Self-Cleaning Cement-Based Building Materials

Abstract

The modern rhythm of human life leads to well-known problems, which are air, water and soil pollution and climate warming. An increase in the power of industries and vehicles leads not only to atmospheric pollution by-products of incomplete fuel combustion but also to various microscopic particles that form aerosols, which carry an obvious danger to human health and also pollute the buildings’ facades. An environmentally friendly building material with a hybrid method “Nano-titania gradient” was developed. This method consists of forming a gradient of n-TiO₂ particles concentration in the composite since the physical properties of the composite are always inextricably linked to the geometry. To increase the efficiency of the photocatalytic process, a method of surface sensitization of titanium dioxide with the use of graphene oxide was proposed, which contributed to an increase in the overall photosensitivity. Thus, the decomposition of nitrogen oxide by volume with the modified surface increased by 27% in comparison with the classic titanium dioxide, and the decomposition of volatile organic substances increased by 32%. It was found that for the facade plate made with surface-sensitized TiO₂, the process of self-cleaning is completed after 3 h after the irradiation start. The modern rhythm of human life leads to well-known problems, which are air, water and soil pollution and climate warming. Using the theory of percolation, the concentration range of the photocatalyst content was calculated. To facilitate the material, waste cellulose was introduced. To increase the efficiency of the photocatalytic process, a method of surface sensitization of titanium dioxide (SS TiO₂) with the use of graphene oxide was proposed. The analysis of the experimental-statistical models of the compressive strength shows that the optimum content of TiO₂ was in the range from 0.8 to 1.1%, and cellulose from 0.4 to 0.8%, the optimum content of SS TiO₂ was in the range from 0.7 to 1.1%, and cellulose from 0.4 to 0.8%. Analysis of the experimental and statistical model of the bending strength shows that the optimal content of TiO₂ and SS TiO₂ was in the range of 0.6 to 1.0%, and cellulose from 0.4 to 0.8%. When studying the structure of composites, it was found that titanium dioxide was sorbed on the surface of swollen cellulose fibers and remained there after the process of cement hydration. The effectiveness of the method of surface sensitization of titanium dioxide by combining it with graphene oxide was shown. Thus, the decomposition of nitrogen oxide by volume with the modified surface increased by 27% in comparison with the classic titanium dioxide, and the decomposition of volatile organic substances increased by 32%. It was found that for the facade plate made with surface-sensitized TiO₂, the process of self-cleaning was completed after 3 h after the irradiation start.

1. Introduction

The modern global economy is focused mainly on obtaining energy by burning organic fossil fuels, which leads to well-known problems such as atmospheric pollution and climate warming [1]. At the same time, adding the vehicles’ power and increasing the intensity of their operation lead to the pollution of the atmosphere by the products of incomplete fuel combustion and various microscopic particles. These particles might form aerosols, which carry an obvious danger to human health and also pollute buildings’ facades, small architectural objects, vegetation, etc. Regardless, the tightening of environmental requirements for the industrial sector and the emergence of internal combustion engines with a higher environmental class, air pollution by NOx, SOx, CO oxides remains an acute environmental problem. These pollutants inhibit the growth of certain plants and at the same time, cause the formation of photochemical smog in the urban environment, which contributes to significant pollution of the buildings’ facades. In addition, the presence of these pollutants in the air has a detrimental effect on human respiratory function, causing respiratory diseases. Trace volatile compounds emitted by furniture, equipment and inhabitants themselves, along with microbiological contamination, lead to mucous membrane irritation, headaches, nausea and allergic reactions. Moreover, along with mold fungi, these compounds can cause chronic diseases and “sick building” syndrome [1]. It is estimated that metropolitan areas such as New Delhi, Shanghai face serious environmental problems caused by smog [2], which sometimes interferes with normal human activities, the functioning of schools, businesses and transport [3].

All this leads to increased scientific interest in recent years in the issues of environmental safety and sustainable development of urbanized areas. Therefore, the actual direction is to develop, not only environmentally friendly materials and technologies for use in the construction of civil and industrial facilities, but also building materials that reduce the environmental load on urbanized areas [4,5,6]. To solve the problem of atmospheric air pollution, it is possible to use photocatalytic self-cleaning smart building materials that enable reducing the concentration of pollutants in the atmosphere with the simultaneous function of self-cleaning.

The potential photocatalytic degradation scheme of the pollutants during photocatalysis can be reported in the following form, Figure 1. Titanium dioxide in anatase allotropy, being a semiconductor in the nano-dispersed state, is now often used as such a catalyst. During the photoexcitation process, n-TiO₂ absorbs a quantum of ultraviolet radiation (sunlight or fluorescent lamps), which leads to excitation of the valence zone electrons (e⁻) and their “jumping” into the conduction zone, forming a hole (h+) [7]. The phenomenon of molecular adsorption and subsequent charge transfer has a strong dependence on the surface properties of the photocatalytic material and can vary for polymorphs and even for the same material due to different orientations [8]. During redox reactions, highly reactive superoxide anion radicals (O₂⁻) hydroxyl radicals (OH⁻) are formed in the valence and conduction zones on the n-TiO₂ surface that provides the breakdown of organic and inorganic pollutants and microorganisms [9], Figure 1.

The self-cleaning function is generally thought to be due to the fact that n-TiO₂ has high amphiphilic (hydrophilic or hydrophobic) characteristics [10], which change the wetting contact angle on the surface of photocatalytic concrete under the action of sunlight ranging from about 80 to 0° and back with a duration of a few days. This ensures first complete coverage of the surface of the smart concrete with water and then the collection of water into droplets, followed by their detachment from the surface. In this way, self-cleaning of the photocatalytic concrete is carried out (Figure 2).

