ICHMT-03452; No of Pages 8

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

Contents lists available at ScienceDirect

International Communications in Heat and Mass Transfer

journal homepage: www.elsevier.com/locate/ichmt

1 Effect of particle size and viscosity on thermal conductivity enhancement of graphene

2 oxide nanofluid☆

F
3Q1 Milad Rabbani Esfahani a, Ehsan Mohseni Languri b,⁎, Maheshwar Rao Nunna b

O
4 a
Center for the Management, Utilization and Protection of Water Resources, Tennessee Technological University, Cookeville, TN 38505, USA
5 b
Department of Mechanical Engineering, Tennessee Technological University, Cookeville, TN 38505, USA
6

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

R
8
9 Available online xxxx In this study, first, graphene oxide nanosheets were synthesized based on the modified Hummers methods. The 15
14
13
12
11
10 physicochemical properties of fabricated graphene oxide were characterized using X-ray diffraction analysis 16

P
34 Keywords: (XRD), a scanning electron microscope (SEM), and UV–Vis spectrophotometry. Second, graphene oxide 17
35 Nanofluids nanofluids were prepared at different concentrations (0.01, 0.05, 0.1 and 0.5 wt.%) in water (base fluid). 18
36 Graphene oxide Particle-size distribution and stability of the colloidal solution of the graphene oxide nanofluids were investigated 19
37 Viscosity
using dynamic light scattering and zeta potential techniques. Also, rheological behavior of graphene oxide
D 20
38 Thermal properties
39 Aggregation
nanofluids was studied at different temperatures (25 °C, 40 °C, and 60 °C) and different shear rates (10–100 1/s). 21
40 Dynamic light scattering Results show that both particle size and viscosity of graphene oxide nanofluids increased linearly by increasing 22
the graphene oxide concentration from 0.01 to 0.1 wt.%, but there was a very sharp increment on average particle 23
E
size and viscosity by increasing the concentration to 0.5 wt.%. The thermal conductivity of graphene oxide 24
nanofluids was measured at different temperatures. Thermal conductivity of graphene oxide nanofluids depends 25
T

on both particle-size distribution of graphene oxide and viscosity of graphene oxide nanofluids. All graphene 26
oxide nanofluids showed enhanced thermal conductivity compared to the base fluid, water. Increasing the 27
28
C

graphene oxide concentration from 0.01 wt.% to 0.1 wt.% resulted in 8.7% and 18.9% thermal conductivity enhance-
ment at 25 °C, respectively. However, further concentration increment to 0.5 wt.% increased thermal conductivity to 29
19.9%. This behavior shows that there is an optimal concentration of graphene oxide at which particle size and vis- 30
E

cosity of graphene oxide nanofluids exhibit significant thermal conductivity enhancement, and further concentra- 31
tion increments have no significant effect on thermal conductivity enhancement. 32
© 2016 Published by Elsevier Ltd. 33
R

44
42
41
R

43
45 1. Introduction discovering the optimal conditions of fabrication, characterization and 60
implications of NF to achieve their maximum advantages. 61
O

46 Nanotechnologies have been proven to enhance efficiency in re- Nanoparticle physicochemical properties such as size, shape and 62
47 search areas including pharmaceuticals [1], environmental processes surface charge and also interaction between nanoparticles and base 63
C

48 [2,3], petroleum [4], automobiles [5] and power generation [6]. The crit- fluid are controlling factors for NF thermal conductivity. Hong and 64
49 ical demand for highly efficient thermal transport solution has become a Yang [10] compared the effect of Fe and Cu nanoparticles less than 65
50 major challenge for industries, especially in terms of energy and power 10 nm in particle size on thermal conductivity and reported that sus- 66
N

51 supply [7]. Heat transfer plays a major role in many types of industries, pension of highly thermal conductive materials is not always effective 67
52 such as transportation, air conditioning, power generation, process en- to improve thermal transport property of NF. Li et al. [11] investigated 68
U

53 gineering and electronics [8]. Therefore, there is a need to improve the effect of particle size on thermal conductivity of NF. They compared 69
54 heat transfer capabilities. The advantages of nanoparticle properties 36 and 47 nm particle size Al2O3 nanoparticles in water base nanofluids 70
55 have resulted in the introduction of a nanofluid (NF). In general, an NF and reported that each of the nanoparticles showed different thermal 71
56 is a suspension of ultra-fine particles, rods or tubes with an extremely conductivity increment trend due to their particle size. They concluded 72
57 high thermal conductivity compared to the conventional base fluid that thermal conductivity of Al2O3/water has a nonlinear relationship 73
58 [9]. NFs could enhance thermal conductivity and heat transfer efficiency with temperature, volume fraction and nanoparticle size, and there is 74
59 and also diminish process costs. There is a universal motivation for an optimal combination condition for each of these factors at a certain 75
bulk temperature. However, there is an inconsistency in reported data 76
☆ Communicated by Dr. W.J. Minkowycz
regarding the effect of particle size on thermal conductivity. For exam- 77
⁎ Corresponding author. ple, Chopkar et al. [12] showed that Al2Cu nanoparticles with smaller 78
E-mail address: Elanguri@tntech.edu (E.M. Languri). particle size (~30 nm) significantly increased NF thermal conductivity 79

http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
0735-1933/© 2016 Published by Elsevier Ltd.

