Home Search Collections Journals About Contact us My IOPscience

Effects of micro- and nano-structures on the self-cleaning behaviour of lotus leaves

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2006 Nanotechnology 17 1359

([Link]

View the table of contents for this issue, or go to the journal homepage for more

Download details:
IP Address: [Link]
The article was downloaded on 21/06/2013 at 20:15

Please note that terms and conditions apply.

INSTITUTE OF PHYSICS PUBLISHING NANOTECHNOLOGY
Nanotechnology 17 (2006) 1359–1362 doi:10.1088/0957-4484/17/5/032

Effects of micro- and nano-structures on

the self-cleaning behaviour of lotus leaves
Y T Cheng1,4 , D E Rodak2 , C A Wong3 and C A Hayden3
1
Materials and Processes Laboratory, General Motors Research and Development Center,
Warren, MI 48090, USA
2
Ricardo Meda Technical Services, LLC, Southfield, MI 48075, USA
3
Chemical and Environmental Science Laboratory, General Motors Research and
Development Center, Warren, MI 48090, USA

E-mail: [Link]@[Link]

Received 12 October 2005

Published 10 February 2006
Online at [Link]/Nano/17/1359
Abstract
When rain falls on lotus leaves water beads up with a high contact angle. The
water drops promptly roll off the leaves, collecting dirt along the way. This
self-cleaning ability or lotus effect has, in recent years, stimulated much
research effort worldwide for a variety of applications ranging from
self-cleaning window glasses, paints, and fabrics to low friction surfaces.
What are the mechanisms giving rise to the lotus effect? Although chemical
composition and surface structure are believed important, a systematic
experimental investigation of their effects is still lacking. By altering the
surface structure of the leaves while keeping their chemical composition
approximately the same, we report in this study the influence of micro- and
nano-scale structures on the wetting behaviour of lotus leaves. The findings
of this work may help design self-cleaning surfaces and improve our
understanding of wetting mechanisms.

1. Introduction to rationalize the lotus effect [19–23]. Water can interact with
surfaces in two ways, either by contacting only the peaks of
When water drops fall upon the surface of a lotus leaf they the roughened surface or by wetting the entire surface, both
form high contact angle drops. These drops can move freely the peaks and valleys. In the first case, the apparent contact
on the surface of the leaf, collecting dirt and other particles as angle, θCB , is given by the Cassie–Baxter equation for wetting
they roll off the surface. This lotus effect [1, 2] is thought to on composite surfaces made of the solid and air [17],
be caused by the complex surface structure of the leaf. When
viewed under a scanning electron microscope, two levels of cos θCB = fs cos θ + f s − 1, (1)
surface structure are seen: micro-scale mound-like structures
where f s is the fraction of projected planar area of the drop
of several micrometres tall protruding from the leaf and nano- in contact with the solid. In the limit of f s → 0, the
scale hair-like structures covering the leaf surface (figures 1(a) macroscopic contact angle θCB approaches 180◦ , leading to
and (b)). This two-level roughness in conjunction with the superhydrophobic behaviour. In the second case, where the
leaf’s waxy surface composition are thought responsible for the water wets both peaks and valleys of a rough surface of uniform
lotus leaf’s water repellency [1, 2], though no studies isolating chemical composition, the apparent contact angle of the drop,
these structural and compositional effects have been reported. θw , is given by Wenzel’s formula [18],
Nevertheless, there have been many attempts [3–16] to mimic
the lotus effect for a wide range of applications, such as self- cos θw = γ cos θ, (2)
cleaning windshields, paints and fabrics.
Theoretical models for liquid drops on topologically rough where γ is the ratio of the actual area of liquid–solid contact to
and chemically inhomogeneous surfaces have long been in the projected area on the horizontal plane of the rough surface,
existence [17, 18] and these models have recently been used and θ is the equilibrium contact angle of the liquid drop on
the corresponding flat surface. Wenzel’s equation predicts that
4 Author to whom any correspondence should be addressed.
roughness will amplify the intrinsic wetting behaviour of a

0957-4484/06/051359+04$30.00 © 2006 IOP Publishing Ltd Printed in the UK 1359

Y T Cheng et al

(a) (b)

10 µm 10 µm

(c) (d)

3 µm 3 µm Figure 2. TGA data for untreated and heat treated lotus leaves, as
well as carnauba wax. The data that the wax disintegrates at
temperature above 220 ◦ C which is substantially higher that the
Figure 1. SEM images of untreated lotus leaf (a) and (c), and 150 ◦ C used for annealing the lotus leaves.
annealed lotus leaf (b) and (d).

