Construction and Building Materials 58 (2014) 138–146

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier .com/locate /conbui ldmat

Effect of combined carbonation and chloride ion ingress by anaccelerated test method on microscopic and mechanical propertiesof concrete

http://dx.doi.org/10.1016/j.conbuildmat.2014.01.1020950-0618/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +98 912 504 4602; fax: +98 021 7751 2440.E-mail addresses: aaramce@aut.ac.ir (A.A. Ramezanianpour), ghahary@aut.ac.ir

(S.A. Ghahari), m_esmaeili@iust.ac.ir (M. Esmaeili).

A.A. Ramezanianpour a, S.A. Ghahari a,⇑, M. Esmaeili b

a Department of Civil and Environment Engineering, Concrete Technology and Durability Research Center, Amirkabir University of Technology, Tehran, Iranb Department of Railway Engineering, Iran University of Science and Technology, Tehran, Iran

h i g h l i g h t s

� Lower mechanical properties obtained in combined carbonation–chloride ion ingress.� Combined carbonation–chloride ion ingress has no effect on nano structure.� SEM shows less cubic, crystalline shape particles formed in the combined condition.� Hexagonal portlandite (Ca(OH)2) crystalline cubes observed in the nano graphs.� XRD reveals that carbonation produces more CaCO3 than the combined condition.

a r t i c l e i n f o

Article history:Received 29 December 2013Received in revised form 21 January 2014Accepted 29 January 2014Available online 4 March 2014

Keywords:CarbonationChloride ion ingressMechanical propertiesMicroscopic structureAutomatic tidal cycleSurface resistivityCO2 consumptionAccelerated test method

a b s t r a c t

In this study microscopic and mechanical properties of ordinary concrete exposed to CO2 gas, salinewater, and the combination of CO2 gas and saline water were investigated by an accelerated test method.Specimens with different water to cement ratio of 0.35, 0.4, and 0.45 were retained in an apparatus whichwas developed to provide an environment to simulate tide cycles. The CO2 gas pressure, NaCl solutionconcentration, temperature, and relative humidity were kept constant and controlled, and tide cycleswere executed automatically each 6 h. Specimens were retained in CO2 gas and in NaCl solution withenvironment characteristics tantamount to that of the apparatus. Microscopic structure and interfacialtransition zones of the specimens maintained in all three conditions were studied by implementing scan-ning electron microscope (SEM). In order to identify crystalline phases and morphological and structuralcharacteristics, nanographs were obtained by transmission electron microscopy (TEM), and phase changedue to carbonation and chloride ion ingress were studied along with X-ray diffraction analysis (XRD).Moreover, in order to signify the mechanical properties of specimens, compressive strength, surface resis-tivity, and CO2 consumption were measured. It was found that more C–S–H gel and CH crystals has beenformed in the presence of sole CO2 gas, and higher compressive strength is achieved compared to thecombined CO2 gas and chloride ion ingress, and separate saline water. Besides, due to the presence ofmoisture in the pore solution of specimens maintained in the combined condition, less CO2 gas and chlo-ride ion ingresses were observed.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Deterioration of concrete structures, especially exposed struc-tures, can happen because of several phenomena namely, carbon-ation, chloride ion attacks, sulfate attacks, freezing and thawing

cycles, etc. Among such deleterious environmental conditions, car-bonation and chloride ion attacks are more probable, especially inurban, and marine environment like large harbors, where the rateof carbonation and chloride penetration is high due to pollutionand deicing salts used in winters, and sea water respectively. Usingchloride-based deicing salts in urban regions has caused deleteri-ous effects on infrastructures [1–3]. Its effect can be described inthe reactions as follows [4]:

CaðOHÞ2 þMgCl2 !MgðOHÞ2 þ CaCl2 ð1Þ

Table 1Characteristics of the cement.