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
 2 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

80 compared to Al2Cu nanoparticles with larger particle size (~ 150 nm) in prevent the overheating. Sodium nitrate (1.75 g) was added slowly to 141
81 water and ethylene glycol based NFs. In a similar study, Moghadasi et al. the mentioned solution and mixed for 1 h. Then, 7.5 g of potassium per- 142
82 [13] reported that as the particle size of CuO decreased, the NF thermal manganate was added to the solution and stirred for 12 h. The resulting 143
83 conductivity improved. The findings of Beck et al. [14] showed that ther- solution was diluted by adding 1250 ml of DI water under vigorous stir- 144
84 mal conductivity of Al2O3-water NFs increased by increasing the particle ring. Hydrogen peroxide (12.5 ml, 30%) was added to the solution and 145
85 size from 8 nm to 282 nm. Also, Every et al. [15] observed that the ther- stirred for 2 h to make sure the potassium permanganate reaction was 146
86 mal conductivity of zinc-sulfide increased by adding large particles of complete. The final mixture was washed with 57 ml of concentrated hy- 147
87 highly conducting diamond, but lowered by addition of sub-micron drochloric acid and 5 l of DI water, respectively. Finally, the mixture was 148
88 size particles of diamond. filtered and the graphene oxide cake was collected and dried for 12 h at 149
89 There is a relationship between rheological, aggregate size and NF 40 °C. In general, the mixture of sulfuric acid, hydrochloric acid, and po- 150
90 thermal conductivity. Available literatures show that NFs can exhibit tassium permanganate oxidize the natural graphite powders in the 151
91 both Newtonian and non-Newtonian rheological behaviors, depending water-based solution. During this process, bonding of hydroxyl, carbon- 152
92 on particle size, shape and concentration [16]. Kwak et al. [17] studied yl and epoxy groups to the graphite structure and insertion of H2O and 153
93 the relationship between viscosity and thermal conductivity of copper ions into the graphene layer result in an interlayer spacing increment 154

F
94 oxide NFs dispersed in ethylene glycol and reported that based on the [24–26]. 155
95 non-Newtonian rheological property of NFs, substantial enhancement Graphene oxide NFs were prepared by dispersing different masses of 156

O
96 in thermal conductivity was attainable only when particle concen- graphene oxide in DI water. Graphene oxide in amounts of 0.01, 0.05, 157
97 tration is below the 0.002 volume fraction. Namburu et al. [18] tested 0.1 and 0.5 g was dispersed in an appropriate mass of DI water, 158
98 the rheological properties of copper oxide nanoparticles in a mixture Table 1, and stirred for 5 h. Then, graphene oxide NF colloidal solutions 159

O
99 of ethylene glycol and water and showed that for the particle volume were sonicated at 130 W and 42 kHz using an ultrasonication instru- 160
100 concentration of 0% to 6%, NFs exhibited the Newtonian behavior in ment (Branson Ultrasonics, Danbury, CT, USA) for 1 h. Table 1 shows 161
101 temperature ranging from − 35 to 50 °C. graphene oxide NF sample specifications. 162

R
102 Recently, graphene-based particles received notable attention due to
103 their large intrinsic thermal conductivity and low density in comparison 2.3. Dynamic light scattering and zeta potential 163

P
104 to metals or metal oxides nanoparticles [19]. For example, Ghozatlo
105 et al. [20] reported that adding 0.075 wt.% of graphene to the base The dynamic light scattering method was used for investigating 164
106 fluid resulted in an improvement of thermal conductivity up to particle size and particle size distribution of graphene oxide particles
D 165
107 31.83%. Yu et al. [21] developed the ethylene-glycol based nanofluids in DI water (base fluid). The DLS measurements were performed 166
108 containing graphene nanosheets that showed up to an 86% thermal con- using the Malvern Zetasizer ZS instrument (Malvern Instruments 167
109 ductivity increment. In contrast to the theoretical study of the thermal Ltd., Malvern, United Kingdom) equipped with an He–Ne laser oper- 168
E
110 conductivity of graphene, the experimental work about the heat trans- ated at λ = 633 nm [27]. A 173o angle was set for the detector posi- 169
111 fer property of graphene-based materials is rare due to their unique tion to take advantage of the back scattering mode. Automatic 170
T

112 physicochemical properties and their different compatibility with position of the attenuator was selected to achieve the optimum 171
113 other compounds [21]. Also, very few research studies have been laser intensity based on the sample concentration. The intensity- 172
C

114 performed on the effects of viscosity and particle size on thermal con- based particle size distribution was obtained from fitting the record- 173
115 ductivity of graphene oxide nanofluids [22]. It can be seen from the ed data to a single or multi-exponential correlation function, mea- 174
116 literature that there is a need for a more in-depth understanding of suring the mean size (z-average diameter), and estimating the 175
E

117 the relationship among particle size, rheological behavior and thermal width of the distribution (polydispersity index) according to the In- 176
118 conductivity improvement using graphene oxide. ternational Standard ISO13321, Methods for Determination of Parti- 177
R

119 In this study, graphene oxide was fabricated and then characterized cle Size Distribution. 178
120 using scanning electron microscope (SEM), X-ray diffraction, and Ultra- The same Malvern Zetasizer ZS instrument was used for zeta po- 179
121 violet–Visible (UV–Vis) spectroscopic methods. The graphene oxide NFs 180
R

tential measurement of graphene oxide NFs. The scattered light was

122 were prepared at 0.01, 0.05, 0.1 and 0.5 wt.% concentrations, and ther- detected at an angle of 13° by the detector. The zeta potential (ζ) was 181
123 mal conductivity of each sample was measured at three temperatures calculated from electrophoretic mobility (EM) of particles using the 182
O

