A novel pressure-fluctuation-based method is proposed for measuring the size of microdroplets without the need for visualization through a microscope. In the present work, numerical simulations are carried out in a co-flow geometry to verify this concept. First, the droplet formation frequency is determined by applying the fast Fourier transform to measured pressure fluctuation data with respect to time at any point on the outer wall. Then, the size of dispersed phase microdroplets is determined using a relationship between dispersed-phase flow rate and the droplet formation frequency. The droplet size obtained using the pressure fluctuation method is compared with that from the volume fraction method, and it is found that the error is less than 5%. The deviation is attributed to the formation of satellite droplets in the simulations. The relationship between the nondimensional parameters flow-rate ratio, capillary number, and normalized droplet diameter is investigated systematically, and empirical relations are obtained through power-law regression. The effects of interfacial tension, flow-rate ratio, and viscosity ratio on the magnitude of pressure oscillations and the corresponding droplet size are studied. All the parameters are found to have significant effects on droplet size. The ability of the proposed method to predict microdroplet size is significant with regard to potential applications to biomedical systems and drug delivery.

Generating microdroplets has become an essential task for applications such as drug delivery,1 chemical engineering,2 biomedicine,3 micro-mixing,4 and research in multiple disciplines.5–7 In drug delivery, maintaining a balance between speed of droplet production and precision of droplet volume is essential for releasing the required amount of drug, and extensive research is going on in this area to improve treatment efficiency.8,9 Conventional techniques combining high-resolution imaging with advanced image processing enable precise analysis of droplet behavior and properties within microfluidic systems for various applications in fields such as cosmetics,10 flow cytometry,11 and other areas of biology and medicine.12 

Even though microfluidic experiments are effective in the biomedical field, detecting droplet size is the most time-consuming process. Therefore, researchers now prefer to employ manual measurements together with software such as Image J because of its simplicity of use.13 In the case of a ferrofluid, the effects of an applied magnetic field on the fluid flow field during droplet formation and on the final droplet size have been investigated both numerically and experimentally.14 A method for detection and quantification of droplets has been proposed based on measurements of impedance changes induced by droplets in microstrip transmission lines or thin film electrodes.15,16

A deep-learning method, Segfit, based on a convolutional neural network (CNN), has been proposed as an efficient method for rapid and accurate analysis of microfluidic droplet size.17 Another deep-learning application, DropTrack, provides rapid and automated monitoring of the size distributions and structure of emulsions in microfluidic systems by tracking droplets in a fast-moving series of images.18 Through the use of an adaptive neural fuzzy inference system (ANFIS), it is possible to analyze complex correlations between input and output parameters to predict droplet sizes.19 In this application, the ANFIS model is trained and validated using experimental data obtained under different conditions to determine the droplet size inside a microdevice. A combination of numerical techniques and artificial neural networks has been utilized for the generation and analysis of droplet size.20 A machine learning algorithm has also been developed to increase droplet production efficiency by applying an automated droplet tracking technique to microfluidics videos, thereby improving the accuracy in achieving desired droplet sizes.21,22

Determining emulsion droplet size is crucial in a number of biomedical applications, since droplet size has a direct influence on kinetics and dynamics, thus affecting the efficiency of biological interactions and drug delivery.23 The size of the droplets in a microemulsion system determines its stability and effectiveness for a wide range of applications.24 

A method has been proposed for determining the pressure drop across droplets using an optical tweezer to monitor the displacement of beads.25 The squeezing mechanism of droplets in a T-junction was investigated experimentally by analyzing pressure oscillations created in the continuous-phase fluid due to blockage of the channel by dispersed-phase fluid during droplet formation.26 Similar experiments were performed in a flow-focusing channel, and it was found that the pressure fluctuations were nearly identical to those in the T-junction.27 Directly investigation of pressure fluctuations during drop production in a co-flow microfluidic system has provided essential insights with important implications for microfluidic device design in various domains, including drug delivery, chemical synthesis, and biotechnology.28 

In the present work, we simulate the behavior of uniform-sized droplets in a co-flow geometry, and we explore a pressure-fluctuation-based method for determining droplet size and the effects of parameters including capillary number, flow-rate ratio, density ratio, and viscosity ratio.

The remainder of the paper is structured as follows. First, we establish the geometry for the numerical simulations together with a mathematical model. A mesh-independence study is performed, followed by the simulations themselves. The droplet size determined by our pressure-fluctuation method is then compared with that from the volume fraction method.

As shown in Fig. 1, we construct a co-flow 2D axisymmetric geometry in COMSOL Multiphysics software. The lengths and radii of the outer and inner tubes are denoted by L1, L2, r1, and r2, respectively. The outer and inner tubes contain continuous-phase (CP) and dispersed-phase (DP) fluids, respectively. The flow is assumed to be isothermal and incompressible. The effect of gravity is considered insignificant, owing to the micrometer scale of the vertical gaps in these microchannels.

FIG. 1.
Schematic representation of axisymmetric co-flow geometry created in COMSOL Multiphysics.
View largeDownload slide

Schematic representation of axisymmetric co-flow geometry created in COMSOL Multiphysics.

FIG. 1.
Schematic representation of axisymmetric co-flow geometry created in COMSOL Multiphysics.
View largeDownload slide

Schematic representation of axisymmetric co-flow geometry created in COMSOL Multiphysics.

