REVIEW OF FACTORS AFFECTING THE DURABILITY

OF REPOINTING MORTARS FOR OLDER MASONRY

MADELEINE ROUSSEAU INSTITUTE FOR RESEARCH IN CONSTRUCTION, NATIONAL RESEARCH COUNCIL CANADA, OTTAWA, ON

B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 R OUSSEAU • 7 7

ABSTRACT In older masonry structures, the mortar properties need to be compatible with the exist­ ing mortars and be “sacrificial” to the masonry. Inappropriate repointing materials and exe­ cution practices can result in permanent damage to older masonry assemblies, affecting their aesthetics qualities, historical value, performance, and service life. Public Works and Government Services Canada (PWGSC), which is faced with the maintenance and refur­ bishing of the Canadian government’s older building stock, including the Parliament build­ ings, has been grappling with this problem for years. PWGSC and the National Research Council of Canada’s Institute for Research in Construction (NRC­IRC) have carried out con­ siderable research in the last two decades to evaluate mortar mixes properties for applica­ tion in cold climates susceptible to freeze­thaw action and salt damage. This paper will describe critical elements of mix selection, on­site execution, and curing to obtain a durable mortar that has no detrimental effects on the masonry.

SPEAKER

MADELEINE ROUSSEAU – INSTITUTE FOR RESEARCH IN CONSTRUCTION, NATIONAL RESEARCH COUNCIL CANADA, – OTTAWA, ON

Madeleine Rousseau is an architect with 25 years experience in the area of moisture migration in exterior walls at the National Research Council Canada (NRC). For the last 10 years, Madeleine has participated in NRC research for durable repointing mortars that do not have a detrimental effect on older masonry, in the context of cold climates experiencing freeze­thaw action. In 1998, Madeleine also initiated a multi­disciplinary forum of masonry restoration specialists aimed at enhancing exchange of technical information on repointing mortar technologies. The working group, made up of trades, design professionals, building owners and managers, and manufacturers involved in masonry repointing work, has met twice a year since 1998.

7 8 • R OUSSEAU B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6

REVIEW OF FACTORS AFFECTING THE DURABILITY

OF REPOINTING MORTARS FOR OLDER MASONRY

INTRODUCTION WHAT IS REPOINTING MORTAR? strength mortars are often not appropriate The selection of repointing mortars for Repointing mortars replace the outer and may cause damage. older masonry in cold climates is a subject deteriorated or damaged mortar in masonry Challenges to mortar selection include: of debate. Standards for modern masonry joints. The deteriorated mortar needs to be • Original mortar materials are no construction are not necessarily appropri­ carefully removed without damaging the longer available. ate for the needs of older masonry struc­ masonry units. The resulting gap is then • Difficulty in determining the exact tures, which can have unique performance cleaned and filled (repointed) with a com­ composition and properties of the requirements. As such, determining an patible repair mortar to stop further deteri­ existing mortar. appropriate mortar mix to use in the repair oration and water penetration (Figure 2). • Issues of historic authenticity. of masonry buildings can be a challenge. Repointing mortar mixes are drier and • Lack of trades specializing in old Repointing mortars should be durable, stiffer than bedding mortars to reduce materials and techniques. Many of practical in application (e.g., workmanship shrinkage and avoid staining of the mason­ the materials and techniques for and quality control), and not have a nega­ ry. For the repair of older brick and stone producing and applying traditional tive effect on the durability of the existing masonry walls, there is an increasing use of mortars were lost and substitutes masonry. Durability is not only dependent traditional mortar mixes, largely driven by may have to be found and tech­ on the mortar mix used, but also on how it the wish to have compatibility with the orig­ niques learned again. is installed and cured (workmanship) and inal mortar used and the existing bedding • Little published information on the on the severity of the environmental expo­ mortar (from historic, aesthetic, and mater­ performance of traditional mortars sure, which in turn depends on weather, ial property perspectives). Modern, higher­ in the Canadian climate. design, construction, operation, and main­ tenance. Public Works and Government Services Canada ­ Heritage Conservation Directorate (PWGSC ­ HCD) is responsible for the main­ tenance and repair of a considerable num­ ber of older masonry buildings, including the Parliamentary Precinct in Ottawa (Figure 1). For the past two decades, PWGSC­HCD has teamed up with NRC­IRC to carry out research projects which will help to define performance objectives of repointing mortars and assess a variety of mortar mixes in order to determine appro­ priate mixes for use in the repair of older masonry buildings. This ongoing work has been carried out to help fill a need for clear performance objectives for repointing mor­ tars so that an appropriate repointing mor­ tar can be chosen for a particular building. Performance objectives then need to be translated into set criteria that in many cases are still a subject of debate and research. This paper reviews literature dealing with the selection and performance of mor­ tars used in the repair of older masonry. Particular emphasis is given to factors af­ fecting the resistance to freeze­thaw dam­ age, a major consideration in a cold climate. Examples from Parliament Hill in Ottawa will be given as illustration of the text. (See