124 (25, 40 and 60 °C). The rheological behavior and aggregation behavior Henry equation by considering the Smoluchowski approximation 183
125 of each sample were measured, and their effects on thermal conductiv- [27,28]. Three replicates of each sample were tested and an average 184
126 ity were studied. value reported. 185
C

127 2. Materials and methods 2.4. Characterization techniques 186

N

128 2.1. Materials The crystal structure of graphene oxide was investigated by an X-ray 187
diffractometer (XRD), Rigaku Ultima IV X-Ray Diffractometer, Rikago, 188
U

129 Graphite powder (99.99%, 45 m) was purchased from Bay Carbon, Texas USA. Data was collected from 5° to 80° at a scan rate of 3 de- 189
130 Inc. (Bay City, Michigan. USA). Sulfuric acid (H2SO4), hydrochloric acid gree/min. Morphological characterization of fabricated graphene oxide 190
131 (HCl), sodium nitrate (NaNO3), potassium permanganate (KMnO4) was examined by an environmental scanning electron microscope 191
132 and hydrogen peroxide 30% (H2O2) were purchased from Fisher Scien- (ESEM), FEI Quanta 200, 149 FEI Company, Hillsboro, OR, USA. 192
133 tific (Pittsburgh, PA, USA). All the chemicals were of analytical reagent
134 grades and used as received, without further purifications. Deionized
135 (DI) water was produced by the Millipore DI system Direct-Q 3 UV
Table 1 t1:1
136 (18.2 MΩ cm) (Massachusetts, USA).
Specification of graphene oxide nanofluid samples. t1:2

137 2.2. Fabrication of graphene oxide and preparation of nanofluids Sample name Graphene oxide (wt.%) Deionized water (wt.%) t1:3

GO-NF-0.01 0.01 99.99 t1:4

138 Graphene oxide was prepared from graphite powder based on the GO-NF-0.05 0.05 99.95 t1:5
139 modified Hummers methods [23]. Briefly, 2.5 g of graphite powder GO-NF-0.1 0.1 99.90 t1:6
GO-NF-0.5 0.5 99.50 t1:7
140 was treated with 57 ml of concentrated sulfuric acid in an ice bath to

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 3

193 UV–Vis spectra of the graphene oxide NFs were collected by a UV–
194 visible spectroscopy (Varian 190 Cary 3E, Agilent, Santa Clara, CA,
195 USA). Graphene oxide nanofluid samples (2 ml of each) were filled in
196 a quartz cuvette, and air bubbles were removed before the measure-
197 ments. Rheological behavior of graphene oxide NFs was measured
198 using an AR 500 rheometer (TA instruments, New Castle, DE, USA)
199 at different shear rates and temperatures. Thermal conductivity of
200 graphene oxide NF samples was measured based on the hot-wire tech-
201 nique using a KD2 Prothermal properties analyzer (Decagon Devices,
202 Inc., Pullman, WA, USA). Thermal conductivity of all samples was
203 measured at three temperatures of 25 ± 1, 40 ± 1, and 60 ± 1 °C. A
204 temperature-controlled bath was used to maintain the temperature at
205 the desired set point. All samples were kept in the bath for enough
206 time to reach the equilibrium temperature. Numbers of thermal con-

F
207 ductivity measurements were conducted for each sample, and an aver-
208 age with less than 5% standard deviation was reported here. Fig. 2. SEM micrograph of graphene oxide.

O
209 3. Results and discussion an absorbance of light and the properties of a material through which 236
light is passing. The law can be expressed in the following way: 237

O
210 3.1. Graphene oxide structure
A ¼ α:l:c ð1Þ

R
211 Fig. 1 shows the XRD pattern of a fabricated graphene oxide. The 239
212 sharp diffraction peak observed at 2θ = 10.5°, corresponding to the A, α and l are the absorbance, absorption coefficient, and the distance
213 d-spacing of 0.8 mm [7]. The same crystallinity structure of graphene that light travels through material, and c is the concentration of absorb- 240

P
214 oxide fabricated with the modified Hummer's method was reported ing species in the material, respectively. The linear relationship between 241
215 by Shahriary et al. [29]. Fig. 2 shows the SEM image of amorphous and light absorption and concentration of graphene oxide NFs in the range 242
216 disordered structure of the fabricated graphene oxide. It can be ob- of 0–0.1 wt.% indicates that graphene oxide was dispersed well in the
D 243
217 served from the SEM image that graphene oxide has a layer structure, base fluid. However, for a high concentration, 0.5 wt.%, there is a devia- 244
218 which affords ultrathin and homogeneous graphene films [29]. The tion from Beer–Lambert's that might be justified due to agglomeration 245
219 thick structure of graphene oxide is due to the presence of covalently of nanoparticles in the base fluid [33]. The aggregation behavior of NFs 246
E
220 bound oxygen and the displacement of the sp3 hybridized carbon was discussed in the following section. 247
221 atoms slightly above and below the original graphene plane [30,31].
T

222 UV–Vis spectrophotometer analysis was performed to ensure the 3.2. Particle size distribution and zeta potential measurements 248
223 purity of graphene NFs. Fig. 3 shows the adsorption spectrum of each
C

224 NF sample in the range of 200–800 nm wavelength. It should be noted Particle size distributions of all graphene oxide NFs at room temper- 249
225 that due to high concentration of some samples, very high absorbance ature were presented in Fig. 4(a–d). Graphene oxide potentially can 250
226 values were observed due to saturation of the detector. Therefore, all form well-dispersed aqueous colloids due to its polar and hydrophilic 251
E

