Enhancement of the thermal performance of perforated clay brick walls through the addition of industrial nano-crystalline aluminium sludge

Construction and Building Materials 101 (2015) 227–238

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Enhancement of the thermal performance of perforated clay brick walls through the addition of industrial nano-crystalline aluminium sludge Paulo Santos a,⇑, Cláudio Martins a, Eduardo Júlio b a b

ISISE, Department of Civil Engineering, University of Coimbra, Coimbra, Portugal CEris, DECivil, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, Portugal

h i g h l i g h t s
Characterization of innovative eco-thermo-efficient hollow clay bricks.
Recycling of industrial nano-crystalline aluminium sludge.
Eco-efficient bricks with improved thermal performance (U-value almost 10% lower).
Combined experimental/numerical approach to assess thermal performance.
Importance of mortar joints in the overall wall thermal transmittance value.

a r t i c l e

i n f o

Article history: Received 24 April 2015 Received in revised form 17 August 2015 Accepted 15 October 2015

Keywords: Clay brick Thermal performance Aluminium sludge Nano-crystalline Industrial waste Eco-efficient brick

a b s t r a c t One of the most common approaches to contribute to the sustainability of the construction sector consists in adding industrial by-products to raw materials. However, this generally presents the major drawback of leading to a loss of the product’s properties. The study herein presented is part of a research project, developed in co-promotion with a manufacturer of perforated clay bricks, with the final goal of incorporating industrial nano-crystalline aluminium sludge in the raw material, but with the additional goal of improving the thermal properties of the original product. A combined experimental/numerical approach has been used to assess the thermal performance of walls built with the new eco-efficient perforated clay bricks. First, tests were conducted on laboratorial prototypes and relevant parameters were measured. Next, a numerical analysis was performed using three-dimensional finite element models, calibrated and validated using the experimental results. Furthermore, the results were compared with other simplified models (2D) and its reliability assessed. The obtained results show that the new eco-efficient perforated clay bricks exhibit better thermal performance than the original product, leading to almost 10% improvement of its thermal transmittance. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The improvement of the thermal performance of buildings’ envelope is essential to ensure the energy efficiency of buildings, responsible for 40% of energy consumption in developed countries [1]. A major share of the latter concerns the ambient conditioning inside (space heating and cooling) which presents an increasing trend given people’s growing needs in terms of comfort. One of the most important components regarding heat losses in a building are the external walls, due to their very significant exposed area. ⇑ Corresponding author at: Departamento de Engenharia Civil, Faculdade de Ciências e Tecnologia da Universidade de Coimbra – Pólo II, Rua Luís Reis Santos, 3030-788 Coimbra, Portugal. E-mail address: [email protected] (P. Santos). http://dx.doi.org/10.1016/j.conbuildmat.2015.10.058 0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.

For all these reasons, the development of new solutions (for external walls) with improved thermal performance is a highly relevant and up-to-date subject. Furthermore, the European policies (e.g. the energy performance building directive – European Directive 2010/31/EU [2]) are pushing for the development of more efficient buildings and, therefore, for more efficient building components and materials. According to this directive all new buildings will be ‘nearly zeroenergy’ by 2020. Consequently, both European and member states standards are being updated to copy with these more challenging targets. The following examples can be presented: the Portuguese buildings energy certification system [3], the basic document for limitation of energy demand of the Spanish building technical code [4], and the technical guidance ‘document L’ of the Irish building regulation [5].

228

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

Previous studies have addressed the improvement of the thermal performance of external walls using different strategies with different aims: (i) improving the thermal transmittance coefficient or U-value; (ii) increasing the thermal mass (or thermal inertia); and/or (iii) correcting thermal bridges. Regarding the thermal transmittance coefficient, the following solutions have been studied: (i) the use of alveolar bricks [6,7], and (ii) the adoption of lightweight steel framed walls [8], among others. Furthermore, other influencing parameters have also been analysed in this scope: (i) the number of wall panes [9], (ii) the number, relative position and thickness of insulation layers [10], and (iii) the adoption of an air–gap, and the ventilation level of the latter [11]. In what concerns the relevance of the thermal mass (or thermal inertia) of external walls for their thermal performance, this is mainly due to the fact that they are subjected to thermal dynamic loads [6]. Previous studies have been conducted aiming at improving the thermal performance of walls by adding some supplementary thermal mass, e.g. using Phase Change Materials – PCMs [12– 14]. Regarding thermal bridges, there are also published studies (e.g. [15]) on the effect of mortar joints across insulation layers in masonry walls, under dynamic conditions. According to the latter work [15], thermal bridges can reduce the thermal resistance value in 53% and increase the transmission loads by 103%. Naturally, the significance of the increase of thermal bridges with the increase of the thermal conductivity of the material crossing the insulation layer is particularly relevant in the case of steel frames [16]. Focusing on the use of perforated clay brick units in external walls, there are two main approaches to improve their thermal behaviour. The first one consists in increasing the thermal path of heat flow by changing the perforation geometry of the bricks. This issue was addressed by [17] and it was found that a rhomboid layout of voids with the longer diagonal at right angles to the heat flux is the best internal layout. The second one consists in adding other materials to the raw material (clay or cement) during the production process in order to reduce its thermal conductivity. In [18] recycled paper mill residue and rice husk ash were used in the production of lightweight cement bricks. Besides the obvious environmental benefits, these authors also found out that the increased porosity of the resulting bricks, which led to their lightweight and thus lower thermal conductivity, did not change the needed compressive strength requirements. In [19], the properties of new cement bricks made of fly ash, quarry dust and billet scale are studied. It is reported that the use of these industrial wastes allows obtaining bricks with good and promising performance, not only regarding the ecological and environmental gain, but also in what concerns the obtained mechanical properties and durability. The study herein described refers to the thermal characterization of building walls made of perforated clay bricks containing sludge from the aluminium industry in nano-crystalline form. The main motivation for the research project that supports this study was to solve an environmental problem by adding this industrial waste to the clay brick raw material, with the additional advantages of both improving their thermal behaviour and eliminating the costs associated to the waste disposal. The aluminium sludge is the waste that results from the anodising and lacquering processes of the aluminium alloys. These techniques are used to protect the metallic materials from corrosion and to provide some aesthetic effects. Currently, the aluminium sludge is classified as non-toxic and inert, but the deposition in landfill of considerably high amounts of this industrial waste is quite expensive. This paper is organised in the following sections: (i) description of the wall system and components, (ii) experimental set-up and instrumentation for assessing both thermal conductivity and ther-

