Available online at www.sciencedirect.com

www.elsevier.com/locate/solener

ScienceDirect

Solar Energy 115 (2015) 757–769

Thermophysical properties of Single Wall Carbon Nanotubesand its effect on exergy efficiency of a flat plate solar collector

Z. Said a,b, R. Saidur d,⇑, M.A. Sabiha b, N.A. Rahim c, M.R. Anisur e

a Department of Engineering Systems and Management (ESM), Masdar Institute of Science and Technology, Abu Dhabi, United Arab Emiratesb Department of Mechanical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

c UM Power Energy Dedicated Advanced Centre (UMPEDAC), Level 4, Wisma R&D, University of Malaya, 50603 Kuala Lumpur, Malaysiad Centre of Research Excellence in Renewable Energy (CoRE-RE), King Fahd University of Petroleum and Minerals (KFUPM), Dhahran, 31261,

Saudi Arabiae Department of Mechanical & Aerospace Engineering, Monash University, Clayton, Victoria, 3168, Australia

Received 31 March 2014; received in revised form 18 February 2015; accepted 24 February 2015

Communicated by: Associate Editor Brian Norton

Abstract

In order to enhance thermal efficiency of a flat plate solar collector, the effects of thermo-physical properties of short Single WallCarbon Nanotubes (SWCNTs) suspended in water was investigated in this study. Sodium dodecyl sulphate was used as a dispersantfor preparing a stable nanofluid. Subsequently, the nanofluid was comprehensively characterized by particle size measurement and spec-troscopic technique. Specific heat with the increase of particle loading and temperature was investigated. Thermal conductivity incrementalso showed a linear dependence on particle concentration and temperature. Viscosity of the nanofluids and water reduced with theincrease of temperature and increased with the particle loading. Using improved thermo-physical properties of the nanofluid, the maxi-mum energy and exergy efficiency of flat plate collector was enhanced up to 95.12% and 26.25% compared to water which was 42.07%and 8.77%, respectively. This low exergy efficiency shows that flat plate collectors still require substantial enhancement. To the authors’knowledge, SWCNTs–H2O was used as the functioning fluid for the first time to investigate both the thermos-physical properties as wellas the increase in thermal efficiency of a flat plate solar collector.� 2015 Elsevier Ltd. All rights reserved.

Keywords: Specific heat; Thermal conductivity; Viscosity; Nanofluid; Exergy; SWCNTs

1. Introduction

Nanofluids are new addition to the family of fluids pre-pared by immersing nanoparticles in conventional fluidssuch as water, oils, ethylene glycol or coolants. In general,

http://dx.doi.org/10.1016/j.solener.2015.02.037

0038-092X/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Centre of Research Excellence in RenewableEnergy (CoRE-RE), King Fahd University of Petroleum and Minerals(KFUPM), Dhahran, 31261, Saudi Arabia. Tel.: +966 13 860 4628; fax:+966 13 860 7312.

E-mail addresses: saidur@kfupm.edu.sa, saidur912@yahoo.com(R. Saidur).

these nanoparticles used in nanofluids are metals, metaloxides or carbon nanotubes (CNTs), in diverse allotropicforms. Choi et al. (2001) first reported studies on nanoflu-ids and also explored the potentials of these nanofluids,precisely in heat conduction applications. With regards tothermal engineering applications, enhancement of upto60% in thermal conductivity for water based nanofluidswas reported in literature (Keblinski et al., 2008; Yuet al., 2008).

One of the utmost extraordinary findings of the last dec-ade are carbon nanotubes (CNTs) (Iijima and Ichihashi,

Nomenclature

Ac collector area (m2)Cp specific heat (J/kg K)d diameter of pipe (m)_Exin exergy rate at inlet (W)Gc global solar irradiationD difference_Exout exergy rate at outlet (W)_Exdest rate of irreversibility (W)_Exheat exergy rate received from solar radiation (W)_Exwork exergy output rate from the system (W)_Exmass;in exergy rate associated with mass at inlet (W)_Exmass;out

exergy rate associated with mass at outlet (W)_Sgen entropy generation rate (W/K)_Qsun;in energy gain rate (W)s shear stressI intensity of solar radiation (W/m2)Pnf nanofluidkp thermal conductivity of nanoparticle (W/m K)K loss coefficient (dimensionless)_m mass flow rate (kg/s)_W work rate or power (W)

g collector efficiencyP fluid pressure (Pa)q convective heat transfer rate (W)k thermal conductivity (W/m K)

_Qo heat loss rate to the ambient (W)_Qs energy rate engrossed (W)Ta ambient temperature (K)R ideal gas constant (J/K mol)hin specific enthalpy at inlet (J/kg)hout specific enthalpy at outlet (J/kg)l coefficient of viscosityTout output temperature (K)bf basefluidTs sun temperature (K)Tsur surrounding/ambient temperature (K)M viscosity (N s/m2)s transmittance coefficient of glazingF absorptance coefficient of platesa effective product transmittance–absorptanceU nanoparticles volume fraction (%)sa entropy generation to surrounding (J/kg K)sin entropy generation at inlet (J/kg K)sout entropy generation at outlet (J/kg K)Pm density (kg/m3)R overall entropy production (J/kg K)F friction factorH specific enthalpy (J/kg)_c shear strain rateTin input temperature (K)P nano particle

758 Z. Said et al. / Solar Energy 115 (2015) 757–769

1993; Choi et al., 2001). Depending on their structure, theyhave several unusual properties, such as high electrical andthermal conductivities. In particular, thermal properties ofCNTs have attained a great deal of dedication (Tans et al.,1997; Saito et al., 1998; Mizel et al., 1999; Hone et al., 2000;Zhang et al., 2003; Wen and Ding, 2004; Duong et al.,2008; Sun et al., 2008; Harish et al., 2012). Both experimen-tally and numerically high thermal conductivities of CNTshave been reported in literature (Berber et al., 2000; Kimet al., 2001; Maruyama, 2003; Yu et al., 2005; Pop et al.,2006). Therefore, CNTs are naturally expected to havehigher thermal conductivity enhancements in nanofluidsas compared to other nanoparticles. However, this unusualincrease could not be supported by consequent studiesreported (Xie et al., 2003; Wen and Ding, 2004; Assaelet al., 2005; Liu et al., 2005; Ding et al., 2006; Garget al., 2009). Since unique mechanical, electrical and struc-tural properties are possessed by Single Wall CarbonNanotubes (SWCNT), they have attracted the attentionof the researchers (Dresselhaus and Avouris, 2001;Baughman et al., 2002). SWCNT possesses outstandingthermal and chemical stabilities with high-tensile strengthand extremely light weight (Jha and Ramaprabhu, 2012).The specific heat of SWCNT has also been investigatedby several researchers (Mizel et al., 1999; Hone et al.,2000; Zhang et al., 2003; Pradhan et al., 2009). Most of

the presented reports in literature are focused on multiwalled carbon nanotubes (MWCNTs), whereas limitedstudies are found to be conducted on thermo-physicalproperties of SWCNTs based nanofluid.

