Microfluid Nanofluid (2015) 18:19–27
DOI 10.1007/s10404-014-1414-y
RESEARCH PAPER
Active control of ferrofluid droplet breakup dynamics
in a microfluidic T-junction
Yining Wu • Taotao Fu • Youguang Ma •
Huai Z. Li
Received: 10 February 2014 / Accepted: 28 April 2014 / Published online: 9 May 2014
Ó Springer-Verlag Berlin Heidelberg 2014
Abstract This work aims at investigating the breakup
dynamics of the ferrofluid droplet under magnetic fields in
a microfluidic T-junction. The whole breakup processes of
the ferrofluid droplets under uniform magnetic field and
non-uniform magnetic field were exhaustively considered.
The influence of both the flow rates and magnetic flux
density on the formation and breakup were extensively
studied. It was found that the type of breakup process can
be changed and the breakup frequency of ferrofluid drop-
lets can be actively modulated by applying a uniform
magnetic field. While under non-uniform magnetic field,
the asymmetric breakup of ferrofluid droplets can be real-
ized. Both the magnetic field intensity and hydrodynamic
force affect the asymmetry of the breakup process. The
non-uniform magnetic field can exert attraction on ferro-
fluid droplets in a dissymmetrical way to avoid breakup.
Keywords Droplet-based microfluidics

Ferrofluid droplet
Breakup
Multiphase flow

Interfacial phenomenon
Abbreviations
UM Uniform magnetic field
NM No magnetic field
Y. Wu
T. Fu (&)
Y. Ma
State Key Laboratory of Chemical Engineering, School of
Chemical Engineering and Technology, Tianjin University,
Tianjin 300072, People’s Republic of China
e-mail: ttfu@tju.edu.cn
Y. Wu
H. Z. Li (&)
Laboratory of Reactions and Process Engineering, University of
Lorraine, CNRS, 1, Rue Grandville, BP 20451,
54001 Nancy Cedex, France
e-mail: Huai-Zhi.Li@univ-lorraine.fr
BPO Breakup with permanent obstruction
BTO Breakup with temporary obstruction
BNO Breakup without obstruction
NB Non-breakup
NS Unstable breakup
1 Introduction
The droplet-based microfluidics is capable of precisely
handling minute amounts of fluidic elements (Belloul et al.
2011; Fu et al. 2009; Jullien et al. 2009; Leshansky et al.
2012). Under microscale, these systems possess high
interfacial areas and short diffusion distance which facili-
tate the heat and mass transfer (Shao et al. 2010; Tan et al.
2012). Thus, much attention has been paid to their appli-
cations in chemical engineering (Abate and Weitz 2011;
Hou et al. 2011), biochemistry analysis (Engl et al. 2008),
materials synthesis (Duncanson et al. 2012) and so on. To
achieve successfully these applications, the control of
droplets’ size is of prime importance. In the present studies,
the droplet size is controlled during droplet generation and
the breakup/coalescence processes downstream (Fu et al.
2012; Leshansky et al. 2012; Salkin et al. 2013; Wu et al.
2014). Most of these studies take advantage of geometry-
based passive methods (Salkin et al. 2013). The formations
of droplets take place usually in T-/Y-junction (Carrier
et al. 2014; Glawdel et al. 2012; Leshansky and Pismen
2009; Steegmans et al. 2010; Wang et al. 2011), co-flow
(Cramer et al. 2004; Ganan-Calvo and Riesco-Chueca
2006) and flow-focusing (Fu et al. 2012; van Hoeve et al.
2011; Ward et al. 2010) microdevices. The generations of
droplets under different operating parameters and various
123
20
flow regimes such as squeezing, dripping and jetting were
studied (Christopher et al. 2008; Cordero et al. 2011; Fu
et al. 2012; Li et al. 2011; Steegmans et al. 2010; Utada
et al. 2007). Some physical models were proposed, and
previous works show that the droplet size increases with
the flow rate ratio of dispersed phase to continuous phase.
The breakup and coalescence process downstream can be
regarded as competing mechanism: The former decreases
the droplet size, while the latter increases it. To breakup
droplets in a microfluidic device, various geometries were
employed. Besides T-/Y-junction, obstacles can be placed
in microchannels (Bedram and Moosavi 2011; Leshansky
et al. 2012; Salkin et al. 2013). In these devices, the size
ratio of daughter droplet can be controlled (Bedram and
Moosavi 2011; Hoang et al. 2013; Leshansky et al. 2012;
Salkin et al. 2013; Samie et al. 2013) by changing the
length ratio, cross-section ratio of the branches of the
junction or by changing the position/length of obstacles. To
realize the coalescence between droplets (Bremond et al.
