International Journal of Thermal Sciences xxx (2011) 1e7
Contents lists available at ScienceDirect
International Journal of Thermal Sciences
journal homepage: www.elsevier.com/locate/ijts
Effects of nanowire height on pool boiling performance of water on silicon chips
Z. Yao a, Y.-W. Lu b, S.G. Kandlikar c, *
a
Rochester Institute of Technology, Rochester, NY, USA
National Taiwan University, Taipei, Taiwan
c
Mechanical Engineering Department, Rochester Institute of Technology, Rochester, NY 14623, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 March 2011
Received in revised form
17 June 2011
Accepted 18 June 2011
Available online xxx
A new technique is developed to directly grow Cu nanowire (CuNW) on Si substrate with electrochemical deposition to produce height-controlled hydrophilic nanowired surfaces for enhancing pool
boiling performance. For broader heat transfer applications, CuNW and Si nanowires (SiNW) with various
nanowire heights were fabricated and examined under pool boiling with water. The heat transfer
performance of the samples with NW arrays is enhanced with increasing NW heights regardless of
the NW materials. The surface with the tallest NW structure (35 mm-tall SiNW) yielded a heat flux of
134 W/cm2 at 23 K wall superheat, about 300% higher than a plain Si surface at the same wall superheat.
Ó 2011 Elsevier Masson SAS. All rights reserved.
Keywords:
Nanowire
Pool boiling
Wettability
Direct growth
1. Introduction
The generation of ultra-high heat flux from high performance
electronic devices has motivated a number of investigations related
to advanced heat transfer, especially in the area of pool and flow
boiling performance. For most integrated circuit and logic chips,
a cooling system is needed to maintain a relatively constant
component temperature below 85 C. Thus, primary issues related
to the chip cooling with pool boiling are the enhancement of
nucleate boiling, increasing the critical heat flux (CHF) and heat
transfer coefficient (HTC).
Active studies have been conducted for several decades on
enhancing the boiling heat transfer by surface modification. The
recent surface treatment methods include increasing the surface
area with micro pin-fins [1], applying wicking structures to
promote the liquid supply by capillary pumping [2,3], and depositing nano-particles or coating with nanomaterials [4e6]. Among
those developments, surface modification by incorporating nanomaterials and nanostructures has proven to be a promising technique for providing a more effective boiling heat transfer surface.
Table 1 gives a summary of recent studies that are related to the
enhancement of boiling heat transfer by use of nanomaterials and
* Corresponding author. Tel.: þ1 585 475 6728; fax: þ1 585 475 7710.
E-mail addresses: [email protected] (Z. Yao), [email protected] (Y.-W. Lu),
[email protected] (S.G. Kandlikar).
nanostructures. Wu [7] reported a coating of hydrophilic titanium
oxide nanoparticles on the heating surface that increased the
critical heat flux (CHF) by 50.4% in pool boiling with FC-72.The
enhancement is attributed to the hydrophilicity of the nanoporous layer. Chen et al [8]studied the boiling performance of
a surface covered with super-hydrophilic titanium oxide nanotube
array. The surfaces yielded approximately half the values of wall
superheats during boiling at a given heat flux compared to the bare
Ti surface. It is concluded that the nanotube array introduced
a large number of active nucleation sites that promoted bubble
generation. Carbon nanotube (CNT) is also studied for pool boiling
applications [9e11]. The CNT coated surface is highly effective in
improving both the CHF and HTC in the low heat flux region due to
an increased surface cavity density and enhanced roughness.
