Construction and Building Materials 225 (2019) 302–310

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Effect of dolomite powder on the hydration and properties of calcium

sulfoaluminate cements with different gypsum contents
Jiangtao Xu a, Jingwei Chen a, Duyou Lu a,⇑, Zhongzi Xu a,⇑, R.D. Hooton b
a
College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, China
b
Department of Civil Engineering, University of Toronto, Toronto, ON M5S 1A4, Canada

h i g h l i g h t s

Dolomite promotes the hemicarbonate formation and ettringite stabilization.

The strength loss of CSA cements is mitigated by dolomite additions.

The reactivity of dolomite in CSA cements is lower than that of limestone.

It exhibits the feasibility to utilize dolomite in CSA cementitious materials.

a r t i c l e i n f o a b s t r a c t

Article history: The present study investigates the effect of dolomite powder on the hydration and properties of calcium
Received 20 February 2019 sulfoaluminate (CSA) cements with different gypsum contents, with limestone powder as a comparison.
Received in revised form 17 June 2019 Hydration kinetics was tested using isothermal calorimetry. The hydration assemblage was examined
Accepted 7 July 2019
using X-ray diffraction and thermal analysis. Results indicate that the hydration of CSA cements is accel-
erated with the addition of either dolomite or limestone powder. Similar to limestone, dolomite can par-
ticipate in hydration reaction to form hemicarbonate and to stabilize ettringite when gypsum is
Keywords:
insufficient, but dolomite is less reactive due to its slow dissolution rate. The reaction of the carbonates
Calcium sulfoaluminate cement
Hydration
is suppressed with the increase in gypsum content. Both compressive strength and expansion of CSA
Dolomite powder mortars decrease with the addition of dolomite, regardless of the gypsum contents. In general, the dolo-
Limestone powder mite and limestone modified mortars show similar strength and expansion, however except a higher
expansion in the dolomite mortar without gypsum. Moreover, the strength loss of CSA mortar is miti-
gated when dolomite or limestone is used.
Ó 2019 Elsevier Ltd. All rights reserved.

1. Introduction footprint but also has technical benefits in concrete. Limestone

Using supplementary cementitious materials (SCMs) to replace
cement clinker is an economical and practical method to reduce
CO2 emissions and promote use of sustainable cementitious mate-
rials. However, the current supply of common SCMs such as fly ash
and slag are often inadequate relative to the large scale of cement
production [1]. Therefore, new sources of SCMs beyond those cur-
rently available need to be developed.
Limestone has been commonly used as a mineral addition to
cement around the world. It is allowed to replace up to 15% clinker
in North America [2], and even up to 35% in Europe [3]. The use of
portland-limestone cement not only lowers the environmental
⇑ Corresponding authors at: College of Materials Science and Engineering,
Nanjing Tech University, No. 30 Puzhu Road(S), Nanjing, China.
E-mail addresses: duyoulu@njtech.edu.cn (D. Lu), zzxu@njtech.edu.cn (Z. Xu).
may act as a filler to provide more nucleation sites for precipitation
of hydrates and to accelerate cement hydration [4,5]. In the pres-
ence of limestone, hemi- and monocarbonate are formed, which
prevent the conversion of ettringite to monosulfate and therefore
contributes to strength development [6–9]. According to both
EN197-1 and CSA A3001, the limestone powder used should con-
tain not less than 75% calcite content. This requirement has limited
the use of carbonate sources in cement because some limestone
deposits, such as dolomitic limestone, cannot meet these
standards.
Dolomite CaMg (CO3)2 is an abundant carbonate mineral, and it
frequently coexists with calcite in nature. Unlike limestone, dolo-
mite is currently not accepted as a filler or component of cement
because some forms of dolomitic aggregates might cause concrete
damage due to dedolomitization. However, it has already been
reported that the addition of dolomite in powder form is not

https://doi.org/10.1016/j.conbuildmat.2019.07.050
0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.
 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310 303