2011; Christopher et al. 2009; Wang et al. 2013), divergent
and convergent structures are mainly used to facilitate
contact and collision between droplets.
Within geometry-based passive microdevices, the
droplet size is mainly determined by the device geometry
(Bedram and Moosavi 2011; Christopher et al. 2008;
Cordero et al. 2011; Fu et al. 2012; Hoang et al. 2013;
Leshansky et al. 2012; Li et al. 2011; Salkin et al. 2013;
Samie et al. 2013; Steegmans et al. 2010; Utada et al.
2007). For a given geometry, it is difficult to casually
modulate the size of droplets. It is then necessary to make
use of external factors to regulate the size of droplets.
Several works have elaborated local control coupled with
the flow rate ratio to extending the droplet size distribution
(Kim et al. 2007; He et al. 2010; Ho et al. 2011; Cheung
and Qiu 2011; Liu et al. 2011a; Tan et al. 2010; Tan and
Nguyen 2011; Wu et al. 2013). He et al. transformed the
flow front of the dispersed phase to a Taylor cone in turn
causing Rayleigh capillary instability to control the
breakup processes. Ho et al. controlled the viscosity and
interfacial tension by local heating to adjust the sizes of
droplets. Cheung and Qiu utilized the acoustic actuation to
bring periodic oscillation to control the sizes of droplet.
Recently, the magnetic manipulation has attracted much
attention (Liu et al. 2011a; Tan et al. 2010; Tan and
Nguyen 2011; Wu et al. 2013). The magnetic field will
neither induce thermal effect nor affect pH, surface charge
and ionic concentration (Nguyen 2012). In contrary to
laser, acoustic actuation and electric field, it is simple and
practicable to endow a fluid with the paramagnetic char-
acter by adding a minute amount of nanoscale magnetic
particles (Afkhami et al. 2010; Nguyen 2012; Wu et al.
2013). Therefore, it is straightforward to manipulate the
formation of ferrofluid droplets by an external magnetic
123
Microfluid Nanofluid (2015) 18:19–27
field (Liu et al. 2011a; Tan et al. 2010; Tan and Nguyen
2011; Wu et al. 2013). The effects of uniform and non-
uniform magnetic fields on the droplet formation of fer-
rofluids are studied (Liu et al. 2011a, b). The first work in
T-junction concerns the active control of droplets’ size by
changing the magnet positions (Tan et al. 2010). When the
magnet is placed at the downstream of the T-junction, the
formation of bigger droplets is favoured. Inversely, it forms
smaller droplet. Some similar observations were also made
in flow-focusing device, in particular under a uniform
magnetic field (Liu et al. 2011a, b; Tan and Nguyen 2011;
Wu et al. 2013). Under a radial uniform magnetic field, the
magnetic force acting on the droplet offsets some part of
the impact force of the continuous phase during expanding
process of droplet. While under axial magnetic field, the
magnetic force stretches the droplet to enlarge the flow
section for the continuous phase within the microchannel to
expel. The impact force of the continuous phase on droplet
is then reduced. To our best knowledge, only a very few
studies focus on the breakup dynamics of ferro-droplets.
Our present work aims to study the breakup of ferro-
droplets in T-junction under uniform and non-uniform
magnetic fields.
2 Experimental
2.1 Details of microchannel
Figure 1 shows the layout of the microchannel that consists
of a flow-focusing geometry to form droplet and a sym-
metric loop. The continuous phase (mineral oil) flows into
the two identical lateral branches with the equal flow rate
Qc/2, and the dispersed phase (ferrofluid) flows into main
channel at a flow rate of Qd; these three flow streams focus
at the intersection downstream forming new droplets. And
then, the droplet flow toward the symmetric loop is illus-
trated in the right-hand side diagram of Fig. 1. This
structure guarantees the same pressure drop in two sym-
metric branches, which reduces maximally the influence of
the distribution of continuous phase to the breakup of ferro-
Fig. 1 Schematic diagram of the microfluidic channel. Ferrofluid
droplets breakup into two daughter droplet at the microfluidic
T-junction divergence. The cross-section of the formation part of
the microchannel is 400 9 400 lm
Microfluid Nanofluid (2015) 18:19–27
droplets. At the T-junction divergence of the loop, the
microchannel is divided into two branches of equal length;
here, droplet could breakup into two daughter droplets. At
the T-junction convergence, the two streams recombine
and exit into the outlet channel. The microfluidic device
was fabricated in a plate (90 9 30 9 8 mm) of polymethyl
methacrylate (PMMA) by precision milling and sealed
with another PMMA plate (90 9 30 9 3 mm). The mi-
crochannel has a square cross-section of 400 lm wide and
400 lm high.
