International Journal of Heat and Mass Transfer 93 (2016) 862–871

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer

journal homepage: www.elsevier .com/locate / i jhmt

Experimental investigation on thermo physical properties of singlewalled carbon nanotube nanofluids

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2015.10.0710017-9310/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +60 3 7967 6851; fax: +60 3 7967 5316.E-mail addresses: sabiha.akter89@gmail.com (M.A. Sabiha), saad@um.edu.my

(S. Mekhilef).

M.A. Sabiha a, R.M. Mostafizur a, R. Saidur b, Saad Mekhilef c,⇑aDepartment of Mechanical Engineering, Engineering Faculty, University of Malaya, 50603 Kuala Lumpur, MalaysiabCentre of Research Excellence in Renewable Energy (CoRE-RE), King Fahd University of Petroleum and Minerals (KFUPM), Dhahran 31261, Saudi ArabiacPower Electronics and Renewable Energy Research Laboratory (PEARL), Department of Electrical Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia

a r t i c l e i n f o

Article history:Received 23 May 2015Received in revised form 16 October 2015Accepted 16 October 2015

Keywords:Single walled carbon nanotubeThermal conductivityViscosityRheologySpecific heat

a b s t r a c t

This experimental study is aimed to measure and analyze the thermal conductivity, viscosity, and specificheat of water based single walled carbon nanotube (SWCNT) nanofluids in presence of sodium dodecylsulfate (SDS) surfactant. The surfactant was used to prepare stable nanofluids and the stability ofSWCNT nanofluids of five volume concentrations (0.05–0.25 vol%) is observed good. The measured valuesof thermal conductivity in the range of 0.615–0.892 W/m K, viscosity in the range of 0.67–1.28 mPa s, andthe specific heat in the range of 2.97–3.90 kJ/kg �C, were observed for temperature rising from 20 to 60 �Cwith an interval of 10 �C as the volume concentration increased from 0.05 to 0.25 vol%. The maximumthermal conductivity enhancement of 36.39% compared to water is observed for 0.25 vol% at 60 �C.The viscosity of SWCNT nanofluids exhibited a non-Newtonian shear-thinning behavior due to the align-ment of nanotube clusters and agglomerates with increasing shear rate. The temperature and volumeconcentrations have effect on specific heat as well and it decreases with particle loadings while increaseswith temperature.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Carbon nanotube (CNT), a wonder material in multidisciplinaryfields including material science, automotive, optical, electrical,aerospace, and energy conversion, possess outstanding electrical,thermal, mechanical, chemical and optical features. Carbon nan-otubes (CNTs) have the capability to conduct electricity and heatefficiently and can act like metals or semiconductors. CNTs are fas-cinating as they can be used in lithium ion batteries; polymerbased composite materials; nanoelectronics as diodes and transis-tors, and in super-capacitors like electromechanical actuators andsensors [1]. Advanced membrane technology is another field whereCNTs are used for water desalination process. Due to the uniquestructure, high surface to volume ratio and high chemical stability,carbon nanotubes have emerged as new class nanomaterials whichpossess properties of individual components with synergistic effectwhen integrated with some other materials [2]. Thus research oncarbon nanotubes has become essential to accelerate the innova-tion of advanced technologies in diversified fields.

CNTs are tubular in shape as they are composed of cylindricalsheet form with carbon which is rolled up in a tube like structurewith the appearance of lattice work fence. There are three types ofCNTs for instance single walled, double walled and multi walled.Single-walled carbon nanotube (SWCNT) consists of one cylindricalgraphite sheet whereas multi-walled carbon nanotube (MWCNT)contains multiple layers of graphene sheets.

Experimental and theoretical studies have demonstrated highthermal conductivity of cylindrical structured nanoparticles com-pared to spherical nanoparticles [3]. Spherical nanoparticles arethe metallic and oxide nanomaterials such as aluminum and alu-minum oxide have thermal conductivity of 237 W/m K and 40 W/m K respectively [1]. In contrast, CNTs have thermal conductivityin a range of 2000–6000W/m K. Specifically, the values for thermalconductivity of single walled carbon nanotube (SWCNT), doublewalled carbon nanotube (DWCNT) and multi walled carbon nan-otube (MWCNT) are 6000 W/m K, 3986W/m K and 3000 W/m K,respectively [4,5]. The thermal conductivity of DWCNT andMWCNT decreases respectively due to the increase of nanotubewall layers [6].

High thermally conductive materials have attracted researchersto investigate the performance of existing heat transfer system byadding highly thermally conductive particles like carbon nan-otubes, metal, metal oxides into heat transfer fluid to improve

Abbreviations and Nomenclature

SWCNT single walled carbon nanotubeMWCNT multi walled carbon nanotubeSDS sodium dodecyl sulfateTC thermal conductivityEG ethylene glycolwt% weight percentTHWM transient hot wire methodD diameter of nanoparticle, nmL length of nanoparticle, lmm weight, kgl viscosity, mPa s

Cp specific heat, kJ/kg �CK thermal conductivity, W/m Kq density, kg/m3

/ volume concentration, vol%

SubscriptNf nanofluidNp nanoparticleBf base fluideff effective

M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871 863

the overall thermal conductivity. In 1995, researchers at ArgonneNational Laboratory first demonstrated the dilute liquid particlemixtures, called nanofluids, can exhibit thermal conductivity val-ues of about 20–150% higher than the base heat transfer fluids[4]. The term introduced as ‘nanofluid’ because the particles dis-persed in the base fluid are in range of 1–100 nm size.

CNT nanoparticles have high thermal conductivity, large speci-fic surface area (SSA), high aspect ratio and low specific gravity.Therefore, CNT nanofluids expected to exhibit excellent thermalfeatures, long term stability and rheological properties comparedto traditional working fluids [7]. The first study regarding CNTnanofluid was conducted by Choi et al. [8] by dispersing MWCNTnanoparticles in synthetic poly oil base fluid. They reported 160%enhancement in thermal conductivity for 1.0 vol% of MWCNTnanoparticles. According to literature, no further studies havereported such improvement in thermal conductivity using MWCNTnanofluid. Xie et al. [9] dispersed 1.0 vol% MWCNT nanoparticles inthree different base fluids and the enhancement of thermal con-ductivity was observed as 7.0%, 12.7%, and 19.6% for distilled water(DW), ethylene glycol (EG), and decene (nonpolar liquid) respec-tively. Though a small amount of CNT nanoparticles in nanofluidis capable of enhancing thermal conductivity significantly, theCNT nanoparticles tend to agglomerate due to hydrophobic natureof CNTs, high surface area and the van der waals forces. Sonicationis the most common method used by the researchers to improvethe stability of CNT nanofluids by breaking the agglomerations ofnanoparticles. Beside sonication, various surfactants such as sodiumdodecyl sulfate (SDS), sodium dodecyl benzene sulfonate (SDBS),cetyltrimethyl ammonium bromide (CTAB), hexamethyldisi loxane(HMDS), sodium deoxycholate (DOC), poly-vinyl pyrrolidone (PVP)and gum Arabic (GA) have been used to obtain desired stability ofsample nanofluids. Functionalization of CNT nanofluid is anotherwaytoachievebetterstabilitywithhigher thermalconductivity[10].

