Where Did Ion Selective Electrodes Come From? The Story of Their

Waters Symposium: Ion Selective Electrodes

waters symposium: ion selective electrodes

Where Did Ion Selective Electrodes Come From? The Story of Their Development and Commercialization Martin S. Frant Orion Research, Inc., 500 Cummings Way, Beverly, MA 01915 In 1967 and 1968, in a small refurbished warehouse in the unfashionable part of Cambridge, new “specific ion electrodes” were being invented, developed, patented, and released for sale at such a rate that insiders talked about the “Electrode of the Month Club”. The need for such devices had existed for a long time, and the technology to use them (i.e., pH meters) had also been widely available. The literature was cluttered with tempting clues, inventions that were ignored, interesting routes that turned out to be dead ends, near misses, and electrodes that sounded wonderful but that no one could reproduce. In the usual portrayal of the progress of technology, one development neatly builds the groundwork for the next, and the historian, looking back, can show a logical progression of concepts and discoveries. This was not the way it was with ion-selective electrodes (ISEs), and even those of us who were deeply involved in it were surprised at the way it unfolded so rapidly and dramatically. The Founding of Orion Research To explore the history of ISEs, we must begin by looking at the formation and history of Orion Research1 (1). In 1961, John Riseman was a 27-year-old electrical engineer who had started a consulting business in the medical instrumentation field, operating out of his home. Using the newest transistors that had just become available, he developed a dual-input pH meter. From his con-

sulting experience he concluded that, to sell his meter to some company, he needed sodium and potassium electrodes to go with it. For sodium, he searched the literature and came upon the work of George Eisenman. Eisenman was a medical researcher who had been studying the transmission of cations through cell membranes and had tried to make sodium and potassium electrodes for this work. In 1957, he published the results of his group’s investigation into the effect of glass composition on selectivity for sodium and potassium ions (2). They had come up with what today is still the best glass composition for sodium, designated as NAS 11-18, 2 as well as a moderately usable potassium glass, NAS 27-4. Neither of the Eisenman glasses was commercially available, and Riseman hired Eisenman as a consultant to what was now called “Riseman Development Laboratories”. Riseman knew very little chemistry, and he went to the MIT Industrial Liaison Program for help in developing electrodes. They referred him to a young electrochemist on the staff, Dr. James Ross. Ross and Eisenman first met in Riseman’s kitchen. Eisenman’s NAS 11-18 sodium glass was high in alumina and was extremely high-melting. There was no way it could be made in Riseman’s kitchen, and they turned to the Corning Glass Research Department for help in preparing test samples. Samples of capillary sodium electrodes made from the Eisenman glass were enthusiastically used by Dr.

The Annual James L. Waters Symposia at Pittcon The objectives of the annual James L. Waters Symposia at Pittcon are different from those of other symposia at either Pittcon or other conferences. Waters, founder of the well-known Waters Associates, Inc., and currently president of Waters Business Systems, Inc., arranged with the Society for Analytical Chemists of Pittsburgh (SACP) in 1989 to offer annual symposia at Pittcon to explore the origin, development, implementation, and commercialization of scientific instrumentation of established and major significance. The main goals were and still are to ensure that the early history of this cooperative process be preserved, to stress the importance of contributions of workers with diverse backgrounds, objectives, and perspectives, and to recognize some of the pioneers and leaders in the field. Important benefits of these symposia are creation of awareness of the way in which important new instruments and, through them, new fields, are created,

and promotion of interchange among inventor, development engineer, entrepreneur, and marketing organization. The topics of the first six Waters Symposia, beginning in 1990, were gas chromatography, atomic absorption spectroscopy, infrared spectroscopy, nuclear magnetic resonance spectroscopy, mass spectrometry, and high-performance liquid chromatography. Publication of the papers presented at the Waters Symposia is a high priority of the SACP. The papers of the first symposium were published in LC.GC Magazine and those of the next four symposia were in Analytical Chemistry. The sixth Waters Symposium on high-performance liquid chromatography was published in the January 1997 issue of this Journal. Ion selective electrodes was the topic for the seventh Waters Symposium, held in March of 1996. Administration of the Symposium, including selection of the topic and speakers, is handled by the SACP. J. F. Coetzee University of Pittsburgh Waters Symposium Coordinator

Vol. 74 No. 2 February 1997 • Journal of Chemical Education

159


Edward Moore (3) at the Boston VA Hospital (Fig. 1). His endorsement set the stage for Corning’s entrance into the story and for the beginning of the search for other ISEs that could be made from glass. (As it would turn out, glass as the basis of a family of ISEs would be a dead end: no other commercially viable glass ISEs have been found since Eisenman’s now-classic work.)

