nanostructured surface

international journal of refrigeration 62 (2016) 207–221

Available online at www.sciencedirect.com

ScienceDirect 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 / i j r e f r i g

Experimental research on wetting behavior of refrigerant–oil mixture on micro/nanostructured surface Hao Peng a, Lingnan Lin b, Guoliang Ding b,* a

Department of Energy and Power Engineering, Beijing University of Civil Engineering and Architecture, Beijing 100044, China b Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200240, China

A R T I C L E

I N F O

A B S T R A C T

Article history:

The wettability of micro/nanostructured surface is a key property for its application in en-

Received 15 September 2015

hancing the boiling heat transfer of refrigerant–oil mixture. The objective of this research

Received in revised form 28 October

is to experimentally investigate the wetting behavior of refrigerant–oil mixture on micro/

2015

nanostructured surface. Three types of surfaces including plain copper surface (PS), micro/

Accepted 30 October 2015

nanostructured surface (MNS) and micro/nanostructured surface with fluorinated self-

Available online 4 November 2015

assembled monolayer (MNFS) were fabricated; and the wetting behavior of pure R141b as well as R141b-NM56 mixtures with different oil concentrations on three types of surfaces

Keywords:

was measured. The experimental results show that the protuberant liquid film is formed

Micro/nanostructure

during the wetting of refrigerant–oil mixture on MNS or PS, but does not exist on MNFS;

Oil

the presence of F-SAM or micro/nanostructure modified by F-SAM reduces the surface

Refrigerant

wettability, while the presence of micro/nanostructure increases the surface wettability; oil

Self-assembled monolayer

increases the wettability of refrigerant on MNS, while it reduces the wettability of refrig-

Wetting

erant on MNFS. © 2016 Elsevier Ltd and International Institute of Refrigeration. All rights reserved.

Recherche expérimentale sur le comportement de mouillage d’un mélange frigorigène-huile sur une surface micro/nano-structurée Mots clés : Micro/nano structure ; Huile

* Corresponding author. Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai 200240, China. Tel.: +86 21 34206378; Fax: +86 21 34206814. E-mail address: [email protected] (G. Ding). http://dx.doi.org/10.1016/j.ijrefrig.2015.10.033 0140-7007/© 2016 Elsevier Ltd and International Institute of Refrigeration. All rights reserved.

208

international journal of refrigeration 62 (2016) 207–221

Nomenclature EFF EFMN EFMNF EFoil H h L l Ma Ra vCL yCL

F-SAM effect factor micro/nanostructure effect factor micro/nanostructure combined F-SAM effect factor oil effect factor rising liquid height [mm] vertical distance between contact line and horizontal liquid level in CCD image [pixel] liquid film length [mm] vertical distance between contact line and top of meniscus in CCD image [pixel] magnification of image [mm pixel−1] arithmetic mean roughness [μm] contact line velocity [mm s−1] vertical coordinate of contact line in CCD image [pixel]

Greek symbols Δt time interval between two frames [s] θ contact angle [°]

1.

Introduction

Micro/nanostructured surface, formed by fabricating micro/ nanostructures on a conventional surface, has shown great potential for enhancing the boiling heat transfer (Dong et al., 2014; Kim et al., 2015; Kruse et al., 2015; Launay et al., 2006; Shojaeian and Kosar, 2015). Application of this type of surface for enhancing the boiling heat transfer of refrigerant–oil mixture might become a new method for improving the energy efficiency of vapor compression refrigeration systems. In order to realize the application of micro/nanostructured surface in vapor compression refrigeration systems, the influence mechanism of micro/nanostructured surface on the boiling heat transfer of refrigerant–oil mixture should be known. A micro/nanostructured surface affects boiling heat transfer performance through changing the heating surface properties. The surface wettability is an important surface property (Attinger et al., 2013, 2014), and has significant influences on active nucleation site density and bubble departure frequency (Gong and Cheng, 2015; Jo et al., 2011; Li et al., 2013, 2014). In order to understand the influence mechanism of micro/nanostructured surface on the boiling heat transfer of refrigerant–oil mixture, the wetting behavior of refrigerant–oil mixture on micro/nanostructured surface needs to be known. The effect of surface modification should be considered during the investigation of the wetting behavior of refrigerant– oil mixture on micro/nanostructured surface based on the following reasons. The micro/nanostructure itself modifies the surface morphology of plain surface, and causes the increase of surface roughness. For the fluids with high surface wettability, the increase of surface roughness will increase the surface wettability; the increase of surface roughness enhances the

ω

oil concentration [wt%]

Subscripts a time-averaged r pure refrigerant ro refrigerant–oil mixture Abbreviations CCD charge-coupled device FESEM field emission scanning electron microscopy fps frames per second F-SAM fluorinated self-assembled monolayer FTIR Fourier transform infrared LED light-emitting diode MNFS micro/nanostructured surface with F-SAM MNS micro/nanostructured surface PFOTS 1H, 1H, 2H, 2H – perfluorooctyltrichlro silane PS plain copper surface UV ultraviolet

boiling heat transfer, while the increase of surface wettability deteriorates the boiling heat transfer (Li et al., 2015). The negative effect of surface wettability on the boiling heat transfer enhancement could be eliminated by the control of surface wettability (Zhang et al., 2012), and the surface modification is usually used to control the surface wettability (Li et al., 2015). In order to develop the surface wettability control method for refrigerant–oil mixture on micro/nanostructured surface, the effect of surface modification should be evaluated. The effect of oil concentration should also be considered during the investigation of wetting behavior of refrigerant– oil mixture on micro/nanostructured surface due to the following reasons. Firstly, the saturation vapor pressure of oil is much less than that of pure refrigerant (Ermolaev et al., 1972), causing the negligible evaporation of oil during the boiling process of refrigerant–oil mixture, which leads to the nonuniform mass transfer and the dynamic wetting (Sefiane et al., 2008). Secondly, the thermophysical properties of refrigerant– oil mixtures are changed with the oil concentration, and the variations of thermophysical properties will change the wetting behavior. The existing studies on the wetting behavior on micro/ nanostructured surfaces are mainly focused on water, in which the surfaces were fabricated by thermal oxidation (Nam and Ju, 2013), chemical oxidation (Kim et al., 2011; Köroglu et al., 2013; Nam and Ju, 2013; Zhu et al., 2012), chemical oxidation combining UV-photolithography (Kim et al., 2009), chemical oxidation combining self-assembled monolayer (SAM) (Chen et al., 2009), SAM (Lee et al., 2008, 2012, 2013), etching (Lee et al., 2013), coating materials (Lee et al., 2013; Zhang et al., 2015), electro-deposition (Khorsand et al., 2015), photolithography (Zhong et al., 2006), colloidal lithography combining plasma etching (Park et al., 2011) or ultraviolet nanoimprint lithography (Jo et al., 2014). The existing literatures have also reported

