Study About the Influence of Diatoms on the Durability of Monumental Limestone

Abstract

This study focuses on the evaluation of the effects of a natural treatment of limestone rock samples using microalgae known as diatoms. A total of 18 samples in the form of 50 mm cubes, carved from limestone rock from Boñar (Spain), were analyzed, divided into experimental and control groups with an equal number of samples. Through various tests evaluating porosity, water absorption, frost resistance, and salt crystallization, diatom-treated samples were found to show higher porosity and water absorption compared with the control samples, especially when the entire sample was analyzed as a whole. However, in tests focusing on the surface side most exposed to biodeposition, reduced water absorption was observed in the treated samples, suggesting an improvement in their antiabsorption properties. In addition, slightly higher frost resistance was detected in the treated samples. For this reason, this study provides valuable information on the potential of diatoms to influence the properties of limestone rocks, which can serve as a basis for future research in this field and for the development of more effective treatments to improve the characteristics of rocks used in various applications.

1. Introduction

The protection of historical heritage has become a major concern in recent years, particularly regarding the preservation of historical structures and artwork. The increase in pollutants in the atmosphere and, consequently, in rainwater has accelerated the deterioration caused by natural processes; factors such as growing air pollution levels have accelerated the rate at which these ancient structures and works decay [1,2,3,4]. To address this concern, new techniques need to be researched and developed to prevent the deterioration of works and historical buildings as much as possible and to improve the durability of new ones in a sustainable manner. In this regard, new surface treatments with compounds such as graphene oxide or employing microorganisms have already been considered to achieve protective surface coatings. Promising results have been obtained from innovative processes that involve the use of organisms to enhance properties through biomineralization processes [5,6,7].

Biomineralization is a biochemical process by which living organisms stimulate the formation of minerals. Currently, a large number of organisms are known to be involved in mineralization processes:

• Plants in the formation of silicates.
• Invertebrates in the formation of carbonates.
• Vertebrates in the formation of phosphates and carbonates.
• Microbes in the biomineralization of gold, copper, and uranium.

Bacteria are powerful organisms for biomineralizing carbonates, phosphates, sulfates, and silicates. So far, the most studied biomineralization processes to obtain improvements in building materials have been:

• Precipitation of calcium carbonate by microbial action [8,9,10,11,12,13].
• Precipitation of biogenic silica by microalgae [14,15].

In the present study, a series of tests are performed on limestone samples subjected to precipitation of biogenic silica by diatom microalgae. The goal is to achieve a natural surface treatment by harnessing the silicon biomineralization produced by these microalgae, known as diatoms. These microalgae play a fundamental role in oxygen production on Earth and present characteristics that render them suitable candidates for this type of experimentation. Their cell wall is almost entirely silicon and, therefore, they can easily colonize surfaces by creating biofilms; after their death, the remains composed of silicon are left. Diatoms are a group of microalgae generally unicellular, although they can also coexist, forming cellular chains, fans, or other types of groupings. These microalgae include about 20,000 species and represent one of the most common types of phytoplankton. In addition, diatoms can also live colonizing underwater or even land surfaces known as benthic life forms.

These types of microalgae can develop in both salt and fresh water and are able to survive in environments under extreme temperature and salinity conditions [16]. When diatoms die, their frustules are left at the mercy of water currents and, over time, they commonly accumulate and compact, mainly in sedimentary environments, forming the rock diatomite. Due to the nature of its formation, diatomite is a siliceous sedimentary rock widely used for filtration processes in different industries, and diatomaceous earth is extracted from it for various uses, such as fertilizer, insecticide, or building material [17,18,19,20]. Silicon deposition can be achieved thanks to the ability of diatoms to colonize surfaces by adhering to them through the creation of an initial biofilm [21]. Here, the aim is to take advantage of the silicon deposition produced by the frustules of dead diatoms and obtain limestone rock samples with a natural surface treatment, which enable the analysis of any improvements in surface properties.

The present work focuses on the engineering involved in the planning and execution of these experimental tests and the analysis of the results obtained. However, the underlying ideas and elements that generated both the experiment and the samples studied cannot be overlooked. Therefore, the motivations behind the experimentation, the concepts on which this is based, what diatoms are, and their special characteristics need to be properly comprehended.

2. Materials and Methods

2.1. Materials

A total of 18 samples were used for the tests: 9 experimental samples and 9 control samples. All samples were made from limestone rock originating from Boñar, Spain, cut into cubes with 50 mm edges. Limestone has been widely used throughout history as a construction material, either in its natural form or crushed for use as aggregate. Many monuments and historic buildings were constructed using limestone, and this remains an important component in the production of cement and stucco today.

Limestone is a sedimentary rock formed by the accumulation of sediments that undergo a series of physical and chemical processes, leading to their consolidation into coherent and solid rock. This rock is primarily composed of calcium carbonate, often containing varying amounts of magnesite and other carbonates, along with small quantities of minerals such as clay and quartz. The action of water, in particular when containing carbonic acid, causes limestone dissolution through a chemical weathering process known as karstic weathering [22,23,24].

The Boñar stone used in the present work is a fine-grained dolomitic limestone from the Campanian age, with cream to beige-ochre tones. This is a compact, uniform, sedimentary rock extracted in large regular blocks. The stone shows some sedimentary structures such as laminations, bioturbations, and trace fossils, indicating a lagoon-type depositional environment. Its appearance is homogeneous and finely granular, with reddish grains of varying sizes and occasional intergranular spaces filled with coarse calcite, sometimes associated with fossil remains. Its texture is crystalline and fine-grained, with an equigranular granoblastic structure and idiomorphic to subidiomorphic dolomite crystals showing a relict clastic texture. Physical and mechanical specifications are presented in detail in

Table 1. Physical and mechanical properties of Boñar stone.

