Research ArticleResearch on Sulfate Attack Mechanism of Cement ConcreteBased on Chemical Thermodynamics

Peng Liu 12 Ying Chen 123 Zhiwu Yu12 Lingkun Chen4 and Yongfeng Zheng 5

1School of Civil Engineering Central South University 22 Shaoshan Road Changsha 410075 China2National Engineering Laboratory for High Speed Railway Construction 22 Shaoshan Road Changsha 410075 China3School of Civil Engineering Central South University of Forestry and Technology 498 Shaoshan Road Changsha 410004 China4School of Civil Engineering Southwest Jiaotong University 6 Jingqu Road Chengdu 610031 China5Key Laboratory of Building Structural Retrofitting and Underground Space Engineering Shandong Jianzhu UniversityMinistry of Education Jinan Shandong 250101 China

Correspondence should be addressed to Ying Chen cheny83csueducn and Yongfeng Zheng zhyf-fb163com

Received 18 September 2019 Revised 16 December 2019 Accepted 28 January 2020 Published 9 March 2020

Academic Editor Mohammad A Hariri-Ardebili

Copyright copy 2020 Peng Liu et al0is is an open access article distributed under theCreative CommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Based on principles of chemical thermodynamics the relationship between temperature and the Gibbs free energy of erosionproducts generated during the sulfate attack on cement concrete was deduced 0e orientation of chemical reactions ofsulfate attack on cement concrete was theoretically determined as well as the critical sulfate ion concentration and theformation conditions of erosion products 0e phase composition microstructure crystal form and morphology of erosionproducts before and after sulfate attack were investigated by environmental scanning electron microscope and energyspectrum analysis (ESEM-EDS) and X-ray diffraction (XRD) 0e results show that the effects of sulfate ion concentrationand temperature on cement concrete sulfate attack are significant and different influencing factors correlate with eachother 0e crystal transition temperature between the anhydrite and dihydrate gypsum is 42degC and the correspondingconcentration of sulfate ion is about 23 times10minus 3 molL Simultaneously the crystal transition temperature between thethenardite and mirabilite is 324degC Moreover the theoretical upper limit temperature and sulfate ion lower limit con-centration of thaumasite are 44degC and 00023 molL respectively 0e ESEM-EDS and XRD results imply that the chemicalthermodynamics can be used to reveal the erosion mechanism of sulfate attack on cement concrete 0e major erosionproducts of sulfate attack on cement concrete are rod-like ettringite with a larger slenderness ratio plate-like gypsumgranular sulfate salt incompletely corroded calcium hydroxide and residual skeleton of calcium silicate hydrate 0e sulfateattack has double effects on mechanical properties of specimens which can affect the microstructure phase compositiontype and morphology of erosion products

1 Introduction

Sulfate attack on cement concrete can degrade the durabilityand reduce the service life of concrete building structures[1 2] 0ere are plenty of influence factors of sulfate attackon concrete which are mainly focused on the kind andconcentration of sulfate solution temperature pH valuecement composition admixtures and erosion form [3ndash7]Among them the temperature concentration and type ofthe sulfate solution are considered as the most importantfactors of sulfate attack [1 8 9] For example Valencia et al

[10] indicated that the MgSO4 had more influence on thedeterioration of concrete than that of Na2SO4 the resultsshowed that after attacked for 360 days in 5 MgSO4 so-lution the concrete expanded by 004 and its mechanicalresistance decreased by 33 Axel et al [11] regarded theerosion product as monosulfate when the concentration ofsulfate ion was less than 1 gL Lawrence [12] proposed thatthe highest stable temperature of the ettringite (AFt) wasabout 65degCsim70degC and the AFt was decomposed when thetemperature was higher than 70degC [13] Hekal et al [14]investigated the variation of the strength of the concrete

HindawiAdvances in Materials Science and EngineeringVolume 2020 Article ID 6916039 16 pageshttpsdoiorg10115520206916039

under different circumstances and they declared that thehigh temperature of 60degC played a significant effect onmechanical properties of concrete Dehwah [15] investigatedthe effect of sodium sulfate concentrations (to yield 125 and 4 [SO4

2minus ]) on concrete deterioration andmorphological changes in cement hydrates their resultsindicated that the deterioration was noted in the plain andfly ash cement concrete specimens exposed to sodiumchloride admixed with magnesium sulfate In general thesulfate solution concentration and temperature have re-markable influence on sulfate erosion mechanism crystalform and type erosion rate spontaneity of reaction andstable existence condition [16] 0erefore it is very im-portant to investigate the effects of sulfate solution con-centration and temperature on sulfate attack of concrete

Although a lot of achievements of sulfate attack on cementconcrete have been made [17] some divergences of sulfatesolution concentration and temperature on sulfate attack alsoexist For example Biczok [18] considered that the erosionproducts of sulfate attack on concrete were AFt and gypsumwhen [SO4

2minus ] was less than 1 gL and more than 8 gL re-spectively and gypsum and AFt could be observed when[SO4

2minus ] ranged from 1 gL to 8 gL Bellmann et al [19]pointed out that the portlandite reacted to gypsum at aminimal sulfate concentration of approximately 1400mgL(pH 1245) Zhang et al [20] conducted the acceleratedlaboratory tests on resistance of concrete exposed to sulfateattack with high concentration sulfate solutions or underdrastic drying-wetting cycle conditions their results showedthat the exposure regime of full immersion in 21 sulfatesodium solutions subjected to natural drying-wetting cyclescan well reproduce the field exposure condition of concreteunder certain sulfate-rich environments Both concentrationand exposure type affect the nature of sulfate attack mech-anism on concrete along with the evolution of physical andmechanical properties Yu et al [21] investigated the evolu-tion properties of the cement mortar fully immersed in so-dium sulfate solutions of different concentrations (0 5and 15) their results showed that the properties deterio-ration at the late stage was accelerated with increase of theconcentration of the sodium sulfate solution Santhanam [22]carried out an experimental investigation to understand theinfluence of the solution concentration of seawater on per-formance of the cementmortars and the results indicated thathigher concentrations of seawater did not cause the same levelof damage as higher concentrations of groundwater relative tothe typical seawatergroundwater solutions 0e temperaturenot only affects the reaction rate and diffusion of ions but alsochanges the erosion mechanism and products in cementconcrete Xie et al [23] proposed a novel theory concerningsulfate expansion phenomena in cement concrete systemsbased on the principles of chemical thermodynamics andtheir results showed that temperature had a significant effecton the concrete expansion Blanco-Varela et al [24] regardedthat the low temperature of less than 15degC as one of theessential conditions for the formation of thaumasite(CSCSH) Hartshorn et al [25] demonstrated that plenty ofthaumasite could be generated at 5degC and a little of thau-masite could also be formed at 20degC which caused a

deterioration of concrete However Santhanam [5] proposedthat the forming condition of thaumasite was temperature lessthan 20degC and the carbonation was another essential con-dition Moreover Brown et al [26] employed the scanningelectron microscope (SEM) to investigate the thaumasitegenerated in concrete attacked by sulfate and their resultsindicated that the formation of thaumasite was related toconcrete carbonation

Although many achievements regarding the effects ofsulfate solution concentration and temperature on theerosion mechanism of sulfate attack have been made theresearches of sulfate attack on cement concrete based onprinciples of chemical thermodynamics have no break-through progress as yet Furthermore there exists a devi-ation between the theoretical analysis and the in-site testresults 0e objective of this study is to investigate the re-lationship between temperature and the Gibbs free energy ofthe erosion products generated during the sulfate attackbased on the principle of chemical thermodynamics 0eorientation of chemical reactions of sulfate attack on con-crete was theoretically determined as well as the criticalsulfate ion concentration and forming conditions of erosionproducts Moreover the phase composition microstructurecrystal form and morphology of erosion products of cementbefore and after sulfate attack were studied by ESEM-EDSand XRD 0e research results in this study can provide asupport for the erosion mechanism of sulfate attack crystalform and forming conditions of erosion products in cementconcrete

2 Theoretical Analysis

0e sulfate attack on cement concrete is a complex physicaland chemical process [4 5] 0e temperature and sulfatesolution concentration have remarkable effects on thethermodynamic equilibrium state of the reaction so it is veryimportant to establish their international relationship Inorder to simplify the theoretical analysis we assume that thetemperature and concentration of various ions are uniformand unchangeable and the theoretical analysis of thermo-dynamic equilibrium state of the sulfate attack on cementconcrete is conducted as in the following section

21 Qualitative Analysis of the Erosion Products of SulfateAttack on Cement Concrete 0e relationship betweenthermodynamic equilibrium constant and the Gibbs freeenergy of the reaction can be used to characterize thespontaneity of the generation of erosion products [27]written as equation (1) Based on GibbsndashHelmholtz equation[27] the relationship between Gibbs energy and temperaturecan be obtained written as equation (2)

ΔGΘ minus RT lnKΘ

(1)

dlnKΘ

dTΔHΘ

RT2 (2)

where ΔGΘ is standard Gibbs free energy ΔHΘ is thestandard enthalpy change of reaction T is temperature KΘ

2 Advances in Materials Science and Engineering

stands for thermodynamic equilibrium constant and R isideal gas constant with a recommended value of 8314 J(molmiddotK)

Kirchhoff equation is usually used to describe therelationship between ΔHΘT and T formulated as equation(3)

d ΔHΘT1113872 1113873 ΔCpdT (3)

ΔCp 1113944ΔCp minus products minus 1113944ΔCp minus reactants

Δa + ΔbT + ΔcTminus 2

(4)

where ΔCp minus products and ΔCp minus reactants are the differences inthe specific heat at constant pressure between the productsand reactants respectively ΔCp is the difference in theheat capacity of the reaction ie the difference betweenthe sum of molar heat capacity under constant pressure ofproducts and the sum of molar heat capacity underconstant pressure of reactants Δa Δb and Δc are fittedconstants respectively

0e change of the molar heat capacity Cp under constantpressure of products with temperature T can be approxi-mately described by equation (5) Hence equation (4) can berepresented as equation (6)

Cp A1 + A2 times 10minus 3T + A3 times 105Tminus 2

+ A4 times 10minus 6T2

+ A5 times 108Tminus 3

(5)

ΔCp ΔA1 + ΔA2 times 10minus 3T + ΔA3 times 105Tminus 2

+ ΔA4

times 10minus 6T2

+ ΔA5 times 108Tminus 3

(6)

where A1 A2 A3 A4 and A5 are the fitted data respectivelyΔA1 ΔA2 ΔA3 ΔA4 and ΔA5 are the differences in thecorresponding fitted data respectively

Integrating equations (3) and (5) the expression ofthe heat function of the reaction can be determined by[28]

ΔHΘT ΔA1T +12ΔA2 times 10minus 3

T2

minus ΔA3 times 105Tminus 1+13ΔA4

times 10minus 6T3

minus12ΔA5 times 108Tminus 2

(7)

If the standard molar formation heats of various sub-stances at room temperature ΔHΘif298 are known thecorresponding reaction heat of the system at T 298K canbe determined as

ΔHΘT 1113944 niΔHΘif2981113872 1113873reactants minus 1113944 niΔH

Θif2981113872 1113873products

(8)

Substituting the ΔHΘ298 calculated by equation (8) andT 298K into equation (7) the corresponding integrationconstant A6 can be determined

A6 ΔHΘ298 minus ΔA1T minus12ΔA2 times 10minus 3

T2

+ ΔA3 times 105Tminus 1

minus13ΔA4 times 10minus 6

T3

+12ΔA5 times 108Tminus 2

(9)

Using a numerical integrating of equation (2) thecorresponding equation can be obtained as equation (10)Integrating equations (7) and (9) the standard Gibbs freeenergy of the reaction at T can be expressed as equation(11)ΔGΘT

T minus 1113946ΔHΘT

T2 dT (10)

ΔGΘT minus ΔA1T lnT minus12ΔA2 times 10minus 3

T2

minus12ΔA3 times 105Tminus 1

minus16ΔA4 times 10minus 6

T3

minus16ΔA5 times 108Tminus 2

+ A6primeT + A6

(11)

where A6prime is the integration constant of the GibbsndashHelmholtzequation

If the change of the standard enthalpy ΔHΘ298 and thethermal entropy ΔSΘ298 of the reaction at room temperature(298K) are known the corresponding standard Gibbs freeenergy of the reaction at 298K can be determined byequation (12) Substituting the ΔGΘ298 calculated by equation(12) and T 298K into equation (11) the correspondingintegration constant A6prime can be determined by equation (13)

ΔGΘ298 ΔHΘ298 minus 298ΔSΘ298 (12)

A6prime ΔGΘ298

T+ ΔA1 lnT +

12ΔA2 times 10minus 3

T

+12ΔA3 times 105Tminus 2

+16ΔA4 times 10minus 6

T2

+16ΔA5 times 108Tminus 3

minus A6Tminus 1

(13)

A conclusion can be drawn from equation (1) that thesulfate attack on cement concrete can be spontaneous whenthe Gibbs free energy is less than zero Conversely thereaction cannot progress spontaneously Equation (11)shows the relationship between temperature and the Gibbsfree energy of the erosion products generated during thesulfate attack on cement concrete which can be used toreveal the direction of chemical reaction existing conditionand chemical stability of erosion products and spontaneityof the erosion reaction

22 Critical Ion Concentration of Erosion Products of theSulfate Attack on Cement Concrete Some researchers [27]proposed the expression of the thermodynamic equilibriumconstant written as

lgKΘeq A + BT +

C

T+ Dlg(T) +

E

T2(14)

Advances in Materials Science and Engineering 3

where KΘeq stands for the thermodynamic equilibriumconstant at standard conditions A B C D and E areconstant respectively

0e main erosion products of sulfate attack on cementconcrete may be gypsum ettringite and sulfate salt [29 30] and[31] and the corresponding reactions can be written as [19 32]

Ca2+(aq) + SO2minus

4(aq) Crystallize

DissolveCaSO4(s) (15)

Ca2+(aq) + SO2minus

4(aq) + 2H2O(l) Crystallize

DissolveCaSO4 middot 2H2O(s) (16)

2Na+(aq) + SO2minus

4(aq) Crystallize

DissolveNa2SO4(s) (17)

2Na+(aq) + SO2minus

4(aq) + 10H2O(l) Crystallize

DissolveNa2SO4 middot 10H2O(s)

(18)

6Ca2++ 2Al3+

+ 3SO2minus4 + 32H2O(l)

Crystallize

Dissolve3CaO middot Al2O3

middot 3CaSO4 middot 32H2O(s)

(19)

CaSiO3 middot CaSO4 middot CaCO3 middot 15H2O(s) + 3H+ Crystallize

Dissolve3Ca2+

+SO2minus4 + HCOminus

3 + H4SiO4 + 14H2O(l)

(20)

where subscripts s aq and l stand for solid state ionic statein solution and liquid respectively