Oxide elements (% by mass) Cement (C)

SiO2 21.10CaO 62.08Fe2O3 2.74Al2O3 3.81SO3 2.6MgO 3.22K2O 0.73Na2O 0.12P2O5 0.22TiO2 0.24

Loss on ignition (%) 2.95Physical properties Specific gravity 3.17

Blaine fineness (cm2/g) 3519

A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146 139

3CaðOHÞ2 þ CaCl2 þ 12H2O! 3CaOþ CaCl2 þ 15H2O ð2Þ

Regarding the reaction (1), CaCl2 and Mg(OH)2 can cause a reduc-tion in compressive strength. Many researches have been con-ducted on carbonation of concrete structures due to atmosphericCO2 gas and pozzolans [5]. It is investigated that CO2 gas affectsthe durability of reinforced concrete structures, and causes corro-sion in the long run since the pH of pore solution may be reduced[6], and it may cause an increase in the concentration of chlorideions in the pore solution as well [7]. Moreover, due to the presenceof CO2 gas, concrete shrinkage happens when CO2 penetrates intoconcrete structure: it polymerizes the silicate chains in C–S–Hwhich can decrease the volume, and consequently causes cracks[8]; however, CO2 gas can improve the compressive strength ofspecimen [9,10] because of developing homogeneous micro struc-ture and reduction in porosity [11]. The following reactions happenin concrete structure in the proximity of CO2 gas:

CaðOHÞ2 þ CO2 ! CaCO3 þH2O ð3Þ

C� S�Hþ 2CO2 ! SiO2 þ 2CaCO3 þH2O ð4Þ

According to the reactions above, CO2 reacts with Ca(OH)2, and byconsuming (OH)� ion, the pH in the pore solution may increasewhich can increase the rate of corrosion [12], and reduce the poros-ity [11,13]. In fact, by the dissolution of CO2 in water, reaction (7),according to the following reactions, CaCO3, reaction (8) may beproduced in concrete pore solution:

CO2 þH2O! H2CO3 ð5Þ

H2CO3 þH2O! HC�3 þH3Oþ ð6Þ

HCO�3 þH2O! CO2�3 þH3Oþ ð7Þ

H2CO3 þ CaðOHÞ2 ! CaCO3 þ 2H2O ð8Þ

Regarding the reaction (7), the main cause of reduction in pH aftercarbonation, is the liberation of H3O+ [14]. Many other researcheshave been performed on chloride ion ingress in the vicinity of seashores and harbors [15–17]. It is well known that Kuzel’s salt(3CaO�Al2O3�0.5CaSO4�0.5CaCl2�10(11)H2O) and Friedel’s salt(3CaO�Al2O3�CaCl2�10H2O) may be formed when chloride ions pre-sents while hydration of cement is continuing. Both of the saltsare considered as AFm phases, i.e. hydrated calcium aluminatephases [18]. Depth of chloride penetration [19,20], and diffusionof chloride ions [21] have been studied as well. When chloride ionsreact with C3H, Friedel’s salt is formed. The chloride ion reactionwith C4AF leads to the formation of Kuzel’s salt; Kobayashi et al.[22] found that Friedel’s salt is formed because of mono sulfate(AFm) reaction with chloride ion. After the carbonation of Friedel’ssalt, some chloride ions fixed by Friedel’s salt may dissolute intopore solution, and an increase in chloride ions may lead to more in-gress of chloride ions. Such ions may reach the reinforcement barsaccording to the following reaction that can cause corrosion:

Fe2þ þ 2Cl� ! FeCl2 ð9Þ

FeCl2 þ 2H2O! FeðOHÞ2 þ 2HCl ð10Þ

The effect of carbonation on chloride ion diffusion has beeninvestigated [23–25], and it has been found that chloride bindingreduces when exposed to CO2 gas [26]. Since high amount of boundchlorides may release when pH goes below 12, such chlorides areconsidered as a potential risk for corrosion [27], however, carbon-ation may reduce the capacity of chloride binding [28] which maylead to an increase in the rate of chloride ion ingress [29].

In order to consider conditions that are more analogous to thereality, effect of combined carbonation and chloride ion ingress

should be investigated so that a mixture design that has moredurability in such conditions may be achieved; consequently, few-er budgets would be allocated to ‘‘bring the infrastructure to agood condition’’ [30]. However, few studies have been performedon the effect of combined carbonation and chloride ion attackson concrete although both of those factors affect the durability ofconcrete structures [31], and even accelerate each other [32].

In this research, the effect of simultaneous carbonation andchloride ion ingress on mechanical properties and micro structureof concrete have been investigated. In order to perform an acceler-ated test method of simultaneous carbonation and chloride ion in-gress, an apparatus was made to automatically maintain thetemperature, humidity, concentration of CO2 gas and execute wet-ting and drying cycles.

