History of Chemical Engineering - American Chemical Society

---

17 The History of Chemical Engineering

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

at Exxon EDWARD J. GORNOWSKI Exxon Research and Engineering Co., P.O. Box 101, Florsham Park, NJ 07932

From W. K. Lewis, one of our most valued consultants, we learned early on to go for the fundamentals, to understand what we're doing, to build on a solid foundation—but not to wait until we understand everything before being willing to make a decision. Major chemical engineering contributions from Exxon include continuous thermal cracking processes; putting fractional distillation on a sound basis; processes for tetraethyllead, 100-octane avgas, and raw materials for syn­ thetic rubbers and chemicals; fluid cat cracking, fluid coking, and the Flexicoking process. Fluid hydroforming was a disappointment. Reactor engineering and separa­ tions technology have seen major advances. Coal conver­ sion is again a focus of activity. This chapter deals with refining technology although chemical engineering has been essential to many other Exxon activities.

hemical Engineering at Exxon goes back to the founding of its corporate predecessors and underlies much of the technology that pro­ vides the basis of Exxon's multi-faceted operations. Today Exxon is in­ volved in chemicals, minerals, nuclear fuel assemblies, solar photovoltaic cells, and in many other fields—and chemical engineering has played an important role in the history of each of them. I wish that I could delve into the historical contributions of chemical engineering and chemical engineers to all of these fields but space limitations have persuaded me to limit myself to just one—the refining of fossil fuels. Relevant commercially oriented technology existed even before Drake drilled thefirstoil well in 1859. In 1850 James Young, a scientist from Glasgow, Scotland, patented a distillation process to produce 0-8412-0512-4/80/33-190-303$05.00/l © 1980 American Chemical Society In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

304

HISTORY O F C H E M I C A L E N G I N E E R I N G

naphtha, kerosene, lubricating o i l , and paraffin wax from coal tar and oil shale. W e shouldn't forget that what we now call the "potential syn­ thetic fuel industry" actually predates the petroleum industry. The infant petroleum industry soon saw the benefit of technological expertise and the Standard O i l Alliance, formed i n 1875, engaged William G . W a r d e n , inventor of an oil-distilling process and an improved railroad tank car, H e n r y Rogers, inventor of a widely used petroleum distillation process and a naphtha separator, and E l i Hendrick, a noted lubricating oil specialist. B y the time the Standard O i l Company was organized in 1882, it held over 20 patents for general refining processes and equipment as well as some 30 lubricating oil patents. Although several extended R & D programs added to this base of technical knowledge after 1882, the work was directed towards the solution of very specific problems. F . W . Arvine set up the company's first engine lab, known as a "power and machinery room," at 128 Pearl St., New York City, in 1882 to help improve lubricating oil and grease quality. George M . Saybolt set up his standard inspection lab at the company's headquarters at 26 Broadway in N e w York C i t y i n 1883 to develop product quality tests and uphold the meaning of the word "standard" in the company's name. Herman Frasch set up a lab at the Whiting, I N Refinery in 1886 to develop a process for converting the sour, sulfur-laden crude oil from the L i m a , O H field into marketable petroleum products. W h e n the old Standard O i l Company was split into 34 companies by the Supreme Court antitrust decision of 1911, the Standard O i l Company (New Jersey)—which later changed its named to Exxon—lost the services of the research laboratory at the Whiting Refinery, then headed by W i l l i a m M . Burton. It also lost the benefit of a new cracking process developed at that lab i n 1913. D u r i n g W o r l d War I, Jersey Standard's management was too busy supplying petroleum products to the allied armies to press far into new fields. But by 1919 the company's top management had become con­ cerned about the technological base for its operation. Jersey Standard President Walter C . Teagle wrote an associate in June, 1919, " I have felt more than ever before the need for a thoroughly organized and competent research department under an able executive, such a department not to be confined to chemical research, but to general research i n connection not only with production and refining of our products, but with the sales end of our business as w e l l . " A formal notice announcing the formation of a development department was issued on September 27, 1919. The department's 26 employees worked in labs at the Bayway Refinery in New Jersey. W h e n we undertook to set up an organization to apply chemical engineering to the oil industry as rapidly and efficiently as possible, it was

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

17.