mal transmittance, (iii) numerical approach, including the discretization domain, the boundary conditions, and the modelling of air layers, (iv) results and discussion, including a comparison between experimental and numerical values, and (v) conclusions. 2. Wall system and components In this section the new hollow clay brick units are presented and the construction of the single pane masonry wall system is described. 2.1. Brick units In the present study, two types of hollow brick units were analysed: (1) the ‘original brick’, made of clay without any addition of aluminium sludge; and (2) ‘ecological brick’, made of clay with an addition (5% in weight) of nano-crystalline aluminium sludge. The original brick units are manufactured by Preceram, S.A., under the commercial label ‘‘Thermal and Acoustic Brick 30  19  24”. The geometry of the new ‘ecological brick’ is exactly the same of the original one (Fig. 1). 2.2. Wall construction system The wall construction system consists of a single pane type, plastered in both faces with cement mortar (18 mm in each surface). Only the horizontal joints between bricks are filled with cement mortar, although a 100 mm air gap (splitting the mortar joint into two) is considered (Fig. 2), in order to minimize the thermal transmission through this material. 3. Experimental tests The experimental tests performed and the results obtained, used to calibrate and validate the numerical study, are briefly presented in this section. First, the thermal conductivity of both ceramic materials (‘original’ and ‘ecological’) and mortars used in the finishing and joint layers of the wall are presented. Next, physical properties such as the density and the percentage of voids of the perforated ‘original’ and ‘ecological’ brick units are listed. Finally, it is explained how the wall thermal transmittance was assessed using the calibrated hot box apparatus and results for two test specimens – one produced with ‘original’ and the other with ‘ecological’ brick units, and considering readings before and after the mortar finishing layer was applied – are presented.

3.1. Ceramic material and mortar Nowadays, there are several standards available with tabulated design values regarding hygrothermal properties of materials. Some examples are: ISO 10456 (2007) [20], EN 12524 (2000) [21] and DIN 4108-4 (2013) [22]. However, given the new materials being studied herein and the large variability associated with the properties referred to, it was decided to measure the hygrothermal properties of all materials used in this work.

Fig. 1. Hollow clay brick unit geometry and dimensions.

229

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238 Table 2 Measured brick unit physical properties. Properties 3

Real density [kg/m ] Apparent density [kg/m3] Water absorption [%] Percentage of voids [%]

Original brick

Ecological brick

2030 890 9.5 56.0

1920 770 12.5 60.0

4. Numerical simulations

Fig. 2. Masonry brick wall test specimen assembly: air gap within the horizontal mortar joint.

A three-dimensional numerical simulation was first conducted with the main objective of reproducing the experimental test results and thus of calibrating and validating the model. Next, several simplified approaches were used in order to assess its reliability and limitations. 4.1. Three-dimensional approaches

The tests to measure the thermal conductivity of the original ceramic material, the new eco-efficient ceramic material, and the mortar were performed using a Thermal Constants Analyser test system (Hot Disk TPS 2500s), using the Transient Plane Source (TPS) method, in accordance with ISO 22007-2 (2012) [23]. For each type of brick three test-specimens were analysed, being the obtained results displayed in Table 1 with a standard deviation of 0.03 W/(m K). As expected, the thermal conductivity is higher for the cement mortar than for the ceramics (Table 1). Additionally, the new ceramic material, incorporating nano-crystalline aluminium sludge, shows a significantly lower thermal conductivity (15.6% reduction) when compared with the original one. This lower thermal conductivity is expected to enable an increased thermal performance of the ‘ecological brick’ unit and consequently of the resulting masonry wall.

3.2. Hollow brick units Several experimental tests were performed in order to obtain and compare the physical properties of both ‘original brick’ and ‘ecological brick’. The physical properties of the hollow bricks were characterised using standardized tests: real and apparent density [24], water absorption [25] and percentage of voids [26]. The obtained results are listed in Table 2. The lower density of the ecological brick unit is due to the higher porosity of the brick material, given the nature of the added nano-crystalline sludge. This fact is the rationale for the increase registered in thermal performance, i.e., lower thermal conductivity of the new ceramic material and consequent expected lower thermal transmittance of the brick units and of the assembled masonry wall.

