ionic liquid nanofluids with excellent dispersion stability

Solar Energy Materials and Solar Cells 170 (2017) 219–232

Contents lists available at ScienceDirect

Solar Energy Materials and Solar Cells journal homepage: www.elsevier.com/locate/solmat

Preparation and photo-thermal conversion performance of modified graphene/ionic liquid nanofluids with excellent dispersion stability ⁎

Jian Liu, Chao Xu, LeiLei Chen, Xiaoming Fang , Zhengguo Zhang

MARK

⁎

Key Laboratory of Enhanced Heat Transfer and Energy Conservation, the Ministry of Education, School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China

A R T I C L E I N F O

A B S T R A C T

Keywords: Nanofluid Dispersion stability Ionic liquid Graphene Direct absorption solar collector

Dispersion stability has been long considered as a critical issue for applying nanofluids in various fields, especially for the applications at elevated temperatures. Herein a novel route is explored to improve the dispersion stability of graphene (GE)/ionic liquid (IL) nanofluids for use as working fluids in medium- and high-temperature direct absorption solar collectors (DASCs), which involves modifying GE according to the molecular structure of the IL. Specifically, GE was modified using the reagents and process for synthesizing [HMIM]BF4, followed by dispersing the modified GE (MGE) into [HMIM]BF4. It is verified that the molecular chains similar to [HMIM]BF4 have been grafted on the nanosheets of GE, and the MGE/[HMIM]BF4 nanofluids exhibit much better dispersion stability than the one containing the unmodified GE, even at elevated temperatures. Moreover, the temperature profiles of the nanofluids containing MGE and GE were obtained both from the experimental measurement and the theoretical prediction using a one-dimensional transient heat transfer model. It is shown that the experimental data are in good agreement with the numerical ones for the MGE nanofluids, while a large deviation between them is found for the one containing the unmodified GE. And the MGE nanofluid shows enhanced receiver efficiency as compared to the GE one due to its much improved dispersion stability. Further, the transient model was used to predict the performance of the MGE nanofluid based DASCs under high solar concentrations. And by integrating the MGE concentration and the receiver height into a parameter, namely optical thickness, the optimization of the MGE nanofluid based DASC was carried out varying solar concentration, MGE concentration, nanofluid height and exposure time. It is revealed that the photo-thermal conversion performance of nanofluids greatly depends on its dispersion stability at elevated temperatures, and the MGE/[HMIM]BF4 nanofluids possess excellent dispersion stability and show great potentials for use as the working fluids in DASCs. This work sheds light on effective routes for improving dispersion stability of nanofluids as well as numerical investigations on nanofluid based DASCs.

1. Introduction Nanofluid refers to a new kind of heat transfer fluids (HTFs), prepared by dispersing a little amount of nanomaterials into conventional working media. Since this kind of HTFs was coined by Choi [1] in 1995, nanofluids have been demonstrated to exhibit enhanced thermal conductivity [2–5] and convective heat-transfer coefficient [6–8] as compared to the corresponding base fluids. And these good thermal characteristics make the nanofluids show potentials for use in various fields, such as vehicular and avionics cooling systems in the transportation industry [9,10], hydronic heating and cooling in buildings [11,12], and industrial process heating and cooling systems in petrochemical [13], textile [14], pulp [15], food [16] and other processing plants. However, due to the thermodynamic instability inherent in suspensions,

⁎

Corresponding authors. E-mail addresses: [email protected] (X. Fang), [email protected] (Z. Zhang).

http://dx.doi.org/10.1016/j.solmat.2017.05.062 Received 28 February 2017; Received in revised form 17 May 2017; Accepted 26 May 2017 Available online 13 June 2017 0927-0248/ © 2017 Elsevier B.V. All rights reserved.

nanofluids suffer from the aggregation of the dispersed nanoparticles, which results in the decrease in their thermo-physical characteristics along with the clogging in practical heat transfer equipment [17–19]. Consequently, dispersion instability has long been a critical issue for nanofluids, greatly limiting their practical applications. Apparently, developing nanofluids with good dispersion stability, even at elevated temperatures, is a fundamental and challenging subject. In order to improve the dispersion stability of nanofluids, several approaches have been explored. Addition of surfactant is a general route for increasing the dispersion stability of nanoparticles in aqueous suspensions. Popular surfactants that have been used in literature include sodium dodecylsulfate (SDS) [20–22], sodium dodecyl benzene sulfonate (SDBS) [23], and cetyltrimethylammoniumbromide (CTAB) [24]. With the aid of the surfactants, the hydrophobic surfaces of

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Nomenclature Ρl Cpl Cpn Cpb kl q η T η1 t m

Gs A Qnanofluid Qbasefluid H mp,n mp,b Kn,λ Kb,λ Mf ƒ τ

density of fluid [kg m−3] specific heat of liquid [J g−1 K−1] specific heat of nanofluid [J g−1 K−1] specific heat of basefluid [J g−1 K−1] thermal conductivity of liquid [W m−1 K−1] heat flux [W m−2] relative thermal storage capacity temperature [K] receiver efficiency [%] time [s] weight of nanofluid [kg]

irradiance of the solar simulator [W m−2] surface area of receiver [m2] heat storage capacity of the nanofluid [J] heat storage capacity of the basefluid [J] nanofluid height [m] weight of nanofluid [kg] weight of basefluid [kg] extinction coefficient of nanofluid [cm−1] extinction coefficient of basefluid [cm−1] slope mass fraction of MGE optical thickness

properties and photo-thermal performance, it is urgent to prepare the GE/IL nanofluids with good dispersion stability at elevated temperatures for their practical applications. In this work, to make GE disperse stably into 1-hexyl-3-methylimidazolium tetrafluoroborate ([HMIM]BF4), a novel method for preparing the nanofluids was explored, which is quite different from the previously-reported approaches. First, GE was modified according to the molecular structure of [HMIM]BF4 by grafting the molecular chains similar to [HMIM]BF4 on its surface using the reagents and the process for synthesizing HMIM]BF4. The morphology and structure of MGE were characterized. Second, the HMIM]BF4 based nanofluids were prepared by dispersing the modified GE (MGE) sample into the IL at different loadings. The dispersion stability of the HMIM]BF4 based nanofluids was investigated in the temperature range from room temperature to 180 ℃, and their thermal conductivity, specific heat and transmission spectra were measured. Then, the performance of the DASCs based on the MGE/HMIM]BF4 nanofluids were theoretically and experimentally investigated. Specifically, a transient one-dimensional numerical model was used to predict the temperature distribution of the MGE nanofluids under an irradiation of 2300 W m−2 based on the extinction coefficients of the nanofluids. And a cylindrical simulative receiver was set up to experimentally measure the temperature profiles of the DASCs under the irradiation with the same incident light intensity. The obtained experimental results were compared with the aforementioned numerical ones to validate the model. Further, the model was employed to predict the temperature profiles of the MGE/ [HMIM]BF4 based DASCs with various geometrical parameters under different operating conditions. And thus the effects of the parameters including the MGE concentration, the receiver height and the solar concentration on the thermal storage capacity of the DASCs were systematically investigated. What’s more, the MGE concentration of the nanofluid and the receiver height of the DASC were integrated into one parameter, namely optical thickness, and the optical thickness was optimized under the fixed solar concentration. Finally, the exposure time was optimized based on the optical thickness, and the systematical investigation along with optimization on the MGE/HMIM]BF4 based DASCs were conducted under the concentrated solar incident radiation. It is found that the MGE/HMIM]BF4 nanofluids possess good dispersion stability at elevated temperatures due to the good compatibility of MGE with HMIM]BF4, and the optimized nanofluid based DASC exhibits large thermal storage capacity, making it show great potentials for practical applications. This work sheds light on the approaches for increasing dispersion stability of nanofluids and makes an improvement on numerical investigations on nanofluids based DASCs.

