Influence of Hydrophobic Coating on Freeze-Thaw Cycle Resistance of Cement Mortar

Due to the porous characteristics of cement-based materials, they are often corroded by salt solutions, which results in decreased durability, especially against damage under freeze-thaw cycles (FTCs). Improving surface properties is an effective way to improve the durability of these materials. In this study, a hydrophobic coating was applied to the surface of cement mortar by chemical modification of low surface energy materials. Fourier transform infrared spectroscopy (FT-IR) showed that low surface energy substances are linked to hydration products through chemical bonds. A water contact angle test indicates that the surface of cement mortar changed from hydrophilic (θ�14°) to hydrophobic (θ�140°) after chemical modification. .e cumulative water uptake of hydrophobic samples decreased by 90%. Meanwhile, the wear resistance of the hydrophobic coatings was excellent. Compared with the baseline sample, mass loss rate, flexural strength, and compressive strength of hydrophobic coating samples increased several-fold in the FTC test. Microstructural changes of the mortar were characterized by scanning electronmicroscopy. .e results show that a hydrophobic coating can significantly improve the freeze-thaw resistance of cement-based materials. .e formation of a hydrophobic layer on the surface of cement-based materials can improve their durability. .e research results not only have applications in civil engineering but will also have great impact in the restoration of historic structures.


Introduction
Concrete is a type of building material based on cement with highly effective mechanical properties; it is widely used as a structural material for buildings, bridges, undersea tunnels, etc. In ancient times, cementitious materials made from calcium hydroxide and clay were often used to build what have become today's world-famous historical buildings, such as the Pantheon in Rome. However, both modern and historical buildings are usually corroded by salt solutions, by which water penetrates the concrete, which is a factor contributing to concrete degradation. e freeze-thaw cycle (FTC) in severely frozen regions will cause sustained damage to concrete due to osmotic pressure, water expulsion, and inpore crystallization during the FTC process [1][2][3][4][5][6]. Researchers have proposed many methods to improve the FTC resistance of cement-based materials, such as adding air-entraining substances [7][8][9][10], pozzolanic minerals, or fiber admixtures [11][12][13][14][15][16][17][18][19][20][21][22][23][24]. e first method can relieve crystallization pressure in the FTC, while the latter method can improve the compactness of the concrete. However, the aforementioned methods lead to negative impacts on the concrete, such as deteriorating mechanical properties, difficult workability, and increased drying shrinkage.
Super-hydrophobic phenomena exist widely in nature [25][26][27][28][29][30][31][32]. Previous studies have pointed out that two basic requirements must be met for the surface of a solid material to be super-hydrophobic: (1) microscale and nanoscale rough structures and (2) lower surface free energy. Researchers have accurately expressed this theory through the Wenzel model [33] and Cassie-Baxter model [34].
Based on the above research, super-hydrophobic coatings have been applied to concrete surfaces for waterproofing, deicing, and self-cleaning [35][36][37][38][39][40]. Super-hydrophobic coatings can be prepared by bonding low surface energy materials to the concrete surface. Materials such as polytetrafluoroethylene (PTFE), polyether ether ketone (PEEK), and silanized diatomaceous earth (DE) are bonded to the concrete surface by epoxy resin to obtain a super-hydrophobic surface [41]. Besides, super-hydrophobic surfaces can also be obtained by bonding super-hydrophobic rice husk ash [42], paper sludge ash [43], or nanosilica gel [34] to the concrete surface. Another way to obtain superhydrophobic surfaces is the template method, where the features of micropillared molds made of polydimethylsiloxane (PDMS) are replicated immediately after demolding, and then siloxane-based compounds are sprayed to form a low energy surface [44].
Due to the excellent waterproofing effect of super-hydrophobic coatings, the water absorption of concrete decreases significantly, but the durability of such coatings is insufficient and they can easily fall off. Up to now [37], there has been a lack of research on the mechanical stability of super-hydrophobic coatings; therefore, the application of super-hydrophobic coatings in engineering is limited. To solve this problem, a vacuum impregnation process was adopted in this study. Such technology is more suitable for performance improvement of prefabricated concrete structures, similar to the anticorrosion treatment of steel structures.
rough this technology, low surface energy materials (iso-octyltriethoxysilane) can penetrate the cement mortar and combine with the cement hydration products, such as calcium hydroxide and ettringite, to form a continuous self-assembled molecular film layer. is molecular film layer reduces the surface energy of the mortar, thus achieving chemical modification of the rough surface of the mortar to form a hydrophobic coating. e wetting property was characterized by a water contact angle (WCA) test. A water absorption test and FTC resistance test were used to evaluate the protective effect of the hydrophobic coating on mortar blocks. Building structures are often subjected to external forces, which may result in surface wear and tear. us, the wear resistance was tested by sandpaper polishing under a certain pressure, after which the change of water absorption was tested. e microstructure of the cement mortar was characterized by scanning electron microscopy (SEM). Interface chemical reactions were characterized by Fourier transform infrared spectroscopy (FT-IR).

Materials.
Ordinary Portland cement (OPC) was used as the binding material in all mortar specimens. e chemical composition of the OPC is shown in Table 1. e aggregates were acquired from Xiamen ISO Standard Sand Co., Ltd., with particle diameters ranging from 0.5 to 2.0 mm. Iso-octyltriethoxysilane was acquired from Wacker Chemicals. Tap water was used in the preparation of mortar samples. Anhydrous ethanol was acquired from Cormio Inc, China. To ensure permeability, iso-octyltriethoxysilane was used as a low surface energy material for surface treatment with a concentration of 2%, with the remainder composed of ethanol (28%) and water (70%).

Preparation.
e proportions used to prepare the mortar samples and their properties after 28 days of curing are shown in Table 2. To ensure the uniformity of all mortar blocks, the following mixing process was adopted: (1) 450 g of cement and 225 g of water were added to the mixer and stirred for 60 s at slow speed; (2) 1350 g of sand was added evenly for 30 s; (3) the mixture was stirred at high speed for 30 s; (4) the mixer was stopped for 90 s for manual mixing; and (5) the mixture was further stirred for 60 s at high speed. After the concrete mixture was mixed, it was horizontally poured into a cuboid mold (40 mm × 40 mm × 160 mm) and a cubic mold (40 mm × 40 mm × 40 mm). After molding, all specimens were wet cured for 24 h (RH � 100% and T � 21 ± 1°C) and then demolded and cured for 28 days in water (21 ± 1°C). e cuboid specimen was used for the FTC test, and the cubic specimen was used for water absorption and wear resistance tests.
To ensure the effective penetration of modifiers, the hydrophobic coating on the surface of the mortar specimen was prepared using an osmosis vacuum degassing device ( Figure 1). After curing for 28 days, the sample was dried to a constant weight and placed in the vacuum tank. When the vacuum was below 20 kPa, the iso-octyltriethoxysilane solution was slowly infused into the vacuum tank until it inundated the entire sample. en, the hydrophobic coating was applied to the mortar surface after drying at 60°C for 12 h.

Test Methods.
e WCA was measured using a contact angle tester (KRUSS, K100, Germany). It was determined using deionized water (2.5 μL deposited with a micropipette) and by calculating the average of three measured values on the surface. e capillary water absorption test was used to quantify the ability of the concrete to absorb water by capillary suction. To evaluate the hydrophobicity, the baseline samples and hydrophobic samples were placed in water under atmospheric pressure (101 kPa) and vacuum (20 kPa) for two days each. e weights of the samples were recorded before and after immersion to calculate the mass uptake of water. An MPD-2 metallographic polisher (Shanghai Zhongyan Instrument Co., Ltd., China) was used to determine the wear resistance of the mortar. As the polisher rotates, the sandpaper begins to slide and creates friction with the surface of the sample under pressure, such that the microstructure of the sample surface will be destroyed. After polishing, the mass loss rate was calculated and the mass uptake of water was measured. 240-grit sandpaper was fixed on the turntable of the polisher, and the turntable rotated at 500 rpm. e longer the polishing time, the higher the wear degree of the test block. In this study, the lengths of polished mortar blocks indicate the degree of wear. e KDR-A rapid freeze-thaw circulator (Beijing Kangluda Test Instrument Co., Ltd., China) was used to determine the FTC resistance ( Figure 2). e sample was immersed in water for two days and then put into the rubber sleeve of the circulator, which was filled with water. e temperature cycle consisted of freezing and heating stages and took approximately four hours in total. During the freezing stage, the water temperature dropped from 5°C to − 17°C after 2.5 h. During the melting stage, the water temperature rose from − 17°C to 5°C after 1.5 h. With an increase in the FTCs, the damage degree of the samples increased. After FTC damage, the mass loss rates and flexural and compressive strengths of the samples were measured to evaluate their degrees of damage. e microscopic morphologies of the samples were observed by SEM (MAIA3, TESCAN, Czech Republic). Fourier transform infrared spectroscopy (FT-IR) spectra were acquired in the range of 400-4000 cm − 1 with an IR spectrophotometer (380FTIR, ermo Fisher Scientific, America). e flexural and compressive strengths of the mortar were evaluated on the same testing machine (SANS CMT5105, Shenzhen, China) at a loading rate of 2400 ± 200 N/s.

Wetting Properties and Water Contact Angle.
Without changing the surface microstructures of the mortar specimens, the surface of the mortar shows good hydrophobicity only through iso-octyltriethoxysilane solution modification, as shown in Figure 3(b). is phenomenon can be explained by the Wenzel theory that a hydrophobic surface can be obtained by modifying the rough mortar sample with low surface energy materials. First, iso-octyltriethoxysilane is hydrated to produce silanols (Si-OH). Secondly, silanol is combined with quartz sand, hydrated C-S-H gel, ettringite, and calcium hydroxide through -OH group reactions. Finally, the two -OH groups of iso-octyltriethoxysilane form Si-O-Si bonds by condensation while   Advances in Materials Science and Engineering releasing water. After the above reaction, iso-octyltriethoxysilane forms a continuous self-assembled molecular film on the surface of the hydrated products. Iso-octyltriethoxysilane contains a -CH 3 group and -CH 2 group which effectively reduce the surface energy of the cement mortar. erefore, the rough surface structure modified by low surface energy materials exhibits excellent hydrophobic properties. Figure 3(c) shows that the contact angle of the baseline sample is approximately 14°, indicating that the porous, rough surface of the mortar belongs to the hydrophilic surface. Figure 3(d) shows that the WCA of the modified mortar surface increases to 140°, which proves that a hydrophobic surface can be obtained by modifying the rough hydrophilic surface with low surface energy materials. e WCA of the modified mortar surface did not reach a super-hydrophobic state (θ > 150) because the surface roughness of the mortar itself did not conform to the Cassie-Baxter model. e fragile micro/ nanostructure is not conducive to the wear resistance of the coating itself. erefore, the obtained WCA (θ �140°) is sufficient to improve the waterproof performance of the mortar.

Water Absorption.
e influence of the hydrophobic coating on the water absorption of the mortar samples is shown in Figure 4. e results show that the cumulative water uptake of the baseline sample increased gradually from the beginning to achieve equilibrium and remained at a stable level thereafter, while the water absorption of the hydrophobic sample remained at a low level. After 15 days of immersion, the cumulative water uptake of the hydrophobic samples was reduced by 90%. is excellent waterproofing effect is equivalent to the waterproofing effect of a nanocomposite waterproof coating [45]. e microscale rough structure of the mortar surface is modified by the low surface energy material to reach a Wenzel state, which shows an excellent waterproofing effect. e unmodified samples retain the hydrophilic properties of the cement-based materials.

Wear Resistance and
ickness of the Hydrophobic Coating. Fragile micro/nanostructures on hydrophobic surfaces are susceptible to damage, leading to degradation of the hydrophobic properties. In this research, to test the wear resistance of the hydrophobic coating, the mortar sample was placed on a burnisher and polished with 240-grit sandpaper at 500 rpm, and then the water absorption was tested. e lengths of the mortar blocks after polishing indicate their degrees of wear. Table 3 lists the length reductions corresponding to different polishing times.
As shown in Figure 5(c), after polishing, the WCA (θ � 77°) of sample 5 decreased but remained between that of the baseline sample as shown in Figure 5(a) and that of the unpolished hydrophobic sample as shown in Figure 5(b). Because the thickness of the chemically modified mortar reaches the range of 1-3 mm, the hydrophobic sample remains hydrophobic even after the rough surface structure is destroyed. In the next section, the mass loss and water absorption after polishing are discussed. Figure 6(f ) shows that the mass loss rate of the samples increases with an increase in length reduction. e mass loss rate of the sample is the largest, and its WCA (θ � 77°) is still significant. Samples 1, 3, and 5 were immersed in water for 15 days with the baseline sample ( Figure 6(a)) and the unpolished hydrophobic sample (Figure 6(b)). When the samples were immersed in water, many bubbles were observed on the surface of the baseline sample (Figure 6(a)). In contrast, we could see only a few bubbles on the surface of the hydrophobic sample (Figure 6(b)), even if it was reduced by over 10 mm by the sandpaper (Figure 6(e)). Surface bubbles are formed when water enters the sample and displaces air from the sample. A few bubbles on the surfaces of the polished samples show that the hydrophobic coating maintains an excellent waterproofing performance even after polishing. In the cumulative water uptake test, it was also proved that the hydrophobic coatings have excellent wear resistance, as shown in Figure 7.
In Figure 7, the water absorption curve shows that the cumulative water uptake of the polished hydrophobic samples remained at a low level. is phenomenon shows that the wear resistance of the hydrophobic coating is outstanding. e surface of the hydrophobic coating is worn after polishing, which leads to decreased WCA, but it still maintains excellent waterproof properties. is phenomenon can be explained by Figure 8. Figure 8 illustrates the thickness of the hydrophobic coating by wetting the cross section with water. In this figure, the light hydrophobic coating can be observed continuously around the perimeter of the dark central area, which indicates that a continuous hydrophobic coating was formed on the surface of the sample by vacuum impregnation. e thickness of the hydrophobic coating is within the range of 1-3 mm.
is is a reasonable explanation for the hydrophobic properties of the polished hydrophobic samples decreasing after polishing but the waterproof properties remaining excellent. When iso-octyltriethoxysilane penetrates the mortar sample, self-assembled membranes will form on the surface of the hydrated particles. As the interior of the mortar is rough and porous, a stable hydrophobic network structure with a certain thickness is obtained. In this way, even if the surface of the microscale rough structure is destroyed, the network structure can still play a perfect waterproofing role. Figure 9, steep curves for the mass loss rate and flexural and compressive strengths were observed for the baseline sample after the FTC tests. On the contrary, the corresponding curves for the hydrophobic samples change more smoothly.

FTC Resistance Analysis. As shown in
is phenomenon shows that the baseline sample was severely damaged after the FTCs, while the hydrophobic sample was much less damaged due to the protection of its hydrophobic coating. After 36 FTC tests, the mass loss rate of the baseline sample was approximately 48.0 wt.%. Meanwhile, the flexural and compressive strengths of the baseline sample were reduced to 0.3 MPa and 11.0 MPa, respectively. After 36 FTC tests, the mass loss rate and flexural and compressive strengths of the hydrophobic sample were 0.8 wt.%, 7.5 MPa, and 38.2 MPa, respectively. After 48 FTC tests, the baseline sample lost its original morphology and size ( Figure 10) because its mass loss rate was over 62 wt.%. In contrast, the original morphology and size of the hydrophobic sample remained after 72 FTC tests ( Figure 10). e test results show that the mass loss rate and flexural and compressive strengths of the hydrophobic sample are 10.4 wt.%, 1.0 MPa, and 16.5 MPa, respectively, which are very close to the values of the baseline sample after 24 FTCs. e test results indicate that the hydrophobic coating not only has an excellent waterproofing effect but it also has excellent anti-FTC performance. Due to the protective effect of the hydrophobic coating, water cannot impregnate the sample, thus alleviating damage from FTCs. Exfoliation of the surface coating of the hydrophobic sample occurred at the 36 th cycle. is is because the sample had been immersed in water at below 0°C. e sample was encapsulated by external ice, which produced certain stresses and destroyed the hydrophobic coating. It is predicted that increasing the hydrophobic coating thickness will effectively improve the FTC resistance of the samples. is will be further explored in future research. e improvement of frost resistance was also observed in concrete modified with metakaolin and Advances in Materials Science and Engineering nanoparticles [46]. is is an entirely different technical concept from the hydrophobic coating, but it can be combined to achieve better frost resistance in future research.

Microscopic Analysis and Chemical Characterization.
To study the effect of the FTC on the internal structure of the mortar, the sectional microstructures of the hydrophobic and the samples were compared after enduring FTC damage ( Figure 11). As shown in Figure 11(a), after the 12 th FTC test, a visible crack could be seen on the baseline sample. Moreover, the crack was further extended due to the continuous damage from FTC after the 36 th FTC test. e width of the crack increased from 2.1 μm to 4.3 μm (Figure 12(b)). Significantly, only cracks less than 1 μm (Figure 11(c)) could be observed on the hydrophobic sample after the 12 th FTC test. After the 48 th FTC test, a crack becomes evident with a width of approximately 1 μm (Figure 11(d)). As exhibited in Figure 11(e), the width of the crack is still below 4 μm after the 72 nd FTC test. Figure 11(f ) displays the microstructure of the hydrophobic coating on the mortar block after enduring 72 FTCs, indicating that the hydrophobic coating could effectively reduce the damage to the mortar caused by FTCs. Based on this, we can confidently predict that the FTC resistance of the mortar block would be improved by increasing the thickness of the hydrophobic coating.
As shown in Figure 12(a), the hydrophobic coating appeared after the sample section was wet with water. In Figure 12(b), by magnifying by 20,000 times, the SEM images show that the interior of the mortar is filled with acicular or flaky hydration products.
is micron-scale rough structure is one of the conditions for the formation of the hydrophobic coating. e chemical modification of the     Advances in Materials Science and Engineering hydration products was characterized by the FT-IR. As shown in Figure 12(c), the FT-IR wavenumbers ranged from 3750 cm − 1 to 1000 cm − 1 . e absorption peak at 3644 cm − 1 was attributed to the stretching vibration of -OH from Ca (OH) 2 . e -OH stretching vibration peaks of Ca(OH) 2 were only observed in the baseline sample. is indicates that hydroxyl groups are consumed in the reaction of calcium hydroxide with iso-octyltriethoxysilane. Absorption peaks at 2970 cm − 1 and 2920 cm − 1 were observed in the hydrophobic coating, corresponding to -CH 3 and -CH 2 groups, respectively, implying that chemical bonds are formed between the coating and cement hydration products. e hydrophobic coating peak at 1130 cm − 1 corresponds to the Si-O-Si group, which shows that continuous self-assembled molecular films are formed on the surface of hydration products.

Conclusions
is study aimed to transform a porous, hydrophilic cement mortar surface into a hydrophobic surface by chemical modification.
rough hydrolysis and condensation, isooctyltriethoxysilane forms continuous self-assembled molecular films on the surface of hydrated products, thus producing a hydrophobic coating with a thickness of 1-3 mm on the surface of the mortar. e WCA of the hydrophobic coating was 140°, and it had good waterproofing and wear resistance. A water absorption test showed that the cumulative water uptake of the hydrophobic samples decreased by 90%. Compared with a baseline sample, the mass loss rate and flexural and compressive strengths of the hydrophobic sample increased several-fold in the FTC test stages. Chemical bonding between iso-octyltriethoxysilane and cement hydration products ensures excellent wear resistance of the hydrophobic coating. In conclusion, the hydrophobic coating prepared by vacuum impregnation has an excellent protective effect on cement-based materials, and this technology has a wide range of applications in the building industry. Future work will focus on the preparation of hydrophobic coatings with higher thicknesses and better waterproofing performances via a simpler process.

Data Availability
All the data in this study are original.

Conflicts of Interest
e authors declare that they have no conflicts of interest.