Improvement of Water Erosion Resistance of Gypsum Mortars in the Historic Buildings for Conservation Purpose

Abstract

Gypsum mortar is widely used in structures of architectural heritage. However, due to the high solubility of gypsum in water, it is easily destroyed by water erosion in outdoor environments, leading to the instability or even failure of the ancient buildings constructed with it. To improve the water erosion resistance of gypsum mortar, the alcoholic solution of barium hydroxide was explored as the protective agent in this study. The method involves treating the gypsum mortar with the alcoholic solution of barium hydroxide and water in sequence. The mechanism of its action and protective properties were studied by infrared spectroscopy, X-ray diffraction, scanning electron microscopy, energy-dispersive spectroscopy, conductivity meter, colorimeter, etc., and conclusions were made that the alcoholic solution of barium hydroxide has high permeability and its subsequent conversion to insoluble barium sulfate and calcium carbonate helps to increase the water erosion resistance of the solution. Additionally, the positive results such as the increase in mechanical strength from 20.80 HD to 60.94 HD, the reduction in water absorption from 18.37% to 15.75%, and a total color difference (ΔE*) of less than 3.0 indicated the application prospects of the proposed method in the conservation of the historical buildings made from gypsum mortar.

1. Introduction

1.1. Application Background of Gypsum Mortar in Ancient Buildings

As a traditional inorganic cementitious material, the application of gypsum mortar in construction dates back several thousand years [1]. Owing to its rapid setting, high strength, and good plasticity, it has been widely used as joint mortar or plaster in architectural heritage structures [2]. However, after prolonged exposure to the outdoor environment, gypsum mortar undergoes varying degrees of weathering, characterized by pulverization and loss of adhesion. It is mainly due to the high solubility of gypsum (2090.00 mg/L) in water, which is 149 times larger than that of calcium carbonate (14.00 mg/L), which is the main component of lime mortar. Consequently, gypsum can be dissolved and leached away by liquid water such as rainwater and dew. The weathering of gypsum mortar poses a significant risk, as it may lead to structural instability or even the collapse of the affected buildings [3].

1.2. Research Status of Waterproof Materials for Gypsum Mortar

Numerous methods have been developed to protect gypsum and gypsum-based materials from water-induced dissolution damage. Based on their mechanism of action, these methods can be categorized into two types: surface coating and chemical conversion. In coating methods, organic materials such as organosilicone [4], acrylic acid, and epoxy resins [5] are brushed or sprayed on the surface of gypsum substrates to form a waterproof film [6]. Unfortunately, organic polymer films have been found to exhibit low durability. As organic polymers, they are susceptible to degradation by oxygen, light, heat, microorganisms, and other environmental factors in outdoor conditions [7]. In conversion methods, gypsum is transformed into water-insoluble substances by chemical agents [8]. For instance, gypsum can be converted into calcium oxalate and hydroxyapatite by oxalate and phosphate [9], respectively. However, soluble sulfate is produced simultaneously, which introduces the risk of salt damage and thus poses adverse effects. The transformation products of gypsum by barium hydroxide in the open air are calcium carbonate and barium sulfate—both of which are water-insoluble compounds. Importantly, no soluble salts are produced in this reaction. This indicates that barium hydroxide-based protectants possess favorable safety profiles and promising application prospects in academic research [10].

However, in practical applications, it has been found that aqueous barium hydroxide solutions exhibit poor penetration into the interior of gypsum substrates, and the transformation reaction remains far from complete. This issue is closely associated with the high reactivity of barium hydroxide in aqueous systems [11]. When exposed to air, barium hydroxide rapidly absorbs the carbon dioxide from the atmosphere to form barium carbonate. Similarly, when in contact with gypsum, it reacts rapidly to generate barium sulfate and calcium carbonate as well [12]. Due to the generation of these precipitants, the permeable pores within the gypsum substrate are filled and the further penetration of aqueous solution of barium hydroxide is blocked.

