseen instrumental developments, external pressures of law or regulation, and sometimes by a powerful dose of publicity. George Bernard Shaw said, "All evolution in thought and conduct must first appear as heresy and misconduct." Evolution, especially at the rapid pace evident in MS, often made its practitioners uncomfortable, and sometimes somewhat less welcoming of new ways of doing things propounded by the heretics. Resistance may be natural, but it has always been futile, and after 50 years of inevitable evolution in MS, we might start to characterize it and strive to understand some of the driving factors.

Kenneth L. Busch

We previously described the S-curve of innovation as it describes the development of inductively coupled plasma–mass spectrometry (ICP-MS). In 1962, Everett M. Rogers published a book entitled Diffusion of Innovations, in which he described the spread of innovations through an S-curve model (Figure 1). Early adopters select and explore the technology first, followed by the larger majority, with the process continuing until a technology or innovation becomes commonplace. The figure shows labels that might appeal more to Mr. Shaw and to us — heretics, pioneers, adopters and adapters, and finally the beneficiaries. The general validity of the S-curve as a model of technical innovation is widely accepted, and it fits the evolution of MS in the United States. The curve in MS is particularly steep and especially evanescent, as one curve soon gives way to the next, as new instruments, new ionization methods, new interpretations, and new demands for information sweep aside old MS to carve out new audiences. The curve describes an inevitable evolution and does not provide a specific forecast. As we assess details of MS development at a particular time, we can gauge the placement of that point in time on the curve.

Figure 1

Many resources describe the evolution and history of MS; Michael Grayson's presentations for ASMS are always insightful and complete. Adopting a wider perspective, I describe the evolution of MS in terms of simple synergistic factors that can be described and evaluated — who did it, where it was done, how it was paid for, what it was done with, when it was presented, and finally why.

We start with Who, and the pictures of a few pioneers are composited in Figure 2. These names should convey a concordant appreciation of the significance of their accomplishments. Even at the foot of the S-curve, there are many more names that should be included, and the overlapping work of multiple scientists with careers that span decades defies a simple linear progression. In the United States, many of the very first users of MS were in the petroleum industry. The early publications of Johnsen (at Monsanto in Texas City, Texas); of Brown, Taylor, Melfolder, and Young (at Atlantic Refining Company in Philadelphia, Pennsylvania); and of Sy Meyerson (at Amoco in Whiting, Indiana) include, as you might expect, applications of MS to hydrocarbon analysis. The tempting apple of organic analysis was also soon tasted. The field of organic MS was greatly advanced in 1960 by the monumental books (1,2) of Bieman (at MIT) and Beynon (who had not yet traveled to the colonies). A list of early mass spectrometrists includes the name of Rick Honig, working for Sarnoff Laboratories of RCA, who pioneered the use of secondary ion mass spectrometry (SIMS) in an industrial setting. There were still other applications of MS. Claire Cameron Patterson of the California Institute of Technology (building on his experience at the University of Chicago and later with the Manhattan Project at Oak Ridge) received an NSF grant in 1961 to support the construction of a mass spectrometer. Patterson had impeccable credentials in the field of lead isotope analysis; his uranium–lead dating process was used to determine one of the first reliable dates for the age of the Earth.

Figure 2

The breadth of interests in these early researchers is prescient. Recall that the American Society for Mass Spectrometry presciently refers to its annual meeting as a conference on "mass spectrometry and allied topics," reflecting the genesis of ASMS in committee E-14 of the American Society for Testing of Materials. The first petroleomics folk that formed a core group in the early ASMS meetings were soon complemented by a cadre of (eventually) academic researchers in other areas. Foremost among the early researchers is Al Nier, who first worked for industry associated with the Manhattan Project, and then joined the University of Minnesota (Minneapolis, Minnesota). Likewise, Fred McLafferty started at Dow and then joined Cornell University (Ithaca, New York). Jean Futrell started a scientific career in a petroleum company in Baytown, Texas, and then detoured through Wright Patterson Air Force Base, where he modified a CEC instrument there for analytical MS in the early 1960s, almost 50 years ago. He ultimately found his way to Utah, then Delaware, and is now at the Pacific Northwest National Laboratories (Richland, Washington). Al Sharkey was a physicist by training who first worked with Westinghouse when Westinghouse was building mass spectrometers, and was then later with the Department of Energy, and also at the University of Pittsburgh (Pittsburgh, Pennsylvania), where he taught one of the first graduate level courses on MS in 1964. Kenneth Rinehart is an example of an individual who started in organic chemistry within an academic setting but was soon drawn to the allure of MS for structural elucidation in the early 1960s. Rinehart's work in MS and natural products chemistry at the University of Illinois (Urbana, Illinois) extended over decades, as did the work of many others from this period.