2. Materials and methods

2.1. Materials

All specimens were made with type I Portland cement, which meets ASTM C150[33] specification, with different water/cement ratios. The physical and chemicalcharacteristics of cement are presented in Table 1. Local natural river sand as fineaggregate, and crushed stone as coarse aggregate with maximum aggregate sizeof 4.75 mm, and 19 mm were used respectively. The water absorption, and specificgravity of fine and coarse aggregates are 2.5% and 1.6%, and 2490 and 2550 kg/m3

respectively. The specimens were cast and cured with potable water. In order tomaintain the workability, a liquid polycarboxylic-ether base with solid content of45% and specific gravity of 1.2 was used as superplasticizer.

2.2. Mixture design and specimens preparation

The mixture proportion of all specimens is illustrated in Table 2. For all mix-tures, by adding adequate amount of superplasticizer, slump was kept constant atabout 80 mm. All materials were batched in a 60 l capacity vertical axis mixer with3 different water/cement ratios: 0.35, 0.4, and 0.45 shown as S35, S40, and S45respectively. In order to eliminate entrapped air, after casting fresh concrete, spec-imens were vibrated for 3–5 s and maintained at the temperature of 24 ± 1 �C. Hav-ing cured in moist room for 24 h, they were demolded and cured in water at24 ± 2 �C before placing them in the apparatus for testing under carbonation andsaline water. Since early carbonation can alter hydration procedure and calciumhydroxide content at the surface of concrete [32], specimens were cured for 7 daysso that the test has less impact on the normal procedure of hydration. After this per-iod, 50% of all specimens were placed in the apparatus for simultaneous carbon-ation and chloride ion ingress and only carbonation, 25% of them were cured inwater for standard curing, and rest of them were maintained under saline water.

2.3. Apparatus

The apparatus for simultaneous carbonation and chloride ion ingress, and wet-ting and drying cycle simulation is illustrated in Fig. 1 schematically. In chamber 1(Ch.1), by using a water pump and electric valve, saline water (0.3% NaOH and 3%NaCl [34]) was pumped and drained from chamber 2 (Ch.2) every 6 h to simulatethe tide process and provide an environment in which we have tidal zone like mar-ine structures; shorter or longer duration for cycles are achievable by commandinga programmable logic controller (PLC) which interpret all commands through a

Table 2Mixture proportion of specimens.

Sample ID W/C Water Cement content Coarse aggregate Fine aggregate Superplasticizerkg/m3 kg/m3 kg/m3 kg/m3 % by paste volume

S35 0.35 140 400 1018 833 1.6S40 0.4 160 400 990 810 1.3S45 0.45 180 400 963 788 1.2

Fig. 1. Apparatus for simultaneous carbonation and chloride ion ingress.

140 A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146

human machine interface (HMI) presented in Fig. 2. Besides, by using a pneumaticvalve, the interior pressure was kept constant at 1.5 bar (CO2 content of7500 ± 100 ppm).

The temperature and relative humidity were controlled and kept constant at23 ± 2 �C and 70 ± 5% by a warming element and a dehumidifier respectively. Allcommands for the cycles were executed automatically by implementing a PLC (Pro-grammable Logic Controller). At the bottom of Ch.1, saline water was circulated tosimulate the tidal zone, and specimens at the top of Ch.1 were not exposed to thesaline water in order to have pure carbonation. The real model of the registered pat-ent apparatus is illustrated in Fig. 3; the overall capacity of each chamber is 1000 l,and both of the chambers can be filled with specimens, if needed; however in thisresearch the capacity of Ch.1 was enough for the overall number of specimens. Inaddition, the CO2 tank contains 40 l of high pressure CO2 gas. All connections werechecked to be sealed in order to apply cycles in an isolated environment.

3. Experimental procedure

3.1. Mechanical properties

In order to measure compressive strength, according to ASTMC39 [35], 10 � 10 cm concrete cubes were cast, and specimenswere tested at ages of 28 and 90 days.

Fig. 2. The human machine interface of the apparatus.

3.2. Surface resistivity

The surface resistivity of specimens was measured at four qua-ternary longitudinal locations by four point wenner array probetechnique [36] on 10 � 20 cm cylinder specimens at each test ages,i.e. 28 and 90 days. This non-destructive test shows the conductiv-ity and ion penetration of specimens.

3.3. CO2 consumption

In order to compare CO2 consumption of specimens through acertain period, i.e. 28 days, which represents the capacity of

Fig. 3. The real model of the apparatus.

