The role of saline solution properties on porous limestone salt weathering by magnesium and sodium sulfates

Saline solution properties, viscosity in particular, are shown to be critical in salt weathering associated with sodium and magnesium sulfate crystallization in porous limestone. The crystallization of sodium and magnesium sulfate within a porous limestone has been studied at a macro- and microscale using different techniques, including mercury intrusion porosimetry, environmental scanning microscopy and X-ray computed tomography. Such analysis enabled the visualization of the crystallization process in situ, and at high magnification, yielding critical information as to where and how salts crystallize. Sodium sulfate decahydrate (mirabilite) tends to crystallize in large pores as euhedral micron-sized crystals formed at low supersaturation near to the surface of the stone. In contrast, magnesium sulfate heptahydrate (epsomite) tends to precipitate as anhedral wax-like aggregates formed at high supersaturation and distributed homogeneously throughout the stone pore system filling large and small pores. While the former crystallization behavior resulted in scale formation, the latter led to crack development throughout the bulk stone. Ultimately, the contrasting weathering behavior of the two sulfates is explained by considering differences in flow dynamics of solutions within porous materials that are mainly connected with the higher viscosity of magnesium sulfate saturated solution (7.27 cP) when compared with sodium sulfate saturated solution (1.83 cP). On the basis of such results, new ways to tackle salt weathering, particularly in the field of cultural heritage conservation, are discussed.     

Performance of limestones laden with mixed salt solutions of Na2SO4–NaNO3 and Na2SO4–K2SO4

The behaviour of two types of limestones having a different porosity, Maastricht and Euville limestone, laden with aqueous solutions of equimolar mixtures of sodium sulphate/sodium nitrate or sodium sulphate/potassium sulphate was investigated. At 50 % RH, the efflorescences on Maastricht samples during the first 30 h of drying consisted of similar amounts of thenardite and darapskite in case of an equimolar mixture of sodium sulphate/sodium nitrate while those on Euville samples under the same conditions contained mainly darapskite. After drying at 20 °C and 85 % RH, thenardite, formed through the precipitation and dehydration of mirabilite, was mostly detected in the efflorescences on both Maastricht and Euville samples. Re-wetting by increasing the RH from 50 to 85 % resulted in substantial damage on Maastricht stone laden with an equimolar mixture of sodium sulphate/sodium nitrate as a consequence of high supersaturation of mirabilite. In case of a contamination with equimolar amounts of sodium sulphate and potassium sulphate, the efflorescence on both limestones during drying at 50 % RH contained predominantly aphthitalite. The observed crystallisation behaviour is compared to the theoretical behaviour. The results indicate a strong influence of stone properties on the crystallisation behaviour of salt mixtures.     

Crystallization behavior of Na2SO4–MgSO4 salt mixtures in sandstone and comparison to single salt behavior

We report on the crystallization behavior and the salt weathering potential of Na₂SO₄, MgSO₄ and an equimolar mixture of these salts in natural rock and porous stone. Geochemical modeling of the phase diagram of the ternary Na₂SO₄–MgSO₄–H₂O system was used to determine the equilibrium pathways during wetting (or deliquescence) of incongruently soluble minerals and evaporation of mixed electrolyte solutions. Model calculations include stable and metastable solubilities of the various hydrated states of the single salts and the double salts Na₂Mg(SO₄)₂·4H₂O (bloedite), Na₂Mg(SO₄)₂·5H₂O (konyaite), Na₁₂Mg₇(SO₄)₁₃·15H₂O (loeweite) and Na₆Mg(SO₄)₄ (vanthoffite). In situ Raman spectroscopy was used to study the phase transformations during wetting of pure MgSO₄·H₂O (kieserite) and of the incongruently soluble salts bloedite and konyaite. Dissolution of kieserite leads to high supersaturation resulting in crystallization of higher hydrated phases, i.e. MgSO₄·7H₂O (epsomite) and MgSO₄·6H₂O (hexahydrite). This confirms the high damage potential of magnesium sulfate in salt damage of building materials. The dissolution of the incongruently soluble double salts leads to supersaturation with respect to Na₂SO₄·10H₂O (mirabilite). However, the supersaturation was insufficient for mirabilite nucleation. The damage potential of the two single salts and an equimolar salt mixture was tested in wetting–drying experiments with porous sandstone. While the high damage potential of the single salts is confirmed, it appears that the supersaturation achieved during wetting of the double salts at room temperature is not sufficient to generate high crystallization pressures. In contrast, very high damage potentials of the double salts were found in experiments at low temperature under high salt load.1     

Can drying and re-wetting of magnesium sulfate salts lead to damage of stone?