a Au–Pd sputtered coating to prevent charging during SEM

surface as determined by its surface chemistry. If the contact
observations. The scans were performed in a Hitachi S4800
angle on the smooth surface is larger than 90◦ , roughness
field emission SEM (FE-SEM) with accelerating voltages from
will further increase the observed contact angle. If it is less
300 V to 5 kV. SEM images reveal that the normal, untreated
than 90◦ , roughness will reduce the observed contact angle.
lotus leaf indeed had two levels of structures. However,
Equations (1) and (2) represent two possible equilibrium states
annealing the leaves for 1 h at 150 ◦ C removed all the nano-
of liquid drops on rough surfaces. Drops in the Cassie–Baxter
scale hair-like structures from the leaf, leaving behind a
state can easily roll because of low resistance from the air
smooth leaf surface interrupted frequently by 5 to 10 µm tall
pockets [20–22]. In contrast, drops in the Wenzel state can
mounds.
become ‘sticky’ from contacting the rough surface [20–22].
To determine the temperature effect on the leaf
For the lotus surface, it is generally believed that the two-
materials, high-resolution thermogravimetric analysis (TGA)
level roughness, in particular, the nano-scale hairs, amplifies
the apparent contact angle and is responsible for the rolling was performed using a TA Instruments TGA Q500. Both
behaviour of the drops. the dry and annealed lotus leaves were cut into small pieces
In this paper, we examine the micro- and nano-scale and placed in a platinum pan. Once loaded with the leaf
roughness effects on wetting by intentionally altering the pieces the pan was inserted into the device automatically. The
surface structure of the lotus leaves while keeping the temperature was ramped to 700 ◦ C at a rate of 20 ◦ C min−1
chemical composition approximately the same. Specifically, and a resolution of 6 ◦ C in a nitrogen atmosphere. The TGA
we use thermal annealing to remove the nano-scale hair-like curves of the dry and annealed leaves were essentially the same
features while leaving the micro-scale roughness and chemical (figure 2), suggesting that no large quantities of chemicals,
composition nearly intact. As a result, we are able to separate such as waxes, were lost by annealing the leaves in air at
the effects of nanometre features from micrometre roughness 150 ◦ C. The TGA curves show that the wax is stable, i.e., there
on the wetting behaviour of lotus leaves. is no significant mass loss, up to 150 ◦ C. Thus, the wax from
the melted hair-like structures is believed to be still present on
the leaf surface after annealing at 150 ◦ C.
2. Experimental details
The chemical structure of the dry and treated leaves
Dry lotus leaves were purchased from a local Chinese grocery was examined using Fourier transform infrared spec-
store and, for comparison, freshly cut lotus leaves and a live troscopy (FTIR) and the attenuated total reflectance (ATR)
lotus plant were obtained from a nursery. Since there is little method on a ThermoNicolet Magna 760-IR spectrometer
discernable difference in the wetting behaviour between the ([Link]). Two pieces of leaf were cut and clamped
live and dried leaves, the results reported here were mostly against the sides of a germanium ATR crystal (25 mm×5 mm ×
from experiments with the dry leaves. Leaf samples were cut 2 mm, 45◦ angle) and 32 scans were signal-averaged at 4 cm−1
from similar regions of the plant, avoiding the large plant veins spectral resolution. With this method, the depth of penetration
so as to give a relatively consistent surface. Some of the leaf is wavelength dependent, but ranges from about 0.4 to 1.0 µm.
samples were annealed by baking the leaves in an oven in FTIR measurements were performed on the dry and annealed
air at 150 ◦ C for 1 h. For comparison, smooth carnauba wax lotus leaves to identify possible changes in chemical composi-
films were made by casting the wax (supplied by Poth Hille) tion, specifically in the surface wax type and content. Refer-
on smooth Pyrex glass surfaces and air cooling. The reason ence IR spectra for untreated and annealed carnauba wax were
for choosing the carnauba wax will be made clear later in this obtained via transmission. The data (figure 3) show similarity
paper. between the treated and annealed lotus leaves. When compared
The microstructures of the dry and annealed leaves with reference spectra, the resultant spectra appear to represent
were examined by SEM (figure 1). Samples were given a combination of cellulose and a wax similar to carnauba wax,

1360
 Effects of micro- and nano-structures on the self-cleaning behaviour of lotus leaves

Figure 3. FTIR data for (a) untreated lotus leaves, (b) annealed lotus leaves, (c) untreated carnauba wax, (d) annealed carnauba wax, and
(e) a standard plot for cellulose.

