Flow Condensation on Copper-Based ... - ACS Publications

Dec 19, 2012 - ... flow condensation conditions before drawing conclusions regarding their .... Internal convective jumping-droplet condensation in tu...
0 downloads 0 Views 3MB Size
Subscriber access provided by FORDHAM UNIVERSITY

Article

Flow condensation on copper based nanotextured superhydrophobic surfaces Daniele Torresin, Manish K. Tiwari, Davide Del Col, and Dimos Poulikakos Langmuir, Just Accepted Manuscript • DOI: 10.1021/la304389s • Publication Date (Web): 19 Dec 2012 Downloaded from http://pubs.acs.org on December 25, 2012

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

1

Flow condensation on copper based nanotextured

2

superhydrophobic surfaces Daniele Torresin1,2, Manish K. Tiwari1*, Davide Del Col2, Dimos Poulikakos1*.

3 4

1

5

Engineering Department, ETH Zurich, 8092 Zurich, Switzerland

6

2

Laboratory of Thermodynamics in Emerging Technologies, Mechanical and Process

Dipartimento di Ingegeneria Industriale, Università degli Studi di Padova, 35131 Padova, Italy

7 8

KEYWORDS. Dropwise Condensation, Copper, Superhydrophobicity, Nanostructuring.

9

ABSTRACT

10

Superhydrophobic surfaces have shown excellent ability to promote dropwise condensation with

11

high droplet mobility, leading to enhanced surface thermal transport. To date, however, it is

12

unclear how superhydrophobic surfaces would perform under the stringent flow condensation

13

conditions of saturated vapor at high temperature, which can affect superhydrophobicity. Here,

14

we investigate this issue employing all-copper superhydrophobic surfaces with controlled

15

nanostructuring for minimal thermal resistance. Flow condensation tests performed with

16

saturated vapor at a high temperature (110 °C) showed the condensing drops to penetrate the

17

surface texture i.e. attain the Wenzel state with lower droplet mobility. At the same time, the

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

18

vapor shear helped ameliorate the mobility and enhanced the thermal transport. At the high end

19

of the examined vapor velocity range, a heat flux of ~600kWm-2 was measured at 10K

20

subcooling and 18 m s-1 vapor velocity. This clearly highlights the excellent potential of a

21

nanostructured superhydrophobic surface in flow condensation applications. The surfaces

22

sustained dropwise condensation and vapor shear for five days, following which mechanical

23

degradation caused a transition to filmwise condensation. Overall, our results underscore the

24

need to investigate superhydrophobic surfaces under stringent and realistic flow condensation

25

conditions before drawing conclusions regarding their performance in practically relevant

26

condensation applications.

Page 2 of 30

27

28 29

INTRODUCTION

30

Vapor condensation is an important phenomenon in nature and is directly relevant to many

31

technical applications ranging from the chemical to the power industries. Enhancing thermal

32

transport during condensation has been the subject of numerous scientific investigations. It is

33

well established that the dropwise condensation mode can lead up to several hundred percent

34

enhancement in heat transfer coefficients compared to the film condensation.1,2 The sweeping

35

and renewal mechanism present in the droplet growth process in dropwise condensation

36

augments both the heat and the mass transfer coefficients. On the other hand, during film

37

condensation the liquid layer adjacent to the wall introduces an additional thermal resistance and

38

adversely affects the thermal and mass transport. The surface properties, in particular the surface

ACS Paragon Plus Environment

2

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

39

energy and the related phenomena, play a crucial role in determining the condensation mode.

40

Typically, hydrophobic surfaces are expected to promote dropwise condensation, while

41

hydrophilic ones induce filmwise condensation.3 To this end, superhydrophobic surfaces, where

42

a combination of low surface energy and surface texturing is used to enhance the hydrophobicity,

43

have recently been proposed as a promising approach to promote dropwise condensation. 4-7 An

44

attractive feature of using superhydrophobic surfaces is to facilitate easy roll-off of the droplets

45

as they form during condensation. Such facile mobility of the drops could lead to a significant

46

improvement in the heat transfer associated with the condensation process.8-12 When

47

condensation of humid air occurs, the mobility of droplets on superhydrophobic surfaces

48

however, is promoted by trapping air between the droplet and the texture of the surface, i.e. by

49

promoting the so called Cassie state of a droplet on the superhydrophobic surface.13 The trapped

50

air, although beneficial for droplet mobility, adds a thermal resistance, which adversely

51

influences the thermal transport.14 During condensation of a pure vapor, it is assumed that the

52

vapor trapped in the texture of the surface condenses in contact with the cold surface. Allowing

53

the liquid to penetrate the asperities of the surface texture, i.e. by forming the so-called Wenzel

54

state of the droplet, could at least partially ameliorate the thermal resistance problem. However,

55

this leads to an increase in the droplet adhesion to the substrate and a corresponding reduction in

56

droplet mobility. The increase in the droplet adhesion is also associated with an increase in the

57

droplet contact angle hysteresis, which is the difference in the advancing and the receding

58

contact angles of the droplet. The Wenzel state with its high hysteresis is especially favored by a

59

reduction in the liquid surface tension15,16 and increase in the ambient humidity.17,18 For practical

60

applications, high operation temperature − with the associated reduction in water surface tension

61

− and saturated conditions are very relevant. Most of the previous reports on condensation

ACS Paragon Plus Environment

3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

62

experiments performed on superhydrophobic surfaces have focused on using ideal environments

63

of environmental scanning microscope6,14,19-22 or condensation tests in unsaturated and room

64

temperature conditions in the absence of any vapor shear flow5,21.5,21 In addition, these works

65

involve condensation of humid air (or gas). In our study, we deliberately avoided presence of any