Several researchers have proposed different techniquesand models for obtaining stable nanofluid suspensions(Li et al., 2007; Jiang et al., 2010; Ghadimi et al., 2011;Said et al., 2013; Sajid et al., 2014). Some importantparameters such as the length of the CNTs, the purity level,preparation method and pH of the solution andthermophysical properties should be known in order tomake a direct comparison between experimental and theo-ritical results. In this context, stable suspension ofSWCNTs based nanofluid using SDS surfactant was pre-pared to get more accurate results.

Heat transfer enhancement using nanofluids in solarthermal collectors is one of the main issues in savingenergy. Several studies related to nanofluids and its usesin solar collectors are reported (Link and El-Sayed, 2000;Kameya and Hanamura, 2011; Mercatelli et al., 2011a,b,2012; Sani et al., 2011; Saidur et al., 2012). Tyagi et al.(2009) reported the efficiency enhancement for low valuesof the volume fraction of nanoparticles. However, for avolume fraction higher than 2%, the efficiency stayed close-lypersistent. Otanicar et al. (2010) found that the additio-nof a little amount of nanoparticles enhanced the

Z. Said et al. / Solar Energy 115 (2015) 757–769 759

efficiency until a volume fraction of approximately 0.5%.However, further addition of volume fraction levels off oreven slightly reduced the efficiency. Taylor et al. (2011)showed an efficiency enhancement of 10% using nanofluids.He et al. (2011) reported that the CNT–H2O nanofluidaremore suitableas compared to TiO2–H2O in a vacuum tubesolar collector. Yousefi et al. (2012c) reported an energyefficiency of 28.3% with 0.2 wt.% as compared to water ina flat plate solar collector. Yousefi et al. (2012b) with thesimilarsetup reported an efficiency enhancement of 35%with 0.4 wt.% of MWCNT-H2O nanofluid. Again, withsame setup in Yousefi et al. (2012a, 2012c) studied theeffects of pH variation of the MWCNT–water nanofluidon the efficiency of the flat plate collector. Tiwari et al.(2013) showed an efficiency improvement of 31.64% usingAl2O3 nanofluid in flat plate solar collector.

Studies using SWCNTs based nanofluid as a workingmedium was not reported anywhere in literature. In thisstudy, SWCNTs was characterized using TEM, ZetaSiezer, UV–Vis spectroscopy as well as visual recordings.Thermal conductivity with respect to different volume frac-tion using a KD2 pro was measured. Specific heat and vis-cosity were measured as well. The energy and exergyefficiencies of a flat plate solar collector using SWCNTsbased nanofluid are examined experimentally to evaluatethe performance enhancement.

2. Theoretical background

Theoretical studies on energy and exergy analyses arereported below in the sub sections.

2.1. Energy analysis

The thermal efficiency of the flat plate solar collector (g)is defined in Eq. (1) (Sukhatme, 2008).

g ¼ _mCpðT out � T inÞ=IAc ð1Þ

2.2. Entropy analysis

In this analysis, the system is assumed to be steady flowand steady state operation. Work transfer from the systemand heat transfer to the system are also considered positive.Loss coefficient is only considered for the entrance effect. Ifthe influences of potential and kinetic energy deviations areignored, the typical exergy stabilities can be expressed inthe rate form as in Eq. (2) (Ucar and Inallı, 2006).

_Exheat � _Exwork � _Exmass;in � _Exmass;out ¼ _Exdest ð2ÞThe rate of the general exergy balance can also be com-posed as in Eq. (3).

X1� T a

T sur

� �_Qs� _m½ðhout�hinÞ�T aðsout� sinÞ� ¼ _Exdest

ð3Þ

Solar energy _Qs is the energy absorbed by the collectorabsorber surface (Esen, 2008). The enthalpy and entropy

deviations of the nanofluid in the collector are expressedin Eq. (4) (Ucar and Inallı, 2006).

1� T a

T sur

� �IT ðsaÞAc � _mCp;nf ðT f ;out � T f ;inÞ

þ _mCp;nf T a lnT f ;out

T f ;in� _mRT a ln

P out

P in¼ _Exdest ð4Þ

The exergy destruction (or irreversibility) rate, _Exdest can besincerely appraised from the subsequent Eq. (5).

_Exdest ¼ T a_Sgen ð5Þ

As reported by Bejan (1996), in a non-isothermal solar flatplate collector, the overall rate of entropy generation canbe written as in Eq. (6).

_Sgen ¼ _mCp lnT out

T in�

_Qs

T sþ

_Qo

T að6Þ

In order to measure the total heat loss to the ambient, Eq.(7) can be used.

_Qo ¼ _Qs � _mCpðT out � T inÞ ð7Þ

Finally, the exergy efficiency is calculated from Eq. (8).

gex ¼ 1� T a_Sgen

½1� ðT a=T sÞ� _Qs

ð8Þ

3. Materials and methods

3.1. Materials and data collection

Short SWCNTs (90% CNTs, 60% SWCNTs) of length1–3 lm and diameter 1–2 nm were purchased fromNanostructured & Amorphous Materials, Inc, USA.Sodium dodecyl sulphate (SDS, 92.5–100.5%, Sigma–Aldrich) as surfactants, was used. Distilled water was usedas a base fluid.

TEM was used to characterize SWCNTs nanoparticles.A Zeta-seizer Nano ZS (Malvern) was used to obtain theaverage diameter of the nanoparticles immersed in the basefluids. DLS (dynamic light scattering) approach is used togive the hydrodynamic radius of the particles in solution.Mettler toledo pH meter was used to measure the pH ofthe solution. The Density Meter DA-130 N from KyotoElectronics is used to measure the density of the nanofluids.Viscosity of nanofluid was measured using Brookfield vis-cometer (DV-II + Pro Programmable Viscometer) whichwas connected with a temperature-controlled bath.

