Effect of various curing methods and addition of silica aerogel on mortar properties

Mechanical and thermal properties and porosity of aerogel-incorporated mortars exposed to various curing conditions (curing by wetting-drying, curing by magnesium sulphate (MgSO4), and water curing) are experimentally investigated in this study. Maximum compressive strengths at 0.5 % aerogel content under the effects of wetting-drying and MgSO4 curing conditions amounted to 60.8 MPa and 44.3 MPa, respectively. In addition, compared to the other curing methods, the gel pores formation in mortars exposed to MgSO4 effects increased with an increase in aerogel content.


Introduction
With the high rate of scientific and technological changes in the construction materials sector and in all other sectors, the production of technological materials with superior features, and the development of existing materials, has become quite inevitable. In this respect, the idea of enhanced thermal insulation through the developed pore structure in the cement based materials has an important place in innovation strategies as a means to limit energy consumption considering the decrease in energy resources. Numerous experimental studies on additive materials such as mineral wool, expanded polystyrene, extruded polystyrene, polyurethane, aerogels, nano insulation materials etc., are currently conducted for this purpose [1]. The improvement of thermal insulation through contribution of silica aerogels to C-S-H structure of materials has become an interesting and attractive subject in recent years. Silica aerogels, first produced by Kistler in 1931, and considered as one of the world's lightest solid materials, have become scientifically attractive materials that have a wide range of applications due to their high surface area, total porosity values that can be found in 99 % and low density [2,3]. Aerogel is generally used as a thermal insulator additive in mortar and concrete applications. Experimental studies carried out with silica aerogel admixtures have generally focused on large volume scale (50-90 %) aerogel replacement with sand in mortar or concrete mixtures. The scope of these studies, conducted with regard to the decrease in thermal conductivity coefficient, is often restricted to the production of non-loadbearing elements having low mechanical strengths and the use of insulating plastics. A meaningful correlation between the thermal insulation target and the mechanical strength exists, and this relation limits the structural use of the produced cementitious materials [1,4,5]. The first experimental study in which silica aerogels with granular form were placed in cement matrix was carried out by Ratke in 2008 [6]. Ratke preferred, in his study, the cement types of CEM II 32.5 R, CEM I 42.5R and CEM I 52.5R in mixtures containing 50-70 % of aerogel by volume. When silica aerogel was used at 70 % by the volume, compressive strengths ranged from 0.6 to 1.5 MPa corresponding to a thermal conductivity coefficient of 0.010 W/ mK [7]. Hub et al. obtained the compressive strengths in the range of 1.4 -2.5 MPa from the mixtures including 65-75 % aerogel by the volume corresponding to a thermal conductivity coefficient interval of 0.10-0.14 W/mK [8]. Gao et al. investigated the change of mechanical strength and thermal conductivity coefficient by producing aerogelincorporated concrete mixtures at the change interval of 0-60 % in volume. They determined that the thermal conductivity coefficients decline as the density and mechanical strengths decrease, while the aerogel content increases. They found a 8. 3 MPa compressive strength value corresponding to the thermal conductivity of 0.26 W/mK at 60 % aerogel content rate by volume [4]. Fickler et al. prepared various mixtures in which aerogel ratio varied from 60 % and above by volume for high-performance aerogel concrete. They determined a compressive strength of 10.0 MPa corresponding to the thermal conductivity coefficient value of 0.17 W/mK in the mixture that gives optimum results [6].
Serina et al. determined a 20 MPa compressive strength corresponding to the thermal conductivity of 0.55 W/mK by testing mortar mixtures including 50 % aerogel rate by volume. However, the cases in which aerogel ratio exceeds the range of 50-60 % by volume are not recommended in terms of mechanical strength [9]. In their experimental work, Julio et al. prepared lightweight aggregated cementitious aerogel-based renders composed of 60 % aerogel and 40 % granular expanded cork, expanded clay, and perlite mixtures. A 0.92 MPa value of compressive strength was found corresponding to the thermal conductivity of 0.084 W/mK and 60 % total porosity values of the samples [10]. Kim et al. prepared gel-typed aerogels through methanol to reduce the difficulties encountered in mixing the aerogels in granular form and the capillary cracks during hydration. In the mortar mixtures, they used aerogels in gel-form at the rates of 0.5 %, 1.0 %, 1.5 % and 2.0 % by weight and obtained compressive strengths of 13.1 MPa, 8.0 MPa, and 5.9 MPa, respectively, while the compressive strength of the reference sample amounted to 26.3 MPa [11]. It can be deduced from these studies that the pore structures of mortar must also change in addition to an increase in thermal insulation properties resulting from aerogel effect. Generally, a porous structure is needed in the cement matrix for high thermal insulation properties. However, an exact opposite pore structure is required for high mechanical strength [1,4,5]. Similarly, when the effect of aerogel contribution on the concrete pore structure was investigated, it was found that the aerogel used at 20 % by volume can increase the total porosity value by 8.63 % through an additional pore volume created, especially at the porosity range of 10 -30 nm [12]. Studies that examine the change of pore structure in aerogelincorporated mortars under durability conditions are currently quite limited. It was established in [13] that the freezingthaw cycles exert a significant negative effect on the thermal conductivity coefficient of aerogel-incorporated plaster mortars. In the study conducted by Ng et al., 2016, it was established that curing conditions exert a significant effect on thermal conductivity coefficient and compressive strength of aerogelincorporated samples. In samples with aerogel admixtures subjected to the first 24-hour moulding process at around 80 °C, followed by 28 days of curing at 80 °C in water, covered with aluminium foil, the compressive strength increased by 10 % compared to the reference sample, and a 35 % decrease in thermal conductivity coefficient was observed [14]. GRAĐEVINAR 71 (2019) 8, 651-661 Effect of various curing methods and addition of silica aerogel on mortar properties Magnesium sulphate, which is found in groundwater, sea water, or industrial liquid wastes, affects the hydration process and formation of hydration products in calciumsilicate-hydrate (C-S-H) based materials by Mg and SO 4 ions existing in its structure. Considering its negative effect on the hydration process, MgSO 4 is one of the dangerous salts that are effective in reducing service life of structural elements. In previous experimental studies, significant losses were observed in the modulus of elasticity, stiffness, compressive and flexural strengths of the structural elements exposed to MgSO 4 effect. As a result of hydration process, the formation of the C-S-H bond structure, which has quite a vital role in the gain of mechanical strength, is reduced by Mg salts and the formation of C-S-H is partially replaced by magnesium-silicate-hydrate (M-S-H) structure which is very fragile and does not have the capability of binding [15]. In order to reduce the negative effect of MgSO 4 on C-S-H structure, low porosity material designs are frequently examined through alternatives such as designs with high cement dosage, low water/binder ratio, and various saltresistant mineral added binders. Porosity-controlled designs, which are provided with additional additives in mortar mixtures in order to improve their durability properties under the effect of MgSO 4 , generally cause losses in mechanical strength of mortar despite the change in total porosity of mortar. The originality of this investigation lies in revealing significant mechanical, thermal conductivity and porosimetric properties of silica aerogel -incorporated mortars under durability curing conditions such as wetting-drying and MgSO 4 .