Close modal

For droplet generation and tracking of the interface, various methods are available, such as the volume of fluid (VOF) method29 and the level set method (LSM).30 In the present work, the LSM is used for droplet formation and the results are compared with those of the VOF method.31 

The governing equations for velocity and phase fields in the LSM are given below under the assumptions that the fluids are incompressible, Newtonian, and have a thin interface at the two-phase boundary. Thus, both the continuous and dispersed phases are considered here as immiscible and incompressible fluids.

The mathematical equations governing the behavior of the fluids are determined on the basis of mass and momentum conservation, respectively, as follows:32 
(1)
(2)
where u is the velocity vector, ρ(ϕ) is the density, t is the time, p is the pressure, τ is the shear stress, Fσ is the interfacial tension force, and Fg is the force due to gravity (which, as already mentioned, is neglected in the microfluidic channels). The shear stress τ is related to the shear rate, depending on the type of fluid. In this simulation, the fluids are considered to be Newtonian fluid, and so the shear stress is linearly proportional to the shear rate, with the proportionality constant being the viscosity. The interfacial force Fσ depends on the surface tension and the radius of curvature of the interface. In this study, it is taken to be proportional to the surface tension and inversely proportional to the radius of curvature. The latter is determined from the second and first derivatives of the interface shape.
The physical properties of the continuous- and dispersed-phase fluids are described by the following equations:
(3)
(4)
The motion of the interface between the two liquids is controlled by the conservative level set method (CLSM), employing the Lagrangian approach in the following manner:32 
(5)
where ɛls is the interface thickness controlling parameter, which determines the width over which the level set function ϕ transitions from one phase to another, and γ is a numerical stabilization parameter used to control the stability and convergence properties of the numerical scheme.
The flow-rate ratio is defined as the ratio of dispersed-phase and continuous-phase flow rates:
(6)
The normalized droplet diameter is defined as the ratio of droplet diameter to outer channel diameter:
(7)
The continuous-phase fluid capillary number is given by
(8)
The discrete fast Fourier transform formula used to find the droplet formation frequency is given by
(9)
The droplet diameter is determined using the following formula:
(10)
where d2, vd, Qd, and f are the inner tube diameter, dispersed-phase velocity, dispersed-phase flow rate, and droplet formation frequency, respectively.

The fluid and flow properties used in the simulations are given in Table I.

TABLE I.

Properties of fluids and flow used in simulations.

ParameterValue
Dynamic viscosity of continuous phase μc 0.067 Pa s 
Dynamic viscosity of dispersed phase μd 0.001 Pa s 
Density of continuous phase ρc 916 kg/m3 
Density of dispersed phase ρd 998.2 kg/m3 
Surface tension coefficient σ 0.05 N/m 
Wetted angle θw π/2 rad 
Velocity of continuous phase (constant) Vc 0.1 m/s 
Velocity of dispersed phase Vd 0.05, 0.1, 0.15, 0.2 m/s 
Radius of outer tube r1 50 µm 
Radius of inner tube r2 15 µm 
Length of outer tube L1 800 µm 
Length of inner tube L2 100 µm 
ParameterValue
Dynamic viscosity of continuous phase μc 0.067 Pa s 
Dynamic viscosity of dispersed phase μd 0.001 Pa s 
Density of continuous phase ρc 916 kg/m3 
Density of dispersed phase ρd 998.2 kg/m3 
Surface tension coefficient σ 0.05 N/m 
Wetted angle θw π/2 rad 
Velocity of continuous phase (constant) Vc 0.1 m/s 
Velocity of dispersed phase Vd 0.05, 0.1, 0.15, 0.2 m/s 
Radius of outer tube r1 50 µm 
Radius of inner tube r2 15 µm 
Length of outer tube L1 800 µm 
Length of inner tube L2 100 µm 
View Large

After the geometry has been constructed in COMSOL Multiphysics, the various physical parameters, such as water as the dispersed-phase fluid and oil as the continuous-phase fluid, and the inlet and outlet boundary conditions are assigned as shown in Fig. 1. The mesh is then generated, as shown in Fig. 2. Before applying the finite element method to obtain the solution for the phase field ϕ, it is necessary to perform a mesh-independence study. A previous mesh independence study in the case of droplet generation was done using the droplet size as the key parameter.33 

FIG. 2.
Mesh generated using COMSOL Multiphysics.
View largeDownload slide

Mesh generated using COMSOL Multiphysics.

FIG. 2.
Mesh generated using COMSOL Multiphysics.
View largeDownload slide

Mesh generated using COMSOL Multiphysics.

Close modal

The mesh size, which is the distance between two neighboring nodes, is varied from 8 to 4 µm as shown on the horizontal axis of Fig. 3. It is found that the droplet size determined using the volume fraction method does not vary once the mesh size is 5 µm or less. Hence, a 5 µm mesh size is considered for all further simulations. The mesh used here is of tetrahedral shape, which offers better solution convergence with the COMSOL Multiphysics solver using the finite element method. To study 3D droplet formation, the simplest geometry is chosen, namely, a coaxial and axisymmetric one. When the solution is rotated about the central axis, this corresponds to 3D or axisymmetric droplet formation. Since parameters such as velocity and pressure variations can be considered insignificant in the θ direction of a circular channel, in the current work, a 2D axisymmetric geometry is adopted instead of a 3D geometry, to reduce computational time. In future work, fully 3D simulations will be carried out to examine droplet non-sphericity.

FIG. 3.
Mesh-independence study.
View largeDownload slide

Mesh-independence study.

FIG. 3.
Mesh-independence study.
View largeDownload slide

Mesh-independence study.