Most of the studies in literature used water or EG to disperseMWCNT nanoparticles and mainly focused on their stability.Experimental studies to determine the thermal conductivity ofCNT nanofluid using different methods, different base fluids, differ-ent surfactants, and different volume concentrations of nanoparti-cles having different length and diameter have been conductedwhich is summarized in Table 1.

Due to outstanding thermal conductance capability, CNT nano-fluid is privileged nowadays as an excellent heat transfer fluid indifferent applications. Kathiravan et al. [11] used MWCNT nano-fluid for pool boiling to investigate the heat transfer behavior ofthe fluid. The MWCNT nanoparticles were dispersed by 0.25, 0.5,and 1.0 vol% with water and also with the suspension of waterand SDS. The heat transfer coefficients of nanofluid were enhanced1.7 times compared to water.

Park and Kim [12] used nanofluid, formulated with hydroxylradicals combined with oxidized multi-walled CNTs (MWCNTs)

to enhance the heat-transfer utility of the heat pipe in a solar col-lector. The thermal conductivity is found to be 12.6% higher at90 �C than that of the base fluid and the viscosity is 11% lowerdue to oxidation. Therefore, oxidized MWCNT nanofluid as work-ing fluid can provide increased operating temperature range aswell as total heat in a heat pipe of a collector.

Karami et al. [13] introduced CNT nanofluid as an excellentworking fluid for direct absorption solar collector (DASC) due tohigh thermal conductivity, good optical properties, and dispersionstability. Functionalized CNT (f-CNT) nanoparticles were dispersedin water and the thermal conductivity, optical property and stabil-ity were observed for six different volume concentration (5, 10, 25,50, 100, 150 ppm) of f–CNT particle. The extinction coefficient ofnanofluid having 150 ppm CNT increased by 4.1 cm�1 and thermalconductivity increased by 32.2% compared to water. They alsofound that the thermal conductivity is mainly dependent on tem-perature than the volume concentration which is an advantagefor solar collector applications. Therefore, they reported CNT nano-fluid as a very suitable working fluid for increasing overall effi-ciency of DASC.

Chougule [14] conducted a study on FPC consists of heat pipeand compared the performance using water and CNT nanofluid.The performance of collector using nanofluid found to be better.The average collector efficiencies using water and nanofluid for tiltangle 31.5� are 25% and 45%, for tilt angle 50� are 36% and 61%respectively. The maximum instantaneous efficiency obtained byusing CNT nanofluid is 69% for 50� tilt angle.

To investigate the effect of nanofluid on heat transfer character-istics of an intertube falling film, Ruan and Jacobi used MWCNTnanofluid based on water and EG due to very high thermal conduc-tivity (3000 W/m K) of MWCNTs [15]. Nanofluids using both basefluids were prepared with volume concentration of 0.05, 0.14,and 0.24. About 9% enhancements in thermal conductivity wereobserved for 0.24 vol% MWCNT nanofluid based for both base flu-ids. However, 0.24 vol% MWCNT nanofluid based on EG were cap-able of enhancing the heat transfer up to 20%.

Compared to MWCNT and DWCNT nanoparticles, SWCNTnanoparticles exhibit higher thermal conductivity and better opti-cal properties such as raman, fluorescence and absorption spectra.Therefore, it is necessary to conduct more studies on SWCNT nano-fluid to find out their suitability in various heat transferapplications.

Table 1 summarizes the previous studies regarding thermalproperties of CNT nanofluid and it is remarkable that most of thestudies conducted on MWCNT and mainly focused on thermal con-ductivity. Viscosity and specific heat capacity are the key parame-ters to calculate enthalpy in various thermal processes, todetermine heat transfer rates under flow conditions and other heattransfer properties, and to evaluate heat storage capacity of ther-mal management systems. However, Table 1 illustrates that only

Table 1Summary of previous studies on CNT nanofluids.

References Nanofluid Particlesize

Surfactant Concentration Sample preparation andmeasurement method

Findings

Estellé et al. [16] MWCNT/DI water D 9.2–11.4 nmL 1–1.5 lm

SDBSLigninSodium polycarboxylate

1 wt.% nanotube2 wt% surfactant

Two stepKD2 pro

–Very low volume concentration of nanoparticles is ableto enhance TC– TC enhancement is weakly affected by CNT aspectratio and surfactant type

Sadri et al. [5] MWCNT/DW D 20–30 nmL 10–30 lm

SDSSDBSGA

0.25 wt.%0.5 wt.%

Two stepTC-KD2 Proviscosity –rotational Rheometer

–TC increased 22.31%– MWCNT nanofluid containing GA exhibits higher TCthan SDS and SDBS.

Wusiman et al. [17] MWCNT/DW D 5–20 nmL � 5 lm

SDBSSDS

0.5 wt.% CNT with0.25 wt.% SDBS (pH � 9)

Two stepTHWM

–MWCNT nanofluid containing SDBS performs better TCthan SDS–TC increased 2.8%

Wusiman et al. [7] MWCNT/DW + EG D 20–30 nmL 10–30 lm

Chitosan dispersant 0.1 wt.%0.2 wt.%0.3 wt.%

Two step(Temp range 80–95 �C)KD2 Pro–TCViscosity meter for viscosity andDSC for specific heat

–TC ratio increased 35.1–40.2% compared to EG/DW andviscosity ratio increases by 3.8–5.1%– TC increased 35.4–45.5%–TC increased 40.4– 49.6%– Viscosity ratio increased 4.8–6.4%– Specific heat increased with temperature butdecreased with increasing concentrations

Kumaresan et al. [18] MWCNT/DI water(70 vol%)/EG (30 vol%)

D 30–50 nmL 10–20 lm

SDBS (0.1 vol%) 0.15 vol%0.3 vol%0.45 vol%

TC–KD2 ProViscosity– Bohlin CVOrheometer

–Heat transfer coefficient enhanced by 92% for 0.15 vol%MWCNT–Heat transfer coefficient enhanced by 150% for 0.45 vol% MWCNT

Haris et al. [3] SWCNT/DW D 0.8–1.6 nmL 0.1–0.6 lm

DOC 0.3 vol% –TC increased 16% (60 �C)–Viscosity ratio increases 8.8–9.3%

Haris et al.[19] SWCNT/EG DOC 0.2 vol.% THWM –TC increased 14.8%Wavekar et al. [20] MWCNT/EG D 20 nm