66

non-condensable gases, which are known to affect condensation based thermal transport across a

67

solid surface. In addition, the majority of these studies have used top-down microfabrication as a

68

means to obtain the nanotextured superhydrophobic surfaces. Recently, a novel approach was

69

introduced, in which an immiscible liquid was impregnated in to a nanotextured solid surface. 23

70

With carefully selected immiscible liquids, having desirable surface energy and spreading

71

coefficient with respect to water, the authors reported remarkable increase in the mobility and

72

decrease in the size of the shedding, condensed water droplets. These works have produced

73

valuable results regarding drop formation and mobility, but only few 14,24 have actually focused

74

on enhanced heat transfer rates. Miljkovic et al.14 showed that the ‘partial wetting’ state of

75

droplets, i.e. a state which is intermediate between the adhesive Wenzel and the mobile Cassie

76

states19, is best suited to improve thermal transport on superhydrophobic surfaces. On the other

77

hand, Zhong et al.24 reported an underperformance of a superhydrophobic oxidized copper

78

surface compared to a mirror polished one. The thermal resistance of the oxidized surface layer

79

and degradation of polished metallic copper due to spontaneous oxidation (as is typical for

80

metallic copper) can lead to substantial degradation in thermal properties of a substrate. To the

81

best of our knowledge, no previous work has investigated the role of vapor flow on drop

82

condensation on superhydrophobic surfaces. Apart from its obvious relevance in practical

83

applications, the inherent unique feature of such flow condensation tests is that vapor shear can

84

facilitate droplet mobility on a surface. In addition, as can be inferred from a broad body of

ACS Paragon Plus Environment

4

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

85

classic literature dealing with flow condensation on non-superhydrophobic surfaces, 25-28 it also

86

augments thermal transport. A practical new challenge of using the vapor shear is the

87

requirement of mechanical stability of the nanostructures in the surface texture when subjected to

88

mechanical stress from the shear flow. Mechanical stability also influences the life-time of a

89

surface in operation. These issues are directly addressed in this work through flow condensation

90

experiments on nanoengineered superhydrophobic copper surfaces. The experimental results are

91

used to analyze the condensation mechanism, the thermal transport and the role of vapor flow

92

rate thereof. The choice of copper as substrate and wet chemical nanotexturing is motivated by

93

the prevalence of copper in real life applications, which seek to exploit its high thermal

94

conductivity. We use technically relevant test conditions of saturated vapor and high temperature

95

(110°C) during which penetration of droplet liquid interface into the surface texture occurs. This

96

minimizes the thermal resistance. We then exploit the vapor shear effect to enhance the droplet

97

mobility. A wet-chemical fabrication process was used to nanostructure the copper surfaces by

98

oxidation and nanowire formation29, followed by surface functionalization to render them

99

superhydrophobic. The surfaces were then evaluated using flow condensation tests at various

100

vapor flow rates. The surfaces sustained dropwise flow condensation for five consecutive days of

101

tests. This is an important first step for the potential viability of these nanostructured surfaces in

102

practical applications given that a polished copper sample got readily oxidized after only a day of

103

exposure to our test conditions and showed filmwise condensation. Therefore, the nanotextured

104

superhydrophobic surfaces demonstrate a 5 fold improvement in their life time with respect to

105

sustenance of dropwise condensation compared to unprotected and uncoated copper. The

106

approach also fares well in comparison to other state-of-the-art approaches for different

107

applications. For example, very recently studied graphene coatings for dropwise condensation,

ACS Paragon Plus Environment

5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

108

30

109

n-alkanethiol based self-assembled monolayers is also known to degrade with a similar time

110

frame. 32 Overall, the present results indicate that whereas there are benefits to using

111

nanotextured superhydrophobic samples in flow condensation applications, their durability under

112

realistic conditions must be evaluated through long term tests. This is apparent from their

113

susceptibility to mechanical degradation due to shearing vapor flow, and saturated conditions

114

and temperature involved in flow condensation.

115

MATERIALS AND METHODS

116

yielded a 7 fold improvement corrosion protection, 31 and the corrosion protection offered by

Thermosyphon flow loop

117

A two-phase flow loop operating as thermosyphon was designed to study the condensation

118

phenomenon and determine the heat transfer coefficient. The vapor condensation tests were

119

performed at a saturation temperature of 110 °C and pressure of 1.43 atmospheres. Supporting

120

Information Section I reports the details of the different flow loop components and their control.

121

Special care was taken to avoid any non-condensable gases, which could introduce errors in the

122

measurement of heat transfer coefficients 2. All tests were performed with DI water. For testing

123

samples over consecutive days, the system was kept in overpressure overnight after nearly 10-12

124

hours of sustained testing during the day.

125

Test section

126

The minichannel test section was designed to measure the heat transfer coefficients during vapor

127

flow condensation. The main part of the channel was made out of teflon due to its high thermal

128

stability and very low thermal conductivity (0.23 W m-1 K-1), which helped ensure one

129

dimensional steady state heat conduction through the copper samples. The latter helped in

ACS Paragon Plus Environment

6

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

130

accurate assessment of the heat transfer coefficients reported here. The minichannel was 22 mm

131

wide and 2 mm high and had a length of 230 mm. A glass plate with built in heaters (to avoid

132

vapor condensation and facilitate visualization) was used to cover the channel from top. Each

133

copper specimen had a nominal diameter of 20 mm and a length of 38 mm. On the side of the

134

specimens opposite to the test surface (gray surface in Figure 1a), the copper specimens were in

135

contact with a cooling system in order to dissipate the heat. The water bath 1 in Figure S1 was

136

used for heat removal. The water flow rate was measured using an electromagnetic flow meter,

137

while a thermocouple fitted to the inlet if the minichannel was used to measure the temperature.

