*2.2. Methods*

2.2.1. Manufacturing and Curing Process of the Mortars and Acid Attack Simulation

The different tests were performed on normalised mortar specimens—4 × 4 × 16 cm. The Xypex powder admixture was previously dry-mixed with the cement for 1 min with a laboratory mortar mixer. After that, the mortars were manufactured following the procedure of the standard UNE-EN 196-1 [40]. A total of 48 specimens were manufactured, 12 per each type of mortar (A, B, C and D). The specimens were cured in a humidity chamber for 28 days at 95% ± 2% of relative humidity and 20 ± 2 ◦C to complete the hardening process. After 28 days of curing, 24 of the specimens (representing half of the total of 48, that is, 6 specimens for each type of mortar) were exposed to a sulphuric acid solution for 90 days, and the other 24 were immersed in water for the same amount of time for reference purposes.

There are no European standards to test the chemical resistance of cement-based materials. However, Sokolowska et al. [41] studied the tests on cement-based materials according to ASTM (American Society for Testing and Materials) standards and they concluded that there is a lack of clear criteria for the evaluation of research results via ASTM methods. Due to the absence of standardised tests, in this study, the sulphuric acid attack was performed by immersing the specimens into an H2SO4 solution (3% w/w) in hermetically closed containers. This procedure has been used in previous studies to analyse the effect of acidic environments on cement-based materials [23,24]. A high concentration of sulphuric acid was chosen in order to accelerate their effects on the mortars and obtain the same degradation in less exposure time [12]. The performance of mortars against the acid attack was evaluated taking into account common parameters used in the literature for this purpose, such as mass variation and mechanical properties decrease [41]. The ultrasonic pulse velocity and the capillary water absorption were also studied [24]. The volume of solution was approximately four times the volume of the samples, as suggested by the ASTM C 1012-04 standard [42]. The H2SO4 solution was replaced weekly for a new solution (3% w/w) so that the concentration of sulphuric acid had minimal variation. After removing the specimens from the acid solution, they were brushed under a flow of water in order to remove the superficial layer of adhered material. After that, the specimens were introduced into a new solution. Once the acid exposure was finished, the specimens were dried at 105 ± 2 ◦C for 24 h, and then they were kept for 1 h at laboratory conditions before continuing with the tests.

#### 2.2.2. Scanning Electron Microscopy (SEM) Examination

The microstructural changes on the mortars, due to the action of the crystalline admixture and the effect of the sulphuric acid attack, were examined with scanning electron microscopy (SEM, Hitachi High-Technologies Canada, Inc., Toronto, ON, Canada) (Hitachi S3000N). Before the examination, small fragments from the mortar surfaces were removed, then softly dried at 60 ◦C for 24 h and finally metallised with Au–Pd (30 nm) in order to improve the image quality. The images were taken with the following conditions: secondary electrons mode, ultrahigh vacuum, 15 kV of accelerating voltage and variable working distance.

#### 2.2.3. Physical and Mechanical Properties of the Mortars after the Sulphuric Acid Attack

The impact of a sulphuric acid attack on cementitious material can be studied by evaluating its mass loss over the acid exposure time [43]. In this work, the mass loss was studied in seven stages of the acid simulation. To do this, the samples were weighed at the following intervals: (i) after completing their curing (i.e. at 28 days after their manufacture and 0 days of acid attack (t28(0))); (ii) at 7 days of acid exposure (t35(7)); (iii) at 14 days (t42(14)); (iv) at 21 days (t49(21)); (v) at 28 days (t56(28)); (vi) at 56 days (t84(56)); and (vii) at 90 days (t118(90)) of acid exposure. The percentages of mass loss of the mortars were calculated taking into account the initial weights.

Compressive strength is a characteristic of the cement-based materials commonly used in the literature for analysing their performance against a chemical attack [12,43,44]. The reference mortars (that were kept in a nonaggressive environment) and the mortars exposed to the sulphuric acid attack were tested at t56(28) and t118(90). The compressive strength tests were performed according to the UNE-EN 196-1 standard [40]. The conventional mortar testing machine used had a load cell of 20 T capacity and was operated at a speed of 2.4 kN/s until failure.

Ultrasonic pulse velocity is a parameter that can be correlated with the elasticity modulus and therefore provides information about the stiffness of the material [45]. The ultrasonic pulse velocity test was performed according to the UNE-EN 12504-4 standard [46]. A total of four determinations were made per sample and the mean value was adopted. The test consisted of measuring the propagation time of the ultrasonic waves when crossing the longest dimension of the specimen (160 mm). Contact transducers emitting ultrasonic pulses at 54 kHz were coupled to the end sides of the specimens using a coupling agent. The wave speed was obtained from the propagation time and the length of the sample.

