Catalytic Hydrogenation - NOPR

Indian Journal of Chemical Technology Vol. 12, March 2005, pp. 232-243

Catalytic Hydrogenation Jaime Wisniak* Department of Chemical Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel 84105 Development of catalytic hydrogenation is one of the most significant chapters in theoretical and applied chemistry, which led to the opening of a whole series of new industries, particularly in the petrochemical area. The mechanism for a catalytic reaction involving the presence of an intermediate complex formed by the catalysts and one of the reagents, which eventually led to our present understanding of the phenomenon was suggested by Paul Sabatier. For his achievements in the development of catalytic processes Sabatier was awarded the 1912 Nobel Prize of chemistry, together with Victor Grignard.

Catalytic hydrogenation—a new technique, was contributed to science, by Paul Sabatier (1854-1941). The basic work of P Sabatier in this fundamental scientific and industrial subject forms the basis of our modern theories about catalysis and catalysts, as well as of the processes for the thermal and catalytic cracking of the heavy fractions of petroleum, isomerisation and polymerization of hydrocarbons, hydroforming, synthesis of ammonia, methane, methanol, a very large number of intermediates and fine chemicals, hydrogenation of liquid fats, dye intermediates, and the Fischer-Tropsch process for the manufacture of synthetic fuels. Here, first the work of P Sabatier that led to the discovery of catalytic hydrogenation and the postulation of a mechanism for heterogeneous reactions, is being described and then some details about the life and career of P Sabatier, that will shed light on the road that led him to the Nobel Prize is being given. Inorganic chemistry During his doctoral thesis Sabatier prepared sodium monosulphide (Na2S) anhydrous and hydrosulphides in the pure state (NaSH, NaSH.2H20, NaSH.3H2O); he established the formula of a hydrated potassium hydrosulphide and showed that a number of alkaline polysulphides that had been described as definite chemical species were actually mixtures containing free sulphur. He developed an original method for the preparation of the sulphides of calcium, barium, and strontium in the pure state based on passing a stream of hydrogen over the corresponding carbonates heated to live red (about 500°C). He —————— *E-mail: [email protected]

described for the first time a method for the preparation of aluminum sulphide pure and crystallized, by reacting hydrogen sulphide with alumina heated to the temperature of red in a carbon boat3. After moving to Toulouse he continued his studies of sulphur and sulphides. Carl Wilhelm Scheele (1742-1786) had shown in 1777 that alkaline and alkaline-earth polysulphides treated with a diluted acid did not liberate hydrogen sulphide, like they did with the corresponding sulphides, but generated an oily liquid having an unpleasant smell, from which no compound of definite composition could be separated. Berthollet4 and Louis-Jacques Thénard (17771857)5,6, among others, had tried to determine the composition of this substance that seemed to be composed by a mixture of hydrogen polysulphides, accompanied by hydrogen sulphide and sulphur, but the analysis of these products had proven very difficult because they decomposed easily in contact with many substances, particularly glass. Sabatier solved the problem by distilling the oil under vacuum and isolating a liquid having a composition very close to hydrogen disulphide, H2S2, which he named persulphure d’hydrogène (hydrogen persulphide). The errors in relation to the theoretical composition were due to the presence of a small amount of dissolved sulphur. Sabatier studied the properties of the persulphide, in particular its ability to decompose violently under the action of light or the presence of substances that reacted with it forming unstable combinations (eg, water, alcohols, ethers and alkaline sulphides). In the presence of water it formed a rather unstable form of amorphous sulphur, insoluble in carbon disulphide, while the action of ether led to the formation of crystalline variety of sulphur known as soufre nacre (pearl sulphur) or soufre de Gerne3.

EDUCATOR

Afterwards, he developed a new preparation method for silicon disulphide SiS2, which had been synthetised before by Frémy: Treatment of crystalline silicon heated to red with hydrogen sulphide. He found the concurrent transport of amorphous and crystalline silicon, observation that led him to assume the simultaneous formation of a sublimable subsulphide of silicon, SiS, stable only at high temperature, and decomposing slowly on cooling. In fact, rapid cooling of the vapours generated by the reaction allowed him to isolate this metastable sub-sulphide at room temperature and study its properties3. By reacting hydrogen sulphide with boron at the red temperature he was able to develop a new method of preparation of boron sulphide, B2S3, which he isolated in two amorphous forms, one white and opaque, the other transparent and vitreous, and in a crystalline form obtained by sublimation at 200°C of the white amorphous sulphide. The formation of a volatile boron sulphide was accompanied by that of subsulphide, to which Sabatier assigned the formula B4S, and that of a hydrosulphide of probable composition B(SH)3. He also obtained the sub-sulphide by the action of hydrogen at red temperature on the normal sulphide, and described its properties and its compounds3. He then went on to study selenides, He isolated for the first time a silicon selenide, SiSe2, having a metallic aspect, unstable under the action of water, and a yellow boron selenide, B2Se3, sublimable, and destroyed by water. In addition, he recognized the formation of a sub-selenide of boron, non-volatile3. From 1881 onwards he studied different hydrates of metallic chlorides, determining their heats of hydration, stability, the possibility of their dehydration under cold, and their reaction with cold concentrated hydrogen chloride. In particular, he studied the hydrates of ferric chloride and cupric chloride and the conditions for their formation and dehydration. He showed that the absorption of hydrogen chloride by a solution of cupric chloride decreased the solubility of this salt yielding crystals of hydrated cupric chloride that dissolved under the action of an additional amount of hydrogen chloride, leading to the formation of complex hydrochlorides. Sabatier was one of the first to use spectroscopy of absorption to study hydrates, particularly those of cupric bromide7. Led by the indications of the absorption spectra of solutions of cupric bromide in hydrogen bromide, he succeeded in isolating a complex bromhydrate, which crystal-

