Renewable Energy 81 (2015) 785e794

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Thermal performance of a double pane window using glazing

available on the Mexican market

J.O. Aguilar a, b, *, J. Xaman a, G. Alvarez a ndez-Pe
, I. Herna rez a, C. Lo
 pez-Mata c
a n y Desarrollo Tecnolo
Centro Nacional de Investigacio gico, CENIDET-TNM-SEP, Prol. Av. Palmira s/n. Col. Palmira, Cuernavaca, Morelos CP 62490, Mexico
b
Universidad de Quintana Roo, Boulevard Bahía s/n, Col. Del Bosque, Chetumal Quintana Roo CP 77019, Mexico
c rez Auto
Universidad Jua noma de Tabasco, UJAT, Avenida Universidad s/n, col. Magisterial, Villahermosa Centro, Tabasco CP 86040, Mexico

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

Article history: A numerical study of a double pane window (DPW) with three types of commercial glass available in
Received 14 December 2014 Mexican market is presented. The DPW consist of two vertical semitransparent walls separated by an air
Accepted 23 March 2015 gap. The effect of varying the gap width (b) between glasses, the room temperature and the incident
Available online
solar radiation is analyzed. Simulations were conducted for three configurations; case 1: clear glass þ air
gap þ clear glass; case 2: clear glass þ air gap þ absorbent glass; and case 3: clear glass þ air
Keywords:
gap þ reflective glass. Optical transmittance and specular reflectance of each case were measured. Two
Double pane window
climatic conditions were analyzed, warm and cold climate. The results showed that, in order to increase
Thermal performance
CFD
or reduce the indoor heat gains, from b  0.02 m, the heat fluxes remains almost constant for both
climate conditions. For cold climate, the case 1 reached the highest energy savings (~10.5 and ~28.5%
higher than cases 2 and 3, respectively), however in warm climate it had the worst behavior (~105 and
~177% higher than cases 2 and 3, respectively). Finally, considering the case 1 as reference, the case 3 had
the best combined energy saving ($17.64 USD-kWh/year) and case 2 presents a combined energy saving
of $7.16 USD-kWh/year. Therefore is highly recommended the use of reflective double pane window, like
to case 3, in Mexican warm and cold climates.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction window. A variety of studies considering DPW have been devel-

oped over the years. For example, double glazing with air at at-
In regions where the climate causes large temperature gradients mospheric pressure [2,3] or with some inert gas [4,5], with water as
between indoor and outdoor ambient on buildings, the glazing are thermal fluid [6,7], with natural convection [8], DPW with enclosed
the component that have greater heat gain in these [1]. If a building slats [9], with transparent solar cells [10], and those that use phase
has glassed facades, which is the current trend in architecture, an change materials (PCM) [11]. Among DPW mentioned above, those
air conditioner is required to maintain the thermal comfort inside manufactured with air enclosed between the panes are the most
the building. Such equipment will consume a large amount of commonly used because their low cost with respect to other DPW
electricity. To overcome this problem, building's designer have used configurations.
different window configurations that use a wide variety of glass A large number of studies related to the thermal performance of
(monolithic, tempered and laminated) like absorbent, reflective DPW have been conducted over the years. The most relevant are
and low emissivity among the most used. those conducted by Lampert [12], who carried out an extensive
In other configurations, a double pane window (DPW) has a literature review about glass with solar control films. In other
better thermal performance than the traditional single layer similar work, Rubin [13] developed a general procedure for calcu-
lating the net energy flux through the glazed area of multipane
windows filled with air or other gases. Besides, the methodology
used by Rubin to evaluate multi glazing, with or without solar
control film, considering environmental variables such a tempera-
* Corresponding author. Universidad de Quintana Roo, Boulevard Bahía s/n, Col.
ture and wind velocity. The main finding is that with a gap width
Del Bosque, Chetumal Quintana Roo CP 77019, Mexico.
E-mail addresses: [email protected], [email protected] (J.O. Aguilar). larger than 2 cm, the thermal conductance (U ¼ 3 Wm2 K1) for

http://dx.doi.org/10.1016/j.renene.2015.03.063
0960-1481/© 2015 Elsevier Ltd. All rights reserved.
786 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794

Nomenclature Tg glass temperature, K

x, y dimensional coordinates, m
b width of the cavity, cm
Cp specific heat, J kg1 K1 Greek symbols
G solar radiation, W m2 a* absorptivity
*
H height of the glass, m 3 emissivity
q heat flux, W m2 l thermal conductivity, W m1 K1
qconv
ext convection heat flux towards the exterior of the room, r *
reflectivity
W m2 r density, kg m3
qconv
int
convection heat flux towards the interior of the room, t *
transmissivity
W m2
qrad
ext radiation heat flux towards the exterior of the room, Subscripts
W m2 conv convection heat transfer
qrad
int
radiation heat flux towards the interior of the room, ext external ambient
W m2 g glass
qt transmitted heat flux through the double pane int internal ambient
window, W m2 rad radiation heat transfer
sg extinction coefficient, m1 1, 2 glass number
T temperature, K

