This research programme was funded by Historic Scotland. The research programme was carried out at The Robert Gordon University and at the Building Research Establishment.. The aim of the research was to examine the effects of various consolidants and water repellents on Scottish building sandstones. Copies of the research results can be obtained from Historic Scotland.Historic Scotland
For further information contact Dr. Maureen Young at firstname.lastname@example.org
The following information is a summary from the results in the research report. This information is copyrighted to The Robert Gordon University, The Building Research Establishment and Historic Scotland. If quoted the information should be referenced as:Young, M.E., M. Murray and P. Cordiner, 1999.
Water repellents and consolidants are used in an attempt to minimise the rate of stone decay and to strengthen decayed stone where there has been a failure of the natural stone cement through the normal processes of weathering, the effects of atmospheric pollution or through inappropriate intervention. In the past 20 years many materials have been used with varying degrees of success and no single method has been found to be successful on all stone types.
Where decay has occurred the aim is to preserve the stone, as far as is practicable, in its original state. For this purpose recent research has focused on developing suitable materials for consolidating stone or imparting water repellency. Water repellents are intended to prevent or reduce water penetration into stonework and therefore minimise the rate of decay. Consolidants are intended to strengthen weakened stone and slow the rate of surface loss by binding loosened grains. Knowledge of the medium and long term performance of consolidants and water repellents is minimal and there is a lack of agreement between experts in this field as to which materials can be appropriately used. This casts doubt on the advisability of using these materials on important structures without extensive research and testing.
The aim of this programme was to research the effects of modern consolidants and water repellents on Scottish sandstones and their success in reducing the rate of stone decay and of consolidating existing decayed stone surfaces. Specific objectives of the programme were:
Sandstone characteristics can vary widely between sandstone types and even between blocks of the same stone from different parts of the quarry. The results of different treatments may also vary widely. As a result, the research focuses on selected consolidants and water repellents which are or have been used in Scotland and tests them on a range of sandstones (both soiled and fresh) which are commonly used as building stone in Scotland.
The research programme was commissioned by Historic Scotland and was a collaborative project involving both The Robert Gordon University (Aberdeen) and the Building Research Establishment (Watford).
In general, if a consolidant or water repellent treatment is to perform adequately there are certain criteria which must be met:
Water repellents are intended to prevent or reduce water penetration into stonework and so reduce the rate of decay. Consolidants are intended to strengthen weakened stone and slow the rate of surface loss by binding loosened stone through bridging of gaps between grains. Many treatments have a mixture of consolidant and water repellent characteristics. Some are predominantly water repellents with little consolidative effect, others are mainly consolidants but also have a water repellent effect. The treatments tested and used on stone are normally of a single material type. However, mixtures of treatments are also occasionally used.
Silane-based materials: These include a wide variety of organosilicon compounds which polymerise to form networks of silica gel. As the gel dehydrates it forms deposits of silica in the substrate. Some types (e.g. tetraalkoxysilanes) have no water repellent properties, others (e.g. alkyl trialkoxysilanes) have a degree of water repellency which can be controlled by altering the properties of starting material. Polysiloxanes are partially polymerised silanes. They are less volatile than silanes and are soluble in organic solvents. Silane-based materials are generally able to penetrate relatively deeply into a porous substrate although results can be variable. Significant strength improvements have been reported using silanes. The polymer is very stable and will not break down in UV light or when exposed to acid rain. However, it may break down in the presence of silane vapour during re-application of consolidant if the previous application has not fully cured.
Acrylic consolidants: Acrylic resins which have been used on stone include methylacrylate, methyl methacrylate, ethyl methacrylate and butyl methacrylate. They can achieve substantial increases in the strength of the substrate. When it is free from impurities, polymethyl methacrylate (PMMA) is stable to degradation by heating, oxygenation and ultra-violet radiation and will not yellow with age. Other acrylics are less stable.
Vinyl consolidants: Vinyl consolidants include polyvinylchloride (PVC) and polyvinylacetate (PVA). These have generally been unsuccessful as stone consolidants since they tend to give the surface a glossy appearance and it is difficult to achieve good penetration.
Epoxies: Epoxy resins are formed by cross-linking of low molecular weight epoxy polymers. They can achieve substantial increases in the strength of a substrate. However, epoxy resins have generally been found to be impractical for use in the consolidation of sandstone due to their low penetration.
Polyurethane: Polyurethanes are introduced into stone in solvents and deposited as the solvent evaporates Polyurethanes can achieve substantial increases in stone strength but are vulnerable to decay caused by exposure to heat or light, they may therefore include stabilisers.
Polyesters: Polyesters have poor resistance to UV radiation, acids and alkalis. Their long term performance is poor and they have not been found to be useful in stone conservation.
