**1. Introduction**

Population growth and the consequent demand of more infrastructure have increased the consumption of ordinary Portland cement (OPC), which is the main cementitious material in concrete technology. In spite of the benefits of concrete, there are several factors, such as high energy consumption, high CO2 emissions, and limited limestone reserves that have led to the reconsideration of the use of this material [1]. Owing to the above, measures have been taken both to improve the production processes of Portland cement, via the use of alternative fuels and different sources of raw materials to develop new cementitious materials that equal or exceed the properties of Portland cement and, in turn, exhibit better environmental performance. Geopolymers have been considered as an alternative at a global level; they are considered to be the third generation of cement, after lime and OPC. It is estimated that, depending of the precursors and activators, the production of geopolymers results in ~70% less greenhouse gas emissions than the production of cement, which makes some geopolymers environmentally friendly [2–4].

Geopolymers are obtained from the optimum blend of a material, mineral or industrial by-product based on SiO2 and Al2O3 (precursors) with a chemical agent (alkaline activator), which, through a series of reactions at low temperature (<100 ◦C), leads to the formation of a product with cementitious characteristics [5]. The main criterion for the stable development of a geopolymer is that the source of the aluminosilicate material is highly amorphous and contains sufficient reactive glass content. The main alkaline activators used are sodium hydroxide (NaOH), potassium hydroxide (KOH), sodium silicate (Na2SiO3), and potassium silicate (K2SiO3). Compared with NaOH, KOH exhibits a higher degree of alkalinity; however, it has been reported that NaOH has a greater ability to separate the monomers of silicates and aluminates in the phase of dissolution of the precursor [5]. Geopolymer

cements are considered materials that can replace Portland cement owing to properties such as high resistance at early ages, resistance to chemical attack, low thermal and acoustic conductivity, and high temperature and fire resistance; these properties, depending of the type of raw materials and mix formulation of the geopolymer, can be similar, or even superior, to those of traditional Portland cement.

As in the case of Portland cement, the addition of nanoparticles to geopolymers has focused on improving the mechanical properties. For example, Assaedi et al. added SiO2 nanoparticles in a proportion of up to 3 wt % to a fly ash-based geopolymer, finding that the addition of these materials improved the mechanical performance of the material by 27%, while decreasing the porosity and water absorption [6]. Riahi and Nazari synthesized geopolymers from rice husk ash and fly ash, to which SiO2 and Al2O3 nanoparticles were added; they also obtained an improvement in the mechanical properties of the material [7]. However, the best performance was for the samples with added nano-silica, owing to their greater amorphous character, unlike the alumina nanoparticles, which, because they were crystalline, did not participate in the process of geopolymerization [7]. Recently, Duan et al. developed geopolymers from fly ash, focusing on the mechanical performance and durability to carbonation when nano-titanium oxide (TiO2) particles were added; the authors reported that the addition of these up to 5 wt % improves the mechanical properties and durability and attributed this behaviour to pore refinement and densification of the geopolymer microstructure [8]. Similarly, Yang et al. studied the mechanical and physical behaviour of an alkaline activated slag to which TiO2 nanoparticles were added at 0.5 wt % and reported a decrease in porosity and material shrinkage; in addition, the material was densified and had a greater resistance to compression [9]. On the other hand, Zhang et al. evaluated the effect of adding titanium dioxide and hollow glass microspheres to a geopolymer coating from metakaolin (MK) on the geopolymer's optical and thermal properties [10]. The study found that by adding hollow glass microspheres at 6 wt % and titanium dioxide based on the cementant at 12 wt %, increases of 12% and 90% of the thermal insulation and reflectivity, respectively. The authors attributed this behaviour to the pigment character of the titanium dioxide, for which the high coefficient of refraction enabled it to reflect wavelengths close to infrared and visible light [10].

In other studies that were not particularly focused on geopolymer materials, it has been found that nanoparticles of TiO2, because they are a semiconductor substance, are able to confer photocatalytic properties to the system to which they are added [11], and this results in self-cleaning and air-purifying properties, which is highly desirable for building materials.

