Iceland State Cement Works

Initiator in production of ternary blended pozzolanic cements

( Dr. Gudmundur Gudmundsson previous manager of Iceland State Cement Works in Akranes Iceland )

Ternary cements:

An emerging technology shows that 50 to 70% Portland cement can be replaced with one or several complementary cementing materials, such as coal, fly ash, granulated blast furnace slag, natural pozzolans, silica fume and rice husk ash, with dramatic improvements in properties of concrete. This can be done either by blending in a cement plant or during concrete batching.

Ternary cements are multi-component cements that are made by blending Portland cement (clinker + gypsum) with two complementary cementing materials, such as coal fly ash, natural pozzolans, granulated blast-furnace slag, silica fume and reactive rice-husk ash. The properties of ternary-blended cements depend on chemistry, mineralogy, particle size distribution and surface area of the three components which make up the cementing material. Differences in physical and chemical characteristics of individual components and their proportions can be exploited for producing concrete mixtures with desired properties.

A concrete structure that remains water-tight for an indefinite period under service conditions should remain durable on exposure to commonly known causes for loss of durability. Therefore, ternary blended cements with high volume of complementary cementing materials will have a long service life and a low life-cycle cost.¹

The experience of the last one or two decades has shown that the blend of clinker with two supplementary materials (ternary mix) gives very good concrete quality and is preferably used in high performance concrete. In the blend silica fume has showed the best results to obtain strong and durable concrete owing to its very fine particle size and high surface area. In Iceland no fly ash or granulated blast furnace slag is available but instead volcano materials or natural pozzolana.

Ternary Cements in Iceland, introduction:

Cement and concrete making i Iceland is in many ways unique. The country is geological young so that calcareous raw material are only to be found as off-shore seashell deposits and for argillaceous material rhyolitic material must be used. Because of this icelandic cement has low silica and is high in alkali content.

As a preventative measure against deleterious alkali aggregate reactions in concrete occurring in Iceland in the years 1965 to 1980, the Iceland State Cement Works ( ISCW ) in cooperation with the Building Research Institute ( BRI ) in Reykjavik started research on the use of natural pozzolan in cement. Built on the results of the research, the ISCW started production of pozzolanic cements in 1975. As pozzolanic material used was rhyolite, also used as raw material for the clinker production in the factory.

In 1979 a ferrosilicon factory was erected about 10 km away from the ISCW. Already in 1972 research was started in cooperation with BRI in Reykjavik to investigate, if silica fume generated by the ferrosilicon production could be used in stead of rhyolite as pozzolanic material in Icelandic cement. The silica fume turned out to be very good pozzolan and showed excellent ability as inhibitor to alkali expansion. Further it improved other qualities of the concrete such as water impermeability, frost resistance and chloride impermeability. The ferrosilicon factory was started 1979 and at the same time the production of silica fume cement was started at ISCW.

This was the first factory made pozzolanic cement in the world where silica fume was used as pozzolan.^{2, 3}

In Iceland many important constructions have to withstand the exposure of severe conditions caused by cold whether, heavy winds and water, especially sea water. This called for cement types with stronger pozzolanic activity. In 1983 building of a hydro electrical powerplant “ Blanda” in North of Iceland was planned. This construction had extensive water tunnels and other constructions exposed to water and severe weather conditions. The concrete aggregate which was to be used for this construction was highly alkali active so new type of cement had to be developed. ISCW delegated this project to the BRI.

The cement research carried out earlier by the BRI had showed that silica fume had excellent pozzolanic capacity and therefore it was first considered to use higher proportion of it in the cement. But a higher proportion than 10% showed some negative effect for the concrete such as high water requirement. But the fine ground rhyolite had also strong and effective pozzolanic properties. Then tests were made to grind the three materials, 65% clinker (+ gypsum), 25% rhyolite and 10% silica fume together. The results of this blend were very satisfactory for the use in the concrete for the hydro electrical power plant. The early strength was lower than with the ordinary cement type but the 28 days strength similar. The ability to diminish alkali aggregate reaction was much stronger by this new cement and the heat of hydration lower.

The production of this new cement type was then started in1985 and sold to the power plant for use in the mass concrete and in concrete made for severe outside conditions. Looking through the literature it seems that this was the first time that a factory made ternary blend cement was produced and put on the market. Anyway it was presented on the 9^th International Congress on the Chemistry of Cement in New Delhi 1992 ⁴. Landsvirkjun ( the National Power Company of Iceland ) has used Blanda-cement and other types of pozzolanic cements produced by ISCW in many of their power plants and has confirmed the quality and the durability of the concrete made of it .⁵

This is so far interesting as ternary or higher blend cements are to day regarded the best solution to produce cements for use in high performance concretes.

In the late nineties ISCW and BRI started research on the use of other pozzolanic material blend for the cement production such as fly ash and ground granulated blast furnice slag ( ggbfs ). The mixture with the best quality to withstand severe environmental attack was a blend of 40% ggbfs, 5% silica fume and 55% clinker+gypsum. This blend had excellent capacity to withstand alkali aggregate reaction and had also lower heat of hydration than the others. This cement type was used for concrete in one of the sewage treatment plant owned by the City of Reykjavik in 2002 and also for some repair work at harbors and bridges at the sea side with good success.