3.3. Masonry brick walls The thermal transmittance of the single pane masonry brick walls was measured using a calibrated hot box apparatus. The equipment requirements and the test procedure are defined in ISO 8990 (1996) [27] and ASTM C1363 (2011) [28]. Two different walls were tested: (1) ‘original brick’ wall; and (2) ‘ecological brick’ wall. Each one was tested at two construction stages: (a) before and (b) after applying the mortar finishing layer. Fig. 3 illustrates the test-specimen at stage (a) and its dimensions. The temperature difference, in steady-state, between the cold and the hot sides of the hot box apparatus during the tests was 20 °C. Table 3 shows the thermal transmittance values measured using the calibrated hot box technique. As expected, given the above-mentioned thermal conductivity results, the ‘ecological brick’ wall exhibits lower thermal transmittance values (1.5% and 3.2%, before and after the mortar finishing, respectively).

Table 1 Measured thermal conductivities (k) of materials. Material

k [W/(m K)]

Ceramic

Original Ecological

0.64 0.54 (15.6%)

Mortar

Finishing Joint

0.80 1.80

The three-dimensional (3D) model was implemented using a computational fluid dynamics (CFD) finite element method (FEM) based software: ANSYS CFX (2013) [29]. The geometry of the 3D model of the wall was defined according to the geometry of the test-specimen previously described (Fig. 3). The FEM brick model comprised 2 218 488 nodes, a number that was found to be sufficient to provide good convergence and beyond which the results did not change. In order to reproduce the steady-state temperature difference between the two sides of the wall in the calibrated hot box, different temperatures (20 and 0 °C) were assumed in the 3D model, respectively in the hot and cold side of the apparatus. These ambient temperatures are average values measured during the experimental tests. The film coefficients values for interior surface is 7.69 and 25 W/(m2 K) for the exterior surface. The following difficulties were identified in modelling the wall test-specimen: (1) the high number of brick units needed to model the entire wall; (2) the complex geometry of each brick unit, containing a solid material (clay) and a fluid (air); (3) the additional horizontal mortar layers between each brick unit level, including an air gap; and (4) the surrounding vertical and horizontal mortar layers along the wall perimeter. Since to include all these details in the model would require huge computation resources and time, it was decided to perform several simplifications in the brick model and in the test-specimen wall, as described in the following subsections and illustrated in Table 4. Nevertheless, both the precision loss originated by each simplification and the reliability of the obtained results were assessed. 4.1.1. 3D Brick Model A: fluid airspaces In this first approach (the most accurate one: Model A), the air was directly modelled as a fluid using the CFD option in the 3D hollow clay brick model as illustrated in Fig. 4. Fig. 4a displays the detailed 3D brick unit model and Fig. 4b shows the finite element mesh, which provided convergence, i.e., no results difference by implementing further refinement. In this approach, the buoyancy model was selected for convection and the Monte Carlo model for radiation. The value of the air thermal conductivity was 0.0261 W/(m K). The obtained brick unit U-value, assuming ‘original’ and ‘ecological’ ceramic material, was 0.830 and 0.748 W/(m2 K), respectively. 4.1.2. 3D Brick Model B: solid equivalent airspaces In this simplified approach, the air was modelled assuming an equivalent thermal conductivity of a solid computed according to ISO 6946 (2007) [30] for small unventilated airspaces. For minor airspaces in building components, other than glazing, which have

230

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

0.298

0.052 0.130 0.018

1.240

0.190

[m] 1.480

(a) Wall test-specimen before applying the mortar finish

(b) Sketch of the test specimen and dimensions

Fig. 3. Adopted test-specimen of masonry brick wall.

Table 3 Overall measured thermal transmittance values of the walls. Brick units

Mortar finishing

U-value [W/(m2 K)]

Original

Measured before Measured after

1.31 1.24

Ecological

Measured before Measured after

1.29 1.20

air gaps with a width less than 10 times its thickness, the standard gives the following equation to compute the thermal resistance (Rg ):

Rg ¼

1 ha þ hr

ð1Þ

where ha is the conduction/convection coefficient and hr is the radiative coefficient. The latter is computed using the following expression:

hr ¼

hr0 1 2  pffiffiffiffiffiffiffiffiffiffiffiffiffi þ  2 þ e1 e2 2 2 1

1þ

1þd =b d=b

ð2Þ



where d is the thickness of the airspace (m); b is the width of the airspace (m); e1 and e2 are the values of the hemispherical emissivity of the surfaces on the warm and cold surfaces of the airspace (with e1 = e2 = 0.9); ha and hr0 are calculated by section B.2 of ISO 6946 (2007) [30], being hr0 evaluated at 10 °C. Fig. 5 illustrates the brick unit model and the different airspace geometries (Ai). The obtained results for these brick unit airspaces, assuming horizontal heat flow, are presented in Table 5.

Table 4 Implemented numerical models.

a

These equivalent thermal conductivity values were input in the 3D model of an original clay brick and the obtained thermal transmittance value was 0.864 W/(m2 K). Comparing the results of both approaches, the difference between these is quite small (4.1%). Given this reduced difference and taking into account the substantially less computing time of the solid equivalent air method, this simplified approach was adopted in the present study. Furthermore, assuming the thermal conductivity for the ecological ceramic material (see Table 1), this 3D model also allows to obtain the U-value for this brick type: 0.788 W/(m2 K). These values imply a U-value decrease of 8.8%, when replacing the usual clay material of the brick units by the new ecological clay containing nano-crystalline aluminium residual sludge.