Test	Standard	Unit	Value
True density	UNE-EN 1936 [25]	kg/m3	2832
Apparent density	UNE-EN 1936 [25]	kg/m3	2510
Compressive strength	UNE-EN 1926 [26]	MPa	117
Static elastic modulus	UNE-EN 14580 [27]	MPa	34 × 103
Flexural strength	UNE-EN 12372 [28]	MPa	16

.

The samples were evenly distributed across the surface of a pond to minimize potential variations among them. These specimens were submerged at a depth of 15 ± 5 cm below the water surface, with the test face oriented parallelly to the bottom to promote the growth of diatoms. This positioning is crucial, as previous studies on concrete [14,15] have demonstrated that orientation significantly influences the biodeposition process.

For this reason, it is important to clearly determine the position of the faces or surfaces of the samples submerged in the pond throughout the entire experimental work. As shown in Figure 1, samples are placed with their upper face marked in red. In this manuscript, the different surfaces of these samples are referred to as:

• Upper face: marked with a red square.
• Lateral faces: the four faces adjacent to the upper one and marked with a blue line.
• Lower face: the face opposite to the upper face.

2.2. Methodology

The main objective of this work is to verify whether or not a positive influence appears on the surface properties and behavior of the limestone rock samples after submerging them in the pond. In addition, the impact on the surface properties that the experiment itself may have caused to obtain the diatoms to act on the samples was also evaluated.

To achieve these goals, several test methods were selected to identify and quantify possible differences in behavior between the pond samples and the control samples. Given the limited number of specimens available, the priority was to maximize their use and obtain as much information as possible.

For this reason, the noninvasive tests were performed first, as these do not alter either the samples or the biodeposition of silica on their surface and leave the more invasive and destructive tests for the final stages, as these do cause significant surface wear. Thus, the use of the samples was carefully planned, alternating their use to minimize their deterioration and avoid interference with the biofilm of the diatom biodeposition.

These details are presented in

Table 2. Classification of the noninvasive and destructive tests used.

Noninvasive Tests	Destructive Tests
Porosity test	Spray water absorption test
Water absorption test at atmospheric pressure	Water absorption test by pipette method
Water absorption test by capillarity	Frost resistance rest

, which offers a detailed classification of the noninvasive and destructive tests used.

To achieve this natural treatment, nine limestone rock samples were introduced in a pond with the presence of diatoms. All the samples introduced were cube shaped, with a side of 50 mm, and originated from Boñar (Spain).

Diatom colonies are known to grow on submerged substrates in as little time as 4 weeks [29]. For the present work, samples were immersed in an experimental pond, as illustrated in Figure 2a, and exposed to natural environmental conditions for a period of 1 year and 2 months in León, Spain. This time period was stipulated because, during this time, it was considered that a sufficiently dense overlay would potentially be generated due to the biodeposition of the silica frustules of the diatoms on the samples. This process would lead to a surface sealing that would possibly improve the surface properties, as illustrated in Figure 2b. In addition, it should be noted that the key environmental factors influencing diatom growth are the temperature and hours of light received [30], which coincide with positive photoperiods in which hours of light exceed hours without light. For this reason, data on temperature and solar radiation were collected throughout the entire period during which the samples were submerged in the pond (Figure 3), using records obtained from the Spanish State Meteorological Agency (AEMET) [31], specifically from the meteorological station at Virgen del Camino, located at approximately 1 km from the pond.

As a result of all the aforementioned factors, in particular due to the higher exposure to solar radiation, the upper side, which receives more direct light, shows a higher concentration of diatoms. Conversely, the side faces, with less exposure, show significantly lower growth due to the limited availability of light for photosynthesis and colony development. Therefore, the upper side, which presents the highest silica biodeposition, is studied.

2.2.1. Porosity Test

The porosity tests performed to study the effect of diatoms include:

• Open or communicated porosity, also called efficient or effective.
• Closed porosity.
• Total porosity (sum of the previous two).

The study of the open porosity of the samples was performed following the UNE EN 1936:2007 [25] standard that describes the test method to determine the real and apparent density and the open and total porosity of natural stone. To ascertain the real density, each specimen should be ground separately, so this part was left aside to be able to use the samples in subsequent tests. Six cubic samples with 50 mm edges were used: three with diatoms (D1, D2, and D3) and the other three as control samples (C1, C2, and C3). Figure 4 shows the samples used for the open-porosity test painted and tagged as control or diatom.

All specimens were waterproofed with fiberglass paint on the face opposite to the surface under analysis. This measure was taken as this side did not show the presence of diatom biodeposition.

Once the samples were painted and after drying the paint in the air for 1 day, they were placed in a ventilated oven at 68 °C for another 4 consecutive days to achieve maximum drying and constant masses. The samples were dried and weighed one by one, and the value of their mass was noted as dry mass (m_d); then, they were allowed to cool in a desiccator until reaching room temperature. The next step of the test consists in obtaining the water-saturated mass of the samples (m_s). Note that, even when the saturation of the sample is reached, the smaller the difference between two consecutive weighings, the greater the amount of air in the open pores has been removed. The same sequence was followed with each sample: the sample is removed from the water, quickly dried with a damp cloth to remove only the water from the surface, weighed and reimmersed, as shown in Figure 5a. When saturation is reached in all samples, the next step is to obtain the value of the mass of the test cube submerged in water (m_h). A hydrostatic weighing was used, as shown in Figure 5b; the sample was placed in a basket under water attached to the hydrostatic balance, and its mass value was tared. At this point, values of the mass of the dry test piece, the mass of the sample saturated in water, the mass of the test piece submerged in water, and the density of water in kg/m³ at 20 °C (998 kg/m³) were known.

