Ocean Engineering 157 (2018) 1–12

Contents lists available at ScienceDirect

Ocean Engineering
journal homepage: www.elsevier.com/locate/oceaneng

Behavior of mortar exposed to different exposure conditions of sulfate attack

Xiao-tong Yu a, Da Chen a, *, Jia-rui Feng a, Yan Zhang b, Ying-di Liao a
a
College of Harbour, Coastal and Offshore Engineering, Hohai University, Nanjing, 210098, China
b
College of Mechanics and Materials, Hohai University, Nanjing, 210098, China

A R T I C L E I N F O A B S T R A C T

Keywords: The sulfate deterioration of cement-based materials has become severe durability problems for the offshore
Mortar structures. Considering the various service conditions such as splash and tidal zone with/without sulfate attack,
Sulfate exposures the behavior of mortar samples under full immersion and dry-wetting cycles with 0% and 5% Na2SO4 solution,
Water permeability i.e., four exposure conditions, was investigated in this research. Compressive strength, elastic modulus, perme-
Compressive strength ability (water permeability coefficient), expansion behavior and SEM images were tested and analyzed. The re-
In-situ test
sults showed that the behavior of mortar evolved in two stages, initial improvement followed by later
degradation, determined by two competitive effects under each condition. Different deterioration mechanisms
worked in the different exposure conditions. We proposed binomial formulas to describe the evolutions of the
compressive strength and water permeability, and we further applied them in other researches well. Meanwhile,
there was a linear relationship between elastic modulus and square root of compressive strength. Furthermore, the
relation between water permeability and compressive strength was highly dependent on exposure types and time.
Study showed that compressive strength and water permeability coefficient can be selected as controlling pa-
rameters in in-situ tests, and a possible approach to evaluate the sulfate deterioration level of concrete structures
in-situ was represented.

1. Introduction cracking, softening and degradation of concrete (Whittaker and Black,

Cement-based materials have been widely applied in marine and
offshore engineering (say piers, abutments and offshore turbines) due to
their good performance and low cost. However, the aggressive marine
environment causes serious durability problems on them, shortening the
service life (Yu et al., 2017). The detrimental ions in seawater, like Cl

(chloride ions), SO2
4 (sulfate ions), Mg
2þ
(magnesium ions), induced
reinforcement corrosion and concrete deterioration (Glasser et al., 2008).
While the temperature and humidity changes in splash and tidal zones
further aggravate the durability (de Medeiros-Junior et al., 2015).
Compared with the indirect effect from chloride ions, sulfate ions dete-
riorate the concrete directly through reacting with cement hydration
products (Yin et al., 2017). And it is an adverse environmental factor in
the service of concrete structures, second to reinforcement corrosion
(Ghafoori et al., 2015). According to the deterioration features in
different exposure conditions, there are chemical attack and physical
attack. In chemical sulfate attack, sulfate ions penetrate into internal
pores and react with cement hydration products to form expansive
products (ettringite, gypsum or thaumasite), leading expansion,
2015). In physical sulfate attack (PSA), also termed as salt weathering,
humidity and temperature changes induce sulfate salt crystalline and
pressures against pore wall, causing surface scaling and flaking of con-
crete (Haynes et al., 2008). In marine environment, both attacks might
happen simultaneously to concrete structures. For instance, the structure
foundations sited in sulfate-bearing ground and seawater suffer from the
chemical sulfate attack under the water level. Whereas parts of them
undergo physical sulfate attack from dry-wetting cycles in splash and
tidal zone (Yoshida et al., 2010). Thus, it is essential to investigate the
behavior of cement-based material under different exposure conditions
of sulfate attack, for the sulfate deterioration assessment of structures
serving in marine environment.
So far, chemical sulfate attack, in form of full-immersion in sulfate
solutions, on cement-based materials has been paid much attention.
Experimental results reveal the expansion and degradation mechanism
(Whittaker and Black, 2015; Lothenbach et al., 2010; Roziere et al.,
2009). Various affecting factors, e.g., water-cement ratio, cement type,
sulfate types, temperature, pH, on the deterioration process have been
investigated (Santhanam et al., 2002; Wee et al., 2000; Monteiro and

* Corresponding author.
E-mail address: chenda@hhu.edu.cn (D. Chen).

https://doi.org/10.1016/j.oceaneng.2018.03.017
Received 25 September 2017; Received in revised form 12 January 2018; Accepted 3 March 2018
Available online 19 March 2018
0029-8018/© 2018 Elsevier Ltd. All rights reserved.
X.-t. Yu et al. Ocean Engineering 157 (2018) 1–12