Close modal

First, the 2D axisymmetric model is established, and parameters are assigned accordingly. The droplets are generated by the co-flow method at different capillary numbers of the continuous phase. The droplet formation frequency with respect to the continuous-phase capillary number is compared with the results of Wu et al.,34 as shown in Fig. 4. Wu et al. applied a constant dispersed-phase inlet velocity and varied the continuous-phase velocity in different simulation runs. In the present work, we vary the dispersed-phase inlet velocity and keep the continuous-phase inlet velocity constant. Nevertheless, the nondimensional numbers, such as capillary numbers, are matched between the studies. From the comparison, it can be concluded that out present approach of generating droplets in simulations using the level-set method (LSM) is valid and appropriate for further study with different parameters.

FIG. 4.
Validation of present simulation work with data from Wu et al.34
View largeDownload slide

Validation of present simulation work with data from Wu et al.34 

FIG. 4.
Validation of present simulation work with data from Wu et al.34
View largeDownload slide

Validation of present simulation work with data from Wu et al.34 

Close modal

The flow chart in Fig. 5 gives an overview of the procedure employed to determine droplet size and empirical relations between flow-rate ratio, capillary number, and normalized droplet size.

FIG. 5.
Flow chart illustrating the steps in determining droplet size by the pressure fluctuation method and the dependence of droplet size on flow rate and capillary number.
View largeDownload slide

Flow chart illustrating the steps in determining droplet size by the pressure fluctuation method and the dependence of droplet size on flow rate and capillary number.

FIG. 5.
Flow chart illustrating the steps in determining droplet size by the pressure fluctuation method and the dependence of droplet size on flow rate and capillary number.
View largeDownload slide

Flow chart illustrating the steps in determining droplet size by the pressure fluctuation method and the dependence of droplet size on flow rate and capillary number.

Close modal

Droplets are generated for different flow-rate ratios, and a visual representation of this process is shown in Fig. 6, where the dispersed-phase fluid (gray) is surrounded by continuous-phase fluid (blue).

FIG. 6.
Visual representation of droplet formation for various flow rates: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

Visual representation of droplet formation for various flow rates: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

FIG. 6.
Visual representation of droplet formation for various flow rates: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

Visual representation of droplet formation for various flow rates: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

Close modal

In the present work, since we are determining the droplet size on the basis of the pressure exerted on the wall, we will first examine the reason why pressure fluctuations occur on the wall during the passage of a droplet through the channel. As shown in the inset of Fig. 7, for each flow-rate ratio, there are two pressure dips, with the smaller dip corresponding to the passage of the droplet through the cross-section at the pressure measuring point and the larger dip corresponding to the pressure recorded at the same cross-section during the exit of the same droplet from the channel.

FIG. 7.
Pressure fluctuations on the wall at a point 600 μm from the channel entrance measured during droplet passage: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

Pressure fluctuations on the wall at a point 600 μm from the channel entrance measured during droplet passage: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

FIG. 7.
Pressure fluctuations on the wall at a point 600 μm from the channel entrance measured during droplet passage: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

Pressure fluctuations on the wall at a point 600 μm from the channel entrance measured during droplet passage: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

Close modal

However, the passage of a single droplet from its formation point to the channel exit results in a set of smaller and larger dips in pressure. The peak phase of each pressure curve corresponds to the passage of continuous-phase fluid at the exit of the channel, while the trough corresponds to the presence of dispersed-phase droplets. The FFT of the pressure–time results provides the details of the number of droplets passing through the microchannel in unit time.

The inset of Fig. 7 shows that the number of pressure fluctuations occurring on the wall increases as the flow-rate ratio increases from Q1 to Q4, indicating an increase in the droplet formation frequency. Pressure measurements in the upstream and downstream directions of the channel give the high and low magnitudes of the pressure fluctuations, which enables an experimentalist to adjust the pressure measuring point according to the operating range of the available pressure sensor.

An FFT is applied to the measured pressure–time data, and the droplet formation frequency is calculated. The droplet formation frequencies corresponding to the pressure oscillations are found to be 950, 1250, 1350, and 1400 s−1 for flow-rate ratios of 0.5, 1, 1.5, and 2, respectively, as shown in Fig. 8. It can be observed that an increase in the dispersed-phase flow rate at constant continuous-phase flow rate results in a higher droplet formation frequency.

FIG. 8.
FFT analysis of pressure fluctuations for various flow-rate ratios: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

FFT analysis of pressure fluctuations for various flow-rate ratios: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

FIG. 8.
FFT analysis of pressure fluctuations for various flow-rate ratios: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.
View largeDownload slide

FFT analysis of pressure fluctuations for various flow-rate ratios: (a) Q1 = 0.5; (b) Q2 = 1; (c) Q3 = 1.5; (d) Q4 = 2.

Close modal

In our approach, the pressure measured near the exit of the microchannel is recorded continuously with a very small time interval of the order of a fraction of a millisecond and then subjected to an FFT. The droplet formation frequency (DFF) can thus be obtained from measurements of the pressure signal itself, as the frequency of maximum amplitude in the FFT, and therefore it can be determined without any need to capture droplet images using a high-resolution microscope and high-speed camera.

The pressure fluctuation method uses the relationship between the dispersed-phase flow rate and DFF given by Eq. (10) to accurately determine the size of the droplets produced in the microfluidic system. The droplet diameter is found to be 41.41, 47.62, 53.13, and 57.77 μm for flow-rate ratios of 0.5, 1, 1.5, and 2, respectively. The droplet diameter is also determined by the volume fraction method in COMSOL Multiphysics. As shown in Fig. 9, the distance between the two opposite extreme points on the interface between dispersed- and continuous-phase volumes in any direction will give the actual diameter of the dispersed-phase volume. For flow-rate ratios of 0.5, 1, 1.5, and 2, the droplet diameter according to the volume fraction method is found to be 39.71, 45.44, 50.77, and 55.44 μm, respectively.

