The use of lime mortar in building conservation

A selection of sands and other aggregates used for preparing mortars at Nidaros Cathedral, Norway (Photo: James Miller).

Choose lime; favour site mixing but do not reject pre-mixed solutions; do not focus on strength; take care in choosing the sand; if appropriate, analyse the original mix; and other good advice.

1 Introduction
2 Conservation remedial work
3 Modern codes of practice
4 Movement
5 Mortar specification
6 Summary: how to choose a mortar
7 Related articles on Designing Buildings

Introduction

It is often said by engineers engaged in conservation that masonry mortar is only there to even out the surfaces of bricks and stone: to stop any hard points from spalling when they press together and make walls easier to build.

This principle certainly applies in mass walls where the whole structure is in compression, which accounts for the majority of historic buildings and structures. As we approach the modern age, walls become thinner and mortar is required to act in tension and shear, as well as compression, but this article considers the older forms of construction. For these, the physical strength of the mortar has very little relevance to the function of the wall.

In the UK and across Europe our early mortar technologies were based on lime, as opposed to the gypsum and occasionally bitumen mortars found elsewhere. The quality of early burnt limes and the consistency of excavated sands were variable, limited by the technology available at the time. Unlike today, kilns did not fire at a uniform temperature and the lime produced would have different properties, depending on in which parts of the kiln it was burnt. The limestones would vary considerably across the country, resulting in particular in varying hydraulic properties. Some might set in water, some might not.

As it was impractical to transport sands and other aggregates long distances before the canals or railways were built, most of those found in historic mortars were locally sourced. The properties of the resulting mortar would be dependent on the skill of the artisans in using the materials at their disposal.

Conservation remedial work

While it is tempting to hold up historic mortars as fine examples that are well-formulated, in fact many were made with poorly graded sands. The conservation engineer, looking at the internal structure of the mortar matrix, is faced with the philosophical question: ‘do we replicate a poor original or do we replace with a well-designed mortar?’ There is no simple answer; every case must be considered in its own right.

Although the focus is generally on the lime binder, the aggregate is the most significant element in the matrix of a mortar. Sand is usually the principal component, but the aggregate is often found to include fragments of waste masonry materials: crushed stone, brick and old mortar. This ‘filler’ has greater influence on durability, colour, frost resistance and strength than the choice of lime.

The burnt limestone will have varied across the country. Reference to the geological maps will point to whether it may be a chalk, blue lias or a grey lime. Unfortunately, changing use and economic fortune have removed many of the historic limestones from production. On a commercial scale we are now left in this country with relatively pure calcium carbonate which is not hydraulic, such as found in Buxton, or foreign hydraulic limes.

There are many good historic references for mortars and plasters, dealing with what was thought good practice and appropriate in areas, and with the materials available at that time. The buildings these applied to were generally thick walled and the mortar very rarely acted in tension.

One of the best reference books on historic mortars was written by WJ Dibdin and published by the RIBA in 1911. This lists the results of a whole range of tests on different mortars, limes, sands and additives such as the natural pozzolan trass. This information, useful for those wanting a fuller understanding of the properties of mortars, puts into context the limited scope of the current codes of practice. For example, it sets out typical ratios of binder to filler, a common one being 1:2.5.

Modern codes of practice

The 20th century saw an increasing codification of the production of materials and a requirement that materials met a standard to be considered acceptable. These codes are influenced by the manufacturers of the materials that they are codifying and so might not always be independent.

The production of the lime is covered by the code EN459, which sets out the properties of all the different limes from CL70 through to NHL 5.0. The code requires the tests to be carried out with a ‘standard’ sand, one that is not well graded. EN459 also stipulates that the strength tests should be carried out after 28 days, whereas lime mortars develop strength over at least 90 days, generally up to 360 days. This means that, some of the basic premises of the code do not align well with the formulation of historic mortars.

BS5268 and Eurocode 6 are the codes used for the design of masonry, including both mass masonry and thin-walled cavity construction. They are used to check the strength of walls, which is only occasionally required in historic restoration because of the relatively low masonry stresses found in traditional forms of architecture and building. The codes, based on the minimum strength of the different mixes, require a partial factor to be applied for both the skill of the tradesman and the production of the masonry unit. These factors can result in the wall being up to 10 times stronger than the specified strength. Arguably these partial factors can be adjusted in the case of highly-skilled artisans and very close attention to production.

