Influencia de la textura, del sistema poroso y del acabado superficial en la durabilidad de areniscas y travertino explotados en andalucía y utilizados en construcción

  1. MOLINA PIERNAS, EDUARDO
Dirigida por:
  1. Giuseppe Cultrone Director/a

Universidad de defensa: Universidad de Granada

Fecha de defensa: 12 de junio de 2015

Tribunal:
  1. José Francisco Rodríguez Gordillo Presidente/a
  2. Carlos Rodríguez Navarro Secretario/a
  3. Javier Martínez Martínez Vocal
  4. Paolo Mazzoleni Vocal
  5. María Ángeles García del Cura Vocal

Tipo: Tesis

Resumen

Influence of texture, porous system and surface finish on the durability of Andalusian sandstones and travertine used as building materials Introduction and objectives In recent decades there has been increasing interest in the intrinsic properties and the behavior of traditional construction materials. These studies have tried to identify the factors that contribute to the decay of these materials, which in many cases have resulted in serious economic losses and damage to our artistic heritage. One of these traditional building materials is natural stone, which has played a very important part in the development of different cultures due to its many advantages, the most important of which are its relative abundance and the fact that in many cases it is more resistant to decay than other building materials. The durability of stone depends on a wide variety of factors such as external atmospheric agents and biological and anthropic attacks as well as the intrinsic characteristics of the stones themselves and the mechanical stresses that they have to withstand once laid in the building. Nowadays there are a wide range of analytical techniques and standardized tests with which scientists can characterize natural stone. By comparing the results obtained with these different techniques, we obtain a full and very detailed picture of the characteristics and the factors that influence the durability of a rock. In this way, scientific and technical research can provide very useful information for professionals working in a range of different fields, such as construction, conservation and restoration. It can also be used to improve the quality of the products produced by stone companies, so helping create brands with a high reputation in the construction sector. Sedimentary rocks have a wide variety of functions as construction materials, ranging from their use as load-bearing masonry to other more delicate decorative aspects of building work. A large number of quarries produce this kind of rock, which makes them of great interest from both an economic and an artistic or monumental point of view. In this Thesis we have selected four kinds of sedimentary rock used in the construction of historic and more recent buildings. The main objective is to identify the different factors that influence the quality of these four rocks as building materials. To this end we have: 1. Related the petrographical and physical characteristics of the rock with its surface properties, especially those referring to porosity. 2. Assessed the influence of the surface finishes on the decay process of each type of rock. 3. Assessed the importance of texture and surface finish in the durability of rocks using accelerated aging tests. 4. Analyzed the behavior of rocks when exposed to real decay conditions in the city of Granada after exposure to the elements on an open roof terrace over a period of four and a half years. 5. Studied the changes in the porous system caused by different decay agents. Our ultimate objective is to obtain quality criteria for these sedimentary rocks when used for construction and ornamental purposes. Indeed one of the most efficient means of competing with ornamental stones from other countries is to create prestigious brands with a reputation for high quality. Structure of the Thesis This Thesis is divided into six parts, each of which is made up of one or more chapters. First part (Chapter 1) is an introduction about the use of sedimentary rocks in construction and the research that has been done into their physical behavior and durability. Second part contains a description of each variety of rock and its particular geological context (Chapter II) and the methodology and analytical techniques used to carry out this research (Chapter III). Third part studies the volume (Chapter IV) and surface (Chapter V) properties of the four types of rock in order to identify the intrinsic properties that influence their durability. In Fourth part (Chapter VI) we perform accelerated aging tests and other tests involving exposure to the elements in order to analyze the processes that cause decay. In Fifth part (Chapter VII) we assess the effectiveness of a phosphate-based consolidant in these rocks. Finally Sixth part (Chapter