Significant enhancements in both the CHF and the HTC have also
been obtained from surfaces coated with Cu nanorod and nanowire
(NW) arrays [12e14]; the reported CHF (220 W/cm2) for the CuNW
surface is one of the highest for pool boiling heat transfer with
water [14]. The nanowire structures are preferred in pool boiling
enhancement due to their unique properties. The NWs enhance
surface wettability, which helps in increasing CHF and delaying the
dry-out condition [13]. Due to the pin fin effect, NW structures
greatly increase the active heat transfer area. In addition, the NW
height was found to directly affect the heat transfer [15]. By
incorporating the hydrophilic NW structure with hydrophobic
surfaces, the pool boiling performance can be further enhanced
[16]. These studies have shown that most of the surface
1290-0729/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ijthermalsci.2011.06.009
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
2
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
Table 1
Summary of pool boiling enhancement by the surfaces with nanostructures.
Researchers
Heater/working
fluid
Nanostructure
Surface
characterization
Vemuri et al. [6]
Flat plate heater/FC-72
Pore diameter ¼ 50e250 nm
Wu et al. [7]
Cu block/FC-72 and
water
Plane heater/FC-72
Aluminum oxide particle
thin layer
Titanium oxide
nanoparticle coating
CNT coated on Si and
Cu surface
CNT coated on plain Si
and 3D Si micro-structure
MWCNT forest
Cu nanorod array
CNT height ¼ 25 mm
Nanorod height ¼ 450 nm
Ujereh et al. [9]
Launay et al. [10]
Results
Layer thickness ¼ 1 mm
CNT height ¼ 40 mm
CNT height ¼ 40e100 mm
Mechanism/comments
Incipient superheat
reduced by 30%
CHF increased by
50.4% for water
CHF and HTC enhanced
on CNT surface
Moderate enhancement
at low heat flux region
CHF improved by 28%
High heat flux coupling
with low incipience
Ahn et al. [11]
Li et al. [12]
Cu heater/PF5060
and water
Cu heater/H2O
Cu heater/H2O
Chen et al. [14]
Si chip heater/H2O
Cu/Si nanowire
(CuNW, SiNW)
CuNW, SiNW
height ¼ 50 mm
CHF and HTC enhanced
by 100%
Im et al. [15]
Cu heater/FC-72
Cu Nanowire
(CuNW)
CuNW height ¼ 2, 4,
6, 8 mm
200% CHF enhancement
on 2 mm NW
nanostructures were effective in decreasing the wall superheat at
boiling incipience, and enhancing the nucleate boiling heat transfer
and critical heat flux. The factors leading to enhanced boiling
surfaces are the surface micro-roughness and porous nanostructure
which provide sufficient active nucleation sites at low wall superheats, the evaporation of liquid film within a very small confined
space, and the increase in the effective heat transfer area. In addition, the enhanced surface wettability induced by nanostructures
also promotes pool boiling by staving off CHF.
This work is aimed at investigating the boiling enhancement
mechanism of nanostructured surfaces and the effect of NW height on
pool boiling heat transfer with water. Two types of NWs made of the
materials commonly used in boiling application, copper and silicon,
with different heights are utilized to examine their pool boiling
performances. Particularly, both NWs are directly grown on the Si
substrates without any interfacial layer in between, making the two
kinds of NW structures more comparable in the pool boiling performance. Traditionally, for metallic NW made by template-assisted
electro-chemical deposition method, an epoxy layer is needed to
bond NW structure on the testing surface [14,15], which introduces an
additional interfacial layer and becomes troublesome in practical
application. The thermal resistance of a typical epoxy adhesive containing boron nitride fillers is in the range of 0.7e1.6 C/W, and it
increases proportionately as the layer becomes thicker. When
temperature rises, the resistance furthermore increases. The epoxy
adhesive also traps air bubbles and forms voids, creating a non-
Porous structure and
enhanced surface wettability
Enhanced surface wettability
Increased surface cavity density
and enhanced surface roughness
Roughness enhancement
Surface roughness enhancement
Enhanced wettability and
coupling effects of micro/
nanostructure
Capillary effect, enhanced
wettability and active
nucleation sites
Enhanced micro-scale cavities
and surface roughness
uniform surface due to its high viscosity during the application. As
a result, the epoxy layer becomes a major source of uncertainty in the
metallic NW application in boiling and it is hard to make it comparable to other NW structures in pool boiling performance. In this
study, a direct CuNW growth technique is employed to produce
a uniform CuNW array on the Si substrate without an additional
bonding layer (i.e. epoxy layer) in between, thereby eliminating the
interfacial thermal resistance and results in similar testing fixture as
SiNW, which is also directly fabricated on the Si surface.
