- Email: [email protected]
Experimental research on particle aggregation behavior in nanorefrigerant–oil mixture
Applied Thermal Engineering 98 (2016) 944–953
Contents lists available at ScienceDirect
Applied Thermal Engineering j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a p t h e r m e n g
Research Paper
Experimental research on particle aggregation behavior in nanorefrigerant–oil mixture Lingnan Lin a, Hao Peng b, Guoliang Ding a,* a b
Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200240, China Department of Energy and Power Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
H I G H L I G H T S
• • • • •
Time evolutions of the hydrodynamic diameter of aggregates in nanorefrigerant–oil mixture were measured. The presence of lubricating oil inhibits the aggregation of nanoparticles. The inhibition of oil on aggregation is strengthened with the increase of oil concentration. Larger primary particle size or concentration leads to larger hydrodynamic diameter of aggregate. The hydrodynamic diameter of aggregate is enlarged with the rise of temperature.
A R T I C L E
I N F O
Article history: Received 10 October 2015 Accepted 11 December 2015 Available online 14 January 2016 Keywords: Aggregation Nanorefrigerant Oil Size Temperature
A B S T R A C T
The objective of this study is to experimentally investigate the effects of primary particle size, primary particle concentration and temperature on particle aggregation behavior in nanorefrigerant–oil mixture. The nanoparticles, refrigerant and lubricating oil for experiments were TiO2, R141b and ATMOS NM56, respectively. Experimental conditions included primary particle size from 25 to 100 nm, primary particle concentration from 50 to 500 mg L−1, temperature from 6 to 27 °C, and oil concentrations of 1, 3 and 5 wt%. Time evolutions of the hydrodynamic diameter of aggregates were measured by dynamic light scattering (DLS) method. It is shown that the presence of lubricating oil inhibits the particle aggregation, and the inhibition is strengthened with the increase of oil concentration; the hydrodynamic diameter of aggregates is enlarged with the increase of primary particle size, primary particle concentration or temperature, and the enlargement increases with the increase of oil concentration. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Nanorefrigerant–oil mixture is a special kind of nanofluid which is prepared by dispersing nanoparticles in refrigerant–oil mixture. As the addition of nanoparticles remarkably increases the thermal conductivity and the boiling heat transfer coefficient, nanorefrigerant–oil mixtures have better heat exchange performance compared to conventional refrigerant–oil mixtures [1–3], and have great potential for improving the energy efficiencies of vapor compression refrigeration systems [4–8]. However, the nanoparticles in refrigerant–oil mixture are likely to form aggregates due to strong Brownian motion and high surface free energy. The aggregation of nanoparticles accelerates the sedimentation process and then reduces the long-term stability, which will hinder the actual application of nanorefrigerant–oil mixtures. In order to develop the technology for maintaining the long-term stability of nanorefrigerant–oil
* Corresponding author. Tel.: +86 21 34206378; fax: +86 21 34206814. E-mail address: [email protected] (G. Ding). http://dx.doi.org/10.1016/j.applthermaleng.2015.12.052 1359-4311/© 2015 Elsevier Ltd. All rights reserved.
mixture, the particle aggregation behavior in nanorefrigerant–oil mixture should be known. The effects of primary particle parameters (size and concentration) and temperature should be considered during the investigation on the particle aggregation behavior in nanorefrigerant–oil mixture due to the following reasons. Firstly, the primary particle parameters (size and concentration) affect the interaction between particles. Secondly, the variation of temperature alters the thermophysical properties of refrigerant–oil mixture as well as Brownian motion of nanoparticles, which leads to the change of particle aggregation behavior. The existing researches on particle aggregation behavior focus on the nanofluids with the base fluids of water [9–17], aqueous solution [9,13,18], ethylene glycol [12,19,20], silicon oil [15], gear oil [21] and refrigerant [22–24], as summarized in Table 1. The experimental methods for evaluating the particle aggregation behavior include dynamic light scattering (DLS) [9,11–24], zeta potential analysis [9,11,13–15,19,23], UV–vis spectrophotometer [10,24], transmittance measurement [22], sedimentation picture capturing [11,14,19,22,24], hydrometer [12], scanning electron microscope
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
945
Table 1 Existing studies on particle aggregation behavior in nanofluids. Researchers (year)
Base fluid
Nanoparticle
Evaluation method
Investigated factors
Ref.
Hong et al. (2006) Bi et al. (2007)
EG R113, R141b, R123
Fe TiO2
Time elapse Temperature, time elapse
[20] [22]
Li et al. (2007)
Water
Cu
Water, silicon oil Water
Carbon black, Ag TiO2 (anatase)
Suttiponparnit et al. (2010)
Water
TiO2 (0–100% anatase), TiO2 (P25)
DLS, zeta potential analysis
Pastoriza-Gallego et al. (2011)
Water
CuO
Kole et al. (2011) Gharagozloo and Goodson (2011)
Gear oil Water
CuO Al2O3
UV–vis spectrophotometer, TEM, SEM, DLS DLS
Ismay et al. (2012)
Water
TiO2 (rutile)
pH, surfactant type and concentration Preparing method, pH pH, ionic strength, surface charge and surface coating Primary particle size, pH, ionic strength, surface area, crystal phase Primary particle concentration, pH, time elapse Primary particle concentration Primary particle size and concentration, temperature, time elapse pH, sonication time
[11]
Hwang et al. (2008) Jiang et al. (2009)
DLS DLS, sedimentation picture capturing, transmittance measurement DLS, zeta potential analysis, sedimentation picture capturing DLS, zeta potential analysis, TEM DLS, zeta potential analysis
Lee et al. (2013)
NaCl aqueous solution
Al2O3
[18]
Witharana et al. (2013)
PG, EG, water/PG, water/EG
TiO2, Al2O3, ZnO
Primary particle concentration, temperature, ultrasonication time Base fluid type, day light
Lee et al. (2014) Gomez-Merino et al. (2014) Peng et al. (2015)
Water, EG Water R141b
Al2O3, Ag, ZnO TiO2 (anatase) TiO2 (anatase)
[12] [16] [23]
Lin et al. (2015)
R141b
MWCNTs
Type of nanofluids, time elapse Time elapse Primary particle size and concentration surfactant, time elapse Surfactant type and concentration, ultrasonication time, time elapse, primary particle concentration
(SEM) [10,18] and transmission electron microscope (TEM) [10,14–16]. The results indicated that the aggregation behavior are affected by some factors such as primary particle size and concentration, temperature, base fluid type, time elapse, pH value, surfactant type and concentration, ionic strength, and ultrasonication time. The particle aggregation behavior in nanorefrigerant-oil mixture may be different from those in the oil-free nanorefrigerant and other non-refrigerant-based nanofluids revealed by the above existing researches, based on the following two reasons. Firstly, the interaction between the oil molecules and the refrigerant molecules could change the particle aggregation behavior. Secondly, the physicochemical properties (viscosity, Hamaker constant, etc.) of refrigerant– oil mixture are different from the base fluids in the existing researches, causing different motion of nanoparticles and interaction between nanoparticles, which leads to the change of aggregation behavior. In order to know the particle aggregation behavior in nanorefrigerant–oil mixture, the experiments using refrigerant– oil mixture as base fluid are carried out in the present study, considering the effects of primary particle parameters size, primary particle concentration and temperature.
DLS, zeta potential analysis, sedimentation picture capturing, TEM Sedimentation picture capturing, DLS, SEM DLS, zeta potential analysis, sedimentation picture capturing DLS, hydrometer DLS, TEM DLS, zeta potential
Sedimentation picture capturing, UV–vis spectrophotometer, DLS
[15] [13] [9]
[10] [21] [17]
[14]
[19]
[24]
2. Experimental materials and method 2.1. Preparation of nanorefrigerant–oil mixture The materials used for preparing the nanorefrigerant–oil mixtures include nanoparticles, pure refrigerant and lubricating oil. Titania (TiO2) nanoparticles with primary particle sizes of 25, 40, 60 and 100 nm (purchased from Aladdin Industrial Corporation, Shanghai, China) were used, which are in accordance with those used by Peng et al. [23]. Titania nanoparticles could enhance the heat transfer performance of refrigerant and have stable chemical and physical properties, so they have been widely used in nanorefrigerant researches [5,22,25–28]. The properties of the TiO2 nanoparticles are listed in Table 2. R141b (1,1–Dichloro–1–fluoroethane) was used as the base refrigerant, based on the following reasons. As the nanorefrigerant is usually prepared by two-step method under ambient condition [2,3], it is better to choose those refrigerants in liquid state under ambient condition. However, the common refrigerants such as R410A and R134a are in vapor state under ambient condition, so the experimental nanorefrigerant cannot be prepared by these refrigerants.
Table 2 Properties of TiO2 nanoparticle. Primary particle size (nm)
Specific surface area (m2 g−1)
Molecular weight (g mol−1)
Density (kg m3)
Purity (%)
Crystal phase
Appearance
25 40 60 100
250–400 100–200 30–60 15–20
79.9
3780
100
Anatase
White powder
946
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Table 3 Properties of R141b at different temperature under ambient pressure. Properties
At 6 °C/13 °C/20 °C/27 °C
Chemical formula Molecular weight (g mol−1) Boiling point (°C) Critical temperature (°C) Critical pressure (Mpa) Liquid phase density (kg m3) (1 atm) Liquid phase viscosity (μPa s) (1 atm)
C2H3Cl2F 116.95 32.05 204.5 4.25 1271.7/1254.8/1245.3/1235.8 517.43/463.63/437.11/412.64
Among the refrigerants in liquid state under ambient condition, R141b is still in use and the physicochemical properties are close to those of the common refrigerants, so it is widely used as the base fluid in nanorefrigerant researches [23,24,28–31]. The R141b was purchased from Zhejiang Juhua Co., Ltd., China, and its properties at different temperature under ambient pressure are listed in Table 3. The lubricating oil ATMOS NM56 was used in preparing the nanorefrigerant–oil mixture, which is a mineral oil with kinematic viscosity of 56.2 mm2 s−1 at 40 °C and density of 0.928 g cm−3 at 15 °C, as reported by the manufacturer. The oil concentrations (ω) include 1, 3 and 5 wt%, covering the oil concentrations used in the actual refrigeration systems. The lubricating oil NM56 is completely miscible with R141b over the entire range of oil concentration and temperature under present experimental conditions. The nanorefrigerant–oil mixture (TiO2/R141b/NM56) was prepared by the following steps: (1) weighing the required mass of TiO2 nanoparticles, lubricating oil NM56 and refrigerant R141b, respectively, via a digital electronic balance with measurement range of 10–210 mg and maximum error of 0.1 mg; (2) mixing the weighed TiO2 nanoparticles, lubricating oil NM56 and refrigerant R141b together to form the nanorefrigerant–oil mixture; (3) vibrating the nanorefrigerant–oil mixture by an ultrasonic processor with output power of 110 W and amplitude of 40% for 20 min to fully disperse the nanoparticles. As the R141b is volatile under ambient condition, it was compensated every 5 min during ultrasonication process to keep the primary particle concentration unchanged. The photograph of the prepared nanorefrigerant–oil mixtures (TiO2/R141b/NM56) with different oil concentrations just after ultrasonication are shown in Fig. 1, and it can be seen that the nanoparticles are all well dispersed in the nanorefrigerant–oil mixtures. The experiments on the oil-free nanorefrigerant (TiO2/R141b) were also performed in the present study. The preparation procedure of oil-free nanorefrigerant (TiO2/R141b) is identical with that of nanorefrigerant–oil mixture (TiO2/R141b/NM56) except for not adding lubricating oil. 2.2. Hydrodynamic diameter measurement The particle aggregation behavior is quantitatively evaluated by the time evolution of hydrodynamic diameter of aggregates, which is measured based on the dynamic light scattering (DLS) technique. The principle of the hydrodynamic diameter measurement by DLS could be found in Ref. [23]. The DLS apparatus used is Particle Sizer NICOMP™ 380 ZLS (PSS–NICOMP Inc., California, USA), which consists of a He–Ne laser generator, an optical lens, a sample cell, a detector, a signal processor, a Peltier thermoelectric element, a chamber and a computer, as shown in Fig. 2. The He–Ne laser generator (35 mW) is used to emit a laser beam with wavelength of 639 nm. The optical lens is used to focus the laser beam on the sample cell. A glass, cubic sample cell (inner size: 10 mm × 10 mm × 45 mm) with a lid on its top is used to contain the sample of nanorefrigerant. The particles of the nanorefrigerant in the sample cell are illuminated by the laser beam
Fig. 1. Photograph of the prepared nanorefrigerant–oil mixtures (TiO2/R141b/ NM56, dp = 40 nm, c0 = 250 mg L−1, T = 20 °C) with different oil concentration just after ultrasonication: (a) 1 wt%; (b) 3 wt%; (c) 5 wt%.
