What Is Off-Gas Analysis?
Off-gas analysis is the continuous measurement of O2 and CO2 concentrations in bioreactor exhaust gas to calculate oxygen uptake rate (OUR), CO2 evolution rate (CER), and respiratory quotient (RQ) in real time. It is the most information-rich non-invasive measurement available in fermentation, providing a direct window into cellular metabolism without removing a sample from the vessel.
By comparing the gas composition entering the bioreactor (typically air: 20.95% O2, 0.04% CO2, 79.01% N2) with the exhaust gas leaving it, off-gas analysis quantifies how much oxygen cells consume and how much carbon dioxide they produce each second. The ratio of these two rates, the respiratory quotient, reveals whether cells are in fully oxidative metabolism, overflow metabolism, or anaerobic fermentation.
Off-gas analysis is standard practice at pilot and production scale (10 L and above) and is increasingly common at bench scale (1-5 L). It detects metabolic events 2-4 hours earlier than offline metabolite sampling because gas exchange responds immediately to intracellular metabolic changes, while extracellular metabolite concentrations lag behind.
Measurement Setup and Gas Analyzers
A complete off-gas analysis system consists of five components connected in series: the bioreactor headspace, an exhaust condenser to remove water vapor, a particulate filter, the gas analyzer(s), and a data acquisition system. Accurate off-gas analysis depends on proper conditioning of the exhaust gas before it reaches the analyzer.
Paramagnetic O2 analyzers exploit the fact that oxygen is one of very few gases with strong paramagnetic properties. A sample cell positioned in a non-uniform magnetic field measures the deflection force proportional to O2 concentration. Accuracy is typically ± 0.01-0.05% v/v with a measurement range of 0-25% O2.
NDIR (non-dispersive infrared) CO2 analyzers pass infrared light at 4.26 µm through the gas sample. CO2 absorbs at this wavelength according to the Beer-Lambert law, and the detector measures the attenuation. Accuracy is ± 0.01-0.03% v/v with measurement ranges from 0-5% or 0-15% CO2.
The exhaust condenser is critical: water vapor in the exhaust gas (saturated at 37°C, ∼6.2% v/v) dilutes O2 and CO2 readings. Cooling the gas to 4-10°C in a Peltier or glycol condenser reduces the water content to < 0.5% v/v. A downstream Nafion dryer or desiccant can reduce it further. Failure to remove moisture is the single most common source of off-gas measurement error.
The N2 Balance Method
The inlet and outlet gas flow rates of a bioreactor are not equal. In aerobic fermentation, cells consume O2 and produce CO2, but typically consume more O2 than they produce CO2 (when RQ < 1). This means the total moles of exhaust gas leaving the bioreactor are fewer than the moles entering it.
The N2 balance corrects for this volume change by using nitrogen as an inert tracer. Since N2 is neither consumed nor produced by the culture, the moles of N2 entering must equal the moles leaving:
N2 Balance Derivation
Fin × yN2,in = Fout × yN2,out
Therefore:
Fout = Fin × (yN2,in / yN2,out)
Where yN2 is calculated as the remainder after measuring O2 and CO2:
yN2 = 1 - yO2 - yCO2 - yH2O - yAr
For dry gas after the condenser, and absorbing Ar (∼0.93%) into the N2 fraction:
yN2 ≈ 1 - yO2 - yCO2
Skipping the N2 balance and assuming Fout = Fin introduces a systematic error of 5-15% in OUR, depending on cell density and metabolic rate. At high cell densities where O2 depletion exceeds 3-4% of inlet concentration, this error becomes significant for process control decisions.
How to Calculate OUR and CER
OUR and CER are calculated from the difference between inlet and outlet gas compositions, corrected by the N2 balance. All gas volumes must be at standard conditions (0°C, 1 atm) or normalized accordingly.
