1 Introduction

1.1 Research background

The preservation of buildings of high historical value is in the public interest due to their cultural value. Such buildings are often several hundred years old, and during their operation, they are subjected to various impacts that contribute to the degradation of their structure. To slow down the destruction processes, it is necessary to undertake various types of protective and restoration measures. Such activities should be preceded by tests carried out on the building. The improper selection of protective treatments and the undertaking of substitute actions often worsen the situation and result in more frequent interventions in the future.

In the case of buildings, the walls of which are built of clay solid bricks, the destructive processes caused by the adverse effects of the natural environment are largely related to the presence of excessive moisture in the pores of the ceramic material. Investigations of the causes and the degree of moisture of such excessively damp and saline brick walls are carried out, but it seems that they should be accompanied by analyses of ceramic porosity, which in turn are performed very rarely. Visual inspections of local buildings erected in different historical periods show that brick walls have survived to the present times in very different conditions. Some of these walls, despite the moisture and salinity found in them, bear visible traces of frost destruction, while others have been preserved to the present day in a good condition. The conducted investigations most often aim to assess the state of preservation of historical bricks, and to determine the properties of the repair bricks that will be used for the conservation work. Moreover, the knowledge of the structure of porosity could also be useful, for example, in the process of selecting a chemical agent that can be used to make (using injection) horizontal damp-proof membranes to prevent the capillary rise of water from the ground to the wall. These agents are currently used indiscriminately, i.e., without linking their mode of action (e.g., hydrophobizing capillary walls, agents that narrow the diameter of capillaries, etc. [1, 2]) with the structure of porosity, which does not guarantee the effectiveness of the membrane that was made using them.

1.2 Material characteristics

In the past, clay solid brick was a basic building material, which, like stone, was used to erect masonry buildings. This traditional building ceramic is still produced and used, but with the passage of time, the technology of its production has been improved [3]. It is made by forming a clay–sand molding mass, which is then subjected to a high-temperature firing process. This material is classified as being capillary porous, which means that with a high apparent density of about 1.8 g/cm3, it is characterized by a relatively high open porosity of about 20–30% [4]. This in turn means that it is susceptible to wetting with water.

Clay solid brick is considered to be one of the most durable building materials. However, if such brick, after being embedded in walls, becomes excessively damp for a long time, it also starts being susceptible to destruction caused by the adverse effects of the natural environment. The physical impact of water absorbed by ceramics, in the form of long-term cyclical freezing and thawing, is of fundamental importance with regards to the progress of destruction [5,6,7]. The absorption of water takes place primarily from the ground. It results from a direct and long-term contact of a wall devoid of anti-moisture insulation (which was not used in the past) with the ground. Rainwater, together with pressurized wind gusts, is another important factor responsible for the brick building facades water penetration [8, 9].

The mechanism of capillary rise, which is known for capillary-porous materials and is described for example in Refs. [10, 11], is responsible for the transport of water in brickwork. Both components of masonry, clay brick and mortar, usually lime in historic buildings, are capillary-porous materials. Both of these masonry components are involved in water transport, with clay brick being the main building block and accounting for about 80% of the masonry volume. Together with water, salts that are soluble in it are also transported inside the wall, the sources of which are, e.g., fertilizers, de-icing agents, and acid rain. These salts crystallize in the pores located in the near-surface zone during drying. When the crystallization pressure in the pores exceeds the tensile strength of the brick, the structure is destroyed. This is a chemical interaction, the mechanism of which is described, among others, in papers [12,13,14,15]. Such a mechanism intensifies the destruction caused by any physical impact. Moreover, damaged and excessively moist clay solid brick, as a result of the described physical and chemical interactions, become susceptible to the action of biological factors that are associated with the development of various microorganisms [7, 16,17,18] (Fig. 1). Therefore, in natural conditions, environmental impacts stimulated by excessive moisture are synergistic.