nanoparticles could be modified to become hydrophilic or vice versa for non-aqueous liquids. However, the bonding between surfactants and nanoparticles would be damaged at the temperatures above than 60 ℃ [25]. Consequently, the nanofluids lose their stability, leading to the sedimentation of nanoparticles at the elevated temperatures. Besides, controlling pH is another method to improve the dispersion stability of nanofluids, since the stability of nanofluids directly links to their electrokinetic properties. Xie et al. [26] reported that a carbon nanotube suspension gained a good dispersion stability in water through a simple acid treatment on the carbon nanotubes and found that the good stability was attributed to a hydrophobic-to-hydrophilic conversion on the surfaces of the carbon nanotubes. Lee et al. [27] worked on the nanofluids containing Al2O3 at different pH values. Their experimental results indicated that, when the nanofluids had a pH of 1.7, the agglomerated particle size was reduced by 18%; and when the nanofluids had a pH of 7.66, the agglomeration size was increased by 51%. However, although the addition of surfactants and the adjustment of pH play an effective role in improving the dispersion stability of nanofluids, the damage of the bonding between surfactants and nanoparticles at the temperatures above than 60 ℃ and the decrease in zeta potential with an increase in temperature make these two routes inapplicable to the nanofluids for medium- and high-temperature applications [28]. And it has been reported that high temperatures intensified the agglomeration of nanoparticles, resulting in the remarkable decrease in thermal conductivity of the nanofluids [29]. Therefore, new approaches need to be explored for preparing nanofluids with good dispersion stability at elevated temperatures. Ionic liquids (ILs) are a group of molten salts with a wide liquid temperature range, high density and heat capacity, good thermal and chemical stability and low vapor pressure [30,31]. These favorable thermophysical properties make ILs show great promise to be used as HTFs for medium- and high temperature applications. And it has been reported that the ILs based nanofluids, prepared by dispersing nanoparticles into ILs, show enhanced thermal conductivity, making them more promising for use as HTFs [32,33]. In our previous work, graphene (GE), a 2D carbon nanomaterial with extraordinarily high thermal conductivity [34], was dispersed into ILs to prepare the GE/IL nanofluids, and their thermophysical properties were measured in wide temperature range [35]; Further, the optical absorption property and photo-thermal performance of the GE/IL nanofluids as well as the performance of the GE/IL nanofluid based direct absorption solar collectors (DASCs) were experimentally and numerically investigated [36,37]; The obtained results show that graphene is an excellent nanoadditive for preparing nanofluids, and the GE/IL nanofluids are a kind of promising working fluid for medium- and high temperature DASCs; However, it has been also found that the agglomeration of GE increased with temperature during the GE/IL nanofluids were irradiated, leading to the decrease in the photo-thermal performance of the GE/IL nanofluid. Therefore, in view of the strong influence of dispersion stability of the IL-based nanofluids on their thermophysical

2. Experimental section 2.1. Materials and reagent Carboxyl graphene (GE-COOH) was purchased from Nanjing 220

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

and binding energies were corrected for specimen charging by referring to the C1s at 284.8 eV. The thermal and dispersion stability of the MGE/[HMIM]BF4 and GE-COOH/[HMIM]BF4 nanofluids were evaluated with the aid of a thermotank with temperatures ranging from room temperature to 180 ℃. A thermal infrared imager was used to monitor the temperatures of the MGE/[HMIM]BF4 and GE-COOH/[HMIM]BF4 nanofluids. The zeta potentials of MGE/[HMIM]BF4 nanofluidS before and after heating were measured by Zetasizer Nano (Nano-ZS90, Malvern Instruments, England). Room temperature transmission spectra of the MGE/[HMIM]BF4 nanofluids were recorded in the wavelength range from 200 to 2200 nm with a double-beam UV–vis–NIR spectrophotometer (PerkinElmer Lambda 950) using a cuvette with an optical path length of 10 mm. The effect of multiple reflections at the interfaces between air, glass and fluid is assumed to be negligible. Wavelength accuracy: ± 0.02 nm, stray light ≤ 0.00007% T. The thermal conductivity of the MGE/[HMIM]BF4 nanofluids was measured using the transient hot plane method on a thermal constant analyzer (TPS2500, Hot Disk, Sweden). During the measurement, the experimental temperatures ranged from 25 to 145 °C, with the aid of a cyclic oil bath. The specific heat of the MGE-dispersed nanofluids were measured using the differential scanning calorimeter (DSC) at temperatures ranging from 30 to 160 °C.

XFNANO Materials Tech Co, Ltd. (purity, ~99.8%; thickness, 0.8–1.2 nm). [HMIM]BF4 (CAS: 244193-50-8) was purchased from Lanzhou Institute of Chemical Physics. 1-(3-Aminopropyl) imidazole (98%), 1-bromohexane (99%), NH4BF4 (97%), tetrahyrofuran (THF, anhydrous, 99.8%) were purchased from Alfa Aesar (Ward Hill, MA, USA), and Thionyl chloride (SOCl2, AR, 99%) was from Aladding Reagent (Shanghai) Co., Ltd. 2.2. Synthesis The modification of GE-COOH was performed according to the previously documented protocols [38], with several small modifications. As illustrated in Fig. 1, the procedure could be divided into four steps. Step 1, 500 mg of GE-COOH was suspended in 150 mL of SOCl2 under nitrogen atmosphere with stirring for 24 h, and then the solid was collected by centrifugation, washed with anhydrous THF, and subsequently dried under vacuum at room temperature for 2 h to obtain graphene-COCl (F-1). Step 2, a mixture of graphene-COCl and 1-(3aminopropyl)imidazole (300 mL) was stirred at 120 ℃ for 24 h under nitrogen atmosphere. And then the solid was separated by centrifugation, washed and dried to obtain F-2. Step 3, the mixture of F-2 (450 mg) and 1-bromobutane (20 mL) was stirred at 80 °C for 24 h under nitrogen atmosphere. And then the solid was separated by centrifugation and thoroughly washed with anhydrous THF for three times to remove the excess 1-bromobutane. The obtained solid was dried under vacuum at room temperature for 24 h to generate F-3. Step 4, excess amount (500 mg) of solid NH4BF4 was added into a solution of F3 (400 mg) dispersed in water (20 mg) under stirring at room temperature for 24 h until the black solid was precipitated. And then the solid was separated by centrifugation and washed with water to remove the water-soluble species such as NH4BF4 residue and NH4Br, followed by washing with water for three times. The resulting solid was dried overnight in a vacuum oven to obtain modified graphene (MGE). The MGE/[HMIM]BF4 nanofluids were prepared by dispersing different mass fractions (0.01%, 0.03% and 0.05%) of MGE into [HIMIM] BF4 via mildly stirring (without ultrasonication) at room temperature. For comparison, 0.01% of the GE-COOH/[HMIM]BF4 nanofluid was also prepared.

2.4. Experimentally evaluating photo-thermal conversion performance To experimentally demonstrate the concept of the nanofluid based DASCs, an experimental system was set up for testing temperatures. Fig. 2 shows the main components of the system, including a solar simulator, a beam-down mirror, a receiver, a thermocouple array and a data acquisition system. A solar simulator (700 W, SOLAREDGE700, Perfectlight Inc.) was used as the radiative source, which can provide a solar spectrum match with the AM1.5 one well and meet the ASTM Class A standards [39]. The average radiative heat flux incident into the upper surface of the MGE/[HMIM]BF4 nanofluids was measured to be 2300 W m−2 using an irradiatometer (ST-80C Peifbnu Inc.). A custom thin-walled cylinder insulated with a kind of low-density foam was used to hold the nanofluids, and its top was sealed with a high-purity quartz window (Perfectlight Inc.). During the experiments, the radiation from the solar simulator was beamed down to the receiver through the quartz window, and the temperatures of the receiver containing the nanofluids under the irradiation for different times were monitored using eleven type-K thermocouples positioned along the centerline of the receiver from the bottom to the top at intervals of 1 cm. A data acquisition system (Agilent 34970A) was used to record the temperature variations at an interval of two seconds. The thermocouples were calibrated using

2.3. Characterization and measurements Transmission electron microscopy (TEM) observations of GE-COOH and MGE were conducted on a Hitachi Model H-7650 (Japan) electron microscope. Fourier transformation infrared (FT-IR) spectra were recorded on a spectrometer (Bruker, Germany), using the KBr disk method. X-ray photoelectron spectroscopy (XPS) analysis was performed on an Axis Ultra DLD (Kratos, England) multifunctional XPS,

Fig. 1. Diagram for preparing MGE.

