Anal. Chem. 2000, 72, 1930-1933
Room-Temperature Imprinting Method for Plastic Microchannel Fabrication Jingdong Xu,† Laurie Locascio,‡ Michael Gaitan,§ and Cheng S. Lee*,†
Department of Chemistry and Biochemistry, University of Maryland, College Park, Maryland 20742, and Analytical Chemistry Division and Semiconductor Electronics Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899
A new plastic imprinting method using a silicon template is demonstrated. This new approach obviates the necessity of heating the plastic substrate during the stamping process, thus improving the device yield from ∼10 devices to above 100 devices per template. The dimensions of the imprinted microchannels were found to be very reproducible, with variations of less than 2%. The channel depths were dependent on the pressures applied and the materials used. Rather than bonding the open channels with another piece of plastic, a flexible and adhesive poly(dimethylsiloxane) film is used to seal the microchannels, which offers many advantages. As an application, isoelectric focusing of green fluorescence protein on these plastic microfluidic devices is illustrated. In the past decade, the field of microfluidics has developed into one of the most dynamic fields in analytical chemistry.1-10 Choices of material and fabrication procedure are critical aspects of commercial microfluidic devices. Currently, the majority of the microfluidic devices are made from glass or silicon.1-15 These materials are often not chosen because of their appropriateness †
University of Maryland. Analytical Chemistry Division, National Institute of Standards and Technology.§ Semiconductor Electronics Division, National Institute of Standards and Technology. (1) Harrison, D. J.; Manz, A.; Fan, Z.; Ludi, H.; Widmer, H. M. Anal. Chem. 1992, 64, 1926. (2) Harrison, D. J.; Fluri, K.; Seiler, K.; Fan, Z.; Effenhauser, C. S.; Manz, A. Science 1993, 261, 895. (3) Chiem, N.; Harrison, D. J. Anal. Chem. 1997, 69, 373. (4) Effenhauser, S. F.; Manz, A.; Widmer, H. M. Anal. Chem. 1993, 65, 2637. (5) Effenhauser, C. S.; Manz, A.; Widmer, H. M. Anal. Chem. 1995, 67, 2284. (6) Jacobson, S. C.; Hergenroder, R.; Koutny, L. B.; Warmakc, R. J.; Ramsey, J. M. Anal. Chem. 1994, 66, 1107. (7) Waters, L. C.; Jacobsen, S. C.; Krotchinina, N.; Khandurina, J.; Foote, R. S.; Ramsey, J. M. Anal. Chem. 1998, 70, 158. (8) Woolley, A. T.; Mathies, R. A. Anal. Chem. 1995, 67, 3676. (9) Woolley, A. T.; Sensabaugh, G. F.; Mathies, R. A. Anal. Chem. 1997, 69, 2181. (10) Van den Berg, A., Bergveld, P., Eds. Micro Total Analysis Systems; Kluwer Academic Publishers: London, 1995; p 253. (11) Volkmuth, W. D.; Austin, R. H. Nature 1992, 358, 600. (12) Murakami, Y.; Takeuchi, T.; Yokoyama, K.; Tamiya, E.; Karube, I.; Suda, M. Anal. Chem. 1993, 65, 2731. (13) Wilding, P.; Pfahler, J.; Bau, H. H.; Zemel, J. N.; Kricha, L. J. Clin. Chem. 1994, 40, 43. (14) Furlan, R.; Zemel, J. N. Sens. Actuators A 1995, 51, 239. (15) Terry, S. C.; Jerman, J. H.; Angell, J. B. IEEE Trans. Electron Devices 1997, 26, 1880. ‡
1930 Analytical Chemistry, Vol. 72, No. 8, April 15, 2000
for the applications at hand but because the technology is readily transferable from previously established procedures. The Achilles’ heel of glass or silicon-based microfluidic devices is the exorbitant expense of fabrication and the brittleness of the material. Plastics offer a very promising alternative enjoying advantages of being less expensive and easier to manipulate. One salient advantage is that the pliability of plastics lends itself readily to replication techniques such as casting, embossing, injection molding, and imprinting. McCormick et al.16 reported an injection-molding protocol for microfabricated separation devices. In this case, a wet-etched silicon template served as the master, from which nickel electroforms were made and employed as the injection mold insert. The acrylic copolymer resin was then molded on these nickel electroforms to make separation devices. The research groups of Whitesides,17-20 Biebuyck,21-22 and Effenhauser23 have demonstrated the