In the mid-1950s and early 1960s, as MS became established in U.S. universities, credit also must be given to researchers and educators such as R.W. Kiser (at Kansas State at the time, and later at Kentucky), whose book Introduction to Mass Spectrometry and Its Applications appeared in 1965 (3). This book can be viewed as one of the first books on analytical MS, complementing the earlier Bieman (1) and Beynon (2) texts on organic MS. The book has been reprinted by ASMS. In addition, much credit is due analytical scientists such as Herbert Laitinen and C.N. Reilly, who defended analytical chemistry as a profession over a period of decades and provided opportunities for publication of substantial new developments and new applications. Even early in the 1950s, it was clear that MS blended physical chemists, organic chemists, physicists, engineers, and information scientists, and also those magical folk that could actually make the machine work. "Analytical chemistry" continues to be the perfect descriptive umbrella for this diverse group.

The next factor is Where. From an academic point of view of the past 50 years, a great deal of research in MS was carried out at the land-grant universities of the United States. Land-grant colleges were established by the Morrill Act of 1862, passed by the 37th Congress and signed by Abraham Lincoln (this was also the Congress that passed the Internal Revenue Act.). With an emphasis on practical arts and sciences, chemistry became a core component of the educational curricula at land-grant colleges. With the expectation that a college education should be available to more citizens, enrollment jumped. More students meant more professors, more chemistry professors, more analytical chemistry professors, and ultimately, more mass spectrometrists. MS took root in places sometimes far afield from early venues such as MIT or Chicago or CalTech. Graham Cooks started at Kansas State University (Manhattan, Kansas). The honorable Charles Judson was at the University of Kansas. A minimum size of college or university might be needed to support research in MS. Industrial concerns come and go, even if they are of the size and quality of Sarnoff Laboratories, where Honig worked. But larger land-grant universities tend to be more stable, and academic researchers in MS have flourished and continue to multiply there. Of many similar academic laboratories that maintain an extended lineage, the Aston Laboratory at Purdue University (West Lafayette, Indiana) keeps a list of students and post docs who have carried out research there. At last check, the list extends to number 276. Many graduates now work for industry; many became academic researchers who have since graduated their own Ph.D. students, who have in turn become professors doing research in MS. The lineage from the Aston Laboratory, as in many similar examples, can be traced through a third generation and now into the fourth. This expanding population of capable, inquisitive, and demanding researchers has catalyzed the evolution of MS.

Certainly, universities were not the only sites for MS research. A paper from Berry and Walker of Consolidated Engineering Corporation (CEC) (4) documents hundreds of industrial laboratories with mass spectrometers by 1955. Government laboratories, and in particular the national laboratories such as Oak Ridge National Laboratory (ORNL), have a long and parallel history of development in MS, in applications but also in leading innovation. Certainly at ORNL, most of the early work was with isotopes and MS, but significant advances in trace detection and organic analyses were also achieved. Industrial laboratories, faced with challenging analytical problems, also were early adopters and adapters of MS. Development of MS in industrial laboratories is more difficult to document. The early close collaboration between industrial users and instrument manufacturers apparent in the earliest days of CEC evolved into a more constrained arrangement in which the protection of intellectual property and trade secrets was paramount. However, industrial users have always been prodigious purchasers of new instrumentation. Ph.D. students moved into industry with increasingly more sophisticated expectations for instrumentation. Instruments were purchased to allow these users to meet the increasing regulatory burden and increasing competition.

The next factor is How. Research in MS has to be paid for. Government laboratories and industrial laboratories have their inherent specific mechanisms for funding. Academic research seems sometimes to depend upon the kindness of strangers — that is, those strangers on the review panel for proposals. In the evolution of MS in the U.S., it was the establishment of mechanisms for federal funding of research that became an important factor. NIH funding has its focus areas and is to be credited for establishing centers for MS. Mission-oriented federal agencies provided specific research support. MS has always been part of the National Science Foundation (NSF) funding portfolio. We can search the public awards database and pull up some of the earliest awards in MS. These awards, dating back to 1953, went to the heretics — er — pioneers of the field, in the expected locations, and even in a few unexpected locations. Two early awards went to Northwest Nazarene College in Idaho. As unexpected as that location might seem, the PI for those awards was a researcher with experience from the MS work in the Manhattan Project.