Raw materials for the ternary pozzolan cement in Iceland:

The typical average chemical composition, fineness and specific gravity of the binary and ternary blend were as follows: (A= ordinary Icelandic Portland cement, B= “Blanda” cement, ternary blend.)⁴

A B

Clinker+ Clinker +

7.5% silica 10% silica+25% rhyolite

SiO₂ % 24 36

Al₂O₃ % 4 5

Fe₂O₃ % 4 3

Free CaO % 1 1

Fineness cm² / g 3900 6600

Specific gravity g/cm³ 3.0 2.7

The cement clinker used for the cement blend and produced in a small, wet kiln ( type Folax from F.L Smidth in Copenhagen Denmark, 300 tons per day ) had rather low C₃S content of 59%, C₂S=23%, C₃A=5% and C₄AF=13% , and high alkali content ( Na₂0_eqv. = 1.5%. )

The rhyolite of type dactic pitchstone with high glass content was blasted out together with the rhyolite used as raw material for the cement clinker in a pit located in Hvalfjord ca. 30 km. from the factory. There the rhyolite is ground to a maximum size of 25 mm, then partly dried before it is transported to the cement mill.

The silica fume is a very fine dust (ca. 20.000 m²/kg as produced) and very difficult to handle in bulk transport systems (pneumatic transport). By compacting this extremely fine material a pneumatic transport is possible. The compaction consists in that the as-produced silica fume is placed in a silo and compressed air is blown in from the bottom of the silo, causing the particles to tumble. As the particles tumble, they agglomerate. The heavier agglomerates fall to the bottom of the silo and are periodically removed. Because the agglomerates are held together relatively weakly, they break down in the mill during the grinding process. The as-produced silica fume has bulk density as low as 200 kg. /m³ but after the compaction it was possible to get it over 600 kg. /m³.

In collaboration with Elkem AG in Norway a special equipment arrangement was established at the factory to receive the silica fume and transport it over an automatic weight device to the cement mill. The rhyolite and the silica fume were then interground with the cement clinker in the cement mill (ball mill, type Unidan ).

The production of Portland cement containing 7.5% silica fume started at ISCW in 1979 and the production of Blanda-cement started 1985.

At that time the European Cement Standard EN 197 was under development. As ISCW was the first producer of silica fume cement it was important to get this type of cement put in the new standard. The standard was ready for use in 1990 and ISCW succeeded in getting the silica fume and rhyolite blended cements in the EN 197. The mark in the standard for the two cements was CEM II / A-D, 42.5R for the ordinary Portland cement and CEM IV /B, 42.5 N for Blanda-cement.

The rhyolite is a soft and glassy material. Therefore the grindability is much higher for rhyolite than for the clinker. This diminishes the early strength of the Blanda- cement but improves the pozzolanic property. This is illustrated in Fig. 1 and 2

Fig 1: The compressive strength of ordinary Portland cement and Blanda - cement at various ages

Fig. 2: Alkali expansion for ordinary Portland cement and Blanda-cement (ASTM C 1260)

Fig. 1 and 2 illustrate that the intergrinding of the three materials, clinker,rhyolite and silica fume result in a cement with good pozzolanic properties and appropriate for the use in mass concrete and concrete in aggressive environments. On the other hand these properties are not in demand by the common building industry where cements with high early strength are most popular. Having in mind the different grindability of the soft rhyolite against cement clinker separate grinding of the two materials should be advisable when cement of this type were brought on the common market. Then the silica fume would be interground together with the clinker but the rhyolite ground separately and thereafter mixed to the other. The research on production of such ternary blended cement through separate grinding has not yet been carried out in Iceland.

As mentioned earlier the ISCW and BRI started in the late nineties on request of the National Road Authority research on blended cement type which would have better quality to be used in concrete for constructions under severe environmental conditions. The best blend found did consist of 40% ggbfs, 5% silica fume and 55% clinker+gypsum. The ggbfs was imported from Belgium, first dried a then interground with the cement clinker and silica fume. The cement was classified in EN 197 as CEM III/A with permission to blend to it up to 5% silica fume. The fineness of the ggbfs-cement was 4050 cm²/g but 5618 cm²/g for the Blanda-cement.

The properties of the ggbfs-cement by contrast with the Blanda-cement shows that it is more qualified for use in mass concrete or in concrete where high protection against environmental attack is demanded.

Fig.3: Compressive strength of the ggbfs-cement versus Blanda- cement

Fig.4: Alkali expansion of ggbfs-cement versus Blanda-cement (ASTM C 1260)

Fig.5: Heat of hydration, ggbfs-cement versus Blanda-cement.

(NT-Build 388).

.

References:

1. P.Kumar Mehta 50^th Brazilian concrete congress Salvador Bahia, September 6. 2008. A glimpse into sustainable ternary-blended cements of the future.

2. Guide for the Use of Silica Fume in Concrete reported by Guide for the Use of Silica Fume in Concrete*, Reported by ACI Committee 234, ACI 234R-06 page 4

3. Detwiler R.J. Blended Cement now and for the Future, Emerging Technologies, Symposium on Cements for the 21.Century, March 15,1995, Chicago, Illinois,

4. Gudmundsson G. and Möller J. 9^th International Congress on the Chemistry of Cement New Delhi 1992

5. Arnalds S. and Sveinbjörnsson S.,”Kárahnjúkar” Hydroelectric Project, Norsk betongdag, Oslo, November 2004