Heat Mass Transfer (2017) 53:3487–3499

DOI 10.1007/s00231-017-2059-7

ORIGINAL

Modeling thermal performance of exterior walls retrofitted

from insulation and modified laterite based bricks materials
Elvis Wati 1 & Pierre Meukam 2 & Jean Claude Damfeu 2

Received: 11 April 2016 / Accepted: 11 May 2017 / Published online: 15 June 2017
# Springer-Verlag Berlin Heidelberg 2017

Abstract Uninsulated concrete block walls commonly found layer at the interior surface. For intermittent cooling space, the
in tropical region have to be retrofitted to save energy. The best wall configuration from the minimum energy consumption
thickness of insulation layer used can be reduced with the help depends on total insulation thickness. For the total insulation
of modified laterite based bricks layer (with the considerably thickness less than 8 cm approximately, the best wall configura-
lower thermal conductivity than that of concrete block layer) tion is placing the half layer of insulation material at the interior
during the retrofit building fabrics. The aim of this study is to surface and the other half between the two different massive
determine the optimum location and distribution of different layers with the modified earthen material at the exterior surface.
materials. The investigation is carried out under steady peri- Results also showed that, the optimum insulation thickness cal-
odic conditions under the climatic conditions of Garoua in culated from the yearly cooling transmission (estimated only
Cameroon using a Simulink model constructed from H- during the occupied period) and some economic considerations
Tools (the library of Simulink models). Results showed that slightly depends on the location of that insulation.
for the continuous air-conditioned space, the best wall config-
uration from the maximum time lag, minimum decrement Keywords Multi-layered wall . Optimum configuration .
factor and peak cooling transmission load perspective, is di- Insulation thickness impact . Optimum insulation thickness
viding the insulation layer into two layers and placing one at
the exterior surface and the other layer between the two dif-
ferent massive layers with the modified laterite based bricks Nomenclature
As Annual energy savings ($ . m−2)
C Specific heat (J . kg−1 . K−1)
Highlights C Cost ($)
• Thermal retrofit solutions of walls using laterite and insulation layers
were studied.
COP Coefficient of performance of air-conditioning system
• Optimum configuration of fixed resistance and capacitance walls was G Inflation rate (%)
determined. h Convection and/or radiation heat transfer coefficient
• Comparative performance of some configurations depends on insulation (W . m−2 . K−1)
thickness.
• Insulation thicknesses of walls were optimized for intermittent air-
L Wall thickness (m)
conditioned space. Lop Optimum insulation thickness (m)
I Interest rate (%), order of node
* Pierre Meukam I Total solar radiations on the horizontal surface (W. m−2)
pierre_meukam@[Link] Ib Direct solar radiations on the horizontal surface
(W.m−2)
1
Environmental Energy Technologies Laboratory, University of
Id Diffuse solar radiations on the horizontal surface
Yaoundé I, Yaounde, Cameroon (W . m−2)
2
Laboratory L3E:Energy, Water and Environment, National
N Number of nodes
Advanced School of Engineering, University of Yaoundé I, Po Box n Lifetime of building (years)
8390, Yaounde, Cameroon M Number of layers of composite wall
3488 Heat Mass Transfer (2017) 53:3487–3499

pb Payback period (years) that will not require large amounts of energy in the construc-
Qc Annual cooling transmission load (MJ . m−2) tion or deconstruction processes [1].
Rd Diffuse radiation conversion factor Many authors studied the embodied energy or the life cycle
T Time (s) assessment of many building materials. Results of some of
T Temperature (°C) these studies show that earthen materials have less environ-
x Coordinate direction normal to wall (m) ment impact. Indeed, Morel et al. [2] compared the embodied
Greek symbols energy of a typical house made from stone masonry, rammed
Α Solar absorptivity of outside surface of wall soil and concrete. In their study, only stone and soil are locally
β Tilted surface angle (°) available. Results show that the concrete house consumes
δ Declination angle (°) 146% more energy than the house made with stone masonry,
Λ Thermal conductivity (Wm−1 K−1) and 240% more energy required for the rammed soil house.
ϕ Latitude (°) The transport impact on the environment is significantly less
ω Hour angle (°) for the stone masonry building (480% lower) and the rammed
ωs Sunset-hour angle for a horizontal surface (°) soil building (640%). Bribián et al. [3] studied the life cycle
ρ Density of material (kg.m−3) assessment of some building materials considering as impact
ρr Ground reflectivity categories the primary energy demand according to the cumu-
Subscripts lative energy demand method, the global warming potential
El Electricity according to the IPPC (Intergovernmental Panel on Climate
Enr Energy Change) 2007 methodology and water demand (in liters).
I Inside Results show that the use of light clay bricks (85% clay and
in Interior 15% straw) clearly reduces the impact. Indeed, 45% of the
Ins Insulation primary energy demand which is relatively high comes from
o Outside biomass, due to the straw content. In addition, light clay bricks
Sa Solar-air have a practically neutral CO2 balance. The energy used in a
t Total rammed earth office building (two storeys) in a hot and dry
climate (Wodonga, Australia) were studied by Taylor et al. [4].
Results show that, rammed earth building used more energy
for heating than another office building in the same location.
1 Introduction In spite of the fact that earth block is an environmental
friendly material as descript above, its lower insulation level
With the increase in worldwide energy consumption, due to leads to important energy consumption in conditioned build-
the population growth, urbanization, industrialization, im- ing. So, many authors have mixed others material with earth
provement in standard of living, deterioration of the thermal material in order to improve its thermal and/or mechanical
quality of recent constructions etc., which usually increases properties. Mixing the earth with straw is traditionally done
the greenhouse gas emissions, it has become necessary to take in the cob and the wattle and daub techniques [5] and signif-
immediate action to avoid dangerous consequences for future icantly reduces the thermal conductivity [6]. The effect of the
generations. One can understand why, at the World Summit on addition of plastic fibres, straw or polystyrene fabric in the
Sustainable Development in Johannesburg in 2002, the mud brick stabilised with cement, basaltic pumice or gypsum
world’s nations reaffirmed their commitment to fostering sus- on the thermal properties was investigated by Binici et al. [7].
tainable development. In order to ensure this in the building Results show that the thermal conductivities of bricks made of
sector, green construction as well as passive techniques for basaltic pumice are lower than that made with other stabi-
reduction of building energy consumption should be promot- lizers. Meukam et al. [8] determined the effect of the addition
ed. A passive system is an assembly of natural and architec- of some local materials in cement stabilised lateritic soil brick
tural components, such as shading, orientation, thermal mass, on the thermal properties. Results show that the addition of
insulation and optimum location and distribution of material natural pozzolan (45%) or sawdust (10%) to the laterite based
in multi-layered wall of the building which decrease the total brick leads to maintain their strength while their thermal con-
energy consumption. ductivity significantly decrease (from 0.87 W/mK to 0.50 W/
Buildings are major consumers of energy throughout their mK and 0.40 W/mK for pozzolan and sawdust, respectively).
life cycle. Sustainable building may create healthy indoor en- Bal et al. [9] showed that adding millet waste to the laterite
vironment pollutants, feature landscaping that reduces water based brick may strongly decrease their thermal conductivity
usage and/or incorporate sustainable materials in their con- (from 1.4 for dry pure laterite blocks to 0.29 W/mK for dry
struction. Sustainable building materials are those that will laterite blocks with 0.122 kgmi/kg mass content of millet
provide a thermally efficient wall, while at the same time walls waste). The thermal conductivities of the above modified
Heat Mass Transfer (2017) 53:3487–3499 3489