2. Material and methods
There have been numerous fabrication techniques for creating
nanostructures to alter the surface properties for boiling applications, including the use of lithographic techniques, porous materials, nanotubes, nanowires and nanoparticles [17]. Among them,
the fabrication approach used here mainly resorts to two methods:
(i) nanoporous template-based fabrication for Cu nanowires
(CuNW) and (ii) electro-less Si nanowire (SiNW) etching due to
their excellent properties of being scalable, cost-effective, and able
to produce uniform geometrical features at a large scale.
2.1. Direct growth of CuNW on Si chip
To directly deposit CuNW arrays on a substrate, the NWs were
first synthesized via an electro-chemical deposition described
Fig. 1. (a) The fabrication process of growing CuNW on the substrate: (i) an Ag layer was sputtered, (ii) AAO was hermetically applied onto the substrate, (iii)electro-chemical
deposition was performed, and (iv) AAO is removed. (b) The technique allows the AAO template hermetically contact the substrate.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
Table 2
Deposition conditions for CuNW synthesis.
NW height
Potential
Duration
20 mm
10 mm
5 mm
2 mm
0.9e1.2 V
0.6e0.75 V
0.45Ve0.5 V
0.4 V
900
900
900
600
s
s
s
s
elsewhere [18]. Briefly, Anodized Alumina Oxide(AAO) membranes(Anodisc, Whatman) were employed as the templates to
form the nanowires of 200 nm diameter with 50% porosity. In order
to directly deposit CuNW on the substrate without epoxy bonding,
a novel technique was developed. ASi substrate was first coated
with a 500 nm thick silver layer by sputtering. The silver layer
serves three purposes: (i) it provides the counter electrode for
electro-chemical deposition,(ii) it acts as the seeding layer for
CuNW synthesis and (iii) it reduces the surface roughness for better
contact with the AAO template. Then a liquid film of deionized (DI)
water, whose surface tension provides necessary adhesion forces,
was applied in between the AAO membrane and the silver layer.
The electro-chemical deposition was then conducted in a threeelectrode potentiostat scheme as shown in Fig. 1(a). Platinum (Pt)
gauze was attached onto the AAO membrane and served as the
working electrode, as depicted in Fig. 1(b). The Pt gauze was
employed since it permitted high and stable diffusion flux of ions
migrating through the AAO membrane, creating a uniform NW
growth rate. In addition, the gauze allows the attraction forces to
sandwich the AAO membrane between working and counter
electrodes (e.g. the Pt gauze and substrate). After the electrochemical deposition, the sample was immersed in a NaOH solution to remove the AAO membrane template, and then vacuum
dried.
The free-standing CuNW arrays with different deposition
conditions are detailed in Table 2. The deposition electrical potential
3
and time duration were the major parameters used to control the
height of CuNW. Four different kinds of CuNW samples, with the
average NW height of 20 mm, 10 mm, 5 mm, and 2 mm were
successfully fabricated as shown in Fig. 2(a)e(d). When the NWs
reached 20 mm, micro/nano scale cavities and openings were
observed. These cavities were created during the AAO removal
process, in which the surface tension of the etchant caused the NW
arrays to bundle together. These cavities and openings are desirable
for enhanced boiling due to serving as bubble nucleation sites at low
superheats [19,20] and tunnels for liquid transport. Non-uniform
distributions of the NW heights, caused by the growth rate variation of CuNW across the substrate, were observed and shown in
Fig. 2(b)e(d). The non-uniform NW height distribution became the
surface waviness at a larger scale and may also promote bubble
nucleation. As a result, the samples with longer nanowire arrays had
better boiling performance than the ones with shorter arrays shown
in Fig. 2 (c) and (d), as the former provided more active nucleation
sites and wider passages for liquid transport.
