Subscriber access provided by University of Texas Libraries

Article

A Dry Process for Production of Microfluidic Devices Based on the Lamination of Laser-Printed Polyester Films
Claudimir Lucio do Lago, Heron Dominguez Torres da Silva, Carlos Antonio Neves, Jos Geraldo Alves Brito-Neto, and Jos Alberto Fracassi da Silva
Anal. Chem., 2003, 75 (15), 3853-3858 DOI: 10.1021/ac034437b Publication Date (Web): 20 June 2003 Downloaded from http://pubs.acs.org on February 27, 2009

More About This Article

Additional resources and features associated with this article are available within the HTML version: Supporting Information Links to the 10 articles that cite this article, as of the time of this article download Access to high resolution figures Links to articles and content related to this article Copyright permission to reproduce figures and/or text from this article

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036

Anal. Chem. 2003, 75, 3853-3858

A Dry Process for Production of Microfluidic Devices Based on the Lamination of Laser-Printed Polyester Films
Claudimir Lucio do Lago,* Heron Dominguez Torres da Silva, Carlos Antonio Neves, and Jose Geraldo Alves Brito-Neto

Departamento de Qu mica Fundamental, Instituto de Qu mica, Universidade de Sa o Paulo, Av. Prof. Lineu Prestes 748, CEP 05508-900, Sa o PaulosSP, Brazil
Jose Alberto Fracassi da Silva

Escola Polite cnica, Universidade de Sa o Paulo, Av. Prof. Luciano Gualberto 158, travessa 3, CEP 05508-900, Sa o PaulosSP, Brazil

A new microfabrication process based on a xerographic process is described. A laser printer is used to selectively deposit toner on a polyester film, which is subsequently laminated against another polyester film. The toner layer binds the two polyester films and allows the blank regions to become channels for microfluidics. These softwareoutlined channels are 6 m deep. Approximately twice this depth is obtained by laminating two printed films. The resulting devices were not significantly damaged after 24 h of exposure to aqueous solutions of H3PO4, NaOH, methanol, acetonitrile, or sodium dodecyl sulfate. Electric tests with an impedance analyzer and microchannels filled with KCl solution demonstrated that (1) wide channels suffer from deformation of the top and bottom walls due to the lamination of the polyester films and (2) the toner walls are somewhat porous. Although these drawbacks limit the maximum width of a channel and the minimum distance between two channels, the process is an attractive option to other expensive, laborious, and timeconsuming methods for microchannels fabrication. The process has been used to implement devices for electrospray tip and capillary electrophoresis with contactless conductivity detection.
Miniaturization of devices and systems has been a permanent objective of the analytical community. Portability, low consumption of reagents, improved performance, and disposability are features that motivate this search. Additionally, the reduction of scale causes an increase of the importance of the interfacial region over the behavior of the fluids that can lead to new strategies for analytical systems. An approach used since the 1970s is to implement microchannels on the basis of microelectronics processes to pattern on silicon. However, microfluidics for chemistry has continuously received new contributions, because the microelectronics processes do not match all the needs of the field.
* Fax: +55 11 3815 5579. E-mail: claudemi@iq.usp.br. 10.1021/ac034437b CCC: $25.00 Published on Web 06/20/2003 2003 American Chemical Society