2.2 Experimental setup and procedure
The scheme of the experimental setup is shown in Fig. 2.
The microfluidic device was placed under an inverted
microscope (Leica, Germany) which was connected to a
high-speed camera up to 100,000 frames per second
(CamRecord600, Optronis GMBH, Germany). Three frame
rates (500 fps with the resolution 1,280 9 1,024 pixels,
1,000 fps with the resolution 800 9 500 pixels and 2,000
fps with the resolution 400 9 248 pixels) and a shutter
speed of 1/6,000 s were used. The dispersed phase (ferro-
fluid) and the continuous phase were injected to the mi-
crochannel from two independent syringes (5 and 10 mL,
Hamilton, Germany) driven by syringe pumps (Harvard
Apparatus, USA). The uniform magnetic field was realized
by placing two neodymium–iron–boron (NdFeB) magnet
(50 9 15 9 15 mm, Supermagnete, France) at each side of
the loop as shown in Fig. 2. The T-junction divergence
where the droplets fragment was placed in the middle
height of magnets. The direction of the uniform magnetic
field is parallel to the branches of the T-junction diver-
gence. The non-uniform magnetic field is achieved by
placing only one boron (NdFeB) magnet at one side of the
T-junction divergence. The magnetic flux density could be
easily adjusted by changing the separation distance
between the T-junction divergence and magnets. The
magnetic flux densities were measured using a Gaussmeter
(GM07, Hirst, UK) with an accuracy of 0.01 mT. A cold
fiber illumination (Jeulin SA, France) was placed on the top
Fig. 2 Scheme of the experimental setup. The magnets are added in
the two sides and the left side to form uniform and non-uniform
magnetic field, respectively
21
of the microdevice as light source. The images were ana-
lyzed by customized image processing software Matlab.
2.3 Materials
In order to form ferrofluid droplets, water-based ferrofluid
(EMG 807, Ferrotech, Germany) and oil phase (mineral
oil) were, respectively, introduced as the dispersed phase
and continuous phase. At 27 °C, the density and viscosity
of the ferrofluid were qd = 1,100 kg m-3 and ld = 2 -
mPa s, respectively. As the magnetic particles had an
average diameter of 10 nm, the magnetoviscous effect
could be negligible (Pamme 2006). The particles volume
concentration was 1.8 % in this ferrofluid with the initial
susceptibility v = 0.39. The mineral oil (CAS 8042-47-5,
Sigma-Aldrich, Germany) with 4 % (wt) surfactant sorbi-
tan lauric acid ester (Span20, Sigma-Aldrich, Germany)
was used as the continuous phase. The density and vis-
cosity of the continuous phase were qc = 838 kg.m-3 and
lc = 35.2 ± 0.2 mPa s, respectively. The interfacial ten-
sion between the ferrofluid and mineral oil measured by a
dynamical tensiometer (IT Concept, France) was
r = 1.5 ± 0.1 mN m-1.
3 Results and discussion
3.1 Droplet breakup and non-breakup in the T-junction
under uniform magnetic fields
Previous studies indicate the behaviors of droplets at the
T-junction depend on the droplets’ size L or the dimen-
sionless length e = L/pWc (where Wc denotes the width of
the channel) and the capillary number Ca (Bedram and
Moosavi 2011; Leshansky et al. 2012; Salkin et al. 2013).
In our experiments, the influence of these two parameters
and the magnetic flux densities will be addressed. To gain a
comprehensive insight into the effect of a uniform mag-
netic field (UM), it is suitable to study different breakup
regimes at T-junction.
There are four main breakup regimes for ferro-droplet at
the T-junction both under uniform magnetic field (UM) and
no magnetic field (NM) as illustrated in Fig. 3a–d. The
breakup processes without magnetic field (NM) are intro-
duced as a reference for comparison. These different
regimes provide a visual description of the influence of a
magnetic field.
1. Breakup with permanent obstruction (BPO) as shown
in Fig. 3a
Droplets penetrate T-junction with two branches
obstructed by the droplet under UM and NM. The differ-
ence between the breakup processes under UM and NM is
123
22
Fig. 3 Breakup process of ferrofluid droplets in the T-junction
divergence
inconspicuous. Within optical resolution limits, there is no
free flow section between the droplet and the walls adjacent
the feed channel during the entire breakup process. The
two dumbbell-like parts in each branch obstruct the con-
tinuous flow and finally breakup symmetrically into two
daughter droplets.