138

More details of the cooling system design are provided in the Supporting Information Section II.

139

Each copper specimen was fitted with 5 T-type thermocouples in precisely drilled holes into the

140

sample specimens (see Figure 1a and Supporting Information Section III). The reported heat

141

fluxes were determined from the slope of the measured temperature profile33 (Eq. 1) and the

142

average coefficient of condensation heat transfer was computed using the extrapolated specimen

143

wall temperature (Eq. 2):

144

q = −k

145

h=

146

In Eqs. 1 and 2 k is the copper thermal conductivity, z is the axial position inside the cylindrical

147

copper specimen, Tsat is the measured saturation temperature and Twall is the extrapolated surface

148

temperature of the specimen where condensation takes place. For each heat flux measurement

149

presented in the result section, the flux calculated from Eq. 1 was compared with the one

150

obtained from the energy balance of the cooling system. For the latter, the cooling water flow

dT , dz

(1)

q . Tsat − Twall

(2)

ACS Paragon Plus Environment

7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

151

rate and temperature change were used to compute the caloric heat flux received by the water.

152

For each data point reported (as a function of vapor flow rate and degree of supercooling), a

153

minimum of 60 minutes was allowed in order to let the measured temperature to stabilize and

154

reach a steady state condition. For life time testing, on each day the samples were subjected to a

155

continuous testing for 10-12 hours. Overnight, the flow loop was kept in overpressure in order to

156

avoid any seepage of non-condensable gases. More details on the calibration process are

157

presented in Supporting Information Section III.

a

b

158 159

Figure 1. Test specimen and surface nanotexturing. a) Schematic of the copper made tested

160

samples. The top (gray) surface of the sample was nanotextured and rendered superhydrophobic.

161

b) SEM images showing the development of the nanotexture with increase in solution immersion

162

time. Each scale bar is 200 nm.

163

ACS Paragon Plus Environment

8

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

164

Langmuir

Superhydrophobic copper sample preparation

165

We used a wet chemical approach to develop superhydrophobic surface29, in order to develop an

166

all-copper, low temperature surface treatment. The test surface of each copper specimen was first

167

mechanically polished to a mirror like finish. Then it was ultrasonically cleaned in ethanol (99%

168

Solvent grade) and Millipore DI water for 5 min each.

169

A two step process was followed to form nanostructured superhydrophobic samples. First, the

170

polished copper surface was immersed in a mixture solution containing 2.5 mol L-1 sodium

171

hydroxide (Merck) and 0.1 mol L-1 ammonium persulphate (Sigma Aldrich) at room temperature

172

for 12 min. The solution mixture was continuously stirred with a magnetic stirrer during the

173

entire immersion process. This step served to nanostructure the copper surface through a

174

combination of oxidation and etching.29 Subsequently, the surface was thoroughly rinsed with DI

175

water and dried in a nitrogen stream. In the second step, to impart surface energy change, a self-

176

assembled monolayer was formed by a two hour immersion of the nanostructured surface in an

177

ethanol solution of 1H,1H,2H,2H-perfluorodecanethiol (Sigma Aldrich). The surface

178

functionalization was performed at room temperature and followed by washing in ethanol for

179

one hour.

180

Contact angle measurement and droplet visualization

181

A Data physics OCA20A system was used in backlight configuration15 for contact angle

182

measurements of 5 µL droplets. The sliding angle measurements were performed by stage tilting

183

and recording. The condensation process during each test, was recorded with a Phantom V 9.1

184

high speed camera. The videos were digitized into image sequences for analyzing the departing

185

drop diameters with image processing.

ACS Paragon Plus Environment

9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

186

RESULTS AND DISCUSSION

187

Surface Characterization

188

The scanning electron microscope (SEM) images shown in Figure 1b capture the surface

189

morphology of the samples obtained by varying the immersion time during the first step. As

190

reported first by Chen et al. 29, clearly the oxidation and etching lead to formation of Cu(OH)2

191

nanowires, which gradually cover the entire surface of the copper substrate. The nanowires are

192

crucial to form the rough superhydrophobic surface. However, the thermal conductivity of

193

Cu(OH)2 is one order of magnitude lower than that of copper. Therefore, it was important to find

194

the optimal solution immersion time that yielded superhydrophobicity with minimal degradation

195

in the substrate thermal resistance due to the increased thickness of Cu(OH)2 nanowire layer.

196

Clearly, the nanostructuring is rapid and after only 30 seconds some nanowires already appear in

197

the surface. Increasing the immersion time results in an increase of the nanowire length and in

198

uniformity of the surface coverage. After 15 minutes the surface is uniformly covered with

199

nanowires whose length is almost the same as at 450 seconds.

200

At the end of the optimization step, a 12 minute immersion time was found sufficient for

201

complete surface coverage with a layer of nanowires and also resulted in a superhydrophobic

202

sample behavior (see the next subsection) after functionalization. Figure 2 shows the

203

morphology of the resulting surface. Determining the optimal immersion time was also crucial

204

for the mechanical durability of the nanostructured surfaces. In fact, a surface immersion time of

205

20 minutes led to a fragile nanowire layer, which could be readily removed through gentle

206

rubbing of the surface. Mechanical strength is crucial for stability against vapor shear, which is

207

the main focus of the current work. The nanowire layer with 12 minute immersion had sufficient

208

mechanical strength to easily withstand the range of vapor shear rates tested here.

ACS Paragon Plus Environment

10

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

209 210

Figure 2. SEM images of the copper surface structures at 12 min immersion time.

211

Contact angle measurements

212

The initial polished copper surface was hydrophilic with a contact angle of 86±2° (Figure 3a).

