Bendable fuel cells: on-chip fuel cell on a flexible polymer substrate

Abstract

A novel concept of fuel cells, i.e., bendable fuel cells, is proposed and demonstrated by the fabrication of on-chip fuel cells on cycloolefin polymer films. This approach overcomes the brittleness and cost problems, i.e., a troubling aspect of miniaturized devices, of on-chip fuel cells.

---

Broader context

Ultra small fuel cells, which are applicable as power sources for on-chip devices and micro-electronics, have been developed in our laboratory. Recently, we proposed on-chip fuel cells of an air-breathing, membraneless and monolithic design, and have fabricated them on silicon thin films so far. However, alternatives to the silicon films are strongly needed because of their brittleness and expensiveness. In view of this, in this communication, we fabricated the cells on cheap flexible materials, cycloolefin polymer films. Consequently, we succeeded in obtaining bendable on-chip fuel cells without any power decrease compared with those on silicon films. This bendable feature is of significant importance for on-chip fuel cells, because it overcomes the most troubling aspect of miniaturized devices, i.e., brittleness. This achievement opens up the possibility of use of on-chip fuel cells as power sources for future electronics, e.g., health care chips. As future works, it is naturally suggested that we still have to improve their power, and developments of new catalysts for these on-chip fuel cells will be a key issue to make on-chip fuel cells as practical devices.

Recent progress in micro-electromechanical systems (MEMS) and on-chip devices, e.g., micro-sensors,^1,2 inevitably requires the developments of on-chip power sources, including micro-solar cells,³ micro-batteries,^4,5 and micro-fuel cells.^6–8 Among them, micro-batteries seem the most developed one,^9–11 but other micro-power sources also need to be developed because of the restricted application of micro-batteries due to their necessity for recharging. On the other hand, micro-fuel cells are expected to fill a different niche because of their possible operation as long as fuel is supplied. To turn micro-fuel cells into practical devices, their systems including microelectrodes and microchannels need to be designed based on micro-fluidics and physicochemistry. Since miniaturization alters physical properties, especially associated with fuel supply and water removal, as derived from the scaling law,^12,13 their optimum system is dependent on the application of choice.

In view of this, we have developed on-chip fuel cells of a novel design, an air-breathing, membraneless and monolithic design, where all the components necessary for power generation were integrated on a single substrate (Fig. 1).^7,8 The cell operates on a liquid fuel solution supplied through the microchannel by capillary force and oxygen in the air without pumps. So far, we demonstrated their operation on a methanol, ethanol or 2-propanol solution containing an acidic or neutral-pH salt.⁸ As for their applications, we have indeed focused on the application as power sources for miniature devices, but we also aim to use them as power sources for general electronics, e.g., laptops and cell phones, by integrating a large number on a single chip.⁸ We believe that this approach will lead fuel cell developments out of the constraint of using membrane electrode assemblies (MEAs), which have been a main component of fuel cells and constrained fuel-cell designs and applications. The fascinating on-chip cells, however, still have difficulties in spreading into a broad range of applications due to their brittleness and process cost originating from the use of silicon substrate.

Fig. 1 Schematic illustration explaining the design of our on-chip fuel cell. Protons are generated by fuel oxidization on the anode, deposited on the bottom of a microchannel, transferred to the cathode through a fuel solution containing ion conductors, and then reacted with oxygen to become water.

Silicon films have been used as substrate materials for microdevices because of their ease of microfabrication; however, applications of such devices are limited due to their brittleness and cost, and their long-time expensive fabrication processes. To overcome these problems, we use a flexible polymer film as a substrate and replicate the on-chip fuel cells onto it by hot embossing, which is a rapid and inexpensive process.^14,15 Compared to general fuel cells using graphite plates both as current collectors and channels for feeding reactants, the use of polymer substrates is also advantageous, because the fabrication cost of a graphite plate is also expensive.

Use of flexible components, especially current collectors, is important for miniaturized fuel cells in view of facile assemblies and flexibility in design. For example, Wheldon et al.¹⁶ and Hahn et al.¹⁷ developed flexible fuel cells by preparing flexible membrane electrode assemblies, which are the energy-converting components, by choosing flexible current collectors. Moreover, a Swedish company, myFC AB, developed excellent flexible fuel cells, FuelCellSticker™, as power sources for portable electronics. These approaches seem quite sophisticated and innovative in view of fuel cells for portable electronics. On the other hand, our polymer-based fuel cell reported here has potentiality as power sources for other organic devices, including microfluidic devices, organic light emitting display and organic batteries,^18,19 because of the compatibility based on its size and monolithic design with nothing on the underside.