209

international journal of refrigeration 62 (2016) 207–221

the wetting behavior of some organic fluids (e.g. ethylene glycol, hexadecane, and diiodomethane) on micro/nanostructured surfaces, in which the surfaces were fabricated by chemical oxidation (Zhu et al., 2012), SAM (Lee et al., 2013), etching (Lee et al., 2013), coating (Lee et al., 2013) or ultraviolet nanoimprint lithography (Jo et al., 2014). The results indicated that the change of fabrication method (e.g. surface treatment condition and time) causes the different contact angles, and the surface wettability can be adjusted by SAM. The liquids in the above studies are water and organic fluids, which have different surface tension, density and viscosity from the refrigerant– oil mixture. As the surface tension affects the contact angle, and the density as well as the viscosity influence the motion of contact line (Wang et al., 2007), the wetting behavior of water and organic fluids in the existing studies may not be extended to the refrigerant–oil mixture. The existing researches on the wetting behavior of refrigerant–oil mixture are mainly focused on plain surface (Lin et al., 2016). The influences of oil concentration and surface roughness on the wetting behavior have been investigated. The results showed that the protuberant liquid film exists in front of meniscus; the addition of oil enhances the surface wettability of refrigerant, and the enhancement increases with the increase of oil concentration or surface roughness. As the morphology and chemical properties of plain surface are different from those of micro/nanostructured surface, the wetting

behavior of refrigerant–oil mixture on plain surface may not be same as that on micro/nanostructured surface. The objective of this research is to understand the wetting behavior of refrigerant–oil mixture on micro/nanostructured surface, and to evaluate the effects of surface modification and oil concentration on the wetting behavior.

2.

Experimental materials and method

2.1.

Technical approach of experimental research

The technical approach, as shown in Fig. 1, includes the following three steps.

Research objective

(1) Fabrication of test surfaces. For evaluating the effect of surface modification on the wetting behavior, three types of surfaces, i.e. plain copper surface (abbreviated as PS), micro/nanostructured surface without surface modification and that with surface modification, are fabricated. The PS is used as the substrate for fabricating MNS. The fluorinated self-assembled monolayer (F-SAM) is used for surface modification because F-SAM is a widely used surface modification method for reducing the surface wettability (Chen et al., 2004; Wu et al., 2005).

Experimental research Step 1: Fabrication of test surfaces Plain surface (PS) Surface modification

Effect of surface modification

Micro/nanostructured surface (MNS)

Test surface

Surface modification Micro/nanostructured surface with F-SAM (MNFS)

Step 2: Preparation of liquid samples Pure refrigerant Liquid sample

on

Effect of oil concentration

Different oil concentration

Refrigerant-oil mixture

Step 3: Measurement of wetting behavior

on

Method: capillary rise method

forming Meniscus capturing

qualitative observation

Wetting behavior of refrigerant-oil mixture on micro/nanostructured surface

Meniscus shape getting

quantitative evaluation

Characteristic parameters of wetting behavior

Fig. 1 – Road map of technical approach.

Experimental apparatus

210

international journal of refrigeration 62 (2016) 207–221

(2) Preparation of liquid samples. For evaluating the effect of oil on the wetting behavior, the refrigerant–oil mixtures with different oil concentrations are prepared, and the pure refrigerant is used as the benchmark. The oil concentrations should be in accordance with the oil concentration range in actual vapor compression refrigeration systems. (3) Measurement of wetting behavior. For accurately measuring the wetting behavior of the refrigerant–oil mixture on micro/nanostructured surface, a suitable measurement method should be selected, and the experimental apparatus based on the selected measurement method should be designed. In the present study, the capillary rise method is applied to measure the wetting behavior due to its good repeatability and high accuracy (Kwok et al., 1995; Vadgama and Harris, 2007). As required by the capillary rise method for measuring the wetting behavior, the experimental apparatus have three functions including adjusting the positions of test surface and liquid sample to form the meniscus, capturing the meniscus shape, and getting the characteristic parameters of wetting behavior. The first step mentioned above will be introduced in detail in section 2.2. The second and third steps are based on the methods reported by Lin et al. (2016) and will be introduced in sections 2.3 and 2.4, respectively.

2.2.

Fabrication of test surfaces

The test surfaces include PS, micro/nanostructured surface (abbreviated as MNS), and micro/nanostructured surface with

(a) CuO and Cu(OH)2 hierarchical structure (low magnification, 500×)

