Nội dung

The Haber-Bosch Heritage: The Ammonia Production Technology Haber Bosch Mittasch Max Appl 50th Anniversary of the IFA Technical Conference September 25 – 26th 1997, Sevilla, Spain which was far beyond equilibrium. BASF found the reason for his erroneous data and – irony of history – he withdraw his patent application, not knowing how important that application could have been later when indeed iron became the basis of the commercial ammonia synthesis catalyst. Introduction Based on the fundamental research work of Fritz Haber, Carl Bosch and his engineering team developed the ammonia synthesis to technical operability using the promoted iron-based catalyst found by Alwin Mittasch and co-workers. Since then there has been no fundamental change in the synthesis reaction itself. Even today every plant has the same basic configuration as this first plant. A hydrogen-nitrogen mixture reacts on the iron catalyst (today’s formula differs little from the original) at elevated temperature in the range 400 – 500 °C (originally up to 600 °C), operating pressures above 100 bar, and the unconverted part of the synthesis gas is recirculated after removal of the ammonia formed and supplemented with fresh synthesis gas to compensate for the amount of nitrogen and hydrogen converted to ammonia. 3H2 + N2 2NH3 H0298 = – 92.4 kJ/mole ∆F0298 = – 32.8 kJ/mole First systematic measurements were made by Haber in 1904/05 but they yielded too high figures as a consequence of problems with exact analysis of the low concentrations values attained at atmospheric pressure and 1000 °C using iron for catalysis. As this figures did not comply with the Heat Theorem, W. Nernst made own measurements at 75 bar, which were actually the first experiments at elevated pressure. From the results he concluded that a technical process, which he probably anticipated as a once-through process, should not be feasible as the much higher pressures needed in this case seemed to be beyond the technical possibilities of the time. Haber continued with his investigations now also including pressure experiments. (1) From the more reliable equilibrium data now available it was obvious that at normal pressure the reaction temperature should be kept well below 300 °C in order to obtain even a small percentage of ammonia. For this temperature level no catalyst was Of course, progress made in mechanical and chemical engineering and increased theoretical knowledge have led to improvements in efficiency, converter design and energy recovery in the synthesis section, but really dramatic changes happened over the years in the technology of synthesis gas generation. As the synthesis is the very heart of every ammonia production and is also from an historical point of view the most interesting section, it is probably appropriate to start our review with this section. The synthesis The ammonia equilibrium and the recycle concept The reaction proceeds with a reduction in volume and is also exothermic, so the equilibrium concentrations of ammonia are higher at high pressure and low temperature, but at the turn of the last century a quantitative knowledge of chemical equilibrium was not available, and this might explain why early experiments aimed at the ammonia synthesis were unsuccessful. A famous victim of the lack of thermodynamic data was Wilhelm Ostwald. He offered in 1900 BASF a process in which nitrogen and hydrogen were passed over heated iron wire at atmospheric pressure, claiming several percent of ammonia, a concentration Figure 1: Equilibrium conversion and space time yield for NH3 and SO3 production 2 co-workers to develop in an unprecendented effort a commercial process in less than five years. The production facilities for 30 t/d were erected on a new site near the village Oppau (now a part of the city of Ludwigshafen), the first production was in September 1913 and full capacity was reached in 1914. available. By increasing the pressure for example to 75 bar the equilibrium conditions were improved, but with catalysts active at 600 °C only low ammonia concentrations were attained. So Haber concluded that much higher pressures had to be employed and that, perhaps more importantly, a recycle process had to be used, an actually new process concept at that time, and thus he overcame his collegues’ preoccupation which resulted from the unfavorable equilibrium concentrations and the concept of a once-through process. The ammonia catalyst In BASF Alwin Mittasch was responsible for the catalyst search. Osmium, used by Haber showed excellent catalytic activity but was difficult to handle, the main disadvantage, however, was that the world’s stock of this rare material was only a few kilograms. Mittasch started a systematic screening program, covering nearly all elements of the periodic table. Until 1910 more that 2500 different formulas were tested in 6500 runs. For these experiments special small test reactors containing easily removable cartridges holding about 2 g of catalyst were developed. The amount of ammonia formed in a single pass of the synthesis gas over the catalyst is indeed much too small to be of interest for an economic production. Haber therefore recycled the unconverted synthesis gas: after separating the ammonia formed by condensation under synthesis pressure and supplementing it with fresh synthesis gas to make up for the portion which was converted to ammonia, the gas was recirculated by means of a circulation compressor to the catalyst containing reactor. In November 1909 a sample of magnetite from a place in Sweden showed exceptionally good yields, which was surprising because other magnetite types