Carbonation of Water Repellent-Treated Concrete

Water repellent treatment has been considered an e1ective preventive method against water and aggressive ions penetration into concrete and consequently can improve the durability of concrete structures. In reality, many concrete structures are exposed to conditions with high risk of carbonation. In this contribution, one type of ordinary concrete had been prepared and surface impregnated by 400 g/m silane cream and 100 g/m and 400 g/m silane gel. In addition, integral water repellent concrete was produced by adding 2% silane emulsion. +en, the specimens were exposed to accelerated carbonation for 7, 28, and 72 days. +e e1ect of water repellent treatment on carbonation of concrete has been investigated.+e results indicate that surface impregnation reduced carbonation depth of concrete under RH 70%, but integral water repellent concrete increased carbonation. Carbonation reaction started behind the hydrophobic layer in the surface-impregnated concrete. +e coe;cient of carbonation can be described better by a hyperbolic function of time. Treatment by 400 g/m silane gel and silane cream showed better e;ciency on reducing carbonation than usage of 100 g/m. Coe;cient of water capillary suction was decreased signi=cantly by both surface impregnation and integral water repellent treatment. It is an e1ective method to protect concrete from water penetration into the material.


Introduction
Ordinary cement concrete is a cheap and versatile building material which has been applied for nearly 200 years with great success and in most diverse structures.It is hardly possible to imagine the construction of the infrastructure in all industrialized nations today without concrete.But pure cement concrete is rather brittle with a small ultimate strain capacity. is often leads to early crack formation.Moreover, cement concrete is porous, which allows aggressive substances to migrate into the material.is is at the origin of early deterioration of concrete in aggressive environment and of limited protection of embedded steel reinforcement.
Moisture transport in cement-based materials is an important physical process with direct in uence on service life and durability.If the surface of concrete is in direct contact with water-soluble compounds, hydration products may be washed out by leaching.In this way, near-surface zones are weakened and become more frost sensitive.If the liquid in contact with the surface is an acid, this deteriorating process is even accelerated.In case water gets in contact with a predried surface of concrete, it will be absorbed quickly and deep into the pore space.Aggressive compounds dissolved in water can be transported deep into the pore space of concrete.Drying of concrete always imposes shrinkage of the dry zones and leads to crack formation.Gases of the environment may be dissolved in the aqueous pore solution of concrete.In this way, acids are formed, for instance, by dissolution of CO 2 , which could react rapidly with the hydration products of cement.
It is obvious that all these deterioration mechanisms act from the surface of concrete.roughout history, a range of protective materials have been applied on the exposed surfaces of structural concrete elements to prevent the ingress of water.Nowadays, water repellent treatment has proved to be e ective to prevent concrete from water penetration and consequently can improve the durability of concrete structures [1][2][3][4][5][6][7][8].Application of hydrophobic treatment with an approved product has even become standard practice on new bridge decks in the Netherlands [9].For broader information about the studies on water repellent treatment, the proceedings of the Hydrophobe conference series (Hydrophobe I-VIII) can be referred [10][11][12][13][14][15][16][17].
When applying the water repellent agents on actual structures, however, many concrete surfaces are exposed to carbonation condition, and some had been carbonated already before the surface treatment.Heinrichs et al. [18] and Gerdes [19] reported that water repellent treatment alters the drying attitude of concrete.is in turn mainly in uences the process of carbonation.ey had simulated the transport, storage, and chemical reaction of the main in uencing variables for carbonation of treated concrete.It was believed that the carbonation process can be delayed by surface impregnation under proper conditions.e study from Courard and Lucquiaud [20] also showed that the application of a water repellent delayed the propagation of the carbonation front.However, there is some concern that the gradual drying out due to hydrophobic treatment causes deeper carbonation [9].Besides, very little study has focused on the in uence of integral water repellent treatment on the carbonation process of concrete.erefore, it has to be clari ed if the cement-based materials could be protected by water repellent treatments any more when carbonation reaction occurs simultaneously.
In this contribution, one type of ordinary concrete with a water-to-cement ratio of 0.5 has been produced.Two types of water repellent treatment were applied.One was surface impregnation.e surface of the samples was covered by di erent amounts of silane, in this study, 400 g/m 2 silane cream and 100 g/m 2 or 400 g/m 2 silane gel.Another type was integral water repellent treatment by adding 2% of silane emulsion in fresh concrete during mixing.All surfaceimpregnated and integral water repellent concrete were exposed to accelerated carbonation for 7, 28, and 72 days.In this way, the e ect of water repellent treatment on carbonation of concrete has been investigated.Another emphasis had been concentrated on the water capillary absorption behavior of water repellent-treated concrete after carbonation.

