Bendix first Gas Chromatograph/Mass Spectrometer capable of handling 25 ml/min flow rate directly into MS source from GC. Bendix model MA-2/015/2500 with (Luchte) male technician. CHF Collections.


DuPont 21-103C MS

Diagram of the DuPont 21-103C Mass Spectrometer. CHF Collections. Click here for full size

Mass-spectrometry instrumentation certainly has evolved since the days of Thomson’s experiments and Aston’s spectrometer. The mass spectrometer has changed over the years, becoming more efficient and more precise. But it has also been combined with other technologies to create an even more powerful instrument capable of the most sensitive analysis of a wide variety of samples. Mass spectrometers are highly customizable, capable of being used both in tandem and combined with a wide number of other analytical technologies. Intrepid scientists have consistently taken advantage of the customizable nature of the instrumentation, and the resulting changes demonstrate the fast pace of dynamic research and innovation, a hallmark of the field. Klaus Biemann notes:

That is one of the things that make chemistry and instrumentation nowadays so rapidly changing because it gets so highly developed, and in the process so expensive. People think of doing it better, cheaper, and faster, and the moment that [it] succeeds of course makes previous expensive, complex, very sophisticated, and very elegant instrumentation overnight useless. It is left for some very narrow fringe things in which you can do some detailed studies. (51)

While there are many technologies that could be featured here, we’re focusing on four particular developments—quadrupole mass spectrometry, gas chromatography–mass spectrometry (GC-MS), time-of-flight (ToF) mass spectrometry, and matrix-assisted laser desorption/ionization (MALDI)—that the CHF interviewees talk about at length.


It also gave me a chance to see various laboratories in Germany, and at a different time, in England. In fact, we stopped in Bonn [Germany] and I saw the first prototype mass spectrometer, well, quadrupole with four long wires strung in, from the first down to this—from the second floor down to the first floor. . . . It was a large instrument. (Richard Honig, 21)

Quadrupole mass spectrometers, which were developed in the 1950s by Wolfgang Paul, as observed by Richard Honig on a trip to Germany, offered researchers more of the same basic technology as previous mass spectrometers but with a few new incentives. There was increased resolving power in a more compact piece of equipment; no large electromagnet was required; and the instrument could be modified and specialized through combination with other quadrupole instruments or other analytical tools, such as gas or liquid chromatography.

At its most basic a quadrupole mass spectrometer consists of four rods in parallel through which stable ions flow. The mass-to-charge ratio of the desired ions is specified by modifying the applied electric field. But quadrupole had potentially wide-reaching applications. One example is in the area of process control. Robert Finnigan explained,

Hear Robert Finnigan: So what we were doing from EAI [Electronics Associated Incorporated] was saying, “We really think that there is a need for analog controls that would really do the job better. They’re relatively inexpensive and you could have, ultimately, a digital decision-maker back here—it would be sort of hybrid. But you need different and better sensors. You could really improve your process, if you had better sensors.” And ultimately we saw a quadrupole mass spectrometer as one of those instruments which would give the composition of the products of a process plant, allow you to control levels of certain organic components, and so on. I think they’re doing a lot of those things now forty years later. There was a colleague of mine at the Air Force Institute of Technology, Ted [Theodore] Williams, who built the first all-digital controls after he left the Air Force. He went to Monsanto [Company] and pioneered direct-digital control. It was considered really revolutionary and encountered great resistance. Today, it would be considered passé. I’d say a lot of the things, including quadrupoles, are used in process control now. But this was thirty-something years ago. And, then, that was too far out. (49)

All these years later nothing is “too far out,” and quadrupole mass spectrometers have become commonplace.

