Chapter 3
Simulating Droplet Microfluidic
Networks
In order to design a droplet microfluidic network, a huge number of parameters
have to be considered, which finally have to implement the desired functionality.
This results in a complex task as design parameters often depend on and affect
each other. In order to handle this complex task, models and simulation methods
can be employed in the corresponding design process. These models and simulation
methods allow for deriving the design, for validating the functionality of the design,
and for exploring alternative designs.
Throughout the design process, the objectives which have to be fulfilled by the
design are getting more and more precise. For example, in early stages of the design
process, basic objectives have to be considered, e.g., that the microfluidic channel
network ensures that the droplets flow through the desired paths. Later in the design
process, more details have to be considered, e.g., that droplet contents are properly
mixed. In order to incorporate that in the microfluidic design, means of abstraction
levels are required—in early stages higher abstractions and in later stages lower
abstractions.
Since the automatic design methods proposed in this book address design tasks
early in the design process, a high abstraction level is of particular interest in the
following. Here, especially the one-dimensional (1D) analysis model [93, 101] is
suited as it allows for deriving the design, for checking intermediate results, and for
identifying possible errors as early as possible (i.e., even before a physical design is
drawn or a physical prototype is fabricated and experiments are conducted).
However, state-of-the-art simulation tools on the 1D analysis model come
with severe shortcomings, which prevent their utilization for practically relevant
applications. More precisely, they are often not dedicated to droplet microfluidics,
cannot handle the required physical phenomena, are not publicly available, and
can hardly be extended. To address these shortcomings, this chapter introduces
an advanced simulation framework (based on [52]), which, eventually, allows to
simulate practically relevant applications. To this end, the proposed simulation
© Springer Nature Switzerland AG 2020
A. Grimmer, R. Wille, Designing Droplet Microfluidic Networks,
https://doi.org/10.1007/978-3-030-20713-7_3
21
Simulating Droplet Microfluidic
Networks
In order to design a droplet microfluidic network, a huge number of parameters
have to be considered, which finally have to implement the desired functionality.
This results in a complex task as design parameters often depend on and affect
each other. In order to handle this complex task, models and simulation methods
can be employed in the corresponding design process. These models and simulation
methods allow for deriving the design, for validating the functionality of the design,
and for exploring alternative designs.
Throughout the design process, the objectives which have to be fulfilled by the
design are getting more and more precise. For example, in early stages of the design
process, basic objectives have to be considered, e.g., that the microfluidic channel
network ensures that the droplets flow through the desired paths. Later in the design
process, more details have to be considered, e.g., that droplet contents are properly
mixed. In order to incorporate that in the microfluidic design, means of abstraction
levels are required—in early stages higher abstractions and in later stages lower
abstractions.
Since the automatic design methods proposed in this book address design tasks
early in the design process, a high abstraction level is of particular interest in the
following. Here, especially the one-dimensional (1D) analysis model [93, 101] is
suited as it allows for deriving the design, for checking intermediate results, and for
identifying possible errors as early as possible (i.e., even before a physical design is
drawn or a physical prototype is fabricated and experiments are conducted).
However, state-of-the-art simulation tools on the 1D analysis model come
with severe shortcomings, which prevent their utilization for practically relevant
applications. More precisely, they are often not dedicated to droplet microfluidics,
cannot handle the required physical phenomena, are not publicly available, and
can hardly be extended. To address these shortcomings, this chapter introduces
an advanced simulation framework (based on [52]), which, eventually, allows to
simulate practically relevant applications. To this end, the proposed simulation
© Springer Nature Switzerland AG 2020
A. Grimmer, R. Wille, Designing Droplet Microfluidic Networks,
https://doi.org/10.1007/978-3-030-20713-7_3
21