2.2. Si nanowires synthesis
The SiNWs were synthesized by a metal particle-assisted electro
less chemical etching method on B-doped p-type (100) Si wafers in
AgNO3 and HF aqueous solution. The SiNW synthesis was based on
a galvanostatic reaction between Agþ and Si0. Due to the higher
positive redox level of Ag/Agþ, the deposited Agþ was reduced into
Ag0, and consequently, Si was oxidized into SiO2, which was later
dissolved by HF as shown below [21,22]:
Agþ þ e /Ag
(1)
Si þ 2H2 O/SiO2 þ 4Hþ þ 4e
(2)
2
SiO2 þ 2HF
2 þ 2HF/SiF6 þ 2H2 O
(3)
Fig. 2. SEM images of CuNW on Si substrate with average heights of (a) 20 mm, (b) 10 mm, (c) 5 mm and (d) 2 mm.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
4
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
Table 3
Deposition conditions for SiNW synthesis.
NW height
Solution
Duration and temperature
35 mm
20 mm
10M HF with 0.02 M AgNO3
5M HF with 0.02 M AgNO3
60 min at 50 C
60 min at 24 C
Table 3 shows the process conditions of temperature, HF
concentration, and time durations for the SiNW fabrication. The
heights of Si nanowires increased approximately linearly with the
etching time and temperature [23]. Fig. 3(a) and (b) show the SiNW
with heights of 35 and 20 mm. Because of the bundling effect, the
micro-scale cavities were again observed; they increased in both
number and size as the NW height increased.
Due to the differences in their synthesis processes, the morphology
of CuNW is slightly different from that of SiNW. CuNW arrays are more
uniform in dimension as the array spacing and single pore diameter
are constrained by the AAO template. However, the NW height varies
along with the locations due to an unavoidable deposition rate variation, which could result from surface roughness, interfacial contact
intimacy and so on. For SiNW structure, the dimension depends on the
etchant concentration and reaction temperature. A higher AgNO3
concentration and reaction temperature would result in more single
NW arrays and less branch structures. In addition, as the etching is
anisotropic along (100) orientation, no surface waviness is observed
on the SiNW surface, the average height is more uniform than that of
the CuNW.
Fig. 4. Cavity density and maximum size of NW samples with different heights.
are hydrophilic, the surfaces with nanowire structures become
more hydrophilic. The hydrophilic surface is preferred in pool
boiling as it can promote bubble nucleation and increase the CHF by
preventing dry out. At the same time, the availability of larger
cavities with taller micro/nanostructures allow the onset of
nucleate boiling at lower wall superheats.
2.5. Boiling test setup
2.3. Cavity characterization
In order to characterize the surface cavity of samples with
different NW heights, a Matlab program was used to analyze the
cavity density and size based on the SEM images. The SEM images
were first converted to binary by thresholding, and then the
segmentation between the NW area and cavity was made. In this
way, the area percentage and size of cavity can be measured. Fig. 4
shows the cavity density and average size change as NW height
increases. The cavity density of 20 mm CuNW is 58% and for 35 mm
SiNW, it is 65%. The maximum cavity size also increases as NW
height increases, the maximum cavity was found on the 35 mm
SiNW surface, with a size of 2.5 mm. This explains how NW height
affects the boiling performance.