Figure 1.) Figure 1 – The Center Block of Parliament Hill in Ottawa, Canada. B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 R OUSSEAU • 7 9 and at the mortar/unit interface). These, in turn, are caused by frost action, salt crys­ tallization (sulphates and chlorides – Figure 3), movement (settlement, differential, ther­ mal, and moisture), dissolution, environ­ mental pollution (acid rain), water migra­ tion (Figure 4), and biological attack in wet areas of the building. Moisture is the environmental factor commonly associated with most of these problems, although temperature also has a significant influence on the rate and extent of damage caused by the moisture. Temperature and moisture also directly affect expansion and contraction move­ ments in the masonry. Mortar can also

cause problems for the masonry units such Figure 2 – Example of a mortar joint with repointing mortar. as lime leaching out of the mortar causing staining (Figure 5) or spalling with too

• Traditional mortars, usually weaker POSSIBLE FORMS OF FAILURE strong a pointing mortar. than modern mortars, are less for­ Potential problems with mortar include Traditional mortars with high lime con­ giving of poor construction prac­ spalling, crumbling, efflorescence, biologi­ tent have more initial flexibility and higher tices; good quality control and site cal growth, and cracking (within the mortar porosity than modern mortars, and hence, supervision are needed they can better accom­ to ensure success. modate minor move­ • Difficulty in correlating ments in the wall with­ laboratory tests to actual out cracking. Cracks field exposure condi­ lead to increased in­ tions. gress of water into the • Few standards on use and testing of traditional

mortar mixes. Modern Below: Figure 4 –

standards have often Water­shedding

deleted reference to details deflect water

older mortars but this off the upper surface

trend is beginning to of the masonry.

reverse. Unfortunately, the

water is deflected

Mortars are basically com­ onto lower surfaces. posed of a binder, aggregate (sand), water, and additives. Traditional mortars usually have lime as the major compo­ nent of the binder. Today, many restoration mortars have a cement/lime binder – usually with more lime than cement. There is also increasing interest in the use of pure lime mortars, hydraulic lime mortars, and proprietary pre­mix mortars to which only water needs to be added. In contrast, binders for modern mortars have cement as the major component with either lime added in equal or less proportion, or with propri­ etary additives (e.g., masonry cement). Figure 3 – Salt damage to stone and mortar is apparent beside an

entryway where salt is used to remove ice in the winter. 8 0 • R OUSSEAU B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 masonry that in turn hang has less risk of dam­ affects durability. If age than mortar in a cracks do occur, they are chimney). more likely to occur along North America has a the weaker mortar joints large climatic diversity. and not through the ma­ Climates vary from hot sonry units (this is prefer­ summers to cold winters. able, simply because it is In winter, some areas can easier and cheaper to be cold and relatively dry, repoint mortar joints than while others have large to repair or replace dam­ amounts of precipitation. aged brick or stone). Fine Regions with more precip­ cracks may reseal due to itation (snow and rain) redeposition of lime with­ and many freeze­thaw cy­ in the crack. cles are more severe for the masonry than regions

