There are a large number of mechanisms by which masonry structures can deteriorate; these can be categorised into:
- Chemical/biological attack on either the mortar or the units or both, due to water and waterborne acids, sulphates, pollution and chemicals released by growing plants.
- Corrosion of embedded metal (usually steel) components, particularly ties, straps, reinforcing rods, hangers, etc. – a special case of chemical attack.
- Erosion of units or mortar by particles in flowing water and wind, by frost attack and by salt crystallisation.
- Stress-related effects due to movement of foundations, movement/consolidation/washout of in-fill, vibration, overloading, moisture movement of bricks and blocks, thermal movement, growth of woody plants.
- Staining due to efflorescence, lime, iron, silica, vanadium and biological growth.
Deterioration of Masonry
Because they must be finely divided to be able to bind together the sand grains in aggregates and to react fairly rapidly to set to a hard adhesive mass, binders are usually more chemically reactive than the other components of masonry. Their chemical reactivity is their weakness, in that they often react with chemicals in the environment with resultant deterioration.
Mortar is generally the least durable of the concrete-like materials because it contains binders, usually has a relatively high connected-porosity that allows water to percolate through it, and usually has only a modest hardness and abrasion resistance.
Dense concrete units are durable because they are hard and resist percolation and lightweight concrete units because they have a mixture of large and fine pores. Well-fired clay units are generally very resistant to chemical attack. The durability of natural stone is very variable, ranging from the highest performance – given by dense impermeable granites and marbles – to the quite poor performance of porous limestones and lime and clay-bound sandstones.
Water and Acid Rain
Water percolating into masonry is always a potential source of damage and, where possible, the structure should be designed to throw falling rain away from façades and to channel absorbed water away, or at least to allow it to escape via weepholes then drip away from the face.
Absolutely pure water will have no direct chemical effect but some of the constituents of mortar are very slightly soluble and will dissolve very slowly. Rainwater containing dissolved carbon dioxide is a very mild acid that dissolves calcium carbonate by formation of the soluble bicarbonate via the reactions:
CO2 + H2O → H2CO3 ……(1)
CaCO3 + H2CO3 → Ca(HCO3)2 …..(2)
This means that lime mortars, weak ordinary Portland cement (OPC), lime mortars, porous limestones, porous lime-bonded sandstones and porous concrete blocks made with limestone aggregate will eventually be destroyed by percolating rain water because calcium carbonate is a key constituent.
Strong OPC mortars with well-graded sand and most concrete and sandlime units are less susceptible partly because the calcium silicate binder is less soluble but mainly because they are less permeable, so free percolation is prevented. Typical visible effects of water leaching on mortar are loose sandy or friable joints, loss of mortar in the outside of the joints giving a raked joint appearance, and in serious cases the loss of units from the outer layer of masonry, particularly from tunnel/arch heads.
The process will sometimes be accompanied by staining due to re-precipitation of the dissolved materials. Stones lose their surface finish and may develop pits or rounded arrises. Sulphur dioxide reacts with water to form initially sulphurous acid but can oxidise further in air to sulphuric acid:
SO2 + H2O → H2SO3 ……..(3)
2H2SO3 + O2 → 2H2SO4 …….(4)
There is no systematic evidence that rain acidified by sulphur dioxide from flue gases at the normal levels has a particular effect on mortar, but very clear evidence that sulphur dioxide and its reaction products do attack limestones, usually with the formation of black crusts in partly-sheltered smoky environments followed by surface spalling. Lime-bound sandstones are also attacked and suffer contour scaling, as shown in Fig. 1.
One mechanism of failure is the expansive conversion of calcium carbonate to gypsum:
CaCO3 + H2SO4 + 2H2O → CaSO4.2H2O + H2CO3 ………(5)
Gaseous carbon dioxide (CO2) at humidities between about 30 and 70% neutralises any alkalis present. This process occurs for all lime and Portland cement binders with the conversion of compounds such as sodium hydroxide (NaOH) and, most commonly, calcium hydroxide (Ca(OH)2) to their respective carbonates.
In lime mortars this process probably increases the strength and durability. In Portland cement-based materials the key effect of the process is to reduce the pH from around 12–13 down to below 7, i.e. converting the material from highly alkaline to slightly acid. This can have a profound effect on the durability of embedded steel components.
There is also some evidence that there is a slight associated shrinkage that may reduce the strength of very lightweight concrete units. Very dense concrete and calcium silicate units will carbonate slowly and may take 50–100 years or more to carbonate through.
Sulphate attack is the next most common problem, and is due to the reaction between sulphate ions in aqueous solution and the components of hardened Portland cement to form ettringite. The resulting expansion, which can be on the order of several per cent, causes both local disruption of mortar beds and stresses in the brickwork, but only in wet or saturated conditions and where there is a source of a water-soluble sulphate compound. It will never occur in dry or slightly damp masonry.