Magnesium sulfate salts have been linked to the decay of stone in the field and in laboratory experiments, but the mechanism of damage is still poorly understood. Thermomechanical analysis shows that expansion of stone contaminated with magnesium sulfate salts occurs during drying, followed by relaxation of the stress during dehydration of the precipitated salts. We applied thermogravimetric analysis and X-ray diffractometry to identify the salt phases that precipitate during drying of bulk solutions. The results show the formation of 11 different crystal phases. A novel experiment in which a plate of salt-laden stone is bonded to a glass plate is used to demonstrate the existence of crystallization pressure: warping of the composite reveals significant deformation of the stone during re-wetting of lower hydrates of magnesium sulfate. Environmental scanning electronic microscope (ESEM)/STEM experiments show that hydration of single crystals of the lower hydrates of magnesium sulfate is a through-solution crystallization process that is only visible at a small scale (~μm). It is followed by growth of the crystal prior to deliquescence. This demonstrates that crystallization pressure is the main cause of the stress induced by salt hydration. In addition, we found that drying-induced crystallization is kinetically hindered at high concentration, which we attribute to the low nucleation rate in a highly viscous magnesium sulfate solution.     

Crystallization of sodium sulfate phases in porous materials: The phase diagram Na2SO4-H2O and the generation of stress

Mechanical disintegration by crystal growth of salts in pores is generally considered as an important mechanism of rock breakdown both on Earth and on Mars. Crystal growth is also a major cause of damage in porous building materials. Sodium sulfate is the most widely used salt in accelerated weathering tests of natural rocks and building materials. This paper provides an updated phase diagram of the Na₂SO₄-H₂O system based on a careful review of the available thermodynamic data of aqueous sodium sulfate and the crystalline phases. The phase diagram includes both the stable phases thenardite, Na₂SO₄(V), and mirabilite, Na₂SO₄·10H₂O, and, the metastable phases Na₂SO₄(III) and Na₂SO₄·7H₂O. The phase diagram is used to discuss the crystallization pathways and the crystallization pressures generated by these solids in common laboratory weathering experiments and under field conditions. New crystallization experiments carried out at different temperatures are presented. A dilatometric technique is used to study the mechanical response of sandstone samples in typical wetting-drying experiments as in the standard salt crystallization test. Additional experiments with continuous immersion and evaporation were carried out with the same type of sandstone. Both, the theoretical treatment and the results of the crystallization experiments confirm that the crystallization of mirabilite from highly supersaturated solutions is the most important cause of damage of sodium sulfate in porous materials.     

Hydrogen and oxygen isotope fractionation during crystallization of mirabilite and ice

The equilibrium fractionation factors between mirabilite (Na₂SO₄·10H₂O) and saturated sodium sulphate solution at 25°C and 0°C and between ice and 2·5 molal sodium chloride solution at ?10°C have been measured. For mirabilite, the deuterium factors are 1·017 and 1·019, and the oxygen-18 factors are 1·0014 and 1·0020 at 25°C and 0°C, respectively. For ice, the factors are 1·024 for deuterium and 1·0022 for oxygen-18 at ?10°C. These fractionation factors are used to estimate the fractionation factors between ice and mirabilite and concentrated sea water at ?10°C. It is concluded that the average binding strengths of hydrogen in ice and mirabilite are very similar.     

The influence of environmental conditions on weathering of porous rocks by gypsum: a non-destructive study using acoustic emissions

Gypsum is one of the most universally distributed salts in weathered materials but little is known about the influence of environmental conditions on the damage generated by gypsum in stones. To quantify the damage induced by gypsum crystallization acoustic emission techniques are employed to record the elastic energy released during salt crystallization cycles in a limestone. Different environmental conditions have been established during the cooling and drying periods in traditional salt crystallization cycles. During drying two different temperatures (50 and 25 °C) and relative humidity (low, 25 % at 50 °C and 65 % at 25 °C, and high, 99 %) have been applied. The acoustic emission signals are filtered by a frequency analysis in order to eliminate signals corresponding to external noise or artifacts. Our experimental results show that acoustic emission activity is higher under high relative humidity conditions than under low relative humidity conditions, and also higher when drying at 50 °C than at room temperature. Microscopic observations on the weathered samples indicated that under high relative humidity conditions and at room temperature, gypsum crystallizes not on the sample surface, like in the other samples, but deeper in the inner part of the sample, in good agreement with previously published data. We show that using acoustic emissions as usually done in rock mechanics is also very useful in the study of stone decay and weathering processes in the laboratory.     