(a) (b)

Figure 4. Graphical representation of FTIR measurement. The

4mm
leaves and mounds are covered with wax; for the untreated leaf, the
wax coating is thicker due to its porous, hair-like makeup, whereas
the coating is thin and smooth for the heat-treated leaf. Figure 5. Images of 2.5 mm diameter drops of water: (a) placed on
an untreated dry lotus leaf, and (b) placed on an annealed lotus leaf,
then tilted to an angle of 90◦ .

which is an exudate from leaves of the wax palm, Coperica

cerifera. Further chemical analysis similar to tests performed Taking the SEM, TGA and FTIR results as a whole, it is
by Koch et al [25] and Matas et al [26] would be required to reasonable to conclude that the major effect of annealing at
fully identify the wax present on the leaf. However, due to this
150 ◦ C is to melt the nano-scale hair structure and keep the
similarity and its availability, carnauba wax was subjected to
wax composition and quantity nearly unchanged on the lotus
the same contact angle measurements as the lotus leaves for
leaf surface.
comparison.
For water contact angle measurements, dry and annealed
The wax present on the surface of the dry and annealed
leaves of 2.5 cm squares were mounted on smooth metal plates
lotus leaves does not appear to undergo any gross chemical
with double-sided foam tape to keep the leaves flat. Contact
change, based on the IR data. The primary C–H stretching
angle measurements were taken using a Krüss DSA10L system
peaks present at 2915 and 2850 cm−1 remain the same (except
with DSA1 drop shape analysis software. Drops of 18 M
for relative intensity), and there is no obvious growth of new
deionized water with a diameter of 2.5 mm were deposited
peaks. Additionally, the ratio of intensities of the carnauba-like
on the leaf surface and the software determined the static
peaks versus cellulose peaks appears to suggest less wax on the
contact angle by using a sessile drop fitting routine (Krüss). A
surface of the baked leaf. However, this relative decrease in
programmable tilting table was used to observe the water drops
signal from the wax is most likely due to the shallow detection
on inclined leaf surfaces. All contact angle measurements were
depth of the FTIR measurement (∼1 µm), which is only deep
carried out at 20–23 ◦ C and 45–55% relative humidity.
enough to probe the composition of the top of the mounds.
The combination of the melting of the wax by heat treatment
and the shallow depth of the FTIR measurement result in the 3. Results and discussion
detection of a lower ratio of wax to cellulose (3400 cm−1 ) as
visualized in figure 4. Furthermore, the FTIR results show The measured static contact angle of water drops placed on the
that the structure of the carnauba wax is also unchanged after dry leaf is 142.4 ± 8.6◦ (figure 5(a)). In contrast, the static
thermal annealing at 150 ◦ C. contact angle of water drops placed on the heat-treated leaf is