1.3. Research Content

In this study, a methanol-based barium hydroxide solution was explored for the conservation of gypsum mortar. Unlike in aqueous solutions, barium hydroxide solutions in methanol exhibit sufficient stability [13]. Research has indicated that barium hydroxide forms a solvated complex, [Ba(OH)₂(MeOH)₂] (MeOH), with methanol, where “Me” denotes a methyl group, and cannot dissociate into barium ions (Ba²⁺) and hydroxide ions (OH⁻) [14]. Thus, it reacts neither with gypsum nor atmospheric CO₂ and remains in a liquid state when the solution drips onto the gypsum block. This stability translates to excellent permeability of the methanol-based barium hydroxide solution in gypsum substrates.

The protective mechanism of barium hydroxide methanol solution was investigated using FTIR, XRD, and SEM-EDS. Meanwhile, protective efficiency was evaluated through a series of tests, including surface cohesion measurement, surface hardness testing, porosity analysis, capillary water absorption testing, erosion resistance assessment, and visual appearance evaluation.

The technical route adopted in this paper is shown in the following Figure 1:

2. Materials and Methods

2.1. Chemicals

Barium oxide (analytical reagent, AR, ≥98.0%), barium hydroxide (AR, ≥98.0%), methanol (AR, ≥99.5%), and calcium sulfate hemihydrate (guaranteed reagent, GR, ≥97.0%) were all purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All chemicals were of analytical grade and used without further purification. River sand with a median particle size of 257.50 μm [15] was supplied by Fujian Nonmetallic Mineral Co., Ltd. (Fuzhou, China).

2.2. Preparation of Gypsum Mortar Specimens

First, calcium sulfate hemihydrate, sand, and water were weighed at a mass ratio of 1:9:2.2 [8] and blended evenly to form a mixture. Then, the mixture was poured into a steel mold with internal dimensions of 5.0 × 5.0 × 2.0 cm³. After 20 min, the specimens were demolded and air-dried naturally to a constant weight prior to use. In the subsequent experiments, three parallel specimens were used for each test in the experimental group, and each experiment was repeated twice.

2.3. Preparation of Barium Hydroxide Solution

First, 3.89 g of barium hydroxide was dissolved in 100.00 mL of deionized water at 25 °C to prepare a saturated solution with a concentration of 3.89% (w/w). Second, 12.00 g of barium hydroxide was added to 88 mL of anhydrous methanol and stirred at 25 °C for 5 h to prepare a 3.89% (w/w) solution, less than one quarter of its saturation concentration in methanol [14].

2.4. Methods

The application procedure is as follows. Firstly, the methanol-based barium hydroxide solution was infiltrated into the gypsum mortar. Then, water was introduced further to initiate the reaction between barium hydroxide and gypsum. The protection mechanism of the proposed method was investigated using X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), and Scanning Electron Microscopy/Energy Dispersive Spectroscopy (SEM/EDS).

3. Treatment Methods

3.1. Selection of Treating Agents

Both aqueous and methanol-based barium hydroxide solutions were applied to the gypsum mortar specimens by surface brushing. After only one treatment with the aqueous barium hydroxide solution, obvious surface whitening was observed on the specimens, as presented in Figure 2a. It indicated that the aqueous solution of barium hydroxide is not a suitable protectant for gypsum mortar. However, almost no visible changes were detected on the specimens treated by the methanol-based barium hydroxide solution for three times instead, as shown in Figure 2b. Therefore, the methanol-based barium hydroxide solution was adopted in the subsequent experiments.

3.2. Treatment of the Mortar Specimens

The gypsum mortar specimens were brushed with the methanol-based barium hydroxide solution from their upper surface until reaching saturated adsorption. After the solvent (methanol) had completely evaporated, this treatment process was repeated. Specimens subjected to 1, 2, and 3 treatment cycles were labeled as T1, T2, and T3, respectively. After treatment, 3.0 g of water was applied to the specimens by surface brushing. After that, the specimens were placed in a controlled environment with a relative humidity of 60 ± 5% and a temperature of 20 ± 2 °C for 7 days before all kind of investigations.