Model

2D/3D

Element

Remarks

A B C D E F G H

3D

Brick unit

Fluid airspaces Solid equivalent airspaces Homogeneous equivalent brick Typical cross-section Complete test-specimen Solid equivalent airspaces Typical cross-section ‘Complete’ test-specimena

Masonry wall 2D

Vertical cross-section.

Brick unit Masonry wall

4.1.3. 3D Brick Model C: homogeneous equivalent brick In the previous approach, the brick perforations were modelled assuming solid equivalent thermal conductivities of the air voids as explained before. In order to reduce more the time consumption to compute an entire wall using all the detailed airspaces, it was decided to make an additional simplification in the model, by using an equivalent brick unit thermal conductivity. This way, it was assumed a homogeneous brick unit, i.e., with no perforations as illustrated in Fig. 6. Making use of the original brick U-value early obtained (0.864 W/(m2 K)) and taking into account the thickness of the brick unit (240 mm), the obtained equivalent thermal conductivity was 0.241 W/(m K). Using a similar approach for the ecological brick, it was obtained an equivalent thermal conductivity of 0.217 W/(m K), i.e., there is a decrease of about 10% in the equivalent thermal conductivity of this brick unit.

4.1.4. 3D Wall Model D: typical cross-section In order to assess the thermal performance of the wall, a typical cross section including a horizontal mortar joint and a vertical hollow brick joint was modelled as illustrated in Fig. 7. Since to calculate an entire wall using all the detailed air voids could be very time-consuming, it was decided to verify the accuracy of a simpler model, assuming a homogeneous brick unit with an equivalent thermal conductivity, i.e., without perforations, as explained before for the 3D Brick Model C. To evaluate the differences in the results for a typical cross-section of a wall, two different wall models were compared: (1) solid equivalent brick units (Fig. 7a); and (2) homogeneous equivalent brick units (Fig. 7b). It should be noticed that in

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

(a) Geometry view

231

(b) Finite elements mesh

Fig. 4. Brick unit FEM model with fluid airspaces: 3D Brick Model A.

Fig. 5. Brick unit FEM Model B and solid equivalent airspace references.

all of the previous models (brick models) the mortar joints were not considered. A similar approach was performed for the wall model with mortar finishing applied in both surfaces, as illustrated in Fig. 8. Table 6 displays the U-values obtained by the previously presented numerical models. It is clear that the difference in results, between exact and equivalent brick wall model, is much reduced, presenting a maximum value of 2.5%. Having as reference the most accurate values (Brick Model B), it can be stated that the ecological brick unit allows to reduce the U-value of the typical wall crosssection in 6.8% and 6.4%, respectively without and with the mortar finishing layer. Furthermore, applying a mortar finishing layer in the wall leads to a thermal transmittance reduction of 4.4% and 3.9%, respectively for the original and ecological brick units. The last column of Table 6 shows the influence of the horizontal mortar layer between brick units in the thermal performance of the wall. The U-value increased 17.6% and 20.6%, respectively for the ‘original’ and the ‘ecological’ brick units.

4.1.5. 3D Wall Model E: complete test-specimen To evaluate the influence of the joint mortar applied in the experimental test setup perimeter, a 3D model containing an entire wall test-specimen (1.48  1.24 m2) including the abovementioned perimeter mortar was built (Fig. 9). Two wall execution stages were analysed: one without mortar finishing (Fig. 9a), and another with mortar finishing (Fig. 9b). In both models the equivalent homogenous brick unit simplification (Model C) was used. The obtained U-values are presented in Table 7. Comparing these values with the previous ones (Table 6: Model D – typical wall cross-section), a major increase in the computed thermal transmittance values (+23.4% in average) is observed. This is due to the increased thermal transmission originated by the mortar at the wall perimeter. This is not only related with the high thermal conductivity of this mortar layer, but also due to two more additional factors illustrated in Fig. 3, namely: (1) the high thickness of this mortar layer, mainly in the top of the wall; and (2) the reduced area of this wall test specimen, leading to a higher importance of this flanking thermal transmission. Moreover, the reduction of the U-value when using the ecological brick units instead of original bricks is now slightly lower, when compared with Model D (typical wall cross-section): 4.8% and 2.8%, before and after the finishing mortar layers. The use of these two mortar finishing layers allows to reduce the wall thermal transmittance in 6.0% (original clay bricks) and 4.0% (ecological brick units).

Table 5 Thermal resistances (R) and solid equivalent thermal conductivities (keq) of clay brick airspaces. Airspace reference

Thickness [mm]

R [m2 K/W]

keq [W/(m K)]

A1 A2 A3 A4 A5 A6

9.50 12.90 22.20 14.50 32.80 2.00

0.156 0.178 0.240 0.206 0.225 0.062

0.061 0.072 0.092 0.071 0.146 0.032

Fig. 6. Homogeneous equivalent brick unit FEM model: 3D Brick Model C.

232

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

(a) Solid equivalent airspaces (Model B)

(b) Homogeneous equivalent bricks (Model C)

Fig. 7. Typical cross section of the wall: 3D Wall Model D without mortar finishing.

(a) Solid equivalent airspaces (Model B)

(b) Homogeneous equivalent bricks (Model C)

Fig. 8. Typical cross section of the wall: 3D Wall Model D with mortar finishing.

