Experimental investigation of thermophysical properties of Al2O3–water nanofluid: Role of surfactants

Accepted Manuscript Experimental investigation of thermophysical properties of Al2O3–water nanofluid: Role of surfactants

Pritam Kumar Das, Nurul Islam, Apurba Kumar Santra, Ranjan Ganguly PII: DOI: Reference:

S0167-7322(17)30851-6 doi: 10.1016/j.molliq.2017.04.099 MOLLIQ 7255

To appear in:

Journal of Molecular Liquids

Received date: Revised date: Accepted date:

25 February 2017 19 April 2017 20 April 2017

Please cite this article as: Pritam Kumar Das, Nurul Islam, Apurba Kumar Santra, Ranjan Ganguly , Experimental investigation of thermophysical properties of Al2O3–water nanofluid: Role of surfactants. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Molliq(2017), doi: 10.1016/ j.molliq.2017.04.099

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT EXPERIMENTAL INVESTIGATION OF THERMOPHYSICAL PROPERTIES OF Al2O3– WATER NANOFLUID: ROLE OF SURFACTANTS Pritam Kumar Das, Nurul Islam, Apurba Kumar Santra,* Ranjan Ganguly Department of Power Engineering, Jadavpur University, Kolkata, 700 098, India

T

ABSTRACT:

IP

Nanofluids are colloidal suspension of nanoparticles that have attracted much attention due to their enhanced heat transfer characteristics. Stability of the nanofluid is important for industrial

CR

applications. Although several researchers studied thermophysical properties of nanofluids, the literature is replete with contradictory results. The aim of the present work is to experimentally investigate the

US

stability and properties of Al2O3-water nanofluid and the effect of surfactant on those; the effects of both temperature (20–60 °C) and solid volume fractions (0.1–2.0%) of nanofluids are also analyzed. Stability the

nanoparticle

suspensions

is

investigated

with

different

surfactants,

when

sodium

AN

of

dodecylbenzenesulfonate (SDBS) is found to offer the best stabilization. TEM images show presence of rod-like particle of 20-70 nm size. DLS measurements indicate that the presence of SDBS surfactant

M

reduces particle clustering and polydispersity index of the suspension. Rheological measurements show that viscosity of Al2O3-water based nanofluids increases with solid volume fraction ( = 0.1–1.0%) and

ED

decreases with temperature, while the thermal conductivity increases with the increase of both solid volume fraction (0.1–2.0%) and temperature. The observed values of thermal conductivity and viscosity

PT

are compared with different established models. Finally, a sensitivity analysis is performed, which reveals that, for a given temperature, the sensitivity of thermal conductivity increases with the particle loading.

CE

The results provide comprehensive thermophysical property database of water-based Al2O3 nanofluids for

AC

different engineering applications.

KEYWORDS: Al2O3-water nanofluid; SDBS surfactant; Dispersion and stability; Thermal conductivity; Viscosity.

*

Corresponding author, e mail: [email protected]

1

ACCEPTED MANUSCRIPT 1. INTRODUCTION: Conventional heat transfer fluids such as water, ethylene glycol (EG), oil, etc. are widely used in various industries, including power generation, transportation, chemical processes, heating or cooling process, etc. But the heat transfer capacity is often poor due to the low thermal conductivity of such conventional fluids. This often limits their use in high heat flux devices, e.g., electronics cooling, material processing or solar thermal collectors. In quest of fluids of higher thermal conductivity, researchers

T

attempted to develop nanofluids possessing improved thermal properties. Nanofluids are colloidal

IP

suspension of solid particles such as TiO2, Al2O3, Cu, Si, CuO and SiO2 – with sizes of nanometer range (1–100 nm) – in conventional liquids such as water, oil, ethylene glycol (EG), etc. Since the first report of

CR

Choi [1], nanofluids have been used in many systems, such as chemical reactions [2], solar energy systems [3], thermal energy storage media [4], medical applications, cooling of electronics,

US

electrochemical energy storage systems [5], nuclear reactors [6], sensors and photocatalysis applications [7], optical devices, and many more. However, to understand the suitability of these nanofluids and use

AN

them in practical thermal applications, precise knowledge of their thermophysical properties is necessary. Till date, many experimental studies of thermal conductivity and viscosity of nanofluids have been performed, albeit the results have often diverged from one group to other [8, 9]. Hence, a comprehensive

M

and systematic characterization of nanofluids has still remained an important issue for their reliable use in thermal applications. In this study, aluminum-oxide (Al2O3) nanoparticle has been chosen as the dispersed

ED

phase of the nanofluid, since it has high chemical and physical stability, low cost, easy availability and feasibility of large-scale, industrial production. Al2O3 nanofluid in water has also been pitched as a

PT

promising heat transfer medium [6]. Our literature survey indicates a paucity of research data on the stability of water-based Al2O3 nanofluids. Therefore, the present study attempts to characterize the effect Al2O3 nanofluids.

CE

of surfactants and particle concentration on the stability and thermophysical properties of water-based

AC

Preparing a stable and homogeneous nanofluid is an important prerequisite to achieve an improved thermal conductivity. Das et al. [10] measured the thermal conductivity of Al2O3-water and CuO-water nanofluids using temperature oscillation technique at different temperatures (21°C–51°C) and volume fractions ( = 1–4%). The results showed that with increasing temperature and  of nanofluid, its thermal conductivity could be enhanced up to 24.3% compared to the base fluid. Chon et al. [11] measured the thermal conductivity of Al2O3 nanofluid in the temperature range of 21–71 °C and nanoparticle size ranging between 11 – 150 nm (nominal diameters) using the transient hot wire method. They found that the thermal conductivity increased with temperature and decreased with nanoparticle size. A similar study by Zhu et al. [12] showed that the thermal conductivity of Al2O3-water nanofluids increased up to 10.1% with 0.15 wt% nanoparticle compared to the host fluid. Li and Peterson [13]

