Durability considerations for cements containing approved secondary main constituents and additions, The

future specification of cement type, class and water/cement ratio: Part 2: Durability considerations for cements containing approved secondary main constituents and additions, The

Owens, Philip

BS EN 197-1: 2000 Cement -Part 1: Composition, specifications and conformity criteria for common cements(1) has made provision for incorporating several approved secondary main constituents (SMCs) in cement, covering the same strength class range as CEM I (i.e. 32.5, 42.5 and 52.5). Thus, with these new cement types, knowledge of cement performance is needed to predict concrete strength and durability, particularly as both CEM I and composite cements have the same standard strength classes.

Cement types

Table 1 of BS EN 197-1: 2000 illustrates the emphasis placed on manufacture of cements containing constituents other than clinker. This implies that the most effective method of using a high-performance clinker is by combination with an approved SMC (see Table 1). SMC materials may be combined with any clinker to produce a range of ‘blended’ cements. These may be either as composites produced at the cement factory, or combinations manufactured at the concrete plant. However, more constituents are permissible, particularly in factory-produced cements, including minor additional constituents, set regulators and additives.

However, the restrictions on types and proportions of SMCs stated in Table 1 of BS EN 197-1: 2000(1) are not based on a scientific understanding of their rheological and hydraulic properties. Their inclusion has been based on experience relating to SMC use, derived from various EU countries.

UK cements

The UK cement industry has been reluctant to incoiporate SMCs in composite cements, other than limestone (L and LL), due to their cost and availability in the required quantities. Nevertheless, ggbs and pfa may significantly increase the resistance of concrete to sulfates and chlorides. Table 1 of BS 8500: 2002 Concrete -Complementary British Standard to BS EN 206-1(7) states the performance of various cements in relation to sulfate resistance and exposure to chlorides. This information has been arranged in Table 2 in cement group order, which is akin to Table? of BS 5328-1: 1997 Concrete. Guide to specifying concrete(8).

Table 2 enables selection by the specifier of the most appropriate cement or combination to resist either sulfates or chlorides. The resistance increases as the cement group number increases.

UK combination cements

Cement combinations are currently used in the UK according to the product standards established in BS 5328-1: 1997(8). By following the conformity procedure for combinations in Annex A (normative) of BS 8500-2: 2002(7), it is possible to produce combinations with CEM I incorporating major constituents such as ggbs to BS 6699: 1992 Specification for ground granulated blastfurnace slag for use with Portland cement(10) or pfa to BS 3892-1: 1997 Specification for pulverized-fuel ash for use with Portland cement(11). However, BS 8500-2: 2002(7) increases the number of combinations that include pfa according to BS EN 450: 1995 Fly ash for concrete – definitions, requirements and quality control(12) and powdered limestone to BS 7979: 2001 Limestone fines for use with Portland cement^. Although metakaolin is classified as a natural calcined pozzolana (Q) in BS EN 197-1: 2000(1) ,and may be used in CEM II/A-Q within Groups 1 and 4, it does not yet have a British Standard product specification in Table 2. This prevents its application in the cement combination procedure. Conversely, sil- ica fume has a product Standard (prEN 13263: 1998 Silica fume for concrete – Definitions, requirements and conformity control(14)). It is used in CEM U-D within Groups 1 and 4 but not included in the normative conformity procedure for combinations in BS 8500-2: 2002(7).

The equivalence of these combinations to a particular cement type is established by comparing the CEM I strength and addition combination with the relevant cement strength class. For example, the combination of CEM I and a pfa to B S 3 892-1: 1997(11) is considered equivalent to CEMII/B-V 32.5, as stated in Table 1 of BS EN 197-1: 2000(1). It is indicated that the concrete user should be concerned about the omission in BS EN 196-1: 1995 Methods of testing cement. Determination of strength (15) relating to the effect of cement on consistence. This has been justified on the basis that the ‘consistence test’ on a neat cement and water paste is an integral part of the initial setting time test in BS EN 196-3: 1995 Methods of testing cement. Determination of setting time and soundness(16). However, consistence does not have to be declared. As this is essential for its application, the authors believe that the determination for fresh mortar to establish the strength should be according to BS EN 196-1: 1995(15). Such tests could be as specified in BS EN 1015-4: 1999 Methods of test for mortar for masonry. Determination of consistence of fresh mortar (by plunger penetration)(17).

