J Mater Sci (2012) 47:3690–3697 DOI 10.1007/s10853-011-6217-x
A low-cost method to produce superhydrophobic polymer surfaces J. J. Victor • D. Facchini • U. Erb
Received: 14 November 2011 / Accepted: 17 December 2011 / Published online: 7 January 2012 Ó Springer Science+Business Media, LLC 2012
Abstract Here, we introduce a novel and inexpensive template-based structuring process to create superhydrophobic polymer surfaces adapted from the naturally occurring micro/nano structured surfaces found on the superhydrophobic leaves of the quaking aspen tree. Electroformed nanocrystalline nickel coupons were sandblasted and chemically etched to create a negative reproduction of the aspen leaf surface structure. These nanocrystalline nickel samples were then employed as re-useable templates and pressed against various polymers at elevated temperatures, transferring the desired superhydrophobic structure to their surfaces. This structuring process resulted in water contact angles above 150° and tilt angles below 5° for polyethylene, polypropylene and polytetrafluoroethylene samples. In addition, the effects of temperature, water drop size and surfactant concentration on these pressed polymer surfaces were investigated to assess potential application limitations for these surfaces. Introduction Within the last decade researchers have attempted to reproduce the non-wetting surface effect found on the Lotus leaf and other superhydrophobic and self-cleaning biological surfaces such as insect wings. It has previously been shown that this extreme non-wetting effect on leaves can be J. J. Victor U. Erb (&) Department of Materials Science and Engineering, University of Toronto, 184 College St. Room 140E, Toronto, ON M5S 3E4, Canada e-mail:
[email protected] D. Facchini Integran Technologies Inc., 1 Meridian Rd., Toronto, ON M9W 4Z6, Canada
123
attributed to a complex dual-scale surface structure, consisting of micro-scale papillae and low surface energy nanoscale wax platelets [1–6]. Potential applications for these types of surfaces include, but are not limited to, self-cleaning windows, glasses, paints, textiles and fabrics; low-friction surfaces that minimize flow resistance in macro-scale pipelines or micro-fluidic channels; and surfaces with locally tailored electrical properties for MEMS/NEMS (micro/nano electric mechanical systems) components. For self-cleaning applications these surfaces will act as an important labour saving device and result in a reduction of the use of environmentally harmful cleaning agents. For many other potential applications the introduction of this technology will result in more effective and efficient components. Specifically for MEMS/NEMS, which require both hydrophobic surfaces and interfaces with low adhesion and friction, these types of fabricated surfaces would be very attractive. There have been many successful attempts at artificially reproducing the type of surface structure found on the Lotus leaf. Many of these processes are time consuming and very expensive or have size/scaling restrictions that limit their use for practical applications. Currently, the five most common methods to create self-cleaning surfaces are lithography [7, 8], sol–gel [9, 10], plasma enhanced CVD (chemical vapour deposition) [11], nanocasting [12] and chemical or plasma etching techniques [13–15]. All of these fabrication techniques have one important commonality: they all create rough patterned or porous surfaces to which a thin hydrophobic layer can be applied. Lithography, sol–gel and plasma enhanced CVD are expensive methods that use nanomasks to produce highly structured surfaces. Nanocasting is the most direct reproduction approach, but the resulting self-cleaning surface is limited in size by the dimensions of an actual Lotus leaf. This method involves creating a template by casting liquid
J Mater Sci (2012) 47:3690–3697
PDMS (polydimethylsiloxane) onto a Lotus leaf, and using this template to subsequently structure other surfaces [12]. The objective of this research is to reproduce the naturally occurring surface structure found on one specific superhydrophobic leaf: the quaking aspen leaf. Here, we introduce a novel template-based method to quickly and inexpensively produce superhydrophobic polymer surfaces using these leaves’ surface structures as a blueprint. The main advantage to this method over currently available techniques is its ease of scalability to allow for a high throughput of large and/or complex-shaped products. The surface structures and wetting properties of polymer samples structured using this process have been characterized. In addition, the effects of temperature, water drop size and surfactant concentration on the wetting properties have been studied to explore a range of potential application conditions for these structured surfaces.