The impact of the sulphuric acid attack on the mortars was also evaluated studying the capillary water absorption of the specimens. The tests were conducted following the UNE-EN 1015-18 standard [47] for all the mortars (nonattacked and attacked) at t56(28) and t118(90). According to the standard, the water absorption coefficient is the slope of the line that joins the points corresponding to 10 and 90 min in the curve, representing the mass variation of water absorbed per unit area as a function of the square root of time; that is, the coefficient was computed using the formula

$$C = \frac{M\_2 - M\_1}{A\left(t\_2^{0.5} - t\_1^{0.5}\right)}\tag{1}$$

where:

*<sup>C</sup>* is the capillary water absorption coefficient, kg/(m2·min0.5); *M*<sup>1</sup> is the specimen mass after the immersion for 10 min, kg; *M*<sup>2</sup> is the specimen mass after the immersion for 90 min, kg; *A* is the surface of the specimen face immersed in the water, m2; *t*<sup>2</sup> = 90 min; *t*<sup>1</sup> = 10 min.

#### **3. Results and Discussion**

#### *3.1. Compressive Strength*

The results of the compressive strength tests showed that for a nonaggressive environment (Figure 3a) at t56 (56 days from the manufacture of the mortars), the mortars with the highest compressive strength were the C (54.3 MPa) and D (53.7 MPa) types compared with the compressive strength of the A (42.2 MPa) and B (40.8 MPa) mortars. Besides the obvious advantages in the use of the admixtures for mortars, there were drawbacks to take into account. One of those was the lowered final compressive strength compared with nonmodified concretes and mortars using some types of admixtures [48]. In the interval from t56 to t118, the only mortar that increased its compressive strength was the reference mortar (A mortar, without crystalline admixture), from 42.2 to 42.8 MPa, whereas in that same period, the three mortars with the crystalline admixture decreased their compressive strength. Usually, admixtures with an accelerated setting effect can reduce the strength of the concrete at later ages. In the high-strength mortars, a mechanism of deterioration of the hardened cement paste phase at the microscopic scale seems to lead to reduced strength at long ages [49]. The C and D mortars clearly had more compressive strength than the reference mortar at t56 (B mortar had 28.7% more compressive strength than the A and C mortars (27.7%)), although at t118, this difference was smaller (C mortar: 2%, D mortar: 2.3%) (Figure 3c). It was observed that the C and D mortars exhibited similar compressive strength to the reference mortar at t118, which was in accordance with previous research [33], where the crystalline admixture does not significantly affect the compressive strength of the concretes studied. However, the compressive strength of the B mortar was lower than that of the reference mortar, both at t56 and t118.

**Figure 3.** (**a**) Compressive strength in a nonaggressive environment at t56 and t118. (**b**) Compressive strength in an aggressive environment (sulphuric acid exposure) at t56(28) and t118(90). (**c**) Compressive strength differences of the mortars with the crystalline admixture with respect to the reference mortar in a nonaggressive environment. (**d**) Compressive strength differences of the mortars with the crystalline admixture with respect to the reference mortar exposed to sulphuric acid.

The compressive strength of the mortars exposed to 28 days of acid attack (56 days after manufacture) followed a pattern similar to that of mortars without acid attack (Figure 3b). The mortars with the highest compressive strength were types C (48.9 MPa) and D (46.3 MPa), and the lowest were types A (36.0 MPa) and B (36.9 MPa). The compressive strength at t56(28) decreased in the same way for all mortar types (due to the acid attack), given as a reference the compressive strength without acid attack, for each mortar with the same curing time (t56). The reduction in compressive strength was a direct effect of the acid attack due to the microcracking caused by the formation of expansive compounds [50,51]. As expected, the compressive strength of mortar type B slightly exceeded the compressive strength of mortar A. When increasing the exposure time of the attack to 90 days (t118(90)), the compressive strength of all mortars decreased due to the effects of the attack. In this case, mortars with the admixture (B, C and D mortars) clearly had higher compressive strength compared with the reference mortar (B mortar had 28% more strength than mortars A, C (15.8%) and D (28.8%)) (Figure 3d).

On the other hand, a behavioural change in the compressive strength of the mortars was observed for 90 days of acid exposure. B mortar had the highest compressive strength, whereas for a 28-day exposure, the one with the highest compressive strength was C mortar.

