15th International Congress on the Chemistry of Cement

Prague, Czech Republic, September 16–20, 2019

The real world of physical attack on concrete

zanqun Liu 1,a, Dehua Deng1,b, Geert De Schutter2,c

1School of Civil Engineering, Central South University, Changsha, China

2Magnel Laboratory for Concrete Research, Department of Structural Engineering, Ghent University,

Ghent, Belgium

[email protected]
[email protected]
c [email protected]

ABSTRACT
According to an extensive literatures review, the the real world of physical salt attack on concrete was
summarized. The physical salt attack on concrete can be divided into two categories according to the
physical and chemical relationship between salt and the hydration products of cement:
(1) a typical physical salt attack is attributed to the damage mechanism for the evaporative concrete
where the salt is inert to the hydration products of cement, such as Portland cement concrete partially
immersed in Ca(NO3)2 solution and Calcium Sulfoaluminate cement concrete in Na2SO4 solution.
The concrete will be damaged in a quiet short time.
(2) a complicated physical and chemical salt attack combining concrete carbonation is attributed to the
damage mechanism for the evaporative concrete where the salt is reactive with the hydration products
of cement. The Portland cement concrete partially immersed in the sulfate environment belongs to the
second category and its damage mechanism is that the chemical sulfate attack will be the only
mechanism causing the evaporative concrete damage if there is no the effect of carbonation. With the
progress of exposure, the outer layer of evaporative concrete will be neutralized by carbonation and
become inert to sulfate, then the physical sulfate attack occurs in the neutralized concrete layer. In the
inner area isolated from carbonation, the chemical sulfate attack is still the damage mechanism. The
evaporative concrete damage is collectively controlled by the susceptibility to carbonation of concrete
and the factors in favour of chemical sulfate attack.
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

1. INTRODUCTION
The term “physical attack” (or salt weathering and crystallization distress) is usually used to describe
the deterioration of stones or bricks when partially exposed to a salt environment. The typical damage
observation lies in a region of failure within the evaporation zone, but with a sound part in direct
contact with the salt environment [Benavente et al. 2001]. In 2010, the American Concrete Institute
(ACI) also attributed physical salt attack to concrete damage, since a similar damage observation is
found in the evaporation zone of concrete elements partially exposed to groundwater or to soil
containing salts. In order of decreasing aggressiveness, these salts are: sodium sulfate, sodium
carbonate, and sodium chloride [ACI 201.2R 2016]. At present, pphysical sulfate attack on concrete is
a key area of research due to its aggressive nature [Mark et al. 2014].
However, following an extensive literature review [Liu et al. 2014], it seems extremely simplistic to
attribute concrete damage to just physical sulfate attack. This is due to the complicated physical and
chemical relationship between concrete and sulfates. Many different test results could not be
satisfactorily explained using the classical theory of salt weathering. Two main issues can be
illustrated as follows:

· The effect of pore structure on concrete damage. Field tests [Stark 2002, Irassar et al.
1996] and indoor tests [Ahmed et al. 2017, Semion 2017a] all showed two common results,
namely that (1) concrete with a lower water/cement (W/C) ratio was more resistant to
sulfate attack; and (2) pozzolanic material (fly ash, slag and silica fume) significantly
accelerated the degradation of evaporative concrete. As for the first result [Ahmed et al.
2017], the authors indicated that a low W/C ratio can significantly reduce the total volume
of the pores, as well as their connectivity, leading to less capillary rise and surface scaling.
Concerning the second result, the authors attributed the pore refinement caused by
pozzolanic materials to lead to larger crystallization distress [Stark 2002, Irassar et al.
1996]. However, it should be noted that using low W/C ratio and pozzolanic materials can
both reduce the total volume of pores and refine pore size
· The effect of relative humidity on concrete damage. The paper [Ferraris et al. 2006]
indicated an interesting result, namely that concrete cylinders partially exposed to a
Na2SO4 solution at 80% relative humidity (RH) show a larger damage area than at 30%
RH. However, it is common sense that lower RH could result in more aggressive
crystallization distress [Navarro et al. 1999].