(c) micro-flower

fluorinated self-assembled monolayer (abbreviated as MNFS). The arithmetic mean roughness (Ra) of PS was measured to be 0.098 μm by SJ-210 portable surface roughness tester (Mitutoyo Corporation, Japan). MNS was fabricated by performing the alkali assistant surface oxidation on PS, and the procedure includes the following steps, similar to the procedure proposed in the existing literatures (Chen et al., 2009; Zhu et al., 2012). Firstly, PS was cleaned by acetone in an ultrasonic cleaner for about 3 min. Secondly, the cleaned PS was immersed in an aqueous solution of sodium hydroxide (NaOH) and ammonium persulphate ((NH4)2S2O8) at room temperature for 30 min, and the concentrations of NaOH and (NH4)2S2O8 were 2.5 mol L−1 and 0.1 mol L−1 respectively. Thirdly, with the elapse of time, the color of surface gradually varied from reddish brown to faint blue, light blue, and a little light black, indicating that the CuO/Cu(OH)2 micro/ nanostructures have been formed on the surface. Finally, the surface was removed from the solution, washed by deionized water, and then dried by nitrogen stream. Fig. 2 shows the morphologies of MNS, which were observed by field emission scanning electron microscopy (FESEM, Sirion 200, FEI Company, USA). Fig. 2a represents the low magnification image of CuO and Cu(OH)2 hierarchical structure, and indicates that the micro/nanostructures cover the surface densely and uniformly. Fig. 2b represents the high magnification image of CuO and Cu(OH)2 hierarchical structure, and shows that the micro/nanostructures are composed of CuO micro-flowers on the upper side and Cu(OH)2 nanorod arrays on the bottom. Fig. 2c and d represents the structures of microflowers and nanorod arrays, respectively, and shows that the micro-flower has several rough petals and the diameter of

(b) CuO and Cu(OH)2 hierarchical structure (high magnification, 4000×)

(d) nanorod arrays

Fig. 2 – Morphologies of micro/nanostructured surface (MNS) observed by FESEM.

211

international journal of refrigeration 62 (2016) 207–221

100

Table 1 – Information of 1H, 1H, 2H, 2H – perfluorooctyltrichlrosilane used in the experiment. 1H, 1H, 2H, 2H – perfluorooctyltrichlrosilane

Abbreviation CAS registry number Chemical formula Molecular weight Structural formula

PFOTS 78560-45-9 C8H4Cl3F13Si 481.5 g mol−1

80

Transmission (%)

Name

951.1

60 1022.1

40 1071.5 1238.9

20 1206.1

Supplier

1143.6

TCI (Shanghai) Development Co., Ltd., China 0 3600

micro-flower is about 1.5 μm, while the nanorod arrays form a compact network, and the nanorod has the average size of about 1 μm in length and 100 nm in diameter. The procedure for fabricating MNFS includes the following steps, similar to the procedure reported in the existing literature (Chen et al., 2009). Firstly, the as-prepared MNS was immersed in a 0.5 vol% 1H, 1H, 2H, 2H – perfluorooctyltrichlrosilane (detailed information is listed in Table 1)/n-hexane solution at room temperature for 15 min. Secondly, with the elapse of time, 1H, 1H, 2H, 2H – perfluorooctyltrichlrosilane was hydrolyzed and chemically adsorbed on the surface. Thirdly, the surface was removed from the solution, and washed by n-hexane for 3 min. Finally, the surface was heated in a vacuum oven at 100 °C for 1 hour in order to promote silane hydrolysis and condensation, and the

3200

2800

2400

2000

1600

1200

800

Wavenumber (cm-1) Fig. 4 – FTIR spectra of micro/nanostructured surface with F-SAM (MNFS).

stable F-SAM was formed on MNS. The mechanism of F-SAM formation on MNS is presented in Fig. 3. The formation of F-SAM could be determined by Fourier transform infrared (FTIR) spectra because the characteristic bands in FTIR spectra correspond to the vibrations of bonds (Brassard et al., 2012; Xu et al., 2008). In order to confirm the formation of F-SAM on MNS, FTIR spectra of MNS after F-SAM modification was obtained by FTIR Microscope (Nicolet iN10 MX, Thermo Scientific, USA), as shown in Fig. 4. The bands at 1238.9, 1206.1 and 1022.1 cm −1 represent the stretching

Fig. 3 – Mechanism of F-SAM formation on the micro/nanostructured surface (MNS).

212

international journal of refrigeration 62 (2016) 207–221

Fig. 5 – Schematic diagram of experimental apparatus.

vibration of C-F bond; the bands at 1143.1 and 1071.5 cm−1 represent the stretching vibrations of Si-O-C bond and Si-O-Si bond, respectively; and the band at 951.1 cm−1 represents the Si-OCu vibration. The occurrence of these bands indicates that F-SAM has been formed on MNS.

2.3.

Preparation of liquid samples

The liquid samples include the refrigerant–oil mixtures with different oil concentrations, and the pure refrigerant. For conveniently measuring the wetting behavior, R141b was used, which is in liquid state at room temperature and atmospheric pressure, and has the properties close to the widely applied refrigerants. The mineral oil ATMOS NM56 was used, and the oil concentrations (ω) include 1, 3, 5 and 10 wt%. NM56 has kinematic viscosity of 56.2 mm2 s−1 at 40 °C, density of 0.928 g cm−3 at 15 °C, and surface tension of 25.5 mN m−1 at 15 °C; while it is miscible with R141b at 15 °C at ω from 1 to 10 wt%.

2.4.

Measurement of wetting behavior

The experimental apparatus for measuring the wetting behavior of refrigerant–oil mixture on micro/nanostructured surface includes three units (i.e. sample control unit, optical imaging unit, and image processing unit), as schematically shown in Fig. 5. The sample control unit is used for adjusting the positions of test surface and liquid sample, in which the test surface bracket and two retaining nuts can fix the test surface on the elevator, the elevator can adjust the position of test surface, the spilled liquid collector is put under the liquid sample pool, and the x-y-z movable platform can adjust the position of liquid sample. The optical imaging unit is used for capturing the meniscus shape, in which the light source is the monochromatic light-emitting diode (LED) cold light source with adjustable illumination, the ultraviolet (UV) optical filter can filtrate the ambient UV light, the microscopic lens (fold magnification: 0.7–4.5, optical distortion: 0.01%) can realize the