VIII) offers the general conclusions to this Thesis, suggestions as to possible uses for these rocks and future lines of research. The results of this Thesis have been partially published in the following journals: - The pore system of sedimentary rocks as a key factor in the durability of building materials. Engineering Geology 118, 110-121 (2011). - 3D analysis of the porous system of stone building materials using X-ray computerized axial tomography. Science and Technology for Cultural Heritage 8, 9-16 (2011). - Evaluation of the petrophysical properties of sedimentary building stones in order to establish quality criteria. Constructions and Building Materials 41, 868-878 (2013). - Evaluation of stone durability using a combination of ultrasound, mechanical and accelerated ageing tests. Journal of Geophysical and Engineering 10, 035003, 1-18 (2013). - The influence of rock fabric in the durability of two sandstones used in the Andalusian Architectural Heritage (Montoro and Ronda, Spain). Engineering Geology (second review). Materials and geological setting of the quarries Three sandstones and one travertine were selected for this Thesis. Three of the four varieties (Santa Pudia Calcarenite, Arenisca Ronda and Molinaza Roja) have been quarried for at least five centuries and have been used in the building and decorative ornamentation of various monuments in our architectural heritage. Quarrying of Travertino Amarillo only began more recently and it can therefore only be found in modern buildings. The four materials we chose are all porous sedimentary rocks that have more or less rough surfaces which influence their behavior when exposed to decay agents. Therefore as well as investigating their mineral composition, texture and porous system, in each type of rock we have applied different surface finishes (those typically applied in the workshop or factory) in order to carry out the petrophysical characterization of both the rock in block form and of its surface parameters. The following surface finishes were analyzed: "saw-cut (S)", "pumice-polished (P)" and "bush-hammered (J)". Santa Pudia Calcarenite (SP) is a Tortonian bioclastic calcarenite (fig. II-1) from Escuzar (Granada, fig. II-5) and is basically composed of calcitic bioclasts (foraminifera, red algae, echinoderms, bryozoans, bivalves and serpulids) and a small amount of siliceous fragments of about 0.5 mm in size. The rock fabric is clast-supported (fossil fragments make up over 95%) with little cohesion, while the matrix is microsparitic with sporadic amounts of cement. Porosity varies between 30% and 36%. Most of the larger pores are intergranular (over 0.8 - 1cm), while the smaller ones are the result of the intraclastic porosity associated with skeletal fragments. The Santa Pudia Calcarenite (SP, fig.II-5) quarries are a few kilometers south of the village of Escúzar in the province of Granada, Spain (Fig. II-6). In geological terms, these materials belong to a calcarenite unit that dates from the Lower Tortonian. This unit outcrops in the southern and eastern edge of the neogene basin of Granada developed on the collision zone between the Internal and External Zones of the Betic Cordillera. These are platform areas with abundant supplies of calcarean limestones and calcarenites, fundamentally bioclastic, with on-lap type structures on the metamorphic materials from the Internal Zones due to a transgression in the Tortonian. These sediments were deposited in coastal areas in which different resedimentation processes took place that fixed some of the properties of the material. Travertino Amarillo (TA) from Albox (Almeria, Fig. II-5) is a Pleistocene travertine. It has alternating yellow and brown bands due to the presence of iron and manganese oxides/hydroxides. The yellow bands are characterized by macrocrystals of calcite, in some cases with fibrous morphology Fig. II-2). The pores are usually elongated in shape and they can be quite large (over1 - 2 cm). Brown bands are composed of calcitic microcrystals and the pores are normally small and rounded. Travertino Amarillo (TA, Fig. II-5) is quarried near the estate known as Cortijo de Los Marcelinos (Fig. II-8), north of Albox (Almeria). These travertine rocks are part of an extensive area of Pleistocene materials which stretches north of the depression of Albox in the Internal Betics geological area. This is a line of travertine masses running WNW-ESE which is brusquely interrupted on its western and eastern boundaries and is related with paleo-upwellings associated with faults. The Arenisca Ronda (AR) from Ronda (Malaga, Fig. II-5) is a carbonatesandstone with a mean clast size of around 1-2 mm (Fig. II-3). It is a light pink-whitish stone, the rock fabric is massive and only in some cases can clast orientation be detected.Arenisca Ronda