Consistence improvements achieved by incorporating such materials as pfa are important as they significantly reduce water demand. However, other SMCs may not have these advantages. The omission of information on the relationship between cement type and class and consistence in BS EN 197-1: 2000(1) make it even more difficult to design concrete mixes from first principles. In reality, the physical and chemical characteristics of SMCs significantly differ, particularly the interaction between particle shape, particle density and solution chemistry. For instance, CEM I clinker has angular particles and a particle density close to 3.1Mg/m^sup 3^, whilst pfa is mainly spherical and can have a density of 2.2Mg/m^sup 3^. Combined in proportions by mass, 65/35 CEM I/pfa has a composite cement density of approximately 2.8Mg/m^sup 3^. This combination, used on a mass-for-mass basis, will increase in volume by 11 %. Consequently, CEM I in concrete with a minimum cement content of 300kg/m^sup 3^ occupies 97 litres/m^sup 3^, whereas the composite cement occupies about 108 litres/m^sup 3^. When there is no increase in water demand, concrete consistence achieved with the composite cement improves due to particle shape and increased powder volume.


BS EN 197-1: 2000(1), which includes SMCs, omits several important requirements in relation to CEM I. Further development is required in order to demonstrate cement performance, including SMCs. The main points for consideration include the following:

* BS EN 197-1: 2000 does not refer to strength development after 28 days

* there is a requirement to measure the fresh mortar density in BS 196-1: 1995″5).

* declaration of the clinker strength class or CEM I used in the manufacture of a given composite cement so the effects of dilution by the SMC may be evaluated in relation to the strength class

* a relationship exists between the properties of mortar and concrete, provided the density is controlled

* introduction of a consistence test for BS EN 196-1: 1995 mortar made at a fixed total-water/cement ratio of 0.50.

Descriptions of SMCs and restrictions on the amounts to be combined within cements specified in Clause 5 of BS EN 197-1: 2000(1) indicate that the list is comprehensive and there is a similarity between the different SMCs. This has not been demonstrated and it is not obvious why the maximum 10% has been chosen for silica fume, compared with 95% for ggbs.

It will be necessary to develop a test for evaluation of each SMC as a potential cementitious constituent to calculate free-water/ cement ratio for each cement type, related to the amount of SMC incorporated.



In contrast to BS EN 197-1: 2000 and its liberal approach to SMCs in cement, BS EN 206-1 : 2000 Concrete: Specification, performance, production and conformity^ makes restrictive and limited provision for additions. In BS EN 206-1: 2000, concrete additions are defined as finely divided mineral fillers and pigments to improve or achieve special properties. Additions qualifying as concrete constituents are classified as ‘nearly inert’ (Type 1), pozzolanic or latent hydraulic (Type II).

In practice, although described as ‘nearly inert’, Type I additions are treated as filler aggregates conforming to BS EN 12620: 2002 Aggregates for concrete(19) or pigments conforming to BS EN 12878: 1999 Pigments for the colouring of building materials based on cement and/or lime. Specifications and methods oftesf1^. Type II additions are restricted in BS EN 206-1: 2000(I8) to pfa A, as cited in BS EN 450: 1995(12) and silica fume to prEN 13263: 1998(I4), although the UK has incorporated more additions in BS 8500: 2002(7).

The k-value concept

In BS EN 206-1: 2000(18), both pfa and silica fume form part of the minimum cement content, when used with CEM I, via the k-value concept. This technique is used to calculate minimum cement contents and maximum free-water/cement ratios for various exposure classes. However, it is not supported by a test method. The expression ‘water/cement ratio’ is replaced by ‘water/(cement + k x addition) ratio’. Complicated restrictions apply to the use of pfa. Use is limited, according to BS EN 450: 1995(12), where:

i) it is limited in concrete to 25% by mass of CEMI (or 33% of fly ash + CEM I)

ii) k-value is related to the CEM 1 class, namely 0.2 for CEM I 32.5 and 0.4 for CEM142.5 and higher

iii) the minimum CEMI content can be reduced by up to k × (minimum cement content -200)kg/m^sup 3^

iv) (CEM I + pfa) must be greater or equal to the specified minimum cement content

v) if a proportion of pfa greater than in (iii) and (iv) is used, the excess shall not be taken into account in calculating minimum cement content and maximum water/cement ratio

vi) the concept is not recommended for combinations of pfa and sulfate-resisting cement CEMI + SR, when sulfates are the aggressive substances.

