PCA R&D Serial No.

2658

The Use of Limestone in Portland Cements:

Effect on Thaumasite Form of
Sulfate Attack
by R. Doug Hooton and Michael D. A. Thomas

Portland Cement Association 2002

All rights reserved
This information is copyright protected. PCA grants permission to electronically share this document with other professionals on the condition
that no part of the file or document is changed

The Use of Limestone in Portland Cements:

Effect on Thaumasite Form of Sulfate Attack
R. Doug Hooton1 and Michael D.A. Thomas 2
1

University of Toronto and 2 University of New Brunswick

INTRODUCTION
This paper discusses the use of limestone as a component of portland cements, particularly in
relation to the risk of sulfate attack due to thaumasite formation. This form of sulfate attack
requires a source of carbonate ions, and concerns have been raised that the use of limestone in
portland cement could increase the vulnerability of concrete to attack.

USE OF LIMESTONE IN PORTLAND CEMENT

The use of limestone as a component of portland cement is now common practice in Canada and
Europe. In Europe limestone contents can be as high as 35%. Such practice is not permitted by
standards governing cement manufacture in the U.S.A.; however, it is currently being proposed
to allow up to 5% limestone in cements meeting ASTM C 150.

Use of Limestone in Canadian Portland Cements

After testing (unpublished data from CSA committee members) of both cement and concrete
properties in the late 1970s, to evaluate the influence of 5% limestone in portland cements, the
1977 version of CSA A5-M77 was amended in November 1980 to allow a maximum of 5%
addition of limestone to normal portland cement (CSA Type 10, equivalent to ASTM Type I).
Use of limestone at some plants started immediately. In the next revision to the standard,
limestone was permitted in high-early strength cement (CSA Type 30, equivalent to ASTM
Type III).
After 22 years of extensive use in the cold Canadian climate, which would tend to favor any
potential thaumasite problems, there have been no cases of Thaumasite Sulfate Attack (TSA)
related to the use of limestone in cement. In fact, the only reported case of TSA in Canada from
concretes made since 1980 (Bickley, et al 1990) did not involve the use of cement with
limestone, but rather a sulfate-resisting cement meeting CSA Type 50 and API Type G cement.

Limestone in European Cements

The use of limestone in portland cement was common and accepted practice in France in the
1980s (Hawthorn, 1989), whereas even minor amounts of other constituents was not permitted
by British Standards until 1991. A BRE Working Party was formed to examine the effect of

limestone on the performance of cement and concrete (Matthews, 1989). A comprehensive

testing program was initiated by the members of the Working Party, to determine the effect of
limestone at levels of 5% and 25% on the performance of concrete. A paper reporting the 5-year
data (Matthews, 1994) concluded that the performance of cements containing 5% limestone
is, overall, indistinguishable from that of OPC without additions, vindicating the decision to
permit such additions under British Standards.
In the European Standard, EN 197-1 Cement Part 1: Composition, specifications and
conformity criteria for common cements, all 27 common cement products defined are permitted
to contain up to 5% of minor additional constituents (macs) which can be limestone.
In addition to permitting limestone as a mac in other types of cement, EN 197-1 also covers
portland- limestone cement, which may contain up to 20% limestone when designated as a CEM
II/A cement or up to 35% limestone as a CEM II/B cement. It is not permitted to use limestone as
a mac in portland-limestone cement.
The European Standard covering concrete, EN 206-1 Concrete. Part 1: Specification,
performance, production and conformity, does not specify the types of cement that are permitted
in various classes of chemical (sulfate) exposure. Instead, it simply refers to sulfate-resisting
cement, which is intended to cover all cement types recognized as being sulfate resistant.
However, there is as yet no European Standard for sulfate-resisting cements, as it has not been
possible to achieve consensus among the member countries regarding the cement types to be
included. This issue is therefore dealt with on a national basis. In the UK, blastfurnace cements,
CEM III/B, with 66-80% slag, portland- fly ash cements, CEM II/B-V, with more than 25% fly
ash, and pozzolanic cements, CEM IV/B, with no more tha n 40% fly ash, are all permitted in the
most severe sulfate class, although fly ash cements are excluded in some situations where the
magnesium ion concentration in the groundwater exceeds 1 g/L. The use of macs ( 5%) is
permitted in the manufacture of all these cement types. Sulfate-resisting portland cement (SRPC)
is also permitted in this exposure class. In the U.K. the governing standard for SRPC is BS 4027
(1996) Specification for sulfate-resisting portland cement, which does not permit any additional
constituents. Portland- limestone cements (e.g. cements with more than 5% limestone) are only
permitted for use in the lowest (i.e. most benign) sulfate exposure class in Europe and the U.K.