DURABLE that tend to stay cold or

REPOINTING mild over the entire win­ Repointing mortars ter. Building orientation should be as durable as affects the local climate possible without causing (e.g., prevailing driving damage to the existing rain and solar radiation). masonry (a dense mortar Not much can be done could retard the drying of about the weather, but the masonry assembly the design of the building, and cause frost or salt especially building de­ crystallization damage in tails, can have a large the bedding mortar or influence on the local masonry units). It is environment experienced preferable to repoint mor­ by the masonry. tar joints at more frequent Attention to water­ intervals than to have to shedding architectural Figure 5 – Lime leached out of the mortar joints onto the masonry repair damaged masonry features, including their surface leaves an unsightly stain. Had the lime had chance to units. A building and its maintenance, will reduce carbonate before being exposed to the weather, this might not have components should be the risk of occurrence of occurred. durable enough to per­ high moisture levels need­ form the required func­ ed for frost damage tions in its service environment over the will be used. The severity of the environ­ (Maurenbrecher, 1998). Melted water from design service life without unforeseen cost ment depends both on the local weather snow is a major cause of frost damage in for maintenance or repair (CSA 1995). The and the exposure of the masonry elements. freezing climates (snow melt during the day design service life for masonry is usually 50 For example, relatively weak mortars can is absorbed by the masonry underneath, to 100 years. A normal expectation for survive well in freezing climates, provided which freezes at night; this cycle may be pointing mortar is at least 30 years, but they are protected from excessive moisture kept up for many days). Cracks, including preferably 50 to 100 years (Mack & Speweik (mortar in a wall protected by a roof over­ hairline cracks at the mortar masonry unit 1998). If a weak mix is required in more exposed areas, weathering of the mortar and more frequent repointing should be accepted as part of the maintenance of the masonry (BBA 1999). The durability of mortar not only depends on the materials used, but on how they are installed and cured (workmanship) and on the severity of the exposure, which in turn depends on the local climate, design, construction, operation, and main­ tenance (Figure 6).

SEVERITY OF THE ENVIRONMENT Mortar selection must take into account the severity of the environment in which it Figure 6 – Factors affecting durability. B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 R OUSSEAU • 8 1 interface, will also allow increased water Historic Scotland (1995) recommends sharp added to the quicklime to produce lime ingress. This highlights the need for regular sand with a balanced range of particle putty or, in modern production, a dry, maintenance and attention to shrinkage shapes (coarse sand is also referred to as hydrated lime powder, Ca(OH)2. Lime in a and bonding properties of pointing mortars. sharp). Although sharper sand makes the mortar mix hardens by carbonation; the

Moisture can also come from inside the mix less workable, it provides more inter­ lime recombines with CO2 in the air and building. For example, in cold weather, air locking between particles, reducing shrink­ reverts back to limestone (to harden, lime exfiltration of indoor air of humidified build­ age. On the other hand, Mack & Speweik mortars need access to both air and mois­ ings can lead to interstitial condensation in (1998) recommend rounded sand particles ture). Lime mortars take much longer to localized areas of the masonry assembly because they give the mortar better plastic­ gain strength than Portland cement­based (based on the path and exit locations of the ity, making it easier to compact into the mortars. Pozzolanic additives can be added air). In addition to the risks associated with joint. Rounded sand was often used in his­ to the mortar; these react directly with lime high moisture levels within the masonry, toric mortars. The easier compaction makes and water, allowing part of the lime to hard­ the subsequent outward drying of the mois­ it more likely the joint is filled properly en faster without the need for carbon diox­ ture may carry efflorescence salts to the (good contact between the pointing mortar ide from the air (hydraulic property). surface. and the existing bedding mortar improves In the late 1700s, it was discovered that moisture transfer and durability). The need burning limestone with clay impurities

MATERIAL CHARACTERISTICS for less pressure to fill the joint is also resulted in limes, which could also gain Without going in detail about the char­ preferable from the mason’s point of view, some of their strength by reacting directly acteristics of the materials that make mor­ reducing the risk of repetitive strain injury. with water (hydraulic limes). This also tars, here is a discussion of important ele­ The moisture content of the sand affects required increased firing temperatures (up ments that can affect the site execution as its volume. Damp sand (2­6% water mois­ to 1250˚C). Hydraulic limes range from well as the in­service performance. ture content) occupies more volume than weakly hydraulic to eminently hydraulic dry sand. A mix based on damp sand would (they are still made in Europe and by one