double glass unit remains constant, while adhering a solar control configuration was the second one. In similar works and after
coating on polyester film the U value reduces a 33%.The above re- conduct a brief overview of state of the art on solar window tech-
sults provided the basis for further research to find the optimal air nology, Chow and Li [8] proposed a double glass unit with enclosed
gap width in double glazing. Related works are presented below. water. Three different configurations, for forced flow (5  103 m/s)
Aydin [14] presented a numerical study of the heat transfer and natural flow (1  103 m/s), were used: clear/water flow/clear,
through a double pane window to determine the optimal gap width clear/water flow/reflective on clear, and tinted/water flow/clear.
between two panes for different climates. The study consider The results demonstrated that the best configuration for both cases
several climate conditions for Turkey locations, with Dt ranging were those that use tinted glass.
19e49  C and a cold wall temperature fixed in 21  C. The optimal Gueymard et al. [19] studied the SHGC and the behavior of the
gap width was found to be in the range of 12e21 mm for the lo- visible transmittance in vertical and inclined glazing as function of
cations proposed. Years later, Aydin [2] extended his work to a spectral solar radiation. They used 37 type of glass, single, double
conjugate conduction and natural convection heat transfer in and triple, with and without solar control film. The thermal eval-
double pane windows. The author's conclusion are that variation of uation was conducted with commercial software window 5.2.17
heat flux through the inner pane in all cases analyzed is constant for and Optics 5.1. The obtained results show SHGC values from 0.247
air gap 20 mm. Ismail and Henríquez [15] conducted a study of to 0.678 for double glass windows with low-e solar control films,
conduction and convection heat transfer in a double pane window presented the better behavior.
with gap with from 0.5 to 10 cm. The authors concluded that with Chow and Li [7] extended their previous works to analyze the
the boundary conditions proposed, the gap width has no significant thermal characteristics of double glass unit with enclosing water
impact on the SHGC (Solar Heat Gain Coefficient) neither on the SC considering outdoor and indoor temperatures, solar irradiance,
(shading coefficient). beam angle as well as flow velocity on the heat flow. In other work
Arici and Karabay [16] determined the optimal gap width of a [6], report the integrative thermal performance of a double glass
double pane window using the degree-day method. The study was unit with enclosing water as compared to the conventional single
conducted for four different climates zones of Turkey and the and double pane absorbing glazing. In both work, authors
heating cost was calculated for five types of fuels. The base tem- concluded that double pane window with enclosing water is suit-
peratures were 18, 20 and 22  C and the optimal gap width varies able for applications in warm climates. The most recent results
between 12 and 15 mm, according to climate zone, fuel type and reported about double pane window are focused on the effect of
base temperature. The results showed that it is possible achieve up varying the gap width between the glasses, indoor temperature,
to 60% of energy saving. solar radiation [3], emissivity and gas fill [20] and entire wave
Referent to thermal performance of double windows glazing, length heat radiation [21].
reported works like Ismail et al. [17] conducted an analysis to From the literature review it can be concluded that there are
compare thermal efficiency for two types of double pane window, plenty studies related to thermal performance of double pane
one filled with an absorbing gas and the other filled with a phase windows with and without solar control films. However, several
change material (PCM) and exposed to solar radiation. They found works use correlations for convective and radiative heat transfer
that double pane window filled with absorbing gas mixture and coefficients for convection and radiation between the glass and
using reflective glass is more efficient and has an SHGC between 55 fluid.
and 65% meanwhile the double pane window filled with PCM has The objective on this paper is to evaluate three types of glass
an SHGC ranging 60e80%. available in the Mexican market in a double pane window config-
Ismail et al. [18] conducted a comparative study of their previ- uration for using in cold and warm climates. The results will allow
ous work on three different systems: single glass, double glazing one to know which glass presents a better thermal performance
unit filled with an absorbent gas and double glass unit with natural and to know the optimal gap width between glasses aiming to
ventilation. The authors concluded that the most effective
 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794 787

control the energy flow towards the inside environment of the glazing-2 facing the external environment at temperature (Text) and
room. exposed to solar radiation (G). There are two horizontal adiabatic
walls. The gap between the layers of glass has a length (b). The three
2. Case study heat transfer mechanisms are considered in the study. Heat transfer
through the semitransparent walls occurs by conduction, convec-
2.1. Mathematical model tive heat transfer occurs inside the gap, and radiative exchange
among the cavity walls is also present.
The geometry of the double glazing window is shown in Fig. 1. Fig. 1(b) shows the heat transfer mechanisms in the double glass
Fig. 1(a) presents the configurations for the warm and cold climate. windows for warm climate. Glass-2 is a reflective glass, solar ra-
The physical model representing the double glazing window con- diation (G) impacts on this wall, part of this energy is reflected,
sists of two vertical semitransparent walls with a height “H”: absorbed by the glass and transferred towards inside to the cavity.
glazing-1 facing the room with an inside temperature (Tint), and The energy going through the Glass-2 (G1) impacts directly on the

Fig. 1. Physical model of the double pan window.

