ICHMT-03350; No of Pages 10

International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

journalhomepage:www.elsevier.com/locate/ichmt

1 Viscosity of nanofluids: A review of recent experimental studies☆

a b b c d
2Q1 Kazem Bashirnezhad , Shahab Bazri , Mohammad Reza Safaei , Marjan Goodarzi , Mahidzal Dahari ,
b, e f,
3 Omid Mahian , Ahmet Selim Dalkõlõça , Somchai Wongwises
4 a
Department of Mechanical Engineering, Islamic Azad University, Mashhad Branch, Mashhad, Iran

5 b
Young Researchers and Elite Club, Islamic Azad University, Mashhad Branch, Mashhad, Iran

6 c
Faculty of Computer Science and Information Technology, Department of Software Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia

7 d
Faculty of Engineering, Department of Mechanical Engineering, University of Malaya, Kuala Lumpur 50603, Malaysia

8 e
Heat and Thermodynamics Division, Department of Mechanical Engineering, Faculty of Mechanical Engineering, Yildiz Technical University (YTU), Besiktas, Istanbul 34349, Turkey

9 f
Fluid Mechanics, Thermal Engineering and Multiphase Flow Research Laboratory (FUTURE), Department of Mechanical Engineering, Faculty of Engineering,

10 King Mongkut's University of Technology Thonburi, Bangmod, Bangkok 10140, Thailand

11

12 article info abstract

13
14 Available online xxxx During the past decade, nanotechnology with its rapid development has grabbed the attention of scientists, 20 scholars, and
198765 engineers. Nanofluids are one of the surprising outcomes of this technology that could increase 21 the efficiency of thermal
30 Keywords: systems remarkably. Nanofluids containing solid nanoparticles have a higher viscosity 22 than common working fluids; hence,
31 Nanofluids measuring the viscosity is necessary for designing thermal systems and es- 23 timating the required pumping power. In the
32 Viscosity
33 Experimental studies current review study, an attempt has been made to cover the latest 24 experimental studies performed on the viscosity of
nanofluids. An experimental investigation is very vital for the 25 analysis since the theoretical models usually underestimate the
nanofluid viscosity. Through experiments, the 26 real effects of volume fraction, temperature, particle size, and shape on the
viscosity of nanofluids will be 27
determined. 28 © 2016 Published by Elsevier Ltd. 29

3754
36

38 1. Introduction viscosity [7]. Many parameters affect the nanofluid viscosity including 54 preparation
method, base fluid type, temperature, particle size and 55 shape, volume concentration,
acidity (pH value), shear rate, surfactants, 56 and particle aggregation [2]. The aim of this
39 The basic idea of particle-dispersed fluid can be traced back to paper is to review the latest 57 experimental studies conducted on viscosity of nanofluids
40 Maxwell's study in 1873 [1]. Afterward, in 1904, he indicated that the
by consider- 58
41 molecule that has nano-scaled diameter could be considered as starting
ing the abovementioned parameters. 59
42 the striking technology of nano. From that time up to now, the rapid
2. Nanofluid definition and applications 60
43 development of nanotechnology has been seen in all aspects [2,3].
44 Accordingly, nanofluid introduced as a new term, indeed, defined for
45 the first time as liquids having nanometer-sized particles, emerged
Nanofluids, as an innovative material, are solid–liquid mixtures in 61 which
46 after Choi innovative work in 1995, which has been the pioneer one
the solid particles have usually a size more than 1 nm and less 62 than 100 nm
47 [2,4–7]. Before design a thermal system in which a nanofluid is the
[8–9]. The particles could be metal particles such as Al, 63 Cu, and Ni; oxides
48 working fluid, it is necessary to know the thermophysical properties
49 of nanofluid including thermal conductivity, viscosity, heat capacity, such as Al2O3, TiO2, CuO, SiO2, Fe2O3, and Fe3O4; 64 and some other
50 and density. Among the nanofluid properties, viscosity is an important compound materials such as AlN, SiC, and graphene 65 [2,10–11]. Nanofluids
51 property since it indicates the fluid's resistance. With increasing the vis- are new generation of heat transfer fluids with 66 an anomalous behavior, which
52 cosity, the required energy for pumping and mixing increases. Also, are taken from stably suspending colloi- 67 dal nanoparticles in the original fluids
53 pumping power and pressure drop are two key factors that depend on (conventional heat transfer liq- 68 uids). Nanofluids can be applied in various
devices and systems such 69 as cooling electronic components, transportation,
industrial cooling, 70 heating buildings, medical systems, reducing pollution,
☆ Communicated by Dr. W.J. Minkowycz. nuclear systems 71 cooling, atomic engineering, space and defense, energy
Corresponding authors. storage, solar 72 absorption, friction reduction, magnetic sealing, antibacterial
E-mail addresses: omid.mahian@gmail.com (O. Mahian), somchai.won@kmutt.ac.th
activity, 73
(S. Wongwises).

http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
0735-1933/© 2016 Published by Elsevier Ltd.
Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 2 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

T1:1 Nomenclature up to 4.0% to be useful for reasonable particle volume concentrations 114
[23]: 115
T1:2 μr Relative viscosity
" #
μ
T1:3 μeff Effective viscosity nf ¼ 1 : ð2Þ
T1:4 μbf Base fluid (liquid) viscosity μ ð1−ϕÞ
2:5
bf
T1:5 μnf Nanofluid viscosity 117

T1:6 ϕ Volume fraction 3- Krieger and Dougherty (K-D) (1959): Krieger [26] derived a semi-empirical
T1:7 ϕm Maximum volume fraction relation for the shear viscosity that covered the full 118 range of particle
T1:8 T Temperature volume concentration, known as K-D model [27]: 119
:
T1:9 γ Shear rate
μnf ¼ μbf 1− ϕ −½η&ϕm
T1:110 ra Particle aggregate
ð3Þ
ϕm
where φm is the maximum particle fraction, which varies from 121 0.495 to
74 nano-drug delivery, intensify micro reactors, microbial fuel cells, and so 0.54 under quiescent conditions and is approximately
75 on [12–15]. 0.605 at high shear rates. 122
4- Frankel (1967): In 1967, Franken and Acrivos [28] developed a 123
76 3. Preparation and characterization mathematical expression [29]: 124

2 3
77 3.1. Preparation of nanofluids μ 6
1
ϕ 3 7
¼8 ðÞ

μ ϕ
bf ϕm −ϕ 3

nf 96 m 7 4
78 There are some significant primary steps before using nanofluids in a
ð 1 Þ

1 :

6 7
79 specific application, first of all, is preparation step, and latter is the mea- ϕm 3

80 surement of properties to estimate the well verification of nanofluids' 6 7

4 5
81 performance. In other words, the method of nanofluids' preparation 126
82 has a momentous effect on the properties. There are two fundamental 5- Nielsen (1970): Nielsen [30] proposed the power law model, in 1970, to
83 methods for the fabrication of nanofluids: one-step method or chemical determine the viscosity of nanofluids corresponding to 127 the particle
84 synthesis technique or single-step direct evaporation method in which volume concentration more than 0.02, and here is the 128
85 particles are formed directly in the base liquid. Later is two-step method suggested mathematical equation [27,31]: 129
86 in which the nanoparticles are synthesized by different methods and μnf ¼ ð1 þ 1:5ϕÞe
ϕ
ð5Þ
87 then are dispersed into the base liquid. Two most imperative character-
ð1−ϕm Þ μnf :
88 istics of the prepared nanofluid are (i) stable suspension without 131
89 sedimentation during a long time and (ii) non-agglomeration [16–19]. 6- Lundgren (1972): Lundgresn [32] proposed the following formula
to predict the suspension viscosity under the form of a Taylor series 132
90 3.2. Characterization techniques in terms of [29]: 133
25 2 3 ð6Þ
μnf ¼ μbf 1 þ 2:5ϕ þ ϕ þf ϕ :
91 The methods for characterization of the structural properties of 4