Fig. 1
figure 1

Effects of physical (a), chemical (b), and biological (c) interactions on an excessively moist clay solid brick

Full size image

1.3 Review of literature

An important parameter that characterizes building ceramics, and which largely determines its durability and resistance to physical, chemical, and biological destructive influences, is porosity. This porosity is defined as the ratio of the volume occupied by pores to the total volume of the porous material. The pores, which are voids of various shapes and sizes (large, medium, and small) in the solid material, are largely interconnected and form an extensive and irregular structure. However, when analyzing porosity, the ratio of open pores to closed pores should also be taken into account. As can be concluded from Refs. [5, 19, 20], the initial distribution of pores in terms of their size in building ceramics depends both on the composition of the molding mass (in addition to clay, the presence of various types of mineral and non-mineral inclusions is also very important), the molding process, and the firing temperature. When analyzing the porosity, one should not only determine the pore structure in terms of their size, but also the ratio of open pores to closed pores.

Among the researchers dealing with the subject, there is a consensus that the presence of large pores in ceramic material, i.e., with diameters greater than 3.0 µm, is beneficial for the durability of this material [21,22,23,24,25]. These pores are kinds of chambers that compensate for the stresses that occur during the freezing of water and the crystallization of salts. This is due to the fact that the pressure exerted on the walls of large pores is lower than in the case of pores with a smaller diameter. Medium pores, i.e., with diameters ranging from 3.0 to 0.1 µm, are of the greatest importance for the durability of ceramics, and are therefore considered critical. On the other hand, small pores, i.e., with diameters below 0.1 µm, are considered to be of less importance [6]. This is due to the fact that, in these pores, water freezes at a lower temperature than in medium pores and this is well below 0 °C.

The available literature lacks information concerning what proportions of different sizes of pores that are present in building ceramics—large, medium, and small—can be considered beneficial from the point of view of durability. The tests described in Ref. [6] only show that if the ratio of the content of pores with diameters of 3–10 µm to the content of pores with diameters of 0–10 µm is over 70%, then the building ceramics can be considered to be resistant to frost destruction. This paper also shows that if the share of the binder (clay) in the ceramic composition is at a level of 50% or more, then such a ceramic material is resistant to physical influences.

It should also be noted that the research concerning the structure of building ceramics, which is described in the available literature, does not generally apply to a material that is for a long time excessively moist and simultaneously exposed to environmental conditions. For example, the study described in Ref. [7], despite the fact that it concerns the effect of water on a historical brick wall, was carried out on small samples that were made according to traditional construction techniques in a laboratory. A certain exception to this rule is Ref. [26], which concerns low-density ceramics. Other studies described in the available literature are case studies of selected buildings, which aimed to assess the state of preservation of historical bricks. Moreover, they also tried to determine the properties of the repair bricks that will later be used in conservation works (e.g., [5, 27, 28]), or the relationship between the properties of the bricks and the type of frost damage. These concern brick samples taken from buildings and tested in the laboratory [6].

1.4 Formulating the problem and research goal

To fill the indicated research gap concerning the lack of research in the literature on the structure of clay solid bricks from different historical periods that have been subjected to long-term, excessive moisture and exposed to environmental conditions at the same time, it seems reasonable to quantitatively know at least some of the parameters that describe the structure of such bricks. Such knowledge can be the basis for a scientific discussion on the state of preservation of the bricks in question, for drawing conclusions about their further durability, and to make conscious decisions regarding protective and renovation activities.

This paper aims to enrich the current state of knowledge with the results of preliminary research and comparative analyses of the structure of clay solid bricks from historical objects that are erected in Poland in various historical periods and which have been exploited in natural conditions for about 100 to even more than 700 years. The origin of the tested material from various objects and historical periods distinguishes the research described in this paper from other studies in which the tested material usually came from one object.

2 Characteristics of the historical objects from which the brick samples were taken

Samples of clay solid bricks, which were intended for the structural tests, were taken from the walls of six historical buildings, the concise characteristics of which are presented in Table 1. The oldest of the objects comes from 1300, while the youngest from 1910. The table also indicates the elements of the buildings from which the samples were taken, as well as the designations adopted for the samples. The location of the buildings from which the samples were taken is shown using the map of Poland in Fig. 2.