233

lized into black crystals. His discoveries in this subject led to the development of the technique for detecting cupric compounds in the presence of hydrogen bromide: they yield a purple coloration, easily observable and having a characteristic wave length in the visible absorption spectra3. In 1896, he observed that the reaction of all copper compounds with a nitro sulphuric solution (nitrosulphuric acid), obtained by dissolving nitric acid in sulphuric acid, yielded an intense blue purple solution due to the reduction of the nitric acid to a new acid, which he named nitrosodisulphonic acid (today: nitrosisulphonic acid). By studying the absorption spectra he established that the colouration was not due to the copper but corresponded to the new acid formed. He also found that this reduction could be obtained with the help of other metallic reducing agents or with organic substances. He then proceeded to the direct synthesis of nitrosisulphonic dark blue, by the reaction between nitric oxide, oxygen, and sulphur dioxide, in the presence of a small amount of water. He proved that nitrosisulphonic acid could produce several metallic salts, particularly a blue cupric salt and a pink ferric salt3. Heterogeneous reactions What makes Sabatier’s discoveries even more sensational is the simplicity of the equipment he built for his studies of heterogeneous reactions: The reactor consisted of a glass tube filled with catalyst and connected to a oxygen generator, a mechanism for adding the reagents, and a receptacle for collecting the reaction products. The hydrogen generator, developed by Deville, operated continuously, and was based on the reaction between diluted hydrogen chloride over granulated zinc. The hydrogen generated was washed by passing it through caustic soda, sulphuric acid, and through tubes filled with copper turnings heated to about 500°C. The reaction tube had a diameter of 1418 mm and length 60 to 100 cm and was positioned within a bed of fine sand to assure constant temperature. The reactor was heated by either a gas burner or an electrical heater8. In 1890 Mond, Langer, and Quincke announced that by the direct action of carbon monoxide on very finely divided nickel, prepared by the reduction from its oxide, they had obtained nickel carbonyl, Ni(CO)4, a volatile compound resulting from the fixation of CO on the metal. They also reported that reduced iron yielded a similar compound9. Their procedure was

234

INDIAN J. CHEM. TECHNOL., MARCH 2005

very simple: It involved passing a current of CO over finely divided metallic nickel at a temperature between 350 and 450°C, yielding CO2 gas and a solid mixture of a black amorphous powder of nickel and carbon. The composition of the powder varied widely with the temperature employed and still more with the time the reaction was carried on. In this manner they obtained a product containing as much as 85 mass percent of carbon and 15% nickel. When a finely divided nickel was obtained, for example, by reducing nickel oxide by hydrogen at about 400°C was allowed to cool in a slow current of carbon monoxide, the gas was readily absorbed as soon as the temperature descended to about 100°C. If the current of CO was continued or was replaced by one of inert gases (such as carbon dioxide, nitrogen, hydrogen, or even air), a mixture of gases was obtained which contained upwards of 30% mass of nickel-carbon oxide. Analysis of the gas indicated that it corresponded to the formula Ni(CO)4. After Berthelot and Mond in 1891, independently, had also succeeded in making iron carbonyl, Fe(CO)510,11, Sabatier and his doctoral student Serendens speculated about the possibility that other unsaturated gaseous molecules such as nitric oxide, nitrous oxide, nitrogen peroxide, acetylene, and ethylene could also be fixed on nickel or on reduced iron, giving well-defined, stable, and volatile products comparable to nickel carbonyl. They first tried unsuccessfully to fix nitric oxide (NO) on nickel, cobalt, iron, and copper; at high temperatures nitric oxide was reduced to nitrogen, with formation of NiO, CoO, FeO, and CuO. Similar results were obtained with N2O, but when passing vapours of NO2 over copper freshly reduced from its oxide, they observed that at room temperature (25 to 30°C) there was a regular fixation leading to a definite compound, solid, black, unstable, and nonvolatile, of nitrated copper, Cu2NO2. In all cases, the experimental technique involved reducing with a current of hydrogen the finely divided metal oxide placed inside a heated glass tube and then passing the unsaturated gas through the tube. Sabatier and Senderens made a detailed study of the properties of the new compound: It was not altered by dry, cold air, but when heated in the presence of pure dry nitrogen, it decomposed at about 90°C releasing NO2, NO, and N2. Carrying the reaction in a Faraday tube (a closed bent tube cooled at one end) it was possible to obtain liquid NO2. Nitrated copper was decomposed violently by water (or humid air) with release of NO;

it was not decomposed by cold hydrogen but on heating to about 180°C, it generated ammonia and ammonium nitrite. Reduced cobalt, reduced nickel, and reduced iron gave similar reactions, but the products were less stable12-15. Sabatier and Senderens decided now to repeat their experiments, this time with ethylene and acetylene. Then, in 1896 they learned that Moissan and Moureu had recently tried the fixation of acetylene on the same metals16. They had passed a current of acetylene on slivers of iron, nickel, or cobalt freshly reduced from their oxides by hydrogen and chilled in this gas, and observed a brilliant incandescence. The high temperature thus produced decomposed the greater part of the acetylene into hydrogen and a large amount of carbon, which would eventually block the tube. The remaining acetylene was converted into liquid hydrocarbons (such as benzene and styrene), which closely resembled those obtained by Berthelot by heating acetylene to dull redness inside a bell inverted over mercury. According to Moissan and Moureu “cette reaction est due à un phénomène physique” (this reactions is due to a physical effect): reduced iron, nickel, or cobalt being extremely porous, absorbed the acetylene with production of enough heat to cause its spontaneous destruction. The reaction being endothermic, incandescence was reached and maintained as long as the acetylene entered. The incandescence also determined the polymerization of the acetylene into liquid products. However, Moissan and Moureu neglected to analyze the free gas, which they judged to consist of hydrogen, and examined the liquids only sufficiently to recognize the presence of benzene. Similar results were obtained with platinum black12. Sabatier made some discreet inquiries whether Moissan and Moureu would be continuing these experiments, and after learning they would not, he and Senderens repeated the experiments but using ethylene instead of acetylene, a hydrocarbon less violent in its reactions. The procedure the followed was similar to the one used before: They directed a current of ethylene upon slivers of reduced nickel and noticed that this time no reaction occurred at room temperature. Raising the temperature progressively, a brilliant incandescence of the metal took place at about 300°C, which disappeared in a voluminous deposit of black carbon, proving the destruction of ethylene. At around 300°C they, too, observed a blockage of the tube containing finely divided nickel, and the production of free hydrogen. The gases released did not react