92 nanoparticles to estimate the chemical nature, size, and the morphology

93 as well as the size of agglomerations are called characterization tech- 135
94 niques. Here are all usable methods in most of the studies, including 7- Batchelor (1977): Batchelor [33] considered the effect due to the Brownian
95 transmission electron microscopy (TEM) using wet-TEM technique to motion of particles for an isotropic suspension of rigid 136
96 assess the dispersion state, x-ray diffraction (XRD), vibration sample and spherical particles [29]: 137
97 magnetometer (VSM), energy-dispersive x-ray spectroscopy (EDX), μ ¼h i ð7Þ
bf 1 þ 2:5ϕ þ 6:5ϕ :
98 thermal analysis TG-DTA, UV–Vis spectroscopy, Fourier transform infra- μ
nf 2
99 red spectroscopy (FTIR), infrared absorption spectroscopy, scanning 139
100 electron microscopy (SEM), and inductively coupled plasma–optical
101 emission spectroscopy (ICP-OES) [6,17,20]. 8- Graham (1981): Graham [34] developed a comprehensive outline
of Frankel–Acrivos model by introducing particle radius and 140
102 4. Predicted models and theories of nanofluids viscosity inter-particle spacing which has good agreement with Einstein's 141
formula for small φ. The model is expressed as follows [29]: 142
103 There are some existing formulas and models to estimate the viscos- 0 2 1 31 : ð8Þ
μnf ¼ μ f 1 þ 2:5ϕ þ 4:5
104 ity of nanofluids.
105 1- Einstein model (1906): The Einstein model [21–22] is the pioneer 2
h h h
106 theory and mostly referred equation to predict the viscosity of B 6 2 1 dp 7C
þ dp þ

6
B dp 7C
107 nanofluids, which predicts and assumes only very low nanoparticle @ 4 5A 144
108 concentrations (φ ≤ 2%) and linearly viscous fluid having dilute,
109 suspended, and spherical particles for. The model is stated as [23]: 9- In 1981, Kitano et al. [35] proposed a simple correlation to predict
the viscosity of a two-phase suspension [27]: 145
μ ¼ ½1 þ 2:5ϕ&: ð1Þ
: ð9Þ
nf μnf ¼ μf
μ
bf ϕ 2
h 1− ϕ i
m
147
111 In 1911, Einstein also presented a correlation of zero-shear 10- White (1991): Notice that all above correlations or models were
viscosity [24]. developed only based on viscosity as a function of volume fraction; 148
112 2- Brinkman (1952) (by modifying Einstein Model): Brinkman [25] ex- to clarify; there is no consideration of temperature dependence. 149
113 tended and modified the Einstein's model to volume concentration However, viscosity has a strong function with the temperature; in 150

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 3

151 other words, fluids have higher viscosity near their freezing point
152 and moderately low viscosities near their boiling point. White
153 [36] presented a model for pure fluids between viscosity and
154 temperature which is given by [27]:
155 2

ln μ 0 ≈ aþb
T þc
T :
ð10Þ

μ
bf T0 T0

157
Here the viscosity and temperature parameters with zero index are
158 related to reference values; a, b, and c vary from fluid to fluid that
159 were given by White. For example, for water:

161 a ¼ −2:10; b ¼ −4:45; c ¼ 6:55 5.

Experiments

162 Experimental studies are very significant for the analysis and valida-
163 tion of theoretical models proposed by researchers. These experimental
164 studies over the viscosity of nanofluids reveal the rheological behavior
165 that has equal importance in experimental as well as practical heat
166 transfer applications. Several experiments on this issue show that Fig. 1. Comparison of measured viscosity with proposed correlation. (Reprinted from Sundar et
167 some parameters such as volume fraction, temperature, particle shape al. [38], with permission from Elsevier.)
168 and size, surfactants, and acidity (pH) have direct impacts on the viscos-
169 ity of nanofluids. In this section, the attempt has been made to classify Zyla et al. [39] presented results of new data of rheological experi-ments on
170 all collective papers in order of the usage materials in each of them. viscosity under anisotropic pressure and in the electric field 205 of diethylene
glycol-based MgAl2O4 nanofluids. They had prepared 206 nanofluids in a two-step
171 5.1. SAl2O3 and other alumina nanoparticles preparation method. The dynamic viscosity 207 of nanofluids with various mass
concentrations of nanoparticles was 208 measured in the range of shear rates from
172 To begin with, most of the papers usually have disbursed the effec- −1 −1
10 s to 1000 s in the con- 209 stant temperature under the pressure of 7.5
173 tive consequences of alumina base nanofluids in this field. Aladag
MPa. It showed an increase 210 in dynamic viscosity in comparison with the
174 et al. [37] investigated experimental effects of temperature and shearing
results obtained at com- 211 mon atmospheric pressure, which did not show any
175 time on viscosity of Al2O3/water-based nanofluids at low concentration
change in the na- 212 ture of the viscosity curve. They measured the influence of
176 and low temperatures. They collected the data of viscosity by using a
177 stress-controlled rheometer equipped with parallel plate geometry pressure 213 and electric field on the viscosity of MgAl2O4-DG nanofluids with
178 under up and down shear stress ramp. The experiments showed that the 214 use of a HAAKE MARS 2 rheometer. They stated that the electric field
179 when the stress is gradually loaded and unloaded, Al2O3 water-based 215 has not any effect on the rheological properties of the MgAl2O4-DG 216
180 nanofluids exhibited hysteresis behavior, also depending on shearing nanofluids. It is so valuable information for the related industrial 217 applications,
181 time. It also proved that the nanofluid suspensions indicated either in addition to the fact that one can use such nanofluids 218 without worrying about
182 Newtonian or non-Newtonian behavior, depending on shear rate. changing in viscosity of the material under 219 the presence of an electric field.
183 They investigated Al2O3 water-based nanofluid was non-Newtonian Few numbers of studies have focused 220 on nanorefrigerant. John et al. [40]
184 within the range of low temperatures; they also studied carbon demonstrated enhanced viscosity 221 in comparison with the base fluid and
185 nanotube/water-based nanofluid that behaves as Newtonian fluid at investigated the variation of vis- 222 cosity of ethylene glycol-based alumina and
186 high shear rate. Sundar et al. [38] investigated viscosity experiments copper oxide nanofluids, 223 considering particle volume fraction and
temperature. The effective 224 viscosity of nanofluids was much higher than the
187 of nanofluids by dispersing Al2O3 nanoparticles in different base fluids
values predicted by 225 Einstein and Batchelor model. There was no observation
188 such as 20:80%, 40:60% and 60:40% by mass of ethylene glycol (EG)
about the con- 226 sistent tendency for the temperature dependence of relative
189 and water (W) mixtures in temperatures between 0 °C and 60 °C and
effective 227 viscosity of nanofluids. According to their paper, the stability of 228
190 in volume concentrations between 0.3% and 1.5%. The viscosity enhance-
nanofluids decreased when the particle volume concentration in- 229 creased and
191 ment for 1.5% particle concentration of 20:80% EG/W nanofluid is 1.37
enhanced with increasing in viscosity of the base fluid. 230 Also, Utomo et al. [41]
192 times, for 40:60% EG/W nanofluid is 2.75 times, and for 60:40% EG/W
investigated the viscosity of water-based alumina 231 nanofluids. Compared to the
193 nanofluid is 2.58 times at a temperature of 0 °C, respectively, compared prediction of Einstein–Batchelor model, 232 viscosity of alumina nanofluids was
194 to base fluid. Nanofluid prepared in higher viscosity base fluid exhibits higher than them due to aggrega- 233 tion. Experimental heat transfer coefficients
195 more enhancement compared to low viscosity base fluid. They men- in alumina nanofluids, as 234 well as related wall temperatures, agree within
196 tioned the traditional models, which have been presented by Einstein, ±10% with the numer- 235 ical simulations' values employing homogeneous flow
197 Brinkman, and Batchelor (Formulae 1, 2, and 7). The Einstein models model with in- 236 fluential thermophysical properties of nanofluids. On the other
198 failed to predict the viscosity of nanofluids with the influence of temper- hand, 237 some researchers such as Nguyen et al. [42] have hinted the augmenta-
199 atures. Therefore, some correlations have been proposed for the estima- 238 tion of viscosity from the specific quantity of some parameters. Nguyen 239 et
200 tion of nanofluids viscosity based on 135 data points. The following al. [42] investigated experimentally the effect of both the tempera- 240 ture (up to
201 correlation was developed by assuming the nanofluid viscosity increases 75 °C) and the particle size (36 nm and 47 nm) on the dy- 241 namic viscosity of
202 exponentially with particle concentrations (Fig. 1). Al2O3/water nanofluids. They measured nanofluid 242 dynamic viscosity by using
a ‘piston-type’ calibrated viscometer based 243 on the Couette flow inside a
μ
nf ¼AeBϕ cylindrical measurement chamber. It is 244 composed of two magnetic coils,
μ which is installed inside a 316 stainless 245 steel sensor body. Those coils are
bf
used to generate a magnetically- 246
A¼0:9396 and B¼24:16; 20 : 80%EG=W nanofluid ð11Þ
A¼0:9299 and B¼67:43; 40 : 60%EG=W nanofluid
204 A¼1:1216 and B¼77:56; 60 : 40%EG=W nanofluid