221

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Table 1 Thermophysical properties of [HMIM]BF4 and nanofluids. Thermophysical properties

Correlation with temperature (K)

Thermal conductivity (W m−1 K−1)

[HMIM]BF4：y = 0.1625 + 2.13 × 10−4T 0.01% MGE: y = 0.1704 + 2.31 × 10−4T 0.03% MGE: y = 0.1740 + 2.31 × 10−4T 0.05% MGE: y = 0.1744 + 2.25 × 10−4T 0.01% GE-COOH: 0.1694 + 2.45 × 10−4T [HMIM]BF4：y = 1.6416 + 8.582 × 10−4T + 3.408 × 10−4T2-3.923 × 10−6T3 + 1.174 × 10−8T4 0.01% MGE: y = 2.1311-0.017T + 3.988 × 10−4T2-2.880 × 10−6T3 + 6.340 × 10−9T4 0.03% MGE: y = 1.8023-0.002T + 7.149 × 10−5T2-7.497 × 10−8T3 + 2.340 × 10−9T4 0.05% MGE: y = 1.1812 + 0.026T-4.211 × 10−5T2 + 3.474 × 10−6T3-1.006 × 10−8T4 0.01% GE-COOH: y = 1.8895-0.012T + 3.256 × 10−4T2 −2.329 × 10−6T3 + 5.205 × 10−9T4

Specific heat (J g−1 K−1)

Fig. 2. Experimental setup for evaluation the photo-thermal conversion performance of nanofluids.

attenuation of radiative heat flux along y-direction. Last, the radiative heat flux function was incorporated into Eq. (1) to determine the temperature profile at a subsequent time (t) with boundary conditions. This new temperature profile was subsequently reintroduced into the aforementioned step, followed by circularly repeating with t. In this section, the thermal conductivity k(T) and specific heat Cp(T) were measured, as shown in Table 1. The temperature distributions along the y-direction for the DASCs at varying times was obtained by numerically solving the energy equation.

a temperature-controlled bath with 0.01 K resolution (C50p, HAAKE). 3. Theoretical model 3.1. Energy equation The energy equation and boundary condition for a nanofluid-based DASC, accounting the thermophysical and radiative properties of the working fluid for the volumetric heat generation and release, can be deduced as follows, according to our previous work [37]:

ρl Cpl

4. Results and discussion

∂T ∂ 2T = kl 2 + q (y ) ∂t ∂y

(1)

4.1. Morphology and structure of MGE

Where ρl is the density, Cpl is the specific heat, and kl is the thermal conductivity of the nanofluid, t is the time. The solar receiver is designed by an important metrics for solar thermal applications: heat storage capacity. The receiver efficiency of a nanofluid-based DASC can be calculated according to the following formula [39].

η1 =

m ∫ Cp (T ) dT Gs At

× 100%

The molecular chains similar to [HMIM]BF4 were grafted on the nanosheets of GE using the reagents and process for synthesizing [HMIM]BF4, as illustrated in Fig. 1, with the purpose of improving the compatibility of GE with [HMIM]BF4. In order to identify the chemical structure of the obtained MGE, the characterizations by FT-IR and XPS have been conducted. Fig. 3a shows the FT-IR spectra of GE-COOH and MGE. The peak at 1402 cm−1 is the C-O stretching vibration peak, and the one at 1036 cm−1 is the O-H out of plane bending vibration peak. The peak at 1635 cm−1 corresponding to C = O stretching vibrations of the carboxyl group is observed in the spectrum of GE-COOH [40], which shifts to 1643 cm−1 in the spectrum of MGE. The reason for this shift could be attributed to C = O stretching vibrations and C = C asymmetric stretching vibrations, originating from the modifications. Moreover, in the spectrum of MGE, there are the peaks at 2917 and 2848 cm−1 belonging to symmetric and asymmetric stretching vibrations of C-H [41], the one at 1464 cm−1 corresponding to the C-N stretching vibration peak and the characteristic peak at 719 cm−1 corresponding to long chain methylene plane bending vibration, all of which demonstrate that GE have been successfully modified. The XPS full-scan spectrum of MGE and the high-resolution spectra of corresponding elements are shown in Fig. 3b. The peak of N1s (402.1 eV) suggests the attachment of the imidazolium cation-based ILs. And B1s (193.8 eV, inserted in Fig. 3b) and F1s (684.8 eV), with a mole fraction ratio of B1s to F1s ~1:4, verify the existence of the anion BF4- [42]. On the basis of the above FT-IR and XPS spectra, it could be inferred that the surfaces of MGE have been successfully grafted with the molecular chains similar to [HMIM]BF4, which makes the obtained MGE has good compatibility with [HMIM]BF4. In addition, the TEM images of GECOOH and MGE are shown in Fig. 4. It can be seen that the modifications did not deteriorate the structural integrity of GE. Compared with GE-COOH, the nanosheets of MGE exhibit a decrease in transparency, because of the grafting of the molecular chains.

(2)

Where η1 is the receiver efficiency, m represents the mass of the nanofluid, Cp(T) is the specific heat of the nanofluid, T is the temperature of the nanofluid, Gs is the irradiance of the solar simulator, which was 2300 ± 20 W m−2 in the current work, A is the surface area of the receiver, and t is the time. Moreover, another metrics, namely relative heat storage capacity, is employed to evaluate the enhanced performance of a nanofluid, which is the ratio of the collected thermal energy of the nanofluid to that of its basefluid.

η=

Qnanofluid − Qbasefluid Qbasefluid

× 100%

T

=

T

mp, n ∫T =n290 Cp, n (T ) dT − mp, b ∫T =b290 Cp, b (T ) dT T

mp, b ∫T =b290 Cp, b (T ) dT

× 100% (3)

Where Qnanofluid and Qbasefluid, Cp,n and Cp,b, and mp,n and mp,b are the heat storage capacity, specific heat, and mass of the nanofluid and its basefluid, respectively. 3.2. Solution methodology The solution methodology of the energy equation was separated to two steps: solar radiative performance and heat transfer performance. First, the extinction coefficients of the MGE/[HMIM]BF4 nanofluids were calculated. Second, the heat source item q(y) were explicitly solved in MATLAB through numerical integration, which resulted in the 222