laterite based bricks are in the same order as that of stuff stone BiAC^ space under steady periodic conditions using the
(0.22–0.40 W/mK) used in ref. [10] as exterior insulation of climatic data of Garoua, Cameroon. The compressed
concrete block wall. stabilised laterite block mixing with sawdust and pozzolan
In Cameroon such as in many other sub-Saharan countries, are taken as the modified laterite based bricks while the
uninsulated concrete block walls (presenting a very low ther- expanded polystyrene is taken as insulation material.
mal resistance) as external wall of buildings are found. These
walls have to be retrofitted to save energy. Since many insu-
lation materials are expensive and/or have high environmental 2 Methodology
impact, theirs thicknesses should be reduced with the help of
environmental friendly and low cost materials such as the 2.1 Mathematical formulation
modified laterite based bricks during the retrofit building fab-
rics. The optimum location and distribution of insulation and The wall facing south of the room model used in this
modified laterite based brick layers around the existing wall study is exposed to the outdoor climatic conditions while
should be determined and constitutes one of the aims of the the other walls, ceiling and roof are adiabatic since the
present study. adjacent rooms are assumed to be in the same thermal
The optimum location and distribution of insulation conditions as the studied room (Fig. 1a). The indoor
layers on buildings’ walls consisting of masonry/ space is continuously air-conditioned (cAC) or intermit-
insulation structure was investigated in many studies ac- tently air-conditioned (iAC). Thus, heat flows through
cording to different operating modes of the heating/ the south-facing wall. Since the wall thickness is gener-
cooling system. For continuous air-conditioned space, ally small according to its width and height, it is as-
the optimum location is which ensures maximum time sumed that the heat flows in only one direction. So, the
lag and minimum decrement factor. Some studies showed temperature T(x, t) at x coordinate point of the wall at
that placing the insulation layer at the outer wall surface time t (Fig. 1b) verifies Eq. (1).
provides better thermal performance than placing it at the
inner surface [11–13]. Other studies found that placing ∂T ðx; t Þ ∂2 T ðx; t Þ
c jρ j ¼ λj ð1Þ
insulation layers both at the inside and outside wall sur- ∂t ∂x2
faces is the best [14, 15]. Fewer studies have dealt with
where the subscript j denotes the layer j of the studied
optimizing the layer composition of exterior walls in in-
multilayer wall and λj, cj and ρj are the thermal conduc-
termittently air-conditioned spaces. Results showed that
tivity, the specific heat and the density of the same layer,
positioning the insulation at the inner wall surface is the
respectively.
best [13, 16, 17]. For a climate in which the sol-air tem-
To solve Eq. (1), an initial condition and two bound-
perature crosses the thermal comfort zone, Barrios et al.
ary conditions need to be specified. The initial tempera-
[18], showed that, locating the insulation layer at the ex-
ture is taken arbitrary since the system is run until the
terior wall surface presents better performance. As it can
steady state reached. The limit condition at the inside
be seen in the literature survey, there are no information
surface is shown in Eq. (2).
on the case where the wall, in addition of insulation layer


