Construction and Building Materials 75 (2015) 359–367

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

The effects of site practice on the physical properties of proprietary stone

restoration mortar
Clare Torney a,⇑, Alan M. Forster b, Phillip F.G. Banfill b, Ewan M. Szadurski b
a
Historic Scotland Conservation Directorate, 7 South Gyle Crescent, Edinburgh EH12 9EB, United Kingdom
b
School of Energy, Geoscience, Infrastructure and Society, Heriot-Watt University, Edinburgh EH14 4AS, United Kingdom

h i g h l i g h t s

 Laboratory research simulates site modification of a formulated restoration mortar.

 Increased water content (120%) decreases compressive strength by 50%.
 Increased mix time (up to 6 min) increases moisture transmission of material.
 Surface laitance reduces vapour permeance by 13%.
 Deviation from manufactures guidelines produces unknown properties, increasing risk.

a r t i c l e i n f o a b s t r a c t

Article history: Commercially-produced ‘restoration mortars’ are increasingly being used in stone masonry conservation.
Received 8 July 2014 The convenient ‘mix and go’ approach of these materials is opening up the area of masonry repair to a
Received in revised form 6 October 2014 wider, lesser skilled, consumer base. Pragmatic site practice with restoration mortars often leads to
Accepted 12 November 2014
the modification of materials with the aim of providing enhanced workability, fitting with varying
Available online 5 December 2014
weather conditions and project timescales. This work aims to establish the resilience of one such proprie-
tary restoration mortar to variations in its preparation and finishing. The properties of the material, and
Keywords:
therefore its performance in service, are significantly influenced by variations in mixing regime, surface
Stone repair
Restoration mortar
finish and pigmentation. Results highlight the impact of workmanship on the material’s properties and
Lithomex the need for a thorough understanding of the product prior to specification, preparation and application.
Hybrid mortar Ó 2014 Elsevier Ltd. All rights reserved.
Laitance
Pigmentation

1. Introduction including; natural stone replacement (indenting), consolidation

Weathering and deterioration of historic stone masonry is a
worldwide problem. This situation is likely to be exacerbated by
current and predicted future climate change [1], particularly
associated with intensifying precipitation patterns [2–4]. Increas-
ing rapidity of, and prevalence in, masonry degradation and the
associated loss of historic building fabric is creating an urgent need
for practitioners to make efficient and effective conservation
decisions surrounding the repair process. A number of repair
options may be considered in cases of masonry deterioration,
⇑ Corresponding author. Tel.: +44 (0) 131 334 6367.
E-mail addresses: Clare.Torney@scotland.gsi.gov.uk (C. Torney), A.M.Forster@
hw.ac.uk (A.M. Forster), P.F.G.Banfill@hw.ac.uk (P.F.G. Banfill), E.Szadurski@hw.ac.uk
(E.M. Szadurski).
of existing masonry, or ‘plastic’ repair with mortars [5–8]. Each
of these repair approaches brings with it a number of benefits
and drawbacks relating to both technical and philosophical aspects
of masonry conservation [6,7].
‘Restoration mortar’ is a term used to describe a group of
commercially produced, proprietary products specifically designed
for the repair of masonry units, usually natural stone and/or brick
[9,10]. These materials fall into the broader and less well defined
category of ‘plastic’ repairs (repairs using any workable material
that will adhere to a substrate and will harden after application),
but should not be confused with traditionally prepared bespoke
lime ‘mortar’ repairs. The term ‘restoration mortar’ does not in
itself indicate a material of a specific composition. Many ‘off the
shelf’ products are available on the market, with significantly
different compositions [10]. Ideally, restoration mortars will fulfil
a number of requirements relating to both technical and visual