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

kept at 180 ℃ for one month. At the same loading of 0.01%, a heavy sedimentation is observed for the GE-COOH/[HMIM]BF4 nanofluid, while no sedimentations are found for the MGE/[HMIM]BF4 nanofluid. And the other nanofluids containing 0.03% and 0.05%MGE also exhibits good dispersion stability. It is indicated that the modification on GE-COOH by grafting the molecular chains similar to [HMIM]BF4 greatly improve the dispersion stability of the nanofluids, due to the excellent compatibility of MGE with [HMIM]BF4. Furthermore, the dispersion stability of the nanofluids was elucidated by measuring their zeta potentials before and after being kept at 180 ℃ for one month, and the obtained results are displayed in Fig. 5c. Note that a well-dispersed suspension can be acquired with a high zeta potential to get strong repulsive forces for the suspended particles, and the dividing line between stable and unstable nanoparticle suspensions is generally taken at either + 30 or −30 mV. So nanofluids with zeta potentials of more positive than + 30 mV or more negative than −30 mV are normally considered stable, which is one of the most common methods for evaluating the dispersion stability of nanofluids [43]. From Fig. 5c it can be seen that, the zeta potentials of all the MGE/[HMIM]BF4 nanofluids are positive, and their ζ values are greater than 30 mV before heating. After being kept at 180 ℃ for one month, their zeta potential just slightly decreases to 29 mV, not far away from 30 mV. It is indicated that the MGE/[HMIM]BF4 nanofluids can keep stable even after being kept at 180 ℃ for one month. While for the GE-COOH/[HMIM] BF4 nanofluid, its zeta potential is as low as 13.8 mV before heating, and remarkably decreases to 2.6 mV after being kept at 180 ℃ for one month. It is suggested that the MGE/[HMIM]BF4 nanofluids has much better stability duration than the one containing the unmodified GE at the high temperature owing to the good compatibility of MGE with [HMIM]BF4. Moreover, the nanofluids experienced 500 heating-cooling cycles, and their infrared thermography and photographs were taken after every 100 cycles. As shown in Fig. 5c, the temperatures of the nanofluids are close to 180 ℃. It can be seen from Fig. 5d that, the GECOOH/[HMIM]BF4 nanofluid has heavy sedimentation after 100 cycles, while no sedimentations are found for the MGE/[HMIM]BF4 nanofluid at the same loading even after experiencing 500 heating-cooling cycles. It is indicated that the GE-COOH/[HMIM]BF4 nanofluid is unstable even for the as-prepared one. And the other nanofluids containing 0.03% and 0.05%MGE also exhibits good dispersion stability after they have experienced 500 heating-cooling cycles. All these results suggest the excellent dispersion stability of the MEG nanofluids, even at elevated temperatures, making them feasible to be used as working fluids in DASCs. And it is revealed that the route developed in the current work is very effective for improving the dispersion stability of the GE/ [HMIM]BF4 nanofluids, which involves grafting the molecular chains similar to [HMIM]BF4 on the nanosheets of GE.

Fig. 3. (a) FT-IR spectra of GE-COOH and (b) MGE and XPS spectra of GE-COOH and MGE.

4.2. Dispersion stability of MGE/[HMIM]BF4 nanofluids The dispersion stability of the nanofluids were investigated by keeping them at 180 ℃ for one month along with making them experience 500 heating-cooling cycles (180 ℃-30 ℃-180 ℃). When the nanofluids were placed on a heating stage with a temperature of 200 ℃0 they reached a temperature of around 180 ℃, as shown in Fig. 5a. Fig. 5b shows the photographs of the nanofluids before and after being

Fig. 4. TEM images of GE-COOH (a) and MGE (b).

223

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Fig. 5. Infrared thermography (a), photographs (b) and zeta potentials (c) of nanofluids before and after being kept at 180 ℃ for one month, together with infrared thermography (c) and photographs (d) of nanofluids experiencing different heating-cooling cycles (Ⅰ: 0.01%-MGE, Ⅱ: 0.01%-GE-COOH Ⅲ: 0.03%-MGE, and Ⅳ: 0.05%-MGE).

224

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Fig. 6. Thermal conductivity (a) and specific heat (b) of [HMIM]BF4 and its nanofluids.

Fig. 7. Transmittance spectra (a) and extinction coefficient (b) of [HMIM]BF4 and its nanofluids before and after being heated.

4.3. Thermophysical properties of MGE/[HMIM]BF4 nanofluids

nanofluid containing 0.03% MGE reaches a reduction in specific heat by approximately 5% at tested temperatures, which is increased to about 7% for the one containing 0.05% MGE. The reduction in specific heat is due to the fact that the specific heat of MGE is lower than that of [HMIM]BF4. It has been reported that the specific heat of the [C4mmim][NTf2]-based nanofluids containing 0.5% carbon black (4 nm) diminishes by 12.4% at 80 ℃ and 24% at 200 ℃ [44]. Interestingly, the peaks showing the obvious increase in specific heat appear at the temperatures between 120 and 160℃for the MGE/[HMIM]BF4 nanofluids, higher than around 90℃for [HMIM]BF4. The reason for this may be attributed to the adsorption of the nanosheets for the water in the IL, leading to its evaporation taking place at higher temperatures. As shown in Fig. 9b, the specific heat of the GE-COOH/[HMIM]BF4 nanofluid increases from 1.75 to 2.15 J g−1 K−1 as the temperature is increased from 30 to 160 ℃, which is similar to that of the 0.03% MGE/ [HMIM]BF4 nanofluid at the same temperatures. It is suggested that the surface modification has no effect on the specific heat of the nanosheets.

The thermal conductivity of [HMIM]BF4 and the MGE/[HMIM]BF4 nanofluids with different mass fractions of MGE is shown in Fig. 6a, together with that of the 0.01% GE-COOH/[HMIM]BF4 one. The thermal conductivity of [HMIM]BF4 increases from 0.167 to 0.193 W m−1 K−1 at the temperatures ranging from 20 to 145 ℃. When dispersing 0.01% of MGE into the ionic liquid, the thermal conductivity of the obtained nanofluid increases from 0.177 to 0.205 W m−1 K−1 in the same temperature range, indicating an enhancement by 5% in thermal conductivity. With further increasing mass fraction of MGE, the MGE/[HMIM]BF4 nanofluids exhibit a slight increase in thermal conductivity at the same temperatures. And an enhancement by 10% in thermal conductivity is achieved by the 0.05% MGE/[HIM]BF4 nanofluid. For the 0.01% GE-COOH/[HMIM]BF4 nanofluid, its thermal conductivity enhancement reaches 6.2%, slightly higher than 5% of the 0.01% MGE/[HMIM]BF4 one. The reason for this may be attributed to the grafting of the molecular chain on the surfaces of the nanosheets, which leads to a decrease in thermal conductivity of the nanosheets. Fig. 6b shows the specific heat of [HMIM]BF4 and the MGE/[HMIM] BF4 nanofluids with different mass fractions of MGE, together with that of the 0.01% GE-COOH/[HMIM]BF4 one. For [HMIM]BF4, its specific heat first increases with temperatures rising from room temperature to around 90 ℃ and then drops with the further increase in temperature to 160 ℃. The appearance of the peak centered at 90 ℃ may be attributed to the evaporation of the water in the IL, resulting in the remarkable increase in specific heat. The specific heat of the MGE/[HMIM]BF4 nanofluids decreases with the increase in mass fraction of MGE, which is slightly lower than that of [HMIM]BF4 in the temperature range from room temperature to 120 ℃. Specifically, the MGE/[HMIM]BF4

4.4. Radiative properties of MGE/[HMIM]BF4 nanofluids Fig. 7 respectively show the transmittance spectrum and extinction coefficient of the MGE/[HMIM]BF4 nanofluids. From the spectra in Fig. 7a, the favorable effect of MGE as direct light absorbers is immediately evident: for a light path of 10 mm in the MGE/[HMIM]BF4, even for the nanoparticle concentration level of 0.01%, the average transmittance in the 300–850 nm range (the region of the higher transparency of the [HMIM]BF4) decrease of more than 40% with 225

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Fig. 8. Absorption power fraction of [HMIM]BF4 and its nanofluids.