includes two masonry layers presenting considerably dif- ∂T
−λM ¼ hc ðT x¼L −T in Þ þ hr ðT x¼L −T s Þ ð2Þ
ferent thermal conductivities. ∂x x¼L
A study conducted by Al-Sanea and Zedan [14] on build-
ing envelop research aims to simultaneously determine the where λM is the thermal conductivity of the composite
optimum thicknesses and positions of insulation layers for wall layer facing the indoor conditions, hc and hr are the
continuously air-conditioned space (cAC). The studied walls convection and radiation heat transfer coefficient at the
have the same thermal mass and R-value. The results showed inside surface, respectively, L is the thickness of the
that the optimum thickness(es) of insulation layer(s) was in- composite wall, Tin is the indoor air temperature and T s
dependent on their location. The same result was obtained by is the average value o (T s ¼ ∑ Ai T i = ∑ Ai ). For the con-
i i
Ozel [19]. Similar study deserves more attention for intermit-
tinuously air-conditioned space (cAC) case, T s is as-
tent air conditioned space (iAC).
sumed to be equal to Tin. Thus, Eq. (2) becomes:
The objectives of this work are to determine (a) the


optimum location and distribution of insulation and mod- ∂T
ified laterite based bricks layers around the concrete block −λM ¼ hi ðT x¼L −T in Þ ð3Þ
∂x x¼L
wall as energy retrofit solution and (b) the influence of
location of insulation layer(s) on its total optimum thick- where hi is the combined convection and radiation heat
ness. The investigation is carried out for BcAC^ and transfer coefficient at the inside surface.
3490 Heat Mass Transfer (2017) 53:3487–3499

(a) (b)
N
W Room air at
E
design
S
temperature

Fig. 1 Plan of room zone (a) and M-layered composite wall with boundary conditions and grid arrangement (b)

At the outside surface, the limit condition is given by 2.2 Assessment parameters
Eq. (4).

 For the case of continuously air-conditioned space, the
∂T
−λ1 ¼ ho ðT 0 −T x¼0 Þ þ αI t ð4Þ assessment parameters are the yearly average of time lag
∂x x1 ¼0 (∅) and decrement factor (f) and the peak of the cooling
transmission load during the representative day of
where λ1 is the thermal conductivity of the outer layer of com-
January, one of the hottest months with the large diurnal
posite wall; ho is the combined convection and radiation heat
temperature variation. Time lag (∅) and decrement factor
transfer coefficient at the outside wall; To is the ambient air
(f) are evaluated as following:
temperature; α is the solar absorptivity and It is the total solar
radiation incident on the tilted surface which can be expressed as:
8
< t T x¼L ðmaxÞ −t T x¼0 ðmaxÞ if t T x¼L ðmaxÞ > t T x¼0 ðmaxÞ
cosθ 1−cosβ ϕ¼ t T x¼L ðmaxÞ −t T x¼0 ðmaxÞ þ 24 if t T x¼L ðmaxÞ < t T x¼0 ðmaxÞ
It ¼ I b þ Rd I d þ I:ρr ð5Þ :
cosθz 2 24 if t T x¼L ðmaxÞ ¼ t T x¼0 ðmaxÞ
ð9Þ
where Ib, Id and I are direct, diffuse and global solar radiations
reaching the horizontal surface respectively; ρr is the ground T x¼L ðmaxÞ−T x¼L ðminÞ
f ¼ ð10Þ
reflectivity assumed equal to 0.2 in this work; θ is the incident T x¼0 ðmaxÞ−T x¼0 ðminÞ
angle; θz is the zenith angle; β is the tilted surface angle with
respect to the horizontal and Rd is the diffuse radiation conversion For the intermittent air-conditioned space case, several as-
factor expressed by Perez et al. [20]. For a vertical wall (β = 90°) sumptions are made:
facing south,
& The capacity of the cooling system is sufficiently higher to
cosθ ¼ cosδsinϕcosω−sinδcosϕ ð6Þ immediately meet the load.
cosθz ¼ cosϕcosδcosω þ sinϕsinδ ð7Þ & the room air reaches its set-point temperature in a very
short time
where δ, ϕ and ω are the declination, latitude and hour angle,
respectively. The assessment parameters in the intermittent air-
The indoor air temperature is set to a constant value when conditioned space case are the cooling transmission load (q)
the air-conditioning system is ON. When it is OFF, the indoor during the occupied period (when the air-conditioning system
air temperature Tin verifies Eq. (8). is ON). q is calculated using Eq. (11).

dT in
ρ a cp a V ¼ ∑ hc ðT si −T in Þ ð8Þ
dt i q¼ ∑ ½hc ðT x¼L −T i Þ þ hr ðT x¼L −T s Þ ð11Þ
19h→6h

where Tsi is the temperature of the ith room surface; V is the

volume of the room; ρa and cpa are the density and specific It is assumed in this work that, during each month,
heat capacity of air. The values of some parameters used in the climatic variables on the outside surface are period-
this section are given in Table 1 and Table 2. ic. Thus, the daily cycle of the solar-air temperature,
Heat Mass Transfer (2017) 53:3487–3499 3491

Table 1 Materials properties

[6, 8, 21, 22] Material Dry density kg/m3 Thermal conductivity W/m . K Specific heat J/kg K