A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146 141

specimens to store CO2, the consumption can be obtained inaccordance with Eq. (2):

WCO2ð%Þ ¼ðWafter carbonation þWwater lossÞ � ðWbefore carbonationÞ

Wcementð2Þ

3.4. Scanning electron microscope (SEM)

5 � 5 � 2 mm pieces of all types of specimens, which werecured in water, at age of 28 days, and also specimens that weremaintained in the apparatus were coated with gold–palladium,and scanned with Hitachi U8040 scanning electron microscope.In order to observe the effect of carbonation and chloride ion at-tacks on microstructure and interfacial transition zone, aggregatesand cement paste were positioned at the center of the SEMseparately.

35 Carb 35 Sim 35 Chl 35 PC 40 Carb 40 Si

90 48 47.1 45.8 45.8 42.1 41.9

28 45.2 46 44.2 44.2 40.5 41.2

0

10

20

30

40

50

60

70

80

90

100

Com

pres

sive

str

engt

h (M

Pa)

W/B = 0.35 W

Fig. 4. Results of compressive strength for all specimens main

35 Carb 35 Sim 35 Chl 35 PC 40 Carb 40 Sim

90 31 29.3 26.8 26 27.1 24.1

28 20.3 19.1 18.1 17.2 16 14.7

0

10

20

30

40

50

60

Surf

ace

resi

s�vi

ty (k

Ωcm

)

W/B = 0.35 W/B

Fig. 5. The results of surface resist

3.5. Phase changing monitoring

In order to monitor and quantify calcium hydroxide changes bya non destructive test, 10 mm of the hydrated cement pastes atages of 28 days was powdered, and X-ray diffraction (XRD) analysiswas performed by Philips powder diffractometer. The powder wassituated in Cu ka radiation and Ni filler with 40 kV and 30 mAworking condition for 4–70�, 2h range analyzing with step size of0.02�. A Profile Fit Software (PANalyticalX‘PertHighScore) was thenimplemented for phase quantification.

3.6. Transmission electron microscope (TEM)

Transmission electron microscopy (TEM) and selected area ofelectron diffraction patterns (SAEDP, on powdered cement pastemaintained in CO2 gas , and combined condition of CO2 gas and

m 40 Chl 40 PC 45 Carb 45 Sim 45 Chl 45 PC

41.1 40.9 39.4 38.1 36.4 35

41.7 39.3 38.2 37.4 34.9 33.8 hh/B = 0.4 W/B = 0.45

tained at different conditions at ages of 28 and 90 days.

40 Chl 40 PC 45 Carb 45 Sim 45 Chl 45 PC

20.2 21 21.6 19.3 18.2 17.1

11.5 12.4 12.4 11.8 7.6 7.8

hh= 0.4 W/B = 0.45

ivity at age of 28 and 90 days.

Table 3CO2 consumption results for all specimens maintained in CO2 gas and at tidal zone.

Curing condition

in CO2 gas Label at Tidal zone

Mass afterCarbonation (gr)

Massbefore (gr)

Waterloss (%)

Cementmass (gr)

Uptake fromeq. (2) (%)

Mass afterCarbonation (gr)

Massbefore (gr)

Waterloss (%)

Cementmass (gr)

Uptake fromeq. (2) (%)

2399 2394 13.4 40 12.1 S35 2399 2395 9.9 40 9.22363 2357 14.2 40 13.9 S40 2363 2359 10.3 40 10.52335 2330 15.4 40 14.6 S45 2333 2328 10.8 40 11.0

142 A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146

saline water, were observed in order to identify morphologicalphases in CO2 gas, and the combined CO2 gas and saline water con-dition. A capillary tube was used, and colloidal suspension wasdropped on a 3 mm diameter cupper grid coated with carbon. Phi-lips CM30 TEM with electron diffraction mode was operated at150 kV. Digital nano graph software was then used to identifythe size of particles and the electron diffraction patterns.

Fig. 6. S35 maintained in CO2 gas (30 kX magnification).