and scatter the light. The detector receives the signals of the scattered light, and transfers them to the signal processor. The signal processor converts the signal to data, and send the data to the computer. The Peltier thermoelectric element is used to monitor the temperature of the sample cell and heat/cool the sample cell, and it is fixed on the wall of the chamber for putting sample cell, as shown in Fig. 3. The chamber is well thermal insulated so that the temperature gradient of the liquid sample in the cell can be neglected. In order to investigate the effect of temperature on particle aggregation behavior, the temperature of nanorefrigerant sample should be controlled. The principle of temperature control is described as follows. Before DLS measurement, a temperature value is set in the software of computer. During measurement, the Peltier thermoelectric element constantly monitors the temperature of the sample cell, and transmits the temperature signals to the computer through feedback circuit. If the temperature is below (or above) the set value, the computer will command the Peltier thermoelectric element to heat (or cool) the sample cell through control circuit, until the set value is reached. The control range of temperature is 4–60 °C with accuracy of ±0.2 °C. The hydrodynamic diameter measurement for each condition is repeated for 5 times to calculate the average value and standard deviation of hydrodynamic diameter. 2.3. Experimental conditions The experimental conditions are divided into three categories, in which the first category deals with the effect of primary particle size, the second category deals with the effect of primary particle concentration, and the third category deals with the effect of temperature. Detailed experimental conditions are listed in Table 4. In the first category, four different primary particle sizes including 25, 40, 60 and 100 nm are used to evaluate the effect of primary particle size on particle aggregation behavior. In the second category, four different primary particle concentrations including 50, 100, 250 and 500 mg L−1 are used to evaluate the effect of primary particle concentration on particle aggregation behavior.
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
947
Fig. 2. Schematic diagram of the DLS apparatus.
In the third category, four different temperatures including 6, 13, 20 and 27 °C are used to evaluate the effect of temperature on particle aggregation behavior. The temperature range was chosen to meet two requirements, i.e., within the measurement range of DLS
apparatus (4 ~ 60 °C), and lower than the boiling point of refrigerant R141b at 1 atm (32.1 °C). For each category, four different oil concentrations including 0, 1, 3, 5 wt% are used to evaluate the influence of oil concentration. The experimental data for oil concentrations of 1, 3, 5 wt% are obtained from the present study. The experimental data for oil concentrations of 0 wt% (i.e., oil-free nanorefrigerant) are obtained from Peng et al. [23] (categories 1 and 2) or the present study (category 3).
3. Results and discussion 3.1. Particle aggregation behavior in nanorefrigerant–oil mixture
Fig. 3. Photograph of the chamber for putting sample cell and the Peltier thermoelectric element.
Fig. 4 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% at primary particle size of 40 nm, primary particle concentration of 250 mg L−1 and temperature of 20 °C. It can be seen that with the increase of time from 0 to 4 h, the hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are enlarged by 26.9%, 26.0% and 23.5%, respectively.
Table 4 Experimental conditions for analysis. Objective
Nano-particles
Base fluid
Oil concentration (wt%)
Primary particle size (nm)
Primary particle concentration (mg L−1)
Temperature (°C)
Time range (h)
Data source
1. Evaluate the effects of primary particle size 2. Evaluate the effect of primary particle concentration 3. Evaluate the effect of temperature
TiO2
R141b R141b/NM56 R141b R141b/NM56
0 1, 3, 5 0 1, 3, 5
25, 40, 60, 100 25, 40, 60, 100 40 40
250 250 50, 100, 250, 500 50, 100, 250, 500
20 20 20 20
0–4 0–4 0–4 0–4
Peng et al. (2015) [23] Present study Peng et al. (2015) [23] Present study
R141b R141b R141b/NM56
0 0 1, 3, 5
40 40 40
250 250 250
20 6, 13, 27 6, 13, 20, 27
0–4 0–4 0–4
Peng et al. (2015) [23] Present study Present study
TiO2
TiO2
948
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
eters (dh,a) in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are smaller by 19.7%, 23.5% and 31.7% than that in oil-free nanorefrigerant, respectively, indicating that the presence of lubricating oil inhibits the particle aggregation, and the inhibition is strengthened with the increase of oil concentration. The inhibition of lubricating oil on particle aggregation could be explained in the following three aspects.
350 ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh (nm)
300
3.1.1. Effect of oil on collision frequency of nanoparticles The aggregation rate is positively correlated to the collision frequency of two interacting particles (βij), and βij can be expressed as the following equation [32].
250
βij = 200 R141b/NM56/TiO2 dp = 40 nm, c0 = 250 mg L-1, T = 20 oC
150 0
1
2
3
4
t (h) Fig. 4. Time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant– oil mixture with different oil concentrations.
In order to compare the extent of aggregation under different conditions, and given that the hydrodynamic diameter is timedependent, the time-averaged hydrodynamic diameter (dh,a) is defined as Eq. (1).
d h,a =
1 N ∑ d h (t i ) N i =1
(1)
where ti is the time point, N is the total number of time points. Fig. 5 shows the time-averaged hydrodynamic diameters (dh,a) in nanorefrigerant–oil mixtures with oil concentrations (ω) of 1, 3 and 5 wt% and oil-free nanorefrigerant (ω = 0 wt%) at primary particle size of 40 nm, primary particle concentration of 250 mg L−1 and temperature of 20 °C. The time-averaged hydrodynamic diam-
⎛ 1 1⎞ 2k BT (Ri + R j ) ⎜ + ⎟ 3μ ⎝ Ri R j ⎠
(2)
where kB is the Boltzmann constant (kB = 1.38 × 10−27 J K−1), T is the temperature, μ is the liquid viscosity, Ri and Rj are the effective radii of two interacting particles, and the term of “kBT/μ” reflects the contribution of Brownian motion in the collision frequency. The collision frequency in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are smaller by 10.2%, 23.5% and 25.4% than that in oil-free nanorefrigerant, respectively, which is calculated by Eq. (2) using the viscosities of refrigerant–oil mixtures presented in Ref. [33], and the calculation conditions include dp = 40 nm, c0 = 250 mg L−1 and T = 20 °C. The reduction of collision frequency is caused by the increase of liquid viscosity with the addition of oil. The reduction degree of collision frequency caused by the presence of oil is less than that of time-averaged hydrodynamic diameters (dh,a) shown in Fig. 5, which indicates that the effect of oil on the collision frequency is the main reason accounting for the inhibition of aggregation, but not the only one. 3.1.2. Effect of oil on Hamaker constant The aggregation rate is also positively correlated to Hamaker constant of nanoparticle–liquid–nanoparticle in nanorefrigerant–oil mixture (An–ro–n), and An–ro–n can be calculated by Eq. (3) [34–36].
An-ro-n =
(
An − Aro
)
2
(3-a)
2 3 3h v (ηn2 − 1) ⎛ ε − 1⎞ k BT ⎜ n ⎟ + P e 32 ⎝ ⎠ 4 εn + 1 16 2 (ηn2 + 1) 2
An =
400
(3-b)
R141b/TiO2 (Peng et al., 2015) R141b/NM56/TiO2 (present study)
2 2 − 1) 3 3h v (ηro ⎛ ε − 1⎞ k BT ⎜ ro ⎟ + P e 32 2 ⎝ ε ro + 1⎠ 4 16 2 (ηro + 1) 2
Aro =
350
dh,a (nm)
ε ro = (1 − ω ) ε r + ωε o ηro =
300
250 dp = 40 nm, c0 = 250 mg L-1, T = 20 oC
200 -1
0
1
2
3
4
5
6
ω (wt%) Fig. 5. Time-averaged hydrodynamic diameters (dh,a) for different oil concentrations (ω).
ηrηo
(1 − ω )ηo + ωηr
(3-c)
(3-d)
(3-e)
where An and Aro are the Hamaker constants of nanoparticle and refrigerant–oil mixture, respectively; hP is the Planck constant (6.626 × 10−34 J s); νe is the main electronic absorption frequency in the ultraviolet region (about 3 × 1015 s−1); εn, εr, εo, εro are the dielectric permittivities of nanoparticle, pure refrigerant, oil and refrigerant–oil mixture, respectively; ηn, ηr, ηo, ηro are the refractive indexes of nanoparticle, pure refrigerant, oil and refrigerant– oil mixture, respectively; and the values of εn, εr, εo, ηn, ηr and ηo are listed in Table 5. The Hamaker constant of nanoparticle–liquid–nanoparticle in nanorefrigerant–oil mixtures (An–ro–n) with oil concentrations of 1, 3 and 5 wt% are smaller by 0.2%, 0.5% and 0.9% than that in
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Table 5 Dielectric permittivities and refractive indexes of pure refrigerant, oil and nanoparticle. Meaning
Value
Ref.
εr ηr εo ηo εn ηn
Dielectric permittivity of R141b Refractive index of R14b Dielectric permittivity of lubricating oil Refractive index of lubricating oil Dielectric permittivity of TiO2 Refractive index of TiO2
8.13 1.37 2.5 1.45 86 2.5
[37] [38] [39] [40] [41] [41]
oil-free nanorefrigerant, respectively, which is calculated by Eq. (3) and the calculation conditions include dp = 40 nm, c0 = 250 mg L−1 and T = 20 °C. This result indicates that the effect of oil on Hamaker constant is another reason accounting for the inhibition of aggregation, but the contribution of oil effect on Hamaker constant is much weaker than that of oil effect on collision frequency. 3.1.3. Effect of oil excess layer on van der Waals potential energy Lubricating oil has stronger adhesion on solid surface compared to refrigerant [42], causing the possible existence of oil excess layer around nanoparticles in nanorefrigerant–oil mixture, as shown in Fig. 6. The variation of oil excess layer thickness (δ) will change the van der Waals attractive potential between nanoparticles (UA), and the relationship between δ and UA is as the following equation [43]. 2 R +δ 2 R ⎡ ⎤ Aro − A o + A o − An h h + 2δ ⎥ 1⎢ ⎥ UA = − ⎢ R (R + δ ) ⎥ 12 ⎢ A o − An +4 Aro − A o ⎢⎣ (h + δ ) (2R + δ ) ⎥⎦
(
(
)
(
)(
)
)
(4)
where h is separation distance. The variation of van der Waals potential energy (|UA|) with the oil excess layer thickness (δ) is shown in Fig. 7, in which the results are calculated by Eq. (4) at the condition of R = 20 nm, Ar-o = 4.42 × 10−20 J, Ao = 5.92 × 10−20 J, An = 3.75 × 10−19 J. The calculation results show that with the increase of δ from 0 to 5 nm, |UA| is reduced by 45.9%, 29.6% and 16.9% at h = 10, 20 and 40 nm, respectively. The increase of oil concentration enlarges the oil excess layer thickness (δ), causing the reduction of van der Waals potential energy (|UA|), which inhibits the particle aggregation. 3.2. Effect of primary particle size on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 8 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures for different primary particle sizes and oil concentrations. The experimental conditions
3.0 h = 10 nm h = 20 nm h = 40 nm
2.5
2.0
|UA| (×10-20 J)
Symbol
949
1.5
1.0
0.5
0.0 0
1
2
3
4
5
δ (nm) Fig. 7. Variation of van der Waals potential energy (|UA|) with the thickness of oil excess layer (δ).
