Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 14.2.2 - Numerical Simulation of Two Phase Flow Mixing in Microchannel with Internal Features
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00344
 Material Information
Title: 14.2.2 - Numerical Simulation of Two Phase Flow Mixing in Microchannel with Internal Features Micro and Nano-Scale Multiphase Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Qiu, D.
Tonkovich, A.L.
Fanelli, M.
Silva, L.
Lerou, J.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: gas-liquid two-phase flow
microchannel
surface feature
VOF
 Notes
Abstract: Slug or plug flow is generally considered as major gas-liquid two-phase flow pattern in microchannels. In the present study, numerical simulations show that the new microchannel designs with partial blockages or engraved features enable increased interfacial area density by gas phase breakup, i.e. by the formation of a quasi bubbly-flow pattern. Special blocking features built in the channel cause local flow acceleration, higher shear force between the phases and, in turn, the break-up of gas pockets. In the case of engraved features (surface features), the velocity components of secondary flow induced by the angled microgrooves break the gas phase into small bubbles, where much larger gas pockets/slugs would otherwise dominate in flat or smooth wall microchannels. As such, mixing of the two phases is greatly enhanced, providing potentially shorter average mass transfer distances for reactive applications. Interfacial area density, a key measure of mixing effectiveness, is evaluated through a numerical procedure based on the cell void fraction gradient calculated in the Volume-Of-Fluid (VOF) method that is used in the numerical computations. The interfacial area density values in the simulated cases are discussed and compared with literature. Flow patterns that develop over a range of gas and liquid flow rates and liquid viscosities are presented for blocking and surface features.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00344
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1422-Qiu-ICMF2010.pdf

Full Text


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010



Numerical Simulation of Two Phase Flow Mixing in Microchannel with Internal Features


Dongming Qiu, Anna Lee Tonkovich, Maddalena Fanelli, Laura Silva, Jan Lerou

Velocys, Inc., 7950 Corporate Boulevard, Plain City, OH 43064, USA

E-mail: qiut(ivelocys.com

Keywords: gas-liquid two-phase flow, microchannel, surface feature, VOF



Abstract

Slug or plug Hlow is generally considered as major gas-liquid two-phase flow pattern in microchannels.
In the present study, numerical simulations show that the new microchannel designs with partial
blockages or engraved features enable increased interfacial area density by gas phase breakup, i.e. by the
formation of a quasi bubbly-flow pattern. Special blocking features built in the channel cause local flow
acceleration, higher shear force between the phases and, in tumn, the break-up of gas pockets. In the case
of engraved features (surface features), the velocity components of secondary flow induced by the angled
microgrooves break the gas phase into small bubbles, where much larger gas pockets/slugs would
otherwise dominate in flat or smooth wall microchannels. As such, mixing of the two phases is greatly
enhanced, providing potentially shorter average mass transfer distances for reactive applications.
Interfacial area density, a key measure of mixing effectiveness, is evaluated through a numerical
procedure based on the cell void fraction gradient calculated in the Volume-Of-Fluid (VOF) method that
is used in the numerical computations. The interfacial area density values in the simulated cases are
discussed and compared with literature. Flow patterns that develop over a range of gas and liquid flow
rates and liquid viscosities are presented for blocking and surface features.


Introduction


Microchannel technology provides
opportunities to enhance mass and heat transfer
in various processes. In chemical processes, the
controllability and efficiency can be greatly
increased in a minimized hardware volume.
However, reactions involving multiphase flow
have been notoriously challenging to perform in
microchannel reactors, for example, in
absorptions for gas purifieation using liquid
sorbent and in hydrogenations and syntheses of
liquid fuels, where mass transfer to reactive gas
and liquid interfaces as well as to the catalyst
surface is crucial. Due to the microchannel
dimensions (typically 100 tpm to 1 mm in
diameter), capillary force strongly alters the flow
patterns and impacts the mixing quality between
the phases. Slug Hlow and plug Hlow are common
flow patterns in many chemical process
operation in circular or square microchannels,