SAND likely be stronger and potentially experience manufacturer in the U.S.). Sand is generally selected by what is more shrinkage than a mix with the same Further increases in clay content and locally available and what color and texture proportions of ingredients using dry sand. firing temperatures resulted first in natural are desired for the mortar, but sand cleanli­ Consequently accounting for this property, cements, and finally, Portland cement ness, grading, and particle shape should it is good practice for the mortar mix design­ (named after its resemblance to Portland also be considered, for they can have an er/specifier to review and possibly adjust limestone). Portland cements require a fir­ important influence on overall durability. the mortar mix recipe the first day of mor­ ing temperature of around 1450˚C. They Sand is normally made up of particles tar mixing (as well as regularly afterwards, gain nearly all their strength by reaction with a range of sizes, the smaller sizes fill­ depending on the weather conditions (e.g., with water. Mortars using these cements ing in spaces between the larger sizes. In heat spells), once the dampness of the sand are stronger and gain their strength more well­graded sands, the void space left over available on site is defined (usually with a rapidly which also meant construction is assumed to be about one­third of the test consisting of drying the sand in an could proceed faster. As binders become sand volume (the space varies depending on oven). Trial mortars are mixed with an effort more hydraulic, mortars made using them the sand particle shape and grading). to define the quantity of water to be used become denser and less porous. They are Enough binder paste is then added to fill depending on the aggregate that will be therefore better at keeping water out, but this void and provide a film between the used on site. North American mortar stan­ should water get into the masonry, it will sand particles to give the fresh mortar plas­ dards (ASTM 2006 and CSA 2004) refer to also take much longer to dry out. Portland ticity. This explains the binder to the aggre­ damp sand because that is usually the con­ cements come in a variety of types (normal, gate ratio of 1:3 seen in most specifications. dition of sand on building sites. white, and sulphate­resistant Portland This ratio will vary, depending on the grad­ Crushed limestone, crushed low­fired cements). Portland cement­based masonry ing and particle shape, and on the binder. brick, and expanded vermiculite materials cements have been used in pointing mor­ Mortar with too little binder will have poor are among other aggregates which are tars in Canada. water retention and workability, while too sometimes added to mortar (Historic Hydrated lime powder from different much binder will lead to higher strengths, Scotland, 1995); however, these materials manufacturers can have different bulk den­ increased shrinkage, and risk of cracks. may have different water absorption rates sities (largely due to particle size). Thus, vol­ The effect of sand grading on durability than sand and this must be taken into con­ ume batching results in different weights of needs further investigation. Coarser sand sideration. lime. The finer the lime, the less weight for gradings for pointing mortars in modern a given volume. Will the lower weight of the masonry improved frost resistance (Elsen et BINDERS finer lime still provide the same workability al., 1993). Grading also influences mortar The most common binders in mortar are to the mix because of the increased fine­ shrinkage and the bond between mortar lime and Portland cement. Lime is obtained ness? On the other hand, will the long­term and masonry units. from limestone (CaCO3) when it is burned at strength of the mortar be less because of