3.2. Specific heat

In the measurement of thermo-physical properties, theterm “specific” means the measure is an intensive property,wherein the quantity of substance must be specified. Forspecific heat capacity, mass is the specified quantity (unitquantity). The specific heat capacity determines the convec-tive flow nature of the nanofluid, and it necessarily depends

760 Z. Said et al. / Solar Energy 115 (2015) 757–769

on the volume fraction of the nanoparticles. Consideringthe fact that very limited experimental data on specific heatcapacity values for various water-based nanofluids at dif-ferent concentrations are available, the value of the specificheat capacity is estimated using theoretical models. A heat-flux-type Differential Scanning Calorimeter (PerkinElmer’sDSC 4000) was used to measure the nanofluid specific heatcapacities. The Differential Scanning Calorimeter (DSC)measures the heat flux into a sample as a function of tem-perature for a user-prescribed heating regime. The classicalthree-step DSC procedure was followed to measure specificheat capacity (Hohne et al., 2003). Therefore, this instru-ment is used to measure the experimental values of waterbased nanofluids. The specific heat capacity of nanofluids,calculated at any particle concentration, which is valid forhomogeneous mixtures (Syam Sundar and Sharma, 2008),is given by:

CPnf ¼ð1� /ÞðqCP Þbf þ /ðqCP ÞPð1� /Þqbf þ /qP

ð9Þ

3.3. Thermal conductivity

The thermal conductivities of the tested nanofluids wereextracted by the ‘k’ module which contained a DecagonDevice KD2 Pro thermal property analyser. Equipped withthe optional KS-1 transient hotwire sensor, capable ofreading a fluid’s thermal conductivity from �323 K to423 K with a maximum deviation of 5.0% reported andwas tested for accuracy under the experimentation parame-ters. The sensor’s stem was vertically inserted in a jarthrough the lid of a small container filled with USP glycer-ine, which was in turn completely submerged in thePolyscience Circulating Water Bath. Applying this method,it was possible to accurately test the KD2 Pro at tempera-tures ranging from 298 K to 323 K. Thermal conductivityreadings were found to be within 0.3% deviation from cali-bration values until 323 K. Above this temperature, natural

Fig. 1. The experimental setup used for this study: (a) front vie

heat flux in the glycerine caused micro-convection currentsto affect the hotwire stem surface and amplify the readingsby 4.2%, still within the acceptable ±5.0% tolerance.

3.4. Experimental procedure

The experimental set up of the solar collector and theschematic diagram of the experiment are presented inFigs. 1 and 2 respectively. The dimensions of the solar col-lectors are listed in Table 1. The experiment was carried outat University Malaya, Malaysia. The collector position wasfixed at 22� angle, for the maximum solar radiation absorp-tion. Table 1 presents the specifications of the flat platesolar collector that are considered in this experiment. Forthe force convection system, an electric pump is used inthe solar water heating system. A radiator is used for cool-ing the water inlet temperature. It is shown in Fig. 2 thatthe tank which has a capacity of 50 L absorbs the heat loadfrom the collector cycle. All the data were later transferredinto the computer via interfaces. Calibration of the entiresystem was taken several times.

ASHARE Standard 93-2003 (Standard, 1977) is used toevaluate the thermal performance of the flat-plate solar col-lector. The flow rates of 0.5, 1.0 and 1.5 kg/min are used totest the flat plate solar collector.

3.5. Error analysis in measurements

Two groups of errors are reported in our measurements.One group could come from the direct measurementparameters such as solar radiation flux (DGc), DT, DP

and the second group of errors could come from the indi-rect measurements, such as energy and exergy efficiencies.The following relations can be used based on theLuminosu and Fara (2005) method:

Dgex ¼D_I

_Exheat

þ_I _Exheat

_Ex2heat

ð10Þ

w, (b) back view, (c) left side view and (d) right side view.

Fig. 2. Schematic presentation of the experimental set up.

Table 1Specifications for the flat plate solar collector.

Parameters of collector Value

Frame Aluminum alloyGlazing 4 mm tempered texture glassWorking fluids in flow ducts Water and SWCNTs based nanofluidAbsorption area, Ap 1.84 m2

Wind speed 2–4 m/sCollector tilt, bo 22�Absorption rate 0.94Emittance 0.12Heat transfer coefficient 4.398Header material Copper TP2Header tube size £22 mm � t0.6 mm (2 pcs)Riser tube material Copper TP2Riser tube size £10 mm � t0.45 mm (8 pcs)

Z. Said et al. / Solar Energy 115 (2015) 757–769 761

and

Dgen ¼D _qa

Gcþ _qaDGc

G2c

ð11Þ

where each error component can be evaluated through thefollowing relations:

DExheat ¼DTT sþ T aDT

T 2s

� �AcðsaÞGc þ 1� T a

T s

� �AcðsaÞDGc

ð12Þ

D_I ¼ T aD _Sgen þ _SgenDT ð13Þ

_DSgen ¼ R lnP out

P inþ Cp ln

T in

T outþ Cp

T out þ T in

T a

� �D _m

þ GcAcðsaÞDT

T 2a

þ _mCp1

T outþ 1

T inþ 2

T aþ ðT out þ T inÞ

T 2a

� �DT

þ _mR1

P outþ 1

P in

� �DP

þ AcðsaÞ1

T sþ 1

T a

� �DGc ð14Þ

where Pin and Pout are the pressure difference of the agentfluid with the surroundings at entrance and exit of thesolarcollector.

D _qa ¼ CpD _mðT out þ T inÞ þ 2 _mDT

Ac

� �ð15Þ

The total uncertainties of the measurements are estimatedto be ±3.0% for solar radiation, ±1.60% for the nanofluidand water temperatures, ±3.32% for pressures and±3.02% for massflow rate. Therefore, the maximumerrors for the indirect measuring of energy and exergyefficiencies were estimated to be ±0.1 and ±0.14 usingEqs. (10) and (11).

762 Z. Said et al. / Solar Energy 115 (2015) 757–769

4. Results and discussion

4.1. Nanofluid characterization and stability

The highly hydrophobic (tending to repel or fail to mixwith water) nature of SWCNTs makes very hard to dis-perse them in water. Preparation of a stable and homoge-nous dispersion is a vital prerequisite for a nanofluid. Inthis present work, 0.1 and 0.3 vol.% dispersant was usedto prepare the nanofluid suspension. Fig. 3 shows thevisualization of SWCNTs using transmission electronmicroscope (TEM), whereas Fig. 4 shows SWCNTs nano-fluid with SDS after a period of 30 days. Chemical struc-ture of Sodium dodecyl sulphate is presented in Fig. 5.

Stable nanofluid suspension was prepared by addingnecessary loading of SWCNTs. For this pupose, theSWCNT and SDS density was considered to be 2.1 g/cm3

and 1.01 g/cm3 respectively. Ratio of 1:1 was employed

Fig. 3. TEM visualization of SWCNTs (length 1–3 lm and diameter1–2 nm) nanoparticleimage captured at an acceleration on 200 kV.

Fig. 4. TEM visualization of SWCNTs nanofluid with SDS image after30 days.

Fig. 5. Chemical structure of Sodium dodecyl sulphate (NaC12H25SO4).

for the SWCNT nanoparticles and SDS. The dispersionswere subjected to a tip sonication using an ultrasonic pro-cessor for 1:30 h. Same sonication conditions were used forthe samples of different volume concentrations. It was alsonoticed that about 4–5% of the volume was lost during thetip sonication, and the losses were taken into account dur-ing the preparation of the nanofluid solutions.