Several special aspects of NSF funding played roles in the evolution of MS. NSF sets aside funds in special programs specifically to support instrument development. NSF set aside funds in special programs to support the acquisition of major instrumentation (including mass spectrometers) at colleges and universities. NSF reviewers and program officers were always strongly supportive of the inclusion of support for graduate and undergraduate students in research proposals. They also strongly suggested that research results be published promptly and disseminated widely. While NSF research funding was never guaranteed, there was an expectation that good work could be supported with continued funding. Finally, the quality of the proposed and completed research work was assessed, and the competition for funding predicated, on comparisons with other proposals received from across the nation. The competition was national in scope, and there was therefore a clear and convincing need to do good work, put it out before the community, and keep track of what the other folks in MS were up to. Kelsey Cook prepared a breakdown of NSF funding that gives some measure to MS funding from NSF in more recent years. In 1984–1985, MS-coded awards constituted 10% of the Chemistry Analytical and Surface Chemistry portfolio. Two decades later, MS had risen to 17% of the funded projects. A large slice of awards in "surface and interface analysis" still contains its own subset of MS in the guise of secondary ion MS, among other techniques, and we can expect that MS shows up in the "sensors" area as well.

The What factor refers to the instruments of MS, and the focus is on instruments manufactured in the United States. Many instruments were specially constructed by the first researchers, the heretics and the pioneers, for specific purposes. In the United States, many of the first magnetic sector mass spectrometers were manufactured by Consolidated Engineering Corporation, known as CEC. CEC was founded in 1937 by Herbert Hoover, Jr. CEC's first Vice President for Research was Harold Washburn. Although Washburn's job was to make instruments for petroleum prospecting, his thesis professor was E.O. Lawrence of cyclotron fame. Not surprisingly then, CEC soon began a project to develop a mass spectrometer (4). According to CEC history, the initial product produced was the 21-101 Mass Spectrometer, which was delivered in December 1942 — at a price of $12,000. The name of the company was changed to Consolidated Electrodynamics Corporation in 1955, because some states required that a service engineer for an engineering company be a licensed engineer in that state. There seems to be no escape from regulations and bureaucracy. CEC became a subsidiary of Bell and Howell Corp. in 1960, then became the Electronics Instrument Group, and finally evolved into the "Analytical Instruments Division" of Bell and Howell. The division was sold in about 1975 to the Instrument Division of DuPont, who abandoned the business of making mass spectrometers by 1980.

Time-of-flight (TOF) mass spectrometers have a U.S. origin, and were first described by William E. Stephens of the University of Pennsylvania (Philadelphia, Pennsylvania) in a Physical Review paper in 1946 (5), and patented shortly thereafter. The improvements made by McLaren and Wiley have been well documented, and early TOF instruments were manufactured in the United States by the Bendix Corporation. This made the news. On August 3, 1957, the New York Times had a short article headlined "Desk-Top Device Speeds Analyses; Mass Spectrometer Sorts Out Molecules and Counts and Identifies Them." Reporter William M. Freeman wrote that "Not very long ago an industrial chemist who sought to identify the components of a chemical mixture was obliged to spend many hours, even days, testing for one ingredient after another," and touted the speed of the analysis with this amazing new device called the mass spectrometer. We can also investigate the ad from Science in August 1959 featuring Roland Gohlke (Figure 3). The desktop device does not appear in retrospect to be "desktop" as described, and we might not believe all of the rest of the advertising text.

Figure 3

The evolution of MS has been punctuated by new instruments produced to meet new analytical needs, changing the shape or starting entirely new S-curves. The field of environmental MS will always be linked to the development of practical quadrupole-based gas chromatography (GC)–MS instruments, and the analysis of New Orleans drinking water (6). The instrumental part of the evolution of MS is perhaps the most widely known, since we retain at least a few artifacts of our past. Sometimes special aspects of history or location preserve the instruments that were part of the evolution of MS. Some of the alpha and beta Calutrons are still in place at Oak Ridge, where they were first used for the enrichment of uranium isotopes during the Second World War, and were still in operation in the 1990s. There's also a GC–MS unit associated with the Viking Lander that set down on Mars in 1976, waiting on a service call.

When refers to the chronology of publications and presentations. Before there were specialized journals for MS, there was the journal Analytical Chemistry and its precursors. It is interesting to note that there were so many errors on one of the first articles containing mass spectra of a variety of organic compounds that the entire set of tables had to be republished in a corrected version a few months later. The need for reliable reference mass spectra assembled in a collection was recognized early. First was a collection of mass spectra on McBee "Keysort" cards, then the collection of mass spectra in the American Petroleum Institute project, and then the reference spectra efforts of ASTM Committee E-14. The topic of reference mass spectra has its own intriguing history to be described at a later time.