L 35 lm0.35 wt% SDS 0.6 vol.% Two step THWM –TC increased 21%

MWCNT/GA 0.05 vol% –TC increased 37.4% at 25 �C�TC increased 287.5% at 60 �C

Kumaresan and Velraj[21]

MWCNT/DI water (70vol%) /EG(30vol%)

D 30–50nmL 10–20 lm

SDBS (0.1 vol%) 0.15 vol%0.3 vol%0.45 vol%

TC –KD2 ProViscosity– Bohlin CVOrheometerSpecific heat– DSC(TAinstrument, Q200)

–TC increased by maximum 19.73% at 0.45 vol%–The enhancement in specific heat is maximum forlower concentration– Viscosity increases with increasing vol% but decreaseswhen temperature increases

Aravind et al. [22] MWCNT/DW/EG Functionalized CNT 0.03 vol% Two step Lambda Instruments –TC of water and EG increased up to 33% and 40%respectively

Kim et al. [23] Plasma coated MWCNT/DW

0.01 vol% Two step THWM –TC increased 255%

Garg et al. [24] MWCNT/GA 1.0 wt% –TC increased 20% at 30 �CChen and Xie [25] MWCNT/SO HMDS/functionalization 1.0 vol% Two step TSHWM –TC increased 19%Jiang et al. [26] CNT/R113 None 1.0 vol% One step Thermal constant

analyzer–TC increased 10.4%

Nanda et al. [27] SWCNT/EG/PAO Transient planar sourcetechnique

–TC increased for EG 35% and for PAO 12%

Glory et al. [28] MWCNT/W GA 3.0 wt% Steady state –TC increased 64%Chen et al. [29] MWCNT/W/EG Surface treatment 1.0 vol.% Two step THWM –TC increased 17.5%Amrollahi et al. [30] SWCNT/EG None 2.5 vol% Two step steady method –TC increased 20%Hwang et al. [31] MWCNT D 10–30 nm

L 10–50 lmSDS 1.0 vol% –TC increased 7.0%

MWCNT/Oil 0.5 vol% –TC increased 8.7%Ding et al.[32] MWCNT/GA D 20–60 nm

L few ten lm1.0 wt% –TC increased 79% (30 �C)

Hwang et al. [33] MWCNT/W SDS 1.0 vol.% Two stepTHWM

–TC increased 11.3%

864M.A.Sabiha

etal./International

Journalof

Heat

andMass

Transfer93

(2016)862–

871

Table1(con

tinu

ed)

Referen

ces

Nan

ofluid

Particle

size

Surfactant

Con

centration

Sampleprep

arationan

dmea

suremen

tmethod

Findings

Assae

let

al.[34

]MW

CNT/W

CTA

BTw

ostep

–TCincrea

sed34

%

Triton

X-100

0.6vo

l.%TH

WM

–TCincrea

sed13

%Hwan

get

al.[35

]MW

CNT/W

SDS

1.0vo

l%TH

WM

–TCincrea

sed7%

Assae

let

al.[36

]MW

CNT/W

CTA

B0.6vo

l%TH

WM

–TCincrea

sed34

%Liuet

al.[37

]MW

CNT/EG

D20

–50nm

1.0vo

l%Mod

ified

THW

M–T

Cincrea

sed12

.4%

MW

CNT/en

gineoil

2.0vo

l%–T

Cincrea

sed30

%Assae

let

al.[38

]MW

CNT

D10

0nm

L50

lm

SDS

0.6vo

l%TH

WM

–TCincrea

sed38

%

M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871 865

a couple of studies were conducted regarding viscosity and specificheat of CNT nanofluids. In addition, there are very limited litera-tures on the study of SWCNT nanofluids particularly. Therefore,this study aimed to investigate the stability and thermal properties(thermal conductivity, viscosity, specific heat, and rheology) ofSWCNT nanofluid to fill the research gap.

In this paper, the thermal conductivity, viscosity and specificheat of SWCNT nanofluid of five volume concentrations rangingfrom 0.05 to 0.25 vol% are studied. The higher temperature to mea-sure thermal properties was limited to 60 �C as the bondingbetween the nanoparticles and surfactant can be damaged above60 �C [36] as well as to prevent water evaporation from the sampleduring the experiments.

2. Experimental procedures and apparatus

2.1. Material

SWCNT nanoparticles, distilled water (DW) as base fluid and,SDS surfactants (to stabilize the nanofluid) were used in preparingthe SWCNT nanofluid. The nanotubes (90% CNTs, 60% SWCNTs) oflength 500 nm and diameter 1–2 nm were purchased from Nanos-tructured & Amorphous Materials, Inc., Houston, TX, USA. Sodiumdodecyl sulfate (SDS, 92.5–100.5%) surfactant was sourced fromSigma Aldrich, USA. The properties of SWCNT nanoparticle usedin this experiment are listed in Table 2.

The specific surface areas of SWCNTs were in range of 360–400 m2/g as received. The crystalline phases of the SWCNT weredetermined by X-ray diffraction which is illustrated in Fig. 1.

The characteristic peaks match with the standard Joint Commit-tee on Powder Diffraction Standards (JCPDS) card No. 98-061-7290.Fig. 2 presents the scanning electron microscopy (SEM) imagewhich was used to measure the size and morphology of SWCNTnanoparticles before being dispersed in base fluids.

2.2. Procedures to prepare SWCNT nanofluid

The preparation of stable nanotube suspension requires twosteps. The first one is breaking the agglomeration by mechanicalprocesses and the second one is stabilizing the suspensions. Highpressure ultrasonic homogenizers are most frequently used inorder to overcome the strong cohesion forces between nanotubes.The nanotubes in the suspension segregate and settle down undergravitational forces due to the hydrophobic character without anysurfactant. Therefore, SDS surfactant which gives a hydrophiliccharacter to CNTs is used in the experiment in order to get a stablesuspension of CNTs and water. Two-step method was applied toprepare SWCNT nanofluids of five desired volume concentrationsin the present study. The suspension of SWCNTs, SDS and DWwas then sonicated by a high pressure ultrasonic homogenizerfor one hour as shown in Fig. 3. The desired concentrations used

Table 2Properties of SWCNTs.

CNTs (%) 90SWCNTs (%) 60Density (g/cm3) 2.1 at 20 �CDiameter (nm) 1–2Length (nm) 500Purity (%) 90Melting range (�C) 3652–3697Color BlackOdor OdorlessThermal conductivity (W/m K) 6000 [5]Specific surface area (SSA) (m2/g) 360–400

0

10000

20000

30000

40000

50000

60000

70000

0 20 40 60 80 100

Counts

2-Theta

(002)

(011)

(004) (112)

Fig. 1. XRD pattern of SWCNT nanoparticles.

Fig. 2. SEM image of SWCNT nanoparticles.