SDS-dispersed SWCNTs were further characterizedusing UV–Vis spectroscopyup to the range of 1100 nm,presented in Fig. 6. The nanofluid solutions were dilutedto perform the measurements with the base fluid. Fig. 6shows a typical absorption spectrum obtained fromSWCNTs dispersed in water using SDS. Sharp peaks arewitnessed in the absorption spectrum, which are mainlydue to the characteristic of isolated nanotubes.

The visual observations of sedimentation of SWCNTsare illustrated in Fig. 7. The samples with different volumefraction were poured into transparent cells right after pre-paration. The visual images of the 1st day and the 30thdays are presented and show no visible sedimentation.

The pH of the (SWCNT + SDS)/water nanofluid for0.1 vol.% was obtained to be 7.0. The nanofluids remainedhighly stable showing no visible signs of sedimentationeven after 1 month of incubation.

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

300 400 500 600 700 800 900 1000 1100

Abs

orba

nce

Wavelength (nm)

Fig. 6. UV–Vis absorbance spectrum of SWCNTs dispersed in waterusing SDS.

Fig. 7. Prepared SWCNTs nanofluid solutions (a) samples on the first dayof preparations and (b) samples after 30 days of preparations.

80

100

120

140

160

180

200

0 5 10 15 20 25 30

Dia

met

er (n

m)

Number of Days

SWCNTs+SDS (0.1 vol. %)

Fig. 9. Diameter in relation to the time elapsed from the day ofpreparation, for water containing 0.1 vol.% SWCNTs + 0.1 vol.% SDS.

Z. Said et al. / Solar Energy 115 (2015) 757–769 763

4.2. Particle size measurements

SWCNTs–water nanofluids with SDS as dispersant weretested in several concentrations. An anionic dispersant waschosen based on reports present in literature (Sun et al.,2008). The investigated fluids were as follows.

Water +0.1 and 0.3 vol.% SDS at 0.1 and 0.3 vol.%SWCNTs, respectively.

Fig 8 presents the mean particle diameters as a functionof time for the nanofluid in static mode.

In order to improve the stability of this suspension,same SDS: SWCNTs ratio was tested. The result is shownin Fig. 9. Here, the suspension contains 0.1 vol.% ofSWCNTs and shows a very stable behavior for 30 days,keeping a mean diameter of about 130 nm.

The measured zeta potential was around �42 mV asanticipated in the case of anionic dispersant (Li et al.,2007; Sun et al., 2008) for all the studied SWCNTs-nano-fluids, supporting their non-aggregating tendency. Owingto the strong opacity of the solutions at 0.1 and0.3 vol.%, they were diluted with distilled water to performthe zeta potential measurements. Fig. 10 presents the effectsof zeta potential on the stability of suspension properties.

In conclusion, based on the higher zeta potential value(�42 mV) and maintaining a mean diameter of about130 nm for 30 days, proved the water-based nanofluidscontaining SWCNTs and SDS as a very stable nanofluid.Further investigation on their properties is underway.

Table 2 presents the experimentally measured thermo-physical properties.

4.3. Specific heat

Fig. 11 shows the specific heat of SWCNTs nanoparticlewith respect to changing temperature.

The specific heat of SWCNTs compared to water ismuch lower. Therefore, higher volume fraction ofSWCNTs in water could result in greater reduction in thespecific heat of SWCNTs based water nanofluid. Thisexperimental data agree well with data reported by other

0123456789

10

1 10 100 1,000 10,000

Inte

nsity

(%)

Size (nm)

0.1 vol. % SWCNTs 0.3 vol. % SWCNTs

Fig. 8. Particle size distribution (Z-average = 139.3 d nm) after 3 dayspreparation of sample (0.1 vol.% SWCNTs + 0.1 vol.% SDS)/distilledwater and particle size distribution (Z-average = 135.5 d.nm) afterpreparation of sample (0.3 vol.% SWCNTs + 0.3 vol.% SDS)/distilledwater.

researchers (Hone et al., 2000; Zhang et al., 2003;Pradhan et al., 2009).

In Fig. 12, the result indicates that the specific heatcapacity of SWCNTs nanofluid decreases gradually withincreasing volume concentration of nanoparticles. It isobserved that the specific heat reduces gradually with theincreasing temperature. For 0.3 vol.% of SWCNTs nano-fluid, it is observed that the specific heat reduces by a largemargin compared to 0.1 vol.% of SWCNTs particle load-ing. From Fig. 12, it is noted that, the specific heat dropsalmost linearly until 331 K of temperature and then a sud-den rise is observed in the specific heat beyond this tem-perature. After this point, no further drop is noticed,suggesting the critical point at which the boiling starts totake place and therefore, showing inconsistency in experi-mental data. The reason behind the critical point is theincreased thermal conductivity of SWCNTs based nano-fluidwhich is supported by the findings presented in ther-mal conductivity section. Therefore, by the increase innanoparticle fraction, the portion of heat absorption withthese lower specific heat nanoparticles is increased and leadto the decreasing in nanofluid specific heat. Eq. (9), cannotbe used to predict the tendency for fluid with nanoparticleinclusions. The decline in specific heat for 0.3 vol.% ofSWCNTs nanofluid cannot be explained by the theoreticalmodel. This may be due to the reason that Eq. (9) does nottake temperature into account. Hence further studies needto be carried out to provide models that can explain thissudden decline in specific heat of nanofluids with respectto temperature.

Qualitatively, the solid-liquid interface may change thephonon vibration mode near the surface area of a nanopar-ticle and thus change the specific heat capacity of nano-fluid. The high specific interfacial area of nanoparticlecan adsorb liquid molecules to its surface and form liquidlayers, which will reversely constrain nanoparticle andturns its free-boundary surface atoms to be non-free inter-ior atoms (Wang et al., 2006). Specific heat obtained fromthe experimental measurements is used for our study.

4.4. Thermal conductivity

Thermal conductivity of SWCNTs/water nanofluid withvolume fractions of 0.1 and 0.3 vol.% was experimentally

Table 2Experimental thermo-physical properties of SWCNTs/water and base fluid at room temperature.

Particle and base fluid Average particlesize (nm)

Actual density(kg/m3)

Cp (J/kg K) K (W/mK) Viscosity(m Pa s)

pH

SWCNTs D = 1–2 nm 2100 600 �3500 (Pop et al., 2006)L = 1–3 lm

Water 998.8 4179 0.605 0.89SWCNTs/water (0.1%v/v SWCNTs + 0.1%v/v SDS) 1007 4104 0.651 7.0SWCNTs/water (0.3%v/v SWCNTs + 0.3%v/v SDS) 1024 3845 0.691 7.0

3.100

3.300

3.500

3.700

3.900

4.100

4.300

4.500

280 290 300 310 320 330 340 350

Spec

ific

Hea

t (J

/g*°

C)

Temperature (K)

Specific Heat (J/g*°C) of Water Specific Heat (J/g*°C) of 0.1 vol. % SWCNT

Specific Heat (J/g*°C) of 0.3 vol. % SWCNT

Fig. 12. Specific heat of SWCNTs based nanofluids with increasingtemperature and volume fraction.