problematic and when added to cement has a comparable compres- performance. Moreover, there is relatively little research on the
sive strength to portland-limestone cement [10–12]. Moreover, sim- use of dolomite in CSA cements.
ilar to limestone, dolomite may be involved in hydration of cement,
C4 A3 S   H12 þ 2AH3
 þ 18H ! C3 A  CS ð1Þ
resulting in the formation of carboaluminates and the stabilization
of ettringite. Besides, Mg released from dolomite is considered to
 þ 2CS
C4 A3 S   H12 þ 2AH3
 þ 38H ! C3 A  3CS ð2Þ
interact with the cement aluminates to form hydrotalcite, which
leads to a decrease in porosity and an increase in compressive In this study, the hydration and properties of CSA cements
strength [13,14]. Concerning durability, some studies have reported blended with dolomite at different M values (0, 1 and 2) were
that carbonation, chloride penetration resistance and dimensional investigated. Similar mixtures containing limestone were prepared
stability of portland-dolomite cement is satisfactory, while poor for comparison. Hydration kinetics was assessed by isothermal
freezing-thawing properties have been reported due to slightly calorimetry. The phase developments of the hydrates were moni-
higher affinity for capillary water suction [10,15–17]. tored with X-ray diffraction and thermogravimetric analysis.
Calcium sulfoaluminate (CSA) cement is a type of cement con-
sisting of the dominant mineral ye’elimite, and has attracted
extensive attention because of its manufacture has low-CO2 emis- 2. Experimental
sion and uses less energy [18]. Normally, about 15–25 wt% of cal-
cium sulfate is interground with the clinker for optimizing 2.1. Materials and mix design
setting time, strength development and dimensional stability
[19]. The primary hydration products of CSA cement are crystalline Commercial CSA clinker, dolomite, limestone, and analytical
monosulfate and ettringite as well as amorphous aluminum grade gypsum were used in this study. The chemical compositions
hydroxide (Eqs. (1) and (2)). The molar ratio of calcium sulfate to determined by XRF are shown in Table 1. The mineral composition
ye’elimite, known as the M value, determines the ratio of ettringite of the clinker was determined by quantitative XRD using Rietveld
to monosulfate that forms. Based on the special hydration proper- refinement; it mainly consists of 58% ye’elimite and 24% belite.
ties, CSA cement has outstanding properties such as rapid- The dolomite and limestone powders contain 95.4% CaMg (CO3)2
hardening, high-early strength, shrinkage compensation, sulfate and 95.5% CaCO3 respectively, determined by TG analysis. The par-
resistance, and chloride corrosion resistance as well as its ability ticle size distributions of the CSA clinker, dolomite and limestone
to immobilize hazardous metals [19–21]. Recently several studies determined by laser diffraction are given in Fig. 1. The CSA clinker
have been published showing that limestone addition is beneficial shows a wide particle size distribution. The two carbonate pow-
to many properties of CSA cement, including enhancement of ders have similar fineness, and both show a slightly finer particle
early-age hydration, stabilization of ettringite and reduction of size than the CSA clinker.
strength loss at later ages [22–25]. However, it is unclear whether Three series of CSA cements were prepared at varying M values
the addition of dolomite has a similarly positive effect on cement using a laboratory mixer, i.e. M = 0 (No gypsum), 1 and 2. The
replacement of dolomite and limestone powders was fixed at
20% by mass of total CSA cement. The experimental matrix is
Table 1 shown in Table 2. For each matrix, both cement pastes and mortars
Chemical and mineralogical composition of the used materials. were prepared at a constant w/c ratio of 0.6. River sand (0–0.6 mm)
was used with sand to cement ratio of 2:1. Cement pastes were
CSA clinker Dolomite Limestone CSA clinker
cast in 20 mm cube molds to examine the evolution of hydrates.
Chemical composition (wt%) Phase composition (wt%)
Mortars were cast into 30 mm  30 mm  30 mm (compressive
CaO 43.24 31.21 54.93 
C4 A3 S 57.8 strength) and 20 mm  20 mm  80 mm (dimension stability)
SiO2 11.53 0.89 0.17 C2S 23.6 molds, respectively. After 12 h, the samples were demolded and
Al2O3 28.39 0.22 0.12 C3A 4.8
then cured in tap water at 20 °C.
Fe2O3 3.21 0.10 0.03 C4AF 3.1
MgO 2.55 21.78 1.08 C12A7 3.4
TiO2 1.41 CT 4.2 2.2. Test methods
SO3 8.36 C2AS 2.0
K2O 0.71 0.05 0.01 MgO 1.8
Na2O 0.23 0.07
Setting times were recorded using the Vicat apparatus. The
LOSS 45.67 43.17 samples were measured according to the Chinese standard GB/T
1346–211.