FIG. 9.
Measurement of droplet diameter by the volume fraction method from the LSM in a COMSOL Multiphysics simulation for a flow-rate ratio Q1 = 0.5.
View largeDownload slide

Measurement of droplet diameter by the volume fraction method from the LSM in a COMSOL Multiphysics simulation for a flow-rate ratio Q1 = 0.5.

FIG. 9.
Measurement of droplet diameter by the volume fraction method from the LSM in a COMSOL Multiphysics simulation for a flow-rate ratio Q1 = 0.5.
View largeDownload slide

Measurement of droplet diameter by the volume fraction method from the LSM in a COMSOL Multiphysics simulation for a flow-rate ratio Q1 = 0.5.

Close modal

A comparison between the droplet diameters obtained respectively by the pressure fluctuation method and the volume fraction method is presented in Fig. 10. It can be seen that that the droplet diameter determined by the pressure fluctuation method has slightly higher values than that from the volume fraction method. This discrepancy can be attributed to the formation of satellite droplets along with the primary droplets during droplet generation in the simulations. The droplet size determined by the pressure fluctuation method accounts for the overall volume of droplets, including satellite droplets. By contrast, in the volume fraction method, only the size of the primary droplets is measured. Except for this slight deviation, our analysis suggests that the proposed pressure fluctuation method for determining droplet size is valid.

FIG. 10.
Comparison of droplet sizes determined by the pressure fluctuation method and by the volume fraction method directly from the LSM.
View largeDownload slide

Comparison of droplet sizes determined by the pressure fluctuation method and by the volume fraction method directly from the LSM.

FIG. 10.
Comparison of droplet sizes determined by the pressure fluctuation method and by the volume fraction method directly from the LSM.
View largeDownload slide

Comparison of droplet sizes determined by the pressure fluctuation method and by the volume fraction method directly from the LSM.

Close modal

The consistency between the two methods suggests that the pressure fluctuation technique can be adopted as a practical approach for experimentalist to use in determining droplet size.

The effect of two nondimensional parameters, namely, the flow-rate ratio and the capillary number of the continuous phase, on the droplet size is analyzed. From Fig. 11(a), it can be observed that an increase in flow-rate ratio at constant capillary number leads to an increase in the normalized droplet size. This indicates a proportional relationship between normalized droplet size and flow-rate ratio. It is also noted that as the capillary number decreases, the droplet size increases because of an increase in interfacial tension between the immiscible fluids. The effect of the continuous-phase capillary number on normalized droplet size at a constant flow-rate ratio is shown in Fig. 11(b), from which it can be seen that there is an inverse relationship between normalized droplet size and continuous-phase capillary number, with the droplet size decreasing as the capillary number increases. It can again be noted that the droplet size increases with increasing flow-rate ratio.

FIG. 11.
Variation of normalized droplet diameter dr with (a) flow-rate ratio Qr and (b) capillary number of continuous phase Cac.
View largeDownload slide

Variation of normalized droplet diameter dr with (a) flow-rate ratio Qr and (b) capillary number of continuous phase Cac.

FIG. 11.
Variation of normalized droplet diameter dr with (a) flow-rate ratio Qr and (b) capillary number of continuous phase Cac.
View largeDownload slide

Variation of normalized droplet diameter dr with (a) flow-rate ratio Qr and (b) capillary number of continuous phase Cac.

Close modal
FIG. 12.
Power-law regression fit to analyze the effect of flowrate ratio Qr on normalized droplet size dr at constant continuous-phase capillary numbers Cac of (a) 0.134, (b) 0.067, (c) 0.045, and (d) 0.0382.
View largeDownload slide

Power-law regression fit to analyze the effect of flowrate ratio Qr on normalized droplet size dr at constant continuous-phase capillary numbers Cac of (a) 0.134, (b) 0.067, (c) 0.045, and (d) 0.0382.

FIG. 12.
Power-law regression fit to analyze the effect of flowrate ratio Qr on normalized droplet size dr at constant continuous-phase capillary numbers Cac of (a) 0.134, (b) 0.067, (c) 0.045, and (d) 0.0382.
View largeDownload slide

Power-law regression fit to analyze the effect of flowrate ratio Qr on normalized droplet size dr at constant continuous-phase capillary numbers Cac of (a) 0.134, (b) 0.067, (c) 0.045, and (d) 0.0382.

Close modal

A power-law regression analysis is conducted on the dataset shown in Fig. 11(a) to find the relationship between normalized droplet size and flow-rate ratio at constant capillary number. Figure 12 presents the results of this fit.

The following power-law regression equation describing the relationship between droplet size and flow-rate ratio at constant capillary number is obtained:
(11)
i.e.,
(12)

The regression fit constants are listed in Table II. The R2 values of the fit are all almost equal to one, which indicates a strong correlation between the normalized droplet size and flow-rate ratio at constant capillary number.

TABLE II.

Power-law regression constants determined at fixed capillary numbers Cac.

