1. Introduction
Many different requirements are placed on the properties of both interior and exterior plasters. Nevertheless, their thermal insulation function has gained importance in recently. As the consumption of energy used for the temperature control in both residential and commercial buildings through heating and air conditioning is still increasing [
1], the energy performance of buildings represents the driving force for the improvement and the design of the advanced thermal insulation materials. Worldwide, the buildings and construction sector were responsible for 36% of the total energy use and 39% of energy and process-related CO
2 emissions in 2018 [
2]. In the European Union (EU), the heating and cooling of buildings represents around half of the EU’s final energy consumption and is the biggest energy end-use sector, ahead of transport and electricity. Moreover, 85% of the energy used for heating and cooling is produced from natural gas, coal, and oil products and only 15% is generated from renewable energy sources [
3]. The key targets for 2030 adopted by the European Commission under the 2030 Climate and Energy Framework included at least 40% cuts in greenhouse gas emissions (from 1990 levels) and at least 32.5% improvement in energy efficiency [
4].
The most common way how to reduce the energy demands for heating of older buildings is application of the External Thermal Insulation Composite Systems (ETICS). This solution enables to effectively insulate the whole envelope of the building and reduce thermal bridges that would negatively affect the overall building’s hygrothermal performance [
5,
6,
7]. However, in the case of older and historical buildings, application of ETICS is often forbidden especially due to the requirements of culture heritage authorities that insist on the preservation of the original and decorative appearance of the architectural style of facades and their elements, such as balusters, on brackets balconies, ornamental brackets, pillars and pilasters, gables, etc. Therefore, the compromise solution in the improvement of the hygrothermal performance of such buildings is to enhance their envelopes with appropriate thermal insulation plasters possessing other advanced functional parameters [
8]. Many investigations were carried out focused on the integration of different kinds of insulating materials into the composition of plasters in order to achieve their low thermal conductivity and thus thermal insulation efficiency [
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21]. Generally, the authors reported that the addition of lightweight aggregates or fibers into plaster composition greatly improves thermal conductivity, increases porosity, and decreases the mechanical parameters of the hardened plasters. Improvements in sorption properties and water vapor permeability were also reported.
The main aim of this paper is the comprehensive analysis of three types of novel lightweight thermal insulation plasters with enhanced composition containing expanded perlite and designed for interiors, including areas with increased humidity, such as kitchens and bathrooms, or exteriors. The plasters should find use also in repair of historical masonry, where they should act as durable materials for the moderation of the indoor climate of cultural heritage buildings. During restoration works, the compatibility between the new repair mortars and the original components is essential for an adequate intervention on the monument. As pure Portland cement mortars are incompatible with most of the traditional materials inbuilt in historical structures, lime–, natural hydraulic lime–, lime–pozzolan–, gypsum–, and cement–lime-based mortars are considered as materials applicable for the repair of historical masonry and structures. In this sense, the two cement–lime plasters and one lime–gypsum plaster with expanded perlite were characterized in terms of their structural, mechanical, thermal, and hygric parameters. As the durability of plasters is a very important parameter especially in their use in restoration of salt-laden masonry, the tests of salt transport properties were conducted together with the analysis of salt crystallization resistance. To the best of the authors’ knowledge, such comprehensive analysis of plasters intended to be used for complex solution of thermal insulation and repair problems was not presented yet and can be considered as a further step for the improvement of eco-efficiency of both contemporary and older building stock. Moreover, the increasing interest in the knowledge of the properties of mortars for restoration purposes justifies the research carried out.
3. Results and Discussion
The chemical composition of the employed materials obtained by XRF is introduced in
Table 3.
The particle size distribution curves of the prepared dry plasters mixtures are plotted in
Figure 1. The recorded particle size distribution corresponds to the particle size of raw materials contained in plasters’ composition and their dosage.
The results of the conducted tests and analyses were evaluated using the specifications for masonry and repair mortars summarized in EN 998-1 [
42] and the WTA directive 2-9-04/D [
43].