were total failures. Mittasch concluded that certain impurities in this Gallivara magnetite were important for Haber’s recycle idea changed the static conception of process engineering in favor of a more dynamic approach. For the first time reaction kinetics were considered as well as the thermodynamics of the system. In addition to the chemical equilibrium Haber recognized that for the technical realization reaction rate was a determining factor. Instead of simple yield in a once-through process he concentrated on space time yield. Figure 1 illustrates this consideration of equilibrium concentration in combination with space time yield by a comparison of the ammonia synthesis with the SO2 oxidation process. Also anticipated by him was the preheat of the synthesis gas to reaction by heat exchange with the hot effluent gas from the reactor. In 1908 Haber approached the BASF to find support for his work and to discuss the possibilities for the realisation of a technical process. Early in 1909 he discovered in finely distributed osmium a catalyst which yielded 8 Vol% of ammonia at 175 bar and 600 °C. A successful demonstration in April 1909 of a small labscale ammonia plant convinced the representatives of BASF and the company’s board decided to pursue the technical development of this process with all available resources. In BASF then Carl Bosch, entrusted with extraordinary authority, became project leader and succeeded together with a team of dedicated and very able Figure 2: 3 Ammonia equilibrium and catalyst volume nia synthesis in the pressure range of industrial interest. This success has many fathers, outstanding contributions were made by Brill, Ertl, Somorjai, Boudard, Nielsen, Scholtze, Schlögl and many others. The rate determining step is the dissociative adsorption of the nitrogen at the catalyst surface and the most active sites are the crystal faces (111) and (211), which is probably caused that these are the only surfaces which expose C7 sites, which means iron atoms with seven nearest neighbors. its good performance. So he investigated the influence of various individual additives, which in today’s terminology are called promoters. By 1911 the catalyst problem had been solved. Iron with a few percent alumina and a pinch of potassium yielded a catalyst with acceptable reproducibility and performance and tolerable lifetime. But the research program was continued until 1922 to be certain about the optimum composition. The only additional result was that the further addition of calcium gave a certain improvement. All magnetite based catalysts on the market today have a similar composition to that of the original BASF catalyst. Also the catalyst preparation remained practically the same: Melting natural magnetite from Sweden with addition of the various promoters, cooling the melt, breaking the solidified melt into small particles followed by screening to obtain a fraction with suitable particle size. The primary function of the Al2O3 is to prevent sintering by acting as a spacer between the small iron platelets and it may in part also contribute to stabilize the Fe(111) facets. The promoting effect of the potassium is probably based on two factors. One is the lowering of the activation energy of the dissociative adsorption of nitrogen by an electronic charge transfer effect from potassium to iron which increases the nitrogen bond strength to the iron and weakens the nitrogen-nitrogen bond. The other factor consists in reducing the adsorption energy of ammonia thus easing the desorption of the formed ammonia which avoids blocking the surface and hindering the nitrogen adsorption. From these early days until today an enormous amount of academic research was dedicated to elucidate the mechanism of the synthesis, to study the microstructure of the catalyst and to explain the effect of the promoters. Besides the scientific interest there was of course some hope to find an improved catalyst, which could operate at far lower temperatures and thus at lower pressures saving compression energy, which is in a modern plant still 300 kWh/t NH3. In principle one can operate with the classic magnetite catalyst at 35 – 45 bar in the temperature range of 350 to 450 °C, but needing a trainload of catalyst – about 450 m3 (1300 t) for a plant of 1350 t/d NH3 – to achieve very low ammonia concentrations which would require removal by water-scrubbing instead of condensation by refrigeration. M. W. Kellogg proposed such a process in the early 1980s, but didn’t succeed with commercialization. For a real low pressure catalyst operating at front end pressure to need no compression, an operation temperature well below 300 °C would be required. To illustrate this situation figure 2 shows ammonia equilibrium and catalyst volume. Commonly the term ammonia catalyst is used for the oxidic form consisting of magnetite and the promoters. Actually this is only the catalyst precursor, which is transformed into the active catalyst consisting of αiron and the promoters by reduction with synthesis gas. In the 1980s pre-reduced ammonia catalysts found acceptance in the market as they avoid the relatively long in-situ reduction which causes additional downtime and considerable feedstock consumption without production. These catalysts are reduced at the vendor’s facilities and subsequently passivated at temperatures around 100 °C using nitrogen with a small amount of air. A notable improvement of the magnetite system was the introduction of cobalt as an additional component by ICI in 1984. The cobalt enhanced formula was first used in an ammonia plant in Canada using ICI Catalco’s AMV process with a