450

500

550

600

650

700

750

273 283 293 303 313 323 333 343 353 363

Spec

ific

Hea

t (J/

g*°C

)

Temperature (K)

Specific Heat (J/g*°C) of SWCNTs

Fig. 11. Specific heat of SWCNTs with respect to temperature.

Fig. 10. Effects of zeta potential on suspension properties.

764 Z. Said et al. / Solar Energy 115 (2015) 757–769

measured. Fig. 13 shows the thermal conductivity versusdifferent SWCNTs volume fractions measured at differenttemperatures. Thermal conductivity enhanced with theincreasing volume fraction of SWCNTs in a linear fashion.

Fig. 13 is showing the difference between the thermalconductivity of SWCNTs/water nanofluid and the experi-mental data of Harish et al. (2012) with respect to water.SWCNTs/water nanofluid showed higher thermal conduc-tivity enhancement compared to both water and Harishet al. (2012) data. A linear increase is found forSWCNTs/water nanofluid. The SWCNTs form a saturat-ing network which results in an improved energy transport

thereby increasing the effective conductivity of the fluid.The thermal conductivity enhancement witnessed in thecurrent investigation supports the mechanism of particleclustering in improving the thermal conductivity of thefluid (Harish et al., 2012). It needs to be pointed out thatthe thermal conductivity enhancement stayed nearly thesame for one month.

Fig. 14 presents the effective thermal conductivity fortwo volume fractions at different temperatures. Additional

0.55

0.65

0.75

0.85

0.95

1.05

1.15

1.25

295 300 305 310 315 320 325

The

rmal

con

duct

ivit

y (W

/m.k

)

Temperature (K)

water Harish et al. (2012) SWCNTs+SDS (0.1 vol. %) SWCNTs+SDS (0.3 vol. %)

Fig. 13. Temperature-dependent thermal conductivity in (SWCNTs +SDS)/water nanofluid.

1.051.151.251.351.451.551.651.751.851.95

295 300 305 310 315 320 325The

rmal

Con

duct

ivity

Rat

io (K

eff/K

f)

Temperature (K)

SWCNTs+SDS (0.1 vol. %) SWCNTs+SDS (0.3 vol. %)

Fig. 14. Thermal conductivity increase as a function of fluid temperaturein water.

1.00

1.10

1.20

1.30

1.40

1.50

1.60

295 300 305 310 315 320 325The

rmal

Con

duct

ivity

Rat

io (K

eff/K

f)

Temperature (K)

Heating Up Cooling Down

Fig. 15. Comparisons of thermal conductivity improvement during theheating and cooling process in water (SWCNTs: 0.1 vol.%).

Z. Said et al. / Solar Energy 115 (2015) 757–769 765

increase in effective thermal conductivity is observed withthe increasing temperature. A maximum conductivityenhancement of 91% is obtained at a temperature of323 K for 0.3 vol.% volume fraction, whereas, the minimumenhancement in conductivity is 12% for 0.1 vol.% volumefraction of SWCNTs at 298 K in comparison with water.

Nanoparticles tend to aggregate, with the period oftime, which aretemperature dependent and tend to increasewith the growing size of the aggregates as a substantialamount of time is frequently consumed to heat the fluidduring measurements (Gharagozloo and Goodson, 2010).Effective thermal conductivity with respect to heating andcooling is measured in order to examine this mechanismin Fig. 15.

From Fig. 15, it is evident that the fluid effective thermalconductivity remains the same for both the heating andcooling phase. Hysteresis effect was not observed forSWCNTs/water, which therefore does not support “thetime dependent aggregation” (Gharagozloo andGoodson, 2010; Harish et al., 2012), as a possible mecha-nism for the temperature dependent thermal conductivityenhancement.

The effective conductivity enhancement remains thesame with respect to temperature irrespective of whetherthe fluid is heated or cooled with minor errors. The

agglomeration of nanoparticles and the formation of clus-ters can increase the thermal conductivity. Increment inthermal conductivity was reported to be 7% for a volumefraction of 1% of MWCNT based nanofluid (Xie et al.,2003). Another researcher reported an increment of about40% at a volume fraction of 0.6% of MWCNTs basednanofluid at room temperature (Assael et al., 2005). 80%enhancement was reported by Ding et al. (2006) for1 wt.% of MWCNTs at a temperature of 303 K. Nasiriet al. (2012) reported an enhancement of 35% at a tempera-ture of 323 K using 0.25 wt.% of MWCNTs. Our resultsare supported by these findings.

4.5. Viscosity

The difference in temperature-dependent thermal con-ductivity variation could be a possible indication of thecritical role of Brownian motion in the fluid. Gupta andKumar (2007) reported an enhancement of 6% in thermalconductivity at higher temperatures. Due to improved diffu-sion of heat walkers enhanced thermal conductivity wasreported (Duong et al., 2008). Translational diffusion coef-ficient of SWCNTs/water was reported to be much lowercompared to water, ranging from 0.3 to 6 lm2/s(Tsyboulski et al., 2008). Broersma theory is used to esti-mate the rotational diffusion (Dr) of SWCNTs (Broersma,2004; Tsyboulski et al., 2008; Harish et al., 2012).

Dr ¼3kB

pglnðL=dÞ � c

L3ð16Þ

In Eq. (16), L and d denote the length and diameter of thenanotube, respectively. kB is the Boltzmann constant, T isthe fluid temperature, g is the fluid viscosity and c is theend correction coefficient (usually c is assumed to be0.83). Eq. (16) shows that the rotational diffusion is inver-sely proportional to the cube length of the SWCNTs.Viscosity with respect to changing volume fraction andtemperature is presented in Fig. 16. It is observed fromFig. 16, with the increasing temperature, the viscosity ofthe fluid decreases. The decrease in the fluid viscosity

0.400.500.600.700.800.901.001.101.201.301.40

295 300 305 310 315 320 325 330

Vis

cosi

ty

Temperature (K)

Water 0.1 vol. % SWCNTs+0.1 vol. % SDS 0.3 vol. % SWCNTs+0.3 vol. % SDS

Fig. 16. Viscosity of SWCNTs/water nanofluid with respect to volumefraction and temperature.

0100200300400500600700800900

100011001200

9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00

Sola

r R

adia

tion

(W/m

²)

Time of Day (hour)

Solar insolation on a clear day

Solar insolation on a cloudy day

Fig. 17. Solar insolation recordings on a clear and cloudy day forexperimental study.

766 Z. Said et al. / Solar Energy 115 (2015) 757–769

improves the rotational diffusion of SWCNTs. As men-tioned above, the given length of the SWCNTs used forthis study is from 1 to 3 lm, with a diameter of 1–2 nm.Given Eq. (16) and these parameters, it is therefore, possi-ble to conclude that the enhancement in the thermal con-ductivity is due to the presence of shorter SWCNTs,resulting in higher rotational diffusion.