Fig. 1. Particle size distribution of raw materials (a) differential particle size distribution (b) cumulative particle size distribution.
304 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310

The heat evolution was characterized by isothermal calorimetry
with a TAM 83 Air isothermal conduction calorimeter. For the anal-
ysis, 4.0 g of binder was mixed inside the apparatus with 2.4 g
water. Measurements were evaluated for 72 h at 20 °C.
After the selected period, cement pastes were crushed and
soaked in excess ethanol for 24 h to stop hydration, and they were
then stored in a vacuum desiccator over silica gel. Finally, the sam-
ples were ground to pass a 45 lm (#325) mesh sieve for X-ray
diffraction (XRD) and thermogravimetric analyses (TGA).
X-ray diffraction (XRD) patterns for the sample were obtained
using an X-ray diffractometer (SmartLabTM 3 Kw Powder Diffrac-
tometer; Cu Ka1, k = 1.5406 Å). The X-ray tube was operated at
40 kV and 30 mA, a scanning speed of 5°/min with a step of 0.02°
was used. The evolution of the amount of ye’elimite and gypsum
over time was monitored from XRD patterns by measuring the
areas of corresponding reflections using the Jade 6.5 software.
Thermogravimetric analysis (TGA) was performed via simulta-
neous thermal analyzer (STA 449C, Germany) heated to 1000 °C
at a heating rate of 10 °C/min under an N2 atmosphere, at a flow
rate of 20 ml/s. TGA was conducted for the semi-quantitative
determination. The amount of ettringite in the hydrates was deter-
mined by assuming that the weight loss between 50 °Cand 120 °C
correspond to loss of 20 molecules of crystal water per molecule of
ettringite. Monosulfate content was calculated from the TGA

Table 2
Experimental matrix.

M* = 0 (No gypsum) M* = 1 M* = 2
(wt%) CSA DM LS CSA1 DM1 LS1 CSA2 DM2 LS2
CSA clinker 100 80 80 86 68.8 68.8 75.4 60.3 60.3
Gypsum 14 12.2 12.2 24.6 19.7 19.7
Dolomite 20 20 20
Limestone 20 20 20

M* = the molar ratio of gypsum to ye’elimite.

Table 3
Setting times of tested pastes.

CSA DM LS CSA1 DM1 LS1 CSA2 DM2 LS2

Initial set (min) 21 23 19 24 27 24 26 32 23
Final set (min) 30 32 29 32 38 30 35 43 33

Fig. 2. Heat flow of the specimens with different constituents (a) M0, (b) M1 and (c) M2.
 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310
weight loss between 160 °C and 220 °C [26,27]. Additionally, the
reaction degree of two carbonate powders was also calculated
from the TGA weight loss between 580 °C and 700 °C. Note that
the sample weight at 550 °C was used as the dry weight, which
was assumed to be constant during hydration. To compare the dif-
ference, all quantitative results were normalized to the dry weight.
The compressive strength of mortar cubes (at 1 d, 3 d, 7 d, 28 d,
and 90 d) cured in water were tested in a WHY-200 compression
machine at a loading rate of 1 mm/min. Six replicate specimens
were tested for each batch. The measured strength values were
corrected by a factor of 1.78 to take into account the difference
in area with standard prisms (40 mm  40 mm  160 mm), as
described in [28].
The expansion of mortar prisms was measured periodically. The
results are the average of the measurements on triplicate
specimens.
3. Results and discussion
3.1. Setting times
The setting times of all tested pastes are given in Table 3. It is
observed that in the M0 system without gypsum, the initial and
final setting times are slightly prolonged by replacing CSA clinker
with dolomite, compared to the plain CSA clinker with and without
limestone. When gypsum is added to CSA cement, an increase of
the setting time is found. For the M1 and M2 systems, the retarda-
tion of the initial and final setting is more pronounced in dolomite
samples than in limestone samples. This finding is similar to the
effect of dolomite on the properties of portland cement [12].
Fig. 3. Cumulative heat of the specimens with different constituents (a) M0, (b) M1 and (c) M2.
305
3.2. Calorimetry
Fig. 2 depicts the heat flow of the three systems made with dif-
ferent M values. As shown in Fig. 2a, all samples without gypsum
exhibit a very short dormant period after the initial peak around
10 min of hydration. The initial peak is possibly due to the wetting
and dissolution of solid materials. After the dormant period, a wide
diffused peak corresponding to the hydration of ye’elimite can be
observed. The replacement of CSA clinker by dolomite or limestone
results in a slightly earlier onset of the main peak, with the two
carbonate modified samples showing a similar behavior in heat
curves. The results suggest that similar to limestone, dolomite
can accelerate the hydration of CSA clinker. For the M1 and M2 sys-
tems (Fig. 2b–c), the time of the main peak is obviously shortened
due to the presence of gypsum, and the main peak is followed by a
shoulder, which is ascribed to the depletion of gypsum [29,30].
Moreover, it is noticed that the shoulder peak is shifted to earlier
times and increases in height when dolomite powder is used,
which suggests that the observed acceleration in early-age hydra-
tion is taking place in the two systems with gypsum.
To evaluate the contribution of dolomite to the cement hydra-
tion, the total heat release normalized to CSA cement is shown in
Fig. 3. It can be seen that the dolomite replacement leads to a sig-
nificant increase in the total heat release in the M0, M1 and M2
systems, regardless of the gypsum content. These effects are poten-
tially associated with the addition of dolomite powder, which pro-
vides the additional nucleation sites and high effective w/c ratio for
the precipitation of hydrates, as reported in the use of limestone in
cement [24,31]. Additionally, in the M1 and M2 systems with gyp-
sum, there is a slightly lower total heat release for the dolomite
samples than for the limestone samples. This could be related to
306 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310