The macrostructural parameters of the investigated plasters are summarized in
Table 3. The agreement between the bulk density and total open porosity values is quite obvious. The WTA directive 2-9-04/D [
43] prescribes the bulk density < 1400 (kg·m
−3) for the repair plaster. Similarly, the porosity must be >40%. Both these conditions were safely encountered for all tested materials. In
Table 4, the total open porosity measured by mercury intrusion
ψHg (-) is also presented. Taking into consideration the principles of the applied total porosity assessment methods, low weight of samples for MIP tests, and inhomogeneity of the tested plasters, the difference in the porosity values can be considered as insignificant.
The pore size distribution curves obtained by mercury intrusion porosimetry are given in
Figure 2 and
Figure S1 (Supplementary Materials). Apparently, the microstructural data corresponds with the total porosity data presented in
Table 4. The pore size distribution parameters acquired by MIP are presented in
Table S1 (Supplementary Materials). They clearly characterize highly open porous structure of the examined materials, which is positive with respect to the requirements imposed on the repair plasters. In the whole studied pore diameter range, the volume of pores was the lowest for plaster CLM1. Contrary to that the relative volume of pores in the recorded pore radii was the highest for LGM. The relative volumes of pores of plaster CLM2 were in the middle.
The mechanical resistance of the investigated plasters is apparent from
Table 5. These are the results of three combined effects, the nature of the used binder, use of lightweight admixture (perlite), and porosity. The highest flexural strength and compressive strength exhibited lime–gypsum plaster LGM. On the other hand, the dynamic modulus of elasticity of this material was in the middle. Similar performance of lime–gypsum plasters was reported, e.g., in [
44,
45]. From the practical point of view, all plasters can be classified in category CS II [
42] which well satisfies the condition of the WTA directive 2-9-04/D [
43]. According to this directive, the compressive strength of repair plasters must be in the 1.5–5.0 MPa. This criterion safely met all prepared plastering mortars.
Heat transport and storage in the investigated plasters were characterized by the thermal conductivity
λ (W·m
−1·K
−1), thermal diffusivity
a (m
2·s
−1), and volumetric heat capacity
cv (J·m
−3·K
−1). These parameters are summarized in
Table 6. As there is not any strict requirement on the thermal characteristics of plaster intended for repair applications, the thermal insulation potential was evaluated in respect to EN 998-1 [
42]. According to this standard, all tested plasters can be ranked as lightweight plasters for interior and in the case of CLM materials also for exterior use. Moreover, lime–gypsum plaster satisfied the criteria imposed on thermal insulation plaster of T2 type. The thermal conductivities of both cement–lime plasters were slightly higher than required limit for T2 (
λ < 0.2), but they were still acceptable for the improvement of thermal performance of repaired masonry.
The water vapor transmission properties are important parameters of plasters for repair applications. The results of the cup test that was conducted in both dry cup and wet cup arrangements are presented in
Table 7. All materials had the water vapor resistance factor <12.0, which is strictly limited by the WTA directive 2-9-04/D [
43]. Such plasters enable water vapor release from the interior and possible drying of the structures suffering from the excessive moisture presence. It must be noted, the water vapor resistance factor criterion of the WTA directive is higher than that prescribed for repair mortars in the EN 998-1 [
42] which requires
µ < 15.0. The water vapor permeability was higher for wet cup arrangement of the test than that of assessed in the dry cup analysis. Similar material performance in the different conditions of the cup experiment was observed, e.g., in [
46,
47,
48]. The acceleration of water vapor transmission in the wet cup test can be attributed to the reduced surface binding forces between water vapor molecules and pores due to the filling by water molecules within samples conditioning for the test [
49].
Basically, the water vapor permeability of materials is considered to be dependent on its macrostructural and microstructural parameters and binder nature [
50,
51]. In our case, not only the total pore volume, but also pore size distribution, their shape, and tortuosity played a role in the water vapor transmission process [
52]. Quantitatively, both the examined vapor transport parameters were similar to those published in [
53,
54,
55,
56,
57,
58].