EDUCATOR

significantly with an aqueous solution of bromine or with an ammonia solution of cupric chloride, which showed that they did not contain measurable amounts of ethylenic or acetylenic hydrocarbons. But they also found methane to be almost pure. Sabatier and Senderens concluded that an unstable combination between nickel and ethylene had formed, analogous to nickel carbonyl, which doubled itself into carbon, methane, and nickel, C2H4 → C + CH4, which could then repeat an identical process17. Pursuing this idea, the two tried the reduction of the finely divided nickel oxide at temperatures below 300°C, cooling the reduced nickel first in a current of hydrogen and then of ethylene. After washing the gases leaving the reactor with bromine to absorb any traces of ethylene, they discovered that is was a mixture of ethane, formène (a mixture of equal volumes of ethane and hydrogen), and hydrogen, with traces of hydrocarbons. Raising the temperature above 325°C decomposed the ethane into methane and carbon and the formène into carbon and free hydrogen. The overall ratio between ethane and hydrogen formed varied with temperature, at 325°C it was 75% by volume ethane and 25% hydrogen and at around 390°C it was essentially pure formene. Hydrogen could only be formed from hydrogenation of ethylene, and this hydrogenation had been provoked by the presence of nickel. In other words, it seemed that reduced nickel had the property of hydrogenating ethylene. To test this possibility they repeated their experiments, this time directing a mixture of equal volumes of ethylene and hydrogen upon a bundle of thin slivers of freshly reduced nickel, slightly heated at temperatures from 30 to 40°C. A considerable increase in temperature was observed. The results confirmed their expectations: only one half of the volume of practically pure ethane was obtained, and the reaction continued indefinitely without the necessity of heating and without an appreciable modification of the metal. At a higher temperature (150 to 180°C) the reaction was still very rapid and a catalyst bed of a few centimeters of metallic slivers was sufficient to accomplish it12. This result, they believed, “doit être certainment attribuée à la formation temporaire d’une combinaison directe et spécifique du nickel et de l’éthylène” (ought certainly to be attributed to the temporary formation of a direct and specific combination of nickel and ethylene)18. Cobalt, iron, copper, and platinum black gave similar but less intense results. It was not only the easy hydrogenation of ethylene and acetylene that was extraordinary in

235

their results, but also the use of metals outside the platinum family for catalysts. This result was completely unexpected19. The following year Sabatier and Senderens discovered that nickel is also capable of hydrogenating acetylene, but that reaction could be initiated at room temperature20. Acetylene was first hydrogenated to ethylene and then to ethane, depending on if the reacting mixtures contained the same or double the volume of hydrogen. The reaction was very exothermic and the temperature within the reactor could reach 100 to 150°C, depending on the length of the tube. Again, reduced cobalt, iron, and copper, as well as finely divided platinum sponge or black, slightly heated, led to an analogous but less energetic reaction. With freshly reduced copper, for example, the composition of the exiting gas varied with the temperature; at about 130°C, it contained 11% volume of ethylene and 178% ethane, and at 150°C, 331% ethylene, 20% ethane, and 184% of higher unsaturated hydrocarbons21. Further work examined in more detail the hydrogenation of ethylene and acetylene using other metals22,23. After having studied in depth the hydrogenation of unsaturated hydrocarbons Sabatier and Senderens turned to the next challenging problem: Hydrogenation of benzene. This reaction had been attempted by Berthelot using his universal agent of hydrogenation, a concentrated solution of hydrogen iodide in a sealed tube heated to 250°C (under these conditions hydrogen iodide decomposes into hydrogen and iodine), but instead of cyclohexane, which boils at 81°C, he had only obtained its isomer, methylcyclopentane, which boils at 69°C. Instead, Sabatier and Senderens tested the possibility of using reduced nickel and excess hydrogen at 200°C. The gaseous mixture issuing from the reactor was sent to a U-tube surrounded by ice, within which the vapours of cyclohexane were expected to condense to a liquid product. After boiling the benzene for a rather short time, they noticed that the tube became clogged by colourless crystals, which they assumed to be benzene, solidifying at 4°C, whereas cyclohexane was reported in the literature to crystallize at –11°C. On opening the U-tube they detected, instead of the odour of the original benzene, the special intermediate odour between that of chloroform and that of rose, which belongs to cyclohexane: “It was from cyclohexane obtained practically pure at the first attempt, the fusion of which is in reality 65°C…that hour was one of the greatest joy in my

236


life”. The transformation of benzene had been complete14,24. In the following three years, Sabatier and Senderens hydrogenated unsaturated ethylenic or acetylenic carbides into saturated carbides, nitro compounds and nitriles into amines, aldehydes and ketones into the corresponding alcohols, unsaturated hydrocarbons of cyclic nuclei, homologous with benzene (polyphenyls, naphthalene, anthracene, etc) into the corresponding saturated hydrocarbons, the phenols into the cyclohexanic alcohols, and aniline into cylohexylamine. They also found that they could produce the major types of natural petroleum by modifying conditions of the hydrogenation of ethylene1,14. They also realized the synthesis of methane by hydrogenating carbon dioxide or carbon monoxide25,26. With Mailhe, another doctoral student, Sabatier found that some metal oxides were catalysts not for hydrogenation and dehydrogenation, but instead for hydration and dehydration. While ordinary alcohols directed at 250°C on reduced copper split into hydrogen and aldehyde, the same vapour directed on fine alumina or thoria split into water and aldehyde27,28. They also observed that amorphous oxides were more active catalysts for dehydrogenation or dehydration than the crystalline oxides ones28,29. Calcination of the latter at temperatures higher than 500°C led to a notable agglomeration. The calcinations reduced the active surface by modifying the nature and distribution of the active centers. This was the first example of the sintering effects, which were later used to graduate the activity of catalysts. Catalysis Catalysis is a phenomenon known from very ancient times, although not its theory or characteristics. By the nineteenth century, enough experimental information began to accumulate to call the attention of scientists. In 1811, Sigismund Constantin Kirchhoff (1764-1833) discovered that mineral acids upon heating changed starch into dextrin and sugar, without themselves being modified by the reaction30. In 1833 Anselme Payen (1795-1871) and Jean-François Persoz (1805-1868)31 found that the transformation of starch discovered by Kirchhoff was attributable to the action of a special substance, which they called diastase (amylase), which can be extracted from germinated barley by water and purified by repeated precipitation with ethyl alcohol; they had also found that the activity is eliminated by heating to 100°C. Payen

