Integrated design of a robust superhydrophobic cement mortar layer via sizing sand grains

This work reported a facile route for fabricating super-hydrophobic concrete via sizing sand grains. It was found that mixing the sands with a size ranging from 150–180 μm into cement enabled the formation of a lotus-like surface with a papillary structure at micro-scale. SEM showed that the size of bumper was about 3 μm. When spraying a fluorocarbon solution onto this surface, the porous nature of the cement matrix showed the advantage of taking the fluorocarbon into the internal structure of the concrete via capillary force. As a result, the sub surface up to a depth of ∼1.5 mm were transformed into a thick superhydrophobic layer directly. The contact angle (CA) of water droplets could reach 157° on this surface, and which could remain more than 150° after abrasion 100 cycles under a weight of 300 g at 360 grit sandpaper. This thick hydrophobic layer significantly reduced the corrosion rate of the steel the concrete at the Cl- environment by 620 times. The measurement of British pendulum number and compression strength revealed that this superhydrophobic layer was beneficial for maintaining the friction coefficient of the concrete surface in wet condition without altering the mechanical integrity of the concrete.


Introduction
As a hydrophilic porous material, the concrete suffers from the formation of ice on the surface and the steel bar inside tends to be corroded by the water infiltrating into the bulky phase [1][2][3][4].Construction of a superhydrophobic surface with a contact angle (CA) over 150°like a lotus leaf has been proposed to tackling these challenges [5][6][7][8][9].It is known that the 'lotus effect' is associated with a rough surface with well-designed bumpers at nano/micro-scale and a low surface energy waxy coating [10][11][12][13][14].To making a bumpy surface, the hard particles such as metakaolin [15], polyether ether ketone [16] silica [17], TiO 2 [18] and silver [19] were used to introduce the roughness on the surface of the concrete.However, maintaining the superhydrophobicity remain as the challenge since the concrete always needs to live with harsh and dynamic abrasion [20,21].This triggers the effort on the incorporation of hydrophobic additives into bulk concrete instead of onto the surface only [22,23].However, the additives mixed with concrete precursor affected the micro-structure of concrete, and this made it hard for controlling the mechanical properties of the concrete itself.In this work, we reported a facile route for making a robust hydrophobic surface without altering the mechanical integrity of the concrete.A series of experiments led to a key observation that mixing a sand with a size ranging from 150-180 μm with cement could result in the formation of a rough surface required by 'lotus effect' spontaneously.The contact angle of water droplet on this surface could reach 157°.In addition, the interconnected pores of the concrete offered a pathway for the diffusion of a liquid solution loaded with a hydrophobic additive into the sub surface of the concrete.The concrete showed superhydrophobic nature even when the top layer was scrapped away.The superhydrophobicity could be restored through re-filling the hydrophobic additive into the porous space of the concrete.This strategy was not involved with a bespoke process for achieving superhydrophobic concrete, and therefore demonstrated a strong potential to be scaled up for the civil engineering applications.

Experiments Materials
Portland cement (P.O 32.5) was provided by Anhui Conch Cement Co., Ltd Perfluorooctyltriethoxysilane (PFOS) was purchased from Aladdin Industrial Corporation.Other chemicals including ethanol and hydrochloric acid were provided by Shanghai Lingfeng Chemical Reagent Co., Ltd All materials were used as received without further purification.
Preparation of fluorocarbon solution 0.7 g HCl solution (PH ≈ 3) was slowly added to a flask charged with 2.04 g PFOS, and the mixture was stirred at 70°C for 5 h.After removing residual HCl, the resulting H-PFOS was diluted with ethanol to a solids content of 0.25 wt%.

Preparation of super-hydrophobic concrete
Firstly, 20 g cement, 10 g water, and 60 g quartz sand were uniformly mixed and poured into a 4 × 4 × 2 (cm) PTFE plastic mold.Afterward, the fresh concrete was placed in a conservation box at 30°C, 90 RH% for 28 days.Then, the solution (0.25 wt% H-PFOS/ethanol) was sprayed onto the surface of the fresh concrete, and which could slowly diffuse into the interconnected pore network of the concrete.Finally, the concrete was left at 30 °C for 2 h in order to evaporate ethanol.For comparison, the sand with three different size range (0.425-0.250 mm, 0.250-0.180mm, 0.180-0.150mm) were used in this fabrication process.It was noted that the size of sand grains was determined through the method indicated in ASTM C33/C33M-18 standard [24].The C-1, C-2 and C-3 were used to label the concrete samples associated with different sized sands.

