Dielectrophoretic Force Microscopy

Dielectrophoresis (DEP), the force exerted on a polarizable particle in a nonuniform electric field, has been used with great success in recent years to manipulate and separate cells and biomolecules.^1–3 DEP-based microfluidics platforms have been shown to be capable of separating different species of microorganisms as well as separate viable yeast cells from nonviable yeast cells.^2,3 A scanning-probe microscopy technique has been developed in which DEP forces were incorporated into the feedback mechanism of an atomic force microscope (AFM).^4,5 In dielectrophoretic force microscopy (DEPFM), an ac field is applied between a conducting AFM tip and a counterelectrode. The resulting DEP image yields information on charge mobility and local capacitance with nanometer-scale spatial resolution. DEPFM has two distinct advantages over alternative related dc or quasi-dc electric field-induced force microscopies (i.e., electric force microscopy, polarization force microscopy, etc.) when imaging under aqueous conditions: 1) The peak potentials accessible by DEPFM are significantly greater, and 2) the ac frequency dependence yields information on charge mobility. Localized DEP spectroscopy can be performed by sweeping the ac frequency and recording corresponding changes in DEP force. Additionally, the nature of the DEP interactions reduces mechanical tip–sample contact during imaging. Currently, in situ contact and intermittent contact (tapping mode) AFM studies of biological systems are routinely limited by artifacts associated with tip–sample interactions, including sample deformation/damage⁶ and surface adhesion. The principles and recent applications of this technique are briefly reviewed.

The time-averaged DEP force, F_DEP(ω), on an isolated particle as a function of the ac angular frequency is governed by the following expression.⁷

In the above equation, ε₁ is the electrical permittivity of the medium, V is the volume of the particle, K(ω) is the frequency-dependent Clausius-Mossotti factor of the system, and E is the electric field. K(ω) is a sensitive function of the polarizability, permittivity, and conductivity of the particle and the suspension medium.⁷ For a biological interface such as a cell or lipid layer, the measured Clausius-Mossotti factor can be correlated to biologically interesting properties such as membrane capacitance and surface charge. Since the DEP force depends on the gradient of the squared electric field, the field must be nonuniform for there to be a nonzero DEP force. The field inhomogeneities are particularly large in the region immediately adjacent to sharp conducting tips such as those commonly used in scanning probe microscopy. The high spatial localization of the DEP force can be coupled with the high force sensitivity of AFM to allow DEPFM measurements with nanoscale resolution under aqueous conditions.

Instrumentation

DEPFM experiments were performed by adaptation of a commercial Dimension 3100 AFM (Veeco Metrology, Chadds Ford, PA) operating in the tapping mode. Ti/Pt-coated AFM tips (MikroMasch, Wilsonville, OR) with nominal resonance frequencies of 6–8 kHz in water were used in all experiments. To generate an ac electric field between the tip and the counterelectrode, the signal output from a function generator (Agilent Technologies, Palo Alto, CA) was connected to a conducting tip. The ground lead from the function generator was connected to an n-doped silicon wafer. In this configuration, the spacing between the electrodes (i.e., the AFM tip and the underlying silicon) is equal to the thickness of the wafer’s oxide overlayer plus the tip–surface separation (see Figure 1). This configuration allows for a strong, nonuniform electric field to be established with a relatively low voltage (e.g., no greater than 10 V_pp). All DEPFM experiments were conducted in ~17 MΩ resistivity water (NANOpure, Barnstead/Thermolyne Corp., Dubuque, IA). The range of ac frequencies used was kept above 50 kHz to kinetically limit the extent of electrolysis of water at the tip and to prevent interference with the cantilever motion.

Figure 1 - DEPFM instrument schematic.

When an ac field is established between an oscillating AFM tip and a sample surface during tapping-mode imaging, the resulting DEP force alters the amplitude and phase of oscillation of the cantilever. To maintain constant amplitude of oscillation, the AFM feedback mechanism subsequently adjusts the height of the cantilever above the surface. In this way, changes in the DEP force can be measured as phase shifts in the cantilever motion and/or as field-induced changes in apparent image height.

DEPFM of gold islands

An electrically isolated gold pad microstructure was analyzed via DEPFM (Figure 2). Reactive ion etching was used to deposit a series of circular gold pads onto a Si/SiO₂ wafer. The electrical pad was embedded within an oxide film and electrically isolated from the underlying Si substrate, complicating the use of more traditional scanning probe imaging techniques that require a complete electrical circuit with the sample.

Figure 2 a) - Topography, b) topographic cross-section, c) DEP force map, and d) DEP force cross-section of a gold island embedded in silica imaged in water with an 8-V_pp, 25-MHz ac field.

From the image cross-sections, it can be seen that the resulting DEP force map does not directly track the topography of the sample. The averaged cross-section of the gold island shows a flat plateau ~20 nm above the surrounding oxide layer in the DEP force map. The topography indicates a curved surface that does not extend significantly above the oxide layer. However, the image contrast arising in DEPFM reflects the dielectric properties of the underlying sample. Since the gold pad was significantly more polarizable than the silica in which it was embedded, it is reasonable to expect an average increase in DEP force over the gold region, consistent with the experimental observations.