With the increasing temperature, the viscosity of thenanofluids and water, both reduced. An increase of 39%in viscosity isobserved for SWCNTs and SDS suspensionat a volume fraction of 0.3 vol.%. An increase in viscosityis observed with the increasing volume fraction ofSWCNTs and SDS. This strong increase in viscosity willhave adverse effects in practical applications of suchnanofluids.

4.6. Energy and exergy efficiencies using SWCNTs based

nanofluid

4.6.1. Entropy generation and exergy destruction

Entropy is produced in irreversible processes. Therefore,for the energy optimization analysis, it is essential to mea-sure the entropy generation or exergy destruction due toheat transfer and viscous friction as a function of thedesign variables selected (Onsager, 1931b,a; Kreuzer,1981). Fig. 17 provides the solar insolation data recordedfor a clear and cloudy day used for the performance mea-surement of the solar collector.

Experimental data of solar water heating systems withand without nanofluids are various days and different flowrate is provided in Table 3.

Fig. 18 presents the entropy generation and exergydestruction with respect to mass flow rate and different vol-ume concentrations of water and nanofluid. Eqs. (5) and(6) are used to obtain exergy destruction and entropygeneration, respectively.

As shown in Fig. 18, the entropy generation is reducedup to 32.21 M/K for 0.1 vol.% SWCNTs, for a mass flowrate of 0.5 kg/min. For 0.3 vol.% the entropy generationis reduced to 37.51%, for a mass flow rate of 0.5 kg/min,whereas, for water with similar mass flow rate, the reduc-tion in entropy generation is 43.53 M/K. Therefore, from

the obtained results, it can be said that the entropy canbe reduced with the least volume fraction of SWCNTsused, compared to higher volume fractions.

The other axis of Fig. 18 illustrates the exergy destruc-tion with respect to mass flow rate and changing volumefraction. Similar behavior as that of entropy generation isobserved for exergy destruction as well. With 0.1 vol.% ofSWCNTs and a mass flow rate of 0.5 kg/min, the exergydestruction reduced to 1037.11 W. For 0.3 vol.% ofSWCNTs and the similar mass flow rate as of earlier case;the exergy destruction reduced to 1200.39 W. In case ofwater, the lowest exergy destruction was observed1423.69 W for a mass flow rate of 0.5 kg/min. From theseobservations, SWCNTs based water with as low as0.1 vol.% is very useful in reducing the entropy generationand exergy destruction.

4.6.2. Effect of SWCNTs on the output temperature

Fig. 19 demonstrates the effect of mass flow rate andvolume fraction on output temperature. As recognized,the output temperature is one of the most effectiveparame-ter that affects the energy efficiency of a flat plate solar col-lector directly. It is increased intensely with the growingoutput temperature.

As illustrated in Fig. 19, a greater difference between thewater inlet temperature and ambient temperature results inan enhanced exergy efficiency of the flat plate solar collec-tor. This enhancement is due to the increasing temperatureof the absorber’s plate along with rising inlet water tem-perature. The main reason of exergy loss in a collectoristhe difference between the temperature of the solar radia-tion and the absorber plate temperature, since the risingtemperature of the absorber flat, results in a higher differ-ence and subsequently reduced collector exergy loss.

4.6.3. Energy and exergy efficiencies

To ensure the best results with least error, eachinvestigation was repeated for several days. Different massflow rates and changing volume fraction of nanoparticlesare used to present the energy efficiency and exergy

Table 3Experimental data of the solar water heating system with and without nanofluids at various days.

Local time (h) Volumeconcentration (%v/v)

Solarradiation (W/m2)

Water temperature (�C) Mass flowrate (kg/min)

Ambienttemperature (�C)

Windvelocity (m/s)Inlet Outlet

12:30 Water + 0.1% of SWCNTs 730.2 44.2 58.9 0.5 33.3 2.7513:00 834.2 48.1 59.7 0.5 33.9 2.6413:30 851.4 48.4 65.3 0.5 34.2 3.5514:30 885.4 48.4 67.7 0.5 35.1 3.3812:50 981.1 40.2 48.1 1.5 33.1 2.6613:30 985.2 40.8 48.8 1.5 33.9 2.7514:00 992.4 40.9 47.9 1.5 33.6 3.25

12:30 Water only 716.3 43.4 49.2 1.0 33.1 3.0013:00 759.8 44.2 49.9 1.0 33.3 3.2513:30 765.8 48.5 54.6 1.0 33.4 3.38

0

5

10

15

20

25

30

20

30

40

50

60

70

80

90

100

0.5 0.7 0.9 1.1 1.3 1.5

Exer

gy e

ffici

ency

, %

Ener

gy e

ffici

ency

, %

Flow rate, kg/min

0.30% 0.10% Water 0.10% 0.30% Water

Fig. 20. The energy efficiency (solid lines) and exergy efficiency (dottedlines) at different mass flow rates and different volume fractions forSWCNTs based nanofluid.

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

30333639424548515457

0.5 0.7 0.9 1.1 1.3 1.5

Exer

gy d

estru

ctio

n, W

Entro

py g

ener

atio

n, M

/K

Flow rate, kg/min

0.30% 0.10% Water Water 0.10% 0.30%

Fig. 18. Entropy generation (dotted line) and exergy destruction (solidline) for SWCNTs based nanofluid at different mass flow rate and volumefraction.

45

49

53

57

61

65

69

73

77

0.5 0.7 0.9 1.1 1.3 1.5

Flow rate, kg/min

0.30%

0.10%

Water

Fig. 19. SWCNTs based nanofluid at different mass flow rate and volumefraction and its effect on output temperature.

Z. Said et al. / Solar Energy 115 (2015) 757–769 767

efficiency of the solar collector in Fig 20. This efficiency wasevaluated using Eqs. (10)–(14) and input Tables 2 and 3.

As shown in Fig. 20 and 0.3 vol.% of SWCNTs and amass flow rate of 0.5 kg/min enhanced the energy efficiencyup to 95.12%, whereas for 0.1 vol.% of SWCNTs theenergy efficiency improved up to 89.26%, for the same massflow rate. The highest energy efficiency record for waterwas 42.07% for a mass flow rate of 0.5 kg/min. The otheraxis in Fig. 20 shows the exergy efficiency with respect to

changing mass flow rate and volume fraction. As it isshown, the exergy efficiency enhanced up to 26.25%, using0.3 vol.% of SWCNTs and a mass flow rate of 0.5 kg/min.Under the similar conditions for 0.1 vol.% of SWCNTs theexergy efficiency enhanced up to 22.35%, whereas the high-est exergy efficiency measured for water under similar con-ditions was 8.77%. This highest energy and exergyefficiency is the result of higher thermal conductivity, whichis supported by the results presented in Figs. 13 and 14.

