Df

MK

a

ARRA

KACDE

1

lsrsaihbd[

canelHcsBafi

(

0d

Energy and Buildings 42 (2010) 1773–1778

Contents lists available at ScienceDirect

Energy and Buildings

journa l homepage: www.e lsev ier .com/ locate /enbui ld

etermination of optimum thickness of double-glazed windowsor the climatic regions of Turkey

üslüm Arıcı ∗, Hasan Karabayocaeli University, Department of Mechanical Engineering, Umuttepe Campus, 41380 Kocaeli, Turkey

r t i c l e i n f o

rticle history:eceived 10 March 2010

a b s t r a c t

In this study, the optimum air layer thickness of double-glazed windows is determined using the degree-days method. Calculations are obtained for Iskenderun, Kocaeli, Ankara and Ardahan which are in different

eceived in revised form 24 April 2010ccepted 13 May 2010

eywords:ir layer thicknesslimate zones

climate zones of Turkey. Heating cost of the objective function is calculated for natural gas, coal, fuel-oil,electricity and LPG. The optimum air layer thickness is obtained for three different base temperatureswhich are 18, 20 and 22 ◦C. The results show that the optimum air layer thickness varies between about12 and 15 mm depending on the climate zone, fuel type and base temperature. The effect of the fuel typeand the base temperature on the optimum air layer thickness diminishes in cold zones. It is shown that

zed w

ouble-glazed windownergy saving

with a well-optimized gla

. Introduction

The world energy demand is growing with increasing of popu-ation and growing economy. It is particularly important for Turkeyince energy sources met only about 25% energy requirement; theest is imported [1]. Therefore, energy saving has become compul-ory. In Turkey, about 30% of total energy is used in the residentialnd commercial buildings [2]. Therefore, the energy saving in build-ngs should be focused on by a designer. 82% of energy is used foreating in the buildings in Turkey [3]. Energy loss from a typicaluilding occurs 40% through external walls, 30% through win-ows, 17% through doors, 7% through roof, and 6% from floors2].

Heat losses from external walls, roofs and floors can be reducedonsiderably using insulation materials. In the literature, therere a great number of studies to determine the optimum thick-ess of the insulation material depending on climate zones andnergy sources. A systematic approach for optimization of insu-ation material thickness applied to Palestine was developed byasan [4]. It was based on the life cycle cost analysis. Generalizedharts for selecting the optimum insulation thickness were pre-ented as a function of degree-days and wall thermal resistance.

olattürk [5,6], Dombaycı et al. [7], Kaynakli [8], Sisman et al. [9]nd Ucar and Balo [10] obtained the optimum insulation thicknessor various cities from different climate zones of Turkey consider-ng different fuel types. Growing number of similar studies were

∗ Corresponding author. Tel.: +90 532 365 6880; fax: +90 262 3033003.E-mail addresses: muslumarici@kocaeli.edu.tr, muslumarici@gmail.com

M. Arıcı), hkarabay@kocaeli.edu.tr (H. Karabay).

378-7788/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.enbuild.2010.05.013

indow, up to 60% energy saving can be achieved.© 2010 Elsevier B.V. All rights reserved.

done for different countries [11–14]. The annual energy require-ment of buildings for heating or cooling can be obtained by meansof degree-days method which is broadly used in energy consump-tion to plan and predict heating and cooling loads of buildings. Thismethod can be defined simply as a quantitative reflection of thedemand for energy needed to heat or cool spaces. Turkey has beendivided into four different climate zones as defined in Turkish Stan-dard Number 825 (TS 825) [15] and Buyukalaca et al. [16] presenteddegree-days data of the climate zones.

Although there are a great number of studies about the optimumthickness of the insulation material for the walls, there are fewstudies published about the optimization of glazing in buildings.As well-known, about one-third of the total heat loss of build-ings occurs through windows which makes undesirable from anenergy conversation point of view. One way to reduce the energyloss through window is to install double or triple pane windows.The energy saving can be further enhanced by replacing the airfilled in the cavity between the two panes with an inert gas suchas argon, krypton or xenon. Nearly half of the heat loss through adouble-pane window is by radiation. Coating glass surfaces witha low-emissivity material is another method to reduce the energyloss from double-glazed windows.