EB 1942 0.897 1000

ES 1200 0.417 878
EP 1716 0.503 943
HCB 1250 0.67 880
C 2000 1.70 1000
Cement mortar 1950 1.15 880
RCB 2300 2.30 1000
Polystyrene 24 0.034 1400
VPB 837 0.125 960

representing the monthly average, is repeated on consec- function method is not suitable for heavy walls having
utive days during a month. Before evaluating any above high thermal mass when using the time step of an hour
parameters, the system is run until the steady state is [6], the finite difference method is used in this work. The
reached. simulation results of EnergyPlus have been validated
through analytical, comparative, and empirical tests in
2.3 Calculation method and validation of the Simulink ref. [30].
model results Figure 2a and b present the variations of the indoor air
temperature and the inside surface temperature of the
To solve the above non-linear problem, a Simulink model south-facing wall during the representative day of
is constructed from H-Tools. One can note that H-tools is January calculated with EnergyPlus and Simulink model
a library of models developed by Chalmers University of constructed from H-tools. All walls are constituted of
Technology (Sweden) in partnership with Technical 150 mm of concrete block while ceiling and floor are
University of Denmark. It is available for free made with 150 mm of reinforced concrete block. Walls,
downloading in ref. [27]. The Bmaterial^ notes blocks ceiling and floor are covered on both sides with 15 mm of
used to construct the wall model used in this study were coated mortar. Properties of used materials are presented
validated by comparing its results with those from Match in Table 1. The convection, radiation and combined heat
(commercial and Breliable^ HM program) in ref. [28]. transfer coefficients at the inside and outside wall surfaces
However, the results from the complete Simulink model used in this work are presented in Table 2. One can note
constructed in this study are compared to those obtained that, the results from EnergyPlus are in good agreement
by EnergyPlus [29] version 7.0.0. This validated software with those obtained with the Simulink model constructed
for building simulation was developed by the United from H-tools
States Department of Energy. As many other dynamic The above Simulink model is used to simulate tem-
software used to simulate a building thermal behavior, perature variations at some points of an experimental
EnergyPlus calculates thermal loads of zones by the heat setup used by Samri [22] (Fig. 3a). The results of simu-
balance method. The heat balance method takes into ac- lation (Fig. 3c) are compared to those measured experi-
count all heat balances on outdoor and indoor surfaces mentally by Samri [22] (Fig. 3b). It can be observed that,
and transient heat conduction through building fabric. the amplitude of the experimental temperature variations
Transfer function method or finite difference method can is slightly higher than that of the simulated temperature
be used by EnergyPlus for the calculation of the heat- variations. In the simulation study, the indoor tempera-
transfer through the building envelope. As the transfer ture was assumed constant (Tinit = 24.5 °C). In fact, it
changes with time in the experimental study. That can
explain the different observed. In spite of the fact that
Table 2 Convective, the moisture content is not taken into account, simulated
radiative, combined Parameters Values
temperatures at C and C2 points are in good agreement
convective and radiation
heat transfer coefficients ho 22 W/m2K [23, 24, 25] with those obtained experimentally.
used in this study hi 9 W/m2K [23, 24, 25] In the lack of hourly measured climatic variables for
hr 6 W/m2K [26] Garoua required by the above models, the same techniques
hc 3 W/m2K [26] used by Wati et al. [24] were employed to interpolate the
available climatic data.
3492 Heat Mass Transfer (2017) 53:3487–3499

(a) (b)

Fig. 2 Comparative results from EnergyPlus software and Simulink model built from H-tools for continuously air-conditioned space (a) and intermittent
air-conditioned space (b)

2.4 Description of studied walls and materials presentation and the compressed stabilised earth block with 45% of
pozzolan (EP). These blocks are the modern descendants
The thermal performances of two groups of twelve wall con- of moulded earth. More details on the ES and EP blocks
figurations are investigated in this study. In each group, all are given by Meukam et al. [8]. These materials present
wall configurations have the same thermal mass and R-value the following economic and environmental advantages:
(Fig. 4). They differ only by the location and distribution of
different materials. & laterite, pozzolan and sawdust (from local factory where
The earthen materials used in this work are the com- wood is being sawed) are locally available resources in
pressed stabilised earth block with 8% of sawdust (ES) most area in Cameroon such as in many other tropical

Vertical Perforated
Brick (VPB) wall
(a) Climatic chamber Laboratory
C
A =0
B D E
= 15
C2
= 21.6
= 30
= 30
(b)
(c)

Fig. 3 Experimental setup (a), experimental temperature variations obtained by Samri [22] (b) and simulated temperature variations obtained from
Simulink model used in this study (c)
Heat Mass Transfer (2017) 53:3487–3499 3493

C X PS X C PS PS C X

Outdoor
Outdoor
Outdoor

Indoor

Indoor

Indoor
15 15 Li 15 15 Li Li 15 15

C|X|PS X|C|PS PS |C|X

PS X C X PS C C PS X

Outdoor

Indoor
Outdoor

Indoor

Indoor
Li 15 15 15 Li 15 15 Li 15

PS|C|X X|PS|C C|PS|X

X PS C PS PS X PS C PS C PS X

Outdoor

Outdoor
Outdoor

Indoor
Indoor
Indoor
15 2 15 2 2 15 2 15 2 15 2 15

X|PS| C|PS PS|X|PS| C PS|C|PS| X

C PS X PS PS C X PS PS X C PS

Outdoor
Outdoor

Outdoor

Indoor
Indoor
Indoor

15 2 15 2 2 15 15 2 2 15 15 2

C|PS| X|PS PS|C|X|PS PS|X|C|PS

X=ES/EP; C=concrete block; ES=CSEB + sawdust; EP=CSEB + pozzolan; PS=expanded polystyrene