There have been many research papers [Bassuoni MT et al. 2016, Nehdi ML et al.2014, Semion Z et
al. 2017a,b, Najjar MF et al. 2017, Ahmed RS et al. 2017, Semion Z et al. 2017] after 2010, though
these papers always ignored the two issues mentioned above and stubbornly insisted on the effect of
physical salt attack. In this review paper, a clear concept for “physical salt attack on concrete” will be
put forward. The paper has been divided into 4 Parts as follows:

· Part I: damage to concretes made with different types of cements partially exposed to
different salt solutions.
· Part II: field case investigations
· Part III: the effect of carbonation depth on the damage to evaporative concrete exposed to
sulfate solutions.
· Part IV: quantitative analysis of products in Portland cement paste partially exposed to
Na2SO4 solution

Figure 1. Relationship between the 4 parts

 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

The purpose of Part I is to check if typical physical attack damage can be observed in all tests. There
will be two possibilities: either the typical damage observation can be found in every test or it cannot.
A new viewpoint for physical salt attack on concrete will then be proposed based on the test results. In
Part II, some real evidences will be deduced from field case investigations to support and improve the
new viewpoint. In Parts III and IV, several additional tests will be designed and carried out to further
improve the proposed viewpoint. The relationship between the four parts is shown in Fig.1.

2. PART I: DAMAGE TO CONCRETES MADE WITH DIFFERENT TYPES OF CEMENTS

PARTIALLY EXPOSED TO DIFFERENT SALT SOLUTIONS
The different concretes, made with Portland cement (PC) and Calcium Sulfoaluminate cement
(CSAC), were partially immersed in Na2SO4, MgSO4, Na2CO3 or Ca(NO3)2 solutions. These tests
focused on two points: (1) whether the typical damage performance could be found in the concretes
made with the different cementitious materials; and (2) whether different salt crystals can be identified
in the damaged concrete as direct evidence of crystallization distress. The above two points illustrate
the classic damage mechanism due to salt weathering of porous materials [George WS 2004, Ruiz-
Agudo E et al. 2007]. If the physical salt attack can be attributed to this damage mechanism, then the
classic damage performance should be found in every test.
Unfortunately, the tests results were quite complicated.
2.1 Test 1: PC specimens partially immersed in Na2SO4 solution [Liu Z et al. 2010, 2011]
The results showed that those parts of the specimens in the evaporation zone were severely
damaged. However, Na2SO4 and MgSO4 crystals were not identified as being replaced by the
chemical sulfate attack products of ettringite and gypsum (shown in Figure 2).
[ZL-17.raw]

4000
I(CPS)

2000

0
84-1265> Ca(OH)2 - Calcium Hy droxide

72-1937> Calcite - CaCO3

41-1451> Ettringite - Ca6Al2(SO4)3(OH)12!26H2O

33-0311> Gy psum - CaSO4!2H2O

Na2SO4 33-0302> Larnite - Ca2SiO4

Solution 83-0539> Quartz - SiO2

10 20 30 40 50 60
2-Theta(°)

Figure 2. Damage of PC specimens partially immersed in Na2SO4 solution

2.2 Test 2: CSAC specimens partially immersed in MgSO4 solution [Liu Z et al. 2016 a]
The results showed that those parts of the specimens immersed in the solution were damaged due to
gypsum formation, but with an undamaged part in contact with the air; this is indicative of a typical
chemical sulfate attack (shown in Figure 3).

Air

MgSO4
Solution

Figure 3 Damage of CSAC specimens partially immersed in MgSO4 solution

2
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

2.3 Test 3: CSAC specimens partially immersed in Na2CO3 solution [Liu Z. et al. 2018]
The results showed that those parts of the specimens in the evaporation zone were severely damaged
by typical Na2SO4 crystallization distress. Na2SO4 was the chemical reaction product of the CSAC
paste and the Na2CO3 solution (shown in Figure 4).

NaCO3
Solution

Figure 4. Damage of CSAC specimens partially immersed in NaCO3 solution

2.4 Test 4: CSAC specimens partially immersed in Na2SO4 solution [Liu Z. et al. 2016 b]
The results showed that those parts of the specimens in the evaporation zone were severely
damaged, and that Na2SO4 crystals were clearly identified as the cause of specimen failure. This is
indicative of a typical sulfate salt weathering (shown in Figure 5).