The use of nanoscale additives in cement-based composite materials has the potential to solve other problems, such as low crack resistance and tensile strength, long curing time, high water absorption and low plasticity [11,12]. Nano-dispersed titanium dioxide (n-TiO₂) was widely used as a photocatalyst thanks to its availability and the ability to be a photocatalyst not only under the action of solar radiation but also under fluorescent lamps [13]. It is preferable to use it in a polyform of anatase since this form has the greatest photocatalytic activity. However, at the same time, the main challenge of its production is that it is a temperature-unstable form [14].

The effectiveness of the photocatalytic process depends on many factors, which can be divided into external and internal ones [15,16]. Internal factors include the combination of catalyst and cement hydrates, photocatalyst morphology [17], the ratio of cement concrete components (water-binder ratio, coarse aggregate to binder ratio) and curing time [18], phase composition, electronic structure, particle size and degree of n-TiO₂ aggregation [19], etc. External factors can be divided into two subgroups: environmental factors such as humidity, temperature, illumination, radiation spectrum, pollutant level, wear level, etc. [20,21,22,23,24], and the contact characteristics of the composite product, such as surface roughness, color, porosity, and microstructure of the substrate, as well as its hydrogen exponent pH [24,25,26,27,28,29].

To increase the efficiency of the photocatalytic process, many researchers [30,31] recommend using the technology of surface application of n-TiO₂ particles. However, at the same time, it was found that when applying this placement technology, a photocatalyst has a short period of activity [32], especially if the composite is exposed to abrasion [33]. In another study, when simulating operational exposures through abrasion, it was found that a large value of residual photocatalytic activity was characteristic of composites prepared using a technology where n-TiO₂ particle powder was introduced and dispersed in the volume of the cement material [15]. This makes this technology of n-TiO₂ insertion more promising for the long-term operation of building products made of such composite materials. However, it is worth noting that higher absolute values of photocatalytic activity were recorded for composites with the surface way of introduction and distribution of n-TiO₂ particles.

An important factor is also the carrier (Support Material) of the n-TiO₂ particles. It was found that n-TiO₂ particles deposited on a concrete carrier exhibit a greater photocatalytic effect over a longer period of time compared to glass and asphalt concrete support materials [34,35]. This indicates a strong influence of the substrate material on the photocatalytic performance and longevity of the composite. There is a study that demonstrates that using a substrate with a harder surface will improve the fixation of n-TiO₂ particles [33]. In addition, enhancement of photocatalytic performance can be achieved by increasing the surface area of the support material and catalyst particles [25], as well as by the presence of chemical interaction between them [36,37]. The use of materials providing high porosity of concrete will contribute to more intensive adsorption of NO₂ [38,39]. The formation of a basic medium type (pH) on the concrete surface facilitates the oxidation of NO₂ to nitrates [40]. This process can be regulated by introducing various additives such as lime, alumina and CaCO₃ [28,38].

Summarizing the studied information [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40], we can conclude that several methods of introducing titanium dioxide (n-TiO₂) into a building material are currently known. The authors of the study propose the following explanation and classification: surface “Nano-titania outside,” internal “Nano-titania inside.” The former is based on the application and distribution of a prepared suspension containing n-TiO₂ particles on the surface of the cured composite material. The latter is based on the introduction and dispersion of n-TiO₂ powder in the matrix material in the process of composite preparation. It can be carried out directly in a cement batter or at a stage of introduction of aggregates, whereas the spraying of particles n-TiO₂ was preliminarily carried out on the surface of aggregates. The advantages of the surface method “Nano-titania inside” can include a higher value of photocatalytic performance, but at the same time, a short duration of operation due to loss of photocatalyst particles. The advantages of the second method can include a longer period, but a lower photocatalytic performance of the building products. Thus, a relevant scientific direction is considered the development of a new way of introducing n-TiO₂ particles, devoid of the shortcomings of existing methods and simultaneously retaining their advantages. To solve the problem, the authors propose to develop a hybrid method, “Nano-titania gradient,” which consists of forming a gradient of n-TiO₂ particles concentration in the composite since the physical properties of the composite are always inextricably linked to the geometry. A schematic representation of the formation of the concentration gradient of n-TiO₂ particles “Nano-titania gradient” in the facade concrete slab is shown in Figure 3. The proposed method will be thoroughly explained in Section 2.

It is also worth mentioning the existing shortcomings in the application of titanium dioxide. The main one was that only the UV part of the spectrum could be used in catalysis, i.e., only 3–4% of the radiation, whereas scattered daylight does not cause a catalytic effect, or the effect is insignificant. In this regard, there was a need to modify titanium dioxide in such a way that the main part of the sunlight spectrum can be used to induce this effect, namely, the visible part of the spectrum from approximately 400 nm to 700 nm. For this purpose, the most promising method, from the point of industrial implementation, was surface sensitization, which would increase the overall photosensitivity and shift the spectral sensitivity zone to the region of visible light [41]. Recently, carbon materials, including fullerene C₆₀ [42], carbon nanotubes [43], and activated carbon [44], were actively studied as titanium dioxide sensitizers. The authors of the study propose to consider graphene oxide as a potentially effective material for surface sensitization. In view of the fact that initially graphene was characterized by a zero forbidden zone and high electron mobility, being the result of delocalized conjugated p-electrons. This will also improve the resulting photocatalytic characteristics [45].

As a practical application of self-cleaning smart concretes, let us consider concrete tile as cladding for the installation of a hinged ventilated facade. Recently, titanium dioxide nanoparticles were used for the production of concrete blocks. This contributes to the relevance of the present study. However, less was reported about the role of this chemical in air purification [46], and there were no results of studies on the self-cleaning of concrete. Consequently, it is necessary to investigate the effect of the “Nano-titania gradient” method of forming photocatalytic concrete prepared using n-TiO₂ with surface sensitizing on the composite material properties and on the air purification processes and self-purification processes.