Kurtis, 2003). Further, diffusion - chemo - mechanical models consid- Table 2

ering sulfate ions diffusion, chemical reactions and mechanical damages Particle size distribution and the physical properties of the fine aggregate.
are established (Bary, 2008; Tixier and Mobasher, 2003). PSA on con- Sieve size (mm) Cumulative pass amount (%)
crete was often overlooked and received little attention until recent years
5 100
(Haynes et al., 1996). Many long-term field tests showed that the con- 4.75 100
crete foundation parts exposed to air were severely deteriorated, and 2.36 100
various efflorescence salts remained on the evaporating surface (Stark, 1.18 75.7
1989; Li et al., 1995). In laboratory tests, two exposure regimes are often 0.6 46.2
0.3 36.1
applied, one is partial immersion in sulfate solution with cyclic temper- 0.15 9.0
ature and humidity condition (Haynes et al., 2008; Bassuoni and Rah- Fineness modulus 2.33
man, 2016; Nehdi et al., 2014; Aye and Oguchi, 2011), the other is Physical properties
full-immersion in sulfate solution under dry-wetting cycles (Sahmaran Specific gravity (kg/m3) 2590
Apparent density (kg/m3) 2608
et al., 2007; Yuan et al., 2016).
Bulk density (kg/m3) 1438
Currently, the performance of concrete subjected different sulfate Water absorption (%) 0.64
environments is receiving increasingly attention. Also, the concrete re-
sistances have been modified using fly ash, silica fume, blast furnace
slag, nanoparticles for those exposure conditions (Bassuoni and Rah- 2. Experimental procedure
man, 2016; Nehdi et al., 2014; Nehdi and Hayek, 2005). Jiang et al.
(2014); Jiang and Niu, (2016) explored the effect of solution concen- 2.1. Materials and samples preparation
trations and sulfate types on the deterioration process of concrete sub-
jected to dry-wetting cycles in different sulfate solutions. Haynes et al. The Ordinary Portland cement (P.O. 42.5) supplied by Anhui Conch
(2008) submerged concrete samples partially in 5% Na2SO4 solutions Cement Company, China was used in this study. The chemical compo-
with temperature and humidity cycles simulating 5 ambient exposures. sitions and the mineral compositions are presented in Table 1. The ISO
Aye and Oguchi (2011) investigated the resistance of cement mortar standard sand complying with GB 178-1977 (SAC, 2006) was used as fine
blended with pozzolans in 10% Na2SO4/MgSO4 solutions under four aggregate. Its particle size distribution and physical properties are given
exposure conditions, including full-immersion and partial-immersion at in Table 2. The mortar mixture was prepared with the water/cement
constant temperature/dry-wetting cycles. However, the mentioned ratio of 1:2 and cement/sand ratio of 1:3. The mixture proportion is
exposure conditions are always the coupled actions between sulfate shown in Table 3.
immersion and dry-wetting cycles, the comparisons between individual Cylinders (Φ 50 mm  100 mm) and prismatic bars
effect of immersion and dry-wetting cycling on the behavior of concrete (25 mm  25 mm  285 mm) were prepared. The mortar was prepared in
is limited in literature. In dry-wetting process, drying removes water in cement mixer as follows:
concrete and leads shrinkage and microcracks (Hansen, 2010). Though
re-wetting regains water, some microstructural changes and microcracks (1) The cement and fine aggregates were premixed for 0.5 min.
in the previous drying are irreversible (Wu et al., 2017). Such effect (2) The entire amount of water was added and mixed for 1 min.
should not be ignored in the coupled action. In addition, establishing (3) After a pause for 1 min, the mortar mixture was mixed against for
the real sulfate deterioration level in-site is a hot topic for scientists. 2 min.
However, few researches above concern the damage evaluation of (4) The fresh mortar was filled into the mold by three times.
concrete structures on spot. It is essential to extend those experimental (5) The fresh samples were compacted by a mechanical vibrating
and modelling results to the practical assessment for deteriorated table for 2–3 min at a frequency of 50 Hz and the vibration
concrete. amplitude of 0.5 mm.
The permeability of concrete controls the transport of aggressive (6) After compacting, the samples were cured at 23  C for 24 h in
ions/water, primarily affecting the structure durability in severe envi- molds covered with polyethylene film to prevent moisture loss.
ronments. Permeability has been suggested to be taken into account for (7) After 1 day, the samples were demolded and transferred to a
mix design and quality control of concrete (Aldea et al., 1999). Never- standard curing box (20  2  C and 95  3% RH) for 28 days ac-
theless, the evolutions of water permeability of concrete subjected to cording to ASTM C192 (ASTM, 2007).
sulfate exposures are rarely reported (Hadjsadok et al., 2012). In our
study, besides the mechanical properties (compressive strength, static
2.2. Exposure conditions
elastic modulus), expansion behavior and SEM images, the permeability
(water permeability coefficient) of the deteriorated mortar (susceptible
After the standard curing of 28 days, the mortar samples were divided
matrix of concrete) was tested and analyzed. The samples were prepared
into 4 groups and subjected to 4 different exposure (list in Table 4)
under 4 different exposure conditions, i.e., full-immersion and
conditions up to 270 days. The solutions were refreshed every 2 months.
dry-wetting cycles in 0% and 5% Na2SO4 solutions up to 270 days. The
There were various dry-wetting cycles in the literature (Nehdi et al.,
deterioration mechanisms and property evolutions were then discussed
2014; Sahmaran et al., 2007; Wu et al., 2017). The dry-wetting cycles of
and described by binomial functions. Finally, we proposed a possible
this study were referred to GB/T 50082-2009 (SAC, 2010), A
solution to establish real sulfate deterioration level in-situ.
CABR-LSB/Ⅱdry-wetting machine designed in accordance with this code

Table 1
Chemical compositions and mineral compositions of the OPC.
Chemical compositions (%) Mineral compositions (%)

SiO2 Al2O3 CaO Fe2O3 MgO SO3 K2O Na2O TiO2 C3S C2S C3A C4AF

21.35 4.94 60.16 2.71 0.46 1.96 0.48 1.00 0.15 60.74 16.18 6.66 14.17

2
X.-t. Yu et al.
Table 3
Mix proportion of the mortar.
Cement (kg/m3) Fine aggregate (kg/m3) Water (kg/m3)
450 1350 225
Table 4
Four different exposure conditions.
Exposure 1 Full-immersion; 0% Na2SO4 solution; 20  2  C
Exposure 2 Dry-wetting cycles; 0% Na2SO4 solution; 20  2  C/80  2  C
Exposure 3 Full-immersion; 5% Na2SO4 solution; 20  2  C
Exposure 4 Dry-wetting cycles; 5% Na2SO4 solution; 20  2  C/80  2  C
was used for dry-wetting cycling. It was equipped with two chambers,
one front chamber loaded with mortar specimens and the other back
chamber stored solution (in drying stage). One dry-wetting cycle (24 h)
consisting of inflowing, immersion, draining, drying and cooling was set
as presented in Fig. 1.
2.3. Mechanical test
At the exposure durations of 0, 30, 60, 90, 120, 150, 180, 210, 240,
270 days, three replicate Φ 50 mm  100 mm cylinder samples under
each conditions were prepared for the uniaxial compression test. On the
test day, specimens full-immersed (or under wet state of dry-wetting
cycles) were taken out from solution, and the surface of each spec-
imen was thoroughly washed and wiped. Then the samples were dried
at 60  C until a constant mass, later cooled to room temperature before
tests.
The uniaxial compression tests were performed in auto-compensated
and auto-equilibrated triaxial cell system (Fig. 2) of TOP INDUSTRY,
France. In this cell system, there are three high - pressure generators
equipped for deviatoric stress (σ1-σ3) load, confining stress (σ3, σ2 ¼ σ3)
load and drainage stress load (Pin, Pout), respectively. In the uniaxial
compression test (σ2 ¼ σ3 ¼ 0), the deviatoric stress generator with a
maximum capacity of 375 MPa is used for axial stress load (σ1), and the
axial displacement of the mortar was measured with two linear variable
displacement transducers (LVDT) placed at two opposite sides of the
sample.
The compression test was conducted in the axial displacement control
mode with the rate of 0.001 mm/s. The values of stress (σ1) and strain
(ε1) in the total process were collected every 5 s until the sample reached
fractured.
3
Ocean Engineering 157 (2018) 1–12
2.4. Water permeability test
The auto-compensated and auto-equilibrated triaxial cell system was
also used for the water permeability test of the mortar. Three cylindrical
samples were taken out and prepared as above in mechanical test after
the exposure for 0, 30, 60, 90, 120, 150, 180, 210, 240, 270 days. In this
test, the confining pressure was set to 5 MPa to prevent side leakage
between the samples and membrane, and then an initial water pressure of
2 MPa was applied to the surface of the sample for 24 h (Pin ¼ 2 MPa,
Pout ¼ 0 MPa). The total water volume penetrating the specimen was
recorded. The water permeability coefficient of the sample was deter-
mined according to Darcy's law.
μLV ρg LV ρg
K¼ ⋅ ¼ (1)
AΔPt μ AΔPt
where K is the water permeability coefficient (m/s); μ is the coefficient of
dynamic viscosity of water (N⋅s/m2); L is the thickness of the sample (m);
V is the total water volumes penetrating into the sample during the
certain time (m3); t is the time under water pressure (s); A is the measured
cross-sectional area (m2); ΔP (Pin - Pout) is the difference of seepage
pressure between the two sides of the sample (Pa); ρ is the density of
water (kg/m3); g is the acceleration of gravity (m/s2).
2.5. Expansion test
Twelve 25 mm  25 mm  285 mm prism bar subjected to 4 different
exposure conditions were used for the expansion test as described by
ASTM C 1012 (ASTM, 2004). After the standard curing of 28 days, the
initial length of the mortar bar was tested. Then, the sample was
full-immersed or placed in the dry-wetting circulation machine for
further length measurement. The length variation was monitored
through a high-accuracy digital length comparator. The expansion of
mortar was obtained according to equation (2):
Lx
Li
ΔL ¼  100% (2)
Lg
where ΔL is the length change of the mortar bar at x days (%); Lx is the
length of the mortar bar at x days exposed to different conditions (mm); Li
is the initial length of the mortar bar curing for 28 days (mm); Lg is the
nominal gage length, or 250 mm as applicable.
The visual inspection of the mortar sample was conducted by taking
the sample from the full-immersion solution or dry-wetting circulation
Fig. 1. One dry-wetting cycle.
X.-t. Yu et al. Ocean Engineering 157 (2018) 1–12