Materials
CEM I 42.5 R type Portland type cement produced by Limak Cement Co. and standard CEN sand were used in mortar mixtures produced for the experimental works. Chemical composition and physical properties of the cement are presented in Table  1. In experimental studies, silica aerogel, which is preferred as additive material for improving thermal insulation properties of mortars, was provided by Alison Aerojel Co. The aerogel was produced according to technical specifications given in Table 2. Physical properties of aerogel used in the study are presented in Table 2.

Mixing and preparing specimens
In order to investigate mechanical, thermal and porosimetric properties of silica aerogel-incorporated mortars, aerogel added mortars mixtures for five content ratios in total, and three  Table 3.
In the process of mixture preparation, a dry mixture containing cement, standard sand, and aerogel particles was initially formed. Water was subsequently gradually added to the dry mixture in order to provide for a uniform distribution of aerogel particles. A 0.50 water/cement ratio was adjusted in the mixtures. The total amount of cement and aerogel in the mixtures was considered as the total binder material. After the end of the mixing process, cement mortar was placed in steel casts in which standard flexural specimens of 40 mm x 40 mm x 160 mm are produced and an effective compacting operation was carried out in order to prevent separation of mix components. Mixture samples were kept at 21 ± 2 o C for 24 hours and then removed from the casts. Samples were subjected to three different curing conditions, i.e. the wettingdrying effect, MgSO 4 effect, and water-curing.