where elongated gas bubbles fill the whole
channel cross section and are separated by liquid
slugs [1]. Thus, the interfacial area density -
surface area per unit volume at the interface- is
low limiting mass transfer and reaction rates. In
the current paper, an alternate approach to
efficient multiphase mixing has been developed
to disrupt gas-liquid Hlow and increase interfacial
area density in microchannels [2]. Numerical
simulations show that disrupted Hlow and large
amounts of small bubbles are created by the
integration of micro features into a rectangular
microchannel. The micro features include
blocking features and angled microgrooves
(surface features). The mixing performance of
gas-liquid two phase flow in flat and micro
feature modified microchannels is compared
through Hlow patterns from numerical prediction
and literature.









Recent advancements found in the
literature have suggested special microstructures
to enhance contact and mixing between the
different phases.
Kreutzer et al [3] described the industrial
benefits and practical pitfalls of monolith
reactors for gas-liquid reactions. It is stated that
the honeycomb of parallel capillaries in the
reactor allows the advantages of microfluidics to
be applied on an enormous scale, making this
type of reactor well suited for scale-up of
chemical processes. An interfacial area density
of 2,000-4,000 m2/m3 is reported. However,
flooding is described as an issue in counter-
current flow of gas and liquid, which prevents
"rigorous" interfacial contacting.
Abdallah et al [4] investigated the
microstructured mesh contactor for liquid-gas-
solid or gas-liquid hydrogenation, mainly for
application to catalyst/chiral indicator screening
and kinetic data acquisition. Mesh structures
coated with catalyst are used to separate the gas
and liquid bulk flows but allow mass transfer
and reaction between the phases. An interfacial
area of approximately 2,000 m2/m3 based on
liquid volume is achieved.
The microchannel concept is also applied
to packed-bed and post-bed reactors in order to
improve the mixing between phases. Losey et al
[5] built microfabricated multiphase reactors for
hydrogenation of cyclohexane. Catalyst particles
are filled into array of microchannels 625x300
tpm2 in Size. A 10 fold rate increase in reaction is
reported. It is seen from the flow pattern that
bubbles form slugs that occupy the full channel
cross section. This indicates that the gas slug is
not broken up into small bubbles by the packed
bed. "Pulsing flow" is also found in the single
channel.
Yue et al [6] experimentally studied
cocurrent gas-liquid Hlow through a horizontal
rectangular microchannel with a hydraulic
diameter of 667 tpm by absorbing pure CO2 into
water and a 0.3M NaHCO3 / 0.3M Na2CO3
buffer solution. They report slug Hlow and
annular flow as major Hlow patterns in their
system, i.e. no break-up of gas bubbles are
observed. Interfacial areas were determined by
absorbing pure CO2 into a 1M NaOH solution.
An interfacial area density as high as 9000
m2/m3 iS reported.


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010

From the cited literature we found that
the major two-phase Hlow patterns in these new
generation microstructures are either film flow
(annular or falling film) or slug Hlow including
rectangular microchannels. In addition,
interfacial area density in a wide range is
reported 2000 -9000 m2/m .

NUMERICAL SIMULATION APPROACH
Other than for the global two phase flow
parameters such as gas volume fraction or
pressure drop, it is difficult to directly measure
local flow characteristics such as the interfacial
area, the slip, the three dimensional (3D) bubble
shapes and their dependence on operating
conditions. Thus, computational simulation is
helpful in gaining insight into these details. In
this paper, in order to evaluate the disrupting
effects of the surface features and blocking
features on the Hlow patterns and mixing
enhancement in microchannels, numerical
simulations are conducted on a flat
microchannel, angled surface features, and
blocking features in comparable parameter
ranges.
Flow of gas and liquid phases is
simulated using VOF formulation of the
software package Fluent where a control-
volume-based technique is used to solve the
mass and momentum conservation equations in
each computational cell. The volume fraction of
gas phase, u, is a variable determined in the
continuity equation for one of the phases [7]:
88
-+ V (a u) = 0 (1)

where n is velocity vector. The value a = 0or
a 1 in a cell represents a cell full of liquid
phase or gas phase, respectively. When the cell
contains an interface between the two phases,
the average void fraction is 1> a > 0. A single
momentum equation for incompressible fluids
throughout the domain involving the two phases
is solved, and the resulting velocity field is
shared among the phases:
(pu) + V (puu) = -Vp + V [pu(Vu + VuT)] + pg (2)


where p is pressure, g is gravity vector, and p is
density (averaged by void fraction in each cell).
It is noted Eq. (2) only applies to the non-