4. Results and discussion

4.1. Mechanical properties

The results of compressive strength as a mechanical propertyfor all specimens maintained in saline water (Chl), CO2 gas (Carb),standard curing condition (PC), and at tidal zone (Sim) are pre-sented in Fig. 4. The results show that specimens maintained inCO2 gas have about 13% higher compressive strength comparedto specimens maintained in the tidal zone containing both CO2

gas and saline water. Although it is expected to have synergy effectin the tidal zone, less compressive strength than specimens cure inCO2 gas could be due to the moisture that may available near thesurface of the specimens which hinders CO2 from penetrating intothe pores of specimens. In fact, saline water has washed the sur-face, so the process of forming carbonation products has been de-layed. During carbonation, CO2 reacts with cementitious materialswhich may be the main reason of higher compressive strength. Animportant carbonation product is calcium carbonate (CaCO3)which is harder than Ca(OH)2. On the other hand, calcium hydrox-ide is weaker than C–S–H gel, hence with carbonation, Ca(OH)2 haschanged to CaCO3 which is harder so that the specimen would be-come harder at the surface. It can be seen that more carbonationdepth can cause harder proportion of specimen, and higher com-pressive strength. Moreover, different mixture designs have causeddifferent values of compressive strength since there is less porosityin specimens with low water/cement ratio. The compressivestrength for specimens with water/cement ratio of 0.35 is at most23% higher than that of 0.4, and 0.45. As can be seen from Fig. 4, thehighest compressive strength for S35 maintained in CO2 gas, isabout 17% and 22% higher than that of S40 and S45 maintainedin the standard condition at age of 90 days respectively. However,no significant difference in compressive strength of specimensmaintained in salinewater was observed since chloride ions donot alter pore structures.

Fig. 7. S35 maintained in tidal zone (30 kX magnification).

4.2. Surface resistivity

Relative permeability of specimens can be evaluated by thisnon-destructive test [37]. Regarding the results of SR values forall specimens maintained in saline water (Chl), CO2 gas (Carb),standard curing condition (PC), and at tidal zone (Sim) presentedin Fig. 5, carbonated specimens have higher SR value; However,specimens maintained at tidal zone did not reach that value of sur-face resistivity which could be due to the moisture available at the

surface of specimens during the simulation of four tide cycles of sixhour-period per day. Carbonation products can fill the pores, andalso with CaCO3 more strength can be obtained from the carbon-ated area. According to the results, there is a slight difference inSR values of the reference and specimens maintained in salinewater because saline water cannot change any phase in the hydra-tion products. According to the Fig. 5, S35 maintained in CO2 gashas the highest surface resistivity which is about 40% and 80%higher than that of S40 and S45 maintained in the standard condi-tion at age of 90 days respectively.

4.3. CO2 consumption

The CO2 consumption for all specimens was not constant.Regarding the CO2 consumption results presented in Table 3,

A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146 143

obtained from equation (2), the maximum and minimum CO2 con-sumption for specimens maintained in CO2 gas are 14.6% and 12.1%for S45 and S35 respectively. It can be concluded that lower water/cement ratio could fill the pores more than higher water/cementratio so that CO2 could not penetrate into the specimen. That islower water/cement ratio which connotes less porosity is animportant element that could reduce CO2 penetration. Moreover,the maximum and minimum CO2 consumptions for specimenmaintained at tidal zone are 11% and 9.2% for S45 and S35 respec-tively, the same as what mentioned for CO2 gas maintainedspecimens. However, because of the moisture available at thesurface that has caused less room for carbonation products to beformed, the CO2 consumption for specimen maintained at tidalzone is at least about 5% lower than that of specimen maintainedin CO2 gas.

4.4. Scanning electron microscope results

Gold coated small portion of specimens were placed in thescanning electron microscope (SEM) stubs. The micrographs ofS35 for hydrated concrete specimens maintained in different con-ditions at age of 28 days are illustrated in Figs. 6–9. In order to dis-tinguish C–S–H gel, in the micrographs, they are slightly shownbrighter, and aggregates are shown darker [38]. Sands can be de-scried by their semi-dark hue. The specimen S35 micrographs (at30 kX and 15 kX magnification) maintained in CO2 gas, at tidalzone, in saline water, and in standard curing condition are pre-sented in Figs. 6–9 respectively. According to the SEM micrographs,S35 maintained in CO2 gas has many cubic, crystalline shape parti-cles that have been formed because of carbonation [39]. Besides,CaCO3 is available in various forms which may increase compres-sive strength and surface resistivity. When compared to Fig. 7,S35 maintained at tidal zone, the cubic crystalline shape particlesand CaCO3 have been reduced. It is also seen that carbonated spec-imens has less porosity than specimens maintained at tidal zone.Furthermore, in both curing conditions, capillary porosity has beenreduced which may lead to lower sorptivity coefficient. In Fig. 8,S35 maintained in saline water, no perceptible difference can beobserved compared to S35 (Fig. 9) maintained in standard curingcondition. It can be concluded that by curing in saline water, nochemical or physical reaction has been taken place.