227 samples were diluted 10 times for the sake of better comparison. The character, arising from a number of reactive oxygen functional groups 252
228 maximum absorption peak of the graphene oxide 0.01 wt.% was at including epoxide, hydroxyl, and carboxylic acid [34]. However, 253
R

229 ~ 227 nm, which could be ascribed to π–π* transition of the atomic graphene oxide NFs particle size distribution at all concentrations can 254
230 C\\C bonds. Shoulder peak at ~300 nm could be ascribed to n–π* tran- be categorized into three particle size groups: particle sizes less than 255
231 sitions of aromatic C\\C bonds [29,32]. The value reasonably agrees 200 nm, particle sizes greater than 200 nm but less than 1000 nm, and 256
R

232 with the absorption spectrum data of graphene oxide NFs reported by finally particle sizes greater than 1 μm. It should be noted that these par- 257
233 Zubair et al. [7]. The relationship between nanofluids concentration ticle size distribution measurements were conducted after 5 h mixing 258
O

234 and light absorption based on the Beer–Lambert's law was shown in and 2 h sonication. This trend agrees with the data reported by Gupta 259
235 Fig. 3(b). The Beer–Lambert law shows a linear relationship between et al. [35]. Other similar studies did not report the particle size distribu- 260
tion of fabricated graphene oxide in base fluid but just reported the av- 261
C

erage particle size or particle size of solid nanoparticles (not in aqueous 262
phase), which does not reveal the specific information about the parti- 263
N

cle size distribution of graphene oxide in aqueous phase [30]. 264

Fig. 4(e) shows the average particle size of all graphene oxide NFs. It 265
can be seen that average particle size of graphene oxide NFs increased 266
U

by increasing the graphene oxide concentration. Increasing the concen- 267

tration from 0.01 wt.% to 0.1 wt.% resulted in average aggregate size in- 268
crement from 600 nm to 1200 nm. Further concentration increment to 269
0.5 wt.% resulted in a large aggregate size in the range of 2800 nm. Com- 270
paring the aggregation trend at different concentrations reveals that the 271
sever aggregation phenomenon occurred by increasing the concentra- 272
tion from 0.1 to 0.5 wt.%. Graphene oxide aggregation in water involved 273
five forces including: van der Waals attraction, electrostatic interaction, 274
hydrogen-bond interaction, π–π stacking and the collision of water mol- 275
ecules [36]. The pH of solution, electrical surface surge and functional 276
groups of graphene oxide impact its aggregation behavior. Duch et al. 277
[37] reported the graphene oxide fabricated by the modified Hummers 278
Fig. 1. XRD pattern of fabricated graphene oxide at following XRD conditions: X-Ray: methods contained hydroxyl and carbonyl functional groups. The in- 279
40 kV, 44 mA, Scan speed/Duration time: 5.0000 deg./min. creased particle size resulted from a reduction in the electrostatic 280

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
4 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

F
Fig. 3. (a) UV–vis spectrophotometers of graphene oxide nanofluids at various concentrations, (b) The relationship between light absorption and concentration of graphene oxide

O
nanofluid at wavelength 229 nm.

O
R
P
D
E
T
C
E
R
R
O
C
N
U

Fig. 4. Intensity based particle size distribution of (a) GO-0.01%, (b) GO-0.05%, (c) GO-0.1% and (d) GO-0.5%, (e) Z-average size of all graphene oxide nanofluids at different concentrations.

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 5

[43]. Another observation of viscosity behavior is that non-Newtonian 326

behavior of graphene oxide NFs at lower shear rates was significant at 327
NFs with higher concentrations. In general, NFs can exhibit both Newto- 328
nian and non-Newtonian behaviors, depending on particle size and 329
shape, concentration, base liquid viscosity, and solution chemistry 330
[16]. As reported by Mehrali et al. [46], the structure of the fluid mole- 331
cules changes temporarily as the spindle rotates in the fluid at low 332
shear rates, and gradually molecules aligns themselves in the direction 333
of increasing shear, resulting in less resistance and reduction in viscosi- 334
ty. However, when the shear rate is high enough, the maximum amount 335
of possible shear ordering was attained. The larger aggregates were bro- 336
ken down to smaller sizes, resulting in less friction and therefore less 337
viscosity. 338

F
3.4. Thermal conductivity of graphene oxide nanofluids 339

O
Fig. 5. Zeta potential of graphene oxide nanofluids at different concentration.
Thermal conductivity of all NFs at different temperatures was 340
presented in Fig. 7. The calibration tests for DI water were verified 341

O
281 repulsive force between graphene oxide as predicted by colloidal theory by data reported by other researchers [47,48], and the results were 342
282 [38,39]. obtained within 2–4% accuracy. All graphene oxide NFs showed 343

R
283 Fig. 5 shows the zeta potential of graphene oxide NFs at different higher thermal conductivity compared to DI water at all tempera- 344
284 concentrations. A pervious study of graphene oxide by Lee et al. [40] re- tures. Graphene oxide NFs at 0.01 and 0.5 wt.% showed 8.7% and 345
285 ported that a graphene oxide colloid solution at a pH range of 2–4 shows 19.9% thermal conductivity enhancement with respect to DI water 346