Although the initial development of micrototal analytical systems (TAS) was focused on glass or quartz devices,1-3 the introduction of new types of materials offered a new way for fast prototyping and disposable devices. Indeed, plastic materials have been extensively used for this purpose. Of course, the substitution of silicon and glass by polymers renders new features that are often more important than fast production and low cost. Duffy and co-workers4 established a rapid prototyping procedure for poly(dimethylsiloxane) (PDMS). This technique was recently reviewed,5 and several applications of PDMS microchips have been demonstrated.6-10 Manica and Ewing11 used the rapid PDMS prototyping as a step for glass microchip fabrication. Unfortunately, this technique still requires the use of silicon wafers and the development of photoresists. Alternative methods of fabrication of plastic microdevices include UV laser ablation and hot embossing techniques. UV laser ablation is capable of micromachining many polymeric substrates, such as polycarbonate, polystyrene, cellulose acetate, and poly(ethylene terephthalate) (PET), generating channels with controlled rugosity and increased hydrophilicity.12-14 In hot embossing
(1) Harrison, D. J.; Fluri, K.; Seiler, K.; Fan, Z.; Effenhauser, C. S.; Manz, A. Science 1993, 261, 895-897. (2) Effenhauser, C. S.; Manz, A.; Widmer, H. M. Anal. Chem. 1993, 65, 26372642. (3) Jacobson, S. C.; Hergenro der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem. 1994, 66, 1114-1118. (4) Duffy, D. C.; McDonald, J. C.; Schueller, O. J. A.; Whitesides, G. M. Anal. Chem. 1998, 70, 4974-4984. (5) McDonald, J. C.; Duffy, D. C.; Anderson, J. R.; Chiu, D. T.; Wu, H.; Schueller, O. J. A.; Whitesides, G. M. Electrophoresis 2000, 21, 27-40. (6) Martin, R. S.; Gawron, A. J.; Lunte, S. M.; Henry, C. S. Anal. Chem. 2000, 72, 3196-3202. (7) Gawron, A. J.; Martin, R. S.; Lunte, S. M. Electrophoresis 2001, 22, 242248. (8) Martin, R. S.; Ratzlaff, K. L.; Huynh, B. H.; Lunte, S. M. Anal. Chem. 2002, 74, 1136-1143. (9) Ocvirk, G.; Munroe, M.; Tang, T.; Oleschuk, R.; Westra, K.; Harrison, J. D. Electrophoresis 2000, 21, 107-115. (10) Herbert, N. E.; Kuhr, W. G.; Brazill, S. A. Electrophoresis 2002, 23, 37503759. (11) Manica, D. P.; Ewing, A. G. Electrophoresis 2002, 23, 3735-3743.

Analytical Chemistry, Vol. 75, No. 15, August 1, 2003 3853

and wire-imprinting techniques,15-20 metallic templates containing the channel patterns or wires, respectively, are heated and pressed against the polymeric material. Using wires, only linear channels are obtained. Weigl and co-workers21 have demonstrated that lamination of plastic films renders a low-cost process to produce several kinds of microfluidic devices. Using polyester, Uchiyama and co-workers22 built an on-chip fluorescence detector by incorporating a blue-light-emitting diode and an optical fiber and direct polymerization on a glass template. Recently, Tan and co-workers23 proposed the use of a photocopying machine to produce a master for molding of PDMS devices. The master is formed by photocopying the desired image over a transparency. The deposited toner layer defines the regions that will result in channels in the PDMS device. For the equipment used in that work, channel depth in the range of 8 to 14 m was obtained by using this toner-transparency master. In this paper, toner deposited on polyester transparency is used in a different fashion. Instead of producing a master, the microfluidic device is directly formed by using two transparencies as the base, and the cover and the toner are used to define the channels. This easy and fast process is described as well as its characterization and some devices. EXPERIMENTAL SECTION Materials and Reagents. Transparency film CG3300 (3M, Sumare -SP, Brazil) was used as the base material. All solvents and reagents were of analytical grade or better. Deionized water was produced by a Nanopure UV system (Barnsted, Dubuque, IA). Fabrication Process. Although the process should be tailored for each application, the basic steps are shown in Figure 1. A cover plate is prepared by drilling a hole through a piece of transparency film (I). The layout of the device is printed on a transparency film, using white for the regions where the microfluidic channels will be formed (II). The cover and the base are laminated together, producing the channels with access holes (III). Reservoirs are glued on the channel input holes (IV). The above procedure is called the single toner layer (STL) method. An alternative method is the double toner layer (DTL) that consists of laminating two printed films with mirrored images.
(12) Roberts, M. A.; Rossier, J. S.; Bercier, P.; Girault, H. Anal. Chem. 1997, 69, 2035-2042. (13) Rossier, J. S.; Schwarz, A.; Reymond, F.; Ferrigno, R.; Bianchi, F.; Girault, H. H. Electrophoresis 1999, 20, 727-731. (14) Rossier, J. S.; Ferrigno, R.; Girault, H. H. J. Electroanal. Chem. 2000, 492, 15-22. (15) Martynova, L.; Locascio, L. E.; Gaitan, M.; Kramer, G. W.; Christensen, R. G.; MacCrehan, W. A. Anal. Chem. 1997, 69, 4783-4789. (16) Locascio, L. E.; Perso, C. E.; Lee, C. S. J. Chromatogr., A 1999, 857, 275284. (17) Chen, Y.-H.; Wang, W.-C.; Young, K.-C.; Chang, T.-T.; Chen, S.-H. Clin. Chem. 1999, 45, 1938-1943. (18) Chen, D.-C.; Hsu, F.-L.; Zhan, D.-Z.; Chen, C.-H. Anal. Chem. 2001, 73, 758-762. (19) Grass, B.; Siepe, D.; Neyer, A.; Hergenro der, R. Fresenius J. Anal. Chem. 2001, 371, 228-233. (20) Galloway, M.; Soper, S. A. Electrophoresis 2002, 23, 3760-3768. (21) Weigl, B. H.; Bardell, R.; Schulte, T.; Battrell, F.; Hayenga, J. Biomed. Microdevices 2001, 3, 267-274. (22) Uchiyama, K.; Xu, W.; Qiu, J.; Hobo, T. Fresenius J. Anal. Chem. 2001, 371, 209-211. (23) Tan, A.; Rodgers, K.; Murrihy, J. P.; OMathuna, C.; Glennon, J. D. Lab Chip 2001, 1, 7-9.