Curing methods
Curing conditions involving the wetting-drying effect and MgSO 4 effect were provided in addition to conventional water-curing. The samples are prepared separately for each mixtures and three samples were produced in order to obtain the arithmetic mean of mixture samples for three different curing methods. The samples produced for water curing were immersed in the curing tank at a temperature of 21 ± 2 o C and the curing was carried out continuously for 14 weeks. Unlike the conventional water curing process, an alternative curing method was prepared to test the wetting-drying effect. Aerogel-incorporated mortar samples, which completed the wetting-drying curing process in the first week, were removed from the curing tank and kept at room temperature of 21 ± 2°C for the following week. The curing process for 14 weeks in total was performed in seven consecutive cycles. As a second alternative to the conventional water curing process, the curing method involving MgSO 4 was conducted. The specimens were kept in a curing tank, which contained 13 % MgSO 4 by weight, and kept for one week at a temperature of 21 °C ± 2 °C. Then in the following week they were oven-dried at 105°C. The curing process performed as seven consecutive cycles were completed in 14 weeks.

Testing
The samples, which completed the 14-week curing process for water curing, wetting-drying, and MgSO 4 effects, were subjected to flexural and compressive strength tests, thermal conductivity coefficient measurement, and mercury porosimetry (MIP) tests, respectively. The experimental test programme is shown in

Compressive strength results
Compressive strengths of aerogel-incorporated mortars, associated with the curing process in water, wetting-drying, and MgSO 4 testing, are presented in Figure 1. Regarding the compressive strengths, the strengths in the sample groups, which are cured in water at the aerogel content rates increasing from 0.

Figure 1. Relation between aerogel content and compressive strength test results of aerogel -incorporated mortars
The highest compressive strength results are generally obtained from the curing with the wetting-drying effect. In this curing method, similar compressive strengths, which are around 60 MPa, were obtained at 0.1-0.7 % aerogel content ratios, and the compressive strength decreased to 36.3 MPa, which means 40 % strength loss at 1.0 % content rate.
In the sample group that completed the curing process in water, the compressive strength was determined as 56.5 MPa at the lowest content rate (0.1 %). As the content rate was increased up to 0,5 %, 0,7 %, and 1,0 %, the compressive strengths were not adversely affected by the increased aerogel additive and the strengths for the corresponding content rates amounted to 54.5, 54.9 and 57 MPa, respectively. The minimum compressive strength of 37.4 MPa was obtained at the 0.3 % aerogel content ratio, which means that the strength loss was 33.8 % compared to the reference sample. The strength values are generally close to those that were obtained from the wetting-drying effect.
However, they were found to be lower. Although a significant decrease in strength (max. 2.75 %) does not appear for the samples which completed the curing process under MgSO 4 effect, the strengths are lower than the ones obtained by the remaining two curing methods.

Flexural strength results
Flexural strengths of aerogel-incorporated mortar samples, which completed the curing process in water and under the wetting-drying and MgSO 4 effects, are shown in Figure 2.
Examining the flexural strength results, in the sample group which is cured in water at the aerogel content rates increasing from 0.1 % to 1.0 %, the strengths varied in the range of 6.

Thermal conductivity test results
The thermal conductivity coefficients of aerogel-incorporated mortar samples that completed the curing process in water, under the wetting-drying, and MgSO 4 effects, are presented in Figure 3. The lowest thermal conductivity coefficient for curing in water amounted to 1.56 W/mK at the 0.3 % aerogel content. The minimum thermal conductivity coefficient resulting from the wetting-drying effect amounted to 1.80 W/mK at the 0.3 % aerogel content, and the lowest thermal conductivity coefficient for samples cured in MgSO 4 was found to be 1.61 W/mK at the 0.1 % aerogel content.

Figure 3. Thermal conductivity test results for aerogel-incorporated mortars
In the sample group that completed the curing process in water, the thermal conductivity coefficient amounted to 1.75 W/mK at the 0.1 % aerogel content. As a result of using 0.