is quarried (AR, fig. II-5) about 10 km ENE of the city of Ronda in the province of Malaga (Fig.II-10). The Ronda Basin is one of the largest piggyback basins in the Western Betics. It is located over the north-western Subbetic Units with a continuous structure, the Subbetic Chaotic Complexes and the Flysch Units. The Setenil Formation on the top is Late Tortonian to Late Messinian in age and crops out in most of the depression. This formation is divided into two members, the limestone member and the calcarenite member, in which AR is quarried. The Molinaza Roja (MR) is a sandstone quarried near the town of Montoro (Cordoba, fig. III-5). The color of the stone varies from red to purple (fig. II-4), due to the alternating lamination rich in iron oxides and hydroxides. The clasts are less than 1 mm in size and the pores are difficult to distinguish to the naked eye. Molinaza Roja is quarried 2-3 km north-east of Montoro (Córdoba) (MR, Fig. II-5), by companies from Montoro and nearby Adamuz. This material is a red continental sandstone deposited in fluvial and alluvial environments during the Buntsandstein that can be classified as Red Bed facies. They are laid out in alternating strata between clays and sandstones. The arkose layers are of higher quality for use in construction than the greywackes layers, which is why we chose the sandstone from the arkose layers for our analysis. Analytical methods Chemical, mineralogical and petrographic characterization - Chemical analysis: performed using x-ray fluorescence (XRF) using a S4 Pioneer Bruker AXS spectrometer. -Mineralogical analyses: performed by means of x-ray diffraction (XRD) using an X'Pert PRO system. The results were interpreted with the XPowder© software (Martín Ramos, 2004) and the total fraction and the clay fraction were analyzed using the disoriented powder method and the oriented aggregates method respectively. -Petrographic study: this was conducted using polarized optical microscopy (MOP, Carl Zeiss Jenapol-U equipped with a Nikon D7000 digital microphotography system) and scanning electron microscopy (FESEM, LEO GEMINI-1530 and ESEM, Quanta 400; both equipped with an EDX microanalysis system). Petrophysical parameters - Porous system: we assessed the open and total porosity, the pore size distribution and the specific surface area using mercury injection porosimetry (MIP, Micromeritics Autopore III model 9410), a helium pycnometer (PHe, Micromeritics AccuPyc 1330 equipment), the absorption of nitrogen (N2, Micromeritics Tristar 3000 and Micromeritics Flowprep systems) and X-ray computerized microtomography (µCT, equipment from the University of Ghent). -Hydric behavior: we calculated the coefficients for water absorption at atmospheric pressure and for forced water absorption in a vacuum, the degree of pore interconnection, the desorption rate, capillarity and capillary rise, continuous capillarity, the degree of saturation, the apparent and real density and the open porosity. Finally in order to characterize the hydric behavior according to the different surface finishes, we performed water drop adsorption (WDA) and water vapor permeability (KV) tests. Mechanics and dynamics properties - Compressive strength and flexural strength tests under a concentrated load using an Instron 4411 hydraulic press. - Abrasion and anti-slip resistance: for the abrasion resistance test we used an Ibertest device with a corundum abrasive and the anti-slip resistance was measured with a TRRL-Wessex pendulum skid tester with a rubber brake shoe. - Resistence to drilling: we used a DRMS Cordless drill equipped with a bit with a diamond-covered tip (Sint Technology). - Ultrasound propagation (US), which we measured with a Panametrics HV Pulser/Receiver 5058 PR system coupled with a Tektronix TDS 3012B. Thermal properties -Thermal conductivity and thermal dilation: the machines used to assess these parameters were respectively a C-Therm TCi thermal conductivity analyzer supplied by Mathis Instruments Ltd with a universal sensor and a TMA Q400 thermal mechanical analyzer (TA Instruments). Color and roughness - Color analysis: color was analyzed using a Minolta CM-700d spectrophotometer with a xenon lamp and diffuse reflectance geometry. The chromatic parameters were obtained using the CIELab system and the color difference ¿E*94. -Roughness analysis: we used a Leica DVM2000 video-microscope and data and images were processed using the Leica Application Suite v.3.8.0 and Leica Maps Start v.6 (Leica Microsystems©) software packages. Decay tests - The following accelerated ageing tests were performed in the laboratory: the freeze-thaw test, the salt crystallization test and ageing by SO2 in the presence of moisture. - Test samples exposed to the elements (real conditions). In order to assess the behavior of these