Fig. 2. Auto-compensated and auto-equilibrated triaxial cell system.

machine at the given exposure durations of 90 and 270 days. All the
changes on the surface of the mortar specimen, i.e. surface color,
expansion, spalling and delamination, were thoroughly checked and
recorded.

2.6. SEM analysis

The microstructure of the deteriorated mortar was evaluated by

scanning electron microscopy (SEM). An FE SEM S-4800 with an accel-
erating voltage of 0.5–30 kV and 25-800,000 magnification was used in
this research. The deteriorated cylinder mortar at 270 days was cut into
several portions using diamond wire cutting machine (Switzerland Well
6234). 4 Portions of sample were obtained and dried at 60  C. The
selected sample portion was then fixed on copper stubs using carbon
paste, polished and coated (by evaporation) with gold.

3. Results and discussion

3.1. Compressive strength
The evolution of compressive strength of mortar subjected to four
exposure conditions are illustrated in Fig. 3, and it exhibits two distinct
stages, an initial increase stage and a decrease stage followed.
Exposure 1: In 0% Na2SO4 solution (tap water), the compressive
strength increased before early 120 days, and then the growth rate
slowed down even reduced slightly at later stage. The further hydration
of cement particles refined the microstructure and enhanced the
compressive strength at early stage. Later the increase rate reduced and
the compressive strength kept around 39 MPa. Furthermore, the strength
slightly reduced after 240 days. Two reasons may be given to this phe-
nomenon. One is that the value at 270 days fell in the margin of errors of
the test. Considering the standard deviations are 2.03 at 240 days and
1.92 at 270 days, it is possible that the value may remain stable at 270
days. The other is leaching, in which the concentration differences be-
tween pore solution and water caused reduction of calcium ions, then led
to a dissolution of the portlandite (CH) and calcium-silica-hydrates (C-S-
H) (Jain and Neithalath, 2009), eventually triggering the loss of strength
(Torrenti et al., 2013).
Exposure 2: The cyclic effect caused an accelerating and greater
Fig. 3. Compressive strength of mortar subjected to different expo-
sure conditions.
variation in compressive strength of mortar. Water evaporated from
mortar in drying, capillary pressure and moisture gradient generated.
The capillary pressure acted like a prestressing to strenghten the mortar
(Bartlett and MacGregor, 1994; Okajima et al., 1980). Moisture gradient
induced the differential contraction between external and internal parts
of the sample, not only confined the internal part, but led to micro-
cracking (Wu et al., 2017; Baǎnt and Raftshol, 1982). Later on re-wetting,
the loss of water and microcracking partially recovered (Wu et al., 2017;
Sakata, 1983), leaving residual drying effect after one cycle. Before 180
cycles, prestressing in the residual drying and hydration took dominant
and increased the compressive strenght from 28.8 MPa to 40.9 MPa.
Afterwards, the accumulative microcracing became preponderant and
declined the strength to 38.0 MPa.
Exposure 3: As expected, sulfate immersion caused more severe
deterioration. This agreed with other research results and can be attrib-
uted to the chemical reaction between sulfate ions and hydration prod-
ucts (Lothenbach et al., 2010; Çavdar and Yetgin, 2010; Zhou et al.,

4
X.-t. Yu et al.
2016). At first, gypsum and ettringite, the primary reaction products,
filled the pores and compacted the microstructure, resulting in the
compressive strength increment. Then they exceeded the pore limitation
and expansive stress sprouted. Once the stress outstripped the tensile
strength of mortar, microcracking developed and deteriorated the
strength. Also, leaching and consumption of hydration products further
weakened the cohesion.
It is noteworthy that the compressive strength of sample full-
immersed in 5% Na2SO4 solution had more significant changes than
that under drying-wetting cycle without sulfates (Exposure 2). The
mortar gained 40.6% strength increase within 150 days, shorter than that
in Exposure 2. Also, it lost 25.3% strength at the end of immersion, more
than the reduction in dry-wetting cycles. Lee et al. (2008) reported
around 18.2% and 26.8% decrease in compressive strength of mortar
immersed in 5% Na2SO4 solution for 240 days and 360 days. Brown
(1981) found a nearly 40.0% decrease in strength for mortar after 84
days’ immersion in a similar exposure.
Exposure 4: Under sulfate immersion and dry-wetting cycles, the dual
sulfate attack (physical and chemical sulfate attack) deteriorated the
samples more seriously compared with chemical sulfate attack of Expo-
sure 3. Under the residual drying effect and positive effect of expansive
products, the compressive strength grew sharply to 40.6 MPa from 0 day
Table 5
Fitting parameter and R-square of the empirical formulas for compressive
strength.
Exposure A b R2