Fig. 4 Different wall configurations composed of concrete block, modified earth block and insulation layer(s) (Li = 4 cm)

countries, thus manufacturing and transport are low- where Cins is the cost of insulation material per unit
energy consuming and therefore, low carbon dioxide volume ($ . m−3), PWF is the present worth factor, Lins
emission, is the insulation thickness, Q c is the annual cooling
& a few cement quantity is used to stabilised these blocks so transmission need per unit area of external wall (kWh .
that its embodied energy does not significantly increase, m −2 ), COP is the coefficient of performance of air-
& laterite is a recyclable material. conditioning system and Cel is the cost of electricity
($ . kWh−1).
The thermal properties of materials used in this study are
presented in Table 1. ð1 þ rÞn −1
PWF ¼ ð13Þ
r ð r þ 1Þ n
2.5 Optimization model of the insulation’s thickness
where n is the lifetime of the building and r is a param-
As the thickness of the insulation layer within the exterior eter depending on the inflation (g) and the interest (i)
wall increases, the heat gain/loss through the building rates [31] given as:
envelope decreases, and thus, the energy consumption
cost of the building decreases. Nevertheless, this increase 8
> i−g
in insulation layer thickness increases the initial cost of < for i > g
r¼ 1 þg ; for i ¼ g; PWF ¼
n
ð14Þ
the structure. Therefore, an economic analysis should be >
: g−i 1þi
for i < g
performed in order to estimate the optimum insulation 1þ1
thickness which minimizes the total cost [23].
The total cost (Ct) is expressed as: The payback period (pb) can be obtained by solving the
following equation:

Qc C el C ins Lop
Ct ¼ PWF þ C ins Lins ð12Þ ¼ PWFðpb Þ ð15Þ
COP As
3494 Heat Mass Transfer (2017) 53:3487–3499

where AS is the annual energy savings obtained by dividing

the energy savings by the PWF. The parameters used in this
section are given in Table 3.

3 Results and discussion

3.1 Environmental conditions

The study is carried out in the climatic conditions of Garoua

(latitude: 9° 19′, longitude: 13° 22′, and standard meridian:
15°). When the air-conditioning system is ON, the indoor air
temperature is maintained constant at Tin = 26°C which is
within the thermally acceptable temperature range [34]. The
outdoor temperature and solar radiation are assumed to be the
average diurnal for each month. The solar absorptivity of the Fig. 5 Hourly solar-air temperature for January 15th and July 15th
wall exposed to the effects of short waves radiations is as-
sumed to be equal to 0.6 (for medium color [35]). The hourly and high thermal conductivity (concrete block) presents the
variations of solar-air temperature for south-facing wall on lowest time lag. These results are in harmonic with those
January 15 and July 15 are depicted in Fig. 5. The peak of
the solar-air temperatures are 51.48 °C and 35.15 °C during (a)
the representative day of January and July (the coldest month),
respectively.

3.2 Continuous air-conditioning space

In this section thermal characteristics such as annual average

of time lag and decrement factor are calculated under dynamic
thermal conditions. The wall is assumed to face south.

3.2.1 Thermal characteristics of uninsulated wall

configurations

The thermal characteristics of three uninsulated walls (con-

crete block: C, compressed stabilised earth block: EB, ES
and EP) coated on both sides by 15 mm layer of cement
mortar are studied with respect to different thicknesses. (b)
The yearly average time lag of different materials with
respect to their thickness is presented in Fig. 6a. It is observed
for all material that as the thickness increases, the time lag
increases. One can see that material with high heat capacity

Table 3 Economic
parameters used for Parameters values
calculations
Electricity for cooling
Cost 0.16 $/kWh
COP 2,5
Expanded polystyrene
Cost 164.32 $/m3
Inflation rate, g 2.9% [32]
Interest rate, i 5% [33]
Life time, n 30 years Fig. 6 Time lag (a) and decrement factor (b) of some uninsulated wall
with respect to their thicknesses
Heat Mass Transfer (2017) 53:3487–3499 3495

obtained by Asan [36]. It is also observed that the effect of X = EP and X = ES, respectively. The corresponding values
adding sawdust (10%) or pozzolan (45%) in the compressed for the worst case (C|X|PS wall) are 10.8 h and 10 h for X = EP
stabilized block significantly increases the time lag. Time lags and X = ES, respectively.
obtained for 30-cm-thick ES and EP walls are 4.21% and
16.14% in excess of that of 30-cm-thick EB wall, respectively.
In Fig. 6b is depicted the yearly average decrement factor 3.2.3 Effect of wall configurations on decrement factor
against the wall thickness for the studied walls. It is seen that
as the wall thickness increases, decrement factor decreases for The annual average of decrement factor for different wall con-
all building materials. This result is similar to that obtained by figurations is shown in Fig. 8. It is seen that, wall configura-
Asan [36]. Concrete block presents the highest decrement tions made from ES present better performance than those
factor for all wall thicknesses. The difference in decrement made from EP. It is also seen that inverting C and X layers
factor for ES and EP is not significant for all wall thicknesses. noticeable influence the decrement factor. For Bmid/int^, Bext/
However, the decrease in decrement factor with the wall thick- mid^ and Bint^ insulation configurations, placing X layer at
ness for EP is slightly significant than that for ES. The addi- the exterior or interior wall surface presents better perfor-
tion of the sawdust (10%) or pozzolan (45%) in compressed mance than locating C layer at the same positions. For exam-
building block as described by Meukam et al. [8] leads to a ple, fC|ES|PS = 2.36 fES|C|PS. For Bmid^ insulation configura-
decrease in decrement factor by up to 35.22% for wall thick- tion, placing X layer at exterior surface provides better perfor-
ness of 15 cm. mance than placing it at the interior surface. Invert C and X
By the end, thermal performance of wall configurations layers’ position has no effect for Bext/int^ insulation configu-
presented in Fig. 4 will be investigated. ration. For all, the best configuration in minimum decrement
factor perspective is the PS|C|PS|X wall.