H.W. Washburn (from CEC) was the first author on the two seminal papers on MS that appeared in the Industrial Engineering Chemistry, Analytical Edition, in 1943 and 1945. In the newly named journal Analytical Chemistry, one paper on MS by the Atlantic Refining Group (Brown and colleagues) appeared in all of 1946, and then one paper by Johnsen in 1947 that established a quantitative basis for MS analysis. Then, suddenly, the 1948 volume of Analytical Chemistry contained seven papers on MS, including the first suggestions that MS could be used for general organic analysis. We were already ascending the S-curve. Within a few decades, there were journals specializing in MS, and these have evolved over the years, merging as needed, but almost uniformly growing larger and with a bigger audience. The list of specialist journals includes the Journal of the American Society for Mass Spectrometry. ASMS has embraced both the evolution and the revolutions of our field, and benefited from the service of many prominent mass spectrometrists, including perhaps a few of the heretics, in the ranks of its officers. The steady growth of attendance at the conference is only rarely matched by other U.S. professional societies.

Finally, the Why factor. To simply state "because it works" underestimates the seductive nature of metrology. Once the capability to perform a measurement is in place, it seems to be human nature to adopt the metric and expect that the measurement will be made routinely. In weather, it does not suffice to be hot or cold; we have to know the exact temperature. We can't have an earthquake without a Richter-scale value, or a hurricane without a category classification. The public might not know all the details of the analytical MS. But the public wants environmental analysis, forensic analysis, certified pesticide-free produce, or certified dopant-free athletes. MS is often introduced as a "sensitive" method of analysis; we progressed from micrograms to nanograms from 1950 to 1960, and from nanograms to picograms by 1970, to femtograms by 1980, to attograms now. Now we root around in samples and sample solutions to track down and analyze only a few molecules and ions.

In the past 50 years, because of increased sensitivity and increased capabilities, the world of application of MS has expanded by a thousandfold. The seemingly magical capabilities of just a few years ago are routine and expected, whether it be pediatric screening by MS-MS, or the use of MS in the search for viable disease state markers. The late Arthur C. Clarke said, "Any sufficiently advanced technology is indistinguishable from magic." If we understand the evolution of MS, any illusion of "magic" fades away, and our vision grows clearer. Core competencies in MS continue in their evolutionary steady growth and broadening applications. Leading-edge capabilities show the expected series of stops and starts, trial and error, but the revolutionary developments in this field continue. In hindsight, we accept them all – so much so that we might have to modify Clarke's quote to "Any sufficiently commonplace technology is in danger of becoming indistinguishable from magic." We have an obligation to continue to educate all current and potential users about MS so that potential illusion and deception do not gain a foothold.

Almost 51 years to the day after the New York Times commented on the rapid analysis of mixtures by MS with the Bendix time-of-flight mass spectrometer, the Times carried another article about the use of desorption electrospray ionization headlined "Fingerprint Test Tells What a Person Has Touched," referring to a report in the journal Science (7). These two simplified newspaper articles, 51 years apart, illustrate the evolution of MS. MS data have almost always been used previously to tell us about things — elements and their isotopes, organic molecules and their structures, mixtures and their components. First we had data, and then data collected into databases, and then we began to take full advantage of all the higher dimensions of data through informatics. Now, our evolution has taken us to the edge of a great abyss, and neither our pioneers nor our heretics have yet to be found. MS is now being used to tell us about ourselves, and about single identifiable individuals — their health, their diseases, their potential diseases, what they have eaten, what they smoked, and what they have touched. Information generated by mass spectrometry now has to be evaluated in an ethical environment that considers factors of confidentiality, control, certification, and even culpability — this is informethics. We are at the foot of the S-curve of this brand new field. It's maybe a bit uncomfortable, but it's an exciting place to be.

Kenneth L. Busch owns a few artifacts of earlier mass spectrometry, such as the manufacturer name plate from a Dupont 21-490 mass spectrometer, and an early ZAB-2F instrument. These have meaning because he actually worked on these instruments. He also visited the Science Museum in London solely to see one of Aston's first spectrographs. This article expresses only the opinion of the author and not those of the National Science Foundation. He can be reached at WyvernAssoc@yahoo.com

References

(1) K. Bieman, Mass Spectrometry — Organic Chemical Applications (McGraw Hill, New York, 1962).

(2) J.H. Beynon, Mass Spectra and Its Application to Organic Chemistry (Elsevier, New York, 1960).

(3) R.W. Kiser, Introduction to Mass Spectrometry and Its Applications (Prentice Hall, Englewood Cliffs, NJ, 1965).

(4) C. Judson, Consolidated Electrodynamics Corporation CEC 100 Series Mass Spectrometers, available at http://www.asms.org/portals/0/CEC_103C_HRes.pdf

(5) W.E. Stephens, Phys. Rev. 69, 691 (1946).

(6) B. Dowty, D.R. Carlisle, and J.L. Laseter, Environ. Sci. Technol. 9(8), 762–765 (1975).

(7) D.R. Ifa, N.E. Manicke, A.L. Dill, and R. Graham Cooks, Science 321(5890), 805 (2008).