continuous zoom, and the charge-coupled device (CCD) camera (resolution: 752 × 480, maximum frame rate: 87 fps) can capture the images. The image processing unit is used for getting the characteristic parameters of wetting behavior, in which the software CAST 3.0 (KINO Industry Co., Ltd., USA) can determine the contact angle. The experimental procedure includes the following steps: (1) fixing the test surface between two retaining nuts on the test surface bracket, and filling the liquid sample pool with the liquid sample; (2) adjusting the positions of test surface and liquid sample by the sample control unit to form the meniscus; (3) adjusting the brightness of light and the focus of microscopic lens to obtain the clearest image of meniscus; (4) recording the images by the CCD camera at a speed of 0.5 fps; (5) cleaning the test surface with acetone in an ultrasonic cleaner for 3 minutes after recording the images, and then drying the test surface by nitrogen stream; and (6) getting the characteristic parameters of wetting behavior through the image processing unit. The characteristic parameters for quantitatively describing the wetting behavior include contact line velocity (vCL), contact angle (θ), rising liquid height (H) and liquid film length (L), as defined below. Contact line velocity (vCL) is the moving speed of contact line, i.e. the line where solid, gas and liquid meet. Contact angle (θ) is the angle between solid surface and tangent line of liquid/vapor interface at contact line. Rising liquid height (H) is the vertical distance between the contact line and the horizontal liquid level, and is used to evaluate the surface wettability. Liquid film length (L) is the vertical distance between contact line and top of meniscus. The data reduction and uncertainties of these four parameters are listed in Table 2.

2.5.

Experimental conditions

Three categories of experimental conditions were formulated to investigate the wetting behavior on PS, MNS and MNFS

213

international journal of refrigeration 62 (2016) 207–221

Table 2 – Data reduction and uncertainties of characteristic parameters. Parameter

Data reduction y (t ) − yCL (t − Δt ) vCL (t ) = Ma CL Δt Spline curve fitting H = Ma ⋅ h L = Ma ⋅ l

Contact line velocity (vCL) Contact angle (θ) Rising liquid height (H) Liquid film length (L)

respectively, as listed in Table 3. For each category, the liquid samples include pure R141b and R141b-NM56 mixtures with the oil concentrations (ω) of 1, 3, 5 and 10 wt%. The images at 2 to 30 s with the time interval of 2 s are processed to get the characteristic parameters of wetting behavior. The experiments

Serial

Category 1

Test surface

Plain copper surface (PS)

Category Micro/ 2 nanostructured surface (MNS) Category Micro/ 3 nanostructured surface with F-SAM (MNFS)

Liquid sample

Pure R141b R141b-NM56 mixture (ω = 1, 3, 5, 10 wt%) Pure R141b R141b-NM56 mixture (ω = 1, 3, 5, 10 wt%) Pure R141b R141b-NM56 mixture (ω = 1, 3, 5, 10 wt%)

Time corresponding to the processed images Range: 2–30 s Interval: 2 s

Range: 2–30 s Interval: 2 s

Range: 2–30 s Interval: 2 s

Source

0.0057 mm s−1

Lin et al. (2016)

0.1° 0.0081 mm 0.0081 mm

Present study

were carried out at temperature of 15 °C in open-air environment with relative humidity of 49%.

3. Table 3 – Experimental conditions.

Uncertainty

Results and discussion

3.1. Meniscus shapes of pure refrigerant and refrigerant–oil mixture on PS, MNS and MNFS Fig. 6 shows the meniscus shapes of pure refrigerant and refrigerant–oil mixture (ω = 10 wt%) wetting on PS, MNS and MNFS. The meniscus of pure refrigerant wetting on PS, MNS and MNFS are concave, indicating that PS, MNS and MNFS are all refrigerant-philic; while the liquid level decreases with the time due to the evaporation of pure refrigerant, as shown in Fig. 6a–c. The meniscus shapes of refrigerant–oil mixture on PS and MNS are different from that on MNFS, as shown in Fig. 6d–f. For PS and MNS, the protuberant liquid film is formed in front of meniscus, climbs up and is elongated with the elapse of time; while for MNFS, the protuberant liquid film does not exist. The wetting behavior of pure oil on PS, MNS and MNFS was also investigated to understand the effect of oil, and the images of oil drop on PS, MNS and MNFS are shown in Fig. 7. The static contact angles of pure oil on PS, MNS and MNFS are 29.15°,

Fig. 6 – Meniscus shapes of pure refrigerant and refrigerant–oil mixture (10 wt%) wetting on PS, MNS and MNFS.

214

international journal of refrigeration 62 (2016) 207–221

Fig. 7 – Images of oil drop on: (a) PS, (b) MNS, (c) MNFS.

16.05° and 129.11°, respectively; meaning that PS and MNS are oleophilic, while MNFS is oleophobic. Fig. 8 schematically shows the physical mechanism of refrigerant–oil mixture wetting on PS, MNS and MNFS, which could explain the existing of protuberant liquid film on PS and MNS, and the nonexistence of protuberant liquid film on MNFS. The existing of protuberant liquid film on PS is caused by the upward Marangoni convection, which is attributed to the faster evaporation of refrigerant and the difference of surface tension between refrigerant and oil (Lin et al., 2016), as shown in Fig. 8a. The reason for the existence of protuberant liquid film on MNS is similar to that on PS, as shown in Fig. 8b. As

the surface free energy and the frictional force on MNS are different from those on PS, the differences of protuberant liquid film characteristics exist between these two surfaces, and the differences will be discussed in detail in sections 3.2 and 3.4. As MNFS is refrigerant-philic and oleophobic (as illustrated in Fig. 7), the oil molecules tend to move away from the contact line, while the refrigerant molecules tend to move closer to the contact line. These two tendencies cause the oil concentration near the contact line to be lower than that in the interior area despite the faster evaporation of refrigerant in meniscus. The lower oil concentration and lower surface tension of refrigerant–oil mixture near the contact line result in the

Fig. 8 – Physical mechanism of refrigerant–oil mixture wetting on: (a) PS, (b) MNS, (c) MNFS.

international journal of refrigeration 62 (2016) 207–221

downward Marangoni convection, i.e. the liquid flows from the contact line to the interior area. The liquid flow will cause the downward movement of contact line, thus the protuberant liquid film cannot be formed, as shown in Fig. 8c.