Characterization
The surface of the concrete was imaged by a scanning electron microscopy (SEM, JEM-6510, Tokyo, Japan).Element analysis of the cross-sectional surface was measured by energy dispersive x-ray spectroscopy (EDS).The crystalline structure of the concrete was analyzed by an x-ray diffraction (XRD, Rigaku, Tokyo, Japan) at a scanning rate of 10 °/min and using Cu Kα radiation.The drops water contact angle was measured by contact angle tester (DSA-100, Kruess, Germany) with the use of 2 μl droplets.White light interferometer (WLl, BMT, Germany) was used to measure 3D topography of the concrete surface.Abrasion resistance test was performed on a sample with the area of 4 × 4 cm 2 by using a 360 grit sandpaper under a certain load (300 g and 1000 g).In anti-icing test, an ice cube was formed on the top surface of concrete in the refrigerator at −15°C.A universal testing machine (CMT-525, Sans, Shenzhen, China) was used to test the longitudinal shear force between the ice and concrete.For measuring the corrosion of steel bar, self-corrosion potential (E corr ), self-corrosion current density (i corr ) and electrochemical impedance spectroscopy (EIS) were tested by using an electrochemical workstation (EW, Solarton-1200, Hampshire, UK).The British pendulum number (BPN, Hebei, China) was tested for evaluating the surface friction coefficient of concrete according to ASTM E303-93 standard [25].The pressure tester (AEC-201, Shanghai, China) was used to measure the compressive strength according to the ASTM C109/C109M-16a standard [26].

Result and discussion
Figure 1 illustrates the steps of fabricating the superhydrophobic concrete.This process was facile since the creation of lotus-like surface was free of using any other particles.Through a series of experiments, it was found that the surface contact angle was highly associated with the size of sand mixed into the cement.With using properly sized sands, the lotus-like surface could be directly formed when mixing sand and cement together.This is well-known method for making cement mortar in construction industry.After hardening, resulting porous network of the cement mortar was open and which could take H-PFOS/ethanol solution into the subsurface of the concrete.This avoided the mix of the low surface energy additive prior to hardening and hence minimized its impact on the mechanical strength of the concrete.The SEM images in figures 2(a)-(c) show that all three samples have rough surfaces with micro-scale sized bumpers.However, the texture of the surface is highly associated with the size of sands mixed with the cement.XRD pattern in figure 2(d) shows that the concrete is composed of quartz (SiO 2 ), calcium silicate, calcium silicate hydrate, calcium hydroxide, ettringite and calcium carbonate.This was in the agreement with previous publications [27,28].In the SEM images, it could be seen that the size of the bumpers on the surface of C-3 sample is less than 3 μm, which is smaller than of that of C-1 and C-2 samples.Figure 2(e) shows that the contact angles of C-1, C-2 and C-3 samples are 133 ± 2.5°, 141 ± 3.4°and 157 ± 3.6°, respectively.All these pointed out that the use of a quartz sand with a size of 150 μm could directly yield a louts-like bumpy surface.
For measuring the thickness of the superhydrophobic layer, the sample was sliced off by immersion wire cutter for taking the image of the cross-sectional area as shown in figure 3(a).A clear boundary could be observed between the off-white (top) and gray-black (down).The thickness of the off-white area on the vertical direction is about ∼1.5±0.2 mm owing to the diffusion of the fluorocarbon additive into the internal structure of concrete.Figure 3(b) presents the SEM image of this cross-sectional area.Figures 3(d), (e), (f) shows the EDS spectra of Si, Ca, and F elements of the area highlighted in the image (c).The phase with Si elements represents the sand  dispersed in the phase of the cement rich of Ca elements.The F element of the H-PFOS was evenly distributed in the cross-sectional area, and this indicates the penetration of hydrophobic additive into the internal structure of the concrete.
The robustness of the superhydrophobic layer was measured under a designed abrasion test illustrated in figure 4(a).The C-3 sample demonstrates a slight decrease on the contract angle from 157°to 152°by 3.28% after 100 cycles of forth and back abrasion under a weight of 300 g.Even under a much higher weight of 1000 g, figure 4(d) shows that the contact angle of C-3 sample decreases from 157°to 151°with a 4% decrease after 10  cycles of abrasion, then further drops to 141°after 50 cycles of abrasion.It was noted that de-icing force for hydrophilic concrete is more than 815 KPa. Figure 4(d) shows that the initial de-icing force is as low as 25 KPa for the C-3 sample, which increases to 221 KPa after 50 cycles of abrasion.However, this de-icing force remains much smaller than that between the ice and hydrophilic concrete.
The penetration of water into the pore network of concrete as illustrated in figure 5(c) would cause the internal steel bars to be corroded by chlorides, sulfates and other ions from natural environment [29,30].As shown in figure 5(a), a carbon steel bar (Φ: 2.75 mm) is inserted into C-3 sample (2 × 2 × 2 cm) for the simulation of reinforced concrete.Figure 5(b) shows a diagram of three electrode in electrochemical working station including the steel rod in the concrete as a working electrode (WE, area: 0.19 cm 2 ), graphite as a counter electrode (CE) and calomel as a reference electrode (RE).The corrosion rate of steel is evaluated following the formula of ν = (M/nF) × i corr [31], where ν is corrosion rate (g/m 2* h), i corr is self-corrosion current density (μA/cm 2 ), M is the molar mass of Fe, n is metal valence and F is faraday constant (C/mol).All the samples were immersed in a 0.5 mol/l NaCl solution for 24 h before testing.Figure 5(d) shows the polarization curves of the treated and pristine samples in a 0.5 mol/l NaCl solution (25°C) at scanning voltage of 0.5 mv/s for 60 s.Table 1 shows that E corr and ν of the untreated sample and treated sample were −0.692 v, 7.418 × 10 -2 g * m −2* h −1 , −0.03 v, 1.196 × 10 -4 g * m −2* h −1 , respectively which indicated the treated sample is significantly lower than that of pristine sample.The corrosion rate is reduced by 620 times, which implies the substantial improvement of corrosion resistance.The ohmic resistance is another key parameter for evaluating the corrosion resistance since the hydrophobic layer is able to slow the transfer of ions in the concrete [32][33][34].Figures 5(e)-(f) shows the EIS Nyquist plot of the samples at a rate of 0.1 Hz to 1 MHz frequency 20 mv amplitude perturbation condition, in which the high-frequency region of the x-axis is defined as ohmic resistance.It is observed that ohmic resistance of the treated sample is extremely lager than pristine sample.Figure 5(c) shows the schematic diagram of anticorrosion mechanism of super-hydrophobic concrete, the air cushion formed between the concrete and water, and this indicates that the super-hydrophobic layer can effectively resist the intrusion of Cl − and ensure a prolonged service life of steel bars.The coefficient of friction of the concrete surface is evaluated by BPN testing as illustrated in figure 6(a).Under dry condition, figure 6(b) shows the BPN value of the treated samples is almost equal to that of pristine concrete.The situation is not the same in the wet condition.The BPN value of the treated sample is unchanged in compared with that under dry condition.But, the wet condition leads to a decrease in the BPN value of the pristine sample by 13%.This is associated with the formation of a layer of water membrane on the surface of the hydrophilic concrete [15].The compressive strength of concrete before and after treatment was measured.Figure 6(c) shows that the treatment has little affect on the compressive strength, which means this treatment has an advantage in terms of reserving the integrity of the concrete.