use of an elastameric polymer, poly(dimethylsiloxane) (PDMS), to form microchannels from photoresist images on silicon template. Furthermore, a laser ablation method for plastic microchannel fabrication was presented by Roberts et al.24 We have previously reported an imprinting method for the fabrication of plastic microfluidic devices.25 This imprinting technique required heating during both the imprinting and bonding steps. Heating was thought to be useful to soften the plastics so that the transfer of the image during imprinting would be more effective. The disadvantage is that the silicon templates tend to break upon cooling due to the different thermal expansion properties of the silicon template and the plastic substrates. Practically, a silicon template could only be used to imprint ∼10 (16) McCormick, R. M.; Nelson, R. J.; Alonso-Amigo, M. G.; Benvegnu, D. J.; Hooper, H. H. Anal. Chem. 1997, 69, 2626. (17) Kim, E.; Xia, Y.; Whitesides, G. M. Nature 1995, 376, 581. (18) Kim, E.; Xia, Y.; Whitesides, G. M. J. Am. Chem. Soc. 1996, 118, 5722. (19) Jackman, R. J.; Duffy, D. C.; Ostuni, E.; Willmore, N. D.; Whitesides, G. M. Anal. Chem. 1998, 70, 2280. (20) Duffy, D. C.; McDonald, J. C.; Schueller, O. J.; Whitesides, G. M. Anal. Chem. 1998, 70, 4974. (21) Delamarche, E.; Bernard, A.; Schmid, H.; Michel, B.; Biebuyck, H. A. Science 1997, 276, 779. (22) Delamarche, E.; Bernard, A.; Schmid, H.; Michel, B.; Biebuyck, H. A. J. Am. Chem. Soc. 1998, 120, 500. (23) Effenhauser, C. S.; Bruin, G. J. M.; Paulus, A.; Ehrat, M. Anal. Chem. 1997, 69, 3451. (24) Roberts, M. A.; Rossier, J. S.; Bercier, P.; Girault, H. Anal. Chem. 1997, 69, 2035. (25) Martynova, L.; Locascio, L. E.; Gaitan, M.; Kramer, G.; Christensen, R. G.; MacCrehan, W. A. Anal. Chem. 1997, 69, 4783. 10.1021/ac991216q CCC: $19.00
© 2000 American Chemical Society Published on Web 03/17/2000
devices before it had to be discarded. A commercial vendor (Jenoptik AG, Jena, Germany) has recently adapted an improvement of the technique and utilized a blast of air to release the plastic from the template in order to prevent breakage.10 In this work, we demonstrate the imprinting of plastic microfluidic channels from silicon templates at room temperature under the application of high pressures. Since no external heating is involved, one silicon template can be used to imprint scores of microdevices. Furthermore, overall fabrication time per device is shortened from about 30 to 5 min. These characteristics can potentially make this new fabrication technique practical for mass production of plastic microfluidic devices. EXPERIMENTAL SECTION Fabrication of Silicon Micromachined Template. The silicon template was fabricated as described previously.25 Briefly, the schematic of the channels was first drawn using a standard CAD software. The image was printed onto a transparency with high resolution (1200 dpi). This transparency was placed on top of a silicon wafer with crystal orientation 〈100〉 . The wafer surface was first coated with silicon oxide and then a layer of photoresist. Upon exposure to the UV light source, the photoresist was developed, revealing the transferred image. The exposed silicon oxide layer was removed with hydrofluoric acid solution, followed by the removal of the photoresist with acetone. The wafer was anisotropically etched in ethylenediaminepyrocatechol (Transene Co., Raleigh, NC) to establish a raised three-dimensional image of the channels. The height and the width of the positive image were controlled by the amount of time the wafer was etched. The silicon template was later bonded to another silicon wafer with epoxy. This procedure provided additional strength to the template and improved the longevity of the template for imprinting. Imprinting Channels with Micromachined Template. Pieces of plastic substrates, 76 mm by 25 mm, were cut from the poly(methyl methacrylate) (PMMA) sheet (Lucite CP, ICI Acrylics, Memphis, TN) or the copolyester sheet Vivak (DSM Engineering Plastic Products, Sheffield, MA) for imprinting channels with a silicon template. First, holes were