The common sulphates found in masonry are the freely soluble sodium, potassium and magnesium salts and calcium sulphate, which is less soluble but will diffuse in persistently wet conditions. Sulphates may be present in groundwater and can affect masonry in contact with the ground such as in foundations, retaining walls, bridges and tunnels. In this situation porous concrete units are also at risk – see BRE Special Digest 1 (2005).
Soluble sulphates are also present in some types of clay brick and will be transported to the mortar in wet conditions. Examples are any clay bricks with unoxidised centres (‘blackhearts’), some Scottish composition bricks and semi-dry pressed bricks made from Oxford clay (Flettons), which have high levels of calcium sulphate. Sulphates may also attack lime mortars by conversion of the lime to gypsum in a similar reaction to that shown in equation 5.
Sulphate-resisting Portland cement is deliberately formulated to have a low C3A content but may be attacked in very extreme conditions. Another compound, thaumasite Thaumasite Experts Group (1999), may form by reaction between dicalcium and tricalcium silicate, sulphate, carbonate and water. This process can disrupt mortar beds.
Visible effects of sulphate attack on mortar are expansion of the masonry where it is unrestrained and increase in stress where it is restrained. Surface spalling is common. Typically the mortar is affected more within the body of the wall than on the surface, so small horizontal cracks are sometimes visible in the centre of each bed joint – as in Fig. 2 – and vertical cracks may appear on the external elevations of thick masonry. Rendered masonry often exhibits a network of cracks termed ‘map-cracking’ or cracking that follows the mortar joints.
The susceptibility of mortar to sulphate attack (and frost attack or a combination of the two) can be tested using the Harrison technique, which has been standardised by RILEM (1998a). Sulphates rarely affect the units but precautions are advisable when building in ground containing sulphates (BRE Special Digest 1 (1996)) or constructing flumes and tunnels to carry contaminated effluents (WRC, 1986).
A special type of ‘engineering quality’ concrete brick is available which is designed to be stable in effluents (Concrete, 1986). These units are manufactured to a high strength and low permeability with sulphate resisting Portland cement as the binder. Mundic concrete blocks made using tin-mine tailings as aggregate in southwest England have suffered attack from indigenous sulphates (Bomley and Pettifer, 1997; RICS, 1997).
The effects of acids, e.g. rain run-off from peat moors, industrial or agricultural pollution, on cement-based products has been discussed in formar articles. Fired clay products are normally resistant to acids.
Chlorides can have a weakening effect on calcium silicate units but have little effect on mortars, clay units or concrete masonry units. They also catalyse the rusting of embedded steel even in alkaline conditions (see below).
Corrosion of Embedded Metals
Figure 3 shows a corroded masonry wall tie; de Vekey (1990a, 1990b) has given specific coverage of wall-tie corrosion. More details on all the chemical processes described above are given in Yu and Bull (2006).
Erosion processes such as wind and water scour attack both units and mortar but erode the softer of the two at a faster rate. Freeze–thaw frost attack and salt crystallisation are complex cyclic erosion processes where the susceptibility is dependent on pore size distribution and the number of cycles, and not simply on hardness, strength and overall porosity.
Freeze–thaw damage is one of the principal eroding agents of porous materials, including masonry units and mortar, exposed to normal exterior conditions.
Quite clearly it will not affect masonry buried more than a few feet and so will not affect foundations, the insides of tunnels away from portals or buried culverts. It may affect any masonry exposed on the exterior of structures but is more likely to affect exposed components that become saturated.
Typical problem areas are civil engineering structures such as bridges, canal locks, earth-retaining walls and exposed culverts and parts of buildings such as parapets, copings, chimneys, freestanding walls and masonry between the ground and the damp-proof course.
Freeze–thaw attack is due to the stresses created by the 8% expansion of water on freezing to ice in the pore systems of units and mortars and thus only occurs in water-saturated or near-saturated masonry. A pictorial explanation of one of the classic mechanisms is given in Yu and Bull (2006). Typical effects are the spalling (scaling) of small pieces of either the unit or the mortar or both forming a layer of detritus at the foot of the wall, as in Fig. 4.
Clay bricks, particularly lightly fired examples of types made from some shales and marls, are especially susceptible and tend to delaminate. Semi-dry pressed bricks made from Oxford clay tend to break down into grains the same size as they were made from (e.g. around 2–4 mm diameter).
Old types of solid clay brick with ellipsoidal drying microcracks tend to spall from the boundary between the heart and the outside of the brick. Some natural stones with a large proportion of fine pores are also susceptible.
Modern perforated clay bricks are generally more frost-resistant because the more open structure allows more even drying and firing with less chance of drying/firing cracks forming. It can be puzzling as to why some building materials are susceptible to freeze–thaw attack while other, apparently similar, materials are not. Clearly materials such as glass, plastic and metals, which are totally non-porous to water, are not affected.