Crystallization kinetics of alkali feldspars in cooling and decompression-induced crystallization experiments in trachytic melt

Cooling and decompression experiments have been carried out on trachytic melts in order to investigate crystallization kinetics of alkali feldspar, the effect of the degree of undercooling (ΔT = T _liquidus ? T _experimental) and time on nucleation and crystal growth process. This experimental work gives us new data about crystallization kinetics of trachytic melts, and it that will be useful to better understand the natural system of Campi Flegrei volcanoes. Experiments have been conducted using cold seal pressure vessel apparatus, at pressure between 30 and 200 MPa, temperature between 750 and 855 °C, time between 7,200 and 57,600 s and redox condition close to the NNO +0.8 buffer. These conditions are ideal to reproducing pre- and syn-eruptive conditions of the Campi Flegrei volcanoes, where the “conditions” pertain to the complete range of pressures, temperatures and time at which the experiments were performed. Alkali feldspar is the main phase present in this trachyte, and its abundance can strongly vary with small changes in pressure, temperature and water content in the melt, implying appreciable variations in the textures and in the crystallization kinetics. The obtained results show that crystallization kinetics are strictly related to ΔT, time, final pressure, superheating (?ΔT) and water content in the melt. ΔT is the driving force of the crystallization, and it has a strong influence on nucleation and growth processes. In fact, the growth process dominates crystallization at small ΔT, whereas the nucleation dominates crystallization at large ΔT. Time also is an important variable during crystallization process, because long experiment durations involve more nucleation events of alkali feldspar than short experiment durations. This is an important aspect to understand magma evolution in the magma chamber and in the conduit, which in turn has strong effects on magma rheology.     

Assessment of the Effects of Freeze–Thaw and Salt Crystallization Ageing Tests on Anahita Temple Stone,Kangavar, West of Iran

Building stone of Anahita Temple seriously suffers from weathering due to long term freezing-thawing and salt crystallization processes. This article investigates possible changes of physical and mechanical characteristics of this stone subjected to freeze–thaw and salt crystallization ageing tests. Fresh samples obtained from the Chelmaran quarry (the main quarry supplying for Anahita Temple stone) were tested under freeze–thaw and salt crystallization experiments. The freeze–thaw and sodium sulfate salt crystallization are suggested to be the most effective factors affecting in apparent deterioration of the stone in compare to the magnesium sulfate salt crystallization test. Significant decreases in mechanical properties of the stone were observed after freeze–thaw and salt crystallization tests. However, more mechanical losses were recorded after the salt crystallization cycles than the freeze–thaw cycles. This is probably due to crystallization pressure of salt crystals in compare to ice wedging force, which promoted more development of micro-fractures in the specimens. Probably, intrinsic factors of the stone such as frequent calcite veins and stylolites, are the main factors that control the durability of Anahita Temple stone. Preferential weakening along these features during freeze–thaw and salt crystallization cycles led to physical destruction and strength loss of the stone. Based on comparison between experimentally induced damages and field observations, reasonably freeze–thaw process is major factor in weathering of Anahita Temple stone. It should be noted that recorded 102 frozen days for the region imply high destruction potential of the stone during freeze–thaw cycles.     