Installing double or triple pane window introduce stationaryair layer in the glazing unit which decreases the conduction heattransfer. However, increasing the air layer thickness further a crit-ical value initiates convection currents in the enclosed air space,

which increases the heat transfer and thus defeats the purpose [17].Korpela et al. [18] published a numerical study about heat trans-fer through a double-pane window using finite difference methodand presented detailed plots of the stream patterns and isotherms.Aydın [19,20] also presented numerical studies to determine the

1 nd Buildings 42 (2010) 1773–1778

ceTttfcSsnvwee

ethd2ssHwTssptiwbwtcmietlttotS

2

sfi

q

Hadst

E

we

E

774 M. Arıcı, H. Karabay / Energy a

ritical value of air layer thickness between two panes for differ-nt climate zones of Turkey considering only the amount of energy.he calculations given in [19] showed that the optimum air layerhickness between panes varies from 12 to 21 mm depending onhe climate zone. The effect of the convection boundary conditionsor the outer surfaces was investigated in [20]. The new boundaryonditions reduced the optimum air layer thickness given in [19].oylemez [21] presented a thermoeconomic optimization analy-is yielding a simple algebraic formula for estimating the optimumumber of panes for windows. The overall heat transfer coefficientalues were correlated for single, double, triple and quadruple paneindows that are used in HVAC and refrigeration applications. The

ffect of the gas type between panes and frame materials on thenergy loss was investigated by Weir and Muneer [22].

To the best knowledge of authors only energy loss is consid-red on the determination of the optimum air layer thickness inhe literature. Increasing the air layer thickness decreases energy;owever, it increases the investment cost of the framing of the win-ow. The heat transfer rate of a single-glazed window is about.5 times higher than which of a double-glazed window. Con-equently, installing double-glazed windows can reduce both aignificant amount of energy and an environmental pollution.owever, approximately 87% of the buildings have single-glazedindows, 9% have double-glazed, and only 4% have low-e glass in

urkey [23]. Therefore, properly designed double-glazed windowshould become widespread all over the country. Some parametersuch as investment cost, indoor and outdoor temperatures, fuelrice for the heating and the present worth factor (PWF) influencehe optimum value of the air layer thickness. The aim of this studys to determine the optimum air layer thickness of double-glazed

indows considering the both heating and investment costs. It haseen found that there are commonly two types of double-glazedindow which has 12 and 16 mm air layer thickness in the produc-

ion list of Turkish companies. Each type of double-glazed windowsan be seen easily in every city of Turkey which shows that the cli-ate conditions are not considered as a decision parameter in the

nstallation of the double-glazed windows. In this study, to see theffect of the climate conditions on the determination of the air layerhickness for the glazing, Iskenderun, Kocaeli, Ankara and Ardahanocated in different climate zones are chosen as model cities. Also,he effect of the five different fuel types is considered to determinehe optimum air layer thickness. To do so, an objective functionf the economical model is outlined in Section 2 and the parame-ers are given in Section 3. The calculated results are compared inection 4.

. Method

In this study, the optimum air layer thickness is determined con-idering the investment cost and the energy cost due to heat lossrom window. Heat loss (q) from per unit area through the windows determined according to Eq. (1):

= U(Tb − To) (1)

ere U is the overall heat transfer coefficient, Tb is the base temper-ture and To is mean daily temperature. The summation of Eq. (1)uring heating season will be the heating energy. In this study, forimplicity, Eq. (2) is used to obtain the heating energy according tohe degree-day method [16]:

= 86400 DDU (2)

here DD is the degree-day sum. Considering the heating systemfficiency, �s, the energy demand is given as:

A = 86400 DDU

�s(3)

Fig. 1. Geometry of a double-glazed window.

Schematic representation of a double-glazed window is shownin Fig. 1 and the overall heat transfer coefficient (U) of a double-glazed window is defined in Eq. (4):

UD = 1(1/hi) + (t/kglass) + (1/hspace) + (t/kglass) + (1/ho)

(4)

here hi (=8.29 W m−2 K−1) and ho (=34 W m−2 K−1) are the heattransfer coefficients at the indoor and outdoor surface of win-dow, respectively [17], t is the thickness of the glass which istaken as 4 mm for both single and double-glazed configurationsand kglass (=0.92 W m−1 K−1) is the thermal conductivity of the glass.Nearly half of the energy passes through the window is transferredby radiation and the rest by conduction or convection. So, hspace

(=hradiative + hconvective) is the combination of radiative and convec-tive heat transfer coefficients in the air layer between the panes.The term 1/hspace in Eq. (4) represents the overall thermal resis-tance of the trapped air layer. Disregarding the thermal resistancesof glass layers in Eq. (4), the heat transfer coefficient at the centerof glass given by [24] is as shown in Fig. 2.

