Instruments4Chem

High Performance, Low Cost…

Stacks Image 4101

Piezoelectric Micropumps


The Bartels Mikrotechnik MP6 is a miniature microfluidic device based on a piezoelectric (PZE) diaphragm that has passive check valves to control the flow direction. Voltage applied to a piezo-ceramic mounted on a coated brass membrane results in deformation, causing upstroke/downstroke motion that alternatively draws in/pushes out liquid from the pump chamber. The achievable liquid flow rate of these devices is 0.1-7 ml/min, controllable by the drive amplitude and frequency.

An MP6-OEM microprocessor-compatible control module (in a convenient DIP-14 package) simplifies the generation of PZE drive waveforms (up to several hundred volts peak-to-peak) while running from a low voltage supply of just 3-5V DC. Importantly, the current requirement of a single channel system (30 mA) is very modest, making battery operation quite feasible.

A Dual Channel MP6 Controller Shield for Flow Analysis


An XMOS Startkit shield housing two MP6-OEM modules is shown in the photo below. As can be seen, the shield plugs into the Startkit PCB via two long single row IDC headers at J7 and J8. Various additional headers are brought out on the shield - providing connections to two MP6 pumps, digital control signals to control a 6 port switching valve (VALVE), a connection to a VODS detector (which can be fitted with either a light-to-frequency or light-to-voltage converter), a connection to an LCD display and a detector analog out (ADC). The latter enables an analog detector signal from a VODS to then be routed to the Startkit’s on-board ADC.

This functionality affords flexibility, allowing an XMOS Starkit and this shield to be used in a versatile flow analysis system having two streams of liquid delivery, some on-line chemistry and a detection system based on colorimetry, fluorometry or even chemiluminescence.

During operation, two digital lines from the XMOS Startkit are dedicated to generating PWM waveforms for each MP6-OEM module. One input waveform sets the amplitude and the other the frequency of the pump’s drive waveform.

Using a Pair of MP6 Micropumps in a Flow-Injection Colorimetric Detection System


To carry out a preliminary evaluation of these Bartels micropumps, a two channel MP6-based flow system was assembled, as shown in the photo below. The XMOS Startkit with its dual channel MP6 shield can be seen upper left of centre. One pump channel transports water from a drinking glass (carrier) while a second channel transports red food dye solution from another glass (sample).

The two liquid streams then merge at a Y-junction whose outlet goes to a spectrophotometric flow cell that is inserted into a VODS detector housing. The VODS unit is fitted with a green LED and a TSL235 light-to-frequency converter. A blue peg holds the cuvette at the correct height for the light to pass through. The outlet from the cuvette then passes to a waste container seen on the right hand side of the photo.

LabVIEWTM 2-Channel Pump Controller and Detector Logger


The LabVIEWTM front panel shown below runs pump1 (the carrier; here water) continuously, while logging the light intensity on the TSL235 detector (green trace). Intensity readings are also converted to absorbances (lower orange trace). Periodically, pump 2 (sample) is turned on briefly to inject small aliquots of red food dye solution.

Two time settings control this action and both are measured in units of the detector period. Here, the detector period is set to 100 msec, the pump 2 off time to 600 (60 sec) and the pump 2 on time to 10 (1 sec). This results in a 1 second injection of dye solution every minute while during the run, 10 detector readings are taken each second.

As noted earlier, each pump channel has an amplitude and frequency setting that are used to control the flow rate. Separate sets of these parameters are used for pump 2 to accommodate the off and on cycles.

In the data logged in this experiment, the system was run for 13 injections. As the food dye enters the colorimetric detector there is a dip in light intensity as seen in the top (green) trace. The absorbance in the bottom (orange) trace changes in the opposite direction from zero (carrier only) to ~ 0.20 (carrier + dye). The individual peak absorbance values are consistent to within 5-10% over the course of this experiment.

The use of micro pumps for sample delivery is very attractive as it avoids the need for a costly 6-port switching valve and further optimisations are currently underway to characterise long term performance in this type of application. For analytical work the percent relative standard deviation (% rsd) of peak absorbance readings measured over the course of a run needs to be improved to around 1% before this approach can be considered a serious option, however.