Figure 5. Damage of CSAC specimens partially immersed in NaSO4 solution

2.5 Test 5: PC specimens partially immersed in Ca(NO3)2 solution [Lee BY et al. 2017]

Figure 6. Damage of PC specimens partially immersed in Ca(NO3)2 solution

The results showed that those parts of the specimens in the evaporation zone were severely
damaged, and that NaNO3 crystals were clearly identified as the cause of specimen failure. This is
also indicative of a typical sulfate salt weathering (shown in Figure 6).
2.6 Summary
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

Although the above test results are complicated, specimen damage can nonetheless be clearly
divided into two types: (1) the typical physical salt attack of test 4 and test 5; (2) the chemical salt
attack of tests 1-3. If we check the physical and chemical relationship between the cement hydration
products and the salt, it can correspondingly be found that: (1) the salts were inert to the cement
hydration products in tests 4-5 (in test 4, the hydration products of CSAC cement contained significant
amounts of AFt, C-S-H and an Al gel but did not include the Ca(OH) 2 phase [Wang Y et al. 1999]; the
stability of AFt, C-S-H and Al gel were immune to Na2SO4 [Damidot D et al. 1992, Shi C et al. 2000]. In
test 5, the hydration products of PC cement (AFt, AFm, C-S-H, the gypsum and Al gel) were also inert
to the Ca(NO3)2 solution); (2) the salts reacted with the cement hydration products in tests 1-3,
specifically Ca(OH)2 with SO42- in test 1, AFt with Mg2+ in test 2 and AFt with CO32- in test 3.
Therefore, a viewpoint can be deduced that a typical physical salt attack can only be attributed to
the damage mechanism for evaporative cement concrete where the salt is inert to the hydration
products of the cement. As for PC concrete partially immersed in sulfate solution, the damage
mechanism for the evaporative concrete damage is still the chemical sulfate attack.
However, sulfate crystallization distress has been clearly confirmed to cause concrete scaling in many
field cases. What is the reason for this discrepancy between field cases and indoor tests? The
following field case investigation will present a more detailed explanation.

3. PART II: FIELD CASE INVESTIGATIONS

3.1 Case 1: Residential concrete foundation investigation [Yoshida N. et al. 2010]
A number of concrete samples taken from concrete foundations in Japan were investigated by means
of XRD (X-ray diffractometer), XRF (X-ray fluorescence), DSC (differential scanning calorimetry) and
EPMA (electron probe microanalyzer). After combining the simulation tests results, a deterioration
mechanism for residential concrete foundations was presumed as shown in Figure 7.

Figure 7. Presumed mechanism of Na2SO4 weathering in residential concrete foundation

According to the presumed deterioration mechanism, evaporative concrete can be divided into a
carbonated area and a non-carbonated area. In the outer carbonated area, Na2SO4 crystal growth led
to concrete scaling. In the non-carbonated area, ettringite was identified with no traces of Na2SO4.
This means that the Na2SO4 crystals only occurred in the carbonated area, where the chemical
interaction between sulfate ions and the cement paste was restrained. It could perhaps be argued that
the Na2SO4 crystallization should be found in the same location before carbonation, for it to be able to
cause similar damage. In order to answer this question, the effect of carbonation on evaporative
concrete scaling needs further research. If a larger carbonation depth results in more severe concrete
scaling by Na2SO4 crystallization distress, this will mean that the Na2SO4 crystallization could not
occur in the same location before carbonation and could only occur in the carbonated area.

3.2 Case 2: Concrete lining investigation [Liu Z et al. 2017]

4
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

Several concrete cores from the concrete lining of two railway tunnels in the southwestern regions of
China were investigated by means of a Carbonation Depth (CD) test, EPMA, XRD, Ion
Chromatography (IC), and Atomic Absorption Spectrophotometry (AAS). The XRD results in Figure 8
provide some interesting evidence.

