5942
Langmuir 2008, 24, 5942-5948
Dynamic Control of Surface Energy and Topography of Microstructured Conducting Polymer Films Xianjun Wang,*,† Magnus Berggren,† and Olle Ingana¨s‡ Organic Electronics, Department of Science and Technology (ITN), Linko¨ping UniVersity, SE-601 74 Norrko¨ping, Sweden, and Biomolecular and Organic Electronics, Department of Physics, Chemistry and Biology (IFM), Linko¨ping UniVersity, SE-581 83 Linko¨ping, Sweden ReceiVed December 22, 2007. ReVised Manuscript ReceiVed March 5, 2008 Microstructured polymer surfaces, including conducting and insulating polymers, have been prepared to achieve electrochemical control of the surface energy and topography. The reported surface switches include pillar- and mesh-like surface patterns of polypyrrole (PPy), poly(3,4-ethylene-dioxythiophene) (PEDOT), and photoresists. The structures have been evaluated by contact angle measurements and optical and scanning electron microscopy to determine the surfaces characteristics. These microstructured polymer surface switches can be electrochemically modified from dewetting to wetting conditions, with a maximum associated change of the water contact angle from 129° to 44°. This contact angle switching was observed for samples in which dynamic control of the surface topography and surface tension was coupled. Control of topography was achieved with a dynamic height-switching range of more than 3 µm. In addition, dynamic control of anisotropic wetting is reported. Our experiments were carried out under conditions that are suitable for a biointerface, implying potential application in biotechnology and cell science. In particular, switching of the energy, chemistry, and topography of the surface, along with their associated orientation, are interesting features for dynamic (electronic) control of the seeding and proliferation for living cells. The technology reported promises for electronically controlled cell-growth within Petri dishes, well plates, and other cell-hosting tools.
1. Introduction Design and manufacture of surfaces with controllable wettability and topography is an attractive topic in material science1 because of its importance in many practical applications, such as for self-cleaning surfaces,2 antibiofouling,3 protective coatings,4 and printed electronics.5 Further, dynamic control of the surface character is also of great interest for biotechnology since the surface chemistry and topography are highly potent cues that define the cell seeding and growth characteristics along artificial surfaces.6–8 Surface switches have been explored as a dynamic biointerface to regulate the attachment, growth, differentiation, and release of living cells.9,10 Wettability is governed by the surface chemistry and surface topography in combination.11,12 Dynamical changes in either one or two of these factors may modulate surface wettability. The shape of a liquid droplet standing on a solid surface depend on the surface chemistry, which is caused by many interactions between the molecules at * Corresponding author. † Department of Science and Technology (ITN). ‡ Department of Physics, Chemistry and Biology (IFM).
(1) Feng, X. J.; Jiang, L. AdV. Mater. 2006, 23, 3055–3062. (2) Yuan, Z.; Chen, H.; Tang, J.; Gong, H.; Liu, Y.; Wang, Z.; Shi, P.; Zhang, J.; Chen, X. J. Phys. D 2007, 40, 3485–3489. (3) Zhong-Yi, J.; Yan-Qiang, W.; Yan-Lei, S.; Qiang, S.; Ma, X. L. J. Membr. Sci. 2006, 286, 228–36. (4) Berridge, M. J.; Bootman, M. D.; Lipp, P. Nature 1998, 395, 645. (5) Wang, J. Z.; Zhang, Z. H.; Li, H. W.; Huck, W. T. S.; Sirringghause, H. Nat. Mater. 2004, 3, 171–176. (6) Recknor, J. B.; Recknor, J. C.; Sakaguchi, D. S.; Mallapragada, S. K. Biomaterials 2004, 25, 2753–2767. (7) Ateh, D. D.; Vadgama, P.; Navsaria, H. A. Tissue Eng. 2006, 12, 645–655. (8) Gomez, N.; Schmidt, C. E. J. Biomed. Mater. Res., Part A 2006, 81, 135– 149. (9) Argun, A. A.; Aubert, P. H.; Thompson, B. C.; Schwendeman, I.; Gaupp, C. L.; Hwang, J.; Pinto, N. J.; Tanner, D. B.; MacDiarmid, A. G.; Reynolds, J. R. Chem. Mater. 2004, 16, 4401–4412. (10) Wong, J. Y.; Langer, R.; Ingber, D. E. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 3201–3204. (11) De Gennes, P. G.; Brochard-Wyart, F.; Quere, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, WaVes; Springer: New York, 2004. (12) Que´re´, D. Rep. Prog. Phys. 2005, 68, 2495–2532.