drilled through the plastic pieces ∼3 cm apart. The plastic substrate was placed over the silicon template and the whole assembly was sandwiched between two polished aluminum plates (see Figure 1). A hydraulic press (Carver Inc., Wabash, IN) was employed to apply various pressures between 450 and 2700 psi at room temperature. After the pressure was released, the open channels on the plastic substrate were sealed with a layer of PDMS film made from a Sylgard 184 silicone elastomer kit (Dow Corning, Midland, MI). Characterization of Microchannels Using Interferometry. The dimensions of imprinted microchannels were examined using a WYKO RST optical profilometer (WYKO Co., Tucson, AZ). Data obtained from the profilometer were used to determine both the depth and the morphological features of the surfaces by scanning the optical focus through a precalibrated distance and recording the positions where interference fringes appeared. Isoelectric Focusing of Green Fluorescence Protein. Green fluorescence protein (GFP) fluoresces bright green upon exposure to blue or UV light.26 An engineered GFP variant, EGFP (excitation, 488 nm; emission, 510 nm), which fluoresces 35 times (26) Misteli, T.; Spector, D. L. Nat. Biotechnol. 1997, 69, 4783.
Figure 1. Room-temperature imprinting of microfluidic channels using silicon template.
more intensely than wild-type GFP, was obtained from Clontech (San Francisco, CA). A 3-cm-long microchannel was filled with a solution containing 0.01 mg/mL EGFP and 2% Pharmalyte 3-10 (Pharmacia Biotech, Uppsala, Sweden). Focusing was performed at electric field strength of 500 V/cm with the use of 20 mM phosphoric acid and 20 mM sodium hydroxide as the anolyte and the catholyte, respectively. Protein focusing was monitored using a Zeiss (Carl Zeiss Mikroskopie, Jena, Germany) fluorescence microscope. RESULTS AND DISCUSSION Structural Integrity of Silicon Templates. A single silicon template was used for multiple presses, and the template was characterized using the optical profilometer to determine the effect of the stamping procedure on the template definition. After the 30 imprints were made on the PMMA substrates, the template was entirely intact with the dimensions of 58 µm height and 100 µm width. The difference in the height measurement of the silicon template was within 1% before and after the 30 imprints. This difference is within the experimental variance of interferometry measurements, indicating the absence of detrimental effect on the template profile as the result of room-temperature imprinting. Characterization of Imprinted Plastic Microchannels. A top view of microchannels that were fabricated by the roomtemperature imprinting process is shown in Figure 2A. A cross section profile in a trapezoidal shape is presented in Figure 2B. The channel profiles are similar to the ones produced by the heatassisted imprinting procedures as described previously.25 To compare the results obtained from two different plastic substrates of PMMA and copolyester, several microchannels were fabricated under the same imprinting conditions. Channels were imprinted for 5 min at 2700 psi. The depth and the width of the resulting imprints were clearly material-dependent (see Table 1). The dimensions of the microchannels, in general, had a relative standard deviation of less than 2%. The device variability was even lower in the microchannels made from copolyester than from PMMA. The effect of imprinting time was evaluated by stamping multiple channels in a single plastic substrate at the same pressure but varying the pressing time. Four microchannels were imprinted onto the copolyester substrate at 450 psi for 2.5, 5, 10, and 15 Analytical Chemistry, Vol. 72, No. 8, April 15, 2000
1931
A
Figure 3. Effect of press time on channel depth using roomtemperature imprinting method. Reproducibility of channel depth over three replicates at each press time is ∼2%.
B
Figure 4. Effect of press pressure on channel depth using roomtemperature imprinting method. Reproducibility of channel depth over three replicates at each press pressure is ∼2%.