http://dx.doi.org/10.1016/j.conbuildmat.2014.11.040
0950-0618/Ó 2014 Elsevier Ltd. All rights reserved.
360 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367
compatibility including: high vapour permeance and water
absorption; good adhesion; and the ability to be coloured and
tooled in sympathy with adjacent masonry.
The specification and application of mortars for stone repair
work can be extremely complex [11]. An understanding of
individual site conditions, substrate properties and aftercare
requirements are all important [12–16]. Critically an assessment
of the compatibility of new and old materials [11,17–19] is essential
prior to the works. However, the ‘case-by-case’ assessment and
specification approach once adopted for mortar repairs is no longer
seen as practicable in all cases [20] and regrettably the aesthetics of
repairs are viewed as a higher ranked measure of success than
compatibility of materials. This situation is in direct contrast to
the well considered approaches suggested by various authors
[11,18,19].
The use of ‘plastic’ repair materials including restoration
mortars [9] for restoration of historic masonry is increasingly com-
mon at a time when short-term cost savings are viewed by some as
higher priority than longevity of repairs [21]. The emergence of
proprietary ‘off the shelf’ restoration mortars, specifically designed
for the repair of masonry in conservation projects and for the
simulation of masonry in new-build situations, is likely to play a
significant role in this increase in use [9]. Such prescribed products
claim to offer a number of advantages over both natural stone
replacement and repair using lime mortar, including: ease of
preparation and application; universal suitability for a wide range
of substrates; and visual ‘matching’ to substrate colours.
The ‘just add water’ convenience of restoration mortars
undoubtedly makes them attractive to less-skilled workers who
might not otherwise attempt to carry out masonry repair work.
These products also have favourable qualities for specifiers who
lack the knowledge, experience and/or confidence to use the more
complex traditional bespoke mixes tailored to the individual
project; restoration mortars simplify specification by obviating
the assessment and selection of suitable binders and aggregates
for each repair project. This is somewhat at odds with the skills
required for conservation of historic masonry structures where
increasing technical understanding of materials characterisation
and performance has developed [22]. The potential decrease in
skills levels associated with the widened target consumer base of
these products is a cause for concern. It is unclear what impact
variations in workmanship will have on the material’s performance
as a higher number of inexperienced operatives utilise the materi-
als. Whereas strict manufacturer’s guidelines on the appropriate
use of products and step-by-step instructions on their preparation
and application might be seen as a route to ensure success, this is
not necessarily the case. Site practice is often influenced by time
and financial constraints, as well as weather and workmanship,
and variations undoubtedly impact upon the materials’ ultimate
physical properties. The growing availability of such products from
specialist builders’ merchants and general trade suppliers is also
stimulating the market. The potential consequence of these trends
is the accumulation of latent defects, with future adverse effects on
the built heritage.
Proprietary restoration mortars for masonry repair [10] and
commercially produced pre-mixed ‘lime-based’ repair products
for pointing/bedding [20] can significantly differ in their
composition and physical properties. Variations in curing regime
and aftercare of lime-based mortars can lead to further significant
variations in materials’ properties and their success rates [22–25].
The objective of the present work is to investigate the impact of
variations in site practice (i.e. mix regime modification and
workmanship variations) upon the properties and performance of
a proprietary restoration mortar. The resilience of a material to
modification in site practice is key to the overall robustness of
any specification.
1.1. Restoration mortar
Commercially available restoration mortars are typically dry
packaged (anhydrous) materials consisting of binder, aggregate
and, in some cases, other fillers in predetermined ratios. Such
products are optimised during manufacture and are similar in form
to some ready-mixed mortars utilised for other functions such as
pointing, bedding [20] and rendering. This pre-batching is assumed
to reduce the risk of failure associated with drying shrinkage and
inaccurate proportions [9], as well as providing ‘just add water’
convenience. It must be emphasised that some products require
mixing of multiple dry components prior to addition of water.
Additionally, some restoration mortars may be ‘formulated’,
containing a number of natural or synthetic additives that may
or may not be declared by the manufacturer. These additives aim
to enhance the fresh properties and/or the hardened properties
[23], for example by entraining air (Fig. 1).
A previous study of two commercial restoration mortars
highlighted significant differences in physical properties [10] that
have also been observed between ready-made lime mortars for other
applications [20]. One material was found to be essentially a natural
hydraulic lime mortar, and the other a formulated hydraulic lime–
cement hybrid mix [10]. The hydraulic lime restoration mortar offers
no obvious technical advantages over a site blended mortar mix apart
from the convenience of pre-batched components which eliminates
the need for careful batching by the end-user. Although the manufac-
turer provides detailed guidance on appropriate background prepara-
tion and application of the mortar, little technical advice was
provided on how best to mix the material (i.e. water content and
mixing duration) at the time of purchase in 2012. It can be assumed
that the ‘best practice’ guidelines that apply to hydraulic lime mortars
in general (e.g. [16,23,26]) are applicable to this restoration mortar.
The properties of hydraulic lime mortars are relatively well under-
stood and the impact of workmanship and site practice on their
properties has been dealt with in other studies [23,26].
The present study focuses on a single lime–cement hybrid res-
toration mortar ‘Lithomex’, a material produced by Chaux et
Enduits St. Astier (CESA, France), based on a St. Astier natural
hydraulic lime binder. Lithomex is supplied with detailed guide-
lines on its use and mixing regime [27,28]. We explore the impact
on technical performance of deviation from these guidelines,
giving mortars of different water content, mixing duration and
pigmentation, and also the impact of surface finish.
2. Materials and methods
Unpigmented Lithomex and pre-pigmented Lithomex were obtained from a
local supplier and stored in airtight containers until ready for use. Manufacturer’s
data [27] indicates that the material consists of the following components
(expressed as percentage of binder): calcium hydroxide 620%; hydraulic binder
(Portland cement) 620%; filler (vermiculite) 65%. In addition, previous research
has established that Lithomex also contains fine grained quartz and calcite aggre-
gate and talc filler [10].
The manufacturer’s preparation guidelines state that the materials should be
mixed (mechanically or by hand) for between three and five minutes, with a water
content of 4.5–5.5 L of water per 25 kg of dry material (giving a water/solid ratio of
0.18–0.22). The test programme extended these two ranges and, in addition,
assessed the effect of surface finish, specifically the presence of laitance, on the
hardened properties of the material. Laitance is a surface coating, ‘skin’ or ‘scum’
that forms when fine lime particles held in suspension migrate to the outer surface
of the wet material when the surface is being trowelled smooth [29]. This is
believed to hinder the vapour permeability of lime-based materials [30], and
negatively impact upon the substrate beneath by causing accelerated masonry
decay associated with entrapment of moisture.
2.1. Specimen preparation
Specimens were prepared as 40 mm  40 mm  160 mm rectangular prisms;
mortar was mixed using a Hobart 5 L bench top paddle mixer, and cast in polysty-
rene moulds. These specimens were utilised for determination of carbonation
depth, water absorption and compressive strength. Cast cylindrical mortar discs,
 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367 361

Fig. 1. The presence of large voids in hardened Lithomex, a proprietary restoration mortar, identified in (a) thin section and (b) electron microscope imaging, indicates the use
of an air entraining additive. Blue areas in (a) represent porosity; entrained air voids are highlighted with white dashed lines. The ‘matrix’ of the mortar between aggregate
grains also appears blue due to its microporous nature. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)