Thus, we propose here bendable on-chip fuel cells which were fabricated on flexible COP films. Such fuel cells fabricated on a flexible polymeric substrate were reported for the first time by Ito et al., who fabricated tiny MEA structures in holes of 0.5 mm in diameter drilled on a polysulfone substrate.²⁰ Comparing their approach to ours, the use of polymeric substrate is the same, but we aim to develop a new class of fuel cell, “on-chip fuel cell”, which was precisely designed to be compatible with other micro-devices.^7,21 Thus, our bendable fuel cells can be applied in a variety of fields, including lab-on-a-chip, micro-sensors and portable electronics. Furthermore, since our fuel cell has nothing on the underside, we are aiming to use the cell as a power source that can be pasted everywhere (like Scotch tape).²²

As a polymer of choice, cycloolefin polymer (COP) is a promising candidate in terms of its high chemical stability, low water absorbency, and outstanding heat resistance.^14,23,24 This polymer is widely available and has been used both as a optical film for the reason of its high transparency and as a substrate for microfluidic devices for the reason of its molding processability. To use this polymer as a substrate for electrochemical devices, since adhesiveness between the polymer surface and electrode materials is of primary importance, surface modification is needed. The surface of COP has been modified by using typical surface modification treatments, including acid treatment,²⁵ UV-ozone treatment,²⁶ and oxygen plasma treatment,²³ to introduce polar functional groups, e.g., C–OH and C=O, to the surface. In this work, we chose oxygen plasma treatment, which is known as a green process, to improve the adhesiveness of gold current collectors.

The fabrication procedures are briefly illustrated in Scheme 1 and detailed as follows. First, the microchannel was replicated onto a COP film (ZeonorFilm^® ZF16-188 from Zeon Corp., glass transition temperature 163 °C, thickness 188 µm) by hot embossing using a Ni mold prepared from a silicon-based cell by electroplating.²⁷ The Ni mold was pressed at ∼1 MPa onto the COP film for 10 min with heating the upper plate of a hot-press machine at 185 °C and the lower plate at 160 °C (details of the machine configuration is available in our previous report).²⁷ Then, the temperature was gradually decreased down to <100 °C before releasing the pressure. To deposit current collectors, the film with microchannels was spray-coated (EVG101 from EV Group) with a photoresist (AZ^® P4620 from AZ Electronic Materials) diluted ten-times by weight with acetone and then was patterned with UV lithography. The exposed COP surface was modified with an oxygen plasma treatment (PR500 from Yamato Scientific Co., Ltd.) at 300 W, 40 Pa for 20 min, a 200-µm Au layer with a 30-nm Ti adhesion layer was deposited by electron-beam evaporation. This strong plasma treatment is needed to increase adhesiveness of the current collectors to be strong enough for electrodepositing rough catalyst layers, which cause a large stress. On the Au current collectors, (i) Pd–Co alloy was electrodeposited as cathode catalyst at −200 mA cm⁻² for 60 s in a solution containing an ammonia complex of Pd(II) and a malonic acid complex of Co(II), and (ii) Pt–Ru alloy was electrodeposited as anode catalyst from chloride electrolytes by applying pulse current, as reported previously.^7,8,28

Scheme 1 Fabrication procedure of a flexible on-chip fuel cell onto a polymer substrate.

The cell was evaluated in terms of morphology, catalyst condition, and cell performance. First, the cell was observed with a real colour confocal microscope (H300 from Lasertech). Second, electrochemical responses from the catalysts were evaluated by cyclic voltammetry in nitrogen-saturated 0.5 M H₂SO₄ at room temperature. Third, the cell performance was evaluated at room temperature by dropping a 2 M methanol solution containing 0.5 M H₂SO₄ onto the end of a microchannel to be supplied by capillary force. Current–voltage (I–V) curves were recorded on an electrochemical instrument (HZ-3000, Hokuto Denko). During the I–V measurements, potentials of each individual electrode were measured using a silver/silver chloride (Ag/AgCl) reference electrode set in the fuel solution at the end of the channel.

The photograph shown in Fig. 2a shows that the on-chip fuel cell fabricated on a COP film closely resembles that on a silicon substrate,⁷ except for the transparency of the substrate. The micrographic 3D image (Fig. 2b) shows that the channel dimensions (width: 200 µm, depth: ∼100 µm) were very consistent with those of a silicon cell, as discussed in detail in our previous report.²⁷ Though we were concerned about applicability of the lithography process to the COP films, the patterning of the Au current collectors were successful. Interestingly, the COP film was thin enough for bending of the cell as shown in Fig. 2c, and the Au layer did not peel off from the film. This feature will fit a new niche of power sources, as described above.