2.2.2. Water Absorption Test at Atmospheric Pressure

The next part of the study consists in determining the amount of water the test cubes absorb at atmospheric pressure and detecting the differences in behavior between the samples with diatoms and the control samples. The test method used is described in the UNE-EN 13755:2008 [32] standard and based on completely immersing the test pieces in water at atmospheric pressure to study their absorption capacity. The process consists in drying the test cubes until they reach their constant mass and then completely submerging them in a tank or container with water until all the samples reach saturation. Knowing the difference between the mass of the test tube saturated in water and the mass of the dry test tube, the amount of water absorbed by each of them was determined expressed as a percentage according to the expression (1):

$$A_b={\displaystyle \frac{m_s-m_d}{m_d}}.100$$

(1)

For this test, six cubic specimens with 50 mm edges were used: three treated with diatoms (d1, d2, and d3) and three as control specimens (c′1, c′2, and c′3). All specimens were waterproofed with fiberglass paint on the face opposite to the surface under analysis.

Initially, the specimens were dried in a ventilated oven at 70 ± 5 °C until reaching a constant mass; specifically, they were kept for 5 days at 68 °C to prevent potential damage to the deposited biofilm by a higher temperature. As in the previous test, diatom-treated samples with the lower surface isolated were used to compare surfaces potentially containing diatoms with those known to be free of them. Also, a random face was isolated in the control samples. Following this initial drying phase, the specimens were returned to the oven for an additional 4 days at 68 °C, again until constant mass was achieved. Once dried, each specimen was weighed to determine its dry mass (m_d) and then allowed to cool in a desiccator.

The specimens were placed on supports (plastic grids), maintaining a minimum distance of 15 mm among them. Tap water at 20 ± 10 °C was then added to reach half the height of the specimens, and the container was sealed. This moment is defined as t₀, marking the start of the test.

After approximately 60 min (t₀ + [60 ± 5] min), water was added to reach three-quarters of the specimen height, and the container was resealed. After about 120 min (t₀ + [120 ± 5] min), additional water was added until the specimens were fully submerged, with a 25 ± 5 mm layer of water above them, and the container was closed once again. The specimens remained submerged in the closed container, as shown in Figure 6. After 48 ± 2 h (t₀ + 48 h), the specimens were removed, lightly cleaned with a damp cloth, weighed, and their masses recorded. Once weighed, the specimens were returned to the water, and the container was resealed. From this point, the process of removing, drying, and weighing was repeated every 24 ± 2 h to track mass changes.

The test was considered as complete when all specimens reached their saturated mass in water (m_s), defined as the point at which the difference between two consecutive weighings is less than or equal to 0.1% of the first of these two recorded masses.

2.2.3. Water Absorption Test by Capillarity

The water absorption test by capillarity was performed following the method described in the UNE-EN 1925:1999 standard for the determination of the coefficient of water absorption by capillarity in natural stone [33]. The test procedure described in this standard consists in immersing the test surface of the samples in water to a depth of 3 ± 1 mm. As time passes, measurements are taken to verify the increase in mass that each sample experiences due to the absorbed water.

According to the standard, the samples used for the present work were cubic with an edge of 50 mm; three samples contained diatoms and were named D1, D2, and D3, and three samples were used as control samples and named C1, C2, and C3. As shown in Figure 7, the samples were arranged in closed containers to start the capillarity test. Previously, the specimens had first been dried for 5 days in a ventilated oven at 68 °C. In this case, to isolate as much as possible the upper face of the samples with diatoms and to prevent water entry through the four lateral faces adjacent to the studied face (which are in contact with water), they were waterproofed with fiberglass paint to a height of 1 cm.

This is necessary because diatom biodeposition occurs most significantly on the side parallel to the pond surface (the upper face). Similarly, four faces adjacent to a random face were partially covered in the control samples.

After allowing the paint to air dry for 1 day, the samples were placed in a ventilated oven for an additional four days at 68 °C to achieve a constant dry mass before starting the test. The dry samples were then positioned in a container or tank on supports that ensure a separation among them and allow the test face to be immersed without being in full contact with the tank surface. In this way, the samples rest partially on their base, minimizing any interference with their absorption capacity. The same plastic grid used in the previous test was employed for this one. Water was added until the test face of each sample was submerged to a depth of 3 ± 1 mm, and timing began immediately. After initially short and then progressively longer intervals, the samples were removed from the water, and the slightly submerged base was dried with a damp cloth, carefully removing only surface water drops without removing moisture from the sample. The samples were then weighed immediately. Once the mass values had been recorded, the samples were returned to the container to continue the test.