and Persoz studied in detail the transformation of starch into dextrin and then into sugar by the action of diastase and proved how starch, once rendered soluble, went from one tissue to another, as much as to accumulate again, as much as to bind in strong aggregation and participate in this form in the formation of cellular membranes in the tissue. In 1823, Johann Wolfgang Döbereiner (1780-1849) obtained a spongy platinum material by calcinating ammonium chloroplatinate32. This material was shown to be able to absorb hydrogen at room conditions and on heating up to ignite a stream of air directed to it. Water was formed as a result without the spongy mass changing its aspect or its weight, and being capable of repeating the process after cooling. In those days, when there was no simple way to produce fire, Döbereiner’s discovery led immediately to its application, the hydroplatinic lamp, also called briquet à hydrogene (hydrogen lighter). Platinum did not behave in this manner only when made as a sponge; it did also when finely divided as filings, wire, or turnings, as long as it was first heated slightly. Many experiments led to think that this activity increased the more the platinum was divided; it even increased more if before calcination the aqueous solution of chloroplatinate was boiled with a little of sodium carbonate and sugar. The chloroplatinate was completely decomposed and the metal precipitated as a black powder. This powder was much more active than the sponge; it absorbed hydrogen rapidly and the smallest particle led to the instantaneous ignition of a mixture of hydrogen and air. Thénard then found that black platinum decomposed hydrogen peroxide rapidly and with violence into oxygen and water, without absorbing the gas releasing or losing its activity33. By 1900 this property of platinum was found also in other metals (such as copper and iron), metallic oxides (such as manganese dioxide), carbon, certain acids, etc. The only difference with platinum was that these additional materials had to be heated, sometimes to red temperature34. In his 1836 Annual Survey35, Jöns Jacob Berzelius (1779-1848) summarized the findings of different scientists on the formation of ether from alcohols; on the enhanced conversion of starch to sugar by acids; the hastening of gas combustion by platinum, the stability of hydrogen peroxide in acid solution but its decomposition in the presence of alkali and such metals as manganese, silver, platinum, and gold; and the observation that the oxidation of alcohol to acetic

EDUCATOR

acid was accomplished in the presence of finely divided platinum. In a brilliant stroke, he was able to understand that all these processes, although seemingly different, had a common denominator, which he called catalysis (either catalysis of inorganic reactions by metals or of biological reactions by enzymes). In the Annual Survey he wrote: ”In inorganic nature when compounds arise through the interaction of several substances, the available combining units strive for a state of better satisfaction. Thus, the substances endowed with strong affinities combine readily on the one hand, while those weakly endowed form combinations among themselves on the other. The agent causing the conversion of substances does not participate in the new compounds formed but remains unchanged, thus operating by means of an internal power, the nature of which is still unknown, although it was in this way that it revealed its existence. Thus, it is certain that substances, both simple and compounds, in solid form as well as in solution, have the property of exerting an effect on compound bodies which is quite different from ordinary affinity in that they promote the conversion without necessarily participating in the process. This is a new power to produce chemical activity belonging to both inorganic and organic natures. It will also make it easier for us to refer to it if it possesses a name of its own I shall call it the catalytic power of substances, and decomposition by means of this power catalysis“(καταλγω from the Greek kata-, "down" and lyein "loosen”). Mechanism of catalysis

When Sabatier commenced his investigations on catalysis there were two theories of heterogeneous catalysis, a physical and a chemical one. In 1833, after studying the data on the catalytic combination of hydrogen and oxygen on the surface of platinum, Faraday suggested a physical theory in which one or more of the reacting gases were condensed by attraction on the surface of the metal. He wrote: “The course of events when platinum acts upon and combines oxygen and hydrogen may be stated according to these principles as follows. From the influences of the circumstances mentioned, ie, the deficiency of elastic power and the attraction of the metal for the gases, the latter, when they are in association with the former, are so condensed as to be brought within the action of their mutual affinities at the existing temperature, the deficiency of their elastic power not only subjecting them more closely to the attractive influ-

237

ence of the metal, but also bringing them into more favourable states for union by abstracting a part of that power (upon which depends their elasticity) which elsewhere in the mass is opposing their combination. The consequence of their combination is the production of the vapour of water and an elevation of temperature. But as the attraction of the platina for the water formed is not greater than for the gases, if so great (for the metal is scarcely hygrometric), the vapour is quickly diffused through the remaining gases. The platina is not considered as causing combination of any particles with itself but only associating them closely around it and the compressed gases are as free to move from the platina being replaced by other particles as a portion of dense air upon the surface of the globe or at the bottom of a deep mine is free to move by the slightest impulse into the upper and rarer parts of the atmosphere”. Berzelius argued that the catalytic force acted on the polarity of atoms through some phenomenon of temperature elevation. The physical theory was supported by the work of Jacques Duclaux (1877-1978) and Moissan on the absorption of gases by finely divided metals36,37. Wilhelm Ostwald (1853-1932; 1909 Nobel Prize for Chemistry) and others had also assumed that catalyzed gas reactions resulted from the absorption of gases in the cavities of the porous metal, where compression and local temperature elevation led to chemical combination. Ostwald believed that a catalyst did not induce a reaction but rather accelerated it but not with formation of intermediate compounds. He argued it was necessary to prove that the succession of assumed reactions required less time than the direct reaction itself. William Charles Henry (1774-1836) in 182438 and August de la Rive (1801-1873) in 182839 proposed a chemical theory where intermediate compounds, for example, oxides of metals, were formed and decomposed. Sabatier did not accept this purely physical view of the function of the catalyst, remarking that if it was true then charcoal should be almost a universal catalyst, whereas it proved to be somewhat mediocre except for the formation of carbonyl chloride40. While finely divided metals were able to absorb substantial quantities of gas, these absorptions were somewhat specific, being “characterized by a sort of selective affinity”. Not only that, some catalytic reactions were extremely specific, for example, zinc oxide decom-