OUR Formula (Off-Gas Mass Balance)
OUR = (Fin × yO2,in - Fout × yO2,out) / VL
Substituting the N2 balance for Fout:
OUR = (Fin / VL) × [yO2,in - yO2,out × (yN2,in / yN2,out)]
Units: mmol O2 L-1 h-1 (when Fin in mmol/h and VL in L)
CER Formula (Off-Gas Mass Balance)
CER = (Fout × yCO2,out - Fin × yCO2,in) / VL
Substituting the N2 balance for Fout:
CER = (Fin / VL) × [yCO2,out × (yN2,in / yN2,out) - yCO2,in]
Units: mmol CO2 L-1 h-1
The specific rates (qO2 and qCO2) normalize OUR and CER by biomass concentration:
- qO2 = OUR / X (mmol O2 gDCW-1 h-1)
- qCO2 = CER / X (mmol CO2 gDCW-1 h-1)
Typical qO2 ranges: E. coli 10-20 mmol g-1 h-1 (glucose-limited growth), CHO cells 0.05-0.4 mmol per 106 cells per hour, S. cerevisiae 5-12 mmol g-1 h-1.
What Is the Respiratory Quotient (RQ)?
The respiratory quotient (RQ) is the molar ratio of CO2 produced to O2 consumed, calculated as RQ = CER / OUR. It is a dimensionless number that directly reflects the metabolic state of the culture because different catabolic pathways produce different ratios of CO2 to O2.
For the complete oxidation of glucose (C6H12O6 + 6 O2 → 6 CO2 + 6 H2O), six moles of CO2 are produced per six moles of O2 consumed, giving RQ = 1.0. Any deviation from 1.0 indicates that pathways beyond simple glucose oxidation are active.
RQ can be calculated at any time point where valid O2 and CO2 measurements are available. During lag phase or at very low cell densities, the difference between inlet and outlet gas compositions may be within analyzer noise (ΔO2 < 0.1%), making RQ unreliable. A minimum OUR of ∼ 1 mmol L-1 h-1 is typically needed for meaningful RQ values.
| Substrate / Metabolic State | Stoichiometry | Theoretical RQ | Typical Measured |
|---|---|---|---|
| Glucose oxidation (aerobic) | C6H12O6 + 6 O2 → 6 CO2 | 1.00 | 0.95-1.05 |
| Lipid/fatty acid oxidation | C16H32O2 + 23 O2 → 16 CO2 | 0.70 | 0.65-0.75 |
| Ethanol oxidation | C2H6O + 3 O2 → 2 CO2 | 0.67 | 0.60-0.70 |
| Glycerol oxidation | C3H8O3 + 3.5 O2 → 3 CO2 | 0.86 | 0.80-0.90 |
| Acetic acid oxidation | C2H4O2 + 2 O2 → 2 CO2 | 1.00 | 0.95-1.05 |
| Citric acid production | Glucose → citrate + CO2 (partial) | 1.33 | 1.20-1.40 |
| Ethanol overflow (S. cerevisiae) | Glucose → ethanol + CO2 (Crabtree) | > 1.5 | 1.5-4.0 |
| Anaerobic fermentation | Glucose → products + CO2 (no O2) | ∞ | > 5.0 |
RQ Values for Different Metabolic States
RQ serves as a real-time metabolic fingerprint. Values near 1.0 confirm fully oxidative glucose metabolism. Values significantly above 1.0 indicate that overflow or fermentative pathways have activated, producing reduced byproducts like ethanol or organic acids. Values below 1.0 indicate that the cells are oxidizing substrates more reduced than glucose, such as fatty acids or ethanol.
- RQ = 0.65-0.75: Lipid or ethanol oxidation. Seen when S. cerevisiae consumes ethanol after a diauxic shift, or in CHO cells catabolizing lipid stores.
- RQ = 0.95-1.05: Balanced oxidative metabolism on glucose. The target operating range for most aerobic fermentations.
- RQ = 1.05-1.20: Early overflow metabolism. Acetate formation begins in E. coli; mild ethanol production in yeast. Corrective action (reduce feed rate) is still effective.
- RQ > 1.5: Strong overflow or Crabtree effect. Significant ethanol production in S. cerevisiae, or severe acetate overflow in E. coli. Carbon is being diverted to fermentative products.
RQ-Based Feedback Control of Feeding
RQ-based feed control is the most powerful application of off-gas analysis. By monitoring RQ in real time and adjusting the glucose feed rate to maintain RQ below a critical threshold, overflow metabolism is prevented without offline sampling. This approach was pioneered for S. cerevisiae and is now standard for E. coli high-cell-density fed-batch processes.