As with any mass-spectrometry instrumentation, modifications and combinations with other technology would soon follow. Quadrupole modifications include the triple quad, which contains a linear series of quadrupole mass spectrometers to increase fragmentation of analyte molecules. Marvin Vestal explained,

Around that time, I had been doing this photo disassociation work with the triple quad, and I believe I built the first triple quadrupole. Jim Morrison built his very nearly the same time, but he was in Utah when I built mine, and he finished his [and] up and went back to Australia. That’s the one Yost and Enke used for their work. And I built a triple quad to do photo disassociation, and I considered collision disassociation to be a royal pain in the ass. (43)

While there is no controversy regarding the origin of the triple quad, Vestal further explained,

I don’t know if he [Morrison]’s really credited with it. But certainly the one that my friends, Enke and Yost used was the one he built. I don’t even think they were even aware that I had one, to tell you the truth. I published the first paper for triple quad. It’s a photo dissociation paper, Jean [Futrell is] the coauthor on that. . . . Well, I don’t claim that Morrison got the idea from me, and I don’t think he claims I got the idea from him. I’m sure we did discuss it. (46)

The quadrupole mass spectrometer offers analytical tools that previous mass spectrometers did not. However, the real gain with the quadrupole is in the ways it can be modified and suited to an individual researcher’s needs.


Developed in the 1950s, gas chromatography can be combined with mass spectrometry to create a powerful analytical tool. Gas chromatography by itself is an analytical method for separation and purification based on turning samples into the gas phase. The vaporized components of the GC mixture reach the detector at different times, allowing for separation. By combining the two instruments into one process, the gas chromatograph is able to ionize volatile compounds and separate out the heavier components for subsequent ionization and analysis in the mass spectrometer. Over the years such instruments have been used in individual laboratories, in industry, and in a variety of applications ranging from proteomics to drug screening to environmental research.


Bendix first Gas Chromatograph/Mass Spectrometer capable of handling 25 ml/min flow rate directly into MS source from GC.  Bendix model MA-2/015/250, ca. 1970. CHF Collections. 

One of the first GC-MS instruments was made by Roland Gohlke and Fred McLafferty at Dow Chemical Company. Although their instrument was not a commercial one, it was described by Robert Finnigan:

They built a GC-MS using a time-of-flight mass spectrometer some time in the late 1950s, I think. I can’t remember the exact date. But it was really a breadboard instrument, a jerry-rig, one-of-a-kind sort of a deal. But they had run an instrument early on. (62)

McLafferty explained how he and Gohlke came to have a GC-MS and the relationship they had with the instrumentation company Bendix, located in nearby North Detroit.

Roland [and I], we had very close relations with Bendix . . . because they were in North Detroit and they were only an hour and a half drive from Midland. And, that’s when we did our first GC-MS. We took Roland’s GC and put it in the backseat of the car, and we drove down to Bendix—and I think that was February 1956. And, and so that made it very easy because they had the instrument. No. Yeah. February 1956? Yeah, that’s right. But we got the first laboratory time-of-flight, except that it arrived just about the time I had left. (58)

Similarly, Earl Lumpkin at Humble Oil in Texas was using GC-MS. He explained that the data analysis component had no prewritten computer programs as exist today.

I also helped write our own computer programs. In the latter few years, when mass spectrometers had a dedicated computer program of their own, we used the computer program provided for us only for data acquisition. For the handling of the data on down the line, we wrote our own programs. We’d [taken] the data, fed it into a tape read-off, and selected the masses that we wanted and what we wanted to do with them. (25–26)

Klaus Biemann of MIT employed mass spectrometry in his studies of organic chemistry, and he eventually began studying peptides and proteins using the technique. He was also involved in creating the GC-MS instrument for the Viking program’s 1976 Mars launch.