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010

the interfacial surface area in a cell can be
calculated from the gradient of the cell property.
In the VOF model with constant fluid properties,
the only cell property is void fraction a. Thus,
the surface area vector of the interface within a
cell is given by


interface fluid zone. The continuum surface
force (CSF) model proposed by Brackbill et al [8]
is applied at the interface, which provides a
source term in the momentum equation. In the
CSF model, the surface curvature is computed
from local gradients in the surface normal at the
interface. Let n be the surface normal, defined as
the gradient of a,

n = Vu (3

The curvature, k, is defined in terms of the
divergence of the unit normal, i,

k- V- ii (4

where

Ai = n / In I (5

The force (the pressure jump across the
surface due to interfacial tension) at the interface
can be expressed as a volume force using the
divergence theorem. It is this volume force that
is the source term which is added to the
momentum equation. It has the following
reduced form in a two-phase system:


Ai=-V* (Va)


3) As described above, because the VOF
model uses the apparent contact angle, the model
is unable to capture the extremely thin liquid
film between wall and gas bubbles. Thus, In
calculating the gas-liquid interfacial area, the
)gas-wall contact area is also taken into account.
Because sufficient discretization is
required to represent the smallest gas bubble size,
a grid size sensitivity study on the flow results
was conducted. It was found that a grid size
smaller than 1/3 of the bubble size ensures the
grid independence of the results. Figure 1 shows
examples of two grid structures in the grid size
sensitivity study. It is noted that smaller grid size
drastically increases the computation time and
computer resources. Thus, the grid size was
chosen that bubbles larger than 1/3 of the
microgroove s width would be accurately
simulated. The grid example is for a case of 45 o
surface features defined by the bold lines in
6) Figure 1(a) : thus the grids are made along the
angle direction.





.2
.2


pkVa


(P, + Pa)


where p is the volume-averaged density.
Equation (6) shows that the interfacial tension
source term for a cell is proportional to the
average density in the cell. Constant interfacial
tension a is used in the simulations.
At the contact lines between the interface
and the solid wall of the microchannel and the
features (three-phase-line), contact angle is
specified in order to reflect the wall wettability
of the studied fluid. Contact angle is generally
obtained from photographic measurement and is
actually the so-called "apparent contact angle",
i.e. at the three-phase-line there is no transition
between bulk liquid and a thin liquid film
between gas phase and the wall.
Because the gradient of a cell property
can be determined by adding up the property
values at the cell's bounding faces multiplied by
the face area and dividing by the cell volume V,


(a) Coarse grad






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010

simulated gas bubble size range, number of
bubbles in a size, bubble shape and spacing are
similar to those captured in the testing. It is
noted that the gas bubbles are all flattened, i.e.
the thickness of the bubbles spans the
microchannel height (178 tpm). The interfacial
area density of this case is 2851 m /n? (based on
total volume). This value is aligned with that of
the reviewed literature.


