Cement Kilns

"Special" clinkers

Home About this site About cement List of plants Raw materials Kilns Other technical advances Trends in innovation Sources
About clinker Clinker mineralogy Clinker thermochemistry Clinker analysis

Although this page is often accessed as a stand-alone piece, it is part of a work on the history of the British and Irish cement industries, and where statements are made about the historical development of techniques, these usually refer only to developments in Britain.

Special Clinkers

This website addresses only the production of Portland clinker, which contains some combination of the four main phases mentioned in the page on clinker minerals. Within that definition, however, no particular distinction is made as to the precise nature of the clinker made. All clinkers are "special"! Clinkers that are at the extremes of the composition range may sometimes be more difficult and expensive to make, involving higher temperatures and greater energy consumption, but incompetently-made clinkers of "normal" composition can - and do - suffer similar difficulties. Having said this, the type of clinker made may have significant effects upon the operation of the plant, and needs to be considered.

Sulfate Resisting Clinkers

This, historically, has been the most commonly made "special" clinker. It was found, from early times, that concrete exposed to sulfate solutions would sometimes deteriorate by expansion, causing spalling of the surface, and eventually exposing the reinforcement, followed by rapid failure. Typically, this would take place in foundations exposed to sulfate-rich ground waters, or some marine structures.

Research during the 1920s showed that the problem was due to reaction with sulfate of the hydration products of tricalcium aluminate. Infiltrating sulfate ions react with "monosulfate" to form ettringite. A combination of expansion due to "imbibing" of a large amount of water, and orientated growth of the needle-like ettringite crystals causes disruption of the matrix.

Ca4(Al(OH)6)2.(SO4).6H2O + 2 CaSO4 + 20 H2O → [Ca3(Al(OH)6).12H2O]2.(SO4)3.2H2O

Having understood this mechanism, it became clear that to reduce the propensity of concrete to undergo this form of attack, the amount of tricalcium aluminate in the cement should be reduced. Sulfate resisting clinker began to be produced after WWII. The British Standard for sulfate resisting cement required that the calculated tricalcium aluminate content should be below 3%. A simple way of reducing the amount of this phase is to increase the amount of iron in the clinker, thus ensuring that alumina is tied up in the aluminoferrite phase, and the amount available for making tricalcium aluminate is reduced.

In practice, the simple addition of sufficient iron oxide to achieve this causes a considerable increase in the amount of non-strength-producing phases and consequent loss of strength. To counteract this, in most instances, sand was also added to the mix to restore the silicate content of the clinker. Mixes which include insufficiently ground quartz are difficult to burn, and for this reason sulfate resisting mixes gained a reputation for hard burning, high kiln fuel consumption and low kiln output. However, where the quartz component is properly ground, sulfate resisting mixes have no appreciable effect on kiln performance.

A typical good-quality sulfate resisting clinker might contain 76% alite, 5% belite, 2% tricalcium aluminate, 16 % tetracalcium aluminoferrite, and 1% free calcium oxide. Because it usually has a high ferrite content, it is dark in colour. Because alkali-bearing clay is displaced from the mix by sand and iron oxide, such clinkers usually have reduced alkali contents. Alkali contents are further reduced if the mix is hard to burn, because with higher burning zone temperature, the evaporation of alkalis increases.

Beginning just after WWII, production of sulfate resisting clinker increased to a peak in the 1970s, but more recently production has greatly decreased, because sulfate resistance is more easily obtained by use of granulated blast furnace slag as a cement component. This provides an excess of alumina which ensures that any infiltrating sulfate ions produce only "monosulfate".

Specially formulated SRPC clinkers have often been used as a basis for Oilwell Cement, used for sealing the linings in oilwells.

Plants that have produced SRPC clinker as part of their product mix have included Holborough, Humber, Johnsons, Ketton, Magheramorne, Masons, Pitstone, Platin, Rhoose, Ribblesdale, Rochester, Rugby, South Ferriby, Swanscombe, Weardale, Wilmington and Wouldham.

Low Heat Clinkers

In the late 1920s, the elucidation of the phases in clinker was paralleled by research into the heat evolved when cement hydrates. It was found that most of the heat evolved is produced by the hydration of tricalcium aluminate, and to a lesser extent, alite. At about this time, a number of big concrete dams were being constructed in the USA - notably the Hoover Dam on the Nevada-Arizona border. When large structures are made from normal concrete, the huge amount of heat produced during curing can't be dissipated, and as a result the concrete temperature rises to a point at which the structure is damaged. The latest hydration research was therefore used to produce cements that would minimise this effect, simply by reducing the tricalcium aluminate and alite content.

