Construction and Building Materials 159 (2018) 376–389

Contents lists available at ScienceDirect

Construction and Building Materials

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

Performance of a lime-based insulating render for heritage buildings

Yves Govaerts a,⇑, Roald Hayen b, Michael de Bouw c, Ann Verdonck a, Wendy Meulebroeck d, Stijn Mertens e,
Yves Grégoire e
a
Vrije Universiteit Brussel, Department of Architectural Engineering, Pleinlaan 2, B-1050 Brussels, Belgium
b
Royal Institute for Cultural Heritage, Monuments Lab, Parc du Cinquantenaire 1, B-1000 Brussels, Belgium
c
Belgian Building Research Institute, Lab of Renovation, University of Antwerp, Department of Design Sciences, Mutsaardstraat 31, B-2000 Antwerp, Belgium
d
Vrije Universiteit Brussel, Department of Applied Physics and Photonics, Brussels Photonics (B-PHOT), Pleinlaan 2, B-1050 Brussels, Belgium
e
Belgian Building Research Institute, Division of Materials, Avenue P. Holoffe 21, B-1342 Limelette, Belgium

h i g h l i g h t s

 Moisture is stored for a longer time when multiple NHL top layers are applied.
 The considered insulating lime render has got a low risk to develop frost damage.
 Replacing imitative renderwork by lime-perlite mortar encourages moisture migration.
 The render’s thermal insulation efficiency is lower in winter compared to summertime.

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

Article history: Insulating heritage buildings in order to provide a comfortable indoor climate, to reduce energy con-
Received 1 April 2017 sumption and avoid thermal stresses into the structure, is generally not in accordance with architectural
Received in revised form 16 October 2017 preservation guidelines. Without compromising historic features, insulation renders may however
Accepted 26 October 2017
upgrade rendered heritage of which the original materiality is considered inferior compared to its design.
This study evaluates the hygrothermal performance of 49 insulating lime-perlite render configurations
on brick masonry, by combining characterisation experiments with numerical modelling. The effects of
Keywords:
the coupled heat-moisture transport are finally validated by ageing cycles and bond strength tests. It
Insulating render
Material characterisation
is found that insulating renders cause smaller shifts in temperature in the structure compared to historic
Hygrothermal modelling renders, but the moisture level of the support may increase significantly. In addition, the insulation layers
Ageing cycles store a lot of moisture during winter, affecting the expected insulation capacity.
Stone imitation Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction render layers based on cement [3,4]. Whereas the decorative

hybrid lime-cement renders were extensively applied for stone
Throughout hundreds of years, a high number of historic imitation purposes during the interwar period [5], pure Portland
façades has been finished with mortar layers. Although these ren- cement renders were frequently used for finishes like glass-
ders were primarily employed as a protecting envelope against sprinkled cimorné and bright coloured tyrolean mortars for Moder-
weathering to provide a longer life span for the underlying building nist architecture [6,7]. Such finishes were commonly obtained as a
materials, their decorative aspects were also of great importance to thin top layer on a less expensive rendering layer of 5–30 mm
strive for aesthetical perfection [1]. Literature shows that the older thickness, to ensure an adequate adhesion and levelling in relation
mortars consist of air lime or hydraulic lime mixed with sand or to the masonry structure underneath. Despite the valuable decora-
crushed aggregates derived from stone or brick [2], while a signif- tive aspects of these traditional render systems, their thermal con-
icant share of our young architectural heritage is associated with ductivity is high compared to the EN 998-1 definition of a thermal
insulating render (k  0.20 W/mK), and typically ranging from 0.12
to 0.80 W/mK in dry condition [8–11]. As a consequence, their
⇑ Corresponding author.
insulation capacity is rather low and insufficient at the given thick-
E-mail addresses: [email protected] (Y. Govaerts), [email protected]
(R. Hayen), [email protected] (M. de Bouw), [email protected]
ness, especially when the air volume within the pores is largely
(A. Verdonck), [email protected] (W. Meulebroeck), [email protected] replaced by rainwater, which leads to an increase of the thermal
(S. Mertens), [email protected] (Y. Grégoire).

https://doi.org/10.1016/j.conbuildmat.2017.10.115
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 Y. Govaerts et al. / Construction and Building Materials 159 (2018) 376–389 377