2

ACCEPTED MANUSCRIPT evaluated experimentally the effect of particle size (36 nm and 47 nm diameter) and temperature (27–37 ˚C) on thermal conductivity of Al2O3-water nanofluid for  = 0.5–6.0%. They found that at  = 6%, the thermal conductivity enhancement (over the base case of pure host fluids and no particles) rose from 26% to 28% when the nanoparticle size was decreased from 47 nm to 36 nm. Murshed et al. [14] measured the thermal conductivity and viscosity of aqueous TiO2 and Al2O3 nanofluids having 15 and 80 nm particles, respectively. For different extents of solid volume fractions (15%), their result showed that the effective

T

thermal conductivity and viscosity of nanofluids increased with temperature as well as . Experimental

IP

and theoretical investigations of effective thermal conductivity and viscosity of Al2O3-water nanofluid in the range of  = 0.33–5% were performed by Chandrasekar et al. [15]. They found that thermal

CR

conductivity and viscosity of nanofluid increased with . They also proposed a new model that could be used for the prediction of thermal conductivity and viscosity of Al2O3-water nanofluids. From the above

US

reports, it can be stated that thermal conductivity of nanofluid depends strongly on , temperature, shape and diameter of nanoparticles. It has been largely agreed by the researchers that thermal Brownian motion

AN

of the nanoparticles plays a very important role in increasing thermal conductivity of nanofluids [16]. Several researchers, who observed that effective thermal conductivity of nanofluid increases with increase

M

in temperature and nanoparticle concentration, suggested appropriate models [17, 18, 19] to explain such variations.

ED

Besides thermal conductivity, viscosity is also an important property of the nanofluids as it is relevant for estimating the pumping power requirement to ensure the flow of nanofluid in the flow circuits of the relevant engineering applications [20]. If the viscosity of the nanofluid increases, the pressure drop

PT

across the flow section also increases, resulting in a decrease of the flow and convective heat transfer rate. Nguyen et al. [9] investigated the influence of both temperature and particle size on the viscosities of

CE

Al2O3 (average particle sizes of 36 and 47 nm), and CuO (29 nm) water-based nanofluids where temperature varied from ambient to 75 °C. The results showed that for  < 4%, the viscosity of Al2O3-

AC

water nanofluid was independent of particle size. However, for higher , viscosity increased with increased particle size. Similar types of experiments [21] showed that viscosity increased with  and decreased with temperature. On contrary, Chen et al. [8] found that the relative viscosity for TiO2-water and Al2O3-propylene glycol nanofluids showed temperature-independent behavior over a wide temperature range (20-60 °C). Sekhara and Sharma [22] reported the effect of variation of temperature (25 – 45 °C) on viscosity of Al2O3-water nanofluid at low  (0.01–1.0%) and found a nonlinear increment of viscosity with , which they attributed to the aggregation of particles. Wang et al. [23] also measured viscosity of Al2O3-water nanofluids at  = 3% and observed 20–30% higher in viscosity compared to that of the base fluid.

3

ACCEPTED MANUSCRIPT Effects of surfactant on the nanofluid thermal conductivity and viscosity have also been investigated by several researchers. Wang et al. [24] reported that an optimal concentration of sodium dodecylbenzenesulfonate (SDBS) can result in an enhanced thermal conductivity in water-based Cu and Al2O3 nanofluids. Assael et al. [25] prepared nanofluids by dispersing multi walled (MWCNT) and double walled (DWCNT) carbon nanotube suspensions in water, where cetyl trimethyl ammonium bromide (CTAB) and Nano-sperse AQTM [26] were employed as surfactants. The maximum thermal

T

conductivity enhancement obtained was 34% for a 0.6 vol.% MWCNT suspension in water with CTAB.

IP

Hung et al. [27] found that addition of chitosan (dispersant) in MWCNTs-water nanofluids increased the viscosity of those nanofluids up to 233% compared to that of water. Li et al. [28] investigated the

CR

surfactant concentration on the viscosity of magnetic nanofluids and found that the viscosity of nanofluids increased with increase in concentration of the surfactant. Phuoc et al. [29] and Wang et al. [30] also

US

found that the addition of surfactant in MWCNT-water nanofluid changed its flow behaviour from Newtonian to non-Newtonian.

AN

The literature lacks in a comprehensive study on stability and characterization of nanofluids and the influence of surfactants; the results so far reported are often inconclusive, sometimes even contradictory. Therefore, a systematic characterization of the thermophysical properties of nanofluid is

M

necessary for its deployment in industrial applications. In this present experimental work, preparation and characterization of water-based Al2O3 nanofluids, both with and without surfactant, have been performed.

ED

A two-step method for nanofluid synthesis has been followed as it offers a facile and rapid method for preparation of large volumes of nanofluids with greater control over nanoparticle concentration and size

PT

distribution than a single-step synthesis. Nanoparticle size distribution has been measured using dynamic light scattering (DLS) technique. The microstructures and particle size have also been examined using

CE

transmission electron microscopy (TEM). Thermal conductivity and viscosity of Al2O3-water nanofluids have been measured at different temperatures (2060 °C) and  (0.12.0% for thermal conductivity, and

AC

0.11.0% for viscosity measurement), using different surfactants for stabilization. The experimental results of thermal conductivity and viscosity of Al2O3-water and Al2O3-water-SDBS nanofluids have been compared with the major correlations from the literature. A sensitivity analysis has been performed to assess the influence of particle loading on thermal conductivity at different temperatures. The results provide comprehensive thermophysical property database of water-based Al2O3 nanofluids, which can be used for computational fluid dynamics (CFD) analysis and design of different engineering applications deploying Al2O3 nanofluids.

4

ACCEPTED MANUSCRIPT 2. MATERIALS AND METHODS: 2.1 Materials: Distilled water (Merck Millipore) of 99.7% purity was used as base fluid. Al 2O3 nanoparticles, with an average particle diameter of ~ 50 nm (Sigma Aldrich Chemicals Limited, Germany), were used as the dispersed phase. Different surfactants (Merck Millipore), viz., cetyl trimethyl ammonium bromide (CTAB), sodium dodecylbenzenesulfonate (SDBS) and sodium dodecyl sulfate (SDS) were used to help

T

disperse the particles and improve the stability of suspensions.