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 4 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

247 induced force on a cylindrical piston that moves back and forth over a size. Moreover, Masoumi et al. [46] introduced a new equation in 313 order to
248 distance of 5 mm. The elapsed time corresponding to a round trip of measure the viscosity of nanofluids with respect to the 314 Brownian motion of
249 the piston can then be measured by alternatively powering the coils nanoparticles by using limited experimental data 315 consisting of 13 and 28 nm-
250 with a constant force. According to their work, such measurement is ac- sized Al2O3 nanoparticles in water. Their 316 equation presented the nanofluid
251 curately related to the viscosity of the fluid sample contained in the viscosity as a function of the temper- 317 ature, the mean nanoparticle diameter,
252 chamber through an accurate calibration process. The system forces the nanoparticle volume frac- 318 tion, the nanoparticle density, and the base fluid
253 the fluid at the bottom to flow around the piston toward the sensor physical properties 319 with the basis datum of the relative velocity between the
254 opening while the piston is pulled toward the bottom of the measure- base fluid 320 and nanoparticles. They compared their model with the other
255 ment chamber. On the other hand, on the upward piston stroke, fresh theoreti- 321 cal models that are available in the literature, and, eventually, they
256 fluid is pulled around the piston to thoroughly mix the contents of the indi- 322 cated that their presented model had generally a higher accuracy. In 323
257 measurement chamber. Such a mixing effect ensures the consistency addition, based on their experiments, the maximum quantity of relative 324
258 of fluid temperature inside the chamber. Also, because the measure- viscosity for 28 nm-sized is higher than 13 nm-sized nanoparticle, at the 325 same
259 ment of the piston motion is in two directions, variations due to gravity volume fraction. Yiamsawas et al. [47] determined Al2O3 nanopar- 326 ticles'
260 or flow forces are annulled. In addition, by reason of the very small mass viscosity experimentally, which suspended in a mixture of ethyl- 327 ene
261 of the piston, the induced magnetic forces greatly exceed any distur- glycol/water (EG–water, 20/80 wt%). Those experiments were hold 328 at
262 bances due to vibrations. Because the fluid viscosity may considerably different volume concentrations between 0% and 4% and a tempera- 329 ture
263 vary with temperature, it is important to know the exact temperature range between 15 °C and 60 °C. Their results showed that the the- 330 oretical
264 of the measurement chamber. Hence, the temperature of the fluid models are not appropriate to predict nanofluids' viscosity. 331 According to their
265 sample inside the sensor chamber is continuously monitored using a study, a useful correlation was presented to predict 332 the viscosity of Al2O3
266 precision Platinum RTD that is internally mounted at the base of the nanoparticles suspended in the aforementioned 333 mixture for practical
267 chamber. The viscometer system was factory calibrated and delivered applications. In their correlation, the viscosity is a 334 function of volume fraction
268 ready for operation. They mentioned, according to the manufacturer, of nanoparticles, temperature, and the 335
269 the temperature accuracy and repeatability of the RTD probe were esti-
270 mated to be ±0.2 °C and ±0.1 °C, while the viscometer accuracy and re- viscosity of base fluid: 336
271 peatability were, respectively, ±1% and ±0.8% for the range of 0–20 cP B C D ð12Þ
μnf ¼Aϕ T μ f
272 (Fig. 3). According to their observation, for particle volume fractions
273 lower than 4%, 36 nm and 47 nm particle size alumina–water
274 nanofluids, viscosities are almost the same, and for higher particle frac- where the A, B, C, and D coefficients are listed for the Al2O3 nanoparticle 338 in
275 tions, 47 nm particle sizes, viscosity is obviously higher than those of Table 1. Although the power of temperature (C coefficient) in the above
276 36 nm size. Viscosities corresponding to water-oxide copper are the correlation is positive, they mentioned that the viscosity of base 339 fluid
277 highest among the nanofluids tested. Based on the results, the applica- decreases with an augmentation in temperature; thus, the viscos- 340 ity of
278 tion of Einstein's model and those derived from the linear fluid theory nanofluid totally decreases. The viscosity of the EG/water mixture 341
279 sounds not to be suitable for nanofluids. The hysteresis phenomenon is obtained by the following relation: 342
2 ð13Þ
280 over the viscosity measurement had been raised serious concerns μ f ¼ 0:0003T −0:0461T þ 2:3775
281 using of nanofluids for enhancement purposes of heat transfer. Some
282 papers have showed their results in a good agreement with the predict- 344
283 ed classical models. Chandrasekar et al. [43] reported experimental in- According to their study, they used a capillary tube viscometer (Cannon
284 vestigations and theoretical determination of the effective viscosity of Instrument, USA) to measure the nanofluids' viscosities. That 345 system has
285 Al2O3/H2O nanofluid with a nominal diameter of 43 nm at different vol- composed of a 0.75 mm viscometer tube with a measure- 346 ment range of 1.6–8
286 ume fractions ranging from 0.33% to 5%. The nanofluid had been pre- cSt. Fig. 4 shows a schematic view of the viscome- 347 ter. The tube is immersed
287 pared by synthesizing Al2O3 nanoparticles using microwave-assisted in a transparent temperature-controlled basin 348 connected to a temperature-
288 chemical precipitation method. After that, they dispersed in distilled controlled water tank. A stopwatch with an 349 accuracy of 0.01 s is used to
289 water using a sonicator. The viscosity of nanofluids increases with the record the time of passing the fluid from a 350 specified section of the viscometer,
290 nanoparticle volume fraction. Classical models are developed to predict and a thermometer with the accura- 351 cy of 0.1 °C is used to measure the
291 viscosity of nanofluids, which showed practically acceptable agreement temperature. The tests are repeated 352 three times for any volume concentration.
292 with the experimental results. Yiamsawas et al. [44] measured the vis- One can find that the 353 nanofluid is Newtonian or not by using the viscometer.
293 cosity of the prevalent nanofluid of Al2O3/water at high concentrations By referring to 354 Fig. 4, the sample is Newtonian if the efflux time in bulbs A
294 and high temperatures. The range of volume fraction of nanoparticles and B are ap- 355 proximately the same. Based on their observation, the
295 varied from 1% to 8% while the temperature was between 15 °C and nanofluids have a 356 Newtonian behavior for volume fractions less than 4%.
296 60 °C. They stated that the viscosity decreases with a growth in temper- They mentioned 357 that the maximum uncertainty was 1.62%, and average
297 ature. Also, they compared their observations with the theoretical and uncertainty was 358
298 experimental reports in the literature. Jarahnejad et al. [45] investigated 0.82% for viscosity. 359
5.2. Fe2O3 or other related oxide nanoparticles 360
299 the effect of temperature, volume concentration, and size of nanoparti-
300 cles and the addition of surfactants on dynamic viscosity of nanofluids
301 involving alumina (Al2O3) nanoparticles with water as a base fluid.
302 They measured viscosity by two viscometers, a capillary, and a falling Abareshi et al. [48] prepared magnetic nanoparticles of hematite, 361 α-Fe2O3,
303 ball, in the temperature ranging from 20 °C to 50 °C and the particle con- by the solvothermal method using Fe (NO 3)3 as a starting ma- 362 terial to make
304 centration of 3 to 14.3 wt.%. The results showed that the viscosity of ferrofluid. They characterized nanoparticles by x-ray dif- 363 fraction and
305 nanofluids decreases by increasing the temperature, similar to their transmission electronic microscope. They focused on the 364 rheological
306 base fluids. In addition, surfactants, which were used to enhance the properties of nanofluids of α-Fe2O3 smart-nanoparticles 365
307 shelf stability of nanofluids, most likely increase their viscosity. The
308 equations derived from the linear fluid theory such as Einstein and Table 1 t1:1
309 Batchelor, were not well-matched enough to predict viscosity, especial- Coefficients for correlations predicting viscosity [47]. t1:2
310 ly for solid concentration above 1.5 wt.%; on the other hand, the devel- Nanoparticle A B C D t1:3
311 oped Krieger–Dougherty model estimated viscosity of nanofluids with t1:4
Al2O3 0.891842 0.739192 0.099205 0.9844
312 good agreement in a specific range of solid particle size to aggregate