788 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794

Glass-1, part of this energy is reflected and/or absorbed for this vTg2
 
glass and transmitted to inside environment. The transmitted en- lg ¼ hext Tg2  Text þ r*g 3 *g Tg2
4 4
 Text for x ¼ x1 þ x2
vx
ergy arriving to the inner medium is the product of solar irradiance þ x3 ; 0  y  H
(G) and the transmittance of each element of the double glass
(8)
system. The total absorbed energy is the sum of the absorbed en-
ergy by each glass layer according to its optical properties. The
reflected energy is the energy reflected of the double glass system, vTg2
lg ¼ qconv
2 þ qrad
2 for x ¼ x1 þ x2 ; 0yH (9)
as is shown in the figure. Due to the absorbed energy, the two glass vx
layers have a variation of their internal energy manifested in a
On the Glass-1 for the warm and cold climate configuration, the
change of their temperature; as a consequence, both glasses ex-
horizontal boundaries are thermally insulated; at x ¼ 0 the vertical
change thermal energy by radiation and convection with the sur-
boundary exchange heat by convection and radiation to the inside
roundings, towards the interior (qconv and qrad ), the exterior (qconv
int int ext environment at Tint, at x ¼ x1 on the opposite vertical boundary is
and qrad conv , qrad , qconv and qrad ) between
ext ) and the environment (q1 1 2 2 made an energy balance considering conduction on the glass and
them. convection towards the air between glasses. The boundary condi-
The mathematical models for natural convection inside the tions are written as:
rectangular cavity with vertical semi-transparent walls (double
glazing window) are the conservative equations of mass, mo- vTg1
¼0 for y ¼ 0; 0 < x < x1 (10)
mentum and energy for laminar flow. The boundary conditions at vy
the solid walls for the fluid are assumed zero velocity. For the
thermal boundary conditions, adiabatic conditions are given at the vTg1
top and bottom walls and on the vertical boundaries an energy ¼0 for y ¼ H; 0 < x < x1 (11)
vy
balance between the corresponding glass and the air is carried out.
Mathematically the boundary conditions are given by:  
vTg1

lg ¼ hint Tg1  Tint þ r*g 3 *g Tg1
4 4
 Tint for x ¼ 0; 0  y  H
vx
qconv
3 þ qrad
3 ¼0 for y ¼ 0; x1 < x < ðx1 þ x2 Þ (1)
(12)

qconv
4 þ qrad
4 ¼0 for y ¼ H; x1 < x < ðx1 þ x2 Þ (2) vTg1
lg ¼ qconv
1 þ qrad
1 for x ¼ x1 ; 0yH (13)
vx
vTg1
qconv
1 þ qrad
1 ¼ lg for x ¼ x1 ; 0yH (3) The convective and conductive mathematical models described
vx
above were solved using the finite volume method [23]. The
convective terms are discretized applying the hybrid scheme and
vTg2
qconv
2 þ qrad
2 ¼ lg for x ¼ x1 þ x2 ; 0yH (4) the diffusive terms with the central difference scheme. Coupling
vx between the momentum and continuity equations is made with the
where qconv and qconv are the convection heat transfer fluxes from SIMPLEC algorithm [24]. The results in steady state are computed
1 2
the inside surface to the adjacent fluid on semitransparent walls 1 using the false transient method. Global convergence is achieved
and 2, respectively. The terms qrad rad rad and qrad are the net when the mass balance on each control volume is within a value of
1 , q2 , q3
1010 and the residuals for every different variable are about 1010.
4
radiative heat transfer fluxes at walls 1, 2, 3 and 4. Finally,
lg ðvTg =vxÞ is the heat flux by heat conduction through the corre- A radiative balance at the walls is solved using an iterative approach
sponding semi-transparent wall. in order to couple laminar natural convection to surface thermal
The net radiative method was used to calculate the resulting radiation effect at the boundaries. A summary for the numerical
heat fluxes from the radiation exchange in the rectangular cavity solution and the grid independence study was presented by the
[22]. The two horizontal surfaces of the cavity are assumed to be authors in a previous work [3].
opaque and diffuse, whereas the vertical walls are semitransparent.
The surface thermal radiation model was presented by Xama n et al.
2.2. Numerical verification and validation
2014 [3].
As a result of the energy balance on the semitransparent wall
The numerical code was developed in Fortran computer lan-
(Glass-1 and Glass-2), the differential equation for the temperature
guage and it was verified comparing the results against the
distribution over the glass wall is:
benchmark problem results for natural convection in a differen-
    tially heated square cavity reported by Ref. [25]. Comparisons for
v lg vTg v lg vTg 1 vQðxÞ
þ þ (5) this problem for a Ra ¼ 106 are shown in Table 1. We compare the
vx Cpg vx vy Cpg vy Cpg vx
values of average Nusselt numbers, maximum and minimum on the
hot wall, as well as the maximum velocity components with their
where QðxÞ ¼ G exp½sg ðw  xÞ is function of the glass extinction
respective coordinates. A maximum difference of 6.07% for the
coefficient sg [22] and w (x1 or x3) is the glass thickness.
minimum Nusselt number is obtained. Also, the numerical code
The boundary condition on the Glass-2 for the warm and cold
was verified comparing our results against the results for natural
climate configuration are written as:
convection and surface thermal radiation in a square cavity heated
vTg2 from below [26]. The convective and radiative Nusselt numbers of
¼0 for y ¼ 0; ðx1 þ x2 Þ < x < ðx1 þ x2 þ x3 Þ (6) the hot wall were compared with those due to [26], obtaining good
vy
agreement. Table 2 shows the comparison results for Ra ¼ 106 and
3 ¼ 0.5, for Nurad, the maximum difference was 5.31%; and for
vTg2 Nuconv, the minimum difference was 0.63% for Ra ¼ 2  106 and
¼0 for y ¼ H; ðx1 þ x2 Þ < x < ðx1 þ x2 þ x3 Þ (7)
vy 3 ¼ 0.
 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794 789