Table 6 Thermal transmittances: 3D Wall Model D – typical cross-section. Mortar finishing

Brick units

U-value [W/(m2 K)] a

a b c d

DUDB-DCc

DUD-Cd

b

Solid equivalent airspaces

Homogeneous equivalent brick

Without

Original Ecological

1.038 0.967 (6.8%)

1.016 0.950 (6.5%)

2.1% 1.8%

+17.6% +20.6%

With

Original Ecological

0.992 0.929 (6.4%)

0.967 0.906 (6.3%)

2.5% 2.5%

– –

Using Brick Model B. Using Brick Model C. Differences between two versions of Model D. Differences between Wall Model D with homogeneous equivalent brick and Brick Model C.

4.2. Two-dimensional approaches The two-dimensional (2D) approach has the advantage of being simpler and faster to perform. However, since it is not possible to model the three-dimensional effects, in certain cases the results can be significantly different from reality. Therefore, the main objective of the 2D simulations undertaken was to quantify the error of this approach applied to these single veneer masonry walls. The 2D numerical simulations referred to were performed using the THERM software [31]. This is based on the finite element method and it allows obtaining the temperature and heat flux distribution inside the element, as well as the thermal transmission value, among other important parameters. As previously illustrated in Table 4, three different 2D models were built and the corresponding results analysed: (1) a brick unit

assuming solid equivalent airspaces; (2) a typical vertical wall cross-section; and (3) a more complete wall model, taking into account the test-specimen vertical cross-section dimensions and the perimeter support conditions.

4.2.1. 2D Brick Model F: solid equivalent airspaces A 2D Brick Model of a single brick unit was built corresponding to a horizontal cross-section. In Fig. 10, the blue colour gradients represent the different airspaces geometries in the brick, previously presented in Table 5, assuming solid-equivalent thermal conductivities. As before, two brick materials were analysed: original and ecological bricks, leading respectively to the following U-values: 0.853 and 0.778 W/(m2 K), representing a difference of 1.3% between 2D and 3D models.

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

(a) Without finishing

233

(b) With mortar finishing

Fig. 9. Complete wall test-specimen: 3D Wall Model E.

Table 7 Thermal transmittances: 3D Wall Model E – complete test-specimen. Mortar finishing

Brick units

U-value [W/(m2 K)]

DUE-Da

DU b

Without

Original Ecological

1.272 1.211 (4.8%)

+22.5% +25.2%

– –

With

Original Ecological

1.196 1.163 (2.8%)

+20.6% +25.2%

6.0% 4.0%

a Difference between 3D Wall Model E (complete test-specimen) and D (typical cross-section). b Difference between 2D Wall Model E with and without mortar finishing.

4.2.2. 2D Wall Model G: typical cross-section Similarly to what has been considered in the 3D models, typical vertical cross-section were also built in the 2D models of the wall, as illustrated in Fig. 11. These models do not take into account the edge effects of the test specimen in the wall thermal performance. Four distinct situations were considered: two for the original brick and two for the ecological brick, in both cases one simulating the situation before applying the mortar finishing layer (Fig. 11a), and the other one afterwards (Fig. 11b). In these models, the equivalent thermal conductivity for the brick unit is used, obtained from the 2D Brick Model F (Fig. 10), since the vertical cross-section within the brick unit changes from point to point.

Results are presented in Table 8. Regarding the difference between original and ecological bricks, a maximum value of 7.2% was found in the U-value without mortar finishing. The major difference seen in Table 8 is related with the thermal bridge effect of the horizontal mortar joint, leading to an average U-value increase of 19.3%. As before, the influence of the mortar finishing layer is very reduced given the small thickness (2  18 mm) and its high thermal conductivity (0.8 W/(m K)). 4.2.3. 2D Wall Model H: ‘complete’ test-specimen For a more complete wall analysis, considering only the top and bottom edge effects, the 2D models illustrated in Fig. 12 were built. As in previous models, four situations were analysed to have into consideration the two types of bricks, as well as the mortar finishing layer of the wall. Table 9 displays the obtained results. The ‘ecological’ brick allows a reduction in the wall thermal transmission value of 5.1% and 5.5%, with and without the mortar finishing layer. Comparing these U-values with the previous ones (Table 8) there is an average increase of 16% in the U-value, given the top/bottom flanking increase of the heat transmission in the wall test-specimen. However, this increased percentage is not enough to provide reliable results (as expected), being significantly lower in comparison with the one obtained in 3D Model E (Table 7). In order to quantify this, Table 9 also presents the differences between the U-value provided by 2D Model H and 3D Model E (last column), allowing concluding that there is an average decrease of 8.1% in the bi-dimensional wall model thermal transmittance. These differences confirm that the mortar in the vertical perimeter joint also has a significant influence in the overall thermal performance of the test-specimen wall and therefore should not be neglected. 5. Results and discussion

Fig. 10. Brick unit with solid equivalent airspaces: 2D Brick Model F (horizontal cross-section).

An overview of the main results obtained from both experimental tests and numerical simulations of bricks and walls are displayed in Table 10. The 3D and 2D numerical models present similar values, being differences minor and explained by the three-dimensional effects that exist in the heat flow path. For this reason, and as expected, 3D models provide more precise results, since these three-dimensional effects are included in the models. This can be verified by comparing the error between numerical

234

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

Fig. 11. Typical (vertical) cross-section: 2D Wall Model G.