GORNOWSKI

Chemical Engineering at Exxon

305

absolutely necessary to turn to others for help. The original group which advised and assisted in creating the technical organization was made up of Ira Remsen, President Emeritus of John Hopkins University and perhaps the leading organic chemist of his day, Robert A . Millikan of the Cali­ fornia Institute of Technology, one of our greatest physicists and later a Nobel Prize winner, and Warren K . Lewis, professor of chemical engineering at M . I . T . W e also relied heavily on Charles A . Kraus of Brown University and Carelton Ellis of Montclair, N J . The first major chemical engineering refining problem we faced was the development of a satisfactory continuous cracking process for con­ verting heavier oils into gasoline to replace the Burton process. Two processes were developed. Both were engineering applications of the principle that the cracking of petroleum is a function of time and tem­ perature. The first process was the Double C o i l Process, in which the oil was heated rapidly to a high cracking temperature in pipes exposed to the radiation i n the furnace, and then "soaked" or held at this high temperature for a further period in pipe coils that were exposed only to the m i l d heating effect of low-temperature combustion gases. The heat flux was so small that there was little danger of burning the tubes when they became coated with carbon inside. The second was the Tube-and-Tank Process, in which a relatively large pressure tank on the exit end of the pipe coil provided the time necessary for the desired cracking reaction to complete itself. These two operations almost immediately became the standard operations for the cracking of heavy oils for Exxon, and also were used widely by licensees, replacing the older batch-type cracking stills and effecting enormous economies, both in the first cost and in the operating cost, as well as greatly extending the range of practical feedstocks that could be used. The attention of the world's entire oil industry was focused on this development, and there were endless patent controversies. O u r next important problem was the modernization of the basic operation of petroleum refining—fractional distillation. The oil refinery of the early 1920's was made up of batteries of horizontal cylindrical batch stills with a capacity of 200 to 1,000 bbl each, sometimes arranged for operation singly, and sometimes connected in cascade by overflow pipes. The stills were mounted in brick settings that exposed the bottom third of the still to the direct radiant heat of the furnace. To assist the separation carried out by the distillation itself, the stills usually were equipped with a series of partial condensing towers in which heavier components of the vapors condensed seriatim before the final condensation in the watercooled worm or pipe coil. The separations effected by this combination of fractional distillation and fractional condensation were very poor, and it was necessary for the refiner to do an enormous amount of redistillation. A refinery such as Bayway would have several batteries of stills engaged

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

306

HISTORY O F C H E M I C A L E N G I N E E R I N G

i n redistillation for every battery engaged in crude distillation. The solution to this problem was found during the 1920's by combining the pipe coil type of furnace first used on a large scale in the continuous cracking processes with fractionating columns designed in accordance with principles of fractional distillation, which were understood well in the scientific world, but which the oil industry never had applied before on a large scale. The oil industry i n general owes more to Warren K . Lewis than to any other individual for the quick and successful application of the scien­ tific principles of fractionating column design to the oil industry. We have continued to develop the chemical engineering technology for frac­ tional distillation and Exxon now has continuous distillation units capable of handling up to 275,000 bbl/day (40,000 tons/day) of crude oil. One of the lessons that Doc Lewis taught us, along with teaching it to whole generations of chemical engineers, was to go for the funda­ mentals, to understand what you're doing, to build on a solid foundation— but not to wait until you understand everything before you're willing to make a decision. This has been as important a lesson for us as any other. W e still are shooting for that optimum balance of theoretical under­ standing, engineering knowhow, common sense, intuition, and guts that characterize a good chemical engineer. D u r i n g this same period, i.e. the 1920's, we were engaged in another extremely important chemical engineering development. Messrs. Kettering and Midgley of General Motors had investigated the phenome­ non known as knocking in internal combustion engines and identified it as detonation of the fuel charge in the engine cylinder. Finding a way to upgrade gasoline to prevent engine knock became an important research goal during this period as the horseless carriage gained in popularity and as automobile engines became more powerful. Midgley discovered that a mixture of tetraethyllead (TEL) and an alkyl halide, in the astonishingly small proportions of .1 of 1%, would prevent knocking of ordinary gaso­ line. But there was no known method of making T E L economically in the required large amounts. With the help of D r . Kraus, our chemical engineers solved this problem by developing a simple and practical method of producing T E L from an alloy of sodium and lead treated with ethyl chloride. The ethyl chloride required for the production of T E L was made also from refinery ethylene by a process that our chemical engineers worked out. D u r i n g the 1920's fears of a future petroleum shortage caused Jersey Standard to become interested in synthetic fuels. This interest led to an agreement in 1927 between Jersey Standard and the German firm, I. G . Farben, for a cooperative research program to develop a coal hydrogé­ nation process in the United States. A group of 18 American oil com­ panies was organized. This group exchanged technical information but