2. Breakup with temporary obstruction (BTO) as shown
in Fig. 3b
At the beginning of neck forming process, the two
dumbbell-like parts enter into contact with the walls
adjacent the feed channel and the tunnel between them is
123
Microfluid Nanofluid (2015) 18:19–27
not observable. With the development of the thinning
process of the neck, the two dumbbell-like parts detach
from the walls to give rise to flow tunnels. It is obvious that
under UM the tunnel is much bigger than the case under
NM as the ferrofluid droplet is stretched by the magnetic
force which is perpendicular to the feed channel by posing
two magnets on both sides of the device as shown in Fig. 2.
Then, after a moment, the droplet breaks into two identical
daughter ones.
3. Breakup without obstruct (BNO) as shown in Fig. 3c
After reaching at the T-junction, the droplet stretches
but the neck is not formed immediately. The droplet
approaches to the wall opposite to the feed channel,
replaces the former droplet stops near the stagnation point
until the arrival of the next droplet. After the impact of the
new coming droplet, it breaks into two daughter ones.
During the whole process, the flow tunnels are clearly
visible between droplet and the side walls of the branches,
allowing the continuous phase to flow through during the
breakup. Moreover, there is no obvious difference between
two breakup process under UM and NM.
4. Non-breakup (NB) as shown in Fig. 3d
Droplets enter into the T-junction and then flow out
from either of the bifurcations depending on the feedback
of the former droplets which still flow in the loop (Cybulski
and Garstecki 2010; Wu et al. 2012). This regime occurs
under relatively small flow rate and small droplet size.
The breakup of these droplets at the T-junction depends
on the droplet size L or the dimensionless droplet size e and
the capillary number Ca (Leshansky et al. 2012; Link et al.
2004), as illustrated in Fig. 4a–d. Non-breakup occurs at
small droplet size. Between non-breakup (NB) and breakup
without obstruct (BNO), there exists an unstable breakup
(NS) region. In this region, the mother droplets are sensi-
tive to the distribution of the continuous phase in the two
branches. The feedback effect at T-junction due to impact
of daughter droplets from two branches will bring a fine
difference of flow rates between two branches. This could
lead to a non-breakup of the mother droplet. Subsequently,
the following will not breakup until the new balance is
reached. Thus, the droplets arrive at T-junction breaking
into two daughter ones or not in an irregular way. Cer-
tainly, this configuration occupies relatively narrow
domain. With the increase in the dimensionless droplet
size, the BNO, BTO and BPO scenarios occur sequentially.
It can be observed in Fig. 4a–d that the BTO region
gradually enlarges toward BPO with the increasing mag-
netic flux densities. Figure 4e depicts the displacement of
the frontier between BTO and BPO. This is caused by the
elongation of ferrofluid droplets due to magnetic force
(Afkhami et al. 2010; Liu et al. 2011b; Wu et al. 2013).
Microfluid Nanofluid (2015) 18:19–27
Fig. 4 a–d Operating diagram organizing droplet behavior as a
function of capillary number Ca and the dimensionless bubble size e
under four different magnetic flux densities. The dotted line
The ferrofluid droplet is stretched in the direction of the
uniform magnetic field. This direction is parallel to the
branches of T-junction; thus, the two dumbbell-like parts
detach from the walls offering larger flow sections of the
continuous phase in two branches as shown in Fig. 3b.
According to the previous works (Afkhami et al. 2010; Liu
et al. 2011b; Wu et al. 2013), the magnetic force that
stretches the droplet is proportional to the square of the
magnetic field strength H. The stable shape of the droplet is
determined by the ratio of interfacial tension and magnetic
force in unbounded flow. The magnetic Bond number was
used to represent the ratio between magnetic force and
surface tension Bo = l0vWcH2/r (Nguyen 2012). Here,
l0 = 4p 9 10-7 N/A2 is the permeability of the free
space. For B increase from 0 to 24.1, 52.1 and 89.8 mT
23
represents the critical capillary number for the transformed from
BPO to BTO region. e The displacement of the critical capillary
number as the increase in the magnetic flux densities
(B = l0H), Bo ranges from 0, 47.7 and 222.8–661.9. The
droplet aspect ratio b/a (b is the major axis of droplet and
a the minor axis) increases with the magnetic Bond number
Bo. Therefore, the increase in magnetic intensity enlarges
the area of BTO region toward BPO region. As shown in
Fig. 4, the area for BNO is barely changed. This is due to
the small droplet size and the tunnels are inherent. The
external magnetic field expands only the opening of the
tunnels.