In the practical application of self-cleaning smart concretes, special attention should be paid not only to the technology of production of concrete tiles according to the proposed composition but also to the technology of its device as a facing material for a hinged ventilated facade. If the concrete tile is installed incorrectly in the design position, the function of self-cleaning of concrete may be disrupted. It will ensure the quality and correctness of the device of the work performed by carrying out construction control, which consists of input, operational and acceptance control. The entrance control includes the following control operations: checking the documents of suppliers and marking for compliance with the working documentation; checking the quality of the concrete tile surface, dimensions and shape correctness. The sheets should not have delaminations, splinters, cracks or other defects that violate the integrity of the product. Operational control consists of checking the conformity of fixing concrete tiles to the guides (the number of rigid and movable fasteners); checking the number of kleymers and fastening to their guides with rivets. When installing concrete tiles from smart concrete, it is necessary to carry out timely verification of gaps between the cladding elements and the thickness of the seams; verticality and horizontality of the plane, the presence of ledges between the cladding elements, ensuring the wetting angle; straightness of the seams.

2. Materials and Methods

In order to produce a cement-based smart material, Portland cement CEM 42.5H manufactured by “Eurocement Group” was used as a binder, whereas quartz river sand of the deposits “Vakino2”, Kolomna, Moscow region, was used as a fine aggregate, with bulk density of 1540 kg/m³, true density of 2650 kg/m³, and fineness modulus of 2.43.

Commercially available titanium dioxide in an anatase modification, brand “A 200,” manufactured by Shanghai Yujiang Titanium Chemical Manufacturer Co., Ltd., Shanghai, China, was used as a catalyst. The initial average equivalent diameter of TiO₂ particles determined by the laser diffraction on Microtrac S3500 device was 492 nm, with a specific surface area of 3800 m²/kg and bulk density 760 kg/m³. Qualitative and quantitative analyses of the photocatalytic additive were implemented on the X-ray diffractometer ARL X’TRA by ARL company, Switzerland, by means of X-ray powder diffraction method. The results of the analyses are presented in

Table 1. Oxide chemical composition of photocatalysts TiO₂.

Oxide	TiO2	V2O5	SiO2	P2O5	CaO	K2O	Al2O3	Other
Content, %	97.96	0.56	0.37	0.32	0.18	0.17	0.13	0.31

.

It was found that the photocatalyst contained silicon and aluminum oxides, which were surface coating substances. This enables TiO₂ manufacturers to reduce its photoactivity. The oxides of these elements were washed off the surface of titanium dioxide grains during mechanical treatment in aqueous medium. The executed analysis of the results obtained by X-ray diffractometric method of the investigated photocatalyst showed the presence of TiO₂ crystals with tetragonal syngony in the form of anatase modification—98.3% and rutile—1.7%.

Surface sensitization of titanium dioxide was carried out by means of graphene oxide produced by LLC “NanoTechCenter”, Tambov, Russia. Graphene oxide was obtained using an original modification of the Hummers–Offeman method. The need to use sodium nitrate was eliminated. In the final product, the impurity of underexposed graphite was removed, which is usually present when producing graphene oxide using the classical Hummers–Offeman method. Graphene oxide is characterized by the data presented in Figure 4 and Figure 5.

The Raman spectroscopy spectrum obtained by Raman spectroscopy corresponds to graphene oxide. The excitation laser wavelength was 532 nm. Microphotographs were obtained by scanning electron microscopy, Figure 5. The main morpho structural characteristics of graphene oxide were studied with their help.

When analyzing the SEM Tescan MIRA3 LMU microscope by Tescam Orsay Holding, Brno, Czech Republic image of graphene oxide (Figure 5), we can conclude that its structure consists of scales of large lateral size, organized in the folds. According to the energy dispersive analysis, the air-dry product had the following elemental composition, taking the sum of C + O + S = 100% exclusive of other elements: carbon—54.6%, oxygen—43.5%; sulfur—1.9%. In the dry form, graphene oxide forms flexible, sufficiently strong films, consisting of tightly packed graphene-oxide layers oriented parallel to the substrate.

Surface sensitization of titanium dioxide included 2 stages: (1) joint grinding of TiO₂ and graphene oxide (in dry form) to a finely dispersed state, Figure 6a; (2) subsequent heat treatment of TiO₂ with graphene oxide in the presence of oxygen at 190 °C, Figure 6b. This temperature was chosen according to the information from the manufacturer for this product: the decomposition temperature of graphene oxide was over 200 °C. Earlier, in [47] it was found that a higher photoactivity was observed at the content of carbon materials from 0.4 to 0.8% with reference to TiO₂. Therefore, the graphene oxide content was chosen as 0.6% with reference to TiO₂. In the case of surface sensitization, only the surface layer will contain the graphene oxide particles.

During the thermal treatment, the color changes from white to brown and finally to beige, which indicates the completion of the surface sensitization process, Figure 6c.

A recoverable waste cellulose was used as a pore-forming agent for additional formation of a strong and stable frame for placing and supporting titanium dioxide particles, as well as for more intensive adsorption of harmful emissions. To obtain it, waste paper was used, intended for recycling purposes, Figure 7. The size of the cellulose in the dry residue was studied by means of the Nikon Eclipse MA200 optical microscopy complex by Nikon Group, Melville, NY, USA.

Five different samples of dried recoverable waste paper were studied, the sampling was conducted at 5 different points, whereas at least 5 objects were sized on each micrograph. Thus, more than 130-dimensional points were analyzed, and it was found that the diameter of secondary cellulose fibers ranged from 11 microns to 33 microns, and the length of secondary cellulose fibers ranged from 500 microns to 2000 microns.