include the primary particle sizes of 25, 40, 60 and 100 nm, and the oil concentrations of 1, 3 and 5 wt%; the primary particle concentration and the temperature are fixed at 250 mg L−1 and 20 °C, respectively. From Fig. 8, it can be seen that for the nanorefrigerant–oil mixtures with primary particle sizes (dp) of 25, 40 and 60 nm, the hydrodynamic diameter (dh) increases with time, and the increasing rate is decreased as dp increases; for the nanorefrigerant–oil mixture with dp of 100 nm, dh decreases with time. This phenomenon could be explained as follows: (1) For dp of 25, 40 and 60 nm, dh is determined by the aggregation process of nanoparticles, which leads to the increase of dh with time. With the increase of dp, the interaction energy barrier increases and the surface free energy decreases, causing lower aggregation rate, which leads to the decrease of the increasing rate of dh [44]. (2) For dp of 100 nm, the aggregates are large enough to make the sedimentation process obvious, which leads to the decrease of dh with time. Fig. 9 shows the time-averaged hydrodynamic diameters (dh,a) for different primary particle sizes (dp) and oil concentrations. It can be seen that dh,a is enlarged with the increase of dp, and the enlargement increases with the rise of oil concentration. With the increase of dp from 25 to 100 nm, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 140.2%, 208.3%, 229.8% and 272.1%, respectively; the enlargement for 5 wt% is 94.1% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the increase of primary particle size (dp) is as follows. The hydrodynamic diameter of aggregates at ti (i = 1, 2, …, N) can be described by the following equation
d h (t i ) = d p ⋅ n (t i )
1 DF
(5)
where n is number of primary particles contained in an equivalent aggregate; DF is fractal dimension, which is a constant. By substituting the Eq. (1) into Eq. (5), dh,a can be described as
d h,a =
Fig. 6. Schematic diagram of oil excess layer.
dp N 1D ∑ n (t i ) F N i =1
(6)
Since the variation of dp has less impact on n as well as DF, dh,a increases with the increase of dp.
950
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Fig. 8. Time evolutions of hydrodynamic diameters (dh) of aggregates in nanorefrigerant–oil mixture for different primary particle sizes (dp) and oil concentrations (ω).
The reason for the phenomenon that the enlargement of dh,a with respect to dp increases with the rise of oil concentration is as follows: (1) With the rise of oil concentration, the viscosity of refrigerant– oil mixture increases and then the settling velocity decreases, causing the sedimentation of aggregates to be weakened. (2) The inhibition of the sedimentation of aggregates has positive effect on the enlargement of dh,a for large primary particle size [12], while it has insignificant effect on the enlargement of dh,a for small primary particle size, which causes the enlargement of dh,a with respect to dp to be increased with the rise of oil concentration. 3.3. Effect of primary particle concentration on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 10 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures for different primary particle concentrations and oil concentrations. The experimental con-
700 R141b/TiO2 (Peng et al., 2015): ω = 0 wt%
600
R141b/NM56/TiO2 (present study): ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
500 400 300 200 100
c0 = 250 mg L-1, T = 20 oC
ditions include the primary particle concentrations of 50, 100, 250 and 500 mg L−1, and the oil concentrations of 1, 3 and 5 wt%; the primary particle size and the temperature are fixed at 40 nm and 20 °C, respectively. From Fig. 10, it can be seen that the hydrodynamic diameter (dh) of aggregates in nanorefrigerant–oil mixture increases with time gradually, and the increasing rate is larger for higher primary particle concentration (c0). This phenomenon is caused by the following reason. With the increase of c0, the interparticle distance decreases at fixed liquid volume, causing more frequent collisions of particles and larger aggregation rate, which leads to the larger increasing rate of dh. Fig. 11 shows the time-averaged hydrodynamic diameters (dh,a) for different primary particle concentrations (c0) and oil concentrations (ω). It can be seen that dh,a is enlarged with the increase of c0, and the enlargement increases with the rise of oil concentration. With the increase of c0 from 50 to 500 mg L−1, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 7.5%, 12.8%, 12.1% and 20.3%, respectively; the enlargement for 5 wt% is 170.7% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the increase of primary particle concentration (c0) is as below. The increase of c0 leads to higher frequency of particle collision and higher aggregation rate, causing the number of nanoparticles contained in an aggregate (n) to be larger. As dh,a can be described by the Eq. (6), the increase of n will result in the enlargement of dh,a. The reason for the phenomenon that the enlargement of dh,a with respect to c0 increases with the rise of oil concentration is as follows: (1) The oil excess layer becomes thicker with the rise of oil concentration. (2) At fixed liquid volume, the increase of c0 will lead to the decrease of interparticle distance. (3) The smaller interparticle distance and the thicker oil film may lead to the formation of “oil bridge” [45,46]. (4) The “oil bridge” links the nanoparticles together, promotes the particle aggregation, and greatly increases dh,a, which causes the enlargement of dh,a with respect to c0 to be increased with the rise of oil concentration.
0 0
20
40
60
80
100
120
dp (nm) Fig. 9. Time-averaged hydrodynamic diameters (dh,a) for different primary particle sizes (dp) and oil concentrations (ω).
3.4. Effect of temperature on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 12 shows the time evolutions of the hydrodynamic diameters of aggregates in oil-free nanorefrigerant and
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
951
Fig. 10. Time evolutions of hydrodynamic diameters (dh) of aggregates in nanorefrigerant–oil mixture for different primary particle concentrations (c0) and oil concentrations (ω).
450 R141b/TiO2 (Peng et al., 2015): ω = 0 wt%
400
R141b/NM56/TiO2 (present study): ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
350
300
250 dp = 40 nm, T = 20 oC
200 0
100
200
300
400
500
600
-1
c0 (mg L ) Fig. 11. Time-averaged hydrodynamic diameters (dh,a) for different primary particle concentrations (c0) and oil concentrations (ω).
nanorefrigerant–oil mixtures. The experimental conditions include the temperatures of 6, 13, 20 and 27 °C, and the oil concentrations of 0, 1, 3 and 5 wt%; the primary particle size and the primary particle concentration are fixed at 40 nm and 250 mg L−1, respectively. From Fig. 12, it can be seen that the hydrodynamic diameter (dh) of aggregates in either oil-free nanorefrigerant or nanorefrigerant– oil mixture increases with time gradually, and the increasing rate is higher for higher temperature (T). This phenomenon is caused by the following reason. The increase of temperature enhances the Brownian motion of nanoparticle, and decreases the liquid viscosity, causing the increase of collision frequency between nanoparticles, which leads to the increase of aggregation rate. Fig. 13 shows the time-averaged hydrodynamic diameters (dh,a) for different temperatures (T) and oil concentrations. It can be seen that dh,a is enlarged with the rise of T, and the enlargement increases with the rise of oil concentration. With the rise of T from 6 to 27 °C, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 14.5%, 15.7%, 20.0% and 22.6%, respectively; the enlargement for 5 wt% is 55.9% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the rise of temperature (T) is as below. The rise of T enhances the Brownian motion of
Fig. 12. Time evolutions of hydrodynamic diameters (dh) of aggregates for different temperatures (T) and oil concentrations (ω).
952
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
3. The lubricating oil enhances the effects of primary particle size, primary particle concentration and temperature on the aggregation of nanoparticles, and the experimental results show that these three effects are respectively enhanced by 94.1%, 170.7% and 55.9% when the oil concentration rises from 0 to 5 wt%.
450 R141b/TiO2:
dp = 40 nm, c0 = 250 mg L-1
ω = 0 wt%
400
R141b/NM56/TiO2: ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
350
Acknowledgements The authors gratefully acknowledge the support from the National Natural Science Foundation of China (Grant No. 51376124).
300 Nomenclature
250
200 0
5
10
15
20
25
30
o
T ( C) Fig. 13. Time-averaged hydrodynamic diameters (dh,a) for different temperatures (T) and oil concentrations (ω).
nanoparticles and then increases the collision frequency between nanoparticles, causing the number of nanoparticles contained in an aggregate (n) to be larger, which leads to the enlargement of dh,a. The reason for the phenomenon that the enlargement of dh,a with respect to T increases with the rise of oil concentration (ω) is as follows: (1) The viscosity of refrigerant–oil mixture decreases with the increase of temperature, and the decrease degree increases with the rise of oil concentration. (2) The decrease of viscosity of refrigerant–oil mixture leads to the increase of collision frequency between nanoparticles, causing the increase of dh,a with respect to T to be increased with the rise of ω. It should be noted that the evaporation or condensation of refrigerant has not occurred under the present experimental conditions. In the actual refrigeration systems, the nanorefrigerant undergoes the evaporation and condensation processes. The effects of evaporation and condensation on the particle aggregation behavior of nanorefrigerant should be studied in the future. 4. Conclusions The aggregation behavior of nanoparticles in nanorefrigerant– oil mixture was studied experimentally by measuring the hydrodynamic diameter of aggregates. The nanoparticles, refrigerant and lubricating oil for experiments were TiO2, R141b (liquid state) and ATMOS NM56, respectively. Experiments were conducted under ambient pressure, with primary particle size from 25 to 100 nm, primary particle concentration from 50 to 500 mg L−1, temperature from 6 to 27 °C, and oil concentrations of 1, 3 and 5 wt%. The following conclusions were obtained from the experiments at above conditions:
A c0 DF dh dh,a dp h hP kB N n R T t UA ve
Hamaker constant Primary particle concentration Fractal dimension Hydrodynamic diameter Time-averaged hydrodynamic diameter Primary particle size Separation distance Planck constant Boltzmann constant Number of time points Number of primary particles contained in an equivalent aggregate Nanoparticle radius Temperature Time van der Waals attractive potential Main electronic absorption frequency in the ultraviolet region
Greek symbols β Collision frequency ε Dielectric permittivity δ Thickness of oil excess layer μ Viscosity ω Oil concentration η Refractive index Subscripts l n n–l–n o r ro
Liquid Nanoparticle Nanoparticle–liquid–nanoparticle Oil Refrigerant Refrigerant–oil mixture
Abbreviation DLS Dynamic light scattering EG Ethylene glycol MWCNTs Multi-walled carbon nanotubes PG Propylene glycol SEM Scanning electron microscope TEM Transmission electron microscope References
1. The presence of lubricating oil inhibits the aggregation of nanoparticles in nanorefrigerant, and the inhibition is strengthened with the rise of oil concentration. 2. The increase of primary particle size, primary particle concentration or temperature enhances the aggregation of nanoparticles in nanorefrigerant–oil mixtures, and the maximal enhancements under the present conditions are 272.1%, 20.3%, and 22.6%, respectively.