The uniform magnetic field affects not only the type of
breakup, but also the breakup processes. As illustrated in
Fig. 5, the temporal evolution of the minimum width of the
thinning neck Wn in BPO (Fig. 5a), BTO (Fig. 5b) and
BNO (Fig. 5c) regions. The time zero is set at the moment
of the droplet entirely penetrates into the branches.
123
24 Microfluid Nanofluid (2015) 18:19–27

consequence, the tunnels as shown in Fig. 5 disappear with

decreasing flow rates in those gutters. But the prerequisite
is invalid in the second stage (Wn \ 0.5Wc), where the neck
starts autonomous pinching. During the second stage, the
flow rate in the gutters cannot be ignored (Hoang et al.
2013). The flow reversal from these gutters accelerates the
pinch off of the neck (Hoang et al. 2013). These explain
why the scaling law deviates from the experimental data in
this stage. In BNO region, the flow section is relatively
larger; thus, the enlargement of the tunnels due to the
elongation of the droplets is slight. The breakup process is
slightly accelerated by the magnetic force. But the external
magnetic field plays an important role while the breakup
processes locate within BTO region. With increasing
magnetic flux densities, the breakup process takes longer.
The deceleration of the thinning rate becomes increasingly
dominant in the middle of the breakup process revealed in
Fig. 5b. It is clear that the scaling law (c-Wn) * t3/7 con-
forms to the data only at the beginning of the breakup
processes. As the increase in the magnetic intensities, the
deviations become more significant. We find that the
appearance of the deviation can be related to the opening of
the tunnels between the droplet and the walls.
Figure 6 describes the temporal evolution of the mini-
mum dimensionless width of the tunnel d/Wc. As shown in
Fig. 6, the magnetic field causes an early appearance of the
tunnels and the enlargement of the flow section. In our
experiments, there is a critical value of the d/Wc. Above
0.08 ± 0.02, the deviations between the data and the
scaling law appear. This reveals the leakage from the
tunnels cannot be ignored when d/Wc exceeds 0.08 ± 0.02.
Considering Figs. 5b and 6, as the difference of d/Wc
increases, the difference of thinning rate becomes gradually
pronounced, due to the augmentation of leakage of the
continuous phase from the tunnels with the increase of
d. Furthermore, it seems that the pressure difference
between droplet cap and neck decreases and at the same
Fig. 5 a–c Temporal evolution of the minimum width of the neck Wn
as a function of time in BPO, BTO and BNO region, respectively,
under four different magnetic flux densities

Obviously, the breakup process in BPO region is barely

influenced by the external magnetic field. The evolution of
the neck thickness Wn in the early stage (Wn [ 0.5Wc)
(Hoang et al. 2013) is approximately in agreement with the
previous data in the literature (Hoang et al. 2013; Le-
shansky et al. 2012). As shown in Fig. 5a, the scaling law
(c-Wn) * t3/7 describes well the data in the early stage
with c as a fitting constant here. Leshansky et al. (2012)
deduced the scaling law based mainly on the mass con-
servation in the depression region. This prerequisite Fig. 6 Temporal evolution of the minimum dimensionless width of
imposes that the volume change in the depression region is the tunnel d/Wc as a function of time under two different magnetic
equal to the inlet flow rate during the breakup. As a flux densities

123
Microfluid Nanofluid (2015) 18:19–27 25

Fig. 7 Temporal evolution of

the asymmetric breakup of a
ferrofluid droplet

time the shear of continuous phase acting on droplet of 630 ± 30 lm. Figure 8a reveals that the ratio VR/VL
increases. This implies that in BTO region, the pressure decreases obviously with the increase in the magnetic flux
difference is the primary factor to determine the breakup density. In general, the volume ratio VR/VL increases with
process. Qt. This implies that the hydrodynamic force plays an
increasing role. To assess this point, we use the velocity
3.2 Droplet breakup and non-breakup in the T-junction ratio of droplets in the right and left branch UR/UL to
under non-uniform magnetic fields represent the flow rate ratio in two branches. As shown in
Fig. 8b, the velocity ratio UR/UL was measured as a func-
Under a non-uniform magnetic field, the magnetic force tion of Qt. With the increase in Qt, the ratio UR/UL
F acting on droplet is not symmetrical (Mefford et al. decreases gradually toward an asymptotic value of one.