Basic Mechanism of Water Repellent Treatment
ere are basically two di erent technologies to make concrete water repellent: (1) surface impregnation and (2) integral water repellent treatment.In the rst case, the surface of hardened concrete is impregnated with a reactive agent, and in the second case, the reactive agent is added to the fresh mix.Di erent organic compounds have been used in the past as water repellent agents.Historically, natural polymers such as oils, fats, and waxes have been used to protect statutes and buildings.Today, in many cases, silanes are applied to make concrete or stones water repellent [15,16].e general composition of higher silanes can be written as H-(SiH 2 ) n -H with n 1, 2, 3, . ... Similar to alkanes, silanes with more than three Si atoms can form structural isomers.In the presence of water, silanes react to form siloxanes and nally silicon resin.
e resin forms a network, which is rmly xed on the huge surface of the hydration products of Portland cement.In Figure 1, the replacement of an adsorbed water layer by the silicon resin network is shown schematically.
In the case of surface impregnation, liquid silane penetrates into the pore space of concrete by capillary action.In order to achieve a su cient penetration depth, the contact time between the concrete surface and the liquid silane must be long enough [21][22][23].On a horizontal surface, this can be achieved easily, but vertical surfaces have to be sprayed or brushed with liquid silane for several times.is problem can be overcome by applying a so-called cream, a silane emulsion which can be applied once with the necessary thickness or silane gel which is a suspension of ne clay articles in silane.e gel is sti enough to be applied with the necessary thickness.If integral water repellent concrete is to be produced, silane emulsion can be added to the fresh mix.After hardening and initial drying, the emulsion breaks and silicon resin is formed on the pore walls.In principle, the distribution of silicon resin in integral water repellent concrete should be uniform all over the volume.is is, however, not automatically achieved.
e e ect of water repellents is essentially based on their low surface tension.
e behavior of water when contacting the surface of a material is governed by the surface tension which can be measured by the contact angle.e intensity of the water repellent property is associated with the contact angle between water and the treated surface.Contact angles of a water droplet of more than 90 °represent hydrophobic property while less than 90 °hydrophilic property.
e organofunctional groups will reduce the critical surface tension of the material surface and thus provide hydrophobicity.

Materials and Preparation of Specimens.
e composition of the concrete matrix has been the same for all specimens used in this project, which is shown in Table 1.Portland cement 42.5 type II with a compressive strength at 28 days of no less than 42.5 MPa, local crushed granitic aggregates with a maximum diameter of 20 mm and density of 2620 kg/m 3 , and river sand with a maximum grain size of e waterto-cement ratio was 0.5.To improve workability, 0.5% superplasticizer of naphthalene type, related to the mass of cement, had been added to the fresh concrete.In addition, some integral water repellent concrete specimens were prepared with addition of 2% (related to the mass of cement) silane emulsion (octyltriethoxysilane) into the fresh concrete during mixing.It shall be noticed that the compressive strength of integral water repellent concrete at 28 days was reduced by approximately 11%, from 40.1 MPa for reference-untreated concrete to 35.7 MPa.
Prismatic specimens with the dimensions of 100 × 100 × 400 mm 3 and cubic specimens with the edge length of 100 mm were produced.All specimens were compacted in steel forms and cured for one day before demoulding.
en, the specimens were stored in the curing room at a temperature of 20 ± 2 °C and relative humidity of 95% until an age of 14 days.After that, all specimens were taken out of the curing room and further stored under laboratory conditions at 20 ± 2 °C and 60% RH for another 7 days before applying surface impregnation.