(b) Finer grid
Figure 1 Grid size sensitivity study examples

SIMULATION RESULTS AND
DISCUSSIONS
A benchmark microchannel
configuration without features is simulated and
compared with experimental visualization. The
experimental set-up consists of a flat channel
0.16 inch (4.06 mm) in width, 0.007 inch (178
tpm) in gap and 3 inch (7.6 cm) in length, while
the simulation domain is of the same width and
gap, but half the length. Figure 2 shows the
computational results of the phase contour (Hlow
pattern) in the mid-plane of the flat
microchannel at flow rates of 10 sccm for gas
and 10 ccm for liquid, a surface tension of 0.036
N/m, and contact angle of 900. It is seen that the


(a) Numerical simulation


(b) Experiment
Figure 2 Flow pattern in flat microchannel

Figure 3(a) shows the three dimensional
(3D) Hlow domain of the microchannel including
the 450 asymmetric surface feature defined in the
right part of the figure. The microchannel is 0.18
inch (4.57 mm) in width, 0.007 inch (178 tpm) in
gap and 1.67 inch (4.36 cm) in length, while the
surface feature has a depth of 0.015 inch (381
pLm).



v \ Dimension of Details


Liquid inlet chamber
Mixing ofthe
/two phases


Mixture outlet


Overall View


Gas inlet
chamber


(a) Geometry of simulated Hlow domain


Flow
Direction


(b) Flow pattern


Figure 3 Model domain of microchannel with the 45 o asymmetric surface feature.






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010


:'~init


Figure 4 Stream line of two-phase flow in microchannel with 45o asymmetric surface feature


It is noted that in this model, the full flow
Hield is simulated (no symmetry boundary
condition is applied). The simulation assumes
flow rates of 10 sccm for the gas and 10 ccm for
the liquid, surface tension of 0.06 N/m, contact
angle of 90o and viscosity of 0.059 kg/m s.
Figure 3(b) shows contours of the two-phase
interface at the middle plane of the microchannel
(cut in the main channel above the surface
features). It is seen that gas phase is broken into
many smaller bubbles. There is no large gas
pocket at a width comparable to the channel
width.
The evaluated interfacial area density is
5655 m2/m3 based on the volume of both channel
and surface features (total volume of the fluid).
This value is higher than that in the monolith
reactor reported by Kreutzer et al [4] and that in
the mesh structure from Abdallah et al [5] but
lower than that of Yue et al [7].
From the streamlines shown in Figure 4, it
is found that two large secondary Hlow streams
are formed near the channel side walls and partly
enter into the surface features from the sides at
an angle. The local velocity inside the
microgrooves and near the bottom is mostly in
the direction along the grooves, which is
approximately normal to the secondary flow.
The difference between the channel bulk Hlow
and the microgroove local Hlow causes shearing
of the gas pockets or bubbles that Hlow by or
partly enter the surface features, greatly
enhancing two-phase mixing.
In the simulation described in Figure
5(a), a microchannel with built-in blocking


features (posts) is defined. The simulation
assumes Hlow rates of 10 sccm for the gas and 10
ccm for the liquid, surface tension of 0.06 N/m,
contact angle of 90o and viscosity of 0.059 kg/m
s. The computational results of the flow pattern
are shown in Figure 5(b) from the top-view of
the mid-plane. It is seen that small bubbles are
formed at the outlet.
The interfacial area density is found to be
8721 m2/m3 based on the total volume of
channel. This value is 55% higher than that in
the microchannel with built-in surface features,
indicating the more effective enhancement of
two-phase Hlow mixing caused by the blocking
features than the surface features.
From the streamline picture in Figure 6, it
is seen that there are multiple secondary flows
along the diagonal directions. The streamlines
near one sidewall move to the opposite sidewall
or to the center regions. From the velocity field
in Figure 7(a), the effect of this kind of
secondary flow can be more clearly seen: the
sharp edged blocking features, posts, face the
incoming flow and split it and the gas pockets.
The split gas pockets are further split into
smaller bubbles, until the bubble size becomes
similar to the spacing between the posts. The
initially uneven velocity profile near the inlet is
made more uniform by the splitting of the flow
from row to row. It is noted that each post is
comprised of a larger and smaller half-post. The
larger and smaller posts alternate from top to
bottom along each row, as detailed in Figure
5(a). This alternating post size design causes
flow components that alternate towards top and






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010

are relatively low, meaning that the shear force
mainly comes from the structures' contribution.