According to the standard, a minimum of seven measurements must be taken, and the choice of time intervals depends on the absorption capacity of the material. For samples with high water absorption capacity, the recommended intervals are 1, 3, 5, 10, 15, 30, 60, 480, and 1440 min. For samples with low absorption capacity, the suggested intervals are 30, 60, 180, 480, 1440, 2880, and 4320 min. For the present study, the following time intervals were used: 1, 3 min, 5 min, 10 min, 15 min, 30 min, 1 h, 2 h, 3 h, 6 h, 11 h, 24 h, and thereafter every 24 h until the end of the test. Time intervals must be measured with a precision of ±5%, and the test is considered complete when the difference between two consecutive weighings is less than or equal to 1% of the mass of water absorbed by the sample.

Similarly, to evaluate the effect of the 1 cm lateral sealing with fiberglass paint—designed to isolate exclusively the upper surface under study—the test was also repeated using samples without this lateral sealing. This comparison was performed considering the concentration of diatoms on the lateral faces to be low or practically negligible compared with that on the upper face.

2.2.4. Spray Water Absorption Test

To assess whether differences exist or not in the properties of the six samples (d1, d2, d3, c′1, c′2, and c′3) when only their top face is considered, favoring in particular those treated with diatoms, a method that maximizes the isolation of the upper side during the analysis was used.

The test consists in spraying pressurized water from a distance onto the surface under examination only. This is done using a compressor (L 100-50 8 bar 1 CV 64 L/min 50 L) and a pressure gun with an attached tank, which ensures a uniform and continuous dispersion of the water in the form of very fine droplets. This process does not damage the biofilm, as the diatoms are strongly attached to the rock substrate through their extracellular polymeric secretions [21]. The aim is to simulate an extreme scenario similar to a very humid environment, light wind-driven rain, or dense fog.

The principle on which this test is based relies on comparing the dry masses of each of the samples with their successive increasing masses during the test to determine the amount of water absorbed. The six cubic samples with 50 mm edges (d1, d2, d3, c′1, c′2, and c′3) were used for this test. All were dried until reaching the constant mass value (m_d) or dry mass. The samples were kept for 4 consecutive days in a ventilated oven at 68 °C. Once all the samples were dry, they were weighed to record the value of the dry mass of each sample and then allowed to cool in a desiccator until reaching room temperature.

After cooling, one sample was selected to start the actual test. The selected sample was placed in a tank or container that isolates all its faces from contact with water except the one to be analyzed. In this case, an insulated wooden box was constructed to prevent moisture transmission from the outside. This box has a square-shaped opening with a side length of 40 mm. The opening was made slightly smaller than the size of the square face of the samples to avoid a tight fit that could cause water to enter the interior of the box or allow the samples to absorb water from their edges during placement.

With the sample inside the closed box, water as sprayed at a pressure of 6 bars perpendicularly to the exposed face of the sample. In Figure 8, the water spray process can be observed. Spraying was performed at a distance of 120 ± 10 cm for 20 s. Subsequently, the sample was removed from the box, ensuring that no water had entered the interior; the exposed face was dried with a damp cloth and weighed to record the new mass. The sample was then reintroduced into the box and closed to repeat the process as many times as desired or until the sample is observed to show minimal changes in mass.

Once finished, the sample was set aside, and the entire process was repeated with the remaining samples. The process was repeated 19 times for each sample, with a spray durations of 3 s, 5 s, and 10 s in the first three cycles and 20 s in the remaining 16 cycles.

2.2.5. Water Absorption Test by Pipette Method

The UNE-EN 16302:2016 [34] standard defines the procedure to measure water absorption using the Karsten tube pipette method, which determines the amount of water absorbed by porous inorganic materials commonly found in cultural heritage. These materials include natural stones, such as limestone and marble, as well as artificial materials such as bricks, mortar, and plaster.

The test aims to measure the degree of water penetration under a pressure similar to that of rainwater. Its main advantages are that it is nondestructive, can be performed both in the laboratory and in situ, and is applicable to untreated materials as well as those that have been treated or subjected to aging processes.

Water absorption is quantified by measuring the volume of water, in ml, transferred from the pipette to the material through a defined test area, expressed in cm². This area is determined by the diameter (D) of the contact surface between the Karsten tube and the sample, which, in this case, is 3 cm.

Results are expressed as ml/cm², indicating the amount of water absorbed by the material’s surface after a fixed time, measured with a stopwatch accurate to 1 s. A sealing material is applied to ensure proper fixation of the Karsten tube to the sample surface. This sealant must prevent water leakage during the test and must not damage or alter the sample surface. For the present work, butyl rubber was used for sealing. Although the test is considered nondestructive, removing the tube may cause slight detachment of surface material, which could affect the behavior of diatom-treated samples in subsequent tests.

Therefore, samples tested with the Karsten tube method were not reused for further analyses. To avoid water leakage at the interface between the sample surface and the Karsten tube, the surface must be uniform, flat, and free of visible cracks. The area tested must be larger than the diameter of the test area (D).

Depending on the material’s heterogeneity, the number and size of test surfaces may vary, but at least three samples or different surfaces must be evaluated per series. For this experiment, six samples were tested: three with diatoms (D1, D2, and D3) and three as control samples (C1, C2, and C3).

Prior to testing, samples were dried until a constant mass was achieved, using a ventilated oven at 68 °C for 4 days. Before attaching the Karsten tube and starting the test, samples were stored in a desiccator until reaching room temperature (23 ± 1 °C). Once cooled, the Karsten tube was fixed to the surface with light pressure, carefully filled with distilled water to the 0 level on the pipette scale, and the timer was started. The procedure was performed carefully to ensure no air bubbles were present within the pipette and no leakage at the connection.