6
3
1
8.75  10 3.70  10 0.990

1.22  10
5 4.45  10
3
2 0.999
3
1.86  10
5 5.09  10
3 0.940
4
2.33  10
5 6.08  10
3 0.928
5
Ocean Engineering 157 (2018) 1–12
to 90 days. Subsequently, it dropped to 28.3 MPa at 270 days, much less
than that under other three conditions. It may be due to the fact that, the
drying process created super-saturation of sodium sulfate, which exerted
high tensile stress against the pore walls and generate microcracks. In
wetting, those microcracks along with capillary absorption gained more
sulfates into mortar, which further accelerated the chemical reaction and
physical crystallization (Sahmaran et al., 2007; Yuan et al., 2016).
Moreover, the scheduled cycling in this study was suitable for the con-
version of anhydrous sodium sulfate (thenardite, Na2SO4) to hydrous one
(mirabilite, Na2SO4⋅10H2O), involved with a 315% volumetric change
(Haynes et al., 2008; Bassuoni and Rahman, 2016). Such cyclic changes,
accompanied by crystallization and chemical reaction, aggravated the
compressive strength.
Based on the results above, a binomial function was proposed to
describe the relationships between compressive strength and exposure
time for 4 different conditions, as Eq. (3)
ft
¼ at2 þ bt þ 1 (3)
f0
Where ft and f0 are uniaxial compressive strength of mortar subjected to
each exposure condition at t days and 0 day, respectively. And the fitting
parameters corresponding to the exposure condition were listed in
Table 5. Fig. 4 (a) - (b) show that these proposed empirical formulas had
good fitting performance.
Considering that 2at þ bcontrolled the evolution rate of compressive
strength with exposure time, the parameter a of 2at þ b played the pri-
mary role in the strength deterioration process. Neglecting the influence
of parameterb, Δa2 (a2
a1 ), Δa3 (a3
a1 ), Δa4 (a4
a1 ) can be consid-
ered as the main effect of dry-wetting cycle, chemical sulfate attack and
dual attack and they are 0.454  10
5, 0.985  10
5 and 1.455  10
5
respectively. The sum of Δa2 and Δa3 is less than Δa4 , indicating that the
dual attack (physical and chemical sulfate attack) on mortar was more
Fig. 4. Fitting curves of compressive
strength for each exposure condition.
X.-t. Yu et al. Ocean Engineering 157 (2018) 1–12

Table 6
Compressive strength variations fitting from reference. (Brown, 1981; Irassar et al., 2000; Zhou et al., 2016).
Sample Exposure regime Compressive strength Fitting (at2þbtþ1) R2
Variation

Days variation a b

Mortar; Cured until 24.1  0.7 MPa; Immersion; 0 1
1.71  10
4 1.08  10
2 0.948
ASTM Type I; 5% Na2SO4 solution. 28 1.229
25 mm cube. 42 1.158
(Brown, 1981) 56 1.027
70 0.878
84 0.738
Mortar; Cured until 303 MPa; Immersion; 0 1 Low C3S: 3.78  10
3 0.985
2 ASTM Type V: 5% Na2SO4 solution. 28 1.069
2.14  10
5
(low C3S, 56 1.146
high C3S); 90 1.176
25.4 mm cube. 180 0.986
(Irassar et al., 2000) 0 1 High C3S: 6.15  10
3 0.901
28 1.039
1.36  10
5
56 1.284
90 1.578
180 1.611
360 1.463
Concrete; Cured for 28 days; 0 1
7.21  10
5 4.64  10
3 0.980
150 mm  300 mm cylinder. Dry-wetting cycling 30 1.129
(Zhou et al., 2016) (2 cycles/day); 60 0.994
7% Na2SO4 solution. 90 0.811
120 0.534

severe than the superposition of dry-wetting cyclic attack and chemical
sulfate attack.
Note that the specific parameters are unique to the mixture propor-
tion and exposure regime of this study. To prove the applicability of the
function form proposed, some compressive strength data of other re-
searches are listed in Table 6, with some detailed information. Also the
good fitting results are presented in Fig. 5. It is conceivable that our
empirical equations can be extended to practical application.
3.2. Elastic modulus
From Fig. 6, it can be observed that elastic modulus under different
exposure tests increases in early exposure period and decreases in later
period, which are similar with the trend of compressive strength.
Due to continuous hydration, the elastic modulus of mortar subjected
to Exposure 1 increased from 14.6 GPa to 17.2 GPa. And then the
increment slowed down and just gained 4.1% increase between 120 days
and 240 days. It even dropped to 17.4 GPa at 270 days, which can be
attributed to either the standard deviation or the effect of leaching and
decalcification of hydration products. Zhu et al. (2013) studied the dy-
namic elastic modulus of mortar immersed in 0% Na2SO4 and found the
value dropped after 330 days. Under the residual drying effect of
dry-wetting cycles in Exposure 2, the mortar took shorter time (180 days)
to the peak value of elastic modulus. Later the damage of accumulative
microcracking overrode the positive residual drying effect, and the
elastic modulus declined from 18.0 GPa to 16.1 GPa.
When being exposed to sulfate solution (Exposure 3, Exposure 4), the
elastic modulus of samples improved with the progress of immersion or
dry-wetting cycles until certain days (150 days, 90 days), subsequently,
the severe sulfate deterioration decreased the elastic modulus by 16.2%
and 20.9%. In addition, the action of dry-wetting cycles of Exposure 4
triggered physical sulfate attack, in the form of crystalline pressure and
volumetric expansion. The joint effect of physical and chemical damage
shortened the time to the peak value of elastic modulus and sped up the
decrease rate in later stage.
Moreover, as represented in Fig. 7, there was a significant correlation
between the elastic modulus and the square root of compressive strength

Fig. 5. Fitting curves of compressive strength variation of other researches

(Brown, 1981; Irassar et al., 2000; Zhou et al., 2016). (For interpretation of the
references to color in this figure legend, the reader is referred to the Web version
of this article.) Fig. 6. Elastic modulus of mortar subjected to different exposure conditions.

