Embed Size (px)
Citation preview
RESEARCH ARTICLE
Drag reduction using superhydrophobic sanded Teflon surfaces
Dong Song • Robert J. Daniello • Jonathan P. Rothstein
Received: 8 January 2014 / Revised: 13 June 2014 / Accepted: 2 July 2014
� Springer-Verlag Berlin Heidelberg 2014
Abstract In this paper, a series of experiments are pre-
sented which demonstrate drag reduction for the laminar
flow of water through microchannels using superhydro-
phobic surfaces with random surface microstructure. These
superhydrophobic surfaces were fabricated with a simple,
inexpensive technique of sanding polytetrafluoroethylene
(PTFE) with sandpaper having grit sizes between 120- and
600-grit. A microfluidic device was used to measure the
pressure drop as a function of the flow rate to determine the
drag reduction and slip length of each surface. A maximum
pressure drop reduction of 27 % and a maximum apparent
slip length of b = 20 lm were obtained for the superhy-
drophobic surfaces created by sanding PTFE with a
240-grit sandpaper. The pressure drop reduction and slip
length were found to increase with increasing mean particle
size of the sandpaper up to 240-grit. Beyond that grit size,
increasing the pitch of the surface roughness was found to
cause the interface to transition from the Cassie–Baxter
state to the Wenzel state. This transition was observed both
as an increase in the contact angle hysteresis and simulta-
neously as a reduction in the pressure drop reduction. For
these randomly rough surfaces, a correlation between the
slip length and the contact angle hysteresis was found. The
surfaces with the smallest contact angle hysteresis were
found to also have the largest slip length. Finally, a number
of sanding protocols were tested by sanding preferentially
along the flow direction, across the flow direction and with
a random circular pattern. In all cases, sanding in the flow
direction was found to produce the largest pressure drop
reduction.
1 Introduction
Superhydrophobic surfaces can be found in nature in a
number of plants and insects around the world (Barthlott
and Neinhuis 1997; Hebets and Chapman 2000; Hu et al.
2003; Nørgaard and Dacke 2010; Shirtcliffe et al. 2006a, b;
Zheng et al. 2007). Superhydrophobic surfaces have large
advancing contact angles, hA [ 150�, and little difference
between their advancing and receding contact angles,
hA & hR. The extreme water repellency of naturally
occurring superhydrophobic surfaces gives the lotus leaves
and other plants their self-cleaning properties (Barthlott
and Neinhuis 1997), allows water striders to walk on water
(Hu et al. 2003) and makes it possible for insects and
spiders to breath under water (Hebets and Chapman 2000;
Shirtcliffe et al. 2006a). There are two important factors
needed to produce superhydrophobic surfaces: low surface
energy and micro/nanoroughness. The combination of
surface roughness and hydrophobicity can trap an air layer
in the depressions on the surface and result in the formation
of an air–water interface that is supported by the peaks in
the surface roughness. This is known as the Cassie–Baxter
state (Cassie and Baxter 1944). The presence of the air–
water interface increases the contact angle and reduces the
contact angle hysteresis (Quere 2008). As a result, droplets
have been shown to move very easily on superhydrophobic
surfaces because the droplet pinning force is known to be
proportional to the contact angle hysteresis (Hammer et al.
D. Song
School of Marine Engineering, Northwestern Polytechnical
University, 127 Youyi Xilu, Xi’an 710072, Shaanxi,
People’s Republic of China
D. Song � R. J. Daniello � J. P. Rothstein (&)
Department of Mechanical and Industrial Engineering,
University of Massachusetts, Amherst, 160 Governors Drive,
Amherst, MA 01003-2210, USA
e-mail: [email protected]
123
Exp Fluids (2014) 55:1783
DOI 10.1007/s00348-014-1783-8
2010; Reyssat et al. 2006; Richard and Quere 2000; Nils-
son and Rothstein 2011, 2012). The presence of the air–
water interface has also been found to change the boundary
condition on a superhydrophobic surface from the classic
no-slip boundary condition to a partial slip boundary con-
dition resulting in substantial drag reduction in both lami-
nar and turbulent flows (Rothstein 2010).
Over the past decade, there have been a large number of
studies investigating the drag reduction properties of su-
perhydrophobic surfaces containing both regular and
irregular patterns of micron and nanometer-sized surface
roughness (Bocquet and Barrat 2007; Choi et al. 2006;
McHale et al. 2009; Ou et al. 2004; Ou and Rothstein 2005;
Rothstein 2010; Tretheway and Meinhart 2002, 2004;
Truesdell et al. 2006). Ou et al. (2004, 2007), and Ou and
Rothstein (2005) were one of the first to demonstrate that
superhydrophobic surfaces could produce drag reduction.