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 5

366 and glycerol (base fluid). Their experimental results showed that the (PG) was the weaker inter-molecular hydrogen bonding in PG. This 422 fact will
367 viscosity of these nanofluids increases with the augmentation of the influence by reducing the values of viscosity for nanoparticle 423 dispersions in
368 particle volume fraction and decreases with the increase of tempera- comparison with the base fluid. Lower temperatures 424 and higher volume
369 ture. Their results also showed that the α-Fe2O3/glycerol nanofluids concentration are the most announcements of 425 viscosity's decrease in the
370 are non-Newtonian shear-thinning and the shear viscosity strongly de- studies. The result of the reduction in viscos- 426 ity is 53% and 32% at 10 °C and
371 pends on the temperature. They compared the experimental data with 28 °C, respectively, for 2 vol% ZnO– 427 propylene glycol nanofluid in
372 some theoretical models and correlations; as a result, the measured comparison with pure propylene glycol. 428 At the high temperature where the
373 values of the effective viscosity of such nanofluids are underestimated hydrogen bonds in the base fluid 429 are substantially weaker, the viscosity of
374 by those theories. Pastoriza-Gallego et al. [49] have carried out the rhe- ZnO–PG dispersions is higher 430 than PG. On the other hand, Li et al. [53]
375 ological behavior of EG/Fe2O3 nanofluids (ethylene glycol-based showed highly agreement 431 with most of the previous work. They prepared
376 nanofluids containing hexagonal scalenohedral-shaped a-Fe2O3 or he- well-dispersed ethylene 432 glycol (EG)-based nanofluids containing ZnO
377 matite) nanoparticles for temperature of 303.15 K and particle mass nanoparticles with differ- 433 ent mass fractions between 1.75% and 10.5% by a
378 fraction up to 25% using a cone-plate Physica MCR rheometer. Based typical two-step meth- 434 od. Structural properties of the dry ZnO nanoparticles
379 on their tests, the studied nanofluids present non-Newtonian shear- were measured 435 by XRD and TEM. They measured experimentally thermal
380 thinning behavior. Moreover, the viscosity at a given shear rate depends transport 436 properties or thermophysical properties, namely thermal
381 on time; to clarify, the fluid is thixotropic. As a final point, all samples conductivity 437 and viscosity. The experimental results showed that viscosity
382 showed viscoelastic nature, symptomatic of the mixture of particle increases 438 with increasing the concentration of ZnO nanoparticles and
383 aggregation and shape influences is the mechanism for its high-shear decreases 439 with temperature. Also, the relative viscosity of ZnO–EG
384 rheological behavior (Fig. 5). nanofluids 440 shows the same dependent on temperature. They stated that ZnO–
EG 441 nanofluids with concentrations of less than 10.5% (by weight) prove 442
385 5.3. SiO2 nanoparticles Newtonian behavior. The nanofluids might turn to non-Newtonian 443 mode at a
peak point if the viscosity increases with volume concentra- 444 tion. As a result,
386 Jamshidi et al. [50] investigated the experimental effect of adding further studies need to focus on the rheological proper- 445 ties of higher volume
387 SiO2 nanoparticles on the viscosity of base fluid. Base fluids were ethyl- concentration nanoparticles (Fig. 7). While some 446 scholars have studied the
388 ene glycol, transformer oil, and water. It was shown that the viscosity of reduction of viscosity, other researchers con- 447 centrate on the enhancement of
389 suspension enhances with addition of nanoparticles. The presented data viscosity. Jeong et al. [54] investigated 448 experimentally the viscosity of ZnO
390 showed that when the temperature increases, the viscosity of suspen- nanofluids with nanoparticle 449 shapes of nearly rectangular and sphere, under
391 sion decreases, also the viscosity increases when the nanoparticles vol- different volume frac- 450 tions of nanoparticles from 0.05 to 5.0 vol.%. Their
392 ume factions in the base fluid increase (Fig. 6). They mentioned that the results indicated 451 that the viscosity of the nanofluids increased from 5.3 to
393 increase of viscosity does not follow the common models such as 68.6% with in- 452 creases in the volume concentration and compared to the base
394 Einstein's formula, and, on the other hand, the increasing rate of viscos- fluid 453 (water). Moreover, the enhancement of the viscosity of the nearly 454
395 ity for fluids with lower viscosity such as water is higher than fluids with rectangular shape nanoparticles was found to be more than 7.7%, rather 455 than
396 higher viscosity such as ethylene glycol. They are also stated that there the spherical nanoparticles. The shape of the particles is found to 456
397 are a little bit differences between the viscosity of nanofluid at cooling
398 and heating cycles in each specific temperature. According to the exper-
399 imental results, the correlations relate the viscosity of nanofluid to the have a significant effect on the viscosity enhancements. 457
400 particle volume fraction and temperature in order to predict the 5.6. SiC nanoparticles 458
401 viscosity. The following correlation has been proposed for use of SiO2/
402 transformer oil nanofluid and particle volume fractions of 0 to 1%.
Lee et al. [55] investigated the dispersion behavior of SiC/deionized 459 water
μnf ¼ expðaTþbÞ ð14Þ (DIW) nanofluids under various pH values and characterized 460 with the zeta
a ¼ ‐0:03959‐0:01523ϕ potential values. Because the properties of the nanofluids 461 depend on the
b ¼ 3:53267 þ 6:3848ϕ morphologies of nanoparticles, transmission electron 462 microscopy and
404 scanning electron microscope images were used to 463 characterize the shape and
In the above correlation, T (°C) is the fluid temperature. size of SiC nanoparticles. Next their viscosity 464 was studied as a function of
volume fraction to assess the potential of 465
405 5.4. TiO2 nanoparticles SiC/DIW nanofluids to use in heat transfer applications. 466
5.7. CuO/copper nanoparticles 467
406 Duangthongsuk et al. [51] reported dynamic viscosity of nanofluids
407 experimentally. They used TiO2 nanoparticles dispersed in water with
408 volume concentration of 0.2 to 2 vol.% and for temperatures ranging There are several studies that are based on the expected results of 468
409 from 15 °C to 35 °C. The Bohlin rotational rheometer (Malvern viscosity augmentation. Nikkam et al. [17] investigated rheological 469
410 Instrument) is used to measure the viscosity of nanofluids. The results characteristics and the fabrication evaluation of nanofluids consisting 470 of
411 showed that the viscosity of nanofluids augmented as the particle con- copper nanoparticles in diethylene glycol base liquid. Copper nano- 471 particles
412 centrations increased and importantly are higher than the base liquid were directly formed in diethylene glycol using microwave- 472 assisted heating,
413 values. On the other hand, the viscosity of nanofluids decreased with in- which provides uniform heating of reagents and 473 solvent, accelerating the
414 crease of nanofluids' temperature. Additionally, the measured viscosity nucleation of metal clusters, resulting in 474 mono-dispersed nanostructures.
415 of nanofluids is quite different from the predicted formulae. Copper nanoparticles displayed an av- 475 erage primary particle size of 75 ± 25
nm from SEM micrographs, yet 476 accumulated to form large spherical particles
416 5.5. ZnO nanoparticles of about 300 nm. They 477 measured the physicochemical properties such as the
viscosity of 478 nanofluids for the nanofluids with nanoparticle concentration
417 There are some investigations based on the reduction of viscosity. between 479 0.4 and 1.6 wt% in the temperature range of 20 °C to 50 °C. Proper
418 Suganthi et al. [52] studied the rheological characteristics of ZnO– 480 theoretical models were applied to compare the experimental results 481 with
419 propylene glycol nanofluid over a temperature range of 10 °C to the estimated values for the viscosity of nanofluids. For these 482 nanofluids with
420 140 °C and nanoparticle concentration range of 0 to 2 vol%. They men- 1.6 wt% nanoparticles, maximum increase in viscosity 483
421 tioned that adding ZnO nanoparticles (35–40 nm) to propylene glycol

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 6 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