213

The contact angle changed significantly upon nanostructuring. Before surface functionalization,

214

the nanostructured samples exhibited a contact angle less than 5°, characteristic of

215

superhydrophilic surfaces. However, after the functionalization with the fluorosilane, the surface

216

free energy of the samples was reduced drastically resulting in a contact angle of 159°±2°

217

(Figure 3b) and a sliding angle of less than 2°. The latter is the resolution of the stage used for

218

sliding angle measurement since it was nearly impossible to place a droplet on the surface

219

without it rolling-off.

220 221

ACS Paragon Plus Environment

11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

222 223

Figure 3. Change in wettability with nanostructuring and functionalization. Water droplets (5

224

µL) on the initial polished copper sample (a) and on nanostructured and functionalized

225

superhydrophobic copper sample (b).

226

Velocity characterization

227

In our setup the vapor mass flow rate could be adjusted by changing the electrical power

228

supplied to the boiling chamber. A fine regulation of the needle valve enabled the control on the

229

vapor flow and its stabilization. The vapor mass flow rate can be obtained from an energy

230

balance of the boiling chamber as

231

& vapor = m

232

where hL,sat and hL,sub are respectively the enthalpies of saturated and subcooled water, at the

233

temperature and pressure measured at the boiling chamber inlet and hLV is the corresponding

234

latent heat of vaporization. In Eq. (3), Qbc is the electric power supplied to the boiling chamber,

235

which was measured with a NORMA AC Power Analyzer D 5255 S. The above calculation

236

neglects the heat losses to the ambient in the part of flow loop extending from the boiling

237

chamber to the test section. This is justified given the fact that we used good foam insulation. A

238

simple estimation of the heat losses using the correlation for natural convection yields a heat loss

( hL,sat

Qbc , − hL,sub ) + hLV

(3)

ACS Paragon Plus Environment

12

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

239

of ~2%, thus justifying our approach to neglect the heat loss while calculating the vapor mass

240

flow rate and velocity. A detailed error analysis for the heat flux and heat transfer coefficient

241

measurements are provided as a separate section in Supporting Information Section VI.

242

The test section was vertically oriented with a downward vapor flow. Therefore, in our setup the

243

vapor shear and the gravitational forces acted together to aid condensate droplet removal. The

244

gravity effect plays a role in droplet removal only if the drop diameter is comparable to the liquid

245

capillary length34

246

κ −1 =

247

where σLV is the water surface tension, ρL is the condensate density and g is the acceleration due

248

to gravity. For water at 1.43 bars (i.e. the test condition in our experiments), the capillary length

249

comes out to be 2.45 mm. The size of departing droplets on nanotextured superhydrophobic

250

surfaces was ~mm, therefore, in our experiments both gravity and the vapor shear induce the

251

droplet mobility on the surfaces.

252

Influence of vapor shear rate

253

The influence of the vapor shear on nanostructured, superhydrophobic copper surfaces was

254

analyzed first. The saturated vapor entered the test section at a controlled temperature of 110 °C.

255

The temperature of the superhydrophobic copper surface was tuned by varying the inlet

256

conditions in the cooling water system on the back side of the samples. Figures 4 and 5 show the

257

variation of heat flux q and the heat transfer coefficient h with the degree of solid surface

258

subcooling i.e. the difference between the water saturation and the solid surface temperatures. To

σ LV , ρV g

(4)

ACS Paragon Plus Environment

13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

259

ensure the accuracy of results, for each measurement the error bars were calculated from a

260

detailed uncertainty analysis described in the Supporting Information Section VI. Note the error

261

bars in heat flux measurements are too small to be visible. For each data point, first the

262

agreement between the calculated (from the measured pressure) and the directly measured

263

saturation temperatures (at the test section inlet) was checked. The measurements were recorded

264

only if the disagreements were within the measurement uncertainty. Once the agreement was

265

verified, the steady state conditions were reached within approximately one hour. Three different

266

vapor shear rates were tested with vapor velocities of 6, 12 and 18 m s-1. For all these

267

measurements, vapor condensed on the nanostructured surfaces in the dropwise mode. As shown

268

in Figure 4, the heat flux increases both with increase in the wall subcooling and the flow rate.

269

This is due to reduction in the size of the departing droplets with increase in the flow rate. The

270

vapor shear imposes a drag force on a condensing droplet. When the drag overcomes the

271

capillary adhesion of the droplet on the surface, the droplet will move. A simple scaling analysis

272

for this concept is discussed in more detail in the Supporting Information Section IV. Clearly, the

273

increase in the shear rate should result in a reduction in the departing droplet size, thereby

274

increasing the rate of thermal transport8,25.

ACS Paragon Plus Environment

14

Page 15 of 30

700

600

Heat flux [kW m-2]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

500

400

300

200

12_03_12 =18G15 m ss-1-1 VVvapor m vapor=18 12_03_12 =12G10 m ss-1-1 VVvapor m vapor=12

100

13_03_12 G5 =6 m m ss-1-1 VVvapor vapor=6

0 0

5

10

15

Tsat-Twall [K] 275 276

Figure 4. Heat flux as a function of wall subcooling during DWC on the superhydrophobic

277

copper sample at three different vapor velocities.

278

The heat transfer coefficient exhibits a different behavior: It remains rather insensitive (in

279

particular at higher subcooling) to increase in subcooling and seems to depend strongly on the

280

vapor velocity (Figure 5). The latter finding clearly underpins the beneficial influence of vapor

281

velocity on the thermal performance of the nanostructured surfaces. Note that in our

ACS Paragon Plus Environment

15

Langmuir

282

measurements the benefits of the vapor velocity are not easily quantifiable at subcooling below 3

283

K, where the measurements for different vapor velocities display larger measurement error bars.