Correspondingly the decomposition reactions of thecement hydration products can be expressed as follows[19 33 34]

Ca(OH)2(s) + 2H+ Ca2++ 2H2O (21)

C3AH6(s) + 12H+ 2Al3++ 3Ca2+

+ 12H2O (22)

C4AH13(s) + 14H+ 2Al3++ 4Ca2+

+ 20H2O (23)

C-S-H(08)(s) + 16H+ 034H2O + 08Ca2++ H4SiO4

(24)

C-S-H(12)(s) + 24H+ 126H2O + 12Ca2++ H4SiO4

(25)

Ca2SiO4(s) + 4H+ 2Ca2++ H4SiO4 (26)

Al(OH)3(s) + 3H+Al3++ 3H2O (27)

Ca6Al2 SO4( 11138573(OH)12 middot 26H2O(s) + 12H+ 2Al3++ 6Ca2+

+ 3SO2minus4 + 38H2O

(28)

0e reference values recommended by researchers[35 36] are listed in Table 1

0e above reaction equations can be formulated asequation (29) and the relationship between the thermo-dynamic equilibrium constant and concentration of dif-ferent substances can be expressed as equation (30)

dD + eE gG + hH (29)

KΘeq

[G]g[H]h

[D]d[E]e (30)

where [D] [E] [G] and [H] mean the concentration ofdifferent substances respectively d e g and h are thestoichiometric numbers respectively

Combining equations (14) and (30) the formationconditions and the critical ion concentration of sulfate ionfor the formation of erosion products can be determinedbased on the thermodynamic parameter of different sub-stances in Table 1

3 Experimental Procedure

31 Raw Materials Portland cement of P O 325 class wasproduced by Pingtang Cement Plant of Hunan and thepolycarboxylic acid series of superplasticizer with solidcontent of 30 provided by Changsha Huangteng ChemicalTechnology Co Ltd was used as a water reducer ISOstandard sand with a fineness modulus of 30 was producedby Xiamen Aisiou Standard Sand Co Ltd Local river sandwith a fineness modulus of 29 continuous grading lime-stone gravel with grains sizes of 5mmsim20mm and tap waterwere used to prepare the specimens Moreover the indus-trial grade sodium sulfate with purity of 99 was purchasedfrom market 0e chemical composition and physicalproperties of cement are listed in Tables 2 and 3

32 Specimen Preparation Experimental Process and TestDevices According to Common Portland Cement and TestMethods for Water Requirement of Normal Consistency SettingTime and Soundness of the Portland Cement [37 38] the pastespecimens of 40mmtimes 40mmtimes 160mmwere cast with a waterto cement ratio of 028 and were used for XRD analysisAccording to Method of Testing CementsndashDetermination ofStrength the mortar specimens of 40mmtimes 40mmtimes 160mmwere cast with a water cement sand ratio of 05 13 by weight[39] According to Standard for Test Method of MechanicalProperties onOrdinary Concrete [40] the concrete specimens of100mmtimes 100mmtimes 300mm were cast and there are threespecimens for each group 0e C20 grade concrete was pre-pared with a weight ratio of cement sand limestone waterwater reducer of 305 831 1102 158 35 All specimens werecured in standard curing pool at a temperature of (20plusmn 1)degCand a relative humidity of (95plusmn 3) 0e specimens weredemolded after 24h and cured in water for 28d at a tem-perature of (20plusmn 1)degC Subsequently the specimens were takenout from water to carry out sulfate attack test

4 Advances in Materials Science and Engineering

0e experimental process of the concrete attacked bysulfate was carried out as follows Firstly the sulfate solu-tions of different concentrations (ie 1 5 10 andsaturated solution) were prepared and various specimenswere immersed into sulfate solution In order to ensure theuniformity of sulfate solution the distance of specimens wasset as no less than 2 cm 0en the sealed solution box wascovered with plastic film to prevent the water from evap-oration Moreover the sulfate solution was changed once aweek Finally the specimens were taken out from sulfatesolution and dried at room temperature of 25degC and arelative humidity of 70 for 3 days when the erosion agesreached the setting time of 2 and 4 months 0e flexuralstrength of the mortar and paste specimens was first testedand then the corresponding compressive strength

0e preparation process of the samples for micro-properties test was as follows Firstly the specimens reachingthe sulfate attack age were cleaned with distilled water andbroken into particles of 2mmsim5mm Secondly the particleswere immersed into alcohol for 24 h to terminate reaction0en the particles were dried at 60degC for 48 h Finally theprocessed particles were placed and stored in a dryer 0eproceeded particles containing surface were selected andtheir cross-sections were sprayed with gold for ESEM-EDSanalysis 0e observed localization of the proceeded particle

for ESEM-EDS analysis was near the outer surface of thecross-sections In order to conduct the XRD analysis thesuperficial region within 2mm from cement sample surfacewas obtained and treated as mentioned above Subsequentlythe proceeded cement paste particles were ground to powderthrough 200 mesh and the scanning step of XRD analysiswas 002deg All samples were prepared and tested according tothe equipment guide

0e main test devices were as follows 0e Quanta-200Environment Scanning Electronic Microscope (ESEM-EDS)produced by FTI Company of Czech Republic with anamplification of 600000 times was used to measure themicrostructure 0e Dmax-2550 X-ray diffraction (XRD)with the angular range of 5degsim120deg and a minimum scanningstep size of 001deg was used to carry out the phase analysis0eYAW-300D produced by Jinan Kesheng Test Equipment CoLtd of China was used to measure the flexural and com-pressive strength of the specimens 0e maximum com-pressive and flexural loads were 300 kN and 10 kNrespectively0e corresponding loading rates of compressiveand flexural strength were (24plusmn 02) kNs and (50plusmn 10)Nsrespectively Moreover the WAW-1000 electrohydraulicservo universal testing machine made by Shanghai Sansi CoLtd of China was used to measure the compressive strengthof the concrete According to Standard for Test Method of

Table 1 0ermodynamic parameters of different substances

Items A B C D ECaSO4 middot 2H2O(s) 162021439e+ 3 257234846e minus 1 minus 891506186e+ 4 minus 587385148e+ 2 534735206e+ 6CaSO4(s) 161807826e+ 3 262044313e minus 1 minus 895853477e+ 4 minus 586632877e+ 2 535893242e+ 6CaSO4(aq) 172034184e+ 3 265734992e minus 1 minus 942553556e+ 4 minus 623563883e+ 2 549729959e+ 6Na2SO4 middot 10H2O(s) 158837182e+ 3 231777424e minus 1 minus 843055786e+ 4 minus 578226172e+ 2 509260164e+ 6Na2SO4(S) 161633032e+ 3 253239679e minus 1 minus 898032147e+ 4 minus 586414685e+ 2 540049408e+ 6AFt(s) Ca6Al2(SO4)3(OH)12middot26H2O minus 667460197e+ 3 minus 104743390 378708023e+ 5 242662966e+ 3 minus 205189885e+ 7Ca(OH)2(s) minus 284930557e+ 2 minus 447108160e minus 2 213801821e+ 4 104205027e+ 2 minus 754252617e+ 5Al(OH)3(s) minus 493752625e+ 2 minus 809001544e minus 2 297138788e+ 4 177903516e+ 2 minus 126765911e+ 6NaAlO2(aq) 704197406e+ 2 111341739e minus 1 minus 474872291e+ 4 minus 253129969e+ 2 218693139e+ 6AlO2

minus(aq) minus 178049448e+ 2 minus 268902416e minus 2 186721225e+ 3 668330936e+ 1 minus 750442968e+ 5

H2O(aq) minus 701957319 minus 0112739992 36168254 25360128 minus 242327306C-S-H (08)(s) Ca08SiO28middot154H2O minus 251027448e+ 2 minus 365449659e minus 2 155601183e+ 4 921931312e+ 1 minus 652576525e+ 5C-S-H(12)(s) Ca12SiO32middot206H2O minus 372034132e+ 2 minus 541579845e minus 2 239845213e+ 4 136559314e+ 2 minus 966181083e+ 5C-S-H(16)(s) Ca16SiO36middot258H2O minus 491723254e+ 2 minus 717628896e minus 2 325370146e+ 4 180399649e+ 2 minus 127964571e+ 6C2AH8(s) Ca2Al2O5middot8H2O minus 173467190e+ 3 minus 238046203e minus 1 110411145e+ 5 623227246e+ 2 minus 418576035e+ 6C3AH6(s) Ca3Al2(OH)12 minus 178581145e+ 3 minus 288036362e minus 1 119706499e+ 5 647583573e+ 2 minus 461172146e+ 6CaSiO3CaSO4CaCO3middot15H2O(s) minus 276283310e+ 3 minus 432736383e minus 1 149848596e+ 5 100850740e+ 3 minus 852601737e+ 6α-Ca2SiO4(s) minus 526934036e+ 2 minus 867703839e minus 2 389768377e+ 4 192577860e+ 2 minus 136961546e+ 6CaSiO3(s) minus 265453934e+ 2 minus 441966587e minus 2 177395216e+ 4 972523635e+ 1 minus 665653039e+ 5

Table 2 Chemical composition of cement ()

CaO SiO2 Al2O3 P2O5 Fe2O3 MgO K2O Na2O Loss6561 1989 473 288 145 201 114 021 198

Table 3 Physical properties of cement

Average diameter(μm)

Specific surface area(m2kg)

Bulk density(gcm3)

Initial setting time(min)

Final setting time(min)

Water requirement for standardconsistency ()

346 3452 135 172 251 28

Advances in Materials Science and Engineering 5

Mechanical Properties on Ordinary Concrete [40] the cor-responding loading rate was in the range of 03MPassim055MPas

4 Results and Discussions

41 Assessment of Erosion Products of Sulfate Attack on Ce-ment Concrete 0e main erosion products of sulfate attackon cement concrete may be gypsum ettringite sulfate saltand so on0erefore the formation spontaneity of the aboveerosion products was deduced firstly based on the principlesof chemical thermodynamics Because the thermodynamicsparameters of different erosion products were affected byenvironmental conditions remarkably the thermodynamicsparameters of different reactions for erosion products atdifferent conditions including dry humid and water en-vironment were calculated theoretically 0e reactions ofdihydrate gypsum and anhydrite and mirabilite and the-nardite can be written as follows (subscript (g) stands for thegaseous state) [19 32 36]

CaSO4(s) + 2H2O(g) CaSO4 middot 2H2O(s) (31)

CaSO4(s) + 2H2O(l) CaSO4 middot 2H2O(s) (32)

Na2SO4(s) + 10H2O(g)Na2SO4 middot 10H2O(s) (33)

Na2SO4(s) + 10H2O(l)Na2SO4 middot 10H2O(s) (34)

0e corresponding thermodynamic parameters of dif-ferent reactions can be determined based on equations(4)sim(13) and Table 4 as listed in Table 5 Moreover theGibbs free energy curves of various erosion products underdifferent conditions as a function of temperature are plottedin Figure 1

Figure 1(a) shows that the theoretical transformationtemperature between dihydrate gypsum and anhydriteunder humid and water conditions is about 91degC that is thedihydrate gypsum can be spontaneously formed when thetemperature is lower than 91degC However the correspondingtemperature under dry condition is about 56degC ie thedihydrate gypsum can be translated into anhydrite when thetemperature exceeds 56degC Figure 1(b) indicates that thetheoretical transformation temperature between mirabiliteand thenardite under dry condition is about 45degC howeverit is about 88degC under humid and water conditions From theabove discussions a conclusion can be drawn that the en-vironment conditions have significant effects on thetransformation of the crystals with combined water0erefore the environmental condition should be consid-ered as an important factor of sulfate attack on cementconcrete during the test process of sulfate attack andspecimensrsquo preparation It is the theoretical transformationtemperature curve based upon which the substances with orwithout crystal water would be converted In Figure 1 theGibbs free energy curves are calculated based on the the-oretical transformation temperature of the substanceswithout crystal water converted into the substances withcrystal water Because the theoretical calculated data do not

consider the activity of substances there exists a differencebetween the theoretical calculated results and the in-sitemeasured data 0e research [41] showed that the solubilityof calcium sulfate at normal temperatures and pressures wasabout 02 and the temperature corresponding to the in-tersection of solubility curves of dihydrate gypsum andanhydrite was 42degC [42] 0e above achievements accordwell with the theoretical calculated data in Figure 1 whichimplies that equation (11) proposed by this study can be usedto determine the mechanism of sulfate attack Because all thesulfate attacks took place in the concrete pore solution thedeterioration mechanism and erosion products of sulfateattack reacting in solution were further investigated

42 Numerical Analysis of Critical Concentration of SulfateIon of Different Erosion Products with TemperatureCalcium hydroxide (CH) in cement and concrete has asignificant effect on the thermodynamic equilibrium ofvarious products so the variation of critical concentration of[OHminus ] and [Ca2+] with temperature is investigated firstlyAssuming the content of hydroxyl ion is generated by CHand dominated by CH saturated solution the correspondingchemical equilibrium based on Hessrsquos law can be expressedas follows [36]Ca(OH)2(s) + 2H+ Ca2++2H2O (1)

H2O H+ + OHminus (2)

⎫⎬

⎭⟹(1) + 2 times(2)⟹

Ca(OH)2(s) Ca2++ 2OHminus

(35)

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the above reaction can bedetermined based on equation (1) written as

lnKΘ

lnKΘ1 + 2 lnK

Θ2 (36)

Assuming the [OHminus ] in the system is generated byCa(OH)2 the concentration of calcium ion is half of thehydroxyl ion 0e concentration of hydroxyl ion withtemperature can be determined based on equation (36) asshown in Figure 2 If the activity of condensed matters is setas 1 the critical concentration of sulfate ion for the for-mation of dihydrate gypsum and anhydrite in equations (15)and (16) can be deduced on the basis of the calcium ionconcentration and thermodynamic equilibrium constantwritten as equations (37) and (38) 0e correspondingcritical concentration curves of sulfate ions for the formationof dihydrate gypsum and anhydrite as a function of tem-perature are plotted in Figure 2

SO2minus41113960 1113961

KΘCaSO4 middot2H2O

a Ca2+[ ]middot cΘ

(37)

SO2minus41113960 1113961

KΘCaSO4

a Ca2+[ ]middot cΘ

(38)

As seen from Figure 2(a) the theoretical concentrationsof [Ca2+] and [OHminus ] corresponding to the thermodynamic

6 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

stands for thermodynamic equilibrium constant and R isideal gas constant with a recommended value of 8314 J(molmiddotK)

Kirchhoff equation is usually used to describe therelationship between ΔHΘT and T formulated as equation(3)

d ΔHΘT1113872 1113873 ΔCpdT (3)