rocks in real conditions, samples were placed on one of the roof terraces of the Faculty of Science at the University of Granada and were exposed to real environmental conditions for four years. Petrographic description and petrophysical properties From a mineralogical point of view the Santa Pudia Calcarenite, Travertino Amarillo and Arenisca Ronda are made up almost exclusively of calcite, while the principal phase of Molinaza Roja is quartz and the secondary phase is potassium feldspar. Our study of the clay fraction revealed the presence of smectites in Santa Pudia Calcarenite and above all in Arenisca Ronda. The petrographic study showed that Santa Pudia Calcarenite is highly variable in terms of the both clast and pore size. There is little cement, which means that the percentage of porosity is high. We observed that Travertino Amarillo had the most crystalline texture of the four varieties. In the hand-held sample it was easy to distinguish light and cream bands: the white band has big pores and large numbers of schist and metamorphic quartz fragments, while the cream-colored band is more compact and the pores are smaller. The Arenisca Ronda rock is formed by reworked micritic clasts with high fossil content. This rock is well cemented although areas rich in clayey matrix were also observed. Molinaza Roja has a markedly laminated texture in which fine beds rich in iron oxyhydroxides can be observed. In this variety of rock we detected three kinds of cement (ferruginous, carbonated and siliceous) and a clayey matrix. The study of the porous system and the hydric properties shows that Santa Pudia Calcarenite has the highest effective porosity (33%), followed by Arenisca Ronda (17%), Molinaza Roja (15%) and Travertino Amarillo (8%). As regards the distribution of the pores, Santa Pudia Calcarenite and Arenisca Ronda have a high percentage of pores in the 0.1-1 µm range, the range most sensitive to damage by salt crystallization. Santa Pudia Calcarenite also has the highest values for the various hydric parameters we assessed (absorption at atmospheric pressure, capillarity and permeability to water vapor), while Travertino Amarillo scored the lowest values; Arenisca Ronda and Molinaza Roja obtained similar values somewhere between those for the other two varieties of stone. The orientation of the test samples shows that the position of the heterogeneities (lamination and clast orientation) has a clear influence on water movement through the stone. Our analysis of the dynamic and mechanical behavior of the rocks confirmed the results of the hydric tests. Thus although in Santa Pudia Calcarenite no heterogeneity (lamination or clast orientation) can be observed with the naked eye, it had a high degree of anisotropy; the same occurred with some of the samples of Arenisca Ronda, in which the orientation of the clasts was not clear to the naked eye, while ultrasound analysis confirmed it had an heterogeneous texture. The heterogeneities were more obvious in Travertino Amarillo and Molinaza Roja. Mechanical tests indicated that Travertino Amarillo and Molinaza Roja are the most resistant rocks while Santa Pudia Calcarenite is the least. We were unable to establish a clear relationship between the surface finish and the properties we analyzed, except in slip resistance and thermal conductivity, which are influenced by roughness. As a result it can be argued that texture and fabric are the factors that most influence the hydric and mechanical behavior of these rocks. In particular, pore size, porosity, crystal size, the degree of cementation of the tocks and the heterogeneities within it (such as laminations or clast orientation) are of fundamental importance. On this question, the indicators we observed suggest that, due to the decay of their texture and the presence of smectite, the Santa Pudia Calcarenite and Arenisca Ronda varieties are more susceptible to damage compared to Molinaza Roja and above all compared to Travertino Amarillo. Evaluating the durability of the stones by decay tests We performed various accelerated aging tests of which the salt crystallization test proved most aggressive both in terms of the amount of damage it caused and of the speed with which this damage appeared. This test caused the sanding and fissuring of the Santa Pudia Calcarenite, Arenisca Ronda and Molinaza Roja samples. The freeze-thaw test however only affected the Santa Pudia Calcarenite. By contrast, Travertino Amarillo was undamaged, at least to the naked eye, by either of these tests. ESEM microscope observation showed however that cement material had been lost and fissures had opened, leading to a fall in compactness measured using US and changes in the porous system manifested by MIP. In the case of Travertino Amarillo, although to the naked eye it behaves well, under