IP

2.2 Nanofluid preparation:

The procedure followed an earlier approach by this group, described elsewhere [31]. Nanofluid

CR

samples were prepared by dispersing the Al2O3 nanoparticles in distilled water (host fluid) by two-step method. Measured masses of nanoparticles were dispersed into appropriate base liquid of 100 ml to

US

prepare nanofluid samples of varying , ranging from 0.1 to 2.0%. Gravimetric measurements of nanoparticles and dispersant were performed using a high-precision electronic balance (Sartorius, BSA

AN

224S-CW, range 220 g, accuracy 0.1 mg). The resulting  of synthesized nanofluid was calculated by using the formula

=

M

,

[1]

1000 kg/m3),

ED

denotes the density of alumina nanoparticles (3960 kg/m3),

where

the weight of base fluid (100 ml), and

the density of base fluid

the weight of nanoparticle. For

PT

preparing the non-surfacted nanofluid, appropriate volume of Al2O3 nanoparticles were added in distilled water and the mixture was stirred in a magnetic stirrer (Remi, model name 2MLH) for 1 hour so that the

CE

particles get dispersed homogeneously. The resulting suspension was sonicated for 15 minutes using a probe sonicator (PCI Analytics, PKS-750FM, at a power input rate of 0.167 J/s) to break the agglomerates remaining in the sample even after stirring. The sonicated suspensions, bearing different particle loadings,

AC

were then observed for several hours to check if the particles settled under gravity. Another set of nanofluid samples was prepared that contained any of the three surfactants, viz., sodium dodecylbenzenesulfonate (SDBS), sodium dodecyl sulfate (SDS), and cetyl trimethyl ammonium bromide (CTAB). For these samples, the concerned surfactant was first mixed with distilled water and stirred in the magnetic stirrer for 1 hour to ensure homogenous mixing of surfactant with the host fluid. Measured quantity of nanoparticles was then added to the surfactant-water mixture and stirred and sonicated following the same protocol described above. For all three types of surfactants, different ranges of particle to surfactant mass ratios were tried; CTAB and SDS did not exhibit satisfactory stabilization with Al2O3 nanoparticle, whereas SDBS (at a particle to surfactant mass ratio of 2:1) showed better stabilization. The

5

ACCEPTED MANUSCRIPT prepared samples using SDBS was found to remain stable for several hours with negligible particle agglomeration and settling; periodic stirring (every 40 – 60 minutes) further ameliorated the issue of particle sedimentation. 2.3 Characterization of nanofluids: Morphology and size of Al2O3 nanoparticles were analyzed by both Transmission Electron Microscope (TEM, JEOL JEM-2100, operating at an acceleration voltage of 200 kV) and dynamic light

T

scattering (DLS, Zetasizer Nano ZS, Malvern, 0.3 nm – 10.0 µm) instrument. The samples for TEM

IP

analyses were prepared on transparent and thin carbon-coated, 300 mesh copper grids by depositing ~1µL droplets of freshly prepared nanofluids. Samples were air-dried for about 2-3 minutes in a controlled

CR

environment and further dried in a desiccator for ~48 h. Figures 1a and 1b show the TEM images of Al2O3-water and Al2O3-SDBS-water nanofluids, respectively, for  = 0.5%. Figure 1a shows Al2O3

US

nanoparticles appearing as black rod-like formations, where the particle size is found to vary in the range of 20  60 nm. The TEM image of SDBS-stabilized Al2O3 nanofluids (figure 1b) shows the black spots of

AN

the particle varying in the size range of 30 nm70 nm. DLS analysis was used to get a quantitative estimate of the particle size distribution. The analysis was conducted at 25 °C, with freshly prepared dilute sample. Particle size distribution for the Al2O3-water nanofluid is shown in figure 2a; the DLS

M

analysis shows a number-averaged particle size of 222 nm while the polydispersity index (ratio of the

ED

mass-weighted average particle diameter to the number-weighted average particle diameter) is 1.23. The observed particle size is much larger than the specified nanoparticle size (50 nm), indicating the

AC

CE

PT

possibility of formation of thermodynamically stable clusters of particles.

(a)

(b)

Figure 1: TEM images showing the microstructure of (a) Al2O3 and (b) Al2O3- sodium dodecylbenzenesulfonate (SDBS) water nanofluid for φ = 0.5%. The elongated dark spots represent the Al2O3 particles.

6

CR

IP

T

ACCEPTED MANUSCRIPT

Figure 2: Particle size distribution through DLS in Al2O3 -water nanofluids (a) without surfactant, and (b)

US

with SDBS surfactant.

Figure 2b shows the number distribution of particles in the SDBS-stabilized Al2O3-water

AN

nanofluid; the plot shows a number-averaged diameter of 75 nm and a polydispersity index is 1.19 – both lower than those of the non-surfacted type. The reduced value of polydispersity index of the surfacted nanofluid is indicative of a better homogeneity and lower clustering; the average particle size appears

M

slightly larger than the specified nanoparticle size possibly because of the surfactant layer.

ED

Figure 3a and 3b shows the variation of pH with  at different temperature for Al2O3-water and Al2O3-water-SDBS nanofluids respectively. While the pH value of both non-surfacted and surfacted nanofluids remains nearly insensitive to particle loading (pH decreases marginally with increased  due to

PT

selective adsorption of the OH- ions at the Stern layer on the Al2O3-nanoparticles [32]), it has a more pronounced effect of temperature for the non-surfacted nanofluid. The ionic dissociation of water

CE

increases with temperature, resulting in higher H+ loading (leading to lowering of pH) [33]. On the other hand, being an anionic surfactant, aqueous SDBS solution exhibits a strong basic nature, which is

AC

observed from the high value of pH (>10) for the surfacted nanofluid (see figure 3b). Since the particle to SDBS mass ratio was held constant (2:1), the amount of surfactant was increased with particle loading, leading to an increase in the pH with . The pervasive presence of SDBS also masks out the effect of temperature on the pH, and hence the isothermal pH curves in figure 3b runs close to each other.

7

CR

IP

T

ACCEPTED MANUSCRIPT

(a)

(b)

US

Figure 3: Variation of pH with φ for (a) Al2O3-water nanofluid (without surfactant) and (b) Al2O3-SDBS stabilized nanofluid at different temperature.