Figure 8. XRD patterns for different concrete cores

The test results indicate that the concrete lining in the two tunnels was almost neutralized to more than
55 mm, after 60 years’ exposure to the atmosphere. In Figure 8, the sample for the XRD pattern of
No. 4 was made up of concrete pieces that had become detached from the concrete core, whereas
the sample for No. 5 was made up of broken concrete pieces from the same concrete core that had
come into contact with the air. The damage in the No. 5 sample might be regarded as the first step in
the damage of the No. 4 sample. CaSO4·2H2O, AFt and Na2SO4 crystals were found in No. 5, but
Ca(OH)2 was not, suggesting an occurrence of chemical sulfate attack in the concrete:

· The Ca(OH)2 consumption means that the chemical reaction between sulfate and paste had
just finished, since (1) the formation of AFt and CaSO4·2H2O requires Ca(OH)2, and (2) the
AFt and CaSO4·2H2O would decompose to form the final product of CaCO 3 (as the XRD
pattern for No. 4) under neutralized conditions [Nishikawa T et al. 1992];
· Na2SO4 crystals were generated in the inert paste without the presence of Ca(OH)2. As a
comparison, Na2SO4 crystals could not be identified in the reactive paste containing Ca(OH)2
as shown in Figure 2.
3.3 Summary
Combining the field investigation results and the test results from Section 2, the proposed viewpoint
can be improved as follows:

· When CSAC concrete is partially immersed in Na2SO4 solution, or PC concrete is partially

immersed in Ca(NO3)2 solution, the typical physical salt attack will be attributed to concrete
damage due to the inert relationship between salts and cement pastes;
· When PC concrete is partially immersed in the sulfate solution, the outer layer of the
evaporative concrete will become inert to sulfates due to the effects of carbonation. The
physical sulfate attack then causes scaling of the outer concrete layer. In the inner area, which
is isolated from carbonation, the damage mechanism is still the chemical sulfate attack. This
represents a real world example of physical sulfate attack on (PC) concrete.
In other words, the damage of the evaporative zone of PC concrete due to sulfate attack is made up of
two parts: (1) the outer carbonated area damaged by the physical sulfate attack; and (2) the inner non-
carbonated area damaged by the chemical sulfate attack. Therefore, the viewpoint can be further
improved as follows:

· A larger carbonation depth would result in more severe concrete scaling due to sulfate
crystallization distress;
· More chemical sulfate attack products would be formed in the evaporative zone than in the
immersed zone, resulting in more severe concrete deterioration. Therefore, the evaporative
concrete will show a larger damage area under high relative humidity conditions than under
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

low relative humidity conditions. This is because the higher humidity can result in a wider
sulfate pore solution zone in the evaporative concrete [Ferraris CF et al. 2006].

4. PART III: THE EFFECT OF CARBONATION DEPTH ON DAMAGE TO EVAPORATIVE

CONCRETE EXPOSED TO SULFATE SOLUTIONS.

4.1 Effect of carbonation on physical sulfate attack on concrete by Na2SO4 [Liu Z et al. 2018b]
PC concrete cylinders with different W/C ratios and with different dosages of fly ash (FA) were rapidly
carbonated for 10 d and then partially exposed to a 10% Na2SO4 solution. After 240 d exposure, it
could be found that:

· Concrete carbonation accelerated the concrete damage, and the larger carbonation depth led
to greater concrete scaling (as shown in Figure 9);
· Na2SO4 crystals could be clearly identified (as shown in Figure 10), and the physical Na2SO4
attack resulted in concrete scaling;
· The susceptibility to carbonation of concretes incorporating fly ash resulted in greater concrete
scaling than PC concrete subjected to Na2SO4 salt crystallization distress (as shown in Figure
9).