[1] O.A. Alawi, N.A.C. Sidik, H.A. Mohammed, A comprehensive review of fundamentals, preparation and applications of nanorefrigerants, Int. Commun. Heat Mass Transf. 54 (2014) 81–95, doi:10.1016/j.icheatmasstransfer .2014.03.001. [2] A. Celen, A. Çebi, M. Aktas, O. Mahian, A.S. Dalkilic, S. Wongwises, A review of nanorefrigerants: flow characteristics and applications, Int. J. Refrig. 44 (2014) 125–140, doi:10.1016/j.ijrefrig.2014.05.009. [3] R. Saidur, S.N. Kazi, M.S. Hossain, M.M. Rahman, H.A. Mohammed, A review on the performance of nanoparticles suspended with refrigerants and
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
lubricating oils in refrigeration systems, Renew. Sustain. Energy Rev. 15 (2011) 310–323, doi:10.1016/j.rser.2010.08.018. S.S. Bi, L. Shi, L. Zhang, Application of nanoparticles in domestic refrigerators, Appl. Therm. Eng. 28 (2008) 1834–1843, doi:10.1016/j.applthermaleng.2007 .11.018. S.S. Bi, K. Guo, Z.G. Liu, J.T. Wu, Performance of a domestic refrigerator using TiO2-R600a nano-refrigerant as working fluid, Energy Convers. Manag. 52 (2011) 733–737, doi:10.1016/j.enconman.2010.07.052. F.S. Javadi, R. Saidur, Energetic, economic and environmental impacts of using nanorefrigerant in domestic refrigerators in Malaysia, Energy Convers. Manag. 73 (2013) 335–339, doi:10.1016/j.enconman.2013.05.013. C.S. Jwo, L.Y. Jeng, T.P. Teng, H. Chang, Effects of nanolubricant on performance of hydrocarbon refrigerant system, J. Vac. Sci. Technol. B Nanometer. Struct. Process. Meas. Phenom. 27 (2009) 1473, doi:10.1116/1.3089373. R.X. Wang, Q.P. Wu, Y.Z. Wu, Use of nanoparticles to make mineral oil lubricants feasible for use in a residential air conditioner employing hydro-fluorocarbons refrigerants, Energy Build. 42 (2010) 2111–2117, doi:10.1016/j.enbuild .2010.06.023. K. Suttiponparnit, J. Jiang, M. Sahu, S. Suvachittanont, T. Charinpanitkul, P. Biswas, Role of surface area, primary particle size, and crystal phase on titanium dioxide nanoparticle dispersion properties, Nanoscale Res. Lett. 6 (2010) 27, doi:10.1007/s11671-010-9772-1. M.J. Pastoriza-Gallego, C. Casanova, J.L. Legido, M.M. Piñeiro, CuO in water nanofluid: influence of particle size and polydispersity on volumetric behaviour and viscosity, Fluid Phase Equilib. 300 (2011) 188–196, doi:10.1016/ j.fluid.2010.10.015. X.F. Li, D.S. Zhu, X.J. Wang, Evaluation on dispersion behavior of the aqueous copper nano-suspensions, J. Colloid Interface Sci. 310 (2007) 456–463, doi:10.1016/j.jcis.2007.02.067. J. Lee, K. Han, J. Koo, A novel method to evaluate dispersion stability of nanofluids, Int. J. Heat Mass Transf. 70 (2014) 421–429, doi:10.1016/ j.ijheatmasstransfer.2013.11.029. J. Jiang, G. Oberdörster, P. Biswas, Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies, J. Nanopart. Res. 11 (2008) 77–89, doi:10.1007/s11051-008-9446-4. M.J.L. Ismay, E. Doroodchi, B. Moghtaderi, Effects of colloidal properties on sensible heat transfer in water-based titania nanofluids, Chem. Eng. Res. Des. 91 (2013) 426–436, doi:10.1016/j.cherd.2012.10.005. Y.J. Hwang, J.K. Lee, J.K. Lee, Y.M. Jeong, S. Cheong, Y.C. Ahn, et al., Production and dispersion stability of nanoparticles in nanofluids, Powder Technol. 186 (2008) 145–153, doi:10.1016/j.powtec.2007.11.020. A.I. Gómez-Merino, F.J. Rubio-Hernández, J.F. Velázquez-Navarro, J. Aguiar, C. Jiménez-Agredano, Study of the aggregation state of anatase water nanofluids using rheological and DLS methods, Ceram. Int. 40 (2014) 14045–14050, doi:10.1016/j.ceramint.2014.05.132. P.E. Gharagozloo, K.E. Goodson, Temperature-dependent aggregation and diffusion in nanofluids, Int. J. Heat Mass Transf. 54 (2011) 797–806, doi:10.1016/ j.ijheatmasstransfer.2010.06.058. J.W. Lee, Y.T. Kang, CO2 absorption enhancement by Al2O3 nanoparticles in NaCl aqueous solution, Energy 53 (2013) 206–211, doi:10.1016/j.energy.2013.02 .047. S. Witharana, I. Palabiyik, Z. Musina, Y. Ding, Stability of glycol nanofluids – the theory and experiment, Powder Technol. 239 (2013) 72–77, doi:10.1016/ j.powtec.2013.01.039. K.S. Hong, T.K. Hong, H.S. Yang, Thermal conductivity of Fe nanofluids depending on the cluster size of nanoparticles, Appl. Phys. Lett. 88 (2006) 1–3, doi:10.1063/ 1.2166199. M. Kole, T.K. Dey, Effect of aggregation on the viscosity of copper oxide-gear oil nanofluids, Int. J. Therm. Sci. 50 (2011) 1741–1747, doi:10.1016/ j.ijthermalsci.2011.03.027. S.S. Bi, L. Shi, L.L. Wang, Dispersion behavior of TiO2 nanoparticles in refrigerant, Chin. J. Process Eng. 7 (2007) 541–545 (in Chinese). H. Peng, L.N. Lin, G.L. Ding, Influences of primary particle parameters and surfactant on aggregation behavior of nanoparticles in nanorefrigerant, Energy 89 (2015) 410–420, doi:10.1016/j.energy.2015.05.116. L. Lin, H. Peng, G. Ding, Dispersion stability of multi-walled carbon nanotubes in refrigerant with addition of surfactant, Appl. Therm. Eng. 91 (2015) 163–171, doi:10.1016/j.applthermaleng.2015.08.011. R.X. Wang, B. Hao, G.Z. Xie, A refrigerating system using HFC134a and mineral lubricant appended with n-TiO2(R) as working fluids, in: Proceedings of 4th
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] [42] [43]
[44]
[45]
[46]
953
International Symposium HAVC, Tsinghua University Press, Beijing, China, 2003, pp. 888–892. R. Krishna Sabareesh, N. Gobinath, V. Sajith, S. Das, C.B. Sobhan, Application of TiO2 nanoparticles as a lubricant-additive for vapor compression refrigeration systems – an experimental investigation, Int. J. Refrig. 35 (2012) 1989–1996, doi:10.1016/j.ijrefrig.2012.07.002. V.M.V. Padmanabhan, S. Palanisamy, The use of TiO2 nanoparticles to reduce refrigerator irreversibility, Energy Convers. Manag. 59 (2012) 122–132, doi:10.1016/j.enconman.2012.03.002. V. Trisaksri, S. Wongwises, Nucleate pool boiling heat transfer of TiO2–R141b nanofluids, Int. J. Heat Mass Transf. 52 (2009) 1582–1588, doi:10.1016/ j.ijheatmasstransfer.2008.07.041. P. Naphon, C. Thongjing, Pool boiling heat transfer characteristics of refrigerantnanoparticle mixtures, Int. Commun. Heat Mass Transf. 52 (2014) 84–89, doi:10.1016/j.icheatmasstransfer.2014.01.014. I.M. Mahbubul, R. Saidur, M.A. Amalina, Influence of particle concentration and temperature on thermal conductivity and viscosity of Al2O3/R141b nanorefrigerant, Int. Commun. Heat Mass Transf. 43 (2013) 100–104, doi:10.1016/j.icheatmasstransfer.2013.02.004. X. Tang, Y.H. Zhao, Y.H. Diao, Experimental investigation of the nucleate pool boiling heat transfer characteristics of δ-Al2O3-R141b nanofluids on a horizontal plate, Exp. Therm. Fluid Sci. 52 (2014) 88–96, doi:10.1016/j.expthermflusci.2013 .08.025. H.H. Liu, S. Surawanvijit, R. Rallo, G. Orkoulas, Y. Cohen, Analysis of nanoparticle agglomeration in aqueous suspensions via constant-number Monte Carlo simulation, Environ. Sci. Technol. 45 (2011) 9284–9292, doi:10.1021/ es202134p. L.N. Lin, H. Peng, G.L. Ding, Influence of oil concentration on wetting behavior during evaporation of refrigerant-oil mixture on copper surface, Int. J. Refrig. 61 (2015) 23–26, doi:10.1016/j.ijrefrig.2015.09.001. J.N. Israelachvili, Van der Waals Forces between particles and surfaces, in: J.N. Israelachvili (Ed.), Intermolecular and Surface Forces, third ed., Elsevier, 2011, pp. 253–289, doi:10.1016/B978-0-12-375182 -9.10013-2. P.C. Sedrez, J.R. Barbosa, Relative permittivity of mixtures of R-134a and R-1234yf and a polyol ester lubricating oil, Int. J. Refrig. 49 (2015) 141–150, doi:10.1016/j.ijrefrig.2014.09.019. J. Glin´ski, Determination of the conditional association constants from the sound velocity data in binary liquid mixtures, J. Chem. Phys. 118 (2003) 2301, doi:10.1063/1.1534579. M.T. Barão, N. de Castro, U.V. Mardolcar, Molecular properties of alternative refrigerants derived from dielectric-constant measurements, Int. J. Thermophys. 18 (1997) 419–438, doi:10.1007/BF02575172. H.B. Chae, J.W. Schmidt, M.R. Moldover, Alternative refrigerants R123a, R134, R141b, R142b, and R152a: critical temperature, refractive index, surface tension, and estimates of liquid, vapor, and critical densities, J. Phys. Chem. 94 (1990) 8840–8845, doi:10.1021/j100388a018. R.J. Liao, J. Hao, G. Chen, Z.Q. Ma, L.J. Yang, A comparative study of physicochemical, dielectric and thermal properties of pressboard insulation impregnated with natural ester and mineral oil, IEEE Trans. Dielectr. Electr. Insul. 18 (2011) 1626–1637, doi:10.1109/TDEI.2011.6032833. F. Haus, J. German, G.-A. Junter, Primary biodegradability of mineral base oils in relation to their chemical and physical characteristics, Chemosphere 45 (2001) 983–990, doi:10.1016/S0045-6535(01)00027-3. W.M. Haynes (Ed.), CRC Handbook of Chemistry and Physics, CRC Press, 2014. J. Yamamoto, G. Yuji, K. Seigo, Wear characteristics of aluminum-silicon sliding against steel in HFC134a, Lubr. Eng. 56 (2000) 32–40. F. Gambinossi, S.E. Mylon, J.K. Ferri, Aggregation kinetics and colloidal stability of functionalized nanoparticles, Adv. Colloid Interface Sci. 222 (2015) 332–349, doi:10.1016/j.cis.2014.07.015. Y.T. He, J. Wan, T. Tokunaga, Kinetic stability of hematite nanoparticles: the effect of particle sizes, J. Nanopart. Res. 10 (2007) 321–332, doi:10.1007/s11051-0079255-1. N.D. Denkov, Mechanisms of action of mixed solid–liquid antifoams. 2. Stability of oil bridges in foam films, Langmuir 15 (1999) 8530–8542, doi:10.1021/ la990214y. N. Takenaka, S. Iwamoto, T. Yabe, R. Yamauchi, K. Ogata, K. Kato, Microscopic observation and characterization of the oil bridge between dehulled-roasted sesame seeds, Colloids Surf. B Biointerfaces 55 (2007) 131–137, doi:10.1016/ j.colsurfb.2006.11.014.