ΔCp 1113944ΔCp minus products minus 1113944ΔCp minus reactants

Δa + ΔbT + ΔcTminus 2

(4)

where ΔCp minus products and ΔCp minus reactants are the differences inthe specific heat at constant pressure between the productsand reactants respectively ΔCp is the difference in theheat capacity of the reaction ie the difference betweenthe sum of molar heat capacity under constant pressure ofproducts and the sum of molar heat capacity underconstant pressure of reactants Δa Δb and Δc are fittedconstants respectively

0e change of the molar heat capacity Cp under constantpressure of products with temperature T can be approxi-mately described by equation (5) Hence equation (4) can berepresented as equation (6)

Cp A1 + A2 times 10minus 3T + A3 times 105Tminus 2

+ A4 times 10minus 6T2

+ A5 times 108Tminus 3

(5)

ΔCp ΔA1 + ΔA2 times 10minus 3T + ΔA3 times 105Tminus 2

+ ΔA4

times 10minus 6T2

+ ΔA5 times 108Tminus 3

(6)

where A1 A2 A3 A4 and A5 are the fitted data respectivelyΔA1 ΔA2 ΔA3 ΔA4 and ΔA5 are the differences in thecorresponding fitted data respectively

Integrating equations (3) and (5) the expression ofthe heat function of the reaction can be determined by[28]

ΔHΘT ΔA1T +12ΔA2 times 10minus 3

T2

minus ΔA3 times 105Tminus 1+13ΔA4

times 10minus 6T3

minus12ΔA5 times 108Tminus 2

(7)

If the standard molar formation heats of various sub-stances at room temperature ΔHΘif298 are known thecorresponding reaction heat of the system at T 298K canbe determined as

ΔHΘT 1113944 niΔHΘif2981113872 1113873reactants minus 1113944 niΔH

Θif2981113872 1113873products

(8)

Substituting the ΔHΘ298 calculated by equation (8) andT 298K into equation (7) the corresponding integrationconstant A6 can be determined

A6 ΔHΘ298 minus ΔA1T minus12ΔA2 times 10minus 3

T2

+ ΔA3 times 105Tminus 1

minus13ΔA4 times 10minus 6

T3

+12ΔA5 times 108Tminus 2

(9)

Using a numerical integrating of equation (2) thecorresponding equation can be obtained as equation (10)Integrating equations (7) and (9) the standard Gibbs freeenergy of the reaction at T can be expressed as equation(11)ΔGΘT

T minus 1113946ΔHΘT

T2 dT (10)

ΔGΘT minus ΔA1T lnT minus12ΔA2 times 10minus 3

T2

minus12ΔA3 times 105Tminus 1

minus16ΔA4 times 10minus 6

T3

minus16ΔA5 times 108Tminus 2

+ A6primeT + A6

(11)

where A6prime is the integration constant of the GibbsndashHelmholtzequation

If the change of the standard enthalpy ΔHΘ298 and thethermal entropy ΔSΘ298 of the reaction at room temperature(298K) are known the corresponding standard Gibbs freeenergy of the reaction at 298K can be determined byequation (12) Substituting the ΔGΘ298 calculated by equation(12) and T 298K into equation (11) the correspondingintegration constant A6prime can be determined by equation (13)

ΔGΘ298 ΔHΘ298 minus 298ΔSΘ298 (12)

A6prime ΔGΘ298

T+ ΔA1 lnT +

12ΔA2 times 10minus 3

T

+12ΔA3 times 105Tminus 2

+16ΔA4 times 10minus 6

T2

+16ΔA5 times 108Tminus 3

minus A6Tminus 1

(13)

A conclusion can be drawn from equation (1) that thesulfate attack on cement concrete can be spontaneous whenthe Gibbs free energy is less than zero Conversely thereaction cannot progress spontaneously Equation (11)shows the relationship between temperature and the Gibbsfree energy of the erosion products generated during thesulfate attack on cement concrete which can be used toreveal the direction of chemical reaction existing conditionand chemical stability of erosion products and spontaneityof the erosion reaction

22 Critical Ion Concentration of Erosion Products of theSulfate Attack on Cement Concrete Some researchers [27]proposed the expression of the thermodynamic equilibriumconstant written as

lgKΘeq A + BT +

C

T+ Dlg(T) +

E

T2(14)

Advances in Materials Science and Engineering 3

where KΘeq stands for the thermodynamic equilibriumconstant at standard conditions A B C D and E areconstant respectively

0e main erosion products of sulfate attack on cementconcrete may be gypsum ettringite and sulfate salt [29 30] and[31] and the corresponding reactions can be written as [19 32]

Ca2+(aq) + SO2minus

4(aq) Crystallize

DissolveCaSO4(s) (15)

Ca2+(aq) + SO2minus

4(aq) + 2H2O(l) Crystallize

DissolveCaSO4 middot 2H2O(s) (16)

2Na+(aq) + SO2minus

4(aq) Crystallize

DissolveNa2SO4(s) (17)

2Na+(aq) + SO2minus

4(aq) + 10H2O(l) Crystallize

DissolveNa2SO4 middot 10H2O(s)

(18)

6Ca2++ 2Al3+

+ 3SO2minus4 + 32H2O(l)

Crystallize

Dissolve3CaO middot Al2O3

middot 3CaSO4 middot 32H2O(s)

(19)

CaSiO3 middot CaSO4 middot CaCO3 middot 15H2O(s) + 3H+ Crystallize

Dissolve3Ca2+

+SO2minus4 + HCOminus

3 + H4SiO4 + 14H2O(l)

(20)

where subscripts s aq and l stand for solid state ionic statein solution and liquid respectively

Correspondingly the decomposition reactions of thecement hydration products can be expressed as follows[19 33 34]

Ca(OH)2(s) + 2H+ Ca2++ 2H2O (21)

C3AH6(s) + 12H+ 2Al3++ 3Ca2+

+ 12H2O (22)

C4AH13(s) + 14H+ 2Al3++ 4Ca2+

+ 20H2O (23)

C-S-H(08)(s) + 16H+ 034H2O + 08Ca2++ H4SiO4

(24)

C-S-H(12)(s) + 24H+ 126H2O + 12Ca2++ H4SiO4

(25)

Ca2SiO4(s) + 4H+ 2Ca2++ H4SiO4 (26)

Al(OH)3(s) + 3H+Al3++ 3H2O (27)

Ca6Al2 SO4( 11138573(OH)12 middot 26H2O(s) + 12H+ 2Al3++ 6Ca2+

+ 3SO2minus4 + 38H2O

(28)

0e reference values recommended by researchers[35 36] are listed in Table 1

0e above reaction equations can be formulated asequation (29) and the relationship between the thermo-dynamic equilibrium constant and concentration of dif-ferent substances can be expressed as equation (30)

dD + eE gG + hH (29)

KΘeq

[G]g[H]h

[D]d[E]e (30)

where [D] [E] [G] and [H] mean the concentration ofdifferent substances respectively d e g and h are thestoichiometric numbers respectively

Combining equations (14) and (30) the formationconditions and the critical ion concentration of sulfate ionfor the formation of erosion products can be determinedbased on the thermodynamic parameter of different sub-stances in Table 1

3 Experimental Procedure

31 Raw Materials Portland cement of P O 325 class wasproduced by Pingtang Cement Plant of Hunan and thepolycarboxylic acid series of superplasticizer with solidcontent of 30 provided by Changsha Huangteng ChemicalTechnology Co Ltd was used as a water reducer ISOstandard sand with a fineness modulus of 30 was producedby Xiamen Aisiou Standard Sand Co Ltd Local river sandwith a fineness modulus of 29 continuous grading lime-stone gravel with grains sizes of 5mmsim20mm and tap waterwere used to prepare the specimens Moreover the indus-trial grade sodium sulfate with purity of 99 was purchasedfrom market 0e chemical composition and physicalproperties of cement are listed in Tables 2 and 3

32 Specimen Preparation Experimental Process and TestDevices According to Common Portland Cement and TestMethods for Water Requirement of Normal Consistency SettingTime and Soundness of the Portland Cement [37 38] the pastespecimens of 40mmtimes 40mmtimes 160mmwere cast with a waterto cement ratio of 028 and were used for XRD analysisAccording to Method of Testing CementsndashDetermination ofStrength the mortar specimens of 40mmtimes 40mmtimes 160mmwere cast with a water cement sand ratio of 05 13 by weight[39] According to Standard for Test Method of MechanicalProperties onOrdinary Concrete [40] the concrete specimens of100mmtimes 100mmtimes 300mm were cast and there are threespecimens for each group 0e C20 grade concrete was pre-pared with a weight ratio of cement sand limestone waterwater reducer of 305 831 1102 158 35 All specimens werecured in standard curing pool at a temperature of (20plusmn 1)degCand a relative humidity of (95plusmn 3) 0e specimens weredemolded after 24h and cured in water for 28d at a tem-perature of (20plusmn 1)degC Subsequently the specimens were takenout from water to carry out sulfate attack test

4 Advances in Materials Science and Engineering

0e experimental process of the concrete attacked bysulfate was carried out as follows Firstly the sulfate solu-tions of different concentrations (ie 1 5 10 andsaturated solution) were prepared and various specimenswere immersed into sulfate solution In order to ensure theuniformity of sulfate solution the distance of specimens wasset as no less than 2 cm 0en the sealed solution box wascovered with plastic film to prevent the water from evap-oration Moreover the sulfate solution was changed once aweek Finally the specimens were taken out from sulfatesolution and dried at room temperature of 25degC and arelative humidity of 70 for 3 days when the erosion agesreached the setting time of 2 and 4 months 0e flexuralstrength of the mortar and paste specimens was first testedand then the corresponding compressive strength

0e preparation process of the samples for micro-properties test was as follows Firstly the specimens reachingthe sulfate attack age were cleaned with distilled water andbroken into particles of 2mmsim5mm Secondly the particleswere immersed into alcohol for 24 h to terminate reaction0en the particles were dried at 60degC for 48 h Finally theprocessed particles were placed and stored in a dryer 0eproceeded particles containing surface were selected andtheir cross-sections were sprayed with gold for ESEM-EDSanalysis 0e observed localization of the proceeded particle

for ESEM-EDS analysis was near the outer surface of thecross-sections In order to conduct the XRD analysis thesuperficial region within 2mm from cement sample surfacewas obtained and treated as mentioned above Subsequentlythe proceeded cement paste particles were ground to powderthrough 200 mesh and the scanning step of XRD analysiswas 002deg All samples were prepared and tested according tothe equipment guide

0e main test devices were as follows 0e Quanta-200Environment Scanning Electronic Microscope (ESEM-EDS)produced by FTI Company of Czech Republic with anamplification of 600000 times was used to measure themicrostructure 0e Dmax-2550 X-ray diffraction (XRD)with the angular range of 5degsim120deg and a minimum scanningstep size of 001deg was used to carry out the phase analysis0eYAW-300D produced by Jinan Kesheng Test Equipment CoLtd of China was used to measure the flexural and com-pressive strength of the specimens 0e maximum com-pressive and flexural loads were 300 kN and 10 kNrespectively0e corresponding loading rates of compressiveand flexural strength were (24plusmn 02) kNs and (50plusmn 10)Nsrespectively Moreover the WAW-1000 electrohydraulicservo universal testing machine made by Shanghai Sansi CoLtd of China was used to measure the compressive strengthof the concrete According to Standard for Test Method of

Table 1 0ermodynamic parameters of different substances

Items A B C D ECaSO4 middot 2H2O(s) 162021439e+ 3 257234846e minus 1 minus 891506186e+ 4 minus 587385148e+ 2 534735206e+ 6CaSO4(s) 161807826e+ 3 262044313e minus 1 minus 895853477e+ 4 minus 586632877e+ 2 535893242e+ 6CaSO4(aq) 172034184e+ 3 265734992e minus 1 minus 942553556e+ 4 minus 623563883e+ 2 549729959e+ 6Na2SO4 middot 10H2O(s) 158837182e+ 3 231777424e minus 1 minus 843055786e+ 4 minus 578226172e+ 2 509260164e+ 6Na2SO4(S) 161633032e+ 3 253239679e minus 1 minus 898032147e+ 4 minus 586414685e+ 2 540049408e+ 6AFt(s) Ca6Al2(SO4)3(OH)12middot26H2O minus 667460197e+ 3 minus 104743390 378708023e+ 5 242662966e+ 3 minus 205189885e+ 7Ca(OH)2(s) minus 284930557e+ 2 minus 447108160e minus 2 213801821e+ 4 104205027e+ 2 minus 754252617e+ 5Al(OH)3(s) minus 493752625e+ 2 minus 809001544e minus 2 297138788e+ 4 177903516e+ 2 minus 126765911e+ 6NaAlO2(aq) 704197406e+ 2 111341739e minus 1 minus 474872291e+ 4 minus 253129969e+ 2 218693139e+ 6AlO2

minus(aq) minus 178049448e+ 2 minus 268902416e minus 2 186721225e+ 3 668330936e+ 1 minus 750442968e+ 5

H2O(aq) minus 701957319 minus 0112739992 36168254 25360128 minus 242327306C-S-H (08)(s) Ca08SiO28middot154H2O minus 251027448e+ 2 minus 365449659e minus 2 155601183e+ 4 921931312e+ 1 minus 652576525e+ 5C-S-H(12)(s) Ca12SiO32middot206H2O minus 372034132e+ 2 minus 541579845e minus 2 239845213e+ 4 136559314e+ 2 minus 966181083e+ 5C-S-H(16)(s) Ca16SiO36middot258H2O minus 491723254e+ 2 minus 717628896e minus 2 325370146e+ 4 180399649e+ 2 minus 127964571e+ 6C2AH8(s) Ca2Al2O5middot8H2O minus 173467190e+ 3 minus 238046203e minus 1 110411145e+ 5 623227246e+ 2 minus 418576035e+ 6C3AH6(s) Ca3Al2(OH)12 minus 178581145e+ 3 minus 288036362e minus 1 119706499e+ 5 647583573e+ 2 minus 461172146e+ 6CaSiO3CaSO4CaCO3middot15H2O(s) minus 276283310e+ 3 minus 432736383e minus 1 149848596e+ 5 100850740e+ 3 minus 852601737e+ 6α-Ca2SiO4(s) minus 526934036e+ 2 minus 867703839e minus 2 389768377e+ 4 192577860e+ 2 minus 136961546e+ 6CaSiO3(s) minus 265453934e+ 2 minus 441966587e minus 2 177395216e+ 4 972523635e+ 1 minus 665653039e+ 5

Table 2 Chemical composition of cement ()

CaO SiO2 Al2O3 P2O5 Fe2O3 MgO K2O Na2O Loss6561 1989 473 288 145 201 114 021 198

Table 3 Physical properties of cement

Average diameter(μm)

Specific surface area(m2kg)

Bulk density(gcm3)

Initial setting time(min)

Final setting time(min)

Water requirement for standardconsistency ()

346 3452 135 172 251 28

Advances in Materials Science and Engineering 5

Mechanical Properties on Ordinary Concrete [40] the cor-responding loading rate was in the range of 03MPassim055MPas