2.4. Surface wettability
Both Cu and Si nanostructured surfaces demonstrated hydrophilic properties. Their contact angles are 28 and 0 respectively
(Fig. 5). As the contact angle of water on the plain Si and Cu surfaces
To perform the boiling test, an experimental setup shown in
Fig. 6 was designed and fabricated [24]. The test samples were
mounted on an insulated block which was sealed around the edges
to prevent water leakage. A 450 W capacity cartridge heater was
used as a heat source. The copper heating section was machined to
provide a 10 mm 10 mm base under the chip. Three K-type
thermocouples, placed along the axis of the copper block, were used
to measure the temperature gradient through the tip. The thermocouples were 8 mm apart and the first one (T1) was 3 mm below
the top copper surface. DI water was used as the boiling liquid. Prior
to each test, the water was boiled for over 30 min in order to reduce
the effects of any dissolved air. Periodically between the tests, more
water was added to the pool for replenishment.
Sufficient time was given to remove dissolved air before
commencing the testing. An auxiliary heater (100 W) was used to
maintain the water temperature in the reservoir at 100 C. The
temperature of the water was monitored by a K-type thermocouple
(T4). After water was kept in saturation temperature for 30 min, the
main heater was started and the power was increased in small
Fig. 3. SEM images of SiNW with average heights of (a) 35 mm and (b) 20 mm.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
5
Fig. 5. Contact angle of (a) Plain Cu surface, (b) Surface with CuNW, (c) Plain Si surface and (d) Surface with SiNW.
increments. The temperature gradient in the copper block was
measured by the three K-type thermocouples. Data was recorded at
every stabilized point. The heat flux was then calculated from the
following equation:
dT
q00 ¼ kcu
dx
(4)
The temperature gradient, dT/dx, was calculated using a threepoint backward-difference Taylor series approximation,
dT
3T1 4T2 þ T3
¼
dx
2Dx
through the copper block, as well as the thermocouple T1 from the
following equation,
TS ¼ T1 q00
LCu
L
þ R00t;c þ Si
KCu
KSi
(6)
In order to reduce the contact resistance between the copper
heating block and the testing chip, a layer of thermal paste was
applied. Rt,c represents the thermal contact resistance of the
interface, which is found to be repeatedly 5 106 m2 K/W, with an
uncertainty less than 4%, as calculated in the previous study [24].
(5)
where T1, T2 and T3 are the temperatures measured by thermocouples located at distances Dx ¼ 8 mm. The surface temperature of
a testing sample, Ts, was obtained by calculating the heat flux
2.6. Uncertainty analysis
The major uncertainties originated from the following aspects:
1) thermocouple calibration accuracy and precision resolution; 2)
Fig. 6. Schematic of boiling test fixture (a) cartridge heater, (b) ceramic block, (c) testing chip, (d) gasket, (e) polycarbonate visualization tube, (f) auxiliary heater, (g) K-type
thermocouples, (h) data acquisition system, (i) compression screws, (j) high speed camera and (k) power supply.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
6
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
Table 4
Pool boiling performance of the surfaces with CuNW and SiNW with different
heights.
Fig. 7. Boiling characterization of the samples with CuNW and SiNW at different
heights.
thermal conductivity of materials being altered under temperature
changes and 3) length measurements, spacing between thermocouples, and thickness of materials. Multiple parameters can lead to
propagation of uncertainty. The method used to find the error
propagation is through partial sums in Eq. (7),
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u n 2
uX vp
Up ¼ t
uai
vai
i¼1
(7)
where p is the calculated parameter, ai is a measured parameter,
and u denotes the uncertainty of the subscripted parameter. For the
boiling setup utilized in this study, the uncertainty of the surface
temperature is 4.5%e4.6%, and the uncertainty of heat flux
increases from 2 W/cm2 to 6 W/cm2 from low to high heat flux as
indicated in the previous study [24]. This is considered to be
acceptable because the sample performance is evaluated at
elevated heat flux.
3. Results
Fig. 7 depicts the boiling curves for the plain Si surface, the SiNW
surface, and the CuNW surface. Boiling on the plain Si substrate
served as the primary control for comparing boiling performance.