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

17.

GORNOWSKI

Chemical Engineering at Exxon

307

left the principal burden of technical effort to us. The technology was to become a cornerstone of the company's chemical engineering expertise. A month after signing the Farben agreement, Jersey Standard took another step which signaled the start of an era of expansion in chemical engineering. O n October 27, 1927, Jersey established the Standard O i l Development Company and transferred the development and general engineering departments, formed in 1913, to the new company. The December, 1927 issue of The Lamp reported that the new company " w i l l have personnel and facilities suited for handling all varieties of engineer­ ing and chemical work of general interest, for carrying on major projects of laboratory research, and for the technical and financial direction of major projects." E v e n today Exxon Research and Engineering C o . , the successor of the Standard O i l Development Company, retains much of the basic structure established in 1927. The magnitude of the technical problems involved in high-pressure coal hydrogénation seemed staggering. Preparing to meet this situation in 1927 we drafted R. T. Haslam, an associate of Doc Lewis at M . I . T . , to organize an entirely new technical group at Baton Rouge. Robert P. Russell, assistant professor of chemical engineering at M . I . T . , became manager of the new labs. Russell recruited a staff composed largely of young M . I . T . faculty members and graduate students. This crew of chemical engineers without any previous refinery experience was greeted with some consternation by the established old-timers on site, but the dedication, willingness to learn, competence, and undoubted successes of the newcomers soon established their credentials as a valuable chemical engineering asset. This role has been maintained for over 50 years by Exxon's Research & Development Labs at Baton Rouge. Discovery of additional prolific oil fields postponed any need for hydrogenating coal, but hydrogénation made 100-octane aviation gasoline possible. A t first, 100-octane fuel was made by the dimerization of refinery isobutylene. This hydrogenated diisobutylene which was, in effect, commercial isooctane, was mixed with selected natural gasoline of the highest quality and with T E L to produce the 100-octane gasoline on which the U . S . A r m y A i r Corps did its original high-compression engine development work. Soon the supply of natural gasoline of this quality became inadequate and it was necessary to resort to hydrogénation to produce a synthetic gasoline of high enough octane number for blending with the synthetic isooctane. A t the outbreak of the war in 1939, the Baton Rouge hydrogénation plant was producing both the synthetic blending agent and the synthetic base, and was the largest single source of 100-octane fuels i n the world. Availability of 100-octane avgas, and the power boost it provided, allowed the Royal A i r Force's Spitfire fighters to outperform the Luftwaffe's planes that were performance limited by Germany's 87-octane fuel.