Water Repellent Treatment.
After drying in the lab for 7 days, all the samples were divided into 5 groups.One group was maintained as reference specimens, that is, without water repellent treatment.One group was integral water repellent concrete, as produced above.
e other three groups of concrete specimens were treated by di erent types and amounts of water repellent agents, as listed in Table 2.When one of the moulded surfaces was treated by covering 400 g/m 2 of silane cream, the series was named C400, while the specimens treated by 100 and 400 g/m 2 of silane gel were called G100 and G400, respectively.After surface impregnation, the specimens were exposed to laboratory atmosphere again for another 7 days in order to allow su cient polymerization of silane.en, both water repellent-treated and -untreated concrete specimens were ready for accelerated carbonation test.

Accelerated Carbonation.
All the water repellenttreated and -untreated specimens were then submitted to accelerated carbonation for 7, 28, and 72 days.According to the method in Chinese standard [24], the concentration of CO 2 gas had been maintained constant at 20 ± 2%, relative humidity in the carbonation box was about 70%, and the temperature was 20 ± 3 °C.Four surfaces except for the treated surface and its opposite surface of concrete specimens had been sealed with wax before placing in the carbonation box.After 7, 28, and 72 days, carbonation depth for each sample was determined by spraying phenolphthalein solution of 1% in ethanol.It was measured every 10 mm along a side length of 100 mm.Except for the two values from the border, the average value of the other nine measurements was the carbonation depth of each sample.For each case, three specimens were measured.en, the average value of these three samples represented the carbonation depth for each case.

Water Capillary Absorption.
In addition, in order to study the e ect of water repellent treatment on water capillary absorption of concrete before and after carbonation, the water capillary absorption test has been performed according to the partial immersion method as explained in the ISO standard [25].
e specimens were cut into two halves rst and then were dried in an aerated oven at a temperature of 50 °C for 7 days until mass equilibrium.When the specimens cooled down to room temperature, they were put in contact with water.e amount of absorbed water was then measured by weighing the specimens at di erent time.In this test, it was at 0.5, 1, 2, 4, 8, and 24 hours in the rst day and then measured every day until 7 days or so.