484 of approximately 5.2% was observed. Yang et al. [56] performed a effects on the viscosity of R123-TiO2 nano-refrigerants for 5 °C to 20 °C 544
485 comparative study of viscosity between viscoelastic-fluid-based Cu temperature and up to 2 vol.%. They mentioned pumping power and 545 pressure
486 nanofluids and distilled water-based nanofluids experimentally. They drop depends on the viscosity. They also investigated the effect 546 of pressure
487 prepared viscoelastic-fluid-based nanofluids with the dispersion of cop- drop with the augmentation of viscosity. According to the 547 analysis, it has been
488 per nanoparticles in viscoelastic surfactant solution (aqueous solution found that the viscosity of nanorefrigerant increases 548 with the increase of
489 of cetyltrimethylammonium chloride/sodium salicylate). Various con- nanoparticle volume concentrations and decreases 549 accordingly with the
490 centrations of viscoelastic base fluid and volume fraction of Cu nanopar- augmentation of temperature. As a result, the pres- 550 sure drop increased
491 ticles were matched to test out their influences on the fluid viscosity. notably with the intensification of volume concen- 551 trations and vapor quality.
492 Their experimental results showed that the viscoelastic-fluid-based Cu Consequently, for better performance of a 552 refrigeration system, low volume
493 nanofluid exhibited a non-Newtonian behavior in its viscosity, and, in concentrations of nanorefrigerant 553 were suggested. Mahbubul et al. [4] studied
494 general, the viscosity increases with the increase of Cu nanoparticle con- the volumetric effects of 554 viscosity of Al2O3/R141b nanorefrigerant for
495 centration and with the decrease of temperature. Some studies have in- different temperature 555 ranges. According to the analysis about nanorefrigerant,
496 vestigated the sub-zero situation. Namburu et al. [57] presented an it is found 556 that, viscosity increases with the enhancement of volume fractions
497 experimental investigation of rheological properties of copper oxide 557 and decreases with the increase of temperature (Fig. 2). As heat transfer 558
498 nanoparticles, which suspended in the mass proportion of 60:40 ethyl- performance increases with the enhancement of thermal conductivity, 559
499 ene glycol and water mixture. They tested nanofluids of particle volume pressure drop and pumping power increase with the augmentation of 560 viscosity
500 fraction ranging from 0% to 6.12%, and with temperature ranging from and density. Therefore, an optimal particle volume concentra- 561 tion of
501 −35 °C to 50 °C to prove nanofluids' applicability in sub-zero regions. nanorefrigerant could improve the performance of a refrigera- 562 tion system,
502 The results showed that the viscosity of nanofluids increases when the considering thermal conductivity, viscosity, and density 563 of nanorefrigerant.
503 volume concentration of nanoparticles increases. For instance, the vis- Some of the scholars have stated the higher quantity 564 of viscosity in
504 cosity of 6.12% copper oxide volume concentration is about four times comparison to the predicted models. Dhindsa et al. [61] 565 experimentally
505 the value of the base fluid at −35 °C. It is clearly the next generation studied on viscosity of Al2O3/R11 nanorefrigerant at 566 various temperature (4
506 of heat transfer fluids in the specific cold regions (Fig. 8). They com- °C to 160 °C) and nanoparticle concentrations 567 (0.01 to 0.05 vol.%). Al2O3
507 pared their results with data from the American Society of Heating, Re- nanoparticles with size of 20 nm were 568 mixed with the refrigerant R-11 at
508 frigerating and Air-Conditioning Engineers Handbook (ASHRAE) [58], different nanoparticle concentration 569 by using ultrasonic vibration method.
509 which were well-matched together with the maximum difference of According to their experimental 570 investigation, it is observed that viscosity
510 ± 2%. In the next step, based on the particle volume concentrations enhanced considerably with 571 the augment of volume concentrations.
511 tests, nanofluids showed Newtonian behavior. According to their exper- Furthermore, they mentioned 572 that low volume concentration up to 0.03% of
512 imental data, an experimental correlation was developed that relates nanorefrigerant, in the 573 temperature range from 4 °C to 160 °C, could make
513 viscosity with particle volume fraction and the nanofluid temperature. much better the 574
514 In other words, by careful statistical analyzes an exponential model
515 was derived. Based on their researches, this equation fits the experi- performance of a refrigeration process and system. 575
2
516 mental data with a correlation coefficient R (the goodness of fitting
517 data with correlations) N 0.99. 5.9. Other materials or comparative papers between theoretical and 576
experimental studies 577
‐BT
Logðμ sÞ¼Ae
2 2
A ¼ 1:8375ðϕÞ −29:643ðϕÞ þ 165:56 with R ¼ 0:9873 ð15Þ Mahbubul et al. [2] compiled and reviewed different characteristics 578 of
‐6 2 2 nanofluids viscosity including nanofluid preparation methods, tem- 579 perature,
B ¼ 4 10 ðϕÞ −0:001ð ϕÞ þ 0:0186 with R ¼ 0:988
519 particle size and shape, and volume fraction effects. Their ex- 580 perimental
Where μs is the copper oxide nanofluid viscosity in centipoises (cP), results about the nanofluids viscosity such as other existed 581 papers showed that
520 T is the temperature in °K. viscosity increased with an augmentation of vol- 582 ume concentration and
521 Li et al. [59] have mentioned the specific situation that viscosity is al- decreased with the temperature rise. However, 583 there are some contradictory
522 most constant. They studied the viscosity of Cu-H2O nanofluid, with and results on the impacts of temperature on 584 viscosity. Additionally, it is shown
523 without dispersant preparation, using capillary viscometers for mea- that particle size has some striking 585 effects over the viscosity of nanofluids.
524 surement. The mass fractions of copper nanoparticles in the experiment Mishra et al. [31] reviewed the 586 effects of nanoparticles' shape and size,
525 varied between 0.04% and 0.16% for temperature ranging from 30 °C to temperature, volume concen- 587 tration, pH, and so on. They mentioned that the
526 70 °C. Their experimental results indicated that the temperature and literature review 588
527 SDBS concentration were the key factors, which affected the viscosity
528 of the nano-copper suspensions, while the effect of mass fraction of Cu
529 over viscosity was not as apparent as that of the temperature and
530 SDBS dispersant for the mass fraction chosen in the experiment. The vis-
531 cosity of the copper nano-suspensions reduced with the temperature
532 augmentation and enhanced with the increase of the mass fraction of
533 SDBS dispersant, and almost kept invariability with increasing the
534 mass fraction of Cu. The effect of SDBS fraction on the viscosity of
535 nano-suspension was moderately large in comparison with the nano-
536 particle concentration.

537 5.8. Nanorefrigerant

538 Nanorefrigerant is the mixture of nanoparticles with refrigerants,

539 and, according to the expectations, it has better heat transfer perfor-
540 mance than traditional refrigerants. Since 2005 Wang scientists have
541 tried to work on nanorefrigerant for better consequences of heat trans-
542 fer performance of refrigeration and air-conditioning systems [60]. Fig. 2. Viscosity as a function of temperature for different volume fraction of nanorefrigerant.
543 Mahbubul et al. [7] studied the volume concentration and temperature (Reprinted from Mahbubul et al. [4], with permission from Elsevier.)

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 7

Fig. 3. Illustration of the viscometer measurement chamber. (Reprinted from Nguyen et al. [42],
with permission from Elsevier.)

:
Fig. 5. Viscosity vs. Shear rate γ dependence of EG/Fe2O3 nanofluids. At 303.15 K and t =
589 shows that the viscosity of nanofluid depends on many parameters such
500 s for different mass concentrations: multiplication sign, EG; solid circle, 5 wt%; solid
590 as base fluids, particle volume fraction, particle size, particle shape, inverted triangle, 10 wt%; solid square, 15 wt%; solid diamond, 20 wt%; and solid triangle, 25
591 temperature, shear rate, pH value, surfactants, dispersion techniques, wt%. The inset shows viscosity vs. time dependence for 25 wt% nanofluids at controlled shear
592 particle size distribution, and particle aggregation. For instance, about stress. (Reprinted from Pastoriza-Gallego et al. [49].)
593 the effect of particle size on the viscosity, they mentioned some contro-
594 versial studied in compared to other researchers, that stated increase, experimental results is that nanofluid viscosity did not show acceptable 597
595 decrease, or even almost stable outcomes with the increase in particle agreement with theoretical models. The good evidence of this difference 598 may
596 size for different materials and conditions. The main point of the be due to the effect of Brownian motion, assumptions made while 599 deriving
the models, mathematical modeling approach, and dispersion 600 techniques.
However, the criterion for verifying their points with exper- 601 imental results
still needs more attention. Ilyas et al. [62] emphasized on 602 the comparison of
previous theoretical and experimental studies of 603 thermophysical properties of
nanofluids. The main results of them 604 were the enhancement of viscosity by
increasing the volume fraction 605 and the reversed relation of viscosity and
temperature (Table 2). Some 606 investigations have had a look of the shear
viscosity. For instance, Li 607 et al. [63] investigated the shear viscosity of the
prepared VFBN with dif- 608 ferent particle volume fractions, temperatures and
concentrations of the 609 base fluid experimentally. They prepared a viscoelastic-
fluid-based 610 nanofluid (VFBN) using viscoelastic aqueous solution of
cetyltrimethyl 611 ammonium chloride/sodium salicylate as base fluid and
multiwalled 612 carbon nanotubes (MWCNTs) as nanoparticles in order to obtain
a 613 novel thermo-fluid with both heat transfer enhancement and turbulent 614
drag reducing, compared with drag-reduced flow abilities. According to 615 their
paper the modified Li–Qu–Feng model (Y.H. Li, W. Qu, J.C. Feng, 616 Chinese
Phys. Lett. 25 (2008) 3319–3322) showed excellent agree- 617 ments with the
measured data. The VFBN with MWCNTs showed a 618

