Off-gas RQ anomaly in an aerobic yeast fed-batch
An 800 L fed-batch culture of recombinant Saccharomyces cerevisiae is aerated at 0.5 vvm with air (20.9% O2, 0.04% CO2). Off-gas analysis at steady state reads 19.0% O2 and 2.50% CO2. The bioreactor's kLa for O2 is 120 h^-1, the calibrated dissolved-oxygen probe reads 1.8 mg/L, and the solubility C* is 7.2 mg/L. Feed is 50% (w/w) glucose at 18 L/h; the culture has been at a constant 800 L working volume for 6 h. The operator notes a rising base demand and a faint ethanol smell. The batch record shows the DO probe was cleaned and polarized 18 h earlier, but not checked against a zero/saturation standard. The off-gas analyzer was calibrated the same morning. At the current feed, biomass is 35 g/L and ethanol is 0.6 g/L. The team must decide whether to increase feed for higher productivity or intervene.
1. Calculate the off-gas OUR and CER (g L^-1 h^-1) and the RQ. Compare the off-gas OUR with the maximum OUR supported by the stated kLa and dissolved oxygen.
Read the worked answer
Gas flow and balances. At 800 L and 0.5 vvm, gas flow is 400 L/min. Using 22.4 L/mol, total molar flow is 400/22.4 = 17.86 mol/min. O2 consumed = 17.86 x (0.209 - 0.190) = 0.339 mol/min; as mass, 0.339 x 32 = 10.85 g/min = 651 g/h; per litre, 651/800 = 0.814 g L^-1 h^-1. Net CO2 = 17.86 x (0.0250 - 0.0004) = 0.439 mol/min; as mass, 0.439 x 44 = 19.3 g/min = 1,160 g/h; per litre = 1.45 g L^-1 h^-1. RQ = CER/OUR = 1.45/0.814 = 1.78.
kLa comparison. The kLa-based OUR is kLa(C* - CL) = 120 x (7.2 - 1.8) = 648 mg L^-1 h^-1 = 0.648 g L^-1 h^-1. The off-gas OUR is about 26% higher. If kLa is correct, the true DO would be 7.2 - 0.814/120 = 0.42 mg/L, not 1.8 mg/L. If the DO probe is correct, the required kLa is 0.814/0.0054 = 151 h^-1. The two OUR estimates are therefore not mutually consistent; the discrepancy must be resolved before changing feed.
2. Interpret RQ = 1.78. Does it prove ethanol production? How do you reconcile the OUR discrepancy, and what is the immediate feed decision?
Read the worked answer
RQ = 1.78 is well above the purely respiratory value near 1.0 and is consistent with fermentative metabolism in yeast, i.e. ethanol production from glucose. It does not by itself prove ethanol production: CO2 stripping from bicarbonate, base addition, sensor cross-sensitivity, or a gas-flow error can also shift apparent RQ. However, the faint ethanol smell and rising base demand support a real fermentative shift.
The OUR discrepancy points to an oxygen-transfer or DO-measurement problem. If the DO probe reads falsely high, true DO is about 0.42 mg/L, which is below the usual 20-30% saturation threshold and would cause oxygen-limited fermentation. If the DO probe is correct, kLa is understated at about 151 h^-1 and the reactor is still close to its oxygen-transfer limit. In either case, increasing feed is the wrong action: it would push glucose uptake beyond respiratory capacity and increase ethanol/acetate formation.
Immediate action: cut the feed by 30-50% (for example to 9-12 L/h), hold the culture, and verify DO and kLa before restoring feed. Do not treat RQ alone as a reason to feed more.
3. Propose a three-step check to establish which measurement is wrong and a criterion for resuming the higher feed.
Read the worked answer
- Check the DO probe off-line. Calibrate at zero using sodium sulfite solution and at saturation using air-sparged medium at the run temperature. Compare the probe reading with a second probe or a portable meter. A bias greater than about 10% invalidates the kLa calculation.
- Check gas flow and kLa independently. Measure inlet air flow with a mass-flow meter and analyse a certified gas standard on the off-gas analyser. Then perform a dynamic gassing-out kLa test in the same vessel and agitation/airflow regime, or compare with a sulfite-oxidation kLa. This distinguishes a gas-flow error from a probe error.
- Check the biology offline. Measure glucose, ethanol, glycerol and acetate by HPLC, and measure OUR by a short respirometry test. If ethanol is rising while glucose is still present, the fermentative shift is real even if the sensors disagree.
Resume the higher feed only when: off-gas OUR and kLa-based OUR agree within about 10%; RQ is below about 1.1-1.2; ethanol is below 1 g/L and not rising; DO is above 30% saturation with the probe cross-checked; and the base demand is stable. If these criteria are not met, keep the reduced feed and fix the measurement or oxygen-transfer limitation first.