Optimal design of cell-friendly microfluidic bio-reactors

Contacts:

Fridolin Okkels, DTU Nanotech, 4525-5749 (Fridolin.Okkels@nanotech.dtu.dk)

http://www.nanotech.dtu.dk/microfluidics

Numerical simulation of the flow and consumption of metabolite through a bio-reactor. The lower boundary is the bottom of the reactor, and the flow is visualized by white arrows. Organisms are immobilized on the wall within the central rectangle, and their consumption of metabolite is seen by a local decrease in the metabolite concentration, shown in colorcoding.

One of the most important elements in the drastically emerging Lab-on-a-Chip technology is the micro-scale chemical reactor, where reagents and products are lead to and from the reactor by micro-channel fluid flow. These micro-reactors have numerous advantages compared to conventional reactors: Lower fluid volume consumption, faster analysis and response time and better process control.

Though, when adding living organisms to create microfluidic bio-reactors, the miniaturization causes problems, as the fluid flow in the micro-channels generates a high shear near the channel-walls, and this inhibits the growth of the organisms.  

Solving this problem leads to a contradiction: A flow is needed to supply the cells with metabolite (nutrition and oxygen), but at the same time this flow creates a shear which inhibits the growth of the cells! The figure illustrates this situation, where the white arrows show the fluid flow near the lower reactor wall, and where the metabolism of the cells cause the reduction in the metabolite concentration, as shown in color-scale (red is inlet concentration).

A way to work around this contradiction is to use diffusive transport of the metabolite to the cells. This will involve a complex geometrical structuring of the reactor, which will be found using the state-of-the-art free form optimization method called topology optimization. The force of this method is that (i) the optimal geometrical structure can be of unlimited complexity, and therefore challenging our intuition. (ii) We have implemented this method in the commercial simulation-tool COMSOL, which give us a strong and flexible numerical environment. (iii) Both the method and the COMSOL-implementation can be used for structural optimization of many Lab-on-a-Chip related devices.

The main objectives of the project will be:

• Formulate appropriate reaction-model
• From theoretical considerations to extract the basic properties of the model
• Optimization the reactor-geometry at different conditions
• From these results, if possible, to extract a general optimization strategy