2007). In our experiments, the magnet was placed at the
left side of the microchannel. Under such a magnetic field,
the breakup of the ferrofluid droplet becomes asymmetrical
due to the magnetic force applied in the left direction. Two
daughter droplets with different sizes flow into two oppo-
site branches as shown in Fig. 7. These daughter droplets
bring additional flow resistances to the branch which they
flow through (Salkin et al. 2013). And the additional flow
resistance due to a droplet increases with the increase in
droplet size (Fuerstman et al. 2007). This gives rise to
feedback effects and affects then the flow distribution of
continuous phase at the T-junction. In our experiments, the
breakup of droplet occurs in a symmetric loop; the breakup
process is mainly controlled by the magnetic field and the
continuous flow. But the T-junction divergence and the
T-junction convergence are nearly of the equipotential line
in the magnetic field. The magnetic field affects only the
breakup processes but does not provide additional energy
to the daughter droplets. The kinetic energy is only pro-
vided by the continuous phase. Thus, the influences of the
magnetic field and continuous phase are coupled to some
extent.
Figure 8a shows the volume ratio of the droplets flowing
in the right branch and in the left branch VR/VL (where
V denotes droplet volume, and subscripts R and L mean
‘right’ and ‘left’, respectively) as a function of the total
flow rate of both phases Qt = Qc ? Qd under three dif- Fig. 8 a Volume ratio of the droplets flowing in the right branch and
ferent magnetic flux densities. The size of mother droplets in the left branch VR/VL as a function of the total flow rate Qt under
three different of magnetic flux density. b The velocity ratio of
is controlled by a fine adjustment of continuous and dis- droplets in right and left branch UR/UL as a function of the total flow
persed phases to maintain the length within a narrow range rate Qt

123
26
Fig. 9 Volume ratio VR/VL as the function of the flow rate ratio QC/
Qd. (Inset) the flow condition when QC/Qd is equal to 1.778
The hydrodynamic force should be dominant when the
total flow rate Qt is very high.
Furthermore, the flow rate ratio between the continuous
phase and dispersed phase QC/Qd exerts also an influence on
the breakup of mother droplets. Keeping a constant total
flow rate Qt at 50 lL/min, the flow rate ratio QC/Qd was
varied. Figure 9 shows the variation of the volume ratio VR/
VL in function of the flow rate ratio QC/Qd. VR/VL decreases
with either the increase in the flow rate ratio QC/Qd or the
augmentation of the magnetic flux density. The decrease
rate grows with the increase in the magnetic flux density,
but the difference between these decrease rates diminishes.
This is probably due to the feedback of daughter droplets in
each branch. As illustrated in the inset of Fig. 9, the dis-
tance between droplets is small for relatively lower QC/Qd.
The daughter droplets in the left branch are larger due to the
applied magnetics force. As a consequence, the flow resi-
dence grows in the left branch to lead to a decrease in the
continuous flow rate in this branch. The velocity of these
droplets pushed ahead by the continuous phase decreases
logically. Moreover, the separation distance between
daughter droplets in left side decreases also to induce some
squeezing effects. During the breakup process of a mother
droplet at the T-junction, the droplet cap in the left branch
was hampered by the daughter droplet. The influence of the
magnetic force is then limited when QC/Qd is relatively
small. Figure 9 shows that under the higher magnetic flux
densities (B = 38.9 ± 0.4 mT, B = 52.1 ± 0.7 mT) and
QC/Qd, the volume ration VR/VL becomes zero: All droplets
go to the left branch without breakup. The non-uniform
magnetic field facilitates the non-breakup of droplets.
4 Conclusions
In the present work, both uniform and non-uniform mag-
netic fields were applied to realize an active control of the
123
Microfluid Nanofluid (2015) 18:19–27
droplet breakup of ferrofluid in T-junction. Under the
uniform magnetic fields, the area of BTO region is
enlarged toward BPO region with a longer time required
for the breakup process. The frequency of the breakup
processes can be easily adjusted by the applied magnetics
force. The pressure drop plays an important role in BTO
region for the droplet breakup process. When the breakup
process occurs under the non-uniform magnetic fields, two
daughter droplets with different sizes can be formed. With
the increase in the gradient of the magnetic field H, the
asymmetry of breakup process is intensified, while the
increase in total flow rate Qt or the decrease in the distance
between droplets weakens the influence of the magnetic
field. Furthermore, the non-uniform magnetic field facili-
tates the non-breakup of droplets. These results throw new
insight into the breakup mechanism and the size control of
ferrofluid droplets under an additional and well-controlled
external force and could provide a new avenue to explore
complex interfacial phenomena at microscale.
Acknowledgments The Chinese group acknowledges the support
by the National Natural Science Foundation of China (21106093,
21276175) and the Program of Introducing Talents of Discipline to
Universities (B06006). Y. Wu appreciates the fellowship of the China
Scholarship Council and the technical help of Dr. D. Funfschilling.