Perfluoropolyethers: Perfluoropolyethers are generally used as water repellents as they have little or no cohesive effect. Perfluoropolyethers are very stable to UV and are theoretically removable from stone through their solubility in fluorinated solvents. These products remain mobile within a stone and their performance therefore gradually declines over time. Degree of mobility is affected by the polarity and size of the functional groups and the porosity of the stone, the decline in protective efficacy being slower in less porous stone. These products therefore generally work best in stone types of lower porosity.
Fluosilicates: Fluosilicates of magnesium, zinc and aluminium have been tested as stone consolidants. However, acids are produced in their formation which will attack calcareous stones resulting in a hardened surface layer through deposition of calcium fluorides. Fluosilicate consolidation has been suspected of being responsible for exfoliation and accelerated deterioration of some treated sandstones.
Barium-hydroxide (limestone treatment): Treatments involving barium hydroxide have been used for the consolidation of decayed limestone. Treatment with barium hydroxide, in addition to the formation of barium carbonate, also results in the transformation of calcium sulphate into the relatively insoluble barium sulphate. Treatments in the past have followed the barium compound with a second acid solution to form an insoluble precipitate with the barium. Silicic, fluosilicic, carbonic, sulphuric, chromic and phosphoric acids have been used. Disadvantages of these treatments include colour or textural changes and the short term nature of any consolidation effects.
Limewater (limestone treatment): Lime has been used as a consolidant for limestones. Lime is converted into calcium carbonate by carbonation in contact with the atmosphere. Results have varied and while some authors report good outcomes others have found results to be poor.
The method of application can be very important with respect to the depth of penetration of consolidant or water repellent treatments. Application methods (both field and laboratory based) include: brushing, spraying, dripwise application, capillary absorption, vacuum impregnation, pressure impregnation and total immersion. Depths of penetration achievable in laboratory settings may not necessarily be achievable in the field. Unless the laboratory method reproduces, as far as possible, conditions likely to be found in the field, results are unlikely to reflect those encountered on a building facade.
The amount of treatment absorbed may vary widely depending on the porosity and permeability of the substrate. The state of decay of the stone is also an important influence on treatment absorption since decayed, weathered stones often take up more treatment than relatively intact stone. Rates of absorption of treatments are also affected by the solvent used and the viscosity of the treatment. Deep penetration of stone cannot be achieved if the preservative substance polymerises too rapidly. Some systems therefore add catalysts to the monomer-solvent mixture which cause rapid, but delayed, setting or gelation at some time after application.
The amount of consolidant or water repellent absorbed by stonework can be an indication of likely depth of penetration and may be estimated by measuring the weight or volume of material applied. Errors may arise due to losses through run-off, differential penetration into stones and mortar or (for some treatment methodologies) due to losses during solvent washing phases of the treatment. Losses of consolidant during the solvent washing stage are likely to be significant where the depth of penetration of the treatment is low.
For the sandstone tested here, the depth of penetration of the treatments varied from close to zero up to about 60 mm depending on treatment type and stone characteristics. Penetrations close to zero were found in sandstones of low porosity and permeability. High penetration depths were achieved on highly porous and permeable sandstones. In mortars, penetration depth varied from a few millimetres up to 80 mm depending on the mortar composition. Penetration was often greater along joints than in either the mortar or sandstone.
There are a number of factors which may be involved in assessing the performance of any consolidant or water repellent treatment. These can include:
If depth of penetration is required to be known accurately then there is no substitute for coring and measurement. However, a rough indication of depth of penetration may be gained from measurements of the volume of treatment applied, water absorption rate, sandstone porosity or permeability measurements, provided that pre-existing data on measured penetration depths are available to provide a comparison. There is no single test which effectively determines the depth of penetration of all treatments. For water repellent treatments, tests for hydrophobicity on core samples can give a fast, effective measure of the depth of penetration. Non-hydrophobic treatments may require other test methods such as strength tests. Different test methods may yield different penetration depths on the same sample since measured characteristics may vary in their effectiveness. For example, while water repellency results may indicate a relatively deep penetration, strength changes may occur only at the near surface. Although it is possible to predict, within a margin of error, the likely penetration of treatments the true penetration depth cannot be predicted with accuracy unless core samples are taken for analysis. Depth of penetration varies between stones with different characteristics and the effects of mortar type and joint condition can have a significant effect on penetration depth on a masonry surface.
Useful test methods for measuring depth of penetration include:
For the consolidants treatments tested here, the results of strength tests on treated sandstones indicated only relatively slight increases in strength. However, these samples included mainly fresh and soiled sandstones, rather than highly decayed stone. Other investigators have found substantial increases in strength on application of silanes, acrylics, epoxies and polyurethane consolidants. This suggests that stone has to be extremely friable if consolidants are to produce any effective increase in strength.