80

70

60 h [kW m-2 K-1]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

50

40

30 Vvapor =18G15 m s-1 12_03_12

20

Vvapor =12G10 m s-1 12_03_12 10

Vvapor =6 G5 m s-1 13_03_12

0 0

5

10

15

Tsat-Twall [K] 284 285

Figure 5. Heat transfer coefficient dependence on the wall subcooling during DWC on a

286

superhydrophobic copper sample at three different vapor velocities.

287 288

ACS Paragon Plus Environment

16

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

289

Lifetime test

290

In order to test their mechanical and chemical robustness, the nanostructured superhydrophobic

291

copper samples were tested for several days. In Figure 6, the heat flux as a function of the wall

292

subcooling is shown for measurements performed over six consecutive days. All the

293

measurements in the figure were obtained at a saturation temperature of 110 °C (1.43

294

atmospheres) and a vapor velocity of 12 m s-1. For the first five days, there are no signs of

295

degradation in the surface performance and the heat flux values are reproduced at different

296

degrees of subcooling. A clear difference in the heat flux trend is observed on the sixth day: At

297

the same subcooling the heat flux is reduced by 35% with respect to the previous days. This is

298

explained by the change in the mode of condensation resulting in the surface almost entirely

299

being covered by a film of the condensate. The decrease in the thermal performances appears to

300

be the result of the deterioration of the hydrophobic monolayer35 combined with the gradual

301

mechanical degradation of the surface (see SEM images in Figure 7). The latter effect results

302

from the sustained exposure to measurement temperatures up to ~110°C and high vapor velocity.

303

Figure 7 shows the SEM images, at two different magnifications, of the sample after five days of

304

testing. Comparing the surface morphology in Figure 7 with the ones reported before in Figure 2,

305

the surface degradation is clear. In fact, after the sustained tests, although some nanotexturing

306

does seem to remain, there is no sign of the nanowires.

ACS Paragon Plus Environment

17

Langmuir

700 Day1 Day2

600

Day3

Heat flux [kW m-2]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

500

Day4 Day5

400

Day6

300

200

100

0 0

5

10 Tsat-Twall [K]

15

20

307 308

Figure 6. Thermal performance of a nanostructured superhydrophobic copper sample over 6

309

consecutive days of condensation tests at 110 °C and 12 m s-1vapor velocity. The condensation

310

mode switched from dropwise to film on the sixth day.

ACS Paragon Plus Environment

18

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

311 312

Figure 7. Low and high magnification SEM images of the superhydrophobic copper sample after

313

five days of experiments.

314

Comparison with non-textured copper

315

For comparison purposes, tests were also performed on a non-textured oxidized copper surface.

316

To obtain oxidized copper surfaces, cleaned, mirror polished copper samples were fitted into the

317

test section and exposed to a sustained vapor flow of 12 m/s at ~110°C temperature. After

318

several hours of exposure, the color of the sample changed and gradually the mode of vapor

319

condensation switched to film wise condensation. This is expected given the fact that copper is

320

susceptible to oxidation and corrosion, which make the surface treatment worthwhile in the first

321

place. The heat flux in filmwise mode on the oxidized, non-textured copper sample was recorded

322

only after the measured temperatures stabilized, which happened after a day of exposure to the

323

present test conditions. The corresponding heat flux variation on the oxidized polished copper

324

with filmwise condensation and a nanotextured superhydrophobic sample with dropwise

325

condensation are plotted in Figure 8. The reason to compare the filmwise condensation on pure

326

copper with dropwise condensation on nanostructured copper sample is to show the expected

327

advantage of using nanostructured surfaces as opposed to the polished copper which only

ACS Paragon Plus Environment

19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

328

sustained dropwise condensation for several hours and at most for a day of tests. The lines are

329

data fits. The figure shows data for days 1-5 and 6 for the superhydrophobic sample. Note that

330

since the measurements on first 5 days were nearly overlapping with each other (see the above

331

discussion in the context of the Figure 6), we only show the data for Day 1 in Figure 8 and label

332

it as Day1-5. Clearly, for the first five days of tests under the same operation conditions (vapor

333

velocity and pressure), the thermal performance of the superhydrophobic sample is better than

334

the oxidized copper sample. At subcooling below 3 K, the two surfaces display no differences in

335

the heat flux values. However, with decrease in the surface temperature the superior performance

336

of superhydrophobic nanostructured surface becomes clear. This is due to the increase of the

337

condensate layer thickness which causes an augmentation of the thermal resistance and so a

338

decrease in the thermal transport in film condensation. In addition, a heat flux as high as 600

339

kWm-2 is measured with 10K subcooling on nanotextured superhydrophobic surface (see Figure

340

4), which is comparable the values reported in the literature on non-textured functionalized

341

surfaces24. The five days of sustained dropwise condensation on superhydrophobic

342

nanostructured samples shows an encouraging first result in terms of the potential application of

343

these surfaces in intense flow condensation applications, which are practically most relevant. The

344

polished copper surfaces degraded after only a day due to oxidation. This result shows a nearly 5

345

fold improvement in the lifetime of stable dropwise condensation on nanostructured copper

346

surfaces as compared to the untreated copper. In Figure 8, we report the measurements on

347

superhydrophobic sample on day 6, which clearly show degradation in the thermal performance.