Patterns of halite (NaCl) crystallisation in building stone conditioned by laboratory heating regimes

The crystallisation of soluble salts within the pores of the stone is widely recognised as a major mechanism causing the deterioration of the stone-built architectural heritage. Temperature, in turn, is one of the main controls on this process, including salt precipitation, the pressure of crystallisation and the thermal expansion of salts. Most laboratory experiments on decay generated by salts are just carried out with convective heating regimes, while in natural environments building stones can undergo radiative and convective heating regimes. The thermal response of stone to these different heating regimes is noticeably different and might influence the crystallisation patterns of a salt within a stone. The aim of this work is to raise awareness on the different patterns of crystallisation of NaCl within a porous stone tested with different heating regimes (convection and radiation) and the implications that this could have on the design of experimental modelling of natural weathering conditions in laboratory simulations. Results show that heating regime affects the sodium chloride distribution within a stone with high percentage of microporosity. In this case, radiation heating facilitates the generation of subefflorescences, while convection heating promotes efflorescences. This has a clear implication both on the stone decay in natural environments and on the methodologies for testing salt decay, as subefflorescences are more destructive than efflorescences. In this sense, the use of convective heating in laboratory experimentation might underestimate the potential damage that sodium chloride may generate. This counsels the use of radiation heating test methods in addition to convection for the laboratory study of salt crystallisation.     

Impact of in-pore salt crystallization on transport properties

Precipitation of salts in confined spaces is the key mechanism for rock weathering and damage to building materials. To date there is no comprehensive study of the parameters influencing the reduction of pore space by salt crystals and the consequences for transport and damage by crystallization pressure. A novel method is presented to quantify pore clogging (i.e., the degree to which crystallization of salts interferes with transport of solution in porous materials). After drying capillary-saturated stone specimens containing salt solutions, the rate of capillary uptake of decane into the salt-contaminated specimens is measured. By treating the salt-contaminated material as a bilayer, the width of the crystallization front and the degree of pore filling can be determined. Two model materials with different pore size distributions (Indiana and Highmoor limestone) and three salts (sodium chloride, sodium sulfate and magnesium sulfate) are selected for this study. It is shown that pore clogging results from the interplay between pore size distribution and salt properties. Different scenarios are discussed to link pore clogging with salt damage.     

Plagioclase nucleation and growth kinetics in a hydrous basaltic melt by decompression experiments

Isothermal single-step decompression experiments (at temperature of 1075 °C and pressure between 5 and 50 MPa) were used to study the crystallization kinetics of plagioclase in hydrous high-K basaltic melts as a function of pressure, effective undercooling (ΔT _eff) and time. Single-step decompression causes water exsolution and a consequent increase in the plagioclase liquidus, thus imposing an effective undercooling (?T _eff), accompanied by increased melt viscosity. Here, we show that the decompression process acts directly on viscosity and thermodynamic energy barriers (such as interfacial-free energy), controlling the nucleation process and favoring the formation of homogeneous nuclei also at high pressure (low effective undercoolings). In fact, this study shows that similar crystal number densities (N _a) can be obtained both at low and high pressure (between 5 and 50 MPa), whereas crystal growth processes are favored at low pressures (5–10 MPa). The main evidence of this study is that the crystallization of plagioclase in decompressed high-K basalts is more rapid than that in rhyolitic melts on similar timescales. The onset of the crystallization process during experiments was characterized by an initial nucleation event within the first hour of the experiment, which produced the largest amount of plagioclase. This nucleation event, at short experimental duration, can produce a dramatic change in crystal number density (N _a) and crystal fraction (?), triggering a significant textural evolution in only 1 h. In natural systems, this may affect the magma rheology and eruptive dynamics on very short time scales.     

Intense fracturing and fracture sealing induced by mineral growth in porous rocks

Mineral precipitation in the pores of a rock may exert a force, which is called crystallization pressure. This process has been studied experimentally and results bring a new look on the way fractures may develop and seal in natural systems. Cylindrical core samples of porous limestone and sandstone were left for several weeks in contact with an aqueous solution saturated with sodium chloride, at 30 or 45 °C, under axial normal stress in the range 0.02–0.26 MPa. The fluid was allowed to rise in the core samples by capillary forces, up to a controlled height where evaporation and precipitation occurred. The uniaxial deformation of the samples was measured using high-resolution displacement sensors. The samples were characterized using computed X-ray tomography, allowing therefore imaging in 3D the intensity and localization of the damage. Two kinds of damage could be observed. Firstly, small rock fragments were peeled from the sample surface. Secondly, and more interestingly, fracture networks developed, by nucleation of microcracks at the interface where evaporation occurred, and propagation to the free surface. Two families of fractures could be identified. A first set of sealed fracture parallel to the evaporation front is directly induced by crystallization pressure. A second fracture network, perpendicular to the evaporation front, accommodates the first set of fractures. An analytical model where fluid flow is coupled to evaporation, vapour transport, and localization of mineral precipitation explains the shape of this fracture network.     