There is a critical value for the air layer thickness which mini-mizes the heat loss as seen from Fig. 2. The heat passes through thewindow by conduction mechanism below a critical air layer thick-ness. Beyond the critical thickness, the heat transfer mechanismtransforms to convection that causes increase in the U-factor. For aspecified glass material, the effect of the air layer thickness on theradiation heat transfer is negligible. However, the total heat trans-fer rate will be affected significantly by the emissivity (ε) of theglass materials. In this study, the calculations are done for an ordi-nary glass material (ε = 0.84) and the effect of the emissivity on theobjective function is not considered. The overall heat transfer coef-ficient given in Eq. (4) can be rewritten for the objective functionof this study by correlating data shown in Fig. 2:

U = 0.0074L2 − 0.2179L + 4.3581 (5)

Here L (mm) is the air layer thickness. The correlation coefficientof Eq. (5) is R2 = 0.98. Overall heat transfer coefficient for a single-

M. Arıcı, H. Karabay / Energy and Buildings 42 (2010) 1773–1778 1775

Fd

g

U

Sg

E

a

m

p

C

AlP

P

i

r

f

r

A

P

Taof

C

Table 1The degree-days of cities for different base temperatures [16].

Climate zone Tb = 18 ◦C Tb = 20 ◦C Tb = 22 ◦C

˙

tricity is also considered as a heating source in this study.Cf = 0.1059$ kWh−1 for electricity and the efficiency (�s) of the elec-trical heater is assumed to be 0.99.

The analysis is done according to the published records of theCentral Bank of Republic of Turkey [26] and the State Institute of

Table 2Prices, lower heating values of fuels and efficiencies of heating systems [25].

ig. 2. Variation of heat transfer coefficient with air layer thickness in verticalouble-glazed windows (ε = 0.84) [24].

lazed window is given in Eq. (6):

S = 1(1/hi) + (t/kglass) + (1/ho)

(6)

ubstituting Eq. (5) into Eq. (3), the annual heating load for a double-lazed window can be rewritten as:

A = 86400DD

�s(0.0074L2 − 0.2179L + 4.3581) (7)

Considering the lower heating value of the fuel (LHV), themount of the annual fuel consumption can be calculated as:

f = 86400DD

�sLHV(0.0074L2 − 0.2179L + 4.3581) (8)

Assuming Cf is the unit price of the fuel, the annual heating coster unit area for a double-glazed window can be obtained as:

A,D = mf Cf (9)

ssuming the lifetime is N years, the total energy cost over theifetime is converted to the present value by multiplying Eq. (9) byWF:

WF = (1 + r)N − 1

r(1 + r)N(10)

f i is the interest rate and g is the inflation rate for i > g:

= i − g

1 + g(11)

or i < g:

= g − i

1 + i(12)

nd for i = g:

WF = N

1 + i(13)

In this study, the investment cost of double-glazed window inurkish market is obtained by a marketing survey. The effect of the

ir layer thickness of the framing for the double-glazed windowsn the prices varies from company to company. The data obtainedrom the marketing survey is correlated as given in Eq. (14):

I = 0.2759L + 20.475 (14)

Iskenderun I 690 1022 1406Kocaeli II 1786 2257 2790Ankara III 2677 3214 3811Ardahan IV 5137 5845 6571

where the correlation coefficient of Eq. (14) is R2 = 0.86. Approx-imately 5 mm increase in the air layer thickness of frame causesabout 6% increase on the price.

The objective function for the double-glazed window is givenby:

CT = CA,DPWF + CIL (15)

Combining Eqs. (8), (9) and (14) into Eq. (15), the objective functioncan be rewritten as:

CT = 86400 DD

�sLHV(0.0074L2 − 0.2179L + 4.3581) Cf PWF

+ 0.2759L + 20.475 (16)

As a result, the optimum air layer thickness between two glassescan be obtained by minimizing Eq. (16) as:

xopt = 86400 DD

�sLHV(0.0148L − 0.2179)Cf PWF + 0.2759 (17)

Similarly, the heating cost for a single-glazed window can bewritten as:

CA,S = 86400 DDUsCf PWF

�sLHV(18)