For E. coli, the critical RQ threshold is approximately 1.05. Above this value, the specific glucose uptake rate exceeds the TCA cycle capacity and acetate begins to accumulate. An RQ-stat controller reduces the feed rate whenever RQ rises above the setpoint and increases it when RQ drops below, effectively tracking the maximum oxidative capacity of the culture as it grows.
For S. cerevisiae, the Crabtree effect produces ethanol when glucose exceeds the respiratory capacity. Maintaining RQ < 1.10 through feedback-controlled feeding achieves biomass yields of 0.45-0.50 g g-1 versus 0.10-0.15 g g-1 in uncontrolled cultures.
OUR/CER/RQ Off-Gas Analyzer
Calculate OUR, CER, and RQ from your bioreactor inlet/outlet gas compositions. Single-point and time-series modes with the N2 balance method.
Worked Example: 10 L E. coli Fed-Batch
This worked example calculates OUR, CER, and RQ from off-gas measurements during early fed-batch phase of an E. coli BL21(DE3) fermentation at 37°C, shortly after glucose depletion and feed start.
Worked Example: OUR, CER, and RQ Calculation
Given:
- Working volume VL = 7.0 L
- Inlet gas flow Fin = 7.0 SLPM (1 VVM at STP)
- Inlet composition (dry): yO2,in = 0.2095, yCO2,in = 0.0004
- Outlet composition (dry, after condenser): yO2,out = 0.1940, yCO2,out = 0.0175
- Biomass concentration X = 3.0 gDCW L-1 (OD600 ≈ 9)
Step 1: Calculate yN2 fractions
yN2,in = 1 - 0.2095 - 0.0004 = 0.7901
yN2,out = 1 - 0.1940 - 0.0175 = 0.7885
Step 2: Calculate Fout (N2 balance)
Fout = 7.0 × (0.7901 / 0.7885) = 7.0 × 1.00203 = 7.014 SLPM
Step 3: Convert flow to mmol h-1
At STP: 1 mol = 22.414 L, so 1 SLPM = 60/22.414 = 2.676 mmol min-1 = 160.6 mmol h-1
Fin = 7.0 × 160.6 = 1124.0 mmol h-1
Fout = 7.014 × 160.6 = 1126.3 mmol h-1
Step 4: Calculate OUR
OUR = (1124.0 × 0.2095 - 1126.3 × 0.1940) / 7.0
OUR = (235.5 - 218.5) / 7.0
OUR = 17.0 / 7.0 = 2.43 mmol O2 L-1 h-1
Step 5: Calculate CER
CER = (1126.3 × 0.0175 - 1124.0 × 0.0004) / 7.0
CER = (19.71 - 0.45) / 7.0
CER = 19.26 / 7.0 = 2.75 mmol CO2 L-1 h-1
Step 6: Calculate RQ
RQ = CER / OUR = 2.75 / 2.43 = 1.13
Step 7: Specific rates
qO2 = 2.43 / 3.0 = 0.81 mmol g-1 h-1
qCO2 = 2.75 / 3.0 = 0.92 mmol g-1 h-1
Interpretation: RQ = 1.13 indicates mild overflow metabolism: the glucose feed rate slightly exceeds the oxidative capacity. The qO2 of 0.81 mmol g-1 h-1 is within the glucose-limited range for E. coli at low specific growth rates. Reduce the feed rate by ∼10% to bring RQ below 1.05 and prevent acetate accumulation.
Troubleshooting Off-Gas Measurements
Off-gas data quality depends on proper calibration, gas conditioning, and system leak checks. The most common problems and their solutions follow.
| Symptom | Likely Cause | Corrective Action |
|---|---|---|
| RQ consistently > 1.2 during balanced growth | CO2 analyzer drift or calibration error | Recalibrate with certified span gas (5% CO2 in N2) |
| OUR reads negative or near zero | Leak in exhaust line (ambient air dilution) | Pressure-test all connections; check condenser drain trap |
| Noisy RQ oscillations (± 0.3+) | Condensation in gas lines or analyzer cell | Verify condenser temperature < 10°C; add Nafion dryer |
| OUR spikes at each aeration change | Pressure transient in the headspace | Install back-pressure regulator; increase headspace volume |
| CER reading delayed by > 2 min vs OUR | CO2 dissolving in liquid phase (carbonic acid equilibrium) | Normal at high pH (> 7.5); apply dissolved CO2 correction |
| Gradual O2 reading decline over weeks | Paramagnetic cell fouling or sensor aging | Clean sensing cell; replace if > 2 years old |
Calibration should be performed at the start of each fermentation using two-point calibration: pure N2 (zero gas) and a certified span gas mixture (typically 15% O2 / 5% CO2 / balance N2). Daily zero-drift checks during long runs catch analyzer aging.