According to the reports, it is obvious that the energyand exergy efficiencies have contradictory behaviors innumerous cases. Rising fluid inlet temperature results in areduced energy efficiency of collector. However, it resultsin an overall increased exergy efficiency even to its maxi-mum. Correspondingly, rising mass flow rate results inimproved energy efficiency of the collector; however, thishas an inverse effect on exergy efficiency. Most of theexergy destructions arise during the absorbing process inthe collector’s absorber plate. Rising inlet water tempera-ture and reducing water mass flow rate can be effectiveon reducing these destructions.

5. Conclusions

Water-based nanofluids, obtained by dispersingSWCNTs nanoparticles are investigated in this study. By

768 Z. Said et al. / Solar Energy 115 (2015) 757–769

using DLS equipment, different preparation techniqueswere compared. The mean diameter distribution variationwith time is measured by the size measurement technique,therefore, showing the tendency of nanoparticles to settledown. Moreover, the tendency of the nanoparticles toaggregate is shown by the zeta potential measurement.Stability analysis of SWCNTs nanofluid is carried outusing all these measurements, coupled with the visualobservation of the suspension. Distilled water containing0.1 and 0.3 vol.% SDS and 0.1 and 0.3 vol.% SWCNTs,respectively, proved to be very stable for at least 30 days.

The increase in density for both the investigated volumefractions was almost negligible. Specific heat increased withthe volume fraction and dropped with the increasing tem-perature. A maximum conductivity enhancement of 91%is obtained at a temperature of 323 K for 0.3 vol.% volumefraction, whereas, the minimum enhancement in conductiv-ity is 12% for 0.1 vol.% volume fraction of SWCNTs at298 K. Further investigation is required to confirm whetherthe enhancement of thermal conductivity is abnormal.With the increasing temperature, a reduction in viscosityof the nanofluids and wateroccurred. An increase of 39%in viscosity was observed for SWCNTs and SDS suspen-sion at 0.3 vol.%.

According to the mentioned results, keeping the inletwater temperature higher than the ambient temperatureas well as a lower flow rate may result in improved overallperformance. Based on the experimental results, the maxi-mum energy and exergy efficiency of the flat plate collectoris close to 95.12% and 26.25% compared to water whichwas 42.07% and 8.77%, respectively. This low exergy effi-ciency shows that the flat plate collectors still need signifi-cant improvement.

For future progression, it is recommended to performan exergoeconomic analysis, which is a combination ofexergy and economics, and provides useful insights intothe relations between thermodynamics and economics.

Acknowledgement

This research is supported by UM High ImpactResearch Grant UM-MOHE UM.C/HIR/MOHE/ENG/40 from the Ministry of Higher Education, Malaysia.

References

Assael, M. et al., 2005. Thermal conductivity enhancement in aqueoussuspensions of carbon multi-walled and double-walled nanotubes in thepresence of two different dispersants. Int. J. Thermophys. 26 (3), 647–664.

Baughman, R.H. et al., 2002. Carbon nanotubes – the route towardapplications. Science 297 (5582), 787–792.

Bejan, A., 1996. Entropy Generation Minimization: The Method ofThermodynamic Optimization of Finite-size Systems and Finite-timeProcesses. CRC Pressi llc.

Berber, S. et al., 2000. Unusually high thermal conductivity of carbonnanotubes. Phys. Rev. Lett. 84 (20), 4613.

Broersma, S., 2004. Rotational diffusion constant of a cylindrical particle.J. Chem. Phys. 32 (6), 1626–1631.

Choi, S. et al., 2001. Anomalous thermal conductivity enhancement innanotube suspensions. Appl. Phys. Lett. 79 (14), 2252–2254.

Ding, Y. et al., 2006. Heat transfer of aqueous suspensions of carbonnanotubes (CNT nanofluids). Int. J. Heat Mass Transfer 49 (1), 240–250.

Dresselhaus, M.S., Avouris, P., 2001. Introduction to Carbon MaterialsResearch. Carbon Nanotubes. Springer, pp. 1–9.

Duong, H.M. et al., 2008. Calculated thermal properties of single-walledcarbon nanotube suspensions. J. Phys. Chem. C 112 (50), 19860–19865.

Esen, H., 2008. Experimental energy and exergy analysis of a double-flowsolar air heater having different obstacles on absorber plates. Build.Environ. 43 (6), 1046–1054.

Garg, P. et al., 2009. An experimental study on the effect of ultrason-ication on viscosity and heat transfer performance of multi-wallcarbon nanotube-based aqueous nanofluids. Int. J. Heat MassTransfer 52 (21), 5090–5101.

Ghadimi, A. et al., 2011. A review of nanofluid stability properties andcharacterization in stationary conditions. Int. J. Heat Mass Transfer54 (17), 4051–4068.

Gharagozloo, P.E., Goodson, K.E., 2010. Aggregate fractal dimensionsand thermal conduction in nanofluids. J. Appl. Phys. 108 (7), 074309.

Gupta, A., Kumar, R., 2007. Role of Brownian motion on the thermalconductivity enhancement of nanofluids. Appl. Phys. Lett. 91 (22),223102.

Harish, S. et al., 2012. Temperature dependent thermal conductivityincrease of aqueous nanofluid with single walled carbon nanotubeinclusion. Mater. Exp. 2 (3), 213–223.

He, Y. et al., 2011. Experimental study on the light-heat conversioncharacteristics of nanofluids. Nanosci. Nanotechnol. Lett. 3 (4), 494–496.

Hohne, G. et al., 2003. Differential scanning calorimetry. Springer.Hone, J. et al., 2000. Quantized phonon spectrum of single-wall carbon

nanotubes. Science 289 (5485), 1730–1733.Iijima, S., Ichihashi, T., 1993. Single-shell carbon nanotubes of 1-nm

diameter. Nature 363, 603–605.Jha, N., Ramaprabhu, S., 2012. Thermal and electrical conductivity

studies of metal oxide-multiwalled carbon nanotubes nanocompositebased nanofluids. J. Nanofluid 1 (1), 63–70.

Jiang, W. et al., 2010. Modeling of nanoparticles’ aggregation andsedimentation in nanofluid. Curr. Appl. Phys. 10 (3), 934–941.

Kameya, Y., Hanamura, K., 2011. Enhancement of solar radiationabsorption using nanoparticle suspension. Sol. Energy 85 (2), 299–307.

Keblinski, P. et al., 2008. Thermal conductance of nanofluids: is thecontroversy over? J. Nanopart. Res. 10 (7), 1089–1097.

Kim, P. et al., 2001. Thermal transport measurements of individualmultiwalled nanotubes. Phys. Rev. Lett. 87 (21), 215502.

Kreuzer, H.J., 1981. Nonequilibrium Thermodynamics and its StatisticalFoundations. Clarendon Press, Oxford and New York, 455 p., 1.