Table 1. Basic information about the historical buildings and masonry parts of these buildings from which clay solid brick samples were taken for testing, together with their designation
Full size table
Fig. 2
figure 2

Location of the buildings from which the brick samples were taken for testing

Full size image

With regard to the samples taken for testing, it should be clarified that it was not possible to collect entire clay solid bricks for testing due to conservation restrictions, and therefore only fragments were collected. Samples were taken at random from sections of the wall agreed with the conservation service. Brick fragments were excavated from the walls using a chisel and hammer. After collecting the samples, they were stored in closed glass containers in the laboratory, at an air temperature of 20 ± 2 °C and a relative air humidity of 65 ± 5%. All samples were taken in rain-free weather with an air temperature between 15 and 20 °C, from the east or north walls of the lowest floors of the buildings. In the case of buildings 1–4, the wall surface was not plastered or waterproofed in the past or present. In the case of buildings 5–6, there was plaster on the walls in the past and present; therefore, the samples were taken in areas of missing plaster. All samples were taken from bricks that showed no surface loss. They were taken from the middle zone of the bricks, from a depth of about 5 to about 10 cm from the face of the bricks.

In five of the mentioned objects, the clay solid brick outer walls of the lowest floor partially sunken into the ground were definitely characterized by excessive moisture and salinity, which was determined during the research. The complete results of these studies were published in Ref. [15], whereas Table 1 only provides the general data.

To demonstrate the comparability of the studied material, the authors also attempted to refer to contemporary standards. Thus, in accordance with Ref. [31], the tested bricks could be assigned to the group of HD masonry elements, while the environmental conditions affecting the buildings from which the bricks were taken could be assigned to exposure class MX3.1, according to Ref. [32].

3 Research methodology

The samples marked in Table 1 with numbers from 1 to 6 were subjected to the following structural tests:

  • Scanning electron microscope (SEM) imaging performed using a JEOL JSM-6610A scanning electron microscope (Fig. 3a), which has a backscattered electron (BSE) detector, a current of 40 nA, an accelerating voltage of 20 kV, and a working distance of 10 mm. The imaging was made using 40-fold magnification, which according to the authors is sufficient for comparative analysis [33, 34],

  • Energy dispersive X-ray spectroscopy (EDS) analysis performed with a JEOL JED-2300 detector that had a counting speed of about 6000 cps. Quantitative analyses were performed using the integrated ZAF method, taking into account corrections for the atomic number (Z) effect, absorption (A) effect, and fluorescence excitation (F) effect of X-ray radiation,

  • Porosimetry analysis using a Micrometrics Autopore IV 9510 mercury porosimeter (Fig. 3b) that was connected to a computer with data processing software.

Fig. 3
figure 3

Scheme and photograph of JEOL JSM-6610A microscope (a)—reprinted with permission from [36], photography of Micrometrics Autopore IV 9510 mercury porosimeter (b), and penetrometer diagram (c)—adapted from Ref. [37]

Full size image

The samples intended for the SEM–EDS scanning tests were embedded in epoxy resin using the vacuum impregnation method, and then ground and mechanically polished with the use of a diamond paste (to a granulation of 1 μm). As some of the samples were very brittle, they were all re-stabilized with epoxy resin (on the polished side under vacuum conditions) to strengthen their cross-section. The conducted procedure was analogous to the methodology described in Ref. [35]. To obtain electrical conductivity and to avoid electrostatic charging, a conductive track made of copper tape was applied to the samples, with everything then being covered with a layer of 40 nm thick graphite coating in a high-vacuum thermal evaporator. The area subjected to the SEM–EDS analyses was 35 mm2 in total for each sample.