σCacm1ec1R2
0.05 0.134 0.46 −0.87 0.9535 
0.1 0.067 0.22 −0.63 0.9999 
0.15 0.045 0.17 −0.54 0.9985 
0.175 0.0383 0.18 −0.50 0.9992 
σCacm1ec1R2
0.05 0.134 0.46 −0.87 0.9535 
0.1 0.067 0.22 −0.63 0.9999 
0.15 0.045 0.17 −0.54 0.9985 
0.175 0.0383 0.18 −0.50 0.9992 
View Large
A power-law regression fit is also performed on the dataset depicted in Fig. 11(b) to determine the correlation between normalized droplet size and capillary number at constant flow-rate ratios. Figure 13 presents the results of this fit. The following power-law regression equation describing the relationship between normalized droplet size and capillary number at constant flow-rate ratios is obtained:
(13)
i.e.,
(14)
FIG. 13.
Power-law regression fit to analyze the effect of continuous-phase capillary number Cac on normalized droplet size dr at constant flow-rate ratios Qr of (a) 0.5, (b) 1, (c) 1.5, and (d) 2.
View largeDownload slide

Power-law regression fit to analyze the effect of continuous-phase capillary number Cac on normalized droplet size dr at constant flow-rate ratios Qr of (a) 0.5, (b) 1, (c) 1.5, and (d) 2.

FIG. 13.
Power-law regression fit to analyze the effect of continuous-phase capillary number Cac on normalized droplet size dr at constant flow-rate ratios Qr of (a) 0.5, (b) 1, (c) 1.5, and (d) 2.
View largeDownload slide

Power-law regression fit to analyze the effect of continuous-phase capillary number Cac on normalized droplet size dr at constant flow-rate ratios Qr of (a) 0.5, (b) 1, (c) 1.5, and (d) 2.

Close modal

The regression fit constants are noted in Table III. The R2 values of the regression fit are again all almost unity.

TABLE III.

Power-law regression constants determined at fixed flow-rate ratio Qr.

Qrm2ec2R2
0.5 −0.46 −2.09 0.9143 
1 −0.28 −1.39 0.9752 
1.5 −0.19 −1.06 0.9960 
2 −0.21 −1.07 0.9668 
Qrm2ec2R2
0.5 −0.46 −2.09 0.9143 
1 −0.28 −1.39 0.9752 
1.5 −0.19 −1.06 0.9960 
2 −0.21 −1.07 0.9668 
View Large
A three-dimensional power-law regression analysis is performed to investigate the relationship between the dependent variable normalized droplet size and the independent variables flow-rate ratio and capillary number. The results are shown in Fig. 14(a). The following power-law regression equation is obtained:
(15)
i.e.,
(16)
FIG. 14.
Three-dimensional power-law regression analysis of normalized droplet diameter, flow-rate ratio, and capillary number.
View largeDownload slide

Three-dimensional power-law regression analysis of normalized droplet diameter, flow-rate ratio, and capillary number.

FIG. 14.
Three-dimensional power-law regression analysis of normalized droplet diameter, flow-rate ratio, and capillary number.
View largeDownload slide

Three-dimensional power-law regression analysis of normalized droplet diameter, flow-rate ratio, and capillary number.

Close modal

The normalized droplet size is found to be in good agreement with the flow-rate ratio and capillary number with power-law regression constant values given by m3 = 0.208 94, m4 = −0.2374, and ec3 −1.2797. The R2 value of the regression is 0.960 01. Figure 14(b) presents a side view perpendicular to the axis representing the normalized droplet size parameter, where the plane looks like a line and offers a visual representation of how closely the data points are aligned with the regression plane.

Since our proposed method is based on pressure measurements on the wall of the outer tube, the influence of the various fluid and flow parameters on the pressure measurements is investigated. The effects of pressure fluctuations and their magnitude measured on the wall during the exit of the droplet from the channel are analyzed for different fluid and flow parameters.

1. Effect of flow-rate ratio

Figure 15(a) shows the pressure exerted on the outer tube wall at a point 600 μm from the entrance for different flow-rate ratios of 0.5, 1, 1.5, and 2. From Fig. 15(b), it can be observed that the amplitude of the pressure fluctuations increases with increasing flow-rate ratio. Figure 15(c) shows that the normalized droplet diameter increases with increasing dispersed-phase flow rate at a constant continuous-phase flow rate. Figure 15(d) depicts the relationship between pressure amplitude and normalized droplet diameter at different flow-rate ratios. Here, the increase in the amplitude of pressure fluctuations indicates an increase in the volume fraction of dispersed-phase fluid.

FIG. 15.
Effect of flow-rate ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of flow-rate ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

FIG. 15.
Effect of flow-rate ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of flow-rate ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

Close modal

2. Effect of viscosity ratio

Figure 16(a) depicts the pressure fluctuations experienced by the channel wall at a particular point for various viscosity ratios in the range of 0.25–2, and Fig. 16(b) shows the corresponding variations in pressure amplitude. From Fig. 16(c) it can be seen that the droplet size is smaller at lower viscosity ratios. The reason for this is that, up to a viscosity ratio of 1, the dispersed-phase fluid with low viscosity will experience less resistance, leading to more stretching and elongation. Under this condition, the continuous-phase fluid can easily break the dispersed-phase fluid into smaller droplets. At a viscosity ratio of 1, the droplet diameter reaches its maximum value, and a further increase in viscosity ratio results in a decrease in droplet size. Figure 16(d) illustrates the relationship between droplet size and pressure amplitude at various viscosity ratios.

