Device Structure and Fabrication
The micropump chip fabrication was outsourced to CMC Microsystems and a schematic of the device structure and fabrication techniques is shown below.
The mask layout was designed using Tanner L-edit. Optimal electrode dimensions of 50 µm were used, with 100 µm or 200 µm channel widths and 40 µm channel heights. Two different iterations of chips were designed and fabricated, each including both straight and serpentine channel geometries. The first chip shown in figure a) below was made using gold electrodes, while the second chip shown in figure b) was made with platinum.
Testing of the pump performance was done using a probing station, microscope with computer controlled digital camera, and a function generator. Pumping solutions were mixed with 10 µm polystyrene beads for imaging, and 3-4 µL of solution was injected at the inlet. Video was recorded and the flow speed of the solution was determined by tracking the distance travelled by several beads in defined time intervals as illustrated in the figure below.
Optimal voltage and frequency parameters were experimentally determined to be approximately 3 Vpp and 8 kHz respectively, using deionized (DI) water as the working solution. The experimental results are shown in the figures below.
In addition, the test results from running the pump under a variety of different conditions is shown below.
Both the gold and platinum based devices performed similarly, with velocities from 25 to 135 µm/s,
which is adequate for many lab-on-a-chip applications. Minimal degradation in performance
was seen over 3 hours of testing. However, the Pt device was more prone to microbead adhesion
and bubble formation. The pump was able to operate with a variety of fluids,
including biological fluids such as saliva. However, differences in max flow
speed were observed, which may be eliminated upon further voltage and frequency
optimization. The serpentine channel showed very low flow rates, due to
increased channel resistance.
Integration for Portability
portable micropump can operate for 6.2 hours using a 9V battery, and fits
within a space of 8cm x 6cm x 5cm. A Frequency Devices DX45 Sinewave Oscillator was used to supply an 8 kHz AC signal. Device
performance is comparable to the bench top device, with a maximum velocity of
131.2 µm/s (63.0 nL/min).
Conclusions and Future Work
integrated, portable micropump was developed based on ACEO. It can run for 6
hours on a 9V battery, with speeds of up to 131 µm/s, and
pump a variety of solutions. Sensitivity to fluid properties and
pressure are still issues, but may not be detrimental, depending on the
application. Future work will focus on
integration of a flow sensor using a resistive temperature detector (RTD) and microheater. This will be based on the principle of using 3 electrodes (shown below) to both detect the incoming solution temperature, apply a heat pulse using the middle electrode, and detect the outgoing solution temperature as it passes over the last electrode. By taking into account the time it takes the heat pulse to travel a known distance, the flow rate can be monitored completely on-chip.