The history of the first commercial FTIR for analytical chemistry can best be written in four phases.  The first phase is the concept and its sensitivity advantages.  The best and most authoritative history of this pre-commercial period is in Mertz’ book [1].  Some “historians” include Michelson, either to add class or to impress the reader with scholarship.  Such inclusion is not only incorrect but is misleading. The issue is not the mathematics of the Fourier Transform. If it were, G. Kuiper’s early paper stating that there is no advantage to FTIR would be correct. Kuiper omitted any consideration of the effect of noise in his paper, thereby missing the seminal fact of FTIR. Any field of information can be represented {coded}by any complete orthonormal set of functions. That is, as far as signal is concerned all measurement codes are equivalent. There’s no difference in information between Dirac delta functions [the measurement code of dispersion spectroscopy] and the Fourier Transform [the measurement code of FTIR].  However in a complete analysis where the limiting noise and its cause and nature is included, all sets [measurement codes] are not equivalent.  The selection of the optimum code for the highest S/N is a method for achieving the greatest gains in instrument design and measurement.  FTIR was the measurement code invented by Fellgett/Mertz for detector noise limited measurement .Although Michelson showed that spectral information could be derived from an interferogram, history shows no activity in FTIR until Fellgett /Mertz, the true beginning of the critical path. Although multiplexing is widely understood, to this day there are many reputable spectrosco-pists who do not understand the difference between measurement codes, which can minimize the noise in a measurement and post data acquisition algorithms, which can at best redistribute the existing noise in the measurement. I know this from some of the current objections raised to Kromoscopy, which I recently invented for sample noise limited measurement. It is also illustrated by the use of FTIR outside of its design intent, e.g., in shot noise limited measurement and in the use of FTIR for in vivo or sample noise limited measurement. 

Most of the second phase consists of my adventures in recognizing that certain new features of FTIR produced advantages in trace constituent analysis. My initial interest was the measurement of the spectra of booster plumes of ICBMs from a satellite for ballistic missile defense. This is not too different from the astrophysical applications of interest to Mertz and Fellgett.  For this it was obvious to use rapid scan FTIR as taught by Mertz, whose sole interest was, and still is, astrophysical application. Less obvious was the second milestone on the critical path. That is, the concept of the application of FTIR to analytical chemistry. Here I built on the advantages of Fellgett/Mertz to obtain greater wavelength accuracy [using fringe reference] and photometric accuracy [high throughput instead of programmed slits with their great mechanical noise].  When I got the idea, I ran to tell Mertz. He said, “why use FTIR when the source is under your control?”  Therefore, I said less obvious.  The issue was not a matter of higher sensitivity for the measurement of weak sources, which was of interest to astrophysicists.  The issue was higher precision, which would be of interest in trace analysis. Fortunately, Don Johnson at DuPont understood these less obvious advantages for the measurement of trace constituents and I proceeded with his encouragement and guidance and with financing from DuPont.   

Our relationship with DuPont started in a curious way. We had a contract to put inst-rumentation on the U-2 at the time when its existence was highly classified [prior to the capture of Gary Powers]. Our work was financed by accounts receivable bor-rowing from a bank. The U-2 program was under Clarence Johnson at Lockheed.  For cover they used blank stationery store invoice forms that were rubber stamped with the name “C & J Sales”. The address was that of their mail drop, a Burbank storefront with soaped up windows. As invoice amounts increased, the bank got concerned and sent someone to visit the storefront. After banging on the door and getting no answer he went to the closest payphone to tell the bank that C & J Sales was a fraud.  The bank terminated our financing arrangement promptly.   I had no money.  I called Clarence Johnson for help.  He didn’t appreciate the call.  He said, “You’re a big boy, solve your own problems, your instrument isn’t worth jeopardizing secrecy.”   I asked to speak to his security officer.  My plan was to find some company that had a sufficiently high clearance that I could tell my sad story to and borrow some money.  He read me the very short list of companies with sufficiently high clear-ance.   When he got to DuPont I asked him to contact DuPont’s security officer to arrange for my contact, which was with Sam Lehner, the number two man at DuPont.  Lehner seemed amused and invited me to DuPont.  At that time our instru-ment was the I4.  The moving mirror was glued to the paper cone of a $0.60 minia-ture loudspeaker driven with a saw-toothed wave.  The scan rate was such that an audio interferogram was recorded on quarter inch magnetic tape, which was spliced into a loop and continuously played into a General Radio audio wave analyzer whose tuning knob was replaced with a gear.  The gear was driven by a chain around anoth-er gear, which replaced the tuning knob of a chart recorder.  The paper advance on the chart recorder drove the frequency scan of the wave analyzer.  I demonstrated its performance by showing them a spectrum of the plastic wrapper of a cigarette pack-age.  Sam Lehner said that DuPont wouldn’t lend me any money but they would buy the company.  I said I cherished my independence and he said DuPont would buy a minority [30%] interest.  In a sense, this was the first commercial sale of an FTIR.  We signed the deal in the General Motors boardroom at DuPont, at which time I was told that this was the first minority interest taken by DuPont since General Motors. 