Table 1 Table 3
Results comparisons for the differentially heated cavity problem for the Ra ¼ 106. Comparison of average Nusselt numbers between the present study and the
experimental results for rectangular cavities.
[25] Present study Error (%)
H/b Rab ElSherbiny et al. [27] Present study Absolute difference (%)
Nuave 8.799 8.783 0.18
Numax 17.925 (0.0378) 17.451 (0.0387) 2.64 20 103 1.000 1.071 7.1
Numin 0.989 (1.00) 1.049 (1.00) 6.07 104 1.312 1.526 16.3
umax 0.077 (0.5, 0.850) 0.078 (0.5, 0.864) 1.30 105 3.002 3.341 11.3
vmax 0.260 (0.038,0.5) 0.262 (0.39,0.5) 0.77 106 6.050 6.521 7.8

Note: The values in () are dimensionless coordinates. 40 103 1.000 1.059 5.9
104 1.303 1.441 10.6
105 3.002 3.142 4.7
In order to validate the numerical code, the problem of laminar 80 103 1.000 1.054 5.4
natural convection of air in a differentially heated rectangular 104 1.303 1.389 6.6
cavity with an aspect ratio of 20, 40 and 80 was solved for different
Rayleigh numbers. The results for the average Nusselt number were
compared with the experimental results reported by ElSherbiny warm conditions, due the multiple reflections at the interior of
et al., 1982 [27]. From the comparison of results, we found a double glass pane.
maximum percentage deviation of 16.3% (Table 3). This comparison
allowed us to consider the numerical code as validated.
3. Results and discussion

2.3. Optical characterization Numerical results of conjugate heat transfer in double pane
windows under warm and cold conditions are presented. Three
The double pane window consists of two glass panes separated different configurations of DPW were used, clear, absorbent and
by air or other gas filled space to reduce heat transfer across a part reflective glass, all them of 6 mm thick. These glass pane obtained
of the building envelope. Because the optical properties of the from a supplier from Me rida Yucata n, Mexico were evaluated ac-
whole assembly are required for the model, the optical properties cording ISO 9050 standard as mentioned in section 2.2. The optical
of the three cases were measured on a Shimadzu Spectrophotom- and thermophysical properties of monolithic glass used are pre-
eter 3100 PC. The double glazing samples were constructed by sented in Table 4.
joining the two panes of interest in this research, with four glass The numerical analysis was conducted for different values of
sheets 2 mm thick attached at their ends with silicone glue. Addi- solar radiation, from 250 to 700 W/m2, in intervals of 150 W/m2.
tionally, aluminum strip was placed on the edge of each sample to The interior temperature used in warm climate was ranged from 15
reduce moisture penetration. Fig. 2 shows the double glazing to 30  C in intervals of 5  C and for cold climate the range was from
samples used. The transmittance and specular reflectance of the 20 to 35  C with same temperature interval. The external ambient
whole double pane window are presented in Fig. 3. temperature was fixed in 35  C for warm climate and 15  C for cold
In addition, the optical transmittance and near-normal inci- climate. The height of window was 0.8 m and the heat transfer
dence (5 ) specular reflectance of the each component of the convective coefficients were hext ¼ 6.8 W/m2 K and hint ¼ 6.2 W/
double pane window were also measured. Air and first surface m2 K, for outdoor and indoor conditions, respectively. Finally, the
mirror were used as reference in transmittance and reflectance gap width between glasses used in numerical simulation was of 0.5,
measurement, respectively. The solar control parameters, such as 1, 2, 4 and 8 cm.
total solar transmittance (tSol) and total solar reflectance (rSol) were The three configurations analyzed will be named hereafter as
evaluated for D65 spectra according to the procedure described in case 1: clear glass þ b þ clear glass; case 2: clear
ISO 9050 standard [28]. glass þ b þ absorbent glass and case 3: clear glass þ b þ reflective
Because of the double pane window configuration, the value of
specular reflectance changes if the thermal analysis is conducted in
warm or cold condition. From Fig. 3(b), it can be observed the
changes in the values of reflectance according the configuration
used. The optical behavior of clear glass is the same for warm and
cold conditions; therefore only one graph is presented. Fig. 3 shows
that total solar reflectance rsol values for absorptive glass (case 2)
and reflective glass (case 3) are higher for cold condition than for

Table 2
Comparison between the present study and the reported results in the literature for
the conjugate heat transfer in a square cavity.