Table 8 Thermal transmittances: 2D Wall Model G – without edge effects. Mortar finishing

Brick units

U-value [W/(m2 K)]

DUG-Fa

DUb

Without

Original Ecological

1.006 0.938 (7.2%)

+17.9% +20.6%

– –

With

Original Ecological

0.973 0.921 (5.6%)

– –

3.3% 1.8%

a Difference between 2D Wall Model G (typical cross-section) and F (single brick unit). b Difference between 2D Wall Model G, with and without mortar finishing.

models with the difference between experimental results, being the average 3D Model E error significantly lower (3.9%) than the error obtained with the 2D Model H (11.7%). It should also

be noticed that all numerical models predict U-values lower than the measured ones. As previously mentioned, the numerical results are similar to the experimental ones, when the entire wall test-specimen is simulated (Table 10). However, and as expected, the results obtained with the typical cross-section models exhibit a higher average difference (22.1% and 23.8% for the 3D Model D and 2D Model G, respectively) in comparison with the measured values for the wall test-specimen. This is mainly due to the flanking thermal transmission originated by the mortar layer applied in the perimeter of the test specimen. Since its thermal conductivity is higher than that of the ceramic material, there is an increase in heat transfer along the perimeter and consequently a higher thermal transmittance is experimentally registered. It should be highlighted that this effect

Fig. 12. 2D Wall Model H: ‘complete’ test-specimen (vertical cross-section).

235

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238 Table 9 Thermal transmittances: 2D Wall Model H – with top/bottom edge effects. Mortar finishing

Brick units

U-value [W/(m2 K)]

DUH-Ga

DUH-Eb

Without

Original Ecological

1.166 1.102 (5.5%)

+15.9% +17.5%

8.3% 9.0%

With

Original Ecological

1.120 1.063 (5.1%)

+15.1% +15.4%

6.4% 8.6%

a

Difference between 2D wall models with and without top/bottom edge effects. Difference between 2D Wall Model H (vertical test-specimen cross-section) and 3D Wall Model E (complete test-specimen). b

is even higher in the adopted experimental setup given to the reduced dimensions of the wall test-specimen (1.48 m  1.24 m) and also due to the large amount of mortar used in the edges of the latter, mainly in the top. With the presented combined experimental/numerical approach and given the reduced differences between measurements and the corresponding 3D wall model results,U-values presented in Table 10 are assumed to be reliable. Therefore, it can be concluded that, being

the difference in theU-value (3D Model B) between the original brick and the ecological one of 0.076 W/(m2 K), there is a reduction of 8.8% in the U-value of the latter. Naturally, this is a very good result since it means that it is possible to incorporate an industrial waste (the nano-crystalline aluminium sludge) into the raw material of the clay bricks reaching in addition a significant improvement of the thermal performance of the latter. The surface temperature distributions in both original and ecological brick models are plotted in Fig. 13a and b. It can be observed that these do not differ significantly, although presenting the ecological brick a slightly better thermal performance. In order to illustrate the differences related with the thermal transmittance values obtained, the wall external surface heat flux retrieved from the 3D cross-section models is shown in Fig. 14. As expected, given its higher thermal conductivity, the horizontal mortal joint between brick units exhibit higher heat flux values, leading to an increase of 20.1% and 22.7% in the U-value of the ‘original’ (Fig. 14a) and ‘ecological’ (Fig. 14b) brick unit, respectively. Furthermore, given its lower thermal conductivity, the ‘ecological brick’ wall (Fig. 14b) exhibits slightly lower heat flux

Table 10 Overall thermal transmittance results. Description

Brick units

Brick unit Typical wall cross-section

Without finishing

With mortar finishing

Complete wall test-specimen

Without finishing

With mortar finishing

a

U-value [W/(m2 K)] Measured

3D FEM (DU)a

Original Ecological

– –

Model B

0.864 0.788

Model F

0.853 0.778

Original

–

Model D

Model G

Ecological

–

Original

–

Ecological

–

1.038 (20.8%) 0.967 (25.0%) 0.992 (20.0%) 0.929 (22.6%)

1.006 (23.2%) 0.938 (27.3%) 0.973 (21.5%) 0.921 (23.3%)

Original

1.31

Model E

Model H

Ecological

1.29

Original

1.24

Ecological

1.20

1.272 (2.9%) 1.211 (6.1%) 1.196 (3.5%) 1.163 (3.1%)

1.166 (11.0%) 1.102 (14.6%) 1.120 (9.7%) 1.063 (11.4%)

2D FEM (DU)a

Difference between experimental and numerical results; FEM – finite element method.

U=0.864 W/(m2.K)

(a) Original brick

U=0.788 W/(m2.K)

(b) Ecological brick

Fig. 13. Surface temperatures: 3D Brick Model B – solid equivalent airspaces.