Corning watched the work that Riseman was doing with considerable interest. Moore’s results strongly suggested the possibility of commercial applications for NAS 11-18, and Ross and Riseman approached Corning about the possibility of sponsoring a research program. They introduced Corning to Bob Garrels, a geologist at Harvard, who told them about the success he and A. H. Truesdell were having with glass electrodes for calcium and other divalent ions—work that was to be published shortly (5). It began to appear possible to make glass electrodes for many ions. (There was another factor that may have influenced Corning’s decision. According to legend, at one point in the 1930s, Arnold Beckman had approached Corning about buying his pH patent, so that he could concentrate on his duties at Cal Tech. Corning turned down the offer, apparently deciding that the electrodes used too little glass to make it worthwhile. The story was well known at Corning, and now another opportunity had appeared for getting into a high-value glass-based instrument business.) Instead of the research contract, a much broader arrangement was negotiated. Riseman and Ross formed Orion Research, a private consulting company. Orion would do research under contract to Corning, and Corning would pay all the bills. Orion would not make a profit from the Corning contract. Instead, Orion would get a royalty from the sales of any new product they developed for Corning. Orion was now ready to enter the field of “specific ion electrodes”.3 The company was incorporated in May 1962, and in July, Ross left MIT. There was no question of where the new company would be located.

Figure 2. In 1906, Haber (7) demonstrated that a glass bulb could be used to follow an acid–base titration, and gave and indentical curve to that obtained using a hydrogen gas electrode. His apparatus is shown above without the external shielding that was required.

Figure 3. Front page of Arnold Beckman’s famous patent on a vacuum-tube pH meter. The measurement had to be made inside a built-in compartment because of electrical shielding problems, but it was a portable instrument, which made field measurement of pH practical.

Figure 1. One of the first uses of an ISE in medicine was this report by Moore (3) showing the feasibility of using a “Riseman electrode” (a sodium capillary using Eisenman’s best glass composition) in biological samples.

160

Journal of Chemical Education • Vol. 74 No. 2 February 1997


Cambridge in the early ’60s was an exciting place, and Riseman felt that it had the right kind of atmosphere for a young company. Work began on new glass electrodes. Glass pH Electrodes Actually, the origins of the glass electrode go back almost a century. In 1903, Bruning (6) investigated electrical cells made with such materials as ivory, porcelain, and animal skins. If the solutions were different on the two sides of the membrane, an electrochemical potential resulted. Beginning in 1906, Cremer (7) used glass membranes and found large differences in potential when there was acid on one side and base on the other. In 1909, Haber (8) made glass bulb electrodes (Fig. 2) and showed that they behaved like the hydrogen gas electrode and gave identical titration curves. All of the early glass electrode measurements were made using a quadrant or gold-leaf electrometer, and required elaborate shielding. For “hydrogen ion concentration” measurements (as they were usually called), the hydroquinone electrode (for acid solutions), and colorimetric methods were the most popular. Interest in HIC or pH (we owe the concept and the nomenclature to S. Sorenson [9] ) grew in the intervening years. In the first 10-year index to Chemical Abstracts (1910–1917), 180 column inches were devoted to HIC. In the 1917–1926 index there were 1020 inches, and in the 1927–1936 index there were 3160 inches. (On a relative scale, compared to the total abstracts and using the 1910–1917 index as the base, the ratio was 1:4:7). In the 1930s, Arnold Beckman would vastly simplify the use of glass electrodes. Beckman’s involvement in pH was almost accidental (10). He had started a company to make printing inks when a friend asked for help in measuring the “acidity” of orange juice, which contained sulfites that bleached the usual colorimetric indicators. His use of a vacuum tube voltmeter instead of the usual galvanometer for the null circuit meant that a thicker glass bulb would work, making glass pH electrodes more practical. In his first full year of sales, he sold 87 pH

Figure 4. December 1964 advertisement for the first “all-purpose” pH electrode, which combined chemical durability with low sodium error. The product was the first commercial result of the collaboration between Orion Research and Corning Glass Works.