FIG. 16.
Effect of viscosity ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of viscosity ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

FIG. 16.
Effect of viscosity ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of viscosity ratio on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

Close modal

3. Effect of interfacial tension

The pressure variations transmitted to the channel wall at various interfacial tensions are shown in Fig. 17(a). The dependence of the pressure fluctuation amplitude on capillary number is shown in Fig. 17(b) for constant viscosity and flow-rate ratios. Figure 17(c) shows the variation of the normalized droplet diameter with the capillary number, and it can be seen that the droplet diameter decreases with increasing capillary number. Figure 17(d) illustrates the variation of the normalized droplet diameter with the pressure fluctuation amplitude.

FIG. 17.
Effect of interfacial tension as represented by the capillary number on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of interfacial tension as represented by the capillary number on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

FIG. 17.
Effect of interfacial tension as represented by the capillary number on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.
View largeDownload slide

Effect of interfacial tension as represented by the capillary number on (a) pressure fluctuations, (b) amplitude of pressure fluctuations, and (c) normalized droplet diameter. (d) Normalized droplet diameter vs pressure fluctuation amplitude.

Close modal

These results suggest that the droplet diameter is primarily influenced by the flow-rate ratio, viscosity ratio, and interfacial tension.

In the present work, we have generated micrometer-sized droplets in a 2D axisymmetric co-flow geometry using the LSM from COMSOL Multiphysics software in combination with MATLAB. We have proposed a new method to determine droplet size based on pressure fluctuations exerted on the outer wall of the channel and on the flow rate of the dispersed-phase fluid. This method has potential applications in the biomedical field, such as for drug delivery systems. It has been found that the pressure fluctuation method gives similar values for the droplet diameter as the volume fraction method, with a small deviation due to the formation of satellite droplets. The proposed method for droplet size measurement will be more useful for monodispersed droplets, emulsions, or other droplet size combinations. The effects of various flow and fluid parameters on the pressure fluctuations has been investigated. An empirical relation has been obtained to determine the droplet size on the basis of the input flow rate and capillary number. In summary, the amplitude of the pressure fluctuations is directly proportional to the flow-rate ratio, viscosity ratio, and interfacial tension between the immiscible fluids.

The authors acknowledge the IIT Tirupati for providing the COMSOL Multiphysics 6.2 software.

The authors have no conflicts to disclose.

Babajan Bakthar Khan: Conceptualization (equal); Data curation (lead); Formal analysis (lead); Investigation (equal); Methodology (equal); Validation (lead); Visualization (lead); Writing – original draft (lead); Writing – review & editing (equal). Sunil Kumar Thamida: Conceptualization (equal); Methodology (equal); Software (equal); Supervision (lead); Validation (supporting); Visualization (supporting); Writing – original draft (supporting); Writing – review & editing (equal). Anil B. Vir: Conceptualization (supporting); Formal analysis (supporting); Methodology (supporting); Software (equal); Supervision (equal); Validation (supporting); Visualization (supporting); Writing – review & editing (supporting).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
J.
Jung
 and
J.
Oh
, “
Cell-induced flow-focusing instability in gelatin methacrylate microdroplet generation
,”
Biomicrofluidics
8
(
3
),
036503
(
2014
).
https://doi.org/10.1063/1.4880375
Google Scholar
Crossref
Search ADS
PubMed
 
2.
H.
Bazyar
,
M. H.
Kandemir
,
J.
Peper
,
M. A. B.
Andrade
,
A. L.
Bernassau
,
K.
Schroën
, and
R. G. H.
Lammertink
, “
Acoustophoresis of monodisperse oil droplets in water: Effect of symmetry breaking and non-resonance operation on oil trapping behavior
,”
Biomicrofluidics
17
(
6
),
064107
(
2023
).
https://doi.org/10.1063/5.0175400
Google Scholar
Crossref
Search ADS
PubMed
 
3.
C.
Shao
,
J.
Chi
,
L.
Shang
,
Q.
Fan
, and
F.
Ye
, “
Droplet microfluidics-based biomedical microcarriers
,”
Acta Biomater.
138
,
21
–
33
(
2022
).
https://doi.org/10.1016/j.actbio.2021.10.037
Google Scholar
Crossref
Search ADS
PubMed
 
4.
H.
Song
,
M. R.
Bringer
,
J. D.
Tice
,
C. J.
Gerdts
, and
R. F.
Ismagilov
, “
Experimental test of scaling of mixing by chaotic advection in droplets moving through microfluidic channels
,”
Appl. Phys. Lett.
83
(
22
),
4664
–
4666
(
2003
).
https://doi.org/10.1063/1.1630378
Google Scholar
Crossref
Search ADS
PubMed
 
5.
Y.
He
,
Y.
Qiao
,
L.
Ding
,
T.
Cheng
, and
J.
Tu
, “
Recent advances in droplet sequential monitoring methods for droplet sorting
,”
Biomicrofluidics
17
(
6
),
061501
(
2023
).
https://doi.org/10.1063/5.0173340
Google Scholar
Crossref
Search ADS
PubMed
 
6.
G.
Yang
,
C.
Gao
,
D.
Chen
,
J.
Wang
,
X.
Huo
, and
J.
Chen
, “
Multiplex fluorescence detection of single-cell droplet microfluidics and its application in quantifying protein expression levels
,”
Biomicrofluidics
17
(
6
),
064106
(
2023
).
https://doi.org/10.1063/5.0179121
Google Scholar
Crossref
Search ADS
PubMed
 
7.
H.
Nazari
,
A.
Heirani-Tabasi
,
S.
Ghorbani
,
H.
Eyni
,
S.
Razavi Bazaz
,
M.
Khayati
,
F.
Gheidari
,
K.
Moradpour
,
M.
Kehtari
,
S. M.
Ahmadi Tafti
,
S. H.
Ahmadi Tafti
, and
M.
Ebrahimi Warkiani
, “
Microfluidic-based droplets for advanced regenerative medicine: Current challenges and future trends
,”
Biosensors
12
(
1
),
20
(
2021
).
https://doi.org/10.3390/bios12010020
Google Scholar
Crossref
Search ADS
PubMed
 