6
X.-t. Yu et al.
Fig. 7. Relationship between elastic modulus and square root of compressive
strength of deteriorated mortar.
of mortar under each exposure condition. The damage process had trivial
effect on the correlation between the compressive strength and elastic
modulus. This is also proved by Zhou et al. (2016).
Various ranges of coarse and fine aggregates induce diverse
compressive strengths and elastic moduli of cement-based materials.
However, since those aggregates are made of silica particles and the in-
gredients of normal concrete and mortar are similar. The intrinsic rela-
tion between compressive strength and elastic modulus of concrete may
be applied to that of mortar. Analogy to the equation suggested for
normal weight concrete (ACI, 2005), the elastic modulus of mortar can be
determined by the following Eq. (4), with R-square of 0.960. The result of
concrete under sulfate attack with dry-wetting cycles (Zhou et al., 2016)
can also be described by a similar equation, Eq. (5), with R2 ¼ 0.940,
shown in Fig. 8.
pffiffiffi
E ¼ 3:4 f
4:12 (4)
pffiffiffi
E ¼ 5:8 f
16:67 (5)
where E and f are elastic modulus and uniaxial compressive strength,
respectively.
Even they are of different proportion and exposed to different expo-
sure conditions, the sulfate attack does not change the intrinsic
Fig. 8. Relationship between elastic modulus and square root of compressive
strength of concrete under sulfate attack (Zhou et al., 2016).
7
Ocean Engineering 157 (2018) 1–12
Fig. 9. Water permeability coefficient of mortar subjected to different expo-
sure conditions.
relationship between compressive strength and elastic modulus. Thus it
may be reasonable to evaluate the mechanical properties of mortar in
different environment via either of the two indicators in field test.
3.3. Water permeability coefficient
Since the deterioration starts with the harmful ions transportation by
water, the transport property, e.g. permeability, is critical in the dura-
bility of mortar under aggressive conditions. Fig. 9 depicts the logarithm
of water permeability coefficient of mortar versus exposure days under
different exposure conditions.
In Exposure 1, the logarithm of water permeability coefficient
decreased with immersion time. The continuous water curing gradually
refined the microstructure and decreased the internal flow channel.
Later, a lower decrease rate of permeability was observed due to the less
hydration after 240 days.
For specimens in Exposure 2, the dry-wetting action led to a two-
staged evolution of water permeability. The permeability coefficient
decreased from 2.03  10
12 m/s to 0.31  10
12 m/s during the first
period of 150 days. Then, it increased rapidly by 341.9% from 150 days
to 270 days, indicating that microcracking held dominance in later stage
of dry-wetting cycle. Cracks appeared in the microstructure (presented in
SEM Analysis) also proved it.
The full-immersion in 5% Na2SO4 solution (Exposure 3) decreased the
permeability coefficient from 2.07  10
12 m/s to 0.38  10
12 m/s after
150 days’ exposure. The hydration products and expansive reaction
products gradually filled the pores of mortar and reduced their connec-
tion. Later, microcracking induced by expansion interconnected flow
paths and increased permeability to 1.52  10
12 m/s at 270 days.
Comparatively, the turning point of water permeability coefficient in
Exposure 3 happened at the same time with that in Exposure 2, but that
point of compressive strength and elastic modulus in Exposure 3
happened earlier than that in Exposure 2. It, on one hand, confirmed the
water permeability was more sensitive to deterioration, on the other,
may indicate that sulfate reaction products blocked pore and micro-
cracks, decreasing the connectivity.
To the mortar subjected to Exposure 4, permeability experienced
more dramatic variation, the magnitude of which can vary largely from
10
13 to 10
11. Within 90 cycles, the filling effect of sulfate crystalline
and chemical reaction products decreased the water permeability of
mortar and the minimum value of 0.32  10
12 m/s reached. Since the
dry-wetting regime includes drying at 80  C, and wetting at 20  C, in
which the sodium sulfate is in anhydrous state (thenardite) for the dry-
ing, and transformed to hydrous state (mirabilite) in the later wetting
(Flatt, 2002). With the cycles, excessive crystalline pressure, expansion
X.-t. Yu et al. Ocean Engineering 157 (2018) 1–12

Table 7 where Kt and K0 are the water permeability coefficient of mortar under
Fitting parameter and R-square of the empirical formulas for water permeability exposure for t days and 0 day. The fitting parameters of four exposure
coefficient. conditions are presented in Table 7. And the fitting effects are depicted in
Exposure p q R2 Fig. 10. There are good agreement between those experiment data and
1
3.11  10
7
4
3.67  10 0.992
present fitting results. Those functions not only revealed the effect of
2
2.61  10
6 7.55  10
4 0.863 each exposure condition on the water permeability coefficient of mortar,
3
2.46  10
6 6.84  10
4 0.864 but conformed to the conclusion obtained in compressive strength.
4
5.57  10
6 1.08  10
3 0.939 Similar with the compressive strength, when neglecting the influence
of parameter q, Δp2 (p2
p1 ), Δp3 (p3
p1 ), Δp4 (p4
p1 ) are 2.199  10
6,
2.149  10
6 and 5.259  10
6, mainly representing the effect of Expo-
pressure and repetitive conversion from thenardite to mirabilite caused
sure 2, 3 and 4. It is noteworthy that Δp4 is larger than the sum of Δp2 and
massive microcrack initiation (Haynes et al., 2008; Bassuoni and Rah-
Δp3 , which confirmed the above conclusion that dry-wetting cycle with
man, 2016). Those cracks allowed more sulfates to infiltrate into the
sulfate caused severe physical sulfate attack on mortar, besides the
mortar, facilitating deterioration and cracks. Such chain reaction
chemical sulfate attack and the accumulative dry-wetting effect.
contributed to the enormous growth in water permeability from
Differently, the parameter values of water permeability coefficient in
0.32  10
12 m/s to 27.0  10
12 m/s.
Exposure 2 were nearly the same with that in Exposure 3, indicating that
From above results, the evolution of water permeability coefficient is
samples under that two conditions experienced similar damage. It further
more sensitive than the mechanical properties of deteriorated samples.
supported that the water permeability coefficient was sensitive to
And can be a good indicator to evaluate the mortar durability under
microcracking, which was the dominant deterioration factor in Exposure
deterioration process. The compressive strength is determined by the
2. In the water permeability coefficient, the deterioration degree of 4
cohesion between the cement paste and aggregates, also relevant with
exposure conditions was in the order of Exposure 4 > Exposure
the microcracks and porosity. While the permeability is highly dependent
3  Exposure 2 > Exposure 1.
on the pore continuity and cracks of the microstructures (Bamforth,
1991).
Moreover, our paper proposed similar empirical formulas (Eq. (6)) 3.4. Expansion
with that for compressive strength to simulate the variation of logarithm
of water permeability coefficient in the 4 exposure tests. Note that the The length changes of mortar bars exposed to 4 conditions are illus-
following equations and parameters were derived from our study. trated in Fig. 11. The dashed line represents the expansion limit of 0.10%
lgKt at 26 weeks (182 days), which is classified as moderate sulfate resistance
¼ pt 2 þ qt þ 1 (6) as required by ASTM C1012 (ASTM, 2004). The corresponding surface
lgK0
deterioration of mortar samples for 90 and 270 days are also shown,
respectively, in Table 8.

Fig. 10. Fitting curves of water perme-

ability coefficient for each exposure
condition.