Contents lists available at ScienceDirect
Applied Thermal Engineering j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a p t h e r m e n g
Research Paper
Experimental research on particle aggregation behavior in nanorefrigerant–oil mixture Lingnan Lin a, Hao Peng b, Guoliang Ding a,* a b
Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200240, China Department of Energy and Power Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China
H I G H L I G H T S
• • • • •
Time evolutions of the hydrodynamic diameter of aggregates in nanorefrigerant–oil mixture were measured. The presence of lubricating oil inhibits the aggregation of nanoparticles. The inhibition of oil on aggregation is strengthened with the increase of oil concentration. Larger primary particle size or concentration leads to larger hydrodynamic diameter of aggregate. The hydrodynamic diameter of aggregate is enlarged with the rise of temperature.
A R T I C L E
I N F O
Article history: Received 10 October 2015 Accepted 11 December 2015 Available online 14 January 2016 Keywords: Aggregation Nanorefrigerant Oil Size Temperature
A B S T R A C T
The objective of this study is to experimentally investigate the effects of primary particle size, primary particle concentration and temperature on particle aggregation behavior in nanorefrigerant–oil mixture. The nanoparticles, refrigerant and lubricating oil for experiments were TiO2, R141b and ATMOS NM56, respectively. Experimental conditions included primary particle size from 25 to 100 nm, primary particle concentration from 50 to 500 mg L−1, temperature from 6 to 27 °C, and oil concentrations of 1, 3 and 5 wt%. Time evolutions of the hydrodynamic diameter of aggregates were measured by dynamic light scattering (DLS) method. It is shown that the presence of lubricating oil inhibits the particle aggregation, and the inhibition is strengthened with the increase of oil concentration; the hydrodynamic diameter of aggregates is enlarged with the increase of primary particle size, primary particle concentration or temperature, and the enlargement increases with the increase of oil concentration. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Nanorefrigerant–oil mixture is a special kind of nanofluid which is prepared by dispersing nanoparticles in refrigerant–oil mixture. As the addition of nanoparticles remarkably increases the thermal conductivity and the boiling heat transfer coefficient, nanorefrigerant–oil mixtures have better heat exchange performance compared to conventional refrigerant–oil mixtures [1–3], and have great potential for improving the energy efficiencies of vapor compression refrigeration systems [4–8]. However, the nanoparticles in refrigerant–oil mixture are likely to form aggregates due to strong Brownian motion and high surface free energy. The aggregation of nanoparticles accelerates the sedimentation process and then reduces the long-term stability, which will hinder the actual application of nanorefrigerant–oil mixtures. In order to develop the technology for maintaining the long-term stability of nanorefrigerant–oil
* Corresponding author. Tel.: +86 21 34206378; fax: +86 21 34206814. E-mail address: [email protected] (G. Ding). http://dx.doi.org/10.1016/j.applthermaleng.2015.12.052 1359-4311/© 2015 Elsevier Ltd. All rights reserved.
mixture, the particle aggregation behavior in nanorefrigerant–oil mixture should be known. The effects of primary particle parameters (size and concentration) and temperature should be considered during the investigation on the particle aggregation behavior in nanorefrigerant–oil mixture due to the following reasons. Firstly, the primary particle parameters (size and concentration) affect the interaction between particles. Secondly, the variation of temperature alters the thermophysical properties of refrigerant–oil mixture as well as Brownian motion of nanoparticles, which leads to the change of particle aggregation behavior. The existing researches on particle aggregation behavior focus on the nanofluids with the base fluids of water [9–17], aqueous solution [9,13,18], ethylene glycol [12,19,20], silicon oil [15], gear oil [21] and refrigerant [22–24], as summarized in Table 1. The experimental methods for evaluating the particle aggregation behavior include dynamic light scattering (DLS) [9,11–24], zeta potential analysis [9,11,13–15,19,23], UV–vis spectrophotometer [10,24], transmittance measurement [22], sedimentation picture capturing [11,14,19,22,24], hydrometer [12], scanning electron microscope
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
945
Table 1 Existing studies on particle aggregation behavior in nanofluids. Researchers (year)
Base fluid
Nanoparticle
Evaluation method
Investigated factors
Ref.
Hong et al. (2006) Bi et al. (2007)
EG R113, R141b, R123
Fe TiO2
Time elapse Temperature, time elapse
[20] [22]
Li et al. (2007)
Water
Cu
Water, silicon oil Water
Carbon black, Ag TiO2 (anatase)
Suttiponparnit et al. (2010)
Water
TiO2 (0–100% anatase), TiO2 (P25)
DLS, zeta potential analysis
Pastoriza-Gallego et al. (2011)
Water
CuO
Kole et al. (2011) Gharagozloo and Goodson (2011)
Gear oil Water
CuO Al2O3
UV–vis spectrophotometer, TEM, SEM, DLS DLS
Ismay et al. (2012)
Water
TiO2 (rutile)
pH, surfactant type and concentration Preparing method, pH pH, ionic strength, surface charge and surface coating Primary particle size, pH, ionic strength, surface area, crystal phase Primary particle concentration, pH, time elapse Primary particle concentration Primary particle size and concentration, temperature, time elapse pH, sonication time
[11]
Hwang et al. (2008) Jiang et al. (2009)
DLS DLS, sedimentation picture capturing, transmittance measurement DLS, zeta potential analysis, sedimentation picture capturing DLS, zeta potential analysis, TEM DLS, zeta potential analysis
Lee et al. (2013)
NaCl aqueous solution
Al2O3
[18]
Witharana et al. (2013)
PG, EG, water/PG, water/EG
TiO2, Al2O3, ZnO
Primary particle concentration, temperature, ultrasonication time Base fluid type, day light
Lee et al. (2014) Gomez-Merino et al. (2014) Peng et al. (2015)
Water, EG Water R141b
Al2O3, Ag, ZnO TiO2 (anatase) TiO2 (anatase)
[12] [16] [23]
Lin et al. (2015)
R141b
MWCNTs
Type of nanofluids, time elapse Time elapse Primary particle size and concentration surfactant, time elapse Surfactant type and concentration, ultrasonication time, time elapse, primary particle concentration
(SEM) [10,18] and transmission electron microscope (TEM) [10,14–16]. The results indicated that the aggregation behavior are affected by some factors such as primary particle size and concentration, temperature, base fluid type, time elapse, pH value, surfactant type and concentration, ionic strength, and ultrasonication time. The particle aggregation behavior in nanorefrigerant-oil mixture may be different from those in the oil-free nanorefrigerant and other non-refrigerant-based nanofluids revealed by the above existing researches, based on the following two reasons. Firstly, the interaction between the oil molecules and the refrigerant molecules could change the particle aggregation behavior. Secondly, the physicochemical properties (viscosity, Hamaker constant, etc.) of refrigerant– oil mixture are different from the base fluids in the existing researches, causing different motion of nanoparticles and interaction between nanoparticles, which leads to the change of aggregation behavior. In order to know the particle aggregation behavior in nanorefrigerant–oil mixture, the experiments using refrigerant– oil mixture as base fluid are carried out in the present study, considering the effects of primary particle parameters size, primary particle concentration and temperature.
DLS, zeta potential analysis, sedimentation picture capturing, TEM Sedimentation picture capturing, DLS, SEM DLS, zeta potential analysis, sedimentation picture capturing DLS, hydrometer DLS, TEM DLS, zeta potential
Sedimentation picture capturing, UV–vis spectrophotometer, DLS
[15] [13] [9]
[10] [21] [17]
[14]
[19]
[24]
2. Experimental materials and method 2.1. Preparation of nanorefrigerant–oil mixture The materials used for preparing the nanorefrigerant–oil mixtures include nanoparticles, pure refrigerant and lubricating oil. Titania (TiO2) nanoparticles with primary particle sizes of 25, 40, 60 and 100 nm (purchased from Aladdin Industrial Corporation, Shanghai, China) were used, which are in accordance with those used by Peng et al. [23]. Titania nanoparticles could enhance the heat transfer performance of refrigerant and have stable chemical and physical properties, so they have been widely used in nanorefrigerant researches [5,22,25–28]. The properties of the TiO2 nanoparticles are listed in Table 2. R141b (1,1–Dichloro–1–fluoroethane) was used as the base refrigerant, based on the following reasons. As the nanorefrigerant is usually prepared by two-step method under ambient condition [2,3], it is better to choose those refrigerants in liquid state under ambient condition. However, the common refrigerants such as R410A and R134a are in vapor state under ambient condition, so the experimental nanorefrigerant cannot be prepared by these refrigerants.
Table 2 Properties of TiO2 nanoparticle. Primary particle size (nm)
Specific surface area (m2 g−1)
Molecular weight (g mol−1)
Density (kg m3)
Purity (%)
Crystal phase
Appearance
25 40 60 100
250–400 100–200 30–60 15–20
79.9
3780
100
Anatase
White powder
946
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Table 3 Properties of R141b at different temperature under ambient pressure. Properties
At 6 °C/13 °C/20 °C/27 °C
Chemical formula Molecular weight (g mol−1) Boiling point (°C) Critical temperature (°C) Critical pressure (Mpa) Liquid phase density (kg m3) (1 atm) Liquid phase viscosity (μPa s) (1 atm)
C2H3Cl2F 116.95 32.05 204.5 4.25 1271.7/1254.8/1245.3/1235.8 517.43/463.63/437.11/412.64
Among the refrigerants in liquid state under ambient condition, R141b is still in use and the physicochemical properties are close to those of the common refrigerants, so it is widely used as the base fluid in nanorefrigerant researches [23,24,28–31]. The R141b was purchased from Zhejiang Juhua Co., Ltd., China, and its properties at different temperature under ambient pressure are listed in Table 3. The lubricating oil ATMOS NM56 was used in preparing the nanorefrigerant–oil mixture, which is a mineral oil with kinematic viscosity of 56.2 mm2 s−1 at 40 °C and density of 0.928 g cm−3 at 15 °C, as reported by the manufacturer. The oil concentrations (ω) include 1, 3 and 5 wt%, covering the oil concentrations used in the actual refrigeration systems. The lubricating oil NM56 is completely miscible with R141b over the entire range of oil concentration and temperature under present experimental conditions. The nanorefrigerant–oil mixture (TiO2/R141b/NM56) was prepared by the following steps: (1) weighing the required mass of TiO2 nanoparticles, lubricating oil NM56 and refrigerant R141b, respectively, via a digital electronic balance with measurement range of 10–210 mg and maximum error of 0.1 mg; (2) mixing the weighed TiO2 nanoparticles, lubricating oil NM56 and refrigerant R141b together to form the nanorefrigerant–oil mixture; (3) vibrating the nanorefrigerant–oil mixture by an ultrasonic processor with output power of 110 W and amplitude of 40% for 20 min to fully disperse the nanoparticles. As the R141b is volatile under ambient condition, it was compensated every 5 min during ultrasonication process to keep the primary particle concentration unchanged. The photograph of the prepared nanorefrigerant–oil mixtures (TiO2/R141b/NM56) with different oil concentrations just after ultrasonication are shown in Fig. 1, and it can be seen that the nanoparticles are all well dispersed in the nanorefrigerant–oil mixtures. The experiments on the oil-free nanorefrigerant (TiO2/R141b) were also performed in the present study. The preparation procedure of oil-free nanorefrigerant (TiO2/R141b) is identical with that of nanorefrigerant–oil mixture (TiO2/R141b/NM56) except for not adding lubricating oil. 2.2. Hydrodynamic diameter measurement The particle aggregation behavior is quantitatively evaluated by the time evolution of hydrodynamic diameter of aggregates, which is measured based on the dynamic light scattering (DLS) technique. The principle of the hydrodynamic diameter measurement by DLS could be found in Ref. [23]. The DLS apparatus used is Particle Sizer NICOMP™ 380 ZLS (PSS–NICOMP Inc., California, USA), which consists of a He–Ne laser generator, an optical lens, a sample cell, a detector, a signal processor, a Peltier thermoelectric element, a chamber and a computer, as shown in Fig. 2. The He–Ne laser generator (35 mW) is used to emit a laser beam with wavelength of 639 nm. The optical lens is used to focus the laser beam on the sample cell. A glass, cubic sample cell (inner size: 10 mm × 10 mm × 45 mm) with a lid on its top is used to contain the sample of nanorefrigerant. The particles of the nanorefrigerant in the sample cell are illuminated by the laser beam
Fig. 1. Photograph of the prepared nanorefrigerant–oil mixtures (TiO2/R141b/ NM56, dp = 40 nm, c0 = 250 mg L−1, T = 20 °C) with different oil concentration just after ultrasonication: (a) 1 wt%; (b) 3 wt%; (c) 5 wt%.
