4.2 Validating the Specification
53
Having this partial specification, it is left to define the required channels, i.e. to
properly dimension the channels. This however constitutes a challenge, since their
resistances significantly affect the flow of the droplets. In the currently applied
design process, the channels are defined based on the designer’s experience. While
doing that, designers have to take a huge number of constraints and dependencies
into consideration—already slightly changing, e.g., the specification of a single
channel may change the behavior of the entire microfluidic system. As a consequence, designers often cannot grasp all effects and dependencies anymore. More
precisely, because of improper specifications of channels, microfluidic networks
may result in which
• droplets flow against the intended direction or
• droplets pass a channel/module too slowly (critical when, e.g., a droplet just
passed a heating module and needs to be analyzed by a following detector module
without cooling down) or too quickly (e.g., when the channel is used as delay
line).
In order to formulate these problems, the 1D analysis model provided in
Sect. 3.2 is employed in the following. Based on that model, two methods which
allow designers to (1) automatically validate whether their chosen specification
indeed works as intended (i.e., avoid problems as discussed above) as well as
(2) automatically conduct the dimensioning to obtain a proper specification are
proposed in Sects. 4.2 and 4.3, respectively.
4.2 Validating the Specification
Using the 1D analysis model provided in Sect. 3.2 as basis, this section describes a
method which automatically checks whether a given specification of a microfluidic
network shows any of the problems discussed above. To describe the method,
recall that, as discussed in Sect. 4.1, the used pump, droplet generators, sorters, and
modules are predefined by the experiment and the designer is mainly confronted
with the task of properly dimensioning the channels—especially their resistances.
Example 4.2 Consider again the partial specification as shown in Fig. 4.1. Let’s
additionally assume that the designer uses syringe pumps for the continuous phase
as well as for the dispersed phase, which overall produce a constant volumetric
flow rate of Q in = 3 μl/min. Furthermore, let’s assume that water is used as
continuous phase (having a viscosity of μ cont = 1 mPa s and a density of ρ = 1 g/ml)
and silicone oil is used as dispersed phase (resulting in an interfacial tension of
γ = 46 mN/m).
53
Having this partial specification, it is left to define the required channels, i.e. to
properly dimension the channels. This however constitutes a challenge, since their
resistances significantly affect the flow of the droplets. In the currently applied
design process, the channels are defined based on the designer’s experience. While
doing that, designers have to take a huge number of constraints and dependencies
into consideration—already slightly changing, e.g., the specification of a single
channel may change the behavior of the entire microfluidic system. As a consequence, designers often cannot grasp all effects and dependencies anymore. More
precisely, because of improper specifications of channels, microfluidic networks
may result in which
• droplets flow against the intended direction or
• droplets pass a channel/module too slowly (critical when, e.g., a droplet just
passed a heating module and needs to be analyzed by a following detector module
without cooling down) or too quickly (e.g., when the channel is used as delay
line).
In order to formulate these problems, the 1D analysis model provided in
Sect. 3.2 is employed in the following. Based on that model, two methods which
allow designers to (1) automatically validate whether their chosen specification
indeed works as intended (i.e., avoid problems as discussed above) as well as
(2) automatically conduct the dimensioning to obtain a proper specification are
proposed in Sects. 4.2 and 4.3, respectively.
4.2 Validating the Specification
Using the 1D analysis model provided in Sect. 3.2 as basis, this section describes a
method which automatically checks whether a given specification of a microfluidic
network shows any of the problems discussed above. To describe the method,
recall that, as discussed in Sect. 4.1, the used pump, droplet generators, sorters, and
modules are predefined by the experiment and the designer is mainly confronted
with the task of properly dimensioning the channels—especially their resistances.
Example 4.2 Consider again the partial specification as shown in Fig. 4.1. Let’s
additionally assume that the designer uses syringe pumps for the continuous phase
as well as for the dispersed phase, which overall produce a constant volumetric
flow rate of Q in = 3 μl/min. Furthermore, let’s assume that water is used as
continuous phase (having a viscosity of μ cont = 1 mPa s and a density of ρ = 1 g/ml)
and silicone oil is used as dispersed phase (resulting in an interfacial tension of
γ = 46 mN/m).