Sands can come from natural deposits 900˚C to drive off the carbon dioxide (CO2), the lower weight of lime? In the case of lime or are manufactured by crushing stone. In resulting in quicklime CaO. (Lime is also putty, it usually contains more lime than an general, crushed sand is considered most obtained from magnesium and dolomitic equivalent volume of dry, hydrated lime. angular (sharpest), while quarry sands vary limestone; dolomitic contains 35 to 46% Lime putties made from hydrated limes from a semi­angular to rounded shape. MgCO3, common in North America.) Water is from different manufacturers were found to 8 2 • R OUSSEAU B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 have 16 to 56% more lime than the equiva­ mortars. From these, specific criteria can be the drying of the masonry assembly lent volume of dry, hydrated lime (Phillips developed. There is no miracle mortar mix through the mortar joints. This is 1994, Maurenbrecher et al., 2000). If no suitable for all masonry; the design of a especially important in masonry attention is paid to these aspects, the quan­ mortar mix should be adapted to the partic­ with dense masonry units. A more tity of lime in a mortar can vary, leading to ular masonry assembly under consideration porous mortar will also encourage mortars with differing properties. The sig­ (e.g., in terms of the existing mortar and any salts in the masonry to migrate nificance of these must be resolved. masonry units, and environmental expo­ out through the mortar instead of Ingredients may also be added to the sure). Typical mixes are given by Mack and the masonry units (salts can cause mortar to change its color or improve work­ Speweik (1998), BRE (1999), Historic crumbling, spalling, or efflores­ ability, water retentiveness, water repellan­ Scotland (1995), and BA (1999). There is lit­ cence). cy, bonding with masonry units, and frost tle recent documented data on the use of • Minimize shrinkage after pointing (a resistance. Additives are usually discour­ pure lime and hydraulic lime mortars in maximum of 1 mm/m was recom­ aged in restoration mortars, except for pig­ Canada, so they must be used with caution mended by Knöfel et al., 1993). Use ments for color and air­entraining agents in until more experience is gained with their well­graded, washed sand with no frost­susceptible areas. Air­entraining use. IRC/NRC, in conjunction with PWGSC­ clay fines; low water­to­binder ra­ agents can enhance frost durability. Ex­ HCD, has done some preliminary investiga­ tios; and proper curing to minimize cessive air entrainment, however, reduces tion on durability of hydraulic, lime­based evaporation in the first days after the bond of mortar to the masonry unit repair mortars (Maurenbrecher et al., the mortar is laid. Sand particle (normal recommended range is 10 ­18% 2005). Initial findings showed that shape may also have an effect (see air). Hydrated lime can be obtained with an hydraulic, lime­based mortars could pro­ section on materials). integral air­entraining agent; masonry and vide a durable repair mortar. They are, how­ mortar cement and most pre­mix mortars ever, much more sensitive to mixing and • Promote good (not necessarily have it added, too. This is often the best way curing conditions, so care must be exer­ strong) bond with full contact be­ to get an air­entraining agent mixed into a cised if a good end result is to be obtained. tween mortar and masonry units mortar. Trying to add dry or liquid air­ When selecting a mortar mix, the follow­ and existing bedding mortar. This entraining agents separately on site can be ing performance requirements should be can be achieved by clean removal of a difficult job, as very small quantities of taken into consideration. the deteriorated mortar when cut­ additive are needed for a typical mortar mix • Ensure the compressive strength is ting out, good compaction of the and it can be difficult to get complete dis­ lower than that of the existing fresh mortar, and proper curing. persion of the agent into the mortar. These masonry units, and similar to or Good bond and low shrinkage agents do not work well in very dry pointing lower than the existing bedding mor­ reduce the risk of fine cracks form­ mortar mixes, as they require a minimum tar (if the pointing mortar is too ing at the interface between the amount of water and good mixing to acti­ strong, stress concentrations could masonry units and mortar. Most vate the entraining agent. cause spalling of the masonry water infiltration through a masonry units). assembly occurs at this interface PRE­MIXED MORTARS In future repairs and restoration, and at poorly filled joints. Examples Use of pre­mixed mortar ingredients to weaker mortars are also easier to of minimum recommended flexural which only water needs to be added gives remove without damaging the bond strengths are 0.2 MPa (Knöfel the greatest control over mortar consistency masonry units. Weaker mortars et al., 1993) and 0.3 MPa (with Ne­ on site. Such mixes are more expensive and have less stiffness and greater pean sandstone; Suter et al., 1998). the exact components are often proprietary. creep, allowing accommodation of Caution should be taken when using pre­ • Provide resistance to frost action larger movements without cracking. mixed mortars to avoid scooping the quan­ where needed. An air­entraining Recommendations for maximum tity of dry material needed for a repointing agent added to the mortar will strength have ranged from 8 MPa job off the “top of the bag.” Finer ingredi­ improve frost resistance. (for Nepean sandstone; Suter et al., ents, such as additives or binders, may High lime content mortars have less 1998) to 10 MPa (Knöfel et al., have settled to the bottom during bagging resistance to freeze­thaw action 1993). This still leaves questions on and shipping. It is usually recommended to when they become saturated (they assessment of strength. Mortar cube mix up all the contents of a single bag for a are most vulnerable early in their life strength can be quite different from project or – if smaller quantities are needed because they take a longer time to the strength of the mortar in the – to blend the dry ingredients of a whole bag harden). On the other hand, the joint. In addition, strength varies in a mixer and then subdivide it into small­ more porous mortars tend to dry with age, the high­lime mortars er sealed containers for smaller repointing faster, thus reducing the risk of gaining strength gradually. jobs. damage. Great care must be taken • Aim for a water absorption and va­ in selecting mortars for areas of por transmission rates similar to or DESIGN CONSIDERATIONS severe exposure such as chimneys, greater than those of the bedding The designer needs to take into account parapets, free­standing walls, exteri­ mortar and masonry units. all the factors affecting durability. or steps, and masonry below or at

Performance requirements provide a good The pointing mortar should facilitate ground level (base selection on expe­ base for assessing appropriate repointing B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 R OUSSEAU • 8 3 rience and/or testing). masonry unit (depending on type of unit, use, provided they are kept damp and air is amount of pre­wetting, and the water reten­ excluded (Historic Scotland, 1995). Mortars • Provide resistance to salts where tion capacity of the mortar). This can affect with hydraulic binders have to be used needed (e.g., sulphates). Use of sul­ the bond of the mortar to the masonry within a certain period. Mortars with phate­resistant Portland cement units. Portland cement need to be used within two reduces this risk. A study on modern bricks and pointing to three hours of mixing. The mortar may If sulphates are present in existing mortars found the water remaining in the also be allowed to stand for a period after masonry, they can react with binder mortar affected the frost durability: too mixing to allow pre­hydration (and thereby components (e.g., uncarbonated much water in the mortar decreased frost reduce shrinkage and improve workability). lime and components of Portland resistance; too little water affected hydra­ ASTM C270 recommends 1­1/2 to 2 hours cement) when the masonry is damp tion and reduced strength (Elsen et al., after mixing with sufficient water to pro­ for extended periods of time (Har­ 1993). duce a damp mix. After that period, further rison, 1990; BRE, 1991). water is added until the right consistency is

