The Synergistic Effect of Water Reducer and Water-Repellent Admixture on the Properties of Cement-Based Material

Abstract

Water reducer and water-repellent admixture are very important in improving the workability and durability of cement-based materials. However, the synergistic effect of the two types of admixtures has not been well investigated. In this study, polycarboxylate ether-based superplasticizer (PCE) and octyltriethoxysilane (OTS) were adopted as water reducer and water-repellent admixture, respectively. Their synergistic effect on the fluidity, compressive strength, and water absorption rate of cement-based materials was investigated. Particularly, the pore structure and hydration state of cement paste were analyzed using ¹H Low-Field Nuclear Magnetic Resonance (¹H LF NMR). The result showed that the fluidity of cement paste containing different dosages of PCE was reduced by 5–10 mm by incorporating 1% OTS, and the compressive strength at the early age of 3 d of mortar containing high PCE dosage of 0.25% decreased up to 15% by using 1% OTS. In contrast, the compressive strength of mortar containing 0.20% PCE was slightly enhanced by the addition of 1% OTS. ¹H LF NMR analysis revealed that the combination of PCE and OTS would increase the pore size and total pore volume of cement paste, and more bleeding water would be generated at high PCE dosage. The intensity-weighted T₂ values of the main peak ($\overline{T_2}$) implied that both PCE and OTS produced a retardation effect on cement hydration. However, the water absorption rate decreased by 46.6% despite the increase in pore size and total pore volume. The conflict phenomenon powerfully revealed that the internal hydrophobic treatment by OTS has been successfully achieved. Overall, the combination of 0.20% PCE and 1% OTS exerted a positive synergistic effect in improving the compressive strength and water-repelling ability of cement-based materials, which is meaningful for improving their durability and service life.

1. Introduction

Cement-based materials (CBM) are the most widely used building materials all over the world, depending on their superior advantages of versatility, high strength, and relatively low cost. During recent decades, great efforts have been made to improve the micro–macro properties of CBM to meet the increasing requirements in construction projects [1,2,3,4]. Regarding the hydrophilic and mesoporous properties of CBM [5,6,7,8], special attention has been paid to increasing their durability and service life, which declined due to the penetration of water and aggressive ions (magnesium, chloride, sulfate, etc.) into their porous microstructures and led to steel bars corrosion, leaching of hydration product, and freeze–thaw action [5,9,10,11,12,13]. The most effective method has been established, which is using silane-based compounds as water-repellent admixtures. Since the alkoxy group can easily hydrolyze in the alkaline environment of CBM, chemical bonds are formed between the silicate of cement and silane-based compounds; thus, a hydrophobic layer on the surface of cement particles is produced, which helps prevent the ingression of water and aggressive ions [7,14,15,16,17,18,19,20,21]. Wang et al. used polydimethylsiloxane (PDMS) for the hydrophobic modification of mortar. They found that the water absorption reduced approximately 92.51%, and the contact angle increased to 157.3° of the modified mortar [17]. Raja et al. studied the effect of octyltriethoxysilane, isobutyltriethoxysilane (IS), and waterborne pure nano-silicone emulsion (NS) as water-repellent admixtures on the water absorption and mechanical properties. The results exhibited that water absorption significantly reduced in the modified samples [22].

There are two main methods to apply silane-based water-repellent admixtures. The first way is the direct surface hydrophobic treatment [6,16,23,24,25,26]. This method can be applied to both new and old CBM surfaces. However, the hydrophobic property of CBM will be lost due to cracks, coating aging, and peels [9,14,17]. The second way is the internal hydrophobic treatment by incorporating silane-based water-repellent admixtures during the mixing procedure of fresh CBM to endow an integral hydrophobic property [9,14,17]. In general, the advantage of the internal hydrophobic treatment is that CBM is hydrophobic even with exposure to deterioration factors such as wear or cracking during service life [14,17,18,19].

Silane-based water-repellent admixtures can be used in the forms of dry powder [27,28] and liquid [16,17]. The most commonly used silane-based water-repellent admixtures include octyltriethoxysilane (OTS) and isobutyltriethoxysilane [20,29,30]. OTS is unique in the hydrophobization of CBM depending on its longer alkyl chain and excellent stability, which leads to destroying the original spatial correlation and weakening the capillary adsorption [31]. Furthermore, at room temperature, OTS is in liquid form that can be utilized directly or converted into an environmentally friendly water-based emulsion [6]. Studies showed that the use of OTS can increase the contact angle, reduce capillary water absorption, and improve the durability of CBM [20,32,33,34]. Xue et al. synthesized a water-based hydrophobic agent with OTS and then impregnated the surfaces of mortar and concrete. The result showed that the capillary water absorption of treated samples decreased by 94.6% compared to the un-modified one [30]. However, K. Grabowska et al. [32] reported that the compressive strength of mortar decreased by 15% using OTS. Y.G. Zhu et al. [34] used OTS for the internal hydrophobic treatment of recycled aggregate concrete. They reported that the compressive strength of concrete was 38% lower than that of un-modified samples. Specifically, the porosity increased at high OTS dosages [32,34,35].

Using a water reducer can be regarded as an effective way to increase the mechanical strength of CBM by effectively reducing the water-to-cement ratio. Polycarboxylate-based ester/ether superplasticizer (PCE) is widely used in CBM as a water reducer, which not only lowers water consumption but also improves workability, reduces cement consumption, increases mechanical strength, and enhances durability [36,37]. Several studies reported that PCE modified by alkoxysilane exhibited good dispersion, flowability retention, and low sulfate sensitivity in CBM [38,39,40]. C.A. Casagrande et al. [40] studied the effect of partial substitution of PCE by silane-based water-repellent admixtures in cement paste. They reported that using tetraethoxysilane (TEOS) for substitution could accelerate cement hydration as well as improve its workability and mechanical strength. However, aminoethylaminopropyl-trimethoxysilane (AEAPTMS) and 3-glycidoxypropyltrimethoxysilane (3-GPTMS) exhibited opposite trends. To the best of our knowledge, the synergistic effect of water reducer and water-repellent admixture on CBM has not been well studied so far.

In this work, a self-synthesized PCE was adopted as a water reducer, and OTS was utilized as a water-repellent admixture to achieve internal hydrophobicity. The synergistic effect of PCE and OTS on CBM was studied by evaluating the fluidity of cement paste, the compressive strength, and the water absorption rate of mortar. Moreover, the ¹H Low-Field Nuclear Magnetic Resonance (¹H LF NMR) was employed to investigate the pore structure and hydration state of cement paste. Through this research, a comprehensive understanding of the synergistic effect of water reducer and water-repellent admixture on CBM can be drawn, which is meaningful to utilizing water-repellent admixture as well as improving the durability of CBM.

2. Materials and Methods

2.1. Materials

To reduce the interference of paramagnetic materials on the ¹H LF NMR test, a white Portland cement (P·W 42.5) with a low content (0.35%) of ferric oxide was used in this study. The oxide composition of cement was determined by an X-ray fluorescence spectrometer (PANalytical Axios), and the result is shown in

Table 1. The oxide composition of P·W 42.5 used in this study (%).

CaO	SiO2	MgO	SO3	Al2O3	K2O	Fe2O3	Na2O	P2O5	Others
60.1	18.2	13.5	2.5	2.2	0.5	0.3	0.3	0.1	2.3

. Self-synthesized PCE was used as a water reducer and characterized with gel permeation chromatography (GPC) by a Waters Alliance 2695 instrument (Waters, Eschborn, Germany), and the result is shown in

Table 2. Molecular masses (M_w and M_n), polydispersity index (PDI), and conversion rate of the macromonomer for the self-synthesized PCE.

Mw (Da)	Mn (Da)	PDI (Mw/Mn)	Conversion Rate of the Macromonomer (%)
58,910	24,980	2.4	90.0

. Octyltriethoxysilane (OTS) in chemical pure grade purchased from Macklin Co., Ltd. (Shanghai, China) was used as a water-repellent admixture. The structures of PCE and OTS are shown in Figure 1. The standard sand was supplied by Xiamen ISO Standard Sand Co., Ltd., Xiamen, China. The tap water was used to prepare cement paste and mortar.

2.2. Sample Preparation

In this study, fluidity test and pore structure determination were applied with cement paste, while compressive strength and water absorption rate were measured with mortar specimens. To prepare cement paste, 300 g of cement was used, and a low w/c of 0.25 was adopted. To prepare mortar, 450 g of cement and 1350 g of standard sand were used, and the w/c was 0.5. The dosages (mass percentage of PCE and OTS to cement) of PCE were 0.15%, 0.20%, and 0.25%, while the dosage of OTS was 1%. Throughout this paper, the following nomenclature was used to identify specimens: PCE_X-OTS_Y; X = 0.15, 0.20, and 0.25; Y = 0 and 1, representing the dosage of PCE and OTS, respectively. The mixing of cement paste and mortar was performed in accordance with Chinese standards GB/T1346-2001 [41] and GB/T 17671-2021 [42]. After the mixing process, the mortar specimens were cast into (40 × 40 × 40) mm³ cubes and covered with a polytetrafluoroethylene film on the surface. For the first 24 h, the specimens were cured under the temperature of (20 ± 3) °C and the relative humidity of 90%. After that, the specimens were demolded and then cured continuously in the standard curing room until a certain age.

2.3. Fluidity Measurement

The fluidity of cement paste was determined using a mini-slump cone, with a height of 60 mm, an upper diameter of 36 mm, and a bottom diameter of 60 mm. For each sample, the fluidity result was taken from the average of three specimens, and the standard deviation was limited to 2%.

2.4. Pore Structure Determination

In this study, the pore structure of cement paste was determined using ¹H LF NMR apparatus (PQ-001, Niumag, Shanghai, China), with a 0.5 T magnetic field and a 25 mm radiofrequency coil. The T₂ transverse relaxation time was measured by the Carr–Purcell–Meiboom–Gill (CPMG) sequence. The number of scans and the echo time were kept constant at 4 and 0.302 ms, respectively. After mixing, fresh cement paste was cast into a small glass bottle and then placed in the apparatus; T₂ and relative amplitude were monitored at the age of 0, 1, 2, 3, 4, 5, 6, and 12 h and 1, 2, 3, 7, and 28 d. T₂ data were fitted to a multi-exponential curve using the Multi ExpInv Analysis software 4.0 (Niumag Electric Corporation, Shanghai, China), which was performed via the inverse Laplace transform algorithm [43]. The intensity-weighted T₂ values of the main peak ($\overline{T_2}$) calculated according to Equation (1) are used to characterize the hydration state, where $I_i$ is the intensity at each T₂; t is the start point at T₂ distinguishing surface water peak and other peaks in T₂ distribution curves. For each sample, the result was taken from one specimen and the accuracy was examined by an extra specimen.

$$\overline{T_2}={\displaystyle \frac{{\sum }_{i=1}^t{T}_{2i}\ast I_i}{{\sum }_{i=1}^t{I}_i}}$$

(1)

2.5. Compressive Strength Test

The compressive strength of mortar specimens at the ages of 3, 7, and 28 d were measured. The compression test machine with a capacity of 2000 kN was used at a loading rate of (2400 ± 200) N/s. For each sample, 3 specimens were measured, and their average value was calculated as the final result. The standard deviation is limited to 15%.

2.6. Water Absorption Rate Measurement

The water absorption rate was measured using the immersion method at the age of 7 d, considering that cement hydration was almost stopped [14,16,44]. The mortar specimen was dried at 50 °C for approximately three days to the constant weight $M_a$. After drying, the whole specimen was immersed in water, and the bottom of the specimen was separated from the container with a 3 mm thick plastic mesh to allow for water absorption. The water level was kept at 2–5 mm above the top of the specimen throughout the immersion time. The mass of the specimen $M_C$ was recorded after immersed for 30 min and 1, 2, 3, 4, 5, 6, 12, 24, 48, and 72 h. Thus, the water absorption rate $W_A$ was calculated according to Equation (2) [14,16]. Each result was the average value calculated with 3 specimens, and the standard deviation was limited to 2%.

$$W_A={\displaystyle \frac{M_C-M_a}{M_a}}\times 100\%$$

(2)

3. Results

3.1. Fluidity

First, PCE dosages of 0.15%, 0.20%, and 0.25% were used to prepare cement paste, respectively. It can be seen from Figure 2 that the fluidity of cement paste increased significantly with increasing PCE. With the addition of 1% OTS, the fluidity of the cement paste decreased depending on PCE dosage. The reduction amount was about 10 mm when PCE dosages were 0.15% and 0.25%, and 5 mm while PCE dosage was 0.20%. The result revealed that introducing OTS could impair the dispersion ability of PCE, thus reducing the fluidity of cement paste. This result is in agreement with the previous study [45]. The possible reason can be attributed to the fact that OTS hinders the adsorption of PCE on the surface of cement particles and hydration products.

3.2. T₂ Relaxation Time Distributions

The transverse relaxation time (T₂) correlates linearly with the volume-to-surface ratio of pores in terms of pore diameter. In addition, the cumulative area under the T₂ curve can quantitatively represent the relative pore volume in the cement paste. Hence, the T₂ distribution reflects the pore structure of cement paste [43,46]. Previous research has revealed that short T₂ values in the range of 0.01–1 ms represent the water in the gel pores of about 3 nm, conceptualized as existing between the gel globules that comprise C-S-H. Longer T₂ values in the range of 1–100 ms correspond to the water in the capillary pores, which initially is interstitial water between the clinker grains but then evolves to become inter-hydrate water. Further, T₂ values in the range of 100–10,000 ms can be attributed to water in the meso and macro pores, which can be regarded as bleeding water [43,47,48].

Figure 3 exhibits the T₂ distributions of cement paste mixed with 0.15%, 0.20%, and 0.25% PCE without and with 1% OTS. It can be seen that the relative amplitude of the T₂ main peaks decreased and shifted to the left overall as a function of time, indicating that the pore volume of cement paste reduced and the pore diameter became smaller. Special attention must be paid to the weak peak appearing around 1000 ms. By comparing Figure 3a,c,e, it can be found that the weak peak around 1000 ms arose earlier, and the intensity increased with increasing PCE dosage, revealing that the bleeding water came out earlier and its amount increased with more PCE added. These weak peaks almost faded before 12 h due to the reabsorption of water by cement paste and its involvement in the hydration reaction. From Figure 3b,d,f, it can be observed that with the addition of 1% OTS, the weak peak around 1000 ms appeared immediately after the mixing procedure and could be captured even after 1 d. Furthermore, the intensity of the weak peak around 1000 ms increased significantly compared to that without the addition of 1% OTS. The result disclosed that with the addition of 1% OTS, a strong water-repelling property was created in cement paste due to the hydrophobic effect of OTS. Therefore, the bleeding water appeared from the very beginning, and its migration from the surface to the interior part was hindered as well. As more bleeding water was presented, the water within the cement paste for dispersion was reduced; thus, the fluidity was decreased with the addition of 1% OTS. Notably, the water-to-cement ratio would be increased on the surface of cement paste, and the large-scale pores would be generated because of the increased bleeding water [43,49], which would further affect hydration product distribution, microstructure development, and mechanical properties [50,51,52].

H. Liu et al. [53] have studied the pore structure development and water migration process of cement paste using ¹H LF NMR by analyzing intensity-weighted T₂ values ($\overline{T_2}$), which were calculated according to Equation (1). Based on their research, the decrease in $\overline{T_2}$ indicates that the average pore diameter of cement paste reduced, which can be attributed to that the hydration product has filled spaces between particles. Further, as more water participates in hydration, free water migrates from macropores to micropores, causing $\overline{T_2}$ decrease as well. To better illustrate the pore structure development and hydration state of cement paste, the $\overline{T_2}$ was calculated further in the same manner as described by H. Liu et al. [53], and the results are shown in Figure 4. As the T₂ distribution curves do not change further after 1 d in Figure 3, the $\overline{T_2}$ were collected in the time range of 0–24 h, correspondingly. Figure 4 shows the $\overline{T_2}$ of cement paste mixed with 0.15%, 0.20%, and 0.25% PCE, respectively, without and with 1% OTS as a function of hydration time. It can be observed that the $\overline{T_2}$ decreased at different rates as hydration time prolonged. Specifically, the $\overline{T_2}$ increased with increasing dosage of PCE at 0–3 h, indicating that the pore diameter was increased. As pores were filled with free water, the result implies that more free water was released by adding more PCE, and less water participated in cement hydration, indicating that the cement hydration was delayed with the addition of more PCE. During (3–12) h, the $\overline{T_2}$ reduced sharply; the more the PCE was added, the higher the reduction rate was, indicating the faster hydration rate. Furthermore, with the addition of 1% OTS, the $\overline{T_2}$ increased obviously, signifying that the pore diameter increased. Therefore, it can be inferred that OTS increased the amount of free water and may cause a delay in cement hydration. After 12 h, the $\overline{T_2}$ almost reached a plateau, indicating that the variation in pore diameter and the migration in free water slowed down.

In general, the cumulative peak area of the T₂ spectrum can relatively represent the amount of free water existing in the pores of cement paste, which further reflects the total pore volume. Figure 5 exhibits the cumulative peak area of the T₂ spectrum of cement paste mixed with 0.15%, 0.20%, and 0.25% PCE, respectively, without and with 1% OTS. It can be seen that the cumulative peak area decreased continuously within (0–24) h, indicating that the amount of free water reduced gradually due to the hydration reaction, in which free water was converted into chemically bonded water, and the total porosity reduced as well. Furthermore, the cumulative peak area increased slightly with increasing PCE dosage as well as the addition of 1% OTS. The possible reason behind this result can be attributed to two aspects. The first reason was that more free water was released due to the enhanced dispersing ability of increased PCE and hydrophobicity of OTS. Another reason was that both PCE and OTS provided a retarding effect to cement hydration; thus, the water consumption was reduced.

Overall, the following ideas can be drawn from the ¹H LF NMR study:

• The bleeding water can be generated due to the addition of PCE and OTS. The higher the dosages of PCE, the more the bleeding water;
• The calculated $\overline{T_2}$ revealed that the addition of PCE and OTS provided a retarding effect upon cement hydration. And this retarding effect weakened gradually with prolonged hydration time;
• The porosity of cement paste would be increased with the addition of PCE and OTS.

3.3. Compressive Strength

The compressive strength of mortar specimens containing 0.15%, 0.20%, and 0.25% PCE were measured without and with the presence of 1% OTS. The result is summarized in Figure 6. By incorporating 1% OTS, the compressive strength decreased at PCE dosages of 0.15% and 0.25%. At a PCE dosage of 0.15%, the compressive strength decreased by 6.2%, 5.5%, and 2.2% at the age of 3 d, 7 d, and 28 d with the addition of 1% OTS, respectively. At low dosages, the dispersion ability of PCE was easily weakened by introducing OTS, thus reducing the workability of mortar and causing defections during the casting procedure, which were harmful to compressive strength. Higher reduction rates in compressive strength by adding 1% OTS were found at the early ages of 3 d and 7 d, while PCE dosage increased up to 0.25%, which were 15.0% and 9.7%, respectively. Inferred from the ¹H LF NMR result, this phenomenon could be attributed to bleeding, which happened in the first 6 h. The large-scale pore size on the surface and the movement of cement particles during the bleeding process led to uneven distribution and more defects, which resulted in a significant decrease in compressive strength. This result is in agreement with previous studies [50,51]. In contrast, an enhancing effect of 1% OTS was observed while PCE dosage was 0.20%, indicating that 0.20% PCE and 1% OTS were an adequate combination, which made mortar obtain a better workability and compact structure, as is revealed in Section 3.1 and Section 3.2. Finally, it can be concluded that the synergistic of PCEs and OTS at adequate contents can remarkably increase the compressive strength of mortar, which allows for more applications of water-repellent-based saline in hydrophobic concrete in different areas exposed to water actions.

3.4. Water Absorption Rate

Figure 7 presents the water absorption rate of mortar containing different dosages of PCE without and with the presence of 1% OTS. It can be observed that the water absorption rate increased continuously with the prolonged immersion duration. When the PCE dosage was increased from 0.15% to 0.20%, the water absorption rate was obviously reduced. However, the water absorption rate increased drastically when PCE dosage increased to 0.25%. The high content of PCEs in mortar leads to a bleeding effect, as is shown in the ¹H LF NMR result. During the bleeding process, the top layer of hydrated mortar is more porous than the lower layer due to bleeding, which then increases the water absorption rate of hardened cement mortar [50,51].

Cheerfully, the addition of 1% OTS provided a significant effect in reducing the water absorption rate. It can be observed in Figure 8 that after being immersed for 72 h, with the addition of 1% OTS, the water absorption rate decreased by 40.3%, 46.6%, and 28.1% in the mortar containing 0.15%, 0.20%, and 0.25% PCE, respectively. As is shown in the ¹H LF NMR result, the addition of OTS increased the pore size and total pore volume of cement paste; it seems that an increased water absorption rate would be produced under these effects. However, the reverse result showed that the water absorption rate reduced significantly with the addition of OTS. The conflict phenomenon powerfully revealed that the internal hydrophobic treatment by OTS has been successfully achieved. When OTS was incorporated into mortar during the mixing procedure, the interaction between OTS molecule and cement, as well as hydration products, led to the formation of a stable self-assembled hydrophobic film on the surface of the pores and on the external surface of the mortar, which provided a high water-repellent ability [7,10]. The lowest water absorption rate was obtained when 0.20% PCE and 1% OTS were used, indicating the most excellent synergistic effect in water repelling. It can be predicted that the synergistic effect of PCE and OTS can remarkably improve the durability of cement-based materials and allow for more applications of water-repellent-based saline in cement-based materials.

4. Conclusions

Water reducer and water-repellent admixture are two important components for improving the workability and durability of CBM, respectively. However, when the two species of admixtures are incorporated at the same time, their synergistic effect may generate a profound influence on the basic properties of CBM, which have not been studied extensively. In this study, by incorporating PCE and OTS in the mixing procedure to prepare CBM, the fluidity, pore structure, compressive strength, and water absorption rate of CBM were investigated. The following main conclusions can be drawn:

• The fluidity of cement paste was reduced with the addition of OTS for about 5–10 mm, especially at a low PCE dosage. On the one hand, OTS impairs the dispersion ability of PCE as OTS may hinder the adsorption of PCE on the surface of cement particles and a hydration product. On the other hand, OTS caused more bleeding water, thus reducing the free-water amount within the cement paste for dispersion;
• The ¹H LF NMR results showed that incorporating PCE and OTS resulted in an increased pore size as well as total pore volume. In particular, more bleeding water was presented by using OTS. The intensity-weighted T₂ values ($\overline{T_2}$) revealed that the addition of PCE and OTS provided a retardation effect on cement hydration;
• Mortar compressive strength decreased by 15.0% at the age of 3 d with the addition of OTS at a high PCE dosage of 0.25%, which could be attributed to a more porous top layer resulting from increased bleeding water amount. Notably, a positive synergistic effect of PCE and OTS was endowed on the compressive strength when PCE dosage was 0.20%;
• Despite the increased pore size and pore volume, as well as the retarded cement hydration, the water absorption rate was still reduced up to 46.6% with the addition of OTS. The conflict phenomenon powerfully revealed that the internal hydrophobic treatment by OTS has been successfully achieved.

The results obtained from this research clearly reveal the synergistic effect of water reducer and water-repellent admixture on the basic properties of CBM and enrich our knowledge in using water-repellent admixture through internal treatment, which is meaningful for improving the durability of cement-based materials.