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

308

HISTORY O F C H E M I C A L E N G I N E E R I N G

In the production of aviation gasoline the original hydrogénation methods soon were supplemented by two other processes—alleviation and catalytic cracking. W e engineered the first commercial alkylation plant and also the first plants for the isomerization of normal paraffins to produce isoparaffins for alkylation purposes. The catalytic cracking proc­ ess was pioneered commercially by the Houdry Company, but an entirely new development, fluid catalytic cracking, which represented the contri­ bution of our chemical engineers to a cooperative effort participated in by several American and foreign companies, enormously advanced the eco­ nomic frontiers of catalytic cracking. The development of fluid catalytic cracking was a real chemical engineering challenge. There were two basic problems: the catalyst was deactivated rapidly by coke deposits and the cracking reaction was very endothermic—huge quantities of heat had to be supplied to the catalyst. Houdry's fixed-bed process had overcome these two problems by switch­ ing the flow through the catalyst bed every few minutes; first hydro­ carbons to crack and deposit coke, then air to burn the coke, regenerate and heat up the catalyst, with steam purges in-between to prevent disasters. W e believed that this cumbersome process, with its constant need to shift large flow streams, could be improved by continuously moving the catalyst between a reaction zone and a regeneration zone. But how to move the catalyst? Bucket elevators? Screw convey­ ors? Early on we thought of the pneumatic transport used to move grain and we considered a powdered catalyst to make the transport easier. Fortunately, or unfortunately, we did some lab work on the coke-burning reaction rate and, assuming pneumatic transport would maintain a solids density of about 1 lb/cu ft (16 g/L), we calculated that we would need a regenerator some 7 ft (2 m) in diameter and about 7 miles (11 km) long. W e set out, therefore, to raise the reaction rate by finding a catalyst that could stand a higher regeneration temperature. A n d we put a long regenerator on our 100 bbl/day (10 L/min) pilot plant in Baton Rouge. In order to fit as much regenerator pipe as we could into a reasonable space, we built what we called a "snake"; it went up and down and up and down and up and down. . . To our surprise, we found that the solids density was higher i n the upflow legs than in the downflow ones. D o c Lewis quickly put some of his students to work measuring solids density as a function of flow direction, gas velocity, and solids flow rate in glass apparatus . . . and shortly the dense fluid bed regenerator became more than a gleam i n the eye. Dense fluid beds were known—the Winkler gas generator had been around for years—but it had always been thought that you had to operate at gas velocities below the free-fall velocity of your particles; that at higher velocities the particles simply would be blown away. O f course,

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

17.

GORNOWSKi

Chemical Engineering at Exxon

309

practical regenerator gas velocities eventually will blow away (or entrain) the fine cat-cracking catalyst particles, but it came as a great surprise that before they blew away they acted just like a Winkler bed. W e capitalized on this surprise and quickly converted the Baton Rouge pilot plant to dense fluid-bed operation. There were still many things we didn't understand, but we went ahead. W e developed a whole host of ancillary technology, from catalyst standpipes that built up hydrostatic pressure to improved cyclones. A n d while we were theo­ rizing and experimenting in our pilot unit, we were designing and build­ ing commercial plants. Commitments were made to build 30 fluid catcracking plants before the first commercial unit started up. Many names are associated with the development of fluid cat crack­ ing. Along with those of our consultants, Doc Lewis and E d Gilliland of M . I . T . , the names that stand out particularly were those of four of our chemical engineers: Homer Z. Martin, C . Wesley Tyson, Donald L . C a m p b e l l , and Eger V . Murphree. It is hard to convey the magnitude of the challenge and the sense of accomplishment that the teams of chemical engineers felt in creating a whole new technology and in seeing it go on-stream practically simultaneously. The cat plants provided more than components for high-octane gasoline. They provided raw materials that were essential to the U . S . Government's synthetic rubber program during W o r l d War II—buty­ lènes directly and butadienes indirectly via dehydrogenation. Chemical engineers at Exxon were involved deeply in the synthetic rubber program (for Buna S as w e l l as butyl rubbers) and in such other wartime programs as the synthesis of nitration-grade toluene and the development of the steam-cracking process to produce chemical raw materials. Again, space limitations do not permit detailing the many chemical engineering con­ tributions made on the chemicals side of the fence. After W W I I , chemical engineering continued to provide opportuni­ ties for technological advances in refining for Exxon—and for an occa­ sional retreat. One of the less successful developments was fluid hydroforming, a process to raise the high-octane aromatics content of naphthas. The desired high conversion and yield levels, coupled with the relatively slow catalyst deactivation rate and modest heat require­ ments, really d i d not make catalytic reforming a good candidate for fluid-bed operation. Fixed-bed reforming with noble metal catalysts has supplanted fluid hydroforming. M o r e successful was the extension of fluidized-bed technology to fluid coking, a process to upgrade heavy petroleum residues. F l u i d coking presented many chemical engineering problems. Coke is laid down i n layers on existing coke particles which thus tend to grow. Therefore, it is necessary to provide small seed particles. There is no natural seed formation as i n crystallizers, so seed particles must be