1361
Y T Cheng et al

126.3 ± 6.2◦ . The average values were the results of measuring falling water to bead up and roll off the leaf. However, by
14 drops placed on the surfaces. For comparison, the static heating the leaf at 150 ◦ C, the nano-scale hair-like structures
water contact angle on the smooth carnauba wax is found to be can be melted, coating the leaf surface. This process, while
74.0 ± 8.5◦ , suggesting that the carnauba wax is hydrophilic. keeping the chemical composition of the surface approximately
This is consistent with the observations by Wagner et al that the same, alters the roughness such that the contact angle is
waxes in general are not very hydrophobic because they have decreased and water droplets no longer roll off the leaf. This
many hydrophilic functional groups [24]. Assuming that the study clearly shows the importance of the lotus leaf’s nano-
chemical composition of the wax on the lotus leaf is similar scale hair-like structure on its self-cleaning ability.
to carnauba wax, these results suggest that the micro-scale
roughness in the form of mounds on lotus surfaces is effective Acknowledgments
in increasing the static contact angle of a likely hydrophilic
wax with a contact angle similar to the carnauba wax’s 74◦ to We would like to thank Wes Capehart, Howard Cox, Mike
the hydrophobic 126◦ seen on the baked lotus leaf, and that Lukitsch, Wangyang Ni, Tom Perry, Yue Qi, Will Rodgers,
the presence of the nano-scale hair-like structure is responsible Jim Spearot, Steve Swarin, John Ulicny, Mark Verbrugge,
for the additional increase of 16◦ in contact angle from 126◦ to Anita M Weiner, and Tao Xie for assistance, comments, and
142◦ seen in the untreated lotus leaf. This decrease in contact suggestions.
angle is similar to a decrease in contact angle often seen when
making replicas of surfaces with multi-levelled roughness. In
References
such a case, the replica is unable to duplicate the nano-scale
roughness leaving only the micro-scale roughness and, as a [1] Barthlott W and Neinhuis C 1997 C. Planta 202 1
result, smaller contact angles [16]. Thus, the high contact [2] Neinhuis C and Barthlott W 1997 Ann. Botany 79 667
angle of water drops on lotus leaves is mainly the result of the [3] Onda T, Shibuichi S, Satoh N and Tsujii K 1996 Langmuir
12 2125
presence of micro-scale surface roughness in conjunction with
[4] Shibuichi S, Onda T, Satoh N and Tsujii K 1996 J. Phys. Chem.
the nano-scale hair-like structures. 100 19512
A dramatic effect is the ability of the water drops to [5] Öner D and McCarthy T J 2000 Langmuir 16 7777
move on the dry and annealed leaf surfaces. While water [6] Erbil H Y, Demirel A L, Avci Y and Mert O 2003 Science
drops placed on the dry leaves roll off at the slightest incline, 299 1377
[7] Miwa M, Nakajima A, Fujishima A, Hashimoto K and
the same size drops become sticky drops on the annealed
Watanabe T 2004 Langmuir 15 5754
leaf, adhering to the surface even when the leaf is tilted 90◦ [8] Bico J, Tordeux C and Quéré D 2001 Europhys. Lett. 55 214
(figure 5(b)) or turned upside down. The annealing has, [9] Wu Y, Sugimura H, Inoue Y and Takai O 2002 Chem. Vapor
therefore, changed the drops from rolling drops to sticky drops. Depos. 8 47
Evidently, the disappearance of nano-scale hair-like structures [10] Feng L, Li S, Li Y, Li H, Zhang L, Zhai J, Song Y, Liu B,
Jiang L and Zhu D 2002 Adv. Mater. 14 1857
is responsible for the transition. These observations show the
[11] Li S, Li H, Wang X, Song Y, Liu Y, Jiang L and Zhu D 2002
critical role of nano-scale hair structure in making lotus leaves J. Phys. Chem. B 106 9274
self-cleaning when water drops are placed on the leaves as [12] Li M, Zhai J, Liu H, Song Y, Jiang L and Zhu D 2003 J. Phys.
strong water adhesion occurs if the leaf surfaces have lost their Chem. B 107 9954
nano-scale hair-like structures. [13] Lau K K S, Bico J, Teo K B K, Chhowalla M,
Amaratunga G A J, Milne W L, McKinley G H and
Strong water adhesion can also occur when water is
Gleason K K 2003 Nano Lett. 3 1701
condensed on the leaf surfaces: the condensed water drops [14] Shiu J-Y, Kuo C-W, Chen P and Mou C-Y 2004 Chem. Mater.
are ‘sticky’ and do not roll [27]. The water condensation 16 561
experiment suggests that small water droplets formed from the [15] de Gennes P-G, Brochard-Wyart F and Quéré D 2004
vapour phase can be trapped in the nano-scale hairy regions D. Capillarity and Wetting Phenomena: Drops, Bubbles,
Pearls, Waves (New York: Springer)
on the lotus leaf surface. The trapped drops remain trapped as
[16] Fürstner R, Barthlott W, Neinhuis C and Walzel P 2005
they grow or become connected to form larger ‘sticky’ drops Langmuir 21 956
with large apparent contact angles [28]. Both the condensation [17] Cassie A B D and Baxter S 1944 Trans. Faraday Soc. 40 546
and annealing experiments can be explained by the intrinsic [18] Wenzel R N 1936 Ind. Eng. Chem. 28 988
hydrophilic behaviour of the wax. Thus, the two-level Wenzel R N 1949 J. Phys. Colloid. Chem. 53 1466
[19] Marmur A 2003 Langmuir 19 8343
roughness, especially the nano-scale hairs, is responsible for
[20] Quéré D, Lafuma A and Bico J 2003 Nanotechnology 14 1109
the rolling behaviour of water drops placed on lotus surfaces. [21] Patankar N A 2004 Langmuir 20 8209
However, the nano-hairs cannot prevent water from becoming [22] Otten A and Herminghaus S 2004 Langmuir 20 2405
‘sticky’ drops when water is condensed on the leaves. These [23] McHale G, Shirtcliffe N J, Aqil S, Perry C C and Newton M I
results confirm the importance and limitation of the nano-scale 2004 Phys. Rev. Lett. 93 036102
[24] Wagner P, Fürstner R, Barthlott W and Neinhuis C 2003 J. Exp.
roughness on the self-cleaning ability of lotus leaves.
Botany 54 1
[25] Koch K, Neinhuis C, Ensikat H-J and Barthlott W 2004 J. Exp.
4. Conclusion Botany 55 711
[26] Matas A J, Sanz M J and Heredia A 2003 Int. J. Biol.
Macromol. 33 31
The lotus leaf is renowned for its ability to shed water due to [27] Cheng Y-T and Rodak D E 2005 Appl. Phys. Lett. 86 144101
its multi-levelled roughness. The combination of the micro- [28] Cheng Y-T, Rodak D E, Angelopoulos A and Gacek T 2005
scale mounds and the nano-scale hair-like structures causes Appl. Phys. Lett. 87 194112

1362