238


posed formic acid into hydrogen and carbon dioxide, but at the same temperature titania gave carbon monoxide and water. The idea that absorption of gases facilitated their liquefaction could not be true since the easily absorbable hydrogen was very difficult to liquefy. The physical theory was unable to explain the development of high local pressure and temperature in the cases where the catalyst was held in suspension, and did not account for the specificity of catalysts and the remarkable diversity of effects they produced, depending on the particular metal or oxide used1,40. Sabatier and his students observed that an extended surface was not necessarily a synonym of catalytic activity. For example, colloidal nickel, contrary to palladium and nickel, was essentially inert. The activity of the catalyst depended particularly of the surface structure; this was the reason why it was preferable to use a chemical method of preparation instead of mechanical ones. Sabatier demonstrated that the manner by which the reducible oxide was prepared played a critical role, and searched which were the best derivatives to prepare the reduced metal. He also showed that these metallic catalysts were extremely fragile and that they became stronger when using high reaction temperatures. Higher temperatures led to an irreversible decrease of the catalytic activity; the activity was not recovered if the operating temperature was lowered8. Sabatier then formulated a chemical theory of catalysis involving the formation of unstable chemical compounds as intermediate stages, which determined the direction and rate of the reaction. He assumed that in hydrogenation various nickel hydrides were involved, whose composition depended on the activity of the nickel. Carefully prepared nickel resulted in the very active NiH2, which would operate on benzene, while impure nickel or nickel prepared at too high temperature gave an impoverished hydride, Ni2H2, which is inactive with benzene, but works with the ethylenic carbides or nitrate derivatives. Sabatier argued that the formation and decomposition of intermediate compounds usually corresponds to a diminution of the Gibbs energy of the system. This reduction is accomplished in several steps and this process is frequently much easier than an immediate and direct decrease of Gibbs energy, just as the use of a staircase facilitates a descent1. He argued that while the presence of catalyst might indeed lower the temperature required by a reaction, the catalyst’s greatest asset was in reacting with a molecular gas in order to pro-

vide a free ion for a reaction which simply would not occur otherwise. Thus, hydrogen peroxide solutions decompose relatively slowly in the cold the same as solutions of chromic acid do, but when the two solutions are mixed there is a rapid decomposition with brisk evolution of oxygen and appearance of an intense blue colouration (reciprocal catalysis). The colour is due to the unstable combination 3H2O2.2H2CrO4, which can be isolated by shaking with ether and evaporating the ether1. Sabatier also addressed himself to the problem of orientation and specificity of the catalyst and claimed that they strongly supported the chemical theory. He showed that different contact masses produced different reactions: At 350°C amyl alcohol in the presence of copper yielded valeric aldehyde and hydrogen; at the same temperature but in the presence of some thoria as catalyst it yielded amylene and water. Chromic oxide acted both in oxidation and in dehydrogenation and dehydration reactions and heated alumina decomposed alcohol into ethylene and steam, while metallic molybdenum and zinc oxide decomposed it into acetaldehyde and hydrogen. He attributed the orientation of the reaction to the metallic or nonmetallic character of the catalyst or, in other words, to the intervention of different electrical contact forces. He understood that this differentiation was not valid in every case. He wrote: “J’ai trouvé avec M Mailhe que les vapeurs d’acide formique en présence d’oxyde de zinc à 250°C, donnent de l’anhydride carbonique et de l’hydrogène, en presence d’oxyde titanique elles se détruisent exclusivement en eau et oxyde de carbone Ici les deux oxides n’ont aucune dissemblance physique et l’intervention d’affinité chimique spéciale s’exerçant à la surface de ces catalyseurs est seule capabe d’expliquer une inversion aussi complete du phénomène…” (I have found with M Mailhe that vapours of formic acid in the presence of zinc oxide at 250°C yield carbon dioxide and hydrogen, in the presence of titania they decompose only into water and carbon monoxide. Here, the two oxides do not have any physical resemblance and the intervention of a special chemical affinity that is exerted at the surface of the catalyst is capable of explaining such a complete inversion of the phenomenon)8. Sabatier postulated the formation of different intermediate compounds, each with its own mode of decomposition, and he also clearly established that some organic reactions are reversible. In cases where the intermediate compounds could not be isolated, he