The samples for the porosimetry analysis were prepared by crushing large fragments of bricks into 2–3 mm granules. The granules were homogenized, and a representative sample weighing about 2 g was taken from them. The granules that were combined with the mortar were omitted. Each sample was initially dried in air at 110 °C for 1 h, and after being placed in the penetrometer (Fig. 3c), degassing was repeated at 25 °C for half an hour at a pressure of 50 μmHg. The penetrometer with the sample was filled with mercury at a pressure of 0.01 MPa. Based on the known volume of mercury in the empty penetrometer and the one containing the sample, its apparent density was calculated. Then gradually the mercury pressure was increased to 414 MPa, allowing it to penetrate the porous structure of the sample (intrusion step). The specific density of sample was calculated at this step of analysis. Next, the mercury pressure was gradually reduced to 0.1 MPa, which resulted in partial removal of mercury from the porous structure (extrusion step). The volume of the pores corresponds to the volume of mercury filled them. The current volume of mercury in the pores is related to its level in the stem of the penetrometer. According to Washburn’s formula, the pressure range used (from 0.01 to 414 MPa) corresponds to pore widths from 150 μm to 3.5 nm.

4 Research results and discussion

The analysis of the results obtained on the basis of the SEM scanning tests of clay solid brick samples 1–6 indicates significant differences in their microstructure (Fig. 4). These results allow for the formulation of the statement that with an increase in the age of the brick, the homogeneity of its structure decreases.

Fig. 4
figure 4

Exemplary images of the structure of the clay solid brick samples from the years 1300–1910 in 40-fold magnification: a sample 1; b sample 2; c sample 3; d sample 4; e sample 5; f sample 6

Full size image

Therefore, in samples 1 and 2, the layering of the ceramic composite and the unidirectional arrangement of elongated air voids of various widths are clearly visible. These samples also contain oval air voids of various sizes, as well as numerous large grains of the clay fraction. In samples 3 and 4, structural layering is not observed, and there are a few elongated air voids. However, large grains of the clay fraction and oval air cavities are still numerous. The homogeneity of the structure of samples 5 and 6 is much better than that of samples 3 and 4, and the size of grains and air voids is also smaller (especially in sample 6). The increasing homogeneity of the brick structure observed in subsequent samples can be explained by the increasingly better homogenization of the clay–sand mixture and the progressive improvement of the brick production process over the years. No mineral or other inclusions were found in the tested samples. It should be noted that the above analysis was carried out on the basis of preliminary tests involving only single samples. To confirm the results obtained, a larger number of samples should be tested.

In turn, Table 2 contains the results of the chemical composition of samples 1–6, with the calculated measurement absolute error being provided. Figure 5 shows an example of the EDS spectral analysis for these samples. Such analysis was carried out to determine whether the material collected was similar in this respect and suitable for further comparisons.

Table 2 Chemical composition of samples 1–6
Full size table
Fig. 5
figure 5

EDS spectral analysis for: a sample 1; b sample 2; c sample 3; d sample 4; e sample 5; f sample 6

Full size image

As can be seen in Table 2, the content of Si in samples 1–6 is high and exceeds 72%. It is the highest in sample 1, which was taken from a brick dating back to 1300, and amounts to 77.47%. In turn, the lowest value of Si (72.15%) can be found in sample 5, which was taken from a clay solid brick dating back to around 1860. The highest contents of Mg, Al, and Fe (amounting to 1.59, 12.68, and 8.91%, respectively) are observed in samples 5 and 6 from the youngest bricks that date back to the nineteenth and twentieth centuries, with the lowest contents (amounting to 1.12, 8.41, and 6.98%, respectively) being seen in samples 1 and 2 from bricks that date back to the fourteenth and fifteenth centuries. The opposite situation can be seen in the case of K. The highest content of this element (5.92%) is found in sample 2, and the lowest (3.24%) in sample 6. In general, the research conducted shows that the chemical composition of all the samples is quite similar, despite the fact that the material that was used to make the bricks comes from different historical periods and different clay deposits—probably one located in Central Poland and another one in southern Poland. As the chemical composition of all samples is similar and no mineral or other inclusions were found, it can be assumed that the tested material is suitable for further comparisons. There is no side variable in the form of a chemical compound that may affect the susceptibility to frost damage.