P
286 negative zeta potential, which agrees with this data. According to the at 25 °C, respectively. This thermal conductivity enhancement can 347
287 ASTM, colloid solutions with zeta potentials higher than − 40 mV are be justified based on the specific properties of graphene oxide, such 348
288 considered to have good stability [41]. The zeta potential of 0.01 and as high aspect ratio and stiffness, that are helpful for increasing the
D 349
289 0.05 wt.% graphene oxide NFs was − 31 ± 0.7 mV and − 36 ± thermal conductivity of graphene oxide NFs [49]. By increasing the 350
290 0.5 mV, which implies the instability of NFs and their intent for aggrega- temperature, thermal conductivity of all graphene oxide NFs in- 351
291 tion. These zeta potential values confirm the aggregation behavior of creased. However, concentration and temperature showed a differ- 352
E
292 graphene oxide by increasing the concentration. Graphene oxide NFs ent trend. The effect of temperature on enhancement of thermal 353
293 at 0.1 and 0.5 wt.% showed zeta potential of − 41 ± 1 mV and conductivity was more significant for graphene oxide NFs at the 354
T

294 −50 ± 3 mV, respectively. It implies that by increasing the concentra- highest concentration, 0.5 wt.%, compared to other concentrations. 355
295 tions of graphene oxide NFs, the negative electrical surface charge of Graphene oxide NFs at 0.01 and 0.5 wt.% showed 8.8% and 55.2% 356
C

296 graphene oxide (corresponding to ionization of carboxylic acid groups) thermal conductivity enhancement at 60 °C compared to their ther- 357
297 increased [42]. mal conductivity at 25 °C, respectively. This trend can be explained 358
by the fact that increasing the temperature resulted in reduction of 359
E

298 3.3. Rheological properties of graphene oxide nanofluids base fluid viscosity, but increment in Brownian motion of nanoparti- 360
cles, consequently convection-like effects, are remarkably increased, 361
R

299 To determine the rheological behaviors of graphene oxide NFs, the resulting in increased conductivities [50]. 362
300 viscosity of all samples at various shear rates was measured at the An interesting point is that for all three temperatures, the thermal 363
301 three temperature points of 25, 40, and 60 °C, Fig. 6. Viscosity of all conductivity enhancements showed linear relationships with con- 364
R

302 graphene oxide NF samples decreases by increasing temperature. This centration where increasing the concentration of graphene oxide re- 365
303 viscosity reduction behavior can be explained by the fact that higher sulted in thermal conductivity enhancement (Fig. 7). However, 366
O

304 temperature increases the Brownian motion, thermal movement of increasing the concentration from 0.1% to 0.5% insignificantly affect- 367
305 molecules and their average speed, resulting in weakened intermolecu- ed thermal conductivity enhancement. Increasing the concentration 368
306 lar interaction and adhesion forces between molecules [43]. Viscosity from 0.01 to 0.1 wt.% resulted in 9.4%, 13.3%, and 40.1% thermal con- 369
C

307 behavior of 0.01 and 0.05 wt.% graphene oxide NFs (relatively lower ductivity enhancement at 25 °C, 30 °C, and 60 °C, respectively, 370
308 concentrations) versus temperature were almost similar to viscosity of whereas, increasing the concentration from 0.1 to 0.5 wt.% resulted 371
N

309 DI water (base fluid) for all the temperatures. However, 0.1 and in 0.7%, 1.8%, and 1.7% thermal conductivity enhancement at 25 °C, 372
310 0.5 wt.% (relatively higher concentrations) viscosities of NFs were in a 30 °C, and 60 °C, respectively. This behavior can be explained based 373
311 different range compared to DI water. These trends indicate that the on the aggregate size of graphene oxide particles in the base fluid. 374
U

312 base fluid has more effect on rheological properties of NFs at relatively As discussed earlier (Section 3.2), graphene oxide aggregate size in- 375
313 lower concentration, but in NFs at relatively higher concentration, the creased sharply by increasing the concentration from 0.1 to 0.5 wt.% 376
314 interaction between base fluid and nanoparticles plays an important (Fig. 4). Graphene oxide NFs at 0.5 wt.% contain micron-size average 377
315 role on rheological properties of nanofluids [44]. The increases in viscos- aggregate sizes that could not show a significant Brownian motion 378
316 ity, (μnf − μbf) / μbf) × 100, for graphene oxide NFs at 0.01 and 0.5 wt.%, enough to enhance the thermal conductivity [51]. In addition, be- 379
317 T = 25 °C, and shear rate of 100 1/s, were 38% and 130%, respectively. cause heat transfer takes place at the surface of the particle, particles 380
318 The increase in viscosity with increase in concentration was also ob- with large surface-to-volume ratios, compared to those with small 381
319 served at higher temperatures but with slight reduction in magnitude. surface-to-volume ratios, have a significantly greater impact on ther- 382
320 This phenomenon can be justified by the fact that increasing the con- mal conductivity. The surface-to-volume ratio is 1000 times larger 383
321 centration of nanoparticles induces the formation of agglomeration in for particles with a 10 nm diameter than for particles with a 10 μm 384
322 suspension [45]. Particle size distribution of NFs confirms this explana- diameter [52]. Therefore, another explanation for insignificant ther- 385
323 tion (Fig. 4). Aggregation of nanoparticles leads to the increase of inter- mal conductivity enhancement between graphene oxide nanofluids 386
324 nal shear stress because the greater force is required to dissipate the at 0.1 and 0.5 wt.% might be the small surface-to-volume ratio of 387
325 aggregated particles in the media and therefore increase the viscosity larger aggregate sizes for graphene oxide at 0.5 wt.%. 388

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
6 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

F
O
O
R
P
D
E
T
C
E
R
R

Fig. 6. Viscosity at various shear rates at 25 °C, 40 °C, and 60 °C for: (a) DI water, (b) GO-0.01%, (c) GO-0.05%, (d) GO-0.1%, and (e) GO-0.5%.
O
C
N
U

Fig. 7. (a) Thermal conductivity of graphene oxide nanofluids at different concentrations, (b) Percentage enhancement of thermal conductivity versus concentration at different
temperatures.