AN

2.4 Thermal conductivity measurements of test fluid:

Thermal conductivity (TC) of the Al2O3 nanofluids, for different  (0.12.0%) and temperature

M

(20 °C60 °C), was measured using a KD2 Pro thermal analyzer (Decagon Devices, WA, USA). Figure 4 describes the experimental set-up for conductivity measurement. A thermostatic bath (Model no.-CTB 07)

ED

is used to maintain the sample temperature (within a range of 0.1 °C of the set value). TC measurement was based on transient hot wire technique using a KS-1 sensor, which has an accuracy of 5%. During

PT

measurement, the 60-mm long and 1.3-mm diameter stainless steel probe was fully immersed in the nanofluid sample, which was kept in a cylindrical glass container (25 mm diameter and 80 mm height). The samples were maintained at different stipulated temperatures by immersing them in the constant

CE

temperature bath. Before analyzing the thermal characteristics of the prepared nanofluids, the sensor was calibrated every time with the base liquid (water) and glycerin at a room temperature of 25 °C. Thermal

AC

conductivity measurement at a given  and temperature lasted for 30 to 40 minutes (excluding the time taken for thermally equilibrating the thermostatic bath and the nanofluid sample), when 5 to 8 runs were taken (and the average values were reported). With every sample of nanofluid (i.e., at a given ), measurements were also made at different sample temperatures. To prevent any possibility of particle sedimentation and the resulting deviation in thermal conductivity data, the nanofluid sample was stirred for 10 to 15 minutes between two consecutive runs (at two different temperatures).

8

CR

IP

T

ACCEPTED MANUSCRIPT

US

Figure 4: Schematic diagram of experimental setup used to measure the thermal conductivity of nanofluids.

AN

An uncertainty analysis has been carried out by using the standard technique [34]. The contributing factors towards the overall uncertainty

in the measured thermal conductivity are the

M

individual uncertainties in measurement of particle volume fraction

and

,

. Therefore,

,

[2]

denote the variation of thermal conductivity with respect to volume fraction and

PT

where

ED

and the device uncertainty of the thermal conductivity meter itself

, nanofluid temperature

temperature, respectively. The maximum uncertainty in measured thermal conductivity of the Al2O3-

CE

water nanofluid samples, with and without SDBS surfactant, was found to be ±5.0%. 2.5 Viscosity measurements of test fluid: A standard programmable viscometer (model DV-II + Pro, Brookfield Instruments) was used to

AC

measure the viscosity of the nanofluid samples. Figure 5 describes the experimental set-up of the viscometer. A spindle (S18, range of 1-2000 cp), was connected to the viscometer and inserted into the corresponding chamber, which contained a minimum of 16 ml of sample. The machine was connected to a computer through an RS232 cable, and RHEOCALC V3.3 software was used for data collection and storage. Measurements were performed at different temperatures and particle loadings of the nanofluids. For each sample, temperature was varied from 20 – 60°C at increments of 5°C while the measurements were taken at a fixed shear rate of 122.3 s-1. The same thermostatic bath (used in TC measurement) was attached to the viscometer cup to maintain the temperature within ±0.1°C of the set value. The experiments were conducted at least ten times for generating the datasets for the analysis. Similar to the

9

ACCEPTED MANUSCRIPT case of thermal conductivity (see Eq. [2]), an uncertainty analysis in measurement of viscosity data has also been carried out. The maximum uncertainty in measured viscosity of the Al2O3-water nanofluid

AN

US

CR

IP

T

samples, both with and without SDBS surfactant, was found to be ±1 %.

Figure 5: Schematic Diagram of experimental setup for measurement of viscosity of nanofluid.

M

3. RESULTS AND DISCUSSIONS:

3.1 Effect of particle loading and temperature on thermal conductivity:

ED

Thermal conductivity of Al2O3 nanofluid was measured at different temperatures (20 – 60C) and

 = 0.1–2.0%. Figures 6a and 6b show the variation of thermal conductivity with , at different temperatures, for Al2O3-water nanofluid samples without surfactant and with SDBS surfactant

PT

stabilization. Thermal conductivity measurements for each  and at each temperature was repeated 5 to 8 times; the largest standard deviation in thermal conductivity data was found to be ±0.047 W/m-K. Results

CE

show that thermal conductivity of nanofluid increases with the temperature as well as . The enhancing effect of temperature on thermal conductivity may be attributed to the increased intensity of the Brownian

AC

motion of nanoparticles with the increase of temperature. Microscopic motion of suspended nanoparticles due to Brownian, van der Waals, and electrostatic forces also play significant role. Nanofluids exhibit higher thermal conductivity than their base fluids even at very low concentration of suspended nanoparticles. Thermal conductivity of liquids enhances due to both static and dynamic factors [35]. Brownian motion is caused by the random bombardment of liquid molecules, causing the particles to randomly move through the liquid, thereby enabling stronger transport of heat. This translates into an enhanced effective thermal conductivity. When the size of the nanoparticles in a nanofluid becomes less than the phonon mean-free path, phonons no longer diffuse across the nanoparticle but move ballistically without any scattering [35]. However, the mechanism responsible for ballistic phonon transport to be

10

ACCEPTED MANUSCRIPT more effective than a very-fast diffusion phonon transport is still nebulous, falling short of explaining the anomalously high thermal conductivity of nanofluids at low concentrations [35]. Jang and Choi [36] also explained the role of Brownian motion in enhancing thermal conductivity of nanofluids and attributed it to the collision between nanoparticles. As can be seen from figure 6, the average value of thermal conductivity of non-surfacted Al2O3-water nanofluid at each temperature and  is higher than that of SDBS-stabilized nanofluid. Although this is in line with prior findings that the thermal conductivity of

T

SDBS-water mixture decreases with the increasing SDBS concentration [37]  the dispersants effectively

IP

suppress the thermal conductivity of the base fluid [38]  the result is somewhat inconclusive as the

6 show the standard deviation of thermal conductivity of data.