An application field of cement-based smart material: facade concrete slabs were used to clad walls of buildings and structures in urban agglomerations. The overall dimensions of the considered facade slabs were: 330 × 270 × 30 mm. For manufacturing, heavy weight fine-grained concrete was used, with a strength class B30 and density of 2.135 kg/m³, water-cement ratio of 0.57, workability grade Zh2. The composition of the heavy weight fine-grained concrete is presented in

Table 2. Consumption of concrete mixture components per 1 m³.

Components	Content by Weight, kg
Cement	457.5
Water	260.0
Sand	1418.3
Theoretical density of concrete mix	2136 kg/m3

.

Obviously, when the photocatalyst was placed volumetrically, the concentration of its particles on the surface could be determined by the formula:

$$N_s={\left(\frac{6{\nu }_c{V}_0}{\pi d_c^3}\right)}^{2/3},$$

(1)

where ${\nu }_c$—the volume fraction of photocatalyst particles; $d_c$—the diameter of photocatalyst particle; $V_0$—volume of material. The surface density of the catalyst particles in the layer of thickness dc will be equal:

$${\rho }_s=\frac{1}{4}\sqrt[3]{\frac{36\pi d_c}{S_0}},$$

(2)

where $S_0$—surface area.

To determine the concentration of photocatalyst particles in advance, the results of percolation theory can be used. According to this theory, the concentration of particles forming an infinite cluster must exceed a certain percolation threshold, which depends on the mesh type. The mentioned approach in solving the materials science problem is of fundamental importance, since the catalyst particles are located in the matrix material that ensures their fixation in the material and, consequently, its prolonged functioning. In general, we can assume that the percolation threshold is equal ${\phi }_c$. Then the volume fraction of photocatalyst particles is equal:

$${\nu }_c=\frac{\pi }{6}{\left(\frac{4{\phi }_c{a}_{\mathrm{p}}}{\pi {\left(1+\frac{h}{d_c}\right)}^2}\right)}^{3/2},$$

(3)

where $a_{\mathrm{p}}$—coefficients factoring in the mesh type; $h$—the thickness of the layer between photocatalyst particles.

The results of calculating the photocatalyst volume fraction depending on the mesh type are presented in

Table 3. Volume fractions of photocatalyst in the matrix material of photocatalytic concrete.

Mesh Type	${\mathit{\phi }}_{\mathit{c}}$	${\mathit{a}}_p$	Ratio h/dc
			0.5	1	2	3	5	10
Cubic	0.59	1	0.10	0.0426	0.0126	0.0053	0.0016	0.0003
Hexagonal	0.70	1	0.13	0.0551	0.0163	0.0069	0.0020	0.0003

. The results of calculating the concentration of photocatalyst particles on the surface are presented in

Table 4. Concentration of photocatalyst particles on the surface (N_s).

Mesh Type	Ratio h/dc
	0.5	1	2	3	5	10
Cubic	3.01 × 1035	5.35 × 1034	4.70 × 1033	8.36 × 1032	7.34 × 1031	1.93 × 1030
Hexagonal	5.02 × 1035	8.94 × 1034	7.85 × 1034	1.40 × 1032	1.23 × 1032	3.23 × 1030

.

Based on the data (Table 3) on the volume fractions of the photocatalyst in the matrix material, let us calculate the concentrations as a percentage of the cement mass,

Table 5. Concentration of photocatalyst particles on the surface (N_s).

Cubic Mesh		Hexagonal Mesh
${\mathit{\nu }}_{\mathit{c}}$	TiO2, %	${\mathit{\nu }}_{\mathit{c}}$	TiO2, %
0.0003	0.11	0.0003	0.11
0.0016	0.52	0.0020	0.67
0.0053	1.74	0.0069	2.26
0.0126	4.14	0.0163	5.34

.

Analysis of the calculation results (Table 5) has shown that the formation of a hexagonal mesh requires a higher content of TiO₂ particles; thus, for further studies, the cubic type of mesh was chosen, as the use of concentrations of photocatalyst particles over 2% is not cost-effective.

2.1. Preparation of Concrete Mixture for the Facade Slab

At the first stage, a suspension “dispersed aqueous solution of recoverable waste cellulose—TiO₂/SS TiO₂” was prepared. TiO₂ powder was introduced into the water, then mixing was conducted by the Homogenizer type MPW-324. Due to the porosity of its surface, cellulose would contribute to the deposition of TiO₂ particles on the fibers, thus preventing the sedimentation of photocatalyst particles in the suspension. Therefore, in this dispersed system, it will perform the function of a solid carrier (Support Material).

Then, a fine-grained concrete mixture was prepared, whereas the components were taken in the ratio shown in Table 2. First, dry mineral components (sand and cement) were mixed. Then, the resulting dispersed system “dispersed aqueous solution of recoverable waste cellulose—TiO₂/SS TiO₂” was introduced into the mixing water, with due account of the amount of water contained in the disperse system. Then, recalculation was made. After that, all the components of the fine-grained concrete mixture were combined and mixed until they reached homogeneity. Several concrete mixtures with different ratios of “recoverable waste cellulose—TiO₂/SS TiO₂” were prepared.

Molding of the facade concrete slab was carried out sequentially in 2 stages: at the first stage, the lower part of the facade slab (2/3 h) with the lowest content of particles n-TiO₂/SS TiO₂ (volumetric degree of filling ${\nu }_c$ = 0.0016, …, 0.0020). Then, after compaction of the lower part, the mold was filled with concrete mixture, the upper part of the slab was molded (1/3 h.) with the highest content of particles n-TiO₂/SS TiO₂ (volumetric fill level ${\nu }_c$ = 0.0053, …, 0.0060)—the second stage. Slabs hardening was carried out under standard hardening conditions: 1 day in the forms at 22 ± 0.5 °C and 90–95% humidity, then 27 days without molds under similar hygrothermal conditions.