synthesis pressure of 90 bar. With similar kinetic characteristics, the volumetric activity is about two times higher than that of the standard iron catalyst. With the modern spectroscopic tools of Surface Science rather detailed information on the reaction mechanism at the catalyst surface was obtained. Kinetics of nitrogen and hydrogen adsorption and desorption were investigated and adsorbed intermediate species could be identified. The results allowed to explain, for the most part, the mechanism of ammo- In October 1990 Kellogg commercialized the Kellogg Advanced Ammonia Process using a catalyst composed of ruthenium on a graphite support, which is 4 claimed to be 10 – 20 times as active as the traditional iron catalyst. According to original patents asigned to BP, the new catalyst is prepared by subliming ruthenium-carbonyl Ru3(CO)12 onto the carbon-containing support which is impregnated with rubidium nitrate. The catalyst has a considerably higher surface than the conventional catalyst and, according to the patent example, it should contain 5 % Ru and 10 % Rb by weight. This catalyst works best at a lower than stoichiometric H/N ratio of the feed gas and it is also less susceptible to selfinhibition by NH3 and has an excellent low pressure activity. The potential of ruthenium to displace iron in new plants will depend on whether the benefits of its use are sufficient to compensate the higher costs. In common with the iron catalyst it will also be poisoned by oxygen compounds. Even with some further potential improvements it seems unlikely to reach an activity level which is sufficiently high at low temperature to allow an operation of the ammonia synthesis loop at the pressure level of the syngas generation. Figure 3: First pilot plant converter with soft iron lining and external heating The ammonia converter and the synthesis loop configuration Bosch’s unconventional solution to the embrittlement problem was to use a carbon steel pressure shell with a soft iron liner. To prevent the hydrogen which had penetrated this liner from attacking the pressure shell, measures had to be taken to release it safely to normal pressure. This was achieved by providing small channels on the outer side of the liner which was in tight contact with the inner wall of the pressure shell and by drilling small holes, later known as “BoschHoles”, through the pressure shell, through which hydrogen could escape to the atmosphere. These holes had no effect on the strength of the shell and the resulting losses of hydrogen were negligible. Figure 3 gives a sketch of such a pilot plant converter. With the catalyst at hand, the next step was to construct somewhat larger test reactors for catalyst charges of about 1 kg. Surprisingly, these reactors ruptured after only 80 hours. Further studies showed that the internal surface had totally lost its tensile strength. This phenomenon had apparently propagated from the inner surface outward until the residual unaffected material was so thin that rupture occurred. With the aid of microscopic investigations by thin section technique Bosch found the explanation. Decarbonization of the carbon steel had occurred, but, surprisingly, the result was not soft iron but rather a hard and embrittled material. Hydrogen diffusing into the steel caused decarbonization by methane formation. This methane, entrapped under high pressure within the structure of the material, led to crack formation on the grain boundaries which finally resulted in embrittlement. Systematic laboratory investigations and material tests demonstrated that all carbon steels will be attacked by hydrogen at high temperatures and that the destruction is just a matter of time. Bosch did not content himself with his liner/hole concept but looked further for alternative solutions for the embrittlement problem. He intitiated in the late 1920s research in the steel industry to develop steels resistant to hydrogen under pressure. Special alloy components as for example molybdenum, chromium tungsten and others form stable carbides and enhance the resistance of steel against this sort of attack considerably. This problem and the related physical hydrogen attack is not restricted to the synthesis but has to be considered carefully also in the synthesis gas production section because of the temperatures and 5 synthesis gas kept the pressure vessel walls cool and rendered the liner-hole concept redundant. Subsequent reactor designs in the technical plant included internal heat exchangers and later the catalyst was placed in separate tubes which were cooled by the feed gas. Another improvement was the introduction of an externally insulated catalyst basket. Because of the low concentrations aqueous ammonia was separated from the loop by water scrubbing. Converters with catalyst tubes had a better temperature control and this led together with an increased pressure to higher ammonia concentrations which now allowed from 1926 onwards the direct production of liquid ammonia. In 1942 the first quench converter was installed and this design gradually has replaced then the converters with the catalyst tubes. Figure 4: nitrogen Soon after the first world war development started also in other countries, partly on basis of BASF’s pioneering work. Luigi Casale built 1920 the first plant in Italy, and based on developments by M. G. Claude the first French plant started to produce in 1922. Both the Casale and the Claude process operated under extreme high pressure. In contrast to this Uhde constructed a plant based on coke oven gas, operating under extreme low pressure. (Mont Cenis process). Futher developments were by