Content uploaded by Caspar Groot
Author content
All content in this area was uploaded by Caspar Groot on Nov 04, 2021
Content may be subject to copyright.
3rd Historic Mortars Conference 11-14 September 2013, Glasgow, Scotland
1
Choosing mortar compositions for repointing of historic
masonry under severe environmental conditions
Caspar JWP Groot 1 & Jos TM Gunneweg 1
1. TU- Delft / Fac of Civil Engineering and Geo Sciences, Stevinweg 1 2628CN Delft / NL
c.j.w.p.groot@tudelft.nl, j.t.m.gunneweg@tudelft.nl
Abstract
The quality of repointing work in historic masonry is to an important degree determined by the
composition of the repair mortar. Apart from this good workmanship is a basic requirement for durable
repointing.Over the past decades awareness has grown that the mortar composition of repointing should
always be considered and applied taking into account the hygric and mechanical properties of the existing
adjacent materials. Often this is easy enough to realize. However, choosing the composition of a
repointing mortar there are situations where various damage risks seem to point at opposing materials
properties.In the paper this problem is approximated analyzing a number of damage cases with the aim to
define more precisely which requirements and to what extent should be maintained.
Subsequently, from lab studies and experiences with the application of natural hydraulic lime (NHL) in
specific repointing projects a set of requirements is proposed.
The context of this repointing study is repair of low-strength historic fired clay brick masonry in the
coastal area of the Netherlands; environmental conditions: sea salt laden, heavy rain load and freeze-thaw
cycling.
Keywords: repointing, damage, requirements, restoration
1. Introduction
Historic masonry in the Netherlands (before 1850) is mostly composed of relatively low-strength
bricks held together with a relatively low-strength mortar (often pure lime mortars). In most cases the
mechanical strength of the masonry, although low, is sufficient to show a satisfactory durability.
Hygrically the low mechanical strength mostly goes together with a high-porosity: 25-40 vol %.
In the past repointing of historic masonry was often done using shell lime. The shell lime was
obtained by burning shells from the sea. This lime was supposed to be weakly hydraulic however more
importantly, these mortars (the lime being relatively coarse) showed excellent drying behaviour. For a
good quality the preparation of the shell lime mortar was rather time-consuming. With time the use of
shell lime mortars for repointing stopped as the production of shell lime came to an end in the
Netherlands. The durability of shell repointing often turned out to be excellent.
In the repointing practice of historic masonry, with the disappearance of shell lime, cement as the
basic binder was introduced. Speed of production (early strength) and the idea that the use of an, in itself,
durable material were important incentives for the application of cement-based repointing mortars in
historic masonry. Subsequently, there was amazement at the occurrence of damage as a result of this type
of repointing. Mostly the problems are caused by incompatibility between the repair material and the
adjacent historic materials. Since the joint material in itself generally remains undamaged, the repointers
doing the job are reluctant to adapt their composition, as a change in general will diminish the durability
of their joint (it’s about who is responsible for an eventual damage).
From a restoration point of view, however, the repair mortar should not have a negative effect on the
durability of the existing masonry and other components in the masonry. With this constraint in mind, the
service life of the repair mortar should be as long as possible (be durable).
2
Service life not only depends on the mortar components but also on how it is installed (workmanship)
and cured, on the compatibility between the masonry unit and the mortar, and on the severity of the
environmental exposure, which in turn depends on weather, design, construction practice, operation, and
maintenance.
2. Requirements related to damage risks
Main causes of damage in or as a result of repointing are: (i) freeze-thaw cycling (ii) salts and (iii)
thermal /moisture expansion/shrinkage or a combination of these three. These types of damage may be
the result of inappropriate materials behaviour and / or unskillfull execution.
2.1 Freeze-thaw damage
Freeze-thaw damage may occur in a repointing (see figure 1) if the binder is an air lime and not
sufficiently carbonated at the onset of winter. Under average weather conditions, after 4 weeks,
carbonation should have occurred to a depth of at least 5 mm (Ratcliff 1997); carbonation depths of 8
mm after 2 months are reported by (Waldum 2009). This means that the application of air lime has to take
place in the right season and at least 2 to 3 months before the frost season. This seasonality also plays a
role with regard to the application of Natural Hydraulic Lime (application at least 1 month before the frost
season).
For many appliers the use of pure air-lime mortars without the addition of hydraulic components is
often considered too risky, especially if the masonry is very exposed. The reason being that non-
carbonated lime has an high solubility and is easily leached out by rain. Given the the relatively long
period of sensitivity to weather conditions pure air lime repointing is in fact not a very favourable option.
