ElectroOsmosis and Particle Trapping in Microporous Membranes

EO tests were continued using 2 micron carboxylated PS Latex beads in a 0.2 mM K2HPO4 solution containing 0.1% tween. The solution had a pH of 7.54 and a conductivity of 67.0 uS/cm. The beads were negatively charged, and so would move towards the positive electrode absent electroosmotic flow. It was suggested that the pump area/fluid volume quotient would play a crucial role in determining the strength of electroosmotic pumping, so this quotient was also investigated.

When placed in a control device (unobstructed channel), the beads in the center of the channel moved towards the positive electrode (owing to their negative charge) while beads near the walls of the channel moved towards the negative electrode (presumably EO). The channel was designed to be 300 microns tall, 800 microns wide, and 8500 microns long. This gives an area/volume quotient of 0.00916 um^-1.

(1)blank control 0,+50u,+50d (video shows initial pressure driven flow, followed by a +50V potential towards the top of the screen, then a +50 volt potential towards the bottom of the screen)

When placed in a channel obstructed by a PDMS sheet containing a small conical hole (80-100 um diameter at its smallest), electroosmosis was observed at all points within the channel. This hole had an area/volume quotient of 0.02 um^-1, making it theoretically twice as effective at driving EO as the open channel.

(2)PDMS 0,+50b,0,+50a,0 (video shows hole in PDMS initially under pressure driven flow. a +50 V potential is then introduced below the hole before being followed by a +50 V potential above the hole)

In a channel with a 3 micron microporous membrane but without the PDMS sheet there was a mixture of EO and DEP as observed in the unobstructed channel as well as in previous experiments. I plan to try this again to be certain the lack of uniform EO was not due to a construction error, but it seems as if the PDMS sheet is sufficient and necessary in order to drive EO. Micropores should have a 0.66 um^-1 area/volume quotient, 33 times higher than that in the channel. If area/volume ratios were completely indicative of electroosmotic efficiency, the membrane should have performed far better than the PDMS sheet it is competing with. If area/volume quotients are at all a valid means to quantify EO strength, they must only work when comparing setups with similar critical dimensions.

In a channel with both a membrane and a PDMS sheet there was electroosmosis throughout the entire device as well as some reversible particle trapping at the membrane (though some beads attached permanently and others passed through). This was observed in two separate devices at different times.

(3)micropore 0,+50 up switch (video shows the responsiveness of the beads to electroosmotic flow as the potential direction is switched quickly)

(4)micropore0,+50a,0,+50b,0 (video shows the microporous membrane, initially with no applied voltage, then with a +50 V potential above the membrane, followed by a +50 V potential below the membrane)

(5)Particle Trapping (video shows a close up of the membrane which appears to exhibit electric trapping when the power is switched on and release when the power is turned back off)

Several attempts were made to create a device with a PDMS sheet on top of a nano pore membrane (17nm pore, 12% porosity) but the membranes appear to be too fragile to make this viable. The nano pore membrane was recorded flexing reversibly under the strain of electrically induced fluid flow before tearing apart. Part of the difficulty of this process stems from the necessity of lining up the tiny hole in the PDMS with the narrow slits of open membrane on the nano pore chips. If the nano membrane working area was as large as the micro pore ones, construction would be more feasible. It might also help keep stress evenly dispersed while pumping, which might help to prevent ripping.

(6)Nanomembrane rip (video of membrane ripping under stress from fluid flow)