Their measurements showed pressure drop reductions of up
to 40 % in laminar flows using superhydrophobic surfaces
consisting of regular arrays of posts and ridges fabricated
through photolithography. In addition, they used micro-
particle image velocimetry (l-PIV) to measure the velocity
profile above a superhydrophobic surface and demonstrated
peak slip velocities along the air–water interface as large as
60 % of the average velocity. The numerous papers that
followed focused primarily on superhydrophobic surfaces
with regularly patterned geometries with the goals of
understanding how the microstructure pattern and size
affect slip (Ybert et al. 2007) and maximizing drag
reduction (Lee and Kim 2009) for superhydrophobic sur-
faces (Rothstein 2010).
It is important to note, however, if we hope to establish
wide, commercial application of superhydrophobic sur-
faces that more drag reduction studies need to be per-
formed on substrates with random microstructures
(Bocquet and Lauga 2013). Such surfaces are often easier
and less costly to fabricate than precisely patterned sur-
faces and they have the potential to be applied over large
surface areas making them ideal for application on ships. A
number of theoretical works including Sbragaglia and
Prosperettii (2007) and Feuillebois et al (2009) have
demonstrated that randomly rough surfaces can produce
significant drag reduction. Additionally, over the last few
years, there have been a number of experiments performed
on randomly rough superhydrophobic surfaces. Gogte et al.
(2005) investigated drag reduction on a hydrofoil coated
with a hydrophobic sandpaper. Their results showed up to
20 % drag reduction. McHale et al. (2009) coated a series
of spheres with a sharp, hydrophobic sand and found as
much as a 30 % reduction in the drag coefficient. More
recently, Srinivasan et al. (2013) spray coated a number of
different surfaces with a mixture of PMMA and fluorodecyl
POSS producing a superhydrophobic surface decorated
with a random array of hydrophobic 20 lm beads. Mea-
surements using a parallel plate rheometer showed slip
lengths of more than 30 lm on flat substrates and much
larger slip lengths when theses coatings were applied to a
series of woven metal meshes.
In this work, an easy, inexpensive method involving the
sanding of polytetrafluoroethylene (PTFE) with different
grit sandpapers was employed to make randomly structured
superhydrophobic surfaces with a wide range of wetting
properties (Nilsson et al. 2010). The pressure drop across a
microfluidic rectangular microchannel was then used to
experimentally characterize the drag reduction ability of
these superhydrophobic sanded Teflon surfaces.
2 Experimental setup
Many methods have been used to make high-quality su-
perhydrophobic surfaces with large contact angles, little
hysteresis and the ability to maintain an air–water interface
under reasonable large static pressures. These techniques
include among others: photolithography (Oner and
McCarthy 2000; Ou et al. 2004), phase separation (Li et al.
2006; Yamanaka et al. 2006), chemical vapor deposition
method (Liu et al. 2004; Yao et al. 2011), wet-chemical
method (Lau et al. 2003), electrospinning (Singh et al.
2005), self-assembly (Lee et al. 2009; Pietsch et al. 2009)
and coated porous aluminum oxide (Jin et al. 2005) among
others. Most of these methods are complicated, require
costly clean room facilities or expensive equipment and are
challenging to apply on large area. In this paper, we use an
inexpensive, uncomplicated technique to fabricate super-
hydrophobic surfaces developed by Nilsson et al. (2010).
Specifically, smooth polytetrafluoroethylene (PTFE),
commercially known as Teflon, is sanded using various
grits of sandpaper to impart random surface roughness to a
low-energy surface.
To fabricate our superhydrophobic surfaces, a series of
Teflon sheets were first glued onto glass slides make sure the
surface was flat and smooth. The Teflon samples were then
sanded by hand for a minimum of 2 min using sandpaper
grits of 120, 180, 240, 360 and 600. Several Teflon surfaces
sanded with each sandpaper grit were fabricated and inde-
pendently tested so that the variation associated with pre-
paring the surfaces by hand could be accounted for in the
statistical experimental uncertainty. Three distinct sanding
protocols were used to illustrate the importance of sanding
direction. These protocols involved sanding the Teflon back
and forth only in the flow direction, sanding the Teflon only
in the cross-flow direction and finally randomly sanding the
Teflon with a circular motion. After sanding, the loose Teflon
particles were removed from the surface with a gentle stream
of compressed air and rinsed with water.