4 Results and Discussions

41 Assessment of Erosion Products of Sulfate Attack on Ce-ment Concrete 0e main erosion products of sulfate attackon cement concrete may be gypsum ettringite sulfate saltand so on0erefore the formation spontaneity of the aboveerosion products was deduced firstly based on the principlesof chemical thermodynamics Because the thermodynamicsparameters of different erosion products were affected byenvironmental conditions remarkably the thermodynamicsparameters of different reactions for erosion products atdifferent conditions including dry humid and water en-vironment were calculated theoretically 0e reactions ofdihydrate gypsum and anhydrite and mirabilite and the-nardite can be written as follows (subscript (g) stands for thegaseous state) [19 32 36]

CaSO4(s) + 2H2O(g) CaSO4 middot 2H2O(s) (31)

CaSO4(s) + 2H2O(l) CaSO4 middot 2H2O(s) (32)

Na2SO4(s) + 10H2O(g)Na2SO4 middot 10H2O(s) (33)

Na2SO4(s) + 10H2O(l)Na2SO4 middot 10H2O(s) (34)

0e corresponding thermodynamic parameters of dif-ferent reactions can be determined based on equations(4)sim(13) and Table 4 as listed in Table 5 Moreover theGibbs free energy curves of various erosion products underdifferent conditions as a function of temperature are plottedin Figure 1

Figure 1(a) shows that the theoretical transformationtemperature between dihydrate gypsum and anhydriteunder humid and water conditions is about 91degC that is thedihydrate gypsum can be spontaneously formed when thetemperature is lower than 91degC However the correspondingtemperature under dry condition is about 56degC ie thedihydrate gypsum can be translated into anhydrite when thetemperature exceeds 56degC Figure 1(b) indicates that thetheoretical transformation temperature between mirabiliteand thenardite under dry condition is about 45degC howeverit is about 88degC under humid and water conditions From theabove discussions a conclusion can be drawn that the en-vironment conditions have significant effects on thetransformation of the crystals with combined water0erefore the environmental condition should be consid-ered as an important factor of sulfate attack on cementconcrete during the test process of sulfate attack andspecimensrsquo preparation It is the theoretical transformationtemperature curve based upon which the substances with orwithout crystal water would be converted In Figure 1 theGibbs free energy curves are calculated based on the the-oretical transformation temperature of the substanceswithout crystal water converted into the substances withcrystal water Because the theoretical calculated data do not

consider the activity of substances there exists a differencebetween the theoretical calculated results and the in-sitemeasured data 0e research [41] showed that the solubilityof calcium sulfate at normal temperatures and pressures wasabout 02 and the temperature corresponding to the in-tersection of solubility curves of dihydrate gypsum andanhydrite was 42degC [42] 0e above achievements accordwell with the theoretical calculated data in Figure 1 whichimplies that equation (11) proposed by this study can be usedto determine the mechanism of sulfate attack Because all thesulfate attacks took place in the concrete pore solution thedeterioration mechanism and erosion products of sulfateattack reacting in solution were further investigated

42 Numerical Analysis of Critical Concentration of SulfateIon of Different Erosion Products with TemperatureCalcium hydroxide (CH) in cement and concrete has asignificant effect on the thermodynamic equilibrium ofvarious products so the variation of critical concentration of[OHminus ] and [Ca2+] with temperature is investigated firstlyAssuming the content of hydroxyl ion is generated by CHand dominated by CH saturated solution the correspondingchemical equilibrium based on Hessrsquos law can be expressedas follows [36]Ca(OH)2(s) + 2H+ Ca2++2H2O (1)

H2O H+ + OHminus (2)

⎫⎬

⎭⟹(1) + 2 times(2)⟹

Ca(OH)2(s) Ca2++ 2OHminus

(35)

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the above reaction can bedetermined based on equation (1) written as

lnKΘ

lnKΘ1 + 2 lnK

Θ2 (36)

Assuming the [OHminus ] in the system is generated byCa(OH)2 the concentration of calcium ion is half of thehydroxyl ion 0e concentration of hydroxyl ion withtemperature can be determined based on equation (36) asshown in Figure 2 If the activity of condensed matters is setas 1 the critical concentration of sulfate ion for the for-mation of dihydrate gypsum and anhydrite in equations (15)and (16) can be deduced on the basis of the calcium ionconcentration and thermodynamic equilibrium constantwritten as equations (37) and (38) 0e correspondingcritical concentration curves of sulfate ions for the formationof dihydrate gypsum and anhydrite as a function of tem-perature are plotted in Figure 2

SO2minus41113960 1113961

KΘCaSO4 middot2H2O

a Ca2+[ ]middot cΘ

(37)

SO2minus41113960 1113961

KΘCaSO4

a Ca2+[ ]middot cΘ

(38)

As seen from Figure 2(a) the theoretical concentrationsof [Ca2+] and [OHminus ] corresponding to the thermodynamic

6 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

where KΘeq stands for the thermodynamic equilibriumconstant at standard conditions A B C D and E areconstant respectively

0e main erosion products of sulfate attack on cementconcrete may be gypsum ettringite and sulfate salt [29 30] and[31] and the corresponding reactions can be written as [19 32]

Ca2+(aq) + SO2minus

4(aq) Crystallize

DissolveCaSO4(s) (15)

Ca2+(aq) + SO2minus

4(aq) + 2H2O(l) Crystallize

DissolveCaSO4 middot 2H2O(s) (16)

2Na+(aq) + SO2minus

4(aq) Crystallize

DissolveNa2SO4(s) (17)

2Na+(aq) + SO2minus

4(aq) + 10H2O(l) Crystallize

DissolveNa2SO4 middot 10H2O(s)

(18)

6Ca2++ 2Al3+

+ 3SO2minus4 + 32H2O(l)

Crystallize

Dissolve3CaO middot Al2O3

middot 3CaSO4 middot 32H2O(s)

(19)

CaSiO3 middot CaSO4 middot CaCO3 middot 15H2O(s) + 3H+ Crystallize

Dissolve3Ca2+

+SO2minus4 + HCOminus

3 + H4SiO4 + 14H2O(l)

(20)

where subscripts s aq and l stand for solid state ionic statein solution and liquid respectively

Correspondingly the decomposition reactions of thecement hydration products can be expressed as follows[19 33 34]

Ca(OH)2(s) + 2H+ Ca2++ 2H2O (21)

C3AH6(s) + 12H+ 2Al3++ 3Ca2+

+ 12H2O (22)

C4AH13(s) + 14H+ 2Al3++ 4Ca2+

+ 20H2O (23)

C-S-H(08)(s) + 16H+ 034H2O + 08Ca2++ H4SiO4

(24)

C-S-H(12)(s) + 24H+ 126H2O + 12Ca2++ H4SiO4

(25)

Ca2SiO4(s) + 4H+ 2Ca2++ H4SiO4 (26)

Al(OH)3(s) + 3H+Al3++ 3H2O (27)

Ca6Al2 SO4( 11138573(OH)12 middot 26H2O(s) + 12H+ 2Al3++ 6Ca2+

+ 3SO2minus4 + 38H2O

(28)

0e reference values recommended by researchers[35 36] are listed in Table 1

0e above reaction equations can be formulated asequation (29) and the relationship between the thermo-dynamic equilibrium constant and concentration of dif-ferent substances can be expressed as equation (30)

dD + eE gG + hH (29)

KΘeq

[G]g[H]h

[D]d[E]e (30)

where [D] [E] [G] and [H] mean the concentration ofdifferent substances respectively d e g and h are thestoichiometric numbers respectively

Combining equations (14) and (30) the formationconditions and the critical ion concentration of sulfate ionfor the formation of erosion products can be determinedbased on the thermodynamic parameter of different sub-stances in Table 1

3 Experimental Procedure

31 Raw Materials Portland cement of P O 325 class wasproduced by Pingtang Cement Plant of Hunan and thepolycarboxylic acid series of superplasticizer with solidcontent of 30 provided by Changsha Huangteng ChemicalTechnology Co Ltd was used as a water reducer ISOstandard sand with a fineness modulus of 30 was producedby Xiamen Aisiou Standard Sand Co Ltd Local river sandwith a fineness modulus of 29 continuous grading lime-stone gravel with grains sizes of 5mmsim20mm and tap waterwere used to prepare the specimens Moreover the indus-trial grade sodium sulfate with purity of 99 was purchasedfrom market 0e chemical composition and physicalproperties of cement are listed in Tables 2 and 3

32 Specimen Preparation Experimental Process and TestDevices According to Common Portland Cement and TestMethods for Water Requirement of Normal Consistency SettingTime and Soundness of the Portland Cement [37 38] the pastespecimens of 40mmtimes 40mmtimes 160mmwere cast with a waterto cement ratio of 028 and were used for XRD analysisAccording to Method of Testing CementsndashDetermination ofStrength the mortar specimens of 40mmtimes 40mmtimes 160mmwere cast with a water cement sand ratio of 05 13 by weight[39] According to Standard for Test Method of MechanicalProperties onOrdinary Concrete [40] the concrete specimens of100mmtimes 100mmtimes 300mm were cast and there are threespecimens for each group 0e C20 grade concrete was pre-pared with a weight ratio of cement sand limestone waterwater reducer of 305 831 1102 158 35 All specimens werecured in standard curing pool at a temperature of (20plusmn 1)degCand a relative humidity of (95plusmn 3) 0e specimens weredemolded after 24h and cured in water for 28d at a tem-perature of (20plusmn 1)degC Subsequently the specimens were takenout from water to carry out sulfate attack test

4 Advances in Materials Science and Engineering

0e experimental process of the concrete attacked bysulfate was carried out as follows Firstly the sulfate solu-tions of different concentrations (ie 1 5 10 andsaturated solution) were prepared and various specimenswere immersed into sulfate solution In order to ensure theuniformity of sulfate solution the distance of specimens wasset as no less than 2 cm 0en the sealed solution box wascovered with plastic film to prevent the water from evap-oration Moreover the sulfate solution was changed once aweek Finally the specimens were taken out from sulfatesolution and dried at room temperature of 25degC and arelative humidity of 70 for 3 days when the erosion agesreached the setting time of 2 and 4 months 0e flexuralstrength of the mortar and paste specimens was first testedand then the corresponding compressive strength

0e preparation process of the samples for micro-properties test was as follows Firstly the specimens reachingthe sulfate attack age were cleaned with distilled water andbroken into particles of 2mmsim5mm Secondly the particleswere immersed into alcohol for 24 h to terminate reaction0en the particles were dried at 60degC for 48 h Finally theprocessed particles were placed and stored in a dryer 0eproceeded particles containing surface were selected andtheir cross-sections were sprayed with gold for ESEM-EDSanalysis 0e observed localization of the proceeded particle

for ESEM-EDS analysis was near the outer surface of thecross-sections In order to conduct the XRD analysis thesuperficial region within 2mm from cement sample surfacewas obtained and treated as mentioned above Subsequentlythe proceeded cement paste particles were ground to powderthrough 200 mesh and the scanning step of XRD analysiswas 002deg All samples were prepared and tested according tothe equipment guide

0e main test devices were as follows 0e Quanta-200Environment Scanning Electronic Microscope (ESEM-EDS)produced by FTI Company of Czech Republic with anamplification of 600000 times was used to measure themicrostructure 0e Dmax-2550 X-ray diffraction (XRD)with the angular range of 5degsim120deg and a minimum scanningstep size of 001deg was used to carry out the phase analysis0eYAW-300D produced by Jinan Kesheng Test Equipment CoLtd of China was used to measure the flexural and com-pressive strength of the specimens 0e maximum com-pressive and flexural loads were 300 kN and 10 kNrespectively0e corresponding loading rates of compressiveand flexural strength were (24plusmn 02) kNs and (50plusmn 10)Nsrespectively Moreover the WAW-1000 electrohydraulicservo universal testing machine made by Shanghai Sansi CoLtd of China was used to measure the compressive strengthof the concrete According to Standard for Test Method of

Table 1 0ermodynamic parameters of different substances

Items A B C D ECaSO4 middot 2H2O(s) 162021439e+ 3 257234846e minus 1 minus 891506186e+ 4 minus 587385148e+ 2 534735206e+ 6CaSO4(s) 161807826e+ 3 262044313e minus 1 minus 895853477e+ 4 minus 586632877e+ 2 535893242e+ 6CaSO4(aq) 172034184e+ 3 265734992e minus 1 minus 942553556e+ 4 minus 623563883e+ 2 549729959e+ 6Na2SO4 middot 10H2O(s) 158837182e+ 3 231777424e minus 1 minus 843055786e+ 4 minus 578226172e+ 2 509260164e+ 6Na2SO4(S) 161633032e+ 3 253239679e minus 1 minus 898032147e+ 4 minus 586414685e+ 2 540049408e+ 6AFt(s) Ca6Al2(SO4)3(OH)12middot26H2O minus 667460197e+ 3 minus 104743390 378708023e+ 5 242662966e+ 3 minus 205189885e+ 7Ca(OH)2(s) minus 284930557e+ 2 minus 447108160e minus 2 213801821e+ 4 104205027e+ 2 minus 754252617e+ 5Al(OH)3(s) minus 493752625e+ 2 minus 809001544e minus 2 297138788e+ 4 177903516e+ 2 minus 126765911e+ 6NaAlO2(aq) 704197406e+ 2 111341739e minus 1 minus 474872291e+ 4 minus 253129969e+ 2 218693139e+ 6AlO2

minus(aq) minus 178049448e+ 2 minus 268902416e minus 2 186721225e+ 3 668330936e+ 1 minus 750442968e+ 5

H2O(aq) minus 701957319 minus 0112739992 36168254 25360128 minus 242327306C-S-H (08)(s) Ca08SiO28middot154H2O minus 251027448e+ 2 minus 365449659e minus 2 155601183e+ 4 921931312e+ 1 minus 652576525e+ 5C-S-H(12)(s) Ca12SiO32middot206H2O minus 372034132e+ 2 minus 541579845e minus 2 239845213e+ 4 136559314e+ 2 minus 966181083e+ 5C-S-H(16)(s) Ca16SiO36middot258H2O minus 491723254e+ 2 minus 717628896e minus 2 325370146e+ 4 180399649e+ 2 minus 127964571e+ 6C2AH8(s) Ca2Al2O5middot8H2O minus 173467190e+ 3 minus 238046203e minus 1 110411145e+ 5 623227246e+ 2 minus 418576035e+ 6C3AH6(s) Ca3Al2(OH)12 minus 178581145e+ 3 minus 288036362e minus 1 119706499e+ 5 647583573e+ 2 minus 461172146e+ 6CaSiO3CaSO4CaCO3middot15H2O(s) minus 276283310e+ 3 minus 432736383e minus 1 149848596e+ 5 100850740e+ 3 minus 852601737e+ 6α-Ca2SiO4(s) minus 526934036e+ 2 minus 867703839e minus 2 389768377e+ 4 192577860e+ 2 minus 136961546e+ 6CaSiO3(s) minus 265453934e+ 2 minus 441966587e minus 2 177395216e+ 4 972523635e+ 1 minus 665653039e+ 5

