The chemical industry: what it makes and how it is organised: A handful of raw materials and building blocks

3.1 Nine raw materials supply almost everything the industry makes

Oil and natural gas dominate. Wittcoff estimates that about 95% by weight of the world's organic chemicals come from them, and the IEA says oil and gas make up about 90% of all chemical feedstock today, with coal and biomass supplying the rest. Coal, the industry's base until the Second World War, now matters mainly as a source of synthesis gas, except in China, where it is the main feedstock for ammonia and methanol.

Five mineral and natural raw materials feed the inorganic side. Salt, with more than 14,000 recorded uses, is the starting point for chlorine, caustic soda, soda ash and sodium chlorate; world output was about 270 million tonnes in 2010, and the US chemical industry took 40% of US consumption. Sulphur now comes almost entirely from cleaning oil and gas: of 48.7 million tonnes of elemental sulphur in 2009, 22.7 million came from oil and 23.3 million from gas, and only 0.3 million was still mined. Phosphate rock, an apatite mineral, is treated with sulphuric acid to make wet-process phosphoric acid, mainly for fertiliser. Air is separated by cryogenic distillation into nitrogen, oxygen and argon, with commercial oxygen production dating from 1902. Water supplies steam and, through electrolysis, hydrogen.

Biomass completes the list. Fats, oils and carbohydrates account for only a few per cent by weight of chemical output, but sugar and starch are the usual food for industrial fermentation, which makes citric acid, amino acids, enzymes and antibiotics. Modules 13 and 14 ask how far biomass could go.

• Wittcoff (2004), Ch 2 Chemicals from Natural Gas and Petroleum · pdf pp. 83-84, 86-87
• IEA (2018), The Future of Petrochemicals · pdf pp. 11, 27-28, 30-31
• Kent (2012), Ch 26 Salt, Chlor-Alkali, and Related Heavy Chemicals · pdf pp. 1021-1022
• Kent (2012), Ch 25 Sulfur and Sulfuric Acid · pdf pp. 1001-1002
• Kent (2012), Ch 23 Phosphorus and Phosphates · pdf p. 945
• Kent (2012), Ch 27 Industrial Gases · pdf p. 1043
• Kent (2012), Ch 30 Industrial Biotechnology: Discovery to Delivery · pdf pp. 1132-1133

3.2 Nine building blocks carry most of the industry's tonnage

Wittcoff organises the whole of industrial organic chemistry around seven basic chemicals or groups made from oil and gas. Three are olefins: ethylene, propylene and the C4 olefins (butadiene, isobutene and the butenes). Three are aromatics: benzene, toluene and the xylenes. The seventh is the simplest alkane, methane, which is turned into synthesis gas, a mixture of carbon monoxide and hydrogen, and from there into ammonia and methanol.

The IEA uses a slightly different seven, which it calls primary chemicals: ammonia, methanol, ethylene, propylene, benzene, toluene and mixed xylenes. Put the two lists together and you get nine building blocks: ethylene, propylene, C4 olefins, benzene, toluene, xylenes, methane via synthesis gas, ammonia and methanol. Their scale is enormous. In the IEA's figures, world ammonia output is about 185 million tonnes a year, methanol about 100 million, ethylene and propylene together about 255 million and the three aromatics about 110 million.

Each building block has one or two dominant uses. More than half of ammonia becomes urea, and agriculture takes about 80% of ammonia overall. About 40% of methanol becomes formaldehyde. The largest derivative of ethylene is polyethylene and of propylene is polypropylene. Wittcoff points out how many large chemicals exist mainly to make one other: ethylene dichloride to make vinyl chloride, ethylbenzene to make styrene, p-xylene to make terephthalic acid, cumene to make phenol and acetone, and cyclohexane to make the nylon raw materials adipic acid and caprolactam. Alongside the organic trunk runs an inorganic one built on sulphuric acid, chlorine and caustic soda, phosphoric acid and the air gases.