Crystallization temperature of wyomingite from Leucite Hills

Wyomingite collected from Leucite Hills is composed mainly of leucite, diopside, phlogopite, and small amounts of apatite, calcite, magnetite and rare amphibole, and is characterized by very high content of potash. Thermal experiments at atmospheric pressure indicate that the liquidus phase is always diopside with liquidus temperature of 1320 °C, and solidus temperature is about 1000 °C. Various kinds of melt inclusions are abundant in all constituent minerals. They comprise mono-phase (glass only), two-phase (gas+glass), three-phase (gas+glass+one crystalline phase) and multi-phase (gas+glass+more than two crystalline phases) inclusions. Thermal experiments have been made on these inclusions in phlogopite, diopside, and leucite in order to estimate the temperature of crystallization by homogenizing these inclusions. The results show that the crystallization of wyomingite began with formation of phlogopite accompanied by diopside at 1270 °C. Although diopside ceased crystallization at 1220 °C recurrent crystallization of phlogopite was noticed between 1120 ° and 1040 °C. Leucite crystallized out abundantly between 1250 ° and 1150 °C. Complete solidification of wyomingite occurred at about 1000 °C.     

Crystallization kinetics of strontianite from Sr(HCO3)2 solutions

The kinetics of crystallization of strontium carbonate (strontianite) from strontium bicarbonate solutions were examined. CO₂ was stripped from a slightly acidic solution of Sr(HCO₃)₂ by stirring resulting in critical supersaturation and precipitation of strontianite. The reduction of the Sr²⁺ concentration was recorded as a function of time by measuring the electrolytic conductivity and the pH value.

Homogeneous primary nucleation is dominant at high supersaturations, whereas heterogeneous primary nucleation prevails at low supersaturations. The crystal growth rate increases with increasing supersaturation. This effect is less pronounced at higher supersaturations. The growth rate is mostly transport-controlled at high supersaturation. At lower supersaturation the crystal growth is mainly determined by integration of ions into the crystal lattice. These results may be used to explain the deposition of strontianite in natural systems.     

Changing climate, changing process: implications for salt transportation and weathering within building sandstones in the UK

Salt weathering is a crucial process that brings about a change in stone, from the scale of landscapes to stone outcrops and natural building stone façades. It is acknowledged that salt weathering is controlled by fluctuations in temperature and moisture, where repeated oscillations in these parameters can cause re-crystallisation, hydration/de-hydration of salts, bringing about stone surface loss in the form of, for example, granular disaggregation, scaling, and multiple flaking. However, this ‘traditional’ view of how salt weathering proceeds may need to be re-evaluated in the light of current and future climatic trends. Indeed, there is considerable scope for the investigation of consequences of climate change on geomorphological processes in general. Building on contemporary research on the ‘deep wetting’ of natural building stones, it is proposed that (as stone may be wetter for longer), ion diffusion may become a more prominent mechanism for the mixing of molecular constituents, and a shift in focus from physical damage to chemical change is suggested. Data from ion diffusion cell experiments are presented for three different sandstone types, demonstrating that salts may diffuse through porous stone relatively rapidly (in comparison to, for example, dense concrete). Pore water from stones undergoing diffusion experiments was extracted and analysed. Factors controlling ion diffusion relating to ‘time of wetness’ within stones are discussed, (continued saturation, connectivity of pores, mineralogy, behaviour of salts, sedimentary structure), and potential changes in system dynamics as a result of climate change are addressed. System inputs may change in terms of increased moisture input, translating into a greater depth of wetting front. Salts are likely to be ‘stored’ differently in stones, with salt being in solution for longer periods (during prolonged winter wetness). This has myriad implications in terms of the movement of ions by diffusion and the potential for chemical change in the stone (especially in more mobile constituents), leading to a weakening of the stone matrix/grain boundary cementing. The ‘output’ may be mobilisation and precipitation of elements leading to, for example, uneven cementing in the stone. This reduced strength of the stone, or compromised ability of the stone to absorb stress, is likely to make crystallisation a more efficacious mechanism of decay when it does occur. Thus, a delay in the onset of crystallisation while stonework is wet does not preclude exaggerated or accelerated material loss when it finally happens.     