Fig. 4. A schematic view of the visometer manufactured by Cannon Instrument company. Fig. 6. The effect of increasing temperature and volume fraction on the viscosity of transformer
(Reprinted from Yiamsawas et al. [47], with permission from Elsevier.) oil nanofluid. (Reprinted from Jamshidi et al. [50].)

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 8 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

Table 2 t2:1
Experimental viscous properties of nanofluids at 20 °C [62]. t2:2

Nanofluids Volume fraction [%] Viscosity [mPa s] t2:3

CuO/EG:water 0 5.38 t2:4

CuO/EG:water 2 6.83 t2:5
CuO/EG:water 4 11.38 t2:6
CuO/EG:water 6 18.75 t2:7

significantly higher than the base fluids' value. The viscosity of 640 nanofluids
increases with the increase of nanoparticle volume fraction. 641 According to the
calibration results from the transient hot-wire method, 642 the measurement error
was estimated up to 2%. Moreover, the mea- 643 sured values of the effective
viscosity of nanofluids were found to be 644 underestimated by predicted models.
Esfe et al. [65] investigated 645 thermophysical properties such as the viscosity of
nanofluid that were 646 developed differently by mixing water as a base fluid with
magnetic 647 nanoparticles. They examined the effects of different nanoparticles'
Fig. 7. Rheological behaviors of ZnO/EG nanofluids. (Reprinted from Li et al. [53], with
648 nominal diameters including, about 37 nm, 71 nm, and 98 nm, and con- 649
permission from Elsevier.)
centrations of nanoparticles over the viscosity of nanofluids. Experi- 650 mental
results indicated that the nanofluid dynamics viscosity ratio 651 increases with
619 non-Newtonian fluid behavior in its shear viscosity, similar to its visco- increasing the particle concentration and also by increas- 652 ing nanoparticles'
620 elastic base fluid, and its shear viscosity increases with the increase of diameter. Some other papers have mentioned the re- 653 markable enhancement
621 particle volume fraction and the decrease of temperature. In turbulent of viscosity in good agreement the predicted 654 classical models. Jo et al. [66]
622 flows, it is anticipated that the prepared VFBNs might also have drag- investigated nanoparticle aggregation 655 that is related to experimental
623 reducing ability. Other scholars have studied the effect of some specific measurements of the viscosity of 656 molten-salts nanofluid containing
624 parameters over viscosity. Garg et al. [20] studied the effect of dispers- multiwalled carbon nanotubes, 657 which were performed for a wide range of the
625 ing energy (ultrasonication) on viscosity. They prepared four samples shear rate and different 658 nanotube volume concentrations. The high
626 of 1 wt% multiwalled carbon nanotube-based (MWCNT) aqueous temperature nanofluid ex- 659 hibited the non-Newtonian behavior in low shear
627 nanofluids via ultrasonication that were thermally characterized. A rate region; however, 660 it was extended to high shear rate region by increasing
628 newly developed wet-TEM technique was used to evaluate the disper- the nanoparticle 661 concentration. Their experimental results showed that the
629 sion state of carbon nanotubes in suspension by direct imaging. Results viscosity of 662 the nanofluid is noticeably enhanced up to 93% with considering
630 indicated that the suspensions showed a shear-thinning behavior, the 663 concentration of 2 wt%. Based on their paper, the results have good 664
631 which followed the Power Law viscosity model. They discussed a num- agreement with the Krieger–Dougherty model using a nanoparticle ag- 665
632 ber of mechanisms which are related to boundary layer thickness, gregation factor. One of the striking points of the experimental results is 666 the
633 micro-convective effect, particle rearrangement, and possible induced variation of them based on the different locations of the laboratories 667 round the
634 convective effects due to temperature and viscosity variations in the ra- world. This issue has been noticed by some researchers in a 668 good way.
635 dial direction and the non-Newtonian nature of the samples. Another Venerus et al. [67] reported viscosity data on a series of col- 669 loidal
636 researcher has hinted the effect of measurement error estimation be- dispersions, which have collected as part of the International 670 Nanofluid
637 cause of the measurement methods. Murshed et al. [64] conducted a Property Benchmark Exercise (INPBE). Data were reported 671
638 combined experimental and theoretical study on the viscosity of
639 nanofluids. The viscosity of nanofluids was measured and found to be

Fig. 8. Relative viscosity and temperature relationship for various concentrations of CuO. (Reprinted from Namburu et al. [57], with permission from Elsevier.)