Figure 1. Schematic of the main parts generated by the single toner layer (STL) microfabrication process: I, perforated polyester cover; II, printed polyester base (a, toner layer); III, cover (I) and base (II) laminated together; IV, the final device (b, liquid reservoirs).

In this case, the resulting structure has about twice the thickness of toner, and thus, the aspect ratio is improved. The layouts of the devices were made in CorelDraw 7.0 (Corel, Ottawa, Canada) and printed out by a LaserJet 6L with toner cartridge C3906A (Hewlett-Packard, Palo Alto, CA), operating at 600 dots per inch (dpi) in the raster mode. The scheme in Figure 1 suggests that the devices were individually prepared. However, since transparency film size A4 was used, dozens of devices can be produced at the same time and subsequently separated. The drilling of the film is accomplished by adapting a paper driller in order to make holes even at the center of an A4 film. Although it is not presented in the above process, it is convenient to print out a drilling guide on the cover film for the STL method. The best performance for the lamination was obtained by using a laminator AC91-230 (Gazela, Divino polis, Minas Gerais, Brazil) at 130 C and 45 cm min-1. This equipment was originally developed for plastification of documents. After the lamination, the devices were sliced using scissors. The reservoirs were formed by gluing pieces of 8-mm o.d. and 5-mm-long tubes of PVC or polystyrene using bicomponent epoxy resin. Characterization. Thermal characterization of the polyester film and toner was carried out in a thermobalance Shimadzu TGA50 and in a differential scanning calorimeter, Shimadzu DSC-50 (Kyoto, Japan). The thickness of the toner layer was measured using a profilometer Dektak 3030 (Sloan, Santa Barbara, CA). Impedance measurements were made by an impedance analyzer HP 4194A (Hewlett-Packard, Palo Alto, CA). Applications. Electrophoresis experiments were carried out in a modified apparatus that has been described elsewhere.24,25 The electrospray experiment was implemented using a high
(24) Fracassi da Silva, J. A.; Lago, C. L. Anal. Chem. 1998, 70, 4339-4343. (25) Fracassi da Silva, J. A.; Guzman, N. A.; Lago, C. L. J. Chromatogr., A 2002, 942, 249-258.