The third phase was a team effort to design and manufacture the first commercial FTIR for analytical chemistry.   I decided to position this first FTIR to compete with PE’s top of the line IR spectrophotometer.  I rationalized this decision to my associ-ates and DuPont with commercial arguments, but it was really done out of mischief, for the joy of beating the best the establishment had to offer.  This history is best writ-ten by members of our development team [e.g., R. Curbello and G. Wyntjes].  Their success is the third milestone.  The general design concept was novel at the time.  On one hand, I saw laboratory instruments growing in the complexity and magnitude of their dedicated electronics, on the other hand, I saw the great growth looming in com-puters.  Therefore, I decided that we should put our effort into the sensing head and buy whatever the state of the art computer would be required.  In recognition of this approach, I named the company in which this effort would be done, Digilab.  In addi-tion to FTIR, Digilab had a sensing head for FT-NMR.  However, the rapid growth of FTIR and difficulties in procuring a magnet doomed our FT-NMR.  The fourth phase was to gain acceptance for this new concept in the market.  This turned out to be more difficult than I had anticipated and I made some false starts.  For example, I despaired of teaching analytical chemists why FTIR was superior, so I bought Sadtler to introduce FTIR as a black box that made the great new spectra in their new library. This was ineffective.  The fourth milestone, crucial to FTIR’s acceptance, was the work of Bill Fateley, an opinion maker who was open to new concepts.  The history of this phase is best written by the pioneers, e.g., Bill Fateley, Don Johnson, and Tom Dunn.  I thought that the efforts of Tomas Hirschfeld, our chief scientist, would give closure to this phase.  What really happened was that we finally gained acceptance of FTIR and the consequent competition, but the concept and guiding principles remain a mystery to all but a precious few.

Some of the difficulty in gaining acceptance was self-inflicted.   Mertz and I shared a disdain for peer review journals. It all began when Mertz’ trail blazing article was rejected for publication in JOSA.  I then purchased monthly ads to publish his article in serial form in JOSA.  Attached are examples of our early ads.  When the Editor, Wallace Brode, belatedly discovered my end run, he refused further publication of our ads.  I appealed to the American Institute of Physics.  After a difficult battle, a compromise was reached.  Brode would publish the ads if encased in a prominent border carrying the consumer warning “advertisement”.  A fine solution for us since our only interest was to get an audience.  Having the warning label only added to the fun.

The next event was even more amusing.  John Strong submitted a paper referencing one of Mertz’ ads.  Brode told Strong that he couldn’t reference an ad and suggested that it be referenced as “private communication”.  Meanwhile John Howard at AFCRL was using our instruments in a classified program.  Brode’s persistent density on FTIR was the impetus for Strong and Howard’s founding the journal Applied Optics.    However by that time Mertz’ book was published and neither the ads nor Applied Optics were needed.  I thought that the final chapter in this story came many years later after the general acceptance of FTIR.  Mary Warga, Brode’s assistant, approached me because JOSA wished to publish a commemorative issue on FTIR, featuring a collection of our ads.  Since ads are protected by copyright, she needed my permission.  The offer of a commemorative issue came when I no longer needed it, and Mertz had his book.  Besides, I had a tittle of bitterness for all the rejections along the way.  I thought my refusal would end my long period of disillusionment with peer review journals with panache. 

However, I was wrong.    I recently invented Kromoscopic analysis [which employed similar guiding principles to FTIR] and Bill Fateley invited me to submit a paper to his journal, Applied Spectroscopy.  By the time I submitted the paper, Fateley was no longer Editor and my paper was, of course, rejected.  This time I didn’t have to use ads, non-peer review magazines had proliferated.  Our seminal articles [except for one paper [2]] have been published in such magazines [3.4].

I would like to attempt to express my lifelong gratitude to my old FTIR associates, beginning in time with Larry Mertz and ending in time with Tomas Hirschfeld.  This team of brilliant individualists was named by Peter Griffiths [5] the “remarkable Block Busters.”   If I were to list each one and his contributions here that list would fill all the pages of this publication.