What Is Oxygen Uptake Rate (OUR)?
Oxygen uptake rate (OUR) is the volumetric rate at which cells consume dissolved oxygen in a bioreactor, expressed in mmol O2/L/h. It quantifies the total respiratory demand of the culture and is one of the most important physiological parameters in aerobic fermentation.
OUR depends on two factors: how fast each cell respires, and how many cells are present. The fundamental relationship is:
where qO2 is the specific oxygen uptake rate (the oxygen consumption per unit biomass per unit time) and X is the biomass concentration. For microbial cultures, qO2 is typically expressed in mmol O2/gDCW/h, while for mammalian cells it is expressed in mol/cell/h or pmol/cell/day.
OUR serves three practical purposes in bioprocess engineering:
- Oxygen supply sizing — the bioreactor must deliver enough oxygen transfer (OTR ≥ OUR) at peak cell density
- Metabolic monitoring — sudden changes in OUR indicate shifts in cell metabolism, nutrient depletion, or contamination
- Biomass estimation — if qO2 is known, OUR measurements can estimate viable cell concentration in real time
OUR vs OTR: The Oxygen Mass Balance
The dissolved oxygen concentration in a bioreactor is governed by the balance between oxygen supply (OTR) and oxygen demand (OUR). At any instant, the rate of change of dissolved oxygen follows:
where CL is the dissolved oxygen concentration (mmol/L), C* is the saturation concentration (~0.21 mmol/L in water at 37°C, 1 atm air), and kLa is the volumetric mass transfer coefficient (h−1).
Three scenarios arise from this balance:
- OTR = OUR — steady state. Dissolved oxygen remains constant at the setpoint. This is normal operating condition.
- OTR > OUR — dissolved oxygen rises. The controller reduces agitation or airflow to maintain the setpoint.
- OTR < OUR — dissolved oxygen drops. If it falls below the critical dissolved oxygen concentration (Ccrit), cells become oxygen-limited and metabolism shifts (acetate overflow in E. coli, lactate accumulation in CHO).
Diagram showing a bioreactor vessel with gas bubbles and cells in the liquid phase. Arrows indicate OTR as oxygen moving from gas to liquid phase, and OUR as oxygen consumed by cells. Labels show the equations OTR = kLa times (C star minus CL) and OUR = qO2 times X. At the bottom, the steady-state condition is shown: dCL/dt = 0 implies OTR = OUR.
The critical dissolved oxygen concentration (Ccrit) is the threshold below which qO2 becomes dependent on CL and metabolism is impaired. Typical Ccrit values are 5–10% of air saturation for bacteria and yeast, and 10–50% for filamentous fungi (depending on pellet size). For most aerobic fermentations, the DO setpoint is maintained well above Ccrit — typically at 20–40% air saturation.
Three Methods to Measure OUR
OUR can be measured experimentally using three approaches, each suited to different scales and equipment configurations. The dynamic method is simplest at lab scale, while off-gas analysis is preferred in production bioreactors.
1. Dynamic Method (DO Drawdown)
The dynamic method measures OUR directly from the rate of dissolved oxygen decline when aeration is briefly stopped. With no gas transfer (kLa ≈ 0), the mass balance simplifies to:
The procedure is:
- Record baseline DO at the current operating conditions
- Stop air flow and agitation (or reduce agitation to minimum to avoid surface aeration)
- Monitor the linear decline in DO over 30–120 seconds
- Calculate OUR from the negative slope of CL vs time
- Resume aeration before DO drops below Ccrit
Advantages: No special equipment needed beyond a DO probe. Simple, fast, and works at any scale.
Limitations: Briefly starves cells of oxygen (must not drop below Ccrit). DO probe response time must be faster than the OUR signal (~5–15 s for optical probes, 30–60 s for polarographic).