1783 Page 2 of 8 Exp Fluids (2014) 55:1783
123
In Fig. 1, the scanning electron microscopy (SEM)
images of smooth and sanded Teflon surfaces are pre-
sented. Nilsson et al. (2010) demonstrated that by sanding
Teflon with a series of different grit sizes, surfaces with a
wide range of hydrophobic properties could be developed.
This includes ideal superhydrophobic surfaces with high
contact angle and low hysteresis for sandpaper grits
between 240 and 320 as well as a range of moderate to high
contact angle and contact angle hysteresis surfaces sanded
with larger or smaller grit sandpapers. Although measure-
ments of contact angle and contact angle hysteresis can
indicate whether a droplet on a surface is in the Cassie–
Baxter state, there is no known direct relationship between
drag reduction and contact angle hysteresis on superhy-
drophobic surfaces (Voronov et al. 2008). By using a wide
range of sandpaper grits to impart different random surface
roughness to Teflon, in this paper, we will test whether it is
possible to correlate contact angle hysteresis to slip length
for randomly rough surfaces.
To test the drag reducing ability of each surface, a
microfluidic device with a rectangular microchannel was
developed to measure the pressure drop across the smooth
and sanded Teflon surfaces. A schematic of the micro-
channel along with an image of the microfluidic device is
shown in Fig. 2. The device contains three parts: the san-
ded/smooth Teflon test surface epoxied to a glass micro-
scope slide; an upper cover made of PDMS bonded to an
acrylic cover slip with a 137-lm-deep, 3.4-mm-wide and
60-mm-long rectangular microchannel formed in the
PDMS; and several clips used to clamp the microfluidic
device together under uniform, consistent pressure. The
acrylic top cover slip functions only as a stiffener to con-
strain the PDMS and distribute the clamping force of the
clips that hold the top and bottom sections of the device
together. The low modulus (*1.8 MPa) of the PDMS
allows the device to make a seal that was experimentally
verified to be leak free with the lower surface but also
necessitates relatively light, even and consistent clamping
force. This is because under pressure the low modulus of
the PDMS means that the microchannel will deform
slightly. The 137 mm depth quoted for the microchannel
was measured after the channel was sealed using micros-
copy and later confirmed by fitting the pressure drop
measurement of the smooth, no-slip Teflon substrate to the
expected pressure drop for Poiseuille flow through a rect-
angular channel (White 1991). Note, however, that at an
aspect ratio of 25:1, the end effects are essentially negli-
gible and the channel can be assumed to be equivalent to
two infinite parallel plates without introducing significant
error.
The microchannel was fabricated from Sylgard 184
PDMS (Dow Corning) using standard soft lithography
techniques. The flow inlet and outlet were located at the far
ends of the channel, spaced 60 mm apart. The pressure
ports were installed 45 mm apart. The test fluid was
ultrapure water driven by a syringe pump (KD Scientific
Model 100) at a constant flow rate. Pressure drop was
measured by recording the height difference between two
manometer columns attached to the pressure ports. The
resolution of the pressure drop measurements was 0.5 mm
of water or equivalently 5 Pa. Experiments were run over a
wide range of flow rates for a series of smooth/sanded
Teflon surfaces all well below Reynolds numbers of
Fig. 1 SEM images of a series of Teflon surfaces sanded with
sandpapers of various grit designations. The images include a smooth
Teflon, b Teflon sanded with 600-grit, c Teflon sanded with 320-grit and
d Teflon sanded with 240-grit. The SEM images have been reproduced
with permission from Nilsson et al. (2010)
Fig. 2 a A schematic diagram of the PDMS microfluidic devices
used in these experiments with the appropriate dimensions and b an
image of the actual device
Exp Fluids (2014) 55:1783 Page 3 of 8 1783
123
Re � 100 to ensure laminar flows. In all cases, the top
wall was smooth PDMS and was assumed to be no-slip.
3 Results and discussion
Five different grits of sandpaper (120, 180, 240, 360 and
600-grit) were used to roughen an initially smooth Teflon
surfaces. In the initial experiments, the Teflon was sanded
with a bias in the flow direction in an attempt to introduce
scratches or grooves along the length of the microchannel
in order to maximize the drag reduction. To obtain a
baseline for comparing the superhydrophobic surfaces, the
pressure drop of smooth Teflon channel flow was measured
for flows with average velocities varying from 24 to
48 mm/s. As seen in Fig. 3, the experimental measure-
ments for flow past a smooth Teflon surface match the
predictions for flow between two infinite no-slip plates.