348

Visual observations confirmed that this coincided with switching of condensation mode to film

349

condensation. Note also that an as prepared surface, without being subjected to condensation test,

350

showed no degradation in the superhydrophobicity even after a shelf storage of over 6 months

ACS Paragon Plus Environment

20

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

351

(tested through contact angle and sliding angle measurements). This shows that practically

352

relevant flow condensation testing under saturated conditions at high temperature can degrade an

353

apparently robust superhydrophobic surface (see Figure 7). Therefore, while developing

354

improved methods and coatings for condensation enhancement is needed and should be

355

encouraged, the longer-term flow condensation performance under realistic conditions must be

356

studied in parallel. This will help place the results into proper perspective, before novel

357

superhydrophobic surfaces, or for that matter any surface treatment, could be hailed as

358

exceptional and robust means for promotion of dropwise condensation.

359

It is also important to point out that thermal performance of oxidized copper becomes better than

360

the superhydrophobic surface after day 6 (see Figure 8). The reason for this is the higher thermal

361

resistance of the nanostructured superhydrophobic surface compared to the non-textured

362

oxidized copper surface.

ACS Paragon Plus Environment

21

Langmuir

363

700 FWC, Oxidized copper DWC, Superhydrophobic copper Days1-5

600

DWC

FWC, Superhydrophobic copper Day6

500 Heat flux [kW m-2]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

FWC

400 300 200 100 0 0

2

4

6

8

10

12

14

16

Tsat-Twall [K]

364

Figure 8. Comparison of the heat fluxes during condensation on a nanotextured

365

superhydrophobic and a naturally oxidized polished copper sample. Both the measurements were

366

performed at 110 °C saturation temperature and 12 m s-1 vapor velocity.

367

Visualization and analysis of dropwise condensation

368

The digitized images sequences from high-speed video recording of the condensation process

369

were used to gain insight into the condensation process on the nanostructured superhydrophobic

370

copper samples. The image sequence shown in the Figure 9 captures the dropwise condensation

371

on a sample subjected to vapor velocity of 12 ms-1 and at a subcooling of 2K. The condensation

ACS Paragon Plus Environment

22

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

372

process was found to be cyclic in nature characterized by drop nucleation, growth and departure.

373

The cyclic condensation process was observed to repeat itself with a frequency that depended

374

upon the surface temperature and the vapor velocity. The cycle started with nucleation of small

375

droplets (0 ms), which grew and coalesced together with time due to continuous vapor

376

condensation. During coalescence the center of mass of the merging drops did not change

377

appreciably before and after coalescence (110-900 ms). After growing to a critical diameter,

378

drops became mobile (900-975ms) and departed downstream (top to bottom in the images shown

379

in Figure 9). The departing drops swept over the surface, thereby wiping small droplets in their

380

path. After the drop sweep, fresh drops started to nucleate (975 ms) and grow, allowing the

381

cyclic process to continue.

382

ACS Paragon Plus Environment

23

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

383 384

Figure 9. Droplet cycle on a superhydrophobic copper sample at 2 K of surface subcooling and a

385

vapor velocity of 12 m s-1. The arrow indicates the vapor flow direction.

386

The effect of the vapor shear rate during DWC is highlighted in Figure 10, which shows the

387

difference in the condensation process on the a superhydrophobic copper sample at the same

388

surface subcooling of 3.5K, but at three different vapor shear velocities of 6, 12 and 18 m s-1.

389

Clearly, with increase in the vapor velocity, the size of the departing droplets becomes smaller

390

(c.f. explanation in Supporting Information Section IV). A close inspection and analysis of the

391

high speed videos of the tests in Figure 10 it was observed that the droplet departure frequency

392

was also observed to increase with vapor shear. It is well established in the literature that even at

ACS Paragon Plus Environment

24

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

393

equal surface coverage, the heat transfer rate increases with decreasing average departing drop

394

diameter. 2,36 This is due to the lower thermal resistance of the smaller drops. Therefore, it

395

becomes obvious that increasing the vapor velocity, with the resulting decrease in drop departure

396

diameter, contributes to enhance the thermal performance surfaces (see Figures 4 and 5).

397 398

Figure 10. Effect of vapor shear velocity on droplet size during DWC on a superhydrophobic

399

copper sample. Surface subcooling was 3.5 K. Videos of the phenomenon at 18 m s-1 is provided

400

in the Movie.

401

For further insight into the experimental results presented above, we note that in our test

402

conditions of 110°C, the water surface tension is ~ 56 mN/m37. In addition, our experiments are

403

performed with pure vapor in saturated conditions. Both these facts (low surface tension15,16 and

404

saturation17,18) are known to favor the Wenzel state of droplets on superhydrophobic surfaces,

405

which is characterized by the liquid meniscus penetrating the surface texture. The recent

406

literature concerning the use of on superhydrophobic surfaces has mostly focused on experiments

407

in environmental scanning microscope6,14,19-22 and/or condensation tests in the absence of

ACS Paragon Plus Environment

25

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 30

408

external flow, performed in unsaturated and room temperature conditions5,21. Water surface

409

tension in these studies should be 72 mN/m or higher, reaching as high as 76 mN/m in the

410

conditions used in ref.14, thereby favoring the Cassie state and its associated droplet mobility.

411

However, the thermal resistance could undesirably increase due to the gas/vapor trapped

412

underneath the droplets in the Cassie state. To this end, the Wenzel state with its associated

413

liquid penetration in to the surface asperities is beneficial to the thermal transport since it avoids

414

the thermal resistance of the trapped gas. The current work is the first attempt at using a scalable,

415

all-copper based nanoengineered superhydrophobic surfaces under realistic and intensive

416

condensation conditions, and evaluating the simultaneous effect of vapor shear velocity on

417

dropwise condensation and thermal transport. We refer to our approach as scalable in order to

418

indicate that the wet chemical approach used for nanostructuring can be easily applied to larger

419

size surfaces than those used in our study. The results clearly indicate that nanotextured

420

superhydrophobic surfaces should be evaluated through prolonged flow condensation tests due to

421

their vulnerability to mechanical degradation under vapor shear flow.

