Modelling for the robust design of layered resonators for ultrasonic particle manipulation.

Several approaches have been described for the manipulation of particles within an ultrasonic field. Of those based on standing waves, devices in which the critical dimension of the resonant chamber is less than a wavelength are particularly well suited to microfluidic, or "lab on a chip" applications. These might include pre-processing or fractionation of samples prior to analysis, formation of monolayers for cell interaction studies, or the enhancement of biosensor detection capability. The small size of microfluidic resonators typically places tight tolerances on the positioning of the acoustic node, and such systems are required to have high transduction efficiencies, for reasons of power availability and temperature stability. Further, the expense of many microfabrication methods precludes an iterative experimental approach to their development. Hence, the ability to design sub-wavelength resonators that are efficient, robust and have the appropriate acoustic energy distribution is extremely important. This paper discusses one-dimensional modelling used in the design of ultrasonic resonators for particle manipulation and gives example of their uses to predict and explain resonator behaviour. Particular difficulties in designing quarter wave systems are highlighted, and modelling is used to explain observed trends and predict performance of such resonators, including their performance with different coupling layer materials.

Spore and micro-particle capture on an immunosensor surface in an ultrasound standing wave system.

The capture of Bacillus subtilis var. niger spores on an antibody-coated surface can be enhanced when that coated surface acts as an acoustic reflector in a quarter wavelength ultrasonic (3 MHz) standing wave resonator. Immunocapture in such a resonator has been characterised here for both spores and 1 microm diameter biotinylated fluorescent microparticles. A mean spatial acoustic pressure amplitude of 460 kPa and a frequency of 2.82 MHz gave high capture efficiencies. It was shown that capture was critically dependent on reflector thickness. The time dependence of particle deposition on a reflector in a batch system was broadly consistent with a calculated time of 35 s to bring 95% of particles to the coated surface. A suspension flow rate of 0.1 ml/min and a reflector thickness of 1.01 mm gave optimal capture in a 2 min assay. The enhancement of particle detection compared with the control (no ultrasound) situation was x 70. The system detects a total of five particles in 15 fields of view in a 2 min assay when the suspending phase concentration was 10(4) particles/ml. A general expression for the dependence of minimum concentration detectable on; number of fields examined, sample volume flowing through the chamber and assay time shows that, for a practical combination of these variables, the threshold detection concentration can be two orders of magnitude lower.