3854 Analytical Chemistry, Vol. 75, No. 15, August 1, 2003

voltage source PS/MJ30P0400 (Glassman, Whitehouse Station, NJ). RESULTS AND DISCUSSION Characteristics of the Polyester and the Toner. According to the material safety data sheet (MSDS),26 the toner inside the cartridge C3906A is mainly composed of a styrene/acrylate copolymer (45-55 wt %) and iron oxide (45-55 wt %). Its solubility in water is negligible, and it is partially soluble in toluene and xylene. The material should soften between 100 and 150 C. The thermogravimetric analysis (TGA) shows a mass loss of 50% at 400 C. A reddish residue is formed at 900 C, corresponding to 48% of the initial mass. The analysis of this residue revealed that it is mainly Fe2O3, corresponding to 45% (w/w) of the toner, which is compatible with the manufacturer information. The result of the CHN elemental analysis of the toner was C44.8%H4.1%N0.1%. Considering the residue of 48.0% at 900 C, there is a difference of 3.0% that was attributed to oxygen. Thus, the organic material has composition C86.2%H7.9%N0.2%O5.7%. For the sake of comparison, polystyrene and poly(methyl methacrylate) (PMMA) have composition C92.3%H7.7% and C60.0%H8.0%O32.0%, respectively. These values suggest that the polymeric base of the toner is composed of 82.1% polystyrene and 17.9% PMMA, which corresponds to an average composition of C86.3%H7.8% N0.0%O5.9%. Differential scanning analysis (DSC) revealed two endothermic events at 70 and 106 C, which should be related to the softening of the two polymers used in the formulation of the toner. The MSDS of the transparency film CG3300 indicates that it is a coated poly(ethylene terephthalate).27 The coating contains silica to improve the laser printing process. The film is 100 m thick and has transparency above 80% in the range from 400 nm to at least 800 nm. There is a transition at 320 nm, and the material becomes strongly absorbing below this wavelength. The DSC analysis of the polyester film shows an endothermic event starting at 200 C and with a peak at 251 C, and the TGA shows the decomposition of the polymer above 400 C. Thus, to prevent the deformation of the polyester base during the process, temperatures bellow 200 C should be used. The Fabrication Process. Figure 2a shows the appearance of the toner deposited by the laser printer on the polyester film. It is clear that the definition of the border of the channels is very poor when compared to photolithography usually employed in the fabrication of silicon or glass devices. The roughness of the toner surface reveals that the printer does not completely melt the toner. In fact, complete melting is not required for the original printing purposes, because the objective of the printer is to glue the particles of the toner to the paper or other base. The resolution of the printer and the toner particle size determine the resolution of the produced channels. The presence of toner particles randomly deposited on the region that would form the channels is also noticeable in Figure 2a. Fortunately, they are not present in an amount enough to obstruct the channels. The thickness of the toner layer, evaluated by profilometry, is 6 m. This is an average value, because the surface is flexible
(26) Material Safety Data Sheet for the HP LaserJet Print Cartridge C3906A, Hewlett-Packard: Boise, ID, 2002. (27) Material Safety Data Sheet for the Laser Printer Transparency Film CG3300, 3M: St. Paul, MN, 1998.

Figure 2. Scanning electron micrographs of a 150-m-wide channel. The toner particles appear as light gray regions. Picture (a) shows a STL before lamination. Picture (b) shows a transverse cut of a DTL channel after lamination and the remelted toner (I). The enlarged view (c) shows a 6.35-m-tall single toner particle deposited inside the 15.0-m-tall channel.

and irregular. However, it is close to the value 7 ( 1 m obtained by measuring the mass of toner deposited by square centimeter and the density of the fused material. Figure 2b shows a 150-m-wide DTL channel after being cut with scissors. The deformations of the polyester film as well as
Analytical Chemistry, Vol. 75, No. 15, August 1, 2003
3855

Figure 3. Device for evaluation of the cross-talk effect between two 200-m-wide parallel channels separated by a 10-mm-long toner barrier (a). In this example, a 400-m-wide barrier was formed. The 8-mm-o.d. PVC reservoirs (b) were glued with epoxy resin.