approximately 70 mm in diameter and 10 mm in thickness, were prepared for

vapour permeability testing using the wet-cup method (following the principles
outlined in BS EN 1015-19:1999 [31]).
All specimens remained in covered moulds for 24 h providing a damp environ-
ment that enables progression of hydraulic reactions. After demoulding, specimens
were cured in a stable laboratory environment of nominally 55% ± 10% relative
humidity and 20 °C ± 2 °C until the time of testing. A detailed account of the sample
preparation, curing and testing regimes can be found in Torney et al. [10]. The
following variables were tested by preparing mortar samples with compositions
altered from the manufacturer’s specification:
Water content: Mortars prepared using 5 L water per 25 kg dry material were
considered to represent 100% water content. The impact of variable water content
on material properties was assessed by also preparing mortars using 80%, 90%, 110%
and 120% of this water content: 90–110% falls within the manufacturer’s recom-
mended range. Mortars were mixed for four minutes in this programme. The Euro-
pean Standard for testing building lime [32] states that the water content of natural
hydraulic lime (NHL) mortars should be that which gives a specific consistency
(determined by flow table); 165 ± 3 mm for NHL 2 and NHL 3.5 and 185 ± 3 mm
for NHL 5. Work by Hanley and Pavía [33] showed that the Standard recommenda-
tions for NHL 2 and NHL 5 produce mortars of optimum workability, whilst also
producing favourable mechanical properties. However, a flow diameter of
185 ± 3 mm was found to be more appropriate for NHL 3.5 mortars. The current
study did not adopt this approach for two reasons: (1) it cannot be assumed that
the same flow values are most appropriate for Lithomex, as this product is a hybrid
mix that contains more than one hydraulic phase as well as a water retaining
component (i.e. vermiculite) and an air entraining additive and (2) given the nature
of this proprietary material it was not considered appropriate by the authors to base Fig. 2. Scanning electron microscope (backscattered electron) image of laitance
water content on flow values, but rather to note variations in flow with changing (indicated by the arrows) on the surface of Lithomex. The surface laitance appears
water content, as different flow diameters may be required for repairs to different dense in comparison to the main body of the sample.
building elements, e.g. a stiffer mortar (with lower flow diameter) may be required
for repair to moulded stone detail.
Mixing duration: To assess the impact of mixing duration on mortar properties,
samples were prepared of mortars mixed for different lengths of time; mortars
were mixed for one, two, three, four, five, six and 20 min. In each case, mixing 2.2. Test methods
was stopped after half the time had elapsed and the mixer bowl sides scraped to
reincorporate any material adhering to the walls back into the mix, so as not to alter Proprietary restoration mortars do not currently fall within the remit of
the mix proportions. Mortar was prepared at 5 L water per 25 kg dry material in this European Standard test methods for lime mortars, therefore a bespoke test
programme (i.e. 100% water content). regime was developed to provide relevant information to users and specifiers
Laitance: The effect of surface laitance (Fig. 2) was determined by comparing the of the product. The properties of flow, carbonation depth, compressive strength,
vapour permeance of mortar discs which had been gently rubbed with a tungsten vapour permeance and water absorption were considered to be of primary
carbide abrasive pad (to remove the laitance) with those on which the laitance performance in assessing the material’s suitability. Flow diameter of fresh
was left intact. This was carried out on unpigmented Lithomex mortars prepared mortar was assessed and carbonation depth, compressive strength and vapour
using different mixing durations and water contents. permeance were determined on hardened Lithomex after 28 days of curing.
Pigmentation: The effect of pigmentation was studied using seven different Properties assessed at 28 days are at least in part attributable to early stage
coloured varieties of Lithomex. The colours chosen represent the colours of some hydraulic reactions (e.g. strength development via hydration of alite). Earlier
British sandstones currently in production [34–37], some of which were also used work [10] has shown that strength development of Lithomex decelerates after
historically [38]. Five varieties of blonde Lithomex and two red Lithomex this point although some minor strength development attributable to belite
corresponding to these stone types were selected for analysis: ‘Bearl’, ‘Blaxter’, hydration is evident up to 182 days, at which point the material was fully
‘Clashach’, ‘Peak Moor’, ‘Stanton Moor’, ‘Locharbriggs’ and ‘St. Bees’. Although carbonated. Testing beyond this point (up to 3 years) did not reveal any further
pigment packs are offered by some suppliers for on-site mixing, pre-pigmented increase in compressive strength of the material, nor did it show any significant
Lithomex was purchased for this study to avoid any variations associated with variation in water absorption. Given the high proportion of strength develop-
incomplete or inconsistent dispersion of pigment. Pigmented samples were tested ment attributable to early stage reactions in Lithomex (86% of maximum
for flow, carbonation depth, compressive strength, water absorption and vapour strength was attained by 28 days), this work is considered to provide a good
permeance. indication of the mortar’s properties, even on longer timescales.
362 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367
The fresh mortar was tested by flow table to determine consistency of the
material that in turn provides an indication of workability [23]; tests follow the
principles outlined in BS EN 1015-3:1999 [39]. Flow diameters stated in the results
are mean values, calculated from four individual measurements.
Carbonation was assessed using phenolphthalein indicator. The indicator turns
from colourless to pink on contact with high alkalinity material (i.e. calcium
hydroxide in the case of an uncarbonated lime mortar). Depth of carbonation was
measured from all four external faces on a freshly fractured surface of each of at
least 3 specimens.
Compressive strength was determined on 40 mm cubes, cut from rectangular
prisms, tested on a calibrated Avery–Denison compressive test machine with a load
cell of 3000 kN applied at a rate of 10 kN/min. A minimum of 6 cubes of each sample
type were tested and mean values calculated.
Vapour permeance was determined by the wet-cup method, following the same
principles as the British Standard [31]. Specimens were sealed in place on the
mouth of a cup containing a saturated potassium nitrate solution. Sample cups
were kept in a TAS Series 3 MTCL environmental test chamber at a constant temper-
ature of 20 °C and relatively humidity of 65%. The weight of the sample cups was
recorded at the beginning of the experiment and periodically over a period of
21 days. A minimum of eight specimens of each sample type were tested and mean
values calculated; three with laitance intact and five with laitance removed. Vapour
permeance (K), expressed in kg m2 s Pa, was calculated using Eq. (1).
^ ¼ 1=ðADp=ðDG=DtÞ  RAÞ
where A is the area of the open mouth of the test cup; Dp is the difference in water
vapour pressure; DG/Dt is the weight loss from the test cup; and RA is the water
vapour resistance of the air gap between the potassium nitrate solution (KNO3
solution = 95% RH partial driving pressure) and the test specimen.
Water absorption (by immersion) of the restoration mortar was assessed.
Specimens were oven dried at 105 °C to constant mass prior to being immersed
in water for five days. Water absorption was calculated as percentage mass gain
of samples. This method serves as a proxy for ‘open’ (or accessible) porosity, assum-
ing that all air within accessible pores of the samples was displaced by water within
the five day test period; repetitive weighing of samples showed that weight gain
had ceased by this point. A minimum of three specimens of each type of sample
were tested and mean values calculated.
3. Results & discussion
3.1. Water content
Fig. 3 shows the effect of water content on the flow diameter of
Lithomex. A 10% increase in water content from the ‘100% mix’
increases the flow diameter by 38 mm, whereas a 10% decrease
reduces flow diameter by only 11 mm. Therefore increasing water
content has a greater effect on flow diameter than does reducing it.
Thus the manufacturer’s recommended water content range of
±10% deviation from the mid-point value could have a significant
impact upon the properties of the material prepared on site, even
when guidelines are adhered to. Bigger variations in water content
will have more severe effects.
Fig. 4 shows the effect of water content on compressive
strength and carbonation of Lithomex. Reducing the water content
has only a minor impact on both the depth of carbonation and
compressive strength. However, increasing water content by 20%
250
200
Flow diameter (mm)
150
100
50
0 of 152 mm. While small, this clearly defined and reproducible
80 85 90 95 100 105 110 115
Water content (%)
Fig. 3. The effects of water content on the consistency of fresh Lithomex
determined by flow table. Vertical bars indicate minimum and maximum values
of the samples tested.
10 10
9 9
Compressive strength (MPa)
8 8
Carbonaon depth (mm)
7 7
6 6
5 5
4 4
3 3
2 2
1 1
0 0
80 90 100 110 120
Water content (%)
carbonaon depth compressive strength
Fig. 4. The effects of water content on the carbonation and compressive strength of
hardened Lithomex at 28 days. Vertical bars indicate minimum and maximum
values of the samples tested.
ð1Þ
increases carbonation depth by over 60% from 5.0 mm to 8.4 mm
and reduces compressive strength from 7.19 MPa to 3.14 MPa.
These results mirror the relationship between water content and
traditionally mixed lime mortar properties [40]. Mean values plot-
ted on Fig. 4 are representative of the data set as a whole and the
vertical bars indicate the minimum–maximum range of the values.
The compressive strength at water contents below 100% is signifi-
cantly more variable, suggesting inconsistency of the material.
It is assumed that the manufacturer’s recommended water
content provides sufficient water for reactions with hydraulic com-
ponents in the mortar to reach completion. In mixes where an
excess of water has been used (i.e. over 100%) the space occupied
by the excess water in the mix is translated into increased porosity
in the hardened material. With a water content above 90% of the
recommended value, water absorption increases approximately
linearly from 6.23% (in samples with 90% water content) to
14.42% (in samples with 120% water content) (Fig. 5). This increase
in water absorption and the inferred increase in porosity results in
a significant increase in vapour permeance of the mortar. A ±10%
change in mix water content changes permeance by over 50%, from
0.97  109 kg m2 s Pa (for 90% water content samples) to
2.25  109 kg m2 s Pa (for 110% water content samples). An
additional 10% increase in water content leads to a further
doubling of permeance; 5.2  109 kg m2 s Pa for samples with
120% water content. However, inconsistency of the material in
relation to permeance increases with water contents of 100% and
above as indicated on Fig. 5.
3.2. Mixing duration
Mixing duration has an impact on the consistency of fresh
Lithomex, although to a much lesser degree than water content.
Mixing for the manufacturer’s specified time (3–5 min) gives an
average flow diameter of 156 mm but a deviation from this regime
results in significant changes to flow diameter (Fig. 6). It was noted
that a shorter mix time had a more pronounced effect on consis-
tency than an extended mix time; even slight variation in the
mix time can have a notable affect on the consistency and there-
fore workability of the material. Three and 5 min mixing gave
157–158 mm flow but 4 min mixing results in a lower flow value
120
‘dip’ in the consistency was evident in repeated tests and this
suggests a transition in the behaviour of the material. This sudden
loss in plasticity of the material is likely to be due to the absorption
of water by vermiculite particles resulting in a reduction in the
interlaminar water between the Ca(OH)2 platelets in the mix,
 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367 363