The amount of the annual heating energy saving can be calculatedby subtracting the annual heating cost of the double-glazed win-dow (Eq. (9)) from which of the single-glazed window (Eq. (18))as:

�C = CA,S − CA,D (19)

3. Analysis

In order to show the effect of the climate conditions on the deci-sion of the double-glazed windows, four sample cities in differentclimate zones of Turkey are chosen as mentioned in Section 1. WhileIskenderun is the warmest city which is located in the first zone,Ardahan is the coldest city located at the fourth zone. Kocaeli is amoderate city in the second zone and Ankara is the most crowdedcity in the third zone. The degree-days of these cities are given inTable 1.

Different fuels such as natural gas, fuel-oil, coal and LPG are usedfor heating systems in Turkey. Prices, LHV of the fuels and efficiencyof the heating systems are given in Table 2.

Since electrical heaters are used in some residences, the elec-

Price LHV �s

Natural gas 0.4809$ m−3 34534500 J m−3 0.93Coal 0.3791$ kg−1 29302000 J kg−1 0.65Fuel-oil 1.2414$ kg−1 41336750 J kg−1 0.80LPG 2.5277$ kg−1 46464600 J kg−1 0.90

1776 M. Arıcı, H. Karabay / Energy and Buildings 42 (2010) 1773–1778

Fn

SfP

4

dehoomcomdssmnssvItiI

pAstTtpdz

nFc

there is about 2.6 mm difference of the optimum air layer thicknessfor DD = 500 where the optimum air layer thickness is 11.3 mm forthe natural gas and 13.9 mm for LPG. As DD increases, the effect ofthe fuel type on the optimum air layer thickness diminishes. Alsoit can be seen from Fig. 8 that the effect of DD on the optimum

ig. 3. Effect of air layer thickness on the costs for Iskenderun (Tb = 18 ◦C; N = 10;atural gas).

tatistics [27], the annual interest rate (i) and the inflation rate (g)or the year 2009 were taken as 10.25% and 6.53%, consequentlyWF = 8.32 where the lifetime (N) is taken as 10 years.

. Results and discussion

In this study, the optimum thickness of a double-glazed win-ow is determined for four different cities and for five differentnergy sources. The computed values of the investment cost, theeating cost and the total cost are shown in Fig. 3. These results arebtained for Iskenderun, using natural gas as a fuel source, lifetimef 10 years and Tb = 18 ◦C. As seen from Fig. 3, while the invest-ent cost increases with increasing air layer thickness, the heating

ost decreases at first and then start to increase. This behaviorf heating cost differs from similar studies related to the opti-um insulation thickness such as in [4–14]. While the heating cost

ecreases in these studies as the insulation thickness increases, ittarts to increase beyond a critical air layer thickness in the presenttudy. Because the convective heat transfer becomes dominantechanism that increases heat loss after the critical air layer thick-

ess. Consequently, the heating cost increases beyond this point ashown in Fig. 3. The behavior of the total cost and the heating costhows similarities. While the air layer thickness of the minimumalue for the heating cost is 15 mm, it is 12 mm for the total cost.t shows that if a user does not consider the investment cost, thehickness of the glazed window should be 15 mm. Considering thenvestment cost, the optimum thickness decreases to 12 mm for˙skenderun.

Similar calculations are shown in Fig. 4 for Ardahan. Com-aring Fig. 3 and Fig. 4, it can be said that the heating cost ofrdahan is about 4 times higher than Iskenderun. The compari-on between the heating cost and the investment cost shows thathe heating cost is more dominant in Ardahan as shown in Fig. 4.he difference between the air layer thickness which minimizeshe total cost and the heating cost decreases for Ardahan com-aring with Iskenderun. Thus, the optimum air layer thickness ofouble-glazed windows is determined by the heating cost in coldones.

The plots shown in Fig. 4 are re-calculated altering the fuel fromatural gas to LPG for Ardahan and shown in Fig. 5. As shown inig. 5, the heating cost and thus the total cost increase dramati-ally when LPG is used instead of natural gas in the heating system.

Fig. 4. Effect of air layer thickness on the costs for Ardahan (Tb = 18 ◦C; N = 10; naturalgas).

However, the effect of fuel type on the optimum air layer thicknessis insignificant for Ardahan.