Ra 3 Nuconv Nurad Nutotal

6
Ridouane 2  10 0 7.617 0 7.617
et al. [26]
Present study 7.665 (0.63%) 0 7.665 (0.63%)
Ridouane 106 0.5 6.267 6.599 12.866
et al. [26]
Present study 6.107 (2.56%) 6.249 (5.31%) 12.356 (3.97%)
Ridouane 4  105 1 4.722 11.462 16.183
et al. [26]
Present study 4.591 (2.78%) 10.871 (5.16%) 15.462 (4.46%)

Note: The values in () are the absolute difference in %. Fig. 2. Double glazing samples used in thermal analysis.
790 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794

100 Table 5
Convective (top) and radiative (bottom) heat flux (W/m2) for warm climate,
90 G ¼ 700 W/m2.

80 b (m) Tint ¼ 15  C Tint ¼ 30  C

Case 1: Clear
Case 1 Case 2 Case 3 Case 1 Case 2 Case 3
70
qconv
Transmittance (%)

int
60 0.005 64.13 102.09 113.72 33.33 69.72 80.88
0.01 59.72 92.85 102.28 32.26 63.91 72.93
50 0.02 57.17 88.34 97.06 31.54 60.62 68.71
0.04 57.22 88.88 97.76 31.30 60.70 69.02
40
0.08 56.97 88.26 96.99 31.29 60.34 68.49
30 qrad
int
0.005 50.23 82.54 92.84 29.59 63.72 74.57
20 Case 2: Absorbent
0.01 46.61 74.49 82.70 28.61 58.13 66.82
0.02 44.52 70.61 78.14 27.97 55.00 62.74
10 Case 3: Reflective
0.04 44.56 71.07 78.76 27.75 55.07 63.03
0 0.08 44.36 70.54 78.09 27.73 54.74 62.53
250 500 750 1000 1250 1500 1750 2000 2250 2500
Wavelength (nm)
(a) each case analyzed and for both climatic condition, warm and cold,
50
are presented.

Case 3: Reflective Cold 3.1.1. Warm climate

40 The average convective and radiative fluxes, as a function of b for
the three cases analyzed are shown in Table 5. Although the nu-
merical analysis was conducted using four values of solar radiation,
Reflectance (%)

30 we only present the results obtained for 700 W/m2 in the table. The
indoor ambient temperature (interior) was varied from 15  C to
Case 3: Reflective Warm
30  C and outdoor ambient temperature (exterior) maintained at
20 35  C. The negative values in the table indicate the amount of en-
Case 1: Clear
ergy entering to the room.
Case 2: Absorbent Cold It can be observed from Table 5 that case 1 shows the lowest
10 values for the convective and radiative heat fluxes to the interior in
both temperature values because its optical absorptance (~26%) is
Case 2: Absorbent Warm lower than the absorptance of case 2 (~65%) and case 3 (~56%). The
0 high absorptance values increase the temperature of the external
250 500 750 1000 1250 1500 1750 2000 2250 2500
glass (in contact with exterior ambient) and therefore convective
Wavelength (nm) and radiative heat fluxes also increase. For case 1, when the air
(b) inside the room has a temperature (Tint) of 15  C and the DWP has
an air gap (b) of 0.005 m, the convective flux is 64.13 W/m2. This
Fig. 3. Optical transmittance (a) and specular reflectance (b) of double pane windows
value diminishes for greater values of b, for example, when
for three cases analyzed.
b ¼ 0.02 m the convective flux is 57.17 W/m2. The table shows that

glass. It is considered that the double pane window is reversible;

therefore it can rotate 180 . Thus, when the DPW is used for warm
climatic conditions, the absorptive and the reflective glasses are in
contact with the outdoor environment. On the contrary, for cold
climatic conditions, the absorptive and the reflective glass are in
contact with the indoor air.

3.1. Thermal evaluation

In this section the average values of convective (qconv

int ), radiative
(qrad
int
) and directly transmitted fluxes (qt ) to the room interior for

Table 4
Thermophysical properties of three types of monolithic glass.

Glass type t*Sol [%] r*Sol [%] a* [%] l [W/m K] Cp [J/kg K] 3

*
[adim]

Clear 6 mm thick 0.84 0.09 0.07 1.0 750.0 0.850

Reflective 6 mm 0.08 0.30 0.62 1.0 750.0 0.837
thick
Absorbent 6 mm 0.47 0.07 0.46 1.0 750.0 0.840
Fig. 4. Energy transmitted directly toward the interior as function of G ¼ 700 W/m2 for
thick
warm climate.
 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794 791

-150 convective heat flux. For case 1, when Tint ¼ 15  C and b ¼ 0.005 m,
the radiative heat flux is 50.24 W/m2. If the air gap is increased to
0.02 m, the radiative flux decreases up to 44.52 W/m2. In general,
-200 for the three DPW configurations (cases 1, 2 and 3) and the two
Case 3
room temperatures (15 and 30  C), it can be observed that both the
convective and radiative heat fluxes have a similar behavior; they
-250 Tint=15°C remain at almost the same value once the length of the air gap (b)
Tint=30°C reaches 0.02 m.
qtot (W/m )
2