20 10 16 9

Vapour permeance (x10-9 kg/m2.s.Pa)

Water absorpon (%weight gain)

14 8

Compressive strength (MPa)

Carbonaon depth (mm)
15 7.5 7
12
6
10
5
10 5 8
4
6
3
5 2.5 4
2
2 1
0 0 0 0
80 90 100 110 120 0 5 10 15 20
Water content (%) Mixing duraon (minutes)
Water absorpon Vapour permeance carbonaon depth compressive strength

Fig. 5. The effects of water content on the water absorption and vapour permeance Fig. 7. The effects of mixing duration on the carbonation and compressive strength
properties of hardened Lithomex at 28 days. Vertical bars indicate minimum and of hardened Lithomex at 28 days. Vertical bars indicate minimum and maximum
maximum values of the samples tested. values of the samples tested.

165 28 days in line with those expected of BS standard laboratory

prepared natural hydraulic lime mortars. It is noted that in reality,
160
site-cured mortars can be considerably weaker than the British
Standard tolerances and by comparison Lithomex has a relatively
Flow diameter (mm)

high 28 day compressive strength. However, the strength gain of

155
150
145
140
0 5 10
Mixing duraon (minutes)
Fig. 6. The effects of mixing duration on the flow of fresh Lithomex. Vertical bars
indicate minimum and maximum values of the samples tested.
which modifies the workability. Alternatively this phenomenon
could be explained by the dissolution and reprecipitation of Ca2+
ions associated with the early stage formation of hydraulic compo-
nents [41], resulting in a stiffening of the mix. The recovery of con-
sistency at five minutes marks the ‘fattening up’ of the calcium
hydroxide component, dispersing the agglomerates of lime [23]
and producing an abundance of smaller calcium hydroxide crys-
tals. This is the process associated with the maturation of lime
putty [42,43] albeit on a shorter timescale.
Increased mixing time from 1 min to 2 min reduces the average
28 day compressive strength of Lithomex, from 7.20 MPa to
5.92 MPa, but further mixing by 1 min sees this recover consider-
ably to 6.91 MPa (Fig. 7). The relatively high compressive strength
Lithomex is relatively short lived due to the nature of the hydraulic
components it contains [10] and over longer timescales it is
unlikely to exceed the strength of NHL 5 mortars, which continue
to gain strength over a number of years [23].
Average 28 day carbonation depth varies little with mixing
duration (Fig. 7). However, it is evident that these small variations
in carbonation depth display a relationship with compressive
15 20
strength; high strength correlates with low carbonation depth
indicating that at this early stage, products of hydration have more
influence over strength than carbonation. Prolonged mixing (for
20 min) slightly increases carbonation and reduces the strength
(Fig. 7), possibly associated with an altered pore structure. Vertical
(minimum/maximum) bars in Fig. 7 show significant variation in
the strength of identically prepared samples; there is less variation
in carbonation depth.
Average water absorption and vapour permeance values of
hardened Lithomex progressively increase with mixing time up
to a maximum at 6 min (8.01% and 2.35  109 kg m2 s Pa respec-
tively), decreasing thereafter (Fig. 8). The increase in permeance of
more than 100% with an increase in mix duration from 1 min to
6 min is the largest observed effect and may be indicative of an
increase in porosity due to the continued production of air voids
by the air-entraining agent in the mix. The decrease at longer times
(beyond 6 min) suggests that air bubbles are knocked out of the
Vapour permeance (x10-9 kg/m2.s.Pa)