3.3. Equipment Characteristics

Fourier transform infrared spectroscopy (FT-IR; TENSOR 27, BRUKER, Ettlingen, BW, Germany) was employed for compositional analysis, with the following operating parameters: scanning wavenumber range of 4000–400 cm⁻¹, resolution of 4 cm⁻¹, and 16 scans. X-ray diffraction (XRD; SmartLAB, RIKEN, Ltd., Saitama, JP-11, Japan) was used for crystalline phase identification, using Cu Kα radiation and the following settings: scanning range of 10–90°, step size of 0.01°, and scanning speed of 10°/min. Scanning electron microscopy (SEM; XE-GA-3XMU, TESCAN, Brno, Czech Republic) was utilized for morphological observation, operating in secondary electron (SE) mode with an acceleration voltage of 10.00 kV. Energy dispersive X-ray spectroscopy (EDS; Genesis 2000 XMS, EDAX, Inc., Mahwah, NJ, USA) was applied to characterize the elemental distribution.

The water erosion resistance of the specimens was evaluated by a soaking experiment as following. First, the specimens were immersed in 200 mL of ultrapure water, and the conductivity change in the water was recorded with a conductivity meter (model: DDSJ-308F, manufacturer: Rex, Shanghai, China). After that, the soaking solutions were diluted fivefold for further determination of Ca²⁺ and SO₄²⁻ concentrations by Ion Chromatography (IC). The IC analysis employed the following components: a cation separation column (Dionex IonPac CS12A, 4 × 250 mm, Sunnyvale, CA, USA), a cation suppressor (Dionex CERS 500 4 mm, Sunnyvale, CA, USA), an anion separation column (Dionex IonPac AS22, 4 × 250 mm, Sunnyvale, CA, USA), and an anion suppressor (Dionex ADRS 600 4 mm, Sunnyvale, CA, USA).

Scotch tape test (STT) [16,17] was used to test the surface adhesion of the samples as follows. A tape (2.5 cm × 5.0 cm) was pressed onto the sample and then peeled off with a constant force of 100 N. Five replicate tests were conducted at the same testing area. The peeling amount of mass was inversely related to the surface cohesion. Shore D hardness tester [18] (model: LX-D-1, manufacturer: Dongguan Sanliang Measuring Tool Co., Ltd., Dongguan, China) was used to measure the surface hardness of the specimens.

An electronic densitometer (model: MZ-C300, manufacturer: Mayzun, Shenzhen, China) was used to determine the open porosity and capillary water absorption of the specimens. The water absorption (Wa) [19] and open porosity (OP) [20] were calculated using the following formulas:

$$Wa\left(\mathrm{\%}\right)={\displaystyle \frac{M_1-M_0}{M_0}}\times 100$$

(1)

$$OP\left(\mathrm{\%}\right)={\displaystyle \frac{M_2-M_0}{M_2-M_1}}\times 100$$

(2)

M₀ is the dry mass of the specimen, and its saturated mass was recorded as M₁. M₂ is the suspended mass in water of the saturated sample. A colorimeter (model: WSC-2B, manufacturer: Shanghai Inresa Physical Optical Instrument Co., Ltd., Shanghai, China) was employed to determine the color difference in the specimens induced by the conservation treatment. The color difference (ΔE) was calculated using the formula provided below [21]:

$$∆E=\sqrt{∆{L^2+∆a^2+∆b}^2}$$

(3)

ΔE is the color difference, ΔL is lightness, Δa is the red-green component, and Δb is the yellow-blue component.

4. Results and Discussion

4.1. Action Mechanism

4.1.1. Fourier Infrared Spectrum Analysis

For the blank gypsum mortar (Figure 3a), the absorption bands at 3548, 1685, 1145, 668, and 600 cm⁻¹ are indicative of the presence of gypsum [22]. The absorption bands at 1080 and 796 cm⁻¹ are attributed to silicon dioxide (SiO₂), the main component of river sand [23]. After 7 days of treatment with the methanol-based barium hydroxide solution followed by water introduction, new absorption bands appeared in Figure 2b. The peaks at 1430, 1085, 874, and 711 cm⁻¹ originate from calcite-type calcium carbonate (CaCO₃) [24]. Additionally, the appearance of peaks at 1080, 637, and 610 cm⁻¹ is ascribed to barium sulfate (BaSO₄) [25]. These results confirm that gypsum in the mortar is converted into barium sulfate and calcium carbonate.