the interface of solid and liquid phases, including electrostatic, van der Waals interaction, dipole-dipole interaction, and hydrogen bonding. On the other hand, the surface topography may induce an enhanced interfacial area. The established theories11,12 of surface science demonstrate that an increase of the surface roughness leads to an increased contact angle (CA), if the CA is already on the dewetting side. Young’s equation for a flat surface is
γL cos θ + γSL ) γS
(1)
where γL, γSL, and γS are the interfacial surface tension of the liquid/air, liquid/solid, and solid/air interfaces, respectively. θ is Young’s CA on the flat surface at equilibrium. If the surface is not flat, consideration of the surface thermodynamics as the liquid phase wets the rough surface everywhere leads to the Wenzel equation11,12
cos θ/ ) r cos θ
(2)
where r is the surface roughness, defined as the ratio of real surface area over the projected surface area normal to the substrate. θ is Young’s CA as described in eq 1, while θ* is the apparent CA of the liquid on the rough surface. When the surface contains domains of dewetting material, or when the surface is sufficiently rough, we arrive at the fakir situation. Here, the drop of the fluid is carried on pillars, and air/vapor is trapped in cavities beneath the liquid, as described by the Cassie-Baxter equation11,12
cos θ/ ) fS cos θS + fV cos θV
(3)
where fS and fV are the area fractions of the solid and vapor, respectively. Polypyrrole (PPy) is a conducting polymer commonly explored and used in various electrochemical devices particularly targeting applications in biology. It exhibits excellent biocompatibility and biostability characteristics in biologically relevant environ-
10.1021/la704006z CCC: $40.75 2008 American Chemical Society Published on Web 05/03/2008
Dynamic Control of Microstructured Polymer Films
Langmuir, Vol. 24, No. 11, 2008 5943
ments. As the oxidation state of PPy is altered, its volume and bulk chemical characteristics change, providing a principle of operation for novel electrochemical actuators.13,14 For instance, PPy-based robotic devices is a promising platform of tools for future surgery and prosthetic applications.15 In the form of porous electrodes, it has served as a membrane for controlled release of anionic drugs.16,17 Electropolymerization (EP) of its monomer can be achieved in an aqueous electrolyte solution.14 Electrodes of PPy combined with a common electrolyte can be reversibly oxidized and reduced according to
P+(A) + C+ + e- T P0(AC)
(4)
P+(A-) + C+ + e- T P0 + A- + C+
(5)
where P+ and P0 denote the oxidized and neutral states of PPy, respectively. P+(A-) represents an anion A- that pairs with PPy as a dopant ion. An added property of the PPy is that its volume increases and its chemical character is altered upon reduction of the polymer from the oxidized to the neutral state.18 The increase of thickness in thin PPy films can be as large as 40% upon the first reduction switch, whereas repeated switching typically provides a 20% reversible change of the volume (thickness) in expansion-contraction experiments.19 As the oxidation state is altered, not only do the bulk properties of PPy change, but its surface chemistry is also expected to be changed. This leads to active control of the surface tension. Several groups have demonstrated various kinds of electrochemical wettability switches, including polythiophenes, PPy, and polyaniline as the active surface switch material.20–23 For instance, a wettability switch was recently reported that exhibited reversible switching from superhydropobicity to superhydrophilicity (CA switching from 150 to ∼0°).22 Such large dynamic control of the water CA can only be achieved if active control of the surface tension is combined with control of surface structure. Here we report the results of electrochemical control of the wettability and topography switching from surfaces incorporating microstructures of PPy and insulating polymers, also including thin films of poly(3,4-ethylene-dioxythiophene) (PEDOT)-poly(styrene sulfonate) (PSS) or gold as the bottom electrode. The combined operation of the active microstructured polymer and the passive structural component leads to a major change in the CA for water, going from near hydrophobic (CA of 129°) to hydrophilic (CA ≈ 44°) conditions.