and scatter the light. The detector receives the signals of the scattered light, and transfers them to the signal processor. The signal processor converts the signal to data, and send the data to the computer. The Peltier thermoelectric element is used to monitor the temperature of the sample cell and heat/cool the sample cell, and it is fixed on the wall of the chamber for putting sample cell, as shown in Fig. 3. The chamber is well thermal insulated so that the temperature gradient of the liquid sample in the cell can be neglected. In order to investigate the effect of temperature on particle aggregation behavior, the temperature of nanorefrigerant sample should be controlled. The principle of temperature control is described as follows. Before DLS measurement, a temperature value is set in the software of computer. During measurement, the Peltier thermoelectric element constantly monitors the temperature of the sample cell, and transmits the temperature signals to the computer through feedback circuit. If the temperature is below (or above) the set value, the computer will command the Peltier thermoelectric element to heat (or cool) the sample cell through control circuit, until the set value is reached. The control range of temperature is 4–60 °C with accuracy of ±0.2 °C. The hydrodynamic diameter measurement for each condition is repeated for 5 times to calculate the average value and standard deviation of hydrodynamic diameter. 2.3. Experimental conditions The experimental conditions are divided into three categories, in which the first category deals with the effect of primary particle size, the second category deals with the effect of primary particle concentration, and the third category deals with the effect of temperature. Detailed experimental conditions are listed in Table 4. In the first category, four different primary particle sizes including 25, 40, 60 and 100 nm are used to evaluate the effect of primary particle size on particle aggregation behavior. In the second category, four different primary particle concentrations including 50, 100, 250 and 500 mg L−1 are used to evaluate the effect of primary particle concentration on particle aggregation behavior.
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
947
Fig. 2. Schematic diagram of the DLS apparatus.
In the third category, four different temperatures including 6, 13, 20 and 27 °C are used to evaluate the effect of temperature on particle aggregation behavior. The temperature range was chosen to meet two requirements, i.e., within the measurement range of DLS
apparatus (4 ~ 60 °C), and lower than the boiling point of refrigerant R141b at 1 atm (32.1 °C). For each category, four different oil concentrations including 0, 1, 3, 5 wt% are used to evaluate the influence of oil concentration. The experimental data for oil concentrations of 1, 3, 5 wt% are obtained from the present study. The experimental data for oil concentrations of 0 wt% (i.e., oil-free nanorefrigerant) are obtained from Peng et al. [23] (categories 1 and 2) or the present study (category 3).
3. Results and discussion 3.1. Particle aggregation behavior in nanorefrigerant–oil mixture
Fig. 3. Photograph of the chamber for putting sample cell and the Peltier thermoelectric element.
Fig. 4 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% at primary particle size of 40 nm, primary particle concentration of 250 mg L−1 and temperature of 20 °C. It can be seen that with the increase of time from 0 to 4 h, the hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are enlarged by 26.9%, 26.0% and 23.5%, respectively.
Table 4 Experimental conditions for analysis. Objective
Nano-particles
Base fluid
Oil concentration (wt%)
Primary particle size (nm)
Primary particle concentration (mg L−1)
Temperature (°C)
Time range (h)
Data source
1. Evaluate the effects of primary particle size 2. Evaluate the effect of primary particle concentration 3. Evaluate the effect of temperature
TiO2
R141b R141b/NM56 R141b R141b/NM56
0 1, 3, 5 0 1, 3, 5
25, 40, 60, 100 25, 40, 60, 100 40 40
250 250 50, 100, 250, 500 50, 100, 250, 500
20 20 20 20
0–4 0–4 0–4 0–4
Peng et al. (2015) [23] Present study Peng et al. (2015) [23] Present study
R141b R141b R141b/NM56
0 0 1, 3, 5
40 40 40
250 250 250
20 6, 13, 27 6, 13, 20, 27
0–4 0–4 0–4
Peng et al. (2015) [23] Present study Present study
TiO2
TiO2
948
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
eters (dh,a) in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are smaller by 19.7%, 23.5% and 31.7% than that in oil-free nanorefrigerant, respectively, indicating that the presence of lubricating oil inhibits the particle aggregation, and the inhibition is strengthened with the increase of oil concentration. The inhibition of lubricating oil on particle aggregation could be explained in the following three aspects.
350 ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh (nm)
300
3.1.1. Effect of oil on collision frequency of nanoparticles The aggregation rate is positively correlated to the collision frequency of two interacting particles (βij), and βij can be expressed as the following equation [32].
250
βij = 200 R141b/NM56/TiO2 dp = 40 nm, c0 = 250 mg L-1, T = 20 oC
150 0
1
2
3
4
t (h) Fig. 4. Time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant– oil mixture with different oil concentrations.
In order to compare the extent of aggregation under different conditions, and given that the hydrodynamic diameter is timedependent, the time-averaged hydrodynamic diameter (dh,a) is defined as Eq. (1).
d h,a =
1 N ∑ d h (t i ) N i =1
(1)
where ti is the time point, N is the total number of time points. Fig. 5 shows the time-averaged hydrodynamic diameters (dh,a) in nanorefrigerant–oil mixtures with oil concentrations (ω) of 1, 3 and 5 wt% and oil-free nanorefrigerant (ω = 0 wt%) at primary particle size of 40 nm, primary particle concentration of 250 mg L−1 and temperature of 20 °C. The time-averaged hydrodynamic diam-
⎛ 1 1⎞ 2k BT (Ri + R j ) ⎜ + ⎟ 3μ ⎝ Ri R j ⎠
(2)
where kB is the Boltzmann constant (kB = 1.38 × 10−27 J K−1), T is the temperature, μ is the liquid viscosity, Ri and Rj are the effective radii of two interacting particles, and the term of “kBT/μ” reflects the contribution of Brownian motion in the collision frequency. The collision frequency in nanorefrigerant–oil mixtures with oil concentrations of 1, 3 and 5 wt% are smaller by 10.2%, 23.5% and 25.4% than that in oil-free nanorefrigerant, respectively, which is calculated by Eq. (2) using the viscosities of refrigerant–oil mixtures presented in Ref. [33], and the calculation conditions include dp = 40 nm, c0 = 250 mg L−1 and T = 20 °C. The reduction of collision frequency is caused by the increase of liquid viscosity with the addition of oil. The reduction degree of collision frequency caused by the presence of oil is less than that of time-averaged hydrodynamic diameters (dh,a) shown in Fig. 5, which indicates that the effect of oil on the collision frequency is the main reason accounting for the inhibition of aggregation, but not the only one. 3.1.2. Effect of oil on Hamaker constant The aggregation rate is also positively correlated to Hamaker constant of nanoparticle–liquid–nanoparticle in nanorefrigerant–oil mixture (An–ro–n), and An–ro–n can be calculated by Eq. (3) [34–36].
An-ro-n =
(
An − Aro
)
2
(3-a)
2 3 3h v (ηn2 − 1) ⎛ ε − 1⎞ k BT ⎜ n ⎟ + P e 32 ⎝ ⎠ 4 εn + 1 16 2 (ηn2 + 1) 2
An =
400
(3-b)
R141b/TiO2 (Peng et al., 2015) R141b/NM56/TiO2 (present study)
2 2 − 1) 3 3h v (ηro ⎛ ε − 1⎞ k BT ⎜ ro ⎟ + P e 32 2 ⎝ ε ro + 1⎠ 4 16 2 (ηro + 1) 2
Aro =
350
dh,a (nm)
ε ro = (1 − ω ) ε r + ωε o ηro =
300
250 dp = 40 nm, c0 = 250 mg L-1, T = 20 oC
200 -1
0
1
2
3
4
5
6
ω (wt%) Fig. 5. Time-averaged hydrodynamic diameters (dh,a) for different oil concentrations (ω).
ηrηo
(1 − ω )ηo + ωηr
(3-c)
(3-d)
(3-e)
where An and Aro are the Hamaker constants of nanoparticle and refrigerant–oil mixture, respectively; hP is the Planck constant (6.626 × 10−34 J s); νe is the main electronic absorption frequency in the ultraviolet region (about 3 × 1015 s−1); εn, εr, εo, εro are the dielectric permittivities of nanoparticle, pure refrigerant, oil and refrigerant–oil mixture, respectively; ηn, ηr, ηo, ηro are the refractive indexes of nanoparticle, pure refrigerant, oil and refrigerant– oil mixture, respectively; and the values of εn, εr, εo, ηn, ηr and ηo are listed in Table 5. The Hamaker constant of nanoparticle–liquid–nanoparticle in nanorefrigerant–oil mixtures (An–ro–n) with oil concentrations of 1, 3 and 5 wt% are smaller by 0.2%, 0.5% and 0.9% than that in
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Table 5 Dielectric permittivities and refractive indexes of pure refrigerant, oil and nanoparticle. Meaning
Value
Ref.
εr ηr εo ηo εn ηn
Dielectric permittivity of R141b Refractive index of R14b Dielectric permittivity of lubricating oil Refractive index of lubricating oil Dielectric permittivity of TiO2 Refractive index of TiO2
8.13 1.37 2.5 1.45 86 2.5
[37] [38] [39] [40] [41] [41]
oil-free nanorefrigerant, respectively, which is calculated by Eq. (3) and the calculation conditions include dp = 40 nm, c0 = 250 mg L−1 and T = 20 °C. This result indicates that the effect of oil on Hamaker constant is another reason accounting for the inhibition of aggregation, but the contribution of oil effect on Hamaker constant is much weaker than that of oil effect on collision frequency. 3.1.3. Effect of oil excess layer on van der Waals potential energy Lubricating oil has stronger adhesion on solid surface compared to refrigerant [42], causing the possible existence of oil excess layer around nanoparticles in nanorefrigerant–oil mixture, as shown in Fig. 6. The variation of oil excess layer thickness (δ) will change the van der Waals attractive potential between nanoparticles (UA), and the relationship between δ and UA is as the following equation [43]. 2 R +δ 2 R ⎡ ⎤ Aro − A o + A o − An h h + 2δ ⎥ 1⎢ ⎥ UA = − ⎢ R (R + δ ) ⎥ 12 ⎢ A o − An +4 Aro − A o ⎢⎣ (h + δ ) (2R + δ ) ⎥⎦
(
(
)
(
)(
)
)
(4)
where h is separation distance. The variation of van der Waals potential energy (|UA|) with the oil excess layer thickness (δ) is shown in Fig. 7, in which the results are calculated by Eq. (4) at the condition of R = 20 nm, Ar-o = 4.42 × 10−20 J, Ao = 5.92 × 10−20 J, An = 3.75 × 10−19 J. The calculation results show that with the increase of δ from 0 to 5 nm, |UA| is reduced by 45.9%, 29.6% and 16.9% at h = 10, 20 and 40 nm, respectively. The increase of oil concentration enlarges the oil excess layer thickness (δ), causing the reduction of van der Waals potential energy (|UA|), which inhibits the particle aggregation. 3.2. Effect of primary particle size on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 8 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures for different primary particle sizes and oil concentrations. The experimental conditions
3.0 h = 10 nm h = 20 nm h = 40 nm
2.5
2.0
|UA| (×10-20 J)
Symbol
949
1.5
1.0
0.5
0.0 0
1
2
3
4
5
δ (nm) Fig. 7. Variation of van der Waals potential energy (|UA|) with the thickness of oil excess layer (δ).
