The standard mortar composition involves use of the sulphate-resistant Portland cement CEM I 42.5R-SR 3 as a binder, a siliceous (more than 98% SiO₂) standard CEN sand according to the European Standard EN 196-1:2016 (NORMASAND) (23) and two types of nanomaterials as additions. The chemical and physical properties of the cement are presented in Table 1. A commercial nano-SiO₂ Levasil 200/40%, distributed by Obermeier and a commercial Iron III oxide nano-particles (nano-Fe₂O₃) supplied by Tecnología Navarra de Nanoproductos S.L. (TECNAN) were used (Table 2). Tap water from Canal de Isabel II (located in Madrid) was used to prepare the mortars. A modified polycarboxylate-based superplasticizer or high-range water-reducing additive (HRWRA) according to EN 934-2 (24), Sika ViscoCrete^®-5720 (pH = 4; density = 1.09 kg/l and 36% solid content) was employed with a relatively low dosage of 0.15–0.3% per cement weight (Table 2). Due to its chemical structure, it enabled good particle dispersion to be obtained.

The particle size of the commercial nano-SiO₂ Levasil was 10-20 nm (this value was supplied by the producer). The specific surface area of the commercial Iron III oxide nano-particles (nano-Fe₂O₃) was 55.9 m²/g, and the mean particle size 1.95 µm, with both being experimentally measured.

Prior to testing the mortars, the mix design shown in Table 3 was selected in order to maximise the information recorded regarding the combined use of nano-Fe₂O₃, nano-SiO₂ in mortars.

All the mortar mixes had the same water/cement ratio of 0.5. The water provided by the Nano-SiO₂ solution was deducted from the mixing water of the mortar. In addition, a high-range water-reducing admixture (HRWRA) was adopted in order to obtain similar slump in all mixes.

Flexural and compressive strength tests were performed in 40x40x160mm mortars at two, seven and 28 days according to the European Standard EN 196-1:2016 (23). The specimens were cured at 95% relative humidity in a humidity-testing cabinet at 20.0±1.0ºC.

The curing process of the specimens was carried out according to the European Standard EN 196-1:2016. This standard method allows results to be compared under the same curing conditions for all the specimens. In accordance with Sajedi and Razak, (29) the compressive strength of the mortar is highly dependent on the curing conditions.

Flexural strength was measured on three specimens for each mortar mix. The six samples obtained after the flexural strength testing were used for compressive strength testing. Flexural strength values were calculated according to the following equation [1].

where R_f is flexural strength (MPa), F_f is load applied in the middle of the specimen (N), l is side of the prism (mm) and b is distance between the two steel supporting rollers (mm).

A mortar porous system (with total open porosity and pore-size distribution) was studied at 28 days by means of mercury intrusion porosimetry (MIP) in "Phi mayúscula"12x40 mm cylindrical samples, in a range of pore radius between 0.005 and 180 μm following an internal procedure based on the ASTM D4404-04. The samples were oven-dried at 40±5°C for four days and then analysed with use of a Micromeritics AutoPore IV 9599 porosimeter.

The porosity of a material affects its physical properties and, subsequently, the mechanical strength and durability performance. In particular, the total porosity and pore-size distribution (PSD) give information regarding the open porosity. Therefore, it could serve as an indirect indicator of the permeability of the sample in certain cases. In essence, porosity measurement is a suitable tool in enabling an understanding of microstructure evolution and potential use of nano-oxide-made mortars.

By measuring the volume of mercury that intrudes into the sample material with each pressure change, the volume of pores in the corresponding size interval is obtained. Then, the total porosity can be calculated according to equation [2].

where P_t is the total porosity (%), V_p is the porous volume (mm³) and V_m is the sample volume (mm³).