Conclusion
This work presented a facile route for making a concrete with robust and repairable super-hydrophobic layer.The lotus-like surface was directly realized via mixing the sand with a size range of 0.18-0.15mm into the cement.Taking the advantage of the pore network in the cement, the fluorocarbon solution (hydrophobic additive) could diffuse into the sub-surface with a thickness of by ~1.5 ± 0.2 mm.This led to the formation of a thick superhydrophobic layer with a contact angle of 157 o .This thick layer significantly reduced the corrosion rate of the steel the concrete at the Cl -environment by 620 times.More interestingly, this thick layer demonstrated strong durability when exposing to external abrasion.The contact angle still exceeded 150°after abrasion 100 cycles under a weight of 300 g at 360 grit sandpaper.The BPN and compressive strength revealed that this thick layer would not decrease the friction coefficient of the concrete surface in wet condition and was unlikely to reduce the mechanical strength of concrete.

Figure 3 .
Figure 3.The section morphology of super-hydrophobic concrete.(a) section of concrete after immersion wire cutting.(b)-(c) SEM image of section concrete; (d)-(f) The F, Si, Ca element spectrum of area image (c).

Figure 4 .
Figure 4.The super-hydrophobic concrete surface of the mechanical strength, anti-icing, and 3D topographic profiles.(a) The schematic diagram of abrasion test; (b) The CA result of abrasion distance from 0 to 100 abrasion cycles at the weight of 300 g; (c) The schematic diagram of deicing ice test; (d) The drops water CA and deicing ice force result of abrasion cycles from 0 to 50 at the weight of 1000 g.

Figure 5 .
Figure 5. Anti-corrosion of super-hydrophobic concrete.(a) The super-hydrophobic (left) and hydrophilic concrete (right) sample with steel bar; (b) Schematic of electrochemical corrosion test; (c) The schematic diagram of anti-corrosion of super-hydrophobic concrete; (d) The polarization curves of the super-hydrophobic and hydrophilic concrete in a 0.5 mol/l NaCl solution, 25 °C; (e)-(f) The EIS spectra of the super-hydrophobic and hydrophilic samples in a 0.5 mol/l NaCl solution, 25 °C.

Table 1 .
The E corr , J corr , ν of function coating treated and untreated of concrete.