respect to the pure [HMIM]BF4, whereas the MGE/[HMIM]BF4 transmittance practically falls to zero on the whole wavelength range of interest with the sample(MGE concentration of 0.05%). This means at higher graphene concentrations, about 99% sunlight absorption is obtained within the thin layer of the liquid flowing in the typical solar thermal receiver. For comparison, the transmittance spectrum of 0.01% GE-COOH/[HMIM]BF4 nanofluid was conducted. As shown in Fig. 7a, the average transmittance in the 300–850 nm range decreases to 20–30%, meaning that the GE-COOH/[HMIM]BF4 nanofluid has a better optical absorption capacity the MGE/[HMIM]BF4 nanofluid with the same concentration. To better characterize of optical properties of the samples, extinction coefficient Ke was calculated from the spectral transmittance T(λ), according to the Beer–Lambert law. The spectral extinction coefficient, that includes the spectral absorption and scattering coefficients of the samples, is shown in Fig. 7b. It was also found that the presence of 0.03% MGE increases the extinction coefficient of [HMIM]BF4 by about 3 cm−1. This result is important to improve the compactness of solar concentrator devices, as MGE/[HMIM]BF4 allow the use of higher receiver, also with a subsequent lower concentration of the heat exchange liquid flowing through the whole plant and reduced clogging of nanoparticles [45]. For a deeper understanding of the promising potential of MGE/ [HMIM]BF4 as direct sunlight absorbers, an important information for the optimal system dimensioning is the spatial distribution F of the absorbed energy within the absorber. As shown in Fig. 8, it is noticed that in all cases the absorbed energy distribution of MGE/[HMIM]BF4 samples and GE-COOH/[HMIM]BF4 nanofluid obtained are higher than that of pure [HMIM]BF4. Moreover, for the higher concentration MGE/ [HMIM]BF4 the energy is mainly concentrated in thin layer. The joint effects of the high overall absorption level and the energy localization in the receiver result in a strong distribution gradient for the fluids with high MGE concentration level. On the contrary, as the nanoparticle concentration decreases, the absorbed energy distribution penetrates more deeply in the sample, producing a more uniform sharing in the receiver. This information is very important for the overall system optimization of solar collector designer. In fact, suitable MGE concentration and fluid height could be altered to maintain a constant absorption fraction of sunlight. As aforementioned, the GE-COOH/[HMIM]BF4 nanofluid has better optical absorptive capacity than MGE nanofluid with the same mass fraction. To absorb the same incident solar light, the fluid height of the GE-COOH/[HMIM]BF4 nanofluid is smaller than MGE/[HMIM]BF4 nanofluid. Hence, the two samples (MGE/[HMIM] BF4 and GE-COOH/[HMIM]BF4 nanofluid with the same mass fraction of 0.01% with the fluid height 5.9 cm and 3.2 cm, respectively, as shown in Fig. 8) with the same absorption fraction of 0.99 were chosen to experimentally and numerically investigate their behaviors in DASCs.

Fig. 9. Comparison of experimental and numerical temperature profile of nanofluid with 0.01% MGE (a) or 0.01% GE-COOH (b).

4.5. Numerical and experimental investigations on nanofluids based DASCs The two MGE/[HMIM]BF4 and GE-COOH/[HMIM]BF4 nanofluids at the same loading of 0.01% were contained within a 10-centimeter cylindrical receiver, and their temperature profiles were obtained both from the numerical model (as described in Section 3) and the experimental investigation. As for the numerical simulation, the initial temperatures of the two nanofluids were set to ambient temperature (290 K), and their absorptive fractions (calculated via integration of extinction coefficient over the wavelength range, shown in Fig. 8) were kept at 0.99 by altering the MGE or GE-COOH concentration along with the nanofluid height. To keep the absorptive fraction at 0.99, two parameters (mass fraction: 0.01%, corresponding to fluid height, MGE/[HMIM]BF4: 5.9 cm and GE-COOH/[HMIM]BF4: 3.2 cm) were obtained by numerically solving the energy equation under the incident solar heat flux 2300 W m−2. As shown in Supplementary Video S1, 2, the temperature of MGE/[HMIM]BF4 increases from room temperature to 380 K as the exposure time ranges from 100 to 12000 s. The temperature at the top of the receiver is higher than the bulk and slightly increases with the exposure time. While for GE-COOH/[HMIM]BF4, its temperature increases from room temperature to 440 K under the same exposure time. The temperature at the top of the receiver is little lower than the bulk. In both cases, the incident solar heat flux is attenuated along the ydirection, heat generation decreases sharply at the bottom of the receiver, resulting in tiny temperature enhancement of MGE/[HMIM]BF4 nanofluids. Supplementary material related to this article can be found online at 226

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

http://dx.doi.org/%2010.1016/j.solmat.2017.05.062. Since one goal of this work is to experimentally demonstrate the concept of two nanofluids volumetric receiver rather than replicating the environmental conditions and solar concentration levels, the numerical results were compared to the experimental values to examine the mechanism of extinction of incident light by nanofluid. Fig. 9 shows the experimental results and their comparisons with the numerical ones. The numerically predicted temperature profiles are shown for the same time increments as the experimental data. To predict these profiles, the numerical model was solved using the solar simulator spectrum as the input radiation and the experimental thermophysical properties and extinction coefficients in Sections 4.3 and 4.4. Convective heat loss and thermal re-radiative loss were calculated analytically and added in this section. Also, because of the transient nature of the experiment and the comparable vertical (H = 10 cm) and horizontal lengths (10 cm inner diameter) of the receiver, the numerical model was modified for comparison with the experimental results. By design, the walls and optics directly in contact with the nanofluids are thin enough (2 mm) to be approximated as being at the same temperature as the fluid. Based on a measured weight and heat capacity, the total heat capacitance of the walls is 556 J K−1, while that of the MGE/[HMIM]BF4 are 1480 J K−1 and GE-COOH/[HMIM]BF4 are 560 J K−1. Thus, a significant portion of the thermal energy is stored in the sidewalls. Using a lumped capacitance model, it is estimated that 72.7% and 50.2% of the incident energy results in heating the MGE/ [HMIM]BF4 nanofluid, while the rest of the energy is stored in the sidewalls. By incorporating this scaling factor, the numerical and experimental results show good agreement for MGE/[HMIM]BF4, as shown in Fig. 9a. The temperature discrepancy could be attributed to the relatively large difference in temperature between the top and the bottom of the receiver, which induces a re-distribution of heat via the conductive side-walls, this conjugate heat transfer is not captured in the model. In total, the numerical and experiment results of the MGE/ [HMIM]BF4 nanofluid have a deviation within 10% as the temperature ranges from room temperature to 380 K. The good agreement between the model and experiment firstly demonstrates that the incident solar heat flux is attenuated along the y-direction, and the good dispersion stability of MGE/[HMIM]BF4 nanofluid has an advantage of gathering incident solar light at high temperature. In comparison, the numerical and experimental temperature of the GE-COOH/[HMIM]BF4 nanofluid show good agreement at room temperature, and have a discrepancy with an increase in temperature, as shown in Fig. 9b. As the temperature increases to 440 K, the deviation between the model and experiment is higher than 33.3%. The large discrepancy for GE-COOH/ [HMIM]BF4 is due to its poor dispersion stability at high temperatures, as mentioned in Section 4.2. The GE-COOH nanosheets aggregate together with the increase in temperature, and the photo-thermal performance of GE-COOH/[HMIM]BF4 decreases sharply. As a result, the experimental data are much lower than the numerical results. To provide more evidences to elucidate the difference between the two MGE/[HMIM]BF4 and GE-COOH/[HMIM]BF4 nanofluids at the same loading of 0.01%, their transmittance spectra were measured after they had been experienced the photo-thermal experiment and compared with those of the samples before the experiment. As shown in Fig. 10, the transmittance of the MGE/[HMIM]BF4 nanofluid keeps constant before and after the photo-thermal experiment. While the transmission of the 0.01% GE-COOH/[HMIM]BF4 nanofluid increases from 20–30% to 80% in the visible range, implying that the optical absorption property of the nanofluid decrease dramatically with the increased temperature. The phenomenon verifies the large deviation of the GE-COOH/[HMIM]BF4 nanofluid. It is indicated that the dispersion stability has a great effect on the photo-thermal performance of nanofluids. Increasing the dispersion stability takes an advantage of acquiring higher receiver temperature. Furthermore, in order to quantitatively evaluate the photo-thermal performance of the MGE/[HMIM] BF4 and GE-COOH/[HMIM]BF4 nanofluids at the same loading of

Fig. 10. Transmittance spectra of nanofluid with 0.01% MGE or 0.01% GE-COOH before and after being heated.