8
X.-t. Yu et al.
Fig. 11. Expansion of mortar subjected to different exposure conditions.
In Exposure 1, mortar bars were emerged into water without water
loss, which prevented from drying shrinkage. The expansion had a slight
increase during the early period and then kept stable. The surface of
mortar had no remarkable change and remained smooth during the
whole immersion.
By comparison, for mortar under dry-wetting cycle with water
(Exposure 2), shrinkage took place at first, and then the length turned
stable. Drying in one cycle vaporized the surface water, exerting prestress
to contract the volume of sample. Meanwhile, shrinkage was alleviated to
some extents due to the subsequent re-immersion. The residual drying
effect made the length slightly decrease to 0.01% after 78 cycles. The
surface turned whiter after 90 cycles. With the cycles going on, the
contraction was mitigated in the effect of microcracking, and the length
remained unchanged. The initial surface defects also enlarged in the
sample surface, as shown in Table 8.
The mortar bar in Exposure 3 showed the two-staged evolution of
expansion, which agreed with other studies (Roziere et al., 2009; San-
thanam et al., 2002). During the first 22 days, the expansion of the mortar
Table 8
Pictures of mortar under 4 exposure conditions at 90 and 270 days.
Exposure 90 days
9
Ocean Engineering 157 (2018) 1–12
bar was insignificant, which was often termed as “induction stage”. After
that, the mortar started to expand and an accelerating expansion appears
after 212 days’ immersion. The dark erosion pits, spalling and white
powdery material started to appear on the surface after 90 days. Wider
and thicker white powdery material precipitated, larger and more
erosion pits with spalling and delamination can be inspected at the end of
immersion. It is conceivable that sulfate ions diffused through the surface
pores from outside to inside, and the contents near the surface was
relatively higher. Thus the surface pores were inclined to be erosion pits
and spalling. Though being washed, the white salt deposited can be
observed owing to the high concentrations of sulfates in the pores near
surface.
Mortar subjected to Exposure 4 suffered most severe expansion. At
early stage, the filling process occurred and led trivial expansion. Then
dry-wetting cycle not only induced physical sulfate attack, but promoted
chemical sulfate attack. 1.44% expansion is thus generated in later stage.
The erosion pits and white powdery material emerged excessively on the
surface, followed by progressive spalling and delamination. This phe-
nomenon was also observed in previous study (Haynes et al., 2008;
Bassuoni and Rahman, 2016; Nehdi et al., 2014). The temper-
ature/relative humidity cycling can induce abundant white precipitated
substances, the combination of hydrated and unhydrated forms of
Na2SO4 efflorescence (Haynes et al., 2008; Bassuoni and Rahman, 2016).
The coupled effect of physical and chemical process further amplified
other deterioration signs, compared with single chemical attack.
3.5. SEM analysis
Fig. 12(a)–(d) show SEM micromorphology of mortar subjected to 4
exposure conditions for 270 days, respectively. In Fig. 12(a), clusters of
spinulose C-S-H crystals and thin hexagonal flake calcium hydroxide
crystals can be observed. Also they were in tight connection, representing
compact internal microstructure of mortar. This can well explain the
improved mechanical properties and reduced water permeability of
mortar full-immersed in water.
In the effect of 270 dry-wetting cycles, cracks appeared with some
270 days
X.-t. Yu et al.
white flocculent substances near the crack edge in Fig. 12(b). The re-
petitive dry-wetting action led to microcrack development, which is the
channel for leaching. It is conceivable that the leaching of calcium hy-
droxide (Ca(OH)2) caused the decalcification of C-S-H gels, in the form of
white substances near the cracks.
For the mortar immersed in 5% Na2SO4 solution, there was a large
amount of needle-like ettringite randomly oriented on the surface of the
selected mortar portion, with loose flocculent C-S-H crystals (Nehdi and
Hayek, 2005; Ye et al., 2014). Growth of the columnar or tabulate gyp-
sum and consumption of Ca(OH)2 were also found (Ye et al., 2014).
These expansive reaction products were responsible for the sudden
expansion of mortar in the later stage. Cracks caused by the expansive
products nearby can also be detected. The loose microstructure was the
main reason for the deterioration of mechanical property and increase of
permeability.
When exposed to Exposure 4, the mortar portion surface was covered
with mass of rod-like sodium sulfate crystals, which were mainly the
combinations of thenardite and mirabilite in the action of dry-wetting
cycle with sodium sulfate (Nehdi and Hayek, 2005). Beside the phys-
ical sulfate crystalline, the white acicular ettringite and columnar and
sheet gypsum can be observed. Both physical and chemical deterioration
triggered the cracks, but the large amount of physical sulfate crystals was
believed to be the leading deterioration factor in the Exposure 4 (Haynes
et al., 2008; Bassuoni and Rahman, 2016; Nehdi et al., 2014).
3.6. Application in in-situ tests
Evaluating the behavior of concrete in situ has become a hot topic for
engineering structure assessment and maintenance. Since mechanical
property and durability are two main aspects in concrete structures, it is
worthwhile to determine the relationship between permeability and
mechanical property of deteriorated concrete for a more comprehensive
10
Ocean Engineering 157 (2018) 1–12
Fig. 12. SEM micromorphology of
mortar under different exposure
conditions.
damage assessment. Though the proportion of concrete in situ is different
from that of mortar in laboratory, both of them are cement-based ma-
terials. Santhanam et al. (2002) concluded that the pattern of deterio-
ration and the type of reaction products formed were similar in most of
the cementitious mortar systems. As the main matrix of concrete, mortar
is susceptible to sulfate attack. The applicability of the empirical formulas
for compressive strength has been verified by other types of mortar and
concrete above, it is reasonable to investigate the water permeability
coefficient - compressive strength relationships of mortar under different
exposure conditions.
As shown in Fig. 13, the exposure can be categorized into two types,
one is immersion with only one phase; the other is dry-wetting cycling
with two phases. In immersion type, the relationship between water
permeability and compressive strength of mortar in early stage approx-
imately equals to that in later stage, defined as Phase 1. One of the two
indexes can be used in the practical assessment.
In dry-wetting cycling type, such relationship in first stage differed
from that in second stage of exposure. It can be explained by that the
compressive strength was not as sensitive as permeability to micro-
cracking. The dry-wetting cycling contributed to the burst of micro-
cracking, especially in the later stage. Thus the variation of permeability
was more dramatic than the compressive strength, not consistent with
that in first stage. Both of the two properties should be detected for
assessment in-situ in Phase 2 (deterioration stage).