Fig. 2 Images of the on-chip fuel cell fabricated on a polymer film. (a) Photograph of two cells integrated on a single chip. (b) Real colour 3D images of the microchannel fabricated on a COP film before (top) and after (bottom) the Au deposition. These images were obtained using a confocal microscope. (c) Photographs showing a bending test of the cell.

The electrodeposition of the catalysts did not peel off the Au current collectors deposited with a Ti adhesion layer on the surface-modified COP film. When using a pristine COP film, though the adhesiveness between the COP surface and a Au layer was reported to be relatively strong,²⁹ the current collectors had been peeled off after the electrodeposition. By using both the Ti adhesion layer and the oxygen plasma treatment, the adhesiveness became strong enough for the catalyst electrodeposition. Since oxygen plasma treatment was reported to introduce polar functional groups to COP surface within 1 min and to increase the surface roughness with process time,²³ the main origin of the strong adhesiveness of our cells can be rationalized in terms of the increase in surface roughness.

The electrochemical responses from the electrocatalysts deposited on a COP cell were confirmed to be almost the same as those on a silicon cell (Fig. 3), this indicating that no negative effects of the use of COP films on the electrocatalysts occurred. Regarding the anode, the CVs in Fig. 3a are typical for Pt–Ru catalysts, judging from the peak-less hydrogen region (−0.2 to 0.2 V vs.Ag/AgCl) and the relatively large double-layer currents (0.2 to 0.6 V).³⁰ Likewise, regarding cathode, the CVs in Fig. 3b are typical for Pd–Co catalysts, judging from the peaks in the hydrogen region (−0.2 to 0.1 V) and the potential of oxide reduction peak at ∼0.51 V, which is lower than that of pure Pd (ca. 0.56 V).²⁸ These results indicate that differences of the microchannels did not influence the electrodeposition of the catalysts, even though the Pt-Ru alloy was deposited on the bottom of the channel having a strong capillary force.

Fig. 3 Cyclic voltammograms for comparison of the electrocatalysts deposited on the on-chip fuel cells of a different substrate (green: polymer film; black: silicon substrate). (a) Anode Pt–Ru catalyst. (b) Cathode Pd–Co catalyst. The voltammograms were obtained by scanning the electrodes in a nitrogen-saturated 0.5 M H₂SO₄ at room temperature, 50 mV s⁻¹.

Finally, the cell performance of the COP cell was evaluated and compared with that of a silicon cell. The performance was identical to that on a silicon substrate (Fig. 4), this proving that the cell was successfully fabricated on a COP film. In detail, the I–V curves and the current–power (I–P) curves were identical, respectively (Fig. 4a). In addition, the potentials of each individual electrode are also close (Fig. 4b). Though detailed discussion on the durability is needed, it is concluded that COP films are a promising candidate for a substrate of on-chip fuel cells.

Fig. 4 Cell performance of the on-chip fuel cells of a different substrate (green: polymer film; black: silicon film). (a) Current–voltage curves and current–power curves. (b) Current–potential curves of the anode and the cathode. The footprint area of a single device was 0.4 mm wide × 6 mm long (0.65 mm wide × 6 mm long for a cathode, 0.12 mm wide × 6 mm long for the anode, and some margins for avoiding short-circuiting).

As for the maximum power of the cell, 2.0 µW, this value is relatively high for on-chip fuel cells, as we reported previously.⁷ But, compared with general direct methanol fuel cells, the power density, 83 µW cm⁻², is found to be low.³¹ This low power density is due to the small surface areas of the electrodeposited catalysts (their roughness factors are less than 100, but those of general ones are over 1000). Thus, deposition of electrocatalysts, i.e., Pd–Co alloy and Pt–Ru alloy, with large surface areas onto microelectrodes of on-chip fuel cells is our next challenging issue.

In summary, the on-chip fuel cell, which had been fabricated on a silicon film, was successfully fabricated on a flexible polymer film. The performance of the bendable cell was identical to that of a brittle silicon cell. This achievement is of critical importance for the practical applications of the on-chip fuel cells, because it solved the problems associated with the substrate, i.e., brittleness and cost. As a future work, we are aiming for use of carbon nanotubes as current collectors for this bendable fuel cells because of its suitableness for flexible organic devices.

Acknowledgements

We thank ZEON Corp. for providing the cycloolefin polymer films. This work was partly supported by the Grant-in-Aid for Specially Promoted Research “Establishment of Electrochemical Device Engineering” and by the Global COE program “Practical Chemical Wisdom” from MEXT, Japan.