2. Off-Gas Mass Balance (Exhaust Analysis)
The off-gas method uses a paramagnetic or zirconia O2 analyzer on the bioreactor exhaust to calculate OUR from the difference between inlet and outlet oxygen:
where Fin and Fout are the molar gas flow rates (mmol/h), yO2 is the mole fraction of O2, and VL is the liquid volume (L). The outlet flow rate can be estimated using an inert gas balance (nitrogen or argon tie).
Advantages: Continuous, non-invasive, no disruption to the culture. Also gives the carbon dioxide evolution rate (CER) and respiratory quotient (RQ = CER/OUR).
Limitations: Requires a calibrated off-gas analyzer (~$5,000–$15,000). Condensation in exhaust lines can cause drift. Small headspace volumes amplify measurement noise.
3. Steady-State DO Balance
If kLa is known from prior characterization, OUR can be calculated from the steady-state dissolved oxygen:
Advantages: No additional equipment — uses only the DO probe reading and the known kLa.
Limitations: kLa must be accurately determined under actual broth conditions (viscosity, antifoam, and cell density all affect kLa). Assumes true steady state.
| Method | Equipment needed | Measurement type | Best suited for | Accuracy |
|---|---|---|---|---|
| Dynamic (DO drawdown) | DO probe only | Intermittent | Lab scale (0.5–50 L) | ±10–15% |
| Off-gas mass balance | O2/CO2 analyzer, MFC | Continuous, real-time | Pilot & production (>10 L) | ±3–5% |
| Steady-state DO balance | DO probe + known kLa | Continuous (derived) | Any scale with kLa data | ±15–25% |
Specific Oxygen Uptake Rate (qO2): Typical Values
The specific oxygen uptake rate (qO2) quantifies the respiratory intensity of individual cells or biomass. It varies by organism, growth phase, carbon source, and temperature. Microbial cells respire 10–100× faster per gram of biomass than mammalian cells, which is why microbial fermentations demand much higher kLa values.
| Organism | qO2 (mmol O2/gDCW/h) | qO2 (pmol/cell/day) | Growth phase |
|---|---|---|---|
| E. coli (glucose, 37°C) | 5–15 | — | Exponential |
| E. coli (glucose-limited fed-batch) | 3–8 | — | Fed-batch |
| S. cerevisiae (aerobic, glucose) | 1–8 | — | Exponential |
| Pichia pastoris (methanol) | 4–12 | — | Induction |
| CHO cells (37°C) | — | 2–8 | Exponential |
| CHO cells (33°C shift) | — | 1.5–4 | Production |
| HEK293 (37°C) | — | 3–10 | Exponential |
| Sf9 insect cells (27°C) | — | 4–12 | Pre-infection |
| Corynebacterium glutamicum | 3–9 | — | Exponential |
| Streptomyces spp. | 1–5 | — | Growth & production |
A key observation: microbial qO2 values of 5–15 mmol/g/h combined with high cell densities (10–100 gDCW/L) produce volumetric OUR values of 50–300 mmol/L/h. In contrast, CHO cultures at 10–30 × 106 cells/mL typically have OUR of 0.5–3 mmol/L/h — two orders of magnitude lower. This is why E. coli fermentations require kLa values of 200–1,000 h−1, while CHO cultures operate comfortably at 5–20 h−1.
OTR & kLa Estimator
Calculate whether your bioreactor can deliver enough OTR to meet OUR demand at your target cell density.
Worked Example: OUR from Off-Gas Data
The off-gas mass balance is the most accurate continuous method for OUR determination. Here is a step-by-step calculation from a 50 L E. coli fed-batch fermentation.