Tables 3 and 4, and also Fig. 6, show the test results of samples 1–6 obtained using the mercury porosimetry method. Table 3 contains a summary of the following averaged parameters that characterize the samples: apparent and specific density, total intrusion volume, total porosity, and the median of the pores’ width. The apparent density is the ratio of the mass of the material to its volume, including pores, while the specific density is the ratio of the mass of the material to its volume without pores. The volume of total intrusion is the volume of mercury injected into the sample at a given pressure (for the tests in question, the maximum pressure was 414 MPa). And the total porosity is the ratio of the difference in specific and apparent density to the specific density. In turn, Table 4 contains the percentage of pore width according to the adopted ranges: 100–200 µm, 6–100 µm, 3–6 µm, 1–3 µm, and below 1 µm. Figure 6 supplements the results presented in Tables 3 and 4, and contains graphs of the pores’ width distribution within the range from 200 µm to 3 nm for the individual tested samples.

Table 3 List of the averaged parameters that characterize samples 1–6
Full size table
Table 4 Percentage share of the pores’ width in samples 1–6 according to the ranges adopted by the authors
Full size table
Fig. 6
figure 6

Charts of the pores’ width distribution in the tested samples 1–6 within the range from 200 µm to 3 nm

Full size image

The analysis of the test results presented in Table 3 shows that:

  • Sample 4 has the highest apparent density of 1.878 g/cm3

  • The total porosity of all the tested samples is high and close to 30%; the highest total porosity has sample 3, and is equal to 31.1%, and the lowest total porosity has sample 4, and is equal to 26.7%

  • The median of the pores’ width in the tested samples is also very diverse: the largest, amounting to 15.10 µm, is observed in the case of sample 4, and the smallest, amounting to 0.88 µm, is observed in the case of sample 3

In turn, when analyzing the test results presented in Table 4, large disproportions in the percentage share of the pores’ width within the individual ranges can be observed in the tested samples. For example, the percentage share of pores with a width of less than 1 µm and pores within the range of 1–3 µm is the smallest in sample 4 (amounting to 3.5 and 6.3%, respectively). The highest percentage share of these pores is in sample 3 (amounting to 54.7 and 32.0%, respectively).

As already mentioned, according to papers [6, 21,22,23,24,25], pores with diameters greater than 3 µm are considered to be advantageous, whereas pores with diameters ranging from 3.0 to 0.1 µm are crucial due to the destructive impact of the natural environment on building ceramics (see Fig. 6). Therefore, to compare the test results obtained for the individual samples, and which are presented in Table 4, the following calculation was performed, the result of which for the purposes of this work was tentatively called the P-index:

$$P{ - }index = \frac{{P_{0 - 3\mu m} }}{{P_{0 - 200\mu m} }}$$
(1)

where P0–3μm—the total percentage share of pores with a width of up to 3 µm (inclusive) in the tested structure, P0–200μm—the total percentage share of pores with a width from 0 to 200 µm in the tested structure.

It should be noted that the P-index can take a value between 0 and 1 (1 when all pores are up to 3.0 µm).

The values calculated for samples 1–6 according to Formula (1) are presented graphically in Fig. 7.

Fig. 7
figure 7

P-index values calculated for samples 1–6

Full size image

Figure 7 shows that the lowest value of the P-index (0.098) is found in sample 4, and the highest (0.867) in sample 3. As can be seen in Table 1, both samples were taken from clay solid brick that was taken from a heavily damp and highly saline wall, with the difference being that the wall from which sample 4 came from did not show signs of frost destruction, while the wall from which sample 3 was taken showed clear signs of frost destruction. Therefore, when comparing the information concerning the technical condition of the wall from which the ceramic material was taken with the values of the P-index calculated for individual brick samples, it can be assumed that the lower the P-index value, the lower the susceptibility of the clay solid brick to destruction caused by the natural environment (and vice versa). At present, it is not possible to define the P-index value limit above which damage resistance can be considered insufficient. It is also difficult to match the qualitative assessment indicators to the individual P-index values. It seems that this will be possible once more data are acquired.