2400
Figure 9. The relationship between concrete damage and carbonation depth

2000
S-Sodium Sulfate-Na2SO4
C-Calcium Carbonate-CaCO3
1600
Si Si-Silicon Oxide -SiO2
Counts

1200
S

800 CSS
Si S
400
SC SSi
C
C S
0
5 10 15 20 25 30 35 40 45 50 55 60 65 70
2

Figure 10. XRD pattern of the detached concrete piece from carbonated PC0.55 specimens
4.2 Effect of carbonation on physical sulfate attack on concrete by MgSO 4 [Liu z et al. 2018c]
ACI pointed out that calcium, magnesium, potassium, iron, and ammonium sulfates can only
participate in a chemical sulfate attack on concrete, but do not appear to damage concrete through
physical salt attack [ACI 201.2R 2016]. However, the occurrence of MgSO4 crystallization can be
found not only in the rapid carbonated concrete cylinders but also in normal concrete cylinders. The
damage observation is shown in Figure 11.
Based on the visual damage observation of Figure 11, it can be confirmed that the MgSO4
crystallization could cause evaporative concrete scaling due to carbonation. Compared to CC-SG50,
with its narrow carbonation depth, CC-FA30 and CC-LP2 showed more severe concrete scaling due to
their distinctly wide carbonation depth. Since FA30 and LP20 are susceptible to carbonation, the

6
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

referenced cylinders were also distressed by MgSO 4 crystallization. Figure 12 shows a micro-analysis
of the detached pieces in the red dotted line cycle of NC-FA30 as shown in Figure 11. It can be
confirmed that the outer layer of NC-FA30 had been carbonated, and that Mg2SO4 crystallization
distress caused the detachment.

C20- FA30 CC- FA30 NC- FA30

C20- LP20 CC- LP20 NC- LP20

C20- SG50 CC- SG50 NC- SG50

Figure 11. Damage observation of normal concrete (NC) and carbonated concrete (CC) partially
exposed to 10 MgSO4 solution for 150d (LP20: 20% limestone powder +80% Portland cement,
FA30: 30% fly ash +70% Portland cement; SG50: 50% slag +50 %Portland cement. the same
W/C and sand percentages)
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

Figure 12. XRD pattern, ESEM image and EDS of detached concrete pieces of NC-FA30
4.3 Summary
The effect of carbonation on sulfate crystallization distress of concrete is thus confirmed:

· The larger carbonation depth results in more severe concrete scaling, not only due to Na2SO4
but also due to MgSO4 crystallization distress;
· The susceptibility to carbonation of concretes incorporating mineral admixtures resulted in
greater concrete scaling than sulfate crystallization distress on PC concrete.

5. PART IV: QUANTITATIVE ANALYSIS OF PRODUCTS IN PORTLAND CEMENT PASTES

PARTIALLY EXPOSED TO SULFATE SOLUTION [Liu Z et al. 2018d]
If the chemical reaction is attributed to the damage of evaporative concrete, a greater number of
chemical reaction products should be formed in the evaporative zone than in the immersed zone, thus
resulting in more severe deterioration. Cement pastes with different W/C ratios and different Fly ash
dosages were partially immersed in 5% Na2SO4 solution for 50 d and 100 d under N2 conditions. The
N2 condition was chosen to simulate the inner area of the evaporative concrete isolated from
carbonation. In order to directly compare the degree of chemical sulfate attack occurring in between
the evaporation zone and the immersed zone, the difference in mass loss ( △M) of ettringite or gypsum
formed in the evaporation zone and in the immersed zone of the paste was calculated based on the
TG results and shown in Figure 13. The △M of ettringite and the △M of gypsum of different pastes
were respectively calculated and shown in Figure.14.

Figure 13. The effect of W/C ratios, FA dosages and exposure ages on the total △M of ettringite
and gypsum in different pastes

Figure 14. The effect of W/C ratios and FA dosages on the △M of ettringite and gypsum in
different pastes after 100 d partial exposure
It can be observed that:

8
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

· The total △M of ettringite and gypsum was positive after 50 d exposure and after 100 d
exposure, meaning that the more severe chemical sulfate attack occurred in the evaporation
zone than in the immersed zone;
· An obvious increase in △M between 50 d and 100 d exposure could be seen, especially for
PC0.55 and FA30 paste. This suggests that the chemical sulfate attack became more severe
with more prolonged exposure;
· Compared with PC0.45, the replacement of cement with different dosages of FA noticeably
increased the △M of pastes. After 100 d exposure, the increase in dosage of FA resulted in
the larger △M, meaning that an increase in the dosage of FA could accelerate the chemical
sulfate attack in the evaporation zone.
· The main chemical sulfate attack product was gypsum. The △M of gypsum was significantly
larger than the △M of ettringite.
· The △M of gypsum for PC0.55 paste was almost double that of PC0.35 and PC0.45.
· The △M of ettringite and gypsum for FA paste was larger than that of PC0.45 with the same
W/C ratio.