These measurements gave us confidence in the quality of
our experimental setup and provided a good no-slip base-
line data set to compare against the results of sanded
Teflon. In Fig. 3, the pressure drop for a series of sanded
Teflon surfaces in the microchannel is also presented as a
function of flow rate. As seen in Fig. 3, the pressure drop of
measured for each of the superhydrophobic sanded Teflon
surfaces was found to be smaller than that of the smooth
Teflon. The percentage of pressure drop reduction for each
of the sanded Teflon surface was found to be independent
of flow rate but strongly dependent on the grit designation.
The maximum pressure drop reduction of 27 % was
measured for the Teflon surface sanded with the 240-grit
sandpaper.
To observe the trends more closely, the data in Fig. 3
were recast as a pressure drop reduction in Fig. 4. Here, the
average pressure drop reduction is presented with error bars
superimposed over the data indicating the 95 % confidence
interval of the average. Each data point represents a min-
imum of five measurements with the microfluidic device
broken down and reassembled for each experiment to
insure repeatability and robustness of the measurement. As
seen in Fig. 4a, the average pressure drop reduction was
found to be very sensitive to the grit designation of the
sandpaper used to create the superhydrophobic surfaces. As
the grit designation was decreased from 600- to 240-grit,
the mean particle size of the sandpaper increased from 25
to 58 lm. Over that range, the RMS roughness on the
Teflon surface increases from 7.6 to 13.7 lm (Nilsson et al.
20 25 30 35 40 45 50400
600
800
1000
1200
1400
Pres
sure
Dro
p [P
a]
Flow Velocity [mm/s]
Fig. 3 Measurement of pressure drop as a function of flow velocity
for a series of smooth and sanded Teflon in microchannel. The
experiment data include both (times) smooth Teflon surface as well as
Teflon surfaces sanded with a series of different sandpaper with grit
designation of (filled diamond) 120-grit, (triangle) 180-grit, (filled
square) 240-grit, (filled circle) 320-grit and (inverted triangle)
600-grit. Superimposed over the data is the theoretical prediction
for flow through the microchannel assuming the walls are no-slip (-)
0
5
10
15
20
25
30
Pres
sure
Dro
p R
educ
tion
[%]
Grit Designation of Sandpaper
(a)
120 180 240 320 600
120 180 240 320 6000
10
20
30
40
50
60
70
Con
tact
Ang
le H
yste
resi
s (
AR)
[deg
]
Grit Designation of Sandpaper
(b)
Fig. 4 Pressure drop reduction (a) and contact angle hysteresis (b) as
a function of the grit designation used to sand the Teflon surfaces. The
contact angel hysteresis in b has been reproduced with permission
from Nilsson et al. (2010)
1783 Page 4 of 8 Exp Fluids (2014) 55:1783
123
2010) and the pressure drop reduction increased from 5 %
to a maximum of about 27 %. As the grit designation
decreased further to 180-grit, the mean particle size of the
sandpaper increased to about 80 lm, and the pressure drop
reduction was found to decrease sharply to only about
12 %. At 120-grit, the pressure drop reduction falls off
even further.
As seen in Fig. 1, the spacing between raised Teflon
features on the sanded surface scales with the particle size
of the sandpaper. As a result when the grit designation is
240-grit or larger, the structures produced by sanding are
separated by gaps small enough to support an air–water
interface on the superhydrophobic surface without col-
lapsing into the features. In the Cassie–Baxter state, a
maximum static pressure exists that can be supported by
the air–water interface before it collapses into the Wenzel
state. That pressure, pwater - pair = -2c cos hA/w,
decreases with increasing feature spacing, w. For a static
pressure of 3,600 Pa, which is typical in our experiments, a
spacing at w = 80 lm can be supported. This corresponds
with the average particle size on the 180-grit sandpaper.