Fig. 3. Sonication of (SWCNT-SDS + DW) suspension.

866 M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871

in this study are 0.05, 0.1, 0.15, 0.2 and 0.25 vol%. The nanoparticlevolume concentrations were calculated using Eq. (1).

Volume concentration; / ¼ mnp=qnp

mnp=qnp þmbf =qbf

!ð1Þ

where, / is the nanoparticle volume concentration (%), mnp and mbf

are the weight of nanoparticles and the base fluid respectively;whereas qnp and qbf are the density of nanoparticles and base fluidsrespectively.

It is understood that the highly agglomerated nanoparticleswere able to be easily broken by the combination of strong shearforce and cavitation generated by the high-pressure homogenizer.The SWCNT nanofluids preparation conditions are enlisted inTable 3.

2.3. Stability

The applications of nanofluids depend on their properties andstability which can be determined by the quality of the dispersionand suspension of nanoparticles in the base fluid. The stability ofthe nanofluids is influenced by the surface charge of the nanopar-ticles and zeta potential is used as the index of the surface charge

Table 3Nanofluids preparation conditions.

Base fluid Water

Nanoparticle SWCNTNanoparticle type CylindricalSurfactant SDSVolume concentration (vol%) 0.05, 0.1, 0.15, 0.2, 0.25

Ultra-sonicatorTime (min) 60Power (W) 500Frequency (kHz) 20Pulse (s) 2Term (s) 2

of the nanoparticles. The zeta potential value is high for a stablenanofluid as the dispersed nanoparticles are stable due to highelectrostatic repulsion force between them. Aggregated particlesdue to weak repulsive force resulting in high collision betweennanoparticles provide lower zeta potential. If the absolute valueof zeta potential is over 30 mV, the nanofluid is generally consid-ered as a stable nanofluid [39,40]. The suspension shows excellentstability when the absolute value of zeta potential is above 60 mVwhile a suspension below 20 mV has limited stability [41]. The zetapotential was measured at 25 �C using a Zetasizer Nano Zs manu-factured by Malvern Instruments Ltd. Fig. 4 shows the absolutezeta potential with respect to the volume concentration of SWCNTnanoparticles. The minimum (lower bar) and maximum (upperbar) values for the measurement are represented by the errorbar. It can be seen that the SWCNT nanoparticles are negativelycharged due to the addition of SDS dispersant. The measured zetapotential was �49.4, �53.1, �55.5, �50.3, and �54.5 mV for 0.05,0.1, 0.15, 0.2 and 0.25 vol% respectively, negative as anticipatedin the case of anionic dispersant. Therefore, the stability of SWCNTnanofluids of all volume concentration can be considered as good.

Fig. 5 demonstrates the average particle size of different volumeconcentrations of SWCNT nanoparticles. It can be observed that theaverage particle sizes of SWCNT are in a range of 122–147 nm. TheSWCNT nanoparticles in nanofluids contact each other and formclusters and the cluster sizes are comparatively larger for 0.2 vol% and 0.25 vol%.

-65

-60

-55

-50

-45

-40

-35

-30

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Abs

olut

e ze

ta p

oten

tial (

mV

)

Volume concentration (vol%)

Fig. 4. Absolute zeta potential of SWCNT nanofluids as a function of volumeconcentration.

100

115

130

145

160

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Ave

raze

par

ticle

size

(nm

)

Volume concentration (vol%)

Fig. 5. Average particle size in SWCNT nanofluids as a function of volumeconcentration.

100 nm

Fig. 6. TEM image of 0.05 vol.% SWCNT nanofluid (SDS surfactant).

M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871 867

To predict the nanofluid stability correctly, it is important tocharacterize the morphology of the nanotubes after they are dis-persed in their respective base fluids. This was accomplished usingtransmission electron microscopy (TEM) as shown in Fig. 6. It isobserved from Fig. 6 that the nanoparticles are tubular in shape.The nanoparticles in this study were dispersed well and noagglomeration was detected in nanofluids. However, the existenceof some clusters among the nanotubes can be seen in Fig. 6.

2.4. Thermal conductivity measurement

The thermal conductivity of nanofluids was measured usingKD2 Pro thermal property analyzer (Decagon Device, USA) whichworks based on the principle of THWM with maximum deviationof 5.0%. This device measures thermal conductivity over the rangeof 0.02–2.00 W/m K with ±0.001% accuracy of the equipment. Thethermal conductivity of the SWCNT nanofluid were measured at20, 30, 40, 50, and 60 �C for five volume concentrations (0.05, 0.1,0.15, 0.2 and 0.25 vol%). All the data were recorded for four timesand the corresponding average values were used for analysis. Themeasured thermal conductivity values of SWCNT nanofluids werethen compared with the model developed by Nan et al. [42], Patelet al. [43], Timofeeva et al.[44] and Haris et al. [3]’s experimentalresults.

The model developed by Nan et al. [42] which is specifically forcarbon nanotube based composites where the thermal conductiv-ity of CNT (Knp) is much larger than the thermal conductivity ofthe base fluid (kbf).

keffkf

¼ 3þ /knp=knf3� 2/

ð2Þ

Patel et al. [43] developed a model to measure thermal conduc-tivity of CNT nanofluids considering combined parallel paths forheat, one through the base fluids, the other through the CNTs. Thismodel depends on volume concentration and both base fluid andnanoparticle radius rbf and rnp respectively. In absence of measure-ment, the radius of base fluid here is considered as the moleculeradius of water.

keffkf

¼ knprbf/kbf rnpð1� /Þ ð3Þ

To investigate the effects of particle shape on thermal conduc-tivity, Timofeeva et al. [44] developed a correlation which is statedas Eq. (4).

knf ¼ 1þ CShape þ Csurface� �� �

kbf ¼ ð1þ Ck/Þkbf ð4Þ

2.5. Viscosity measurement

A Brookfield (LVDV III ultra-programmable) viscometer wasused in this experiment to measure the viscosity and the shearstress at different temperatures, volume concentrations, and shearrates. A computer was connected with the viscometer to collectand store data. The spindle of the viscometer was submerged intothe nanofluids. The viscous effect was developed against the spin-dle due to deflection of calibrated spring with the help of Ultra LowAdapter (ULA). To control temperature, a refrigerated circulatingbath was connected with ULA attached with the rheometer. Therheometer was connected with the computer into which rheocalc32 software has been installed to obtain the rheological data ofSWCNT nanofluids. In this experiment, the viscosity and shearstress of the samples were measured within a shear rate range of24.46–293.5 s�1 while the spindle rotation was 20–240 rpm.

The viscosity of SWCNT nanofluid at five volume concentrations(0.05, 0.1, 0.15, 0.2 and 0.25 vol%) was measured at 20, 30, 40, 50,and 60 �C respectively. The data were recorded four times and thecorresponding averaged values were plotted.