0.01%, their receiver efficiency was calculated according to formula 2. The receiver efficiency of the MGE/[HMIM]BF4 nanofluid decreases slightly from 89% to 85% as the temperature increases from room temperature to 370 K, as shown in Fig. 11. Significantly, the receiver efficiency of the MGE/[HMIM]BF4 nanofluid could maintain at high levels with the increased temperature. While the receiver efficiency of the GE-COOH/[HMIM]BF4 nanofluid decreases drastically from 89% to 52% as the temperature increases from room temperature to 370 K. The reason for this can be attributed to the agglomeration of the GE-COOH at elevated temperatures, which decreases the optical absorption capacity of the nanofluid and thus leads to the rapid reduction in photothermal performance. These results reveal that the high-temperature dispersion stability of the nanofluids is crucial in DASCs, and the surface modification of GE-COOH presented in this work is in favor of preparing the stable [HMIM]BF4 based nanofluid with high photothermal performance.

4.6. Predicted performance of concentrated DASCs based on MGE nanofluids In the Section 4.5, the numerical and experimental results of the MGE/[HMIM]BF4 nanofluid demonstrates that the incident solar heat flux is attenuated along the y-direction to heat the HTF in the receiver, suggesting that the numerical model can be used to investigate volumetric receiver efficiency under concentrated solar intensity. In fact, the heat flux of solar simulator was low to replicate the high solar incident flux ambition like parabolic trough collector [46]. Hence, the effect of solar concentration, on the temperature distributions of

Fig. 11. Receiver efficiency of nanofluid with 0.01% MGE or 0.01% GE-COOH.

227

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

Fig. 12. Predicted temperature curves (a), average temperature (b) and receiver efficiency (c) of nanofluids with 0.01% MGE, nanofluid height H = 6 cm and varying solar concentration.

Fig. 13. Predicted temperature curves (a), average temperature (b) and receiver efficiency (c) of nanofluids with solar concentration C = 20, nanofluid height H = 6 cm and varying MGE concentration.

receiver was numerically investigated rather than experimental test under concentrated incident solar intensity. The temperature profiles over time under concentrated incident solar heat flux(C = 20) is shown in Supplementary Video S3. At first, the temperature of MGE/[HMIM] BF4 increases faster at the top of receiver than the bottom, because the solar light attenuate along y-direction to generate heat. The temperature is increased with exposure time, then, the temperature gradient between the DASC and ambient sharply increase that the convective heat loss from the top surface was enhanced. As the temperature range

to high temperature region, the MGE/[HMIM]BF4 begins to re-radiate and the temperature profiles begin to smoothen. Thus, the temperature at the top of the receiver is initially increased greatest but quickly asymptotes to state value, allowing the temperature inside the receiver to surpass the surface. After exposure to the solar light, an inverted ‘S’shape profile is developed inside the receiver. Figs. 12–14 illustrates the temperature profiles over time for the case of varying solar concentration, MGE concentration or receiver height after exposure to the solar light. Besides, the relative heat storage capacity under different conditions are shown in this section. 228

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

concentration range from C = 5 to 20 when undergoing the same exposure time of 12000 s, implying higher solar concentration has an advantage to obtain higher temperature which improve the Carnot efficiency [47] in the solar generating electricity system. As it is shown in Fig. 12b, the maximum mean temperature of receiver range from 400 to 675 K as the solar concentration increasing from C = 5 to 20, which allow the receiver to maintain a high efficiency over a wider range of temperatures and delay the sharp decrease in efficiency. For example, the relative thermal storage capacity of MGE/[HMIM]BF4 increases from 70% to 130% as solar concentration range from C = 5 to 20 after 12000 s of exposure, as shown in Fig. 12c. Increasing the solar concentration help improve the temperature of MGE/[HMIM]BF4, indicating the MGE/[HMIM]BF4 could be used to efficiently absorb and storage the solar incident radiation along its liquid temperature range under high level solar concentration. It also suggests that increasing the solar concentration ratio has an advantage in extending the high relative thermal storage capacity ranges over the work temperature of MGE/[HMIM]BF4. The effect of solar concentration on receiver efficiency is usually reported for surface receiver [48], the results also show that increasing solar concentration help improve the mean fluid temperature of heat transfer fluid in such isothermal systems. 4.6.2. Effect of MGE concentration As shown in Fig. 13a, under the same solar concentration C = 20 and receiver height H = 6 cm, the temperature profiles of MGE/ [HMIM]BF4 show unexpected trend when varying the MGE concentration. First, the maximum temperature of MGE/[HMIM]BF4 increases from 400 to 725 K as the MGE concentration range from 0.01% to 0.05% after 6000 s of exposure to sunlight, and the maximum temperature located below the top surface of receiver. As it is shown in Fig. 13b, the maximum mean temperature of receiver range from 420 to 690 K as the MGE concentration n increasing from 0.01% to 0.03%. When keep increasing the graphene concentration to 0.05%, the maximum mean temperature of MGE/[HMIM]BF4 increase slightly to 695 K. That is because of the remarkable light absorptive capacity of graphene, tiny graphene could largely enhance the extinction coefficient when dispersed in [HMIM]BF4. However, keep increasing the graphene concentration to a certain concentration, for example, 0.05%, the dispersed graphene would form a reflective surface at the top of receiver, makes a volumetric receiver more closely resemble a surface absorber because the high concentrated graphene reflect proportion of sun light, which will decrease the depth of the photons and heat generation. That means in direct solar thermal collector, appropriate concentration of graphene should be chosen to obtaining higher temperature instead of reflecting the solar light to ambient or insufficiently absorbing the sunlight. Fig. 13c shows the effect of MGE concentration on the relative thermal storage capacity. As discussed above, increasing MGE concentration could help increase the thermal storage capacity. For instance, the relative thermal storage capacity of MGE/[HMIM]BF4 increases from 110% to 300 after 6000 s of exposure as the MGE concentration range from 0.01% to 0.03%. while keep increasing the MGE concentration to 0.05%, the relative thermal storage capacity of MGE/ [HMIM]BF4 keep constant with the 0.03%. This trend suggest that appropriate MGE concentration should be considered to design this kind of DASCs, because MGE has remarkable absorptive solar spectrum capacity, as mentioned in Section 4.4.

Fig. 14. Predicted temperature curves (a), average temperature (b) and receiver efficiency (c) of nanofluids with 0.01% MGE, solar concentration C = 20 and varying nanofluid height.

4.6.3. Effect of receiver height Fig. 14a illustrates the temperature profiles of MGE/[HMIM]BF4 with solar concentration C = 20, MGE concentration 0.01% and varying receiver height. It is clearly seen that the maximum temperature of the receiver increase slightly from 610 to 645 K as the receiver height increase from 3 to 7.5 cm. Moreover, the maximum temperature inside the receiver was risen in sequence with the receiver height. Since MGE/[HMIM]BF4 with 3 cm height could not completely absorb the solar light, the temperature profile shows more uniform in the receiver.

Supplementary material related to this article can be found online at http://dx.doi.org/%2010.1016/j.solmat.2017.05.062. 4.6.1. Effect of solar concentration Fig. 12a shows the effect of solar concentration on the temperature profiles inside the receiver. As mentioned above, the temperature on the top is closely matched the heat release. With the same MGE concentration 0.01% and receiver height H = 5.9 cm, the maximum temperature of the receiver increases from 425 to 750 K as the solar 229

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

As the receiver height gradually increase to 7.5 cm, the whole range of solar spectrum is absorbed by MGE/[HMIM]BF4, resulting larger heat generation, therefore improving the maximum temperature of receiver. In addition, heat losses are dominated by convective heat lost and reradiation at the surface of the receiver. Therefore, the integration of heat generation and losses increases the maximum temperature site in the DASC and maximum mean temperature, as shown in Fig. 14b. In general, higher receiver height makes better absorptive capacity, which will increase the thermal storage capacity. Fig. 14c shows how the thermal storage capacity increases with increasing receiver height for temperatures range from 385 to 560 K. As the receiver height increases from 3 to 7.5 cm, the relative thermal storage capacity was increased from 82% to 130%. Theoretically, larger receiver height allow to increase the thermal storage capacity, but in practical, larger height required longer exposure time to gain the necessary temperature, means that larger amount of solar heat flux is required to heat the transfer fluid to reduce exposure time as the usable solar radiation is limited in a day. Hence, increasing receiver height was usually together with increasing solar concentration in the optimization of DASC. 5. Optimization of MGE/[HMIM]BF4 nanofluid based DASCs 5.1. Optimization of optical thickness In the DASCs, the solar radiation intensity is attenuated by the nanofluid along the y-direction. Hence, the solar absorption fraction directly depends on the MGE concentration and receiver height. In the DASCs, the MGE concentration and receiver height can be integrated to a parameter, optical thickness (τ) [39]. The optical thickness is physically integrated of the extinction coefficient along the solar light path in the DASCs. The extinction coefficient of the nanofluid is mathematically related with the mass fraction of the MGE. As shown in Fig. 15a, the experimental extinction coefficient of the nanofluid at three particular wavelengths (500, 1000, 1500 nm) was fitted and found that the extinction coefficient is determined linearly with the extinction coefficient of basefluid (Kb,λ) and the mass fraction of MGE(f).