Methods and materials Quaking aspen leaf surfaces Several quaking aspen leaves were harvested from a forest near Peterborough, Ontario. Fresh leaves were cut into 1 inch 9 1 inch sections and mounted on flat Plexiglas coupons using double sided tape. Care was taken to ensure that mounted sections contained no major leaf veins or areas that had previously been mechanically damaged. Two leaf samples were selected for scanning electron microscopy and were subsequently carbon coated (Edwards Coating System—E306A) to increase electrical conductivity. Water contact and tilt angle measurements were performed on the remaining mounted leaf surfaces using 5 lL droplets and 25 lL droplets, respectively. Template synthesis Flat 1 inch 9 1 inch sections of electroformed nanocrystalline nickel (Integran Technologies Inc.) were ground, finished with a 1 micron polishing compound (Buehler) and ultrasonically cleaned in ethanol. Subsequently, each sample was sand blasted at 87 psi using 180 grit alumina particles (Chandler Industrial Supplies Ltd.), ultrasonically cleaned in ethanol and chemically etched in a 5% nitric acid solution (Fisher Scientific) for 30 min. The resulting structured nanocrystalline nickel samples were then employed as reusable templates to structure the surfaces of softened polymers. Polymer pressing Commercially available 1 inch 9 1 inch sections of polyethylene (PE), polypropylene (PP) and polytetrafluoroethylene
3691
polymer template restraining device
Fig. 1 Pressing apparatus
(PTFE) (McMaster-Carr) were heated above their respective softening temperatures in contact with a structured nanocrystalline nickel template and a steel restraining device, all inside a stainless steel press (Fig. 1). Once the desired pressing temperature was reached (150 °C for PE, 160 °C for PP and 280 °C for PTFE), the press (with the polymer and template inside) was removed from the furnace and immediately tightened, compressing the structured template into the softened polymer. The template and polymer were then removed from the press and easily separated using a set of pliers. A schematic summary of the complete pressing and surface structuring procedure is given in Fig. 2. Surface characterization and contact angle measurements The surfaces of structured templates and pressed polymers were characterized using an optical profilometer (WYKO interferometric profilometer) and an SEM (Hitachi S-4500 field emission scanning electron microscope). Polymer surfaces were carbon coated (Edwards coating system— E306A) to introduce optical reflectivity and electrical conductivity necessary for both characterization techniques. Several 5 lL drops of de-ionized water with varying surfactant (sodium dodecyl sulphate, SDS—Bioshop Canada Inc.) concentrations (up to 100 g/L) were placed on multiple areas of each structured polymer surface using a controlled dispensing micropipette (Clonex Corporation) and their contact angles (CA) were measured using a standard goniometer. The effect of temperature on the wetting properties of un-pressed and pressed surfaces was investigated by heating each sample from room temperature to 95 °C, in a furnace and measuring their water contact angle at 10 °C intervals. To study the effect of droplet size, water contact angle measurements were carried out for drop sizes ranging from 5 to 25 lL. In addition, to investigate the adhesion between droplets and the solid surface, tilt angles (TA—angle a surface must be tilted to initiate water droplet roll-off) for all polymer samples were determined using a tilting stage and 25 lL droplets.
123
3692
J Mater Sci (2012) 47:3690–3697
Fig. 2 Schematic diagrams summarizing the surface structuring procedure
Results Quaking aspen leaf surface Figure 3 shows an example of the surface of a superhydrophobic quaking aspen leaf consisting of an array of micro-scale protrusions (left) covered by a finer nano-scale surface roughness (middle). This dual-scale surface structure was used as a biological blueprint in the polymer structuring process. The water contact angles, using 5 lL droplets, and tilt angles, using 25 lL droplets on these leaves are 166° ± 3° and \5°, respectively. Template characterization After sandblasting and chemical etching in nitric acid (step 2 in Fig. 2), all structured nanocrystalline templates showed very similar surfaces, consisting of larger microscale pits and finer nano-scale roughness features. SEM micrographs showing both roughness scales found on these
templates are presented in Fig. 4. It is important to note that the etched nanocrystalline template essentially shows the ‘negative’ structure features found on the quaking aspen leaf surface for both the micrometer and nanometer sized surfaces roughness scales. An example of an optical profilometry image of a structured template that was used to measure the size and spacing of the micro-scale pits found on these surfaces is shown in Fig. 5; numerical results from this analysis are presented in Table 1. Characterization of structured polymer surfaces All structured polymer surfaces (after step 6 in Fig. 2) were characterized in the same manner as the templates. Examples of SEM and optical images of a pressed PP surface are shown in Figs. 6 and 7, respectively. The size and spacing of the micro-scale protrusions found on all structured polymer surfaces are presented in Table 1. By comparing Figs. 4, 5, 6, and 7 and the