meters, and the following year sold 444 meters. The era of activity measurements by electrode had begun, and with it a developing faith and trust in electrode measurements. Interestingly, the Beckman patent (11) (Fig. 3) is limited to certain electrical features of the meter, and does not claim the idea of using a high-impedance amplifier with a glass pH electrode. Credit for that idea really belongs to L. W. Elder and W. H. Wright at the University of Illinois, who published “pH Measurement with the Glass Electrode and Vacuum Tube Potentiometer” in a major journal in 1928 (12). The Beckman patent does mention, as common knowledge at that time, that the mechanism of the glass electrode is that “only hydrogen ions can pass through the thin glass bulb…”. This is a simplification of what we know today, but we can now see in retrospect that it was enough to suggest to anyone in the intervening years that the materials which would make other electrodes were ionic conductors through which only one kind of ion could pass. Beyond Glass Orion’s start at investigating new glass electrodes was suddenly halted. A business decision at this point by (of all companies!) Beckman Instruments had a profound effect on the start-up company. Beckman, since its inception, had been selling pH electrodes and meters through laboratory supply dealers such as Fisher Scientific and A. H. Thomas. They decided instead to sell directly, using their own sales force. This left the distributors without a major source of pH electrodes and meters. Corning management decided that if they could move quickly enough, this was a golden opportunity to get into the business. The decision was made in July 1962, and they turned to Orion to develop the new products. By January 1963, Ross, working with Norm Hebert at Corning, had developed a new glass pH composition, making the first “all-purpose” pH glass (before that, glass either had been useful between pH 1 and 12 or had been

Figure 5. A diagram used at the time to explain the construction of liquid membrane electrodes. A porous inert membrane is saturated with a water-insoluble organic solvent containing an organic carrier that is highly selective for a particular ion. The membrane separates an aqueous solution containing a fixed amount of that ion and the sample, which contains a variable amount. The potential that develops across the membrane is measured by a pair of reference electrodes.


161


Table 1. Data Obtained by Beutner in 1912 on KCl Solutions Initial Final Theoretical Guaiacol in Apple as Concentration Concentration Response Oleic Acid Divider (M) (M) (mV) 0.5

0.1

42

28

20

0.1

0.02

42

28

28

0.02

0.004

42

32

34

0.004

126

88

83

Overall: 0.5

Figure 6. In 1912, Loeb and Beutner used an apple as we might use an electrode, and measured KCl solutions. They got a consistent, if less than theoretical, response.

Figure 7. Mimicking the apple, Loeb and Beutner substituted some organic solutions, such as cresol in oleic acid, which yielded a construction electrochemically similar to modern liquid membrane electrodes. They were looking for the origins of electrical potentials in living cells and went no further towards an analytical device.

Figure 8. Before the commercially viable construction in Figure 5 was devised, rapid screening of solvents and ion exchangers (ionophores) was done by using one of the construction techniques shown above.

a high-resistance, special-purpose glass for the range pH 12–14) (Fig. 4). Orion also developed a new transistorbased pH meter (Corning Model 12), improving on the technology Riseman had worked on earlier. In a little over six months, Corning was in the market with innovative new products. Orion turned back to the problem of making glass calcium electrodes. Ross soon made the disheartening discovery that the glasses used in the Garrels and Truesdell electrodes were nonconducting. The observed responses were apparently due to cracks in the glass seal forming a liquid junction, and there was no real selectivity for divalent ions. A few further tries at other glasses were equally disappointing. Ross concluded that, if in the monovalent glasses, hydrogen ions or sodium ions moved through the silica network by hopping from one negative site (Si–O–{ ) to another, the probability of having enough energy at room temperature to get a divalent ion to move from two sites simultaneously was extremely small. It was here that he made a remarkable leap. “If we can’t get the ion to move from the site, why not cut the bonds holding the site, and let the site move, too?” This was the concept of the liquid membrane electrode (13), the true starting point of the ISE era. It did not come from reading the literature, although afterwards, when Orion began to file for patents, we discovered that people had been within a hair’s breadth of the concept.

