From crude oil and natural gas to the C1-C4 building blocks: C1 chemistry: synthesis gas, methanol, ammonia and coal

4.1 Steam reforming turns methane into synthesis gas, and the mix is tuned to the product

Synthesis gas (syngas) is a mixture of carbon monoxide and hydrogen. Different uses need different ratios, Wittcoff explains. Methanol and Fischer-Tropsch synthesis need CO:H2 of 1:2. Hydroformylation needs 1:1. Ammonia needs N2:H2 of 1:3 and no CO at all.

The dominant route is steam reforming of methane. First the gas is desulphurised over zinc oxide at 360-400 C to below 2 ppm sulphur, because sulphur poisons the nickel catalyst. It is then mixed with 2.5-3.5 moles of steam per mole of methane. The mixture passes through catalyst-filled tubes at about 800 C and 35 bar: CH₄ + H₂O → CO + 3H2. The reaction absorbs about 227 kJ per mole. The heat comes from burners outside the tubes, so the reformer is a large furnace. About 7% of the methane passes through unchanged. Shreve (1984) gives a reformer range of 760-980 C. His typical product contains about 75% hydrogen, 8% CO and 15% CO₂.

The water-gas shift then adjusts the mix: CO + H₂O → CO₂ + H2. It releases a little heat and is favoured at lower temperature, so it runs at about 315-370 C over iron oxide. For ammonia, air is added in a second reformer to supply nitrogen. CO is shifted to CO₂, and the CO₂ is scrubbed out with an amine solution. For a CO-rich gas, CO₂ can be added to push the shift backwards.

Partial oxidation is the alternative. The feed is burned with about 35% of the oxygen needed for full combustion, at 1300-1400 C and 60-80 bar. It accepts heavy oils and needs no desulphurisation, but it needs a costly oxygen plant.

• Wittcoff (2004), Ch 10 sections 10.4-10.4.5 · pdf pp. 379-384
• Shreve (1984), Ch 7 Industrial Gases (hydrogen) · pdf pp. 125-127

4.2 Methanol: a low-pressure copper catalyst, a big recycle loop and giant plants

Methanol is made by CO + 2H2 → CH₃OH over a catalyst. The reaction gives off heat and the gas shrinks from three moles to one. So the equilibrium favours methanol at low temperature and high pressure. Shreve's table shows the effect. At 300 C, one-pass equilibrium conversion rises from 8% to 62% as pressure is raised sixfold, from 500 to 3,000 in the table's units.

The industry history is a story of catalysts allowing gentler conditions. BASF's original process used zinc-chromium oxide at 320-380 C and 340 bar. In 1972 ICI introduced a low-pressure process using copper-zinc on alumina at 240-260 C and 50-100 bar. Shreve describes a Lurgi low-pressure plant at 5-8 MPa and 250-260 C. The copper catalyst is more active and selective, giving purer methanol, but it needs syngas almost free of sulphur and chlorine. Lower pressure saves a great deal of compression energy and allows centrifugal compressors. Because conversion per pass is limited, unreacted gas is recycled many times. A small purge removes inert gases. Crude methanol is distilled in two columns. The main by-product is dimethyl ether.

The economics are dominated by the front end. Shreve states that making the syngas accounts for nearly 75% of the final cost of methanol. That is why licensors developed "megamethanol" plants of 5,000-10,000 t a day. Wittcoff explains these as a way to use cheap stranded gas at remote sites and still ship the product to distant markets.

Uses: the IEA (2018) says about 40% of methanol becomes formaldehyde. Wittcoff puts formaldehyde at about 35% excluding MTBE, followed by acetic acid.