4.1.2. X-Ray Diffraction Analysis

For the blank specimen, the characteristic diffraction peaks of calcium sulfate dihydrate (2θ = 11.6°, 20.7°, 23.4°, 29.1°and 33.3°) [26] and quartz (2θ = 20.9°, 26.6°, and 39.5°) [27] are readily observed in Figure 4a, which are the main components of the gypsum mortar. After treatment with the methanol-based barium hydroxide solution followed by water introduction, new diffraction peaks appear in Figure 4b. The peaks at 22.4°, 25.8°, 28.8°, 31.5°, and 42.6° [28] correspond to barium sulfate (BaSO₄). The peaks at 23.0°, 29.4°, 39.4°, and 48.5° [29] are attributed to calcite (CaCO₃). These results are in good agreement with those from the FT-IR analysis, confirming that gypsum in the mortar is converted into barium sulfate and calcite.

4.1.3. Scanning Electron Microscopy/Energy Dispersive Spectroscopy Analysis

The surface of the blank specimen (Figure 5a) shows a loose and needle-like morphology with an elemental composition of Ca, S, and O, indicating the existence of gypsum. After the methanol-based barium hydroxide solution is followed by water introduction, the surface becomes dense, and the needle-like morphology is no longer observed (Figure 5b). In addition, the contents of Ba and C increase, while those of S and Si decrease. The changes in morphology and elemental composition of the cross-sections (Figure 5c,d) are highly similar. Consistent with the FT-IR and XRD results, these changes are all attributed to the transformation reaction between gypsum and barium hydroxide in the open air. Thus, the final dense structure is actually the morphology of barium sulfate and calcium carbonate products.

4.2. Performance

4.2.1. Permeability

The results of permeability of the methanol-based barium hydroxide solution in gypsum mortar are presented in Figure 6. Figure 6a is the cross-section image of the blank specimen. For the specimen after only one cycle of surface infiltration treatment, the entire cross-section appears wet (as indicated by the uniformly darker color) in Figure 6b, indicating that the specimen was fully infiltrated. For improved visibility, the methanol-based phenolphthalein solution was brushed onto the cross-section. The presence of the bright red color [30] throughout the cross-section (Figure 6c) further verifies complete infiltration of the specimen by the methanol-based barium hydroxide solution. The penetration depth of 20.00 mm demonstrates the good permeability of methanol-based barium hydroxide solution in gypsum mortar.

4.2.2. Water Erosion Resistance

In the soaking experiment, gypsum in the mortar is dissolved by water and dissociates into calcium ions (Ca²⁺) and sulfate ions (SO₄²⁻) [31], leading to the conductivity increase in the soaking solution. Thus, conductivity can be used as an indicator to evaluate the water erosion resistance of the specimens. The conductivity values of the specimens are presented in Figure 7. The blank gypsum mortar has a conductivity as high as 1750 μS/cm. After 1, 2, and three treatment cycles, the conductivity of the specimens decreases to 710 μS/cm, 290 μS/cm, and 65 μS/cm, indicating a significant improvement in water erosion resistance. This improvement is attributed to the reaction between barium hydroxide and gypsum in the mortar: Gypsum is gradually converted into less soluble barium sulfate (BaSO₄) and calcium carbonate (CaCO₃) [32], and the water erosion resistance of the specimens is improved accordingly. This can be proved further by the results of ion chromatography (IC) in Figure 8. For the blank specimen, the concentrations of Ca²⁺ and SO₄²⁻ are 183 μS and 125 μS, respectively. After 3 treatment cycles, they are reduced to 20 μS and 5 μS. The conductivity results are in good agreement with the IC results.

4.2.3. Mechanical Strength

The surface cohesion [33] of the specimens was determined by Scotch Tape Test (STT). From Figure 9, the blank specimen has the largest mass loss up to 8.7 mg/cm² in peeling test cycles, indicating the weakest surface adhesion. After 1, 2, and 3 treatment cycles with the barium-based protectant, the mass loss decreases to 6.3, 3.9, and 2.7 mg/cm², respectively. This result indicates that the surface cohesion of the treated specimens is obviously enhanced. Figure 10 also shows an increase in the surface hardness of the treated specimens. The surface hardness of the blank specimen (UT) is only 20.80 HD. For specimens subjected to 1, 2, and 3 treatment cycles, the surface hardness increases to 41.88 HD, 52.16 HD, and 60.94 HD, respectively. These improvements in mechanical strength can be attributed to the application of the barium-based protectant, which fills the weathering-induced cracks and pores and makes the mortar more compact, as can be seen from Figure 5.