3.2. Contact line velocities of pure refrigerant and refrigerant–oil mixture on PS, MNS and MNFS Fig. 9a and b shows the time evolution of contact line velocity (vCL) on PS, MNS and MNFS for pure refrigerant, and that for refrigerant–oil mixtures (oil concentrations include 1, 3, 5 and 10 wt%), respectively. The contact line velocities (vCL) of pure refrigerant on PS, MNS and MNFS are all less than zero (the average values are −0.0116, −0.0119 and −0.0124 mm s−1, respectively), as shown in Fig. 9a. The receding contact line is caused by the evaporation of pure refrigerant. The small difference of vCL among PS, MNS and MNFS is caused by the following reasons. The influences of micro/nanostructures without and with F-SAM on the receding contact line are reflected in the frictional force (Lee et al., 2012). The viscosity of pure R141b is very low (0.459 mPa s at 15 °C and 1 atm) and then the frictional force is small, causing the small difference of vCL among PS, MNS and MNFS.

The contact line velocities (vCL) of refrigerant–oil mixture on PS and MNS are both larger than zero, and vCL on PS and MNS averagely increase respectively by 225.6% and 161.6% with the increase of ω from 1 to 10 wt%, as shown in Fig. 9b. The contact line velocities (vCL) of refrigerant–oil mixtures on MNS are 48.1%, 29.5%, 41.9% and 19.0% larger than those on PS at the oil concentrations (ω) of 1, 3, 5 and 10 wt%, respectively. The possible reasons for this phenomenon are as follows. (1) The upward Marangoni convection occurring on PS or MNS causes the liquid film to climb up, resulting in the advancing contact line. (2) With the increase of ω, the surface tension gradient between the interior area and the meniscus increases, and then the Marangoni convection is enhanced, causing the increase of vCL. (3) The presence of micro/nanostructure increases the surface area, causing the increase of surface free energy, which leads to the enhancement of capillary force and the increase of vCL. The contact line velocity (vCL) of refrigerant–oil mixture on MNFS remains almost unchanged at zero for the first few seconds, meaning the pinning of contact line; then decreases to a certain negative value, meaning the downward movement of contact line, as shown in Fig. 9b. Meanwhile, the onset of contact line movement becomes later with the increase of oil concentration (ω). This phenomenon could be explained as

0.000

R141b

PS MNS MNFS

vCL (mm s-1)

-0.005

-0.010

-0.015

-0.020 0

8

215

16

24

t (s) (a) pure refrigerant

32

(b) refrigerant-oil mixture Fig. 9 – Contact line velocity for: (a) pure refrigerant, (b) refrigerant–oil mixture.

216

international journal of refrigeration 62 (2016) 207–221

follows. (1) The MNFS is refrigerant-philic and oleophobic, causing the contact line to have the tendency to move downward. (2) The liquid with higher surface tension and viscosity has larger contact angle hysteresis, and the positive correlation exists between surface tension and critical hysteresis tension (Wang et al., 2003). (3) Compared to the pure refrigerant, the refrigerant–oil mixture has higher surface tension and viscosity, causing the pinning of contact line for the first few seconds. (4) The increase of oil concentration leads to higher surface tension of refrigerant–oil mixture and then the larger critical hysteresis tension, causing the onset of contact line movement to be later.

3.3. Contact angles of pure refrigerant and refrigerant–oil mixture on PS, MNS and MNFS Fig. 10a and b shows the time evolution of contact angle (θ) on PS, MNS and MNFS for pure refrigerant, and that for refrigerant–oil mixtures (oil concentrations include 1, 3, 5 and 10 wt%), respectively.

The average values of contact angle (θ) of pure refrigerant on PS, MNS and MNFS are 22.19°, 10.51° and 23.01° respectively, as shown in Fig. 10a. The contact angle on MNFS is close to the contact angle on PS, and is about 2.2 times of contact angle on MNS. This phenomenon could be explained as follows. (1) The wetting of pure refrigerant on PS, MNS and MNFS are in the Wenzel state; meaning that the apparent contact angle increases with the intrinsic contact angle, while it decreases with the increase of surface roughness (Wenzel, 1936). (2) The presence of micro/nanostructure significantly increases the surface roughness and has no influence on the intrinsic contact angle, causing the apparent contact angle on MNS to be smaller than that on PS. (3) The fluorine in F-SAM is effective for reducing the surface free energy due to its small atomic radius and large electronegativity (Shang et al., 2005), causing the F-SAM to increase the intrinsic contact angle. (4) The presence of F-SAM increases the intrinsic contact angle and hardly has influence on the surface roughness, causing the apparent contact angle on MNFS to be larger than that on MNS.

40 PS MNS MNFS

35

R141b

θ (o)

30 25 20 15 10 5 0

8

16

t (s)

24

32

(a) pure refrigerant

(b) refrigerant-oil mixture Fig. 10 – Contact angle for: (a) pure refrigerant, (b) refrigerant–oil mixture.

international journal of refrigeration 62 (2016) 207–221

The contact angles (θ) of refrigerant–oil mixture on PS and MNS both increase with time to a maximum value and then decrease, θ on PS and MNS averagely increase respectively by 176.3% and 88.8% with the increase of ω from 1 to 10 wt%, as shown in Fig. 10b. The contact angles (θ) of refrigerant–oil mixtures on MNS are 63.0%, 45.8%, 53.3% and 11.4% larger than those on PS at the oil concentrations (ω) of 1, 3, 5 and 10 wt%, respectively. The possible reasons for this phenomenon are as follows. (1) The oil concentration in the protuberant liquid film increases with the evaporation of the refrigerant, and the oil has higher surface tension and lower surface wettability than the refrigerant on PS and MNS (as shown in Fig. 7), which leads to the increase of contact angle of refrigerant–oil mixture with time. (2) After the entire evaporation of the refrigerant, the protuberant liquid film is pure oil film, and then the tendencies of contact angle changed with time and ω are consistent with those of contact line velocity (vCL), i.e. decreasing with time and increasing with the increase of ω. (3) The surface roughness of MNS is larger than that of PS, and the positive correlation exists between dynamic advancing contact angle and surface roughness (Wang et al., 2003), causing the contact angle on MNS to be larger than that on PS. The contact angle (θ) of refrigerant–oil mixture on MNFS decreases for the first few seconds, and then remains almost unchanged at a certain value, as shown in Fig. 10b. The transition onset of contact angle is in accordance to that of contact line velocity. Meanwhile, θ averagely increases by 27.0% with the increase of oil concentration (ω) from 1 to 10 wt%. This phenomenon could be explained as follows. (1) The contact line is pinning for the first few seconds (as illustrated in section 3.2), but the liquid in meniscus keeps to be evaporated, causing the decrease of contact angle. (2) While the contact line is moving with a nearly constant velocity, the contact angle remains almost unchanged. (3) With the increase of ω, the surface tension of refrigerant–oil mixture increases, causing the increase of contact angle.