2. Experimental Section 2.1. Manufacturing of a Surface with Two Patterned Materials. PEDOT-PSS, ∼200nm thick, coated plastic foils (also named Orgacon, supplied by Agfa N. V.,9 or a gold (150 nm thick) coated (13) Jager, E. W. H.; Inganas, O.; Lundstrom, I. Science 2000, 288, 2335– 2338. (14) Jager, E. W. H.; Smela, E.; Inganas, O. Science 2000, 290, 1540–1545. (15) Immerstrand, C.; Holmgren-Peterson, K.; Magnusson, K. E.; Jager, E.; Krogh, M.; Skoglund, M.; Selbing, A.; Inganas, O. MRS Bull. 2002, 27, 461–464. (16) Kontturi, K.; Murtomaeki, L.; Pentti, P.; Sundholm, G. Synth. Meta 1998, 92, 179–85. (17) Wallace, G. G.; Adeloju, S. B.; Shaw, S. J. In Electroassembly of Smart Polymer Structures (Role of Polyelectrolytes); SPIE: Bellingham, WA, 1997. (18) Pei, Q. B.; Ingana¨s, O. J. Phys. Chem. 1992, 96, 10507–10514. (19) Smela, E.; Gadegaard, N. AdV. Mater. 1999, 11, 953–956. (20) Isaksson, P. K. J.; Tengstedt, M.; Robinson, N. D.; Berggren, M. AdV. Mater. 2004, 16, 316–320. (21) Robinson, L.; Isaksson, P. K. J;.; Robinson, N. D.; Berggren, M. Surf. Sci. 2006, 600, 148–152. (22) Xu, L. B.; Chen, W.; Mulchandani, A.; Yan, Y. S. Angew. Chem., Int. Ed. 2005, 44, 6009–6012. (23) Wang, G. X.; Deng, X. Y.; Tang, C. J.; Liu, L. S.; Xiao, L.; Xiang, L. H.; Quan, X. J.; Legrand, A. P.; Guidoin, R. Artif. Cells, Blood Substitutes, Immobilization Biotechnol. 2006, 34, 11–25.
Figure 1. Schematic illustration of patterning two materials on a surface using lithography and EP (upper). Chemical structures of PEDOT-PSS, PPyDBS, and SU-8 used for preparing surfaces (lower).
silicon wafer were used as electrode substrates (see Figure 1). An insulating polymer (SU-8 or Shipley181824) was first deposited and patterned by lithography, giving a partially exposed electrode surface. The conducting openings were used to grow PPy by EP, and the substrate served as the working electrode immersed in the aqueous electrolyte including pyrrole monomer (0.1 M pyrrole, 0.1 M dodecylbenzenesulfonate acid, sodium salt [NaDBS]). EP was carried out, using a General Purpose Electrochemical System 10 (EchoChemie, Netherlands) (GPES), at current densities from 0.2 to 0.4 mA/ cm2 and at voltages between 0.55-0.80 V versus the Ag+/AgCl reference electrode in the galvanostatic mode. A platinum plate was used as the counter electrode, and a BASI MF-2052 Ag+/AgCl was used as the reference electrode. In Figure 2, the photoresist was patterned on the PEDOT-PSS surface (Figure 2a), with circular openings in a square lattice. The wells were 20 µm in diameter and ∼1.8 µm deep. The current density during the EP of the PPy pillars was set to 0.3-0.5 mA/cm2, and the resulting voltage on the sample, e.g., the working electrode potential, was set to 0.5-0.6 V relative to the Ag/AgCl reference electrode. Investigations of the PPy-photoresist micropatterns implied that stirring of the solution during EP and also lowering the electrolyte concentration is helpful in suppressing inhomogeneous film properties and surface roughness. However, there is no influence on the growth rate when the concentration of the aqueous (NaDBS) electrolyte is changed between 0.01 and 0.1 M. In addition, the synthesis situation, such as material purity, may also influence surface roughness or nonuniformity after extended periods of EP (longer than 30 min). Reference samples of flat PPy films on the corresponding flat conducting surface were also synthesized under the same EP conditions. After completed EP to generate micropatterned surface switches, some of the samples were also polished with a milling tool to achieve completely planar switches. The polishing process was performed using a milling machine, Universal, Gol Matic (MO-23). 2.2. Evaluation of the Micropatterned Surface Switches. Atomic force microscopy (AFM, Nanoscope Digital 3100) was used to determine the film thickness and roughness. A scanning electron microscope (SEM, LEO 1553) was also used for determining the (24) Semiconductor Lab Website of the University of Oregon. http:// darkwing.uoregon.edu/∼hutchlab/semilab/shipleymat.htm (accessed Oct 2007).