The results were consistent with experimental results reported
previously [25]. The NW surfaces are able to provide more active
nucleation sites than the plain Si surface, making more bubbles
generated at a given heat flux, as shown in Fig. 8. In addition, the
incipient wall superheat on the NW surfaces was found to be lower
than that on the plain Si surfaces. Particularly, the wall superheat at
low heat fluxes decreased as the nanowire height increased. This
NW material
NW height
Maximum heat flux
(W/cm2)/superheat (K)
CuNW
CuNW
CuNW
CuNW
SiNW
SiNW
20 mm
10 mm
5 mm
2 mm
35 mm
20 mm
122 W/cm2/24 K
105 W/cm2/29 K
95 W/cm2/33 K
85 W/cm2/39 K
134 W/cm2/23 K
115 W/cm2/25 K
result is in agreement with Hsu’s model [26] which states, for the
sizes considered in this study, the surface superheat decreases as
the cavity size increases. The observed results suggested that
boiling heat transfer was greatly enhanced by NW surfaces, and the
enhancement was further improved by increasing the average NW
heights for both Cu and Si NW samples. Table 4 summarizes the
boiling performance of NW samples with different heights. The
35 mm SiNW sample reached a heat flux of 134 W/cm2 at 23 K wall
superheat, which is among one of the highest recorded for Si
surfaces at this wall superheat. In addition, the CuNW grown on the
Si substrate without an intermediate epoxy layer proves to be
beneficial in pool boiling application as the samples were repeatedly tested, no degradation in performance was observed.
It can also be seen that boiling characteristics of SiNW and
CuNW appear similar to each other although the thermal conductivities of Cu and Si are quite different (400 W/m K for Cu and
150 W/m K for Si at 25 C [27]). This is understandable as the boiling
behavior is more dependent on the surface morphology than the
material properties at micro/nano-scale. However, for a given
height, the CuNW outperforms SiNW. The HTC of 20 mm CuNW is
much higher than that of SiNW at a given wall superheat. The local
heat transfer coefficient of CuNW and SiNW surfaces measured at
25 K superheat are shown and compared with the published data
[14] (Fig. 9). The heat transfer coefficients of the NW surfaces
increases proportionally as NW height increases for both CuNWs
and SiNWs. By incorporating Chen’s results [14] for SiNW and
CuNW with a 50 mm height, the linear relationship between NW
height and the HTC for water pool boiling performance can be
further extended, as shown in Fig. 9. It is observed that the microscale cavities increase in number and size as the NW height
increases, but it is still unclear if the cavities alone are responsible
for the heat transfer enhancement. The enhanced capillary forces
due to higher NW structures could supply the liquid and prevent
dry-out of the heater surface. However, there must be an optimal
NW height beyond which the benefit brought from increasing
height would become limited. Im et al. [15] suggested that the 2 mm
CuNW structure has the best boiling performance for FC-72, and
the higher ones are not preferred due to increased flow resistance
on the surface. For pool boiling with water, although the optimal
NW height is not identified yet, our study suggests that the water
Fig. 8. Bubble images of (a) Plain Si surface, (b) CuNW surface and (c) SiNW surface at w 40 W/cm2 heat flux.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009
Z. Yao et al. / International Journal of Thermal Sciences xxx (2011) 1e7
7
The CuNW array without an intermediate epoxy layer proves to be
beneficial in pool boiling application. It is a simple and compatible
technology for future semiconductor cooling, thermal management,
and high heat flux energy conversion applications. The study of NW
height effects on pool boiling provides a new insight on enhancing
boiling heat transfer more effectively.
Acknowledgement
The work was performed at Microsystems Engineering Program
at RIT and supported by a National Science Foundation EPDT Grant
(#0802100).
References
Fig. 9. Heat transfer coefficients of the samples with CuNW and SiNW (incorporated
with Chen’s results [14]) at given wall superheat (25 K).
pool boiling performance can be enhanced by increasing NW
height to at least 50 mm.