The effect of fuel type on the optimum air layer thickness for dif-ferent cities is shown as more detailed in Figs. 6 and 7 for Tb = 18 ◦Cand Tb = 22 ◦C, respectively. As seen from these figures, the opti-mum air layer thickness for Ardahan is almost the same for all thekind of fuel types. While the effect of fuel type on the optimum airlayer thickness is significant for Iskenderun, it is less significant forKocaeli and Ankara. As the base temperature increases, the opti-mum air layer thickness increases for all the cities as well. Whilethe optimum air layer thickness varies from 12.2 to 14.6 mm forTb = 18 ◦C, it varies from 13.5 to 14.7 mm for Tb = 22 ◦C regarding allthe climate zones of Turkey.

The variation of the optimum air layer thickness as a function ofdegree-days is shown in Fig. 8 for all the fuel types considered in thisstudy. As seen from Fig. 8, fuel type has a significant effect on theoptimum air layer thickness for the small DD values. For instance,

Fig. 5. Effect of air layer thickness on the costs for Ardahan (Tb = 18 ◦C; LPG).

M. Arıcı, H. Karabay / Energy and Buildings 42 (2010) 1773–1778 1777

Fig. 6. Effect of the fuel type on the optimum air layer thickness for different cities(Tb = 18 ◦C).

Fig. 7. Effect of the fuel type on the optimum air layer thickness for different cities(Tb = 22 ◦C).

Fig. 8. Variation of the optimum air layer thickness as a function of degree-days.

Fig. 9. Energy saving for fuel type versus air layer thickness for Iskenderun(Tb = 18 ◦C).

Fig. 10. Energy saving for fuel type versus air layer thickness for Ardahan (Tb = 22 ◦C).

air layer thickness is less than 1 mm for the LPG and electricity.However, this effect causes 1.3, 2.4 and 3.2 mm increase in the airlayer thickness for the fuel-oil, coal and natural gas, respectively.

The amount of the energy cost saving (Eq. (19)) as a functionof the air layer thickness for Iskenderun is shown in Fig. 9 for dif-ferent fuel types. The trend of the saving with air layer thicknessshows a maximum value. The magnitude of the saving depends onthe fuel type which is maximum for LPG and minimum for nat-ural gas. Similar results can be drawn for Ardahan, as shown inFig. 10. For instance, referring to Fig. 10, the calculated heating costof the single-glazed window is about 1850$ m−2 and the heatingcost of the double-glazed window at the optimum air layer thick-ness is about 788$ m−2. Consequently, 58% of the heating cost canbe saved by the double-glazed window having air layer thickness of15 mm. Comparing Fig. 9 and Fig. 10, it is seen that the magnitudeof saving in Ardahan increases dramatically for all the kind of fueltypes.

5. Conclusion

The optimum air layer thickness of double-glazed window isdetermined for different climate zones of Turkey considering five

1 nd Bu

dfi

•

•

•

•

•

•

•

•

ca

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

778 M. Arıcı, H. Karabay / Energy a

ifferent fuel types and three different base temperatures. The mainndings of this study are outlined as follows:

The optimum air layer thickness of double-glazed window variesbetween about 12 and 15 mm depending on the base temperatureand fuel type for the climate zones of Turkey.While the effect of fuel type on the optimum air layer thicknessis not significant for cities in cold climate zones such as Ardahan,it is significant for warm cities such as Iskenderun.As DD increases, the effect of the fuel type on the optimum airlayer thickness diminishes for all the fuel types.In cold zones such as Ardahan, the effect of the investment coston the determination of the optimum air layer thickness is notsignificant. In another word, the optimum air layer thickness isdetermined by the heating cost in cold climates.In terms of heating cost, the natural gas is the most suitable fuelfor all the climate zones of Turkey.As the base temperature increases, the optimum air layer thick-ness also increases. While the optimum air layer thickness variesfrom 12.2 to 14.6 mm for Tb = 18 ◦C, it varies from 13.5 to 14.7 mmfor Tb = 22 ◦C regarding all the climate zones of Turkey.The trend of the energy saving shows a maximum at the optimumair layer thickness. The magnitude of the energy saving dependson the fuel type which is maximum for LPG and minimum fornatural gas for all cases. The amount of energy saving in coldzones increases dramatically for all the kind of fuel types.Up to 60% energy saving can be achieved by a well-optimizeddouble-glazed window.