the residues left by the cutting process are apparent. The clear material (I) at the border of the channel is the toner that acts as glue and spacer between the sheets. Figure 2c is an enlarged vision of the channel, which shows an isolated 6.35-m particle of toner deposited inside the channel. Although the cutting process might modify the edge of the device, the height of the channel (15.0 m) is comparable to the expected value for a DTL channel. The TGA and DSC analyses of the toner and polyester suggest that it is possible to remelt the toner in the range of 110 to 150 C without significant thermal decomposition of the toner or softening of the polyester film. In fact, our studies have demonstrated that, even somewhat below 100 C, the toner remelts, and thus, the sealing process occurs. This ample range of temperature is desirable, because it makes the lamination a flexible process that can be implemented by different approaches. Indeed, a good lamination is obtained by the correct combination of temperature, time, and pressure. The excess of one or more of these factors can lead to an excessive flow of the toner and the consequent blocking of the channel. Chemical Resistance. Pieces of printed polyester film and laminated devices were submitted to solutions commonly used in capillary electrophoresis. The following aqueous solutions were employed: H3PO4 (pH 2), NaOH (pH 12), methanol 20% (v/v), acetonitrile 20% (v/v), sodium dodecyl sulfate (SDS) in phosphate buffer (pH 7), and SDS in borate buffer (pH 9). The qualitative test consisted of visual evaluation of the polyester film (color, brightness, flexibility) and toner adhesion as well as the solution leakage in laminated devices. In all cases, no significant evidence of deterioration was found, even after 48 h. Tests were carried out at 60 C for 12 h, and the same result was obtained. Of course, because of the nature of the materials used, pure organic solvents should not be used. Electrical Characterization. Some test devices were prepared, and the channels were filled with 1 mol L-1 KCl solution in order to evaluate the electrical behavior from 100 Hz up to 5 MHz in an impedance analyzer. Figure 3 shows an example of the device to evaluate the crosstalk between vicinal channels, in which a toner barrier separates two parallel 200-m-wide channels along 10 mm. Both the process STL and DTL were investigated. In an ideal case, only capacitive cross-talk would be expected, because the toner barrier would behave as a dielectric. The spectrum in Figure 4 suggests that a 400-m DTL barrier behaves as a dielectric, because the phase is about -90, and the logarithm of the modulus linearly decreases as function of the logarithm of the frequency, as would be expected for a capacitor. The transition observed when frequency approaches 1 MHz is due to capacitive leakage of the setup and
3856 Analytical Chemistry, Vol. 75, No. 15, August 1, 2003

Figure 4. Impedance spectra of parallel channels with 400-m (a) and 50-m (b) toner barriers. |Z| and are the impedance modulus and phase angle, respectively, and f is the frequency. Noisy regions of the traces for the 400-m barrier are due to the high impedance at low frequency.

should not represent a real limitation. On the other hand, the low and constant impedance modulus and phase 0 suggest that the 50-m DTL barrier behaves as a resistor, i.e., there is flow of solution between both channels. This experiment was carried out in duplicate with devices containing 50-, 100-, 200-, or 400-m toner barriers. For the DTL process, the leakage was observed for 50- and 100-m barriers. For the STL process, even a 200-m toner barrier was not enough to perfectly isolate parallel channels. These results suggest that the porosity of the toner layer is not completely eliminated after the lamination step. The DTL process seems to be more effective, probably because the two toner layers remelt together, forming a more compact material. In another test, straight 24-mm-long STL and DTL channels were prepared with different widths (50, 100, 150, 200, 400, and 800 m). Often, obstruction of the 50-m-wide channels was observed, and the devices were discarded. The other devices were used to estimate the depth of the channels using their geometry, the conductivity of the filling solution, and the impedance of the channels. Table 1 shows the average of two evaluations of the depth when the STL and DTL processes are used to produce channels in the range from 100 to 800 m. For both processes, the depths of the 100-, 150-, and 200-mwide channels agree with those obtained by profilometry (6 m) and the surface density of the toner layer (7 m). However, the depth of the 400- and 800-m wide channels are considerably smaller.

Table 1. Channel Depths Evaluated by Impedance Analyses channel width m 100 150 200 400 800 STL channel deptha m 6(2 5(2 5(1 3.5 ( 0.2 2.58 ( 0.05 DTL channel deptha m 14 ( 5 14.6 ( 0.1 12 ( 1 9(2 6.5 ( 0.2

a Average of two measurements and the resulting estimated standard deviation.

