Colour and protection: dyes, pigments, paints and coatings: Pigments: titanium dioxide, carbon black and organic pigments

4.1 Hiding power comes from refractive index and particle size

A white paint hides because its pigment particles bend light much more than the binder around them, so light entering the film is scattered back before it reaches the surface beneath. Kent explains that hiding power depends on the difference in refractive index, a measure of how strongly a material bends light, between pigment and binder. Titanium dioxide's is 2.70 to 2.76 depending on the source, well above zinc oxide (2.08) or white lead (2.0). Particle size matters too. Kent puts the optimum at 0.2 to 0.4 micrometres, about half the wavelength of light, where scattering peaks; titanium dioxide pigment is made at 0.2 to 0.3 micrometres.

Shreve rates relative hiding power at 125 to 135 for rutile, the crystal form used in almost all paint, 100 for anatase, 20 for zinc oxide and 15 for white lead. Yet 1981 prices were close: $1.65 per kg for rutile, $1.54 for white lead. Dividing price by hiding power, rutile hid at about an eighth of white lead's cost, before counting lead's toxicity. Kent notes that a little titanium dioxide gives good opacity without adding much to paint cost.

Extenders are the opposite: cheap minerals with low refractive index, such as calcium carbonate (whiting), talc, china clay, barytes and mica. They hide poorly but cut raw material cost, control sheen and texture, and help stop settling, so they are used alongside hiding pigments.

Coloured inorganic pigments complete the palette. Kent calls iron oxides low-cost, high-opacity, durable and easy to disperse, used for reds, yellows, browns and primers. Chrome yellows and greens are bright but restricted for their lead, and zinc and lead chromates protect steel from corrosion.

• Kent (2012), Ch 29 Pigments, Paints, Polymer Coatings, Lacquers, and Printing Inks · pdf pp. 1128-1129
• Shreve (1984), Ch 24 Surface-Coating Industries · pdf pp. 449-454

4.2 Titanium dioxide by the sulphate route: cheap ore, heavy waste

The sulphate route is the older of the two, and Shreve says it uses the cheaper ore, ilmenite (FeO.TiO₂, an iron titanium oxide). Dryden lists the steps: digest in sulphuric acid, remove iron as ferrous sulphate, concentrate in vacuum evaporators, precipitate titanium hydrate, then calcine, grind and bag.

In digestion, ground ilmenite (47% TiO₂ in Shreve's flowsheet) reacts with 96% sulphuric acid in a bank of 22 tanks: TiO₂ + H₂SO₄ → TiOSO₄ + H₂O, while the iron forms iron sulphate. Scrap iron is added as a reducing agent to keep the iron ferrous. The liquor is clarified and cooled in a crystalliser, where much of the iron drops out as green crystals of copperas (FeSO₄.7H2O). The titanyl sulphate is then hydrolysed by boiling: TiOSO₄ + H₂O → TiO₂.xH2O + H₂SO₄. Seed crystals set the crystal form; Shreve gives 6 hours of boiling with anatase seeds or 3 with rutile seeds. The hydrate is filtered, washed and calcined; anatase converts to rutile at 700 to 950 C.

The chemistry dictates the waste. A tonne of TiO₂ (molar mass 80) is 12.5 kmol, which takes at least 12.5 kmol of acid, 1.23 t, to dissolve. Pure ilmenite brings one iron atom per titanium, tying up as much acid again. So the minimum is roughly 2.5 t of sulphuric acid per tonne of pigment, and the iron can leave as up to 3.5 t of copperas (12.5 kmol at 278 kg each). The acid freed in hydrolysis emerges as spent acid of 15 to 22%, too weak to reuse cheaply. Shreve adds that violent batch digestion threw acid mist into the air, and that NL Industries' continuous process uses 25 to 60% acid to tame the reaction and recycle more of it.

• Shreve (1984), Ch 24 Surface-Coating Industries · pdf pp. 451-452
• Dryden (1997), IIIC Paints and Varnishes (including Pigments) · pdf p. 322

4.3 The chloride route, and why India stayed mostly with sulphate

The chloride route treats rutile ore, natural or synthetic, with chlorine and coke to form titanium tetrachloride, TiCl₄, a liquid at room temperature. Any iron leaves as volatile ferric chloride and is separated. The TiCl₄ is distilled to purify it, then burned with oxygen in a flame at about 1,500 C: TiCl₄ + O2 → TiO₂ + 2 Cl2. Out come very fine titanium dioxide and chlorine, which goes back to the chlorinator. Dryden calls the burner engineering formidable.