Table 2 Chemical composition of cement ()

CaO SiO2 Al2O3 P2O5 Fe2O3 MgO K2O Na2O Loss6561 1989 473 288 145 201 114 021 198

Table 3 Physical properties of cement

Average diameter(μm)

Specific surface area(m2kg)

Bulk density(gcm3)

Initial setting time(min)

Final setting time(min)

Water requirement for standardconsistency ()

346 3452 135 172 251 28

Advances in Materials Science and Engineering 5

Mechanical Properties on Ordinary Concrete [40] the cor-responding loading rate was in the range of 03MPassim055MPas

4 Results and Discussions

41 Assessment of Erosion Products of Sulfate Attack on Ce-ment Concrete 0e main erosion products of sulfate attackon cement concrete may be gypsum ettringite sulfate saltand so on0erefore the formation spontaneity of the aboveerosion products was deduced firstly based on the principlesof chemical thermodynamics Because the thermodynamicsparameters of different erosion products were affected byenvironmental conditions remarkably the thermodynamicsparameters of different reactions for erosion products atdifferent conditions including dry humid and water en-vironment were calculated theoretically 0e reactions ofdihydrate gypsum and anhydrite and mirabilite and the-nardite can be written as follows (subscript (g) stands for thegaseous state) [19 32 36]

CaSO4(s) + 2H2O(g) CaSO4 middot 2H2O(s) (31)

CaSO4(s) + 2H2O(l) CaSO4 middot 2H2O(s) (32)

Na2SO4(s) + 10H2O(g)Na2SO4 middot 10H2O(s) (33)

Na2SO4(s) + 10H2O(l)Na2SO4 middot 10H2O(s) (34)

0e corresponding thermodynamic parameters of dif-ferent reactions can be determined based on equations(4)sim(13) and Table 4 as listed in Table 5 Moreover theGibbs free energy curves of various erosion products underdifferent conditions as a function of temperature are plottedin Figure 1

Figure 1(a) shows that the theoretical transformationtemperature between dihydrate gypsum and anhydriteunder humid and water conditions is about 91degC that is thedihydrate gypsum can be spontaneously formed when thetemperature is lower than 91degC However the correspondingtemperature under dry condition is about 56degC ie thedihydrate gypsum can be translated into anhydrite when thetemperature exceeds 56degC Figure 1(b) indicates that thetheoretical transformation temperature between mirabiliteand thenardite under dry condition is about 45degC howeverit is about 88degC under humid and water conditions From theabove discussions a conclusion can be drawn that the en-vironment conditions have significant effects on thetransformation of the crystals with combined water0erefore the environmental condition should be consid-ered as an important factor of sulfate attack on cementconcrete during the test process of sulfate attack andspecimensrsquo preparation It is the theoretical transformationtemperature curve based upon which the substances with orwithout crystal water would be converted In Figure 1 theGibbs free energy curves are calculated based on the the-oretical transformation temperature of the substanceswithout crystal water converted into the substances withcrystal water Because the theoretical calculated data do not

consider the activity of substances there exists a differencebetween the theoretical calculated results and the in-sitemeasured data 0e research [41] showed that the solubilityof calcium sulfate at normal temperatures and pressures wasabout 02 and the temperature corresponding to the in-tersection of solubility curves of dihydrate gypsum andanhydrite was 42degC [42] 0e above achievements accordwell with the theoretical calculated data in Figure 1 whichimplies that equation (11) proposed by this study can be usedto determine the mechanism of sulfate attack Because all thesulfate attacks took place in the concrete pore solution thedeterioration mechanism and erosion products of sulfateattack reacting in solution were further investigated

42 Numerical Analysis of Critical Concentration of SulfateIon of Different Erosion Products with TemperatureCalcium hydroxide (CH) in cement and concrete has asignificant effect on the thermodynamic equilibrium ofvarious products so the variation of critical concentration of[OHminus ] and [Ca2+] with temperature is investigated firstlyAssuming the content of hydroxyl ion is generated by CHand dominated by CH saturated solution the correspondingchemical equilibrium based on Hessrsquos law can be expressedas follows [36]Ca(OH)2(s) + 2H+ Ca2++2H2O (1)

H2O H+ + OHminus (2)

⎫⎬

⎭⟹(1) + 2 times(2)⟹

Ca(OH)2(s) Ca2++ 2OHminus

(35)

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the above reaction can bedetermined based on equation (1) written as

lnKΘ

lnKΘ1 + 2 lnK

Θ2 (36)

Assuming the [OHminus ] in the system is generated byCa(OH)2 the concentration of calcium ion is half of thehydroxyl ion 0e concentration of hydroxyl ion withtemperature can be determined based on equation (36) asshown in Figure 2 If the activity of condensed matters is setas 1 the critical concentration of sulfate ion for the for-mation of dihydrate gypsum and anhydrite in equations (15)and (16) can be deduced on the basis of the calcium ionconcentration and thermodynamic equilibrium constantwritten as equations (37) and (38) 0e correspondingcritical concentration curves of sulfate ions for the formationof dihydrate gypsum and anhydrite as a function of tem-perature are plotted in Figure 2

SO2minus41113960 1113961

KΘCaSO4 middot2H2O

a Ca2+[ ]middot cΘ

(37)

SO2minus41113960 1113961

KΘCaSO4

a Ca2+[ ]middot cΘ

(38)

As seen from Figure 2(a) the theoretical concentrationsof [Ca2+] and [OHminus ] corresponding to the thermodynamic

6 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

0e experimental process of the concrete attacked bysulfate was carried out as follows Firstly the sulfate solu-tions of different concentrations (ie 1 5 10 andsaturated solution) were prepared and various specimenswere immersed into sulfate solution In order to ensure theuniformity of sulfate solution the distance of specimens wasset as no less than 2 cm 0en the sealed solution box wascovered with plastic film to prevent the water from evap-oration Moreover the sulfate solution was changed once aweek Finally the specimens were taken out from sulfatesolution and dried at room temperature of 25degC and arelative humidity of 70 for 3 days when the erosion agesreached the setting time of 2 and 4 months 0e flexuralstrength of the mortar and paste specimens was first testedand then the corresponding compressive strength

0e preparation process of the samples for micro-properties test was as follows Firstly the specimens reachingthe sulfate attack age were cleaned with distilled water andbroken into particles of 2mmsim5mm Secondly the particleswere immersed into alcohol for 24 h to terminate reaction0en the particles were dried at 60degC for 48 h Finally theprocessed particles were placed and stored in a dryer 0eproceeded particles containing surface were selected andtheir cross-sections were sprayed with gold for ESEM-EDSanalysis 0e observed localization of the proceeded particle

for ESEM-EDS analysis was near the outer surface of thecross-sections In order to conduct the XRD analysis thesuperficial region within 2mm from cement sample surfacewas obtained and treated as mentioned above Subsequentlythe proceeded cement paste particles were ground to powderthrough 200 mesh and the scanning step of XRD analysiswas 002deg All samples were prepared and tested according tothe equipment guide

0e main test devices were as follows 0e Quanta-200Environment Scanning Electronic Microscope (ESEM-EDS)produced by FTI Company of Czech Republic with anamplification of 600000 times was used to measure themicrostructure 0e Dmax-2550 X-ray diffraction (XRD)with the angular range of 5degsim120deg and a minimum scanningstep size of 001deg was used to carry out the phase analysis0eYAW-300D produced by Jinan Kesheng Test Equipment CoLtd of China was used to measure the flexural and com-pressive strength of the specimens 0e maximum com-pressive and flexural loads were 300 kN and 10 kNrespectively0e corresponding loading rates of compressiveand flexural strength were (24plusmn 02) kNs and (50plusmn 10)Nsrespectively Moreover the WAW-1000 electrohydraulicservo universal testing machine made by Shanghai Sansi CoLtd of China was used to measure the compressive strengthof the concrete According to Standard for Test Method of

Table 1 0ermodynamic parameters of different substances

Items A B C D ECaSO4 middot 2H2O(s) 162021439e+ 3 257234846e minus 1 minus 891506186e+ 4 minus 587385148e+ 2 534735206e+ 6CaSO4(s) 161807826e+ 3 262044313e minus 1 minus 895853477e+ 4 minus 586632877e+ 2 535893242e+ 6CaSO4(aq) 172034184e+ 3 265734992e minus 1 minus 942553556e+ 4 minus 623563883e+ 2 549729959e+ 6Na2SO4 middot 10H2O(s) 158837182e+ 3 231777424e minus 1 minus 843055786e+ 4 minus 578226172e+ 2 509260164e+ 6Na2SO4(S) 161633032e+ 3 253239679e minus 1 minus 898032147e+ 4 minus 586414685e+ 2 540049408e+ 6AFt(s) Ca6Al2(SO4)3(OH)12middot26H2O minus 667460197e+ 3 minus 104743390 378708023e+ 5 242662966e+ 3 minus 205189885e+ 7Ca(OH)2(s) minus 284930557e+ 2 minus 447108160e minus 2 213801821e+ 4 104205027e+ 2 minus 754252617e+ 5Al(OH)3(s) minus 493752625e+ 2 minus 809001544e minus 2 297138788e+ 4 177903516e+ 2 minus 126765911e+ 6NaAlO2(aq) 704197406e+ 2 111341739e minus 1 minus 474872291e+ 4 minus 253129969e+ 2 218693139e+ 6AlO2

minus(aq) minus 178049448e+ 2 minus 268902416e minus 2 186721225e+ 3 668330936e+ 1 minus 750442968e+ 5

H2O(aq) minus 701957319 minus 0112739992 36168254 25360128 minus 242327306C-S-H (08)(s) Ca08SiO28middot154H2O minus 251027448e+ 2 minus 365449659e minus 2 155601183e+ 4 921931312e+ 1 minus 652576525e+ 5C-S-H(12)(s) Ca12SiO32middot206H2O minus 372034132e+ 2 minus 541579845e minus 2 239845213e+ 4 136559314e+ 2 minus 966181083e+ 5C-S-H(16)(s) Ca16SiO36middot258H2O minus 491723254e+ 2 minus 717628896e minus 2 325370146e+ 4 180399649e+ 2 minus 127964571e+ 6C2AH8(s) Ca2Al2O5middot8H2O minus 173467190e+ 3 minus 238046203e minus 1 110411145e+ 5 623227246e+ 2 minus 418576035e+ 6C3AH6(s) Ca3Al2(OH)12 minus 178581145e+ 3 minus 288036362e minus 1 119706499e+ 5 647583573e+ 2 minus 461172146e+ 6CaSiO3CaSO4CaCO3middot15H2O(s) minus 276283310e+ 3 minus 432736383e minus 1 149848596e+ 5 100850740e+ 3 minus 852601737e+ 6α-Ca2SiO4(s) minus 526934036e+ 2 minus 867703839e minus 2 389768377e+ 4 192577860e+ 2 minus 136961546e+ 6CaSiO3(s) minus 265453934e+ 2 minus 441966587e minus 2 177395216e+ 4 972523635e+ 1 minus 665653039e+ 5

Table 2 Chemical composition of cement ()

CaO SiO2 Al2O3 P2O5 Fe2O3 MgO K2O Na2O Loss6561 1989 473 288 145 201 114 021 198

Table 3 Physical properties of cement

Average diameter(μm)

Specific surface area(m2kg)

Bulk density(gcm3)

Initial setting time(min)

Final setting time(min)

Water requirement for standardconsistency ()

346 3452 135 172 251 28

Advances in Materials Science and Engineering 5

Mechanical Properties on Ordinary Concrete [40] the cor-responding loading rate was in the range of 03MPassim055MPas

4 Results and Discussions

41 Assessment of Erosion Products of Sulfate Attack on Ce-ment Concrete 0e main erosion products of sulfate attackon cement concrete may be gypsum ettringite sulfate saltand so on0erefore the formation spontaneity of the aboveerosion products was deduced firstly based on the principlesof chemical thermodynamics Because the thermodynamicsparameters of different erosion products were affected byenvironmental conditions remarkably the thermodynamicsparameters of different reactions for erosion products atdifferent conditions including dry humid and water en-vironment were calculated theoretically 0e reactions ofdihydrate gypsum and anhydrite and mirabilite and the-nardite can be written as follows (subscript (g) stands for thegaseous state) [19 32 36]

CaSO4(s) + 2H2O(g) CaSO4 middot 2H2O(s) (31)

CaSO4(s) + 2H2O(l) CaSO4 middot 2H2O(s) (32)

Na2SO4(s) + 10H2O(g)Na2SO4 middot 10H2O(s) (33)

Na2SO4(s) + 10H2O(l)Na2SO4 middot 10H2O(s) (34)

0e corresponding thermodynamic parameters of dif-ferent reactions can be determined based on equations(4)sim(13) and Table 4 as listed in Table 5 Moreover theGibbs free energy curves of various erosion products underdifferent conditions as a function of temperature are plottedin Figure 1

Figure 1(a) shows that the theoretical transformationtemperature between dihydrate gypsum and anhydriteunder humid and water conditions is about 91degC that is thedihydrate gypsum can be spontaneously formed when thetemperature is lower than 91degC However the correspondingtemperature under dry condition is about 56degC ie thedihydrate gypsum can be translated into anhydrite when thetemperature exceeds 56degC Figure 1(b) indicates that thetheoretical transformation temperature between mirabiliteand thenardite under dry condition is about 45degC howeverit is about 88degC under humid and water conditions From theabove discussions a conclusion can be drawn that the en-vironment conditions have significant effects on thetransformation of the crystals with combined water0erefore the environmental condition should be consid-ered as an important factor of sulfate attack on cementconcrete during the test process of sulfate attack andspecimensrsquo preparation It is the theoretical transformationtemperature curve based upon which the substances with orwithout crystal water would be converted In Figure 1 theGibbs free energy curves are calculated based on the the-oretical transformation temperature of the substanceswithout crystal water converted into the substances withcrystal water Because the theoretical calculated data do not

consider the activity of substances there exists a differencebetween the theoretical calculated results and the in-sitemeasured data 0e research [41] showed that the solubilityof calcium sulfate at normal temperatures and pressures wasabout 02 and the temperature corresponding to the in-tersection of solubility curves of dihydrate gypsum andanhydrite was 42degC [42] 0e above achievements accordwell with the theoretical calculated data in Figure 1 whichimplies that equation (11) proposed by this study can be usedto determine the mechanism of sulfate attack Because all thesulfate attacks took place in the concrete pore solution thedeterioration mechanism and erosion products of sulfateattack reacting in solution were further investigated

