spacer
Insert ALT tag here The University of Adelaide Australia
Chemical Engineering Home
About the School
News
Current Students
Future Students
International Students
Business and Industry
Visitors
Staff
Programs
Research
  Publications
  Research Grants
  Postgraduate Students
  Postgraduate Projects
  Research Reports
  Food Safety Alliance
  Micro-algal Biotechnology
Seminars
Staff Only
text zoom: S | M | L

School of Chemical Engineering
Engineering North Building
THE UNIVERSITY OF ADELAIDE
SA 5005
AUSTRALIA
Email

Telephone: +61 8 8303 5446
Facsimile: +61 8 8303 4373


You are here: Chemical Engineering > Research > Research Reports Print View

Control of a High Cell Density Anoxic Fermenter

Author:Jordan Rogers
Supervisors:JordanDr Chris Colby
A/Prof Brian O'Neill

Aim

To determine whether nitrate, glutamate and counter ion concentration, as well as the tank level, can be adequately controlled.

Background

diagram

Figure 1 Block Diagram of the fermenter system

Denitrification, where nitrate or nitrite is reduced to nitric oxide, nitrous oxide or nitrogen gas, is commonly used in waste water treatment as a feasible and economical means of removing nitrates. Denitrification can occur either aerobically, where oxygen is the terminal electron acceptor, or anoxically, where nitrate is the terminal electron acceptor. One of the biggest problems with aerobic fermentation is oxygen limited growth. Due to the low solubility of O2 in water and the reactor broth, the growth rate of cells is often well below the maximum. Using anoxic fermentation may overcome this problem as nitrates are liberally soluble in water.
Previous work has determined that Pseudomonas Denitrificans, a true denitrifier, can grow both anoxically and aerobically. It has also been shown that reduced fermentation times can be achieved when grown anoxically. The optimum nitrate concentration was found to be 2.5 mg/L.

The Model

The batch model consisted of 12 differential equations, 6 of which are ordinary differential equations. The model is valid for aerobic, anoxic and transient conditions and is taken from Kornaros and Lyberatos (1998). Carbon substrate, nitrate, nitrite and oxygen concentration, as well as cell mass, are accounted for. A block diagram (see Figure 1) and mass balances were used to adapt the batch equations into fed batch equations. The resulting equations were constructed in Matlab Simulink® and control loops added to determine if the nitrate, carbon substrate and counter ion concentrations could be adequately controlled while maintaining a fixed height.

Conclusion

  • Anoxic and combined growth result in reduced fermentation times over aerobic growth.
  • Feeding nitrate at 7g/L and using magnesium as the counter ion will avoid excessive build up of the counter ion.
  • PI control was required to adequately control nitrate concentration with Kc>50 and TI<5 minutes.
  • Both P and PI control were able to control glutamate to with in ±1%.
  • The fresh feed required is up to 115 L/hr for nitrate and 0.5 L/hr for glutamate.
  • When set point deviations were considered, 10 mg/L was the optimum nitrate concentration, requiring 28 hours to reach 100 g/L of cells as shown in Figure 2.
graph

Figure 2: Effect of nitrate set point on error and fermentation time