The picture changed when Franklin Veatch, a research supervisor reporting to E.C. Hughes, director of research, proposed that converting light refinery gases such as the aliphatic hydrocarbon propane to oxygenatescompounds containing oxygencould be profitable. At the time, oxidation of aliphatic hydrocarbons was primitive and expensive. Veatch's idea was to use metal oxides to convert hydrocarbons to oxygenates. Funding was approved for this effort beginning in 1953.

In addition to starting new research, Sohio ventured into the petrochemical business by building ammonia and nitrogen plants in Lima, Ohio, and near Joplin, Missouri, to use by-products from its petroleum refinery. It was a conservative move, but it encouraged Sohio to view chemicals as a commercial enterprisea venture that would lead to remarkable success.

Early experiments in Veatch's research yielded no major developments, and he was given a six-week deadline. The resulting crash program succeeded when a test run was made on propylene over a modified vanadium pentoxide oxidant, and the resultant odor was instantly recognizable as acrolein. Veatch knew that one more oxidation step would take acrolein to acrylic acid--an important, expensive, fast-growing monomer. For the next two years, several researchers, including Ernest C. Milberger, James L. Callahan, Robert W. Foreman, James D. Idol, Jr., Evelyn Jonak, and Emily A. Ross, were involved in this development effort.

In 1955 the team began testing oxidants as direct oxidation catalysts. In an experiment designed by Jim Callahan and performed by Emily Ross, bismuth phosphomolybdate produced acrolein in yields of 40 percent or more. This was a first-magnitude discovery: propylene to acrolein in a single catalytic reaction step. Acrylic acid could be made in a subsequent step. Callahan, Foreman, and Veatch secured key patents on the bismuth phosphomolybdate catalyst, and from then on, things were destined to happen fast.

Jim Idol suggested acrylonitrile as a derivative of acrylic acid and successfully carried out catalytic conversion of the ammonium salt of acrylic acid. Next, acrylonitrile was made by feeding acrolein, ammonia, and air over the catalyst that produced acrylic acid from acrolein. This success suggested that acrylonitrile might be made directly from propylene by carrying out the entire reaction in a single step with bismuth phosphomolybdate. The experiment, designed by Idol and performed by Evelyn Jonak in March 1957, resulted in ammoxidation, a process that produced acrylonitrile in about 50 percent yield with acetonitrile and hydrogen cyanide as co-products.

With the capacity to make acrolein, acrylic acid, and acrylonitrile by efficient, revolutionary new processes, Veatch pressed for a strong development and commercialization effort. The Patents and Licensing Department went to work on securing an iron-clad patent position. Because manufacturing both acrylic acid and acrylonitrile proved to be too ambitious, acrylonitrile production became the priority.

Sohio's process economics for acrylonitrile were so positive that the decision was made to proceed with commercialization even though early market development efforts were discouraging. Major users were unsure that Sohio acrylonitrile would satisfy their needs. One major chemical company declined an opportunity for a joint venture. Another company announced plans for a new 100-million-pound-per-year acrylonitrile plant based on the old acetylene technology, at a cost of $100 million.

Still, Sohio commissioned the design of a detailed acrylonitrile plant. A pilot plant was constructed under the direction of Gordon G. Cross at Sohio's new laboratory in Warrensville Heights, a Cleveland suburb, where Ernie Milberger was instrumental in designing large laboratory-scale reactors and obtaining process design and development data from them.

In a bold move, it was decided to design the commercial plant on the basis of bench-scale laboratory development data rather than wait for pilot plant results. The time gained by eliminating this stage of development offset the added risk. Milberger's bench-scale unit, which required about four pounds of catalyst, generated the key data for the design of commercial reactors holding 40 tons.

By early 1958, the commercial design was going forward under the direction of Edward F. Morrill; a pilot plant was in operation; the catalyst was in final development by Callahan and his team with provisions for large-scale manufacture; and advancement work on reactor operation, product purification, and waste disposal was being coordinated. A key innovation was the successful development of a fluidized bed catalyst to allow for removal of the heat produced by the ammoxidation reaction.

By mid-winter 1959-60, the Lima, Ohio, plant, which cost $10 million to build, was complete. In less than four years since the discovery of bismuth phosphomolybdate as the direct propylene oxidation catalyst and the discovery of propylene ammoxidation, a full-scale commercial plant designed to produce 47.5 million pounds of acrylonitrile per year was ready to go.

There was but one challenge left -- an economic one. Soon after Sohio's entry, a major manufacturer cut its price in half. Sohio met the lower price and still managed to make a profit. The competitor scrapped its own expansion plans and took a license from Sohio. Other acrylonitrile producers soon became licensees of the Sohio process, and within a few years, acetylene-based acrylonitrile production had been replaced by the Sohio process.

To gain a larger share of the overall market, Sohio decided to promote the licensing of the process rather than keep the manufacturing to itself. Sohio's license to The People's Republic of China in 1973 was the first transaction by an American company after China opened its doors to U.S. investment. Today, following Sohio's lead, BP has licensed 42 companies to produce acrylonitrile in 77 plants in 21 countries. Annual worldwide production of acrylonitrile has grown from 260 million pounds in 1960 to more than 9 billion pounds in 1995.

Since 1960 BP Chemicals has developed and commercialized seven improved catalyst formulations, most of them based on the original bismuth phosphomolybdate catalyst. BP's current research focuses on further improvements to the Sohio Acrylonitrile Process and on new technology using the less expensive propane as feedstock.