OTR & kLa Estimator
Estimate oxygen transfer rate and kLa for your bioreactor. Compare OTR capacity against the OUR demands measured by off-gas analysis.
Frequently Asked Questions
What is the respiratory quotient (RQ) and why does it matter in fermentation?
The respiratory quotient (RQ) is the molar ratio of CO2 produced to O2 consumed (RQ = CER / OUR). It reveals the metabolic state of the culture without offline sampling. Fully oxidative glucose metabolism gives RQ = 1.0. Values above 1.0 indicate overflow metabolism or anaerobic byproduct formation (e.g., ethanol at RQ > 1.5), while values below 1.0 suggest lipid oxidation (RQ ∼ 0.7) or gluconeogenesis.
How do you calculate OUR from off-gas data?
OUR is calculated using the inert-gas (N2) balance method: first determine the outlet gas flow rate from Fout = Fin × (yN2,in / yN2,out), then calculate OUR = (Fin × yO2,in - Fout × yO2,out) / VL. This accounts for the change in total gas volume caused by unequal O2 consumption and CO2 production. Units are typically mmol O2 per liter per hour.
What gas analyzers are used for bioreactor off-gas analysis?
Paramagnetic analyzers measure O2 (accuracy ± 0.01-0.05% v/v, response time 3-10 seconds) by exploiting oxygen's strong paramagnetic susceptibility. NDIR (non-dispersive infrared) analyzers measure CO2 (accuracy ± 0.01-0.03% v/v, response time 2-5 seconds) based on CO2 absorption of 4.26 µm IR radiation. Both are standard in benchtop and production-scale bioreactor systems.
Why is the N2 balance needed for off-gas calculations?
The N2 balance corrects for the fact that inlet and outlet gas flow rates differ. In aerobic fermentation, more O2 is consumed than CO2 is produced (when RQ < 1), causing the total exhaust gas volume to shrink. Without the N2 balance correction, OUR calculations contain systematic errors of 5-15%. Since N2 is metabolically inert, its mole fraction change reflects only the flow rate change.
Can off-gas analysis detect contamination or process deviations?
Yes. Sudden RQ shifts, unexpected OUR spikes, or CER increases outside the normal growth phase profile can indicate contamination, substrate feed failures, or metabolic stress. Off-gas analysis detects these events 2-4 hours earlier than offline metabolite sampling because gas exchange responds immediately to metabolic changes.
Related Tools
- OUR/CER/RQ Off-Gas Analyzer — Calculate OUR, CER, and RQ from inlet/outlet gas compositions with the N2 balance method
- OTR & kLa Estimator — Estimate oxygen transfer capacity and compare it against measured OUR
- Gas Mixing Calculator — Calculate gas blending ratios for O2 enrichment and CO2 overlay strategies
References
- Doran PM. Bioprocess Engineering Principles, 2nd ed. Academic Press, 2013. Chapter 6: Mass Transfer. doi:10.1016/C2009-0-22348-8
- Villadsen J, Nielsen J, Lidén G. Bioreaction Engineering Principles, 3rd ed. Springer, 2011. Chapter 5: Stoichiometry. doi:10.1007/978-1-4419-9688-6
- Losen M et al. Effect of oxygen limitation and medium composition on Escherichia coli fermentation in shake-flask cultures. Biotechnol Prog. 2004;20(4):1062-1068. doi:10.1021/bp034282t
- Sonnleitner B, Käppeli O. Growth of Saccharomyces cerevisiae is controlled by its limited respiratory capacity: Formulation and verification of a hypothesis. Biotechnol Bioeng. 1986;28(6):927-937. doi:10.1002/bit.260280620
- Heinzle E, Biwer AP, Cooney CL. Development of Sustainable Bioprocesses: Modeling and Assessment. Wiley, 2006. Chapter 4: Measurement and Control. doi:10.1002/9780470058916