Electrochemical wettability switches gate aqueous liquids in microfluidic systems

Abstract

We demonstrate a simple low-voltage technique for gating the flow of aqueous liquids in microfluidic systems employing the electrochemically-controlled surface energy of the conjugated polymer poly(3-hexylthiophene).

---

As fluidic systems are miniaturized, as has happened with electronic and mechanical technology, control of the interfacial energy between fluids and solids becomes increasingly important. Inertia and buoyancy become insignificant at the micrometer lengthscale, leaving surface tension as the dominant force. In “system-on-a-chip” analytical tools, pico-liter volumes of analytes are guided through narrow micro-channels to mix with reagents before finally reaching an analysis reservoir. To guide the liquids, valve-like micro-mechanical arrangements and micro-pumps have been explored.¹ Cascades of hydrophobic “gates” have been utilized to halt liquids flowing under the influence of specific pressure. As the pressure increases, the liquid moves to the next gate.² Such passive systems are useful, but face severe limitations in that no decisions can be made regarding where to route the fluid. Controlling the interfacial energy of surfaces in microchannels is a relatively straightforward concept for manipulating fluids, but electronic control has yet to be developed beyond electrowetting, which requires a large applied potential on the analyte itself.³

Thus, to control liquids in very small channels, interfacial energy must be controlled. The wetting properties of a solid surface are predicted by its polar, structural and charging properties. Electronic control and addressing of wettability, realized in a surface switch, enables digital electronics to interface directly with microfluidics, i.e. digital circuits can select and change the wettability locally at the junctions in microfluidic channels.

We have done this with a special class of materials called conjugated polymers, which often contain molecular side groups attached to the monomer to make the polymer soluble (for simplified processing), but also affect the polymer's electronic and optical properties. These electronic properties, and therefore also the chemical properties, can also be controlled by switching the oxidation state of the conjugated polymer. Various types of solid-state electrochemical devices have been explored and reported in the past, e.g. organic electrochemical transistors,⁴ electrochromic windows⁵ and paper displays, etc. Electrochemical switching often exhibits a strong degree of bi-stability, meaning that the device “remembers” which state it is in for some time, making electrochemical devices suitable for complex 2-dimensional matrix-addressed memory systems. In addition, devices can be manufactured in a few steps on flexible substrates, a strong advantage conjugated polymers have over metal conductors and crystalline semi-conductors in many applications.

To illustrate the change in the surface energy in such a system, images of water on poly(3-hexylthiophene) (P3HT) films are shown in Fig. 1, with contact angles of 102° (reduced) and 89° (oxidized).⁶ It is this difference in contact angle that we employ to gate water transport.

Fig. 1 Water droplets on oxidized (left) and neutral (right) P3HT films.

We refer the reader to the literature^6,7 for details regarding the physics behind the wettability switch. In summary, the origin of electrochemical control of water contact angle along P3HT surfaces is likely due to the change in dipole strength of the thiophene backbone that occurs with electrochemical doping. We find that increasing the alkyl side-chain length decreases the difference in water contact angle between the neutral and oxidized states and increases the water contact angle in general.⁶ As the polymer is oxidized, anions migrate from the electrolyte into the P3HT film, stabilizing polaronic states via the oxidation

P3HT + X⁻ → P3HT⁺X⁻ + e⁻ (1)

where X⁻ denotes the anion from the electrolyte and e⁻ represents an electron. Since the oxidized state is non-volatile the water contact angle can be measured without a continuously-applied voltage. After oxidation, the doped thiophene contains positive charges along its backbone balanced locally by anions, resulting in dipoles, favored by the water. It should be noted that the surface roughness of the films plays a role in wettability, and may be partly responsible for the difference in wettability observed.

The P3HT wettability switch, shown in Fig. 2, was constructed on a glass substrate (silicon or flexible plastic can also be used) with a cast Ag/AgCl paste (DuPont 5000 Silver Conductor) electrode. Then, a polymer electrolyte layer, approximately 400 µm thick, was cast on top of the silver, followed by a 7 min anneal at 60 °C. P3HT (3 mg ml⁻¹ in CHCl₃) was spin-coated at 1200 rpm for 20 s on the electrolyte and heated at 60 °C for 1 min to evaporate the chloroform solvent, resulting in a ∼170 Å thick film (measured on silicon). The P3HT used in this work (from Sigma Aldrich) is regioregular with greater than 98.5% head-to-tail regiospecific conformation. The electrolyte was made of 25.6 wt% poly(sodium 4-styrenesulfonate) (MW 70 000, Aldrich), 8 wt% D-sorbitol (97% Lancaster), 8 wt% glycerol (87% Merck), 20 wt% magnesium sulfate, MgSO₄ (Merck) and 38.4 wt% deionized water. Electronically isolated P3HT electrode films were then patterned simply by carefully scratching/cutting the P3HT thin film using a scalpel, without cutting through the electrolyte layer. Each P3HT surface, the common electrolyte layer and the Ag counter electrode form an individual wettability switch. PDMS microfluidic channels were generated via a standard templating process. The PDMS layer was 2 mm thick, and the patterned channels had a width of 1000 microns and a height of 60 microns.