Figure 1:Horizontal layering (see arrows) in an uncarbonated air lime repointing as a result of freeze-thaw
cycling
Freeze-thaw damage may also occur indirectly: applying a very dense (e.g. cement-based or
repointing mortars containing water repellents) repointing on a historic air-lime bedding mortar may have
serious consequences for the drying conditions in the masonry. With an open porous structure in the
repointing, moisture may easily move from the bedding mortar to the repointing; with a dense structure in
the repointing the moisture will necessarily pass through the adjacent brick; this is a slower process,
causing longer periods of high moisture content in the bedding mortar; as a result of this the bedding
mortar may become frost prone. Cases are known of centuries-old sound bedding mortars showing this
type of frost damage caused by the application of the wrong restoration practice (a typical example of an
incompatible use of a repair material).
2.1.1. Requirements - materials properties
When considering freeze-thaw the most important material characteristics are:
strength development
3
frost resistance
drying behaviour (moisture transport, porosity)
2.1.2. Requirements - execution techniques
Related damages in addition to the execution technique may play a role in the further occurrence of
freeze thaw damage. For instance: a hollow between repointing and bedding, as a result of insufficient
filling of the joint may result in push out by frost (where a period of heavy rain is directly followed by a
frost period); other examples may be insufficient depth of the repointing, a V-form instead of a
rectangular form of the the joint etc.
2.2. Salt damage
Salts may have a harmful influence on the durability of the mortar as well as the brick. The influence
of sea salt (containing a high content of NaCl) can be observed in many buidings, especially on the North
sea coast of the Netherlands. In figure 2 two examples are presented of damage caused by NaCl.
Figure 2 (left), shows damage to a pure air-lime repointing mortar. In this case the highly soluble
chemical compound calcium chloride (CaCl2) is formed with calcium from the mortar binder and chloride
from the seasalt. Rain, subsequently, causes the leaching out of the mortar. The binder choice is obviously
an important parameter in limiting salt damage.
In practice the start situation of the substrate on which the repointing is applied is important: high
contents of salt may limit the use of lime-based mortars such that only cement seems to be appropriate.
For other reasons this maybe very inappropriate (too dense, too strong). Desalination of the masonry may
then be a solution.
Also crystallization-dissolution cycling, especially around 75% Relative Humidity, resulting in
swelling-shrinkage cycling, may cause considerable damage [Lubelli 2006].
Figure 2:Cases of NaCl damage. Left: Voids in the repointing stemming from leaching of calcium
chloride from the air lime repointing. Right: weathering of underfired brick through NaCl crystallization-
dissolution cycling.
Figure 2 (right), presents damage to fired clay bricks by NaCl. In this case the underfired bricks are
apparently not strong enough to resist the crystallization-dissolution cycling caused by NaCl. The
stronger bricks don’t show damage (compressive strength values of 10-15 Mpa are suggested by practice
as strong enough to withstand salt damage through chrystallization-dissolution cycling).
2.2.1. Requirements - materials properties
When regarding salt damage the most important materials characteristics are:
the choice of the binder
chemical composition such that no soluble compounds are formed
resistance against salt crystallisation requires a certain degree of mechanical strength
2.3. Deformation damage
In practice there is often more of an awareness regarding the risks of damage as a result of thermal
and moisture deformation on a macro scale, than there is for that on a meso scale.
4
Differences in macro deformation behaviour of modern (left) and traditional masonry (right) is shown
in figure 3. The fear of cracks, apparently a risk in modern cement masonry, is solved by extensively
applying dilations. The long wall constructed with a tradtional natural hydraulic lime mortar, NHL3.5
type St.Astier, (right) does not need any dilation.
Figure 3: Left, the yellow lines indicate the dilations applied in a modern masonry. Right, a wall
without dilation in traditional masonry.
At meso scale thermal resp. moisture deformation may as well be a serious cause of damage (see
figure 4). Not only do material properties such as the (linear) thermal resp moisture deformation
coefficient play a role in the detachment process, but so does the exceution technique in for example the
pointing.
Figure 4: Detachment of repointing caused, in particular, by thermal deformation enhanced by a
wrong cross-sectional form (triangular in stead of rectangular) of the repointing.
2.3.1. Requirements - materials properties
(linear) thermal resp moisture deformation coefficient
stiffness (E-modulus)
2.3.2. Requirements - execution technique
rectangular cross section of the repointing (no V-cross section form!!)
2.4. Overview - Requirements
In order to facilitate the analysis of the (in)consistency of the materials and execution requirements
of the different damage risks, they have been collated in Table 1.