• Wittcoff (2004), Ch 10 section 10.5.2 · pdf pp. 388-389
• Shreve (1984), Ch 38 Petrochemicals (methanol) · pdf pp. 781-782
• Brennan (2020), Ch 4, Table 4.2 · pdf p. 106
• Kent (2012), Ch 20 Natural Gas · pdf p. 829
• IEA (2018), Ch 1 · pdf p. 24

4.3 Ammonia, formaldehyde and acetic acid: what syngas and methanol become

Wittcoff calls ammonia by far the most important chemical made from syngas. It consumes about 5% of the world's natural gas, against about 1% for methanol. In the Haber-Bosch process, N2 + 3H2 react over promoted iron oxide at about 450 C and 250 bar. Only about 10% converts per pass, so most of the gas is recycled. Early plants ran at 1,000 bar with reciprocating compressors. Modern plants accept lower pressure because centrifugal compressors driven by waste-heat steam are cheaper and need less maintenance. That switch let plant size grow from about 200 t a day in 1960 to 1,500-1,850 t a day in the early 2000s. Ammonia links back to carbon chemistry too. Its process CO₂ is fed to the urea plant next door: 2NH3 + CO₂ → urea + water. The IEA (2018) says more than half of ammonia becomes urea and about 80% of ammonia use is agricultural.

Formaldehyde is methanol's largest outlet. In the main route, methanol vapour and air pass over a silver catalyst at about 700 C and near atmospheric pressure. Partly it is dehydrogenated and partly oxidised, so the heat roughly balances. Molar yields are 83-92%. A second route uses iron-molybdenum oxide at 300-400 C. The product is sold mostly as a 37% water solution. It goes into urea-formaldehyde resins for particle board, phenolic resins and polyacetal plastics.

Acetic acid shows how a catalyst can reshape an industry. In the Monsanto process, methanol and CO combine (CH₃OH + CO → CH₃COOH) over an iodide-promoted rhodium catalyst at about 200 C. Selectivity is over 99%. BP's Cativa process uses iridium at lower water content, cutting energy and capital. By the early 2000s about 88% of US acetic acid came this way.

• Wittcoff (2004), Ch 10 sections 10.4.5-10.5.2.2 · pdf pp. 383-393
• IEA (2018), Ch 1 · pdf pp. 16, 24
• Brennan (2020), Ch 4, Tables 4.2 and 4.7 · pdf pp. 106, 126

4.4 Coal can replace gas through gasification, and China does so at scale

Coal was the chemical industry's first raw material. Heating coal to about 1,000 C without air in a coke oven gives, by weight, about 80% coke, 12% coke oven gas, 3% tar and 1% light oil. The light oil is mostly benzene, toluene and xylenes. Wittcoff notes that coal tar supply is tied to steel's need for coke, so it cannot grow to replace petrochemicals.

The larger route is gasification. Coal reacts with steam and oxygen to give syngas (C + H₂O → CO + H2), and the syngas is then used exactly as gas-based syngas would be. The hard part is handling solids. Gasifiers use moving beds (Lurgi), fluid beds (Winkler) or entrained flow (Texaco, Shell, Koppers-Totzek). Their product gases range from H2/CO of about 0.5 to 1.7. The IEA (2018) adds that coal-based ammonia equipment costs more than twice as much as gas-based. Even the best coal plants use about twice the process energy of gas plants.

China makes these routes pay because it has abundant cheap coal and little gas rich in ethane. The IEA (2018) gives China 20% of world HVC capacity, 33% of ammonia and 55% of methanol, with coal dominant for both ammonia and methanol. About 20% of Chinese methanol uses coke oven gas. Methanol-to-olefins (MTO) is commercial only in China. Those plants sit inland near the coal, while naphtha crackers sit on the coast beside refineries. Before the 2014 oil price slump, Chinese MTO undercut the cheapest naphtha cracker by nearly $300 per tonne of HVC. By 2017 that discount had almost vanished. Kent (2012) records how unstable coal-to-liquids policy was. In 2008 China's planners required approval for all coal-to-liquids projects except two run by Shenhua.

• Wittcoff (2004), Ch 12 Chemicals from Coal · pdf pp. 425-434
• Wittcoff (2004), Ch 10 section 10.5.2.5 · pdf pp. 397-398
• IEA (2018), Ch 2 · pdf pp. 31, 33, 40-41, 44, 46
• Kent (2012), Ch 19 Coal Technology · pdf pp. 777-778, 806-807

See it in the table

Check your understanding

Steam reforming of methane is run at about 800 C in externally fired tubes. Why so hot?

Brennan gives 0.71 t of methane per tonne of methanol. Stoichiometry needs 0.5 t. What explains most of the gap?

Why can China profitably make methanol and olefins from coal when most of the world uses gas or naphtha?