the weaker nucleation effect of dolomite than limestone because of
the more negative charges of dolomite [32], which has been
reported to produce a weaker effect in accelerating hydration of
portland cement [32–34].
3.3. The change of ye’elimite and gypsum contents
Table 4 shows the semi-quantitative results based on X-ray
diffraction analysis for ye’elimite and gypsum. It is obvious that
in all systems with or without gypsum, the amounts of depleted
ye’elimite and gypsum increase with the replacement of CSA
cement by dolomite or limestone powder respectively, especially
at early age, establishing that the presence of dolomite and lime-
stone enhances the reaction of ye’elimite and gypsum, which
agrees with the calorimetry results. This effect is slightly more pro-
nounced for the limestone samples than for the dolomite samples.
Besides, consistent with the literature [35], the presence of gypsum
leads to accelerating the beginning of hydration but limiting the
final hydration degree, which may be associated with the denser
cement matrix due to the rapid formation of ettringite.
3.4. X-ray diffraction analysis
Fig. 4 shows the XRD patterns of the three systems with differ-
ent M values cured for 7 days and 90 days. For the M0 system
(Fig. 4a), all samples display the presence of ettringite and mono-

Table 4
The amounts of depleted ye’elimite and gypsum as a function of time.

Ye’elimite (wt%) Gypsum (wt%)

6h 1d 7d 90d 6h 1d 7d 90d
CSA 20.5 73.0 89.7 99.2
DM 25.5 79.8 93.4 100
LS 27.2 80.7 94.1 100
CSA1 51.6 75.7 93.0 97.6 94.9 97.7 100
DM1 57.3 83.5 96.4 99.0 96.8 100
LS1 59.1 84.7 96.1 99.2 97.6 100
CSA2 68.2 81.5 89.8 91.1 74.6 92 93.4 96.3
DM2 75.7 92.5 94.5 96.1 86.2 95.1 95.9 97.7
LS2 77.4 92.0 95.0 95.8 89.6 96.4 96.8 97.6