CR

difference in TC (less than 2%) is smaller than the uncertainty of TC prediction (5%). Error bars in figure

Figure 7 shows the thermal conductivity ratio (the ratio of effective thermal conductivity of the

US

nanofluid to that of the base fluid, i.e, knf /kf) of the Al2O3-water nanofluids, both without and with the surfactant (SDBS), as function of  at various temperatures (viz., 20, 40 and 60 °C). The measured

AN

thermal conductivity of nanofluids were used for comparison with those obtained from the existing correlations proposed by Maxwell [39], Timofeeva et. al. [40], and Yu and Choi [41], respectively, which

[4]

AC

CE

PT

, and

[3]

ED

,

M

describe thermal conductivity of nanofluids as functions of :

(a)

(b)

Figure 6 Variation of thermal conductivity with φ at different temperatures for Al2O3-water nanofluids (a) without surfactant, and (b) with sodium dodecylbenzenesulfonate (SDBS) surfactant.

11

ACCEPTED MANUSCRIPT .

[5]

In Eq. [5], β denotes the ratio of the nano-layer thickness to the original particle radius. Here, β = 0.1 is used to calculate the thermal conductivity of nanofluid [41]. It is observed that for the temperature range and φ reported in figure 7, the models proposed by Maxwell (Eq. [3]), Timofeeva and Yu (Eq. [4]) and Choi (Eq. [5]) underpredicts the thermal conductivity values, particularly at low values of φ. Several factors may be attributed to this observed enhancement of thermal conductivity of the tested nanofluid

T

samples. For example, the size and clustering of the particles, the favorable influence of the nano-layer

IP

between the nanoparticles and base fluids, or even the difference in surface chemistry of the nanoparticles

CR

in the surfacted host fluid. These factors have been considered in the model proposed by Patel et al. [19], which is described as:

where

, and

. Here

is the Brownian velocity of the nanoparticles,

denotes the Boltzmann constant and the factor C is determined from the experimental data.

AN

where,

[6]

US

,

Figure 8 shows the variation of thermal conductivity ratio (knf / kf) of Al2O3-water and Al2O3-SDBS-water

M

nanofluids as a function of  at different temperatures (viz., 20, 40 and 60 °C). The experimental values of thermal conductivity ratio for φ > 1.0% are well predicted by Patel model [19] for C = 25,000.

ED

However, even this model underpredicts the knf /kf ratio for the lower range of  investigated here. A more comprehensive model that takes into account the influence of surfactant and particle size would require

PT

consideration of a larger spectrum of parametric study, involving particles of different size and shape,

AC

CE

nature of surfactant, and host fluid, which is beyond the scope of current work.

12

with φ, both without (marked as Al2O3) and with

ratio with φ, both without (marked as Al2O3) and with (marked as Al2O3-SDBS) surfactant,

AN

(marked as Al2O3-SDBS) surfactant, at different

Figure 8: Variation of thermal conductivity

US

Figure 7: Variation of thermal conductivity ratio

CR

IP

T

ACCEPTED MANUSCRIPT

at different temperatures for Patel model [19]

underpredicts the knf / kf ratio, particularly at low φ.

(considering C = 25000).

M

temperatures; existing thermal conductivity models

ED

3.2 Sensitivity analysis of Al2O3 nanofluids thermal conductivity: The sensitivity analysis describes how much the thermal conductivity changes with the changes

PT

of . Figure 9a and 9b illustrate the results of thermal conductivity sensitivity analysis for  varying in the range of 0.1–1.0% at different temperatures (20 – 60 C) for Al2O3-water and Al2O3-SDBS-water nanofluids, respectively. The sensitivity analysis is performed by measuring the change in thermal

CE

conductivity due to a 100% change in particle loading [42]. For example, at  = 1 % and temperature of

AC

20 C, the sensitivity is evaluated as: [7]

Results indicate that at a particular φ, the sensitivity of thermal conductivity for both nonsurfacted and surfacted types of nanofluids increases significantly with increase of φ. Depending upon the nanofluid temperature, the sensitivity of thermal conductivity of the Al2O3-water nanofluid increases by about 4 to 9 times as  increases from 0.1 to 1%; for Al2O3-SDBS-water nanofluid, a similar increase in  raises the sensitivity to the range of 5-12 times. The trends in figure 9 show the importance of adding nanoparticles at higher . It is also found that thermal conductivity sensitivity of Al2O3-water nanofluids is slightly greater than that of Al2O3-SDBS-water nanofluids.

13

CR

IP

T

ACCEPTED MANUSCRIPT

(a)

(b)

US

Figure 9: Sensitivity analysis of thermal conductivity, at different  and temperatures, for Al2O3-water nanofluids (a) without surfactant, and (b) with sodium dodecylbenzenesulfonate (SDBS) surfactant.

AN

3.3 Effect of particle loading and temperature on viscosity:

From the literature study, it was found that the viscosity of nanofluid depends on , particle size

M

and morphology, and shear rate. Figures 10a and 10b show the variation of viscosity of Al2O3-water nanofluid, without and with the SDBS surfactant, respectively, as functions of temperature (20–60 oC),

ED

for different  (0.1–1.0%) at a constant shear rate (122.3 s-1). Data-points in figures 10 a and 10b were obtained from runs that were repeated for at least 10 times, when the standard deviation was found to be

PT

less than 0.03%. Therefore the error bars in figure 10 are not discernable. The result shows that the viscosity of nanofluid decreases with increase of temperature while the same increases with the . The

CE

viscosity of nanofluid at  = 1.0%, for Al2O3-water and Al2O3-SDBS-water nanofluids are found to be 1.5 mPa.S and 3.23 mPa.S, respectively. Comparing figures 10 a and b, it emerges that the viscosity of Al2O3-SDBS-water nanofluid is greater than the nanofluid without surfactant for all temperature and 

AC

values. Figure 10 a shows that for the non-surfacted nanofluid, the increase of viscosity with  is marginal at 25 C – the viscosity increases from 1.32 cP to 1.46 cP as  increases from 0.1% to 1.0%. On the contrary, the viscosity of the surfacted nanofluid increases from 1.38 cP to 3.05 cP at the same temperature. The presence of micelles in the surfacted fluid may be implicated to the increased viscosity [43]. Increment of surfactant concentration above the critical micelle concentration (CMC) in nanofluids has been found to form multilayer adsorption of surfactant molecules on the surface of nanoparticles, while the surplus molecules scattered in the water forms surfactant micelles [43]. For long chain surfactants, viscosity increases quickly at low or moderate concentrations. In this situation, the size of micelles grows gradually with the increase of concentration [44]. The CMC of SDBS surfactants is 3 mM

14

ACCEPTED MANUSCRIPT at room temperature [45]. In the present experiment, the surfactant is added in proportion with the nanoparticles. The concentration of SDBS surfactant in our nanofluid samples increased from 0.57 mM at

 = 0.1 to 11.62 mM for = 1.0. This clearly indicates that at higher , the micellar concentration

AN

US

CR

IP

T

increases. This is why the viscosity of the surfacted nanofluids significantly increases at higher .