CONCLUSIONS
Surface features, angled microgrooves,
can be built in rectangular microchannels to
disrupt two-phase flow, create more interfacial
area and enhance mass transfer. As shown in the
simulated case, an interfacial area density of
5655 m2/m3 can easily be reached. Post-like
blocking features can also be incorporated in
microchannels, leading to repeated splitting of
gas pockets into many smaller bubbles. The
interfacial area density predicted for a simulated
case exceeded 8700 m2/m Results of numerical
simulations suggest that break-up of larger
bubbles by secondary flows through
microgooves is the major mechanism of bubbly
flow formation and higher interfacial area
density. A Hlow splitting effect appears to be the
major mechanism of bubble break-up in
microchannels with blocking features (posts).



Mixture out


bottom of the channel, applying extra shear on
the gas bubbles.
A comparison of the velocity field
between posts with that of the interfacial
contours in Figure 7(b), showing a vertical cut
through the middle plane of the channel, shows
that gas phase regions have a higher velocity (7-
9 m/s). This indicates a slip between gas and
liquid, as the average liquid velocity is lower (<
6 m/s). Higher local gas velocity has higher
momentum, so that gas pockets are more easily
split by the post, when they hit the sharp edges
of the posts, additionally enhancing two-phase
mixing.
The simulation results discussed above
show that a microchannels built with blocking
features can provide an interfacial area density
of nearly 9000 m2/m3, while a Similar one built
with surface features provides interfacial area
density higher than 5000 m2/m It should be
noted that the simulated flow rates and velocities










Liquiid in




Gas


(a) Geometry of simulated flow domain with blocking feature -posts


M ixture
>out


(b) The two phase flow pattern
Figure 5 Model domain and flow pattern in microchannel of blocking features (posts).


Liquid in






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010




di le


1~tsJ-~



I ,"~?


L-



I;---- I;E





II I
I ~ :W.~
I.:il:~- ..:::l:s~:r.t~3M ~
I1~
I .t

Ia eoiyfedto iwna ne



Iotlcto



Figure Streamline o twophse fltow n mircanlopst









(b~~~~() Velocity Heldd top view atcne ln ne oar isnleith hs at

Fiur 42eoiyfedo topaefo nmircanlo o


REFERENCES






7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 June 4, 2010

[1] J. W. Coleman, and S. Garimella, "Two-
Phase Flow Regime Transitions in
Microchannel Tubes: The Effect of
Hydraulic Diameter," Proc. ASME Heat
Transfer Division-2000 HTD, Vol. 366-4, pp.
71-83, 2000.

[2] A.L. Tonkovich, M. Fanelli, R. Arora, T.
Sullivan and D. Kuhlman, Multi-phase
contacting process using microchannel
technology, US Patent Application
20070085227, April 2007 .

[3] M. T. Kreutzer F. Kapteijn and J. A.
Moulijn, "Shouldn't catalysts shape up?
Structured reactors in general and gas-liquid
monolith reactors in particular," Catalysis
Today 111, pp. 111-118, 2006.
[4] R. Abdallah, V. Meille, J. Shaw, D. Wenn.
and C. de Bellefon, "Gas-liquid and gas-
liquid-solid catalysis in a mesh microreactor,"
Chem. Comm., pp. 372-374, 2004.

[5] M. W. Losey, M. A. Schmidt, and K. F.
Jensen, "Microfabricated Multiphase
Packed-Bed Reactors: Characterization of
Mass Transfer and Reactions," Ind. Eng.
Chem. Res., 40, pp. 2555-2562, 2001.

[6] J. Yue, G. Chen, Q. Yuan, L. Luo, Y.
Gonthier, "Hydrodynamics and mass transfer
characteristics in gas liquid Hlow through a
rectangular microchannel," Chemical
Engineering Science, 62, pp. 2096-2108,
2007.

[7] Fluent User's Guide, 2005. Fluent
Documentation 6.2, www. fuent. com.
Lebanon, NH.

[8] Brackbill, J.U., Kothe, D.B. and C.Zemach.,
(1992) "A Continuum Method for Modeling
Surface Tension". J. Comput. Phys.,
100:335-354




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs