PAPER www.rsc.org/loc | Lab on a Chip Electric field control and analyte transport in Si/SiO 2 fluidic nanochannels Yi Zhang,† a Thomas C. Gamble,† a Alexander Neumann, b Gabriel P. Lopez, a Steven R. J. Brueck b,c and Dimiter N. Petsev* a Received 12th March 2008, Accepted 16th July 2008 First published as an Advance Article on the web 29th August 2008 DOI: 10.1039/b804256j This article presents an analysis of the electric field distribution and current transport in fluidic nanochannels fabricated by etching of a silicon chip. The channels were overcoated by a SiO 2 layer. The analysis accounts for the current leaks across the SiO 2 layer into the channel walls. Suitable voltage biasing of the Si substrate allows eliminating of the leaks or using them to modify the potential distribution of the fluid. Shaping the potential in the fluid can be utilized for solute focusing and separations in fluidic nanochannels. Introduction The transport of fluid, current and solutes in fluidic nanochan- nels presents a significant fundamental and practical interest. 1–9 As the dimensions of the channels reach the range between a few nanometres and a few hundreds of nanometres, they become comparable to the size of the electric double layer that usually forms at channel walls. 10,11 The double layer thickness is a measure of the typical range of the electrostatic potential that propagates from the wall into the liquid and depends on the electrolyte concentration and valence. 12 Thus for an aqueous solution of symmetric 1 : 1 electrolyte, such as KCl, with concentrations varying between 0.01 and 10 -6 M, the double layer thickness would change from 3 to 300 nm. For asymmetric 2 : 1 electrolyte, such as MgCl 2 , and the same concentrations the respective double layer thicknesses vary between 1.8 and 18 nm while for symmetric 2 : 2, such as MgSO 4 , the corresponding range is 1.5 to 15 nm. Channels that are less than a few nanometres in size require a description of the molecular structure of the fluid. 13–22 The ionic and molecular transport in channels that are comparable to the double layer thickness is strongly affected by the electrostatic potential. 10,11,23 For example, electroosmotic fluid flow distribution v eo (x) in such channels follows the shape of the potential in the electric double layer W(x), i.e. v x E x eo () () =- - È Î Í ˘ ˚ ˙ ee z h z 0 1 Y (1) e 0 is the dielectric constant of free space, e is the relative dielectric permittivity of the solution, h is its viscosity, z is the electrokinetic zeta potential and x is the transversal coordinate a Center for Biomedical Engineering and Department of Chemical and Nuclear Engineering, University of New Mexico, Albuquerque, NM, 87131, USA. E-mail: Dimiter@unm.edu; Tel: +1-(505) 277–3221 b Center for High Technology Materials, University of New Mexico, Albuquerque, NM, 87106, USA c Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, 87131, USA † These authors equally contributed to the paper. across the channel. 9 The quantity E denotes the externally applied electric field. At the same time all dissolved ionic species are distributed according to their charge: counterions accumulate in the immediate vicinity of the charged channel wall while the co-ions are repelled toward the center according to the Boltzmann distribution 9 cx c ze x k () exp () = - È Î Í ˘ ˚ ˙ 0 A y T (2) c(x) is the local concentration, c 0 is the concentration in the reservoirs in fluidic contact with the channel, z A is the charge number of the analyte and e is the electron charge. The electrophoretic transport of the analyte is given by v ep c(x) where z A is the electrokinetic zeta potential of the analyte 10,11 v ep A = ee z h 0 (3) The strong double layer effects present in small channels imply that manipulating the electrokinetic z -potential at the channel wall is a convenient tool for transport control. The z - potential modulation can be achieved by applying a voltage bias at the wall by means of an additional electrode. In this way the device operates as a fluidic field effect transistor. This idea was originally suggested by Ghowsi and Gale 24–26 and applied for electroosmosis in micron-sized channels. It has only been recently used to execute transport control and manipulation in fluidic nanochannels. 2–4 Alternatively, if one can manipulate the electric field E along the channel it will also be possible to manipulate the transport of the charged analytes (see eqn (1) and (3)). Nanochannels are often fabricated on silicon wafers utilizing techniques developed in the microelectronic industry such as high resolution lithography and anisotropic etching. 7,9,27–31 The channels are then covered with SiO 2 to insulate the solution from the silicon substrate. However, SiO 2 is not a perfect insulator 32,33 and current may leak across into the underlying substrate (see Fig. 1a). As the channel width approaches the nanometre range, its resistance increases and may become comparable or even exceed that of the oxide layer. Hence, for a wide range of This journal is © The Royal Society of Chemistry 2008 Lab Chip, 2008, 8, 1671–1675 | 1671