References
Abate AR, Weitz DA (2011) Faster multiple emulsification with drop
splitting. Lab Chip 11:1911–1915
Afkhami S, Tyler AJ, Renardy Y, Renardy M, St Pierre TG,
Woodward RC, Riffle JS (2010) Deformation of a hydrophobic
ferrofluid droplet suspended in a viscous medium under uniform
magnetic fields. J Fluid Mech 663:358–384
Bedram A, Moosavi A (2011) Droplet breakup in an asymmetric
microfluidic T junction. Eur Phys J E 34:78
Belloul M, Courbin L, Panizza P (2011) Droplet traffic regulated by
collisions in microfluidic networks. Soft Matter 7:9453–9458
Bremond N, Domejean H, Bibette J (2011) Propagation of drop
coalescence in a two-dimensional emulsion: a route towards
phase inversion. Phys Rev Lett 106:214502
Carrier O, Funfschilling D, Li HZ (2014) Effect of the fluid injection
configuration on droplet size in a microfluidic T junction. Phys
Rev E 89:013003
Cheung YN, Qiu H (2011) Characterization of acoustic droplet
formation in a microfluidic flow-focusing device. Phys Rev E
84(6):066310
Christopher GF, Noharuddin NN, Taylor JA, Anna SL (2008)
Experimental observations of the squeezing-to-dripping transi-
tion in T-shaped microfluidic junctions. Phys Rev E 78:036317
Christopher GF, Bergstein J, End NB, Poon M, Nguyen C, Anna SL
(2009) Coalescence and splitting of confined droplets at
microfluidic junctions. Lab Chip 9:1102–1109
Cordero ML, Gallaire F, Baroud CN (2011) Quantitative analysis of
the dripping and jetting regimes in co-flowing capillary jets.
Phys Fluids 23:094111
Cramer C, Fischer P, Windhab EJ (2004) Drop formation in a co-
flowing ambient fluid. Chem Eng Sci 59:3045–3058
Microfluid Nanofluid (2015) 18:19–27
Cybulski O, Garstecki P (2010) Dynamic memory in a microfluidic
system of droplets traveling through a simple network of
microchannels. Lab Chip 10(4):484–493
Duncanson W, Lin T, Abate AR, Seiffert S, Shah RK, Weitz DA
(2012) Microfluidic synthesis of advanced microparticles for
encapsulation and controlled release. Lab Chip 12:2135–2145
Engl W, Tachibana M, Colin A, Panizza P (2008) A droplet-based
high-throughput tubular platform to extract rate constants of
slow chemical reactions. Chem Eng Sci 63:1692–1695
Fu T, Ma Y, Funfschilling D, Li HZ (2009) Bubble formation and
breakup mechanism in a microfluidic flow-focusing device.
Chem Eng Sci 64:2392–2400
Fu T, Wu Y, Ma Y, Li HZ (2012) Droplet formation and breakup
dynamics in microfluidic flow-focusing devices: from dripping
to jetting. Chem Eng Sci 84:207–217
Fuerstman MJ, Lai A, Thurlow ME, Shevkoplyas SS, Stone HA,
Whitesides GM (2007) The pressure drop along rectangular
microchannels containing bubbles. Lab Chip 7:1479–1489
Ganan-Calvo AM, Riesco-Chueca P (2006) Jetting-dripping transition
of a liquid jet in a lower viscosity co-flowing immiscible liquid:
the minimum flow rate in flow focusing. J Fluid Mech 553:75–84
Glawdel T, Elbuken C, Ren CL (2012) Droplet formation in
microfluidic T-junction generators operating in the transitional
regime. I. Experimental observations. Phys Rev E 85:016322
He P, Kim H, Luo D, Marquez M, Cheng Z (2010) Low-frequency ac
electro-flow-focusing microfluidic emulsification. Appl Phys
Lett 96(17):174103
Ho PC, Yap YF, Nguyen NT, Chai JCK (2011) Thermally mediated
droplet formation at a microfluidic T-junction. Micro Nanosyst
3(1):65–75
Hoang DA, Portela LM, Kleijn CR, Kreutzer MT, van Steijn V (2013)
Dynamics of droplet breakup in a T-junction. J Fluid Mech
717:R4
Hou J, Qian G, Zhou X (2011) Gas-liquid mixing in a multi-scale
micromixer with arborescence structure. Chem Eng J
167:475–482
Jullien MC, Tsang Mui Ching MJ, Cohen C, Menetrier L, Tabeling P
(2009) Droplet breakup in microfluidic T-junctions at small
capillary numbers. Phys Fluids 21:072001
Kim H, Luo D, Link D, Weitz DA, Marquez M, Cheng Z (2007)
Controlled production of emulsion drops using an electric field
in a flow-focusing microfluidic device. Appl Phys Lett
91(13):133106
Leshansky AM, Pismen LM (2009) Breakup of drops in a microflu-
idic T junction. Phys Fluids 21:023303
Leshansky AM, Afkhami S, Jullien MC, Tabeling P (2012)
Obstructed breakup of slender drops in a microfluidic T junction.