4. Conclusions
This study demonstrates that pool boiling heat transfer in water
can be significantly enhanced by introducing nanowire structures
onto the heating surfaces regardless of NW materials. The CuNW
and SiNW surfaces exhibited hydrophilic properties. During
nucleation boiling, more bubbles were generated at the NW surface
than the plain surface at a given heat flux. The effects of NW height
have been explored, showing that the density and size of surface
cavity increase as the NW height increases, making pool boiling
heat transfer more efficient. The main findings of this study are
summarized as follows:
1. A modified synthesis technique is developed to directly grow
CuNW on Si substrates. The CuNW structure was directly
fabricated on a Si substrate without additional interfacial
bonding layers, which allows us to compare the pool boiling
performance of CuNW and SiNW surfaces under the similar
substrate configurations. With this technique, the durability
and interfacial strength of the CuNW structure are enhanced
while the thermal resistance between the NW array and the
substrate, due to an epoxy layer needed in previous methods, is
eliminated.
2. The average height of CuNWs and SiNWs can be controlled by
different synthesis parameters, and NW height is one of the key
characteristics that influence the pool boiling performance.
3. The testing results suggest that for pool boiling with water on
the nanowired surfaces, the heat transfer can be enhanced by
increasing the NW height. Large numbers of cavities and
openings were observed on surface with higher NW arrays,
which contribute to the enhanced pool boiling performance by
providing more stable active nucleation sites.
4. The optimal result was obtained on a 35 mm SiNW surface,
which yielded a heat flux of 134 W/cm2 at 23 K superheat. This
heat flux is almost 3 times higher than that of the plain Si
surface at the same wall superheat.
Cu and Si are two of the most commonly used materials in
the semiconductor industry. The nanowire synthesis techniques
employed here are inexpensive and ready for large scale fabrication.
[1] J.J. Wei, H. Honda, Effects of fin geometry on boiling heat transfer from silicon
chips with micro-pin-fins immersed in FC-72, Int. J. Heat Mass Transfer 46
(2003) 4059e4070.
[2] S.G. Liter, M. Kaviany, Pool-boiling CHF enhancement by modulated porouslayer coating: theory and experiment, Int. J. Heat Mass Transfer 44 (2001)
4287e4311.
[3] C. Li, G.P. Peterson, Parametric study of pool boiling on horizontal highly
conductive microporous coated surfaces, J. Heat Transfer 129 (2007)
1465e1475.
[4] Y. Takata, S. Hidaka, M. Masuda, T. Ito, Pool boiling on a superhydrophilic
surface, Int. J. Energy Res. 27 (2003) 111e119.
[5] M.E. Davis, Ordered porous materials for. Emerging applications, Nature 417
(2002) 813e821.
[6] S. Vemuri, K.J. Kim, Pool boiling of saturated FC-72 on nano-porous surface,
Int. Commun. Heat Mass Transfer 32 (2005) 27e31.
[7] W. Wu, H. Bostanci, L.C. Chow, Y. Hong, M. Su, J.P. Kizito, Nucleate boiling heat
transfer enhancement for water and FC-72 on titanium oxide and silicon oxide
surfaces, Int. J. Heat Mass Transfer 53 (2010) 1773e1777.
[8] Y. Chen, D.C. Mo, H.B. Zhao, N. Ding, S.S. Lu, Pool boiling on the superhydrophilic surface with TiO2 nanotube arrays, Sci. China Ser. E-Tech Sci. 52
(2009) 1596e1600.
[9] S. Ujereh, T. Fisher, I. Mudawar, Effects of carbon nanotube arrays on nucleate
pool boiling, Int. J. Heat Mass Transfer 50 (2007) 4023e4038.
[10] S. Launay, A.G. Fedorov, Y. Joshi, A. Cao, P.M. Ajayan, Hybrid micronanostructured thermal interfaces for pool boiling heat transfer enhancement, Microelectronics J. 37 (2006) 1158e1164.