Ross’s basic configuration is shown in Figure 5, which is from a slide that was used in a talk at the time. Two aqueous solutions are separated by a layer of a waterinsoluble organic solvent. A water-insoluble ionic carrier, which is the site that is free to move, is dissolved in the water-insoluble solvent layer. If there is a difference in concentration (or more exactly, activity) of some species on the two sides, potential will be built up across the membrane by ions that are transported by the carrier. This potential will prevent further net transport, and if we keep one side at a fixed activity, the potential will vary with the log of the activity on the other. In hindsight, if one looks through the old literature, there was all sorts of unusual work that might have provided clues much earlier. Reinhold Beutner was a student of Haber, who continued the biochemical interest in measuring the potentials across animal and vegetable membranes. Loeb and Beutner (14) used an apple as we might use an electrode today (Fig. 6). They went on to try to simulate the apple skin, using a mixture of cresol (or guaiacol) and oleic acid in a setup that is electrochemically equivalent to the “liquid membrane” electrode (Fig. 7). However, we know that guaiacol and cresol have appreciable water solubilities. This is probably why Loeb and Beutner obtained slopes that were only about 70% of theoretical (Table 1); and they went no further. They also observed that without the oleic acid, extremely small slopes were obtained (30 mV in going from 0.1 to 0.0008 M KCl).

162



They certainly were on the right track, but Beutner’s interest was in the analogy with life processes, and he never came closer to making liquid membrane electrodes. Liquid Membrane Electrodes The advantage of the Ross approach was that it gave a basic and reproducible system in which the selectivity depended on the structure of the water-insoluble “carrier” and the nature of the solvent, and it appeared that the limits of the ability to measure any species depended only on the ingenuity of the organic chemist. I joined Orion at this stage and began to explore the effect of structural changes on Ross’s first calcium exchanger and solvent (the calcium salt of 2-ethylhexyl phosphoric acid and dioctyl phenyl phosphonate). He had picked these because phosphate and polyphosphate form stable complexes with calcium, but not with sodium or potassium. We had to learn how to assemble and test the many possible combinations of solvent and carrier in order to quickly screen the new ideas that were being generated almost daily. Three configurations that we used are shown in Figure 8. Interestingly, one idea for a screening technique that we tried and abandoned was to apply to a silver wire a coating of liquid ion exchanger and solvent thickened with Cabosil. We knew very quickly that it would not work unless we could make the wire/ ion exchanger boundary electrochemically reversible. I tried to do this by synthesizing organic silver salts that would be soluble in the liquid membrane phase and would provide a fixed Ag/Ag + interface, but they all proved to be unstable in light.4 At this point, Corning’s success against Beckman in selling glass pH electrodes had unexpected consequences. Corning’s royalty payments to Orion for pH electrodes and meters were becoming substantial and the Corning management, embarrassed, made another effort to buy Orion. For a group used to a totally freewheeling atmosphere and the cultural milieu of the Boston area, Horseheads, NY, and the Corning large-company structure was not a tempting prospect. Instead, Orion offered to give up the royalties if Corning would forget the “future noncompetition” clause in the contract. The two companies separated and became competitors. As part of the separation agreement, Ross and I wrote a lengthy report to Corning Research on “organic electrodes”, telling them everything we knew about ISEs up to that point, including what avenues we thought should be explored next. There was an understanding that the two companies would meet in five years to discuss cross-licensing of patents developed in the interim. This might be viewed as an experiment, in which a little company, with almost no finances, started from the same technical base as a major corporation. There was no meeting in five years, because Corning did not develop any significant patents in the field. Although Corning would market and sell ISEs, those organic devices would not attract the corporate commitment or offer the career paths open to those who worked with glass. “Solid State” Electrodes At the point of separation, Orion had no products and no income. Its directors went to the Foxboro Company, looking for help. Foxboro said they would buy 10% of the company, if Orion could demonstrate their ability to make electrodes by coming up with a hypochlorite or sulfide electrode.

Figure 9. The first sulfide electrodes were made as shown above. This schematic drawing was used for a paper at the Eastern Analytical Conference in 1966, describing the new electrode.