5944 Langmuir, Vol. 24, No. 11, 2008
Wang et al.
Figure 2. Micropatterns combining photoresists and PPy. Left, AFM images of photoresist S1818 patterned with opened wells on the PEDOT-PSS surface (a), and electrochemical polymerization of PPy, grown in the wells for 15 min (b) and 30 min (c), respectively.
topography of the sample surfaces. A profilometer, DekTak-32, was used for profile measurement, and an optical microscope (SMZ1000) equipped with a camera (SSC-DC88P) was used to record optical images. Cyclic voltammograms of the grown PPy films and surface switches were measured in an aqueous electrolyte including NaDBS (0.1 M) with a Pt plate as the counter electrode and Ag+/ AgCl as the reference electrode using the GPES system. 2.3. Electrochemical Experiments. The GPES system was used to carry out the electrochemical switch experiments. Electrochemical reduction-oxidation of two-component samples were undertaken by applying a negative and a positive potential to the samples electrodes (-1 to 0 V) versus the Ag+/AgCl reference electrode in an aqueous electrolyte, NaDBS (0.1 M), with a Pt plate used as the counter electrode. Each electrochemical (or EC) reduction-oxidation switch was performed by applying a constant voltage for 2 min. Consequently, the topography and wettability changes were determined ex situ after a full EC reduction or oxidation switch and after rinsing with deionized water, followed by a quick drying step in nitrogen gas. The lifetime of the device is partly predicted by chemical or electrochemical degradation of the PPy material. It is important that surface switches are not exposed to overpotentials or to air and oxygen for extended time periods. However, for the targeted applications (seeding, growth, and release of cells) only one or a few switch cycles are necessary to achieve proper effects. For these limited number of switches, we do not expect any device malfunction due to chemical degradation of the PPy material.
2.4. Contact Angle Measurements. Ex situ CA measurements performed on sample surfaces were done using a goniometer (CAM 200, KSV Instruments, Ltd.), including deionized water as the probing liquid. The measurements were carried out at room temperature (T ∼ 21 °C), at ambient humidity, and CAs were determined using the image processing software of the instrument. For each sample and each switch state, at least seven identical measurements were performed to achieve proper statistics. 3. Wettability and Topography Switching: Results. 3.1. Planar PPy Surfaces. The wettability characteristics along homogeneous EP-grown PPy surfaces were studied. The reduced surface exhibits a relatively lower CA, θred ≈ 50 ( 5°, and in the oxidized state the surface exhibits a water CA of θox ≈ 80 ( 5°. In the oxidized state, the conjugated polymer includes dipoles along the backbone, and, intuitively, this should promote higher surface tension properties, i.e., a relatively lower water CA. However, in our case, we observe the reverse situation, the reduced state of PPy produces a surface of higher energy. In the oxidized state, the outermost surface (and also the bulk) locks the dodecylbenzenesulfonate acid (DBSA) dopant via the acid groups. In this state, along the PPy surface, the hydrophobic alkyl chains are facing the environment.20 So, for the oxidized state, this causes the dipoles of the bulk to be shielded by the alkyl chains. In the reduced state, however, as the dopants are decoupled from the PPy backbone, they can more easily rotate to expose the acid groups to the approaching water droplets. Consequently, the reduced state gives a relatively lower water CA. The surfaces were grown via electrochemical polymerization, which gave
Dynamic Control of Microstructured Polymer Films
Langmuir, Vol. 24, No. 11, 2008 5945
Figure 4. (a) Thickness of PPy films as a function of growth time at the growth conditions specified in the experimental section, and (b) the typical current-voltage characteristics of the microtextured PPy sample established at 50 mV/s scan speeds. Figure 3. SEM images of patterned PPy on PEDOT-coated substrates, after removal of the photoresist. The bar scales in panels a, b, and c are 5 µm, 30 µm, and 10 µm, respectively.