Materials such as well-fired ‘glassy’ engineering clay bricks, well-compacted concretes, low-porosity granites, marbles and slates are also affected very little by frost since the tiny volume of water that can penetrate will cause only a trivial level of stress on freezing. Materials with water absorptions ranging from about 4 to 50% tend to suffer damage but do not do so invariably.
Closed-pore materials such as aircrete (AAC) and air-entrained mortars, which are not easy to saturate, are generally resistant as are materials such as stock bricks and lightweight concrete blocks, with a wide range of pore sizes from very large to fine. This is probably because it is difficult to fully saturate the mixed pore system, the water filling the finer but not the larger pores.
Providing around 10% of the pore system remains air filled there is sufficient space for the ice crystals to expand into without damaging the structure. Materials – particularly some mortars, clay bricks and natural stones – having a limited range of pore sizes, usually of the finer sizes, tend to fail. Most of the older data on susceptibility to frost attack are based on experience in use but this is a very slow and inefficient way of evaluating new products. To try to speed up the process, accelerated freezing tests have been developed.
A typical example is the panel test (West et al., 1984), which was published as an international standard by RILEM (1998a) and as a draft CEN standard prEN 772-20 (1999). Work has also been done by Beardmore and Ford (Beardmore and Ford, 1986; Beardmore, 1990) and Stupart (1996) to develop maps that indicate the average number of days per year of combined driving rain and freeze–thaw cycling affecting different areas of the UK.
Crypto-efflorescence (Sub-florescence) Damage
This is basically the same process as efflorescence (see below) but at certain temperature/humidity conditions it occurs just below the surface of the masonry unit. The hydrated crystals of compounds such as magnesium and sodium sulphates growing in the pore structure result in a compression force in the surface layers and consequently shear and tensile forces at the boundary with the unaffected core. There needs to be a source of water or water containing salts and a surface that is sufficiently warm or well ventilated (or both) to encourage drying of the salt solution. In UK conditions it is more likely to affect clay brick or natural stonework than other units or mortars.
The typical appearance is similar to that of freeze– thaw damage but it will usually be associated with soluble salt crystals, as shown by Fig. 5. Accelerated test methods have been standardised by RILEM (1998b) and by BS EN 12370 (1999).
Abrasion by particles in wind and water probably acts more in concert with other processes than in isolation. Likely areas for such erosion are bridge columns founded in riverbeds and buildings near road surfaces where splash up can occur from vehicles. All types of marine/hydraulic structures such as dams, culverts, lock walls, flumes, etc., where high-velocity flows can occur, may suffer from localised abrasion/ cavitation damage known as ‘scour’.
Stress effects normally cause cracking of varying types but the effect is on the masonry composite and not on the individual components. There are some problems related to faults in manufacture of the units, particularly under- or over-firing of clay bricks and the inclusion of foreign particles in bricks or other types of unit. A good range of coloured illustrations of problems and corrective strategies is given in BRE Digest 361 (1991) and Cracking in Buildings (1995).
This is a staining process caused by dissolution of soluble salts such as sodium, potassium, magnesium and calcium sulphates within the masonry pores by rain, groundwater or construction water from within brickwork which then crystallise on the surface as an off-white powder or encrustation. The surface may be any surface from which drying occurs.
It is commonly the external facade but may be the interior of solid walls particularly those in contact with earth, e.g. cellar walls and ground-floor walls in older structures without damp courses. The salts commonly derive from clay bricks, but may also come from groundwater or stored materials. Figure 6 shows typical effects.
This is caused by calcium hydroxide leaching from mortar or Portland cement-based concretes and being carbonated at the surface to form a white deposit of calcium carbonate crystals. It may also result from the dissolution of calcium carbonate in carbonated rainwater to form calcium bicarbonate, which reverts to the carbonate at the surface (stalactites grow by this mechanism). It is most commonly seen as white ‘runs’ from mortar joints on earth-retaining walls, see Fig. 6 but can occur on the walls of buildings. It is often seen on hollow/perforated unit masonry and accumulates in the holes as the rainwater seeps slowly out via porous mortar or concrete units.
Iron compounds can be present in bricks, concrete units or mortars. They give little problem if well distributed, as in many red-brown sands, but will give ugly brown run-off stains if present as larger discrete particles. Iron may be present in clay deposits in the form of pyrites and is sometimes a constituent of aggregates and additives to clay.
Coloured deposits, usually white/green to brown/ black or orange, can build up owing to the growth of algae, fungi, lichens, mosses etc. Such deposits only form if the masonry is wet for significant periods and is frequently the result of blocked down pipes or leaking gutters. Differential deposition can result from water streaming unequally off features such as windows and mullions, and this can become noticeable owing to the colour contrast. More information and illustrations of staining are given in BRE Digests 441, 460 and 508.