47187-G5
Solvent Effects on Water Restructuring and Electrokinetics at Water-Hydrocarbon Interfaces
We utilized time-resolved electrokinetic measurements to study the electrokinetic properties of silica and TOPAS microuidic channels as a function of the time history of the fluid-solid interface. In pressure driven flow through TOPAS microchannels, the zeta potential as inferred from streaming potential measurements was observed to decay exponentially by a factor of 1.5 with a characteristic decay time of 3 hours after the initial formation of the fluid-solid interface. A similar exponential decay was observed immediately after water was exchanged for ethanol as the solvent in the system. In electroosmotically driven flow through TOPAS microchannels, the zeta potential as inferred through current monitoring experiments was constant in time. No electrokinetic transients were observed in silica microchannels under these flow conditions.
We measured the zeta potential as a function of time after filling with aqueous solution in both hydrophobic (TOPAS) and hydrophilic (silica) microchannels under pressure driven flow, utilizing the phase sensitive streaming potential technique. In both materials, we flushed a pH 7, 1 mM phosphate buffer solution through the microchannels using an applied pressure sinusoidally varying from 0 to 103 kPa (gauge), with a period of 120 s. In TOPAS, we also measured the zeta potential for a 10 mM, pH 7 phosphate buffer solution with the same pressure wave parameters. The resulting sinusoidal streaming potential signals were monitored continuously and recorded with a sampling rate of 5 Hz for 10 hours. The applied pressure and resulting streaming potential waveforms were analyzed in 12 minute segments to give an inferred zeta potential as a function of time.
In the TOPAS microchannels filled with 1 mM phosphate buffer, the normalized zeta potential was initially high in magnitude at -25 mV, but decayed to an equilibrium value of -16.1 mV. The transient is well fit by an exponential decay, with a time constant of 3.0 hours. The normalized zeta potential for 10 mM phosphate buffer in TOPAS had a very similar exponential decay, with an equilibrium value of -16.0 mV and a time constant of 2.7 hours. The normalized zeta potential in the silica microchannels, on the other hand, was constant over the 10 hour period, with a time average of -27.0 mV that varied by less than 3% for the duration.
We conducted time-resolved measurements of the zeta potential in both hydrophobic TOPAS and hydrophilic silica microchannels under electroosmotic ow using the current monitoring technique. For measurements of the electrokinetic properties of TOPAS, an electric _eld magnitude of 120 V/cm was applied across the microchannel for a half cycle period of 6 min. For the silica microchannels, an electric field magnitude of 50 V/cm was used with a half cycle period of 10 min. A 10 mM, pH 7 solution of phosphate buffer was used in both the silica and the TOPAS experiments. In contrast to our experiments in pressure driven flow, the zeta potential in both TOPAS and silica was constant and varied by less than 7% over a period of 10 hours. The time averaged value of the normalized zeta potential for TOPAS was -14.1 mV with a standard deviation of 1.0 mV or 7%. The time average of the normalized zeta potential in silica was -25.7 mV with a standard deviation of 1.6 mV or 6%.
We examined the effect of an initially applied electric field in disrupting electrokinetic transients in pressure driven flow. TOPAS microchannels were initially filled with a 1 mM, pH 7 phosphate
buffer solution. An electric field of magnitude 500 V/cm was then applied across the microchannel for 12 min so as to actuate electroosmotic flow. The electric field magnitude was chosen so that the shear stress at the fluid-solid interface during the initial period of electroosmotic flow would be roughly a factor of 3 larger than that in the current monitoring experiments. The electric field was then switched off, flow actuated via a sinusoidal pressure waveform, and the streaming potential measured using the same experimental parameters.
In contrast to pressure driven flow without an initially applied electric field, the zeta potential
remained relatively constant from the time the electric field was switched off over a period of 10 hours. The normalized zeta potential decayed from -17.3 mV to -15.1 mV, and had a time average of -15.9 mV, with a standard deviation of 0.8 mV or 5%. The variation is not explained by an exponential decay, as an exponential fit to the data in Figure 4 resulted in a non-sensical time constant and a low coefficient of determination. The datum at t=0 was not used in this analysis, since one of the four trials for that datum was determined to be a statistical outlier by Chauvenet's criterion.