the microscope we observed damage due to cracking and dissolution of the calcite. The SO2 attack test showed gypsum precipitation processes and different kinds of decay amongst the different types of rock. Although the surface finish did not seem to affect the result. The greatest gypsum precipitation took place in AR, in which a crust appeared on the surface. This effect was less observable in MR and TA and practically nil in SP. As regards surface finish, the pumice-polished finish underwent the greatest change. When the gypsum was removed, we observed that in SP and AR, the damage was due to the generation of above all intraclastic fissures, abundant amounts of organic matter, the dissolution of calcite crystals and the disgregation of micritic size crystals. In TA we observed crystal dissolution above all at the edges of the sparite crystals and the disgregation of micrite crystals. In MR we observed a loss of cementing materials, expansion of the phyllosilicates, alteration of the feldspars, appearance of cracks and organic matter. The color changes observed after the acid attack in most cases cannot be perceived by the naked eye. These changes were mainly due to changes in the b* parameter induced by modifications in the chromophore phases rather than to changes in the roughness of the surface. Finally the porous system in all the rocks underwent modifications but continued to behave in the same way as the intact samples when subjected to water vapor permeability tests. The results obtained after exposure to the elements indicate that there was an important difference in the weight and color of the samples between the second and third year, due to a loss of material and a blackening of the surface and changes in the color. From the fourth year onwards the deposition of organic matter was observed. The ESEM study demonstrated that the decay processes suffered by these materials were similar to those observed in the samples subjected to SO2 attack. In general the Santa Pudia Calcarenite variety behaved worse than Travertino Amarillo, the best performing rock due to its intrinsic properties. Arenisca Ronda and Molinaza Roja were shown to be significantly more vulnerable to damage by acid attack than Santa Pudia Calcarenite and Travertino Amarillo. Effectiveness of hydroxyapatite-based consolidant For this experiment we used dibasic ammonium phosphate (DAP), which reacts with the calcite to produce hydroxyapatite (HAP), which acts as a consolidant. Three different tests were performed in order to evaluate the effectiveness of the consolidant in terms of the precipitation of hydroxyapatite (HAP) in the four rocks selected for investigation. In the first test we characterized the phase formed from the solution with DAP, so confirming the precipitation of hydroxyapatite in all the rocks, with the exception of MR, in which we detected the presence of octacalcium phosphate, one of the precursors of hydroxyapatite. ESEM observation revealed differences in the morphology of the HAP amongst the different rock varieties, appearing in crust form especially in SP and TA. One negative result we observed was dissolution processes, above all in sparite crystals in contact with the solution. Lastly the spectrophotometry did not reveal any differences appreciable to the naked eye between the treated and untreated samples. In the second test we analyzed the petrophysical properties such as P wave propagation system, compression strength and flexural strength and resistance to drilling for a number of different surface finishes. We also analyzed the changes that took place in the porous system using the water vapor permeability test and mercury injection porosimetry. In general, the precipitation of hydroxyapatite has enhanced and improved the rocks¿ mechanical behavior, especially when analyzed from the point of view of the surface finish. In terms of compression strength however these samples proved weaker. The consolidant penetrated approximately 2 mm. The addition of the consolidant by capillarity appears to be more effective than by immersion although more research is required to improve its application. In the resistance to drilling test we noticed that the irregular distribution of the consolidant was creating alternate areas of varying resistance. Consolidation via HAP also reduced the number of pores in the 0.1 to 1 µm range, a positive change in that the pores in this range are the most vulnerable to salt crystallization damage. The last test involved accelerated aging by SO2 attack in the presence of moisture and by salt crystallization. At a macroscopic level during the SO2 attack we did not observe any differences between untreated and treated samples. The change in weight observed at the end of the test indicates that more gypsum was deposited compared to