217

3.4. Liquid film lengths of refrigerant–oil mixture on PS and MNS Fig. 11 shows the time evolution of liquid film length (L) of refrigerant–oil mixtures (oil concentrations include 1, 3, 5 and 10 wt%) on PS and MNS. The liquid film lengths (L) of the refrigerant–oil mixture on PS and MNS increase with time, and averagely increase respectively by 108.0% and 103.0% with the increase of oil concentration (ω) from 1 to 10 wt%, as shown in Fig. 11. The liquid film lengths (L) of refrigerant–oil mixtures on MNS are 22.7%, 19.7%, 34.5% and 19.7% larger than those on PS at the oil concentrations (ω) of 1, 3, 5 and 10 wt%, respectively. The above phenomenon could be explained as follows. (1) The front part of the liquid film climbs up due to the Marangoni convection, while the end part of liquid film descends due to the evaporation of the refrigerant, causing the liquid film to be elongated with the elapse of time. (2) With the increase of ω, the climbing speed increases due to the increase of contact line velocity (vCL), and the descending speed decreases due to the decrease of evaporation rate. (3) The increase of the climbing speed is larger than the decrease of the descending speed, which leads to the increase of L with the increase of ω. (4) The presence of micro/ nanostructure increases the climbing speed of the liquid film, causing the liquid film length (L) on MNS to be larger than that on PS.

3.5. Rising liquid heights of pure refrigerant and refrigerant–oil mixture on PS, MNS and MNFS Fig. 12a and b shows the time evolution of rising liquid height (H) on PS, MNS and MNFS for pure refrigerant, and that for refrigerant–oil mixtures (oil concentrations include 1, 3, 5 and 10 wt%), respectively.

Fig. 11 – Liquid film length for refrigerant–oil mixture.

218

international journal of refrigeration 62 (2016) 207–221

1.8 PS MNS MNFS

R141b

H (mm)

1.6

1.4

1.2

1.0 0

8

16

t (s)

24

32

(a) pure refrigerant

(b) refrigerant-oil mixture Fig. 12 – Rising liquid height for: (a) pure refrigerant, (b) refrigerant–oil mixture.

The average values of rising liquid height (H) of pure refrigerant on PS, MNS and MNFS are 1.37, 1.55 and 1.25 mm respectively, as shown in Fig. 12a. This tendency is opposite to the contact angle, which is caused by the negative relationship between rising liquid height and contact angle (Kwok et al., 1995). The rising liquid heights (H) of the refrigerant–oil mixture on PS and MNS increase with time, while averagely increase respectively by 38.4% and 33.4% with the increase of oil concentration (ω) from 1 to 10 wt%, as shown in Fig. 12b. The above phenomenon could be explained as follows. The rising liquid heights (H) of refrigerant–oil mixture on PS and MNS are the sum of liquid film length (L) and meniscus height. With the increase of time or ω, L increases and the meniscus height changes slightly according to the measurement, causing H to increase with the increase of time or ω. The rising liquid height (H) of refrigerant–oil mixture on MNFS increases for the first few seconds, and then remains almost unchanged at a certain value, as shown in Fig. 12b. The transition onset of H is in accordance to those of contact line velocity (vCL) and contact angle (θ). H averagely decreases by 20.6% with the increase of oil concentration (ω) from 1 to 10 wt%, which is caused by the negative relationship between H and θ (Kwok et al., 1995).

3.6.

Effect of surface modification on surface wettability

In order to quantitatively evaluate the effect of surface modification on the surface wettability for pure refrigerant and refrigerant–oil mixture, three effect factors, micro/nanostructure effect factor (EF MN ), F-SAM effect factor (EF F ), and micro/ nanostructure combined F-SAM effect factor (EFMNF), are defined as Eqs. (1)–(3).

EFMN = Ha,MNS Ha,PS

(1)

EFF = Ha,MNFS Ha,MNS

(2)

EFMNF = Ha,MNFS Ha,PS

(3)

where Ha, PS, Ha, MNS, and Ha, MNFS are the time-averaged values of rising liquid heights on PS, MNS and MNFS, respectively. Fig. 13 shows the micro/nanostructure effect factors (EFMN), the F-SAM effect factors (EFF), and the micro/nanostructure combined F-SAM effect factors (EFMNF) for pure refrigerant and refrigerant–oil mixtures with different oil concentrations (ω). EFF and EFMNF range from 0.30 to 0.81 and 0.34 to 0.91, respectively; and decrease respectively by 40.5% and 42.6%

219

international journal of refrigeration 62 (2016) 207–221

3.0

1.2

2.5

1.0

2.0 EFoil

Effect factor

1.4

EFMN

0.8

EFF EFMNF

0.6

0.5

0

2

4 6 ω (wt%)

8

0.0

10

Fig. 13 – Micro/nanostructure effect factor (EFMN), F-SAM effect factor (EFF) and micro/nanostructure combined F-SAM effect factor (EFMNF) changed with oil concentration (ω).

with the increase of ω from 1 to 10 wt%. This phenomenon indicates that F-SAM and micro/nanostructure modified by F-SAM both reduce the surface wettability, and the reduction degree of surface wettability is larger for higher oil concentration. The micro/nanostructure effect factor (EFMN) ranges from 1.13 to 1.20 as shown in Fig. 13, indicating that micro/ nanostructure increases the surface wettability. The increase degree of surface wettability is not significantly influenced by oil concentration.