Fig. 2 Schematic of the wettability switch device.

The patterned wettability switches were integrated with the microfluidic system simply by pressing the active P3HT electrodes against the microfluidic channels so that individual switches coincide with each branch of the y-junction. Immediately before use, the polymer film floor of the channel is electrochemically switched by applying a potential between the Ag/AgCl electrode under the electrolyte and the P3HT electrode, contacted by a metal probe placed directly on the polymer surface outside the PDMS layer nearest the channel route to be oxidized. When a potential of a few volts is applied (in this case 15 V, but functional even at 5 V) a “front” of oxidation radiates in a circular fashion from the point at which the electrode contacts the film.⁸ The front is clearly visible as the polymer changes from the reddish neutral state to the nearly colorless oxidized (doped) state. The oxidation process takes only a few seconds, and can be directed by the placement of the external probe.

By using a wettability switch as the floor in a PDMS microfluidic capillary channel, the device can be programmed to direct water along a desired path in a junction system. Fig. 3a shows the introduction of water (dyed green) into a simple channel system with two bifurcations. In this experiment, the P3HT film was patterned and biased to oxidize the path on the right-hand side of the image, making it the most favorable route for the water. Fig. 3b shows the water exiting the channel through the previously-chosen outlet. Such experiments have been run dozens of times with nearly 100% success (manufacturing defects have caused a few failures). As seen in the left-hand branch of the channel in Fig. 3b, the water eventually crosses the frontier between oxidized and reduced P3HT. This “fuzzy” interface is not optimized in this design, and is only capable of holding a pressure of about 5 mm of water.

Fig. 3 (a) Photograph of the electrochemically switched surface with an empty microchannel on top. (b) Photograph of the same surface after water (dyed green) has been guided through the microchannel by the switched surface. The P3HT under the right-most path (in both images) has been electrochemically doped (oxidized). The rest of the P3HT is undoped (neutral) and is pink in colour.

Electrochemically doping a polymer film to guide water through a microchannel does not require the large potentials used in electrowetting.³ Another advantage is that the polymer “remembers” its state (at least for several minutes), since electrochemical switching in P3HT is bi-stable, allowing the device to be “programmed” by applying a potential and then used later. The potential does not need to be applied while the fluid is in the channel. In the next generation of this wettability-gated microfluidic system, the external probe will be replaced by electrochemical logic circuitry⁹ acting on both external and electrochemical sensor¹⁰ information and/or a matrix addressing scheme similar to that we have demonstrated for electrochromic displays. Thus, the path taken by the aqueous sample will be directed from an external electronic system, and/or “internally” using sensors within the microfluidic channels based on the same (or similar) material used to make the valve. We also intend to investigate other switching materials with the goal of increasing the change in surface tension in the two states. This will allow us to gate liquids inside branched channels at higher pressures and/or in more complex channel systems (e.g. a large number of y-junctions).

In summary, we have demonstrated that it is possible to guide water in microfluidic systems with a polymer-based electrochemical wettability switch, which will allow programmable electronic control of fluid transport in microfluidic devices that can be produced with low-cost printing techniques. We would like to thank The Swedish Research Council, The Royal Swedish Academy of Sciences and The Swedish Foundation for Strategic Research (COE@COIN) for funding this research and the Acreo Institute for helping us to manufacture microfluidic channels.

Notes and references

1. H. Suzuki and R. Yoneyama, Sens. Actuators, B, 2003, 96, 38–45.
2. P. Andersson, H. Derand, M. Spesny and F. Foret, 50th ASMS Conf. on Mass Spec. and Allied Topics, Orlando, FL, USA, 2002, p. 147.
3. R. Digilov, Langmuir, 2000, 16, 6719–6723.
4. J. Huang and M. S. Wrighton, Anal. Chem., 1993, 65, 2740–2746.
5. Q. Pei, G. Zuccarello, M. Ahlskog and O. Inganäs, Polymer, 1994, 35, 1347–1351.
6. L. Robinson, J. Isaksson, N. D. Robinson and M. Berggren, Surf. Sci., 2006, 600, L148–L152.
7. H. C. Ko, M. Kang, B. Moon and H. Lee, Adv. Mater., 2004, 16, 1712–1716; J. Isaksson, C. Tengstedt, M. Fahlman, N. Robinson and M. Berggren, Adv. Mater., 2004, 16, 316–320.
8. T. Johansson, N.-K. Persson and O. Inganas, J. Electrochem. Soc., 2004, 151, 119–124.
9. D. Nilsson, N. D. Robinson, M. Berggren and R. Forchheimer, Adv. Mater., 2005, 17, 353–358.
10. J. T. Mabeck and G. G. Malliaras, Anal. Bioanal. Chem., 2006, 384, 343–353; J. T. Mabeck, J. A. Defranco, D. A. Bernards, G. G. Malliaras, S. Hocd and C. J. Chase, Appl. Phys. Lett., 2005, 87, 013503; Z. T. Zhu, J. T. Mabeck, C. C. Zhu, N. C. Cady, C. A. Batt and G. G. Malliaras, Chem. Commun., 2004, 1556–1557.

---