Beginning in 1932, large amounts of low heat cement were produced, mainly for specific dam projects, in the USA. An ASTM standard was introduced in 1934. A certain amount has been produced in Britain, but only ever for single projects. A rather poor British Standard specification was produced, but compliance with the ASTM standard was usually required. The mix would be made with an increased clay/shale content, some added iron oxide, and perhaps a little sand. As with sulfate resisting clinker, there is no longer any need to make low heat clinker, because composite cements made with ordinary clinker and a large addition of ground granulated blast furnace slag have excellent low-heat properties.

A typical low heat clinker might contain 29% alite, 54% belite, 2% tricalcium aluminate and 15 % tetracalcium aluminoferrite, with very little free lime.

White Clinkers

In contrast to the other "specials", white clinker has had a long and continuing history. It also differs in that manufacturing techniques can differ markedly from those of "grey" clinkers, and white clinker has been produced at very few British sites.

The usefulness of a white cement was recognised from the earliest times. White concrete can greatly improve the appearance of exposed architectural features. Coloured concrete can be made by addition of pigments: in ordinary grey concrete, pigments produce subtle pastel shades, but with white cement, vibrant colours can be obtained. Early attempts at burning "white" mixes always produced rather disappointing off-white shades, and the first seriously white product was probably produced in France in the early years of the 20th century. The first successful attempts in Britain were in the late 1920s, and production escalated rapidly in the 1930s. Production peaked around 1970, and diminished from there on. Although a market remains, white cement is more than others traded internationally, and dogged insistence on wet process production priced British cement out of the market, with production ceasing in 1990. White clinker can be made by dry process very efficiently.

There has never been a British Standard specification for white cement: it differs from ordinary cement only in its colour, and this property is easy to agree on an ad hoc basis between producer and customer without the need for a formal specification. To get a high degree of whiteness, the content of dark tetracalcium aluminoferrite must be minimised. In practice, because the majority of white cement goes into factory-made pre-cast concrete applications, a high-performance, high early strength cement is needed to maximise factory productivity. The composition of white clinkers varies widely, but a typical composition for a high performance clinker might be 76% alite, 15% belite, 7% tricalcium aluminate, no tetracalcium aluminoferrite, and 2% free lime.

Details of white clinker manufacture and history are given in a separate article.

Low-alkali Clinkers

The above "special" clinkers are the result of varying the amount of alumina and iron in the four-oxide system CaO / SiO2 / Al2O3 / Fe2O3. A further category has emerged involving the content of minor elements in the clinker. The alkali oxides Na2O and K2O (the only ones routinely measured) can have a deleterious effect when the cement is used with aggregate containing reactive silica. Alkali in solution in the cement paste attacks the aggregate surface, producing alkali silicate gels that cause expansive disruption of the paste/aggregate interface, causing cracking of concrete and exposure of reinforcement. The alkali content of clinker is conventionally given as "total alkalis as Na2O molar equivalent" (Na2Oeq), calculated as Na2O + 0.658 K2O. Early clinkers were low in alkalis because the alkali chlorides, hydroxides and sulfates are volatile. In static kilns and the early rotary kilns, alkali salts were evaporated in the burning zone and largely emitted to atmosphere. When rotary kilns started to be fitted with electrostatic precipitators from the 1930s onward, most of this alkali fume was caught as part of the "cement kiln dust" (CKD). This, for various reasons, was usually discarded, and represented a waste of raw material, and a significant form of heat loss. From the 1960s, it became more common to find ways of recycling the dust in order improve thermal efficiency, so, increasingly, the alkali content of the clinker produced reflected that of the raw material. At the same time, dry process kilns began to be introduced, and these often lost little dust, so allowing little alkali to escape from the kiln system. As a result, there was a general increase in cement alkali levels. Before 1960, few cements exceeded the typical 0.2-0.4% range of Na2Oeq, but then typical levels started to rise towards 0.6%, while some cements had significantly more. Alkali/aggregate reaction in concrete had first been identified as a problem in the USA, and failures due to alkali/aggregate reaction started to be recognised in Britain in the 1960s.

Mitigation of the problem has followed a number of lines:

  • avoidance of alkali-susceptible aggregates
  • use of cement components such as blastfurnace slag and flyash, which tie up the cement alkalis and prevent reaction with the aggregate
  • modification of the cement raw materials, either as a "special" clinker or across-the-board for a plant's entire clinker production.