The sorption and desorption isotherms are plotted in
Figure 3. Both the sorption and desorption curves obtained for the tested materials are quite different. Based on IUPAC isotherm classification [
59], which provides fundamental guidance how to interpret sorption isotherms for the purpose of structural characterization, the measured sorption/desorption data corresponds to the Type IVa isotherm, typical for mesoporous materials. In the case of a Type IVa isotherm, capillary condensation is accompanied by hysteresis. This occurs, when the pore width exceeds a certain critical width, which is dependent on the adsorption system and temperature.
As expected, the lowest water vapor absorption capacity exhibited plaster CLM1; up to 43% of relative humidity (RH), the gravimetric water content was in the range of detection error. For the higher relative humidity, the moisture content increases slightly versus RH to reach about 0.6%. Plasters CLM2 and LGM were more sensitive to the RH changes of the environment. Since RH > 40%, the capillary condensation arose [
60], the micropores and mesopores were filled by water molecules [
61], and the water content has strongly increased to reach about 5.1% and 3.6% at 98% RH for plasters CLM2 and LGM, respectively.
The hysteresis was well visible for all researched plasters. In case of CLM1, it reached about 0.4% and it was almost constant in the RH range 11–80%. The hysteresis of CLM2 was in the range 4.3–3.3%, and the desorption curve exhibited linearly decreasing character in the RH range 11–75%. The hysteresis of LGM was about 2.5% and was almost unchanged in the RH range 11–75%. The observed hysteresis in desorption process is usually assigned to the capillary condensation hysteresis [
62], the contact angle hysteresis [
63], the ink-bottle effect [
64], or chemical interaction of material with water molecules [
65]. A part of the residual moisture may also be due to the partial lime carbonation that could start at high relative humidity. Similar residual mass of lime-based plasters observed, e.g., Mazhoud et al. [
58].
The 24 h water absorption
Ww and water absorption coefficient
Aw assessed in accordance with EN 1015-18 [
38] are presented in
Table 8. All studied plasters exhibited high water absorption capability and safely met criteria of water absorption rate prescribed in EN 998-1 [
42] and WTA directive 2-9-04/D [
43]. According to EN 998-1 [
42], the repair mortar must have
Wa ≥ 0.3 kg·m
−3, which was well fulfilled. Quantitatively, the water ingress into the studied plasters corresponded with the pore size distribution and microstructural parameters, which were the determining factors affecting the overall water imbibition. Namely the volume of capillary pores (0.01–10 µm) in the particular plasters affected the water absorption rate and the total water ingress. Therefore, not the total open porosity, but the share of the volume of capillary pores on overall porosity was the dominant parameter for moisture transport. Similarly, the transport of the tested salt solutions was also governed by the porous structure parameters.
The differences between the observed hygric parameters assessed for penetration of tap water and 1% NaCl and Na2SO4 solutions were small, but some deceleration in the transport of salt solutions in comparison with the transport of pure water can be distinguished.
In
Table 9, the mass change of samples subjected to the salt crystallization tests is given. The ratios of the compressive strength and dynamic modulus of samples exposed to 10 wetting drying cycles (water, NaCl, Na
2SO
4) and that of the reference samples are also introduced. The examined plasters showed excellent resistance against crystallization of NaCl and Na
2SO
4 solutions. Both CLM plasters exhibited even improved compressive strength after they underwent crystallization tests and wetting/drying cycles. This was assigned to the continuous hydration of cement/lime binder whose positive contribution to the total mechanical strength prevailed against other effects. The resistance against salt crystallization was caused by the high open porosity of prepared plasters that enabled crystallization of salts from applied solutions in the free porous space without causing damage. Quantitatively bigger problem for the durability of the tested materials appeared for the action of NaCl solution that caused the biggest mass loss and drop in dynamic modulus of elasticity ratio for all the plasters. However, the damage parameters were in this case also small. In summary, considering the results of salt crystallization tests and measured plasters’ residual parameters, the prepared plasters can be recommended as plastering materials for salt laden masonry.
The results of light microscopy imaging are introduced in
Figure 4. No cracks or any surface damage was observed on the plasters’ fracture surface, which proves the high salt crystallization resistance of the analyzed materials. Photographical observation has not detected any damage of samples subjected to the salt crystallization test as apparent from
Figure S2 (Supplementary Materials).