Fig. 4. XRD patterns of the specimens with different constituents (a) M0, (b) M1 and (c) M2. E: AFt; M: AFm12; M*: AFm14; Hc: hemicarbonate; S: strätlingite; G: gypsum.
 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310
sulfate (AFm12 or AFm14) phases, presumably due to the absence of
gypsum. The CSA clinker sample shows the formation of strätlin-
gite after 7 days of hydration, which is associated with the reaction
of belite and gibbsite according to Eq. (3) [36]. The formation of
hemicarbonate instead of strätlingite and the reduction of mono-
sulfate can be observed in the samples containing dolomite and
limestone. It is related to the reaction of carbonate, which provides
CO2
3 favoring the precipitation of carboaluminates rather than
strätlingite and monosulfate [34,37]. Moreover, the formed hemi-
carbonate is less pronounced in the dolomite-modified sample,
potentially ascribed to the slow dissolution rate of dolomite com-
pared to limestone [38,39]. Besides, it should be noted that after
90 days the transformation of AFm14 into AFm12 occurs in all sam-
ples, which is accompanied by enhanced formation of ettringite.
For the M1 system with intermediate gypsum (Fig. 4b), strätlingite
is not formed after 7 and 90 days, and all tested samples reveal
similar main hydrated phases, including a large amount of ettrin-
gite and a trace of monosulfate. Similarly, the formation of hemi-
carbonate has also been detected in the two carbonate-modified
samples, but their peaks are weaker than those in the M0 system.
The increased gypsum content of the M2 system (Fig. 4c), after
7 days of hydration, clearly results in producing more ettringite
than for samples hydrated in the M0 and M1 systems and no hemi-
carbonate is found except for residual gypsum, and these hydrates
do not change up to 90 days. According to the above results, it can
be concluded that dolomite can participate in hydration reactions
to form hemicarbonate as does limestone when there is insuffi-
cient gypsum, but it does not react or only acts as a filler in the
presence of sufficient gypsum. The result is consistent with previ-
Fig. 5. Differential thermogravimetric (DTG) curves of the specimens with different constituents for 90 days (a) M0, (b) M1 and (c) M2. E: AFt; M: monosulfate; Hc:
hemicarbonate; S: strätlingite; G: gypsum.
307
ous work that has reported where increasing the gypsum content
significantly reduces the extent of limestone reaction [24,40].
C2 S þ AH3 þ 5H ! C2 ASH8 ð3Þ
3.5. Thermogravimetric analysis
The DTG curves of the three systems with different gypsum
contents are shown in Fig. 5. It can be found that ettringite (at
about 120 °C) and aluminum hydroxide (at about 250 °C) are pre-
sented in all investigated samples. For the M0 system, the presence
of dolomite or limestone decreases the amount of monosulfate and
strätlingite, and results in the formation of hemicarbonate, which
is consistent with XRD analysis. In contrast, no hemicarbonate
could be distinguished in M1-system DTG curves. This contrast
can potentially be explained by the small amount of hemicarbon-
ate and its poor crystallinity. To further analyze the difference in
hydrates between the samples containing dolomite and limestone,
the amount of ettringite and monosulfate, as well as dolomite or
calcite is simply calculated as described previously and the results
are shown in Table 5. For the M0 and M1 systems, the dolomite-
modified samples show a lower ettringite and a higher monosul-
fate content when compared to the limestone samples, which is
accompanied by lower consumption of dolomite. This is attributed
to the low CO23 supply by the reacted dolomite and consequently
causes less hemicarbonate formation (Fig. 4a–b), which favors pre-
venting the transformation of ettringite to monosulfate due to
greater stability of carboaluminate phases [7]. This result indicates
that dolomite is less reactive than limestone, and the ettringite sta-
308 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310

Table 5 3.7. Dimension stability

The amounts of ettringite, monosulfate and carbonates at 90 days.

(wt%) DM LS DM1 LS1 DM2 LS2 Fig. 7 displays the dimension stability of mortars with different
Ettringite 29.1 31.2 47.2 48.5 55.6 55.3 constituents. As shown in Fig. 7, for the M1 and M2 systems with
Monosulfate 9.9 9.2 6.8 6.6 gypsum, all mortars show a rapid expansion up to about 7 days,
*
Dolomite 18.2 19.3 20.1 and then the expansion reaches a plateau throughout the whole
Limestone 17 18.9 20.3* period up to 120 days. As expected, higher expansion is obtained
*
The content is slightly higher than 20 wt% possibly due to the experiment for the mortars of the M2 system due to the presence of high gyp-
errors. sum content. The replacement of CSA cements by dolomite signif-
icantly reduces the expansion. However, it is interesting to note
that CSA mortar without gypsum expands steadily and yields the
bilization effect is less obvious in the dolomite samples. Besides, highest expansion among all investigated mortars, which could
dolomite or limestone powder remains stable when enough gyp- be associated with the formation of strätlingite and retarded
sum is added. ettringite as shown in XRD analysis. This is in agreement with
the finding of Berger [35]. Importantly, a smaller decrease in
3.6. Compressive strength expansion is observed in DM mortars compared to LS mortars,
which can be explained by the difference in dissolution rate
Fig. 6 shows the strength measurements of mortars with differ- between dolomite and limestone powders. The slow dissolution
ent constituents after 1, 3, 28 days and 90 days. For the mortars rate of dolomite leads to less hemicarbonate formation, which in
without carbonates, CSA1 mortar with intermediate gypsum dis- turn should supply more available alumina to form strätlingite
plays the highest long-term compressive strength among all inves- and ettringite, and thus promote larger expansion.
tigated mortars, followed by CSA2 mortar with high gypsum, while
CSA mortar without gypsum shows the lowest strength. This could
be attributable to the optimum ratio of gypsum to CSA clinker.
Whatever the gypsum content, the strength is decreased with
replacement by dolomite powder. Moreover, the dolomite-
modified mortars develop a slightly lower early-age strength when
compared to the mortars with limestone, but the difference
decreases and disappears as the age increases. This is similar to
previous observations in portland cement [12]. Additionally, the
mortars without carbonate exhibit an obvious strength loss after
3 days. However, the mortars with dolomite or limestone show
no strength loss in the M0 system and less loss of strength in the
M1 and M2 systems. For example, between 1 and 3 days, the
strength loss in CSA, CSA1, and CSA2 is about 6.2%, 6.0% and
22.3%, respectively. CSA1-D exhibits no strength loss as well while
the loss of strength in CSA1-L, CSA2-D, and CSA2-L is 2.4%, 15.0%,
and 18.8%, respectively. After 3 days of hydration, the strength
increases continually for all the mortars tested. This indicates that
similar to limestone, dolomite additions may reduce the loss of
strength and improve strength evolution of CSA cement. It can be
possibly explained by the filler effect and chemical interaction,
Fig. 7. Dimension stability of the investigated mortars over hydration time.
which creates more stable hydration products.