in samples mixed for 1 min indicates that the fresh mortar has suf- 10 5
Water absorpon (% weight gain)

ficient water and has undergone sufficient mixing to initiate the 9

early stage hydration of cementitious grains. The dip in compres- 8 4
sive strength from one to 2 min mixing might represent the break- 7
ing of initially formed weak bonds as seen during the dormant 6 3
5
phase and the beginning of acceleratory stage in hydration of
4 2
cement paste [41 pp. 272]; continued mixing allows the initiation 3
of stronger chemical bonds and the associated recovery of com- 2 1
pressive strength. An increase in mixing time beyond 3 min 1
resulted in a decrease in compressive strength to a low of 0 0
0 5 10 15 20
5.98 MPa (20 min mixing duration), perhaps due to an alteration
Mixing duraon (minutes)
in pore structure associated with the use of air entraining agents.
Although the presence of cement in a mortar [10,28] can have a Water absorpon Vapour permeance

significant impact on its physical properties and especially com- Fig. 8. Impacts of mixing duration on the water absorption and vapour permeance
pressive strength, the manufacturer’s recommended mix time of of hardened Lithomex at 28 days. Vertical bars indicate minimum and maximum
3–5 min provides the material with a compressive strength at values of the samples tested.
364 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367
mix; in this case 6 min could be seen as the optimum for air void
production. Vertical (minimum/maximum) bars in Fig. 8 show an
increasing variability in the permeance of identically prepared
samples with increasing mix duration. Despite the clear
connections of water absorption and permeance (i.e. their link with
porosity), water absorption does not show the same trend.
The constancy of the flow diameter, compressive strength, car-
bonation and vapour permeance between six and 20 min mixing
suggests that a lengthening of the manufacturer’s specified mixing
duration would be beneficial. However, the increased spread of the
results (minimum to maximum) implies a less consistent product
with longer periods of mixing, so the case for a longer mixing time
is finely balanced.
3.3. Surface laitance
Table 1 shows that the presence of laitance inhibits vapour flow
in all samples tested; the shaded boxes indicate those specimens
prepared in line with manufacturer’s guidance. The permeance of
samples mixed according to the manufacturer’s guidelines is
reduced by an average of 13.23% when the surface laitance is not
removed. Laitance has the most noticeable effect on samples that
have been mixed for 6 min (displaying a reduction of 35.71%),
and those that have 120% water content (displaying a reduction
of 32.68%) compared with those samples prepared in accordance
with the manufacturer’s recommendations. This apparently large
reduction in vapour permeance may be due to the fact that over-
mixing and the introduction of excess water can both assist the
migration of fine particles in suspension, i.e. helping laitance form.
Additionally, the increased porosity associated with optimum mix-
ing of an air entraining agent, and the incorporation of excess
water produce higher initial permeance values against which the
‘laitance-intact’ samples are compared.
Given the emphasis that is placed on vapour permeance, or
‘breathability’, of mortars used in conjunction with stone sub-
strates in particular [5,26,44], these results highlight the impor-
tance of surface finishing and the removal of laitance from this
restoration mortar, especially when applied to permeable sub-
strates. These conclusions are consistent with the difference
between the relatively dense few micron-thick laitance layer and
the open-textured main body of the material, visible in Fig. 2.
3.4. Pigmentation
Pigmentation impacts all the properties tested but the
variations in its effects highlight a degree of complexity not
evident in the unpigmented material.
All pigmented samples except Blaxter show increased flow
compared to the unpigmented material. Lee et al. [45] highlight
Table 1
The effects of surface laitance on vapour permeance of Lithomex tested after 28 days of curing.
Lithomex sample Mean vapour permeance (109 kg m2 s Pa)
Water content (%) Mix time (s) With laitance removed
100 1 1.09
100 2 1.28
100 3 1.35
100 4 1.61
100 5 1.71
100 6 2.35
100 20 2.06
80 4 1.18
90 4 0.97
110 4 2.25
120 4 5.20
the correlation between pigment type/colour and fluidity in
cement mortars; red and yellow pigments decreased the flow of
mortars and green pigments increased the flow (the effects of black
pigments varied); it is possible that similar relationships exist
between lime binders and pigments.
Pigmentation reduces the average 28 day compressive strength
of hardened Lithomex from 7.19 MPa to a minimum of 4.60 MPa
and increases average carbonation depth compared to the unpig-
mented material (Table 2), except in the case of Stanton Moor,
which increases in strength and has a reduced carbonation rate.
The colour change (of phenolphthalein indicator) at the boundary
between carbonated and uncarbonated material is more difficult
to identify in darker samples and this can reduce the reliability
of those measurements.
All pigmented samples have a reduced water absorption capac-
ity relative to the unpigmented mortar (Fig. 9); pigmented samples
have a mean water absorption value of 6.57% compared with 8.42%
for the unpigmented mortar. The value varies between pigmented
samples (minimum of 5.36% in St Bees Lithomex and a maximum
of 8.23% in Locharbriggs Lithomex) but there is no apparent corre-
lation with colour tone. Despite this, vapour permeance typically
increases with addition of pigment, with the exception of Stanton
Moor Lithomex, which reduces to 1.41  109 kg m2 s Pa (com-
pared with 1.61  109 kg m2 s Pa for the unpigmented mortar).
There is a general negative correlation between compressive
strength and vapour permeance as shown in Fig. 9; Peak Moor
Lithomex is the exception to this. The impact of pigmentation on
both these properties suggests that pigmentation influences the
overall hydraulicity, as both these properties are understood to
be linked with hydraulicity [44,46]. This could be as simple as an
alteration in the mix ratios, reducing the relative proportion of
hydraulic components. The opposite trend displayed by Stanton
Moor Lithomex might suggest that there is a component in this
pigmented mix that acts as a pozzolan. Pozzolanic properties of
some pigments have been identified in other studies [47,48].
Pigmented samples do not display the same positive correlation
between water absorption and vapour permeance (Fig. 10) as
unpigmented Lithomex (Fig. 5). This relationship might suggest a
more complex influence of pigments on the microstructure of the
mortar that could only be confirmed through detailed microscopic
examination of the materials. Comparison of these moisture trans-
mission values with those of two natural sandstones and a typical
NHL 3.5 mortar (Fig. 10) put these results into perspective and
allow some assessment of the materials’ likely compatibility.
Although water absorption of Lithomex samples is significantly
lower than NHL 3.5 mortar, vapour permeance is generally higher.
Shading in Fig. 10 marks areas of the graph where moisture
transmission properties are higher than those of the natural sand-
stone (yellow area for Stanton Moor sandstone and red area for
With laitance intact Reduction (%)
0.94 13.30
1.03 18.97
1.19 11.61
1.38 14.45
1.60 6.41
1.51 35.71
1.69 18.18
1.04 10.73
0.81 16.76
1.87 16.93
3.50 32.68
 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367 365