Based on the above, the present article aims to evaluate the effect of the addition of TiO2 particles on the physical and mechanical behaviour of a geopolymer based on MK as a precursor. The variables considered are the liquid/solid ratio (L/S) of the system and the percentage by weight of TiO2 added. Properties in the fresh state, such as the setting time and flowability, and properties of the hardened state, such as the compressive strength, density, absorption, and porosity, are studied. The study is complemented with a microstructural analysis of the geopolymers using the X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy (SEM) techniques. The contribution of the photocatalytic properties of TiO2 to the geopolymer, specifically its self-cleaning and algaecide properties, will be the object of future studies.

#### **2. Materials and Experimental Methodology**

#### *2.1. Materials*

A high-purity commercial MK (Metamax BASF, Florham Park, NJ, USA) was used as the primary source or precursor in the production of the geopolymer material, whose chemical composition is presented in Table 1. The chemical composition was determined via X-ray fluorescence (XRF) using a Phillips MagiX-Pro PW 2440 spectrometer (PANalytical, Tollerton, UK) equipped with a Rhodium tube and a maximum power of 4 kW. The SiO2/Al2O3 molar ratio of the MK was 1.97. High-purity analytical titanium oxide (TiO2; Merck, reference 1008081000) was used as an additive. Particle size

and distribution analysis was performed with a Mastersizer-2000 laser granulometer from Malvern Instruments (Malvern, UK) with a Hydro2000MU dispersion unit; distilled water was used as a dispersant medium. The mean particle sizes D (4,3) of the MK and titanium oxide were 6.57 and 1.59 μm, respectively (Figure 1).


**Table 1.** Chemical composition of the precursor (MK).

**Figure 1.** Particle size distribution of (**a**) titanium dioxide, and (**b**) MK.

Figures 2 and 3 show the XRD and FTIR results for MK and TiO2, respectively. For the XRD analysis, a Bruker diffractometer equipped with a wide-angle goniometer RINT2000 was used, using the Kα1 signal of Cu at 45 kV and 40 mA. A 0.02◦ pitch was used within a range of 5–70◦ at a rate of 5◦/min. Information processing was performed using the X'pert HighScore Plus software package, version 2.2.5. FTIR spectroscopy was performed using a PerkinElmer Spectrum 100 spectrometer (Perkin Elmer, Shelton, United States) in transmittance mode in the frequency range of 450–4000 cm<sup>−</sup>1. Samples were prepared using the KBr compressed method. In Figure 2, it can be observed that MK has a high level of amorphicity due to the halo located in the range between 20 and 30◦ 2θ and presents small traces of a crystalline product identified as anatase (reference pattern 01-078-2486), which corresponds to the observed peaks at approximately 25◦, 38◦, 48◦, 55◦, and 63◦ 2θ. This coincides with the percentage of TiO2 reported in the chemical composition of MK (Table 1). There is also evidence that the TiO2 used is in the anatase phase, which is an important phase because of its photocatalytic potential [11]. In the FTIR spectrum (Figure 3) of the MK, a band at approximately 3437 cm−<sup>1</sup> is observed; this corresponds to the asymmetric vibration of OH–groups. At 1089 cm−1, a Si–O–Al vibrational peak is observed, which corroborates the aluminosilicate character of this raw material. The peaks at 814 and 473 cm−<sup>1</sup> are attributed to the amorphous Al–O stretching vibration and Si–O–Si flexion, respectively [12]. On the other hand, the FTIR spectrum obtained for TiO2 indicates the presence of peaks near 3430 and 1632 cm−1, which correspond to vibration via stretching of the –OH

bonds and vibration via deformation of the bonds in adsorbed surface water molecules, respectively. The peak located near 687 cm−<sup>1</sup> is characteristic of the Ti–O–Ti stretching of the anatase phase, thus confirming the presence of these photoactive species [13].

**Figure 3.** FTIR spectra of the raw materials.
