Industries / Mining and materials / Chemicals and petrochemicals / Refining
Cracking and conversion
Cracking and conversion break large, heavy molecules into smaller ones or reshape them, so that a refinery gets more gasoline, diesel, jet fuel and chemical feedstock from each barrel.
7
Milestones achieved, from 1913 to 2024.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1913
Heat cracks heavy oil
Standard Oil of Indiana started the first two rows of pressure stills at its Whiting, Indiana refinery in 1913. Heat and pressure broke heavy oil into lighter molecules, which the company's magazine said in 1919 gave "enormously more gasoline than is present in the natural crude oil." Penn State's refining course says thermal cracking made about 20% of US gasoline through 1924 and over 50% by 1940.
Credit
William Burton and Robert Humphreys of Standard Oil (Indiana), who began the work in 1908
1942
A catalyst that keeps moving
PCLA No. 1 went on stream at Standard Oil of New Jersey's Baton Rouge refinery on 25 May 1942. The North American Catalysis Society calls it the first commercial circulating fluid-bed reactor: a fine catalyst powder flows like a liquid between the reactor, where it cracks gas oil, and a regenerator, where the coke that builds up on it is burned off. The unit could run without stopping, and PCLA No. 1 ran for 21 years before it was taken down in 1963.
Credit
Warren K. Lewis and Edwin Gilliland of MIT; Donald Campbell, Homer Martin, Egar Murphree and Charles Tyson of Standard Oil Development Co
1958
Hydrogen joins cracking
Hydrocracking breaks heavy oil over a catalyst in a stream of hydrogen, which turns heavy, aromatic feeds that a catalytic cracker handles poorly into diesel and jet fuel. Penn State's course says Chevron started the first modern commercial unit in 1958. Chevron Lummus Global, the company that now licenses Chevron's version, says the first such unit licensed to a US refinery started in 1962. Most US units through the 1990s were set up to make gasoline, while buyers elsewhere wanted diesel and jet fuel.
Credit
California Research Corporation, the research arm of Standard Oil of California, now Chevron
1964
Zeolites in the catalyst
Charles Plank and Edward Rosinski of Socony Mobil filed a patent in 1960, granted on 7 July 1964, for a cracking catalyst containing a zeolite, a crystalline mineral whose pores are about the size of small molecules. The National Inventors Hall of Fame says it was the first cracking catalyst to contain a crystalline zeolite, that it raised the gasoline yield from crude oil sharply, and that it is used in virtually all US cracking units. Penn State's course dates the start of zeolite catalysts in fluid catalytic cracking to 1965.
Credit
Charles J. Plank and Edward J. Rosinski, Socony Mobil Oil
2014
Crude straight to a cracker
ExxonMobil formally opened a steam cracker in Singapore with a capacity of 1 million tonnes a year that can crack crude oil directly. A steam cracker heats hydrocarbons with steam until they break into ethylene and propylene, the building blocks of most plastics, and in Asia it normally runs on naphtha made at a refinery. Reuters reported it as the world's first such unit; it had come online in mid-2013, and ExxonMobil gave no figures for savings or the crudes it ran.
Credit
ExxonMobil Chemical
2024
A refinery turned to fats
Phillips 66 finished converting its Rodeo refinery in California to make only biofuels, chiefly renewable diesel, made by treating fats and vegetable oils with hydrogen. EIA counts the plant's capacity at 767 million gallons a year, up from 180 million the year before, the second-largest renewable diesel plant in the US. It can switch about 150 million gallons a year of that to sustainable aviation fuel.
Credit
Phillips 66
2024
Co-processing stops
In the same year Monroe Energy and Chevron stopped co-processing renewable diesel, that is, running fats and oils through existing refinery units alongside petroleum, at Trainer, Pennsylvania and El Segundo, California. Vertex Energy and Jaxon Energy closed renewable diesel plants in Mobile, Alabama and Jackson, Mississippi. EIA ties the losses to falling biofuel credit values from 2023, as output passed government targets, and to stronger petroleum refining margins.
LessonRefinery conversions to renewable fuels follow credit prices and refining margins, and can be reversed within a few years.