In Fig. 4, the direct transmitted energy (qt) into the room as a

function of the solar radiation is presented. In the three cases, the
-450
transmitted energy to the room increases as a solar radiation (G)
-500 Case 2 increases. For instance, in case 1, qt increases from
-550 187.66e525.44 W/m2 when the radiation increases from 200 to
700 W/m2. For case 2 and case 3 the transmitted energy varies from
-600
109.01 to 305.24 W/m2 and from 18.05 to 50.55 W/m2,
-650 Case 1 respectively. Considering case 1 as reference, with 700 W/m2 and
-700 b ¼ 0.02 m, the cases 2 and 3 can reduce the heat transmitted flux
0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 by 42% and 9.6%, respectively.
b (m) In Fig. 5 the average values for the total heat flux (qtot
¼ qconv þ qrad þ qt) into the room are presented. Total heat flux is
Fig. 5. Total heat flux for warm extreme condition: Tint ¼ 15 , Tint ¼ 30  C and the sum of the average fluxes (convective, radiative and trans-
G ¼ 700 W/m2. mitted) toward the interior ambient. The qtot is shown as a function
of b for G ¼ 700 W/m2 and interior temperatures of 15 and 30  C.
for larger air gaps more than 0.02 the heat flux is almost constant. This figure shows that case 3 has the lowest values of total heat flux
The cases 2 and 3 show an analogous behavior with respect to the compared to case 1 and case 2. Therefore, it allows less energy enter
air gap, the convective heat flux remain almost constant when to the indoor ambient, such effect is desired in a warm climate. The
b  0.02 m. The radiative heat flux has a similar behavior than figure also shows that qtot remains almost constant when

-120 -210
2
G=250 W/m
2
G=400 W/m
-220
-130
-230
-140 -240
Case1 -250 Case1
-150
Case2 Case2
-260
qtot (W/m )

Case3
qtot(W/m )

2
2

-160 Case3
-270
-170 -280

-180 -290
-300
-190
-310
-200 -320
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06 0.08
b(m) b(m)

-300 2
-380
G=550 W/m G=700 W/m
2

-315 -400
-330 -420
-345 Case1 -440
Case2 Case1
-360 Case3 -460 Case2
qtot (W/m )

Case3
qtot(W/m )

2
2

-375
-480
-390
-500
-405
-520
-420
-540
-435
-560
-450
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06 0.08
b(m) b(m)

Fig. 6. Total heat fluxes toward indoor ambient for four solar radiation conditions and Tint ¼ 20  C.
792 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794

b  0.02 m. When the indoor temperature (Tint) is 15  C and gap width (b). When DT (TinteText)  5  C, the differences between
b ¼ 0.02 m, the total heat flux entering to the room is total heat flux to the indoor are minor respect to DT values of 10, 15
627.14, 464.18 and 225.75 W/m2 for cases 1, 2 and 3, or 20  C. For Tint ¼ 20  C and b ¼ 0.02 m, qtot for cases 1, 2 and 3 are
respectively. 190.31, 169.50 and 136.11 W/m2, respectively, whereas for
Tint ¼ 35 and the same b, qtot are 146.94, 125.54 and 92.88 W/
3.1.2. Cold climate m2, respectively. When b  0.02 m the total heat flux remains
Fig. 6 shows the total heat flux (qtot) from cases 1, 2 and 3 with almost constant for all cases, as occurred for warm climate.
four values of solar radiation, 250, 400, 550 and 700 W/m2 and
Tint ¼ 20  C. It can be seen that case 1 allows more energy to enter 3.2. Energy saving cost, hypothetic case
the indoor ambient than the other cases. The case 2 and case 3
allows ~11 and ~28% less heat flux than case 1. This behavior is The yearly cost of electricity that could be saved by using each of
consistent for all G values used. In general, for the three cases it can the three DPW configurations was calculated. Two scenarios were
be observed that for b  0.02 m, the total flux presents a behavior considered, warm and cold climatic conditions. For warm condi-
almost constant. tions, the variables were: Text ¼ 35  C, Tint ¼ 24  C and G ¼ 700 W/
The most critical condition analyzed here is when G ¼ 250 W/m2 m2 and for cold condition: Text ¼ 15  C, Tint ¼ 24  C and G ¼ 250 W/
and Tint ¼ 20  C, because this condition has the lowest energy m2. The results of previous sections showed that the heat flux is
contribution to the indoor ambient and therefore reduced energy almost constant for greater values than b ¼ 0.02 m, therefore this
savings from energy consumption. For b  0.02 m and the above value was considered in the cost analysis.
conditions, case 1 has a qtot ¼ 190 W/m2 whereas cases 2 and 3 The current price authorized by the Mexican Commission of
have 170 and 136 W/m2. In general, results obtained show that Electricity (CFE) for Pacific Zone and average domestic consump-
case 1 has the best thermal performance for cold climate conditions tion (higher than 100 kWh per month) were used. Table 5 shows
due its optical properties, which allow the passage of a large the total heat flux for warm and cold conditions. The cost analysis
amount of heat flux transmitted. takes into account the qtot as total energy transferred to the indoor.
In Fig. 7 it can be observed the thermal behavior of three cases Case 1 is taken as reference in cold climatic conditions because this
analyzed for G ¼ 250 W/m2 and four different values of Tint: 20, 25, case had the highest total heat flux. On the other hand, case 3 is
30 and 35  C. The total heat fluxes are presented as a function of the taken as reference because this case had the lowest total heat flux.