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx 7

389 4. Conclusions [10] T.-K. Hong, H.-S. Yang, C. Choi, Study of the enhanced thermal conductivity of Fe 455
nanofluids, J. Appl. Phys. 97 (6) (2005) 064311. 456
[11] C.H. Li, G. Peterson, The effect of particle size on the effective thermal conductivity of 457
390 In this paper, effects of particle size and viscosity on thermal conduc- Al2O3-water nanofluids, J. Appl. Phys. 101 (4) (2007) 44312. 458
391 tivity enhancement of graphene oxide NFs were examined experimen- [12] M. Chopkar, S. Sudarshan, P. Das, I. Manna, Effect of particle size on thermal conduc- 459
tivity of nanofluid, Metall. Mater. Trans. A 39 (7) (2008) 1535–1542. 460
392 tally. The graphene oxide nanosheets were synthesized based on the [13] A. Moghadassi, S.M. Hosseini, D.E. Henneke, Effect of CuO nanoparticles in enhanc- 461
393 modified Hummers methods, and their physicochemical properties ing the thermal conductivities of monoethylene glycol and paraffin fluids, Ind. 462
394 were characterized using XRD, SEM, and UV–Vis spectrophotometry. Eng. Chem. Res. 49 (4) (2010) 1900–1904. 463
[14] M.P. Beck, Y. Yuan, P. Warrier, A.S. Teja, The effect of particle size on the thermal 464
395 The graphene oxide NFs were prepared at 0.01, 0.05, 0.1, and 0.5 wt.% conductivity of alumina nanofluids, J. Nanopart. Res. 11 (5) (2009) 1129–1136. 465
396 in DI water as the base fluid. [15] A. Every, Y. Tzou, D. Hasselman, R. Raj, The effect of particle size on the thermal con- 466
397 It was observed that graphene oxide particle size distributions at all ductivity of ZnS/diamond composites, Acta Metall. Mater. 40 (1) (1992) 123–129. 467
398 the concentrations could be categorized into three size generations: [16] Y. Ding, H. Chen, Z. Musina, Y. Jin, T. Zhang, S. Witharana, W. Yang, Relationship be- 468
tween the thermal conductivity and shear viscosity of nanofluids, Phys. Scr. 2010 469
399 particle sizes b200 nm, 200 nm b particle sizes b1000 nm, and particle (T139) (2010) 014078. 470
400 sizes N 1 nm. The average particle size of graphene oxide increased [17] K. Kwak, C. Kim, Viscosity and thermal conductivity of copper oxide nanofluid dis- 471
401 linearly by increasing the concentration from 0.01 to 0.1 wt.%, but persed in ethylene glycol, Korea-Australia Rheol. J. 17 (2) (2005) 35–40. 472
473

F
[18] P.K. Namburu, D.P. Kulkarni, D. Misra, D.K. Das, Viscosity of copper oxide nanoparti-
402 significant increments were observed by increasing the concentra- cles dispersed in ethylene glycol and water mixture, Exp. Thermal Fluid Sci. 32 (2) 474
403 tion to 0.5 wt.%. The colloidal stability (Zeta potential) of graphene (2007) 397–402. 475

O
404 oxide NFs increased from − 31 mV to − 50 mV by increasing the con- [19] T.T. Baby, S. Ramaprabhu, Enhanced convective heat transfer using graphene dis- 476
persed nanofluids, Nanoscale Res. Lett. 6 (1) (2011) 1–9. 477
405 centration from 0.01 to 0.5 wt.%, respectively. The slight increment [20] A. Ghozatloo, A. Rashidi, M. Shariaty-Niassar, Convective heat transfer enhancement 478
406 on viscosity of graphene oxide NFs was observed by increasing the of graphene nanofluids in shell and tube heat exchanger, Exp. Thermal Fluid Sci. 53 479

O
407 concentration from 0.01 to 0.1 wt.%, and a severe increment was (2014) 136–141. 480
[21] W. Yu, H. Xie, X. Wang, X. Wang, Significant thermal conductivity enhancement 481
408 observed in viscosity by increasing the concentration from 0.1 to for nanofluids containing graphene nanosheets, Phys. Lett. A 375 (10) (2011) 482

R
409 0.5 wt.%. Viscosity of all the graphene oxide NFs decreased by in- 1323–1328. 483
410 creasing the temperature. [22] I. Mahbubul, R. Saidur, M. Amalina, Latest developments on the viscosity of 484
nanofluids, Int. J. Heat Mass Transf. 55 (4) (2012) 874–885. 485
411 All the graphene oxide NFs showed higher thermal conductivity 486

