(240a) Characterization of Porous Silicon Layers on Ordered Pillar Arrays Located in Flow Channels of Microfluidic Devices

The high surface-to-volume ratio makes porous silicon attractive for microfluidic applications, like carrier matrix for proteins in enzymatic microreactors [1], catalyst support in microreactors for heterogeneous multiphase reactions [2] or in microfluidic devices for liquid chromatography separation [3]. A method to obtain porous silicon is via the anodization of highly doped silicon in hydrofluoric acid solutions [4]. The characteristics of porous silicon layers realized with this electrochemical process ? such as thickness, porosity, pore-size distribution and effective surface-area ? can be determined using gravimetric measurements [5], SEM/TEM inspection and nitrogen adsorption isotherm measurements [6]. Main disadvantages of these conventional techniques are, that they require a lot of porous material (in the order of grams) to obtain reliable data, which can only be comprised by anodization of large surfaces, i.e. wafers with a diameter of 3? or more. This aspect makes these techniques less suitable for the characterization of porous silicon layers in microfluidic devices, where porous silicon is usually locally fabricated in flow channels (i.e. small areas) and onto non-planar shapes and structures (3D instead of 2D). Here we report on fast and easy methods for the characterization of porous silicon films created on the surface of micropillars located in ordered arrays in flow channels. These ordered porous pillar arrays will be functionalized in chips for liquid chromatography separation [3] and in microreactors for studying three-phase (G-L-S) catalytic reactions, both fabricated using micromachining (cleanroom) technologies in silicon and vitreous materials.

Channels containing solid silicon pillars were made using Reactive Ion Etching with the Bosch process (gases: SF₆ and C₄F₈) in highly p-doped silicon (0.01 Ω/cm). Pillars with various shapes and heights were fabricated, with a minimal radius of 3 µm (standard lithography) and heights up to 50 µm (Fig. 1). The exterior of these pillars was anodized using the galvanostatic operational mode, for various conditions (HF-concentration and anodization current). All anodization conditions applied to solid pillars were also applied to flat substrates, in order to compare the characteristics of porous layers on pillars (3D-structures) with the properties of porous silicon realized on flat surfaces subjected to identical experimental conditions.

The characteristics of porous silicon films on flat substrates are investigated with High Resolution SEM imaging (thickness, uniformity), BET/BJH N₂ adsorption measurements (average pore-size distribution, effective surface area) and refractive index measurements (porosity). In figure 2 a typical picture of porous silicon is shown, in table 1 the properties of porous silicon films are summarized. For all investigated conditions mesoporous silicon is formed [7], of which the pore-size and thickness depends on HF-concentration and current density: the higher the HF concentration and the lower the current density, the smaller the pore- size and the porosity. Furthermore, for a fixed HF-concentration the thickness of the porous layer becomes larger for higher current densities, and for a constant current density its thickness increases with increasing HF-concentrations. The porous layers are homogeneous in depth and have a rather narrow pore-size distribution. According to literature the porosity of layers can be measured with ellipsometry [8,9]. The porosity of a porous silicon layer is closely related to its pore-size distribution [5], and indeed different refractive indices are measured for various electrolyte concentrations. However, for fixed HF-concentrations no clear difference in refractive index was found for increasing current densities, despite growing average pore sizes. In figure 3 HR-SEM images are shown of silicon pillars subjected to the electrochemical process, resulting in silicon pillars with a porous cladding. The thickness of the porous cladding is uniform across the height of the pillars, but seems less then on flat substrates, which can be attributed to differences in the E-field density across a pillar and a flat surface. Estimated average pore sizes on pillars yield roughly identical values as for flat substrates (e.g. 5-6 nm for 5 % HF, based on HR-SEM pictures). Currently the pore-size distribution of porous pillar arrays is investigated using HR-SEM imaging in combination with Focused Ion Beam Etching, as well as how to relate the average pore-size and porosity to ellipsometric data obtained from porous pillars arrays.

In conclusion, the characteristics of porous silicon layers are investigated with HR-SEM images, adsorption measurements and refractive index measurements. Since for the investigated anodization times (≤ 10 minutes) no significant differences are found between porous films created on flat surfaces and micropillars, data obtained from flat substrates can be used to quantify the properties of porous micropillar arrays, thereby avoiding destructive methods and/or time-consuming techniques. Channels filled with ordered porous pillar arrays will be implemented in microfluidic devices for liquid chromatography purposes (to enhance the available surface) and in microreactors for studying three-phase (G-L-S) catalytic reactions (as a structured support for the catalytic active phase).

References

[1] J. Drott, K. Lindström, L. Rosengren and T. Laurell, J. Micromech. Microeng., 7, 14-23, 1997.

[2] M.W. Losey, R.J. Jackman, S.L. Firebaugh, M.A. Schmidt and K.F. Jensen, J. Microelectromech. Syst., 11, 709-717, 2002.

[3] W. de Malsche, D. Clicq, V. Verdoold, P. Gzil, G. Desmet and J.G.E. Gardeniers, Lab. Chip, DOI: 10.1039/B710507J, 2007.

[4] R.W. Tjerkstra, M.J. de Boer, J.W. Berenschot, J.G.E. Gardeniers, A. van den Berg and M.C. Elwenspoek, Electrochim. Acta, 1997, 42, 3399-3406.

[5] R. Herino, G. Bomchil, K. Barla, C. Bertrand and J.L. Ginoux, J. Electrochem. Soc., 134, 1994-2000, 1987.

[6] S. Brunauer, P.H. Emmett and E. Teller, J. Am. Chem. Soc., 60, 309-319, 1938.

[7] V. Lehmann, J. Electrochem. Soc., 140, 2836-2846, 1993.

[8] S. Uehara, K. Taira, T. Hashimoto, H. Sasabu and T. Matsubara, Phys. Stat. Sol. (a), 182, 443-446, 2000.

[9] V. Torres-Costa, F. Pászti, A. Climent-Font, R.J. Martín-Palma and J.M. Martínez-Duart, Phys. Stat. Sol. (c), 9, 3208-3212, 2005.