The conditions are equally, if not more, onerous for silica fume. Disparities between the application of SMCs such as pfa and silica fume in BS EN 197-1: 2000(l), relating to cement, and BS EN 206-1: 2000(18), relating to concrete, require clarification. This ‘k-value concept’ does not address the hydraulicity (strength performance) of an addition in concrete. A more appropriate method for evaluating the hydraulicity of additions would be via efficiency factors (c-factor and wd-factor), to be proposed in the third section of this article.

Equivalence testing for additions

Annex A (normative) of BS 8500-2: 2002(7) permits the use of additions with a conformity procedure for combinations. This is already well-established within BS 6699: 1992 Specification for ground granulated blastfurnace slag for use with Portland cement10′ and BS 3892-1: 1997(11). The mix designs incorporate pfa to BS EN 450: 1995 (I2), where loss on ignition is not more than 7%, and limestone fines to BS 7979: 2001(I3). These additions may be fully taken into account within the concrete composition regarding cement content and water/cement ratio if (a) the proportion of addition is within the limits permitted for the appropriate cement type in BS EN 197-1: 2000(1) and (b) the test criteria meet the requirements of Annex A of BS 8500-2: 2002(7). There are differences between the maximum permitted addition proportions combined with CEMI in Table 1 ofBS BS 8500-2: 2002(7) and SMCs in Table 1 ofBS EN 197-1: 2000(l) (see Table 3).

The proportions of available additions and SMCs used are unlikely to be of this level. The procedure specified in BS 8500-2: 2002(7) applies the same principles for cement strength testing, as stated in BS EN 196-1: 1995(15) where strength is classified in a similar way to cements complying to BS EN 197-1: 2000(1). A higher-strength cement such as CEM152.5 can be ‘diluted’ with a permitted addition to obtain a cement of different type and lower strength class, such as CEMII 32.5. It is then permissible to use the resulting combination of CEMI and addition in the same manner as undiluted CEM I to calculate free-water/cement ratio. This procedure is akin to cement testing as it does not incorporate measurements of consistence and hydraulicity.

This conclusion has serious implications when comparing the incompatible approaches in BS EN 206-1: 2000(18) (k-value concept) and BS 8500-2: 2002(7) (conformity procedure for combinations). This particularly relates to pfa in BS EN 450: 1995 (12) and limestone in BS 7979: 2001(13), which has questionable cementing efficiency. It is incongruous that pfa to BS EN 450: 1995 (12) with a k-value of 0.4 according to BS EN 206-1: 2000(18) can only replace up to 25% of the total cement content. This is equivalent to just 10% (25 × 0.4) of the cement content, when the entire 55% could count towards the calculation of water/cement ratio in BS 8500-2: 2002(7). Knowledge of cement and concrete constituent cementing efficiency is essential, yet the core cement and concrete Standards are based on an inconsistent, flawed approach to this property.


* BS 8500-2: 2002(7) has selected cement equivalence testing rather than the ‘k-value concept’ for determining the proportion of additions

* equivalence testing fails to identify the influence of an addition on water demand as consistence measurement is omitted

* k-values, as applied in BS EN 206-1: 2000(18), are neither based on a formal test method nor related to strength.

Durability considerations


Concrete durability is mainly related to cement paste penetrability. This may be measured using permeability testing. Figure 1, after Powers et a/(21), shows how permeability varies with water/cement ratio for a mature hardened Portland cement paste and the time taken before continuous capillary pores become blocked.