a surface roughness of about ∼30 nm. Our tentative explanation for the observed wettability switching effect, induced by changes of the oxidation state of EP-grown PPy electrodes, is the product of switching the anchoring character of the DBSA dopant together with the permanent surface roughness character at the outermost surface. 3.2. PPy Pillars on PEDOT-PSS and Gold. PPy pillars grown by electrochemical synthesis (Figure 2a-c) shows the resulting PPy-photoresist micropattern for 15 and 30 min polymerization times, respectively. As the PPy pillars grew above the top surface of the S1818 photoresist, lateral EP growth occurs (Figure 3) and is also evident from SEM images (Figure 3 a,b) of the samples after stripping of the photoresist. We performed water CA measurements on PPy pillar structures; after that, the photoresist was stripped for pillar heights (t) ranging from 2 to 3 µm (structure 3.2a). In the pristine, as-polymerized state, the water CA was 85° ( 5°. After reduction of the PPy pillars, the resulting CA was 70 ° ( 7°. Also, PPy pillar arrays of sizes from 3 to 5 µm in diameter and aspect ratios (height/ width) from 0.5 to 1 (structure 3.2b) were made (Figure 3c). The smaller sized features could only be manufactured using gold as the bottom electrode materials, due to better adhesion. With PEDOT-PSS we observed severe delamination of the microstructures. After stripping of the photoresist, the associated water CA was 93° ( 8° and 83° ( 5° for the pristine (oxidized) and reduced state, respectively. The growth of the PPy pillars exhibited a linear relationship between the PPy pillar height and EP time (Figure 4a). Cyclic voltammetry (CV) studies performed on the prepared samples, manufactured at different growth times, were identical, implying that similar PPy material characteristics were achieved for each synthesis. The typical CV curve of a sample (see Figure 4b) has a peak of oxidation and reduction potential at approximately -0.6 and -0.3 V, respectively, versus a Ag/AgCl reference electrode.
In comparison to previous reports, the reduction peak is shifted ∼0.1 V,15,25 and the oxidation peak remains unchanged. The water CA was not substantially affected by the change of the oxidation state of the PPy pillars. Our conclusion is that the microstructures simply did not include enough of the electrochemically active PPy material (area) along the surface to considerably affect the binding character of water. 3.3. PPy Mesh-SU-8 Pillars. To further investigate pillar-like structures, we now choose PPy as the mesh, with SU-8 as the pillars, to enhance the area fraction of the active material of the microstructured surfaces. For the flat PPy electrodes, the CA was varied, upon EC reduction-oxidation switching, typically between θox ≈ 80° (oxidized state) to θox ≈ 50°. This variation of ∼30° is sufficiently large to make surface energy control with microstructuring relevant, according to the Cassie-Baxter equation. Encouragingly, with a proper design of the microstructured surface, which now consists of SU-8 square pillars (100 µm wide and 16.8 µm high) surrounded by an EP-grown PPy mesh (12.7 µm thick), the top surface of one of the components along the structured surface, PPy, was turned up and down upon switching of the oxidation state (the 3.3a structure). The PPy phase that expands results in modulation of the cavity depth within the wells. In the first switch cycle, the PPy thickness was varied from 12.7 to 13.8 µm upon reduction to the neutral state. The dynamics of the PPy-mesh-SU-8 pillar structures, upon sequential reduction-oxidation switching, were recorded ex situ by an SEM and a Dektak profilometer (see Figure 5a-d). We clearly observed how application of the cathodic potential leads to the raising of the PPy mesh. The expansion ratio of the initial reduction-oxidation cycle was 9%. This is somewhat lower than the volume/height expansion ratios observed by others19 performed on homogeneous films, which typically reaches above 20%. The increased thickness, reached after the first reduction switch, could not be reachieved in the second oxidation-reduction switch cycle. The shape of water droplets on these micropatterned surfaces at different oxidation states are given as insets in Figure 5a-c. In the (25) Jager, E. W. H.; Immerstred, C.; Peterson, H. K.; Magnusson, K.-E.; Lundstro¨m, I.; Ingana¨s, O. Biomed. MicrodeVices 2002, 4, 177.