236

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

U=1.038 W/(m2.K)

U=0.967 W/(m2.K)

(a) Original brick without finishing

(b) Ecological brick without finishing

U=0.992 W/(m2.K)

U=0.929 W/(m2.K)

(c) Original brick with mortar finishing

(d) Ecological brick with mortar finishing

Fig. 14. Heat flux on the wall external (cold) surface: 3D Wall Model D (typical cross-section).

values when compared with the ‘original brick’ (Fig. 14a), being the differences higher in the brick unit surfaces and in the vertical hollow joints between bricks. Fig. 14c and d show a high reduction in the heat flux with the use of a finishing mortar layer on the vertical surfaces, which can be further improved, for instance by using a lower thermal conductivity mortar. This is related with the heat dissipation provided by the continuous finishing mortar layer, leading to a more uniform heat flux values in this vertical wall surface (Fig. 14c and d). The 3D model of the complete test-specimen (Model E) shows that the mortar joint between brick units and peripheral mortar exhibits even higher heat flux values (Fig. 15), when compared with the typical cross-section of Model D (Fig. 14), leading to an increase of 27.1% and 31.0% in the U-value (without finishing), using ‘original’ and ‘ecological’ brick units, respectively. This demonstrates that, besides the horizontal mortar joints between brick layers, the peripheral mortar joint around the wall testspecimen also has a major influence in the thermal performance. The use of finishing mortar layers on both vertical surfaces (Fig. 15c and d), allows reducing the heat flux and consequently the U-value (average reduction of 5.0%). This surface heat flux reduction is more visible in the vicinity of the horizontal mortar joints between brick layers. The improvement in terms of U-value of the ‘ecological’ brick, both as a single unit (Model B), as a typical cross-section of a wall (Model D) and as a complete wall test-specimen (Model E), with and without mortar finishing, is graphically displayed in Fig. 16. It should be noticed that these results depend on the testspecimen dimensions (1.240  1.480 m2), which are rather small compared to the usual façade dimensions, which therefore leads to U values higher than those that would be registered in reality.

The overall thermal transmittance of the perforated brick wall can be useful for the whole building energy simulation tools (like e.g. EnergyPlus) used in the design stage of the building construction projects. Therefore, additional models were considered, similar to 3D Model E, but with increased dimensions (4.00 m wide by 2.45 m tall). The obtained results for this real sized building wall made with ‘original’ and ‘ecological’ brick units are also plotted in Fig. 16b. As expected, the obtained U-values (1.07 and 1.01 W/m2 K, respectively) are between the results for a typical wall crosssection and for a complete wall test-specimen. The thermal transmittance reduction, due to the better thermal performance of the new ‘ecological’ brick units, is well illustrated herein. Without the mortar finishing layer (Fig. 16a), this U-value reduction changes to 4.8% and 8.8% for the complete wall testspecimen and for a single brick unit, respectively. With the mortar finishing layer the new ‘ecological’ brick units allow a thermal transmittance decrease of 2.8%, 5.6% and 6.4% respectively for the complete wall test-specimen, the real size building wall (4.00  2.45 m2) and the typical wall cross-section.

6. Conclusions Aiming at contributing to the sustainability of the construction sector, a research project was conducted in co-promotion with a clay brick producer to enhance the thermal performance of hollow clay brick walls by adding an industrial by-product, a nanocrystalline aluminium sludge, to the raw material. In the research study herein described a combined experimental/numerical approach was adopted to assess the U-value of both the original and the new ecological bricks, and a good correlation between

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238

U=1.272 W/(m2.K)

(a) Original brick without finishing U=1.196 W/(m2.K)

(c) Original brick with mortar finishing

237

U=1.211 W/(m2.K)

(b) Ecological brick without finishing U=1.163 W/(m2.K)

(d) Ecological brick with mortar finishing

Fig. 15. Heat flux on the wall external (cold) surface: 3D Wall Model E (complete test-specimen).

(a) Without mortar finishing

(b) With mortar finishing

Fig. 16. Thermal transmittance values computed for the original and the ecological brick unit (3D Brick Model B), typical cross-section (3D Wall Model D), complete testspecimen (3D Wall Model E) and real size building wall (4.00  2.45 m2).

numerical and experimental results was observed. As general conclusion, it can be stated that the ‘ecological brick’ shows a better thermal performance than the ‘original brick’, with 8.8% reduction in the U-value. The following specific conclusions are also drawn:
Vertical voided joints between bricks do not significantly affect the thermal performance of the wall;
Contrarily, the horizontal mortar joints between bricks are the dominant parameter in this scope;


Since horizontal mortar joints significantly increase the overall transmission losses, it is important to provide specialized workmanship to reach the optimal thermal behaviour;
Mortar joints are also responsible for significant thermal bridges, even with an air–gap included, being therefore advised to use low thermal conductivity mortars;
Flanking thermal transmission due to the wall perimeter mortar joint is another important factor that needs to be taken in account, due to its high influence in the overall thermal performance of the wall;