include the primary particle sizes of 25, 40, 60 and 100 nm, and the oil concentrations of 1, 3 and 5 wt%; the primary particle concentration and the temperature are fixed at 250 mg L−1 and 20 °C, respectively. From Fig. 8, it can be seen that for the nanorefrigerant–oil mixtures with primary particle sizes (dp) of 25, 40 and 60 nm, the hydrodynamic diameter (dh) increases with time, and the increasing rate is decreased as dp increases; for the nanorefrigerant–oil mixture with dp of 100 nm, dh decreases with time. This phenomenon could be explained as follows: (1) For dp of 25, 40 and 60 nm, dh is determined by the aggregation process of nanoparticles, which leads to the increase of dh with time. With the increase of dp, the interaction energy barrier increases and the surface free energy decreases, causing lower aggregation rate, which leads to the decrease of the increasing rate of dh [44]. (2) For dp of 100 nm, the aggregates are large enough to make the sedimentation process obvious, which leads to the decrease of dh with time. Fig. 9 shows the time-averaged hydrodynamic diameters (dh,a) for different primary particle sizes (dp) and oil concentrations. It can be seen that dh,a is enlarged with the increase of dp, and the enlargement increases with the rise of oil concentration. With the increase of dp from 25 to 100 nm, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 140.2%, 208.3%, 229.8% and 272.1%, respectively; the enlargement for 5 wt% is 94.1% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the increase of primary particle size (dp) is as follows. The hydrodynamic diameter of aggregates at ti (i = 1, 2, …, N) can be described by the following equation
d h (t i ) = d p ⋅ n (t i )
1 DF
(5)
where n is number of primary particles contained in an equivalent aggregate; DF is fractal dimension, which is a constant. By substituting the Eq. (1) into Eq. (5), dh,a can be described as
d h,a =
Fig. 6. Schematic diagram of oil excess layer.
dp N 1D ∑ n (t i ) F N i =1
(6)
Since the variation of dp has less impact on n as well as DF, dh,a increases with the increase of dp.
950
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
Fig. 8. Time evolutions of hydrodynamic diameters (dh) of aggregates in nanorefrigerant–oil mixture for different primary particle sizes (dp) and oil concentrations (ω).
The reason for the phenomenon that the enlargement of dh,a with respect to dp increases with the rise of oil concentration is as follows: (1) With the rise of oil concentration, the viscosity of refrigerant– oil mixture increases and then the settling velocity decreases, causing the sedimentation of aggregates to be weakened. (2) The inhibition of the sedimentation of aggregates has positive effect on the enlargement of dh,a for large primary particle size [12], while it has insignificant effect on the enlargement of dh,a for small primary particle size, which causes the enlargement of dh,a with respect to dp to be increased with the rise of oil concentration. 3.3. Effect of primary particle concentration on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 10 shows the time evolutions of hydrodynamic diameters of aggregates in nanorefrigerant–oil mixtures for different primary particle concentrations and oil concentrations. The experimental con-
700 R141b/TiO2 (Peng et al., 2015): ω = 0 wt%
600
R141b/NM56/TiO2 (present study): ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
500 400 300 200 100
c0 = 250 mg L-1, T = 20 oC
ditions include the primary particle concentrations of 50, 100, 250 and 500 mg L−1, and the oil concentrations of 1, 3 and 5 wt%; the primary particle size and the temperature are fixed at 40 nm and 20 °C, respectively. From Fig. 10, it can be seen that the hydrodynamic diameter (dh) of aggregates in nanorefrigerant–oil mixture increases with time gradually, and the increasing rate is larger for higher primary particle concentration (c0). This phenomenon is caused by the following reason. With the increase of c0, the interparticle distance decreases at fixed liquid volume, causing more frequent collisions of particles and larger aggregation rate, which leads to the larger increasing rate of dh. Fig. 11 shows the time-averaged hydrodynamic diameters (dh,a) for different primary particle concentrations (c0) and oil concentrations (ω). It can be seen that dh,a is enlarged with the increase of c0, and the enlargement increases with the rise of oil concentration. With the increase of c0 from 50 to 500 mg L−1, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 7.5%, 12.8%, 12.1% and 20.3%, respectively; the enlargement for 5 wt% is 170.7% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the increase of primary particle concentration (c0) is as below. The increase of c0 leads to higher frequency of particle collision and higher aggregation rate, causing the number of nanoparticles contained in an aggregate (n) to be larger. As dh,a can be described by the Eq. (6), the increase of n will result in the enlargement of dh,a. The reason for the phenomenon that the enlargement of dh,a with respect to c0 increases with the rise of oil concentration is as follows: (1) The oil excess layer becomes thicker with the rise of oil concentration. (2) At fixed liquid volume, the increase of c0 will lead to the decrease of interparticle distance. (3) The smaller interparticle distance and the thicker oil film may lead to the formation of “oil bridge” [45,46]. (4) The “oil bridge” links the nanoparticles together, promotes the particle aggregation, and greatly increases dh,a, which causes the enlargement of dh,a with respect to c0 to be increased with the rise of oil concentration.
0 0
20
40
60
80
100
120
dp (nm) Fig. 9. Time-averaged hydrodynamic diameters (dh,a) for different primary particle sizes (dp) and oil concentrations (ω).
3.4. Effect of temperature on particle aggregation behavior in nanorefrigerant–oil mixtures Fig. 12 shows the time evolutions of the hydrodynamic diameters of aggregates in oil-free nanorefrigerant and
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
951
Fig. 10. Time evolutions of hydrodynamic diameters (dh) of aggregates in nanorefrigerant–oil mixture for different primary particle concentrations (c0) and oil concentrations (ω).
450 R141b/TiO2 (Peng et al., 2015): ω = 0 wt%
400
R141b/NM56/TiO2 (present study): ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
350
300
250 dp = 40 nm, T = 20 oC
200 0
100
200
300
400
500
600
-1
c0 (mg L ) Fig. 11. Time-averaged hydrodynamic diameters (dh,a) for different primary particle concentrations (c0) and oil concentrations (ω).
nanorefrigerant–oil mixtures. The experimental conditions include the temperatures of 6, 13, 20 and 27 °C, and the oil concentrations of 0, 1, 3 and 5 wt%; the primary particle size and the primary particle concentration are fixed at 40 nm and 250 mg L−1, respectively. From Fig. 12, it can be seen that the hydrodynamic diameter (dh) of aggregates in either oil-free nanorefrigerant or nanorefrigerant– oil mixture increases with time gradually, and the increasing rate is higher for higher temperature (T). This phenomenon is caused by the following reason. The increase of temperature enhances the Brownian motion of nanoparticle, and decreases the liquid viscosity, causing the increase of collision frequency between nanoparticles, which leads to the increase of aggregation rate. Fig. 13 shows the time-averaged hydrodynamic diameters (dh,a) for different temperatures (T) and oil concentrations. It can be seen that dh,a is enlarged with the rise of T, and the enlargement increases with the rise of oil concentration. With the rise of T from 6 to 27 °C, dh,a for oil concentrations of 0, 1, 3 and 5 wt% are enlarged by 14.5%, 15.7%, 20.0% and 22.6%, respectively; the enlargement for 5 wt% is 55.9% higher than that for 0 wt%. The reason for the phenomenon that time-averaged hydrodynamic diameter (dh,a) is enlarged with the rise of temperature (T) is as below. The rise of T enhances the Brownian motion of
Fig. 12. Time evolutions of hydrodynamic diameters (dh) of aggregates for different temperatures (T) and oil concentrations (ω).
952
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
3. The lubricating oil enhances the effects of primary particle size, primary particle concentration and temperature on the aggregation of nanoparticles, and the experimental results show that these three effects are respectively enhanced by 94.1%, 170.7% and 55.9% when the oil concentration rises from 0 to 5 wt%.
450 R141b/TiO2:
dp = 40 nm, c0 = 250 mg L-1
ω = 0 wt%
400
R141b/NM56/TiO2: ω = 1 wt% ω = 3 wt% ω = 5 wt%
dh,a (nm)
350
Acknowledgements The authors gratefully acknowledge the support from the National Natural Science Foundation of China (Grant No. 51376124).
300 Nomenclature
250
200 0
5
10
15
20
25
30
o
T ( C) Fig. 13. Time-averaged hydrodynamic diameters (dh,a) for different temperatures (T) and oil concentrations (ω).
nanoparticles and then increases the collision frequency between nanoparticles, causing the number of nanoparticles contained in an aggregate (n) to be larger, which leads to the enlargement of dh,a. The reason for the phenomenon that the enlargement of dh,a with respect to T increases with the rise of oil concentration (ω) is as follows: (1) The viscosity of refrigerant–oil mixture decreases with the increase of temperature, and the decrease degree increases with the rise of oil concentration. (2) The decrease of viscosity of refrigerant–oil mixture leads to the increase of collision frequency between nanoparticles, causing the increase of dh,a with respect to T to be increased with the rise of ω. It should be noted that the evaporation or condensation of refrigerant has not occurred under the present experimental conditions. In the actual refrigeration systems, the nanorefrigerant undergoes the evaporation and condensation processes. The effects of evaporation and condensation on the particle aggregation behavior of nanorefrigerant should be studied in the future. 4. Conclusions The aggregation behavior of nanoparticles in nanorefrigerant– oil mixture was studied experimentally by measuring the hydrodynamic diameter of aggregates. The nanoparticles, refrigerant and lubricating oil for experiments were TiO2, R141b (liquid state) and ATMOS NM56, respectively. Experiments were conducted under ambient pressure, with primary particle size from 25 to 100 nm, primary particle concentration from 50 to 500 mg L−1, temperature from 6 to 27 °C, and oil concentrations of 1, 3 and 5 wt%. The following conclusions were obtained from the experiments at above conditions:
A c0 DF dh dh,a dp h hP kB N n R T t UA ve
Hamaker constant Primary particle concentration Fractal dimension Hydrodynamic diameter Time-averaged hydrodynamic diameter Primary particle size Separation distance Planck constant Boltzmann constant Number of time points Number of primary particles contained in an equivalent aggregate Nanoparticle radius Temperature Time van der Waals attractive potential Main electronic absorption frequency in the ultraviolet region
Greek symbols β Collision frequency ε Dielectric permittivity δ Thickness of oil excess layer μ Viscosity ω Oil concentration η Refractive index Subscripts l n n–l–n o r ro
Liquid Nanoparticle Nanoparticle–liquid–nanoparticle Oil Refrigerant Refrigerant–oil mixture
Abbreviation DLS Dynamic light scattering EG Ethylene glycol MWCNTs Multi-walled carbon nanotubes PG Propylene glycol SEM Scanning electron microscope TEM Transmission electron microscope References
1. The presence of lubricating oil inhibits the aggregation of nanoparticles in nanorefrigerant, and the inhibition is strengthened with the rise of oil concentration. 2. The increase of primary particle size, primary particle concentration or temperature enhances the aggregation of nanoparticles in nanorefrigerant–oil mixtures, and the maximal enhancements under the present conditions are 272.1%, 20.3%, and 22.6%, respectively.