Fig. 3 SEM micrographs of the adaxial side of a quaking aspen leaf: low magnification image illustrating the array of micro-scale protrusions (left), an individual micro-scale pit covered by a finer nano-scale roughness (middle) and a 5 lL water droplet on this surface (right)
123
J Mater Sci (2012) 47:3690–3697
3693
Fig. 4 SEM micrographs of a structured template: a low magnification image illustrating the array of micro-scale pits, b and c an individual micro-scale pit, and d high magnification image showing finer nano-scale roughness
other words, the filling of the etched depressions on the template by the softened polymers is excellent. Contact angle measurements
Fig. 5 Optical profilometry image of a structured template: surface view
Table 1 Surface structure results Surface
Height/depth (lm)
Diameter (lm)
Protrusion/pit spacing Interspacing (lm)
17.5
24.5
25.0
Pressed PE
15.4
24.0
22.6
Pressed PP
14.7
18.4
28.3
Pressed PTFE
16.7
26.5
23.8
Nanocrystalline nickel template
Protrusion/pit size
quantitative results given in Table 1, it can be seen that this pressing process results in a good transfer of all surface features from the templates to the pressed polymers. In
Water contact angles using 5 lL droplets and roll-off tilt angles using 25 lL droplets for all pressed and un-pressed polymers are given in Table 2. The data show that the pressing process increased the contact angles for PE, PP and PTFE from 96° to 151°, 104° to 153° and 108° to 159°, respectively. In addition, the tilt angles for all polymers dropped from over 30° to below 5°. Clearly, the presented structuring process drastically increased water contact angles and decreased tilt angles for all polymer samples. In other words, all polymer surfaces have effectively been transformed from hydrophobic (CA [ 90°) to superhydrophobic (CA [ 150° and TA \ 5°) from this templatebased surface modification process. The effects of temperature, water drop size and surfactant concentration on the contact angles of pressed and unpressed PP and PTFE are illustrated in Figs. 8, 9, and 10, respectively. For all tested samples, small linear decreases in contact angles were observed with increasing temperature and droplet size (Figs. 8, 9). All samples tested, except for pressed PTFE, showed a large sharp drop in contact angle with small additions of surfactant. Pressed PTFE experienced a relatively small drop in contact angle, though. As the SDS concentration increased, the contact angles for all samples began to level out and remained relatively constant. Pressed PTFE was
123
3694
J Mater Sci (2012) 47:3690–3697
Fig. 6 SEM micrographs of a pressed PP sample: a and b low magnification images illustrating the array of micro-scale protrusions, c an individual micro-scale protrusion, and d high magnification image showing finer nano-scale roughness
180
o
Contact Angle ( )
150
120
90
60
Pressed PTFE Pressed PP
30
Un-pressed PTFE Un-pressed PP
Fig. 7 Optical profilometry image of a pressed PP sample: surface view
0 20
30
40
50
60
70
80
90
100
o
Temperature ( C) Table 2 polymer wetting properties Surface Un-pressed PE
CA (°) 96 ± 5
TA (°) 38
Un-pressed PP
104 ± 5
37
Un-pressed PTFE
108 ± 3
36
Pressed PE
151 ± 3
\5
Pressed PP
153 ± 5
\5
Pressed PTFE
159 ± 4
\5
the only tested sample to retain its hydrophobic property even at the highest SDS concentration tested of 100 g/L. Un-pressed PTFE and both pressed and un-pressed PP lost
123
Fig. 8 Effect of temperature on the water contact angle of pressed and un-pressed polymers
their hydrophobic/superhydrophobic properties at surfactant concentrations as low as 5 g/L.
Discussion Surface structure and wetting properties Many biological surfaces having a dual-scale micro/nano hierarchical structure coupled with a hydrophobic surface chemistry can exhibit extreme non-wetting properties e.g.,
3695
180
Effect of temperature
150
Not surprisingly, the effect of temperature on the contact angles of both pressed and un-pressed PP and PTFE were quite similar (Fig. 8). All samples showed a small linear decrease in contact angles (\10°) over the tested temperature range. The observed reductions in contact angles are likely a direct result of the decreased liquid/vapour surface tension at higher temperatures [18]. This reduced liquid/ vapour surface tension allows water to better wet the polymer surfaces resulting in slightly lower contact angles. However, even at the highest temperatures tested, both pressed polymer samples still exhibited relatively high contact angles indicating that these types of surfaces would be suitable for applications in which they are exposed to elevated temperatures up to 90 °C.