In fact, several other tests confirm this trend, and it is probably related to the deformation of the polyester film caused by the lamination process. Blank regions are not supported by the toner layer and become prone to deformation by the heat and pressure of the lamination. Narrow channels do not significantly suffer from this deformation because of the reinforcement given by the proximity of the toner walls. These tests reveal some important features that one should have in mind when a device is being conceived. Narrow channels (below 100 m) and barriers (below 400 m) should be avoided. The resolution of the HP LaserJet 6L printer is 42 m (600 dpi), which is very close to the 50 m used in the tests, and ultimately determines the limiting resolution of the whole process. On the other hand, wide channels (above 200 m) do not have the desired geometry. Fortunately, wide channels are seldom used. Application Example: Contactless Conductivity Detection in Capillary Electrophoresis. Since the introduction of the contactless conductivity detection to capillary electrophoresis,24,28 several research groups have used it for detection of inorganic and organic species, and recently it has also been applied to separations in microchip format.29-35 Figure 5 shows the layout of a device for a microchip electrophoresis with this kind of detector. The device has the usual layout for microchip electrophoresis with a double-T injection element (500-m long) and a 3.4-cm separation channel. The DTL method was used to produce 150-m-wide channels. Adhesive copper tape (3M, Auckland, New Zealand) strips were glued outside the polyester film near the end of the separation channel to work as the electrodes. The electronics of the detector has been described elsewhere.24,25 Running electrolyte was 30-mmol L-1 MES/His (pH 6.0). The sample (100 mol L-1 KCl, NaCl, and LiCl) was electrokinetically injected at 1.0 kV until current stabilization (1 min). The electropherogram shown in Figure 6 was obtained at 1.0 kV, which corresponds to an electric
(28) Zemann, A. J.; Schnell, E.; Volgger, D.; Bonn, G. K. Anal. Chem. 1998, 70, 563-567. (29) Guijt, R. M.; Baltussen, E.; van der Steen, G.; Frank, H.; Billiet, H.; Schalkhammer, T.; Laugere, F.; Vellekoop, M.; Berthold, A.; Sarro, L.; van Dedem, G. W. K. Electrophoresis 2001, 22, 2537-2541. (30) Lichtenberg, J.; Rooij, N. F.; Verpoorte, E. Electrophoresis 2002, 23, 37693780. (31) Pumera, M.; Wang, J.; Opekar, F.; Jel nek, I.; Feldman, H. L.; Hardt, S. Anal. Chem. 2002, 74, 1968-1971. (32) Wang, J.; Pumera, M. Anal. Chem. 2002, 74, 5919-5923. (33) Wang, J.; Pumera, M.; Collins, G. E.; Mulchandani, A. Anal. Chem. 2002, 74, 6121-6125. (34) Tanyanyiwa, J.; Hauser, P. C. Anal. Chem. 2002, 74, 6378-6382. (35) Wang, J.; Pumera, M. Anal. Chem. 2003, 75, 341-345.

Figure 5. Device layout for microchip electrophoresis with contactless conductivity detection. Double-T element for 500-m-plug injection. Copper strips (a) were used as the electrodes. The PVC reservoirs (b) were glued with epoxy resin.

Figure 6. Electrophoretic separation of K+ (1), Na+ (2), and Li+ (3), 100 mol L-1 each. Running buffer: MES/His 30 mmol L-1 (pH 6.0). Electrokinetic injection at 1.0 kV for 1 min. Separation voltage, 1.0 kV. Contactless conductivity detection at 530 kHz and 10 Vpp.