Worked Example — Off-Gas OUR Calculation
Given:
- Working volume (VL) = 35 L
- Inlet air flow = 35 L/min (1 VVM) at STP
- Inlet O2 mole fraction (yO2,in) = 0.2095
- Outlet O2 mole fraction (yO2,out) = 0.1940 (from paramagnetic analyzer)
- Inlet N2 fraction = 0.7808, outlet N2 = 0.7935 (inert gas tie)
- Temperature = 37°C, pressure = 1 atm
Step 1: Convert inlet flow to molar flow rate
Fin = 35 L/min × (1 mol / 22.4 L) × (273.15 / 310.15) × 60 min/h
Fin = 35 × 0.04464 × 0.8808 × 60
Fin = 82.6 mol/h
Step 2: Calculate outlet flow using nitrogen tie
Fout = Fin × (yN2,in / yN2,out)
Fout = 82.6 × (0.7808 / 0.7935)
Fout = 81.3 mol/h
Step 3: Calculate OUR
OUR = (Fin × yO2,in − Fout × yO2,out) / VL
OUR = (82.6 × 0.2095 − 81.3 × 0.1940) / 35
OUR = (17.30 − 15.77) / 35
OUR = 1.53 / 35
OUR = 43.7 mmol O2/L/h
Step 4: Estimate qO2 (biomass = 25 gDCW/L)
qO2 = OUR / X = 43.7 / 25
qO2 = 1.75 mmol/g/h
This qO2 of 1.75 mmol/g/h is below the typical exponential-phase range of 5–15 mmol/g/h, which is consistent with a glucose-limited fed-batch strategy that restricts the specific growth rate to avoid acetate overflow.
How OUR Changes During Fermentation
OUR is not constant — it rises as cells grow, peaks at maximum cell density, and declines during stationary or death phase. Tracking OUR over time reveals metabolic transitions that are invisible from DO or pH data alone.
In a typical E. coli fed-batch:
- Lag phase (0–2 h): OUR near zero as cells adapt
- Batch exponential phase (2–8 h): OUR rises exponentially, tracking cell growth at μmax. This is the most oxygen-demanding phase per unit biomass.
- Feed start / transition (8–10 h): Glucose depletion triggers a brief OUR dip. The RQ may spike as cells shift to mixed-acid fermentation products before the feed stabilises.
- Fed-batch phase (10–24 h): OUR continues to rise with biomass but qO2 is lower because μ is restricted by the feed rate. Total OUR peaks here.
- Stationary / harvest (24+ h): OUR plateaus or declines as growth stops.
For CHO fed-batch cultures, the pattern is similar but over 10–14 days. OUR peaks around day 5–7 at maximum viable cell density, then declines as viability drops in the production phase. A sudden OUR drop can indicate nutrient depletion, contamination, or loss of viability.
Using OUR for Scale-Up and Process Control
OUR is the bridge between cell physiology and engineering design. The fundamental scale-up requirement for oxygen is simple: the bioreactor must provide OTR ≥ OUR at peak demand. From this single constraint, you can derive the minimum kLa, agitation power, and airflow needed at any scale.
Sizing oxygen supply from OUR
The minimum kLa required to sustain a culture is:
Worked Example — Minimum kLa for Scale-Up
Given: E. coli fed-batch with OURpeak = 100 mmol/L/h, C* = 0.21 mmol/L (at 37°C, air), DO setpoint CL = 0.06 mmol/L (30% air saturation).
kLamin = 100 / (0.21 − 0.06)
kLamin = 100 / 0.15
kLamin = 667 h−1
This is achievable in a well-aerated stirred-tank bioreactor at P/V > 2 kW/m³ and VVM > 1. Oxygen-enriched air would reduce the required kLa by increasing C*.
OUR as a real-time biomass proxy
If qO2 is reasonably constant (often true during fed-batch at a fixed μ), the viable cell concentration can be estimated in real time:
This approach, called an OUR-based soft sensor, provides continuous biomass estimation without sampling. It is particularly useful in single-use bioreactors where inline biomass probes (e.g., capacitance) may not be available. Pappenreiter et al. (2019) demonstrated OUR-based soft sensing for CHO fed-batch cultures, predicting viable cell volume and detecting metabolic transitions in real time.
RQ monitoring for metabolic control
When off-gas analysis provides both OUR and CER, the respiratory quotient (RQ = CER/OUR) reveals the metabolic state:
- RQ ≈ 1.0: Fully aerobic glucose metabolism (C6H12O6 + 6O2 → 6CO2 + 6H2O)
- RQ > 1.0: Overflow metabolism — indicates acetate production in E. coli or ethanol in yeast
- RQ < 1.0: Lipid or amino acid oxidation, or the culture is metabolising overflow products
Scale-Up Calculator
Translate your lab OUR and kLa data to pilot and production scale using constant P/V, tip speed, or kLa strategies.