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 9

672 for seven various nanofluids, which included dispersions of metal-oxide ceramic particles, Arch. Mater. Sci. Eng. 34 (2008) 99–104. 737
[11] V. Trisaksri, S. Wongwises, Critical review of heat transfer characteristics of 738
673 nanoparticles in water, and in synthetic oil. They examined the effects of nanofluids, Renew. Sust. Energ. Rev. 11 (2007) 512–523. 739
674 particle shape and volume concentration on the viscosity of these [12] W. Daungthongsuk, S. Wongwises, A critical review of convective heat transfer of 740
675 nanofluids and compared their data to predictions from theoretical cor- nanofluids, Renew. Sust. Energ. Rev. 11 (2007) 797–817. 741
[13] O. Mahian, A. Kianifar, S.A. Kalogirou, I. Pop, S. Wongwises, A review of the applica- 742
676 relations on suspension rheology. The data were from approximately tions of nanofluids in solar energy, Int. J. Heat Mass Transf. 57 (2013) 582–594. 743
677 ten various laboratories around the world on a series of ten various [14] G. Saha, M.C. Paul, Numerical analysis of the heat transfer behaviour of water based 744
678 nanofluids. Generally, the agreement between various laboratories Al2O3 and TiO2 nanofluids in a circular pipe under the turbulent flow condition, Int. 745
679 was well-matched with variations of approximately ± 20%. Based on Commun. Heat Mass Transfer 56 (2014) 96–108 (ISSN 0735-1933). 746
[15] M. H. Esfe, S. Saedodin, O. Mahian, S. Wongwises, Thermal conductivity of Al2O3/ 747 water
680 the experimental results, two of seven nanofluids exhibited shear- nanofluids: measurement, correlation, sensitivity analysis, and comparisons 748
681 thinning behavior; on the other hand, the rest five showed Newtonian with literature reports, J. Therm. Anal. Calorim. 749
682 behavior. Gaganpreet et al. [68] studied the abnormal augmentation in [16] R. Saidur, K.Y. Leong, H.A. Mohammad, A review on applications and challenges of 750
nanofluids, Renew. Sust. Energ. Rev. 15 (2011) 1646–1668. 751
683 the viscosity of nanofluids, taking clustering as one of the causes. Just [17] N. Nikkam, M. Ghanbarpour, M. Saleemi, E. Bitaraf Haghighi, R. Khodabandeh, M. 752
684 few of these nanoparticles might collide with each other while other Muhammed, B. Palm, M.S. Toprak, Experimental investigation on thermo-physical 753
685 particles might do so along with interfacial layer developed around properties of copper/diethylene glycol nanofluids fabricated via microwave- 754
assisted route, Appl. Therm. Eng. 65 (2014) 158–165. 755
686 them. The conception of aggregation and equivalent volume fraction [18] O. Mahian, A. Kianifar, S. Wongwises, Dispersion of ZnO nanoparticles in a mixture
687 has been used in Kreiger and Dougherty correlation to investigate the 756 of ethylene glycol–water, exploration of temperature-dependent density, and
688 viscosity of nanofluids. Viscosity increases with the increase in particle sensi- 757
689 aggregate (ra) and is found well-matched for ra = 3r at low volume tivity analysis, J. Clust. Sci. 24 (2013) 1103–1114. 758
690 concentrations. [19] K. Bashirnezhad, M.M. Rashidi, Zh. Yang, Sh. Bazri, Wei-Mon Yan, A comprehensive 759
review of last experimental studies on thermal conductivity of nanofluids, J. Therm. 760
Anal. Calorim. (2015). 761
691 6. Conclusion and future works [20] P. Garg, J.L. Alvarado, Ch. Marsh, Th.A. Carlson, D.A. Kessler, K. Annamalai, An exper- 762
imental study on the effect of ultrasonication on viscosity and heat transfer perfor- 763
mance of multi-wall carbon nanotube-based aqueous nanofluids, Int. J. Heat Mass 764
692 A review was done of the latest experimental studies on the viscosity Transf. 52 (2009) 5090–5101. 765
693 of nanofluids containing various nanoparticles. In general, it can be con- [21] A. Einstein, A new determination of molecular dimensions, Ann. Phys. 19 (1906) 766
694 cluded that most of studies justify the increase of viscosity with an in- 289–306. 767
695 crease in nanoparticle volume fraction and a decrease of temperature. [22] A. Einstein, Investigations on the Theory of the Brownian Movement, Dover 768
Publications, New York, 1956. 769
696 In most of experimental studies reported in the literature, just the ef- [23] O. Mahian, A. Kianifar, C. Kleinstreuer, M.A. Al-Nimr, I. Pop, A.Z. Sahin, S. Wongwises, 770
697 fects of two or three important parameters (e.g. temperature and con- A review of entropy generation in nanofluid flow, Int. J. Heat Mass Transf. 65 (2013) 771
698 centration) have been investigated at the same time. Therefore, for 514–532. 772
[24] A. Einstein, A: Berichtigung zu meiner Arbeit: “Eine neue Bestimmung der 773
699 future works, the researchers could perform experimental studies by Moleküldimensionen, Ann. Phys. 34 (1911) 951. 774
700 considering all of important factors including temperature, nanoparticle [25] H.C. Brinkman, The viscosity of concentrated suspensions and solution, J. Chem. 775
701 size, pH, sonication time, aggregation, type of base liquid, and so on at Phys. 20 (1952) 571–581. 776
[26] I.M. Krieger, A mechanism for non Newtonian flow in suspensions of rigid spheres, 777
702 the same time, which leads to generate more accurate correlations to Trans. Soc. Rheol. 3 (1959) 137. 778
703 predict the viscosity. This will help the engineers to design the thermal [27] I.M. Mahbubul, R. Saidur, M.A. Amalina, Latest developments on the viscosity of 779
nanofluids, Int. J. Heat Mass Transf. 55 (2012) 874–885. 780Q2
704 systems with a higher accuracy.
[28] N.A. Frankel, A. Acrivos, On the viscosity of a concentrated suspension of solid 781
spheres, Chem. Eng. Sci. 22 (6) (1967) 847–853. 782
705 Acknowledgment [29] K. Khanafer, K. Vafai, A critical synthesis of thermophysical characteristics of 783
nanofluids, Int. J. Heat Mass Transf. 54 (2011) 4410–4428. 784
706 The authors gratefully acknowledge High Impact Research Grant [30] L.E. Nielsen, Generalized equation for the elastic moduli of composite materials, 785
J. Appl. Phys. 41 (11) (1970) 4626–4627. 786
707 UM.C/HIR/MOHE/ENG/23 and Faculty of Engineering, University of [31] P. Chandra Mishra, S. Mukherjee, S. Kumar Nayak, Arabind Panda, A brief review on 787
708 Malaya, Malaysia, for support in conducting this research work. The viscosity of nanofluids, Int. Nano Lett. 4 (2014) 109–120. 788
709 eighth author would like to thank the “Research Chair Grant” National [32] T.S. Lundgren, Slow flow through stationary random beds and suspensions of 789
spheres, J. Fluid Mech. 51 (02) (1972) 273–299. 790
710 Science and Technology Development Agency, the Thailand Research [33] G.K. Batchelor, The effect of Brownian motion on the bulk stress in a suspension of 791
711 Fund and the National Research University Project for the support. spherical particles, J. Fluid Mech. 83 (1977) 97–117. 792
[34] A.L. Graham, On the viscosity of suspensions of solid spheres, Appl. Sci. Res. 37 (3–4) 793
(1981) 275–286. 794
712 References [35] T. Kitano, T. Kataoka, T. Shirota, An empirical equation of the relative viscosity of 795
polymer melts filled with various inorganic fillers, Rheol. Acta 20 (2) (1981) 796
713 [1] J.C. Maxwell, Electricity and Magnetism, Clarendon, Oxford, UK, 1873. 207–209. 797
[36] F.M. White, Viscous Fluid Flow, McGraw Hill, New York, 1991. 798
714 [2] I.M. Mahbubul, R. Saidur, M.A. Amalina, Latest developments on the viscosity of
[37] B. Aladag, S. Halelfad, N. Doner, Th. Maré, S. Duret, P. Estellé, Experimental investiga- 799
715 nanofluids, Int. J. Heat Mass Transf. 55 (2012) 874–885. tions of the viscosity of nanofluids at low temperatures, Appl. Energy 97 (2012) 800
716 [3] E. Abu-Nada, A.J. Chamkha, Effect of nanofluid variable properties on natural con- 876–880. 801
717 vection in enclosures filled with a CuO–EG–water nanofluid, Int. J. Therm. Sci. 49 [38] L. Syam Sundar, E. Venkata Ramana, Manoj K. Singh, Antonio C.M. Sousa, Thermal 802
718 (2010) 2339–2352. conductivity and viscosity of stabilized ethylene glycol and water mixture Al 2O3 803
719 [4] I.M. Mahbubul, R. Saidura, M.A. Amalinaa, Thermal conductivity, viscosity and den- nanofluids for heat transfer applications: an experimental study, Int. Commun. 804
720 sity of R141b refrigerant based nanofluid, Procedia Eng. 56 (2013) 310–315. Heat Mass Transfer 56 (2014) 86–95. 805
[39] G. Zyła, J. Grzywa, A. Witek, M. Cholewa, Influence of anisotropic pressure on viscos- 806 ity
721 [5] S. Choi, Enhancing Thermal Conductivity of Fluids with Nanoparticles, in: D.A.
and electrorheology of diethylene glycol-based MgAl2O4 nanofluids, Nanoscale 807
722 Siginer, H.P. Wang (Eds.), Developments Applications of Non-Newtonian Flows, Res. Lett. 9 (2014) 170. 808
723 FED—Vol. 231/MD—Vol. 66, ASME, New York 1995, pp. 99–105. [40] T. John, T.S. Krishnakumar, Experimental studies of thermal conductivity, viscosity 809 and
724 [6] R. Saleh, N. Putra, S. Purbo Prakoso, W. Nata Septiadi, Experimental investigation of stability of ethylene glycol nanofluids, Int. J. Innov. Res. Sci. Eng. Technol. 2 810
725 thermal conductivity and heat pipe thermal performance of ZnO nanofluids, Int. J. (2013) 2347–6710. 811