field of 294 V cm-1 along the 2.2-cm path between the injection and detection points. In addition to the low cost and ease of fabrication, the thickness of the polyester film (100 m) provides a suitable barrier to contactless conductometric detection. Other fabrication techniques require that the substrate be etched.34 Application Example: Electrospray Tip. In 1997, Ramsey and Ramsey36 and Karger and co-workers37 introduced microfluidic devices for electrospray ionization, which is one of the most important ionization sources for mass spectrometry. Polymer chips have also been used to this end.38-40 Figure 7 shows an application example of fabrication of a microfluidic device for electrospray generation. The layout is quite simple and consists basically of a reservoir and a straight DTL channel that ends at the edge of the device. This outlet is easily made by cutting the border of the laminated set with scissors.
(36) Ramsey, R. S.; Ramsey, J. M. Anal. Chem. 1997, 69, 1174-1178. (37) Xue, Q.; Foret, F.; Dunayevskiy, Y. M.; Zavracky, P. M.; McGruer, N. E.; Karger, B. L. Anal. Chem. 1997, 69, 426-430. (38) Yuan, C.-H.; Shiea, J. Anal. Chem. 2001, 73, 1080-1083. (39) Rohner, T. C.; Rossier, J. S.; Girault, H. H. Anal. Chem. 2001, 73, 53535357. (40) Gobry, V.; van Oostrum, J.; Martinelli, M.; Rohner, T. C.; Reymond, F.; Rossier, J. S.; Girault, H. H. Proteomics 2002, 2, 405-412.

Analytical Chemistry, Vol. 75, No. 15, August 1, 2003

3857

Figure 7. A device for the electrospray source. The DTL process was used to produce the structure (a). The outlet was made by cutting the border with scissors, resulting in a device (b) with a sharp end. The outlet was dipped in a silicone solution to make the tip hydrophobic.

was applied to the electrode. The counter electrode, consisting of a copper foil, was placed 3 mm distant from the tip. The electrospray phenomenon was observed from a solution drop hanging at the tip (Figure 8). For a 1 mmol L-1 KCl water/ methanol 9:1 (v/v) solution, a stable Taylors cone is established with a current of 170 and 240 nA for 150- and 200-m-wide channels, respectively. This current level and the fact that light emission was not observed suggest that what was taking place was only the electrospray without corona discharge. The steadystate condition is obtained when the amount of solution reaching the tip by pumping equals the sprayed amount. Thus, different pumping pressures should be required for other channel dimensions. CONCLUSIONS The toner-polyester process is quite simple, allowing one to employ easily obtainable materials and equipment. In this paper, we have described only one example, but most of the components can be substituted. For example, there are several alternatives to the drawing software, the laminator can be replaced by a domestic iron, or a photocopying machine can be used instead of a laser printer. Compared to the photolithographic approaches commonly used in other microfabrication process, the presented process has disadvantages regarding the aspect ratio and compactness of the walls. However, to our knowledge, it is the simplest, easiest, fastest, and lowest-cost way to produce prototypes and disposable microfluidic devices. The absence of significant amounts of solutions and materials hazardous to the environment is another quality to be emphasized. ACKNOWLEDGMENT This work was supported by the Conselho Nacional de Desenvolvimento Cient fico e Tecnolo gico (CNPq) and Fundac ao de Amparo a ` Pesquisa do Estado de Sa o Paulo (FAPESP). The authors thank CNPq and FAPESP for the fellowships, Dr. Jivaldo R. Matos for the thermal analyses, and Dr. Z. G. Richter for the English revision. This work was presented at The 8th LatinAmerican Symposium on Biotechnology, Biomedical, Biopharmaceutical, and Industrial Applications of Capillary Electrophoresis and Microchip Technology, Mar del Plata, Argentina, 2002 (OPA23, PP-A41, PP-A42, PP-A43).

Figure 8. Electrospray tip. The drop of KCl solution at the outlet of the channel (a) is distorted by the electrostatic field, generating the Taylors cone (b).

The border was dipped in a silicone OV1 (Carlo Erba, Sa o Paulos SP, Brazil) solution, forming a hydrophobic layer that prevents the aqueous solutions from spreading along the edge. A Plexiglas base was built to hold the device. The solution in the reservoir was pressurized at 50 cm of water while 3-3.5 kV

Received for review April 28, 2003. Accepted May 14, 2003.
AC034437B

3858

Analytical Chemistry, Vol. 75, No. 15, August 1, 2003