#### *3.2. Mass Loss Due to the Sulphuric Acid Attack*

The results of the mass loss test due to the sulphuric acid exposure (Figure 4a) showed that an acid attack caused a mass loss in mortars so that, as the acid attack continued longer, the mass loss of the mortars was higher. The compressive strength loss described above and the increment in mass loss when the exposure time increased were consistent with the results obtained in previous studies [43]. Moreover, a linear correlation (*R*<sup>2</sup> = 0.7724) between the decrease in compressive strength and the mass loss was found (Figure 4b). Progression of the acid attack front caused an increase in porosity and permeability, leading to mass and strength loss [52]. Nevertheless, the mortars with the crystalline admixture (B, C and D mortars) behaved clearly better than the mortars without this admixture (A mortar) since they presented lower mass loss (B mortar: 10.5%; C mortar: 9.5 %; D mortar: 10.1%) than the mortar without the admixture (A mortar: 15%).

**Figure 4.** (**a**) Mass loss due to the sulphuric acid attack. (**b**) Correlation between compressive strength (MPa) and mass loss (%) after the acid attack.

As a result of the sulphuric acid attack, calcium sulphate (gypsum) was formed by the reaction of the acid with the calcium hydroxide (chemical reaction 2) and calcium silicate hydrate (chemical reaction 3) that were present in the hydrated Portland cement [3,5] and limestone sand (chemical reaction 4) [45,53]. The gypsum coating could also be observed with the naked eye (Figure 5b). The formation of gypsum after a sulphuric acid attack has been confirmed in the literature with XRD studies [52,54]. The chemical reactions produced by the acid attack resulted in a profound degradation of the hydrated cement paste, associated with a loss of compressive strength. When the concrete surface in addition to the acid attack was exposed to flowing water, the products of the degradation were carried away to a significant degree, causing a mass loss. Generally, an attack by free sulphuric acid is more severe than any with a neutral sulphate solution [3]. As mentioned in Section 2.2.1, the specimens were brushed and cleaned weekly to remove the gypsum formed on the specimen surface. The mass loss shown in Figure 4a was mainly associated with the amount of gypsum removed from the mortar surfaces during the brushing process. However, as the literature states [54], at the beginning of an acid attack, there is an increase in mass, which can be explained by the generation of gypsum in the pores and the cement–aggregate interface. This gypsum was difficult to remove even though the surface of the samples was brushed; therefore, at the first stage of the attack, a mass increase was found. To corroborate the existence of gypsum on the surface of the attacked mortars, an XRD analysis was performed. The samples were taken from fragments obtained from the surface of the specimens (4 × 4 × 16 cm) that had been broken in the compressive tests. The XRD analysis showed the presence of bassanite (CaSO4·1/2H2O), which was the result of the thermal decomposition of gypsum, and it happened at 110 ◦C (chemical reaction 5) [55]. Therefore, it was consistent to find bassanite (hemihydrate phase of gypsum) after having exposed the specimens at 105 ± 2 ◦C continuously for 24 h. This result confirmed the previous existence of gypsum.

$$\text{Ca(OH)}\_{2} + \text{H}\_{2}\text{SO}\_{4} \rightarrow \text{CaSO}\_{4} \cdot 2\text{H}\_{2}\text{O} \tag{2}$$

$$\text{xCaO} \cdot \text{SiO}\_2 \cdot \text{aq} + \text{xH}\_2\text{SO}\_4 + \text{xH}\_2\text{O} \rightarrow \text{xCaSO}\_4 \cdot 2\text{H}\_2\text{O} + \text{SiO}\_2 \cdot \text{aq} \tag{3}$$

$$\text{CaCO}\_3 + \text{H}\_2\text{SO}\_4 + 2\text{H}\_2\text{O} \rightarrow \text{CaSO}\_4\cdot2\text{H}\_2\text{O} + \text{CO}\_2 + \text{H}\_2\text{O} \tag{4}$$

$$\text{CaSO}\_4\cdot2\text{H}\_2\text{O} \rightarrow \text{CaSO}\_4\cdot1/2\text{H}\_2\text{O} + 3/2\text{H}\_2\text{O} \tag{5}$$

**Figure 5.** (**a**) X-ray diffraction spectra of control mortar and D mortar (with crystalline admixture) after 90 days of sulphuric acid attack; (**b**) Specimens exposed to the acid attack after compressive strength testing at t118(90), where the massive formation of gypsum in the surface of the specimen can be seen.

(b)