Dynamic crystallization of chondrule melts of porphyritic and radial pyroxene composition

Dynamic crystallization experiments in which heterogeneous nucleation is an important variable have been completed on four melts of chondrule composition. Compositions were chosen to best represent chondrules with porphyritic pyroxene and radial pyroxene textures. Experimental results show that heterogeneous nucleation is essential for the formation of porphyritic textures. Without preexisting nuclei, too much supercooling is established before crystallization is initiated and the textures are more likely to be dendritic or radial. In the near total absence of nuclei, radial textures can form at cooling rates as slow as 5°C/hr in this study. By varying the heterogeneous nucleation conditions and having a melt in which the appropriate phases are stable or metastable, analogs to most of the recognized chondrule textures can be produced in a single melt composition. Olivine inclusions in pyroxene can form readily during an experiment from a starting material which did not initially contain olivine crystals. Thus care must be taken in the assumption that olivine inclusions in pyroxene represent preexisting crystals.     

Crystallisation of sodium sulfate: supersaturation and metastable phases

Crystallisation of sodium sulfate solutions by evaporation under controlled climatic conditions has revealed the existence of crystalline hydrated sodium sulfate salts not previously reported. The sodium sulfate phase crystallising and the concentration of the solution at the point of crystallisation depends on the climatic conditions (temperature and evaporation rate). During the rehydration of the anhydrous sodium sulfate phase, thenardite, another previously unreported phase was formed prior to the nucleation of the stable phase, mirabilite Na₂SO₄ · 10H₂O. The addition of organic inhibitors changes both the crystallisation and the rehydration behavior in this system.     

Monitoring the crystallization of water-saturated granitic melts in real time using the hydrothermal diamond anvil cell

Critical aspects of the crystallization dynamics of H₂O-saturated melts of a typical granitic composition as well as granitic melts enriched in lithium (8800 ppm) were investigated in real-time experiments using the hydrothermal diamond anvil cell at 480–700 °C and 220–960 MPa. Complete crystallization of the charges was achieved within 5–118 h with average crystal growth rates ranging from 3 to 41 cm/year for quartz and from 18 to 58 cm/year for alkali feldspars, demonstrating that crystals formed from a silicate melt in the presence of a coexisting aqueous phase crystallize rapidly. The combination of substantial nucleation delays, low nucleation densities, and rapid growth rates for quartz and alkali feldspars led to the formation of euhedral megacrysts of these minerals in the vicinity of clusters comprised of much smaller muscovite or α-spodumene crystals. Subsolidus replacement processes initiated during crystal–fluid interactions after the silicate melt was consumed were directly observed in the experiments. The experimental results underscore the important role of water as a medium for the transport of essential elements such as Si, Al, Na, and K from the silicate melt to the newly formed crystals, and provide important insights into the crystallization of miarolitic pegmatites.     

Differential damage in the semi-confined Munazio Ireneo cubicle in Cagliari (Sardinia): a correlation between damage and microclimate

The Early Christian Munazio Ireneo cubicle in Cagliari (Sardinia) is carved into the rock and is one of the rare monuments of Sardinia belonging to Early Christian Age. It is 166 cm under the planking level and is a semi-confined chamber in which a gate allows exchange with the outside. In 1888, when it was discovered, it was completely painted, but over the years, it suffered a serious damage and all the paintings are nearly disappeared. Now the site shows a high moisture and a differential damage characterised by delamination, powdering, salt crystallisation and biological colonisation. The research offers a multidisciplinary approach to study the salt crystallisation damage, which is a phenomenon still not completely understood today. Studies and analyses highlight that different kinds of damage can be recognised. The cement mortar and the air pollution cause crystallisation of the sulphates. Constant capillary rising and the presence of water infiltration due to the rains, which travelled into the site from the vault, were detected during qualitative inspection. Although the microclimate did not change much in the site and the humidity was almost constant during the monitoring period, the variations in temperature allowed the phase transitions of sodium sulphate, especially in summer. The greatest damage of the porous stone is associated with the phase transitions and crystallisation inside the stone of sodium sulphate, one of the most harmful salts for porous materials because of its high crystallisation pressure.