5
Frost Salt Deformation
damage damage damage
properties
strength development
winter proof 1)
drying (porosity) quick
strength (compressive) ≥ medium low
stiffness (E-mod) ≥ medium low
expansion coefficient low
frost resistance
salt resistance
execution
prep,prewett,curing etc
2)
form joint
depth joint
filling joint
1) strong enough to pass the first w inter w ithout frost damage
2) specialised execution required
Tabel 1: Requirements related to major damage risks of repointing in weak historic masonry
It is quite clear from the table that there is a friction between the required mechanical characteristics
relating to frost and deformation risks, when compared to those relating to the salt damage risk.
The easiecst solution seems to be to opt for the stronger and and less deformable mortar as required
for a better salt damage resistance. This provides an increase of durability in the repointing mortar;
however, such a mortar will also be less porous, leading to slower drying of the bedding mortar, and also
a higher thermal expansion and a higher stiffness, increasing the risk ofdeformation damage.
In fact a compromise should be reached between the durability requirements of the repair mortar and
the compatibility requirements of the adjacent masonry (drying affecting frost resistance) and stresses
(resulting from deformation behaviour).
3. Tests in practice and in the laboratory
In practice as well as in the laboratory a series of site-mixed and prefabricated repointing mortar
compositions were tested. The starting point were recent experiences with repointing mortars used in
historic towers and windmills located on the West coast of the Netherlands. These buildings are
composed of relatively weak masonry which is exposed to sea salts, heavy rain and frost. In such cases
the choice of the binder is essential, as this determines whether the mortar will be salt resistant, have
adequate strength (not too high and not too low), is frost resistent, and shows low thermal deformation.
With regard to the needed salt resistance recent experiences in restoration projects with natural
hydraulic lime (NHL 3,5) showed such good sea salt resistance, that this binder was chosen as a basis for
the site-mixed repointing mortars.
In the laboratory, properties like compressive strength, dynamic E-modulus, free water absorption,
drying characteristics, thermal expansion and frost resistance were determined. In this paper the attention
is focused on freeze-thaw testing and deformation testing, being of major importance with regard to
durability.
3.1 . Mortar compositions
Materials used in the site mixed mortars:
Natural Hydraulic Lime (NHL2 and NHL3,5 from St Astier, France).
BFC: blast furnace cement (HC CEM III/B).
Pozzolan: Trass (Rheinische Trass)
Air limes: ‘Lime (Harl)’ (CL70)
Sand 1: standard repointing (rounded river) sand with Fineness Modulus of 1.8
6
The composition of the prefabricated. repointing mortars is unknown; from the hardening process it
can be concluded that they are cement-based mortars.
Table 2: Repointing mortars used in the laboratory tests
Site-mixed mortars
Prefab
mortars
VB02
NHL3,5
BFC
Sand
1
VB06
NHL2
add
BFC
Sand
1
VP01
A
6
1
17
3
1
10
VB03
NHL2
BFC
Sand
1
VB07
Lime
(Harl)
trass
BFC
Sand
1
VP04
B
3
1
10
1,3
0,4
0,25
3,2
VB04
NHL3,5
Sand
1
VP05
C
1
2,5
VB05
Lime
(Harl)
BFC
Sand
1
VP06
D
5
1,5
16
3.2. Freeze-thaw testing
The freeze-thaw tests were based on the test set-up used in the EU-Pointing project (Wijffels, 2001).
In this test the freeze-thaw cycle is applied from one side of the test specimen, and fresh water is used in
the test.
Precise details on the freeze-thaw cycle tests are given in (Groot, 2012)
Figure 5: The test specimens in the freeze-thaw container
The test specimens were evaluated regarding:
damage of the repointing itself
deterioration of the bond between mortar and brick
damage of the bedding mortar behind the repointing
push out (sound test)
The results are presented within table 3. None of the repointings of the test specimens were pushed
out. Push out may occur in case the depth of the repointing is small (see tests Wijffels 2001); in this
testing series the depth of the repointing was greater than two times the width of the joint.
Frost damage behind the repointing (in the bedding mortar) occured within three of the test specimens
of the cement-based prefab mortars and within only one test specimen of the lime-based site mixed
mortars: a clear indication that the drying conditions of the site mixed specimens are more favorable than
the prefab mortars.