-120 -110
Tint=20°C Tint=25°C
-130 -120

-140
-130
Case1
-150 Case2 Case1
-140 Case2
Case3
qtot (W/m )
qtot (W/m )

2
2

-160 Case3
-150
-170
-160
-180

-190 -170

-200 -180
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06 0.08
b (m) b (m)

-90 -70
Tint=30°C Tint=35°C
-100 -80

-90
-110
-100
-120 Case1 Case1
Case2 -110 Case2
qtot (W/m )

qtot (W/m )
2

-130 Case3 Case3

-120
-140
-130
-150
-140
-160 -150

-170 -160
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06 0.08
b (m) b (m)

Fig. 7. Total heat flux for a solar radiation of 250 W/m2 and four different indoor temperature.
 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794 793

Table 6 ➢ In cold climatic conditions, case 1 achieved the highest energy

Total heat flux (W/m2) for cold and warm conditions. savings, ~12% higher than case 2 and ~30% than case 3. However,
Cold condition Warm condition in warm climatic conditions case 3 shown the major energy
Case 1: Clear 179.23 602.12
saving, ~220 and ~120% higher than case 1 and case 2,
Case 2: Absorbent 158.18 440.06 respectively.
Case 3: Reflective 125.06 199.79 ➢ The global balance costs indicate that case 3 is the best option for
energy saving. It is possible to save up to $17.64 and $10.48 USD
per kWh in a year in comparison to case 1 and case 2,
Table 7 respectively.
Cost analysis for use double pane windows. ➢ A reversible window with a configuration such as case 3 is
Fare Energy Added Energy to maintain Cost Energy recommended for locations with warm and cold seasons. This
(USD kWh) comfort (kWh per day) (USD-kWh/year) recommendation is on basis that in cold conditions the pay or
Condition Case 1 Case 2 Case 3 Case 1 Case 2 Case 3 added energy ($3.64 USD per kWh) is lower than energy saved
in warm condition ($21.28 USD per kWh).
Warm 0.058 2.011 1.201 0 21.28 12.71 0
Cold 0.074 0 0.105 0.270 0 1.41 3.64

Balance $21.28 $14.12 $3.64

Acknowledgments

The first author is grateful to CONACyT-Mexico for the sabbatical

From Table 6 it can be observed that in cold climatic conditions, fellowship awarded. The authors are grateful to Oscar Gomez-Daza
in case 2 it is necessary to add 21.05 W/m2 to maintain the indoor and Jose Campos for the technical support in the evaluation of the
temperature at 24  C with respect to case 1, while in case 3 it is optical properties conducted in the Instituto de Energías Renov-
necessary to add ~54 W/m2. Moreover, in warm condition for cases ables-UNAM.
1 and 2 it is necessary to add ~240 and ~402 W/m2, respectively, to
maintain indoor comfort conditions.
Table 7 shows the yearly energy consumption cost considering References
the use of five hours per day for room heating or cooling to
maintain comfort conditions. The electricity cost has been applied [1] Pal S, Neogi S, Roy B. Heat transfer modelling on windows and glazing under
the exposure of solar. Energy Build 2009;41:654e61. http://dx.doi.org/10.
half year for each climatic condition, as can be observed in second
1016/j.enbuild.2009.01.003.
column of Table 7. [2] Aydin O. Conjugate heat transfer analysis of double pane windows. Build
As mentioned previously, case 1 was the reference in cold con- Environ 2006;41:109e16. http://dx.doi.org/10.1016/j.enbuild.2014.03.015.
[3] Xam an J, Pe 
rez-Nucamendi C, Arce J, Hinojosa J, Alvarez n I.
G, Zavala-Guille
dition, therefore in other cases it is necessary add energy to
Thermal analysis for a double pane window with a solar control film for using
maintain comfort conditions. Table 7 shows that case 2 and 3 are in cold and warm climates. Energy Build 2014;76:429e39. http://dx.doi.org/
$1.41 and $3.64 USD more expensive on annual basis than the 10.1016/j.enbuild.2014.03.015.
reference. The previous results were expected because case 1 has a [4] Muneer T, Han B. Simplified analysis for free convection in enclosures-
application to an industrial problem. Energy Convers Manage 1996;37:
higher optical transmittance (84%) than cases 2 (47%) and 3 (8%). In 1463e7. http://dx.doi.org/10.1016/0196-8904(95)00216-2.
warm condition the reference was case 3, which is configuration [5] Weir G, Muneer T. Energy and environmental impact analysis of double-
that presents the highest energy savings as can be observed in glazed windows. Energy Convers Manage 1998;39:243e56. http://dx.doi.
org/10.1016/S0196-8904(96)00191-4.
Table 7. For case 1 it is necessary to spend $21.28 and $14.12 USD per [6] Chow Tin-Tai, Li Chunying, Lin Zhang. The function of solar absorbing window
kWh on entire year. Although the case 3 has the highest absorp- as water-heating device. Build Environ 2011;46:955e60. http://dx.doi.org/10.
tance (62%), the reflectance value (30%) plays an important role to 1016/j.buildenv.2010.10.027.
[7] Chow Tin-Tai, Li Chunying, Lin Zhang. Thermal characteristics of water-flow
obtain the better thermal performance. double-pane window. Int J Therm Sci 2011;50:140e8. http://dx.doi.org/10.
Conducting a global balance between two climatic conditions 1016/j.ijthermalsci.2010.10.006.
we found that case 3 has the greatest energy savings, because one [8] Chow Tin-Tai, Li Chunying, Lin Zhang. Innovative solar windows for cooling-
demand climate. Sol Energy Mater Sol Cells 2010;94:212e20. http://
would spend only $3.64 USD per kWh on yearly basis (cold con-
dx.doi.org/10.1016/j.solmat.2009.09.004.
dition) while for cases 1 and 2 one would spend $21.28 and $14.12 [9] Chaiwiwatworakul P, Chirarattananon S. A double-pane window with
USD per kWh during a year. It is important to point out that to enclosed horizontal slats for daylighting in buildings in the tropics. Energy
obtain a more accurate cost analysis it is necessary to take into Build 2013;62:27e36. http://dx.doi.org/10.1016/j.enbuild.2013.02.027.
[10] Han Jun, Lu Lin, Yang Hongxing. Numerical evaluation of the mixed convec-
account daily solar radiation conditions for one specific location tive heat transfer in a double-pane window integrated with see-through a-Si
besides considering another variables like average temperatures PV cells with low-e coatings. Appl Energy 2010;87:3431e7. http://dx.doi.org/
and building orientation, which helps to obtain a more reliable 10.1016/j.apenergy.2010.05.025.
[11] Ismail KAR, Morais Lino F, Ribeiro da Silva RC, Salinas CT. Models for thermal
energy saving cost. efficient double glass window. Eur Int J Sci Technol 2014;3:15e32.
[12] Lampert C. Heat mirror coatings for energy conserving windows. Sol Energy
Mater Sol Cells 1981;6:1e41. http://dx.doi.org/10.1016/0165-1633(81)
4. Conclusions 90047-2.
[13] Rubin M. Calculating heat transfer through windows. Energy Res 1982;6:
An evaluation of the thermal performance of three types of glass 341e59. http://dx.doi.org/10.1002/er.4440060405.
[14] Aydin O. Determination of optimum air-layer thickness in double-pane win-
in a double pane window was carried out. Warm and cold climatic dows. Energy Build 2000;32:303e8. http://dx.doi.org/10.1016/
conditions were considered and the following is concluded: j.enbuild.2010.05.013.
[15] Ismail KAR, Henríquez JR. Two-dimensional model for the double glass
naturally ventilated window. Int J Heat Mass Transf 2005;48:461e75. http://
➢ The gap width (b) between glasses, in which the heat flux
dx.doi.org/10.1016/j.ijheatmasstransfer.2004.09.022.
through the indoor remains almost constant is 0.02 m for the [16] Arıcı Müslüm, Karabay Hasan. Determination of optimum thickness of
two climatic conditions and the three cases of study analyzed. double-glazed windows for the climatic regions of Turkey. Energy Build
Using lower values of b implies a higher heat flux gain in warm 2010;42:1773e8. http://dx.doi.org/10.1016/j.enbuild.2010.05.013.
[17] Ismail KAR, Salinas CT, Henríquez JR. Comparison between PCM filled glass
climatic conditions and lower heat flux in cold climatic condi- windows and absorbing gas filled windows. Energy Build 2008;40:710e9.
tions; therefore more energy consumption. http://dx.doi.org/10.1016/j.enbuild.2007.05.005.
794 J.O. Aguilar et al. / Renewable Energy 81 (2015) 785e794