2.2. Determining Physical and Mechanical Properties of Concrete Specimens

Determination of strength characteristics: ultimate compressive strength in the dry and water-saturated state for cube specimens and ultimate compressive and tensile strength for prism specimens of square section, was carried out according to the Russian state standard GOST 10180-2012 “Concretes. Methods of strength determination using reference specimens”. The specimens that were employed for compressive and flexural strength tests were the cubes specimens with a dimension of 50 × 50 × 50 mm³ and the prismatic specimens with dimensions of 40 × 40 × 160 mm³, respectively. The cubes were immersed in the water tank at 25 ± 1 °C until the time of testing and withstood for 1 h. It is worth noting that cubes and prisms were de-molded after 24 h curing in the molds, then the specimens were cured in a water tank for 28 days. The results were obtained from the mean of 3 specimens for each test to decrease the possible errors. Density was determined according to the Russian state standard GOST 12730.1-78 “Concrete. Methods of determination of density”. Determination of water absorption was carried out according to the requirements of the Russian state standard GOST 12730.3-78 “Concretes. Method of determination of water absorption”.

2.3. Determination of Structure-Sensitive Characteristics of Concrete Specimens

The morphology of the microstructure of the cement mortar and the recoverable waste cellulose contained in the cement mortar sample were studied at the age of 28 days using a FEI Quanta 250 scanning electron microscope by FEI, USA. Using an X-ray diffractometer ARL X’TRA by ARL company, Switzerland equipped with a scanning electron microscope, a microanalysis of the microstructure elements of the cement mortar and recoverable waste cellulose was studied.

2.4. Determining the Performance Characteristics of Concrete Specimens

Determining frost resistance, that is, the resistance of concrete specimens to alternate freezing and thawing was conducted according to Russian state standard GOST 10060-2012. The specimens were saturated with water by immersion in water for 1/3 of their height for 24 h, then the water level was raised to 2/3 of the height of the specimens and kept in this state for another 24 h. After that, the specimens were fully plunged into the water for 48 h. The water-saturated specimens were removed and placed in a freezer. The specimens were frozen and thawed for at least 5 h (one cycle). The specimens were held for 15, 25 and 50 cycles. Determining photocatalytic activity in the decomposition (mineralization) reactions of nitrogen oxides and volatile organic compounds (VOC) was carried out according to Russian state standard GOST R 57255-2016. The method was based on the activation of the tested specimen placed in a flow-through photocatalytic reactor by irradiation with ultraviolet light, Figure 8.

When determining the efficiency of photocatalytic concretes in the decomposition (mineralization) reactions of nitrogen oxides, the amount of pollutant adsorbed in “dark” conditions was estimated. Then the amount of pollutant removed in the process of photo exposure and the desorbed amount of pollutant after light treatment was determined. For this purpose, concrete specimens were placed in a photocatalytic reactor and covered with optical glass. Then, a test gas flow was provided through the photocatalytic reactor without photo-irradiation at a rate of (3.00 ± 0.15) L/min. Then, the gas flow was maintained, and the concrete samples were irradiated for 5 h, whereas the changes in the ratio of NO and NO₂ nitrogen oxides in the gas mixture were recorded. After that, irradiation was stopped, “zero” calibration gas was switched on at the same flow rate, the volumetric NOx content was recorded for 30 min, then the flow of “zero” calibration gas into the reactor was stopped and the concrete specimens were removed from the test setup. After that, the samples were placed in distilled water for 1 h, then removed followed by measuring the change in water volume and recording any light variations, then the specimens were again placed in water for 1 h. When the time expired, the specimens were removed, and the water volume and nitrate and nitrite content were determined again.

The self-cleaning ability of concrete specimens was evaluated by changing the contact angle of wetting on the surface of photocatalytic concrete when exposed to ultraviolet radiation. The test procedure was carried out according to the Russian state standard GOST R 57255-2016. The study used ultraviolet radiation with a wavelength of less than 400 nm, which manifests itself at low intensity in the urban environment. This range was chosen on the basis of previous studies proving the lack of activity of TiO₂ particles outside this range [48,49]. Using drops of distilled water, the initial limited wetting angle was measured at 5 points of each sample before irradiation and the average value was calculated. The measurement of the initial limited wetting angle was performed 3–5 s after applying drops of distilled water to the surface of the photocatalytic concrete samples. After turning on the UV irradiator, changes in the contact angle of wetting over time were measured.

3. Results and Discussion

3.1. Physical and Mechanical Characteristics of Concrete Samples

At the first stage of the study, it was necessary to verify the calculated data on the content of titanium dioxide (Table 5), as well as to optimize the content of recoverable waste cellulose. Therefore, the determination of the optimum content of the amount of titanium dioxide and recoverable waste cellulose fibers was carried out. Optimization was conducted on 100 × 100 × 100 mm cube specimens and 100 × 100 × 400 mm square cross-section prism samples made of fine-grained concrete according to the following quality indicators: the ultimate compressive strength in the dry and water-saturated state for cube specimens and the ultimate compressive and tensile strength for square cross-section prism samples.

The method of mathematical planning of the experiment was used for optimization purposes. To establish the limits of variation of the content TiO₂/SS TiO₂ additives and recoverable waste cellulose, a two-factor composite rotatable experiment was compiled. The factors chosen were X1—the content of photocatalytic additive relative to the mass of Portland cement; X2—the content of cellulose (in dry matter) relative to the mass of Portland cement. The main levels of factors X1 and X2 were 1.0 and 0.75, and the step of variation was 0.5 and 0.25, respectively. An experiment planning matrix was made in natural values,

Table 6. Matrix of test planning (Where compositions 1*–9* are prepared using SS TiO₂).

