Service life assessment of three historical building envelopes constructed using different types of sandstone is presented. At first, experimental measurements of material parameters of sandstones are performed to provide the necessary input data for a subsequent computational analysis. In the second step, the moisture and temperature fields across the studied envelopes are calculated for a representative period of time. The computations are performed using dynamic climatic data as the boundary conditions on the exterior side of building envelope. The climatic data for three characteristic localities are experimentally determined by the Czech Hydrometeorological Institute and contain hourly values of temperature, relative humidity, rainfalls, wind velocity and direction, and sun radiation. Using the measured durability properties of the analyzed sandstones and the calculated numbers of freeze/thaw cycles under different climatic conditions, the service life of the investigated building envelopes is assessed. The obtained results show that the climatic conditions can play a very significant role in the service life assessment of historical buildings, even in the conditions of such a small country as the Czech Republic. In addition, the investigations reveal the importance of the material characteristics of sandstones, in particular the hygric properties, on their service life in a structure.
Cultural heritage is comprised of tangible and intangible parts. It can be considered as an image of national advancement, history, or education which should be passed on future generations. Its preservation is therefore highly valued by each nation. Historical monuments and structures represent a major part of tangible cultural heritage. In Europe, they were often made of natural stones, which were the dominant building material until the industrial revolution in the 19th century allowed manufacturing high quality cement for the production of reinforced concrete [
There are several different processes involved in the stone decay [
As it is often impossible to obtain samples for laboratory studies due to the necessary architectural heritage protection, another approach for assessing the quality of stones is based on the application of noninvasive techniques, such as the ultrasonic methods reported in [
In this paper, we focus on the investigation of hygro-thermo-mechanical performance of three different kinds of sandstones that are presently applied in the renovation of historical monuments in the Czech Republic. Since most old quarries that were used for the extraction of building stones in the past are presently not active, a particular attention should be paid to the application of stones whose properties are as close as possible to the original materials. In that respect, detailed preliminary considerations taking into account the material color, structure, general appearance, and mechanical and other physical properties are supposed to be done, in order to find compatible materials with historically applied stones. One should also take into account the durability properties of building stones, that is, their resistance to the harmful climatic effects. In the past centuries, a plaster or another finishing layer was often missing and the stones formed the external parts of historical buildings and bridges. Therefore, the laboratory measurement of basic physical, mechanical, hygric, thermal properties and durability properties of the analyzed building stones is done, in order to get reliable input data for the subsequent computational modeling of their performance under real climatic conditions corresponding to several different Czech regions.
Three different types of sandstone originating from the quarries in the Czech Republic were analyzed. The particular materials are denoted as SK, SL, and SZ; the quarries locations, age, petrography, and mineralogical composition are given in Table
Overview of studied sandstones.
Notation | SK | SL | SZ |
---|---|---|---|
Quarry location | Kocbeře | Libnava | Záměl |
|
|||
Age | Cretaceous (Cenomanian) | ||
|
|||
Petrography | Glauconitic sandstone; |
Glauconitic sandstone; |
Glauconitic sandstone; |
|
|||
Mineralogical composition | Quartz (83%); |
Quartz (71%); |
Quartz (79%); |
Images of investigated sandstones (SK-SL-SZ).
Microscopy images of investigated sandstones (SK—(a), SL—(b), and SZ—(c)).
The chemical composition of studied sandstones was determined by X-ray fluorescence (XRF) analysis. Table
XRF chemical composition of tested materials.