Fig. 8. S35 maintained in saline water (15 kX magnification).

Fig. 9. S35 maintained in standard curing condition (15 kX magnification).

4.5. Phase change analysis

The effect of different conditions on changing in phases can beobserved by means of X-ray diffractometer (XRD). The results ofXRD for S35 maintained in salinewater, CO2 gas, and combinedcondition from top to bottom are shown in Fig. 10. Portlandite(Ca(OH)2), calcite (CaCO3), and silicon dioxide (SiO2), as crystallinephases are the main peaks. According to the results, Ca(OH)2 hasbeen reduced and CaCO3 has been increased consequently in CO2

environment, however, specimens maintained at tidal zone, havelesser amount of CaCO3 which could be due to the preventing effectof moisture available at the surface. Besides that, saline water canwash CO2 that may penetrate into the concrete. The reduction inCa(OH)2 signifies that carbonation consumes Ca(OH)2 and pro-duces CaCO3. Moreover, C2S and C3S have also been reduced, henceproduction of CaCO3 is not only due to Ca(OH)2. XRD results vali-date the fact that carbonation has occurred mostly in the proximityof the surface since the peak of Ca(OH)2 are weak at points far fromthe surface. Specimens maintained in CO2 gas have lower peak forCa(OH)2 and higher peak for C2S, C3S, and CaCO3. Between therange of 2h = 18� and 42�, because of the non-convex shape ofthe XRD results, crystalline materials of carbonation productsmay available.

4.6. Transmission electron microscope results

Transmission electron microscopy (TEM) is a technique that canhelp to identify crystalline phases and morphological and struc-tural characteristics [40]. According to transmission electronmicroscopy (TEM) nano graphs illustrated in Figs. 11 and 12 for ce-ment pastes maintained in CO2 gas and the combined conditionrespectively, the exact phase changes can be obtained. The TEMnano graphs in bright field mode and selected area of electron dif-fraction patterns (SAEDP) for cement paste presented in Figs. 11and 12 at age of 90 days, shows that in specimen maintained inCO2 gas as well as specimens maintained in the combined CO2

gas and saline water, no specific changes in phases are observed.As a result, neither CO2 gas nor the combined condition could alterthe nano structure of the specimen. Hence, nano supplementarymaterials may not have any significant effect on improving thedurability of specimens due to CO2 gas and saline water. Hexagonalportlandite (Ca(OH)2) crystalline cubes observed in the nanographs, however, the congestion of crystals in the cement pastemaintained in CO2 gas nano graph is relatively lower than that ofcement paste maintained in the combined condition. This couldbe due to the consumption of Ca(OH)2 in reaction with carbondioxide. Additionally, in the cement paste maintained in CO2 gas,more aragonite (CaCO3) can be seen. Moreover, corresponds toSAEDP shown in Figs. 11 and 12, both cement pastes still show rel-atively crystalline structure.

Fig. 10. XRD analysis of S35 maintained in salinewater, CO2 gas, and simultaneous CO2 gas and saline water at 28 days.

Fig. 11. TEM and SAEDP (as inset) for S35 maintained in CO2 gas at age of 90 days.

Fig. 12. TEM and SAEDP (as inset) for S35 maintained in CO2 gas and saline water at age of 90 days.

144 A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146

A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146 145

5. Conclusions

In this study the effect of an accelerated test method of simul-taneous carbonation and chloride ion ingress on mechanical prop-erties and durability of ordinary concrete was investigated. In thismatter, an extensive laboratory research was conducted and theprimary findings are as follows:

1. Although it is expected to have synergy effect in the tidal zone,results show that specimens maintained in CO2 gas have highercompressive strength compared to specimens maintained in thetidal zone containing both CO2 gas and saline water.

2. Carbonated specimens have higher SR value; however, speci-mens maintained at tidal zone could not reach that value of sur-face resistivity which could be due to the moisture available atthe surface of specimens during the simulation of four tidecycles of six hour-period per day.

3. S35 maintained in CO2 gas has the highest surface resistivitywhich is about 40% and 80% higher than that of S40 and S45maintained in the standard condition at age of 90 daysrespectively.