The experimentally measured viscosity was then comparedwith Eqs. (5) and (7) developed by Brenner and Condiff [45] andTimofeeva et al. [44] respectively to investigate the effects of par-ticle shape on the viscosity of a suspension. Both models are appli-cable for rod like or tubular shape particles.

lr ¼ ð1þ g/Þ ð5Þ

g ¼ 0:312rln 2r � 1:5

þ 2� 0:5ln 2r � 1:5

� 1:872r

ð6Þ

where r is the aspect ratio of nanoparticles, r ¼ L=D.Here L and D are the length and diameter of the nanoparticles

respectively.

lnf ¼ 1þ 13:5/þ 904:4/2� �lbf ð7Þ

2.6. Specific heat measurement

The differential scanning calorimeter (model: DSC 4000, PerkinElmer, USA) was used to measure the specific heat of SWCNT nano-fluid. The specific heat was measured for every 0.05 vol% increasewithin the range of 0.05–0.25 vol% and every increase of 10 �Cwithin the temperature range 20–60 �C. For each volume concen-tration, the specific heat was measured for three times and the cor-responding average values were plotted.

Considering the fact that very limited experimental data onspecific heat capacity values for various SWCNT nanofluids at dif-ferent concentrations are available, the value of the specific heatcapacities were compared with theoretical models. According to

2.02.53.03.54.04.5

ctiv

ity ra

tio (k

nf/k

bf)

Nan et al. eq. [42]

Patel et al. eq. [43]

Experimental result

868 M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871

available literature, Eq. (8) is mostly used to calculate the specificheat capacity of nanofluids, at any particle concentration [21,46]:

CPnf ¼ð1� /ÞðqCpÞbf þ /ðqCpÞnp

ð1� /Þqbf þ /qnpð8Þ

0.00.51.01.5

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Ther

mal

con

du

Volume concentration (vol%)

Haris et al. [3]

Timofeeva et al. [44]

Fig. 8. Experimental results of thermal conductivity ratio compared to existingcorrelations.

3. Results and discussion

3.1. Thermal conductivity

The thermal conductivity of SWCNT nanofluid for volume con-centration ranging from 0.05 to 0.25 vol% with respect to temper-ature is shown at Fig. 7. Figure also includes the thermalconductivity data of distilled water for comparison. The enhance-ment of thermal conductivity is low at temperature 20–30 �C butlarger enhancement is observed for temperatures above 30 �Ccompared to distilled water. Though Brownian motion is the mainreason for the thermal conductivities change with temperature,higher aspect ratio of the particle, the effects of particle agglomer-ation, the viscosity change with temperature, and the increasedsurface area of suspended nanostructures also contribute to thetemperature dependence of SWCNT nanofluid [47]. Increment inthermal conductivity observed with the increase in the volumeconcentration is due to the inherent high heat transfer capacityof SWCNT nanoparticles.

The highest enhancement in thermal conductivity at 60 �C isobserved to be 9.36%, 13.66%, 23.93%, 30.76% and 36.39% comparedto water for 0.05, 0.1, 0.15, 0.2, and 0.25 vol% SWCNT respectably.Keblinski et al. [48] suggested that the micro or nano-convectiondue to Brownian motion, formation of nanolayer around particles,and near field radiation are the mechanisms behind the thermalconductivity enhancement of naofluids. Some recent studies haverevealed the increase of cluster size has significant impact on theenhancement of thermal conductivity [49–52] but excessive clus-tering results sedimentation which is an adverse effect. From theexperimental result, higher enhancement in thermal conductivityis gained for higher volume concentration and from Fig. 5 largercluster size is observed for higher loading of nanoparticles. There-fore, the nanoparticle clustering in nanofluids can also be the pos-sible reason behind the thermal conductivity enhancement.

The experimental results are then compared with the experi-mental results obtained by Haris et al. [3] at Fig. 8. Similar trendof thermal conductivity enhancement with increasing volume con-centration and temperature of SWCNT nanofluid is observed butHaris et al. [3] reported a maximum conductivity increase of 16%using 0.3 vol% SWCNT at 60 �C whereas the maximum conductivityachieved in this experiment at 60 �C is 36.39% for 0.25 vol%SWCNT. This might happen due to the usage of different surfactantas well as different length and diameter of SWCNT nanoparticles.Another reason could be the purity of SWCNT nanoparticles. The

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

10 20 30 40 50 60

Ther

mal

con

duct

ivity

(W/m

.K)

Temperature (ºC)

0.25 vol%0.20 vol%0.15 vol%0.10 vol%0.05 vol%water

Fig. 7. Thermal conductivity of SWCNT nanofluid as a function of temperature andparticle volume concentrations.

current experiment used commercially available nanoparticleswhereas Haris et al. used synthesized SWCNT nanoparticles. Theexperimental results are also compared with Nan model [42], Patelmodel [43] and Timofeeva model [44]. From Fig. 8, it is observedthat Nan model [42] and Patel model [43] overestimated the ther-mal conductivity of SWCNT nanofluid. The effects of Brwonianmotion and aggregation of the nanotubes and surfactants wereignored for the models which could be the possible reasons behindthe large differences between the theoretical and current experi-mental results. There are other important factors such as differencein particle size, nanofluid preparation methods, and experimentalprocess can cause the discrepancies as well. However, Timofeevamodel [44] shows good agreement with the obtained results.

3.2. Viscosity

Fig. 9 illustrates the viscosity of SWCNT nanofluids, as a func-tion of temperature and various volume concentrations. It showsan increase in viscosity with increasing nanoparticle concentration.At 20 �C, the viscosities of SWCNT nanofluids are 1.18, 1.21, 1.23,1.26, and 1.28 mPa s for 0.05, 0.1, 0.15, 0.2, and 0.25 vol% respec-tively. As particle concentration increases, the internal viscousshear stress increases as shown in Fig. 11 which results in higherfluid viscosity. However, the viscosity decreases with increasingtemperature as the increasing temperature weakens the inter-molecular forces of the particles and fluid itself. At 0.05 vol%, theviscosities of SWCNT nanofluids are 1.18, 0.95, 0.81, 0.75, and0.67 mPa s for 20, 30, 40, 50, and 60 �C respectively. The first studyconducted by Masuda et al. [53] to investigate the concentrationand temperature dependence of viscosity of aqueous nanofluidscontaining Al2O3, TiO2 and SiO2. They also reported that the viscos-ity increases with concentration while decreases with temperaturenonlinearly. Most of the studies in literature regarding viscosity ofnanofluids reported either substantial or nonlinear decrease of vis-cosity with increasing temperature [32,54–57].

0.40

0.60

0.80

1.00

1.20

1.40

10 20 30 40 50 60

Vis

cosi

ty (m

Pas)

Temperature (ºC)

0.25 vol%0.20 vol%0.15 vol%0.10 vol%0.05 vol%water

Fig. 9. Viscosity of SWCNT nanofluid as a function of temperature and particlevolume concentrations.