• Ensure compatibility of the thermal MORTAR MIXING obtained. Another document does not men­

and moisture expansion properties Batching (measuring) of mortar ingredi­ tion this waiting period (BRE, 1999). The

of the repointing mortar to existing ents by weight gives better consistency than cement content and type of lime may also

masonry. batching by volume, but batching by vol­ affect the need for a waiting period. ume is still the most common procedure on

• Encourage good workmanship by site. Volume batching introduces larger MORTAR APPLICATION making mortars practical in applica­ variations in quantities because the level of Good compaction of the pointing mortar tion. compaction in measuring containers varies in the joint is important to good perfor­

• Use contractors and masons experi­ with the individual doing it. In addition, the mance. For deeper joints, it is usually done

enced in the conservation of older volume of sand can vary with moisture con­ in more than one layer. The final finish of

masonry. tent. Pre­mix mortars avoid most of the the mortar joint surface affects its water­ batching problems. shedding capabilities. The surface should When lime is used in the mix, it can be not extend out over the surface of the ON­SITE PRACTICE On­site practice includes mortar joint added in the form of hydrated lime powder, masonry units (thin sections of mortar preparation as well as mortar mixing, point­ lime putty, or “coarse stuff” (lime, sand, and extending onto the face of the unit easily ing, and curing (BRE, 1999; Mack and water mixed ahead of time). The latter two crack and collect water). Standard finishes

Speweik, 1998; Historic Scotland, 1995). All give the best workability because the lime range from a concave finish (best compac­ have an important influence on the long­ particles are fully wetted beforehand. The tion and weathertightness) to a raked joint term performance of the mortar. Super­ difference in lime content in dry lime and (worst) (BRE, 1999). For historic masonry, vision and quality control will help ensure putty lime mentioned earlier must also be many other finishes have also been used. the performance standards are reached. taken into account.

Mortar may be mixed by hand (only CURING CONDITIONS suitable for small amounts), in a standard Curing conditions for freshly pointed JOINT PREPARATION Raking out and cleaning of the joints paddle mortar mixer, or a mortar mill. The joints have led to much debate, especially must be carefully done to ensure no dam­ mortar mill is often used for high lime mor­ when there is the risk of frost (Historic age to the masonry units and ensure a tar mixes. Mixing time must be controlled, Scotland, 1995; Mack and Speweik, 1998; clear, rectangular space for the repointing especially for mortars containing air­ BBA, 1999; BRE, 1999). mortar ( ). Then there is more likely entraining agents (to avoid excessive air Rapid drying out of the mortar should Figure 7 to be a good contact between the pointing content). After mixing, the mortar is often be avoided. It can bring lime to the surface mortar and the masonry unit and the exist­ allowed to stand for a while before use. Pure and increase the risk of shrinkage cracks; it ing bedding mortar. Poor con­ lime mortars can be kept for months before also may not leave enough moisture for cur­ tact inhibits moisture trans­ ing of hydraulic components in the mortar fer. The typical depth to be (Historic Scotland, 1995). Common recom­ raked out is twice the width of mendations include a damp cure of two to the joint, but more material four days (longer for pure lime mortars), or may need to be removed in protection for seven days (Mack and order to get back to “sound” Speweik, 1998; BRE, 1999). Actual times material before applying the will depend on environmental conditions. pointing mortar. Damp curing may be achieved by wet

Before pointing, the joint burlap covered in plastic. Regular misting is is usually pre­wetted to limit an alternative. Water should not run off the its water absorption rate. How joints while doing this; otherwise, staining best to do this is still a sub­ may result from lime leaching out of the ject of debate. At the time of mortar. New repointing should therefore be pointing, water is absorbed protected from rain. from the mortar by the Where there is a risk of frost, protection

Figure 7 – Raking out mortar joints in brickwork. from freezing for a minimum of seven days 8 4 • R OUSSEAU B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 is recommended for weaker mortars (BRE, 1999). Air entrainment will provide added protection. The effect of only seven days’ initial protection on frost durability and long­term strength gain needs further investigation. Pointing should preferably be done well ahead of winter. High lime mortars will slowly gain strength as the lime within the mortar car­ bonates. The outside surface of the joint carbonates within a few days, but within the joint, the process is much slower, taking a year or more, depending on the porosity of the mortar and masonry unit, the depth of the pointing mortar, and the environmental conditions. Favorable conditions for car­ bonation are relative humidities in the range of 60 to 75% or repeated wetting and drying. Full carbonation is likely to improve durability, including resistance to sulphate attack (Harrison, 1990) and frost resistance.