If this information, which is now 45 years old, is used as an illustration, assume that the free-water/cement ratio limits of 0.45-0.65 specified in BS 5328-1: 1997(8) for reinforced concrete under various exposure conditions have been derived from the relationship between permeability and free-water/cement ratio for concrete of a given age. Figure 1 shows that the limits correspond to various maximum permeability values, irrespective of concrete grade or type. The permeability limits for different concrete grades are fixed for each exposure condition. The question is: does the free-water/cement ratio required for a given permeability depend on the cement grade and type and the type and proportion of SMC or addition? For example, if CEM I is combined with a completely inert filler, the filler particles would only contribute to impermeability by occupying space. This would not be influenced by changes in free-water/cement ratio, regardless of concrete maturity. Furthermore, permeability of the system would only be affected by hydration of reactive constituents, irrespective of claims to improve strength by hydration of clinker constituents’22’. The filler should have little or no influence on pore-blocking characteristics. This argument represents an extreme situation, but indicates that the free-water/cement ratio required for a given penetrability will depend on the type and proportion of clinker, the main addition, secondary main addition (SMC) and minor additional constituents (MAC). The latter may be more or less reactive than the clinker.

Durability mechanisms

The strength and durability of well-compacted concrete mainly depend on the chemical and physical characteristics of the hydration products. Different clinkers and additions will produce different products at different rates (see Figure 2 in Part 1 of this article(23). The physical features mainly relate to free-water/cement ratio and its influence on initial spacing of reactive particles; quantity of hydration products required to fill initially water-filled voids; and voids remaining when water surplus to hydration is removed. The risk of ingress by aggressive agents increases when the capillary voids arising from the surplus water are interconnected. The principal mechanisms influencing the long-term performance of hardened concrete are shown in Table 4.

Mechanical property specifications may be based upon strength-related performance testing: this is now reliable as the methods are understood and adequately standardised. Specifying for physical and chemical resistance is more complex as standardised and relevant performance tests remain the province of specialists.

Thus, specifications in Codes are usually based on accumulated experience and are expressed in terms of both prescription (free-water/ cement ratio, cement type and content) and performance (concrete grade).

BS 8500-2: 2002(7) outlines a system for determination of mixes to resist exposure in various environments, based on BS EN 206-1: 2000(18) Annex D. For example, Table A-13 of BS 8500-1: 2002(7) follows the UK traditional practice of establishing a balance between reinforcement cover, concrete strength, water/ cement ratio, cement content (with aggregate size) and cement grouping. A Group 4 cement with a maximum free-water/cement ratio of 0.40 could be appropriate for exposure class XS3 (reinforced concrete in seawater) with a nominal cover of 50mm. However, some cements in Group 4 contain limestones (L or LL), which are non-hydraulic. The specified minimum cement content is (360 + 20)kg/m^sup 3^ for a 20mm maximum aggregate. This requires a water content of 152 litres for a freewater/cement ratio of 0.40. The clinker content is only 323kg/m^sup 3^ when the cement contains 15% limestone, resulting in an increased free-water/cement ratio of 0.47. This increased free-water/cement ratio is not taken into account for durability.

The situation would be clarified if both strength contribution and water demand is known for the individual components. Such information should be obtained for various component combinations at given consistencies and free-water/cement ratios at different ages, particularly beyond 28 days. However, emphasis on 28-day performance is unrealistic as it fails to account for any longer-term improvements (see Figure 2 of Part 1(23)). The data forming the basis of Figure 2 could be considered to be inadequate as it is restricted to two concrete mixes. In 1997 The Concrete Society investigated 96 concrete units with the variables shown in Table 5(24).

Each unit type was temperature-monitored during casting and cores were removed at ages ranging from 28 days to one year. The ratios of core strength to 28-day cube strength are shown in Figure 2.

The relationships for CEM I and Portland limestone cement are similar, whereas those for CEM 1/pfa and CEM I/ggbs show significant increases in in-situ strength with age. The time to achieve an in-situ strength equal to the standard 28-day cube strength is approximately 40 days for both CEM I/ggbs and CEM I/pfa. However, this increases to 65 days for Portland limestone cement and 130 days for solely CEM I. This has major implications for cement type selection to achieve the structural performance level specified in BS 8500-1: 2002(7) for ‘normal’ (30-100 years) and ‘high’ (more than 100 years) levels for concrete structures, also included in Eurocode 2(25).