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

310

HISTORY O F C H E M I C A L E N G I N E E R I N G

generated by grinding. Relations between grinding, growth, and ag­ glomeration rates must be worked out. Coking of tar deposits i n the cyclones must be prevented, but places for tar deposition must be pro­ vided. A n d so on. O u r chemical engineers provided adequate solu­ tions to all of these problems and fluid coking became an established process. F l u i d coking technology has been extended further recently to the Flexicoking process, where the coke is gasified with steam and air to form a l o w - B T U fuel gas and there is only very little net coke production. Two chemical engineering applications based on fluidized-bed tech­ nology are currently under development at Exxon that we believe will make the history books of the future: pressurized fluid-bed combustion ( P F B C ) and magnetically stabilized beds (MSB). P F B C shows particular promise for the clean, compact combustion of coal i n power plants or process heaters. Capital investments should be less and environmental protection easier than with regular coal-fired furnaces. M S B uses ex­ ternally imposed magnetic fields acting on beds of magnetizable particles to collapse gas bubbles as soon as they start to form i n a fluidized bed. A n M S B is thus a calm, quiescent bed without the gas bypassing and the back mixing that gas bubbles ordinarily engender. Our chemical engineers are working to use these characteristics of M S B to help solve some long-standing process problems. I have stressed fluidization technology i n my recounting of chemical engineering at Exxon over the last 40 years. Fluidization has been an important area of chemical engineering activity but it is by no means the only one. W e have been very active in all aspects of what has become known as reactor engineering—the design and operation of reactor sys­ tems to optimize a combination of heat, mass and momentum transfer, chemical kinetics, and control strategy to permit ready, safe, and profit­ able operation with a variety of feed and product constraints. The power of modern computers, along with our innovative software, allows our chemical engineers to handle far more variables than would have been thought possible just twenty years ago. W e have continued to pioneer i n the development and application of many separation techniques from heatless drying (the removal of mois­ ture by intermittent absorption and desorption at different pressure levels), through the use of molecular sieve absorbers, advanced lube oil extraction media, and C 0 absorption promoters, to the use of liquid membranes (containing an encapsulated absorbing phase) and lasers to activate only specific molecules. Some 50 years after the flurry of interest i n coal hydrogénation and 120 years after James Young's patent, we again are immersed in the chemical engineering problems of turning coal into more attractive liquid and gaseous fuels. 2

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.

Downloaded by NORTH CAROLINA STATE UNIV on November 30, 2012 | http://pubs.acs.org Publication Date: June 1, 1980 | doi: 10.1021/ba-1980-0190.ch017

17.

GORNOWSKI

Chemical Engineering at Exxon

311

A n d , of course, chemical engineering at Exxon has, along with the Corporation's expanding interests, branched out into areas far removed from the refining of fossil fuels. There are difficult chemical engineering problems associated with the manufacture of petrochemicals, minerals recovery, advanced batteries and other forms of energy storage, the economic utilization of solar energy to replace other energy forms, and even with information processing technology. Exxon's chemical engi­ neers are active i n all of these areas. But, as I said, I am restricting myself to refining technology. The Exxon Research and Engineering Company has long had a prominent role to play i n the development and application of Exxon's refining technology with the help of chemical engineering. W e con­ tinuously aim to hone our technical skills through challenging assign­ ments, thorough peer review of our work, continuing education, and ample contact with our academic colleagues. I trust that our future accomplishments w i l l warrent Exxon again receiving an invitation 50 years from now to prepare a paper discussing "Chemical Engineering at Exxon from 1979 to 2029." R E C E I V E D May 7,

1979.

In History of Chemical Engineering; Furter, W.; Advances in Chemistry; American Chemical Society: Washington, DC, 1980.