From OpenWetWare
Jump to navigationJump to search

This page is for storing information about operation of a bench-scale chemostat designed by Ben Kirkup. It is in development.

Also see STN Chemostat Design for an alternative, flexible chemostat design.


This image is one of the borosilicate chemostats produced before the switch to quartz. It does not have all the recent design improvements that Finkenbeiner now incorporates (offset air outputs; reinforcements on the struts).

This is an original AutoCAD output for a chemostat design. It is similar to the design originally employed and fabricated by D. Dykhuizen.

The basis of this chemostat is, as designed by Dykhuizen, a siphon which maintains a steady media level passively without need for a second media pump line (an evacuation pump). Mixing is also joined with aeration, as in Dykhuizen's original design. The design has been reinforced, scale changed, and limits on rates redesigned. Glass beads have been added above the air input to promote mixing physically. It is an ~60 ml chemostat as produced now (operating volume actually depends on air pressure and flow rate, at very high flow rates [10 mls/min or greater] or air pressures [which can compensate for the high flow rates]).

This chemostat is best fabricated with the headpiece made from quartz. If minimal media is being used, the body can be of borosilicate and this will protect the cells from 254nm UV, but the UV will penetrate into the head prevent any cells from growing up the feed line, even if they attempt to form a biofilm.


  1. Wash out chemostat. don't lose the little glass beads.
  2. autoclave making sure that all tubes are connected.
  3. setup everything and run media through the device for a little while. this is a good time to calculate the flow rate.
  4. drop some cells in the reactor and off you go.


Normal Operation

  • This Kirkup-style chemostat is assembled using silicone tubing and stainless or plated Luer connectors (Popper and Sons via VWR).
  • The chemostat is autoclaved as a piece, but the head and tail can be autoclaved separately from the body.
  • Glass beads may be autoclaved internal to the chemostat or separately. Every addition after autoclaving increases contamination risk.
  • The media bottle is autoclaved separately, the Luers connected at time of initiation.
  • Luers can be soaked in alcohol just prior to separation/reconnection to reduce contamination risk.
  • The chemostat can be operated by gravity feed or by peristaltic pump. Pumps offer greater control of the flow rate, but also

need to be changed depending on the pump's minimum/maximum rate and tubing required.

Troubleshooting/Common problems

Bryan, anything here?

Temperature consistency in water bath

no problems here. keep the thermometer close to the reactor to make sure you are at 37C where you need to be. use the little surface water balls to prevent evaporation.

Masterflex tubing in the pump

avoid crushing this tubing by removing the clamp on the parastaltic pump when it is not running.


Contamination is a problem with any chemostat, especially if it is run for an extended period. Most of the contamination problems proved to be about the issues of properly autoclaving 10L carboys in a poorly maintained autoclave.

Autoclaving the media thoroughly, being particularly careful to sterilize the Luers when changing parts, using UV to keep aerosols from rising from a very actively aerated chemostat - these are all very helpful. UV can also be used to suppress waste line contamination (so can an ethanol rinse when sampling), though this is only a problem over very long runs (several weeks) it seems. Monitoring the OD of the media as it flows into the chemostat gives notice of media carboy contamination but not prevention. If you suspect contamination, you can always change carboys (and possibly heads) and monitor the suspect carboy for growth. That will confirm contamination of the feed line. UV provides prevention.

Another way to prevent bubbling is to include an anti-foaming agent in the media.