Composition Number	Code Matrix		Additives’ Content, % Relative to the Mass of Portland Cement
	X1	X2	TiO2/SS TiO2	Recoverable Waste Cellulose
1/1*	−1	−1	0.50	0.50
2/2*	+1	−1	1.50	0.50
3/3*	−1	+1	0.50	1.00
4/4*	+1	+1	1.50	1.00
5/5*	−1.42	0	0.30	0.75
6/6*	+1.42	0	1.71	0.5
7/7*	0	−1.42	1.00	0.40
8/8*	0	+1.42	1.00	1.10
9/9*	0	0	1.00	0.75

.

Based on the obtained results on the physical and mechanical characteristics, a mathematical model and surface response functions for the compressive strength in the dry and water-saturated state for cubic specimens with TiO₂ (Figure 9), for cubic specimens with SS TiO₂ (Figure 10), and on compression and tension in bending for prisms-samples with TiO₂ (Figure 11), for prisms-samples with SS TiO₂ (Figure 12) were made. On this basis, the limits of variation of the optimal content of additives “recoverable waste cellulose—TiO₂/SS TiO₂” in the composition to obtain effective photocatalytic composites were determined.

When the concentration of cellulose fibers increases, the ultimate strength is decreased according to the linear law, which corresponds to the already known data on similar composites [50]. From the analysis of the experimental–statistical model of the ultimate compressive strength (Figure 9 and Figure 10), it can be concluded that the optimum content of TiO₂ was in the range from 0.8 to 1.1%, and for cellulose from 0.4 to 0.8%, the optimum content of SS TiO₂ was in the range from 0.7 to 1.1%, for cellulose—has not changed. From the analysis of the experimental and statistical model of the ultimate bending strength (Figure 11 and Figure 12), the optimal TiO₂ and SS TiO₂ content were between 0.6 and 1.0%, and between 0.4 and 0.8% for cellulose.

Titanium dioxide and SS TiO₂ of anatase modification have weak reactivity and the composition of the composite lead to an increase in strength in the zone of optimum due to the compaction of the structure and filling of the pores. Changes in the average density and water absorption values indirectly characterize the porosity parameters of the composites obtained, which, in turn, affects the surface area for photocatalytic reactions. Therefore, in order to establish the dependence of porosity on the composition of the obtained composites, the density and water absorption parameters were recorded,

Table 7. Composition and properties of composite binders (Where compositions 1*–9* are prepared using SS TiO₂).

Composition Number	Additives’ Content, % of the Portland Cement Mass		Output Parameters		Composition Number	Output Parameters
	TiO2/ SS TiO2	Cellulose	Density, kg/m3	Water Absorption by Weight, %		Density, kg/m3	Water Absorption by Weight, %
K	0	0	2064	5.3	K	2064	5.3
1	0.5	0.5	2129	4.8	1*	2130	4.8
2	1.5	0.5	2069	4.6	2*	2071	4.5
3	0.5	1	2057	4.4	3*	2061	4.4
4	1.5	1	2054	4.7	4*	2057	4.6
5	0.3	0.75	2036	3.0	5*	2039	3.0
6	1.7	0.75	2005	3.2	6*	2007	3.1
7	1	0.4	2048	4.8	7*	2050	4.7
8	1	1.1	2026	4.6	8*	2029	4.5
9	1	0.75	2073	4.8	9*	2075	4.8

.

When the content of cellulose exceeds 0.8%, it aggregates in the free state, which leads to a drop in density and strength (Figure 9, Figure 10, Figure 11 and Figure 12) of the obtained composites. Swelling cellulose fibers absorb a significant volume of water, thereby increasing the viscosity of the cement mortar. Then, in the process of hardening the binder, water is separated from the swollen cellulose due to diffusion processes and contributes to the formation of concentrated pores in the composite. Titanium dioxide, in its turn, is a water-insoluble component that reduces the water absorption index by filling the capillary pores, which correlates objectively with the data of the mathematical planning of the experiment. As the results showed (Table 7), surface sensitization of TiO₂ has no significant effect on the indicators of density and water absorption; the changes were less than 1% and 2–3%, respectively.

Thus, there were verified calculated data on the content of titanium dioxide (Table 5), and the content of recoverable waste cellulose was optimized in the formation of the gradient structure of the facade board. The formation of slab layers will be carried out by taking into account the extremes of the optimal range of titanium dioxide content: the bottom layer of the slab will contain 0.6%, and the top layer, 1% of titanium dioxide/surface-sensitized titanium dioxide. The recoverable waste cellulose content on a dry-matter basis will be 0.75% for the two layers of the gradient board.

3.2. Structure-Sensitive Characteristics of Concrete Specimens

The formation of the cement stone structure, the value of total porosity, as well as capillary-porous structure parameters depend not only on the value of the water-binder ratio and the degree of hydration [51] but also on the type of a microfiller, which is the added cellulose. To study the morphology of the modified cement mortar, the microstructure of the binder at 28 days was studied using a FEI Quanta 250 scanning electron microscope, Figure 13. Since the morphology of the specimens prepared with TiO₂ and SS TiO₂ did not differ in terms of visible features, thus we will consider the morphology on the example of TiO₂.

The microstructure of the cement mortar (Figure 13) is characterized by the presence of developed new formations of calcium hydro silicates in the micropores, which contribute to the density. This is confirmed by the data of energy dispersive elemental analysis of the microstructure, Figure 14.

In this case, while studying all the surface areas, titanium dioxide was present in crystalline form on the chips. However, its presence was significantly inferior to the products of Portland cement hydration. The formation of large crater-like pores on the chipped cement stone was explained by interfibrillar and intercrystalline swelling of cellulose in water, which leads to the formation of a more developed surface of cellulose fibers.