Ross remembered a paper he had once read by Kolthoff and wondered if that approach would work for sulfide, using silver sulfide membranes instead of silver chloride. Kolthoff and Sanders (16) back in 1937 had followed up on a paper by Tendeloo (17), who had reported Nernstian responses to Ba2+ and Ca2+ in cells using slices of BaSO4 and CaF 2. They were never able to duplicate the Tendeloo work; but they did report in JACS that a disk of silver chloride, made by melting AgCl, when placed between two silver solutions, gave a theoretically correct response to silver ion. Thirty years later, Ross would suggest trying a disk of silver sulfide instead of the chloride. It took me a month to learn how to make the membranes (by pressing instead of by melting), and we found that Ross’ hunch was correct: Orion had a sulfide electrode (18) (see Fig. 9). Foxboro came on board in early 1966, and Orion was then able to attract other investors. The first money was used to develop an all-electronic digital pH/ISE meter and to start marketing the calcium electrode (Fig. 10). From sulfide we moved to the halides and learned how to make light-resistant membranes (a problem with the Kolthoff approach) for chloride, bromide, and iodide, and how to glue them permanently into epoxy tubes. At the same time, Ross solved the problem of making a liquid-membrane nitrate electrode, and perchlorate fell out as a bonus. We developed a liquid-membrane copper electrode, but later abandoned it when we found the solid copper electrode. Even at this stage, we couldn’t always move fast enough. When we learned from George Eisenman about Pressman’s work (19) with valinomycin (it increased cell permeability to K+ by a factor of 40,000), Ross immediately had Eisenman get him some, and he tried it in an electrode. But with the earlier publication of the Ross calcium paper, the craft of making liquid membrane electrodes was public knowledge. In Switzerland, Willy Simon, who had experience with making pH electrodes, prepared electrodes for ammonium and potassium using a series of antibiotics, among them valinomycin. His patent application would prove to have the earlier date. Ross and I made the first potassium measurements in serum (20); but Simon would have the key patent (21), on which Orion would have to pay royalties, and he


163


Figure 10. The first advertisement announcing “a new series of chemical sensing electrodes”, which appeared in Analytical Chemistry in June 1966. The advertisement stressed the ability to measure calcium activity, but it was soon found that measuring calcium concentration was of much greater commercial importance.

would have universal recognition as the inventor of a very important electrode. Silver-based membrane materials could never yield a fluoride electrode because silver fluoride is soluble. Therefore in searching for a fluoride electrode I had pressed pellets of various fluorides, using the techniques we had developed for the silver-based ones. I was looking for some measure of conductivity and a response to fluoride ion. Hot-pressed lead fluoride had a very high resistance, but seemed to respond. Unfortunately, it was equally responsive to chloride (because of the formation of PbClF) and had a high limit of detection. This was a hopeful sign, but hardly good enough. Bismuth trifluoride was more insoluble and less likely to be affected by chloride; but it partly decomposed during the pressing, since heat was required. When I saw an advertisement in Analytical Chemistry offering something new, rare-earth single-crystal fluorides for use in lasers, I immediately called to ask for some samples. There was disagreement at Orion as to whether these brittle, lightly colored crystals would work—all the ionic conductors known at that point were soft, black materials. But because screening was so easy, I tried samples of a variety of lanthanum, neodymium, and praseodymium fluorides (Fig. 11). The rest is history (22). For Orion, it was extremely important. The electrode could replace a difficult wet analysis, it had almost no interferences, and it was strong and durable. It got almost immediate approval from a giant among analytical chemists, Jim Lingane at Harvard (23), and his paper convinced many skeptics that ISEs were practical

164

Figure 11. A page from the author’s notebook, with the test results on the first fluoride electrode, which was made from a single crystal of neodymium fluoride sealed into a rigid PVC tube with black sealing wax.

devices. It gave Orion an important source of cash flow, and to this date, almost thirty years later, it is still the largest selling ISE. (Our best guess is that about 300,000 fluoride electrodes have been sold worldwide by Orion and its competitors since that first experiment.) The pace of work at this time may be judged from a look at my notebook. In the two months after the first successful fluoride prototype, I was working on the following projects: determining that conductivity in the fluoride crystals was “probably” due to fluoride vacancies (p-type), evaluating various epoxies as sealants, developing a silanizing technique for the crystals to increase epoxy adhesion, evaluating a number of mixed Pr/ La/Ce crystals to find the effect of composition on performance, trying cold-pressed powdered NdF3 (containing 1% polystyrene as the “glue”) for an easier way to make an electrode, investigating rare earth oxides as possible OH{ electrodes, working on the synthesis of tetra-alkyl quaternary ammonium compounds (as exchangers) and beta-diketones (as solvents), looking for a bicarbonate electrode. The next major new electrodes came from the idea of using double solubility products. If Ag2S responded to sulfide because it changed the silver activity at the membrane interface, perhaps a mixture of nickel sulfide and silver sulfide would respond to nickel ions, since the nickel level would, in turn, determine the free sulfide. There do not appear to be any literature references that described making new electrodes this way, and we certainly were not aware of any. It really sprang from our understanding of the mechanism of the sulfide electrode.