(a)

(b)

M

Figure 10: Variation of viscosity with temperature for at different φ for the nanofluids (a) without surfactant, and (b) with SDBS surfactant. Results obtained from at least 10 readings are reported, standard

ED

deviation is below 0.03%.

Figure 11 shows the variation of viscosity ratio nf/f (i.e., the nanofluid viscosity to the viscosity

PT

of its corresponding base fluids) as a function of , for both Al2O3-water and Al2O3-SDBS-water nanofluids, at 30°C. As already mentioned before, that the presence of SDBS-micelles significantly alters

CE

the viscosity of the base fluid. Therefore, for the case surfacted nanofluids, viscosity of water-surfactant mixtures (with the same mass ration of SDBS to water as in the original surfacted nanofluid) was treated

AC

as the base fluid viscosity f. The experimental values of viscosity ratio are compared with two wellestablished models proposed by Buongiorno [46], and Maiga [47], which are represented by Eqs. [8] and [9], respectively. It is clear from the figure that both the models underpredicts the increase in viscosity with .

[8] [9]

15

US

CR

IP

T

ACCEPTED MANUSCRIPT

Figure 11: Variation of viscosity ratio of Al2O3 nanofluids (with and without SDBS surfactant) with φ at

AN

two different temperatures and compared with two models [43, 44]. Figures 12a and 12b show the variation of Prandtl number of the Al2O3-water nanofluids, both

M

without and with SDBS surfactant, respectively, with temperature for different particle concentrations. The Prandtl numbers of both types of nanofluids are found to increase with concentration and decrease

, and

[10]

denote the viscosity and specific heat of the nanofluid, respectively. The viscosity

PT

where

ED

with temperature. Prandtl number is defined as:

AC

mixing rule, i.e.,

CE

value is obtained directly from figure 10, while the specific heat and density (

and

) are evaluated using the

[11]

.

[12]

As apparent from figure 12, Pr decreases with increase in temperature, since the thermal conductivity increases and viscosity goes down with increase in nanofluid temperature. For the nonsurfacted nanofluids, particle loading is found to have little effect on the Prandtl number (figure 12a); for all values of , the Pr decreases nearly to the same extent from ~9.5 at 20 oC to ~ 6.4 at 60 oC. On the contrary, Pr values of the surfacted nanofluid strongly depend on the particle loading (and hence the surfactant concentration). At large surfactant loading (large ), the high value of viscosity (see figure

16

ACCEPTED MANUSCRIPT 10b), results in the high Pr. For lower surfactant loading ( = 0.1), the Pr versus temperature curve

US

CR

IP

T

resembles that of the non-surfacted nanofluid.

(a)

(b)

AN

Figure 12: Variation of Prandtl number with temperature at different φ for Al2O3-water nanofluids (a) without surfactant, and (b) with SDBS surfactant.

M

Figure 12b provides important information regarding selection of the appropriate  of the nanofluid for a particular heat transfer application. It is imperative that for applications warranting high

ED

Pr fluids, SDBS-surfacted Al2O3-water nanofluids may be used, although it might call for higher pressure drop to induce forced convection. For applications involving quick circulation cycles, and where

PT

intermittent sonication or stirring of nanofluid may be provided (to facilitate particle suspension), nonsurfacted nanofluid may be preferred because of their low viscosity.

CE

4. CONCLUSIONS:

In this paper, the thermal conductivity and viscosity of water-based nanofluids containing Al2O3 nanoparticles (~ 50 nm diameter), both with and without surfactant, have been investigated

AC

experimentally. Before measurement, synthesis and characterization of nanofluids have been performed for different particle , temperature, surfactant type and loading. Nanofluid properties at five different  were studied (0.1%, 0.5%, 1.0%, 1.5% and 2.0%), while the temperature was varied in the range of 20–60 °C. Stability of the nanofluids has been tested with different surfactants sodium dodecylbenzenesulfonate (SDBS), sodium dodecyl sulfate (SDS), and cetyl trimethyl ammonium bromide (CTAB) in the base fluid (water), while particle distribution and morphology have been studied using TEM and DLS. Salient feature of the measured thermophysical properties exhibit the following trends: 

SDBS surfactant, at particle to surfactant mass ratio of 2:1, offered the best stabilization for the Al2O3-water nanofluids, while SDS and CTAB yielded poor stability.

17

ACCEPTED MANUSCRIPT 

TEM image of Al2O3-water and Al2O3-SDBS-water nanofluids indicated the presence of rod-like nanoparticles ranging between 20 to 70 nm. DLS measurement of the nanofluid samples showed that the number-averaged particle sizes for the non-surfacted and surfacted nanofluids were 222 nm and 75 nm, respectively, while the polydispersity indices for the samples were 1.23 and 1.19, respectively. Presence of surfactant lowered the extent of clustering and ensured a better homogeneity of particle size distribution; albeit, effective particle diameter was found to be The thermal conductivity of the nanofluids was found to increase monotonically with  and

IP



T

slightly higher due to the presence of surfactant layers. temperature. At low , the measured knf exceeded the predicted values using standard models.

CR

Dynamic contribution arising out of the particle Brownian motion may be attributed to this deviation.

The sensitivity analysis of thermal conductivity showed that at a given temperature, the

US



sensitivity increased with .