Kn, λ = Kb, λ + Mλ f

(4)

Where the Kn,λ is the extinction coefficient of the nanofluid. Hence, the extinction property of the nanofluid is determined by the extinction coefficient of basefluid and the mass fraction. Further, the optical thickness which is integrated of the extinction coefficient and receiver height, could be calculated by a simple iterative algorithm using the semi-empirical approach described in previous paper.

τ=

y=0

∫y=H

Kn, λ dy

(5)

Fig. 15b shows the effect of mass fraction and receiver height on the optical thickness. As discussed previously, the optical thickness of a DASC can be adjusted by varying the mass fraction of the MGE and receiver height. The mass fraction of the MGE and receiver height are inverse correlation with the constant optical thickness. With increasing optical thickness, volumetric absorption will more closely resemble surface absorption because the photon penetration depth will decrease. With shorter penetration depths, the heat release is localized to the top of the receiver where most of the losses occur. Hence, a DASC with τ = 4 is less efficient than one with τ = 1.75. On the other hand, if the optical thickness is too small (as is the case when τ = 1), the receiver is unable to absorb all of the incident solar radiation. Using the relative thermal storage capacity as a metric, the optimum optical thickness was determined to be 1.75, as shown in Fig. 15c. The optical thickness was fixed to this optimum value for the remainder of the study.

Fig. 15. Extinction coefficient of nanofluids under varying wavenumber (a), the nanofluid height vis mass fraction of nanofluids under varying optical thickness (b) and optimization of optical thickness of nanofluids with 0.01% MGE under solar concentration C = 20 (c).

thickness, the effects of solar concentration and MGE concentration on the exposure time were investigated. Each relative thermal storage capacity, displayed in Fig. 16, corresponds to a specific final mean temperature and exposure time. As shown in Fig. 16, the amount of exposure time needed to achieve optimum relative thermal storage capacity decreases with increasing solar concentration and MGE concentration with τ = 1.75. Exposure time is an important quantity in solar thermal plants because a shorter exposure time translates to the capability to store more thermal energy for a limited amount of hours of

5.2. Optimization of exposure time With the efficiency of the DASCs optimized with respect to optical 230

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

References [1] S. Chol, Enhancing thermal conductivity of fluids with nanoparticles, ASME-Publ.Fed. 231 (1995) 99–106. [2] S. Murshed, K. Leong, C. Yang, Enhanced thermal conductivity of TiO2—water based nanofluids, Int. J. Therm. Sci. 44 (2005) 367–373. [3] S.K. Das, N. Putra, P. Thiesen, W. Roetzel, Temperature dependence of thermal conductivity enhancement for nanofluids, J. Heat Transf. 125 (2003) 567–574. [4] S.P. Jang, S.U. Choi, Role of Brownian motion in the enhanced thermal conductivity of nanofluids, Appl. Phys. Lett. 84 (2004) 4316–4318. [5] J.A. Eastman, S. Choi, S. Li, W. Yu, L. Thompson, Anomalously increased effective thermal conductivities of ethylene glycol-based nanofluids containing copper nanoparticles, Appl. Phys. Lett. 78 (2001) 718–720. [6] Y. Xuan, Q. Li, Investigation on convective heat transfer and flow features of nanofluids, J. Heat Transf. 125 (2003) 151–155. [7] D. Wen, Y. Ding, Experimental investigation into convective heat transfer of nanofluids at the entrance region under laminar flow conditions, Int. J. Heat Mass Transf. 47 (2004) 5181–5188. [8] S.Z. Heris, S.G. Etemad, M.N. Esfahany, Experimental investigation of oxide nanofluids laminar flow convective heat transfer, Int. Commun. Heat Mass Transf. 33 (2006) 529–535. [9] K. Leong, R. Saidur, S. Kazi, A. Mamun, Performance investigation of an automotive car radiator operated with nanofluid-based coolants (nanofluid as a coolant in a radiator), Appl. Therm. Eng. 30 (2010) 2685–2692. [10] S.C. Mohapatra, An overview of liquid coolants for electronics cooling, Electron. Cool. 12 (2006) 22. [11] D.P. Kulkarni, D.K. Das, R.S. Vajjha, Application of nanofluids in heating buildings and reducing pollution, Appl. Energy 86 (2009) 2566–2573. [12] R. Strandberg, D.K. Das, Influence of temperature and properties variation on nanofluids in building heating, Energy Convers. Manag. 51 (2010) 1381–1390. [13] B. Farajollahi, S.G. Etemad, M. Hojjat, Heat transfer of nanofluids in a shell and tube heat exchanger, Int. J. Heat Mass Transf. 53 (2010) 12–17. [14] I. Anghel, A.M. Grumezescu, E. Andronescu, A.G. Anghel, A. Ficai, C. Saviuc, V. Grumezescu, B.S. Vasile, M.C. Chifiriuc, Magnetite nanoparticles for functionalized textile dressing to prevent fungal biofilms development, Nanoscale Res. Lett. 7 (2012) 1. [15] M.M.R. Mollick, B. Bhowmick, D. Mondal, D. Maity, D. Rana, S.K. Dash, S. Chattopadhyay, S. Roy, J. Sarkar, K. Acharya, Anticancer (in vitro) and antimicrobial effect of gold nanoparticles synthesized using Abelmoschus esculentus (L.) pulp extract via a green route, RSC Adv. 4 (2014) 37838–37848. [16] X. Li, Y. Xing, Y. Jiang, Y. Ding, W. Li, Antimicrobial activities of ZnO powder‐coated PVC film to inactivate food pathogens, Int. J. Food Sci. Technol. 44 (2009) 2161–2168. [17] Y. Xuan, Q. Li, W. Hu, Aggregation structure and thermal conductivity of nanofluids, AIChE J. 49 (2003) 1038–1043. [18] R. Prasher, P.E. Phelan, P. Bhattacharya, Effect of aggregation kinetics on the thermal conductivity of nanoscale colloidal solutions (nanofluid), Nano Lett. 6 (2006) 1529–1534. [19] W. Evans, R. Prasher, J. Fish, P. Meakin, P. Phelan, P. Keblinski, Effect of aggregation and interfacial thermal resistance on thermal conductivity of nanocomposites and colloidal nanofluids, Int. J. Heat Mass Transf. 51 (2008) 1431–1438. [20] Y. Hwang, J. Lee, C. Lee, Y. Jung, S. Cheong, C. Lee, B. Ku, S. Jang, Stability and thermal conductivity characteristics of nanofluids, Thermochim. Acta 455 (2007) 70–74. [21] L. Jiang, L. Gao, J. Sun, Production of aqueous colloidal dispersions of carbon nanotubes, J. Colloid Interface Sci. 260 (2003) 89–94. [22] Y. Hwang, J.-K. Lee, J.-K. Lee, Y.-M. Jeong, S.-i. Cheong, Y.-C. Ahn, S.H. Kim, Production and dispersion stability of nanoparticles in nanofluids, Powder Technol. 186 (2008) 145–153. [23] D. Zhu, X. Li, N. Wang, X. Wang, J. Gao, H. Li, Dispersion behavior and thermal conductivity characteristics of Al2O3–H2O nanofluids, Curr. Appl. Phys. 9 (2009) 131–139. [24] M. Assael, I. Metaxa, J. Arvanitidis, D. Christofilos, C. Lioutas, Thermal conductivity enhancement in aqueous suspensions of carbon multi-walled and double-walled nanotubes in the presence of two different dispersants, Int. J. Thermophys. 26 (2005) 647–664. [25] S. Murshed, K. Leong, C. Yang, Investigations of thermal conductivity and viscosity of nanofluids, Int. J. Therm. Sci. 47 (2008) 560–568. [26] H. Xie, H. Lee, W. Youn, M. Choi, Nanofluids containing multiwalled carbon nanotubes and their enhanced thermal conductivities, J. Appl. Phys. 94 (2003) 4967–4971. [27] W. Wang, X. Yang, Y. Fang, J. Ding, Preparation and performance of form-stable polyethylene glycol/silicon dioxide composites as solid–liquid phase change materials, Appl. Energy 86 (2009) 170–174. [28] A. Ghadimi, R. Saidur, H. Metselaar, A review of nanofluid stability properties and characterization in stationary conditions, Int. J. Heat Mass Transf. 54 (2011) 4051–4068. [29] J. Jiang, G. Oberdörster, P. Biswas, Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies, J. Nanopart. Res. 11 (2009) 77–89. [30] T. Welton, Room-temperature ionic liquids. Solvents for synthesis and catalysis, Chem. Rev. 99 (1999) 2071–2084. [31] P. Wasserscheid, W. Keim, Ionic liquids—new “solutions” for transition metal catalysis, Angew. Chem. Int. Ed. 39 (2000) 3772–3789.