Substance | SiO2 | Al2O3 | Fe2O3 | CaO | MgO | K2O | P2O5 |
|
---|---|---|---|---|---|---|---|---|
Amount [mass %] | ||||||||
SK | 97.45 | 1.870 | 0.368 | 0.035 | 0.071 | 0.090 | — | 99.887 |
SL | 89.52 | 3.630 | 4.610 | 0.118 | 0.312 | 1.320 | 0.112 | 99.622 |
SZ | 92.92 | 2.710 | 2.420 | 0.173 | 0.533 | 1.080 | 0.008 | 99.844 |
The bulk density, matrix density, and total open porosity were investigated as the basic physical characteristics. Bulk density was measured on the gravimetric principle, using the sample size measured by a digital length meter and its dry mass. For this measurement, 10 cubic samples of side 100 mm were used. The matrix density was determined by helium pycnometry using Pycnomatic ATC (Thermo). At the application of Pycnomatic ATC, a well dried sample of studied material is weighed and placed in a calibrated reference chamber of known volume. Helium is first loaded at known pressure in a calibrated reference chamber and then expanded into the sample chamber. Once the pressure is stabilized, experimental data are collected and the material volume is accessed. The accuracy of the gas volume measurement using this device is ±0.01% from the measured value, whereas the accuracy of used analytical balances is ±0.0001 g. On the basis of bulk density and matrix density measurements, the total open porosity was calculated in a common way [
Mechanical properties of researched stones were characterized by compressive strength and Young’s modulus. The compressive strength was measured on cubic samples of 50 mm side using the standard procedure described in [
Young’s modulus was determined using the pulse ultrasonic method and calculated according to
Freeze/thaw resistance was measured for the investigated materials. It was calculated as a ratio of compressive strength of the frost loaded material and reference material without any freezing/thawing cycles exposure. According to the European standard [
Transport of water in liquid phase was described by moisture diffusivity which was determined as a function of moisture content. For that purpose measurement of moisture profiles was done, using a vertical experimental setup. In the experiment, rod-shaped samples with the dimensions of 20 × 40 × 290 mm were used. Epoxy resin was employed for water- and vapor-proof insulation on the lateral sides to assure 1D moisture transport. In the determination of moisture profiles, the specimens were put in contact with water, whereas the initial state was dry material. After choosing time intervals, the samples were cut into several pieces. The moisture content was then determined in each piece by the gravimetric method. Duration of the suction experiment was 30, 60, and 90 minutes. For the determination of moisture diffusivity, an inverse analysis of experimentally measured moisture profiles was used [
Water vapor transport was characterized by water vapor diffusion permeability
The sorption and desorption isotherms were measured using a DVS-Advantage device (Surface Measurement Systems Ltd.). The instrument measures the uptake and loss of vapor gravimetrically, using highly precise balances having a resolution of 10
Thermal conductivity
Measurement accuracy of ISOMET 2114.
Measurement | Measurement range | Accuracy |
---|---|---|
Thermal conductivity | 0.015–0.700 W/mK | 5% of reading + 0.001 W/mK |
0.7–6.0 W/mK | 10% of reading | |
|
||
Volumetric heat capacity | 4.0·104–4.0·106 J/m3K | 15% of reading + 1·103 J/m3K |
|
||
Temperature | −20–+70°C | 1°C |
The experiments were carried out for both dry and water fully saturated materials at the standard laboratory temperature of 20°C. The particular cubic samples with a side dimension of 150 mm were measured using a surface probe. During the experiment, temperature of probes changed dynamically between approximately 20 and 35°C.
The moisture and heat transport was described by the balance equations formulated in Künzel’s mathematical model [
The calculations were performed using the computer code HMS Transport 1.0 [
A natural stone-based external wall without any external and internal finishes was analyzed. This corresponded to the historical masonry habits, which preferred natural appearance of stones. The thickness of the wall was assumed to be 500 mm. At the solution by the finite-element method, the wall was divided into several segments according to the scheme depicted in Figure
Finite-element scheme of the analyzed external wall.
Dynamic climatic data in the form of test reference years obtained from three meteorological stations in different locations of the Czech Republic were used. These stations are operated by the Czech Hydrometeorological Institute providing official weather data for the Czech Republic. All the weather parameters are gathered every hour for at least thirty years. Only such extensive database can serve as a background for creation of reference year. The data involve long-term hourly average values of temperature, relative humidity, amount of rainfall, wind velocity and direction, and several kinds of sun radiation.
Prague, as one of these locations, represents the capital city of the Czech Republic. It is located approximately in the geographic center of the country and has the average altitude of 250 meters. Prague’s climate is mild with the highest average temperature of 17.6°C in August. The coldest months are December, January, and February with average temperatures between −2.2 and −3.6°C. Annual average amount of rainfalls is 526 mm.
The second location is Holešov in the southeastern part of the Czech Republic. This region is ranked as the warmest region in the country, with the highest average temperature of 19.5°C in August. In winter, only January’s average temperature drops below zero, reaching −2.6°C.
The third locality, Šerák, represents the harshest climatic conditions in the Czech Republic. The meteorological station is located at 1328 meters above the sea level in the mountains of the northeastern part of the country. In a comparison with Holešov, the annual average temperature is almost 6°C lower, while the average relative humidity is about 5% (absolute) higher. A comparison of the temperatures in Šerák and Holešov over a reference year is shown in Figure
A comparison of reference-year temperatures for Šerák and Holešov.
