Cement Kilns

White Cement Manufacture

Snowcrete LogoAPCM's Snowcrete Brand.
Tunnel White LogoTunnel White Brand.

In properties, White Portland cement differs from ordinary Portland cement only in its light colour. However, the details of its formulation and manufacture are quite distinctive. Its position in the context of British cement history is rarely considered, but its story is that of British Portland cement in general, writ large. No white Portland cement is currently made in Britain or Ireland.

Raw Materials

White clinker contains the same phase assemblage as grey clinker - alite, belite, tricalcium aluminate, tetracalcium aluminoferrite and a little free calcium oxide - but the tetracalcium aluminoferrite, which is the only dark coloured phase, is minimised. In practice, the amount of this phase in clinker is roughly proportional to the amount of iron (plus titanium, chromium, manganese, nickel and zinc) in the composition, so to make white clinker, it is necessary to minimise the amounts of these transition elements in the rawmix. Iron oxide, which is by far the most common, has to be controlled to below 0.4% in the clinker, while there are more exacting requirements for the more strongly coloured elements: Mn2O3, which imparts a grey cast, should be below 0.03%, while the intensely green Cr2O3 should be below 0.003%. Precautions have to be taken to minimise contamination with dark materials at all stages of the process.

From the raw material point of view, the most obvious problem is the availability of iron-free limestone. This needs to have less than 0.1% Fe2O3, and preferably half that. In many parts of the world, such white limestones are very hard to find, but most clean Upper Chalk easily meets this requirement, so in Britain, the limestone component is not a problem. All British sites that have made white clinker have used Upper Chalk, although there also exist a number of Carboniferous Limestone quarries that can deliver suitable purity.

Because Portland clinker was traditionally made from chalk and clay, it would appear that it remains only to find a low-iron clay. In order to meet the clinker iron content criterion, the rawmix must contain less than 0.25% Fe2O3, and even with chalk containing 0.05%, the balance of the mix must not exceed 0.95% Fe2O3. Various grades of china clay and ball clay are available in Britain, although a certain amount of transport cost is involved in bringing them to a chalk area. Naturally, the higher-grade, low-iron product is the highest priced, although clays rendered unusable by high organic content can be used satisfactorily in cement mixes. Generally speaking, as the purity increases, so the composition of the clay approaches that of kaolinite - Si2Al2O5.(OH)4. This mineral contains far too little silica for a cement rawmix, so a further addition of sand is needed. There are a few sources of low-iron, high-silica kaolins, but these are of value only if the quartz component is below 45 μm. Spent low-iron alumina catalyst has also been used as an alumina source.

Sand for use in rawmix needs to have Fe2O3 below 0.2%, and preferably half that. This criterion alone is not too difficult to meet - however, the iron that contaminates sand tends to consist of a hydrated oxide film on the surface of the sand grain, so that the amount increases as the sand gets finer, and very fine sands of low iron content are hard - or impossible - to find. As mentioned elsewhere, sand must be ground to below 45 μm for effective chemical reaction in the kiln, and grinding sand is difficult. In conventional grinding equipment when grinding sand, iron is abraded from grinding media and mill linings in large quantities. If abrasion-resistant chrome steel media is used, the contamination, although much less, consists of the much more deleterious chromium. Very often historically, the compromise reached has been to leave the silica relatively coarse, and accept the very high kiln fuel consumption and low kiln output that results. However, fine grinding of sand is possible, using non-metallic media and mill linings, provided that there is a commitment to investment in custom-designed plant.

The lack of iron in the formulation has the disadvantage that the amount of liquid formed in the burning zone is small. This means that the burning zone temperature has to be higher than on ordinary clinker production. It also means that the kiln does not form much coating, and this, combined with the higher processing temperature, leads to relatively high losses of heat by radiation from the kiln shell. The effect of lack of liquid is often counteracted by addition of a mineraliser, usually containing fluoride, such as fluorspar or cryolite. This, in combination with sulfate added to the rawmix or supplied by fuel, lowers the temperature at which alite forms, and is particularly effective in white clinker manufacture.

Fuel

Coal, in general, can't be used as a fuel for white clinker. Even in a "best possible" case, with a dry process kiln consuming 4 MJ/kg, and a high quality coal having 6% ash, 31.3 nett GJ/kg calorific value, and a low 8% Fe2O3 in the ash, the fuel would contribute 0.06% Fe2O3 to the clinker, which would put severe constraints upon the rawmix composition. A more realistic case with wet kilns might contribute 0.3%. Historically, heavy fuel oil was used. Gas can be used, although this involves a certain kiln output penalty because of its less luminous flame, and it is generally more expensive. More recently, petroleum coke has become the fuel of choice, contributing useful sulfur for mineralisation, although some grades can contain more than acceptable amounts of vanadium. Landfill gas has been used to some extent, although, being rather low-grade and lowering flame temperature, it compromises kiln output. Waste solvents can be used, provided that they don't contain chlorine.