42 Numerical Analysis of Critical Concentration of SulfateIon of Different Erosion Products with TemperatureCalcium hydroxide (CH) in cement and concrete has asignificant effect on the thermodynamic equilibrium ofvarious products so the variation of critical concentration of[OHminus ] and [Ca2+] with temperature is investigated firstlyAssuming the content of hydroxyl ion is generated by CHand dominated by CH saturated solution the correspondingchemical equilibrium based on Hessrsquos law can be expressedas follows [36]Ca(OH)2(s) + 2H+ Ca2++2H2O (1)

H2O H+ + OHminus (2)

⎫⎬

⎭⟹(1) + 2 times(2)⟹

Ca(OH)2(s) Ca2++ 2OHminus

(35)

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the above reaction can bedetermined based on equation (1) written as

lnKΘ

lnKΘ1 + 2 lnK

Θ2 (36)

Assuming the [OHminus ] in the system is generated byCa(OH)2 the concentration of calcium ion is half of thehydroxyl ion 0e concentration of hydroxyl ion withtemperature can be determined based on equation (36) asshown in Figure 2 If the activity of condensed matters is setas 1 the critical concentration of sulfate ion for the for-mation of dihydrate gypsum and anhydrite in equations (15)and (16) can be deduced on the basis of the calcium ionconcentration and thermodynamic equilibrium constantwritten as equations (37) and (38) 0e correspondingcritical concentration curves of sulfate ions for the formationof dihydrate gypsum and anhydrite as a function of tem-perature are plotted in Figure 2

SO2minus41113960 1113961

KΘCaSO4 middot2H2O

a Ca2+[ ]middot cΘ

(37)

SO2minus41113960 1113961

KΘCaSO4

a Ca2+[ ]middot cΘ

(38)

As seen from Figure 2(a) the theoretical concentrationsof [Ca2+] and [OHminus ] corresponding to the thermodynamic

6 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

Mechanical Properties on Ordinary Concrete [40] the cor-responding loading rate was in the range of 03MPassim055MPas

4 Results and Discussions

41 Assessment of Erosion Products of Sulfate Attack on Ce-ment Concrete 0e main erosion products of sulfate attackon cement concrete may be gypsum ettringite sulfate saltand so on0erefore the formation spontaneity of the aboveerosion products was deduced firstly based on the principlesof chemical thermodynamics Because the thermodynamicsparameters of different erosion products were affected byenvironmental conditions remarkably the thermodynamicsparameters of different reactions for erosion products atdifferent conditions including dry humid and water en-vironment were calculated theoretically 0e reactions ofdihydrate gypsum and anhydrite and mirabilite and the-nardite can be written as follows (subscript (g) stands for thegaseous state) [19 32 36]

CaSO4(s) + 2H2O(g) CaSO4 middot 2H2O(s) (31)

CaSO4(s) + 2H2O(l) CaSO4 middot 2H2O(s) (32)

Na2SO4(s) + 10H2O(g)Na2SO4 middot 10H2O(s) (33)

Na2SO4(s) + 10H2O(l)Na2SO4 middot 10H2O(s) (34)

0e corresponding thermodynamic parameters of dif-ferent reactions can be determined based on equations(4)sim(13) and Table 4 as listed in Table 5 Moreover theGibbs free energy curves of various erosion products underdifferent conditions as a function of temperature are plottedin Figure 1

Figure 1(a) shows that the theoretical transformationtemperature between dihydrate gypsum and anhydriteunder humid and water conditions is about 91degC that is thedihydrate gypsum can be spontaneously formed when thetemperature is lower than 91degC However the correspondingtemperature under dry condition is about 56degC ie thedihydrate gypsum can be translated into anhydrite when thetemperature exceeds 56degC Figure 1(b) indicates that thetheoretical transformation temperature between mirabiliteand thenardite under dry condition is about 45degC howeverit is about 88degC under humid and water conditions From theabove discussions a conclusion can be drawn that the en-vironment conditions have significant effects on thetransformation of the crystals with combined water0erefore the environmental condition should be consid-ered as an important factor of sulfate attack on cementconcrete during the test process of sulfate attack andspecimensrsquo preparation It is the theoretical transformationtemperature curve based upon which the substances with orwithout crystal water would be converted In Figure 1 theGibbs free energy curves are calculated based on the the-oretical transformation temperature of the substanceswithout crystal water converted into the substances withcrystal water Because the theoretical calculated data do not

consider the activity of substances there exists a differencebetween the theoretical calculated results and the in-sitemeasured data 0e research [41] showed that the solubilityof calcium sulfate at normal temperatures and pressures wasabout 02 and the temperature corresponding to the in-tersection of solubility curves of dihydrate gypsum andanhydrite was 42degC [42] 0e above achievements accordwell with the theoretical calculated data in Figure 1 whichimplies that equation (11) proposed by this study can be usedto determine the mechanism of sulfate attack Because all thesulfate attacks took place in the concrete pore solution thedeterioration mechanism and erosion products of sulfateattack reacting in solution were further investigated

42 Numerical Analysis of Critical Concentration of SulfateIon of Different Erosion Products with TemperatureCalcium hydroxide (CH) in cement and concrete has asignificant effect on the thermodynamic equilibrium ofvarious products so the variation of critical concentration of[OHminus ] and [Ca2+] with temperature is investigated firstlyAssuming the content of hydroxyl ion is generated by CHand dominated by CH saturated solution the correspondingchemical equilibrium based on Hessrsquos law can be expressedas follows [36]Ca(OH)2(s) + 2H+ Ca2++2H2O (1)

H2O H+ + OHminus (2)

⎫⎬

⎭⟹(1) + 2 times(2)⟹

Ca(OH)2(s) Ca2++ 2OHminus

(35)

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the above reaction can bedetermined based on equation (1) written as

lnKΘ

lnKΘ1 + 2 lnK

Θ2 (36)

Assuming the [OHminus ] in the system is generated byCa(OH)2 the concentration of calcium ion is half of thehydroxyl ion 0e concentration of hydroxyl ion withtemperature can be determined based on equation (36) asshown in Figure 2 If the activity of condensed matters is setas 1 the critical concentration of sulfate ion for the for-mation of dihydrate gypsum and anhydrite in equations (15)and (16) can be deduced on the basis of the calcium ionconcentration and thermodynamic equilibrium constantwritten as equations (37) and (38) 0e correspondingcritical concentration curves of sulfate ions for the formationof dihydrate gypsum and anhydrite as a function of tem-perature are plotted in Figure 2

SO2minus41113960 1113961

KΘCaSO4 middot2H2O

a Ca2+[ ]middot cΘ

(37)

SO2minus41113960 1113961

KΘCaSO4

a Ca2+[ ]middot cΘ

(38)

As seen from Figure 2(a) the theoretical concentrationsof [Ca2+] and [OHminus ] corresponding to the thermodynamic

6 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

equilibrium state of the system decrease with increase of thetemperature which accords well with the variation rule ofactual solubility of CH with temperature Figure 2(b) showsthat there exists an intersection of the critical sulfate ionconcentration curves of CaSO4 and CaSO4 middot 2H2O at about42degC 0is indicates that CaSO4 middot 2H2O can be preferentiallygenerated when the reaction temperature is less than 42degC

Conversely it will be CaSO4 Figure 2 also reveals that thecritical concentration of sulfate ion for the formation ofCaSO4 middot 2H2O is more than 23times10minus 3molL which is inaccordance with the solubility of calcium sulfate (about 02wt) [41] under normal temperature and pressure 0echange law of sulfate ion concentration with temperaturealso accords well with the actual solubility of gypsum which

Table 4 0ermodynamic parameters for different products

Items ΔfHΘ298 (Jmol) SΘ298 (JmolmiddotKminus 1)Cp (JKmiddotmolminus 1)

A1 A2 A3 A4 A5

CaSO4(s) minus 1434108 105228 70208 98742 0 0 0CaSO4 middot 2H2O(s) minus 2022629 194138 91379 317984 0 0 0H2O(g) minus 241814 188724 29999 10711 0335 0 0H2O(l) minus 285840 6994 331799 709205 111715 0 0Na2SO4(s) minus 1387205 14962 82299 154348 0 0 0Na2SO4 middot 10H2O(s) minus 4327791 5919 57446 0 0 0 0Ca(OH)4(s) minus 982611 83387 105269 11294 minus 18954 0 0NaOH(s) minus 428023 64434 71756 minus 110876 0 235768 0

Table 5 0ermodynamic parameters of different reactions

Items ΔHΘ298(Jmol)

ΔSΘ298(JmolmiddotKminus 1)

ΔGΘ298(kJmol) ΔA1 ΔA2 ΔA3 ΔA4 ΔA5

ΔCp298(JK

molminus 1)ΔA6 ΔA6prime ΔA6Prime

Equation(31) minus 104893 288538 minus 18908676 minus 38827 19782 minus 067 0 0 19369 minus 10233099 878375 minus 126665

Equation(32) minus 16841 minus 5097 minus 165194 minus 451888 77401 minus 22343 0 0 minus 47283 minus 143091 minus 216018 170829

Equation(33) minus 522446 minus 144496 minus 918479 192171 minus 261458 minus 335 0 0 110484 minus 5692279 265592 minus 246375

Equation(34) minus 82168 minus 25712 minus 556424 160362 863553 minus 111715 0 0 minus 222776 minus 1291187 113664 minus 976278

0 20 40 60 80 100ndash20000

ndash16000

ndash12000

ndash8000

ndash4000

0

4000

Temperature (degC)

Water and high humidity conditionDry condition

∆G k

Jmiddotmol

ndash1

(a)

0 20 40 60 80 100ndash120000

ndash100000

ndash80000

ndash60000

ndash40000

ndash20000

0

20000

Temperature (degC)

Water and high humidityDry condition

∆G k

Jmiddotmol

ndash1

(b)

Figure 1 Gibbs free energy of various erosion products under different conditions versus temperature (a) Dihydrate gypsum andanhydrite (b) Mirabilite and thenardite

Advances in Materials Science and Engineering 7

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

indicates that the principles of chemical thermodynamicscan be used to determine the formation condition andcritical concentration of various ions Moreover the loga-rithm values of the thermodynamic equilibrium constant inFigure 2 are all negative numbers which implies that theGibbs free energy of the reactions is less than zero0ereforethe corresponding erosion reaction is spontaneous

0e variations of critical concentration of sulfate ion forthe formation of AFt AFm and thaumasite (CSCSH) as afunction of temperature were also investigated Assumingthe [Al3+] generated by the decomposition of C3AH6 andC4AH13 can react with [Ca2+] and [SO2minus

4 ] and the con-centration of [Ca2+] and [OHminus ] is also dominated byCa(OH)2 solution the chemical equations of AFt AFm andCSCSH can be expressed as follows [36]

CSCSH(s) + 3H+ HCOminus3 + 3Ca2+ + SO2minus

4 + H4SiO4 + 14H2O (1)

CaCO3(s) + H+ HCOminus3 + Ca2+ (2)

CaSO4 middot 2H2O(s) + H+ Ca2+ + SO2minus4 (3)

CaSiO3(s) + 2H+ + H2O(l) Ca2+ + H4SiO4(s) (4)

⎫⎪⎪⎪⎪⎪⎬

⎪⎪⎪⎪⎪⎭

⟹

(1) minus (2) minus (3) minus (4)⟹CSCSH(s) CaCO3(s) + CaSO4 middot 2H2O(s) + CaSiO3(s) + 13H2O(l)

(39)

C3AH6(s) + 12H+ 2Al3+ + 3Ca2+ + 12H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹C3AH6(s) 2Al3++ 3Ca2+

+ 12OHminus

(40)

AFt + 12H+ 2Al3+(aq) + 6Ca2+

(aq) + 3SO2minus4(aq) + 38H2O (1)

H2O H+ + OHminus (2)1113897⟹

(1) + 12 times(2)⟹AFt(s) 2Al3++ 6Ca2+

+ 3SO2minus4 + 12OHminus

+ 26H2O(41)

0 10 20 30 40 50 60 70 80000

001

002

003

Temperature (degC)

[SO

42ndash] (

mol

L)

ndash60

ndash58

ndash56

ndash54

ndash52

ndash50

[Ca2+][OHndash]lgKeq

lgK e

q(a)

[SO

42ndash] (

mol

L)

lgK e

q

0 10 20 30 40 50 60 70 800000

0001

0002

0003

0004

0005

Temperature (degC)

ndash52

ndash50

ndash48

ndash46

ndash44

ndash42

[SO42ndash] of dihydrate gypsum

[SO42ndash] of anhydrite

lgKeq of dihydrate gypsum lgKeq of anhydrite

(b)

Figure 2 Curves of critical sulfate ion concentration and thermodynamic equilibrium constant with temperature (a) Ca(OH)2 (b) CaSO4and CaSO4 middot 2H2O

8 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

0e corresponding thermodynamic equilibrium con-stant at standard conditions of the reaction can be deter-mined based on equation (1) written as follows

KΘCSCSH

a HCOminus3[ ] middot a3

Ca2+[ ]middot a12

SO2minus4[ ]

middot a H4SiO4[ ] middot a14H2O[ ]

a[CSCSH]

middot a3H+[ ]

(42)

KΘC3AH6

middot KΘH2O1113872 1113873

12

a2Al3+[ ]

middot a3Ca2+[ ]

middot a12OHminus[ ]

a C3AH6[ ](43)

KΘAFt middot K

ΘH2O1113872 1113873

12

a2Al3+[ ]

middot a6Ca2+[ ]

middot a26H2O[ ]

middot a3SO2minus

4[ ]middot a12

OHminus[ ]

a[AFt]

(44)

Assuming the activity of condensed matter equals 1 thecorresponding activity of the sulfate ion for different erosionproducts can be written as follows

a SO2minus4[ ]

KΘCSCSH

a Ca2+[ ](45)

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘC3AH6

3

11139741113972

(46)

Simultaneously the corresponding activity of the criticalsulfate ion for AFt generated by C4AH13 can be representedas follows

a SO2minus4[ ]

KΘAFt middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2middot a2

Ca2+[ ]

3

11139741113972

(47)

If the erosion product of sulfate attack on cement andconcrete is AFm which was generated by [Al3+] decomposedby C3AH6 and C4AH13 the corresponding activity of thesulfate ion can be written as follows

a SO2minus4[ ]

KΘAFmKΘC3AH6

middot a Ca2+[ ](48)

a SO2minus4[ ]

KΘAFm middot a2OHminus[ ]

KΘC4AH13middot KΘH2O1113872 1113873

2 (49)

0e critical concentrations of sulfate ion for differenterosion products calculated by equations (45)sim(49) as afunction of temperature are plotted in Figure 3