For these coarser, smaller grits, many of the larger scrat-
ches introduced into the Teflon surface will not be able to
support an air–water interface and the effectiveness for
drag reduction should decrease. As the grit designation
increases, the particles on the sandpaper become smaller,
and as seen in Fig. 1, the spacing between surface struc-
tures decreases along with their periodicity. For superhy-
drophobic surfaces with same fraction of shear-free air–
water interface, increasing the microstructures spacing has
been shown both experimentally and theoretically to
increase the maximum of the slip velocity (Ou and Roth-
stein 2005) and the slip length (Ybert et al. 2007). Ybert
et al. (2007) showed that for a regular array of posts, which
one can argue most closely resembles our randomly rough
surface, that the slip length is proportional to the spacing
between features, b / L=ffiffiffiffiffi
/s
p
. Here, /s = d2/L2 is the
solid fraction of the superhydrophobic surface, d is the
diameter of the posts and L = w ? a is the pitch of the
surface roughness. The slip length is defined in this context
using Navier’s (1823) partial slip boundary condition, us ¼b ou=oyj j: Here, us is the slip velocity at the wall and qu/
qy is the velocity gradient at the wall. Thus, larger grit
designations which lead to smaller spaced microstructures,
as seen in Fig. 1, are expected to produce less pressure
drop reduction.
The data in Fig. 4a suggest at least a partial transition
from the Cassie–Baxter state to the Wenzel state for grit
designations less than 240-grit. This observation is further
reinforced by the trends in contact angle hysteresis for
sanded Teflon surface made by Nilsson et al. (2010). In
Nilsson et al. (2010), the advancing and receding contact
angles were measured from digitized images of small
droplets as water was slowly added and withdrawn. Their
contact angle hysteresis data are reproduced in Fig. 4b as a
function of grit designation. An inspection of the data
reveals that the drag reduction and contact angle hysteresis
data in Fig. 4a, b are strongly correlated although no clear
scaling relationship can be extracted from the data. As the
grit decreased from 600 to 240, the contact angle hysteresis
is minimized just as the drag reduction is maximized. This
observation is intuitive as a number of studies have shown
that, for a superhydrophobic surface with fixed size mi-
croposts, the contact angle hysteresis decreases as the pitch
between the microposts is increased (Oner and McCarthy
2000). Xu and Choi (2012) recently demonstrated that the
receding contact angle was not solely dependent on peri-
odicity of the surface roughness, but also on feature size.
They showed that as the contact line receded across a su-
perhydrophobic surface consisting of circular posts that the
force needed to de-pin the contact line from the top of the
posts depended on the ratio of the of the posts’ wetted
perimeter to their pitch, FD� d=L ¼ffiffiffiffiffi
/s
p
. Since, as
described above, the slip length is also known to be
dependent on the solid fraction of the surface, b / L=ffiffiffiffiffi
/s
p
,
these observations for grit designations between 600 and
240-grit make sense if increasing the sandpaper particle
size results in both an increase in pitch, L, as expected, and
simultaneously a decrease in the solid surface fraction, /s,
see Fig. 1. For the Teflon surfaces sanded with sandpaper
below a 240-grit designation, Nilsson et al. (2010) argued,
as we have here, that the increase in contact angel hys-
teresis was the result of a gradual transition from a surface
fully in the Cassie–Baxter state to a surface in the Wenzel
state. For laminar flow across superhydrophobic sanded
Teflon surfaces, it thus appears that one can predict the
surfaces that will produce a maximum drag reduction
simply by finding the surface which minimizes the contact
angle hysteresis.
In order to test the sensitivity of the drag reduction
based on sanding protocol, two other sanding procedures,
circular sanding and cross-sanding, were used for the 240-
and 320-grit sandpapers. In Fig. 5, the results of the pres-
sure measurements are presented as a function of sanding
protocol. Sanding primarily in the flow direction was
consistently found to be the best method to reduce drag and
maximize pressure drop reduction as a large fraction of the
scratches introduced into the Teflon surface are oriented in
the flow direction. Sanding in cross-flow direction was
found to be least effective at reducing drag. In fact, the
cross-flow sanding protocol produces approximately half
the pressure drop reduction of sanding in the flow direction.
This observation agrees well with the experimental and
theoretical work for flow over superhydrophobic surfaces
Exp Fluids (2014) 55:1783 Page 5 of 8 1783
123
containing a regular array of ridges. In those cases, flow
across the ridge was found to give only half the pressure
drop reduction of flow along the ridges (Harting et al. 2010;
Lauga and Stone 2003). The circular sanding protocol
resulted in pressure drop reduction between the other two
cases. Thus, using different sanding protocols, we can
optimize the superhydrophobic surfaces structures and
maximize drag reduction, making this technique more
flexible for engineering application.