5946 Langmuir, Vol. 24, No. 11, 2008
Wang et al.
Figure 5. SEM images of (a) SU-8 square pillars on the conducting film surface, (b) Pristine PPy with EC surrounding SU-8 pillars forming a network, (c) the as-prepared surface of panel b after EC reduction. (d) The profile of the sample a (black solid line); the as-prepared sample of b (red dashed line), where PPy(DBS) was in the oxidized state; and sample c, i.e., the sample of b after EC reduction of the first cycle, (green open circles) and the sample after EC reoxidation (blue open trangles). The scale bars in the figure are 200 µm. The insets in the top right corner of the pictures are (a) 148°, (b) 129°, and (c) 44° CAs of the droplets on the as-prepared surfaces. (e) The CA on the surfaces at different EC states. The lines in the boxes from bottom to top present the 25th, 50th, and 75th percentile values, and the error bars represent the 5th and 95th percentile values. The squares in the boxes are mean values, and the symbols below and above the boxes represent the highest and the lowest values.
Figure 6. SEM images of PPy mesh grew around and above the top plane of the SU-8 pillars in the pristine state. There was a (a) 4 µm and (b) 22 µm in height difference between the SU-8 and the PPy top level. The scale bars in the pictures are 200 µm. Table 1. Summary of the Water Contact Angles for PPy Switched to the Oxidized and Reduced Statea θoxidized
surface switch structure 3.1 planar EP-grown PPy 3.2a PPy pillars (w ) 20/100 µm) on PEDOT-PSS, tPPy ) 2-3 µm 3.2b PPy pillars on gold (w ) 5 µm), tPPy ) 2-3 µm 3.3a PPy mesh/SU-8 pillars (tSU-8 ) 16.8 µm) 3.3b PPy mesh/SU-8 pillars (tSU-8 ) 21 µm) 3.4 PPy-mesh/SU-8 wells a
80° ( 7° 85° ( 5°
θreduced 50°( 5° 70°( 7°
93° ( 8° 129 ( 10° (tPPy ) 12.7 µm) 145 ( 5° (tPPy ) 12.7 µm) 88-97°
83°( 5° 44°( 5°(tPPy)13.8µm) 140°( 5°(tPPy)13.8µm) 55-67°
∆θ 30° 15° 10° 85° (∆t ) 1.1 µm) e10° (∆t ) 1.1 µm) ∼30°
t denotes the thickness or height of the specific material features above the supporting electrode surface. w denotes the width of the particular structure.
first reduction, the water CA was found to decrease from ∼129° to ∼44°. In the second switch cycle, the corresponding values are from ∼74° down to ∼26°, respectively. In comparison, there are much
smaller changes in CA if the PPy mesh surface is more than 7 µm below the top surface of the SU-8 pillars (3.3b structure: 21 µm high SU-8 pillars, 100 µm apart and a 12.7 µm thick surrounding PPy
Dynamic Control of Microstructured Polymer Films
Langmuir, Vol. 24, No. 11, 2008 5947
Figure 7. Schematic illustration of a water droplet applied to a PPy mesh/SU-8 pillar surface switch for PPy in its pristine oxidized state and switched to the reduced state. The combinatory effect of reducing the effective height of the SU-8 pillars and changing the surface tension of the PPy surface explains the large difference in water CA.