As seen from Figure 3(a) the upper limit temperaturefor CSCSH to exist steadily is 44degC Once the systemtemperature is larger than this temperature the CSCSHmay be decomposed Simultaneously the critical con-centration of sulfate ion for the formation of CSCSH isabout 00023molL 0e above results change the tradi-tional viewpoint that CSCSH can only exist in low tem-perature [24] and provide theoretical support for theformation of CSCSH under room temperature Figure 3(b)

shows that temperature and types of calcium aluminate hy-drate have significant effects on the critical concentration ofsulfate ion for the formation of AFt and AFm Compared withC4AH13 the C3AH6 reacts more easily with sulfate ion atroom temperature and the corresponding critical concen-tration of sulfate ion is also lower Figure 3 also reveals that thecritical concentration of sulfate ion for the formation of AFtand AFm is lower than that of the gypsum which implies thatthe AFt and AFm can be generated preferentially with betterthermostability [43] It is also the reason that the earlystrength of Portland cement can be enhanced by addinggypsum as well as the improvement in the finial and initialsetting time

Integrating equations (46) and (48) the critical con-centration of sulfate ion for the transformation between AFtand AFm generated by C3AH6 can be expressed as equation(50) 0e theoretical thermodynamic equilibrium constantand sulfate ion concentration as a function of temperatureare plotted in Figure 4

a SO2minus4[ ]

1a Ca2+[ ]

KΘAFtKΘAFm

1113971

(50)

Figure 4(a) shows that the thermodynamic equilib-rium constant of the AFt and AFm increases with tem-perature and that of the AFm is larger 0e criticalconcentration of sulfate ion corresponding to the trans-formation between them at room temperature is very lowbut its variation becomes more significant when thetemperature is higher than 40degC In general the ther-modynamic equilibrium constant of the AFt is less thanthat of the AFm However their thermodynamic equi-librium constants show the cubic and linear functionrelationship with the concentration of sulfate ion re-spectively Hence the AFm can be preferentially gener-ated under a low concentration of sulfate ion 0esubsequent results are the decrease of sulfate ion con-centration and the destruction of the thermodynamicequilibrium state which may result in the decompositionof the AFt 0erefore the AFt and AFm can transformeach other and reach a new thermodynamic equilibriumstate under a certain sulfate ion concentration It may berelated to the thermodynamic stability of the AFt asshown in Figure 4(b) It can be seen from Figure 4(b)that the thermodynamic stability of the AFt is about97degC that is the AFt may be decomposed when thesystem temperature is higher than the critical temper-ature However Lawrence [12] pointed out that themeasured decomposition temperature of the AFt wasabout 60degCsim70degC 0e difference may be due to the factthat theoretical results do not consider the effect ofactivity of various products

Some researches [44 45] revealed the relationship ofthenardite mirabilite and solution expressed as follows

y 1 minus 2304x minus 14618x2 (51)

x atr middot exp(b middot T) (52)

Advances in Materials Science and Engineering 9

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

where T is the temperature y is the activity of liquid waterand x stands for the mole fraction of sodium sulfate solutionatr and b are the fitted constants respectively and therecommended values are listed in Table 6

0e crystal transition point of Na2SO4middot10H2O andNa2SO4 can be determined based on equations (51) and (52)and the corresponding phase spectrum and solubility ofsodium sulfate are plotted in Figure 5

As seen from Figure 5 the crystal transition temperaturebetween Na2SO4 and Na2SO4middot10H2O is about 324degC that isthe Na2SO4 may be crystallized from saturated solutionwhen the temperature is higher than 324degC Correspond-ingly the Na2SO4middot10H2O is the thermodynamical stablephase under high humidity and temperature lower than

324degC Conversely the Na2SO4 salt may be crystallizedunder a lower temperature and humidity conditions Al-though the solubility of sodium sulfate increases with theincrease of the temperature it gradually decreases when thetemperature exceeds 324degC as shown in Figure 5 A con-clusion can be drawn from the above discussion that thetypes and solubility of sodium sulfate are closely related tothe temperature and relative humidity 0erefore the typesof erosion products in cement and concrete attacked bysulfate ie Na2SO4 and Na2SO4middot10H2O are related to thereaction conditions that is they can transform each otherand even coexist at a certain condition0e above theoreticaldeduction changes the traditional perception of the existenceof single salt crystals of Na2SO4 and Na2SO4middot10H2O

0 20 40 60 80ndash3

ndash2

ndash1

0

lgK e

q

Temperature (degC)

000

005

010

lgKeq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

[SO

42ndash] (

mol

L)

0 20 40 60 80

000000

000002

000004

000006

000008

000010

Temperature (degC)

C4AH13-AFtC3AH6-AFt

C4AH13-AFmC3AH6-AFm

(b)

Figure 3 Variation curves of critical ions for different erosion products in system versus temperature (a)CSCSH (b) AFt and AFm

0 20 40 60 800E + 00

1E ndash 41

2E ndash 41

3E ndash 41

10E ndash 29

20E ndash 29

30E ndash 29

40E ndash 29

50E ndash 29

60E ndash 29

K eq ndash

Temperature (degC)

00E + 00

40E ndash 05

80E ndash 05

12E ndash 04

AFt-KeqAFm-Keq[SO4

2ndash]

[SO

42ndash] (

mol

L)

(a)

lgK

ndash

0 20 40 60 80 100ndash12

ndash10

ndash8

ndash6

ndash4

ndash2

0

lgK of AFt and AFm

Temperature (degC)

(b)

Figure 4 Curves of thermodynamic equilibrium constant and critical concentration of sulfate ion versus temperature (a) 0ermodynamicequilibrium constant and sulfate ion concentration (b) 0ermostability of the AFt and AFm

10 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

Moreover the critical concentration of sulfate ion for theformation of different erosion products can be determinedbased on the solubility curve in Figure 5 which providestheoretical support for the types and existence conditions oferosion products in cement

To sum up the generation of erosion products in cementis related to temperature relative humidity and sulfate ionconcentration and these factors affect each other Amongthem the [SO2minus

4 ] is one of the most important factors 0edihydrate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concentration ismore than 23times10minus 3molL However the correspondingformation concentration of anhydrite is more than42times10minus 3molL and the temperature is larger than 42degC0e theoretical thermodynamic stable temperature of AFt isabout 97degC and the critical concentration of sulfate ion forthe AFt is no less than 28times10minus 3molL 0e mirabilite andthenardite may be generated when the concentration ofsulfate ion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperature is lessthan 324degC Moreover the thaumasite can be formed whenthe temperature is less than 44degC and [SO2minus

4 ] concentration ismore than 00023molL 0e above theoretical results ac-cord well with the recommended value proposed by someresearchers [46ndash48] however because the theoretical resultsdo not consider the effect of activity of various products andthe coupled effects of ions in system it is a little bit differentfrom that (less than 20degC) reported by Blanco et al [24] 0eabove results imply that the erosion products in cementattacked by different sulfate conditions can be determined bythe principles of chemical thermodynamics which provides

theoretical support for the determination of erosion prod-ucts in cement concrete

43 ESEM-EDSandXRDAnalysis of SulfateAttackonCementConcrete In order to verify the rationality of the abovetheoretical calculation the sulfate attack on cement andconcrete was studied 0e theoretical formation conditionsincluding temperature and critical concentration of sulfateion were all based on ideal solution the activity of variousions and erosion products was not considered 0ereforethere would be a difference between theoretical calculationresults and actual values In order to determine the con-centration range of sulfate concentration of different erosionproducts the sulfate solution of different concentrations wasprepared to conduct sulfate attack ie 1 5 10 20and saturated solution Figure 6 shows the ESEM-EDSspectra of microstructure types and morphology of erosionproducts of cement concrete attacked by sulfate solution ofdifferent concentrations for 4months

Figure 6 shows that the microstructure types andmorphology of substances of the specimens before and aftersulfate attack are different As seen from Figure 6(a) themajor hydration products of the unattacked specimen arehexagonal plate-like calcium hydroxide (CH) needle-likeettringite (AFt) flocculent and gelatinous calcium silicatehydrate (C-S-H) and calcium aluminate hydrate (C-A-H)Moreover some micro pores and cracks can also be ob-served With the increase of sulfate solution concentration(ie 1 and 5) plenty of rod-like AFt with a larger di-ameter-length ratio is generated in system and the amountof hexagonal plate-like CH decreases as shown inFigures 6(b) and 6(c) Some plate-like erosion products canbe observed from the specimen attacked by 10 sulfatesolution which may be dihydrate gypsum or the partiallycorroded CH as shown in Figure 6(d) Moreover plenty ofdendritic and fibriform erosion products are observed inspecific region of the specimen as shown in Figure 6(e) 0eEDS analysis indicates that the major elements are Ca Si Oand S which may be considered as the residual skeleton ofC-S-H based on its morphology and characteristic as shownin Figure 6(f ) 0e higher the sulfate solution concentrationthe more excellent the crystallinity of granular sulfate saltand plate-like erosion products as shown in Figures 6(g) and6(h) 0e EDS analysis indicates that the major elements areCa O and S which may be dihydrate gypsum based on thecharacteristic of morphology as shown in Figure 6(i) 0econcentration range of sulfate ion for the formation oferosion products in Figure 6 accords well with the theoreticalcalculated value in Section 42 which implies that themechanism of sulfate attack on cement concrete can berepresented by the principles of chemical thermodynamicsMoreover the amount of hydration products including CHC-S-H and C-A-H in the specimens attacked by sulfate isreduced which may be due to the fact that the sulfate attackreduces pH value destroys thermodynamic equilibriumstate and accordingly results in the decomposition of hy-dration products It can be seen that compared with themicrostructure of specimens attacked by sulfate solution of

Table 6 Value of the fitted data

Items atr bNa2SO4 011322 minus 00024978Na2SO4middot10H2O 43751e minus 11 0068435

0 20 40 60 80 10006

07

08

09

10

ThenarditeRela

tive h

umid

ityndash

Temperature (degC)

324degC 085 RH

Solution

Mirabilite

Mirabilite (or solution supersaturatedwrt mirabilite)

0

20

40

60

Solubility of thenarditeSolubility of mirabilite

Solu

bilit

y10

0g w

ater

Figure 5 Phase spectrum and solubility of sodium sulfate

Advances in Materials Science and Engineering 11

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

different concentrations the microstructure of the specimenattacked by 1 sulfate solution becomes denser which iscaused by the generation of expansive substances of the AFtand AFm which fill in pores and reduce part of the porosity0erefore it has a positive effect on the performances of thespecimens attacked by low sulfate solution concentrationHowever if the sulfate attack reduces pH value of the systemand causes the decomposition of hydration products and

deterioration of microstructure it will have a negative effecton the performance of the specimens

In order to investigate the effect of sulfate solutionconcentration on the phase composition of cement concretethe XRD spectra of the cement attacked by sulfate for 4months were tested as shown in Figure 7

As seen from Figure 7 there exist some significantdifferences in the erosion productsrsquo spectra before and after

(a) (b) (c)

(d) (e)

32

26

19

13

06

00100 200 300 400 500 600 700 800 900 100

O NaAl

Si

Au

S

Ca

Fe

Au

KCnt

Energy-keV

C

(f )

(g) (h)

O

SCa

356

285

214

142

71

0375 800 1225 1650 2075 2500 2925 3350 3775

KCnt

Energy-keV

(i)

Figure 6 ESEM-EDS spectra of cement and concrete attacked by sulfate solution of different concentrations (a) Non attacked (b) 1(c) 5 (d) 10 (e) 10 (f ) EDS analysis (g) 20 (h) Saturated solution (i) EDS analysis

12 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

sulfate attack shown as the appearance and disappearance ofsome diffraction peaks and variations of intensity and widthof diffraction peaks Based on the characteristic diffractionpeak of the substances it can be seen that the major hy-dration products of the specimens without sulfate attackedare calcium hydroxide (CH) calcium aluminate hydrate (C-A-H) calcium silicate hydrate (C-S-H) unhydrated trical-cium silicate (C3S) and dicalcium silicate (C2S) 0eir dif-fraction peaksrsquo intensity is stronger and the width isnarrower After being attacked by sulfate solution of dif-ferent concentrations ie 1 and saturated sulfate solutionsome diffraction peaksrsquo intensity of the hydration productsincluding CH C-A-H and C-S-H decreases and thecharacteristic diffraction peaksrsquo intensity belonging to theAFt enhances Moreover some new diffraction peaks cor-responding to the gypsum appear 0is is due to the fact thatthe sulfate ion reacts with hydration products of CH andC-A-H and is generated into AFt and gypsum Furthermoresome characteristic diffraction peaksrsquo intensity weakenswhich may be caused by the decrease of pH value due toreaction of the sulfate ion with CH 0erefore the ther-modynamic equilibrium state of system is destroyed andresults in the decomposition of cementitious hydrationproducts 0e variation of diffraction peaks of the specimensbefore and after sulfate attack accords well with the results inFigure 6 No diffraction speaks of mirabilite and thenarditeare observed in Figure 7 which may be due to the fact thatthe sulfate ion intruding into specimen is reacted and thedehydration of mirabilite resulted from the preparationmethod of XRD analysis

44 Variation of Mechanical Properties of the SpecimensAttacked by Sulfate 0e microstructure types and mor-phology of erosion products and phase compositions of thespecimens can be affected by sulfate attack In order to inves-tigate the effects of sulfate attack on macro properties of thespecimens the mechanical properties of the specimens attacked

by sulfate solution of different concentrations and erosion ages(ie 2 and 4months) were investigated as shown in Figure 8

As seen from Figure 8 the mechanical properties of thespecimens attacked by sulfate for 2months first increase andthen decrease with the increase of sulfate solution concen-tration and the corresponding strength is larger than that ofthe unattacked specimens Compared with mortar and con-crete the variation of the paste strength is more significant0ere may be two primary reasons for the confusion First thewater to cement ratio is lower so the porosity of cement pastespecimen is less Some sulfate ion intruding into specimenscould be reacted and generated into expansive substanceswhich can fill and refine the micro pores Second the formederosion products could enhance the compactness of system andreduce the amount of CH in cement specimen [21 49 50] sothe microstructure and the performance of the specimens areimproved0e higher the concentration of sulfate solution thegreater the concentration gradient between the solution andspecimen surface 0erefore more sulfate ion intrudes intospecimen and plenty of expansive substances are generated Ifthe expansive substances generated by sulfate attack can beaccommodated by pores in system the sulfate attack has apositive effect on performance of the specimens Converselythe sulfate attack destroys the thermodynamic equilibriumstate of the system and results in micro damage so it has anegative effect Macroscopically it is manifested as the decreaseof the mechanical properties of the specimens Although thecompressive strength of paste and mortar attacked by sulfatefor 4 months decreases with the increase of sulfate solutionconcentration the corresponding compressive strength of theconcrete first increases and then decreases 0e maximumdecreasing amount of the compressive strength of the speci-mens can reach up to 30 0e flexural strength of the pastefirst increases and then decreases with increase of the sulfatesolution concentration However the corresponding flexuralstrength of the mortar decreases and the maximum decreasingamount can reach up to 50 0e paste has a better resistanceto sulfate attack which is due to generation of plenty of