Resistivity is a non-destructive test method described extensively in the Spanish Standard UNE 83988-1 (26). The test was performed on 40 x 40 x 160 mm mortar prisms at 60 days for M6Si2Fe and M4Si4Fe specimens and at 90 days for M2.35Si4Fe y M3.5Si2Fe specimens. Electrical resistivity was determined by use of the two-electrode method (Giatec RCON™) that measures electrical resistance with frequencies between 1 Hz and 30 KHz (Figure 1). Equation [3] was used to calculate electrical resistivity.

where R is electrical resistance, ρ is electrical resistivity, V is potential difference, I is current intensity, l is steel plate distance and A is area (Figure 1).

Electrical resistivity measurements give information about the open porosity of the mortar and concrete. The Nernst-Einstein equation expresses the relationship between the electrical resistivity and ion diffusivity for porous materials (27). Therefore, it is possible to estimate the diffusivity of the mortar and concrete by measuring resistivity, as shown in equation [4].

where D_i is diffusivity for ion i, ρ is electrical resistivity and k is a constant value obtained from the slope of the linear correlation between the ion diffusivity and electrical conductivity, which is the inverse parameter of the electrical resistivity (Table 4).

A clear relationship between corrosion rate and electrical resistivity can be found in most of the cement-based materials (28).

The results obtained for flexure and compressive strength at two, seven and 28 days, and their standard deviation in MPa, are shown in Figure 2 and Figure 3, respectively.

All the mortars made with nano-oxides showed compressive strength results below the reference mortar made without additions. These results differ from others reported elsewhere (7-16). Usually, nano-Fe₂O₃ and nano-SiO₂ helps to improve the mechanical performance of mortars and concretes due to the filler effect of both additions and the pozzolanic effect of the last one (22). Therefore, the results reported in this paper suggest that the curing conditions play an important role with regard to the compressive strength.

Nano-SiO₂ performance has customarily been reported as increasing the compressive strength of mortars (7-9) and concretes (11-13) as a result of the pozzolanic and filler effect (22). The finer the SiO₂ in the mortar, the higher is the strength of the cement-based material (5). Moreover, the grain size distribution of the material (GSD) is also found to be of significant importance (5). In contrast, the results obtained in this paper are reversed. The incorporation of nano-SiO₂ and nano-Fe₂O₃ in the mortars leads to less compressive strength, as can be observed in Figure 3. This negative effect on the mechanical strength of mortars can be produced not only by insufficient nano-SiO₂ and nano-Fe₂O₃ dispersion during the mixing operation (7, 22), but also by an unsuitable curing method (in a cabinet at 95% relative humidity). On many occasions, those published studies that identified enhanced mechanical strength cured the mortars in full immersion (7, 9). Such a curing condition also leads to lower mortar compressive strengths than mortars cured at standard ones which are 30 MPa at two days and 61 MPa at 28 days (Table 1).

According to Sajedi and Razak (29), strength loss in Portland cement mortars depends on the cement fineness and regime of curing. They suggest that the main cause of mortar strength loss with the use of fine binders is due to phase separation, given that there is insufficient interlocking on the interfacial area. Another cause related with loss of strength involves the formed hydrates having no time to become arranged suitably. Lastly, the lack of internal water in the mortar could avoid proper hydration.

The earlier mentioned MIP was performed at 28 days and the results are shown in Figure 4. The main peak identified was the critical diameter for each mortar sample, as Table 5 shows.

Mortar M4Si4Fe had the highest critical diameter (1.054μm), followed by mortar M6Si2Fe (0.834μm). This value decreased sharply to 0.183μm for the mortar with 3.5% nano-SiO₂ (M3.5Si2Fe) and to 0.151μm for the mortar with 2.35% nano-SiO₂ (M2.35Si4Fe). Contrary to what was expected, the reference mortar presented the lowest critical diameter (0.063). The curves could be classified in three main groups: with about 8%, 6% or 0% of nano-oxides. The curves moved from the left to the right when nano-oxides were added, showing a pore-size increase. These results were in line with the compressive and flexural strength ones.