Finally, when discussing drag reduction data, it is often
important to present the data as a slip length rather than a
pressure drop reduction because slip length is independent
of channel geometry and flow velocity. For fluid flow
between two infinite parallel plates separated by a distance
H where one wall has a slip length b, the volume flow rate
per unit length, q ¼ H3
4l �dp
dx
� �
13þ b
bþH
h i
; can be expressed
in terms of the channel geometry, the slip length, the
pressure gradient, dp/dx, and the fluid viscosity, l. For a
given fluid and flow rate, the pressure gradient can thus be
significantly affected by slip when the slip length is com-
parable to the height of channel. Using this equation, we
can calculate the slip length directly from the pressure drop
measurements. The pressure drop reduction data from
Fig. 4 is plotted as slip length in Fig. 6 as a function of
mean grit size. As seen in Fig. 6, the slip length was found
to increase approximate linearly with grit size up to a
sandpaper particle size of 60 lm which corresponds to the
240-grit sandpaper. For the 240-grit sandpaper, a maximum
slip length of approximately b = 20 lm was measured.
For a regular array of posts with a small solid surface
fraction, /s, Ybert et al. (2007) showed that
b=L ¼ 0:324=ffiffiffiffiffi
/s
p
� 0:44. Assuming the pitch of the sur-
face features is equal to the average sandpaper particle size,
the approximate solid surface fraction for the 240-grit
sanded Teflon surface can be calculated to be /s & 0.18
which means the superhydrophobic surface created is more
than 80 % air. Alternately, if one assumes the surface is
covered with ridges instead of posts, a slightly larger value
of /s & 0.23 can be calculated for the Teflon surface
sanded with the 240-grit sandpaper. Finally, we note that a
similar value of /s & 0.20 was calculated for the Teflon
surface sanded with the 320-grit sandpaper. Given that the
320-grit sandpaper produces a surface with smaller
roughness spacing, but similar interface coverage, it can be
utilized in flows where the pressure might be too large for
the 240-grit surface to maintain an air–water interface.
4 Conclusion
In this paper, a series of experiments were performed
which demonstrate drag reduction for the laminar flow of
water through microchannels using superhydrophobic sur-
faces fabricated with random surface microstructure. These
superhydrophobic surfaces were fabricated using the sim-
ple, inexpensive technique first developed by Nilsson et al.
(2010) which involves sanding Teflon with various grits of
sandpaper. Teflon surfaces sanded with grit sizes between
120- and 600-grit were studied. A microfluidic device was
used to measure the pressure drop as a function of the flow
Flow Direction Circular Cross Flow Direction5
10
15
20
25
30Pr
essu
re D
rop
Red
uctio
n [%
]
Sanding Protocol
Fig. 5 Pressure drop reduction as a function of sanding protocol. For
each sanding style, flow direction (filled symbols), cross-flow
direction (crossed symbols) and circular sanding protocols (open
symbols) were used with sandpapers with grit designations of (filled
square, open square, crossed square) 240-grit and (filled circle, open
circle, crossed circle) 320-grit
10 20 30 40 50 60 70 80 90 100 110 120 130 1400
5
10
15
20
Slip
Len
gth,
b [
m]
Mean Particle Diameter of Sandpaper [ m]µ
µ
Fig. 6 Slip length, b, measured for a series of sanded Teflon surfaces
as a function of the mean particle diameter of the sandpaper used. The
data include Teflon sanded in the flow direction with (diamond)
120-grit, (triangle) 180-grit, (filled square) 240-grit, (filled circle)
320-grit and (inverted triangle) 600-grit sandpaper; Teflon sanded
with a random circular pattern with (open square) 240-grit and (open
circle) 320-grit sandpaper; and Teflon sanded in the cross-flow
direction with (crossed square) 240-grit and (crossed circle) 320-grit
sandpaper
1783 Page 6 of 8 Exp Fluids (2014) 55:1783
123
rate to determine the drag reduction and slip length of each
surface. A maximum pressure drop reduction of 27 % and
a maximum apparent slip length of b = 20 lm were
obtained for the superhydrophobic surfaces created by
sanding PTFE with a 240-grit sandpaper. The pressure drop
reduction and slip length were found to increase with
increasing grit size up to 240-grit. Beyond that grit size,
increasing periodicity of the surface roughness was found
to cause the interface to transition from the Cassie–Baxter
state to the Wenzel state. This transition was observed both
as an increase in the contact angle hysteresis and simulta-
neously as a reduction in the pressure drop reduction. For
these randomly rough surfaces, a correlation between the
slip length and the contact angle hysteresis was found. The
surfaces with the smallest contact angle hysteresis were
found to also have the largest slip length. Finally, a number
of sanding protocols were tested by sanding preferentially
along the flow direction, across the flow direction and with
a random circular pattern. In all cases, sanding in the flow
direction was found to produce the largest pressure drop
reduction. This work demonstrates that the mechanical
abrasion of hydrophobic surfaces to make them superhy-
drophobic has the potential to establish wide, commercial
application of superhydrophobic surfaces. When compared
to precisely patterned surfaces, they are easier and less
costly to fabricate giving them the potential to be applied
over large surface areas and making them ideal for large-
scale application such as on ocean going vessels.