Fig. 6. Compressive strength of the investigated mortars over hydration time.

 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310
4. Summary
The purpose of this paper was to investigate the effect of dolo-
mite powder on hydration and properties of CSA cements with dif-
ferent gypsum contents, with limestone powder as a comparison.
From the experimental results, the following conclusions can be
drawn:
1. The addition of dolomite to CSA cements delays the initial and
final setting times, regardless of the gypsum contents. Similar to
limestone, dolomite additions can accelerate the hydration pro-
cess, especially at early ages. This is ascribed to the fact that
dolomite powder provides more nucleation sites and a high
effective w/c ratio.
2. The reaction of dolomite in CSA cement is suppressed with
increasing amounts of gypsum. In the system with insufficient
gypsum (the M0 and M1 systems), the formation of hemicar-
bonate takes place in dolomite samples similar to limestone
samples, with less hemicarbonate forming in the dolomite sam-
ples due to its slower dissolution rate. The hemicarbonate
formed stabilizes ettringite and minimizes the precipitation of
strätlingite and monosulfate. In the presence of excess gypsum
(the M2 system), however, dolomite does not react and acts
only as a filler.
3. In all investigated systems, the replacement of CSA cement by
20 wt% dolomite leads to decreasing compressive strength and
expansion. However, the mortars containing dolomite and
limestone have comparable compressive strength, except a
slightly lower early-age strength in the dolomite mortars.
Moreover, both dolomite and limestone can reduce the strength
loss of mortars at 3 days and improve strength development. In
the absence of gypsum, a higher expansion can be observed
with the replacement by dolomite compared with the lime-
stone mortar. When gypsum is added, the two mortars show
similar expansion patterns.
It can be concluded that, in general, dolomite and limestone
show similar effects on the hydration and properties of CSA
cements made with different gypsum contents. This finding makes
possible the utilization of dolomite powder in CSA cementitious
materials.
Declaration of Competing Interest
The authors declared that they have no conflict of interest in the
submission of this manuscript, and manuscript is approved by all
authors for publication. I would like to declare on behalf of my
co-authors that the work described was original research that has
not been published previously, and not under consideration for
publication elsewhere, in whole or in part. All the authors listed
have approved the manuscript that is enclosed.
Acknowledgments
This work is supported by the National Natural Science Founda-
tion of China (51472116), the Graduate Student Scientific Research
Innovation Projects in Jiangsu province and the Priority Academic
Program Development (PAPD) of Jiangsu Higher Education
Institutions.
References
[1] K. Scrivener, Options for the future of cement, Indian Concr. J. 88 (7) (2014)
11–21.
[2] R.D. Hooton, M. Nokken, M.D.A. Thomas, Portland-limestone Cement: State-of-
the-Art Report and Gap Analysis for CSA A 3000, SN3053, Cement Association
of Canada, 2007, p. 59.
309
[3] European Committee for Standardization, Cement: Composition, Specifications
and Conformity Criteria, Part 1: Common Cements, EN 197-1, EN/TC51/WG 6
rev., 2000.
[4] D.H. Wang, C.J. Shi, N. Farzadnia, Z.G. Shi, H.F. Jia, Z.H. Qu, A review on use of
limestone powder in cement-based materials: mechanism, hydration and