Alkalis in cement raw materials are usually associated with the aluminosilicate component of the rawmix. Reduction in the clinker alkali in an efficient kiln process involves either:

  • replacing the rawmix alumina source with another component: this usually means obtaining a more expensive material from a more distant source, or
  • installing an "alkali bleed", which involves removing some of the kiln system's high temperature gases (which contain the alkalis as fume), resulting in some heat wastage.

Rapid Hardening Portland Cement

From WWI onwards, by far the most commonly-produced "special" cement has been a "rapid hardening" grade. However, this has nearly always been made with the plant's ordinary clinker, differentiated from the general-purpose cement by finer grinding. Originally designed for use in reinforced concrete when the latter was a new idea, it became more generally designed for use by precast concrete products manufacturers, who wanted rapid development of early strength so that their moulds could be re-used more rapidly. One-day strength was typically designed to be 50-100% higher than that of the ordinary cement. Sometimes different clinkers were produced, typically with a somewhat higher alite content, and sometimes with a higher alumina (with concomitant higher alkali) content. But since the higher early strength could readily be achieved by finer grinding, such special clinkers were soon abandoned as an unnecessary complication of the plant's operations. From the 1930s, most plants made the product, as typically 5-10% of their production.

Clinker Chemical Analysis

It will be seen from the following table that the different types of Portland clinker described differ only slightly. The chemical data shown are typical, and fairly wide variations can exist in each type, particularly in the minor elements.

Clinker Type SiO2 Al2O3 Fe2O3 CaO MgO SO3 LoI(1) Na2O K2O SrO TiO2 P2O5 Mn2O3 Total FL(2) IR(3) Na2Oeq C3S(4) C2S(4) C3A(4) C4AF(4)
Modern, general purpose 21.43 4.50 3.12 66.18 1.71 1.67 0.05 0.15 0.71 0.07 0.27 0.08 0.07 100.01 1.0 0.07 0.62 63.6 13.3 6.6 9.5
Thames-side, 1960s 22.13 5.88 2.51 66.77 1.19 0.20 0.09 0.28 0.55 0.08 0.26 0.11 0.04 100.09 2.5 0.08 0.64 50.5 25.1 11.3 7.6
Static kiln, 1900 22.51 5.68 2.37 65.88 1.01 1.46 0.37 0.09 0.07 0.12 0.33 0.09 0.09 100.07 4.0 0.06 0.14 35.6 37.5 11.0 7.2
Sulfate resisting 21.22 4.09 4.75 66.67 1.31 0.54 0.19 0.16 0.52 0.07 0.28 0.09 0.13 100.02 1.0 0.15 0.50 71.4 6.5 2.8 14.5
Low heat 24.81 4.44 5.48 63.08 1.15 0.15 0.03 0.14 0.25 0.04 0.25 0.11 0.12 100.05 0.1 0.03 0.30 30.0 48.4 2.5 16.7
White 24.06 3.69 0.32 68.92 0.94 0.55 0.85 0.11 0.08 0.18 0.19 0.12 0.02 100.03 2.0 0.29 0.20 65.0 19.1 9.2 1.0
White, low-alumina 25.30 1.95 0.35 69.99 0.89 0.35 0.71 0.08 0.05 0.15 0.08 0.13 0.01 100.04 1.5 0.03 0.11 72.2 18.0 4.6 1.1
Off-white 22.89 5.27 0.74 67.76 1.67 1.05 0.05 0.08 0.15 0.05 0.26 0.10 0.02 100.09 1.6 0.15 0.18 57.2 22.0 12.7 2.3
High alkali 21.24 5.76 2.26 64.93 2.89 1.07 0.05 0.28 1.10 0.04 0.25 0.15 0.06 100.09 1.1 0.02 1.00 53.7 20.3 11.4 6.9
Low alkali 21.69 5.19 3.39 65.14 2.78 0.61 0.08 0.20 0.57 0.04 0.20 0.10 0.07 100.07 0.8 0.05 0.58 56.0 19.8 8.0 10.3

NOTES
(1) Loss on Ignition
(2) Free lime
(3) Acid-insoluble residue
(4) "Bogue Compounds" defined as follows:

  • C3S = 4.0714 (CaO-FL) - 7.5999 SiO2 - 6.7177 Al2O3 - 1.4298 Fe2O3 - 2.8516 SO3
  • C2S = 2.8666 SiO2 - 0.754387 C3S
  • C3A = 2.6500 Al2O3 - 1.6920 Fe2O3
  • C4AF = 3.0432 Fe2O3

Notes on the derivation of such equations can be found in the Minerals Table.

© Dylan Moore 2013: commenced 20/05/2013: last edit 06/02/2017.