References
Barthlott W, Neinhuis C (1997) Purity of the sacred lotus, or escape
from contamination in biological surfaces. Planta 202:1–8
Bocquet L, Barrat J-L (2007) Flow boundary conditions from nano- to
micro-scales. Soft Matter 3:685–693
Bocquet L, Lauga E (2013) A smooth future? Nat Mater 10:334–337
Cassie ABD, Baxter S (1944) Wettability of porous surfaces. Trans
Faraday Soc 40:0546–0550
Choi C-H, Ulmanella U, Kim J, Ho C-M, Kim C-J (2006) Effective
slip and friction reduction in nanograted superhydrophobic
microchannels. Phys Fluids 18:087105
Feuillebois F, Bazant MZ, Vinogradova OI (2009) Effective slip over
superhydrophobic surfaces in thin channels. Phys Rev Lett
102:026001
Gogte S, Vorobieff P, Truesdell R, Mammoli A, van Swol F, Shah P,
Brinker CJ (2005) Effective slip on textured superhydrophobic
surfaces. Phys Fluids 17:051701
Hammer MU, Anderson TH, Chaimovich A, Shell MS, Israelachvili J
(2010) The search for the hydrophobic force law. Faraday
Discuss 146:299–308
Harting J, Kunert C, Hyvaluoma J (2010) Lattice Boltzmann
simulations in microfluidics: probing the no-slip boundary
condition in hydrophobic, rough, and surface nanobubble laden
microchannels. Microfluid Nanofluid 8:1–10
Hebets EA, Chapman RF (2000) Surviving the flood: plastron
respiration in the non-tracheate arthropod Phrynus marginema-
culatus (amblypygi: Arachnida). J Insect Physiol 46:13–19
Hu DL, Chan B, Bush JWM (2003) The hydrodynamics of water
strider locomotion. Nature 424:663–666
Jin MH, Feng XJ, Feng L, Sun TL, Zhai J, Li TJ, Jiang L (2005)
Superhydrophobic aligned polystyrene nanotube films with high
adhesive force. Adv Mater 17:1977
Lau KKS, Bico J, Teo KBK, Chhowalla M, Amaratunga GAJ, Milne
WI, McKinley GH, Gleason KK (2003) Superhydrophobic
carbon nanotube forests. Nano Lett 3:1701–1705
Lauga E, Stone HA (2003) Effective slip in pressure-driven stokes
flow. J Fluid Mech 489:55–77
Lee C, Kim C-J (2009) Maximizing the giant liquid slip on
superhydrophobic microstructures by nanostructuring their side-
walls. Langmuir 25:12812–12818
Lee JS, Ryu J, Park CB (2009) Bio-inspired fabrication of superhy-
drophobic surfaces through peptide self-assembly. Soft Matter
5:2717–2720
Li XH, Chen GM, Ma YM, Feng L, Zhao HZ, Jiang L, Wang FS
(2006) Preparation of a super-hydrophobic poly(vinyl chloride)
surface via solvent–nonsolvent coating. Polymer 47:506–509
Liu H, Feng L, Zhai J, Jiang L, Zhu DB (2004) Reversible wettability
of a chemical vapor deposition prepared zno film between
superhydrophobicity and superhydrophilicity. Langmuir
20:5659–5661
McHale G, Shirtcliffe NJ, Evans CR, Newton MI (2009) Terminal
velocity and drag reduction measurements on superhydrophobic
spheres. Appl Phys Lett 94:064104
Navier CLMH (1823) Memoire sur les lois du mouvement des fluides.