Table 2
Properties of pigmented Lithomex compared to unpigmented Lithomex.

Lithomex Colour value (Munsell soil Colour description (Munsell Mean flow diameter Mean 28 day compressive Mean 28 day carbonation
sample colour) soil colour) (mm) strength (MPa) depth (mm)
Unpigmented 10YR 9/1 ‘White’ 152 7.19 5.0
Bearl 7.5YR 8/4 ‘Pink’ 160 4.86 6.3
Blaxter 2.5YR 8/2–8/3 ‘Pale brown’ 151 5.93 5.7
Clashach 10YR 7/2 ‘Light gray’ 160 4.60 6.6
Locharbriggs 5YR 6/3–6/4 ‘Light reddish brown’ 159 5.40 4.7
Peak Moor 10YR 7/3–7/4 ‘Very pale brown’ 155 6.08 6.0
Stanton Moor 10YR 8/2 ‘Very pale brown’ 158 7.84 2.1
St. Bees 7.5R 5/3–5/4 ‘Weak red’ 163 5.33 8.8

across a project, these variations could result in a degree of uncer-

Fig. 9. The relationship between compressive strength and vapour permeance of
hardened Lithomex, cured for 28 days.
Fig. 10. Moisture transmission in pigmented Lithomex. Values for two correspond-
ing sandstones, unpigmented Lithomex and an NHL 3.5 mortar are provided for
comparison; sandstone permeance values from [10].
Locharbriggs sandstone). Lithomex samples which fall within these
areas are likely to be compatible with the corresponding stone type
at least in terms of moisture transmission.
3.5. Discussion
The results presented here highlight the importance of consis-
tency in material preparation. Even slight variations in mix regime
can result in quite significant differences in the restoration mortar
used on a single project, and some variation is evident in samples
prepared in a single batch. Given the assumption of specifiers and
operatives that a single product will have consistent performance
tainty regarding mortar performance.
Although an increased water content reduces average compres-
sive strength and increases average vapour permeance of Lithomex
(which may be interpreted as positive attribute), modifying the
mixing regime in this way results in the increased flow of the
material to an unworkable state. It is acknowledged that some
variation in consistency of Lithomex will be required for site appli-
cation due to the material’s use for repair to different building ele-
ments, i.e. a repair to moulded stone will require a stiffer
consistency than a flat surface repair. Lithomex repairs of different
elements are therefore likely to have different properties irrespec-
tive of what may be deemed technically appropriate based on the
substrate. If the material is prepared to reflect differing workability
requirements, then a moulded stone repair is likely to have higher
compressive strength than a flat surface repair, but a lower carbon-
ation rate, water absorption capacity and vapour permeance.
Lower water absorption capacity could in theory lead to a greater
degree of rainwater run-off onto the masonry below. In cases of
isolated small repair, this is unlikely to impact upon the perfor-
mance of the adjacent masonry, but if repairs are extensive this
could prove problematic.
Based on the seven colours tested, pigmentation typically
results in decreased water absorption. This can prove problematic
both visually, creating a contrast with the adjacent stone, and tech-
nically, in areas where large expanses of restoration mortar result
in excessive water run-off. Given that colour matching is standard
practice in the use of Lithomex [9], it is essential that the influence
of pigment use on material properties be understood and consid-
ered prior to its use.
The numerous potential uses and applications of Lithomex on
varying substrates types, of different colours and by operatives of
differing skill levels makes it impossible to determine the full range
of properties of all variations of the product. However, the results
presented here take the first step in highlighting the potential
impacts of site practice of formulated materials. Such factors,
undoubtedly affecting the properties, performance and therefore
suitability of the mortar for different applications should be con-
sidered in the material’s specification and application. To ensure
materials of known properties are used, it is vital to acknowledge
manufacturers’ guidance in mortar preparation and application
and avoid deviation from these where at all possible.
Comparing the properties of Lithomex to those of natural stone
can give some indication of the material’s likely compatibility; of
the properties tested, moisture transmission is perhaps the most
important criterion. The variation in properties of both materials
(i.e. natural stone types and Lithomex) and the comparison
between them (Fig. 10) indicates that not all variations of Lithomex
will have superior moisture transmission characteristics than
stone. More work into the properties of, and degree of variation
between, different historic substrates is required to help asses if
this is likely to be a limiting factor in the use of this proprietary
restoration mortar.
366 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367
This work provides information relating to performance of res-
toration mortar that will influence compatibility with the underly-
ing substrate. This is particularly valuable for those working in the
heritage sector, but is also directly relevant to new build scenarios
where the material is used in conjunction with modern substrates
(e.g. concrete block or reconstituted stone). These results give an
indication of permissible tolerances afforded in the preparation of
this material without causing detriment to the material itself, or
the substrate to which it is applied. It is reasonable to assume that
the findings of this study will prove useful in understanding other
formulated hydraulic lime–cement hybrids (mortars, renders,
etc.) which would be expected to exhibit comparable behaviour.
4. Conclusions
The results of this first detailed study on the post production
modification of restoration mortar will be valuable to those dealing