P
[23] W.S. Hummers Jr., R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc. 80
412 in comparison to DI water (base fluid). However, there is an opti- (6) (1958) 1339-1339. 487
413 mal concentration of graphene oxide at which both particle size [24] G. Wang, J. Yang, J. Park, X. Gou, B. Wang, H. Liu, J. Yao, Facile synthesis and charac- 488
414 and viscosity properties resulted in significant thermal conductivity terization of graphene nanosheets, J. Phys. Chem. C 112 (22) (2008) 8192–8195. 489
[25] C. Hontoria-Lucas, A. Lopez-Peinado, J.d.D. López-González, M. Rojas-Cervantes, R.
D 490
415 enhancement, and further concentration increment did not signifi- Martin-Aranda, Study of oxygen-containing groups in a series of graphite oxides: 491
416 cantly impact the thermal conductivity enhancement. Increasing physical and chemical characterization, Carbon 33 (11) (1995) 1585–1592. 492
417 the concentration of graphene oxide nanofluids to more than [26] H. He, J. Klinowski, M. Forster, A. Lerf, A new structural model for graphite oxide, 493
E
Chem. Phys. Lett. 287 (1) (1998) 53–56. 494
418 0.1 wt.% caused creation of a larger aggregate size (micron size) hav- [27] M.R. Esfahani, H.A. Stretz, M.J. Wells, Abiotic reversible self-assembly of fulvic and 495
419 ing less Brownian motion, smaller surface-to-volume ratio, and humic acid aggregates in low electrolytic conductivity solutions by dynamic light 496
T

420 higher viscosity that resulted in insignificant thermal conductivity scattering and zeta potential investigation, Sci. Total Environ. 537 (2015) 81–92. 497
[28] Malvern Instruments Ltd, Zetasizer Nano Series User Manual, MAN0317 Issue 2.2. 498
421 enhancement. Worcestershire, UK, 2005. 499
C

[29] L. Shahriary, A.A. Athawale, Graphene oxide synthesized by using modified hum- 500
mers approach, Int. J. Renew. Energy Environ. Eng. 2 (1) (2014) 58–63. 501
422 Acknowledgement [30] Z. Hajjar, A.M. Rashidi, A. Ghozatloo, Enhanced thermal conductivities of graphene 502
E

oxide nanofluids, Int. Commun. Heat Mass Transfer 57 (2014) 128–131. 503
[31] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T. 504
423 The authors gratefully thank the Center for the Management, Utiliza- Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via chemical reduction 505
R

424 tion, and Protection of Water Resources; Center for Manufacturing of exfoliated graphite oxide, Carbon 45 (7) (2007) 1558–1565. 506
[32] S. Xu, L. Yong, P. Wu, One-pot, green, rapid synthesis of flowerlike gold nanoparti- 507
425 Research; Department of Mechanical Engineering; and Department of
cles/reduced graphene oxide composite with regenerated silk fibroin as efficient ox- 508
426
Q2 Chemical Engineering at Tennessee Technological University for re- ygen reduction electrocatalysts, ACS Appl. Mater. Interfaces 5 (3) (2013) 654–662. 509
R

427 search funding and support. [33] S. Jana, A. Salehi-Khojin, W.-H. Zhong, Enhancement of fluid thermal conductivity by 510
the addition of single and hybrid nano-additives, Thermochim. Acta 462 (1) (2007) 511
45–55. 512
O

[34] D. Deng, X. Jiang, L. Yang, X. Hou, C. Zheng, Organic solvent-free cloud point 513
428 References extraction-like methodology using aggregation of graphene oxide, Anal. Chem. 86 514
(1) (2013) 758–765. 515
429 [1] H. Devalapally, A. Chakilam, M.M. Amiji, Role of nanotechnology in pharmaceutical 516
C

[35] S.S. Gupta, V.M. Siva, S. Krishnan, T. Sreeprasad, P.K. Singh, T. Pradeep, S.K. Das, Ther-
430 product development, J. Pharm. Sci. 96 (10) (2007) 2547–2565. mal conductivity enhancement of nanofluids containing graphene nanosheets, J. 517
431 [2] M.R. Esfahani, J.L. Tyler, H.A. Stretz, M.J. Wells, Effects of a dual nanofiller, nano-TiO2 Appl. Phys. 110 (8) (2011) 084302. 518
432 and MWCNT, for polysulfone-based nanocomposite membranes for water purifica- [36] H. Tang, D. Liu, Y. Zhao, X. Yang, J. Lu, F. Cui, Molecular dynamics study of the aggre- 519
N

433 tion, Desalination 372 (2015) 47–56. gation process of graphene oxide in water, J. Phys. Chem. C 119 (47) (2015) 520
434 [3] M.R. Esfahani, H.A. Stretz, M.J. Wells, Comparing humic acid and protein fouling on 26712–26718. 521
435 polysulfone ultrafiltration membranes: adsorption and reversibility, J. Water Pro- [37] M.C. Duch, G.S. Budinger, Y.T. Liang, S. Soberanes, D. Urich, S.E. Chiarella, L.A. 522
436
U