Regarding the results summarized above, the companies shouldonsider the fuel type and the climate condition when deciding their layer thickness of the double-glazed windows.

eferences

[1] http://www.enerji.gov.tr/index.php?dil=tr&sf=webpages&b=enerjiverimliligi&bn=217&hn=&id=587 (25.02.2010).

[2] http://www.eie.gov.tr/turkce/en tasarrufu/konut ulas/en tasarruf bina isi.html (25.02.2010).

[3] http://www.eie.gov.tr/turkce/en tasarrufu/konut ulas/bina ulas.html(25.02.2010).

[4] A. Hasan, Optimizing insulation thickness for buildings using life cycle cost,Applied Energy 63 (1999) 115–124.

[

[[[

ildings 42 (2010) 1773–1778

[5] A. Bolattürk, Determination of optimum insulation-thickness for building wallswith respect to various fuels and climate zones in Turkey, Applied ThermalEngineering 26 (2006) 1302–1309.

[6] A. Bolattürk, Optimum insulation thicknesses for building walls with respectto cooling and heating degree-hours in the warmest zone of Turkey, Buildingand Environment 43 (2008) 1055–1064.

[7] Ö.A. Dombaycı, M. Gölcü, Y. Pancar, Optimization of insulation thicknessfor external walls using different energy-sources, Applied Energy 83 (2006)921–928.

[8] O. Kaynakli, A study on residential heating energy requirement and optimuminsulation thickness, Renewable Energy 33 (2008) 1164–1172.

[9] N. Sisman, E. Kahya, N. Aras, H. Aras, Determination of optimum insulationthicknesses of the external walls and roof (ceiling) for Turkey’s different degree-day regions, Energy Policy 35 (2007) 5151–5155.

10] A. Ucar, F. Balo, Effect of fuel type on the optimum thickness of selected insula-tion materials for the four different climatic regions of Turkey, Applied Energy86 (2009) 730–736.

11] K.A. Al-Sallal, Comparison between polystyrene and fiberglass roof insulationin warm and cold climates, Renewable Energy 28 (2003) 603–611.

12] M.J. Al-Khawaja, Determination and selecting the optimum thickness of insu-lation for buildings in hot countries by accounting for solar radiation, AppliedThermal Engineering 24 (2004) 2601–2610.

13] S.A. Al-Sanea, M.F. Zedan, S.A. Al-Ajlan, Effect of electricity tariff on the optimuminsulation-thickness in building walls as determined by a dynamic heat transfermodel, Applied Energy 82 (2005) 313–330.

14] J. Yu, L. Yang, L. Tian, D. Liao, A study on optimum insulation thicknesses ofexternal walls in hot summer and cold winter zone of China, Applied Energy86 (2009) 2520–2529.

15] Turkish Standard Number 825 (TS 825), 1999. Official Gazette Number 23725(in Turkish).

16] O. Buyukalaca, H. Bulut, T. Yılmaz, Analysis of variable-base heating and coolingdegree-days for Turkey, Applied Energy 69 (2001) 269–283.

17] Y.A. Cengel, Heat Transfer: A Practical Approach, 2nd ed., McGrawHill, NewYork, 2003.

18] S.A. Korpela, Y. Lee, J.E. Drummond, Heat transfer through double pane window,Journal of Heat Transfer 104 (1982) 539–544.

19] O. Aydın, Determination of optimum air-layer thickness in double-pane win-dows, Energy and Buildings 32 (2000) 303–308.

20] O. Aydın, Conjugate heat transfer analysis of double pane windows, Buildingand Environment 41 (2006) 109–116.

21] M.S. Soylemez, Thermoeconomical optimization of number of panes for win-dows, Journal of Energy and Engineering-ASCE 135 (2009) 21–24.

22] G. Weir, T. Muneer, Energy and environmental impact analysis of double-glazed windows, Energy Conversation and Management 39 (1998) 243–256.

23] B. Gücyeter, Z.D. Arsan, A research on energy efficient glass types for retrofittingapplications in Turkey, in: 5th International Conference on Sustainable Energy

Technologies, Vicenza, Italy, August 1–6, 2006.

24] ASHRAE Handbook 2001 Fundamentals, American Society of Heating, Refrig-erating and Air-Conditioning Engineers, Atlanta, GA, 2001.

25] http://www.dosider.org/upload/gaz/Tr/32.xls (25.02.2010).26] www.tcmb.gov.tr (25.02.2010).27] www.die.gov.tr (25.02.2010).