Interestingly, at small scales (~30 nm) substantial differences in the DEP force can be seen across the gold sample. The image contrast is intimately related to local changes in topography across the gold. The topography of a conductor represents an isoelectric surface. Consequently, the curvature of the surface affects the local field gradient, and correspondingly the distance dependence of the DEP force. Convex features on the gold pad with length scales comparable to the radius of curvature of the tip resulted in maximum damping in the amplitude of oscillation, appearing as larger apparent DEP forces.

Reduction of mechanical tip–surface contact

Detailed analyses of the topographs acquired in the presence and absence of the DEP force provide compelling evidence supporting DEPFM as a noncontact imaging modality capable of routine operation under aqueous conditions. The field dependent forces adjacent to a polarizable surface (or within a polarizable medium) are relatively long range (~1/distance²). A detailed study of the power spectral densities of DEP force maps and the correlation between those force maps with their corresponding topographs has shown that the presence of a DEP force greatly reduces tip–sample mechanical interactions.⁵ When a rough porous silicon surface (root mean square [RMS] roughness ~10 nm) was imaged with DEPFM, the resulting power spectral density (PSD) of the image showed a decrease in spectral amplitude at high frequencies when compared with the PSD of the complementary topograph (Figure 3a). This preferential loss in high spatial frequencies is consistent with an effective increase in the tip radius of curvature.

Figure 3 a) - Relative difference in PSD between two 1 μm × 1 μm topographs of a porous silicon substrate acquired with no applied field, shown offset by 20%. b) Difference in PSD observed upon application of a 7-V_pp, 100-kHz ac field (adapted from Ref. 5). c) Relative difference in PSD power density between two 1 μm × 1 μm topographs of a porous silicon substrate acquired with a dulled (through repeated use) AFM tip and a sharp AFM tip.

Similar effects were observed when the topograph of a porous silicon surface acquired with a sharp AFM tip was compared with an image acquired with a dull tip (Figure 3b). These results provide evidence supporting the ability of the instrument to maintain feedback for imaging with the average position of the tip ~10–20 nm away from the surface. Compared to alternative noncontact imaging techniques, such as electrostatic, magnetic, and polarization force microscopy, DEPFM has the added benefit of being broadly applicable under aqueous conditions.

DEPFM of biological samples

A DEPFM image and complementary topograph of an Escherichia coli bacterium taken with a 5-V_pp, 100-kHz signal are shown is Figure 4. These images were taken under aqueous conditions using a Si wafer as the counterelectrode. The E. coli DEP force map showed a uniform increase in apparent height over the surface of the cell, consistent with relatively simple cell models predicting homogeneous internal polarizabilities.⁷ These measurements demonstrate that DEPFM can be reliably performed on living systems under aqueous conditions, suggesting new possibilities for ultrahigh-resolution biological microscopy using dielectric properties for image contrast.

Figure 4 a) - Topography, b) topographic cross-section, c) DEP force map, and d) DEP force cross-section of an E. coli bacterium imaged with a 5-V_pp, 100-kHz ac field (adapted from Ref. 4).

Conclusion

DEPFM affords analysis of the dielectric properties of nanoscale electronic and biological systems with the high spatial resolution typical of atomic force microscopy. The reduced tip–sample contact minimizes the sample damage/deformation that is commonly observed when soft surfaces are imaged with AFM. Currently, there are few alternative scanning probe microscopy techniques that are capable of facile noncontact imaging in aqueous media, despite the growing interest in scanning probe analyses of biological systems. In cells, the information acquired in DEPFM experiments is unique from the topography and can be related to functional properties of cells and cell membranes. Because DEPFM experiments can be performed in water and in the tapping mode, this technique should prove to be a simplistic method for the in vitro imaging of cells and lipid layers. In addition, the instrumental modifications to standard scanning probe microscopes required for performing DEPFM are remarkably straightforward.

References

1. Pethig, R.; Markx, G.H. Applications of dielectrophoresis in biotechnology. Trends Biotechnol. 1997, 15(10), 426–32.
2. Markx, G.H.; Dyda, P.A.; Pethig, R. Dielectrophoretic separation of bacteria using a conductivity gradient. J. Biotechnol.1996, 51(2), 175–80.
3. Markx, G.H.; Talary, M.S.; Pethig, R. Separation of viable and non-viable yeast using dielectrophoresis. J. Biotechnol. 1994, 32(1), 29–37.
4. Lynch, B.P.; Hilton, A.M.; Doerge, C.H.; Simpson, G.J. Dielectrophoretic force microscopy of aqueous interfaces. Langmuir2005, 21, 1436–40.
5. Hilton, A.M.; Lynch, B.P.; Simpson, G.J. Reduction of tip–sample contact using dielectrophoretic force scanning probe microscopy. Anal. Chem. 2005, 77, 8008–12.
6. You, H.X.; Lau, J.M.; Zhang, S.; Yu, L. Atomic force microscopy imaging of living cells: a preliminary study of the disruptive effect of the cantilever. Ultramicroscopy2000, 82, 297–305.
7. Jones, T.B. Electromechanics of Particles; Cambridge University Press: London, 1995; p 265.

The authors are with the Dept. of Chemistry, Purdue University, 560 Oval Dr., West Lafayette, IN 47907, U.S.A.; tel.: 765-494-5200; fax: 765-494-0239; e-mail: [email protected].