8.
F.
He
,
M. J.
Zhang
,
W.
Wang
,
Q. W.
Cai
,
Y. Y.
Su
,
Z.
Liu
,
Y.
Faraj
,
X. J.
Ju
,
R.
Xie
, and
L. Y.
Chu
, “
Designable polymeric microparticles from droplet microfluidics for controlled drug release
,”
Adv. Mater. Technol.
4
(
6
),
1800687
(
2019
).
https://doi.org/10.1002/admt.201800687
Google Scholar
Crossref
Search ADS
 
9.
R.
Riahi
,
A.
Tamayol
,
S. A. M.
Shaegh
,
A. M.
Ghaemmaghami
,
M. R.
Dokmeci
, and
A.
Khademhosseini
, “
Microfluidics for advanced drug delivery systems
,”
Curr. Opin. Chem. Eng.
7
,
101
–
112
(
2015
).
https://doi.org/10.1016/j.coche.2014.12.001
Google Scholar
Crossref
Search ADS
PubMed
 
10.
D.
Park
,
H.
Kim
, and
J. W.
Kim
, “
Microfluidic production of monodisperse emulsions for cosmetics
,”
Biomicrofluidics
15
(
5
),
51302
(
2021
).
https://doi.org/10.1063/5.0057733
Google Scholar
Crossref
Search ADS
 
11.
T.
Julian
,
T.
Tang
,
Y.
Hosokawa
, and
Y.
Yalikun
, “
Machine learning implementation strategy in imaging and impedance flow cytometry
,”
Biomicrofluidics
17
(
5
),
051506
(
2023
).
https://doi.org/10.1063/5.0166595
Google Scholar
Crossref
Search ADS
PubMed
 
12.
M.
Zantow
,
R.
Dendere
, and
T. S.
Douglas
, “
Image-based analysis of droplets in microfluidics
,” in Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS) (
IEEE
,
2013
), pp.
1776
–
1779
.
Google Scholar
Crossref
Search ADS
 
13.
A. S.
Basu
, “
Droplet morphometry and velocimetry (DMV): A video processing software for time-resolved, label-free tracking of droplet parameters
,”
Lab Chip
13
(
10
),
1892
–
1901
(
2013
).
https://doi.org/10.1039/c3lc50074h
Google Scholar
Crossref
Search ADS
PubMed
 
14.
J.
Liu
,
S. H.
Tan
,
Y. F.
Yap
,
M. Y.
Ng
, and
N. T.
Nguyen
, “
Numerical and experimental investigations of the formation process of ferrofluid droplets
,”
Microfluid. Nanofluid.
11
(
2
),
177
–
187
(
2011
).
https://doi.org/10.1007/s10404-011-0784-7
Google Scholar
Crossref
Search ADS
 
15.
J.
de Novais Schianti
,
A. L. C.
Serrano
,
D.
Orquiza de Carvalho
,
R. A.
Penchel
,
J.
Mota Pinheiro
,
M. R.
Gongora-Rubio
, and
G.
Pamplona Rehder
, “
Novel platform for droplet detection and size measurement using microstrip transmission lines
,”
Sensors
19
(
23
),
5216
(
2019
).
https://doi.org/10.3390/s19235216
Google Scholar
Crossref
Search ADS
PubMed
 
16.
E. V.
Moiseeva
,
A. A.
Fletcher
, and
C. K.
Harnett
, “
Thin-film electrode based droplet detection for microfluidic systems
,”
Sens. Actuators, B
155
(
1
),
408
–
414
(
2011
).
https://doi.org/10.1016/j.snb.2010.11.028
Google Scholar
Crossref
Search ADS
 
17.
S.
Zhang
,
X.
Liang
,
X.
Huang
,
K.
Wang
, and
T.
Qiu
, “
Precise and fast microdroplet size distribution measurement using deep learning
,”
Chem. Eng. Sci.
247
,
116926
(
2022
).
https://doi.org/10.1016/j.ces.2021.116926
Google Scholar
Crossref
Search ADS
 
18.
M.
Durve
,
S.
Orsini
,
A.
Tiribocchi
,
A.
Montessori
,
J. M.
Tucny
,
M.
Lauricella
,
A.
Camposeo
,
D.
Pisignano
, and
S.
Succi
, “
Measuring arrangement and size distributions of flowing droplets in microchannels through deep learning using DropTrack
,”
Phys. Fluids
36
(
2
),
022105
(
2024
).
https://doi.org/10.1063/5.0185350
Google Scholar
Crossref
Search ADS
 
19.
S.
Mottaghi
,
M.
Nazari
,
S. M.
Fattahi
,
M.
Nazari
, and
S.
Babamohammadi
, “
Droplet size prediction in a microfluidic flow focusing device using an adaptive network based fuzzy inference system
,”
Biomed. Microdevices
22
(
3
),
61
(
2020
).
https://doi.org/10.1007/s10544-020-00513-4
Google Scholar
Crossref
Search ADS
PubMed
 
20.
S.
Naji
,
A.
Rahimi
,
V.
Bazargan
, and
M.
Marengo
, “
Numerical and artificial neural network analysis of an axisymmetric co-flow-focusing microfluidic droplet generator using active and passive control
,”
Phys. Fluids
35
(
6
),
062008
(
2023
).
https://doi.org/10.1063/5.0147790
Google Scholar
Crossref
Search ADS
 