7
Site-mixed mortars Results
NHL3 ,5 HOC Zand interface-damage bedding mortar
6 1 17
NHL2 HOC Zand OK
3 1 10
NHL3 ,5 Za nd interface-damage bedding mortar
1 2.5
ka lk Ha rl H OC Z and frost damage behind repointing
5 1.5 16
NHL2 t o eslag H O C Z a nd OK
3 1 10
ka lk Hrl tr as H OC z and OK
1.3 0.4 0.25 3.2
frost damage behind repointing
frost damage behind repointing
OK
frost damage behind repointing
VP06
D
Prefab mort ars
VP01
A
VP04
B
VP05
C
VB04
VB05
VB06
VB07
VB02
VB03
Table 3: Results Freeze-Thaw testing
3.3. Thermal deformation
Thermal deformation may result into stresses in the masonry. The thermal stress (σ), developing
under restrained conditions is given by the following equation,
σ = α.E.∆Ø [Mpa] [1]
where :
α: linear thermal deformation coefficient [1/0K . 10-6]
E: dynamic E-modulus [Mpa. 103]
Ư: temperature change [0K]
α.E is the materials-dependent stress coefficient, specific for every different type of repointing mortar
(see figure 6)
The significant influence of the choice of material on the possible stress development in masonry can
be shown, by comparing the thermal stresses developed by the cement-based repointing VP01 and the
lime-based repointing VB06 (see as well figure 6)
Stress development under restrained conditions, with a temperature increase of 50 °K (south or west
facade) is as follows:
P01:
Stress coefficient (see figure 6): α.E = 170 . 10-3 MPa/0K
Restrained stress with a temp increase of 50°K: σ = α.E.∆Ø = 170 . 10-3 x 50 = 8.5 MPa
VB06
Stress coefficient (see figure 6): α.E = 34. 10-3 MPa/0K
Restrained stress with a temp increase of 50°K: σ = α.E.∆Ø = 34 . 10-3 x 50 = 1.7 MPa
8
0
2
4
6
8
10
12
14
16
Therm. def.coeff. [(1/K).10-6]
various mortar types
Thermal deformation
lime-based
cement-based
0
20
40
60
80
100
120
140
160
180
200
VB02
VB03
VB04
VB05
VB06
VB07
VP01
VP04
VP05
VP06
α.E(dyn) [MPa x 10-3]
various mortar types
Thermal stress coefficient α.E
cement-basedcement-based
lime-based
α. E
α.E
Figure 6: Materials-dependent stress coefficients α.E
3.4. Guideline requirements
A series of guidelines with the necessary requirements for a suitable repointing mortar could be
developed from the results of the laboratory testing combined with the practical experience obtained in a
series of repointing projects.
The guidelines for requirements applicable to the repointing of low-strength historic fired clay brick
masonry in the coastal area of the Netherlands where conditions encountered are sea salt, heavy rain load
and freeze-thaw cycling , are as follows:
Workability, evaluated as good by an experienced mason/pointer
Choose preferably NHL-based binder (or at least lime-based)
Depth of the joint ≥ 2 times joint width
Compressive strength 3-7 [MPa]
Dynamic E-modulus (6-10) .103 [MPa]
Linear Expansion Coefficient (4-7) .10-6 [1/ 0K]
Water Absorption Coefficient (WAC) 0.3-0.9 [kg/(m2.min0.5)]
Freeze-thaw test : No damage to the joint, No debonding within the joint , No frost damage to
the bedding mortar behind the repointing (acc. to EU-Pointing Project, see ref Wijffels 2001,
Groot 2012).
9
Acknowledgements
The financial support by the Province of Zuid Holland (The Netherlands) for the execution of this project
is gratefully acknowledged.
References
Groot C. and Gunneweg J. (2012). Repointing of Historic Masonry (in Dutch), TU Delft, Institutional
Repository
Hayen R., Van Balen K. (2001). Thermal Expansion of Historic Masonry, looking for Physical
Compatibility. Paper 5.10 . Maintenance of pointing in historic buildings; Decay and replacement; EC
Environment Programme ENV4-CT98-706
Lubelli B.A. (2006). Sodium Chloride Damage to Porous Building Materials, PhD Dissertation, TU-
Delft, ISBN 90-9020343-5
Ratcliffe Tim, Orton Jeff (1997) Success with lime renders, SPAB
http://www.ihbc.org.uk/context_archive/59/limerender_dir/limerender_s.htm
Vermeltfoort, A.T, Groot, C.J.W.P. & Wijen, E. (1999). Thermal strains in repointed masonry:
preliminary investigations using ESPI. RILEM pro 12, paper 21
Waldum Alf M. (2009) Historic Materials and their Diagnostic State of the Art for Masonry Monuments
in Norway, www.arcchip.cz/09
Wijffels T., Van Hees R., and Van de Klugt L. (2001). Pointing Report, ENV4-CT98-706 paper 5.2. TU
Delft, Institutional Repository