cess Eng. 6 (2015) 83–92. Campochiaro, A. Gonzalez, N.S. Chandel, M.C. Hersam, Minimizing oxidation and 523
437 [4] F. Shahrezaei, Y. Mansouri, A.A.L. Zinatizadeh, A. Akhbari, Process modeling and ki- stable nanoscale dispersion improves the biocompatibility of graphene in the 524
438 netic evaluation of petroleum refinery wastewater treatment in a photocatalytic re- lung, Nano Lett. 11 (12) (2011) 5201–5207. 525
439 actor using TiO2 nanoparticles, Powder Technol. 221 (2012) 203–212. [38] I. Chowdhury, M.C. Duch, N.D. Mansukhani, M.C. Hersam, D. Bouchard, Colloidal 526
440 [5] H. Presting, U. König, Future nanotechnology developments for automotive applica- properties and stability of graphene oxide nanomaterials in the aquatic environ- 527
441 tions, Mater. Sci. Eng. C 23 (6) (2003) 737–741. ment, Environ. Sci. Technol. 47 (12) (2013) 6288–6296. 528
442 [6] R. Yang, Y. Qin, L. Dai, Z.L. Wang, Power generation with laterally packaged piezo- [39] E.J.W. Verwey, J.T.G. Overbeek, J.T.G. Overbeek, Theory of the Stability of Lyophobic 529
443 electric fine wires, Nat. Nanotechnol. 4 (1) (2009) 34–39. Colloids, Courier Corporation, 1999. 530
444 [7] M.N.M. Zubir, A. Badarudin, S. Kazi, N.M. Huang, M. Misran, E. Sadeghinezhad, M. [40] S.W. Lee, K.M. Kim, I.C. Bang, Study on flow boiling critical heat flux enhancement of 531
445 Mehrali, N. Syuhada, S. Gharehkhani, Experimental investigation on the use of re- graphene oxide/water nanofluid, Int. J. Heat Mass Transf. 65 (2013) 348–356. 532
446 duced graphene oxide and its hybrid complexes in improving closed conduit turbu- [41] Zeta Potential of Colloids in Water and Waste Water, ASTM Standard D 4187–82, 533
447 lent forced convective heat transfer, Exp. Thermal Fluid Sci. 66 (2015) 290–303. American Society for Testing and Materials, 1985. 534
448 [8] M. Mehrali, E. Sadeghinezhad, S.T. Latibari, M. Mehrali, H. Togun, M. Zubir, S. Kazi, [42] D. Li, M.B. Mueller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous dispersions 535
449 H.S.C. Metselaar, Preparation, characterization, viscosity, and thermal conductivity of graphene nanosheets, Nat. Nanotechnol. 3 (2) (2008) 101–105. 536
450 of nitrogen-doped graphene aqueous nanofluids, J. Mater. Sci. 49 (20) (2014) [43] M.N. Rashin, J. Hemalatha, Viscosity studies on novel copper oxide–coconut oil 537
451 7156–7171. nanofluid, Exp. Thermal Fluid Sci. 48 (2013) 67–72. 538
452 [9] W. Yu, D.M. France, J.L. Routbort, S.U. Choi, Review and comparison of nanofluid [44] M.N. Rashin, J. Hemalatha, Synthesis and viscosity studies of novel ecofriendly ZnO– 539
453 thermal conductivity and heat transfer enhancements, Heat Transfer Eng. 29 (5) coconut oil nanofluid, Exp. Thermal Fluid Sci. 51 (2013) 312–318. 540
454 (2008) 432–460.

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006
 8 M.R. Esfahani et al. / International Communications in Heat and Mass Transfer xxx (2016) xxx–xxx

541 [45] H. Xie, L. Chen, Q. Wu, Measurements of the viscosity of suspensions containing [49] A. Yu, P. Ramesh, M.E. Itkis, E. Bekyarova, R.C. Haddon, Graphite nanoplatelet-epoxy 554
542 nanosized Al2O3 particles, High Temp. High Pressures 37 (2007) 127–135. composite thermal interface materials, J. Phys. Chem. C 111 (21) (2007) 7565–7569. 555
543 [46] M. Mehrali, E. Sadeghinezhad, S.T. Latibari, S.N. Kazi, M. Mehrali, M.N.B.M. Zubir, [50] S.P. Jang, S.U. Choi, Role of Brownian motion in the enhanced thermal conductivity 556
544 H.S.C. Metselaar, Investigation of thermal conductivity and rheological properties of nanofluids, Appl. Phys. Lett. 84 (21) (2004) 4316–4318. 557
545 of nanofluids containing graphene nanoplatelets, Nanoscale Res. Lett. 9 (1) (2014) [51] D. Kwek, A. Crivoi, F. Duan, Effects of temperature and particle size on the thermal 558
546 1–12. property measurements of Al2O3–water nanofluids, J. Chem. Eng. Data 55 (12) 559
547 [47] M. Mehrali, E. Sadeghinezhad, M.A. Rosen, A.R. Akhiani, S.T. Latibari, M. Mehrali, (2010) 5690–5695. 560
548 H.S.C. Metselaar, Heat transfer and entropy generation for laminar forced convec- [52] S. Lee, S.-S. Choi, S. Li, J. Eastman, Measuring thermal conductivity of fluids contain- 561
549 tion flow of graphene nanoplatelets nanofluids in a horizontal tube, International ing oxide nanoparticles, J. Heat Transf. 121 (2) (1999) 280–289. 562
550 Communications in Heat and Mass Transfer 66 (2015) 23–31.
551 [48] M.L. Ramires, C.A.N. de Castro, Y. Nagasaka, A. Nagashima, M.J. Assael, W.A.
552 Wakeham, Standard reference data for the thermal conductivity of water, J. Phys.
553 Chem. Ref. Data 24 (3) (1995) 1377–1381.

F
O
O
R
P
D
E
T
C
E
R
R
O
C
N
U

Please cite this article as: M.R. Esfahani, et al., Effect of particle size and viscosity on thermal conductivity enhancement of graphene oxide
nanofluid, Int. Commun. Heat Mass Transf. (2016), http://dx.doi.org/10.1016/j.icheatmasstransfer.2016.06.006