5.3 Summary

The quantitative analysis of products in the different zones of pastes partially immersed in Na 2SO4
solution supports the idea that:

· More ettringite and a larger amount of gypsum are generated in the evaporative zone of pure
cement paste and cement + fly ash paste than in pure cement paste, thus causing more
severe damage in the evaporative zone.

6. CONCLUSION

According to the literature review, the real world of physical salt attack on concrete can be explained
as follows:
(1) Physical salt attack on concrete can be divided into two categories depending on the physical and
chemical relationship between the salt and the hydration products of the cement;
(2) The first category is the typical physical salt attack, which can be attributed to the damage
mechanism for evaporative concrete. In this case, the salt is inert to the hydration products of the
cement; examples include Portland cement concrete partially immersed in Ca(NO3)2 solution and
Calcium Sulfoaluminate cement concrete partially immersed in Na2SO4 solution. The concrete will be
damaged in quite a short space of time.
(3) The second category is the more complicated physical and chemical salt attack, where concrete
carbonation attributed to the damage mechanism for evaporative concrete is combined with salt
reacting with the hydration products of the cement. In this case, the concrete damage will be a longer
process than in the first one. Portland cement concrete partially immersed in the sulfate environment
belongs to this category and its damage mechanism can be summarized as follows:

· The chemical sulfate attack will be the only mechanism causing evaporative concrete damage
if there is no carbonation. As exposure progresses, the outer layer of evaporative concrete will
be neutralized due to carbonation, and thus become inert to sulfate. The physical sulfate
attack will then occur in the outer neutralized concrete layer. In the inner area of evaporative
concrete, which remains isolated from carbonation, the chemical sulfate attack will still be the
primary damage mechanism.
· The evaporative concrete damage is collectively controlled by the susceptibility of the concrete
to carbonation and the factors in favour of a chemical sulfate attack.
· The larger water to cement ratio, the use of mineral admixtures and a higher relative humidity
can accelerate concrete carbonation and hence evaporative concrete scaling.
· More chemical sulfate attack products will be generated in the evaporative concrete since
there is a greater water to cement ratio here, as well as higher relative humidity and the
presence of a mineral admixture.
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