238

P. Santos et al. / Construction and Building Materials 101 (2015) 227–238


Lastly, it should be highlighted that the use of industrial nanocrystalline aluminium sludge to enhance the thermal performance of perforated clay brick walls is an excellent measure, not only because a significant reduction in the U-value is reached, but also because the (high) disposal costs of the industrial waste can be eliminated. Acknowledgements The authors would like to thank the European Union funding of the research project ‘QREN 13312 SeTiverNano – Production of bricks thermal and structurally more efficient by incorporating nanomaterials’. References [1] L. Pérez-Lombard, J. Ortiz, C. Pout, A review on buildings energy consumption information, Energy Build. 40 (3) (2008) 394–398. [2] European Directive 2010/31/EU, Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings (EPBD), Official Journal of the European Union, Brussels, Belgium, 2010. [3] SCE, Portuguese buildings energy certification system (SCE), Regulation for energy performance of residential buildings (REH) and Regulation for energy performance of commercial and service buildings (RECS), Decree-Law no. 118/ 2013, 20th August, 2013 (in Portuguese). [4] DB-HE1, Spanish Building Technical Code, Basic Document HE 1 – limitation of energy demand. Order FOM/1635/2013, 10th September (Documento Básico DB-HE «Ahorro de Energía» in Spanish), 2013. [5] Part L, Irish Building Regulations, Technical Guidance Document L – Conservation of fuel and energy – Dwellings, ISBN: 978-1-4064-2594-9, 2011. [6] A. de Gracia, A. Castell, M. Medrano, L.F. Cabeza, Dynamic thermal performance of alveolar brick construction system, Energy Convers. Manage. 52 (7) (2011) 2495–2500. [7] L.M. Al-Hadhrami, A. Ahmad, Assessment of thermal performance of different types of masonry bricks used in Saudi Arabia, Appl. Therm. Eng. 29 (2009) 1123–1130. [8] H. Gervásio, P. Santos, L. Simões da Silva, A.M.G. Lopes, Influence of thermal insulation on the energy balance for cold-formed buildings, Adv. Steel Constr. 6 (2) (2010) 742–766. [9] F. Aldawi, F. Alam, A. Date, M. Alghamdi, F. Aldhawi, A new house wall system for residential buildings, Energy Build. 67 (2013) 403–418. [10] S.A. Al-Sanea, M.F. Zedan, Improving thermal performance of building walls by optimizing insulation layer distribution and thickness for same thermal mass, Appl. Energy 88 (9) (2011) 3113–3124. [11] F. Stazi, F. Tomassoni, A. Vegliò, C. Di Perna, Experimental evaluation of ventilated walls with an external clay cladding, Renew. Energy 36 (12) (2011) 3373–3385. [12] A. Castell, I. Martorell, M. Medrano, G. Pérez, L.F. Cabeza, Experimental study of using PCM in brick constructive solutions for passive cooling, Energy Build. 42 (4) (2010) 534–540.

[13] C. Zhang, Y. Chen, L. Wu, M. Shi, Thermal response of brick wall filled with phase change materials (PCM) under fluctuating outdoor temperatures, Energy Build. 43 (12) (2011) 3514–3520. [14] P. Principi, R. Fioretti, Thermal analysis of the application of pcm and low emissivity coating in hollow bricks, Energy Build. 51 (2012) 131–142. [15] S.A. Al-Sanea, M.F. Zedan, Effect of thermal bridges on transmission loads and thermal resistance of building walls under dynamic conditions, Appl. Energy 98 (2012) 584–593. [16] P. Santos, L. Simões da Silva, V. Ungureanu, Energy efficiency of light-weight steel-framed buildings, 1st ed. European Convention for Constructional Steelwork (ECCS), Technical Committee 14 – Sustainability & Eco-Efficiency of Steel Construction, ISBN 978-92-9147-105-8, no. 129, 1st ed., 2012. [17] M.P. Morales, M.C. Juárez, P. Munoz, J.A. Gómez, Study of the geometry of avoided clay brick using non-rectangular perforations to optimise its thermal properties, Energy Build. 43 (2011) 2494–2498. [18] S. Raut, R. Ralegaonkar, S. Mandavgane, Utilization of recycle paper mill residue and rice husk ash in production of light weight bricks, Arch. Civ. Mech. Eng. 13 (2) (2013) 269–275. [19] A.A. Shakir, S. Naganathan, K.N. Mustapha, Properties of bricks made using fly ash, quarry dust and billet scale, Constr. Build. Mater. 41 (2013) 131–138. [20] ISO 10456, Building materials and products – hygrothermal properties – tabulated design values and procedures for determining declared and design thermal values, international organization for standardization, Geneva, Switzerland, 2007. [21] EN 12524, Building Materials and Products – Hygrothermal properties – tabulated design values, European Committee for Standardization, Brussels, 2000. [22] DIN 4108-4, Thermal Insulation and energy economy in buildings – hygrothermal design values, German Institute for Standardization, Berlin, 2013. [23] ISO 22007-2, Plastics. Determination of thermal conductivity and thermal diffusivity – Part 2: transient plane heat source (hot disc) method, International Organization for Standardization, Geneva, Switzerland, 2012. [24] EN 772-13, Methods of test for masonry units – Part 13: determination of net and gross dry density of masonry units (except for natural stone), European Committee for Standardization, Brussels, 2000. [25] EN 772-7, Methods of test for masonry units – Part 7: determination of water absorption of clay masonry damp proof course units by boiling in water, European Committee for Standardization, Brussels, 1998. [26] EN 772-3, Methods of test for masonry units – Part 3: determination of net volume and percentage of voids of clay masonry units by hydrostatic weighing, European Committee for Standardization, Brussels, 1998. [27] ISO 8990, Thermal insulation – determination of steadystate thermal transmission properties – calibrated and guarded hot box, International Organization for Standardization, Geneva, Switzerland, 1996. [28] ASTM C1363, Standard test method for the thermal performance of building assemblies by means of a hot box apparatus, American Society for Testing and Materials, West Conshohocken, USA, 2011. [29] ANSYS CFX, ‘‘ANSYS, Inc.” Canonsburg, USA, 2013. [30] ISO 6946, Building components and building elements – thermal resistance and thermal transmittance – Calculation method, International Organization for Standardization, Geneva, Switzerland, 2007. [31] THERM 7.3, ‘‘Lawrence Berkeley National Laboratory, USA. Finite Element Simulator”, 2013.