The viscosity of Al2O3-water (without surfactant) and Al2O3-SDBS-water nanofluid increased

AN



with  and decreased with temperature. The increase of viscosity with  was far more

M

pronounced in surfacted nanofluids than the non-surfacted ones. This is attributed to the formations of multilayer adsorption of the surfactant molecules on the surface of nanoparticles

ED

and presence of scattered micelle of the surfactant in the nanofluid at high , augmenting the viscosity.

Prandtl number of the nanofluids decreased monotonically with temperature, as viscosity

PT



decreased and thermal conductivity increased with temperature. The Pr of the non-surfacted nanofluids exhibited little dependence on , while that for the surfacted nanofluid increased 

CE

considerably with  (and hence the SDBS concentration). Depending upon the particle loading and SDBS concentration, Al2O3 nanofluids offer a wide

AC

range of Pr to suite specific heat transfer applications. 5. ACKNOWLEDGMENT: The authors gratefully acknowledge the UGC-sponsored DRS (SAP-II) program of Power Engineering Department of Jadavpur University, India, for its support; support provided by TEQIP-JU Phase-II in the form of a research fellowship is also gratefully acknowledged. The authors thank Prof. Kalyan K. Chattopadhyay of School of Materials Science & Nanotechnology, Jadavpur University, for lending the TEM facility. The authors are also indebted to Dr. Aparna Datta and Dr. Abhijit Saha of Inter University Consortium (IUC) for DLS measurements and valuable insight on particle size distribution.

18

ACCEPTED MANUSCRIPT 6. REFERENCES: [1]

S.U.S. Choi, Enhancing thermal conductivity of fluids with nanoparticle, ASME International Mechanical Engineering Congress & Exposition 231 (1995) 99–105.

[2]

S.U.S. Choi, Nanofluids: From vision to reality through research, Journal of Heat Transfer 131 (2009) 033106 (1-9). O. Mahian, A. Kianifar, S.A. Kalogirou, I. Pop, S. Wongwises, A review of the applications of

T

[3]

[4]

IP

nanofluids in solar energy, International Journal of Heat and Mass Transfer 57 (2013) 582–94. S. Cingarapu, D. Singh, E.V. Timofeeva, M.R. Moravek, Nanofluids with encapsulated tin

CR

nanoparticles for advanced heat transfer and thermal energy storage, International Journal of Energy Research 38 (2014) 51−59. [5]

E.V. Timofeeva, J.P. Katsoudas, C.U. Segre, D. Singh, Rechargeable Nanofluid Electrodes for

[6]

US

High Energy Density Flow Battery, NSTI-Nanotech 8 (2) (2013) 363−366. R. Saidur, K.Y. Leong, H.A. Mohammad, A review on applications and challenges of nanofluids,

[7]

AN

Renewable and Sustainable Energy Reviews 15 (2011) 1646–1668. O. Harizanov, A. Harizanova, Development and investigation of sol-gel solutions for the formation

M

of TiO2 coatings, Solar Energy Materials and Solar Cells 63 (2000) 185-195. H. Chen, Y. Ding, C. Tan, Rheological behaviour of nanofluids, New J. of Physics 367 (9) (2007) .

[9]

C.T. Nguyen, F. Desgranges, G. Roy, N. Galanis, T. Mare, S. Boucher, H.A. Mintsa, Temperature

ED

[8]

and particle-size dependent viscosity data for water based nanofluids – Hysteresis phenomenon,

PT

International Journal of Heat and Fluid Flow 28 (6) (2007) 1492–1506. [10] S.K. Das, N. Putra, P. Thiesen, W. Roetzel, Temperature dependence of thermal conductivity

CE

enhancement for nanofluids, Journal of Heat Transfer 125 (2003) 567–574. [11] C.H. Chon, K.D. Kihm, S.P. Lee, S.U.S. Choi, Empirical correlation finding the role of temperature and particle size for nanofluid (Al2O3) thermal conductivity enhancement, Applied Physics Letters

AC

87 (2005) 153107 (1-3).

[12] D. Zhu, X. Li, N. Wang, X. Wang, J. Gao, H. Li, Dispersion behavior and thermal conductivity characteristics of Al2O3-H2O nanofluids, Current Applied Physics 9 (2009) 131-139. [13] C.H. Li, G.P. Peterson, The effect of particle size on the effective thermal conductivity of Al 2O3– water nanofluids, Journal of Applied Physics 101 (2007) 044312 (1-5). [14] S.M.S. Murshed, K.C. Leong, C. Yang, Investigations of thermal conductivity and viscosity of nanofluids, International Journal of Thermal Sciences 47 (2008) 560–568.

19

ACCEPTED MANUSCRIPT

[15] M. Chandrasekar, S. Suresh, A.C. Bose, Experimental investigations and theoretical determination of thermal conductivity and viscosity of Al2O3-water nanofluid, Experimental Thermal and Fluid Science 34 (2010) 210–216. [16] V.I. Terekhov, S.V. Kalinina, V.V. Lemanov, The mechanism of heat transfer in nanofluids: state of the art (review). Part 1: Synthesis and properties of nanofluids, Thermophysics Aeromechanics

T

17 (1) (2010) 1–14.

IP

[17] R.L. Hamilton, O.K. Crosser, Thermal conductivity of heterogeneous two component systems, Industrial & Engineering Chemistry Fundamentals 1 (3) (1962) 182–191.

CR

[18] E.J. Wasp, J.P. Kenny, R.L. Gandhi, Solid–Liquid Flow Slurry Pipeline Transportation, Trans Tech Publication 172 (1977).

US

[19] H.E. Patel, T. Sundarrajan, T. Pradeep, A. Dasgupta, N. Dasgupta, S.K. Das, A micro-convection model for thermal conductivity of nanofluid, Indian Academy of Sciences 65 (5) (2005) 863–869. [20] G. Cheraghian, L. Hendraningra, A review on applications of nanotechnology in the enhanced oil

AN

recovery part B: effects of nanoparticles on flooding, International Nano Letters 6 (2016) 1–10. [21] C.T. Nguyen, F. Desgranges, N. Galanis, G. Roy, T. Mare, S. Boucher, H.A. Mintsa, Viscosity data

M

for Al2O3–water nanofluid—hysteresis: is heat transfer enhancement using nanofluids reliable?, International Journal of Thermal Science 47 (2008) 103–111.