The standard recommends recording the water level change at intervals ranging from 10 s to 1 min initially, and then every 5 min until a steady value is reached, with the test ending after 1 h if equilibrium is not achieved. However, during the experiment, no significant changes in water level were observed at 5 or 10 min intervals for most samples. Consequently, a new criterion was adopted: recording the time taken for each sample to absorb increments of 0.1 mL of water, corresponding to the smallest division on the pipette scale. Additionally, water absorption after 24 h was also recorded.

Figure 9 illustrates the full water absorption test setup, including the Karsten tube fixed to both control and diatom-treated samples.

2.2.6. Frost Resistance Test

The standard UNE-EN 12371:2011 [35] describes a test procedure for determining the freeze–thaw resistance of natural stone. This procedure consists in subjecting stone samples to repeated freezing cycles in air, followed by thawing in water. Since the number of samples was limited, only six cubic samples with 50 mm edges were used in the present study: three with diatoms (D1, D2, and D3) and three as control samples (C1, C2, and C3), with the sides opposite to the test surface waterproofed.

First, the samples were dried until they reached a constant mass; they were then placed in a ventilated oven for 4 days at 68 °C. After drying, the samples were allowed to cool until reaching room temperature, and each imperfection was marked with a marker to monitor potential visible changes during the cycles.

Before starting the first freeze–thaw cycle, the samples remained in contact with water for 48 h inside a closed container to ensure they did not begin the first cycle in a dry state. During this time, measurements were taken to record the increase in mass at two 24 h intervals. This enabled the determination of the amount of water each sample absorbed and the comparison with values obtained under the same immersion conditions after the freeze–thaw cycles.

The freeze–thaw cycle began after the 48 h water immersion. Each cycle consisted in a freezing stage in air at −20 °C for 6 h, followed by a thawing stage in a closed container in contact with a 3 mm-thick layer of water for another 6 h.

After each complete cycle, a visual inspection of each sample was performed to check for signs of degradation. When any damage was observed, the cycle number in which this appeared was recorded. After inspection, the samples were subjected again to the freezing stage, starting a new cycle, as shown in Figure 10. This process can be repeated as many times as necessary; here, a maximum of 45 cycles was chosen. Once the maximum number of cycles had been completed, the samples were fully submerged in water for 24 h to allow any loose material particles to detach. Finally, the samples were dried again in a ventilated oven for 4 days at 68 °C to determine the dry mass of the samples after being subjected to 45 freeze–thaw cycles.

With the samples dry and after undergoing the 45 freeze–thaw cycles, they were once again partially submerged in a 3 mm-thick layer of water for 48 h, in the same manner as at the beginning of the test, before starting the freeze–thaw cycles. Measurements of the increase in mass were taken at two 24 h intervals to enable comparison of their behavior before and after the freezing cycles. After the samples had been subjected to the freeze–thaw cycles, a useful comparison could be made to evaluate the behavior changes between the control samples and those with diatoms, independently, to determine the extent at and manner in which these changes occurred.

The pipette method was used to perform another water absorption test, following the methodology described in Section 2.2.5. Measurements were recorded to determine how long the samples took to absorb each increment of 0.1 mL of water, as well as the total water absorption after 24 h of testing. Figure 11 shows the pipette tubes placed over the samples from the freeze–thaw cycle.

From the results obtained in this complementary part of the frost resistance test, the behavioral differences between the upper face of the samples with diatoms and that of the control samples after exposure to low temperatures can be better understood.

2.2.7. Salt Crystallization Resistance Test

For this final test, the method described in the UNE EN 12370:2020 [36] standard was used to determine the resistance of the samples to salt crystallization. This standard outlines a test procedure for natural stone samples with an open porosity greater than 5%. This procedure consists in immersing the samples in a sodium sulfate solution and then drying them in an oven. After completing the required number of cycles, the mass variation of the samples is evaluated.

For this test, six samples were used: three diatom-treated samples (d1, d2, and d3) and three as control samples (c′1, c′2, and c′3), with the side opposite the test surface waterproofed. First, the samples must be dried in a ventilated oven until they reach a constant mass at the temperature specified by the standard (105 ± 5 °C). For the present work, they were dried for 4 consecutive days at 105 °C. Once dried, the dry mass of each sample was recorded, and then the samples were allowed to cool until reaching room temperature.

Afterwards, the samples were placed in a container where they were in contact with a 3 mm-thick layer of a 14% dehydrated sodium sulfate solution. The samples were spaced at least 10 mm apart from each other and 20 mm from the edges of the container, as shown in Figure 12.

The samples remained in contact with the solution for 2 h before being placed in a ventilated oven at a temperature of 105 ± 5 °C for a minimum of 16 h. The samples were then be cooled until reaching room temperature for 2 ± 0.5 h before starting a new cycle. Once the maximum number of cycles had been completed, the samples were kept completely immersed in water for 24 h and then carefully washed if their condition allowed it. After this, the samples were dried again to a constant mass, and their mass values were recorded in the same way as at the beginning of the test. Mass variation of the samples in percentage could then be evaluated using Equation (2).

$${\displaystyle \frac{\mathrm{F}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{l} \mathrm{d}\mathrm{r}\mathrm{y} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}-\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{d}\mathrm{r}\mathrm{y} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}}{\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{d}\mathrm{r}\mathrm{y} \mathrm{m}\mathrm{a}\mathrm{s}\mathrm{s}}}$$

(2)

Relative humidity during the initial phases of each drying cycle can be achieved by placing a tray of water and heating the oven for about 30 min before introducing the samples. The volume of water to be added should be 2.5 ± 0.5 mL per liter of oven volume. In this case, a 50 × 50 × 50 cm³ oven was used and, therefore, a volume of 300 mL of water was calculated.