The maximum range of wettability switching of the micropatterned surfaces, including PPy doped with DBSA, exhibited a controlled variation of the water CA from 129° to 44°. Dynamic control of the water CA ranging from 150° to almost 0° was reported for PPy including perfluorooctanesulfonate (PFOS) as the dopant ion.22 However, insertion or deinsertion of Na+ dopant in or out of PPy-DBS films in a water-based NaDBS electrolyte solution, in our case, took only about 2 min. This was 10 times faster than that of moving PFOS dopant in or out of PPy-PFOS electrode into the TEAPFOS acetonitrile solution. Our PPy electrodes have been designed to operate in aqueous solutions, thus making them suitable for interfacing with biology. 3.4. PPy-Mesh/SU-8 Wells. Micropatterned structures with thicker PPy meshes were also grown. Here, the EP-grown PPy forms the side walls and top layer of the structures, with SU-8 now as the bottom well floor (see Figure 6a,b). The height of the PPy mesh, above the top plane of the SU-8 pillars, varied from 4 to 16 to 22 µm. Although the topographies varied, i.e., the well depth and also lateral geometry, the water CAs of these surfaces of the oxidized and reduced state were rather close (88-97° and 55-67°, respectively). The difference in water CA, upon switching, was only ∼30°.
4. Discussion
Figure 8. Water droplets of (a) 5 µL on the surface where PPy is in the pristine/oxidized state, and (b) 5 µL and (c) 5 µL more added to the previous 5 µL droplet on the surface in panel b, where PPy is in the reduced state. The white lines in the photos are patterned SU-8, 50 µm in width and 25 µm high; the black line is electropolymerized PPy, 200 µm in width and ∼21 µm high.
mesh), suggesting that the fakir situation11 primarily governs the water adhesion characteristics at all states (see Table 1). In this case, the difference in water CA was found to be less than 10° (θox ≈ 146° and θred ≈ 140°). For the 3.3a structure, after each switch cycle, the trend is that the water CA, of the oxidized as well as the reduced state of the PPy, shifts toward lower values (see Figure 6e). Once switched, the initial water CA of around 129° cannot be reached again. We believe this hysteresis effect in surface energy is, in part, due to the fact that DBSA dopant, along the outermost PPy surface, escapes into the electrolyte solutions primarily when PPy is in the reduced state. So, for every reduction switch, fewer and fewer DBSA molecules along the surface take part in dictating the surface energy. In addition, the initial thin state of the PPy mesh is never reached again after an oxidation switch, which should also contribute to lowering the water CA. In addition, after further EC switch cycles, water might also enter into the film and may introduce serious delamination of the surface switch electrodes. Such delamination was observed for a few surface switches after many repeated switch cycles. To avoid delamination, several strategies are considered, based on patterned film thickness and pattern size and/or materials selection as well.
For the microstructured surface switches here reported, we observe the largest change of the water CA, as the PPy phase is electrochemically switched, when PPy defines the mesh that surrounds distributed SU-8 pillars (see Table 1). The relative dimensions of the microstructures also play an important role. For the 4.1 µm tall SU-8 pillars (100 µm wide and 200 µm apart), the difference in the water CA reaches (∆θ )) 85°. For taller SU-8 pillars (7 µm tall SU-8 pillars above the PPy mesh), a drastically smaller effect on the water CA is observed (∆θ e 10°) as PPy is switched. Also, for planar PPy surfaces and for structures in which PPy defines distributed pillars, the total change in water CA is typically less than (∆θ e) 30°, and is sometimes even as low as (∆θ )) 10°. We interpret these data as follows: For the 4.1 µm tall SU-8 pillars, surrounded by the PPy mesh, the switching effect on the water hosting character depends on two effects. First, as the PPy mesh is reduced, it expands, and the effective SU-8 pillar height is reduced to 3 µm (see Figure 7). Second, the energy of the PPy surface is increased, which promotes stronger binding to water. When water is added to the “3.3a” structures, as PPy is in its oxidized state, it adds onto the surface switch according to the fakir situation, and its shape and binding properties are best described by the Cassie-Baxter theory. As PPy is reduced, the structure becomes shallower, and the PPy surface turns up its surface tension, the water droplets now fill the entire true surface area. In this state, the wetting character is best described by the Wenzel theory, including the wetting characteristics of the PPy and SU-8 materials and the combined structure into the account. So, this combinatory effect of switching
5948 Langmuir, Vol. 24, No. 11, 2008
Wang et al.