According to the theoretical and experimental data of the authors [52], it was found that cellulose macromolecules have a strong tendency to aggregate into highly ordered structural elements due to their chemical structure, spiral conformation and stereoregularity. The high specific surface area of cellulose fibers is a significant factor in filling capillaries, pores and sorption of molecules from aqueous media that leads to its swelling. The change in the size and shape of cellulose fibers is caused by interfibrillar and intrafibrillar swelling of cellulose and includes three main moments: filling the capillary space between the fibrils with liquid, filling the pores on the surface of the fibers and the interaction of liquid with cellulose macromolecules.

When studying the morphology of recoverable waste cellulose contained in a specimen of cement slurry (Figure 15), the presence of mineral substances formed was discovered during the process of Portland cement hydration.

It is known [53] that calcium hydroxide interacts with the surface layer of cellulose and attaches to the hydroxyl groups of cellulose to form additive or molecular compounds under the following scheme:

RcellOH + Ca(OH)₂ → RcellOH·Ca(OH)₂

2RcellOH + Ca(OH)₂ → (RcellOH)₂·Ca(OH)₂

Such compounds are formed as a result of interaction in stoichiometric ratios of two components due to intermolecular bonds, in particular hydrogen bonds. Thus, cellulose forms (C₆H₁₀O₅)₂·Ca(OH)₂ and (C₆H₁₀O₅)·Ca(OH)₂ with calcium hydroxide. Similar compounds are known for other polyatomic spirits: glycerol, mannitol and simple sugars.

Due to the ability of cellulose fibers to swell in liquids, the surface of cellulose fibrils becomes more accessible to microparticles, which are evenly distributed within the fibers. The implemented microanalysis of components (Figure 16) shows that in addition to the formed calcium hydroxide, titanium dioxide is also present.

Thus, when studying the structure of composites, it was found that the recoverable waste cellulose in the form of dispersed solution was able to provide a stable framework for placing and supporting titanium dioxide particles. Consequently, their more intense adsorption on the surface of swollen cellulose fibers remained thereafter the hydration processes.

3.3. Performance Characteristics of Concrete Specimens

Due to the fact that corrosive factors during operation affect the upper layer of the plate, thus the study of the effect of the content of the system “recoverable waste cellulose—TiO₂/SS TiO₂” on the resistance of concrete facade plates to alternate freezing—thawing, was exemplified by the compositions number 9 and 9* as opposed to the reference composition “K” (Table 7). The results are shown in Figure 17.

Having analyzed the obtained results of determining the frost resistance of the studied specimen compositions, it was found that both variants of the introduced additives have a significant increase in this parameter. The increase by up to 33% of photocatalytic concrete resistance to repeated cyclic effects of alternating temperatures was explained by the complex influence of cellulose fibers and titanium dioxide. In the freshly prepared concrete mixture, the cellulose fibers were in a water-saturated state, which led to a significant increase in their volume. Subsequently, the hydration of cement caused the migration of water from the fibers with a subsequent decrease in volume. Due to this process, closed air pores (reserve cells) were formed and penetrated by capillaries after hydration. The pores formed in the space around the cellulose fibers were reserve cells, which contained excess water from freezing and help to compensate for the stress on their walls. At the same time, due to the water-solubility of titanium dioxide, there was additional compaction in the capillary pores of the concrete, which prevents the free movement of water and further restrains the saturation of air pores, increasing the frost resistance in the complex.

When studying the photocatalytic activity and self-cleaning ability of the surface, facade plates with a hybrid structure were used (Figure 18). The plates were manufactured according to the optimized ratio of components; thus that the bottom layer of the plate contained 0.6% and the top layer contained 1% titanium dioxide/surface-sensitized titanium dioxide. The recoverable waste cellulose content on a dry matter basis was 0.75% for the two layers of the gradient plate. The results of determining the photocatalytic activity are presented in

Table 8. Photocatalytic activity of facade concrete plates with a gradient structure.

Parameter	Russian State Standard GOST R 57255-2016	with TiO2	with SS TiO2
Decomposition (mineralization) of air pollutants: - nitrogen oxide NOx,% by volume, not less	60	75	95
- volatile organic substances,% by volume, not less	80	85	112

.

The analysis of the obtained results (Table 8) shows the effectiveness of the method of surface sensitization of titanium dioxide by means of graphene oxide, significantly increasing the photocatalytic activity. Thus, the decomposition of nitrogen oxide by volume with the modified surface increased by 27% in comparison with the classical titanium dioxide, and the decomposition of volatile organic substances by 32%. The increase in photocatalytic activity was due to the fact that graphene performs the rapid transfer of a photoinduced electron between itself and TiO₂, which effectively suppresses the recombination of photogenerated electron-hole pairs in TiO₂. In addition, the presence of residual oxygen-containing groups in the reduced graphene oxide leads to additional fixation of TiO₂ due to Van der Waals interaction, which suggests an increased prolongation of the increased photocatalytic effect. All of this indicates a high potential for surface-sensitized TiO₂ application.

When studying the self-cleaning ability of photocatalytic concrete, the wetting contact angle was determined, and the results are presented in

Table 9. Wetting contact angle in time for the facade concrete plates with a gradient structure.

Intervals between Measurements, Hours	with TiO2	Variation Coefficient, %	with SS TiO2	Variation Coefficient, %
initiating (3–5 s)	35°17′41″	–	37°9′31″	–
1	32°6′1″	15.0	27°7′2″	12.0
2	28°26′56″	13.0	19°44′16″	10.0
3	27°37′40″	11.0	16°35′16″	4.0
4	21°37′40″	5.0	–	–

. The wetting contact angle was measured at five points on each sample of the facade plate by means of distilled water application. The measurement was conducted before irradiation and then at equal time intervals of two hours.