3. Results and Discussions

3.1. Open-Porosity Test Results

The mass of water absorbed by the samples at each time interval is calculated as the difference between the mass of the sample at that interval and its initial dry mass. The mass of water absorbed by the samples between each 24 h interval corresponds to the difference between the mass at a given interval and the mass at the previous interval. It was observed that sample D3 presented a large open crack on one of its lateral faces. Due to this crack, the open-porosity level of the D3 sample was expected to be significantly higher than that of the other two diatom samples (D1 and D2). In addition, when the D3 sample was submerged in water at the beginning of the test, small air bubbles could be seen continuously escaping from the crack.

From the average values shown in Figure 13, it can be verified that the samples with diatoms absorbed more water than the control samples during the first two 24 h intervals, with a greater difference in the first interval than in the second. After this initial phase, the diatom samples slowed their absorption rate. In each subsequent 24 h interval, they absorbed less water than the control samples, although in a very similar manner. Nevertheless, this did not prevent the diatom samples from maintaining a higher total water absorption overall.

Analyzing the average values, it can be observed that the samples with diatoms absorbed more water throughout the test. Finally, after 240 h of immersion, the diatom samples absorbed an average of 11.35 g of water, while the control samples absorbed 9.87 g. Therefore, the diatom samples absorbed on average 1.48 g more water than the controls. Once the values for the mass of water absorbed at each time interval from the start of the test had been analyzed, the behavior of the samples between each 24 h interval could also be assessed.

From the average values, the open porosity of the samples with diatoms is 0.97% higher than that of the control samples, as shown in

Table 3. Average of the results obtained from open porosity.

Average Open Porosity Control Samples (%)	Average Open Porosity Diatom Samples (%)
7.69	8.66

. This may be due to the chemical weathering process caused by water acting on the limestone rock samples during their immersion in the pond for diatom treatment.

Similarly,

Table 4. Average volume of open pores in control and diatom samples.

Average Volume of Open Pores in Control Samples (mL)	Average Volume of Open Pores in Diatom Samples (mL)
9.89	11.37

shows the open pore volume in the diatom samples to be also greater. Once again, the value for sample D3 was excluded from the average. On average, the open pore volume of the samples with diatoms increases by 1.48 mL compared with that of the control samples.

This result was expected since, as abovementioned, the samples with diatoms absorbed an average of 1.48 g more water than the control samples at the end of the test. These additional 1.48 g of water were stored in the 1.48 mL of additional open pore volume present in the diatom samples. Because the density of water is very close to 1 g/mL, the results were expected to be similar.

Regarding the apparent volume, which considers the volume of open pores as part of the total sample volume, all samples present similar dimensions. As shown in

Table 5. Average of the results obtained from the apparent volume.

Average Apparent Volume in Control Samples (mL)	Average Apparent Volume in Diatom Samples (mL)
128.39	128.82

, no significant variation in this value was observed between the diatom and control samples.

The results of the test show that the samples with diatoms present greater open porosity than the control samples. This implies that the diatom-treated samples experienced an increase in open pore volume and were able to be filled with a greater amount of water, and more quickly, compared with the control samples. In addition, it can be observed that the apparent volume of the diatom samples is practically identical to that of the control samples. This confirms that the increase in water absorption and absorption rate in the diatom samples is not due to a larger overall volume that would allow them to hold more water for that reason alone.

Since this test involves total immersion of the samples in water, it allows us to determine the open porosity values of the entire sample, rather than those of specific surfaces. Therefore, it is possible that there was some sealing or reduction of open pores on the upper face of the diatom samples, while open porosity increased on the four lateral faces. This effect may be due to the chemical weathering process caused by the pond water during the immersion period for diatom treatment and the initial exposure phase.

3.2. Results of Water Absorption Test at Atmospheric Pressure

Figure 14a,b illustrate the mass increase of the samples at 24 h intervals. The average values indicate that the diatom-treated samples absorbed more water than the control samples during the first two 24 h intervals. In the third and fourth intervals, the absorption levels of both groups converged. Thereafter, the diatom samples exhibited a reduced absorption rate, ultimately absorbing less water than the controls. This decrease may be attributed to a potential pore-blocking effect in the rock matrix, which could hinder further water penetration.

Figure 15 shows the cumulative average water absorption of both the control and diatom-treated samples throughout the experiment. Analysis of the average values reveals that all samples reached saturation after 120 h of immersion. Moreover, at every stage of the test, the diatom-treated samples consistently absorbed greater amounts of water compared with the control samples. After 240 h of immersion, the diatom samples absorbed an average of 14.10 g of water, while the control samples absorbed an average of 11.17 g. Consequently, the diatom-treated samples absorbed approximately 2.93 g more water than the controls.

It can be verified from the average values in

Table 6. Average of the results obtained from mass of water absorbed.

Average Percentage of Water Absorption in Control Samples (%)	Average Percentage of Water Absorption in Samples of Diatoms (%)
3.49	4.68

that, as expected, water absorption in the diatom-treated samples is 1.19% higher than that in the control samples. This result confirms that the diatom-treated samples absorb more water and at a faster rate during the initial 24 and 48 h intervals. Although their absorption rate subsequently slows, it is not sufficient to prevent them from ultimately reaching a higher total water uptake. This behavior aligns with the increased open porosity observed in these samples.