EDUCATOR

assumed the formation of surface compounds, a phenomenon, which he named fixation, thus linking the physical, and the chemical theories of catalysis41. For example, in the catalytic oxidation of organic compounds with the aid or copper, or the decomposition of carbon monoxide on nickel, the intermediaries, copper oxide and nickel carbonyl, respectively, can be isolated and identified. On his studies of the hydrogenation of ethylene, Sabatier suggested the fixation of hydrogen by the nickel as the intermediary complex. In the case of acetylene, however, he remarked that it was adsorbed more energetically than hydrogen, thus indicating the possibility of organometallic compounds playing a role in heterogeneous catalysis42. During his acceptance speech at Stockholm Sabatier said43: “J’admets que l’hydrogène agit sur le métal en donnant très rapidement sur sa surface une combinaison L’hydrure ainsi engendré est facilement et rapidement dissociable, et s’il est mis en présence de matières capables d’utiliser de l’hydrogène, il le leur cede, en régénérant le métal, qui recommence indéfiniment le même effect La distinction que j’ai faite entre plusieurs sortes de’activité du metal conduirait à admettre qu’il existe plusieurs stades de combinaison” (I admit that hydrogen acting on the metal produces rapidly on its surface a combination. The hydride thus generated is easily and rapidly dissociated; put in contact with substances capable of using hydrogen, gives it to them, regenerating the metal and restarting indefinitely the process)8. Sabatier summarized his views in respect to the mechanism of catalytic action as follows1,40: “As far as I am concerned, this idea of temporary unstable intermediate compounds has been the beacon light that has guided all my works on catalysis; its light may perhaps be dimmed by the glare of light as yet unsuspected which will rise in the better explored fields of chemical knowledge. Actually, such as it is, in spite of its imperfections and gaps, the theory appears to us good because it is fertile and permits us in a useful way to foresee reactions”. We must note that Sabatier was the first to demonstrate, the reversibility of the reaction alcohol⇔aldehyde−hydrogen, which gave place to a large number of thermodynamic and statistical investigations on the Gibbs energy changes of organic reactions. During the Second World War Irving Langmuir’s (1881-1957; 1952 Nobel Prize for Chemistry) published a rival theory called “theory of chemisorption”,

239

according to which a gas adsorbed over a catalyst was fixed thanks to its unsaturated valences yielding a gas-metal compound of the type MxGy, where x is a function of the mass of the catalyst and y varies according to the specific surface and physical conditions of pressure and temperature44,45. This theory was contrary to Sabatier’s hypothesis of distinct, individual intermediates and allowed more importance to physical conditions19. Although Langmuir’s theory retained the concept of fixation of the reagents on the surface of the catalysts, it assigned a predominant importance to the physical conditions that Sabatier had purposedly ignored. For a time, Langmuir’s theory of the fixation of a monomolecular layer on the catalyst gave it a certain advantage because it permitted to address quantitatively the problem of heterogeneous catalysis and played a considerable role in experimental studies8. Langmuir’s theory became particularly important because it permits a first quantitative analysis of the possible mechanism of a reaction. Poisoning

The catalysts sensitivity to poisons, discovered by the work of Rudolph Knietsh (1854-1906) on catalysts used for the synthesis of sulphur trioxide (was shown to be a general phenomenon, which could be used to control catalytic reactions Sabatier, when comparing catalysts to ferments, described poisoning in the following words: “Comme les ferments organisés qui sont tués par des doses infinitésimales de certains toxiques, le ferment minéral qu’est le métal est tué par des traces de chlore, de brome, d’iode, de soufre, d’arsenic, soit qu’elles lui viennent par l’hydrogène soit qu’elles lui soient apportées par la substance qui doit subir l’hydrogenation…le nickel un peu intoxiqué ne pourrait fournir qu’un premier hydrure, comparable à celui du cuivre, et capable d’agir sur les groupes nitrés ou sur la double liason éthylénique; seul le nickel sain pourrait fournir un hydrure capable d’hydrogéner le noyau aromatique” (In the same way that organized ferments are killed by infinitesimal amounts of certain toxins, the mineral ferment which is the metal is killed by traces of chlorine, bromine, iodine, sulphur, or arsenic that are carried by the hydrogen or by the substance to be hydrogenated…nickel, a little poisoned, can only provide a first hydride, similar to that of copper and capable of acting over the nitro groups or on the ethylenic double bond, only nickel cannot provide a hydride able to completely hydrogenate the aromatic ring)1,8. Sabatier and Espil46 made the interesting discovery

240


Paul Sabatier (1854-1941)

that after being used to hydrogenate nitrobenzene, nickel poisoned by chlorine recovers its ability to hydrogenate benzene1. Discussion on the life and scientific career of Sabatier, in the following paras will give an overall view of the background that led to this major contribution. Life and career Paul Sabatier (Fig. 1) was born on November 5, 1854, at Carcassonne, Southern France, one of the seven children of Alexis Sabatier and Pauline Guilhem. He received his first education at the primary school and the local Lycée in Carcassonne. In 1868, he entered the Lycée at Toulouse and a year later joined the Collège Saint-Marie. In June 1872, after receiving his diplomas of bachelier ès sciences (bachelor of sciences) and bachelier ès letter (bachelor of humanities), he departed for Paris to prepare for the entrance examinations to the Grand Écoles (École Polytechnique and École Normale Supérieure). Although he placed highly in the entrance competitions for both Ècoles, he chose the latter, where his brother Théodore has preceded him19,47. At the École Normale he took the courses given by Henry Sainte-Claire Deville (1818-1881), Charles Friedel (1832-1899), Jean Gaston Darboux (1842-1917), and Pierre Auguste Bertin (1818-1884)8.

Three years later (1877), Sabatier received the license de physique and the license de mathématiques at the École Normale and joined the local school at Nîmes as professor of physics. In 1878, he was recommended as normalien to the laboratory of Marcelin Berthelot (1827-1907), then at the Collége de France. At that time, Berthelot was the most outstanding chemist of France and was well connected to the high ranks of the academic and political establishment. Sabatier took advantage of his stay at the Collége de France to successfully prepare for his doctorate, receiving the degree of Docteur ès Sciences (Doctor of Sciences) in 1880 with a thesis on the thermochemistry of sulphur and metallic sulphides48. The fact that Sabatier had been raised at home on a religious and conservative atmosphere soon led to strong divergences of opinion with Berthelot on political and philosophical grounds. This inflexible position explains why Berthelot forced Sabatier to use in his thesis the notation of equivalents (as seen in all the tables), instead of the atomic one. Nevertheless, at various points in his thesis Sabatier mentions that his experiments confirm Berthelot’s principles or predictions8,1. These divergences affected profoundly Sabatier’s career In 1880, after receiving his doctorate Berthelot led him to understand that he should look for a position in the provinces. In France of that time this statement was equivalent to academic exile. Berthelot was known to punish in this harsh manner those who criticized his theories and opposed his ideas. One of the most famous scientists thus penalized was Pierre Maurice Duhem (1861-1916)49. His choices to start an academic career were now limited to three universities: Algiers, Bordeaux, or Lyon. Berthelot recommended Sabatier for the position at Bordeaux, where he stayed for one year as Maître de conferences in physics. Eventually Sabatier’s personality and brilliant scientific achievements overcame the ideological barrier and on December 1905 he became Doyen (dean) of the Faculty of Sciences at Toulouse, a position he occupied for the next twenty-five years. Sabatier remained faithful to Toulouse for all his life, turning down many offers of respectful positions elsewhere. In 1907, he was offered Henri Moissan’s (1852-1907; 1906 Nobel Prize for Chemistry) chair at the Sorbonne and that of Berthelot at the Collège de France (ironies of life!). Although he realized that all candidates for the Académie des Sciences were required to be residents of Paris, he nevertheless chose