conductivity. Lacking insulation is generally accompanied by tee a constant performance in time, but their behaviour is still
pathological effects, generated by surface condensation near ther- unclear under external conditions. Therefore, the aim of this
mal bridges. Long-lasting moisture levels may trigger the growth research is to evaluate the overall performance of multiple insula-
of moulds that affect the air quality of the building occupants. Also tion render configurations on clay units masonry, and compare
the thermal comfort may be undermined due to unpleasant radi- their hygrothermal behaviour with a traditional solution. The ren-
ance from cold inner surfaces, making these valuable historic der configurations are established by combining one particular
buildings less suitable for exploiting a residential function. Because insulation render with different finishing layers or by adding an
they additionally consume a lot of energy, resulting in high green- undercoat. A parameter study should decide which configuration
house gas emissions and high heating costs, the refurbishment of performs best with regard to insulation capacity, moisture trans-
the existing building stock is recently extensively promoted by port and ageing.
the EU government [12,13]. However, it appears to be a challenge
to find the right balance between preservation of heritage values
2. Methodology
and an improvement of the occupant comfort. Within this context,
a considerable number of European research projects were initi-
To understand the moisture balance within the render and its
ated in recent years [14–16].
support, and to assess and mitigate all risks for future degradation,
the necessary experiments are carried out to determine the den-
1.1. Searching compatible interventions
sity, thermal conductivity, capillary water absorption, drying rate
and porosity properties of the individual render layers. Although
In order to retain the aesthetic and historic integrity of a façade,
these experiments provide fundamental information on the beha-
and at the same time to reduce unnecessary heat losses, an internal
viour of each individual render layer, their combined hygrothermal
insulation layer is often preferred. However, this retrofitting strat-
behaviour is difficult to assess experimentally. Therefore, these
egy has an adverse effect on the spatial perception of the inner
parameters are implemented into a numerical model (DELPHIN)
spaces and may lead to moisture accumulation within the façade
to simulate their water uptake, mutual moisture exchange, drying
with potential interstitial condensation. According to the level of
behaviour and change of insulation capacity over a time period of
heritage protection, an external insulation system can be taken
2 years. Different render configurations with variable insulation
into consideration. Historic façades that enjoy integral protection
thicknesses, base and top layers are evaluated in terms of their
are unfortunately not eligible for such interventions, because the
yearly moisture content, indoor temperature and condensation
materiality of their surface is unique and has to be preserved. Nev-
risks.
ertheless, a larger share of our patrimony contributes particularly
In addition, the experiments involve ageing cycles, as well as
to the cultural identity of townscapes or cities, and allows ener-
impact and adhesion tests on wall specimens with combined ren-
getic rehabilitation measures if the façade appearance is main-
der layers to validate their resistance against weathering. Finally,
tained or reconstructed to conserve its identity. Yet, the
the best insulating geometries are assessed against two traditional
application of external insulation layers can be quite encroaching
façade sections: one without renderwork, and one containing an
when the original rendering is replaced by a new render with a dif-
interwar stone imitation finish.
ferent appearance and thickness which distorts the initial façade
proportions. Especially when a façade is characterized by orna-
mental elements, the attachment of traditional insulation panels 3. Materials and samples
as part of an ETICS is very time-consuming. Instead, stylistic ele-
ments may be replicated using insulation mortars without increas- 3.1. Selection of the insulating render system
ing the finishing’s thickness significantly. This radical strategy does
not intend to comply with the current energy standards, because The thermal conductivity of innovative and experimental insu-
the authentic façade concept and design are considered more lation renders is situated between 0.06 and 1.33 W/mK [18–20,23–
important than optimal thermal comfort. Therefore, only a limited 25]. Aerogel mortars form an exception and serve as the best insu-
render thickness is appropriate to reduce energy losses and lators (up to 0.028 W/mK), making their potential towards the
improve the comfort level of occupants. However, before the feasi- building conservation industry a hot topic [26–28]. However, her-
bility level of an upgraded reconstruction strategy towards historic itage agencies often prefer a more traditional render composition,
features can be evaluated, the technical compatibility and durabil- due to compatibility concerns with regard to the water repellent
ity of an insulating render package on a historic support needs to properties of an aerogel render [28]. Therefore, this study concen-
be investigated. The awareness that heritage buildings should ful- trates on a dry premixed commercial render basically composed of
fill certain minimum requirements towards thermal comfort has natural hydraulic lime (NHL5), sand and perlite fragments. Its pre-
recently grown, and the development of improved insulating mor- scribed insulation capacity (k = 0.066 W/mK) is substantially
tars is trending [17]. Researchers have tried to lower the thermal higher in comparison to other insulating mineral renders (Class 1
conductivity of mortars by substituting the traditional aggregates render according to EN 998-1), which makes it convenient for
with lightweight components. Barreca & Fichera performed some obtaining higher thermal comfort levels for identical layer thick-
experiments with the addition of olive stone to lime-cement mor- nesses. One of the advantages is that lime-based mortars can easily
tars [18], Brás et al. considered hydraulic lime renders containing be removed from their substrate, so their reversibility aspect is
different cork percentages instead of conventional aggregates guaranteed.
[19] and Stefanidou incorporated respectively silica sand, glass par-
ticles and glass spheres in different lime-cement renders [20]. Most 3.2. Sample preparation and curing
studies within this topic only focus on the characterisation of the
thermal performance, but the assessment of their general durabil- A common insulating render system is composed of an adhesive
ity and hygric behaviour under weathering conditions is often for- spritz coating (render A), an insulation layer (render B), an inter-
gotten. Only a minority considers an on-site measurement of the mediate layer (render C) and a decorative layer (render D). X-ray
thermal conductivity and takes into account the moisture beha- diffraction (XRD) was used to characterize the crystalline phases
viour of the façade, but unfortunately their focus usually lies on of the ingredients from each render in powdered condition. The
interior applications [17,21,22]. Insulation renders should guaran- XRD patterns indicate that the adhesive render (A), as well as the