On the interior side of the building envelopes, constant values of temperature 21°C and relative humidity 55% were applied. These values are prescribed in the thermal standard [
Central Europe belongs to the geographical areas which are characterized by an alternation of the freezing and thawing periods. In such climatic conditions, a typical damage caused by the external environment is due to the freeze/thaw cycles in the external surface layers of building structures [
Flowchart of the service life assessment process.
The basic physical parameters of investigated materials are given in Table
Basic physical properties.
Sandstone | Bulk density |
Matrix density |
Total open porosity |
---|---|---|---|
SK | 2 227.7 | 2 653.5 | 16.1 |
SL | 2 191.0 | 2 667.8 | 17.9 |
SZ | 2 075.9 | 2 689.2 | 22.8 |
Mechanical properties of sandstones are summarized in Table
Mechanical properties.
Sandstone | Compressive strength |
Young’s modulus |
---|---|---|
SK | 52.9 | 17.4 |
SL | 60.1 | 22.5 |
SZ | 25.5 | 16.8 |
The freeze/thaw resistance (Table
Freeze/thaw resistance.
Sandstone | Freeze/thaw resistance (70 cycles) |
---|---|
SK | 0.70 |
SL | 0.91 |
SZ | 0.77 |
Moisture diffusivity
Moisture diffusivity versus moisture content functions.
The measured water vapor transport properties are summarized in Tables
Water vapor transport properties, dry cup method.
Sandstone |
|
|
|
---|---|---|---|
SK |
|
|
12.7 |
SL |
|
|
11.6 |
SZ |
|
|
11.8 |
Water vapor transport properties, wet cup method.
Sandstone |
|
|
|
---|---|---|---|
SK |
|
|
7.4 |
SL |
|
|
7.4 |
SZ |
|
|
6.9 |
Sorption and desorption isotherms of studied sandstones are given in Figure
Sorption and desorption isotherms.
The results of the measurement of thermal parameters are given in Table
Thermal parameters.
Sandstone | SK | SL | SZ | |
---|---|---|---|---|
Thermal conductivity |
Dry | 3.53 | 2.71 | 2.10 |
Saturated | 5.21 | 4.64 | 3.88 | |
|
||||
Volumetric heat capacity |
Dry | 1.68 | 1.58 | 1.44 |
Saturated | 2.28 | 2.47 | 2.28 |
As the hygric and thermal properties of building materials used as input parameters of the computational model depend on the state variables such as temperature or moisture content, their values in the particular positions in the building envelope can change gradually, based on cyclic fluctuations of moisture and temperature. Therefore, the results of hygrothermal simulations have to be evaluated after certain period of time which is necessary for the properties to achieve a kind of dynamic equilibrium in a sense that the properties used during one test reference year are very close to those used for the subsequent year. According to the previous experience, the time period of five years is mostly long enough for a development of cyclic hygrothermal behavior of building envelopes which is not established yet in the first years of simulation. Therefore, the results of hygrothermal performance of natural stone-based building envelopes presented in this paper will be related to the fifth year of simulation.
Temperature and moisture fields in the analyzed building envelopes were generated in a form of hourly nodal values of moisture content and temperature. These values were used for the construction of moisture or temperature profiles in the building envelope in any moment of the analyzed time period. Figures
Relative humidity profiles, Šerák’s climatic conditions, January 15, 3:00 a.m.
Temperature profiles, Šerák’s climatic conditions, January 15, 3:00 a.m.
Relative humidity profiles, Holešov’s climatic conditions, July 15, 3:00 p.m.
Temperature profiles, Holešov’s climatic conditions, July 15, 3:00 p.m.
The function of relative humidity of SZ sandstone in point 2 mm under the surface versus relative humidity of air during January under Prague’s climatic conditions is captured in Figure
Relative humidity of air for SZ sandstone under Prague’s climatic conditions, January.
It is obvious that temperature differences between different sandstones were not as significant as the differences in moisture content, represented by relative humidity. The highest temperature differences (3.01°C) were achieved on the interior side between SK and SZ sandstones under Šerák’s climatic conditions, which corresponded to their different thermal parameters (Table
The number of freeze/thaw cycles appearing in the investigated sandstones was determined using a parallel evaluation of their hygric and thermal performance. One freeze/thaw cycle in the selected node of the finite-element mesh was counted only if the moisture content was higher than the maximum hygroscopic moisture content, as expressed by the corresponding value of relative humidity of 97.6%, and the temperature at the same time dropped below zero. These two conditions had to be met at least for six hours followed by at least six-hour lag, which was in compliance with the methodology for experimental determination of frost resistance of natural stones [
In the presentation of obtained results, only representative graphs capturing moisture and temperature versus time functions are given, in order to illustrate the process of counting the freeze/thaw cycles in the three different sandstones subjected to the three different climatic conditions.