Figure 2. Microscopic images of (A) top view and (B) cross section of an imprinted PMMA device. Table 1. Dimensions of Imprinted Plastic Microchannels Measured by Profilometry
PMMA copolyester
av depth (µm)
av width (µm)
no. of samples
31.9 ( 0.5 41.7 ( 0.4
90.8 ( 1.6 94.7 ( 0.6
9 10
min. The measured depths were plotted against the imprinting time as shown in Figure 3. Clearly, there was no significant increase in channel depth with an increase in the imprinting time of more than 5 min. Furthermore, the effect of imprinting pressure on channel depth was studied by stamping microchannels at various pressures for 5 min. As shown in Figure 4, the measured channel depth was represented as the percentage of image height on the silicon template. The channel depth appeared to level off at imprinting pressures exceeding 1600 psi. By comparing with the raised image on the silicon template, the established channel dimensions on the copolyester substrate corresponded to 71 and 95% reflection of template structure in height and width, respectively. Evaluation of Channel Hysteresis. To assess the long-term integrity of microchannels imprinted at room temperature, the 1932 Analytical Chemistry, Vol. 72, No. 8, April 15, 2000
channel prepared on the PMMA substrate was reevaluated one week after the imprinting was made. The difference in channel dimensions was less than 2% and was within the variability of interferometry measurements. There was no evidence of hysteresis or relaxation in the PMMA microchannels fabricated by template imprinting at room temperature. Isoelectric Focusing of Green Fluorescence Protein. A precured PDMS film was used to cover and seal the channels imprinted on the PMMA substrate. PDMS is a material that has some unique features19 making it useful for various microchannel applications. PDMS is optically transparent at the wavelengths required for the spectroscopic analysis. The low background fluorescence associated with PDMS implies that it is a better substrate than many other plastic materials for fluroescence detection. Furthermore, PDMS is chemically and physically inert, electrically insulating, and inexpensive. GFP has been a revolutionary reporter protein for various biotechnology applications in recent years.27-32 Isoelectric focusing (27) Chalfie, M.; Tu, Y.; Euskierchen, G.; Ward, W. W.; Prasher, D. D. Science 1994, 263, 802. (28) Prasher, D. C.; Eckenrode, V. K.; Ward, W. W.; Prendergast, F. G.; Cormier, M. J. Gene 1992, 111, 229. (29) Inouye, S.; Tsuji, F. I. FEBS Lett. 1994, 341, 277. (30) Wang, S.; Hazelrigg, T. Nature 1994, 369, 400. (31) Cody, C. W.; Prasher D. C.; Westler, W. M.; Prendergast, F. G.; Ward, W. W. Biochemistry 1993, 32, 1212. (32) Haas, J.; Park, E. C.; Seed, B. Curr. Biol. 1996, 6, 315.
Figure 5. Isoelectric focusing of green fluorescence protein.
of GFP in the PMMA microchannels was studied on the basis of GFP’s strong fluorescence. The image of the focused GFP band was illustrated as fluorescence intensity vs axial distance in the channel (see Figure 5). The GFP peak spreads over 400 pixels in the fluorescence image and corresponded to a bandwidth of 200 µm. CONCLUSION In summary, room-temperature imprinting method reproducibly generates microchannels on a variety of plastic substrates. (33) Tang, Q.; Lee, C. S. J. Chromatogr., A 1997, 781, 113.
In addition to PMMA and copolyester, other polymer substrates including polycarbonate and polystyrene have been successfully employed for room-temperature imprinting of microchannels. This technique imprints channels that exhibit dimensions ranging from a few to 100 µm. For production of deeper channels, injection molding is probably a better alternative. Although the durability of silicon templates works well for our laboratory applications, the silicon images can be transferred to chromium or nickel templates, which provide better mechanical durability than silicon templates in a commercial mass production setting. In the presence of ampholytes, the adsorption of ampholytes onto the channel walls eliminates the electroosmotic flow even in a hybrid device. This is evidenced by the lack of movement of focused protein bands after focusing is complete. The observation of the lack of electroosmosis in isoelectric focusing is further supported by our previous work.33 ACKNOWLEDGMENT Support for this work by the Analytical Chemistry Division of the National Institute of Standards and Technology is gratefully acknowledged. Certain commercial equipment, instruments, or materials are identified in this report to specify adequately the experimental procedure. Such identification does not imply recommendation or endorsement by the National Institute of Standards and Technology, nor does it imply that the materials or equipment identified are necessarily the best available for the purpose. Received for review October 25, 1999. Accepted February 8, 2000. AC991216Q
Analytical Chemistry, Vol. 72, No. 8, April 15, 2000
1933