When I first thought of it, I tried to make electrodes for nickel and zinc, but they didn’t work. Discouraged, I described my experiments to Jim Ross. He suggested that I had a kinetic problem and that copper or cadmium would probably work. They did, as did lead, and we had three new electrodes. What we did then was to push the idea to its limits. Our head start lasted only until we released the new electrodes. We tried to make phosphate and sulfate electrodes via silver as a double solubility product. Then we tried using all sorts of ternary systems. Unfortunately, nothing except thiocyanate, as a double solubility product, turned up. As with the coated wires, we never published on the unsuccessful experiments; and as the same idea would occur to others, we watched the literature with concern to see if we had missed something. So far, no commercially viable ternary systems have worked. Growing Pains The burst of creative activity in new electrodes slowed down considerably at this stage. New work was becoming difficult because new ionophores would require extensive organic syntheses, because our attention was being shifted to the need for new analytical techniques for use with electrodes, and because our small company was suffering growing pains. Somehow, one expects that having invented the better mouse trap, all would be well and the story over. However, as the number of Orion ISEs increased, there began a stream of papers on methods using them and a flood of review and popular articles. This created concern among the traditional instrument manufacturers, who feared that they would be left out of a burgeoning new field. Orion tried to expand into all the obvious openings that had been created, but it no longer had the “deep pockets” of Corning Glass—who in fact was now an opponent. It was a formidable task. ISEs were not only laboratory devices but they had important biomedical applications, and Foxboro had shown that there was a process control market, as well. Each required its own electrode construction and its own instrumentation, and each had its own technology. The inability to commit adequate resources to the biomedical program, as a result of all of these conflicting pressures, was a strong factor in the decision of that group to leave Orion and to form Nova Biomedical. The company was still too new to have manufacturing expertise, and production “fires” usually came to the Research Department (see ref 1 for examples). We also found that we had to devote considerable effort to making it easier for analysts to understand and use the new technology. Our attention turned to devising new analytical techniques: Gran’s Plot Paper (a Riseman idea), electrode indicator titrations (24), TISAB (25), SAOB, known increment methods, etc. Instruction manuals had to be extremely detailed and to contain sections on theory and examples of applications. A highly successful newsletter (Fig. 12) was started and widely mailed, and its articles were cited in published papers. Chemists had to be hired, trained in the research lab, and put on the phones to answer customer questions. Papers using ISEs were read, evaluated, and made available in bibliographies. The first Orion pH meter had preceded the sale of the first ISE and was intended to provide a business that would support the ISE research. The Model 801 was an innovation—an all-electronic digital pH meter. Despite this, the meter business would prove to be difficult and

Figure 12. Orion established a very popular newsletter to quickly communicate new developments in a rapidly expanding field.

expensive, since it went head-to-head with the industry giants. Only by providing significant innovations could Orion compete, but this meant constantly developing new instruments as the competitors played catchup. Adding to this was the burden of patents. These are supposed to reward inventors by giving them a period of exclusive sales, and in the end they did work that way. But in the beginning, they were also a considerable drain on Orion’s resources. It was important for Orion to apply for patents as quickly as possible, but the rate at which new inventions were coming in those first critical years meant that the financial stress had to be balanced against future protection. U.S. patent applications were straightforward but every overseas application required separate filings and translations. As a result, the fluoride electrode was patented in only a limited number of countries. During the time between applying for a patent and its issuance, anyone is legally free to copy the invention and some did. Because we were also scientists, we published refereed papers as quickly as other work would allow. Since these told how we made the electrodes, this, unfortunately, also simplified copying. When the patents began to issue, some major companies (e. g., Corning, Beckman, the Coleman division of Perkin–Elmer) were fearful of losing a toehold in an important new business and continued to infringe, knowing Orion’s size and the high cost and slow pace of litigation. Beckman was convinced to stop only when they were sued in Switzerland, where a deliberate violation of a patent was a criminal offense.