0.8

0.9

1.0

1.1

1.2

1.3

1.4

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Rel

ativ

e vi

scos

ity

Volume concentration (vol%)

Experimental resultBrenner & Condiff [45]Timofeeva et al. eq. [44]

Fig. 10. Experimental results of relative viscosity compared to existing correlations.

0.01.02.03.04.05.06.07.08.09.0

0 50 100 150 200 250 300 350

Shea

r stre

ss (D

/cm

2 )

Shear rate (1/s)

0.25 Vol%0.20 Vol%0.15 Vol%0.10 Vol%0.05 Vol%

Fig. 12. Rheological behavior of SWCNT nanofluids (shear stress as a function ofshear rate).

2.00

2.50

3.00

3.50

4.00

4.50

10 20 30 40 50 60

Spec

ific

heat

(kJ/

kg

ºC )

Temperature (ºC)

water0.05 vol%0.10 vol%0.15 vol%0.20 vol%0.25 vol%

Fig. 13. Specific heat of SWCNT nanofluid as a function of temperature and particlevolume concentrations.

M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871 869

The experimental values are then compared to the existing cor-relations at Fig. 10. The experimental values are observed to behigher than theoretical values of Brenner and Condiff [45] as wellas Timofeeva et al. [44]. The use of SDS surfactant to preventagglomeration due to the hydrophobic in nature of SWCNTnanoparticles could be the reason for higher viscosity whereasthe models were developed to calculate the viscosity of suspensionwithout surfactants.

3.3. Rheology

Rheological behavior of nanofluids has significant impact on thestability and flow behavior of the fluid. Fig. 11 represents the vis-cosity of SWCNT nanofluids of different concentration as a functionof share rate. A shear thinning behavior is observed for all nanopar-ticle concentrations resulting in progressive decrease in viscositywith increasing share rate. A shear thinning behavior can beoccurred due to de-agglomeration of clustered nanotubes orrealignment in the direction of the shearing force which decreasesthe resistance to the flow and therefore, resulting in less viscousfluid. Garg et al. [24] and Nanda et al. [27] have reported similarshear thinning behavior for MWCNT and SWCNT nanofluidsrespectively.

The relation between share rate and share stress for differentvolume concentrations of nanoparticles at 20 �C was studied inorder to verify the rheological behaviors of SWCNT nanofluids.The rheology of SWCNT nanofluid is demonstrated at Fig. 12. Theshear stresses of the fluid increase almost linearly with increasingshear rate for all volume concentrations.

3.4. Specific heat

The effect of temperature on specific heat of SWCNT nanofluidsat different volume concentrations is shown in Fig. 13. The resultsshowed that the specific heat of SWCNT nanofluids increased withtemperature, for instance the values of specific heat obtained in the

1.01.21.41.61.82.02.22.42.62.8

0 50 100 150 200 250 300

Vis

cosi

ty (m

Pas)

Shear rate (1/s)

0.05 Vol%0.10 Vol%0.15 Vol%0.20 Vol%0.25 Vol%

Fig. 11. Rheological behavior of SWCNT nanofluids (viscosity as a function of shearstress).

experiment for 0.05 vol% are 3.66, 3.72, 3.79, 3.83, and 3.90 kJ/kg �Cat temperature 20, 30, 40, 50, and 60 �C respectively. However, thespecific heat decreased gradually with increasing volume concen-tration, means lower heat is required to increase the temperatureof the nanofluid with higher particle volume concentration.According to the experimental results, when the temperature isfixed at 20 �C, the specific heat values are 3.66, 3.43, 3.27, 3.11,and 2.97 kJ/kg �C for volume concentrations of 0.05, 0.1, 0.15, 0.2,and 0.25 vol% respectively. It is also observed that the specific heatof SWCNT nanofluid is lower compared to water. It occurredbecause when the specific heat of nanoparticles is less than thebasic liquid, the specific heat of suspension decreases [58].

Fig. 14 shows the comparison between the experimental resultsof specific heat and the theoretical results obtained by using Eq.(8). The experimental values of specific heat are found to be lowercompared to the theoretical values and the average deviation isfound to be 14.08%. There is no available equation in literature tocalculate the specific heat of carbon nanotube nanofluids ornanofluids containing rod like particles. Therefore, deviation isobserved as Eq. (8) is valid to calculate the specific heat of anyhomogenous mixtures and ignored the crucial factors used in cur-rent experiment such as surfactant effect, particle size and aspectratio of carbon nanotube nanoparticles.

2.0

2.5

3.0

3.5

4.0

4.5

0.00 0.05 0.10 0.15 0.20 0.25 0.30

Spec

ific

heat

(kJ/

kg ºC

)

Volume concentration (vol%)

Theoretical results

Experimental result

Fig. 14. Experimental results of specific heat compared to existing correlations.

870 M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871

4. Conclusion

This study investigated the stability and thermo–physical prop-erties of SWCNT nanofluids. The stability of SWCNT nanofluids isanalyzed with zeta potential and observed good stability due tohigh electrostatic repulsion force. The thermo–physical propertiesof SWCNT nanofluids were measured and compared to the existingmodels. The thermal conductivity enhanced from 2.84% to 36.39%compared to water for 0.05–0.25 vol% and 20–60 �C. Results indi-cate that the viscosity of the nanofluids decreased with theincrease of temperature however increased with particles loading.The viscosity increased from 17.76% to 82.01% for the same exper-imental conditions. The rheological study explained the shear thin-ning behavior of the nanofluids. The specific heat of the fluidsincreased from 6.73% to 28.96% for temperature and volume con-centrations ranging from 20 to 60 �C and 0.05–0.25 vol% respec-tively, however the specific heat of nanofluids are lower thanwater. From the experimental results it can be seen that SWCNTnanofluids have a wide prospect as working fluid to optimize theheat transfer performance.

Acknowledgements

The authors are thankful to the Ministry of Education, Malaysiafor the financial supports of this study, under the project at UM.C/HIR/MOHE/ENG/40, University of Malaya, Kuala Lumpur, Malaysia.The authors would allow themselves to appreciate the enormoussupport of different laboratories in University Malaya speciallyEnergy Lab and Nano Combicat personnel for their outstandingsupports.

References

[1] S. Murshed, C. Nieto de Castro, Superior thermal features of carbon nanotubes-based nanofluids – a review, Renewable Sustainable Energy Rev. 37 (2014)155–167.

[2] I. Capek, Dispersions Based on Carbon Nanotubes-Biomolecules Conjugates,INTECH Open Access Publisher, 2011.