Notes and references

1. D. Niwa, K. Omichi, N. Motohashi, T. Homma and T. Osaka, Sens. Actuators, B, 2005, 108, 721–726.
2. R. Popovtzer, T. Neufeld, N. Biran, E. Z. Ron, J. Rishpon and Y. Shacham-Diamand, Nano Lett., 2005, 5, 1023–1027.
3. J. B. Lee, Z. Z. Chen, M. G. Allen, A. Rohatgi and R. Arya, J. Microelectromech. Syst., 1995, 4, 102–108.
4. H. Nakano, K. Dokko, J. I. Sugaya, T. Yasukawa, T. Matsue and K. Kanamura, Electrochem. Commun., 2007, 9, 2013–2017.
5. D. Golodnitsky, V. Yufit, M. Nathan, I. Shechtman, T. Ripenbein, E. Strauss, S. Menkin and E. Peled, J. Power Sources, 2006, 153, 281–287.
6. S. C. Kelley, G. A. Deluga and W. H. Smyrl, Electrochem. Solid-State Lett., 2000, 3, 407–409.
7. S. Tominaka, S. Ohta, H. Obata, T. Momma and T. Osaka, J. Am. Chem. Soc., 2008, 130, 10456–10457.
8. S. Tominaka, H. Nishizeko, S. Ohta and T. Osaka, Energy Environ. Sci., 2009, 2, 849–852.
9. N. J. Dudney, ECS Interface, 2008, 17, 44–48.
10. B. Dunn, J. W. Long and D. R. Rolison, ECS Interface, 2008, 17, 49–53.
11. E. Takeuchi, ECS Interface, 2008, 17, 43.
12. N. T. Nguyen and S. H. Chan, J. Micromech. Microeng., 2006, 16, R1–R12.
13. S. Tominaka, H. Obata and T. Osaka, Energy Environ. Sci., 2009, 2, 845–848.
14. H. Becker and C. Gartner, Electrophoresis, 2000, 21, 12–26.
15. J. C. McDonald, D. C. Duffy, J. R. Anderson, D. T. Chiu, H. K. Wu, O. J. A. Schueller and G. M. Whitesides, Electrophoresis, 2000, 21, 27–40.
16. J. Wheldon, W.-J. Lee, D.-H. Lim, A. B. Broste, M. Bollinger and W. H. Smyrl, Electrochem. Solid-State Lett., 2009, 12, B86–B89.
17. R. Hahn, S. Wagner, A. Schmitz and H. Reichl, J. Power Sources, 2004, 131, 73–78.
18. T. Suga, H. Ohshiro, S. Sugita, K. Oyaizu and H. Nishide, Adv. Mater., 2009, 21, 1627–1630.
19. H. Nishide and K. Oyaizu, Science, 2008, 319, 737–738.
20. T. Ito, K. Kimura and M. Kunimatsu, Electrochem. Commun., 2006, 8, 973–976.
21. E. Shiells, Chem. Technol., 2009, 6(7), T50, www.rsc.org/Publishing/ChemTech/Volume/2009/07/Chips_and_alcohol.asp.
22. “Fuel cells: Breath of fresh air”, in Asia Materials, Nature Publishing Group, 28th October 2008 (http://www.natureasia.com/asia-materials/highlight.php%3Fid%20%3D%20287).
23. S. J. Hwang, M. C. Tseng, J. R. Shu and H. H. Yu, Surf. Coat. Technol., 2008, 202, 3669–3674.
24. C. K. Fredrickson, Z. Xia, C. Das, R. Ferguson, F. T. Tavares and Z. H. Fan, J. Microelectromech. Syst., 2006, 15, 1060–1068.
25. A. E. Pap, K. Kordas, H. Jantunen, E. Haapaniemi and S. Leppavuori, Polym. Test., 2003, 22, 657–661.
26. G. A. Diaz-Quijada, R. Peytavi, A. Nantel, E. Roy, M. G. Bergeron, M. M. Dumoulin and T. Veres, Lab Chip, 2007, 7, 856–862.
27. S. Tominaka, H. Nishizeko, H. Shinohara, J. Mizuno and T. Osaka, ECS Transactions, 2009.
28. S. Tominaka, T. Momma and T. Osaka, Electrochim. Acta, 2008, 53, 4679–4686.
29. W. Hatke, K.-H. Kochem and T. G. Kreul, U.S. Patent, US6551653, 2003.
30. A. N. Gavrilov, O. A. Petrii, A. A. Mukovnin, N. V. Smirnova, T. V. Levchenko and G. A. Tsirlina, Electrochim. Acta, 2007, 52, 2775–2784.
31. T. Shimizu, T. Momma, M. Mohamedi, T. Osaka and S. Sarangapani, J. Power Sources, 2004, 137, 277–283.

---