G. Fauser who worked together with Montecatini. During the 1920s several plants were built in the USA, some based on European some on American Technology. The successful US company was Nitrogen Engineering Corporation (NEC), the predecessor of Chemico. Converter with pressure shell cooling by hydrogen partial pressures involved there. Extensive research and careful evaluation of operation experiences have made it possible to prevent largely hydrogen attack in modern ammonia plants by proper selection of hydrogen-tolerant alloys with the right content of metals which form stable carbides. Of fundamental significance in this respect was the work of Nelson, who produced curves for the stability of various steels as a function of operation temperature and hydrogen partial pressure. Mechanical design was now already rather advanced but for the process design of converter and loop so far empirical data in form of charts were used as no suitable mathematical expressions for the reaction kinetics were at hand. When better experimental data for the reaction kinetics and other process variables became available in the 1940s and 1950s lay-out of converters received a better quantitative chemical engineering basis. Figure 5 shows reaction rate of ammonia formation and equilibrium. When the temperature is increased (under otherwise constant conditions), the reaction rate increases to a maximum, to decrease with further temperature increase and becomes zero when reaching equilibrium temperature. Joining these points will result in a line giving, for each NH 3 concentration, the temperature for the maximum rate. This curve runs about parallel to the In the small reactors heat losses predominated and continuous direct external heating by gas was necessary and this led to deterioration of the pressure shells after short operation times even without hydrogen attack. With increasing converter dimensions in the commercial plant heating was only necessary for start up. Bosch developed an internal heating by the so-called inversed flame, introducing at the top of the reactor a small amount of air, igniting with an electrically heated wire. Later this was replaced by an electric resistance heater. Subsequently introduced flushing with nitrogen as shown in figure 4 and later with cold 6 first part of the catalyst) should follow this ideal line. For a long time converters were always compared to this “ideal” for optimum use of high-pressure vessel volume. Today the objective is rather to maximize heat recovery (at the highest possible level) and to minimize investment costs for the total synthesis loop. In any case it is necessary to remove the heat of reaction as the conversion proceeds to keep the temperature at an optimal level. For the removal of the reaction two principal configurations are possible: Figure 5 : Tubular converters have cooling tubes within the catalyst bed through which the cooling medium, usually cooler feed gas, flows co-currently or counter-currently to the gas flow in the catalyst bed. Alternatively the catalyst can be placed within tubes with the cooling medium flowing on the outside. The tube cooled converters dominated until the early fifties, but are largely outdated today. Well known examples were the TVA converter (counter-current) and the NEC/Chemico design (co-current, with best approximation to the maximum rate curve). An interesting revival of this principle is the ICI tube cooled converter used in the LCA process and also for methanol production. Reaction rate of ammonia formation In the multi-bed converters the catalyst volume is divided into several beds in which the reaction proceeds adiabatically. Between the individual catalyst layers heat is removed either by injection of colder synthesis gas (quench converters) or by indirect cooling with synthesis gas or via boiler feed water heating or steam raising (indirectly cooled multi-bed converter). equilibrium line and at a about 30 – 50 °C lower temperature. To maintain the maximum ammonia formation rate, the reaction temperature must decrease as the ammonia concentration increases. For optimal catalyst usage the reactor temperature profile (after a initial adiabatic heating zone in the Figure 6: Quench converter 7 In the quench converters only a fraction of the recycle gas enters the first catalyst layer at about 400 °C. The catalyst volume of the bed is chosen so that the gas will leave it at around 500 °C. Before entering the next catalyst bed, the gas temperature is ,,quenched” by injection of cooler (125 – 200 °C) recycle gas. The same thing is done at subsequent beds. In this way the reaction profile describes a zig-zag path around the maximum reaction rate line. A schematic drawing of a quench converter together with its temperature/location and temperature/ammonia concentration profile is presented in figure 6. The catalyst beds may be separated by grids designed as mixing devices for main gas flow and quenchgas (cold shot), or be just defined by the location of cold gas injection tubes as for example in the ICI lozenge converter. A disadvantage is that not all of the recycle gas will pass over the whole catalyst volume with the consequence that a considerable amount of the ammonia formation occurs at higher ammonia concentration and therefore at reduced reaction rate. This means that a larger catalyst volume will be needed compared to an indirect cooled multi-bed converter. On the other hand, no extra space is required for inter-bed heat exchangers, so that the total volume will remain about the same as for the indirect cooled variant. Figure 7: Topsoe Series 200 indirect cooled converter (radial flow) Axial