Credit
US Energy Information Administration, for the analysis
Watch for
A refinery built for chemicals
Huajin Aramco Petrochemical, a joint venture of Chinese companies and Saudi Aramco, received its first cargo of Arabian crude at its complex in Liaoning province, China, and said this brought the project closer to operation. The complex is planned to include a refinery for 300,000 barrels of crude a day, with plants for 1.65 million tonnes a year of ethylene and 2 million tonnes a year of paraxylene, the raw material for polyester. Media had reported that start-up slipped from May or June to September or October.
Achieved whenthe complex runs crude through its refinery and crackers, and an outside party, such as Chinese customs data or a trade publication, reports its ethylene and paraxylene output.
Claimed by
Huajin Aramco Petrochemical Co., October 2026
Open
01
Conversion units, openly listed
A free table of the world's refineries giving, for each, the capacity of its crude unit and of its main conversion units (fluid catalytic cracker, hydrocracker, coker and reformer), its status (running, idled, closed or converted to biofuels), and a source for every number.
The briefCredit · unclaimed02
Crude-to-chemicals claims, one basis
An open model that takes a published crude assay and estimates what share of each barrel ends up as base chemicals (ethylene, propylene, butadiene, benzene, toluene and xylenes) under three designs: a conventional refinery tied to a steam cracker, a crude-to-chemicals complex built around paraxylene, and a design that cracks light crude almost directly.
The briefCredit · unclaimed
Open 01 · the brief
Conversion units, openly listed
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A free table of the world's refineries giving, for each, the capacity of its crude unit and of its main conversion units (fluid catalytic cracker, hydrocracker, coker and reformer), its status (running, idled, closed or converted to biofuels), and a source for every number.
Why it is still open
EIA publishes unit-level capacity for each US refinery, but only for the US. Worldwide unit-level databases are sold by market-intelligence firms such as IIR Energy. Global Energy Monitor's free Global Chemicals Tracker covers plants that make eight base chemicals, not refinery units. A search found no free worldwide table at the level of units.
Built with what exists
EIA's refinery capacity report as a checked US base; company annual reports, licensors' press releases and government statistics; a web agent to find the pages, such as Browser Use 1.0 (see the Note "Deleting 80% to start over"); a language model that drafts each row and quotes the sentence each number came from (see the Note "Findings that cite notes"); two people who check a sample; CSV files on GitHub.
First step
Build rows for the twenty largest refineries outside the US from their owners' own reports, and note which units no public source covers.
Done looks like
Every refinery outside the US with a crude unit of at least 100,000 barrels a day listed, with a source for each capacity; a random sample of 100 numbers checked by a second person, at least 95 of them matching their source; US rows matching EIA's report exactly; and a change log that records each closure or conversion with its date.
Open 02 · the brief
Crude-to-chemicals claims, one basis
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An open model that takes a published crude assay and estimates what share of each barrel ends up as base chemicals (ethylene, propylene, butadiene, benzene, toluene and xylenes) under three designs: a conventional refinery tied to a steam cracker, a crude-to-chemicals complex built around paraxylene, and a design that cracks light crude almost directly.
Why it is still open
The published figures are on different bases. A 2026 industry analysis quotes Wood, the engineering firm, as putting the petrochemical yield of a conventional integrated refinery at 15 to 20% and of a crude-to-chemicals refinery at up to 40%, and quotes Lummus as saying its TC2C process reaches 70% by mass from light crudes. IHS Markit, now part of S&P Global, compared four such designs in a paid report in December 2021 and put their conversion to chemicals at 27 to 76% of the crude fed, but its models are not public. The University of Calgary's free PRELIM model covers refineries that make fuels, and its documentation does not mention steam crackers. A literature search through Crossref found no open model that runs these routes on the same crudes.
Built with what exists
Public assays from the ADIOS database, Equinor or ExxonMobil; published yields for each unit from textbooks and papers; DWSIM, a free, open-source process simulator, for the separation steps; Python; a coding agent that tests its own work (see the Note "Checked without being asked").
First step
Take five public assays and compute the share of each crude that boils in the naphtha range, the usual steam-cracker feed.
Done looks like
The three designs run on five public crudes; every mass balance closing within 1%; every yield assumption cited; results and code public; and a written review by an outside refining engineer published with them.