(283d) Plug Flow of Shear-Thinning Liquids in Microchannels
AIChE Annual Meeting
2016
2016 AIChE Annual Meeting
Engineering Sciences and Fundamentals
Complex Fluids: Polymers and Macromolecules
Tuesday, November 15, 2016 - 9:15am to 9:30am
1 Introduction
The demand for sustainable and efficient continuous
processing makes micro-fluidic devices, with dimensions in the sub millimetre scale, an attractive option. Two-phase
flow chemical operations in
microchannels offer significant advantages compared to large-scale processes, such as lower sample
consumption, increased safety and high surface-to-volume ratio. However, most studies are limited to Newtonian fluids although fluids with non-Newtonian rheology
are very common industrially, including catalytic polymerization reactions, food processing and enhanced oil recovery
[1-2]. The present work investigates the dynamics of plug
formation of a non-Newtonian shear-thinning aqueous
solution and a Newtonian organic fluid
in a circular, glass microchannel using a two-colour micro-PIV system. There are, to the authors
knowledge, no previous μ-PIV studies
in liquid-liquid micro-flows
involving non-Newtonian fluids
where velocity fields are captured simultaneously in both phases.
2 Experimental setup
Two aqueous glycerol solutions containing xanthan gum (1000 and 2000 ppm)
are used as the non-Newtonian fluids
while silicone oil (Sigma-Aldrich) is the
Newtonian phase. Two oils with different viscosities, 5 and 155 cSt have been considered.
The corresponding Newtonian aqueous solution is also used for reference
and comparison. All runs
are carried out in a circular glass microchannel (Dolomite microfluidics) with inner diameter
200 μm at different combinations of flow rates of the two phases,
that varied in the range of 0.01-0.1 ml/min. Using separate syringe pumps (KDS),
the two immiscible fluids
are introduced in the microchannel via a T-junction. In the two-colour micro-PIV experiments
the aqueous phase is seeded with
1 μm
carboxylate-modified microspheres
FluoSpheres with orange
fluorescent colour (540/560 nm) whereas
the organic phase is seeded
with 1 μm blue polystyrene microspheres particles Fluoro-Max (350/440
nm). The illumination is achieved with
a UV double pulsed Nd:YAG laser (Litron Lasers) and the light emitted from the seeded fluids in the test section
is led from the microscope
to a beam splitter (Andor Technology).
Each wavelength of the emitted light is led subsequently to a separate CCD
camera. Both cameras are connected
to a laser pulse synchronizer (TSI)and a PC. Image processing software
(Insight 4G, TSI) is used
for extracting information from
the acquired video images.
The flow is also
visualized using a high speed CCD camera and a light source for the
illumination of the test section.
3 Results
Under the specific
experimental conditions, the organic 5 cSt Newotnian phase always forms the
dispersed plugs irrespective of the side branch of the T-junction used to
introduce it in the main channel. In the case of the viscous oil 155 cSt, the
phase continuity is inversed and the aqueous non-Newotnian phase forms the
plugs whereas the organic phase is now the continuous one.
During
the plug formation different stages of detachment can be observed [3]. Using the two-colour μ-PIV, averaged
velocity profiles were obtained in each phase for different locations of the plug, determined
by the position of the plug tip inside the main channel. The velocity fields for the viscous oil 155 cSt (Figure 1) reveal that the continuous phase resists the flow of the dispersed phase at the rear of
the plug causing a change
in the interface curvature (QC / QD
= 2).
Figure 1: Velocity
fields in both phases for
plug tip position (a) 0.75 (b) 1.25 (Newtonian case).
In every case a thin
filament was produced whose length depends
both on the flow rate and the viscosity
of the fluids. For the same
stage of breakage and same
flow rate (QC
/ QD = 2), the effect
of shear-thinning behaviour
is presented in Figure
2. The dynamics of the filament breakage change depending on the amount of xanthan gum in the aqueous phase and consequently
on rheology of the fluid.
For the 2000 ppm non-Newtonian case, the break-up point has moved
further downstream and the breakage time is increased. This thin film eventually produces small droplets. The droplet diameter inceases with increasing amount of
xanthan gum into the aqueous phase.
Figure 2: Dynamics of filament breakage for 155 cSt silicone oil and (a)
Newtonian aqueous phase (b) non-Newtonian aqueous phase, 2000 ppm.
The experimental results have also been compared with numerical simulations using
the CODE BLUE which is a
massively-parallel Navier-Stokes solver for both Newtonian and
non-Newtonian multiphase flows. The method
for the treatment of the fluid
interfaces and capillary forces uses a parallelized Front Tracking/Level Set technique which defines the interface both by a discontinuous density and viscosity fields as well as by a local triangular Lagrangian mesh. This structure allows the interface to undergo
large deformations including
the rupture and/or coalescence of fluid interfaces. Comparisons between experimental and numerical results on the formation of the plug at the channel inlet with
the 155 cSt silicone oil can be seen
in Figure 3.
Figure
3: Comparison between
experimental results and
simulation.
The effect of the non-Newtonian rheology on the characteristics
of the fully-formed plug flow in the main channel can be
seen in Figure
4 for the less viscous,
5 cSt, oil. For constant flowrates,
the plug shape changes as the amount
of xanthan gum increases. The plug acquires a bullet-shape profile with the
front part curved due to the viscous
forces (Figure 4b)
while the rear
part is similar to the Newtonian case.
Figure 4: Effect of non-Newtonian behaviour on plug curvature (a)
Newtonian fluid (b) non-Newtonian fluid, 2000 ppm (QC = 0.07 mL/min
and QD = 0.01 mL/min).
Acknowledgements
The
project was funded by the UK Engineering and Physical Science Research
Council (EPSRC) Programme Grant MEMPHIS.
References
[1] Salim, A., Fourar,
M., Pironon, J., Sausse, J., 2008. Oil/water two-phase
flow in microchannels: flow
patterns and pressure drop measurements. Can. J. Chem. Eng.
86, 978988.
[2] Tang, G., Lu, Y., Zhang, S., Wang, F., Tao, W., 2012. Experimental investigation of non-Newtonian liquid flow in microchannels. Journal of non-Newtonian Fluid Mechanics. 174, 21-29.
[3]
Garstecki P., Fuerstman MJ.,
Stone HA., Whitesides GM., 2006. Formation
of droplets and bubbles in a microfluidic T-junction scaling and mechanism of
break-up. Lab on a Chip 6(3): 437-446.