In agreement with the mechanical results, critical mean diameter pores determined by MIP could be ordered from the higher to lower size as follows:

M0 < M2.35Si4Fe < M3.5Si2Fe < M6Si2Fe < M4Si4Fe

Often nano-additions, as well as other additions, act by refilling the open spaces between the aggregates and cement paste. This improves the quality of the transition zone. They are located in the capillary pores and refine pore-size distribution (17). In particular, nano-SiO₂ combined with the Ca(OH)₂ produced in the calcium silicates hydration creates a secondary C-S-H gel at later ages, which is deposited in the capillary pores among other parts of the cement paste porous system (8).

Whereas Figure 5 plots the total porosity of the mortars at 28 days, Figure 6 shows the pore-size distribution (PSD). With regard to the total porosity, it may be observed that all the values obtained from mortars with nano-oxides showed high values from 18.8% to 20.9%, double those of the total porosity of the reference mortar (M0). Thus, it is clear that nano-additions in a range of 4-8% of cement replacement in mortars, when mixed and cured following the procedure previously mentioned, produce an increase in total porosity of the mortars with regard to the reference one. This detrimental effect can be explained in two ways: i) the agglomeration effect of nano-oxides in mortar (7, 21, 22) was not fully addressed by using a commercial polycarboxylate-based superplasticizer admixture; ii) pozzolanic materials were highly sensitive to curing conditions, as they require a longer wet period with a continuous water supply (1). Therefore, 95% of relative humidity curing could be considered insufficient for mortars containing 4-8% of nano-oxides.

As a result, in the PSD analyses of mortars (Figure 6) the capillary-pore content increases with the nano-oxide amount which provides agreement with the flexural and compressive strength results.

In theory, nano-particles are uniformly dispersed in mortar and when hydration begins, hydrate products diffuse and cover nano-particles (13) which could restrict the growth of hexagonal Ca(OH)₂ crystals and promote cement hydration nucleation sites. The result is a more homogeneous and compact matrix (13). In this research, however, it was found that the more nano-SiO₂ there was in the mortar, the greater was the capillary-pore formation. Then, the reference mortar showed the lowest capillary porosity. Although such results agree with the compressive strength results, they differ from those provided in other published studies (7-16).

Other researchers (14) have observed that nano-silica (NS) added at ratios of 0.5% to 1.25% improved compressive strength of the mortars at all ages compared with the control specimens (the rate of improvement varied between 5% and 15%). In specimens containing nano-silica powder at a rate of 2.5%, strength reduction appeared to vary between 43% and 60%. In addition, for nano-Fe₂O₃ it is observed that with the decrease in nano-Fe₂O₃ volume fraction (from 10% to 3%) the compressive strength increases at all ages (8).

Resistivity measurements have recently been used to assess the quality of cement-based materials (26). There is a clear relationship with the chloride diffusion coefficient (27) or the corrosion-rate measurements (28). Figure 7 shows two main groups with regard to the nano-oxide performance in the mortar.

Whereas the first group included M2.35Si4Fe and M3.5Si2Fe and exhibited electrical resistivity of around 100 Ω^.m, the second one reached values below 350 Ω^.m. Consequently, these last mortars will perform better in aggressive environments than the first ones (26, 27, 28). Therefore, the higher amount of nano-SiO₂ promoted the increase in durability. By comparing these values with those presented in Table 4, it could be argued that the first group of values corresponded to mortars with moderate chloride diffusivity, while the second one should have had very low chloride diffusivity.

Summarising, electrical resistivity measurements were recorded in order to obtain an indirect durability indicator of the mortars. In this case, as the amount of nano-oxides in the mortar increased, a greater degree of durability was found.

According to Andrade and D’Andrea (30), as resistivity is a property that depends on the concrete porous system and its degree of moisture, it is possible to find relations between diffusivity and resistivity. The authors have also suggested that resistivity could be used as a corrosion indicator.