with repairs to historic masonry, from specification, through to site
preparation and quality control. The relationship between site
practice and material properties is complex. This work focused
upon the evaluation of four principal factors relating to modifica-
tion of Lithomex, namely; water content, mix duration, influence
of surface laitance and pigmentation. There are clear correlations
of mixing time, water content and pigmentation with flow,
compressive strength, water absorption and permeance.
It has been shown that increasing the water content of Litho-
mex impacts the flow of the fresh mortar as well as its hardened
properties of carbonation rate, compressive strength and water
absorption. Mix duration is less influential than water content in
relation to carbonation rate and compressive strength. Addition-
ally, mix duration has a lower impact on average flow but does
result in a greater range of flow values (i.e. a less reproducible
product). Generally, pronounced enhanced workability character-
istics are evident with mixing up to 6 min. Beyond this, flow char-
acteristics are not enhanced. A clear relationship is exhibited
between low-medium duration mixing (66 min) and increased
permeance; although there is increased variability in permeance
in samples mixed for 4 min or more. The increased permeance is
likely to be attributed to the positive influence of air entrainers
and their associated pore structure modification. Beyond this
point, a reduction in permeance was noted as a result of reduced
air content.
The development of laitance has been shown to reduce vapour
permeance and this highlights the need for its removal. Those
materials with higher water ratios were most likely to be prone
to laitance development. This work quantifies the impacts of sur-
face laitance on the material, highlighting the imperative need
for the operative to have an understanding of appropriate surface
finishing techniques.
With respect to the desirable properties of a restoration mortar
(i.e. a material that remains sacrificial and does not inhibit the
transmission of moisture), pigmentation is perhaps the most influ-
ential form of modification as it typically decreases compressive
strength and increases vapour permeance, at least across the range
of colours analysed in this study. Of the samples tested, Stanton
Moor Lithomex seems to be the exception to this, as the only case
in which pigmentation resulted in an increase in compressive
strength and a reduction in vapour permeance; chemical analysis
of pigments may reveal the mechanisms at play and highlight
any potential pozzolanic properties of the pigments. Despite this
one occurrence, Lithomex has a very low strength compared to
the substrates it will be applied to (e.g. Stanton Moor and Loch-
arbriggs sandstones have compressive strengths of 73.33 MPa
[49] and 57.42 MPa [50] respectively, compared to a maximum
of 7.84 MPa for Lithomex in the samples tested). This is unlikely
to cause compatibility issues with sound masonry substrates, but
this work has shown the need for an understanding of other
properties.
Acknowledgements
Mr. Alistair McFarlane and Mr. Graham Sorley (Heriot-Watt
University) are thanked for their technical assistance with physical
testing. Dr. Susanna Kirk (formerly National Museums of Scotland)
is thanked for technical assistance with the SEM. Dr. Alick Leslie
(Historic Scotland) is thanked for providing comments on an ear-
lier version of this manuscript. This research was funded by His-
toric Scotland.
References
[1] Murphy JM, Sexton DMH, Jenkins GJ, Boorman PM, Booth BBB, Brown CC, et al.
UK climate change projections science report: climate change projections:
technical report. Exeter: Meteorological Office Hadley Centre; 2009.
[2] Inkpen R, Viles H, Moses C, Baily B. Modelling the impact of changing
atmospheric pollution levels on limestone erosion rates in central London,
1980–2010. Atmos Environ 2012;61:476–81.
[3] Cassar M, Hawkins C. Engineering historic futures: adapting historic
environments to moisture related climate change; stakeholders
dissemination and scientific research report’ UCL; 2007.
[4] Hulme M, Jenkins GJ, Lu X, Turnpenny JR, Mitchell TD, Jones RG, et al. Climate
change scenarios for the United Kingdom: the UKCIP02 Scientific Report’
Tyndall Centre for Climate Change Research. School of Environmental
Sciences: University of East Anglia; 2002. p. 120.
[5] Odgers D, Henry A. English heritage practical building conservation:
stone. Surrey: Ashgate; 2012.
[6] Forster AM. Building conservation philosophy for masonry repair: Part 1 –
‘‘ethics’’. Struct Surv 2010;28:91–107.
[7] Forster AM. Building conservation philosophy for masonry repair: Part 2 –
‘‘principles’’. Struct Surv 2010;28:165–88.
[8] Torney C, Hyslop EK. Considerations for masonry repairs in traditional
buildings, short guide. Edinburgh: Historic Scotland [in preparation].
[9] Torney C, Forster AM, Kennedy CJ, Hyslop EK. ‘Plastic’ repair of natural stone in
scotland: perceptions and practice. Struct Surv 2012;30:297–311.
[10] Torney C, Forster AM, Szadurski EM. Specialist ‘restoration mortars’ for
building repairs: a comparison of the physical properties of two stone repair
materials. Heritage Sci 2014;2(1).
[11] Schueremans L, Cizer Ö, Janssens E, Serré G, Van Balena K. Characterization of
repair mortars for the assessment of their compatibility in restoration
projects: research and practice. Constr Build Mater 2011;25:4338–50.
[12] Henry A, Stewart J. English heritage practical building conservation: mortars,
renders and plasters. Farnham: Ashgate Publishing Limited; 2011.
[13] Forster AM, Carter K. A framework for specifying natural hydraulic lime
mortars for masonry construction. Struct Surv 2011;29:373–96.
[14] van Hees R. RILEM TC 203-RHM: repair mortars for historic masonry. Repair
mortars for historic masonry. From problem to intervention: a decision