There are three main influential factors to the variations of the re-
lations between water permeability and compressive strength, immer-
sion (0% and 5% Na2SO4), and dry-wetting cycles. The relationship
between them with good fitting is depicted in Fig. 14. Phase 1 of Expo-
sure 1 and 2 can be described a linear relation, in which the effect of dry-
wetting cycling of Exposure 2 is ignored. Since the sulfate reaction
products initially compact the pores and improve the performance of
mortar, the variation is expected to be more significant in Phase 1. As for
X.-t. Yu et al.
Fig. 13. Phase 1 and Phase 2 corresponding to different exposures.
Fig. 14. Relationship between water permeability coefficient and compressive strength of deteriorated mortar.
Phase 2, the dominate factor is dry-wetting cycling. The effect of sulfate
immersion is minor.
Hence, it can be concluded that the logarithm of water permeability
coefficient were related to the compressive strength of mortar subjected
to different exposure, but in some case, the exposure types and phases
needed to be considered.
Based on the above experimental analysis, a possible approach to
evaluate the sulfate deterioration level in-situ for concrete structures was
proposed as follows.
(1) Obtain the referential evolution of compressive strength and
water permeability coefficient. Firstly choose the specific types
from the Exposure 1-4 through the sulfate concentration with/
without dry-wetting cycle on spot. Then prepare laboratory sam-
ples referred to mixture proportions of the concrete structure. Put
them into the specific exposures, and conduct mechanical tests
and water permeability tests. The two indexes are presented as
that in Figs. 3 and 9.
(2) Determine the evaluated property. Investigate the relationship
between compressive strength and water permeability coefficient,
as shown in Figs. 13 and 14. If there is only one phase in the
exposure, i.e., immersion type, any one of the both indexes can be
assessed in concrete structures. Otherwise both of strength and
permeability of concrete are recommend detecting in practical
assessment.
(3) Judge the deterioration stage of concrete structure. Firstly detect
the compressive strength or/and water permeability coefficient of
concrete in-situ regularly (every half a year). The tests can be
conducted on the drill core samples from the structure or through
11
Ocean Engineering 157 (2018) 1–12
non-destructive methods on selected parts, like rebound methods
(Huang et al., 2011; Qasrawi, 2000) and Autoclam system
(Basheer, 1994; Yang et al., 2013). And then compare the data on
spot to the laboratory referential curves in step (1) to judge if this
structures in deterioration stage.
(4) Predict the performance of the structure in the future. The
empirical functions of the data in laboratory like Eq. (3) and Eq.
(6) are proposed. According to those field data, the deterioration
process of the structure can be predicted. Eventually, on the basis
of the deterioration level judgement and prediction, proper mea-
sures for structure maintenance can be drawn up.
4. Conclusion
Considering the effect of sulfate attack and dry-wetting cycles on the
offshore structures in marine environment, the behavior and associated
deterioration mechanism of mortar subjected to four conditions, full-
immersion and dry-wetting cycles in 0% and 5% Na2SO4 solutions,
were investigated. The compressive strength, elastic modulus, water
permeability coefficient of mortar and the relations between them were
analyzed. Also, expansion, visual inspection and SEM analysis were
conducted. The following conclusions were obtained from experimental
results:
1. Mortar samples in each exposure condition generally exhibited two-
staged evolution, the improvement at first stage and deterioration
at later stage. But the process evolved differently corresponding to
exposure conditions. Continuous hydration and slow leaching were
responsible for the mortar full-immersed in 0% Na2SO4 solution
X.-t. Yu et al.
(Exposure 1). Residual drying effect firstly dominated, but then
overrode by the accumulative microcracking in the process of dry-
wetting cycle with water (Exposure 2). Chemical sulfate reaction
took place in the mortar immersed in 5% Na2SO4 solution (Exposure
3), the expansive products (gypsum and ettringite) at first filled the
internal pores, later exceeded accommodation capacity and induced
microcracking. The dry-wetting cycles with sulfate (Exposure 4)
brought physical sulfate attack besides chemical one. Salt growth
pressures and repetitive conversion between anhydrous and hydrous
form of Na2SO4 promoted severe deterioration on mortar.
2. The evolution of compressive strength and water permeability coef-
ficient of damaged mortar can be well described by binomial func-
tions. The function form was also suitable for other researches. The
linear relationship between elastic modulus and square root of
compressive strength was not affected by exposure conditions.
3. The effect of Exposure 4 to the mortar was more severe than the su-
perstition of Exposure 2 and 3. More interestingly, the compressive
strength variation in Exposure 3 was higher than that subjected to
dry-wetting cycles with water (Exposure 2), while water permeability
exhibited different trend, i.e., Exposure 2 and 3 caused similar dam-
age to the mortar.
4. The relation between the water permeability coefficient and
compressive strength was highly dependent on the sulfate exposure
types and time. For the sake of practicality and accuracy, a possible
solution to the in-situ concrete assessment under different exposure
conditions was established.
Acknowledgements
This project was supported by National Natural Science Foundation of
China (51579088, 51779087), Natural Science Foundation of Jiangsu
Province (BK20150037, BK20150811, BK20161507) and the 111 Project
(B12032).
References
ACI 318-05, 2005. Building Code Requirements for Structural Concrete. American
Concrete Institute, Detroit, USA.
Aldea, C.M., Shah, S.P., Karr, A., 1999. Effect of cracking on water and chloride
permeability of concrete. J. Mater. Civ. Eng. 11 (3), 181–187.
ASTM C1012, 2004. Standard Test Method for Length Change of Hydraulic-cement
Mortars Exposed to a Sulfate Solution. ASTM International, West Conshohocken, PA.
ASTM C192/192M-02, 2007. Standard Practice for Making and Curing Concrete Test
Specimens in the Laboratory. ASTM International, West Conshohocken, PA.
Aye, T., Oguchi, C.T., 2011. Resistance of plain and blended cement mortars exposed to
severe sulfate attacks. Construct. Build. Mater. 25 (6), 2988–2996.
Bamforth, P.B., 1991. The water permeability of concrete and its relationship with
strength. Mag. Concr. Res. 43 (157), 233–241.
Bartlett, F.M., MacGregor, J.G., 1994. Effect of moisture condition on concrete core
strengths. Mater. J. 91 (3), 227–236.
Bary, B., 2008. Simplified coupled chemo - mechanical modeling of cement pastes
behavior subjected to combined leaching and external sulfate attack. Int. J. Numer.
Anal. Meth. GeoMech. 32 (14), 1791–1816.
Basheer, P., 1994. The Autoclam-a new test for permeability. Concrete 28 (4), 27–29.
Bassuoni, M.T., Rahman, M.M., 2016. Response of concrete to accelerated physical salt
attack exposure. Cement Concr. Res. 79, 395–408.
Baǎnt, Z.P., Raftshol, W.J., 1982. Effect of cracking in drying and shrinkage specimens.