3.2.2 Effect of wall configurations on time lag

3.2.4 Effect of wall configuration on the peak cooling
The yearly-averaged time lag of each wall configuration is transmission load
presented in Fig. 7. It is noted that, wall configurations made
from EP show larger time lag than the corresponding wall The peak of the cooling transmission load with respect to
configuration made from ES. This is due to the fact the storage different wall configurations during the representative day
capacity of the EP material is higher than that of ES material. of January (January, 15) is presented in Fig. 9. One can
It is also seen that, time lag increases with the insulation note a remarkable effect of wall configurations on the
layers. On can also observes that Bmid/int^, Bext/int^ and peak cooling transmission load. PS|C|PS|ES wall presents
Bint^ insulation configurations present worst performance a better performance from the minimum of the peak
than Bext/mid^ and Bmid^ insulation configurations. In fact, cooling transmission load (3.68 W/m2) while C|X|PS wall
wall configurations where massive layer (C, ES or EP) face presents the worst performance (4.82 W/m2). These re-
indoor conditions present better performance. The best case is sults are consistent with those of the decrement factor
the PS|C|PS|X wall with the time lag of 14.4 h and 13.5 h for presented above.

Fig. 7 Yearly average time lag with respect to different wall structures Fig. 8 Yearly average decrement factor with respect to different wall
(cAC) configuration (cAC)
3496 Heat Mass Transfer (2017) 53:3487–3499

Fig. 11 Influence of the insulation thickness on the relative performance

Fig. 9 Peak of cooling transmission load with respect to different wall of some wall structures (iAC)
structures on January, 15th(cAC)

wall configurations provide 6.12% and 8.67% reduction in

3.3 Intermittent air-conditioning energy consumption than PS|ES|C and C|ES|PS wall configu-
ration, respectively. It is also seen that, for all cases where C
3.3.1 Effect of wall configuration on daily cooling layer is placed at the exterior or interior wall surface, wall
transmission load configurations made from EP layer present better performance
than those made from ES. For the Bext/int^ insulation config-
The daily cooling transmission load during the occupied peri- uration, inverting C and X layer material position does not
od considered from 6 p.m. to 7 a.m. (when the cooling system indicate any impact in energy consumption.
is ON) of the representative day of January for different wall
configurations is depicted in Fig.10. It is seen that for all cases,
placing the insulation at the middle of the wall results in a 3.3.2 Impact of insulation thickness on daily cooling
higher energy consumption. On the other hand, placing the transmission load
insulation and the X layers at the interior and exterior surfaces
of the wall, respectively, results in the least energy consump- The daily cooling transmission loads of different wall config-
tion. For all the cases where insulation layer is placed at the urations are compared in Fig. 10 where calculations are being
exterior (interior) wall surface, placing the X layer at the inte- done from a fixed insulation thickness (40 mm). It is important
rior (exterior) surface of the wall presents better performance
than placing the C layer. For example, PS|C|ES and ES|C|PS

Fig. 10 Daily cooling transmission load in MJ/m2 for different wall Fig. 12 Variations of insulation cost, electricity consumption cost and
configurations (iAC) total cost with the insulation thickness for some wall configurations
Heat Mass Transfer (2017) 53:3487–3499 3497

3.4 Impact of the wall configuration on the optimum

insulation thickness

Al-Sanea and Zedan [14] showed that the optimum insulation

thickness of a wall is independent of the insulation configura-
tion for continuous air-conditioned space. The similar study in
the intermittent air-conditioned space deserves more attention.
Eight wall configurations chosen for the investigation are
those having insulation at the interior face. These wall struc-
tures present better performance in the intermittent air-
conditioning space case [13].
Fig. 12 presents the variation of insulation cost, energy cost
and their sum vs. insulation thickness for two wall configura-
tions (ES|C|PS and ES|PS|C|PS for example). It is seen that as
insulation thickness increases, energy cost decreases while
Fig. 13 Variation of energy savings versus insulation thickness for some
insulation cost increases. The total cost presents a minimum
wall structures
value corresponding to optimum insulation thickness. These
results are in harmonic with those obtained in literature [10,
to know how the insulation thickness impacts the com- 23]. Energy and total cost obtained from the ES|C|PS wall are
parative performance of different wall configurations in lower than that obtained from ES|PS|C|PS wall for all insula-
terms of energy consumption. Fig. 11 presents the daily tion thicknesses.
cooling transmission load of some wall configurations The variation of energy savings versus insulation thickness
with respect to the thickness of insulation layer. for two wall configurations (ES|C|PS and ES|PS|C|PS for ex-
Results are presented only for Bext/int^, Bint^ and ample) is depicted in Fig.13. Energy savings for all wall con-
Bind/int^ insulation configurations (the ones that are figurations present a maximum value at the optimum insula-
more efficient). C|ES|PS is taken as the base case. It tion thickness. It is also seen that, for insulation thickness less
is seen that, cooling transmission load decreases with than 0.066 m approximately, the energy savings obtained from
the insulation thickness for all cases. This result is sim- ES|C|PS wall configuration is slightly lower than that provid-
ilar to those obtained in literature [10, 14, 24]. It is also ed by ES|PS|C|PS wall configuration while the inverse situa-
observed that, as insulation thickness increase the reduc- tion is obtained for insulation thickness higher than 0.066 m
tion in energy consumption increase for all wall config- approximately.
urations excluding the ES|C|PS case where it slightly The payback period, optimum insulation thickness, energy
decreases. It is also seen that, insulation thickness influ- savings and yearly cooling transmission at the optimum insu-
ences the relative performance of some wall configura- lation thickness are summarized in Table 4. It is seen that
tions. For example, C|PS|ES|PS wall shows better per- insulation thickness applying for ES|C|PS wall must be
formance than ES|C|PS wall for the insulation thickness 0.6 cm thicker than of PS|C|PS|ES|PS wall. Energy savings
less than 30 mm. The inverse situation is observed for and payback periods provided by the two walls are signifi-
an insulation thickness higher than 30 mm. By using cantly close. At the optimum insulation thickness, the annual
the EP material, the results are in harmonic with those transmission load provided by PS|C|PS|ES|PS wall is about
obtained with ES material. 5.6% lower than that provided by ES|C|PS wall.