Because chlorine is recycled, only make-up chlorine is consumed, and Shreve says the wastes are easier to dispose of than the sulphate route's. Dryden adds that the flame process gives a superior pigment and is potentially cheaper. By the early 1980s it made over 60% of US output. Its handicap is ore: it needs the dearer rutile, although ilmenite can be upgraded to synthetic rutile and DuPont patented a chloride process that uses ilmenite.

Scale and access to technology decided the choice. Dryden puts the international minimum economic size at 50,000 tonnes a year and says chloride technology was hard for new entrants to obtain for medium projects of 15,000 to 20,000 tonnes, while several firms offered sulphate technology. In the mid-1990s India had four producers with 52,860 tonnes of capacity. Three used the sulphate route and made anatase; only Kerala Minerals and Metals, at 22,000 tonnes, made rutile by chloride. Every plant was below the 50,000-tonne benchmark. There was a twist: India imported the sulphur needed for acid, so assembling raw materials for the sulphate route cost about 11% more than for chloride.

Demand follows paint. Shreve gives 1981 demand of 750,000 tonnes, half into paints, varnishes and lacquers and 23% into paper. Dryden gives paints and coatings 68% of Indian use against 51% in the United States.

• Shreve (1984), Ch 24 Surface-Coating Industries · pdf p. 451
• Dryden (1997), IIIC Paints and Varnishes (including Pigments) · pdf pp. 322-323

4.4 Carbon black and organic pigments: two very different businesses

Carbon black is made by burning oil without enough air. In the oil furnace process, which Shreve says made about 90% of the world's carbon black by the early 1980s, a heavy aromatic oil (usually catalytic cracker residue) is sprayed into a turbulent flame of gas or oil burned with about 50% excess air. Too little oxygen is left to burn the feed, so within microseconds it cracks to carbon at 1,400 to 1,650 C in an alumina-lined reactor. Water sprays 1 to 3 metres downstream quench the smoke to about 200 C, and bag filters recover essentially all the black. Yield on carbon runs from 65 to 70% for coarse grades to about 35% for the finest. Particle size (18 to 75 nanometres) and 'structure', the way particles fuse into chains, are set by nozzle design, temperature and residence time.

Shreve reports that about 94% of US carbon black went into rubber, two-thirds into tyres, at 70 to 84 cents per kilogram in 1981. Pigment grades are a small slice, but Kent calls carbon black probably the most widely used pigment in coatings, from jet car topcoats to conductive primers, and Shreve says most printing inks are based on it.

Organic pigments are a world of small batches, many shades and high value, mostly made with the azo chemistry of Part 2. Shreve's examples: para red couples diazotised p-nitroaniline with beta-naphthol; toluidine red couples m-nitro-p-toluidine with beta-naphthol; Hansa yellow G couples the same amine with acetoacetanilide. Without solubilising groups, or precipitated as metal salts, the products are insoluble. Copper phthalocyanine, from phthalic anhydride or phthalonitrile and a copper salt, gives the most stable blues and, chlorinated, greens. Kent notes organic pigments cost more than inorganic ones, partly offset by higher tinting strength.

• Shreve (1984), Ch 8 Industrial Carbon · pdf pp. 149-153
• Kent (2012), Ch 29 Pigments, Paints, Polymer Coatings, Lacquers, and Printing Inks · pdf pp. 1125, 1129-1130
• Shreve (1984), Ch 24 Surface-Coating Industries · pdf pp. 452-453, 455-456
• Shreve (1984), Ch 39 Cyclic Intermediates and Dyes · pdf pp. 791, 811
• Dryden (1997), IVF Chemicals from Aromatics (including Dyes) · pdf p. 553

See it in the table

Check your understanding

Why does rutile titanium dioxide hide so much better than white lead?

Which waste stream is characteristic of the sulphate route but not the chloride route?

Why did most Indian titanium dioxide capacity in the 1990s use the sulphate route?