The value of the P-index for the other samples (1, 2, 5, 6) is rather high, ranging from 0.529 to 0.705, so it can be assumed that the bricks from which these samples were taken are susceptible to frost destruction. For such destruction to occur, however, an additional condition must be met, namely the bricks must be heavily wet and saline. This is because salts increase the negative effects of excessive moisture on the mechanical strength of the masonry and, as a result of the crystallization and hydration processes in the near-surface zone of the masonry and on its surface, they contribute to the bursting of the masonry structure. Thus, as can be seen in Table 1, the masonry from which samples 1 and 5 were taken was very wet and saline, and therefore showed clear signs of frost destruction. On the other hand, the degree of moisture and salinity in the masonry from which samples 2 and 6 were taken was low, so these masonry walls did not show signs of frost destruction.

The conducted analyses show that the structure of sample 4, which represents clay solid brick from 1500 from the buttress of the outer wall of the post-Cistercian cloister, has the best parameters among all the tested samples. It has the highest apparent density, almost the smallest total porosity, by far the highest median of the pores’ width, and the lowest value of the P-index. Therefore, the brick that was built several hundred years ago into the buttress of the outer wall, despite its long-term excessive moisture and salinity, has survived in good condition until now.

In turn, the structure of sample 3, representing clay solid brick from 1400 from the external wall of a post-Cistercian monastery, is diametrically different. It is characterized by the highest total porosity, the smallest pore width median, and the highest P-index value. This specific porosity may explain the observed visible destruction of the brick wall, but may also be a result of environmental factors. To eliminate this uncertainty, additional tests should be performed on brick taken from the same building, but which is not damp and not saline. However, at the moment, it is not possible to collect an additional sample from the building. The tested material was obtained during the testing of moisture, which was carried out in the building some time ago and which has already been completed. As indicated in the introduction of this paper, taking samples from an object under conservation protection is possible only after obtaining the prior consent of the conservation services, which is not justified due to the already completed research on the discussed object. Bearing in mind the research results described in this paper, the need for such a comparison in future is clearly visible.

Another limitation of the study is undoubtedly the small number of tested samples. To confirm or exclude the existence of the observed patterns, future studies should be carried out on a larger number of samples taken from a larger number of buildings. Perhaps the scope of the study should also be extended and include, for instance, the testing of the mortar, which, representing approximately 25% of the volume of the masonry, may also affect its durability in terms of microstructural damage to the bricks. Bearing in mind all the limitations mentioned above, the presented research should be treated as preliminary.

5 Summary

In general, literature lacks the results of broader studies concerning the structure of clay solid bricks that are subjected to long-term exposure to environmental conditions for several hundred years. According to the authors of this research, such studies may not only be useful for cognitive and comparative purposes, but also for scientific discussion on the existing longevity of historical brick buildings, as well as for forecasting their further durability. Moreover, such results can be very useful for taking more effective protective and restoration measures, e.g., aiming to protect buildings against the capillary action of water from the ground.

The presented results of the experimental studies and comparative analyses of the structure of samples of clay solid bricks from six historical buildings erected in Poland in the years 1300–1910, the collection of which was permitted by conservation services, have allowed to draw preliminary conclusions on the susceptibility to destructive processes caused by environmental impacts. The energy dispersive X-ray spectroscopy showed that the chemical composition of all samples was quite similar and the material collected was similar in this respect and suitable for further comparisons. The test results obtained with a scanning electron microscope and mercury porosimetry showed that as the age of the brick increases, the homogeneity of its structure decreases, and in the tested samples, large disproportions in the percentage of pore width in individual ranges can be observed. On the basis of these results, a P-index based on the knowledge of the share of the pores’ width within the adopted ranges, that can be interpreted as the vulnerability of clay solid bricks to environmental damage, has been tentatively proposed. It should be noted that the presented study has many limitations, which have been mentioned; therefore, it should be treated as preliminary.