[3] S. Harish, K. Ishikawa, E. Einarsson, S. Aikawa, T. Inoue, P. Zhao, M. Watanabe,S. Chiashi, J. Shiomi, S. Maruyama, Temperature dependent thermalconductivity increase of aqueous nanofluid with single walled carbonnanotube inclusion, Mater. Express 2 (3) (2012) 213–223.

[4] A. Nasiri, M. Shariaty-Niasar, A. Rashidi, R. Khodafarin, Effect of CNT structureson thermal conductivity and stability of nanofluid, Int. J. Heat Mass Transfer 55(5) (2012) 1529–1535.

[5] R. Sadri, G. Ahmadi, H. Togun, M. Dahari, S.N. Kazi, E. Sadeghinezhad, N. Zubir,An experimental study on thermal conductivity and viscosity of nanofluidscontaining carbon nanotubes, Nanoscale Res. Lett. 9 (1) (2014) 151.

[6] H. Xie, L. Chen, Review on the preparation and thermal performances of carbonnanotube contained nanofluids, J. Chem. Eng. Data 56 (4) (2011) 1030–1041.

[7] T.-P. Teng, C.-C. Yu, Heat dissipation performance of MWCNTs nano-coolant forvehicle, Exp. Therm. Fluid Sci. 49 (2013) 22–30.

[8] S. Choi, Z. Zhang, W. Yu, F. Lockwood, E. Grulke, Anomalous thermalconductivity enhancement in nanotube suspensions, Appl. Phys. Lett. 79 (14)(2001) 2252–2254.

[9] H. Xie, H. Lee, W. Youn, M. Choi, Nanofluids containing multiwalled carbonnanotubes and their enhanced thermal conductivities, J. Appl. Phys. 94 (8)(2003) 4967–4971.

[10] A. Nasiri, M. Shariaty-Niasar, A. Rashidi, A. Amrollahi, R. Khodafarin, Effect ofdispersion method on thermal conductivity and stability of nanofluid, Exp.Therm. Fluid Sci. 35 (4) (2011) 717–723.

[11] R. Kathiravan, R. Kumar, A. Gupta, R. Chandra, P. Jain, Pool boilingcharacteristics of multiwalled carbon nanotube (CNT) based nanofluids overa flat plate heater, Int. J. Heat Mass Transfer 54 (5) (2011) 1289–1296.

[12] S.S. Park, N.J. Kim, A study on the characteristics of carbon nanofluid for heattransfer enhancement of heat pipe, Renewable Energy 65 (2014) 123–129.

[13] M. Karami, A. Bahabadi, S. Delfani, A. Ghozatloo, A new application of carbonnanotubes nanofluid as working fluid of low-temperature direct absorptionsolar collector, Sol. Energy Mater. Sol. Cells 121 (2014) 114–118.

[14] S.S. Chougule, A.T. Pise, P.A. Madane, Performance of nanofluid-charged solarwater heater by solar tracking system, in: Advances in Engineering, Scienceand Management (ICAESM), 2012 International Conference on, IEEE, 2012, pp.247–253.

[15] B. Ruan, A.M. Jacobi, Heat transfer characteristics of multiwall carbonnanotube suspensions (MWCNT nanofluids) in intertube falling-film flow,Int. J. Heat Mass Transfer 55 (11) (2012) 3186–3195.

[16] P. Estellé, S. Halelfadl, T. Maré, Thermal conductivity of CNT water basednanofluids: experimental trends and models overview, J. Therm. Eng. 1 (2)(2015) 381–390.

[17] K. Wusiman, H. Jeong, K. Tulugan, H. Afrianto, H. Chung, Thermal performanceof multi-walled carbon nanotubes (MWCNTs) in aqueous suspensions withsurfactants SDBS and SDS, Int. Commun. Heat Mass Transfer 41 (2013)28–33.

[18] V. Kumaresan, S.M.A. Khader, S. Karthikeyan, R. Velraj, Convective heattransfer characteristics of CNT nanofluids in a tubular heat exchanger ofvarious lengths for energy efficient cooling/heating system, Int. J. Heat MassTransfer 60 (2013) 413–421.

[19] S. Harish, K. Ishikawa, E. Einarsson, S. Aikawa, S. Chiashi, J. Shiomi, S.Maruyama, Enhanced thermal conductivity of ethylene glycol with single-walled carbon nanotube inclusions, Int. J. Heat Mass Transfer 55 (13) (2012)3885–3890.

[20] R. Walvekar, I.A. Faris, M. Khalid, Thermal conductivity of carbon nanotubenanofluid—Experimental and theoretical study, Heat Transfer Asian Res. 41 (2)(2012) 145–163.

[21] V. Kumaresan, R. Velraj, Experimental investigation of the thermo-physicalproperties of water–ethylene glycol mixture based CNT nanofluids,Thermochim. Acta 545 (2012) 180–186.

[22] S.J. Aravind, P. Baskar, T.T. Baby, R.K. Sabareesh, S. Das, S. Ramaprabhu,Investigation of structural stability, dispersion, viscosity, and conductive heattransfer properties of functionalized carbon nanotube based nanofluids, J.Phys. Chem. C 115 (34) (2011) 16737–16744.

[23] Y.J. Kim, H. Ma, Q. Yu, Plasma nanocoated carbon nanotubes for heat transfernanofluids, Nanotechnology 21 (29) (2010) 295703.

[24] P. Garg, J.L. Alvarado, C. Marsh, T.A. Carlson, D.A. Kessler, K. Annamalai, Anexperimental study on the effect of ultrasonication on viscosity and heattransfer performance of multi-wall carbon nanotube-based aqueousnanofluids, Int. J. Heat Mass Transfer 52 (21) (2009) 5090–5101.

[25] L. Chen, H. Xie, Silicon oil based multiwalled carbon nanotubes nanofluid withoptimized thermal conductivity enhancement, Colloids Surf. A 352 (1) (2009)136–140.

[26] W. Jiang, G. Ding, H. Peng, Measurement and model on thermal conductivitiesof carbon nanotube nanorefrigerants, Int. J. Therm. Sci. 48 (6) (2009) 1108–1115.

[27] J. Nanda, C. Maranville, S.C. Bollin, D. Sawall, H. Ohtani, J.T. Remillard, J. Ginder,Thermal conductivity of single-wall carbon nanotube dispersions: role ofinterfacial effects, J. Phys. Chem. C 112 (3) (2008) 654–658.

[28] J. Glory, M. Bonetti, M. Helezen, M. Mayne-L’Hermite, C. Reynaud, Thermal andelectrical conductivities of water-based nanofluids prepared with longmultiwalled carbon nanotubes, J. Appl. Phys. 103 (9) (2008) 094309.

[29] L. Chen, H. Xie, Y. Li, W. Yu, Nanofluids containing carbon nanotubes treated bymechanochemical reaction, Thermochim. Acta 477 (1) (2008) 21–24.