MIP test results
The results of porosimetry analysis of aerogel-incorporated mortar samples which completed the curing process in water, under the wetting-drying and MgSO 4 effects are presented in Table 5. According to porosimetry results for samples cured in water, the highest total porosity value (15. Effect of various curing methods and addition of silica aerogel on mortar properties nm were obtained for A1 (6.6 MPa) and A5 (7.1 MPa) samples, respectively, which had the lowest flexural strengths in the water curing group. In the sample group cured by the wetting-drying procedure, compressive strengths in the range of 59.4-60. 8   Among the samples that completed the curing process in water, the samples A2 (1.56 W/mK) and A4 (1.70 W/mK), which have the lowest thermal conductivity coefficient, exhibited the highest cumulative pore volume compared to other samples in the diameter range of 3 and 340000 nm throughout the chart.
The A1 sample (1.75 W/mK) exhibited a cumulative pore volume higher than that of the A4 sample only for a diameter range of 3 -3.50 nm. However, this behaviour ended at around 3.50 nm. Because of this behaviour throughout the graph, the lowest thermal conductivity coefficients were determined in samples A2 and A4. In the capillary diameter range from 225 to 9000 nm, the A2 sample exhibited a higher cumulative pore volume than the A4 sample but, at higher diameters, the A4 sample exhibited a higher cumulative pore volume compared to the A2 sample. Therefore, the total porosity value of A2 was higher (15.29 %) than that of A4 and other samples in the curing group within the capillary diameter range of 225 -9000 nm. The median pore diameter-volume (121.4 nm) of the A4 sample was higher than that of A2 and other samples in the cure group due to behaviour in the range from 9000 nm to 340000 nm. In particular, when the diameter distribution at 10000 nm and above is examined, macro pores formation of the highest volumetric ratio (15.6 % and 16.8 %) of the sample group cured in water was observed in A2 and A4 samples due to cumulative pore volume in the upper region of the graph. The formation of macro pores at high levels exerts a considerable influence on the thermal conductivity coefficient. The highest thermal conductivity coefficient (2.13 W/mK) in the water-cured group was found in A3 sample. Therefore, the lowest total porosity value of the curing group ( It was determined that a similar effect cannot be achieved with the A1 -A2 content rate. At this transition rate, the thermal insulation provided with aerogel content and loss of compressive strength are related to a greater macro pores formation (15.6 %) in the pore structure at 0.3 aerogel content, compared to other samples. Although it is known that aerogel particles are stable during hydration and that they maintain their high mechanical properties under high total porosity -high macro-pore structure with low aerogel content, it cannot contribute to the compressive strength of mortar due to poor adhesion. The cumulative pore volume -pore size distribution and the relation between content of pores -thermal conductivitycompressive strength of the samples cured by wetting-drying, are presented in Figure 5a and Figure 5b, respectively.