However, it cannot yet be conclusively determined whether diatoms specifically improve or reduce water absorption on the upper face of the treated samples. As in the test above, this procedure involves complete immersion of the samples, preventing the evaluation of individual surfaces in isolation.

3.3. Results of Water Absorption Test by Capillarity

Figure 16 shows the mass of water absorbed by the diatom-treated samples and the control samples during the capillary water absorption test, in which the lateral surfaces were sealed to ensure absorption occurred only through the treated surface.

Analysis of the results up to 432 h, at which point the values stabilized, reveals that the diatom-treated samples consistently exhibit lower water absorption compared with the control samples, with reductions of up to 32% during the first 96 h. As shown in Figure 16, after 432 h of testing, the diatom-treated samples absorbed 5.078 g/m², compared with 5.419 g/m² for the control samples, representing an overall reduction of approximately 7% in total water absorption.

Examining the water uptake at each time interval (Figure 17), it can be observed that, during the first 5 days (140 h), the diatom-treated samples showed a reduction in water uptake at each interval, with decreases of up to 42%. After 140 h, this trend reversed, and the diatom-treated samples began to absorb more water than the control samples at each subsequent interval. This difference gradually diminished, eventually converging with the control samples toward the end of the test.

When analyzing the results of the samples without lateral coating, as shown in Figure 18, it is observed that, at the end of the test, the diatom-treated samples absorbed a greater amount of water per square meter compared with the control samples. This outcome was expected, since previous tests demonstrated that the diatom-treated samples exhibited higher open porosity and, therefore, a greater capacity for water absorption when all surfaces are exposed. This is particularly relevant considering that the lateral faces of the diatom-treated samples did not receive the biotreatment.

It is important to note that, at the beginning of the test, and up to the first 6 h, the diatom-treated samples absorbed slightly less water per square meter than the control samples. However, this effect was reversed in the later stages of the test, ultimately resulting in higher overall absorption for the diatom-treated samples.

Analyzing the behavior in consecutive time intervals, it can be observed in Figure 19 that the absorption of the diatom-treated samples is slightly lower during the initial intervals. However, after 30 min, the water absorption per interval of the diatom-treated samples exceeds that of the control samples for the remainder of the test. Despite this, a marked similarity can be observed in overall behavior between the diatom-treated samples and the control samples. This observation is consistent with the results of the first two tests discussed above, which demonstrated that the four lateral faces of the diatom-treated samples absorb more water than those of the control samples.

This phenomenon can be attributed to the higher open porosity present on the lateral faces of the diatom-treated samples, which allows them to absorb a greater amount of water per square meter. Nevertheless, it is important to highlight that, during the first 6 h of the test, water absorption per square meter of the diatom-treated samples was lower than that of the control samples.

Analyzing these results, it can be determined that samples with isolation applied exclusively to the study face exhibited lower water absorption compared with those without the fiberglass paint coating. This confirms that diatoms predominantly grow on that surface, sealing the pores and hindering water ingress through capillarity.

3.4. Results of Spray Water Absorption Test

Figure 20 shows the average mass of water absorbed by the control and diatom-treated samples during the spray water absorption test across the different time intervals. In the initial moments, the diatom-treated samples absorbed slightly more water than the control samples. However, this behavior was quickly reversed. The diatom-treated samples followed a trend line with a less pronounced slope, absorbing a smaller amount of water per square meter for the remainder of the test, ultimately resulting in 41% less total absorption.

The average mass of water absorbed per unit area was also analyzed for each time interval for both the control and diatom-treated samples, as shown in Figure 21. The trend in water absorption between each spraying cycle for both sample types was downward. This is a logical behavior, indicating that the greater the number of spraying cycles, the less water the samples can absorb until reaching saturation.

It can be observed that, except for the first three spray intervals (3 s, 5 s, and 10 s), in each subsequent 20 s spray, the diatom-treated samples absorbed a smaller and more stable amount of water per surface area compared with the control samples.

As a result of this test, a favorable difference in behavior can be observed for the diatom-treated samples.

3.5. Results of Water Absorption Test by the Pipette Method

Figure 22 shows the volume of water absorbed (in milliliters) as a function of time. Analysis of the data reveals that the diatom-treated samples required more time to absorb each increment of 0.1, 0.2, 0.3, 0.4, and 0.5 mL of water compared with the control samples. This suggests that water penetration into the diatom samples was slower and less efficient, likely due to increased sealing of the rock pores.

Furthermore, as the volume of water absorbed increased, the trend line for the diatom samples diverged progressively from that of the control samples, with its slope becoming steeper. This indicates that, as the test progressed, the diatom samples required significantly more time to absorb each additional 0.1 mL increment compared with the controls.

By the end of the test, the time needed to absorb 0.5 mL of water was 78% longer in the diatom samples than that in the control samples, demonstrating the waterproofing effect generated by the biotreatment.

In addition, volume of water absorbed after 24 h of testing was examined, as illustrated in Figure 23. It can be seen that the samples with diatoms exhibit a 40% lower water absorption compared with that in the control samples.

3.6. Frost Resistance Test Results

Table 7. Percentages of mass loss experienced by the control and diatom samples after freeze–thaw cycles.