[1] O.A. Alawi, N.A.C. Sidik, H.A. Mohammed, A comprehensive review of fundamentals, preparation and applications of nanorefrigerants, Int. Commun. Heat Mass Transf. 54 (2014) 81–95, doi:10.1016/j.icheatmasstransfer .2014.03.001. [2] A. Celen, A. Çebi, M. Aktas, O. Mahian, A.S. Dalkilic, S. Wongwises, A review of nanorefrigerants: flow characteristics and applications, Int. J. Refrig. 44 (2014) 125–140, doi:10.1016/j.ijrefrig.2014.05.009. [3] R. Saidur, S.N. Kazi, M.S. Hossain, M.M. Rahman, H.A. Mohammed, A review on the performance of nanoparticles suspended with refrigerants and
L. Lin et al./Applied Thermal Engineering 98 (2016) 944–953
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
lubricating oils in refrigeration systems, Renew. Sustain. Energy Rev. 15 (2011) 310–323, doi:10.1016/j.rser.2010.08.018. S.S. Bi, L. Shi, L. Zhang, Application of nanoparticles in domestic refrigerators, Appl. Therm. Eng. 28 (2008) 1834–1843, doi:10.1016/j.applthermaleng.2007 .11.018. S.S. Bi, K. Guo, Z.G. Liu, J.T. Wu, Performance of a domestic refrigerator using TiO2-R600a nano-refrigerant as working fluid, Energy Convers. Manag. 52 (2011) 733–737, doi:10.1016/j.enconman.2010.07.052. F.S. Javadi, R. Saidur, Energetic, economic and environmental impacts of using nanorefrigerant in domestic refrigerators in Malaysia, Energy Convers. Manag. 73 (2013) 335–339, doi:10.1016/j.enconman.2013.05.013. C.S. Jwo, L.Y. Jeng, T.P. Teng, H. Chang, Effects of nanolubricant on performance of hydrocarbon refrigerant system, J. Vac. Sci. Technol. B Nanometer. Struct. Process. Meas. Phenom. 27 (2009) 1473, doi:10.1116/1.3089373. R.X. Wang, Q.P. Wu, Y.Z. Wu, Use of nanoparticles to make mineral oil lubricants feasible for use in a residential air conditioner employing hydro-fluorocarbons refrigerants, Energy Build. 42 (2010) 2111–2117, doi:10.1016/j.enbuild .2010.06.023. K. Suttiponparnit, J. Jiang, M. Sahu, S. Suvachittanont, T. Charinpanitkul, P. Biswas, Role of surface area, primary particle size, and crystal phase on titanium dioxide nanoparticle dispersion properties, Nanoscale Res. Lett. 6 (2010) 27, doi:10.1007/s11671-010-9772-1. M.J. Pastoriza-Gallego, C. Casanova, J.L. Legido, M.M. Piñeiro, CuO in water nanofluid: influence of particle size and polydispersity on volumetric behaviour and viscosity, Fluid Phase Equilib. 300 (2011) 188–196, doi:10.1016/ j.fluid.2010.10.015. X.F. Li, D.S. Zhu, X.J. Wang, Evaluation on dispersion behavior of the aqueous copper nano-suspensions, J. Colloid Interface Sci. 310 (2007) 456–463, doi:10.1016/j.jcis.2007.02.067. J. Lee, K. Han, J. Koo, A novel method to evaluate dispersion stability of nanofluids, Int. J. Heat Mass Transf. 70 (2014) 421–429, doi:10.1016/ j.ijheatmasstransfer.2013.11.029. J. Jiang, G. Oberdörster, P. Biswas, Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies, J. Nanopart. Res. 11 (2008) 77–89, doi:10.1007/s11051-008-9446-4. M.J.L. Ismay, E. Doroodchi, B. Moghtaderi, Effects of colloidal properties on sensible heat transfer in water-based titania nanofluids, Chem. Eng. Res. Des. 91 (2013) 426–436, doi:10.1016/j.cherd.2012.10.005. Y.J. Hwang, J.K. Lee, J.K. Lee, Y.M. Jeong, S. Cheong, Y.C. Ahn, et al., Production and dispersion stability of nanoparticles in nanofluids, Powder Technol. 186 (2008) 145–153, doi:10.1016/j.powtec.2007.11.020. A.I. Gómez-Merino, F.J. Rubio-Hernández, J.F. Velázquez-Navarro, J. Aguiar, C. Jiménez-Agredano, Study of the aggregation state of anatase water nanofluids using rheological and DLS methods, Ceram. Int. 40 (2014) 14045–14050, doi:10.1016/j.ceramint.2014.05.132. P.E. Gharagozloo, K.E. Goodson, Temperature-dependent aggregation and diffusion in nanofluids, Int. J. Heat Mass Transf. 54 (2011) 797–806, doi:10.1016/ j.ijheatmasstransfer.2010.06.058. J.W. Lee, Y.T. Kang, CO2 absorption enhancement by Al2O3 nanoparticles in NaCl aqueous solution, Energy 53 (2013) 206–211, doi:10.1016/j.energy.2013.02 .047. S. Witharana, I. Palabiyik, Z. Musina, Y. Ding, Stability of glycol nanofluids – the theory and experiment, Powder Technol. 239 (2013) 72–77, doi:10.1016/ j.powtec.2013.01.039. K.S. Hong, T.K. Hong, H.S. Yang, Thermal conductivity of Fe nanofluids depending on the cluster size of nanoparticles, Appl. Phys. Lett. 88 (2006) 1–3, doi:10.1063/ 1.2166199. M. Kole, T.K. Dey, Effect of aggregation on the viscosity of copper oxide-gear oil nanofluids, Int. J. Therm. Sci. 50 (2011) 1741–1747, doi:10.1016/ j.ijthermalsci.2011.03.027. S.S. Bi, L. Shi, L.L. Wang, Dispersion behavior of TiO2 nanoparticles in refrigerant, Chin. J. Process Eng. 7 (2007) 541–545 (in Chinese). H. Peng, L.N. Lin, G.L. Ding, Influences of primary particle parameters and surfactant on aggregation behavior of nanoparticles in nanorefrigerant, Energy 89 (2015) 410–420, doi:10.1016/j.energy.2015.05.116. L. Lin, H. Peng, G. Ding, Dispersion stability of multi-walled carbon nanotubes in refrigerant with addition of surfactant, Appl. Therm. Eng. 91 (2015) 163–171, doi:10.1016/j.applthermaleng.2015.08.011. R.X. Wang, B. Hao, G.Z. Xie, A refrigerating system using HFC134a and mineral lubricant appended with n-TiO2(R) as working fluids, in: Proceedings of 4th
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41] [42] [43]
[44]
[45]
[46]
953
International Symposium HAVC, Tsinghua University Press, Beijing, China, 2003, pp. 888–892. R. Krishna Sabareesh, N. Gobinath, V. Sajith, S. Das, C.B. Sobhan, Application of TiO2 nanoparticles as a lubricant-additive for vapor compression refrigeration systems – an experimental investigation, Int. J. Refrig. 35 (2012) 1989–1996, doi:10.1016/j.ijrefrig.2012.07.002. V.M.V. Padmanabhan, S. Palanisamy, The use of TiO2 nanoparticles to reduce refrigerator irreversibility, Energy Convers. Manag. 59 (2012) 122–132, doi:10.1016/j.enconman.2012.03.002. V. Trisaksri, S. Wongwises, Nucleate pool boiling heat transfer of TiO2–R141b nanofluids, Int. J. Heat Mass Transf. 52 (2009) 1582–1588, doi:10.1016/ j.ijheatmasstransfer.2008.07.041. P. Naphon, C. Thongjing, Pool boiling heat transfer characteristics of refrigerantnanoparticle mixtures, Int. Commun. Heat Mass Transf. 52 (2014) 84–89, doi:10.1016/j.icheatmasstransfer.2014.01.014. I.M. Mahbubul, R. Saidur, M.A. Amalina, Influence of particle concentration and temperature on thermal conductivity and viscosity of Al2O3/R141b nanorefrigerant, Int. Commun. Heat Mass Transf. 43 (2013) 100–104, doi:10.1016/j.icheatmasstransfer.2013.02.004. X. Tang, Y.H. Zhao, Y.H. Diao, Experimental investigation of the nucleate pool boiling heat transfer characteristics of δ-Al2O3-R141b nanofluids on a horizontal plate, Exp. Therm. Fluid Sci. 52 (2014) 88–96, doi:10.1016/j.expthermflusci.2013 .08.025. H.H. Liu, S. Surawanvijit, R. Rallo, G. Orkoulas, Y. Cohen, Analysis of nanoparticle agglomeration in aqueous suspensions via constant-number Monte Carlo simulation, Environ. Sci. Technol. 45 (2011) 9284–9292, doi:10.1021/ es202134p. L.N. Lin, H. Peng, G.L. Ding, Influence of oil concentration on wetting behavior during evaporation of refrigerant-oil mixture on copper surface, Int. J. Refrig. 61 (2015) 23–26, doi:10.1016/j.ijrefrig.2015.09.001. J.N. Israelachvili, Van der Waals Forces between particles and surfaces, in: J.N. Israelachvili (Ed.), Intermolecular and Surface Forces, third ed., Elsevier, 2011, pp. 253–289, doi:10.1016/B978-0-12-375182 -9.10013-2. P.C. Sedrez, J.R. Barbosa, Relative permittivity of mixtures of R-134a and R-1234yf and a polyol ester lubricating oil, Int. J. Refrig. 49 (2015) 141–150, doi:10.1016/j.ijrefrig.2014.09.019. J. Glin´ski, Determination of the conditional association constants from the sound velocity data in binary liquid mixtures, J. Chem. Phys. 118 (2003) 2301, doi:10.1063/1.1534579. M.T. Barão, N. de Castro, U.V. Mardolcar, Molecular properties of alternative refrigerants derived from dielectric-constant measurements, Int. J. Thermophys. 18 (1997) 419–438, doi:10.1007/BF02575172. H.B. Chae, J.W. Schmidt, M.R. Moldover, Alternative refrigerants R123a, R134, R141b, R142b, and R152a: critical temperature, refractive index, surface tension, and estimates of liquid, vapor, and critical densities, J. Phys. Chem. 94 (1990) 8840–8845, doi:10.1021/j100388a018. R.J. Liao, J. Hao, G. Chen, Z.Q. Ma, L.J. Yang, A comparative study of physicochemical, dielectric and thermal properties of pressboard insulation impregnated with natural ester and mineral oil, IEEE Trans. Dielectr. Electr. Insul. 18 (2011) 1626–1637, doi:10.1109/TDEI.2011.6032833. F. Haus, J. German, G.-A. Junter, Primary biodegradability of mineral base oils in relation to their chemical and physical characteristics, Chemosphere 45 (2001) 983–990, doi:10.1016/S0045-6535(01)00027-3. W.M. Haynes (Ed.), CRC Handbook of Chemistry and Physics, CRC Press, 2014. J. Yamamoto, G. Yuji, K. Seigo, Wear characteristics of aluminum-silicon sliding against steel in HFC134a, Lubr. Eng. 56 (2000) 32–40. F. Gambinossi, S.E. Mylon, J.K. Ferri, Aggregation kinetics and colloidal stability of functionalized nanoparticles, Adv. Colloid Interface Sci. 222 (2015) 332–349, doi:10.1016/j.cis.2014.07.015. Y.T. He, J. Wan, T. Tokunaga, Kinetic stability of hematite nanoparticles: the effect of particle sizes, J. Nanopart. Res. 10 (2007) 321–332, doi:10.1007/s11051-0079255-1. N.D. Denkov, Mechanisms of action of mixed solid–liquid antifoams. 2. Stability of oil bridges in foam films, Langmuir 15 (1999) 8530–8542, doi:10.1021/ la990214y. N. Takenaka, S. Iwamoto, T. Yabe, R. Yamauchi, K. Ogata, K. Kato, Microscopic observation and characterization of the oil bridge between dehulled-roasted sesame seeds, Colloids Surf. B Biointerfaces 55 (2007) 131–137, doi:10.1016/ j.colsurfb.2006.11.014.