3.3 Advanced Simulation Framework
29
Hence, as soon as any of those events occurs, the current flow state becomes invalid
and has to be re-calculated (i.e., newly added, removed, or changed resistances have
to be incorporated into the equation system which, afterwards, has to be solved
again).
Overall, by considering the droplet states and solving the obtain equation system,
all volumetric flow rates and pressure gradients in the channels can be determined
for the current droplet positions. This allows to determine the current speeds of the
droplets according to Eq. 3.7 (cf. page 26). When a droplet arrives at a bifurcation it
chooses the branch with the instantaneous highest volumetric flow rate [30, 37, 67]
and does not split (which this is true at a low Capillary number). Finally, this allows
to predict the paths of droplets.
3.3 Advanced Simulation Framework
Simulations on the 1D analysis model are perfectly suited for an early validation of
the design prior to fabrication. Simulations allow to predict (1) the droplets’ paths
through the network (this can decide which assay is executed on the droplet), (2) the
flow changes caused by all droplets and the resulting impacts (e.g., distance changes
between droplets, droplet patterns, etc.), as well as (3) the time a droplet takes to
pass through the network.
However, existing simulation solutions on the 1D analysis model are hardly used
in practice yet. This is caused by the fact that, despite their promises, currently
available simulation solutions
1. are not dedicated to microfluidics and, therefore, first require to manually map
the design to an electrical circuit as it is the case for SPICE [93],
2. target only networks which consist of channels which branch and merge [7,
40, 101] or are even limited to networks consisting of a symmetric/asymmetric
loop [4, 21, 37, 67, 105, 106, 109], but ignore essential physical phenomena such
as trapping droplets, checking whether droplets are squeezed through any gaps,
and clogging of channels,
3. are not publicly available (in fact, no tool is publicly available), and
4. are static, i.e. do not allow for further extensions which are essential in order
to support the broad range of application scenarios designers in the microfluidic
domain are faced.
In this section, an advanced simulation framework is introduced which addresses
these shortcomings. To this end, the proposed framework (1) directly works on the
specification of the design, (2) extends the current state of the art with important
physical phenomena which are required for practical designs, (3) is publicly
available at http://iic.jku.at/eda/research/microfluidics_simulation/, and, (4) due to
the availability of the source code and the event-based algorithm, can easily be
extended to support further applications.
29
Hence, as soon as any of those events occurs, the current flow state becomes invalid
and has to be re-calculated (i.e., newly added, removed, or changed resistances have
to be incorporated into the equation system which, afterwards, has to be solved
again).
Overall, by considering the droplet states and solving the obtain equation system,
all volumetric flow rates and pressure gradients in the channels can be determined
for the current droplet positions. This allows to determine the current speeds of the
droplets according to Eq. 3.7 (cf. page 26). When a droplet arrives at a bifurcation it
chooses the branch with the instantaneous highest volumetric flow rate [30, 37, 67]
and does not split (which this is true at a low Capillary number). Finally, this allows
to predict the paths of droplets.
3.3 Advanced Simulation Framework
Simulations on the 1D analysis model are perfectly suited for an early validation of
the design prior to fabrication. Simulations allow to predict (1) the droplets’ paths
through the network (this can decide which assay is executed on the droplet), (2) the
flow changes caused by all droplets and the resulting impacts (e.g., distance changes
between droplets, droplet patterns, etc.), as well as (3) the time a droplet takes to
pass through the network.
However, existing simulation solutions on the 1D analysis model are hardly used
in practice yet. This is caused by the fact that, despite their promises, currently
available simulation solutions
1. are not dedicated to microfluidics and, therefore, first require to manually map
the design to an electrical circuit as it is the case for SPICE [93],
2. target only networks which consist of channels which branch and merge [7,
40, 101] or are even limited to networks consisting of a symmetric/asymmetric
loop [4, 21, 37, 67, 105, 106, 109], but ignore essential physical phenomena such
as trapping droplets, checking whether droplets are squeezed through any gaps,
and clogging of channels,
3. are not publicly available (in fact, no tool is publicly available), and
4. are static, i.e. do not allow for further extensions which are essential in order
to support the broad range of application scenarios designers in the microfluidic
domain are faced.
In this section, an advanced simulation framework is introduced which addresses
these shortcomings. To this end, the proposed framework (1) directly works on the
specification of the design, (2) extends the current state of the art with important
physical phenomena which are required for practical designs, (3) is publicly
available at http://iic.jku.at/eda/research/microfluidics_simulation/, and, (4) due to
the availability of the source code and the event-based algorithm, can easily be
extended to support further applications.