Memoires de l’Academie Royal des Sciences de I’Institut de
France 6:389–440
Nilsson M, Rothstein JP (2011) The effect of contact angle hysteresis
on droplet coalescence and mixing. J Colloid Interface Sci
363:646–654
Nilsson M, Rothstein JP (2012) Using sharp transition in contact
angle hysteresis to move and deflect droplets on a superhydro-
phobic surface. Phys Fluids 24:062001
Nilsson MA, Daniello RJ, Rothstein JP (2010) A novel and
inexpensive technique for creating superhydrophobic surfaces
using Teflon and sandpaper. J Phys D Appl Phys 43:045301
Nørgaard T, Dacke M (2010) Fog-basking behaviour and water
collection efficiency in Namib Desert Darkling beetles. Front
Zool 7:23
Oner D, McCarthy TJ (2000) Ultrahydrophobic surfaces: effects of
topography length scales on wettability. Langmuir 16:7777–7782
Ou J, Rothstein JP (2005) Direct velocity measurements of the flow
past drag-reducing ultrahydrophobic surfaces. Phys Fluids
17:103606
Ou J, Perot JB, Rothstein JP (2004) Laminar drag reduction in
microchannels using ultrahydrophobic surfaces. Phys Fluids
16:4635–4660
Ou J, Moss GR, Rothstein JP (2007) Enhanced mixing in laminar
flows using ultrahydrophobic surfaces. Phys Rev E 76:016304
Pietsch T, Gindy N, Fahmi A (2009) Nano- and micro-sized
honeycomb patterns through hierarchical self-assembly of
metal-loaded diblock copolymer vesicles. Soft Matter
5:2188–2197
Quere D (2008) Wetting and roughness. Annu Rev Mater Res
38:71–99
Reyssat M, Pepin A, Marty F, Chen Y, Quere D (2006) Bouncing
transitions on microtextured materials. Europhys Lett
74:306–312
Richard D, Quere D (2000) Bouncing water drops. Europhys Lett
50:769–775
Rothstein JP (2010) Slip on superhydrophobic surfaces. Annu Rev
Fluid Mech 42:89–109
Sbragaglia M, Prosperetti A (2007) Effective velocity boundary
condition at a mixed slip surface. J Fluid Mech 578:435–451
Exp Fluids (2014) 55:1783 Page 7 of 8 1783
123
Shirtcliffe NJ, McHale G, Newton MI, Perry CC, Pyatt FB (2006a)
Plastron properties of a superhydrophobic surface. Appl Phys
Lett 89:104106
Shirtcliffe NJ, Pyatt FB, Newton MI, McHale G (2006b) A lichen
protected by a super-hydrophobic and breathable structure.
J Plant Physiol 163:1193–1197
Singh A, Steely L, Allcock HR (2005) Poly[bis(2,2,2-trifluoroeth-
oxy)phosphazene] superhydrophobic nanofibers. Langmuir
21:11604–11607
Srinivasan S, Choi W, Park K-C, Chhatre SS, Cohen RE, McKinley
GH (2013) Drag reduction for viscous laminar flow on spray-
coated non-wetting surfaces. Soft Matter 9:5691–5702
Tretheway DC, Meinhart CD (2002) Apparent fluid slip at hydro-
phobic microchannel walls. Phys Fluids 14:L9–L12
Tretheway DC, Meinhart CD (2004) A generating mechanism for
apparent fluid slip in hydrophobic microchannels. Phys Fluids
16:1509–1515
Truesdell R, Mammoli A, Vorobieff P, van Swol P, Brinker CJ (2006)
Drag reduction on a patterened superhydrophobic surface. Phys
Rev Lett 97:044504
Voronov RS, Papavassiliou DV, Lee LL (2008) Review of fluid slip
over superhydrophobic surfaces and its dependence on the
contact angle. Ind Eng Chem Res 47:2455–2477
White FM (1991) Viscous fluid flow. McGraw-Hill, New York
Xu W, Choi C-H (2012) From sticky to slippery droplets: dynamics of
contact line depinning on superhydrophobic surfaces. Phys Rev
Lett 109:024504
Yamanaka M, Sada K, Miyata M, Hanabusa K, Nakano K (2006)
Construction of superhydrophobic surfaces by fibrous aggrega-
tion of perfluoroalkyl chain-containing organogelators. Chem
Commun (21):2248–2250
Yao LJ, Zheng MJ, He SG, Ma L, Li M, Shen WZ (2011) Preparation
and properties of Zns superhydrophobic surface with hierarchical
structure. Appl Surf Sci 257:2955–2959
Ybert C, Barentin C, Cottin-Bizonne C, Joseph P, Bocquet L (2007)
Achieving large slip with superhydrophobic surfaces: scaling
laws for generic geometries. Phys Fluids 19:123601
Zheng Y, Gao X, Jiang L (2007) Directional adhesion of superhy-
drophobic butterfly wings. Soft Matter 3:178–182
1783 Page 8 of 8 Exp Fluids (2014) 55:1783
123