flow through the catalyst in the converters as exclusively used until the early 1970s face a general problem: With increasing capacity the depth of the catalyst beds will increase, as for technical and economical reasons it is not possible to enlarge the pressure vessel diameter above a certain size. In order to compensate for the increasing pressure drop axial flow converters with usual space velocities of 10 –15000 h-1 have to use relatively large catalyst particles and a particle size of 6 –10 mm has become standard. But this grain size has compared to finer catalyst a considerably lower activity, which decreases approximately in a linerar inverse relation. Two factors are responsible for the lower activity of the larger particles. Firstly, the larger grain size retards on account of the longer pores the diffusion from the interior to the bulk gas stream and this will inhibit the dissociative nitrogen adsorption and by this the reaction rate. Secondly, the reduction of an individual catalyst particle starts from the outside and proceeds to the interior. The water formed by removing the oxygen from the iron oxide in the interior of the grains will pass over already reduced catalysts on its way to the outer surface of the particle. This induces some recrystallization leading to the lower activity. The effect is considerable: going from a partide size of 1 mm to one of 8 mm, the inner surface will decrease from 11 –16 to 3 – 8 m2/g. As the quench concept was well suited for large capacity converters it had a triumphant success in the early generation of large single stream ammonia plants constructed in the 1960s and 1970s. Mechanical simplicity and very good temperature control contributed to the widespread acceptance. Multibed converters with indirect cooling. In converters of this category the cooling between the individual beds is effected by indirect heat exchange with a cooling medium, which may be cooler synthesis gas and/or boiler feed water warming and steam raising. The heat exchanger may be installed together with the catalyst beds inside one single pressure shell but an attractive alternative, too, preferentially for large capacities, is to accommodate the individual catalyst beds in separate vessels and have separate heat exchangers. This approach is especially chosen when using the reaction heat for raising high pressure steam. The indirect cooling principle is applied today in almost all large new ammonia plants, and also in revamps an increasing number of quench converters are modified to the indirect cooling mode. 8 Haldor Topsøe’s company solved the dilemma with the pressure drop and small catalyst particles with a radial flow pattern, using a grain size of 1,5 – 3 mm (Figure 7). M.W. Kellogg chose another approach with its horizontal crossflow converter (Figure 8). The catalyst beds are arranged side by side in a carFigure 8: Indirect Cooled Horizontal Converter of M. W. Kellogg tridge which can be removed for catalyst loading and unloading through a full-bore closure of the horizontal pressure shell. Today each new world-size ammonia plant employs the indirect cooling concept raising high pressure steam up to 125 bar. Generally after the first bed an inlet-outlet heat-exchanger is placed and after the second or further beds the reaction heat is used to raise high pressure steam. Brown and Root (formerly C. F. Braun) or Uhde (Figure 9) accommodate the catalyst in several vessels. Figure 9 is a simplified flow sheet of Uhde’s synthesis loop. Actually the concept of separate vessels for the catalyst beds, with heat exchange after the first and waste heat boiler after the Figure 9: Uhde’s synthesis loop with two pressure vessels and three catalyst beds 9 on very fine catalyst particles. The first useful expressions for engineering purposes to describe the reaction rate was the Temkin-Pyshew equation, proposed in 1940. It was widely applied, but today there are improved versions and other equations available. Additional terms are included to model the influence of oxygen-containing impurities on the reaction rate. Although oxygen-containing compounds may be regarded as a temporary poison, severe exposure for an extended period of time leads to permanent damage. For practical application these equations have to be modified to make allowance for transport phenomena (heat and mass transfer), and this is done by so-called pore effectiveness factors. second (nowadays they use also a third one followed by a boiler, too) was already introduced by C. F. Braun at time when most plants still used quench converters. The Ammonia Casale ACAR Converter has a mixed flow pattern. In each catalyst layer the gas flows through the top zone predominantly axially but traverses the lower part in radial direction. This simplifies the design by avoiding special sealing of the top end of the bed to prevent by-passing. Today computerized mathematical models are used for converter and loop lay-out. In principle, these models use two differential equations which describe the steady state behavior of the reaction in the converter. The first gives a concentration-location relationship within the catalyst bed for the reactants and the ammonia. It reflects the reaction kinetic expression. The second models the temperature-position relationship for the synthesis gas, catalyst and vessel internals. The form of this equation is specific to the type of the converter. The kinetics of the intrinsic reaction, that means the reaction on the catalyst surface without any mass transport restrictions, are derived from measurements Figure 10: Simplified flow sheet of a coke-based ammonia plant 10