EDUCATOR

to remain at Toulouse. He retired from his professorship in 1930 after nearly fifty years of uninterrupted service in the Faculty of Science. In 1913, he became the first scientist elected to one of six chairs newly created by the Académie for provincial members. Sabatier’s initial researches were in the field of inorganic and physical chemistry, within the framework of Berthelot’s laboratory. His thesis, which sought to complete the study of sulphides, included accounts of the preparation of various metal and nonmetal, especially alkali and alkaline-earth sulphides and polysulphides, with determinations of their heat of reaction and solution. As described below, Sabatier approached the problem from a variety of possibilities. In the 1879-1897 period, he performed analyses of metallic and alkaline earth sulphides, chlorides, chromates, and notably of copper compounds7, the preparation of hydrogen disulphide by vacuum distillation, the isolation of selenides of boron and silicon, the definition of basic cupric salts containing four copper atoms, and preparation of the deep blue nitrosodisulphonic acid and the basic mixed argentocupric salts. He studied the partition of a base between two acids, using the spectrophotometric change of colouration of chromates and dichromates, as an indicator of acidity50,51, and analyzed the velocity of transformation of metaphosphoric acid52. From 1896 to 1899, he made some in-depth studies of the oxides of nitrogen12,53-55 and of nitrosodisulphonic acid and its salts3,55-58. Sabatier himself has described how his interest in the field of catalysis arose12. In 1890, after learning that Ludwig Mond (1839-1909), Carl Langer, and Friedrich Quincke had succeeded in preparing nickel carbonyl13, a volatile compound, by the direct action of carbon monoxide on finely divided nickel he decided to investigate the possibility that other “incomplete” (unsaturated) gaseous molecules would behave in the same manner, giving well-defined, stable, and volatile products comparable to nickel carbonyl. In collaboration with his doctoral student Jean Baptiste Senderens (1856-1937) they succeeded in preparing nitrated copper, Cu2NO2, by reacting NO2 with reduced copper at room temperature15. Sabatier and Senderens were about to try to fix acetylene on several metals (copper, cobalt, iron, and nickel) when they learned that Moissan and François Charles Léon Moureu (1863-1929) had failed to achieve it16. Sabatier and Senderens tried instead the

241

less violent reaction of ethylene and reduced nickel and found that the reaction product was mainly ethane with a little of hydrogen. Discovery of this hydrogenation reaction opened a new world of chemical synthesis, which was intensively researched by Sabatier and his students. He promptly demonstrated the general applicability of his method to the hydrogenation of unsaturated and aromatic carbides, ketones, aldehydes, phenols, nitriles, nitrate derivatives, carbon monoxide and carbon dioxide, and liquid fats. He also showed that certain metallic oxides, particularly manganous oxide, behave analogously to metals in hydrogenation and dehydrogenation, although at slower rates; and that powdered oxides such as thoria, alumina, and silica possess hydration and dehydration properties. His work revealed also the general increase in catalytic activity arising from the dispersal of the active material on suitable supports. Sabatier’s discoveries lay at the root of most of the giant chemical industries of today (petroleum, petrochemicals, chemical synthesis, synthetic fuels, fat hydrogenation, etc), nevertheless, like Claude-Louis Berthollet (1748-1822) and Michael Faraday (17911867) before him, he pursued science only and did not seek the commercial benefits of his inventions, patenting very few of them. Sabatier passed away in Toulouse on August 1941, at the age of 87. The scientific work of Sabatier was very extensive; in addition to a large number of speeches, reports, and eight patents, he published over 250 scientific memoirs, his famous book La Catalyse en Chimie Organique 1 (1913), as well as Leçons Élémentaires de Chimie Agricole2 (1890), and collaborated in major works such as Edmond Frémy’s (1814-1894) L’Encyclopédie. Charles-Adolph Würtz’s (18171884) Dictionnaire de Chimie, and Moissans’s book Chimie Minérale. La Catalyse en Chimie Organique was first published in 1913; this book on catalysis marks a milestone in the evolution of modern chemistry and still continues to be quoted extensively. Honors and Positions Sabatier received many honours for his contribution to science, industry, and the Nation. He became Correspondent Member of the Académie and was nominated to the Légion d'Honneur. He was awarded the degree of Doctor of Science, Honoris Causa, by several foreign universities and elected member or

242


honorary member of many foreign scientific societies. In 1897, the Académie des Sciences awarded him the Lacaze prize and in 1905 the Jecker prize, jointly with Senderens. The Royal Society of London awarded him the Davy Medal (1915) and the Royal Medal (1918), the Royal Society of Arts the Albert Medal (1926), and the Franklin Institute the Franklin Medal (1933). The maximum award came in 1912 when Sabatier was awarded the Nobel Prize of Chemistry for his method of hydrogenating organic compounds in the presence of finely divided metals. He shared the prize with Victor Grignard (1871-1935), who received it on account of his discovery of the so-called Grignard reagent. Interesting enough, the road of both Sabatier and Grignard to the Nobel Prize of Chemistry was determined by a twist of destiny. As mentioned before, Sabatier went to become a chemist and not an engineer because his acceptance to the École Polytechnique was arrived several days after that to the École Normale Supérieure. In 1887, Grignard graduated with honours the lycée at Cherbourg. At that time the city of Paris offered scholarships to brilliant graduates from the secondary schools in the provinces, to prepare for the entrance examinations to one of her universities. The Cherbourg lycée had received a promise that Grignard would be awarded one of those scholarships in order to prepare for the entrance examinations to the École Normale Supérieure in Paris, to study mathematics. Unfortunately, because of the expenses involved in the preparation of the 1889 World Exposition (that would see the inauguration of the Eiffel Tower) no scholarships were offered at the time of Grignard’s graduation from high school. Whoever took the decisions that affected the careers of Sabatier and Grignard, could have hardly guessed the tremendous impact they would have in the development of modern organic chemistry. Epilogue Sabatier ended his speech at the Nobel Prize award ceremony saying43: “Theories cannot be indestructible. They are only the plough, which the ploughman uses to draw his furrow and which he has every right to discard for another one, of improved design, after the harvest. To be this ploughman, to see my labours result in the furtherance of scientific progress, was the height of my ambition, and now the Swedish Academy of Sciences has come, at this harvest, to add the