422 423

CONCLUSIONS

424

In conclusion, we report the results of flow condensation experiments on nanoengineered

425

superhydrophobic “all-copper” surfaces. A wet-chemical fabrication process was used to

426

nanotexture a copper surface by forming nanowires; the resulting surface was superhydrophilic.

427

Surface functionalization introduced superhydrophobicity to the surfaces. The resulting

428

superhydrophobic samples showed a high contact angle of 159° and nearly zero sliding angle at

429

room temperature. Thermal performances of superhydrophobic nanostructured copper and

430

oxidized polished copper samples were compared via flow condensation tests at 110 °C

ACS Paragon Plus Environment

26

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

431

saturation temperature. The high temperature and saturated vapor caused the condensate

432

droplets on the superhydrophobic surfaces to primarily appear in the Wenzel state. However, the

433

vapor-induced shear forces could clearly improve droplet mobility and reduce the critical size of

434

the departing droplets, thereby enhancing the thermal transport on the nanostructured

435

superhydrophobic surfaces. A heat flux of ~600 kWm-2 was recorded for a vapor shear velocity

436

of 18 ms-1 at 10K wall subcooling. The superhydrophobic samples sustained dropwise

437

condensation for five days of testing. Overall, the present results indicate that, while using

438

superhydrophobicity concepts to develop novel coatings and surfaces for enhanced condensation

439

applications is worthwhile and should indeed be promoted, heralding the superior performance

440

of such surfaces must be preceded by flow condensation tests in realistic high temperature and

441

saturation conditions.

442

ASSOCIATED CONTENT

443

Supporting Information. Detailed description of the thermosyphon flow loop, test section,

444

surface temperature determination, drop size dependence on vapor velocity, measurement

445

sensors, uncertainty estimation procedure and a video showing the shedding of condensed

446

droplets. This material is available free of charge via the Internet at http://pubs.acs.org.

447

AUTHOR INFORMATION

448

Corresponding Authors

449

*Dr. Manish K Tiwari, Email: [email protected]; *Prof. Dimos Poulikakos, E-mail:

450

[email protected]

451

Author Contributions

ACS Paragon Plus Environment

27

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 30

452

All authors were involved in problem conception. DT designed the experimental setup, prepared

453

the surfaces and performed the measurements with direct input and mentoring by MKT. DP and

454

DDC provided overall guidance for all the scientific elements of the research. All authors

455

participated in writing the paper and discussion of the results.

456

ACKNOWLEDGMENT

457

We gratefully acknowledge financial support from the Eng. Aldo Gini fellowship. The authors

458

thank Anja Huch (EMPA, Zurich) for help with the contact angle measurements, Tobias

459

Bachmann (LTNT student, ETH Zurich) for the SEM images and Peter Kocher (MATL, ETH

460

Zurich) for the surface polishing.

461

ABBREVIATIONS

462

SEM: Scanning Electron Microscope; DWC: dropwise condensation; FWC: filmwise

463

condensation.

464

REFERENCES

465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482

1. Schmidt, E.; Schurig, W.; Sellschopp, W., Versuche über die kondensation von wasserdampf in film- und tropfenform. Forschung im Ingenieurwesen 1930, 1, (2), 53-63. 2. Rose, J. W., Dropwise condensation theory and experiment: A review. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy 2002, 216, (2), 115128. 3. Rainieri, S.; Bozzoli, F.; Pagliarini, G., Effect of a hydrophobic coating on the local heat transfer coefficient in forced convection under wet conditions. Experimental Heat Transfer 2009, 22, (3), 163-177. 4. Chen, C.-H.; Cai, Q.; Tsai, C.; Chen, C.-L.; Xiong, G.; Yu, Y.; Ren, Z., Dropwise condensation on superhydrophobic surfaces with two-tier roughness. Applied Physics Letters 2007, 90, (17), 173108. 5. Boreyko, J. B.; Chen, C.-H., Self-Propelled dropwise condensate on superhydrophobic surfaces. Physical Review Letters 2009, 103, (18), 184501. 6. Dietz, C.; Rykaczewski, K.; Fedorov, A. G.; Joshi, Y., Visualization of droplet departure on a superhydrophobic surface and implications to heat transfer enhancement during dropwise condensation. Applied Physics Letters 2010, 97, (3), 033104. 7. Kim, S.; Kim, K. J., Dropwise condensation modeling suitable for superhydrophobic surfaces. Journal of Heat Transfer 2011, 133, (8), 081502.