[18] Ismail KAR, Salinas CT, Henríquez JR. A comparative study of naturally [23] Patankar S. Numerical heat transfer and fluid flow. Washington, DC, USA:
ventilated and gas filled windows for hot climates. Energy Convers Manag Hemisphere; 1980.
2009;50:1691e703. http://dx.doi.org/10.1016/j.enconman.2009.03.026. [24] Raithby G, Van Doormaal J. Enhancement of the SIMPLE method for pro-
[19] Gueymard C, duPont W. Spectral effects on the transmittance, solar heat gain, ducting incompressible fluid flow. Numer Heat Transf 1984;7:147e63.
and performance rating of glazing systems. Sol Energy 2009;83:940e53. [25] De Vahl Davis G. Natural convection of air in a square cavity: a Bench Mark
http://dx.doi.org/10.1016/j.solener.2008.12.012. numerical solution. Int J Num Meth Fluids 1983;3:249e64.
[20] Arıcı Müslüm, Kan Miraç. An investigation of flow and conjugate heat transfer [26] Ridouane EH, Hasnaoui M, Amahmid A, Raji A. Interact between natural
in multiple pane windows with respect to gap width, emissivity and gas convection and radiation in a square cavity heated from below. Numer Heat
filling. Renew Energy 2015;75:249e56. http://dx.doi.org/10.1016/j.renene. Transf Part A 2004;45:289e311.
2014.10.004. [27] ElSherbiny SM, Raithby GD, Hollands KGT. Heat transfer by natural convection
[21] Tian-Peng W, Liang-Bi W. A steady heat transfer model of hollow double across vertical and inclined air layers. J Heat Transf 1982;104:96e102.
glazing under entire wave length heat radiation. Energy Build 2014;81: [28] Glass in buildingddetermination of light transmittance, solar direct trans-
72e83. http://dx.doi.org/10.1016/j.enbuild.2014.06.014. mittance, total solar energy transmittance, ultraviolet transmittance and
[22] Modest M. Radiative heat transfer. New York, USA. 1993. related glazing factorsdISO 9050:2003 standard test method.