TESTING AND STANDARDS Performance requirements and durabil­ ity can be assessed by documenting actual performance in buildings, and by tests in the laboratory and in the field.

PRE­CONSTRUCTION TESTING It is difficult to assess the exact compo­ sition of mortars used in the past; com­ pounding this problem, there is also a lack of information on the properties of such mortars and their influence on performance in the Canadian climate. There is a need for standard tests, so that results from differ­ ent laboratories and countries can be com­ pared. A RILEM committee on the charac­ terization of old mortars with respect to their repair is addressing the issues of sam­ Figure 8 – Regular inspection of the masonry can be conducted using a hydraulic pling, analysis of physical and chemical lift to access the structure. characteristics, damage types, testing, and case studies.

Testing should take into account the QUALITY CONTROL DURING actual buildings should be surveyed, moni­ performance of the mortar in combination CONSTRUCTION tored, and documented in Canadian condi­ with the masonry in which it will be used. There is a need for simple and quick tions.

Testing mortar using small masonry wal­ quality control tests on site. One example, lettes gives a better representation of actual employed by PWGSC­HCD on all of their STANDARDS practice. NRC/IRC, in association with repointing projects, which has worked well The ASTM mortar standard has a short

PWGSC­HCD, is investigating the durability to assess the workability of a particular section on pointing mortar (ASTM C270­ of pointing mortars for stone masonry. mix, is the cone penetration test (ASTM 06). There is currently a draft standard

Small masonry prisms are used to assess C270). This provides an immediate indica­ being put together by ASTM to develop the freeze­thaw durability of pointing mor­ tion of the workability of the mix and pro­ guidelines for repair mortars. The CSA mor­ tars (Fontaine et al., 1998; Maurenbrecher vides a means of controlling consistency of tar standard (A179­04) currently includes a et al., 2000; Suter et al., 1998; Thomson et fresh mortar and prevents excessive non­mandatory appendix that was devel­ al., 1998). Improving the freeze­thaw test to strength development that could occur from oped to provide some guidance on repair more accurately reflect conditions in prac­ a dry mortar mix. mortars for older masonry. More guidance tice is also an objective (e.g., freezing from is needed in both these standards on the one side only, rate of freezing). IN­SERVICE PERFORMANCE repointing of older masonry. The performance of pointing mortars in

B UILDING E NVELOPE T ECHNOLOGY S YMPOSIUM • O CTOBER 2 0 0 6 R OUSSEAU • 8 5

MAINTENANCE REFERENCES Maurenbrecher, A.H.P., G.T. Suter, K. Good durability is not only a design and ASTM. “C270, Standard Specification Trischuk, L. Fontaine. “Contribution construction consideration. Ongoing main­ for Mortar for Unit Masonry.” to Pointing Mortar Durability.” tenance has a large influence on perfor­ American Society for Testing and Historic Mortars: Characteristics and mance as well. Regular visual inspections, Materials. U.S.A. 2006. Tests. Editors P. Bartos, C. Groot, coupled with a maintenance guide, would BBA. “h/2 Blue Lias Hydraulic Lime and J.J. Hughes. RILEM Publica­ be ideal (Figure 8). Failure of water­shed­ Mortar. Detail Sheet 3.” Agrément tions. Pp. 361­369. 2000. ding elements, such a gutters and down­ certificate 99/3581. British Board of Maurenbrecher, A.H.P., K. Trischuk, M. spouts, can result in rapid deterioration, Agrément. P. 5. 1999. Subercaseaux, G.T. Suter. “Prelim­ especially in cold climates. Prompt repair of BRE. “Building Mortar.” Digest 362. inary Evaluation of the Freeze­thaw these elements, along with the damaged Building Research Establishment. Resistance of Hydraulic Lime Mor­ mortar joints, will greatly reduce the extent U.K. P. 8. 1991. tars.” Repair Mortars for Historic of further damage. BRE. “Repointing External Brickwork Masonry – RILEM workshop ­ pro­ Walls. Good Repair Guide 24.” ceedings pending. 2005.