None of the treatments tested here were found to significantly block porosity in the sandstones. However, some treatments did reduce porosity by about 1.5-3%. Other investigators have found some reduction in porosity with silane-based treatments. Measurements of changes in water vapour permeability showed that some treatments tested here could cause reductions in permeability of up to 30%. Other investigators have found evidence of reduction in water absorption for silane, acrylic, epoxy and polyurethane treatments.
All the consolidants and water repellents tested here caused colour changes, some more than others. The degree of colour change may reduce over time and some treatments appear to reduce the soiling rate of the stone. However, although the degree of colour change may vary, the appearance of the stone surface is likely to be permanently altered while the treatment remains present. Treatments mainly resulted in darkening of the surface although slight colour changes may also occur. Light coloured sandstones may be especially vulnerable since colour changes can be most visible on light coloured stone. Colour changes may also differentially enhance small colour differences in the original stone colour. Washing-down with appropriate solvents after application may be necessary to minimise colour changes caused by excessive treatment retention at the surface.
Laboratory results from sandstones treated here indicated that effects on thermal expansion and contraction of treated stone were small but measurable. The implications of these results are difficult to extrapolate in terms of behaviour on a building facade, however, substantial differences in thermal behaviour relative to the underlying stone can result in detachment of a treated surface.
While there are many reports on performance of water repellents and consolidants in laboratory based situations, practitioners should be aware that results from stone samples treated in the laboratory are not necessarily representative of those which would be obtained on real buildings. In the laboratory, samples can be very effectively impregnated. However, on real buildings it is seldom possible to achieve the same level of control. Laboratory based applications therefore produce better, more reproducible results than can generally be expected in the field. Despite these difficulties, laboratory based tests can provide a useful indication of the potential performance of a treatment where the volume of sample and its situation are a good simulation of the building facade.
Laboratory based tests on bulk mortar samples are unlikely to provide an accurate picture of the behaviour of treatments on mortar on building facades. The characteristics of bulk mortar have been found to be significantly different to those of mortar in joints due to differences in compaction and curing and the likely presence of cracks at joints in building facades, especially in cement mortars.
The effective life span of treatments is relatively little studied beyond one or two years. It is therefore not possible to draw any general conclusions from the limited data available. However, some characteristics of particular polymers make them unlikely to be suitable for long term use on buildings. These include polymers which are unstable under ambient conditions of light, temperature or moisture.
Examination of the long-term effects of consolidants and water repellents on field sites where treatment was carried out up to 20 years previously indicates that results can be very variable. Water repellency can, in some circumstances, be retained on a relatively stable, treated surface for over 20 years while the surrounding areas may be very decayed and apparently in a bad state of recession. The treatment may still be detectable (in terms of its water repellency) to a considerable depth. Even where the original surface is lost, water repellency may still be retained inside the stone. In general, better results appear to be obtained using silane-based compounds than with acrylic compounds. This difference is apparently due to the greater penetration depth of the silane materials. Acrylic compounds may, however, perform better where the treated surface is protected from wetting.
Opinion is divided on the use of water repellent materials since they alter the characteristics of the stone with respect to water absorption and evaporation. For both water repellents and consolidants there is no agreement as to what represents an appropriate treatment depth. Some investigators suggest that water repellents might be most appropriately applied only to the immediate surface, whereas others suggest that deep treatments are more appropriate. Water repellents are likely to be effective where the requirement is to stop water penetrating the stone surface. They will be ineffective and potentially damaging where moisture may enter the stone through paths other than via the treated surface.
For consolidant treatments, there is agreement that treatments confined to the outer surface are dangerous since they can result in spalling of the stone but there is no agreement on what would be an appropriate depth of treatment beyond the fact that it is obviously necessary to treat the stone deeply enough to consolidate the full thickness of the decayed zone. Some stone consolidants also have water repellent properties. Since it is dangerous to confine consolidation to the near surface this makes the use of a combined water repellent-consolidant problematic if it is considered best to confine water repellency characteristics to the outer surface.
The ideal polymer for use in stone consolidation would be one which can reverse the degradation of a stone, returning it as nearly as possible to its original condition. In order to achieve this the treated stone should mimic sound stone in as many characteristics as possible. Some characteristics are, however, more important than others. The most important include strength, porosity, permeability, thermal dilation and colour. Of all the polymers tested, silanes seem to hold out the most promise although they may not be suitable in every situation. The theoretical end product of polymerisation of the simplest silanes is silica which is present as a cementing mineral in many sandstones and may mimic the behaviour of a natural cement more closely than many other polymers.
|MCRG Home Page | People | Research Areas | Publications | Activities | Links|
|Robert Gordon University Home Page|