o
Contact Angle ( )
J Mater Sci (2012) 47:3690–3697
120
90
Pressed PTFE
60
Pressed PP Un-pressed PTFE
30
Un-pressed PP 0 0
5
10
15
20
25
30
Drop Size ( µ l) Fig. 9 Effect of water drop size on the contact angle of pressed and un-pressed polymers Pressed PTFE Pressed PP Un-pressed PTFE Un-pressed PP
180
o
Contact Angle ( )
150 120 90 60 30 0 0
20
40
60
80
100
Sodium Dodecyl Sulfate Concentration (g/L) Fig. 10 Effect of SDS concentration on the water contact angle of polymers
[1–6, 16]. In this study, this effect has been demonstrated for the leaves of the quaking aspen. These naturally occurring superhydrophobic surfaces exhibit a high density of micro-scale surface protrusions (papillae) covered by a finer nano-scale roughness (3D wax crystalloids) [6]. All of the pressed polymers surfaces presented here (Figs. 6 and 7) show similar surface structures as the aspen leaves (Fig. 3). On these types of surfaces, micro-scale protrusions enable the formation of trapped air pockets between the droplet and the solid surface, whilst the nanoscale features significantly increase the surface roughness. According to Cassie and Baxter’s heterogeneous wetting equation (cos h = Rf fsl cos ho - fla), for an inherently hydrophobic material (ho [ 90°), increasing the surface roughness (Rf) and maximizing the amount of air trapped under a droplet (fla) results in a superhydrophobic surface with a large contact angle (h) [17].
Effect of drop size Both pressed and un-pressed polymer samples showed small, linear decreases in their respective contact angles with increasing drop size (Fig. 9). Over the tested droplet size range (5–25 lL), the decrease in contact angle for the pressed PP samples (14°) was slightly higher than for the pressed PTFE samples (7°). Whilst only the pressed PTFE samples retained contact angles above 150°, both materials still showed relatively high contact angles for the largest drop sizes, indicating their non-wetting, and potential selfcleaning properties are not lost for larger drop sizes. This important result points out that these types of surfaces would be well suited for many exterior applications, where they will be frequently exposed to rain of different drop sizes. A similar weak droplet size dependence on contact angles was previously observed for quaking aspen and bigtooth aspen leaves [6]. Effect of surfactant concentration All polymer samples tested exhibited a similar trend in their wetting behaviour to increasing surfactant concentrations (Fig. 10). Initially, contact angles sharply dropped with small additions of surfactant (up to 10 g/L) and then levelled out as the concentration was further increased to 100 g/L. This response was expected since surfactants are compounds that promote wetting by collecting at the liquid interface where they act to reduce the liquid surface tension, and ultimately lower the contact angle on any solid surface [19]. However, even at the highest SDS concentration tested (100 g/L), at which all other samples had contact angles less than 90°, the pressed PTFE sample still demonstrated hydrophobic behaviour with a contact angle of *128°. In comparison, at this concentration the contact angle on the un-pressed PTFE sample was only *60°,
123
3696
indicating its loss of hydrophobicity. On the other hand, at this SDS concentration, the contact angle for pressed PP was *72°, whilst the un-pressed PP surface only had a contact angle of *45°. In general, structuring any surface will only result in an enhancement of the pre-existing wetting condition with a given liquid. Since both un-pressed PP and PTFE samples were wetted by the solution containing the highest surfactant concentration (CA \ 90°), it would be expected that the structuring process would enhance the pre-existing wetting condition for PP and PTFE resulting in lower contact angles; however, the opposite effect was observed. This implies that even though the intrinsic contact angles for these polymers are below 90° with this specific solution, a significant amount of air is still trapped between the liquid droplets and the pressed/structured solid surfaces. Theoretically, according to Cassie and Baxter’s heterogeneous wetting equation, a surface with an intrinsic contact angle less than 90° can be structured to increase its contact angle, provided a sufficient amount of air is able to be trapped between the liquid and solid phases [17]. This indicates that these types of structured surfaces are better suited to resist surfactant-induced wetting than their chemically equivalent smooth counterparts. It is interesting to note that the effect of SDS concentration observed in this current study for pressed PTFE is almost identical to what was earlier reported for quaking aspen leaves [6]. In both cases, the contact angle remained relatively high even at the highest SDS concentration of 100 g/L: 128° for pressed PTFE and 122° for quaking aspen leaves. This implies that both surfaces are indeed very similar not only with respect to overall surfaces structure characteristics but also in terms of their response to the wetting agent SDS. Process scalability potential As discussed earlier, one of the main barriers to having micro-nanostrucutred surfaces integrated into large scale production is their relative cost, and the difficulty associated with structuring more complex geometries. For the process described here the starting template material is an electrodeposited metal. It is relatively easy to plate the outside of a large roller and then structure this roller through a sand blasting and etching process to be used as a template. This template roller could then be used in a continuous ‘pressing’ process where large sheets of polymers are fed underneath the roller giving them the desired surface structure. Since template production is relatively inexpensive, this process could result in low-cost, largescale production of superhydrophobic polymeric surfaces. In addition, this non-wetting property can be imparted onto the surfaces of complex shaped polymer articles by changing from a simple pressing process to an injection moulding process. It is possible to electroplate the inside
123
J Mater Sci (2012) 47:3690–3697
surface of a complex shaped polymer mould by an electroforming process, followed by sand blasting and acid etching. This would result in superhydrophobic surfaces on every part coming from this mould.