[30] A. Amrollahi, A. Hamidi, A. Rashidi, The effects of temperature, volume fractionand vibration time on the thermo-physical properties of a carbon nanotubesuspension (carbon nanofluid), Nanotechnology 19 (31) (2008) 315701.

[31] Y. Hwang, J. Lee, C. Lee, Y. Jung, S. Cheong, C. Lee, B. Ku, S. Jang, Stability andthermal conductivity characteristics of nanofluids, Thermochim. Acta 455 (1)(2007) 70–74.

[32] Y. Ding, H. Alias, D. Wen, R.A. Williams, Heat transfer of aqueous suspensionsof carbon nanotubes (CNT nanofluids), Int. J. Heat Mass Transfer 49 (1) (2006)240–250.

[33] Y. Hwang, Y. Ahn, H. Shin, C. Lee, G. Kim, H. Park, J. Lee, Investigation oncharacteristics of thermal conductivity enhancement of nanofluids, Curr. Appl.Phys. 6 (6) (2006) 1068–1071.

[34] M. Assael, I. Metaxa, K. Kakosimos, D. Constantinou, Thermal conductivity ofnanofluids–experimental and theoretical, Int. J. Thermophys. 27 (4) (2006)999–1017.

[35] Y. Hwang, H. Park, J. Lee, W. Jung, Thermal conductivity and lubricationcharacteristics of nanofluids, Curr. Appl. Phys. 6 (2006) e67–e71.

[36] M. Assael, I. Metaxa, J. Arvanitidis, D. Christofilos, C. Lioutas, Thermalconductivity enhancement in aqueous suspensions of carbon multi-walledand double-walled nanotubes in the presence of two different dispersants, Int.J. Thermophys. 26 (3) (2005) 647–664.

[37] M.-S. Liu, M.C.-C. Lin, I.-T. Huang, C.-C. Wang, Enhancement of thermalconductivity with carbon nanotube for nanofluids, Int. Commun. Heat MassTransfer 32 (9) (2005) 1202–1210.

[38] M. Assael, C.-F. Chen, I. Metaxa, W. Wakeham, Thermal conductivity ofsuspensions of carbon nanotubes in water, Int. J. Thermophys. 25 (4) (2004)971–985.

[39] ISO, 14887: 2000(E): Sample preparation dispersing procedures for powders inliquids, International Organization for Standardization, Geneva, Switzerland,2000.

[40] Y. Liu, Y. Liu, P. Hu, X. Li, R. Gao, Q. Peng, L. Wei, The effects of graphene oxidenanosheets and ultrasonic oscillation on the supercooling and nucleationbehavior of nanofluids PCMs, Microfluid. Nanofluid. 18 (1) (2015) 81–89.

[41] J.-H. Lee, K.S. Hwang, S.P. Jang, B.H. Lee, J.H. Kim, S.U. Choi, C.J. Choi, Effectiveviscosities and thermal conductivities of aqueous nanofluids containing lowvolume concentrations of Al2O3 nanoparticles, Int. J. Heat Mass Transfer 51(11) (2008) 2651–2656.

[42] C.-W. Nan, Z. Shi, Y. Lin, A simple model for thermal conductivity of carbonnanotube-based composites, Chem. Phys. Lett. 375 (5) (2003) 666–669.

[43] H. Patel, K. Anoop, T. Sundararajan, S.K. Das, Model for thermal conductivity ofCNT-nanofluids, Bull. Mater. Sci. 31 (3) (2008) 387–390.

M.A. Sabiha et al. / International Journal of Heat and Mass Transfer 93 (2016) 862–871 871

[44] E.V. Timofeeva, J.L. Routbort, D. Singh, Particle shape effects on thermophysicalproperties of alumina nanofluids, J. Appl. Phys. 106 (1) (2009) 014304.

[45] D. Condiff, H. Brenner, Transport mechanics in systems of orientable particles,Phys. Fluids 12 (3) (1969) 539–551 (1958–1988).

[46] L. Syam Sundar, K. Sharma, Thermal conductivity enhancement ofnanoparticles in distilled water, International, J. Nanopart. 1 (1) (2008) 66–77.

[47] L. Yu-Hua, Q. Wei, F. Jian-Chao, Temperature dependence of thermalconductivity of nanofluids, Chin. Phys. Lett. 25 (9) (2008) 3319.

[48] P. Keblinski, R. Prasher, J. Eapen, Thermal conductance of nanofluids: is thecontroversy over?, J Nanopart. Res. 10 (7) (2008) 1089–1097.

[49] J. Gao, R. Zheng, H. Ohtani, D. Zhu, G. Chen, Experimental investigation of heatconduction mechanisms in nanofluids. Clue on clustering, Nano Lett. 9 (12)(2009) 4128–4132.

[50] W. Evans, R. Prasher, J. Fish, P. Meakin, P. Phelan, P. Keblinski, Effect ofaggregation and interfacial thermal resistance on thermal conductivity ofnanocomposites and colloidal nanofluids, Int. J. Heat Mass Transfer 51 (5)(2008) 1431–1438.

[51] C. Pang, J.-Y. Jung, J.W. Lee, Y.T. Kang, Thermal conductivity measurement ofmethanol-based nanofluids with Al2O3 and SiO2 nanoparticles, Int. J. HeatMass Transfer 55 (21) (2012) 5597–5602.

[52] Y. Feng, B. Yu, P. Xu, M. Zou, The effective thermal conductivity of nanofluidsbased on the nanolayer and the aggregation of nanoparticles, J. Phys. D Appl.Phys. 40 (10) (2007) 3164.

[53] H. Masuda, A. Ebata, K. Teramae, N. Hishinuma, Alteration of thermalconductivity and viscosity of liquid by dispersing ultra-fine particles, NetsuBussei 7 (4) (1993) 227–233.

[54] W. Duangthongsuk, S. Wongwises, Measurement of temperature-dependentthermal conductivity and viscosity of TiO2-water nanofluids, Exp. Therm. FluidSci. 33 (4) (2009) 706–714.

[55] N. Jamshidi, M. Farhadi, D. Ganji, K. Sedighi, Experimental investigation onviscosity of nanofluids, Int. J. Eng. 25 (3) (2012) 201–209.

[56] S. Halelfadl, P. Estellé, B. Aladag, N. Doner, T. Maré, Viscosity of carbonnanotubes water-based nanofluids: influence of concentration andtemperature, Int. J. Therm. Sci. 71 (2013) 111–117.

[57] I. Mahbubul, R. Saidur, M. Amalina, Latest developments on the viscosity ofnanofluids, Int. J. Heat Mass Transfer 55 (4) (2012) 874–885.

[58] I. Shahrul, I. Mahbubul, S. Khaleduzzaman, R. Saidur, M. Sabri, A comparativereview on the specific heat of nanofluids for energy perspective, RenewableSustainable Energy Rev. 38 (2014) 88–98.