most brilliant of crowns”. References 1 Sabatier P, La Catalyse en Chimie Organique, C, Béranger, Paris,1913. 2 Sabatier P, Leçons Élémentaires de Chimie Agricole, Mason, Paris, 1890. 3 Champetier G, Bull Soc Chim Fr, 3 (1955) 469. 4 Berthollet C L, Ann Chim, 25 (1798) 233. 5 Thenard L J, Ann Chim, 85 (1812) 132. 6 Thenard L J, Ann Chim, 93 (1831) 79. 7 Sabatier P, Compt Rendus, 106 (1888) 1724; 107 (1888) 40. 8 Wojtkowiak B, Paul Sabatier Un Chimiste Ind?pendant (1854-1941), Jonas Editeur, Argueil, 1989. 9 Mond L, Langer C & Quincke F, J Chem Soc, 57 (1890) 749. 10 Mond L & Quincke F, J Chem Soc, 57 (1890) 604. 11 Berthelot M, Compt Rendus, 112 (1891) 1343. 12 Sabatier P & Senderens J B, Compt Rendus, 114 (1892) 1429. 13 Mond L, Langer C & Quincke F, J Chem Soc, 57 (1890) 749. 14 Sabatier P, Ind Eng Chem, 18, (1926) 1005. 15 Sabatier P & Senderens, J B, Bull Soc Chim Fr, 9 (1893) 669. 16 Moissan H & Moureu, Compt Rendus, 122 (1896) 1240. 17 Sabatier P & Senderens J B, Compt Rendus, 124 (1897) 616. 18 Sabatier P & Senderens J B, Compt Rendus, 124 (1897) 1358. 19 Nye M J, Isis, 68 (1977) 375. 20 Sabatier P & Senderens J B, Compt Rendus, 128 (1899) 1173. 21 Sabatier P & Senderens J B, Compt Rendus, 130 (1900) 1559. 22 Sabatier P & Senderens J B, Compt Rendus, 130 (1900) 1761. 23 Sabatier P & Senderens J B, Compt Rendus, 131 (1900) 40. 24 Sabatier P & Senderens J B, Compt Rendus, 132 (1901) 210. 25 Sabatier P & Senderens J B, Compt Rendus, 134 (1902) 514. 26 Sabatier P & Senderens J B, Compt Rendus, 134 (1902) 689. 27 Sabatier P & Mailhe A, Ann Chim Phys, 20 (1910) 289. 28 Sabatier P & Mailhe A, Compt Rendus, 150 (1911) 823. 29 Sabatier P & Mailhe A, Compt Rendus, 152 (1911) 1212. 30 Kirchhoff S C, Bull Neusten Wiss Naturwiss, 10 (1811) 88. 31 Payen A & Persoz J F, Ann Chim, 53 (1933) 73. 32 Döbereiner J W, Edinburgh Phil J, 10 (1824), 153. 33 Thenard L J, Ann Chim, 9 (1818-1819) 441. 34 Bertrand G, Bull Chim Fr, 473 (1955) 475. 35 Berzelius J J, Jahresbericht 1836, 15, 237, 243. 36 Moissan H, Traité de Chimie Minérale, Masson, Paris, 1904. 37 Duclaux J, Compt Rendus, 152 (1911) 1176. 38 Henry W C, Phil Trans, 114 (1824) 266. 39 De la Rive A, Ann Chim Phys [2], 39 (1828) 297. 40 Sabatier P, Compt Rendus, 152 (1911) 1176. 41 Partington J R, Nature, 174 (1954) 859. 42 Rideal E K, J Chem Soc, Abstracts, (1951) 1640. 43 Anonymous, Nobel Lectures – Chemistry 1901-1921, Novel

EDUCATOR

Foundation, Elsevier, Amsterdam, 1966; pp 216-231. Langmuir I, J Am Chem Soc, 38 (1916) 2221. Langmuir I, J Am Chem Soc, 40 (1918) 1361. Sabatier P & Espil L, Compt Rendus, 158 (1914) 668. Chamichel C, Bull Soc Chim Fr, (1955) 466. Sabatier P, Recherches Thermiques Sur les Sulphures, Thèse de Doctorat, #445, Sorbonne, Paris, 1880. 49 Wisniak J, Chem Educator, 5 (2000) 156. 50 Sabatier P, Compt Rendus, 106 (1888) 1724; 107 (1888) 40.

44 45 46 47 48

51 52 53 54 55 56 57 58 59

243 Sabatier P, Compt Rendus, 103 (1886) 49. Sabatier P, Compt Rendus, 103 (1866) 138. Sabatier P, Compt Rendus, 106 (1888), 63: 108 (1888) 738. Sabatier P, Compt Rendus, 114 (1892) 1476. Sabatier P, Compt Rendus, 115 (1892) 236. Sabatier P, Compt Rendus, 114 (1892) 1429. Sabatier P, Compt Rendus, 122 (1896) 1479. Sabatier P, Compt Rendus, 122 (1896) 1537. Sabatier P, Compt Rendus, 123 (1896) 255.