Table 4 Optimum insulation

thickness, energy savings, Wall structure Optimum Energy savings Payback period Yearly cooling transmission
payback periods and yearly insulation (m) ($/m2) (years) load (MJ/m2)
cooling load at optimum
thickness C|ES|PS 0.076 47.21 4.94 40.27
ES|C|PS 0.072 46.72 4.77 39.46
ES|PS|C|PS 0.071 47.47 4.65 37.97
C|PS|ES|PS 0.073 49.11 4.63 37.68
PS|C|ES|PS 0.071 48.82 4.54 38.07
PS|ES|C|PS 0.071 47.42 4.65 38.07
PS|C|PS|ES|PS 0.070 49.04 4.47 38.04
PS|ES|PS|C|PS 0.070 47.59 4.58 38.07
3498 Heat Mass Transfer (2017) 53:3487–3499

4 Conclusion 3. Bribián IZ, Capilla AV, Usón AA (2011) Life cycle assessment of
building materials: comparative analysis of energy and environ-
mental impacts and evaluation of the eco-efficiency improvement
The main purpose of this work was to determine the best wall potential. Build Environ 46:1133–1140
configuration made from concrete block, modified earthen 4. Taylor P, Fuller RJ, Luther MB (2008) Energy use and thermal
block and one or two insulation layer using the climatic con- comfort in a rammed earth office building. Energy Build 40:793–
800
ditions of Garoua in Cameroon located in tropical region. The
5. Pacheco-Torgal F, Jalali S (2012) Earth construction: lessons from
thermal conductivities of the two masonries layers are signif- the past for future eco-efficient construction. Constr Build Mater
icantly different. A Simulink model of a single room was 29:512–519
constructed to estimate the variations of temperature or heat 6. Collet F, Serres L, Miriel J, Bart M (2006) Study of thermal behav-
flux on the inside surface of wall facing south. The results of iour of clay wall facing south. Build Environ 41:307–315
7. Binici H, Aksogan O, Bodur MN, Akc E, Kapur S (2007) Thermal
this model were compared, with a good agreement, to those
isolation and mechanical properties of fibre reinforced mud bricks
obtained by EnergyPlus (validated software) in the same en- as wall materials. Constr Build Mater 21:901–906
vironmental conditions. The thermal performance of the walls 8. Meukam P, Jannot Y, Noumowe A, Kofane TC (2004) Thermo
was examined taking into account two operating modes of the physical characteristics of economical building materials. Constr
Build Mater 18:437–443
cooling system: continuous and intermittent. For the BcAC^,
9. Bal H, Jannot Y, Gaye S, Demeurie F (2013) Measurement and
the assessment parameters were time lag, decrement factor modelisation of the thermal conductivity of a wet composite porous
and instantaneous peak cooling transmission load. For the medium: laterite based bricks with millet waste additive. Constr
BiAC^ space case, the assessment parameter was the cooling Build Mater 41:586–593
transmission load which was then used to estimate the opti- 10. Ozkahraman HT, Bolatturk A (2006) The use of tuff stone cladding
in buildings for energy conservation. Constr Build Mater 20:435–
mum insulation thickness layer with respect to its location 440
from some economic considerations. 11. Ozel M, Pihtili K (2007) Optimum location and distribution of
The best wall layer configuration depends on the cooling insulation layers on building walls with various orientations.
system operating mode and the assessment parameter used. Build Environ 42:3051–3059
12. Al-Sanea SA, Zedan MF, Al-Hussain SN (2012) Effect of thermal
For the BcAC^, the best wall configuration from the maximum
mass on performance of insulated building walls and the concept of
time lag, minimum decrement factor and peak cooling load energy savings potential. Appl Energy 89:430–442
perspective, is placing a half layer of the insulation at the 13. Ibrahima M, Biwole PH, Wurtz E, Achard P (2014) A study on the
exterior surface and the other layer between the two different thermal performance of exterior walls covered with a recently pat-
massive layers with the modified earthen layer at the interior ented silica-aerogel-based insulating coating. Build Environ 81:
112–122
surface. For BiAC^, the best wall configuration from the min- 14. Al-Sanea SA, Zedan MF (2011) Improving thermal performance of
imum energy consumption depends on the total insulation building walls by optimizing insulation layer distribution and thick-
thickness. For the total insulation thickness less than 8 cm ness for same thermal mass. Appl Energy 88:3113–3124
approximately, the best wall configuration is placing the half 15. Kossecka E, Kosny J (2002) The influence of insulation configura-
tion on heating and cooling loads in a continuously used building.
layer of insulation material at the interior surface and the other Energy Build 34:321–331
half between the two different massive walls with the modi- 16. Tsilingiris PT (2006) Wall heat loss from intermittently conditioned
fied earthen material at the exterior surface. In addition, the spaces- the dynamic influence of structural and operational param-
optimum insulation thickness obtained from the yearly eters. Energy Build 38:1022–1031
cooling transmission calculated only during the occupied pe- 17. Bojic M, Yik F, Sat P (2001) The influence of thermal insulation
position in building envelope on the space cooling of high rise
riod and some economic considerations slightly depends on residential buildings in Hong Kong. Energy Build 33:569–581
their location. The last result is in disagreement with that ob- 18. Barrios G, Huelsz G, Rojas J (2012) Thermal performance of en-
tained for the BcAC B. velope wall/roofs of intermittent air-conditioned rooms. Appl
Therm Eng 40:1–7
19. Ozel M (2014) Effect of insulation location on dynamic heat-
transfer characteristics of building external walls and optimization
Acknowledgements The authors would like to acknowledge the of insulation thickness. Energy Build 72:288–295
Department of National Meteorology of Cameroon for providing the
20. Perez R, Ineichen P, Seals R (1990) Modeling of daylight availabil-
long-term dry bulk air temperature data.
ity and irradiance components from direct and global irradiance. Sol
Energy 44:271–289
21. Claessens J, Coulibaly Y, Kemajou A, (2014) Efficacité energétique
de la climatisation en région tropicale. <[Link].
References org/docs/prisme/ee [Link]>, [accessed 20.09.14]
22. Samri D (2008) Analyse physique et caractérisation hygroscopique
1. Goodhew S, Griffiths R (2005) Sustainable earth walls to meet the des matériaux de construction : approche expérimental et
building regulations. Energy Build 37:451–459 modélisation numérique. Thèse de Doctorat, Institut National des
2. Morel JC, Mesbah A, Oggero M, Walker P (2001) Building houses Sciences Appliquées de Lyon
with local materials: means to drastically reduce the environmental 23. Daouas N (2011) A study on optimum insulation thickness in walls
impact of construction. Build Environ 36:1119–1126 and energy savings in Tunisian buildings based on analytical
Heat Mass Transfer (2017) 53:3487–3499 3499

