Low-cost, open-source cell culture chamber for regulating physiologic oxygen levels.

The physiological oxygen levels for several mammalian cell types in vivo are considered to be hypoxic (low oxygen tension), but the vast majority of in vitro mammalian cell culture is conducted at atmospheric oxygen levels of around 21%. In order to understand the impact of low oxygen environments on cells, oxygen levels need to be regulated during in vitro culture. Two common methods for simulating a hypoxic environment are through the regulation of gas composition or chemical induction. Chemically mimicking hypoxia can have adverse effects such as reducing cell viability, making oxygen regulation in cell culture chambers crucial for long-term culture. However, oxygen-regulating cell culture incubators and commercial hypoxia chambers may not always be a viable option due to cost and limited customization. Other low-cost chambers have been developed, but they tend to lack control systems or are fairly small scale. Thus, the objective of this project was to design and develop a low-cost, open-source, controllable, and reproducible hypoxia chamber that can fit inside a standard cell culture incubator. This design allows for the control of O2 between 1 and 21%, while maintaining CO2 levels at 5%, as well as monitoring of temperature, pressure, and relative humidity. Testing showed our hypoxia chamber was able to maintain CO2 levels at 5% and hypoxic O2 levels at 1% and 5% for long-term cell culture. This simple and easy-to-manufacture design uses off the shelf components, and the total material cost was $832.47 (USD).

Enhancement of microfluidic mixing using time pulsing.

Many microfluidic applications require the mixing of reagents, but efficient mixing in these laminar (i.e., low Reynolds number) systems is typically difficult. Instead of using complex geometries and/or relatively long channels, we demonstrate the merits of flow rate time dependency through periodic forcing. We illustrate the technique by studying mixing in a simple "T" channel intersection by means of computational fluid dynamics (CFD) as well as physically mixing two aqueous reagents. The "T" geometry selected consists of two inlet channel segments merging at 90 degrees to each other, the outlet segment being an extension of one of the inlet segments. All channel segments are 200 microm wide by 120 microm deep, a practical scale for mass-produced disposable devices. The flow rate and average velocity after the confluence of the two reagents are 48 nl s(-1) and 2 mm s(-1) respectively, which, for aqueous solutions at room temperature, corresponds to a Reynolds number of 0.3. We use a mass diffusion constant of 10(-10) m(2) s(-1), typical of many BioMEMS applications, and vary the flow rates of the reagents such that the average flow rate remains unchanged but the instantaneous flow rate is sinusoidal (with a DC bias) with respect to time. We analyze the effect of pulsing the flow rate in one inlet only as well as in the two inlets, and demonstrate that the best results occur when both inlets are pulsed out of phase. In this case, the interface is shown to stretch, retain one fold, and sweep through the confluence zone, leading to good mixing within 2 mm downstream of the confluence, i.e. about 1 s of contact. From a practical viewpoint, the case where the inlets are 180 degrees out of phase is of particular interest as the outflow is constant.

Electroosmotic mixing in microchannels.

Mixing is an essential, yet challenging, process step for many Lab on a Chip (LOC) applications. This paper presents a method of mixing for microfluidic devices that relies upon electroosmotic flow. In physical tests and in computer simulations, we periodically vary the electric field with time to mix two aqueous solutions. Good mixing is shown to occur when the electroosmotic flow at the two inlets pulse out of phase, the Strouhal number is on the order of 1, and the pulse volumes are on the order of the intersection volume.

Parameters influencing pulsed flow mixing in microchannels.

The rapid mixing of reagents is a crucial step for on-chip chemical and biological analysis. It has been recently suggested that microfluidic mixing can be greatly enhanced by simply using time pulsing of the incoming flow rates of the two fluids to be mixed (Glasgow, I.; Aubry, N. Lab Chip 2003, 3, 114-120). This paper elaborates on the latter technique, showing through computational fluid dynamics how the mixing efficiency strongly depends on certain dimensionless parameters of the system, while remaining nearly insensitive to others. In particular, it is demonstrated that higher Strouhal numbers (ratio of flow characteristic time scale to the pulsing time period) and pulse volume ratios (ratio of the volume of fluid pulsed to the volume of inlet/outlet intersection) lead to better mixing. This paper also presents a physical device capable of mixing two reagents using pulsing, which shows improved mixing with greater values of the Strouhal number.