Figure 9. Left: the milled PPy mesh/SU-8 square surfaces, PPy in its pristine (top) and reduced state (bottom). Right: the surface profile before (solid line) and after (dashed line) PPy was reduced.
the shallowness and the overall surface tension of the PPy mesh provides the large swing of water CA seen here. As the electrochemical state is switched, of the conjugated polymer, not only its surface chemistry is controlled but also the surface roughness at the nanometre scale could change. This effect could then add to the control of the surface tension and also contribute to net switching in wettablity. 5. Anisotropic Dynamic Surface Energy on Line Shape Structure. We have also used an alternative geometry to obtain anisotropic dynamic surface energy, where lines of PPy, 200 µm in width and 21 µm high, were grown as defined by the lines of SU-8 photoresist, 50 µm in width and 25 µm high (see Figure 8). A dynamically anisotropic spreading of water is observed. Five microliter water droplets were delivered on the as-prepared/ oxidized surface, and they were asymmetrically distributed as a result of the anisotropic wetting property of the line-shape structure (Figure 8a) and were further spread only along the line’s direction (Figure 8b) after PPy was electrochemically reduced. Furthermore, adding an additional 5 µL water droplet to the former droplet displayed in Figure 8b resulted in water droplet spreading, again only along the lines (Figure 8c), revealing an anisotropic filling of droplets mostly like the liquid spreading in a tube on the surface. 6. Switching the Surface Topography from Planar to Structured. The PPy-mesh/SU-8 pillar structure (structure 3.3a or 3.3b) can be polished using a mechanical milling tool. The resulting structure then becomes a planar intarsia PPy film, ∼20 µm thick, including SU-8 squares (100 µm wide and 200 µm apart) (see Figure 9). The planar structure is also maintained as the surface switches are immersed into a NaDBS (0.1 M) electrolyte solution for 2 min, i.e., very little swelling due to electrolyte exposure is observed. Then, if the PPy mesh is reduced, its thickness increases. The surface then turns into a PPy surface
with approximately 3 µm deep holes, with SU-8 as the bottom floor (see Figure 9). In many studies on cell and tissue growth, it would be of great interest to dynamically convert the planar cell hosting surface into a textured one. Also, this kind of surface switches could be of interest for mechanical stimulation of cells.
7. Conclusions and Summary Micropatterned surface switches have been reported, including the combination of the electrochemically active PPy, other conductors, and insulating polymers. From a series of experiments, performed on geometrically different surface switches, we draw the conclusion that large changes of the water CA, upon switching the PPy phase, is a product of controlling the topography of the surface structure in combination with switching the surface tension of the PPy. In addition, we demonstrate electronic control of anisotropic surface wetting and also switching the surface topography going from being completely planar to becoming textured. Our findings add to the knowledge of understanding the origin of wettability switching along electrochemically active materials. Also, the surface switches described define a powerful toolbox of easily manufactured surface switches that may serve the science fields of cell biology and potentially medicine. Acknowledgment. The authors gratefully acknowledge the financial support from The Strategic Research Center for Organic BioElectronics (OBOE) and Bio-X funded by the Swedish Foundation for Strategic Research (SSF). In addition, the authors acknowledge financial support from the Linko¨ping University, Knut och Alice Wallenbergs Stiftelse, and The Royal Swedish Academy of Sciences. LA704006Z