A result of the improved understanding and control of masonry construction is that local reconstruction, when necessary because of environmental decay, may be undertaken with confidence to the same proportions as the old. If there is no material change of use to building regulation clauses 5 and 6, then the existing may be deemed adequate.

Thankfully, more recent research on the use of lime mortars has been undertaken using well graded, coarser sands. The results of these projects have been used to produce an appendix to BS5628. This appendix is limited but it does give strengths that can be used in calculations undertaken to BS5628, should the engineer need to check.

Movement

In the case of thermal and moisture movements, cracking in brickwork is caused when the tensile stress in the brickwork is exceeded, because the mortar is not flexible enough to allow longitudinal strain in the wall to dissipate. The mechanisms of this cracking vary significantly between thicker, historic walls and modern thin-walled construction. The comments below apply to historic constructions.

The width of any crack depends on the length between points at which these cracks develop. Ideally the cracks should be contained within the joints, formed as the sum of tiny shearing movements, and be of hairline width, so that the wall can reseal itself. The wider the crack, the less likely the free lime in the mortar will dissolve in the rain and reset. On this basis the mortar should be as weak as possible while still capable of sustaining the imposed stresses.

Mortar specification

The durability of a mortar is dependent on its matrix. Lime binders play an important role in this, but the attribute most sought of them is their ability to permit permeability and allow movement. The more exposed the location, the stronger the matrix should be, in particular to prevent surface erosion.

However, it is the aggregate that plays the critical role in achieving durability. The mortar must cure to a mass that has many small voids into which moisture can freeze without cracking the matrix. To achieve this, it must be well graded, meaning that there must be a reasonable proportion of many different sizes of grains, with no more than about a third retained on any one British Standard grading sieve. Unfortunately, the grading curves contained in BS1200 Building Sands from Natural Sources are inadequate to ensure the best mortars, and experience is required in selecting sands.

The process of production and application should be appropriate for the situation. It is probably better to use a mortar that is simple to make and lay than to use one that might be technically slightly better but more difficult to make. This is similar to reducing the partial factors in BS5628.

Questions such as whether to use putty, hydrated, hydraulic or formulated lime should take into account the physical properties required of the mortar. The choice of whether the mortar is hot mixed, cold mixed or left to mature should take into account the skill of the workforce and the site conditions. Any benefit that may be obtained by hot mixing could be completely negated by skills gaps.

The mortar should be more permeable than the masonry units, be they bricks or stone. The set mortar should ideally act as a poultice and preferentially dry out the masonry units. The size of the pore structure will determine the effectiveness of this poultice effect. A pore size of 1 micrometre is a good target. On a large project this can be checked by testing samples petrographically.

Summary: how to choose a mortar

It is generally best to work from the basic principles. A highly technical approach may not be appropriate to all projects.

Choose lime rather than cement. With the exception of exposed civil engineering work, a lime binder is better.
Favour site mixing. This gives flexibility to the mason and allows formulation using local materials.
However, do not reject pre-mixed solutions. Many pre-mixed lime mortars are easy to use and will ensure a well-designed mix. These are valuable if the type of project procurement makes site mixing awkward.
Do not focus on strength. Mortar strength is rarely a significant factor in the historic environment. Be careful not to specify too strong a mix. Use lower strength NHLs (natural hydraulic lime) if hydraulicity is required, except where exposed.
Take care in choosing the sand. The starting point will be a sharp or grit sand. You are looking for a well-graded sand, not dominated by a single size, where no more than 30 per cent is retained on any one BS sieve. The maximum particle size should be about half the width of the joint, although occasionally particles may be up to 8 mm. Reject a soft building sand. Ask for samples.
If appropriate, analyse the original mix. This is well established practice for significant heritage projects. If the original mix is poor, decide whether to use a better new mix or to replicate the original.
Consider whether the masons originally used a hot- or cold-mixed mortar. Encourage hot mixing, but only where appropriate. Consider whether the contractor has the skill to use a hot-mix mortar, or whether a more conventional cold mix would achieve a better result in the circumstances.