Saturated solution

1 sulfate solutionUnattacked specimen

Δ ΔΔΔΔ

Δ Δ Δ Δ

Δ

lowast

lowast lowast

lowastlowast

lowast

lowast

lowastlowast

C3SΔ AFtC-S-H

CHlowast CaSO4middot2H2O

C-A-HC2S

20 30 40 50 60102Θ (deg)

Figure 7 XRD spectra of phase composition of specimens attacked by sulfate solution

Advances in Materials Science and Engineering 13

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

hydration products resulting from the more usage of ce-mentitious minerals per unit volume 0e longer the erosionage the more the sulfate ion intruding into specimens Moreexpansive substances are formed and reacted with CH whichresults in the micro damage and decrease of pH value of thesystem With increasing erosion time more and more hy-dration products are decomposed 0erefore the sulfate attackhas a negative effect on performance of the specimens Mac-roscopically it is manifested as the decrease of strength ofvarious specimens

5 Conclusions

(1) 0e relationship between the temperature and theGibbs free energy of the erosion products generatedduring the sulfate attack was deduced 0eoretical

orientation of sulfate attack on concrete was deter-mined and the critical sulfate ion concentration andforming conditions of erosion products were de-termined 0e results show that the generation of theerosion products in cement concrete is related totemperature relative humidity and sulfate ionconcentration which affect each other 0e dihy-drate gypsum is preferentially generated when thetemperature is less than 42degC and [SO2minus

4 ] concen-tration is more than 23times10minus 3molL However thecorresponding formation [SO2minus

4 ] concentration andtemperature of anhydrite are more than42times10minus 3molL and 42degC respectively 0e theo-retical thermodynamic stable temperature of AFt isabout 97degC and the critical [SO2minus

4 ] concentration ofAFt dominated significantly by temperature is no less

0 10 20 300

40

80

120Co

mpr

essiv

e str

engt

h (M

Pa)

Concentration ()

10

11

12

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

Flex

ural

stre

ngth

(MPa

)(a)

Com

pres

sive s

tren

gth

(MPa

)

Compressive strength for 2 monthsCompressive strength for 4 months

Flexural strength for 2 monthsFlexural strength for 4 months

0 10 20 3020

40

60

Concentration ()

4

5

6

7

8

9

10

Flex

ural

stre

ngth

(MPa

)

(b)

0 5 10 15 20 25 3010

15

20

25

Concentration ()

Axi

al co

mpr

essiv

e str

engt

h (M

Pa)

2 months4 months

(c)

Figure 8 Mechanical properties of different specimens attacked by sulfate solution (a) Paste (b) Mortar (c) Concrete

14 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

than 28times10minus 3molL 0e mirabilite and thenarditemay be generated when the concentration of sulfateion is more than 15times10minus 3molL 0e mirabilite canbe spontaneously crystallized when the temperatureis less than 324degC Moreover the theoretical ther-modynamic stable temperature for thaumasite is lessthan 44degC and the corresponding critical [SO2minus

4 ]concentration is more than 00023molL 0ereforethere is a significant difference in the theoreticaltemperature from that reported by existingdocuments

(2) 0e phase composition microstructure crystalform and morphology of erosion products of ce-ment before and after sulfate attack were investigatedby ESEM-EDS and XRD 0e mechanism of sulfateattack on cement and concrete was investigatedbased on principles of chemical thermodynamics0e major erosion product of cement attacked bylow concentration sulfate solution is rod-like AFtwith a larger diameter-length ratio generated in thesystem but plate-like dihydrate gypsum and gran-ular sulfate salt are the major products when theconcentration of sulfate solution is more than 10Moreover there exist the residual skeletons of C-S-Hor CH 0e XRD spectra show that the phasecomposition intensity and width of diffractionpeaks of erosion products in cement are changedbefore and after sulfate attack 0e diffraction peaksrsquointensity of the AFt and dehydrate gypsum increasesbut the corresponding diffraction peaksrsquo intensity ofthe CH C-S-H and C-A-H decreases and evendisappears 0e sulfate solution concentration anderosion age have significant effects on themechanicalproperties of the specimens Although the com-pressive strength of the paste and mortar attacked bysulfate for 4 months decreases with the increase ofsulfate solution concentration the correspondingcompressive strength of the concrete first increasesand then decreases 0is is due to the double effectsie positive and negative effects of the sulfate attackon performance of the specimens

Data Availability

0e data used to support the findings of this study areavailable from the corresponding author upon request

Conflicts of Interest

0e authors declare that they have no conflicts of interest

Acknowledgments

0is study was funded by the National Natural ScienceFoundation of China (grant nos 51778632 U1434204U1934217 and 51408614) and China Postdoctoral ScienceFoundation (grant nos 2016M600675 and 2017T100647)Author has received research grants from the GuangdongProvincial Key Laboratory of Durability for Marine Civil

Engineering (grant nos GDDCE14-03 15-08 17-2 and18-16) Basic Research on Science and Technology Pro-gram of Shenzhen (JCYJ20170818143541342 andJCYJ20180305123935198) and Natural Science Founda-tion of Hunan Province of China (2017JJ3385)

References

[1] A Neville ldquo0e confused world of sulfate attack on concreterdquoCement and Concrete Research vol 34 no 8 pp 1275ndash12962004

[2] M Yao and J Li ldquoEffect of the degradation of concrete frictionpiles exposed to external sulfate attack on the pile bearingcapacityrdquo Ocean Engineering vol 173 pp 599ndash607 2019

[3] S F Etris Y R Fiorni and K C Lieb ldquoA new test for sulfateresistance of cementsrdquo Journal of Testing and Evaluationvol 2 no 6 pp 510ndash515 1974

[4] R Dhole M D A 0omas K J Folliard and T DrimalasldquoChemical and physical sulfate attack on fly ash concretemixturesrdquo ACI Materials Journal vol 116 no 4 pp 31ndash422019

[5] M Santhanam M D Cohen and J Olek ldquoMechanism ofsulfate attack a fresh look part 2 proposed mechanismsrdquoCement and Concrete Research vol 32 no 6 pp 915ndash9212003

[6] O S B AlndashAmoudi ldquoAttack on plain and blended cementsexposed to aggressive sulfate environmentsrdquo Cement andConcrete Composites vol 24 no 3 pp 305ndash316 2002

[7] L Pel H Huinink K Kopinga R P J van Hees andO C G Adan ldquoEfflorescence pathway diagram under-standing salt weatheringrdquo Construction and Building Mate-rials vol 18 no 5 pp 309ndash313 2004

[8] E Alvarez-Ayuso and H W Nugteren ldquoSynthesis ofettringite a way to deal with the acid wastewaters of alu-minium anodising industryrdquo Water Research vol 39 no 1pp 65ndash72 2005

[9] L Jiang D Niu L Yuan and Q Fei ldquoDurability of concreteunder sulfate attack exposed to freeze-thaw cyclesrdquo ColdRegions Science and Technology vol 112 pp 112ndash117 2015

[10] S Valencia G William A D Eugenia and G R Mejia deldquoFly ash slag geopolymer concrete resistance to sodium andmagnesium sulfate attackrdquo Journal of Materials in Civil En-gineering vol 28 no 12 Article ID 04016148 2016

[11] N C Axel R J Torben and C Jonathan ldquoFormation ofettringite Ca6Al2(SO4)3 (OH)12middot26H2O AFt and mono-sulfate Ca4Al2O6(SO4)middot14H2O AFmndash14 in hydrothermalhydration of Portland cement and of calcium aluminumoxidendashcalcium sulfate dihydrate mixtures studied by in situsynchrotron Xndashray powder diffractionrdquo Journal of Solid StateChemistry vol 177 pp 1944ndash1951 2004

[12] C D Lawrence ldquoMortar expansions due to delayed ettringiteformation Effects of curing period and temperaturerdquo Cementand Concrete Research vol 25 no 4 pp 903ndash914 1995

[13] R H Michael K B Steven and B Ronald ldquo0e evolution ofstructural changes in ettringite during thermal decomposi-tionrdquo Journal of Solid State Chemistry vol 179 no 4pp 1259ndash1272 2006

[14] E E Hekal E Kishar and H Mostafa ldquoMagnesium sulfateattack on hardened blended cement pastes under differentcircumstancesrdquo Cement and Concrete Research vol 32 no 9pp 1421ndash1427 2002

[15] H A F Dehwah ldquoEffect of sulfate concentration and asso-ciated cation type on concrete deterioration and

Advances in Materials Science and Engineering 15

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering

morphological changes in cement hydratesrdquo Constructionand Building Materials vol 21 no 1 pp 29ndash39 2007

[16] P W Brown ldquo0aumasite formation and other forms ofsulfate attackrdquo Cement and Concrete Composites vol 24no 3-4 pp 301ndash303 2002

[17] T Liu S Qin D Zou and W Song ldquoExperimental inves-tigation on the durability performances of concrete usingcathode ray tube glass as fine aggregate under chloride ionpenetration or sulfate attackrdquo Construction and BuildingMaterials vol 163 pp 634ndash642 2018

[18] I Biczok Concrete Corrosion and Concrete ProtectionChemical Publishing Palm Springs CA USA 1967

[19] F Bellmann B Moser and J Stark ldquoInfluence of sulfatesolution concentration on the formation of gypsum in sulfateresistance test specimenrdquo Cement and Concrete Researchvol 36 no 2 pp 358ndash363 2006

[20] Z Zhang X Jin and W Luo ldquoLong-term behaviors ofconcrete under low-concentration sulfate attack subjected tonatural variation of environmental climate conditionsrdquo Ce-ment and Concrete Research vol 116 pp 217ndash230 2019

[21] X Yu Y Zhu Y Liao and D Chen ldquoStudy of the evolution ofproperties of mortar under sulfate attack at different con-centrationsrdquo Advances in Cement Research vol 28 no 10pp 617ndash629 2016

[22] M Santhanam ldquoEffect of solution concentration on the attackof concrete by combined sulphate and chloride solutionsrdquoEuropean Journal of Environmental and Civil Engineeringvol 15 no 7 pp 1003ndash1015 2011

[23] P B Xie ldquoMechanism of sulfate expansion II vali DTA ion ofthermodynamie theoryrdquo Cement and Concrete Researchvol 22 no 5 pp 845ndash854 1992

[24] M T Blanco-Varela J Aguilera and S Martınez-RamırezldquoEffect of cement C3A content temperature and storagemedium on thaumasite formation in carbonated mortarsrdquoCement and Concrete Research vol 36 no 4 pp 707ndash7152006

[25] S A Hartshorn J H Sharp and R N Swamy ldquo0e thau-masite form of sulfate attack in Portland-limestone cementmortars stored in magnesium sulfate solutionrdquo Cement andConcrete Composites vol 24 no 3-4 pp 351ndash359 2002

[26] P Brown R D Hooton and B Clark ldquoMicrostructuralchanges in concretes with sulfate exposurerdquo Cement andConcrete Composites vol 26 no 8 pp 993ndash999 2004

[27] X C Fu W X Shenm T Y Yao and W H Hou PhysicalChemistry Higher Education Press Beijing China 2009

[28] D L Ye Shi Yong Wu Ji Wu Re Li Xu Shu Ju Shou CeMetallurgical Industry Press Beijing China 2002

[29] D Freyer S Fischer K Kohnke andW Voigt ldquoFormation ofdouble salt hydrates I Hydration of quenched Na2SO4_CaSO4phasesrdquo Solid State Ionics vol 96 no 1-2 pp 29ndash33 1997

[30] G W Scherer ldquoCrystallization in poresrdquo Cement and Con-crete Research vol 29 no 8 pp 1347ndash1358 1999

[31] G W Scherer ldquoStress from crystallization of saltrdquo Cementand Concrete Research vol 34 no 9 pp 1613ndash1624 2004

[32] S Jan M Jacques and O Ivan Sulfate Attack on ConcreteSpon press London UK 2002

[33] J Skalny J Marchand and I Odler Sulfate Attack onConcrete Spon Press London UK 2002

[34] J R Clifton and J M Pommersheim Sulfate Attack of Ce-mentitious Materials Volumetric Relations and Expansions(NISTIR 5390) NIST Gaithersburg MA USA 1994

[35] Z D Niu F Q Cheng C B Li and X Chen Shui Yan Ti XiXiang Tu Ji Qi Ying Yong Tianjin University Press TianjinChina 2002

[36] 0ermoddem httpthermoddembrgmfr2019[37] GB175ndash2007 Common Portland Cement China Standards

Press Beijing China 2008 in Chinese[38] GBT 1346ndash2011 Test Methods for Water Requirement of

Normal Consistency Setting Time and Soundness of thePortland Cement China Standards Press Beijing China 2012in Chinese

[39] GBT 17671ndash1999Method of Testing CementsndashDeterminationof Strength China Standards Press Beijing China 1999 inChinese

[40] GBT 50081ndash2002 Standard for Test Method of MechanicalProperties on Ordinary Concrete China Architecture ampBuilding Press Beijing China 2014 in Chinese

[41] S B Wang S X Ji Y J Liu and K Y Hu ldquoEffect of alkali onexpansion of sulfoaluminate cementrdquo Journal of the ChineseCeramic Society vol 14 no 3 pp 31ndash38 1986

[42] Y H Li S Y Gao P S Song and S P Xia ldquoEffect of Pitzermixing parameters on the solubility prediction of the phaesystem HClndashNaClndashH2Ordquo Acta Physico-Chimica Sinicavol 17 no 1 pp 91ndash94 2001

[43] J G Xue ldquoOn the expansion associated with etringite for-mationrdquo Journal of the Chinese Ceramic Society vol 2pp 252ndash256 1984

[44] J F Robert ldquoSalt damage in porous materials how highsupersaturations are generatedrdquo Journal of Crystal Growthvol 242 no 3 pp 435ndash454 2002

[45] N Tsui R J Flatt and GW Scherer ldquoCrystallization damageby sodium sulfaterdquo Journal of Cultural Heritage vol 4 no 2pp 109ndash115 2003

[46] CCES 01ndash2004 Guide to Durability Design and Constructionof Concrete Structures China Architecture amp Building PressBeijing China 2005 in Chinese

[47] GB 50212ndash1991 Specification for Construction and Acceptanceof Anticorrosive Engineering of Buildings China plan pub-lishing house Beijing China 2002 in Chinese

[48] K P Mehta P K Mehta and P J M Monteiro ConcreteStructure Properties and Materials Tongji Press ShanghaiChina 1991

[49] Y Gu R-P Martin O OmikrineMetalssi T Fen-Chong andP Dangla ldquoPore size analyses of cement paste exposed toexternal sulfate attack and delayed ettringite formationrdquoCement and Concrete Research vol 123 p 105766 2019

[50] I Tai H P C Sergio and S Ignacio ldquo0e role of porosity inexternal sulphate attackrdquo Cement and Concrete Compositesvol 97 pp 1ndash12 2019

16 Advances in Materials Science and Engineering