Figure 5. a) Pore size distribution for all samples subjected to wetting -drying cycles obtained from MIP data; b) Relation between content of pores, thermal conductivity and compressive strength for wetting-drying curing group
Among the samples which completed the curing process under the wetting-drying effect, A2 sample with the lowest coefficient of thermal conductivity (1.80 W/mK) showed the highest cumulative pore volume in the range of 3 -340000 nm diameters along the graph, compared to other samples. The A2 sample also does not show deviation in any region throughout the graph and, therefore, the highest total porosity (14.96 %) and the highest median pore diameter-volume (117.4 nm) values in the group are exhibited by this sample. As a result of these porosimetric findings, the lowest thermal conductivity coefficient was obtained at sample A2. If sample behaviour in the diameter range of 10000-340000 nm is examined, it can easily be observed that samples A4 and A5 always exhibit the lowest cumulative pore volume. In particular, median pore diameter-volume (47.2 nm) and the total porosity value (12.17 %) of the A4 sample, which exhibits the lowest cumulative pore volume at all diameters throughout the graph, are the lowest porosimetric values in their curing group. As a result of these porosimetry analysis findings, both samples present the highest thermal conductivity values in their curing groups (1.95 W/mK and 2.01 W/mK). The cumulative pore volume -pore size distribution, and the relation between content of pores -thermal conductivity -compressive strength of the samples that completed the curing process under the MgSO 4 effect, are presented in Figures 6a and 6b. The A1 (1.61 W/mK) sample, which has the lowest thermal conductivity coefficient among the MgSO 4 -cured samples, exhibited the highest cumulative pore volume in the diameter range of 3 -120 nm compared to other samples. The A1 sample showed the second highest cumulative pore volume behaviour after the A2 sample in the diameter range of 120-340000 nm.
The A1 sample has a higher rate of capillary pores distribution (84.6 %) and lower gel pores content (1.74 %) in the mortar pore structure compared to sample A2. GRAĐEVINAR 71 (2019) 8, 651-661 Effect of various curing methods and addition of silica aerogel on mortar properties  , 164 nm, respectively). The increase in gel -pore formation having a high correlation with an increased aerogel content in the MgSO 4 curing group indicates that representation of pore structure would be more accurate using median pore diameter -volume than employing the total porosity. The determination of quite different total porosity values in spite of uniform median pore diameter -volume and compressive strength values under the effect of MgSO 4 shows that the effect of MgSO 4 on the pore structure of the mortar involves formation of pore geometry and connectivity that cannot be explained with pore diameters only. Nevertheless, for the MgSO 4 curing group, highly correlated and controlled effect created by silica aerogel, which is used in experimental studies in diameter range of 8 -10 nm, for a diameter of not more than 10 nm in the gel formation, could provide uniform distribution of compressive strengths although total porosity value was varying. Volumetric pore levels in pore structures of aerogel-incorporated mortars under the effect of different curing conditions are shown in Figure 7. Gel pores formation is at the level of 3.6 -5.8 % at other content rates excluding the rate of 0.7 % for samples which completed the curing process under the effect of wetting-drying. The gel pores formation at the rate of 0.7 % aerogel content reaches up to 26.7 %, which is the maximum value in the curing sample group. Due to the high gel pores formation at this content ratio, the median pore diameter-volume of A4 sample at the aerogel content rate of 0.7 % decreases to 47.2 nm, which is the minimum of the curing group, unlike the other samples. Therefore, the lowest total porosity value (12.17 %) and high compressive strength (60.6 MPa) of the curing group could be achieved at the A4 sample. Due to an increase in aerogel rate in the MgSO 4 curing group, the gel pores formation also increased steadily and finitely. The gel pores formation regularly increased from 1.7 % to 4.4 % with an increase in aerogel content from 0.1 to 1.0 %. Figure 8 shows volumetric capillary porosities of pore structure in aerogel-incorporated mortars under various curing conditions.

Figure 8. Relation between aerogel content and distribution of capillary pores in aerogel-incorporated mortars
Capillary pores formation varies from 72.2 to 85.0 % with an increase in aerogel content in water-cured samples. The highest capillary pores formation was observed at the 0.5 % aerogel content in sample A3.  Figure 9.

Conclusions
The mechanical and thermal conductivity coefficients and porosimetric properties of cement mortars under different curing conditions for the 0.1 -1.0 % aerogel content range are investigated experimentally in this study.
In-water curing group -Compressive strength results are lower than the ones obtained from the wetting-drying experiments even though they are quite close to each other. -The lowest thermal conductivity coefficient measured in this curing group amounted to 1.80 W/mK for 0.3 % aerogel content.
-The minimum capillary pores formation in the mortar pore structure was determined at 0.7 % aerogel added sample. The lowest total porosity value (12.17 %) and the lowest median pore diameter -volume (47.2 nm) of the curing group were determined in this sample. A high compressive strength (60.6 MPa) was also obtained from the sample by means of the aforementioned porosimetric results. -The lowest thermal conductivity coefficient of the curing group amounted to 1.61 W/mK at the 0.1 % aerogel content. This value is the second lowest coefficient of thermal conductivity amongst all mixture samples exposed to different curing conditions. It can be observed that higher flexural and compressive strengths can be obtained for all aerogel contents in the wetting-drying curing method compared to the water curing method. For this reason, in the production of prefabricated structural elements with high mechanical strength expectations, the wetting-drying method seems to be an attractive curing process compared to the traditional curing in water method. The compressive and flexural strength gains that were obtained even at room temperature point to larger mechanical strength gains for the prefabricated structural elements to be produced under higher curing temperature. Similarly, flexural strengths in the MgSO 4 curing method are generally higher than in the conventional water-curing method. Especially under the MgSO 4 effect, it should be noted that the mechanical strength of mortar remains stable despite the change of total porosity value of mortar samples in the range of 2.2-14.49 % with aerogel content. Therefore, aerogel incorporation is an attractive idea in the design of structural elements exposed to the effect of MgSO 4 compared to other traditional additives used in improving durability characteristics of the elements under the influence of sulphate that lead to a loss of mechanical strength.