Gas Mixing Calculator
Optimise your gas blend (air + O2 enrichment) to achieve the target C* and meet OUR demand at scale.
Frequently Asked Questions
What is the difference between OUR and OTR?
OUR (oxygen uptake rate) is the rate at which cells consume dissolved oxygen, while OTR (oxygen transfer rate) is the rate at which oxygen transfers from the gas phase to the liquid. At steady state, OTR = OUR and dissolved oxygen stays constant. When OTR < OUR, DO drops and cells become oxygen-limited.
How do you measure OUR in a bioreactor?
The three main methods are: (1) the dynamic method — briefly stop aeration and measure the slope of DO decline; (2) off-gas mass balance — measure inlet and outlet O2 concentrations to calculate consumed oxygen; and (3) steady-state DO balance using known kLa and DO setpoint. The dynamic method is simplest for lab-scale, while off-gas analysis is preferred at production scale.
What is a typical qO2 for CHO cells?
CHO cells typically have a specific oxygen uptake rate of 0.05–0.35 × 10−12 mol/cell/h, or roughly 2–8 pmol/cell/day. The value varies with growth phase, temperature, and media composition. Exponential-phase cells are at the higher end, while stationary-phase and temperature-shifted (33°C) cells consume less oxygen.
Can OUR be greater than OTR?
Yes — temporarily. When OUR exceeds OTR, dissolved oxygen concentration drops. If this continues, cells become oxygen-limited and metabolism shifts (e.g., overflow metabolism in E. coli produces acetate, CHO cells produce lactate). The bioreactor must be operated so that OTR ≥ OUR to maintain the DO above Ccrit.
Why does OUR matter for scale-up?
OUR determines the minimum oxygen supply your bioreactor must deliver. During scale-up, you need OTR ≥ OUR at peak cell density. Since OUR rises with cell growth while OTR depends on kLa and operating conditions, matching supply to demand is one of the most common scale-up challenges — especially above 500 L where gas-liquid interfacial area per unit volume declines.
Related Tools
- OTR & kLa Estimator — Calculate kLa and maximum OTR for stirred-tank bioreactors at any scale.
- Gas Mixing Calculator — Design gas blends (air + O2 + N2 + CO2) and calculate resulting C* and OTR.
- Scale-Up Calculator — Translate process parameters across bioreactor scales using constant P/V, tip speed, or kLa.
References
- Garcia-Ochoa F, Gomez E, Santos VE, Merchuk JC. Oxygen uptake rate in microbial processes: An overview. Biochemical Engineering Journal. 2010;49(3):289–307. doi:10.1016/j.bej.2010.01.011
- Garcia-Ochoa F, Gomez E. Bioreactor scale-up and oxygen transfer rate in microbial processes: An overview. Biotechnology Advances. 2009;27(2):153–176. doi:10.1016/j.biotechadv.2008.10.006
- Goudar CT, Piret JM, Konstantinov KB. Estimating cell specific oxygen uptake and carbon dioxide production rates for mammalian cells in perfusion culture. Biotechnology Progress. 2011;27(5):1347–1357. doi:10.1002/btpr.646
- Pappenreiter M, Sissolak B, Sommeregger W, Striedner G. Oxygen uptake rate soft-sensing via dynamic kLa computation: Cell volume and metabolic transition prediction in mammalian bioprocesses. Frontiers in Bioengineering and Biotechnology. 2019;7:195. doi:10.3389/fbioe.2019.00195
- Martínez-Monge I, Roman R, Comas P, et al. New developments in online OUR monitoring and its application to animal cell cultures. Applied Microbiology and Biotechnology. 2019;103(16):6903–6917. doi:10.1007/s00253-019-09989-4