7. ACKNOWLEDGMENTS
This study was funded by the National Science Foundation of China (Grant No. 51378508).

8. REFERENCES
ACI 201.2R (2016).:Guide to Durable Concrete, chapter 8, Physical salt attack;
Ahmed R. S., Moncef L. N. (2017), Exploring effects of supplementary cementitious materials in
concrete exposed to physical salt attack, Mag. Concr. Res., 69(11):576-585.
Bassuoni M.T., Rahman M.M. (2016), Response of concrete to accelerated physical salt attack
exposure, Cem Concr Res; 79:395-408;
Benavente D., García Del Cura MA. & Bernabéu A. (2001), Quantification of salt weathering in porous
stones using an experimental continuous partial immersion method. Eng Geol;59(3–4):313–25.
Chen F., Gao J., Qi B., Shen D. (2017), Deterioration mechanism of plain and blended cement
mortars partially exposed to sulfate attack, Constr. Build. Mater.,154:849–856;
Damidot D., Atkins M., Kindness A., Glasser F.P. (1992), Sulphate attack on concrete: limits of the AFt
stability domain, Cem. Concr. Res., 22(2-3): 229-234.
Ferraris CF, Stutzman PE, Snyder Kenneth A. (2006), Sulfate resistance of concrete: a new approach.
Research and development information, PCA R&D, Serial No. 2486;.
George W. S (2004), Stress from crystallization of salt, Cem. Concr. Res. 34 (9) 1613–1624
Irassar EF, Di Maio A, Batic OR. (1996), Sulfate attack on concrete with mineral admixtures. Cem
Concr Res;26(1):113–23.
Lee, B. Y. and Kurtis, K. E. Effect of pore structure on salt crystallization damage of cement based
materials: Consideration of w/b and nanoparticle use. Cem. Conc. Res. 98 (2017) 61-70.
Liu Z., Geert D S., Deng D. (2010), Micro-analysis of the role of interfacial transition zone in “salt
weathering” on concrete, Constr. Build. Mater., 24(11): 2052–2059.
Liu Z., Deng D., Geert De S. (2011), Micro-analysis of “salt weathering” on cement paste, Cem. Concr.
Compo., 2011; 33(1):179-191.
Liu Z., Deng D., De Schutter G. (2014), Does concrete suffer sulfate salt weathering? Constr. Build.
Mater., 66(1): 692-701.
Liu Z., Li X., Deng D., Geert D. S. (2016a), The damage of calcium sulfoaluminate cement paste
partially immersed in MgSO4 solution，Mater. Struct., 49:719–727.
Liu Z., Li X., Deng D., Geert D. S. (2016b), The role of Ca(OH) 2 in sulfate salt weathering of ordinary
concrete, Constr. Build. Mater., 123:127–134.
Liu, Z., Zhang, F., Deng, D., Xie, Y., (2017). Physical sulfate attack on concrete lining–A field case
analysis. Case Stud. Constr. Mater. 6: 206-212.
Liu Z., Hou L., Deng D., Zhang F.(2018a), Na2SO4 salt weathering of CSA cement paste partially
immersed in a Na2CO3 solution, J Mater. Civil. Eng., 30(3): 04017309.
Liu, Z., Hu W., Hou, L. (2018 b), Effect of carbonation on physical sulfate attack on concrete by
Na2SO4, Constr. Build. Mater. 2018 (193): 211–220.
Liu, Z., Hu W., Hou, L. (2018 c), The role of carbonation in the occurrence of MgSO4 crystallization
distress on concrete, Constr. Build. Mater. 2018 (192): 167–178.
Liu, Z., Pei M., Zhang F. (2018 d), Quantitative study on the products in Portland cement pastes
partially exposed to Na2SO4 solution, submitted to J Chin Ceram Soc. (in Chinese)
Mark W., Leon B. (2014), Current knowledge of external sulfate attack, Advances in Cement
Research, 1400089:1-14.
Najjar M.F., Nehdi M.L., A. Soliman M., Azabi T.M. (2017), Damage mechanisms of two-stage
concrete exposed to chemical and physical sulfate attack Constr. Build. Mater.,137: 141–152.
Navarro C. R., Doehne E. (1999), Salt weathering: influence of evaporation rate, supersaturation and
crystallization pattern, Earth Surf Processes and Landforms, 24(2-3): 91-209.
Nehdi M.L., Suleiman A.R., Soliman A.M. (2014), Investigation of concrete exposed to dual sulfate
attack, Cem Concr Res. 64:42–53;

10
 15th International Congress on the Chemistry of Cement
Prague, Czech Republic, September 16–20, 2019

Nishikawa T., Suzuki K., Ito S. (1992), Decomposition of synthesized ettringite by carbonation, Cem.
Concr. Res. 22 (1) :6–14.
Ruiz-Agudo E., Mees F., Jacobs P. (2007), The role of saline solution properties on porous limestone
salt weathering by magnesium and sodium sulfates, Environ. Geol.,52 (2): 305–317.
Stark D. (2002), Performance of concrete in sulfate environments, vol. RD129.Portland Cement
Association;.
Semion Z., Hooton D. H., (2017a), Accelerated testing of cementitious materials for resistance to
physical sulfate attack, Constr. Build. Mater, ,145:98–106.
Semion Z., Hooton D. H. (2017b), Experimental study on physical sulfate salt attack, Materials and
Structures, 50:54: 1-10
Shi C., Day R L. (2000), Pozzolanic reaction in the presence of chemical activators: Part II. Reaction
products and mechanism, Cem Concr Res., 30(4): 607-613.
Wang Y., Shu M., Zhang L. (1999), Sulphoaluminate cement, Beijing University of Technology Press,
Beijing
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):121–34.
Zhang Z., Wang Q., Chen H., Zhou Y. (2017), Influence of the initial moist curing time on the sulfate
attack resistance of concretes with different binders, Constr. Build. Mater.,144: 541–551;
Zhang F. (2018) on the wick action process of concrete partially immersed in sodium sulfate solution.
Changsha, Central South University. (in Chinese).