21.
M.
Durve
,
A.
Tiribocchi
,
F.
Bonaccorso
,
A.
Montessori
,
M.
Lauricella
,
M.
Bogdan
,
J.
Guzowski
, and
S.
Succi
, “
DropTrack—Automatic droplet tracking with YOLOv5 and DeepSORT for microfluidic applications
,”
Phys. Fluids
34
(
8
),
082003
(
2022
).
https://doi.org/10.1063/5.0097597
Google Scholar
Crossref
Search ADS
 
22.
C.
Barnes
,
A. R.
Sonwane
,
E. C.
Sonnenschein
, and
F.
Del Giudice
, “
Machine learning enhanced droplet microfluidics
,”
Phys. Fluids
35
(
9
),
092003
(
2023
).
https://doi.org/10.1063/5.0163806
Google Scholar
Crossref
Search ADS
 
23.
P.
Speranza
,
A. P.
Badan Ribeiro
,
R. L.
Cunha
,
J. A.
Macedo
, and
G. A.
Macedo
, “
Influence of emulsion droplet size on antimicrobial activity of interesterified Amazonian oils
,”
LWT–Food Sci. Technol.
60
(
1
),
207
–
212
(
2015
).
https://doi.org/10.1016/j.lwt.2014.07.022
Google Scholar
Crossref
Search ADS
 
24.
V.
Califano
,
R.
Calabria
, and
P.
Massoli
, “
Experimental evaluation of the effect of emulsion stability on micro-explosion phenomena for water-in-oil emulsions
,”
Fuel
117
(
Part A
),
87
–
94
(
2014
).
https://doi.org/10.1016/j.fuel.2013.08.073
Google Scholar
Crossref
Search ADS
 
25.
Y.
Jin
,
A.
Orth
,
E.
Schonbrun
, and
K. B.
Crozier
, “
Measuring the pressures across microfluidic droplets with an optical tweezer
,”
Opt. Express
20
(
22
),
24450
–
24464
(
2012
).
https://doi.org/10.1364/oe.20.024450
Google Scholar
Crossref
Search ADS
PubMed
 
26.
A. R.
Abate
,
P.
Mary
,
V.
Van Steijn
, and
D. A.
Weitz
, “
Experimental validation of plugging during drop formation in a T-junction
,”
Lab Chip
12
(
8
),
1516
–
1521
(
2012
).
https://doi.org/10.1039/c2lc21263c
Google Scholar
Crossref
Search ADS
PubMed
 
27.
P. A.
Romero
 and
A. R.
Abate
, “
Flow focusing geometry generates droplets through a plug and squeeze mechanism
,”
Lab Chip
12
(
24
),
5130
–
5132
(
2012
).
https://doi.org/10.1039/c2lc40938k
Google Scholar
Crossref
Search ADS
PubMed
 
28.
K.
Xu
,
C. P.
Tostado
,
J. H.
Xu
,
Y. C.
Lu
, and
G. S.
Luo
, “
Direct measurement of the differential pressure during drop formation in a co-flow microfluidic device
,”
Lab Chip
14
(
7
),
1357
–
1366
(
2014
).
https://doi.org/10.1039/c3lc51222c
Google Scholar
Crossref
Search ADS
PubMed
 
29.
R.
Scardovelli
 and
S.
Zaleski
, “
Direct numerical simulation of free-surface and interfacial flow
,”
Annu. Rev. Fluid. Mech.
31
,
567
–
603
(
1999
).
https://doi.org/10.1146/annurev.fluid.31.1.567
Google Scholar
Crossref
Search ADS
 
30.
S.
Osher
 and
J. A.
Sethian
, “
Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton-Jacobi formulations
,”
J. Comput. Phys.
79
(
1
),
12
–
49
(
1988
).
https://doi.org/10.1016/0021-9991(88)90002-2
Google Scholar
Crossref
Search ADS
 
31.
A. S.
Utada
,
A.
Fernandez-Nieves
,
H. A.
Stone
, and
D. A.
Weitz
, “
Dripping to jetting transitions in coflowing liquid streams
,”
Phys. Rev. Lett.
99
(
9
),
094502
(
2007
).
https://doi.org/10.1103/physrevlett.99.094502
Google Scholar
Crossref
Search ADS
PubMed
 
32.
A.
Venkateshwarlu
 and
R. P.
Bharti
, “
Interface evolution and pinch-off mechanism of droplet in two-phase liquid flow through T-junction microfluidic system
,”
Colloids Surf., A
642
,
128536
(
2022
).
https://doi.org/10.1016/j.colsurfa.2022.128536
Google Scholar
Crossref
Search ADS
 
33.
A.
Venkateshwarlu
 and
R. P.
Bharti
, “
Effects of capillary number and flow rates on the hydrodynamics of droplet generation in two-phase cross-flow microfluidic systems
,”
J. Taiwan Inst. Chem. Eng.
129
,
64
–
79
(
2021
).
https://doi.org/10.1016/j.jtice.2021.07.045
Google Scholar
Crossref
Search ADS
 
34.
L.
Wu
,
X.
Liu
,
Y.
Zhao
, and
Y.
Chen
, “
Role of local geometry on droplet formation in axisymmetric microfluidics
,”
Chem. Eng. Sci.
163
,
56
–
67
(
2017
).
https://doi.org/10.1016/j.ces.2017.01.022
Google Scholar
Crossref
Search ADS
 
© 2024 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).