SULFATE ATTACK
The Ettringite (Classical) form of Sulfate Attack
The conventional form of sulfate attack, that occurs when concretes are exposed to ground waters
or soils containing calcium, sodium and/or magnesium sulfates in solution, is reasonably well
understood. All of these sulfates will react with C3 A, hydrated aluminates, or monosulfate
(C 3 ACaSO4 12H2 O) to form ettringite (C 3 A3CaSO4 31H2 O), resulting in expansion and
cracking of mature concrete. In addition, sodium sulfate will react with calcium hydroxide in the
paste to form gypsum, which may cause damage in itself or react with monosulfate to form
ettringite. Magnesium sulfates are far more aggressive to concrete since, in addition to attacking
the aluminates, they will attack and decalcify the C-S-H matrix, forming gypsum, brucite, and
hydrous silica. This results in loss of strength and softening of the affected areas.

Mitigation of these forms of sulfate attack is achieved by designing concrete with low
permeability (i.e. low w/cm, well compacted and cured), and either limiting the amount of C3 A in
the portland cement (i.e. Type II or V) or by use of an adequate cement replacement of tested
pozzolans or slag (or by use of similar materials in the form of blended cements).

The Thaumasite form of Sulfate Attack (TSA)

Thaumasite is a calcium- silicate-sulfate-carbonate hydrate (CaSiO 3 CaCO3 CaSO 4 15H2O) that
can form in concrete by reaction of the calcium- silicate hydrates (C-S-H) with sulfates in the
presence of carbonate ions. The process only occurs in very wet environments and the rate of
formation is increased at low temperatures. It should be noted that in all of the laboratory and
field cases reported, the sulfates for this reaction come from external sources. The conversion of
C-S-H into thaumasite results in a loss of binding material and hence the integrity of the concrete
(or other cement-based material), however, the formation of thaumasite is not always damaging
and a recent report from the Thaumasite Expert Group (1999) in the U.K. introduced the
following terminology:
TSA Thaumasite form of
Sulfate Attack
TF

Thaumasite
Formation

Used when there is significant damage to the matrix of a

concrete or a mortar as a consequence of replacement of
cement hydrates by thaumasite.
Used to describe cases where thaumasite formed in preexisting voids and cracks without necessarily causing
deterioration of the concrete or mortar.

TSA is a relatively rare form of sulfate attack and risk of its occurrence is thought to be
negligible unless the following conditions are met (Thaumasite Expert Group, 1999):
Presence of sulfates and/or sulfides in the ground,
Presence of mobile groundwater,
Presence of carbonate, generally in the concrete aggregates,
Low temperatures (generally below 15C).
However, Sims and Huntley (2002) recently presented a series of cases where TSA had occurred
without all four of the above conditions being fully met, and other workers have provided
evidence that TSA may occur in concrete containing siliceous aggregates where the source of the
carbonate was the groundwater, and not the cement (French, 2000), while TF can occur in
concrete stored at temperatures above 15C (Brown and Hooton, 2002), and possibly even
concrete with no access to external sulfate (Hartshorn and Sims, 1998). Also in some cases the
source of carbonate has been identified as acidic, carbonate-bearing groundwaters (Hagelia et al,
2001). It has been argued (Crammond, 2002b) that the apparent scarcity of TSA is partly
attributed to the following: (i) the conditions of most standardized sulfate resistance tests do not
favor thaumasite formation, (ii) forensic examinations may have failed to detect TSA in
deteriorated concrete, and (iii) buried concrete is rarely inspected. Rogers et al (1997) reported a

case of TSA in an old aqueduct, where previous investigations had wrongly attributed the
deterioration to classical sulfate attack involving ettringite formation.
As with conventional sulfate attack, the rate of TSA can be slowed by producing low w/cm
concrete with low-permeability properties, to reduce the rate of ingress by sulfate solutions.
There is also evidence that use of pozzolans and slag may help, at least in as much as they
contribute to lowering permeability. It is reported that ground- granulated blast- furnace slag as a
partial cement replacement will help mitigate damage due to TSA, at least at 70 % replacement
(Higgins, 2002; Higgins and Crammond, 2002). The use of 7 % metakaolin also appears to help
(Smallwood et al, 2002). In all these cited cases, the carbonate source was from the aggregates.

ROLE OF LIMESTONE AND TSA

The recent discovery in the U.K. of severe deterioration in a number of buried concrete
foundations containing carbonate aggregates has raised concerns about the effect of limestone in
cement on the performance of concrete in sulfate environments. In these cases the concrete
produced would have satisfied the prevailing recommendations for sulfate resistance at the time
(e.g. BRE Digest 363 and BS 5328), but the deterioration was diagnosed as being the result of the
thaumasite form of sulfate attack (TSA) rather than classical sulfate attack. The vulnerability of
the concrete to TSA was ascribed to the presence of carbonate aggregate (Thaumasite Expert
Group, 1999; Crammond, 2002a). However, in some of these cases, it must be noted that high
levels of sulfates were not detected in the original undisturbed soils, but they did contain sulfides,
which oxidized during the construction process to form sulfates. The role of carbonate in
portland cement is discussed in greater detail below.

Performance of Limestone -Filled Cements (more than 5% limestone)

Part of the long-term study performed by the U.K. Working Group on limestone in cement
included studies on sulfate resistance (Matthews, 1994). Concretes were produced with a range of
portland cements, with C3 A contents ranging from 5.3% to 13.1% (by Bogue) and limestone
contents of 0, 5 or 25%, and exposed to solutions of sodium sulfate (1.5% SO3 ) or magnesium
sulfate (0.35% and 1.5% SO3 ) at 20C (68F). There was no consistent difference in the
performance of concrete after 2 and 5 years with the limestone content of the portland cement,
but as expected, a strong correlation was found between performance and the C3 A content of the
cement. However, because of the poor performance exhibited by the cement with the highest C3 A
and at a 25% limestone content, and the detection of thaumasite in most of the concretes
produced with so-called limestone- filled cements (meaning more than 5% limestone), a
recommendation was made that limestone- filled cements not be used beyond Class I sulfate
conditions (i.e. when SO4 content of groundwater exceeds 0.4 g/L or SO4 content of a 2:1 watersoil extract exceeds 1.2 g/L). There is no such restriction on cements containing up to 5%
limestone.
Later studies on concrete stored in sulfate solutions at 5C (41F) indicated little difference in the
performance of concrete produced with high-C3 A (~ 10%) portland cement with and without
15% limestone (Barker and Hobbs, 1999). However, the mode of attack was different with

thaumasite and ettringite being the principle reaction products for the cements with and without
limestone, respectively. Other workers have demonstrated that the presence of amounts of
limestone in the range from 10% to 20% can change the dominant reaction product from
ettringite to thaumasite (Borsoi et al, 2002; Justnes, 2002), but the impact of the limestone on
physical deterioration in these studies is not clear.
The most recent work at the Building Research Establishment indicates that limestone-filled
cements (typically containing much greater than 5% limestone) are the least TSA-resistant
binder types investigated (Crammond, 2002b). These results are confirmed in laboratory studies
conducted at Sheffield University in the UK (Hartshorn et al, 1999; 2001; 2002; Kakali et al,
2002; Torres et al, 2002).

Performance of Portland Cements with 5% Limestone or Less

A study conducted at the Building Research Establishment (BRE) concluded that 5% limestone
had a marginal adverse impact on the sulfate resistance of mortars stored in cold Class 4
magnesium sulfate conditions (Halliwell et al, 1996). After much deliberation over the BRE data,
the Thaumasite Expert Group (1999) concluded that this relatively low level of carbonate
material will not adversely affect the PCs performance in concretes containing either siliceous or
carbonate aggregates. It was pointed out that the use of 5% limestone in portland cement
roughly translates into just 1% limestone by mass of aggregate even in a relatively cement-rich
concrete mixture. Also, given that companies would have to target a lower addition rate to stay
within the 5% proposed limit, the 1% by mass of aggregate value mentioned above would almost
be halved.
From research in North America, the effect of using up to 5% limestone on conventional sulfateresistant properties of portland cements has been shown not to be detrimental. Hooton (1990)
tested a Type II cement with and without 2.6% limestone and found that, while both cements
expanded less than the 6- month 0.10% limit in ASTM C 1012, the 2.6% limestone cement
expanded less and had only half the expansion after one year. Taylor (1998) tested two Type V
cements with and without 3% limestone using ASTM C 452 in two different labs. The results
showed no significant difference in 14-day expansions due to the use of limestone. Equal to
improved results were obtained when these same limestone levels in Type V cements were tested
using ASTM C 1012 (Taylor, 2001).
Taylor (1998) also showed that there was no significant difference in C 452 tests when a Type II
cement had 0, 2.5, 3.0, 3.5, or 5.0% limestone, with 14-day expansions of less than 0.03% in all
cases.. Later, Taylor (1999) found that the performance of these Type II cements using ASTM
C 1012 was also unaffected.
Taylor (1998) also reported on concrete tests done by Matthews in the UK (Matthews 1994). The
residual strengths of 100- mm concrete cubes were measured after 1, 2, and 5 years exposure in
both sodium and magnesium sulfate solutions. In all exposures, after 5 years, the 5% limestone
cement retained higher strengths than the control cement without limestone.

The use of limestone at levels of 2.75 and 5.50% were evaluated as part of PCAs long-term test
program to evaluate concrete performance in sulfate soils (Stark, 2002). The conclusion from this
study was that limestone [in] Type I portland cement resulted in performance similar to the
concrete mixture with no addition. After 8 years the concretes with and without limestone all
recorded a visual rating of 5 (worst rating) and the poor performance is attributed to the high C3 A
content (~ 13%) of the portland cement.
The most recent data published by a group of researchers at Sheffield University (Torres et al.
2002) concluded that even a small amount of limestone, just 5%, significantly impacted the
performance of 8.5% C3 A cement mortars exposed to a strong solution of 1.8% MgSO4 at 5C for
5 years. However, close reading of this report reveals that during the last year the storage
condition was not maintained and the solution was permitted to evaporate exposing the mortar
bars to concentrated solutions and atmospheric CO2 as the bars slowly dried out. Prior to this,
reports from the same group (e.g. Hartshorn et al. 2001; 2002) indicated that only small amounts
of thaumasite were formed in the mortar with 5% limestone and engineering properties were not
much affected. It is not clear to what extent the uncontrolled storage conditions influenced the
performance of these mortars, but it is of interest that the control specimens (with no limestone)
also deteriorated during this last years storage and that there was evidence of thaumasite (or
ettringite-thaumasite solid solution) formation in these bars. Thomas et al (1993; 2002) reported
evidence of deterioration due to TSA in mortar bars produced with CSA Type 50 sulfate-resisting
cement (without limestone) when the bars were exposed to wet-dry cycling in sulfate solutions,
the thaumasite forming in the carbonated surface layers of the mortar.

SUMMARY
Field experience with up to 5% limestone in portland cement in Canada and Europe for over
20 years has not produced any known cases where this has contributed to thaumasite sulfate
attack. Based on the literature reviewed, there does not appear to be any significantly increased
susceptibility to sulfate attack with respect to use of up to 5% limestone in portland cements.
Research does exist concerning much higher levels of limestone (15 to 35%, or where the
carbonate fines originate from the aggregate), where used in cold temperatures combined with
wet and aggressive sulfate environments, that indicates more susceptibility to the thaumasite
form of sulfate attack. Overall, the data available support the proposal to permit use of up to 5%
limestone in portland cements.

ACKNOWLEDGEMENTS
This literature search and review (PCA R&D Serial Number 2658) was conducted by the authors
with the sponsorship of the Portland Cement Association. The contents of this report reflect the
views of the authors, who are responsible for the facts and accuracy of the information presented.

REFERENCES
Barker, A.P. and Hobbs, D.W. 1999. Performance of portland limestone cements in mortar
prisms immersed in sulfate solutions at 5C. Cement and Conc rete Composites, Vol. 21, pp.
129-137.
Bickley, J.A., Hemmings, R.T., Hooton, R.D. and Balinski, J. 1994. Thaumasite Related
Deterioration of Concrete Structures. ACI SP-144, pp.159-175.
Borsoi, A., Collepardi, S. Coppola, L., Troli, R and Collepardi, M. 2002. Sulfate Attack on
Blended Portland Cements. www.encosrl.it/enco%20srl%20ITA/ servizi/pdf/degrado/56.pdf
Brown, P.W. and Hooton, R.D. 2002. Ettringite and Thaumasite Formation in Laboratory
Concretes Prepared using Sulphate-Resisting Cement, Cement and Concrete Composites, Vol.
24, pp. 361-370.
Crammond, N.J. 2002a. The occurrence of thaumasite in modern construction. Cement and
Concrete Composites, Vol. 24, pp. 393-402.
Crammond, N.J. 2002b. The thaumasite form of sulfate attack in the U.K. Proceedings of the
1st International Conference on Thaumasite in Cementitious Materials, Building Research
Establishment, U.K.
French, W.J. 2000. Sub-aqueous carbonation and the formation of thaumasite in concrete.
Proceedings of the Institute of Materials Cement and Concrete Science Conference, Sheffield
University, September.
Hagelia, P., Sibbick, T., Crammond N.J., Gronhaug, A. and Larsen, C.K. 2001. Thaumasite and
Subsequent Secondary Calcite Deposition in Sprayed Concrete in Contact with Sulfate AlumBearing Shale, Oslo, Norway, Proceedings of the Euroseminar on Microscopy Applied to
Building Materials, Athens, Sept.4-7, pp. 131-138.
Halliwell, M.A., Crammond, N.J. and Barker, A.P. 1996. The thaumasite form of sulfate attack
in limestone-filled cement mortars. BRE Laboratory Report - BR307, Construction Research
Communications Ltd (CRC), UK.
Hartshorn, S. and Sims, I. 1998. Thaumasite a brief guide for engineers. Concrete, Vol. 32,
pp. 24-27.
Hartshorn, S.A., Sharp, J.H. and Swamy, R.N. 1999. Thaumasite formation in portland
limestone cement pastes. Cement and Concrete Research, Vol. 29, pp. 1331-1340.
Hartshorn, S.A., Sharp, J.H. and Swamy, R.N. 2001. Engineering properties and structural
implications of portland limestone cement mortar exposed to magnesium sulfate attack.
Advances in Cement Research, Vol. 13, pp. 31-46.
Hartshorn, S.A., Sharp, J.H. and Swamy, R.N. 2002. The thaumasite form of sulfate attack in
portland- limestone cement mortars stored in magnesium sulfate solution. Cement & Concrete
Composites, 24, pp.351-359.

Hawthorn, F. 1989. Ten years experience with composite cements in France. In Performance
of Limestone-Filled Cements Proceedings of a Seminar held at BRE in 1989, BR 245, Building
Research Establishment, Garston, U.K. (published 1993).
Higgins, D.D. 2002, Increased sulfate resistance of ggbfs concrete in the presence of carbonate.
Proceedings of the 1st International Conference on Thaumasite in Cementitious Materials,
Building Research Establishment, U.K.
Higgins, D.D. and Crammond, N.J. 2002, Resistance of concrete containing ggbfs to the
thaumasite form of sulfate attack, Proceedings of the 1st International Conference on Thaumasite
in Cementitious Materials, Building Research Establishment, U.K.
Hooton, R.D. 1990. Effects of carbonate additions on heat of hydration and sulfate resistance of
portland cements, in Carbonate Additions to Cement, ASTM STP 1064, 1990, pp.73-81.
Justnes, H. 2002. Thaumasite Formed By Sulfate Attack on Mortar With Limestone Filler,
Proceedings, Proceedings of the 1 st International Conference on Thaumasite in Cementitious
Materials, Building Research Establishment, U.K.
Kakali, G., Tsivilis, S., Skaropoulou, A., Sharp, J.H. and Swamy, R.N. 2002. Parameters
affecting thaumasite formation in limestone cement mortar. Proceedings of the 1st International
Conference on Thaumasite in Cementitious Materials, Building Research Establishment, U.K.
Matthews, J.D. 1989. Background to the joint working party and scope of programme. In
Performance of Limestone-Filled Cements Proceedings of a Seminar held at BRE in 1989, BR
245, Building Research Establishment, Garston, U.K. (published 1993).
Matthews, J.D. 1994. Performance of limestone filler cement concrete. In Euro-Cements
Impact of ENV 197 on Concrete Construction, (Ed. R. K. Dhir and M. R. Jones), E&FN Spon,
London, pp. 113-147.
Rogers, C., Thomas, M. and Lohse, H. 1997. Thaumasite from Manitoba and Ontario, Canada.
Proceedings of the 19th International Conference on Cement Microscopy, Cincinatti, U.S.A.
Sims, I. and Huntley, S.A. 2002. The thaumasite form of sulfate attack breaking the rules.
Proceedings of the 1st International Conference on Thaumasite in Cementitious Materials,
Building Research Establishment, U.K.
Smallwood, I., Wild, S. and Morgan, E. 2002. The resistance of metakaolin-portland cement
concrete to the thaumasite-type of sulfate attack-programme of research and preliminary results
Proceedings of the 1st International Conference on Thaumasite in Cementitious Materials,
Building Research Establishment, U.K.
Stark, D. 2002. Performance of Concrete in Sulfate Environments, PCA R&D Bulletin RD129,
23p.
Taylor P.C., 1998. Sulfate Resistance Tests on Type V Cements Containing Limestone. PCA
R&D Serial No. 2182a, 8p.

Taylor, P.C. 1999. Performance of low-C3A content cements containing interground

limestone. PCA R&D Serial 2182, 13p.
Taylor P.C., 2001. Sulfate Resistance Tests on Type V Cements Containing Limestone. PCA
R&D Serial No. 2182b, 10p.
Thaumasite Expert Group, 1999. The Thaumasite Form of Sulfate Attack: Risks, Diagnosis,
Remedial Works and Guidance on New Construction, Report of the Thaumasite Expert Working
Group, DETR, London, 180p.
Thomas, M.D.A., Rogers, C.A. and Bleszynski, R.F. 2002. Occurrences of thaumasite in
laboratory and field concrete. Proceedings of the 1st International Conference on Thaumasite in
Cementitious Materials, Building Research Establishment, U.K.
Thomas MDA, Hooton RD, Mukherjee PK. 1993. "Deleterious effects of fly ash and bottom ash
fills on embedded concrete." Proceedings of the Tenth International Ash Use Symposium,
Volume 1, EPRI TR-101774, Electric Power Research Institute, Palo Alto, Ca., pp 16.1-16.13.
Torres, S.M., Sharp, J.H., Swamy, R.N., Lynsdale, C.J. and Huntley, S.A. 2002. Long-term
durability of portland limestone cement exposed to magnesium sulfate attack. Proceedings of the
1st International Conference on Thaumasite in Cementitious Materials, Building Research
Establishment, U.K.