Effect of oil concentration on surface wettability

In order to quantitatively evaluate the effect of oil concentration on the surface wettability on PS, MNS, and MNFS, oil effect factor (EFoil) is defined as Eq. (4).

EFoil = Ha,ro Ha,r

(4)

where Ha, ro and Ha, r are the time-averaged values of rising liquid heights of refrigerant–oil mixture and pure refrigerant, respectively. The oil effect factors (EFoil) on PS, MNS and MNFS are shown in Fig. 14. EFoil on PS and MNS are both larger than 1; and increase respectively by 38.4% and 33.4% with the increase of oil concentration (ω) from 1 to 10 wt%. This phenomenon means that the oil increases the wettability of refrigerant on PS or MNS, and the increase degree of surface wettability is larger for higher oil concentration. The oil effect factor (EFoil) on MNFS is smaller than 1, and decreases by 20.6% with the increase of oil concentration (ω) from 1 to 10 wt%, as shown in Fig. 14. This phenomenon means that the oil reduces the wettability of refrigerant on MNFS, and the reduction degree of surface wettability is larger for higher oil concentration.

1

2

3

4

5 6 7 ω (wt%)

8

9 10

Fig. 14 – Oil effect factor (EFoil) changed with oil concentration (ω) for PS, MNS and MNFS.

4.

3.7.

1.5 1.0

0.4 0.2

PS MNS MNFS

Conclusions (1) The protuberant liquid film is formed in front of the meniscus during the wetting of refrigerant–oil mixture on plain surface or micro/nanostructured surface, but does not exist on micro/nanostructured surface with F-SAM. (2) For refrigerant–oil mixture on plain surface and micro/ nanostructured surface, the contact line velocity is larger than zero, the contact angle increases with time to a maximum value and then decreases, and the liquid film length as well as the rising liquid height increase with the time or the increase of oil concentration. The contact line velocity, contact angle and liquid film length of refrigerant–oil mixture on MNS are all larger than those on PS. (3) For refrigerant–oil mixture on micro/nanostructured surface with F-SAM, the contact line velocity remains almost unchanged at zero for the first few seconds, and then decreases to a certain negative value; the contact angle decreases for the first few seconds, and then remains almost unchanged; and the rising liquid height increases for the first few seconds, and then remains almost unchanged. (4) Surface wettability is reduced due to the presence of F-SAM or micro/nanostructure modified by F-SAM; while it is increased due to the presence of micro/nanostructure. (5) Oil increases the wettability of the refrigerant on plain surface or micro/nanostructured surface, while it reduces the wettability of the refrigerant on micro/nanostructured surface with F-SAM.

Acknowledgments The authors gratefully acknowledge the supports from the National Natural Science Foundation of China (Grant No.

220

international journal of refrigeration 62 (2016) 207–221

51376124) and the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (Grant No. 51521004).

REFERENCES

Attinger, D., Moore, C., Donaldson, A., Jafari, A., Stone, H.A., 2013. Fluid dynamics topics in bloodstain pattern analysis: comparative review and research opportunities. Forensic Sci. Int. 231, 375–396. Attinger, D., Frankiewicz, C., Betz, A.R., Schutzius, T.M., Ganguly, R., Das, A., et al., 2014. Surface engineering for phase change heat transfer: a review. MRS Energy & Sustainability: A Review Journal 1, E4. Brassard, J.D., Sarkar, D.K., Perron, J., 2012. Fluorine based superhydrophobic coatings. Appl. Sci. 2, 453–464. Chen, J.K., Ko, F.H., Hsieh, K.F., Chou, C.T., Chang, F.C., 2004. Effect of fluoroalkyl substituents on the reactions of alkylchlorosilanes with mold surfaces for nanoimprint lithography. J. Vac. Sci. Technol. B 22, 3233–3241. Chen, X.H., Kong, L.H., Dong, D., Yang, G.B., Yu, L.G., Chen, J.M., et al., 2009. Fabrication of functionalized copper compound hierarchical structure with bionic superhydrophobic properties. J. Phys. Chem. C 113, 5396–5401. Dong, L.N., Quan, X.J., Cheng, P., 2014. An experimental investigation of enhanced pool boiling heat transfer from surfaces with micro/nano-structures. Int. J. Heat Mass Transf. 71, 189–196. Ermolaev, M.I., Batishchev, V.V., Goryachev, K.V., 1972. Investigation of the vapor pressure of lubricating oils. Chem. Technol. Fuels Oils 8, 388–390. Gong, S., Cheng, P., 2015. Lattice Boltzmann simulations for surface wettability effects in saturated pool boiling heat transfer. Int. J. Heat Mass Transf. 85, 635–646. Jo, H.B., Choi, J., Byeon, K.J., Choi, H.J., Lee, H., 2014. Superhydrophobic and superoleophobic surfaces using ZnO nano-in-micro hierarchical structures. Microelectron. Eng. 116, 51–57. Jo, H.J., Ahn, H.S., Kang, S.H., Kim, M.H., 2011. A study of nucleate boiling heat transfer on hydrophilic, hydrophobic and heterogeneous wetting surfaces. Int. J. Heat Mass Transf. 54, 5643–5652. Khorsand, S., Raeissi, K., Ashrafizadeh, F., Arenas, M.A., 2015. Super-hydrophobic nickel-cobalt alloy coating with micronano flower-like structure. Chem. Eng. J. 273, 638–646. Kim, D.E., Yu, D.I., Jerng, D.W., Kim, M.H., Ahn, H.S., 2015. Review of boiling heat transfer enhancement on micro/ nanostructured surfaces. Exp. Therm. Fluid Sci. 66, 173–196. Kim, D.H., Kim, Y., Hwang, S.H., Bang, Y.S., Cho, C.R., Kim, Y.K., et al., 2011. Experimental and theoretical evaluation of wettability on micro/nano hierarchically engineered surfaces based on robust micro-post-arrayed- and highly ordered nano-rippled-structures. Appl. Surf. Sci. 257, 8985–8992. Kim, D.S., Lee, B.K., Yeo, J., Choi, M.J., Yang, W., Kwon, T.H., 2009. Fabrication of PDMS micro/nano hybrid surface for increasing hydrophobicity. Microelectron. Eng. 86, 1375–1378. Köroglu, B., Lee, K.S., Park, C., 2013. Nano/micro-scale surface modifications using copper oxidation for enhancement of surface wetting and falling-film heat transfer. Int. J. Heat Mass Transf. 62, 794–804. Kruse, C.M., Anderson, T., Wilson, C., Zuhlke, C., Alexander, D., Gogos, G., et al., 2015. Enhanced pool-boiling heat transfer and critical heat flux on femtosecond laser processed stainless steel surfaces. Int. J. Heat Mass Transf. 82, 109–116.

