There is a growing demand in the micromachining to produce highly selective biological sensors for the detection of small quantities of biological microorganisms. Micromachining technology confers advantages over conventional fabrication techniques due to extreme dimensional control, an ultimate decrease in physical volume, low cost as a result of batch processing, and the feasibility of integrating on-chip circuitry.
In this work, the detection of bacteria using a resonant frequency based mass detection biological sensor has been accomplished. The biological sensor under consideration consists of an array of resonating cantilever beams fabricated, using well established bulk silicon micromachining techniques, from both low pressure chemical deposition (LPCVD) and dual frequency plasma enhanced chemical vapor deposition (PECVD) low stress silicon nitride (SiNx). Signal transduction of the micromechanical oscillators has been accomplished by measuring the out of plane vibrational resonant mode with an optical deflection system (see Fig 1.).
For this experiment an array of cantilevers with dimensions of varying length from 75mm to 400mm, width of 20mm, thickness (t) of 320nm for the LPCVD, and t=600nm for the PECVD nitride, were used. Escherichia coli serotype O157:H7 antibodies were applied to a chip containing 8 cantilevers. Figure 2 shows a tapping mode atomic force micrograph (TMAFM) of antibodies immobilized onto a silicon nitride surface. Devices were exposed to an E. Coli containing solution and loosely bound cells removed. Figure 3a and 3b show respectively a TMAFM and a SEM of cells bound to cantilevers.
Within our experimental study, in order to determine the mass bound to the cantilever, a frequency spectra was taken before and after the cell binding events occurred. Resonant frequency shift resulting from an additional mass loading from the specific binding of the Escherichia coli cells to the Escherichia coli serotype O157:H7 antibodies was found to be in agreement with the predicted theory of bending beams. Under ambient conditions where considerable damping occurs, we were able to detect 16 Escherichia Coli cells (see Fig 4). Resonators exposed to solution without E. Coli cells showed no measurable frequency shift. Methods, utilizing vacuum encapsulation and tailoring of the cantilever dimensions, for single cell detection are presently pursued.
Figure 1. Schematic of the laser deflection setup used to measure the transverse vibrations of the mechanical oscillators.
Figure 2. 1mm x 1mm tapping mode atomic force topograph under ambient conditions of the antibodies immobilized on the SiNx test surface. The image was obtained using a Digital Instruments Dimension 3000 atomic force microscope using standard TESP driven at 280 KHz. Antibodies were imaged for several frames without any visible surface modification by the tip. The test sample consisted of a 1 cm^2 piece of SiNx, identical to the nitride used in the fabrication of the mechanical oscillators, which was immersed for 5 minutes into a prepared antibody solution, the sample was rinsed in deionized water and nitrogen dried. The scope trace showed excellent surface traction during imaging of the antibodies.
Figure 3a. 10mm x 10mm tapping mode image showing the distribution and size of the E. coli cells. Cross sectional analysis showed the cell height of ~300nm.
Figure 3b. Scanning electron micrographs of cells bound to the antibody layer on top of various cantilevers. In order to reduce charging effects during SEM imaging, samples were prepared by evaporating a thin (<10nm) layer of Au/Pd layer. Both samples were prepared in a similar manner (108 E. coli cells/ml) and show a random distribution of the cells.
Figure 4. The measured frequency spectra due to the transverse vibrations of the cantilevers before (black) and after (red) cell attachment. The observed small signal to noise ratio is caused by the low reflectivity of the silicon nitride.