• Wittcoff (2004), Ch 0 How to Use Industrial Organic Chemicals · pdf pp. 28-29
• Wittcoff (2004), Ch 1 The Chemical Industry · pdf pp. 59, 77-78
• IEA (2018), The Future of Petrochemicals · pdf pp. 24-25
• Kent (2012), Ch 22 Synthetic Nitrogen Products · pdf p. 910

3.3 Feedstock choice sets a plant's cost, its product mix and its location

For the trunk of the tree, the IEA finds feedstock the most important cost. Ethane is the main feed for olefins in North America and the Middle East, naphtha in Europe and Asia. Natural gas is the main input for ammonia and methanol everywhere except China, where coal dominates. Feedstock can be anything from 15% to 85% of the levelised cost of high-value chemicals (HVCs: ethylene, propylene and the aromatics), depending on the route. Crude oil costs much the same everywhere, but gas prices differ by up to four times between regions.

The chemistry matters as much as the price. A steam cracker heats a hydrocarbon briefly to very high temperature so that its molecules break into smaller, unsaturated ones. Ethane needs only about 1.2 tonnes of feed per tonne of HVCs, and an ethane cracker yields about 50% more HVCs than a naphtha cracker, but almost all of it is ethylene. Naphtha, a heavier liquid, gives a balanced mix of ethylene, propylene and aromatics at a lower overall yield. A region that cracks ethane therefore has cheap ethylene but must find propylene and aromatics elsewhere, for example by propane dehydrogenation.

Capital costs also differ by route. The IEA assumes about $1,500 of capital per tonne of annual HVC capacity for an ethane cracker, $2,050 for a naphtha cracker and $1,000 for a methanol-to-olefins unit. Wittcoff shows how extreme the feedstock gap could be: in 2001 Saudi Arabia extracted ethane from gas that had no other use at $0.50-0.75 per million British thermal units, while US gas briefly reached $10.

• IEA (2018), The Future of Petrochemicals · pdf pp. 31-33
• Wittcoff (2004), Ch 1 The Chemical Industry · pdf p. 57

3.4 Bigger plants make cheaper tonnes, which drives a race to scale

Commodity chemicals are made in huge continuous plants because cost per tonne falls steeply with size. Wittcoff explains the geometry. The capacity of a storage tank or distillation column depends on its volume, which rises with the cube of its linear size, but its cost depends on the steel needed to enclose that volume, which rises with the square. So cost rises roughly with capacity to the power two-thirds. Heat exchangers do not follow this rule, since their capacity depends on surface area. For a whole petrochemical plant, cost is said to rise with capacity to the power 0.6.

The arithmetic is powerful. Doubling capacity raises the plant's cost by 2 to the power 0.6, about 1.52 times, so the capital per tonne falls by about 24% (1.52 / 2 = 0.76). A plant ten times larger costs about 4 times as much, so capital per tonne falls by 60%. Labour is a small share of cash costs, about 2.5% for high-density polyethylene, 3.5% for benzene, 5.7% for terephthalic acid and 7.3% for acrylonitrile, and doubling a unit may raise its labour cost by only 10-20%.

The result was a steady escalation in size. A typical ethylene cracker grew from 70 million pounds a year in 1951 to 1 billion pounds in 1972, about 450,000 tonnes. In 2001 BASF and Fina started a 2 billion pound plant at Port Arthur, Texas. Scale has a dark side: high fixed and low variable costs make it rational to keep running at low prices, and capacity arrives in big lumps. Part 4 shows the result.

• Wittcoff (2004), Ch 1 The Chemical Industry · pdf pp. 58-59
• Wittcoff (2004), Ch 0 How to Use Industrial Organic Chemicals · pdf p. 28
• Pollak (2011), Ch 2 The Fine Chemical Industry · pdf p. 35

See it in the table

Check your understanding

Where does almost all of the world's elemental sulphur come from today?

If plant cost rises with capacity to the power 0.6, doubling capacity changes capital cost per tonne by about:

Why do ethane crackers dominate in the United States and the Middle East but not in Europe or China?