Cement Concr. Res. 12 (2), 209–226.
Brown, P.W., 1981. An evaluation of the sulfate resistance of cements in a controlled
environment. Cement Concr. Res. 11 (5–6), 719–727.
Çavdar, A., Yetgin, Ş., 2010. Investigation of mechanical and mineralogical properties of
mortars subjected to sulfate. Construct. Build. Mater. 24 (11), 2231–2242.
de Medeiros-Junior, R.A., de Lima, M.G., de Brito, P.C., de Medeiros, M.H.F., 2015.
Chloride penetration into concrete in an offshore platform-analysis of exposure
conditions. Ocean Eng. 103, 78–87.
Flatt, R.J., 2002. Salt damage in porous materials: how high supersaturations are
generated. J. Cryst. Growth 242 (3), 435–454.
GB 178-1977, 2006. Standard Sand for Testing Strength of Hydraulic Cement Mortars.
Standardization Administration of the People's Republic of China, Beijing, China.
GB/T 50082-2009, 2010. Standard for Test Methods of Long-term Performance and
Durability of Ordinary Concrete. Standardization Administration of the People's
Republic of China, Beijing, China.
Ghafoori, N., Najimi, M., Diawara, H., Islam, M.S., 2015. Effects of class F fly ash on
sulfate resistance of Type V Portland cement concretes under continuous and
interrupted sulfate exposures. Construct. Build. Mater. 78, 85–91.
12
Ocean Engineering 157 (2018) 1–12
Glasser, F.P., Marchand, J., Samson, E., 2008. Durability of concrete — degradation
phenomena involving detrimental chemical reactions. Cement Concr. Res. 38 (2),
226–246.
Hadjsadok, A., Kenai, S., Courard, L., Michel, F., Khatib, J., 2012. Durability of mortar
and concretes containing slag with low hydraulic activity. Cement Concr. Compos. 34
(5), 671–677.
Hansen, W., 2010. Drying shrinkage mechanisms in Portland cement paste. J. Am. Ceram.
Soc. 70 (5), 323–328.
Haynes, H., O'Neill, R., Mehta, P.K., 1996. Concrete deterioration from phusical attack by
salts. Concr. Int. 18 (1), 63–68.
Haynes, H., O'Neill, R., Neff, M., Kumar, M.P., 2008. Salt weathering distress on concrete
exposed to sodium sulfate environment. ACI Mater. J. 105 (1), 35–43.
Huang, Q., Gardoni, P., Hurlebaus, S., 2011. Predicting concrete compressive strength
using ultrasonic pulse velocity and rebound number. ACI Mater. J. 108 (4), 403–412.
Irassar, E.F., Gonzalez, M., Rahhal, V., 2000. Sulphate resistance of type V cements with
limestone filler and natural pozzolana. Cement Concr. Compos. 22 (5), 361–368.
Jain, J., Neithalath, N., 2009. Analysis of calcium leaching behavior of plain and modified
cement pastes in pure water. Cement Concr. Compos. 31 (3), 176–185.
Jiang, L., Niu, D.T., 2016. Study of deterioration of concrete exposed to different types of
sulfate solutions under drying-wetting cycles. Construct. Build. Mater. 117, 88–98.
Jiang, L., Niu, D.T., Sun, Y.Z., Fei, Q.N., 2014. Ultrasonic testing and microscopic analysis
on concrete under sulfate attack and cyclic environment. J. Cent. South. Univ. 21
(12), 4723–4731.
Lee, S.T., Hooton, R.D., Jung, H.S., Park, D.H., Choi, C.S., 2008. Effect of limestone filler
on the deterioration of mortars and pastes exposed to sulfate solutions at ambient
temperature. Cement Concr. Res. 38 (1), 68–76.
Li, X., Wang, G., Sun, J., Ma, X., 1995. A study on 33 years corrosion of materials in soil at
three gorges area. Corrosion Sci. Protect. Technol. 7 (1), 1–9 [in Chinese].
Lothenbach, B., Bary, B., Le Bescop, P., Schmidt, T., 2010. Sulfate ingress in Portland
cement. Cement Concr. Res. 40 (8), 1211–1225.
Monteiro, P.J.M., Kurtis, K.E., 2003. Time to failure for concrete exposed to severe sulfate
attack. Cement Concr. Res. 33 (7), 987–993.
Nehdi, M., Hayek, M., 2005. Behavior of blended cement mortars exposed to sulfate
solutions cycling in relative humidity. Cement Concr. Res. 35 (4), 731–742.
Nehdi, M.L., Suleiman, A.R., Soliman, A.M., 2014. Investigation of concrete exposed to
dual sulfate attack. Cement Concr. Res. 64, 42–53.
Okajima, T., Ishikawa, T., Ichise, K., 1980. Moisture effect on the mechanical properties of
cement mortar. Trans. Jpn. Concr. Ins. 2, 125–132.
Qasrawi, H.Y., 2000. Concrete strength by combined nondestructive methods simply and
reliably predicted. Cement Concr. Res. 30 (5), 739–746.
Roziere, E., Loukili, A., El Hachem, R., Grondin, F., 2009. Durability of concrete exposed
to leaching and external sulphate attacks. Cement Concr. Res. 39 (12), 1188–1198.
Sahmaran, M., Erdem, T.K., Yaman, I.O., 2007. Sulfate resistance of plain and blended
cements exposed to wetting–drying and heating - cooling environments. Construct.
Build. Mater. 21 (8), 1771–1778.
Sakata, K., 1983. A study on moisture diffusion in drying and drying shrinkage of
concrete. Cement Concr. Res. 13 (2), 216–224.
Santhanam, M., Cohen, M.D., Olek, J., 2002. Mechanism of sulfate attack: a fresh look:
Part 1: summary of experimental results. Cement Concr. Res. 32 (6), 915–921.
Stark, D., 1989. Durability of Concrete in Sulfate-rich Soils. Research and Development
Bulletin, vol. RD O97. Portland Cement Association.
Tixier, R., Mobasher, B., 2003. Modeling of damage in cement-based materials subjected
to external sulfate attack. I: formulation. J. Mater. Civ. Eng. 15 (4), 305–313.
Torrenti, J.M., de Larrard, T., Benboudjema, F., 2013. Coupling between Leaching and
Mechanical Behaviour of Concrete. Cement-based Materials for Nuclear Waste
Storage. Springer New York, pp. 69–84.
Wee, T.H., Suryavanshi, A.K., Wong, S.F., Anisur Rahman, A.K.M., 2000. Sulfate
resistance of concrete containing mineral admixtures. Mater. J. 97 (5), 536–549.
Whittaker, M., Black, L., 2015. Current knowledge of external sulfate attack. Adv. Cement
Res. 27 (9), 532–545.
Wu, Z., Wong, H.S., Buenfeld, N.R., 2017. Transport properties of concrete after drying-
wetting regimes to elucidate the effects of moisture content, hysteresis and
microcracking. Cement Concr. Res. 98, 136–154.
Yang, K., Basheer, P.A.M., Magee, B., Bai, Y., 2013. Investigation of moisture condition
and Autoclam sensitivity on air permeability measurements for both normal concrete
and high performance concrete. Construct. Build. Mater. 48 (48), 306–314.
Ye, Q., Shen, C., Sun, S., Chen, R., Song, H., 2014. The sulfate corrosion resistance
behavior of slag cement mortar. Construct. Build. Mater. 71 (4), 202–209.
Yin, G.J., Zuo, X.B., Tang, Y.J., Ayinde, O., Wang, J.L., 2017. Numerical simulation on
time-dependent mechanical behavior of concrete under coupled axial loading and
sulfate attack. Ocean Eng. 142, 115–124.
Yoshida, N., Matsunami, Y., Nagayama, M., Sakai, E., 2010. Salt weathering in residential
concrete foundations exposed to sulfate-bearing ground. J. Adv. Concr. Technol. 8
(2), 121–134.
Yu, H.F., Da, B., Ma, H.Y., Zhu, H.W., Yu, Q., Ye, H.M., Jing, X.S., 2017. Durability of
concrete structures in tropical atoll environment. Ocean Eng. 135, 1–10.
Yuan, J., Liu, Y., Tan, Z.C., Zhang, B.K., 2016. Investigating the failure process of concrete
under the coupled actions between sulfate attack and drying–wetting cycles by using
X-ray CT. Construct. Build. Mater. 108, 129–138.
Zhou, Y., Li, M., Sui, L., Xing, F., 2016. Effect of sulfate attack on the stress–strain
relationship of FRP-confined concrete. Construct. Build. Mater. 110, 235–250.
Zhu, J., Cao, Y.H., Chen, J.Y., 2013. Study on the evolution of dynamic mechanics
properties of cement mortar under sulfate attack. Constr. Bulid. Mater. 43 (3),
286–292.