ACS Paragon Plus Environment

28

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528

Langmuir

8. Graham, C.; Griffith, P., Drop size distributions and heat transfer in dropwise condensation. International Journal of Heat and Mass Transfer 1973, 16, (2), 337-346. 9. Dorrer, C.; Rühe, J., Condensation and wetting transitions on microstructured ultrahydrophobic surfaces. Langmuir 2007, 23, (7), 3820-3824. 10. Narhe, R. D.; Beysens, D. A., Nucleation and growth on a superhydrophobic grooved surface. Physical Review Letters 2004, 93, (7), 076103. 11. Varanasi, K. K.; Hsu, M.; Bhate, N.; Yang, W.; Deng, T., Spatial control in the heterogeneous nucleation of water. Applied Physics Letters 2009, 95, (9), 094101. 12. Feng, J.; Qin, Z.; Yao, S., Factors affecting the spontaneous motion of condensate drops on superhydrophobic copper surfaces. Langmuir 2012, 28, (14), 6067-6075. 13. Lafuma, A.; Quere, D., Superhydrophobic states. Nature Materials 2003, 2, (7), 457-460. 14. Miljkovic, N.; Enright, R.; Wang, E. N., Effect of droplet morphology on growth dynamics and heat transfer during condensation on superhydrophobic nanostructured surfaces. ACS Nano 2012, 6, (2), 1776-1785. 15. Tiwari, M. K.; Bayer, I. S.; Jursich, G. M.; Schutzius, T. M.; Megaridis, C. M., Highly liquid-repellent, large-area, nanostructured poly(vinylidene fluoride)/poly(ethyl 2-cyanoacrylate) composite coatings: Particle filler effects. ACS Applied Materials & Interfaces 2010, 2, (4), 1114-1119. 16. Yuyang, L.; Xianqiong, C.; Xin, J. H., Can superhydrophobic surfaces repel hot water? Journal of Materials Chemistry 2009, 19, (31). 17. Yin, L.; Zhu, L.; Wang, Q.; Ding, J.; Chen, Q., Superhydrophobicity of natural and artificial surfaces under controlled condensation conditions. ACS Applied Materials & Interfaces 2011, 3, (4), 1254-1260. 18. Jung, S.; Tiwari, M. K.; Doan, N. V.; Poulikakos, D., Mechanism of supercooled droplet freezing on surfaces. Nature Communications 2012, 3. 19. Rykaczewski, K.; Scott, J. H. J., Methodology for imaging nano-to-microscale water condensation dynamics on complex nanostructures. ACS Nano 2011, 5, (7), 5962-5968. 20. Rykaczewski, K.; Scott, J. H. J.; Rajauria, S.; Chinn, J.; Chinn, A. M.; Jones, W., Three dimensional aspects of droplet coalescence during dropwise condensation on superhydrophobic surfaces. Soft Matter 2011, 7, (19), 8749-8752. 21. Chen, X. M.; Wu, J.; Ma, R. Y.; Hua, M.; Koratkar, N.; Yao, S. H.; Wang, Z. K., Nanograssed micropyramidal architectures for continuous dropwise condensation. Advanced Functional Materials 2011, 21, (24), 4617-4623. 22. Rykaczewski, K.; Landin, T.; Walker, M. L.; Scott, J. H. J.; Varanasi, K. K., Direct imaging of complex nano- to microscale interfaces involving solid, liquid, and gas phases. ACS Nano 2012, 6, (10), 9326–9334. 23. Anand, S.; Paxson, A. T.; Dhiman, R.; Smith, J. D.; Varanasi, K. K., Enhanced condensation on lubricant-impregnated nanotextured surfaces. ACS Nano 2012, Article ASAP. 24. Zhong, L.; Xuehu, M.; Sifang, W.; Mingzhe, W.; Xiaonan, L., Effects of surface free energy and nanostructures on dropwise condensation. Chemical Engineering Journal 2010, 156, (3), 546-552. 25. Tanasawa, I.; Utaka, Y., Measurement of condensation curves for dropwise condensation of steam at atmospheric-pressure. Journal of Heat Transfer-Transactions of the ASME 1983, 105, (3), 633-638. 26. Tanasawa, I.; Ochiai, J.-I., Experimental study on dropwise condensation. Bull JSME 1973, 16, (98), 1184-1197.

ACS Paragon Plus Environment

29

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554

Page 30 of 30

27. Takeyama, T.; Shimizu, S., On the transition of dropwise-film condensation. 1974, 274278. 28. Yamali, C.; Merte Jr, H., Influence of sweeping on dropwise condensation with varying body force and surface subcooling. International Journal of Heat and Mass Transfer 1999, 42, (15), 2943-2953. 29. Chen, X.; Kong, L.; Dong, D.; Yang, G.; Yu, L.; Chen, J.; Zhang, P., Synthesis and characterization of superhydrophobic functionalized Cu(OH)2 nanotube arrays on copper foil. Applied Surface Science 2009, 255, (7), 4015-4019. 30. Rafiee, J.; Mi, X.; Gullapalli, H.; Thomas, A. V.; Yavari, F.; Shi, Y. F.; Ajayan, P. M.; Koratkar, N. A., Wetting transparency of graphene. Nature Materials 2012, 11, (3), 217-222. 31. Prasai, D.; Tuberquia, J. C.; Harl, R. R.; Jennings, G. K.; Bolotin, K. I., Graphene: corrosion-inhibiting coating. ACS Nano 2012, 6, (2), 1102-1108. 32. Laibinis, P. E.; Whitesides, G. M., Self-assembled monolayers of n-alkanethiolates on copper are barrier films that protect the metal against oxidation by air. Journal of the American Chemical Society 1992, 114, (23), 9022-9028. 33. Del Col, D.; Webb, R. L.; Narayanamurthy, R., Heat transfer mechanisms for condensation and vaporization inside a microfin tube. Journal of Enhanced Heat Transfer 2002, 9, (1), 25-37. 34. De Gennes, P. G.; Brochard-Wyart, F.; Quéré, D., Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves. Springer: 2004. 35. Frechette, J.; Maboudian, R.; Carraro, C., Thermal behavior of perfluoroalkylsiloxane monolayers on the oxidized Si(100) surface. Langmuir 2006, 22, (6), 2726-2730. 36. Rose, J. W.; Glicksman, L. R., Dropwise condensation—The distribution of drop sizes. International Journal of Heat and Mass Transfer 1973, 16, (2), 411-425. 37. Vargaftik, N. B.; Volkov, B. N.; Voljak, L. D., International tables of the surface-tension of water. Journal of Physical and Chemical Reference Data 1983, 12, (3), 817-820.

555 556

ACS Paragon Plus Environment

30