165


The End of the Era It was now a different stage in the company’s history. Although a few more important electrodes would be added (cyanide, ammonia, nitrogen oxide, etc.), the technical focus became diffuse, really good new ideas became scarce, and the company became more conventional. Business managers were brought in, and the freewheeling atmosphere ended. In the end, the “Electrode of the Month” era would still be difficult to explain. Orion, as a start-up, had a corporate culture that vigorously supported wild ideas and experimentation; and it also had a few very bright, extremely creative technical people who worked well and easily with each other. They drew very little from the prior art, but developed a good understanding of the mechanisms of ISEs and mastered the important craft of how to build and test them. They had to work quickly, because of the constraints of money and because the head start in time was threatened as the commercial possibilities of ISEs became apparent. Such start-up corporate settings do not last long. The “realities” of the business world and business management soon take over. But they are exciting and fruitful times while they last; and in this case, they left a legacy of a technology that has made life easier for analytical chemists everywhere. Notes 1. This account overlaps the one given in History of the Early Commercialization of Ion-Selective Electrodes ( 1), which should be consulted for additional background material. 2. The name refers to the composition: 11 mol % aluminum, 18 mol % silica, and the balance sodium. It was known (see, for example, Hughes, ref. 3 ) that pH glasses should have a very low aluminum concentration. In effect, what Eisenman did was to make a very bad pH electrode—to have as high a sodium “error” as possible. Even his best glass still has a 10:1 preference for hydrogen ion over sodium ion.

166

3. The term came from the usage in the membrane field at that time: a selective membrane distinguished between cations and anions, a specific membrane distinguished among ions of the same type. Later, a new term, “ion-selective electrodes”, would gain general acceptance, as corresponding to the common meaning of “selective”. Chemical Abstracts has about 150 references to papers using the term “specific ion electrodes”. 4. We abandoned the approach and never published details. Later, after Moody, Oke, and Thomas (15) had suggested making membranes of PVC, there was a series of papers in the literature on the idea of using PVC-coated wires. The question of whether the interface must be stabilized remains somewhat controversial.

Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

Frant, M. S. Analyst 1994, 119, 2293–2301. Eisenman, G.; Rudin, D. O.; Cosby, J. U. Science 1957, 126, 831. Moore, E. W.; Wilson, D. W. J. Clin. Invest. 1963, 42, 295. Hughes, W. S. J. Am. Chem. Soc. 1922, 44, 2860. Garrels, R. M.; Sato, M.; Thompson, M.E.; Truesdell, A. H. Science 1962, 142, 1045; Truesdell, A. H.; Pommer, A. M. Science 1963,142, 1292. Bruning, Z. Physiol. 1903, 17, 622. Cremer, M. Z. Biol. 1906, 47, 562. Haber, F.; Klemensiewicz, Z. Z. Phys. Chem. 1909, 67, 385. Sorenson, S. Biochem. Z. 1909, 21, 131. Stephens, H. Golden Past, Golden Future: The First 50 Years of Beckman Instruments; Claremont Univ. Center: Claremont, CA, 1985. Beckman, A. O.; Fracker, H. E. U.S. Patent 2 058 761, 1936. Elder, L. W., Jr.; Wright, W. H. Proc. Natl. Acad. Sci. U.S.A. 1928, 14, 936. Ross, J. W. Science 1967, 156, 1378. Loeb, J.; Beutner, R. Biochem. Z. 1913, 41, 1. Moody, G. J.; Oke, R. B.; Thomas, J. D. R. Analyst 1970, 95, 910. Kolthoff, I. M.; Sanders, H. L. J. Am. Chem. Soc. 1937, 59, 416. Tendeloo, H. J. C. Rec. Trav. Chim. 1936, 55, 227; J. Biol. Chem. 1936, 113, 333. Frant, M. S.; Ross, J. W. German Patent 1598453, 1971. Pressman, B. C. Proc. Natl. Acad. Sci. U.S.A. 1965, 53, 1076. Frant, M. S.; Ross, J. W. Science 1970, 167, 987. Simon, W. Swiss Patent 479870, 1969. Frant, M. S. U.S. Patent 3 431 182, 1969. Lingane, J. J. Anal. Chem. 1967, 39, 881. Ross, J. W.; Frant, M. S. Anal. Chem. 1969, 41, 1900. Frant, M .S.; Ross, J. W. Anal. Chem. 1968, 40, 1169.


Where Did Ion Selective Electrodes Come From? The Story of Their

Recommend Documents