Li, X. et al., 2007. Evaluation on dispersion behavior of the aqueouscopper nano-suspensions. J. Colloid Interface Sci. 310 (2), 456–463.

Link, S., El-Sayed, M.A., 2000. Shape and size dependence of radiative,non-radiative and photothermal properties of gold nanocrystals. Int.Rev. Phys. Chem. 19 (3), 409–453.

Liu, M.-S. et al., 2005. Enhancement of thermal conductivity withcarbon nanotube for nanofluids. Int. Commun. Heat Mass 32 (9),1202–1210.

Luminosu, I., Fara, L., 2005. Determination of the optimal operationmode of a flat solar collector by exergetic analysis and numericalsimulation. Energy 30 (5), 731–747.

Maruyama, S., 2003. A molecular dynamics simulation of heat conductionof a finite length single-walled carbon nanotube. MicroscaleThermophys. Eng. 7 (1), 41–50.

Mercatelli, L. et al., 2011a. Scattering and absorption properties of carbonnanohorn-based nanofluids for sol. energy applications. J. Eur. Opt.Soc. – Rapid, 6.

Mercatelli, L. et al., 2011b. Absorption and scattering properties ofcarbon nanohorn-based nanofluids for direct sunlight absorbers.Nanoscale Res. Lett. 6 (1), 1–9.

Z. Said et al. / Solar Energy 115 (2015) 757–769 769

Mercatelli, L. et al., 2012. Carbon nanohorn-based nanofluids: character-ization of the spectral scattering albedo. Nanoscale Res. Lett. 7 (1),1–5.

Mizel, A. et al., 1999. Analysis of the low-temperature specific heat ofmultiwalled carbon nanotubes and carbon nanotube ropes. Phys. Rev.B 60 (5), 3264.

Nasiri, A. et al., 2012. Effect of CNT structures on thermal conductivityand stability of nanofluid. Int. J. Heat Mass Transfer 55 (5), 1529–1535.

Onsager, L., 1931a. Reciprocal relations in irreversible processes. I. Phys.Rev. 37 (4), 405.

Onsager, L., 1931b. Reciprocal relations in irreversible processes. II. Phys.Rev. 38 (12), 2265.

Otanicar, T.P. et al., 2010. Nanofluid-based direct absorption solarcollector. J. Renew. Sust. Energy 2, 033102.

Pop, E. et al., 2006. Thermal conductance of an individual single-wallcarbon nanotube above room temperature. Nano Lett. 6 (1), 96–100.

Pradhan, N. et al., 2009. The specific heat and effective thermalconductivity of composites containing single-wall and multi-wallcarbon nanotubes. Nanotechnology 20 (24), 245705.

Said, Z. et al., 2013. Experimental investigation of the thermophysicalproperties of Al2O3-nanofluid and its effect on a flat plate solarcollector. Int. Commun. Heat Mass 48 (0), 99–107.

Saidur, R. et al., 2012. Evaluation of the effect of nanofluid-basedabsorbers on direct solar collector. Int. J. Heat Mass Transfer.

Saito, R. et al., 1998. Physical Properties of Carbon Nanotubes. WorldScientific.

Sajid, M.H. et al., 2014. A time variant investigation on optical propertiesof water based Al2O3 nanofluid. Int. Commun. Heat Mass 50, 108–116.

Sani, E. et al., 2011. Potential of carbon nanohorn-based suspensions forsolar thermal collectors. Sol. Energy Mater. Sol. Cell 95 (11), 2994–3000.

Standard, A., 1977. 93–77, Methods of Testing to Determine the ThermalPerformance of Solar Collectors. American Society of Heating,Refrigeration and Air Conditioning Engineers, New York (USA).

Sukhatme, S.P., 2008. Sol. Energy: Principles of Thermal Collection andStorage. Tata McGraw-Hill Education.

Sun, Z. et al., 2008. Quantitative evaluation of surfactant-stabilized single-walled carbon nanotubes: dispersion quality and its correlation withzeta potential. J. Phys. Chem. C 112 (29), 10692–10699.

Syam Sundar, L., Sharma, K., 2008. Thermal conductivity enhancementof nanoparticles in distilled water. Int. J. Nat. 1 (1), 66–77.

Tans, S.J. et al., 1997. Individual single-wall carbon nanotubes asquantum wires. Nature 386 (6624), 474–477.

Taylor, R.A. et al., 2011. Applicability of nanofluids in high flux solarcollectors. J. Renew. Sust. Energy 3, 023104.

Tiwari, A.K. et al., 2013. Solar water heating using nanofluids – acomprehensive overview and environmental impact analysis. Int. J.Emerg. Technol. Adv. Eng. 3 (3), 221–224.

Tsyboulski, D.A. et al., 2008. Translational and rotational dynamics ofindividual single-walled carbon nanotubes in aqueous suspension.ACS Nano 2 (9), 1770–1776.

Tyagi, H. et al., 2009. Predicted efficiency of a low-temperature nanofluid-based direct absorption solar collector. J. Sol. Energy Eng. 131 (4).

Ucar, A., Inallı, M., 2006. Thermal and exergy analysis of solar aircollectors with passive augmentation techniques. Int. Commun. HeatMass 33 (10), 1281–1290.

Wang, B.-X. et al., 2006. Surface and size effects on the specific heatcapacity of nanoparticles. Int. J. Thermophys. 27 (1), 139–151.

Wen, D., Ding, Y., 2004. Effective thermal conductivity of aqueoussuspensions of carbon nanotubes (carbon nanotube nanofluids). Int. J.Thermophys. Heat Transfer 18 (4), 481–485.

Xie, H. et al., 2003. Nanofluids containing multiwalled carbon nanotubesand their enhanced thermal conductivities. J. Appl. Phys. 94 (8), 4967–4971.

Yousefi, T. et al., 2012a. An experimental investigation on the effect of pHvariation of MWCNT–H2O nanofluid on the efficiency of a flat-platesolar collector. Sol. Energy 86 (2), 771–779.

Yousefi, T. et al., 2012b. An experimental investigation on the effect ofMWCNT–H2O nanofluid on the efficiency of flat-plate solar collectors.Exp. Therm. Fluid Sci. 39, 207–212.

Yousefi, T. et al., 2012c. An experimental investigation on the effect ofAl2O3–H2O nanofluid on the efficiency of flat-plate solar collectors.Renew. Energy 39 (1), 293–298.

Yu, C. et al., 2005. Thermal conductance and thermopower of anindividual single-wall carbon nanotube. Nano Lett. 5 (9), 1842–1846.

Yu, W. et al., 2008. Review and comparison of nanofluid thermalconductivity and heat transfer enhancements. Heat Transfer Eng. 29(5), 432–460.

Zhang, S. et al., 2003. Specific heat of single-walled carbon nanotubes.Phys. Rev. B 68 (7), 075415.