4.2.4. Compatibility

The results of color difference analysis are presented in Figure 11. The color change in the specimens with a different number of treatments was mainly due to significant changes in the a* and b* values, while the luminance (L*) values remained close to the values of the reference samples, and the change in L* values usually implies a change in surface roughness, which means that the reinforcement treatment did not change the lightness or darkness of the samples, i.e., it did not cause reflections and glare on the surface of the artifacts. At the same time the ΔE* values of the specimens after 1, 2, and 3 treatment cycles are 1.78, 2.32, and 2.61, respectively. Generally, a color difference (ΔE*) of less than 3.0 is imperceptible to the naked eye [34]. It means the appearance of the specimens remains nearly unchanged, which is esthetically acceptable. This phenomenon is attributed to the excellent permeability of the methanol-based barium hydroxide solution, prevents the deposition of barium-containing substances on the shallow surface of the specimens and thus avoids surface “whitening”.

The results of open porosity and capillary water absorption are presented in Figure 12. Compared to untreated specimens, gypsum mortar specimens reinforced with the methanolic barium hydroxide solution exhibit a distinct decreasing trend in porosity and water absorption. As shown in Figure 12a, after 1, 2, and 3 treatment cycles, the total open porosity of the specimens decreases from 24.75% to 23.35%, 22.86%, and 19.25%, respectively. A similar trend is observed for capillary water absorption: as displayed in Figure 12b, it reduces from 18.37% to 18.00%, 16.88%, and 15.75%, respectively. The results indicate that the methanolic barium hydroxide solution had little effect on the internal porosity of the specimens, and its influence on the material’s water permeability was within acceptable limits. Additionally, the decrease in both open porosity and capillary water absorption are caused by the application of the barium-based protectant, which fills the weathering cracks and pores. For the mortar, the reduction in porosity and capillary water absorption are beneficial to a certain extent [35]. It means that less water can enter into, and the consequent water erosion is restrained.

5. Conclusions

In this study, a methanol-based barium hydroxide solution was used as a treatment agent to improve the water erosion resistance of gypsum mortar. Its action mechanism and protective performance were investigated, and the research found three conclusions which are as follows:

• Owing to its sufficient stability in the open air, the methanol-based barium hydroxide solution exhibits excellent permeability, achieving a penetration depth of 20 mm in gypsum-sand mortar. The total color difference (ΔE*) of the treated specimens is less than 3.0, which is closely associated with the good permeability of the barium hydroxide solution.
• As indicated by the analysis results of X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR) and scanning electron microscopy/energy dispersive spectroscopy (SEM/EDS), calcium sulfate dihydrate (CaSO₄·2H₂O) in the gypsum-sand mortar is converted into insoluble barium sulfate (BaSO₄) and calcium carbonate (CaCO₃) after sequential treatment with the methanol-based barium hydroxide solution and water.
• The mechanical properties, including surface cohesion and hardness, are significantly enhanced. For specimens subjected to 1, 2, and 3 treatment cycles, the surface hardness increases from 20.80 HD to 60.94 HD and the STT shows that the mass loss decreases from 8.7 mg/cm² to 2.7 mg/cm².Additionally, its open porosity and capillary water absorption are both reduced. After 1, 2, and 3 treatment cycles, the total open porosity of the specimens decreases from 24.75% to 19.25% and capillary water absorption reduces from 18.37% to 15.75%, improving its water erosion resistance. These changes are attributed to the insoluble barium sulfate (BaSO₄) and calcium carbonate (CaCO₃) formed after the introduction of the barium-based protectant, which densify the mortar structure and fill weathering-induced cracks and pores.

During the use of the methanol-based barium hydroxide solution in this study, operators should wear protective equipment, conduct the operation in a well-ventilated environment, and avoid direct skin contact and inhalation of vapors. In addition, this experiment lacks long-term durability tests, which require further improvement and supplementation in subsequent studies.

In conclusion, in view of these positive results, the proposed method using the methanol-based barium hydroxide solution as a protectant shows great potential for the in situ conservation of gypsum-based mortar in architectural heritage.