Carbonation in Water Repellent-Treated and -Untreated
Concrete.Carbonation depth in water repellent-treated and -untreated concrete has been measured after exposure to 7, 28, and 72 days of carbonation.e results are shown in Figure 2. It is obvious that surface-impregnated concrete had lower carbonation depth than untreated samples.But the integral water repellent concrete has comparatively higher carbonation depth.
e possible reason is that the addition of silane emulsion can retard the hydration process of cement [26]. is also explains why Advances in Materials Science and Engineering the compressive strength of integral water repellent concrete was reduced slightly.In this case, the penetration of carbon dioxide would be quicker and then cause higher carbonation depth.
For the surface-impregnated concrete, application of 400 g/m 2 silane cream and silane gel reduced approximately one-half of the carbonation depth compared to the reference-untreated samples.For instance, the carbonation depth of untreated concrete was 7.7 mm after 7 days of carbonation.After surface impregnation, carbonation depth of the samples C400 and G400 was 3.1 mm and 4.1 mm, respectively.e e ciency of reducing carbonation action is better than 100 g/m 2 covering usage of silane.
After surface impregnation, the silane penetrated by capillary transport into the concrete.Consequently, the silicon functional group reacts with water and yields a reactive silanol group (hydrolysis stage); further, condensation by cross linking to the hydroxyl group forms polysiloxane as the active water repellent lm which is linked to the inorganic substance by way of covalent siloxane bounds.
e rate of carbonation process depends very much on the moisture content of concrete.
e neutralization between CO 2 gas and calcium hydrate or C-S-H gel requires the presence of water because carbon dioxide needs to dissolve in water rst forming carbonic acid.If the concrete is too dry, CO 2 gas may enter the material rapidly.But it cannot dissolve, and no carbonation occurs.If, on the other hand, it is too wet, CO 2 is not able to enter the pore and no carbonation occurs, either.
e optimal conditions for carbonation occur at a relative humidity of 50%-80% [27][28][29].In this test, all the treated and untreated specimens were kept in carbonation box under relative humidity of 70%.After 7, 28, and 72 days, the carbonation depth had been measured by spraying the phenolphthalein solution on the fractured surface.e typical results are shown in Figure 3.During the measurements, it has been found that surface-impregnated specimens G400 had three di erent layers, as shown in Figure 3 (the left photo).e rst layer was pink.
is means the near surface region was not carbonated as the water content in this layer was very low [30].Carbonation reaction did not occur in this hydrophobic layer as it is too dry.But CO 2 gas certainly migrated through this layer and reached the inner part where there was necessary water for carbonation reaction.us, the second layer without colour indicates that the carbonation process started behind the water repellent layer.However, the carbonation process in this untreated region was delayed because its drying rate was lowered by the hydrophobic layer, and consequently, a high content of water in the pores hindered the carbonation process.
erefore, the carbonation depth in surface-impregnated concrete was reduced in comparison with the reference untreated concrete.
e carbonation mechanism has been studied by many researchers, and various mathematical models have been developed with the purpose of predicting carbonation depth [31][32][33][34].For the sake of simplicity, however, the carbonation kinetics can be described as a rst approximation to be a di usion process of CO 2 through porous material with given moisture content.By applying Fick's law, (1) can be found after integration: where x is the carbonated depth at time t, D is the e ective di usivity of CO 2 , c is the CO 2 concentration in the atmosphere, and d is the concentration of the reactive compounds.Di usivity D is not a constant but a function of numerous parameters.It mainly depends on time (degree of hydration), location (curing e ect), and on the moisture content.Equation ( 1) can be written in the wellknown form as where K denotes the coe cient of carbonation.Assuming the coe cient of carbonation rate changes with carbonation time, K can be described by a hyperbolic function, as shown in (3), in which P 1 , P 2 , and P 3 are parameters tted from the experimental results.Another exponential model can also be used to describe the decrease of carbonation coe cient, as shown in (4), in which a and b are parameters tted from the experimental results: It has been learned already from Figure 2 that with increasing time, the carbonation depth increases.e relationship between carbonation depth x and the square root of time t is shown in Figure 4.According to (3) and ( 4), the coe cient of carbonation for water repellent-treated and -untreated concrete can be calculated.e results are shown in Figure 5.It can be seen from the results that the carbonation coe cient was decreased signi cantly by surface impregnation.With the increasing time, nally the coe cient values would reach constant.is proves again that

Advances in Materials Science and Engineering
surface impregnation can slow the carbonation process and consequently reduce carbonation depth.Application of 400 g/m 2 silane cream and 400 g/m 2 silane gel shows better e ciency than 100 g/m 2 silane gel.Comparing the two models, it can be found that the exponential model achieves better agreement with the experiment results.