One of the objectives was to look for any signs of biology, past or present, associated with a search for organic compounds because it was not only related to potential biology questions but also, of course, to the origin and past history and present state of the chemistry of Mars. Since no one had an idea what to look for and had to be prepared to find any kind of compound, I proposed to use mass spectrometry because of its wide general applicability and sensitivity: a gas chromatograph for separation and the mass spectrometer for identification. That proposal was accepted in 1969, and the rules of the game required that each experiment had a team of scientists to do the experiment and have the capability of interpreting the data. That team was constituted in 1969, and I was appointed team leader. We were told that the Jet Propulsion Laboratory had a miniaturized gas chromatograph from earlier lunar proposals, but [it] was never sent, and also a miniaturized mass spectrometer. So the instrument side was all set, and all we needed were people who could make sense out of the data. That turned out to be not really the case. They had built a miniaturized double-focusing Nier-Johnson mass spectrometer planned for some atmospheric studies on Earth in rocket flights, and it was just sitting there. Al Nier had done very successful flights in the upper atmosphere with his instrument; ironically he used Mattauch-Herzog geometry, not a Nier-Johnson system. JPL’s instrument had to be redesigned because the miniaturized mass spectrometer part was okay, but the gas chromatograph was not suitable, nor was the data system suitable. The original plans called for doing all the interpretation of the mass spectra automatically on Mars, and send the results back. The data interpretation was to use our low-resolution mass-spectral identification algorithms. I said that wouldn’t work. We had to get all the raw data back because we had by that time developed mass chromatograms, data-interpretation modes which were based on having complete mass spectra of each scan, of each mass spectrum of continuously recorded spectra coming from the effluent of the gas chromatograph. I wanted to have all the data because only that way could we be sure. The reason for not doing that was the data limitation for recording it on the lander and then sending it back to the orbiter, and from there sending it back to Earth. Fortunately, JPL had developed a tape recorder that didn’t use a tape because that couldn’t be sterilized. One problem with the entire Space Program at that time was that in order to go to extraterrestrial bodies, the rule was that you could not contaminate that body, like Mars, with terrestrial living things. Therefore, to get the Viking landers there, they had to be sterilized, and that required heat sterilization. So because you could not heat up and sterilize a tape, JPL had developed for some earlier unmanned lunar missions which were never flown, a recorder that didn’t use a plastic tape. It used a metal wire. They could revitalize that recording system, and it was put on Viking. That also made it possible to record a large number of images, which required much more data than a simple GC-MS data set. (30–31)

Biemann’s laboratory was one of the earliest to use GC-MS for peptide and protein work. He explains his early experience with GC-MS: 

Off-line when I started in 1958, on-line probably 1962. It was published in 1964. At first we used it [GC] on the CEC 21-110B high-resolution instrument. The first examples were on alkaloids where we separated a mixture of alkaloids from a certain plant by gas chromatography directly into the high-resolution mass spectrometer. Since the alkaloids contained carbon, hydrogen, nitrogen, and oxygen, the elemental composition—which you could get from a high-resolution mass spectrometer—was very helpful. (19)

Using the same analytical tools by which the GC-MS helps elucidate proteomics, the technology has been used in screening newborns and young children. Biemann explained,

We developed a procedure for looking for drugs and metabolites in newborn babies and children, mainly the accidental ingestion of things which they found in the bathroom cabinet; overdoses, unintentioned, rarely any foul play was involved. But it was important to identify it quickly and reliably. That was just after we had developed GC-MS and used it for alkaloids, for example. We made a connection with an anesthesiologist at Harvard Medical School and then developed that as sort of emergency service. We generated a computer program that would look through the GC traces and the mass spectra to identify what didn’t belong. We had a 24-hour operation, that if a child was brought into the emergency with some funny symptoms which indicated some toxic problem, they would take the blood and the urine sample, call a cab to send it over to our laboratory, and call someone from my laboratory at home and that person then went to the lab. The cab driver was told exactly where to go. That was at a time when not all the buildings were completely locked, and he was told to go to the basement of building 56 and hand over the samples. The person on call would quickly extract the two samples, run the GC-MS, figure out what it was, and call the physician back and tell them what it was. We did that for two or three years, until it became established. It became so common that analytical laboratories, at least in the greater Boston metropolitan areas, made it a business to run all kinds of analyses, including having mass spectrometrists so that it could be turned over to those commercial laboratories. There was a need for it, with clients that would use it, so they could set up the procedure, and assign or hire someone to do it. (43)

GC-MS has been used extensively for over 60 years and speaks to the versatility of the mass spectrometer as an analytical tool.

 Time of Flight

Time of flight has a long history: along with gas chromatography, it’s one of the earliest technologies to be paired with mass spectrometry. ToF mass spectrometry consists of determining an ion’s mass-to-charge ratio via a time measurement; an electric field accelerates all ions into a field-free drift region; lighter ions have a higher velocity than heavier ions and reach the detector at the end of the drift region more quickly. ToF was developed in the 1950s and looked promising. The mass range, theoretically, was unlimited. Fred McLafferty was one of the many mass spectrometrists who dabbled with ToF early on. His collaborator, Roland Gohlke, proved in 1957 that a Bendix-manufactured ToF instrument was compatible with GC-MS, and the two went to work using both technologies. Klaus Biemann also used ToF and recalled the potential and the drawbacks of the technology. Biemann and Michael Grayson discussed early ToF use:

GRAYSON: The Bendix time-of-flight was used quite a bit or at least looked like it would have been a good instrument for GC-MS. It had the speed in terms of the scan speed.  
GRAYSON: But it was eventually beat out by quadrupole?  
BIEMANN: And the reason was that the early time-of-flight mass spectrometers, and the Bendix was the major commercially available one, had relatively low resolution. When it came out, there were no recorders that were fast enough to record the signal. The electronics were of the 1960s or even 1950s. Gohlke used it in McLafferty’s laboratory to more or less show that you can record and see the mass spectrum of toluene while it comes off the GC column. The mass identification was more or less measuring off the oscilloscope screen using a Polaroid camera. So if you know what it is, you can say, this is mass 92 and mass 91. But when it is completely unknown, it would have been more difficult. The only advantage it had over magnetic instrument was the speed, but it was almost too fast. It could also tolerate a large gas load, which with GC-MS was the main problem to overcome before separators were developed. As history shows, aside from that one paper, maybe two, that was it. There were a number of other ways to try to get mass spectra of GC eluates: by complex trapping and then injecting methods, but they never got beyond the demonstrations that it can be done. 
GRAYSON: It was too cumbersome. (60)


Time of Flight

Bendix Model RGA-1A Time-of-Flight Mass Spectrometer. CHF Collections.

Bendix began manufacturing commercial ToF instruments soon after Gohlke’s demonstration, but as Biemann noted, there were problems with the instrument’s capabilities. The instrument’s resolving power was limited; thus most researchers using mass spectrometers were not willing to invest in the instrument. Keith Jennings noted that ToF wasn’t necessarily recommended in its early incarnation: “Indeed, and people might use it for things like shock-tube work. But serious mass spectroscopists didn’t mess with time-of-flight. . . . John [Beynon] said that so many Bendix instruments were not working, and he didn’t think I should get involved with one of them” (91). The instrument was also expensive and difficult to operate, and thus ToF spent the next 20 years as a forgotten instrument.


Time of Flight

Bendix Model RGA-1A Time-of-Flight Mass Spectrometer. CHF Collections.

ToF experienced a resurgence in the 1970s starting with improvements made on the instrument that increased resolving power and sensitivity.

GRAYSON: I started using time-of-flight instruments when I was at McDonnell-Douglas research labs. When I first started there, that’s what they had. I’ve always been fascinated with the fact that today you can get a time-of-flight instrument with resolving power of 10 or 20,000 and back with a Bendix machine a couple hundred resolving power was the most you could get. It’s just fascinating to me to think that you could improve the technology of that instrument, that analyzer, so much that there’s two orders of magnitude improvement in the resolving power, due primarily to developments in electronics.
BIEMANN: Faster electronics and computer control of all the parameters down to high limits made it possible, while on the magnetic instruments computer control wouldn’t have done much to the performance. It was the ion optics that had to be tuned. Of course computers helped there by calculating the important parameters. First, [Hisashi] Matsuda and then [Takekiyo] Matsuo, his student, refined ion optics for magnetic-sector instruments to achieve higher resolving powers. (60–61)

Use of ToF instrumentation increased even more after it was paired with MALDI. Biemann continued,

There was the big comeback of time-of-flight mass spectrometry when MALDI was developed because it needed a pulsed instrument. . . . They were steadily improved for very special things. But then that ionization method made it necessary to build a high-performance instrument of that type. By that time electronics and computer control had advanced so much that it was relatively easy to build a high-performance instrument almost overnight, and it has been improved since. For example, even at that intermediate time [Boris A.] Mamyrin in Russia had developed the reflection. Nobody was using it, except him and a few other people, until MALDI came around and it was found of great use. So now all MALDI time-of-flight mass spectrometers are reflectrons in one form or another. (60)

ToF also helped MALDI gain wider acceptance. Marvin Vestal commented, “The other side of that is, though, the reason the MALDI didn’t take off faster, I think, is it really only was a good fit in the time-of-flight” (66).