Burning Process

In principle, any type of kiln used for grey clinker production can also be used for white. Internationally, wet, long dry, Lepol, suspension preheater and precalciner kilns are used for white production. Although the process is in general the same as for grey clinker, there are certain distinctive features.

The kiln has to minimise contamination with transition elements: in particular, on wet and long-dry process kilns, contamination by abrasion from chain heat exchangers has to be minimised, so that the potential for good heat recovery is reduced.

Because white clinker is low in iron, and is often low in alumina as well, a comparatively small amount of liquid is formed in the kiln's sintering zone. This has two effects:

  • a higher peak temperature and a longer sintering zone residence time is required to complete combination of free lime with belite to form alite.
  • much less coating is formed on the kiln refractories, so the sintering zone is less well insulated.

These have the effect that white kilns lose much more heat through the kiln shell. This effect can be minimised by using a precalciner process, where there is a small kiln surface area for a given output.

The most distinctive feature of most white kilns is regarding the way the clinker is treated as it leaves the kiln. In grey clinker manufacture, considerable care is taken to ensure that there are no "reducing conditions" (i.e. insufficient air to burn the fuel to CO2) in the sintering zone. In grey clinker, reducing conditions have two deleterious effects:

  • iron is reduced to the Fe2+ state, which leads to undesirable changes in the mineralogy of the clinker phases, and can cause complete disruption of the clinker if the iron subsequently re-oxidises.
  • sulfur is eliminated from the clinker as SO2, leaving alkalis in excess, and causing the formation of undesirable high-alkali aluminate.

However, in the case of white clinker, there is in any case very little iron and not much alkalis, and reduction of the small amount of iron to Fe2+ considerably lightens the clinker colour, provided that subsequent re-oxidation is prevented by very rapidly reducing the temperature of the finished clinker.

In practice, reduction is obtained using a "bleacher" flame directed onto the clinker bed at the kiln exit. The clinker is then immediately "quenched" to below 600°C. In older installations, the bleacher consists of a separate fuel pipe close to the clinker surface. Modern installations use the main burner with a wide flame and a restricted secondary air flow. The quencher may consist of a water bath in which the clinker is saturated before being removed by a screw. Generally speaking, the clinker then has to be passed through a drier. Alternatively, the clinker may fall through a curtain of water sprays, controlled so that the emerging clinker is at around 600°C. This has the advantage that no drier is needed, and some heat can be recovered to warm the secondary air. In either case, exhaust ducting has to be provided to ensure that the enormous amount of steam generated does not end up in the kiln.

Only four sites have made white clinker in Britain, and all were wet process. All are now extinct. A 2002 survey showed that, of European white clinker manufacturing capacity, only 28% was wet process. Britain and Ireland possess abundant raw materials suitable for white clinker manufacture by efficient modern processes.

Production History

In Britain, white Portland clinker has only ever been made at four sites. To the best of my knowledge, the product has never been made commercially in Ireland.

The first British manufacture was on the two small kilns at Beddington. These were due for closure, and were turned over to white manufacture in 1929, as a pilot project during the setting up of the dedicated white kilns at Swanscombe. The big kilns B1-B3 were installed at Swanscombe in 1929, making the old kilns A1-A14 redundant. Parts of the old kilns were re-assembled in the old kiln house as dedicated white kilns. The first two were commissioned in 1932, and the Beddington kilns were shut down shortly afterwards. Two more similar kilns were installed in 1937 and 1949, white manufacture having ceased during WWII. Finally, one of the big kilns (B1 or "S5") was converted to white production in 1958. During the boom years of the 1960s two of the old white kilns were occasionally used to supplement this, but were not used after 1970. Kiln B1 (or S5) continued in operation until 1990, when Blue Circle decided to supply its market with imports, which were by then, no doubt, much cheaper than Swanscombe's production. Blue Circle began its desultory commercial relationship with Aalborg (Denmark) in 1990. Aalborg had at the time the world's largest white cement plant.

The third plant to make white was the FLS plant at West Thurrock, which appears to have started up in 1934. This, it is believed, was done on kiln A3. Except for a wartime intermission, this continued until the kiln was replaced in ?1951. Production was transferred to Pitstone in 1957, made in intermittent "batch runs" on Kiln A1. It finally ceased in 1972, when it became clear that cement could be imported from the FLS plant at Aalborg more cheaply than home production.

© Dylan Moore 2011: last edit 28/06/14.