726 Therm. Sci. 63 (2013) 125–132. [41] A.T. Utomo, H. Poth, Ph.T. Robbins, A.W. Pacek, Experimental and theoretical studies 812 of
thermal conductivity, viscosity and heat transfer coefficient of titania and alumina 813
727 [7] I.M. Mahbubul, R. Saidur, M.A. Amalina, Investigation of viscosity of R123-TiO 2
nanofluids, Int. J. Heat Mass Transf. 55 (2012) 7772–7781. 814
728 nanorefrigerant, Int. J. Mech. Mater. Eng. 7 (2012) 146–151.
[42] C.T. Nguyen, F. Desgranges, G. Roy, N. Galanis, T. Maré, S. Boucher, H. Angue Mintsa, 815
729 [8] O. Mahian, A. Kianifar, A.Z. Sahin, S. Wongwises, Performance analysis of a Temperature and particle-size dependent viscosity data for water-based 816 nanofluids—
730 minichannel-based solar collector using different nanofluids, Energy Convers. hysteresis phenomenon, Int. J. Heat Fluid Flow 28 (2007) 1492–1506. 817
731 Manag. 88 (2014) 129–138. [43] M. Chandrasekar, S. Suresh, A. Chandra Bose, Experimental investigations and 818 theoretical
732 [9] M.H. Esfe, S. Saedodin, O. Mahian, S. Wongwises, Thermophysical properties, heat determination of thermal conductivity and viscosity of Al2O3/water 819
733 transfer and pressure drop of COOH-functionalized multi walled carbon nano-
nanofluid, Exp. Thermal Fluid Sci. 34 (2010) 210–216. 820
[44] Th. Yiamsawas, A.S. Dalkilic, O. Mahian, S. Wongwises, Measurement and correla- 821 tion
734 tubes/water nanofluids, Int. Commun. Heat Mass Transfer 58 (2014) 176–183.
of the viscosity of water-based Al2O3 and TiO2 nanofluids in high temperatures 822
735 [10] I. Tavman, A. Turgut, M. Chirtoc, H.P. Schuchmann, S. Tavman, Experimental inves-
736 tigation of viscosity and thermal conductivity of suspensions containing nanosized
Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005
 10 K. Bashirnezhad et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

823 and comparisons with literature reports, J. Dispers. Sci. Technol. 34 (2013)[57] P.K. Namburu, D.P. Kulkarni, D. Misra, D.K. Das, Viscosity of copper oxide nanoparti- 859
824 1697–1703. cles dispersed in ethylene glycol and water mixture, Exp. Thermal Fluid Sci. 32 860
825 [45] M. Jarahnejad, E.B. Haghighi, M. Saleemi, N. Nikkam, R. Khodabandeh, B. Palm, M.S. (2007) 397–402. 861
826 Toprak, M. Muhammed, Experimental investigation on viscosity of water-based [58] ASHRAE Handbook, Fundamentals, American Society of Heating, Refrigerating and 862
827 Al2O3 and TiO2 nanofluids, J. Therm. Anal. Calorim. 54 (2015) 411–422. Air-Conditioning Engineers Inc., Atlanta, 1985. 863
828 [46] N. Masoumi, N. Sohrabi, A. Behzadmehr, A new model for calculating the effective [59] X. Li 李新芳, D. Zhu, X. Wang, Experimental investigation on viscosity of Cu-H2O 864
829 viscosity of nanofluids, J. Phys. D. Appl. Phys. 42 (2009) 055501 (6pp). nanofluids, J. Wuhan Univ. Technol. Mater. Sci. Educ. 24 (2009) 48–52. 865
830 [47] Th. Yiamsawas, O. Mahian, A.S. Dalkilic, S. Kaewnai, S. Wongwises, Experimental [60] K. Wang, G. Ding, W. Jiang, Development of Nanorefrigerant and its Rudiment Prop- 866
831 studies on the viscosity of TiO2 and Al2O3 nanoparticles suspended in a mixture of erty, 8th International Symposium on Fluid Control, Measurement and Visualiza- 867
832 ethylene glycol and water for high temperature applications, Appl. Energy 111 tion, China Aerodynamics Research Society, Chengdu, China, 2005. 868
833 (2013) 40–45. [61] G.S. Dhindsa, Lalkundan, Experimental investigation of the viscous behavior of Al 2O3 869
834 [48] M. Abareshi, S. Hashem Sajjadi, S. Mojtaba Zebarjad, E.K. Goharshadi, Fabrication, based nanorefrigerant, Int. J. Theor. Appl. Res. Mech. Eng. 2 (2013) (ISSN: 2319 – 870
835 characterization, and measurement of viscosity of α-Fe2O3-glycerol nanofluids, 3182). 871
836 J. Mol. Liq. 163 (2011) 27–32. [62] S. Umer Ilyas, R. Pendyala, A. Shuib, N. Marneni, A review on the viscous and thermal 872
837 [49] M.J. Pastoriza-Gallego, L. Lugo, J. Luis Legido, M.M. Piñeiro, Rheological non- transport properties of nanofluids, Adv. Mater. Res. 917 (2014) 18–27. 873
838 Newtonian behaviour of ethylene glycol-based Fe2O3 nanofluids, Nanoscale Res. [63] F.Ch. Li, J.Ch. Yang, W.W. Zhou, Y.R. He, Y.M. Huang, B.Ch. Jiang, Experimental study 874
839 Lett. 6 (2011) 560. on the characteristics of thermal conductivity and shear viscosity of viscoelastic- 875
840 [50] N. Jamshidi, M. Farhadi, D.D. Ganji, K. Sedighi, Experimental investigation on the fluid-based nanofluids containing multiwalled carbon nanotubes, Thermochim. 876
841 viscosity of nanofluids, Int. J. Eng. Trans. B 25 (2012) 201–209. Acta 556 (2013) 47–53. 877
842 [51] W. Duangthongsuk, S. Wongwises, Measurement of temperature-dependent [64] S.M.S. Murshed, K.C. Leong, C. Yang, Investigations of thermal conductivity and vis- 878
843 thermal conductivity and viscosity of TiO2-water nanofluids, Exp. Thermal Fluid cosity of nanofluids, Int. J. Therm. Sci. 47 (2008) 560–568. 879
844 Sci. 33 (2009) 706–714. [65] M. Hemmat Esfe, S. Saedodin, Somchai Wongwises, D. Toghraie, An experimental 880
845 [52] K.S. Suganthi, N. Anusha, K. Sekar Rajan, Low viscous ZnO–propylene glycol study on the effect of diameter on thermal conductivity and dynamic viscosity of 881
846 nanofluid: a potential coolant candidate, J. Nanoparticle Res. 15 (2013) 1986. Fe/water nanofluids, J. Therm. Anal. Calorim. 119 (2014) (1817-1824). 882
847 [53] H. Li, L. Wang, Y. He, Y. Hu, J. Zhu, B. Jiang, Experimental investigation of thermal [66] B. Jo, D. Banerjee, Viscosity measurements of multi-walled carbon nanotubes-based 883
848 conductivity and viscosity of ethylene glycol based ZnO nanofluids, Appl. Therm. high temperature nanofluids, Mater. Lett. 122 (2014) 212–215. 884
849 Eng. 10 (071) (2014). [67] D.C. Venerus, J. Buongiorno, R. Christianson, J. Townsend, I.Ch. Bang, G. Chen, S.J. 885
850 [54] J. Jeong, Li Chengguo, Y. Kwon, J. Lee, S. Hyung Kim, R. Yun, Particle shape effect on Chung, M. Chyu, H. Chen, Y. Ding, F. Dubois, G. Dzido, D. Funfschilling, Q. Galand, J. 886
851 the viscosity and thermal conductivity of ZnO nanofluids, Int. J. Refrig. 36 (2013) Gao, H. Hong, M. Horton, L. Hu, C.S. Iorio, A.B. Jarzebski, Y. Jiang, S. Kabelac, M.A. 887
852 (2233-224). Kedzierski, Ch. Kim, Ji-Hyun Kim, S. Kim, Th. McKrell, R. Ni, J. Philip, N. Prabhat, P. 888
853 [55] S. Won Lee, S. Dae Park, S. Kang, I. Cheol Bang, J. Hyun Kim, Investigation of viscosity Song, S. Van Vaerenbergh, D. Wen, S. Witharana, Xiao-Zheng Zhao, Sheng-Qi 889
854 and thermal conductivity of SiC nanofluids for heat transfer applications, Int. J. Heat Zhou, Viscosity measurements on colloidal dispersions (nanofluids) for heat trans- 890
855 Mass Transf. 54 (2011) 433–438. fer applications, Appl. Rheol. 20 (2010) 44582. 891
856 [56] J.Ch. Yang, F.Ch. Li, W.W. Zhou, Y.R. He, B.Ch. Jiang, Experimental investigation on [68] Gaganpreet, S. Srivastava, Effect of aggregation on thermal conductivity and viscos- 892
857 the thermal conductivity and shear viscosity of viscoelastic-fluid-based nanofluids, ity of nanofluids, Appl. Nanosci. 2 (2012) 325–331. 893
858 Int. J. Heat Mass Transf. 55 (2012) 3160–3166.
894

Please cite this article as: K. Bashirnezhad, et al., Viscosity of nanofluids: A review of recent experimental studies, Int. Commun. Heat Mass Transf. (2016),
http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.02.005