Figure
Temperature and moisture content versus time functions, SK sandstone, Prague.
Completely different results were obtained when SL sandstone under Šerák’s climatic conditions was assessed (Figure
Temperature and moisture content versus time functions, SL sandstone, Šerák.
The most convenient conditions for the service life of sandstone building envelopes from the point of view of their freeze/thaw resistance were observed in Holešov. Because of the relatively high temperatures and low relative humidity of air in the test reference year, not a single freeze/thaw cycle was counted. As it is presented in Figure
Temperature and moisture content versus time functions, SZ sandstone, Holešov.
Summary of the freeze/thaw cycles appearance for all investigated variations is presented in Table
Summary of freeze/thaw cycles appearance.
Location/type of sandstone | Freeze/thaw cycles | ||
---|---|---|---|
SK | SL | SZ | |
Prague | 6 | 8 | 0 |
Šerák | 29 | 33 | 7 |
Holešov | 0 | 0 | 0 |
Combining the results of numerical simulation of freeze/thaw cycles appearance with the experimental measurements of freeze/thaw resistance, the assessment of service life was performed. The European standard [
The service life estimates of sandstone building envelopes are summarized in Table
Service life estimate of sandstone building envelopes.
Location/type of sandstone | Service life | ||
---|---|---|---|
SK | SL | SZ | |
Prague | 140 | 399 | Unlimited |
Šerák | 29 | 98 | 165 |
Holešov | Unlimited | Unlimited | Unlimited |
The obtained results of service life of sandstone building envelopes cover a relatively wide range of possibilities, because they take into account the differences both in the type of sandstone and in the dynamic climatic conditions applied on the exterior side of building envelope. On the other hand, they are specific concerning the type of rock in general and the climate is restricted to the conditions of Central Europe. Therefore, a comparison with the results presented by other research groups could be done in a limited extent only. In addition, only few researchers combined the experimental and numerical approaches to estimate the service life. They mostly focused just on the determination of the number of freeze/thaw cycles leading to material disruption [
Although the proposed method for service life assessment is very complex and is comprised of building materials research, experimental analysis of heat and moisture transport and storage parameters, and computational analysis using real dynamic weather data, several limitations still remain, which can affect the accuracy of results. The inaccuracies can be generated while service life is estimated by extrapolation of experimentally measured freeze/thaw resistance and calculated number of freeze/thaw cycles. As the material properties are gradually changed due to the effects of freeze/thaw cycles, it would be necessary, in order to obtain more precise results, to determine input material parameters not only as a function of moisture content or temperature but also as a function of freeze/cycles. Theoretically, it would be possible because the applied mathematical model allows including such material dependency. However, from the practical point of view, it would be very time consuming to measure all the material parameters in dependence on two or three independent parameters. Other inaccuracies can be generated by the fact that original building materials can be usually unavailable for experimental analysis (mostly destructive) because of heritage protection. Therefore, often only similar samples from other quarries can be obtained so that it is difficult to ensure that the material properties are really identical.
A noninvasive method for the service life assessment of building envelopes built of natural stones from the point of view of their freeze/thaw resistance was introduced in this paper. Within the frame of this study, three different types of sandstone were investigated under three different climatic conditions in the Czech Republic.
The presented method is based on a combination of experimental analysis and numerical simulation where the experimental measurement provides the material properties of the studied stones as the necessary input parameters for the numerical simulation of hygrothermal performance. The service life of building envelopes is then estimated using the results of hygrothermal simulations and the experimentally determined durability properties.
The results obtained in this paper indicated that hygric parameters of natural stones had the highest influence on the service life of the analyzed building envelopes. It was found that a high value of moisture diffusivity (>10−6 m2/s) of sandstone together with a low value of water vapor diffusion resistance factor substantially contributed to an improvement of freeze/thaw resistance. Therefore, the best performance was achieved for the SZ sandstone with the most favorable combination of hygric properties. On the other hand, SK sandstone provided the worst results.
The service life of the investigated sandstones was very variable and depended on both climatic conditions and the properties of stones. It lied between 29 and 399 years; for certain combinations of climate and stone it was even unlimited. These results underline the necessity to take the service life of natural materials with a great care and investigate it case by case.
The authors declare that there is no conflict of interests regarding the publication of this paper.
This research has been supported by the Ministry of Culture of the Czech Republic under Project no. DF12P01OVV030.