Fig. 16. Optimization of exposure time under varying solar concentration and mass fraction.

sunlight. The results indicate that DASCs are advantageous for applications with high levels of solar concentration such as central receiver designs.

6. Conclusions A novel method for improving the dispersion stability of the ILbased nanofluids has been explored. GE-COOH has been modified with the reagents for use in the synthesis of HMIM]BF4 by the same process. The HMIM]BF4 based nanofluids have been prepared by dispersing the obtained MGE into the IL at different loadings. The characterizations by FT-IR and XPS suggest that the surface of MGE has been successfully grafted with the molecules similar to [HMIM]BF4, which leads to a reduction in transparence of the nanosheets. Different from the GECOOH/[HMIM]BF4 nanofluid, the MGE/[HMIM]BF4 ones exhibit good dispersion stability even after being heated. After being heated, the 0.01% GE-COOH/[HMIM]BF4 nanofluid shows a remarkable increase from 20–30% to 80% in transmittance in the visible range, while no obvious changes in transmittance have been found for the ones containing MGE. A combined numerical and experimental investigation on the photo-thermal performance of the MGE/[HMIM]BF4 and GECOOH/[HMIM]BF4 nanofluid to verify the great effect of dispersion stability on the photo-thermal performance. The MGE/[HMIM]BF4 has much higher receiver efficiency than GE-COOH/[HMIM]BF4 nanofluid at higher temperature. Further, the numerical model was used to predict the receiver temperature of MGE/[HIM]BF4 nanofluid under varying solar concentration, MGE concentration and nanofluid height. The model shows that the thermal storage capacity of the MGE/ [HMIM]BF4 based DASC increases with increasing solar concentration and receiver height, but conversely with the MGE concentration under concentrated incident solar intensity. The optimum optical thickness for a DASC is 1.75. The good dispersion stability at elevated temperatures, excellent optical property and high photo-thermal conversion performance make the MGE/[HMIM]BF4 nanofluids show great potential for used as HTF for DASC in practical applications, and the outcomes of the study provide an important perspective as to how nanofluids can be best utilized as DASCs in concentrated solar applications.

Acknowledgements This work was supported by the National Natural Science Foundation of China (No. 21276088, No. U1507201 and No. U1407132), and the Guangdong Natural Science Foundation (2014A030312009).

231

Solar Energy Materials and Solar Cells 170 (2017) 219–232

J. Liu et al.

[40] D. Li, M.B. Müller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous dispersions of graphene nanosheets, Nat. Nanotechnol. 3 (2008) 101–105. [41] B. Wang, X. Wang, W. Lou, J. Hao, Rheological and tribological properties of ionic liquid-based nanofluids containing functionalized multi-walled carbon nanotubes, J. Phys. Chem. C 114 (2010) 8749–8754. [42] A. Jimenez, M. Bermudez, P. Iglesias, F. Carrion, G. Martinez-Nicolas, 1< i > N < /i > -alkyl-3-methylimidazolium ionic liquids as neat lubricants and lubricant additives in steel–aluminium contacts, Wear 260 (2006) 766–782. [43] M. Mehrali, S.T. Latibari, M. Mehrali, T.M.I. Mahlia, E. Sadeghinezhad, H.S.C. Metselaar, Preparation of nitrogen-doped graphene/palmitic acid shape stabilized composite phase change material with remarkable thermal properties for thermal energy storage, Appl. Energy 135 (2014) 339–349. [44] N.J. Bridges, A.E. Visser, E.B. Fox, Potential of nanoparticle-enhanced ionic liquids (NEILs) as advanced heat-transfer fluids, Energy Fuels 25 (2011) 4862–4864. [45] F. Marquis, L. Chibante, Improving the heat transfer of nanofluids and nanolubricants with carbon nanotubes, JOM 57 (2005) 32–43. [46] E. Zarza, L. Valenzuela, J. León, K. Hennecke, M. Eck, H.D. Weyers, M. Eickhoff, Direct steam generation in parabolic troughs: final results and conclusions of the DISS project, Energy 29 (2004) 635–644. [47] M. Romero, A. Steinfeld, Concentrating solar thermal power and thermochemical fuels, Energy Environ. Sci. 5 (2012) 9234–9245. [48] H. Müller-Steinhagen, F. Trieb, Concentrating solar power, A review of the technology, Ingenia Inform. QR Acad. Eng. 18 (2004) 43–50.

[32] C. Nieto de Castro, M. Lourenço, A. Ribeiro, E. Langa, S. Vieira, P. Goodrich, C. Hardacre, Thermal properties of ionic liquids and ionanofluids of imidazolium and pyrrolidinium liquids, J. Chem. Eng. Data 55 (2009) 653–661. [33] F. Wang, L. Han, Z. Zhang, X. Fang, J. Shi, W. Ma, Surfactant-free ionic liquid-based nanofluids with remarkable thermal conductivity enhancement at very low loading of graphene, Nanoscale Res. Lett. 7 (2012) 1–7. [34] A.A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C.N. Lau, Superior thermal conductivity of single-layer graphene, Nano Lett. 8 (2008) 902–907. [35] J. Liu, F. Wang, L. Zhang, X. Fang, Z. Zhang, Thermodynamic properties and thermal stability of ionic liquid-based nanofluids containing graphene as advanced heat transfer fluids for medium-to-high-temperature applications, Renew. Energy 63 (2014) 519–523. [36] L. Zhang, J. Liu, G. He, Z. Ye, X. Fang, Z. Zhang, Radiative properties of ionic liquidbased nanofluids for medium-to-high-temperature direct absorption solar collectors, Sol. Energy Mater. Sol. Cells 130 (2014) 521–528. [37] J. Liu, Z. Ye, L. Zhang, X. Fang, Z. Zhang, A combined numerical and experimental study on graphene/ionic liquid nanofluid based direct absorption solar collector, Sol. Energy Mater. Sol. Cells 136 (2015) 177–186. [38] M.J. Park, J.K. Lee, B.S. Lee, Y.-W. Lee, I.S. Choi, S.-G. Lee, Covalent modification of multiwalled carbon nanotubes with imidazolium-based ionic liquids: effect of anions on solubility, Chem. Mater. 18 (2006) 1546–1551. [39] A. Lenert, E.N. Wang, Optimization of nanofluid volumetric receivers for solar thermal energy conversion, Sol. Energy 86 (2012) 253–265.

232