Kwok, D.Y., Budziak, C.J., Neumann, A.W., 1995. Measurements of static and low rate dynamic contact angles by means of an automated capillary rise technique. J. Colloid Interface Sci. 173, 143–150. Launay, S., Fedorov, A.G., Joshi, Y., Cao, A., Ajayan, P.M., 2006. Hybrid micro-nano structured thermal interfaces for pool boiling heat transfer enhancement. Microelectron. J. 37, 1158– 1164. Lee, C.Y., Zhang, B.J., Park, J., Kim, K.J., 2012. Water droplet evaporation on Cu-based hydrophobic surfaces with nano- and micro-structures. Int. J. Heat Mass Transf. 55, 2151– 2159. Lee, S., Cheng, K., Palmre, V., Bhuiya, M.D.M.H., Kim, K.J., Zhang, B.J., et al., 2013. Heat transfer measurement during dropwise condensation using micro/nano-scale porous surface. Int. J. Heat Mass Transf. 65, 619–626. Lee, S.H., Ishizaki, T., Saito, N., Takai, O., 2008. Surface characterization on binary nano/micro-domain composed of alkyl- and amino-terminated self-assembled monolayer. Appl. Surf. Sci. 254, 7453–7458. Li, Y.Y., Liu, Z.H., Wang, G.S., 2013. A predictive model of nucleate pool boiling on heated hydrophilic surfaces. Int. J. Heat Mass Transf. 65, 789–797. Li, Y.Y., Chen, Y.J., Liu, Z.H., 2014. A uniform correlation for predicting pool boiling heat transfer on plane surface with surface characteristics effect. Int. J. Heat Mass Transf. 77, 809– 817. Li, Y.Y., Liu, Z.H., Zheng, B.C., 2015. Experimental study on the saturated pool boiling heat transfer on nano-scale modification surface. Int. J. Heat Mass Transf. 84, 550–561. Lin, L.N., Peng, H., Ding, G.L., 2016. Influence of oil concentration on wetting behavior during evaporation of refrigerant–oil mixture on copper surface. Int. J. Refrigeration 61, 23–36. Nam, Y., Ju, Y.S., 2013. A comparative study of the morphology and wetting characteristics of micro/nanostructured Cu surfaces for phase change heat transfer applications. J. Adhes. Sci. Technol. 27, 2163–2176. Park, J., Lim, H., Kim, W., Ko, J.S., 2011. Design and fabrication of a superhydrophobic glass surface with micro-network of nanopillars. J. Colloid Interface Sci. 360, 272–279. Sefiane, K., David, S., Shanahan, M.E.R., 2008. Wetting and evaporation of binary mixture drops. J. Phys. Chem. B 112, 11317–11323. Shang, H.M., Wang, Y., Limmer, S.J., Chou, T.P., Takahashi, K., Cao, G.Z., 2005. Optically transparent superhydrophobic silicabased films. Thin Solid Films 472, 37–43. Shojaeian, M., Kosar, A., 2015. Pool boiling and flow boiling on micro- and nanostructured surfaces. Exp. Therm. Fluid Sci. 63, 45–73. Vadgama, B., Harris, D.K., 2007. Measurements of the contact angle between R134a and both aluminum and copper surfaces. Exp. Therm. Fluid Sci. 31, 979–984. Wang, X.D., Peng, X.F., Lee, D.Z., 2003. Dynamic wetting and stress singularity on contact line. Sci. China Ser. E Technol. Sci. 46, 407–417. Wang, X.D., Peng, X.F., Duan, Y.Y., Wang, B.X., 2007. Dynamics of spreading of liquid on solid surface. Chin. J. Chem. Eng. 15, 730–737. Wenzel, R.N., 1936. Resistance of solid surfaces to wetting by water. Ind. Eng. Chem. 28, 988–994. Wu, K., Bailey, T.C., Willson, C.G., Ekerdt, J.G., 2005. Surface hydration and its effect on fluorinated SAM formation on SiO2 surfaces. Langmuir 21, 11795–11801. Xu, W.G., Liu, H.Q., Lu, S.X., Xi, J.M., Wang, Y.B., 2008. Fabrication of superhydrophobic surfaces with hierarchical structure through a solution-immersion process on copper and galvanized iron substrates. Langmuir 24, 10895–10900.

international journal of refrigeration 62 (2016) 207–221

Zhang, B.J., Kim, K.J., Yoon, H., 2012. Enhanced heat transfer performance of alumina sponge-like nano-porous structures through surface wettability control in nucleate pool boiling. Int. J. Heat Mass Transf. 55, 7487–7498. Zhang, B.J., Kuok, C., Kim, K.J., Hwang, T., Yoon, H., 2015. Dropwise steam condensation on various hydrophobic surfaces: polyphenylene sulfide (PPS), polytetrafluoroethylene (PTFE), and self-assembled micro/nano silver (SAMS). Int. J. Heat Mass Transf. 89, 353–358.

221

Zhong, Y.F., Jacobi, A.M., Georgiadis, J.G. Condensation and wetting behavior on surfaces with micro-structures: superhydrophobic and super-hydrophilic. International Refrigeration and Air Conditioning Conference 2006; Paper 828. Zhu, X.T., Zhang, Z.Z., Xu, X.H., Men, X.H., Yang, J., Zhou, X.Y., et al., 2012. Facile fabrication of a superamphiphobic surface on the copper substrate. J. Colloid Interface Sci. 367, 443–449.