ED

[22] Y. Raja Sekhara, K.V. Sharma, Study of viscosity and specific heat capacity characteristics of water-based Al2O3 nanofluids at low particle concentrations, Journal of Experimental Nanoscience

PT

10 (2) (2015) 86-102.

[23] X. Wang, X.S. Xu, S.U.S. Choi, Thermal conductivity of nanoparticle–fluid mixture, Journal of

CE

Thermophysics and Heat Transfer 13 (4) (1999) 474–480. [24] X.J. Wang, D.S. Zhu, S. Yang, Investigation of pH and SDBS on enhancement of thermal conductivity in nanofluids, Chemical Physics Letters 470 (2009) 107–111.

AC

[25] M.J. Assael, I.N. Metaxa, J. Arvanitidis, D. Christofilos, C. Lioutas, Thermal conductivity enhancement in aqueous suspensions of carbon multiwalled and double-walled nanotubes in the presence of two different dispersants, Int. Journal of Thermo physics 26 (2005) 647–664. [26] http://www.nano-lab.com/nanosperseaq.html. [27] Y.H. Hung, W.C. Chou, Chitosan for suspension performance and viscosity of MWCNTs, International Journal of Chemical and applications 3 (5) (2012) 343–346. [28] Q. Li, X. Yimin, W. Jian, Measurement of the viscosity of dilute magnetic fluids. International Journal of Thermophysics 27(1) (2006) 103–113.

20

ACCEPTED MANUSCRIPT

[29] T.X. Phuoc, M. Massoudi, R.H. Chen, Viscosity and thermal conductivity of nanofluids containing multi-walled carbon nanotubes stabilized by chitosan, International Journal of Thermal Science 50 (2011) 12–18. [30] J. Wang, J. Zhu, X. Zhang, Y. Chen, Heat transfer and pressure drop of nano-fluids containing carbon nanotubes in laminar flows, Experimental Thermal and Fluid Science 44 (2013) 716–721.

T

[31] P.K. Das, A.K. Mallik, A.K. Santra and R. Ganguly, Synthesis and characterization of TiO 2–water

IP

nanofluids with different surfactants, International Communications in Heat and Mass Transfer, 75 (2016) 341–348.

Journal of Experimental Nanoscience (2017) 1745-8080.

CR

[32] R. Choudhary, D. Khurana, A. Kumar, S. Subudhi, Stability analysis of Al 2O3/water nanofluids

US

[33] J.D. Hem, Study and interpretations of the chemical characteristics of natural water. U.S. Geological Survey Water Supply Paper 2254 (1985) 1-263.

[34] J.P. Holman, Experimental methods for engineers, 5th edition, McGraw-Hill Book Company 1989.

AN

[35] S.M.S. Murshed, K.C. Leong, C. Yang, Thermophysical and electrokinetic properties of nanofluids – A critical review, Applied Thermal Engineering 28 (2008) 2109–2125.

M

[36] S.P. Jang and S.U.S Choi, Role of Brownian motion in the enhanced thermal conductivity of nanofluids, Applied Physics Letters 84 (21) (2004) 4316-4318.

ED

[37] I. Tavman and A. Turgut, An investigation on thermal conductivity and viscosity of water based nanofluids, Microfluidics Based Microsystems: Fundamentals and Applications (2010) 139-162.

PT

[38] R. Sadri, G. Ahmadi, H. Togun, M. Dahari, S.N. Kazi, E. Sadeghinezhad, N. Zubir, An experimental study on thermal conductivity and viscosity of nanofluids containing carbon,

CE

Nanoscale Research Letters 2014, 9:151. [39] J.C. Maxwell, A Treatise on Electricity and Magnetism, Oxford, NY, UK, Oxford Clarendon, 1873. [40] E.V. Timofeeva, A.N. Gavrillov, J.M. McCloskey, Y.V. Tolmachev, Thermal conductivity and

AC

particle agglomeration in alumina nanofluids: experiment and theory, Physical Review E 76 (2007) 061203 (1-16).

[41] W. Yu, S. Choi, The role of interfacial layers in the enhanced thermal conductivity of nanofluids: a renovated Maxwell model, Journal of Nanoparticle Research 5 (2003) 167–171. [42] R. Agarwal, K. Verma, N.K. Agrawal, R. Singh, Sensitivity of thermal conductivity for Al2O3 nanofluids, Experimental Thermal and Fluid Science 80 (2017) 19–26. [43] M. Zhou, G. Xia, J. Li, L. Chai, L. Zhou, Analysis of factors influencing thermal conductivity and viscosity in different kinds of surfactant solutions, Experimental Thermal and Fluid Science 36 (2012) 22–29.

21

ACCEPTED MANUSCRIPT

[44] R.G. Laughlin, The Aqueous Phase Behavior of Surfactants, Academic Press, 1994. [45] H.M. Vale, and T.F. McKenna, Adsorption of sodium dodecyl sulfate and sodium dodecyl benzenesulfonate on poly (vinyl chloride) latexes, Colloids and Surfaces A: Physicochemical and Engineering Aspects 268 (2005) 68–72. [46] J. Buongiorno, Convective transport in nanofluids Journal of Heat Transfer 128 (2006) 240–250.

T

[47] S.E.B. Maiga, S.J. Palm, C.T. Nguyen, G. Roy, N. Galanis, Heat transfer enhancement by using

AC

CE

PT

ED

M

AN

US

CR

IP

nanofluids in forced convection flows, Int. J. Heat Fluid Flow 26 (2005) 530–546.

22

ED

M

AN

US

CR

IP

T

ACCEPTED MANUSCRIPT

AC

CE

PT

Graphical abstract

23

ACCEPTED MANUSCRIPT

Highlights 1. Synthesis of Al2O3-water nanofluids with and without surfactants performed. 2. Characterization of nanofluids performed by DLS and TEM method.

IP

T

3. Effect of Surfactant on thermophysical properties investigated

AC

CE

PT

ED

M

AN

US

CR

4. SDBS surfactant renders better stabilization but reduces thermal conductivity and increases viscosity

24