4. The maximum and minimum CO2 consumption for specimensmaintained in CO2 gas is 14.6% and 12.1% for S45 and S35respectively, and the CO2 consumption for specimen main-tained at tidal zone is at least about 5% lower than that of spec-imen maintained in CO2 gas.

5. Results of XRD test reveals that carbonation consumes Ca(OH)2

and produces CaCO3, C2S and C3S have also been reduced, andthat carbonation has occurred mostly in the proximity of thesurface since the peak of Ca(OH)2 are weak at points far fromthe surface.

6. According to the SEM micrographs, S35 maintained in CO2 gashas many cubic, crystalline shape particles that have beenformed because of carbonation. Besides, CaCO3 is available invarious forms which may increase compressive strength andsurface resistivity.

7. According to transmission electron microscopy (TEM) nanographs, specimen maintained in CO2 gas as well as maintainedin the combined CO2 gas and saline water, no specific changesin phases are observed. Therefore, none of the conditions couldalter the nano structure of the specimens.

8. Hexagonal portlandite (Ca(OH)2) crystalline cubes observed inthe nano graphs, however, the congestion of crystals in thecement paste maintained in CO2 gas nano graph is relativelylower than that of cement paste maintained in the combinedcondition.

9. Corresponds to SAEDP, both cement pastes maintained in CO2

gas and the combined condition still show relatively crystallinestructure.

Acknowledgments

This study was funded by Concrete Technology & DurabilityResearch Center. We thank Eng. Alireza Ghahari for his supporting,Dr. Akbar Ghahari from SAFAS Corporation Company in USA andEng. Michael Ghahari from Adcolim Company in England for sup-porting this experiment and Dr. Keivandokht Ghahari from DWGermany for supervising the written manuscript. The authors aregrateful to Dr. Hesam Madani and Eng. Amirreza Pilvar for provid-ing the laboratories for durability tests. Special thanks go toMohammadMehdi HajiGhahremanzade from XRD center for theirhelp with the XRD analyses, and Atousa Khazaei from MicroscopyCenter for her support with the FESEM ad TEM analyses.

References

[1] Mehta PK. Reducing the environmental impact of concrete. Concr Int2001;23:61–6.

[2] Van Dam TJ, Smartz BW. Use of performance specified cements in Coloradotransportation projects: case studies. Washigton DC: Transportation ResearchBoard; 2010.

[3] Samples L, Ramirez J. Methods of corrosion protection and durability ofconcrete bridge decks reinforced with epoxy-coated bars. Phase 1, ReportFHWA/IN/JTRP-98-15. Purdue University; 1999.

[4] Shi X, Fay L, Peterson MM, Yang Z. Freeze–thaw damage and chemical changeof a portland cement concrete in the presence of diluted deicers. J Mater Struct2010;43:933–46.

[5] Sisomphon K, Franke L. Carbonation rates of concretes containing high volumof pozzolanic materials. J Cem Concr Res 2007;37:1647–53.

[6] Mehta PK, Monteiro JP. Concrete microstructure, properties and materials. 3rdedition. McGraw Hill; 2006.

[7] Wan X, Wittmann FH, Zhao T, Fan Hong. Chloride content and pH value in thepore solution of cement under carbonation. J Zhejiang Univ Sci A 2013;14(71–78).

[8] Kropp TH, Hilsdorf HK. The formation of silica gel during carbonation ofcementitous systems containing slag cements. SP114-69 – ACI J1989;114(1413-1428).

[9] Chang CF, Chen JW. Strength and elastic modulus of carbonated concrete. ACIMater J 2005;102:315–21.

[10] Jerga J. Physio-mechanical properties of carbonated concrete. Constr BuildMater J 2004;18:645–52.

[11] Dias WPS. Reduction of cement sorptivity with age through carbonation. J CemConcr Res 2000;30:1255–61.

[12] Chi JM, Huang R, Yang CC. Effects of carbonation on mechanical properties anddurability of concrete using accelerated testing method. J Mar Sci Technol2002;10:14–20.

[13] Verbeck GJ. Carbonation of hydrated portland cement. ASTM Special Tech Pub1958;205:17–36.

[14] Bouchaala F, Payan C, Garnier V, Balayssae JP. Carbonation assessment inconcrete by nonlinear ultrasound. J Cem Concr Res 2011;41:557–9.