Conclusions A technologically viable and economically feasible template process for producing superhydrophobic polymeric surfaces has been developed. PE, PP, and PTFE samples were structured using this process which created a fine nano-scale surface roughness superimposed on larger micro-scale surface protrusions. This resulted in an increase in contact angle from 96° to 151° for PE, 104° to 153° for PP and 108° to 159° for PTFE and a reduction in tilt angles from above 30° to below 5° for all pressed polymer samples. In addition, the effects of temperature, drop size and surfactant concentration on the wetting properties of these polymers were investigated. An increase in droplet size, temperature and water surfactant concentration decreased the measured contact angles for all samples. The effects of drop size and temperature were weak and resulted in all pressed polymer surfaces remaining hydrophobic for droplet sizes between 5 and 25 lL and temperatures up to 90 °C. Initially, all samples except pressed PTFE showed drastic reductions in contact angles with increasing surfactant concentrations. At the highest surfactant concentrations of 100 g/L, all samples had completely lost their superhydrophobicity but pressed PTFE remained hydrophobic with a contact angle of 128°. Overall the wetting characteristics of structured PTFE and quaking aspen leaves are very similar. Acknowledgement The authors thank Sal Boccia and Milos Kucera for help with the SEM and the optical profilometer. This research was sponsored by the Natural Sciences and Engineering Research Council of Canada, the Ontario Research Fund—Research Excellence, the Ontario Graduate Scholarship in Science and Technology, the University of Toronto Open Fellowship and Integran Technologies Inc., Toronto, Canada.
References 1. 2. 3. 4. 5. 6. 7. 8. 9.
Neinhuis C, Barthlott W (1997) Ann Bot 79:667 Koch K, Bhushan B, Barthlott W (2008) Soft Matter 4:1943 Koch K, Bhushan B, Barthlott W (2009) Prog Mater Sci 54:137 Bhushan B, Jung YC (2006) Nanotechnology 17:2758 Cheng YT, Rodak DE, Wong CA, Hayden CA (2006) Nanotechnology 17:1359 Victor JJ, Erb U (2010) Int J Micro Nano Scale Transp 1:323 Oner D, McCarthy T (2000) Langmuir 16:7777 Bico J, Marzolin C, Quere D (1999) Europhys Lett 47:220 Shirtcliffe NJ, McHale G, Newton MI, Perry CC (2003) Langmuir 19:5626
J Mater Sci (2012) 47:3690–3697 10. Erbil HY, Demirel AL, Avci Y, Mert O (2003) Sci 299:1377 11. Lau KKS, Bico J, Teo KBK, Chhowalla M, Amaratunga GAJ, Milne WI, McKinley GH, Gleason KK (2003) Nano Lett 3:1701 12. Sun M, Luo C, Xu L, Ji H, Ouyang Q, Yu D, Chen Y (2005) Langmuir 21:8978 13. Qian B, Shen Z (2005) Langmuir 21:9007 14. Teshima K, Sugimura H, Inoue Y, Takai O, Takano A (2005) Appl Surf Sci 244:619
3697 15. Jansen H, de Boer M, Legtenberg R, Elwenspoek M (1995) J Micromech Microeng 5:115 16. Cheng YT, Rodak DE (2005) Appl Phys Lett 86:144101 17. Cassie A, Baxter S (1944) Trans Faraday Soc 40:546 18. Adam NK (1941) The physics and chemistry of surfaces, 3rd edn. Oxford University Press, Oxford 19. Rosen MJ (2010) Surfactants and interfacial phenomena, 3rd edn. Wiley, New Jersey
123