calculation of cooling and heating transmission loads. Appl Energy 30. Witte MJ, Henninger RH, Glazer J, Crawley DB, (2001) Testing
88:156–164 and validation of a new building energy simulation program,
24. Wati E, Meukam P, Nematchoua KM (2015) Influence of external Proceedings of Building Simulation 2001, IBPSA
shading on optimum insulation thickness of building walls in a 31. Arıcı M, Karabay H (2010) Determination of optimum thickness of
tropical region. Appl Therm Eng 90:754–762 double-glazed windows for the climatic regions of Turkey. Energy
25. Ozel M (2013) Determination of optimum insulation thickness Build 42:1773–1778
based on cooling transmission load for building walls in a hot cli- 32. Inflation, consumer prices (annual %) (2014) <[Link]
mate. Energy Convers Manag 66:106–114 [Link]/indicator/[Link]. [Link]>, [accessed 29.08.14].
26. Kalagasidis AS (2003) The whole model validation for HAM- 33. Prêt Foncier Epargne Logement Ordinaire. (2014) <[Link]
Tools. Case study: hygro-thermal conditions in the cold attic under [Link]/index. php/fr/pret-foncier-epargne-logement/pret-
different ventilation regimes and different insulating materials. foncier-epargne-logementordinaire>, [accessed 29.08.14].
Report R:03–6, 2003. Department of Building Technology, 34. Al-Homoud MS, Abdou AA, Budaiwi IM (2009) Assessment of
Chalmers University of Technology, Gothenburg, Sweden. Also monitored energy use and thermal comfort conditions in mosques
available for free downloading on [Link] in hot-humid climates. Energy Build 41(6):607–614. doi:10.1016/
27. Ibpt library. IBPT-1. (2014) <[Link] [Link].2008.12.005
[accessed 12.05.14].
35. Suehrcke H, Peterson EL, Selby N (2008) Effect of roof solar re-
28. Nielsen TR, Peuhkuri R, Weitzman P, Gudum C. (2002) Modeling
flectance on the building heat gain in a hot climate. Energy Build
Building Physics in Simulink, Report SR-02-03, BYG DTU
40:2224e2235
29. Crawley DB, Lawrie LK, Winkelmann FC, Buhl WF, Huang YJ,
Pedersen CO, Strand RK, Liesen RJ, Fisher DE, Witte MJ, Glazer J 36. Asan H (2006) Numerical computation of time lags and dec-
(2001) EnergyPlus: creating a new-generation building energy sim- rement factors for different building materials. Build Environ
ulation program. Energy Build 33(4):319–331 41:615–620