Phys Rev Lett 108:264502
Li X-B, Li F-C, Yang J-C, Kinoshita H, Oishi M, Oshima M (2011)
Study on the mechanism of droplet formation in T-junction
microchannel. Chem Eng Sci 69:340–351
Link DR, Anna SL, Weitz DA, Stone HA (2004) Geometrically
mediated breakup of drops in microfluidic devices. Phys Rev
Lett 92:054503
Liu J, Tan S-H, Yap YF, Ng MY, Nguyen N-T (2011a) Numerical
and experimental investigations of the formation process of
ferrofluid droplets. Microfluid Nanofluid 11:177–187
Liu J, Yap YF, Nguyen NT (2011b) Numerical study of the formation
process of ferrofluid droplets. Phys Fluids 23:072008
27
Mefford OT, Woodward RC, Goff JD, Vadala TP, St Pierre TG,
Dailey JP, Riffle JS (2007) Field-induced motion of ferrofluids
through immiscible viscous media: Testbed for restorative
treatment of retinal detachment. J Magn Magn Mater
311:347–353
Nguyen NT (2012) Micro-magnetofluidics: interactions between
magnetism and fluid flow on the microscale. Microfluid Nano-
fluid 12:1–16
Pamme N (2006) Magnetism and microfluidics. Lab Chip 6:24–38
Salkin L, Schmit A, Courbin L, Panizza P (2013) Passive breakups of
isolated drops and one-dimensional assemblies of drops in
microfluidic geometries: experiments and models. Lab Chip
13:3022–3032
Samie M, Salari A, Shafii MB (2013) Breakup of microdroplets in
asymmetric T junctions. Phys Rev E 87:053003
Shao N, Gavriilidis A, Angeli P (2010) Mass transfer during Taylor
flow in microchannels with and without chemical reaction. Chem
Eng J 160:873–881
Steegmans MLJ, de Ruiter J, Schroen KGPH, Boom RM (2010) A
descriptive force-balance model for droplet formation at micro-
fluidic Y-junctions. AIChE J 56:2641–2649
Tan SH, Nguyen NT (2011) Generation and manipulation of
monodispersed ferrofluid emulsions: the effect of a uniform
magnetic field in flow-focusing and T-junction configurations.
Phys Rev E 84:036317
Tan S-H, Nguyen N-T, Yobas L, Kang TG (2010) Formation and
manipulation of ferrofluid droplets at a microfluidic T-junction.
J Micromech Microeng 20:045004
Tan J, Lu YC, Xu JH, Luo GS (2012) Mass transfer performance of
gas-liquid segmented flow in microchannels. Chem Eng J
181–182:229–235
Utada AS, Fernandez-Nieves A, Stone HA, Weitz DA (2007)
Dripping to jetting transitions in coflowing liquid streams. Phys
Rev Lett 99:094502
van Hoeve W, Dollet B, Versluis M, Lohse D (2011) Microbubble
formation and pinch-off scaling exponent in flow-focusing
devices. Phys Fluids 23:092001
Wang K, Lu YC, Xu JH, Tan J, Luo GS (2011) Generation of
micromonodispersed droplets and bubbles in the capillary
embedded T-junction microfluidic devices. AIChE J 57:299–306
Wang K, Lu Y, Yang L, Luo G (2013) Microdroplet coalescences at
microchannel junctions with different collision angles. AIChE J
59:643–649
Ward T, Faivre M, Stone HA (2010) Drop production and tip-
streaming phenomenon in a microfluidic flow-focusing device
via an interfacial chemical reaction. Langmuir 26:9233–9239
Wu Y, Fu T, Zhu C, Lu Y, Ma Y, Li HZ (2012) Asymmetrical
breakup of bubbles at a microfluidic T-junction divergence:
feedback effect of bubble collision. Microfluid Nanofluid
13:723–733
Wu Y, Fu T, Ma Y, Li HZ (2013) Ferrofluid droplet formation and
breakup dynamics in a microfluidic flow-focusing device. Soft
Matter 9:9792–9798
Wu Y, Fu T, Zhu C, Ma Y, Li HZ (2014) Bubble coalescence at a
microfluidic T-junction convergence: from colliding to squeez-
ing. Microfluid Nanofluid 16:275–286
123