microstructures, Constr. Build. Mater. 181 (2018) 659–672.
[5] D.H. Wang, C.J. Shi, N. Farzadnia, H.F. Jia, R. Zeng, Y.W. Wu, L.L. Lao, A
quantitative study on physical and chemical effects of limestone powder on
properties of cement pastes, Constr. Build. Mater. 204 (2019) 58–69.
[6] T. Matschei, B. Lothenbach, F.P. Glasser, The role of calcium carbonate in
cement hydration, Cem. Concr. Res. 37 (4) (2007) 551–558.
[7] B. Lothenbach, G. Le Saout, E. Gallucci, K. Scrivener, Influence of limestone on
the hydration of Portland cements, Cem. Concr. Res. 38 (6) (2008) 848–860.
[8] P.D. Tennis, M.D.A. Thomas, W.J. Weiss, State-of-the-Art Report on Use of
Limestone in Cements at Levels of up to 15%, SN3148, Portland Cement
Association, 2011, p. 78.
[9] D.H. Wang, C.J. Shi, N. Farzadnia, Z.G. Shi, H.F. Jia, A review on effects of
limestone powder on the properties of concrete, Constr. Build. Mater. 192
(2018) 153–166.
[10] S. Schöne, W. Dienemann, E. Wagner, Portland dolomite cement as alternative
to Portland limestone cement, in: Proceedings of the 13th International
Congress on the Chemistry of Cement, Madrid, Spain, July 2011, pp. 1-7.
[11] O. Mikhailova, G. Yakovlev, I. Maeva, S. Senkov, Effect of dolomite limestone
powder on the compressive strength of concrete, Proc. Eng. 57 (2013) 775–780.
[12] S.H. Zhang, D.Y. Lu, Z.Z. Xu, Effect of dolomite powders on the hydration and
strength properties of cement mortars, in: Proceedings of the 14th
International Congress on the Chemistry of Cement, Beijing, China, October
2015, pp. 13–16.
[13] M. Zajac, S.K. Bremseth, M. Whitehead, M. Ben Haha, Effect of CaMg(CO3)2 on
hydrate assemblages and mechanical properties of hydrated cement pastes at
40 °C and 60 °C, Cem. Concr. Res. 65 (2014) 21–29.
[14] J.T. Xu, D.Y. Lu, S.H. Zhang, K. Ling, Z.Z. Xu, Pore structures of mortars with
dolomite and limestone powders cured at various temperatures, J. Chin.
Ceram. Soc. 45 (2) (2017) 268–273.
[15] A. Machner, M. Zajac, M. Ben Haha, K. Kjellsen, M. Geiker, K. Weerdt, Chloride-
binding capacity of hydrotalcite in cement pastes containing dolomite and
metakaolin, Cem. Concr. Res. 107 (2018) 163–181.
[16] A. Machner, M. Zajac, M. Ben Haha, K. Kjellsen, M. Geiker, K. Weerdt, Stability
of the hydrate phase assemblage in Portland composite cements containing
dolomite and metakaolin after leaching, carbonation, and chloride exposure,
Cem. Concr. Compos. 29 (2018) 89–106.
[17] J.T. Xu, D.Y. Lu, K. Ling, Z.Z. Xu, Volume stability of Portland dolomite cement
pastes cured at elevated temperatures, in: Proceedings of the 9th International
Symposium on Cement and Concrete, Wuhan, China, October 2017, pp. 213–
220.
[18] M.C.G. Juenger, F. Winnefeld, J.L. Provis, J.H. Ideker, Advances in alternative
cementitious binders, Cem. Concr. Res. 41 (12) (2011) 1232–1243.
[19] F.P. Glasser, L. Zhang, High-performance cement matrices based on calcium
sulfoaluminate-belite compositions, Cem. Concr. Res. 31 (12) (2001) 1881–
1886.
[20] C.A. Luz, J. Pera, M. Cheriaf, J.C. Rocha, Behaviour of calcium sulfoaluminate
cement in presence of high concentrations of chromium salts, Cem. Concr. Res.
37 (4) (2007) 624–629.
[21] S. Peysson, J. Pera, M. Chabannet, Immobilization of heavy metals by calcium
sulfoaluminate cement, Cem. Concr. Res. 35 (12) (2005) 2261–2270.
[22] L. Pelletier-Chaignat, F. Winnefeld, B. Lothenbach, C.J. Müller, Beneficial use of
limestone filler with calcium sulphoaluminate cement, Constr. Build. Mater.
26 (1) (2012) 619–627.
[23] C.W. Hargis, A. Telesca, P.J.M. Monteiro, Calcium sulfoaluminate (Ye’elimite)
hydration in the presence of gypsum, calcite, and vaterite, Cem. Concr. Res. 65
(2014) 15–20.
[24] L.H.J. Martin, F. Winnefeld, C.J. Müller, B. Lothenbach, Contribution of
limestone to the hydration of calcium sulfoaluminate cement, Cem. Concr.
Compos. 62 (2015) 204–211.
[25] Y. Jeong, C.W. Hargis, S. Chun, J. Moon, Effect of calcium carbonate fineness on
calcium sulfoaluminate-belite cement, Materials 10 (8) (2017) 900.
[26] P.M. Wang, N. Li, L.L. Xu, Hydration evolution and compressive strength of
calcium sulphoaluminate cement constantly cured over the temperature range
of 0 to 80 °C, Cem. Concr. Res. 100 (2017) 203–213.
[27] B. Lothenbach, E. Wieland, A thermodynamic approach to the hydration of
sulphate-resisting Portland cement, Waste Manage 26 (7) (2006) 706–719.
[28] G. Alvarez-Pinazo, I. Santacruz, M. Aranda, A. Torre, Hydration of belite-
ye’elimite-ferrite cements with different calcium sulfate sources, Adv. Cem.
Res. 26 (2016) 529–543.
[29] F. Winnefeld, S. Barlag, Calorimetric and thermogravimetric study on the
influence of calcium sulfate on the hydration of ye’elimite, J. Therm. Anal.
Calorim. 101 (3) (2010) 949–957.
[30] D. Jansen, A. Spies, J. Neubauer, D. Ectors, F. Goetz-Neunhoeffer, Studies on the
early hydration of two modifications of ye’elimite with gypsum, Cem. Concr.
Res. 91 (2017) 106–116.
[31] K. Vance, M. Aguayo, T. Oey, G. Sant, N. Neithalath, Hydration and strength
development in ternary portland cement blends containing limestone and fly
ash or metakaolin, Cem. Concr. Compos. 39 (2013) 93–103.
[32] S.H. Zhang, D.Y. Lu, J.T. Xu, K. Ling, Z.Z. Xu, Effects of dolomite powders on the
early hydration properties of blend cement and its mechanism, J. Nanjing Tech
Univ. 4 (2017) 60–66.
310 J. Xu et al. / Construction and Building Materials 225 (2019) 302–310
[33] M. Zajac, W. Dienemann, G. Bolte, Comparative experimental and virtual
investigation of the influence of calcium and magnesium carbonates on
reacting cement, in: Proceedings of the 13th International Congress on the
Chemistry of Cement, Madrid, Spain, July 2011.
[34] S. Krishnan, S. Bishnoi, Understanding the hydration of dolomite in
cementitious systems with reactive aluminosilicates such as calcined clay,
Cem. Concr. Res. 108 (2018) 116–128.
[35] S. Berger, C.C.D. Cournes, P. Le Bescop, D. Damidot, Influence of a thermal cycle
at early age on the hydration of calcium sulphoaluminate cements with
variable gypsum contents, Cem. Concr. Res. 41 (2) (2011) 149–160.
[36] F. Winnefeld, B. Lothenbach, Hydration of calcium sulfoaluminate cements –
experimental findings and thermodynamic modelling, Cem. Concr. Res. 40 (8)
(2010) 1239–1247.
[37] M. Zajac, P. Durdzinski, C. Stabler, J. Skocek, D. Nied, M. Haha, Influence of
calcium and magnesium carbonates on hydration kinetics, hydrate assemblage
and microstructural development of metakaolin containing composite
cements, Cem. Concr. Res. 106 (2018) 91–102.
[38] O.S. Pokrovsky, S.V. Golubev, J. Schott, Dissolution kinetics of calcite, dolomite
and magnesite at 25 °C and 0 to 50 atm pCO2, Chem. Geol. 217 (3–4) (2005)
239–255.
[39] O.S. Pokrovsky, S.V. Golubev, J. Schott, A. Castillo, Calcite, dolomite and
magnesite dissolution kinetics in aqueous solutions at acid to circumneutral
pH, 25 to 150°C and 1 to 55 atm pCO2: New constraints on CO2 sequestration
in sedimentary basins, Chem. Geol. 265 (1–2) (2009) 20–32.
[40] J. Bizzozero, K. Scrivener, Limestone reaction in calcium aluminate cement-
calcium sulfate systems, Cem. Concr. Res. 76 (2015) 159–169.