Value	Average Control	Average Diatom
% mass loss	−0.42%	−0.35%
Standard deviation	±0.19	±0.042

presents the percentages of mass loss experienced by the control and diatom samples after the freeze–thaw cycles. Data indicate that the control samples exhibited approximately 17% greater mass loss compared with the diatom samples. Note that this difference may be even greater, as traces of nondiatom green algal fibers from the biofilm pool were detected in the water used during the freeze–thaw cycles of the diatom samples.

In addition to determining the percentage of mass loss after 45 freeze–thaw cycles, two water absorption parameters were evaluated. First, the capillary water absorption capacity was measured using a water sheet of 3 ± 1 mm thickness, both before and after the freeze–thaw cycles, as shown in Figure 24a,b. Prior to the freeze–thaw test, the control samples exhibited higher water absorption than the diatom samples. However, after the 45 cycles, the results favored the diatom samples, which absorbed 0.28 g less water after 48 h, compared with the control samples, which absorbed 0.73 g more water. This suggests that the presence of diatoms imparts a protective effect against freeze–thaw damage.

The samples were also subjected to the pipette test after the freeze–thaw process. The results, shown in Figure 25, indicate that the diatom samples took longer than the control samples to absorb each increment of 0.1, 0.2, 0.3, 0.4, and 0.5 mL of water. This suggests that water penetration into the diatom samples was slower and more difficult. Furthermore, as the volume of absorbed water increased, the trend line for the diatom samples progressively diverged from that of the control samples, with a steeper slope. This means that, for each 0.1 mL increment absorbed, the time required by the diatom samples to absorb it increased more than that of the control samples. Over time, water penetration and absorption decreased more significantly in the diatom samples compared with those in the control samples.

Focusing on the results obtained after 24 h, Figure 26 shows that the samples with diatoms absorbed an average of 0.90 mL of water, while the control samples absorbed 1.23 mL. This indicates that, after undergoing 45 freeze–thaw cycles, the diatom samples absorbed 27% less water than the control samples. Additionally, the test results reveal that the mass loss in the diatom samples was greater than in the control samples following these cycles.

This test provides a useful comparison between the initial state before the freeze–thaw cycles and the final state after the cycles for each sample type independently. However, to better compare the control and diatom samples after freeze–thaw exposure, a more targeted test focusing only on the upper surface of the diatom-treated samples is advisable. For this purpose, the water absorption test using the pipette method was selected, with results indicating better performance in the diatom samples. Therefore, it cannot be conclusively stated that the freeze–thaw cycles negatively impacted the antiabsorption properties of the diatom-treated samples.

3.7. Salt Crystallization Resistance Test Results

Table 8. Average percentages of mass loss in control and diatom samples.

Value	Average Control	Average Diatom
% mass loss	−0.06%	−0.12%
Standard deviation	±0.15	±0.08

presents the results for the control and diatom samples after exposure to salt cycling. The average values indicate that the diatom samples experienced a higher mass loss of 0.12% compared with that in the control samples, which showed a mass loss of only 0.06%. Visual analysis of the samples revealed that the most affected specimen was a diatom sample, as this was the only one showing material loss on its upper face. Of note, the other samples exhibited minimal visual alterations. Furthermore, during the cycling process, visual inspection showed that salt crystallization in the control samples occurred on the submerged face, whereas, in the diatom samples, crystallization appeared at the edges, as illustrated in Figure 27. This effectively confirms that diatoms deposit on the upper face, sealing pores and producing a protective effect on the samples.

4. Conclusions

The results of this study show the simplicity of the diatom treatment process, which contrasts with its effect on the limestone samples. The treated samples showed significant differences from the control samples. In particular, the treated samples showed significantly higher levels of open porosity and pore volume, indicating significant structural changes. This increase in porosity is related to an increase in water absorption, which was particularly evident when analyzing the whole samples, in which the lateral surfaces performed poorly due to limited or no biofilm development.

However, concentrating only on the upper surfaces, where most of the diatom growth occurs, the samples treated with this biotreatment showed promising results in terms of water absorption resistance properties. This reduction in absorption by capillary, spray and pipette methods suggests a localized improvement in surface resistance, which could be exploited in specific applications requiring protection of surfaces directly exposed to water.

The analysis of the resistance to freeze–thaw cycles was particularly revealing. The treated samples maintained their structural integrity remarkably well, showing a 27% reduction in water absorption after freeze–thaw cycles compared with the control samples. This behavior indicates increased resistance to extreme conditions, which may translate into superior durability against weathering degradation processes.

Although, in general, a higher pore content is associated with lower frost resistance, actual resistance also depends on the size, distribution, and connectivity of pores. In this study, the experimental samples exhibited a greater number of pores, but these were smaller and less connected, which reduces water absorption and retention. Conversely, the control samples presented larger, more open pores, which facilitated water ingress and explain their lower resistance to freeze–thaw cycles.

Despite the challenges observed, such as the possible degradation effect associated with prolonged immersion of the samples during treatment, the benefits of using diatoms are evident in certain areas. The results in terms of resistance to salt crystallization were inconclusive, possibly due to mass variations caused by direct exposure. This aspect highlights the need for more detailed studies to fully understand their potential impact.

Our study therefore represents a significant advance in the understanding of the effects of diatom treatment on limestone properties. Despite its simplicity, the method used here has proved effective in analyzing these effects with limited effort. Analysis of treated samples has provided relevant information on the behavior of the material and its surface properties. However, further research is needed to mitigate the degradation effects observed and to extend the analyses to other types of stone and under different environmental conditions. This research will provide a more robust basis for the development of this innovative and sustainable biotreatment and improve its applicability in the conservation of this type of material.