process. Mater Struct 2012;45:1295–302.
[15] Hughes JJ. RILEM TC 203-RHM: repair mortars for historic masonry. The role of
mortar in masonry: an introduction to requirements for the design of repair
mortars. Mater Struct 2012;45:1287–94.
[16] Snow J, Torney C. Lime mortars in traditional buildings, historic scotland short
guide 6. Edinburgh: Historic Scotland; 2014.
[17] Papayianni I, Pachta V, Stefanidou M. Analysis of ancient mortars and design of
compatible repair mortars: the case study of Odeion of the archaeological site
of Dion. Constr Build Mater 2013;40:84–92.
[18] Mikos E, Schubert P, Knöfel D. Zur Beurteilung von steinergänzungsstoffen für
die restaurierung von natursteinoberflächen. Bautenschutz Bausanierung
1992;15(4):34–40.
[19] Rodrigues DJ, Grossi A. Indicators and ratings for the compatibility assessment
of conservation actions. J Cult Heritage 2007;8:32–43.
[20] Gulotta D, Goidanich S, Tedeschi C, Nijland TG, Toniolo L. Commercial NHL-
containing mortars for the preservation of historical architecture. Part 1:
compositional and mechanical characterisation. Constr Build Mater
2013;38:31–42.
[21] Forster AM, Carter K, Banfill PFG, Kayan B. Green maintenance for historic
masonry buildings: an emerging concept. Build Res Inform
2011;39(6):656–64.
[22] Valek J, Hughes JJ, Groot CJWP. Historic mortars: characterisation, assessment
and repair 2012 RILEM. Springer; 2012.
[23] Allen G, Allen J, Elton N, Farey N, Holmes S, Livesey P, et al. Hydraulic lime
mortar for stone. In: Brick and block masonry. Dorset: Donhead; 2003.
[24] Lanas J, Sirera R, Alvarez JI. Study of the mechanical behavior of masonry repair
lime-based mortars cured and exposed under different conditions. Cem Concr
Res 2006;36:961–70.
[25] Varas MJ, Alvarez de Buergo M, Perez-Monserrat E, Fort R. Decay of the
restoration render mortar of the church of San Manuel and San Benito, Madrid,
 C. Torney et al. / Construction and Building Materials 75 (2015) 359–367
Spain: results from optical and electron microscopy. Mater Charact
2008;59:1531–40.
[26] Gibbons P. The preparation and use of lime mortars: technical advice note
1. Edinburgh: Historic Scotland; 2003.
[27] Setra Marketing Ltd. Health and safety recommendations for Lithomex from
St. Astier, 2008 <http://www.stastier.co.uk/lithomex/lithomex-hs.php> April
2013 [downloaded].
[28] Setra Marketing Ltd. Stone repair mortar – Lithomex from St. Astier. Plastic
repair materials for natural stone and masonry, 2013. <http://
www.stastier.co.uk/lithomex/stonerepair.php> April 2013 [downloaded].
[29] Neville AM. Properties of concrete. 4th ed. England: Longman; 1995.
[30] Frew C. Pointing with Lime. In: The building conservation
directory. Tisbury: Cathedral Communications Limited; 2007.
[31] British Standards Institution, BS EN 1015-19:1999. Methods of test for mortar
for masonry – determination of water vapour permeability of hardened
rendering and plastering mortars. London: BSI; 1999.
[32] British Standards Institution, BS EN 459-2:2010. Building lime – Part 2: test
methods. BSI: London; 2010.
[33] Hanley R, Pavía S. A study of the workability of natural hydraulic lime mortars
and its influence on strength. Mater Struct 2008;41:373–81.
[34] Building research establishment: BRE stone list <http://projects.bre.co.uk/
stonelist/> [accessed 06.2014].
[35] Dunhouse natural stone Website <http://www.dunhouse.co.uk> [accessed
06.2014].
[36] Stancliffe stone Website <www.stancliffe.com> [accessed 06.2014].
[37] Blockstone Limited Website <www.blockstone.co.uk> [accessed 06.2014].
[38] McMillan AA, Gillanders RJ, Fairhurst JA. Building stones of Edinburgh. 2nd
ed. Edinburgh: Edinburgh Geological Society; 1999.
[39] British Standards Institution, BS EN 1015-3:1999. Methods of test for mortar
for masonry. Determination of consistence of fresh mortar (by flow
table). London: BSI; 1999.
367
[40] Costigan A, Pavia S. Influence of mortar water content on the strength of lime
mortar masonry. In: Ní Nualláin, Walsh, West, Cannon, Caprani, McCabe,
editors. Proceedings of BRI/CRI, Cork; 2010. p. 449–56.
[41] Hewlett PC. Lea’s chemistry of cement and concrete. 4th
ed. Oxford: Butterworth-Heinemann; 2004. p. 272.
[42] Rodriguez-Navarro C, Hansen E, Ginell W. Calcium hydroxide crystal evolution
upon aging lime putty. J Am Ceram Soc 1998;81:3032–4.
[43] Hansen EF, Tagle A, Erder E, Baron S, Connell S, Rodriguez-Navarro C, Van Balen
K. Effects of aging on lime putty’. In: Bartos B, Groot C, Hughes JJ (Eds.)
Proceedings of the international RILEM workshop (PRO 12), Historic mortars:
characteristics and tests. Cachan (France): RILEM Publications S.A.R.L; 1999. p.
197–206.
[44] Banfill PFG, Forster AM. A relationship between hydraulicity and permeability
of hydraulic lime, In: Bartos B, Groot C, Hughes JJ (Eds.) Proceedings of the
international RILEM workshop (PRO 12), historic mortars: characteristics and
tests. Cachan (France): RILEM Publications S.A.R.L; 1999. p. 173–83.
[45] Lee H-S, Lee J-Y, Yu M-Y. Influence of inorganic pigments on the fluidity of
cement mortars. Cem Concr Res 2005;35:703–10.
[46] Forster AM. How hydraulic lime binders work. In: Hydraulicity for Beginners
and the Hydraulic Lime Family. Edinburgh: Love Your Building Publishing;
2004.
[47] Pera J, Boumaza R, Ambroise J. Development of a pozzolanic pigment from red
mud. Cem Concr Res 1997;27:1513–22.
[48] Giralso C, Tobón JI, Restrepo Baena OJ. Ultramarine blue pigment: a non-
conventional pozzolan. Constr Build Mater 2012;36:305–10.
[49] Stancliffe stone: Stanton Moor Buff sandstone data sheet <http://
www.stancliffe.com/Content/ProductInformation/PDFs/StoneTypes/Stancliffe-
Stone-Stanton-Moor-Buff-Sandstone.pdf> [accessed 09.2014].
[50] Stancliffe stone: locharbriggs sandstone data sheet <http://
www.stancliffe.com/Content/ProductInformation/PDFs/StoneTypes/Stancliffe-
Stone-Locharbriggs-Red-Sandstone.pdf> [accessed 09.2014].