Concluding remarks

* lower bound limits for free-water/cement

* ratio must be based on evidence that such

* limits produce the appropriate impenetrabilility level for the given durability class

* any specified free-water/cement ratio limits must take account of the proportion and type of clinker, SMC or addition

* evidence relating to longer-term performance of different clinkers, SMCs and additions must be considered on the basis of actual in-situ performance.


1. BRITISH STANDARDS INSTITUTION. BS EN 197-1:2000 Cement – Part 1: Composition, specifications and conformity criteria for common cements

2. BRITISH STANDARDS INSTITUTION. BS146: 2002 Specification for blastfurnace cements with strength properties outside the scope of BS EN197-1

3. BRITISH STANDARDS INSTITUTION. BS 4246: 1996 Specification for high slag blastfurnace cement

4. BRITISH STANDARDS INSTITUTION. BS 6588: 1996 Specification for Portland pulverised-fuel ash cements

5. BRITISH STANDARDS INSTITUTION. BS 6610: 1996 Specification for Pozzolanic pulverised-fuel ash cement

6. BRITISH STANDARDS INSTITUTION. BS 7583: 1996 Specification for Portland limestone cement (withdrawn, superseded)

7. BRITISH STANDARDS INSTITUTION. BS 8500: 2002 Concrete – complementary British Standard to BS EN 206-1: Part 1 – Method of specifying and guidance for the specifier; Part 2 – Specification for constituent materials and concrete

8. BRITISH STANDARDS INSTITUTION. BS 5328-1:1997 Concrete. Guide to specifying concrete

9. BRITISH STANDARDS INSTITUTION. BS 4027: 1996 Specification for sulfate-resisting Portland cement

10. BRITISH STANDARDS INSTITUTION. BS 6699: 1992 Specification for ground granulated blastfurnace slag for use with Portland cement

11. BRITISH STANDARDS INSTITUTION. BS 3892-1:1997 Pulverised-fuel ash: Specification for pulverised-fuel ash for use with Portland cement

12. BRITISH STANDARDS INSTITUTION. BS EN 450: 1995 Fly ash for concrete. Definitions, requirements and quality control

13. BRITISH STANDARDS INSTITUTION. BS 7979: 2001 Specification for limestone fines for use with Portland cement

14. BRITISH STANDARDS INSTITUTION. prEN 13263:7998 Silica fume for concrete -Definitions, requirements and conformity control

15. BRITISH STANDARDS INSTITUTION. BS EN 196-1:1995 Methods of testing cement. Determination of strength

16. BRITISH STANDARDS INSTITUTION. BS EN 196-3:1995 Methods of testing cement. Determination of setting time and soundness

17. BRITISH STANDARDS INSTITUTION. BS EN 1015-4:1999 Methods of test for mortar for masonry: Part 4 – Determination of consistence of fresh mortar (by plunger penetration)

18. BRITISH STANDARDS INSTITUTION. BS EN 206-1:2000 Concrete: specification, performance, production and conformity.

19. BRITISH STANDARDS INSTITUTION. BS EN 12620:2002 Aggregates for concrete

20. BRITISH STANDARDS INSTITUTION. BS EN 12878:1999 Pigments for colouring building materials based on cement and/or lime -specifications and methods of test

21. POWERS, T, COPELAND, L. and MANN, H. Capillary continuity or discontinuity in cement pastes (in) Journal of the Portland Cement Association, Research and Development Laboratories, Vol.1, No.2, May 1959, pp.38-48

22. BOMBLED, J. Influence des fillers sur les proprietes des mortiers et des betons, Ciments, Betons, Plaitres, Chaux, No.738-5/82, pp.282-290

23. OWENS, P. and NEWMAN, J. The future specification of cement type, class and water/cement ratio. Part 1 : Introduction, water and Portland cement, CONCRETE, Vol.38, No.3, March 2004, pp.46-49

24. THE CONCRETE SOCIETY. In-situ concrete strength – an investigation into the relationship between core strength and standard cube strength, Project Report 3 of a Concrete Society Working Party (awaiting publication)

25. COMITE EUROPEEN DE NORMALISATION. prEN 1992-1-2 Eurocode 2: Design of concrete structures, Brussels, 2002


Copyright The Concrete Society Mar 2004

Provided by ProQuest Information and Learning Company. All rights Reserved