CONCLUSION Building Research Establishment. Phillips, M. “A Source of Confusion This paper highlights some of the issues U.K. P. 4. 1999. About Mortar Formulas.” APT affecting the durability of repointing mor­ CSA (R2001). “S478 Guideline on Bulletin. Vol. 25, No 3­4. Association tars. With an increase in popularity of Durability in Buildings. Canadian for Preservation Technology. Pp. 50­ repair and restoration of old masonry build­ Standards Association. Rexdale, 53. 1994. ings, there is pressure to produce design Ont. 1995. RILEM. Historic Mortars: Characteristics and construction guidelines for practition­ CSA. “A179 Mortar and Grout for Unit and Tests. Editors P. Bartos, C. ers to use for best results. Care is needed in Masonry.” Canadian Standards Groot, and J.J. Hughes. RILEM determining which mortar mixes are best Association, Rexdale, Ont. 2004. Publications, France. P. 459. 2000. for a particular building. It is recommended Elsen. J., N. Lens, R. Gérard. “Gélivité SBR­CUR. “De Kwaliteit Van Voegen in to use experienced practitioners in the field des Mortiers de Maçonnerie dans les Metselwerk.” (The Quality of who manage to negotiate the minefield of Façades en Briques (Frost Resis­ Pointing in Brickwork). Report new products, on­site practices, and re­ tance of Masonry Mortars in Brick SBR299/CUR 98­3. Stichting search documentation. This has led to a rel­ Façades).” CSTC Magazine. Bel­ Bouwresearch, the Netherlands. atively young research field exploring the gique. Automne. Pp 3­11. 1993. 1998. interrelation of these factors. Fontaine, L., M.L. Thomson, G.T. Suter. Suter, G.T., M.L. Thomson, L. Fontaine. Over the last 20 years, PWGSC­HCD “In Search of Durable Repair Mortar “Mortar Study of Mechanical has conducted research in conjunction with for Historic Masonry in Northern Properties for the Repointing of the IRC/NRC in an effort to increase the level of Climates: a Prologue.” APT Bulletin, Canadian Parliament Buildings.” understanding of materials and methods of Vol. 29, No 2. Association for APT Bulletin, Vol. 29, No 2. repair of older masonry structures. This is Preservation Technology. Pp 39­40. Association for Preservation the only government branch doing this type 1998. Technology. Pp. 51­58. 1998. of work in an effort to preserve Canada’s Harrison, W.H., T.K. Bowler. (1990). Thomson M.L., G.T. Suter, L. Fontaine. historic masonry buildings. There remains Aspects of mortar durability. Br. “Frost Durability Testing of Stone a need for documented data on perfor­ Ceram. Trans. J. 89, pp 93­101. Masonry Mortars: Restoring the mance and testing of repointing mortars in Historic Scotland. “Preparation and Use Canadian Parliament Buildings.” relation to the Canadian climate and on­site of Lime Mortars.” Technical Advice APT Bulletin, Vol. 29, No 2. Asso­ practices. Without documentation and Note 1. Edinburgh. P. 60. 1995. ciation for Preservation Technology. analysis, codes and standards cannot be Knöfel, D. and M. Huesmann. “Mörtel Pp. 42­49. 1998. improved to meet the current demands. zur Fugeninstandsetzung an His­ This paper is a step towards document­ torischen Bauwerken” (Mortar for ing the current issues and, as can be seen, Pointing Maintenance in Historic there are many gaps. IRC/NRC has made Structures). Bautenschutz + Bau­ efforts to address these gaps through sanierung 16. Pp 30­34. 1993. research and information exchange. A Mack, R.C. and J.P. Speweik. working group of architects, engineers, “Repointing Mortar Joints in His­ materials suppliers, material scientists, toric Masonry Buildings.” Preser­ researchers, contractors, and masons vation Brief 2. Heritage Preservation meets twice a year to discuss current chal­ Services, National Park Service, U.S. lenges and areas requiring further work. To Dept. of the Interior. P. 16. 1998. support this effort, IRC has also developed Maurenbrecher, A.H.P. “Water­shedding a Web site that provides information on its Details Improve Masonry Perfor­ current masonry projects and selected bib­ mance.” Construction Technology liographies on relevant topics (www.nrc.ca/ Update 23, Institute for Research in irc/bes/masonry). Construction, National Research Council, Canada. P. 4. 1998. 8 6 • R O U S S E A U B U I L D I N G E N V E L O P E T E C H N O L O G Y S Y M P O S I U M • O C T O B E R 2 0 0 6