Water Capillary Absorption of Treated and Untreated
Concrete after Carbonation.If the surface of a porous material is put in contact with a liquid, the liquid will be absorbed by capillary action.Within a limited range of time and as rst approximation, the amount of absorbed water as function of duration of contact time can be expressed by where ΔW is the amount of absorbed water; A stands for the coe cient of capillary suction (g/m 2 h 0.5 ), which depends both on the properties of the absorbed liquid and the porous structure; and t is the absorption time.
If the microstructure of concrete changed due to carbonation of CH or C-S-H gel, or the surface of capillary pores was water repellent treated, the capillary suction would be changed consequently.e coe cient of capillary suction is considered to characterize this change.In this 6 Advances in Materials Science and Engineering test, this value for each type of specimen had been obtained experimentally.e results of capillary suction for all the water repellenttreated and -untreated concrete, which were imposed to accelerated carbonation for 7 and 28 days, are shown in Figure 6.With (5), the coe cient of capillary suction within the rst 72 hours can be calculated.e results are compiled in Table 3.
e results clearly indicate that both surface impregnation and integral water repellent treatment can reduce considerably the water absorption of concrete.
In the surface-impregnated concrete samples C400, G100, and G400, the silane molecules (d ≈ 1-3 nm) penetrated into concrete microstructures and made the surfaces of capillary pores hydrophobic.e coe cient of capillary suction was reduced enormously compared to the reference untreated concrete.With respect to the integral water repellent concrete IW, the coe cient of capillary suction also decreased.But the reduction is relatively lower.e origin of this phenomenon is due to the size of silane emulsion.Because the particle size of silane emulsion is approximately 100 nm, nanopores of C-S-H gel are not penetrated.Consequently, the ne pores will not become water repellent by the addition of silane emulsion.If the surface is in contact with water, some water can enter the porous structure of cement-based materials.
It shall be noticed that the carbonation resistance of integral water repellent concrete was lower than the reference untreated concrete mainly because the addition of silane emulsion delayed the hydration process of cement as mentioned previously.But this would not change the situation that the particle of silane emulsion still can break and form a siloxane network on the internal surface of some capillary pores in hardened cement paste.erefore, the water capillary absorption can be still reduced by internal hydrophobic treatment in comparison with the reference untreated concrete.
After exposure to 7 and 28 days of carbonation, water absorption of concrete was also reduced by surface impregnation and integral water repellent treatment.is implies that water repellent treatment can still protect concrete from water absorption e ectively even under carbonation condition.But the carbonation reaction can destroy the hydrophobic lm inside the capillary pores and consequently may cause slightly more water absorption.For instance, the coe cient of capillary suction of the surfaceimpregnated concrete G400 was 15.1 g/m 2 h 0.5 before carbonation.While after 7 days of carbonation, it became 29.0 g/m 2 h 0.5 .However, after 28 days of carbonation, water capillary absorption became lower again since the carbonation products lled the pores in the near-surface zone.

Conclusions
Based on the results presented herein, the following conclusions can be drawn: (1) Carbonation depth of concrete can be reduced by one-half by surface impregnation.But for integral water repellent concrete, carbonation depth increased compared to the untreated concrete.One has to be careful while using integral water repellent concrete under severe carbonation atmosphere concerning steel corrosion.(2) ree layers with di erent colours can be found when determining the carbonation depth of surfaceimpregnated concrete by means of phenolphthalein solution.CO 2 migrated through the impregnated layer quickly without carbonation as it is very dry.Carbonation process started behind the hydrophobic layer in this case.(3) e coe cient of carbonation can be described better by an exponential function than by a hyperbolic function of time.Surface impregnation by silane cream and silane gel with the amount of 400 g/m 2 has better e ciency on reducing carbonation than using 100 g/m 2 silane gel.(4) Coe cient of water capillary absorption of concrete can be reduced signi cantly by both surface impregnation and integral water repellent treatment, even under carbonation condition.It is an e ective method to protect concrete from water penetration despite carbonation reaction.
It must be noticed that in this study accelerated carbonation test was performed with relatively high CO 2 concentration as the natural carbonation process was too slow.ere might be di erent trends between the two types of test. is has to be further clari ed.

Figure 1 :
Figure 1: Schematic representation of (a) adsorbed water layer on the surface of hardened cement paste and (b) silicon resin xed on the surface replacing the water layer.

Figure 6 :
Figure 6: Water capillary absorption of water repellent-treated and -untreated concrete before and after 7 and 28 days of carbonation.(a) No carbonation.(b) 7 days of carbonation.(c) 28 days of carbonation.

Table 1 :
Composition of the concrete used in this project.

Table 2 :
e type of water repellent agents used in this project and their amounts.