[15] Geiker M, Nielsen EP, Herfort D. Prediction of chloride ingress and binding incement paste. J Mater Struct 2007;40:405–17.

[16] Bermudez MA, Alaejos P. Models for chloride diffusion coefficients of concretein tidal zone. ACI Mater J 2010;107:3–11.

[17] Conciatori D, Sadouki H, Bruhwiler E. Capillary suction and diffusion model forchloride ingress into concrete. J Cem Concr Res 2008;38:1401–8.

[18] Matschei T, Lothenbach B, Glasser FP. The AFm phase in portland cement. JCem Concr Res 2007;37:118–30.

[19] Dhir RK, Jones MR, Mc Carthy MJ. Quantifying chloride-induced corrosion fromhalf-cell potential. J Cem Concr Res 1993;23:1443–54.

[20] Saetta AV, Scotta RV, Vitaliani RV. Analysis of chloride diffusion into partiallysaturated concrete. ACI Mater J 1993;90:441–51.

[21] Cao HT, Baweja D, Roper H. Corrosion characteristics of steel in solutionsderived from cements and blended cements. J Cem Concr Res 1990;20:325–34.

[22] Kobayashi K, Shiraki R, Kawai K. Carbon concr Proc JSCE 1991;443:1–14.[23] Ihekwaba NM, Hope BB, Hansson CM. Carbonation and electrochemical

chloride extraction from concrete. J Cem Concr Res 1996;26:1095–107.[24] Feldman RF, Beaudoin JJ, Philipose KE. Effect of cement blends on chloride and

sulfate ion diffusion in concrete. J Il Cem 1991;88:3–18.[25] Tumidajski PJ, Chan GW. Effect of sulfate and carbon dioxide on chloride

diffusivity. J Cem Concr Res 1996;26:551–6.[26] Ho DWS, Lewis RK. The specification of concrete for reinforcement protection-

performance criteria and compliance by strength. J Cem Concr Res1988;4:584–94.

[27] Reddy B, Glass GK, Lim PJ, Buenfeld NR. On the corrosion risk presented bychloride bound in concrete. J Cem Concr Compos 2002;24:1–5.

[28] Neville A. Chloride attack of reinforced concrete; an overview. J Mater Struct1995;28:63–70.

[29] Tritthart J. Chloride binding in cement II. The influence of the hydroxideconcentration in the pore solution of hardened cement paste on chloridebinding. J Cem Concr Res 1989;19:683–91.

[30] America Society of Civil Engineers. Report card for Americas infrastructure.http://www.infrastructurereportcard.org/. [accessed 12.05.13].

[31] Yoon IS. Deterioration of concrete due to combined reaction of carbonationand chloride penetration: experimental study. Trans Tech Pub 2007;348–349:729–32.

[32] Glass GK, Page CL, Short NR. Factors affecting the corrosion rate of steel incarbonated mortars. Corros Sci 1991;32(12):1283–94.

[33] American Society for Testing and Materials. ASTM C-150 Standardspecification for portland cement; 2012.

[34] American Society for Testing and Materials. ASTM C-1202 Standard testmethod for electrical indication of concrete ability to resist chloride ionpenetration; 2012.

[35] American Society for Testing and Materials. ASTM C-39 Standard test methodfor compressive strength for cylindrical concrete specimens; 2012.

[36] Florida Method of Test. FM 5-578 Florida method of test for concrete resistivityas an electrical indicator of its permeability; 2004.

146 A.A. Ramezanianpour et al. / Construction and Building Materials 58 (2014) 138–146

[37] Ramezanianpour AA, Bahrami J. Influence of metakaolin as supplementarycementing material on strength and durability of concretes. Constr BuildMater 2012;30:470–9.

[38] Ramezanianpour AA, Esmaeili M, Ghahari SA, Najafi MH. Laboratory study onthe effect of polypropylene fiber on durability, and physical and mechanicalcharacteristic of concrete for application in sleepers. Constr Build Mater2013;44:411–8.

[39] Xiao J, Gou C, Jin Y, Wang Y. Effect of CaCO3 on hydration characteristics of C3A.J Center South Univ Technol 2010;17:918–23.

[40] Lopez-Arce P, Gomez-Villalbaa LS, Martinez-Ramirez S, Alvarez de Buergoa M,Forta R. Influence of relative humidity on the carbonation of calciumhydroxide nanoparticles and the formation of calcium carbonatepolymorphs. J PowderTechnol 2011;205:263–9.