The effectiveness of the self-cleaning ability of photocatalytic concrete was evaluated by the time of reaching the equilibrium wetting contact angle after the start of irradiation. According to the requirements of the Russian state standard GOST R 57255-2016 to photocatalytic concrete in terms of self-cleaning ability, it should last no more than 80 h. After each measurement of the wetting contact angle, the coefficient of variation was calculated, and when the value was less than 10%, the measurements were terminated.

It is assumed that the surface has hydrophilic properties when a wetting contact angle does not exceed 20. During the test (Table 9), it was found that for the façade plate manufactured with the use of surface-sensitized TiO₂, the self-cleaning process was completed in 3 h after the start of irradiation, as evidenced by the low value of the wetting contact angle of 16°35′16″ and the coefficient of variation equal to 4%. At the same time, in the sample of façade plate manufacture with TiO₂, the coefficient of variation was 5% only after four hours of testing. In this case, the value of the wetting contact angle exceeded 20°.

Thus, in this work, the possibility of applying the method of surface sensitization with graphene oxide to increase the photocatalytic performance and self-cleaning ability was proven. However, the authors will extend this study, which will be aimed at optimizing the content of graphene oxide and conditions of surface sensitization. In further research, the authors plan to substantiate the requirements for the technology of the device of concrete tiles based on the proposed composition in the design position as a facing material of a hinged ventilated facade and propose methods of construction control to ensure the quality and operability of the enclosing element of the building.

4. Conclusions

A classification of titanium dioxide introduction methods is proposed: surface “Nano-titania outside,” internal “Nano-titania inside,” and a new method—hybrid “Nano-titania gradient” is developed.

A method of surface sensitization of titanium dioxide (SS TiO₂) was proposed by means of graphene oxide. The method included two stages: (1) joint grinding of TiO₂ and graphene oxide (in dry form) to a finely dispersed state; (2) subsequent thermal treatment of TiO₂ with graphene oxide in the presence of oxygen at 190 °C.

Using the theory of percolation, the concentration range of the photocatalyst content was calculated. It was shown that the economically optimal mesh type was cubic, whereas the TiO₂ content should be in the range from 0.11 to 1.74% of the cement mass. Verification of calculated data on TiO₂ content and optimization of recoverable waste cellulose content was carried out according to the strength of fine-grained concrete samples in the dry and water-saturated state, as well as on the tensile strength at bending. The analysis of the experimental-statistical models of the compressive strength shows that the optimum content of TiO₂ was in the range from 0.8 to 1.1%, and cellulose from 0.4 to 0.8%, the optimum content of SS TiO₂ was in the range from 0.7 to 1.1%, and cellulose from 0.4 to 0.8%. Analysis of the experimental and statistical model of the bending strength shows that the optimal content of TiO₂ and SS TiO₂ was in the range from 0.6 to 1.0%, and cellulose from 0.4 to 0.8%. Surface sensitization of TiO₂ had no significant effect on the indicators of density and water absorption; the changes were less than 1 and 2–3%, respectively.

Formation of the gradient structure of the facade plate was carried out, taking into account the extreme points of the optimal range of titanium dioxide content: the bottom layer of the plate contains 0.6%, and the top layer, 1% of titanium dioxide/surface-sensitized titanium dioxide. Recoverable waste cellulose’s content on a dry matter basis is 0.75% for the two layers of the gradient plate.

When studying the structure of composites, it was found that titanium dioxide was sorbed on the surface of swollen cellulose fibers and remained there after the process of cement hydration. Due to the increase in porosity of the cement stone, the potential useful surface area for photocatalytic reactions increased. Thus, when studying the structure of composites, it was found that the recoverable waste cellulose in the form of the dispersed solution was able to provide a stable framework for placing and supporting titanium dioxide particles.

The increase by up to 33% of photocatalytic concrete resistance to repeated cyclic effects of alternating temperatures was explained by the complex influence of cellulose fibers and titanium dioxide. In the freshly prepared concrete mixture, the cellulose fibers were in a water-saturated state, which led to a significant increase in their volume. Subsequently, the hydration of cement caused the migration of water from the fibers with a subsequent decrease in volume. Due to this process, closed air pores (reserve cells) were formed and penetrated by capillaries after hydration. The pores formed in the space around the cellulose fibers were reserve cells that contained excess water from freezing and helped to compensate for the stress on their walls. At the same time, due to the water-solubility of titanium dioxide, there was additional compaction in the capillary pores of the concrete, which prevented the free movement of water and further restrained the saturation of air pores, increasing the frost resistance in the complex.

The effectiveness of the method of surface sensitization of titanium dioxide by combining it with graphene oxide was first shown for cement-based materials. Thus, the decomposition of nitrogen oxide by volume with the modified surface increased by 27% in comparison with the classic titanium dioxide, and the decomposition of volatile organic substances by 32%. The increase in photocatalytic activity was due to the fact that graphene performs the rapid transfer of a photoinduced electron between itself and TiO₂, which effectively suppresses the recombination of photogenerated electron-hole pairs in TiO₂. In addition, the presence of residual oxygen-containing groups in the reduced graphene oxide leads to additional fixation of TiO₂ due to Van der Waals interaction, which suggests an increased prolongation of the increased photocatalytic effect.

It was found that for the facade plate made with surface-sensitized TiO₂, the process of self-cleaning was completed after 3 h after the irradiation start, as evidenced by the low value of the contact wetting angle 16°35′16″ and the variant coefficient equal to 4%. Whereas in the sample of facade board prepared with TiO₂, the coefficient of variation was 5% only after four hours of testing. In this case, the value of the contact wetting angle was more than 20°.