the untreated samples. However after washing the samples, the final damage suffered was less. Indeed the layer of consolidant has suffered severe decay, with fragments breaking off, cracks appearing in certain areas and scaling in others. However the impact of the consolidant could be viewed as positive as it has minimized the decay on the rocks, above all in the case of MR, in which even though precipitation of HAP was lower and the decay processes we observed were the same as in untreated samples, they had a less marked effect. The pumice-polished surface finish experienced the greatest loss of weight, especially in AR and MR. We also found that texture had a great influence on damage, in fact much greater than the surface finish. In all the samples we observed the appearance of organic matter associated with the consolidant, which could be a problem in terms of the possible effects of bio-decay. The change in color could not be perceived with the naked eye, except in the case of Travertino Amarillo. Although the roughness of the attacked samples has increased compared to the healthy samples, we were unable to establish a clear relation between variations in color, surface finish and roughness. These changes are above all due to changes in the chromophore phases rather than to changes in the roughness of the surface. Lastly, the water vapor permeability has increased. This change occurred mainly in the non-consolidated part, as an increase in resistance to drilling was detected in the first few millimeters. The consolidation process also proved effective when the test samples, once previously altered with sodium sulphate and with this salt still inside them, were treated with DAP and were then submitted to the salt crystallization cycles again. In this case the consolidant reduced the loss of material in SP and MR, while in AR which suffered a higher degree of decay, material loss stabilized in the final cycles. The ultrasounds measurements confirmed the effectiveness of the consolidant within just 3 days of its application in which an increase in wave velocity was recorded in altered and treated samples. As regards the changes in porosity, µCT confirmed that this is due both to the retention of salts inside the rock sand to the opening of new fissures or the increase in pre-existing interclastic spacing, corroborating in this way the results obtained via MIP. Finally after washing the samples, we observed an increase in the porosity of SP and AR due to the fact that these rocks are more porous than the other two rocks and consolidation was less effective; by contrast the reduction in the total porosity of TA and MR is probably due to higher levels of consolidant being deposited in the smaller pores. In general, the concentration we used of 1M of DAP seems to have been insufficient and more of a problem than an advantage, especially because of the dissolution of the calcite crystals that occurs to enable the hydroxyapatite to precipitate out, along with the fact that the consolidant offers only partial protection as it does not form a homogenous layer on the surface. Nonetheless in spite of the fact that the rock has a low calcium content, this may be sufficient to improve its physical and mechanical performance, as happened in MR. Conclusions In this Thesis, we have characterized the petrographic, physical-mechanical and durability properties of four varieties of rock used as building materials in historic and modern buildings in order to assess the most effective way to use them in the building process. These rocks, namely Santa Pudia Calcarenite, Travertino Amarillo, Arenisca Ronda and Molinaza Roja, are from the Andalusia region of southern Spain, and nowadays are sold with different surface finishes with varying degrees of roughness. On the basis of the results obtained, the four varieties of rock can be ordered from the most resistant to decay to the least as follows: Travertino Amarillo > Molinaza Roja >> Arenisca Ronda >Santa Pudia Calcarenite. The exhaustive study performed in this Thesis of all the petrophysical and durability properties of these rocks, and in particular of their texture and porous system has revealed the most important risk factors for decay, supplying useful, real data for the companies that produce the stone. This information is of great interest in that it is highly reliable and has been checked and confirmed using a range of different analytical techniques. This Thesis is also an example of how research can overcome the barriers between universities and private sector companies which on many occasions obstruct communication and understanding between the two. Without this understanding, it would be difficult to apply research of this kind in either the restoration of Historic Heritage buildings or in modern construction work.