In the 1980s mass-spectrometry technology improvements had tremendous repercussions for the biological research community. Biological researchers had a lengthy history of dabbling in mass spectrometry, but until technologies like MALDI and electrospray (and before those two particular techniques, thermospray), larger biomolecules like proteins were nearly impossible to analyze. MALDI, a “soft” ionization technique, works via a two-step process. A sample to be analyzed is dissolved and dispersed on a matrix material. A laser is applied, and the matrix and sample are vaporized; the matrix is then ionized, and many of the sample molecules are then ionized as well. The mechanism of MALDI—how exactly it achieves ionization—is today still debated. In the mid-1980s the MALDI technique had been shown to ionize a protein. With that breakthrough a rush to use this new technique led to the explosion of proteomics: identifying and understanding proteins could happen increasingly faster and continued to get easier.

Once MALDI’s capabilities were known, researchers saw an immediate need to switch to the new technology. Vestal, whose thermospray technique had been extremely popular and a basis for his instrumentation, recalled that by 1993 “it was clear that MALDI had great potential and it [was] going to be the wave of the future” (70). He noted that others saw this as well:

They had this group set up in England that was going to build a small bench-top californium instrument, and they built this time-of-flight. And part way through the project, that’s when MALDI came out. And [Richard M.] Caprioli and other people said, you’re kidding yourself. That isn’t going to fly at this point. It was dead. You need to switch over to MALDI. So that’s what they did. (69)

Vestal discussed MALDI’s wider significance in the field:

The ionization techniques [electrospray and MALDI] are what revolutionized mass spectrometry. Mass spectrometry was really plateauing before those were developed. We were doing all these things. . . . But you could see it. You were doing all the molecules you could do. You could try to do biological molecules, but it was tough and not many people were really doing it. . . . Once we got these new ionization techniques and what followed, it sparked a revolution. (122)

Vestal responded by switching his company’s focus to MALDI instrumentation and played a significant role in lowering the cost and improving the efficiency of the instrumentation. He remarked, “We became the dominant MALDI company for quite a few years” (70).

MALDI’s possibilities spread quickly by word of mouth. Frank Field explains how he got his lab at Rockefeller University going with MALDI:

FIELD: Then somewhat after that, then it was 1986 or so, I went to Bordeaux [France] to an international meeting. I heard . . .
GRAYSON: Oh, Franz Hillenkamp?
FIELD: . . . talk about laser desorption mass spectrometer, the matrix-assisted laser desorption, and I came back. I was about to retire, but by this time Brian Chait had been working with me for a number of years, and obviously was going to be my successor. So I gave the information to him, and he worked like a beaver and came up very rapidly with the matrix system . . . MALDI, which is an extremely useful device for biochemical work, and all the whole, we were using these instruments to do analyses and so some research. (38)

Field charged Chait with building an instrument, knowing that moving quickly was key in the field. Chait did so, and his achievement influenced other mass spectrometrists. Researchers like Klaus Biemann used Vestal-made instruments. When Biemann himself was considering moving forward with a MALDI instrument, he had several manufacturers to choose from.

Brian Chait at Rockefeller University had built his own instrument, actually put a MALDI ion source into a time-of-flight instrument which he had. So he had more or less the design of a MALDI time-of-flight instrument, which he had built for his own laboratory, and Marvin Vestal made a deal with him to use his design to make it commercially available. So knowing that I have the money, he called me up and said, “By the way, did you buy your time-of-flight instrument yet?” I said, “No, we are still thinking.” He said, “Why don’t I build one for you?” Since Chait had proven that it works and I knew that Marvin Vestal could build instruments, I said, “Okay.” We ordered it, he built it, and in a few months we got that instrument. (49)

Acceptance of MALDI—and therefore the commercialization of the technique—quickly increased.

MALDI’s role in analyzing biomolecules helped garner a shared Nobel Prize in Chemistry in 2001, and the technology has also helped the field of proteomics develop. From polymers to proteins, MALDI coupled with time-of-flight is the preferred method for ionization of natural molecules.

Tell Us About Yourself

Please take a brief survey to help us plan future online exhibits.

Hear It Firsthand

The Center for Oral History captures and preserves the stories of notable figures in chemistry and related fields, with over 425 oral histories that deal with various aspects of science, of scientists, and of scientific practices. For more information please visit CHF’s Oral History Program or e-mail oralhistory@