[11] H.S. Ahn, N. Sinha, M. Zhang, D. Banerjee, S.K. Fang, R.H. Baughman, Pool
boiling experiments on Multi Walled Carbon Nanotube (MWCNT) Forests,
J. Heat Transfer-Transactions ASME 128 (2006) 1335e1342.
[12] C. Li, Z. Wang, P. Wang, Y. Peles, N. Koratkar, P. Peterson, Nanostructured
copper interfaces for enhanced boiling, Smll (2007) 00991.
[13] S. Kim, H.D. Kim, H. Kim, H.S. Ahn, H. Jo, J. Kim, M.H. Kim, Effects of nano-fluid
and surfaces with nano structure on the increase of CHF, Exp. Therm. Fluid Sci.
34 (2010) 487.
[14] R. Chen, M.C. Lu, V. Srinivasan, Z. Wang, H.H. Cho, A. Majumdar, Nanowires for
enhanced boiling heat transfer, Nano Lett. 9 (2009) 548e553.
[15] Y. Im, Y. Joshi, C. Dietz, S.S. Lee, Enhanced boiling of a dielectric liquid on
copper nanowire surfaces, Int. J. Micro-Nano Scale Transport 1 (2010) 79e95.
[16] A.R. Betz, J. Xu, H.H. Qiu, D. Attinger, Do surfaces with mixed hydrophilic and
hydrophobic areas enhance pool boiling? App. Phys. Lett. 97 (14) (2010).
[17] Y.W. Lu, S.G. Kandlikar, Nanoscale surface modification techniques for pool
boiling enhancementeA critical review and future directions, Heat Transfer
Eng. 32 (10) (2011).
[18] Z. Yao, Y.-W. Lu, S.G. Kandlikar, “Direct growth of copper nanowires on
a substrate for boiling applications”, Micro & Nano Lett. 6 (7) (2011) 562e565.
[19] J.W.G. Tyrrell, P. Attard, Images of nanobubbles on hydrophobic surfaces and
their interactions, Phys. Rev.Lett. 87 (2001) 176104.
[20] M. Holmberg, A. Kuhle, J. Garnaes, K.A. Morch, A. Boisena, Nanobubble trouble
on gold surface, Langmuir 19 (2003) 10510.
[21] K.Q. Peng, H. Fang, J. Hu, Y. Wu, J. Zhu, Y. Yan, S.T. Lee, Metal particle induced,
highly localized site-specific etching of Si and formation of single-crystalline
Si nanowires in aqueous fluoride solution, Chem. Eur. J 12 (2006) 7942.
[22] X. Li, P.W. Bohn, Metal-assisted chemical etching in HF/H2O2 produces porous
silicon, Appl. Phys. Lett. 77 (2000) 2572.
[23] S.L. Cheng, C.H. Chung, H.C. Lee, A study of the synthesis, characterization, and
kinetics of vertical silicon nanowire arrays on (001) Si substrates,
J. Electrochem. Soc. 155 (2008) D711.
[24] D. Cooke, S.G. Kandlikar, Pool boiling heat transfer and bubble dynamics over
plain and enhanced microchannels, J. Heat Transfer 133 (5) (2011).
[25] T.G. Theofanous, J.P. Tu, A.T. Dinh, L.J. Zhang, The boiling crisis phenomenon,
Exp. Therm. Fluid Sci. 26 (2002) 775e792.
[26] Y.Y. Hsu, On the size range of active nucleation cavities on a heating surface,
J. Heat Transfer (1962) 207e216.
[27] J.A. Dean (Ed.), Lange’s Handbook of Chemistry, fifteenth ed. McGraw-Hill,
1999.
Please cite this article in press as: Z. Yao, et al., Effects of nanowire height on pool boiling performance of water on silicon chips, International
Journal of Thermal Sciences (2011), doi:10.1016/j.ijthermalsci.2011.06.009