1. What Are Yield Coefficients?
Yield coefficients quantify how efficiently an organism converts substrates into biomass, products, or consumes oxygen. They are the foundation of fed-batch process design: knowing Yx/s (grams of biomass produced per gram of substrate consumed) lets you calculate exactly how much glucose, glycerol, or methanol to feed to reach a target cell density.
The three most important yield coefficients are:
- Yx/s — biomass yield on substrate (g DCW / g substrate). Determines feed rate required for exponential growth.
- Yo/s — oxygen yield on substrate (mol O2 / mol substrate). Determines oxygen demand and whether your bioreactor's kLa is sufficient.
- Yp/s — product yield on substrate (g product / g substrate). Determines process economics and overall efficiency.
These values are not fundamental constants—they vary with growth rate, temperature, media composition, dissolved oxygen, and metabolic state. The tables below provide typical ranges under standard aerobic conditions to serve as starting points for process design.
2. Biomass Yield on Substrate (Yx/s)
| Organism | Substrate | Yx/s (g/g) | Conditions | Source |
|---|---|---|---|---|
| E. coli (K-12, BL21) | Glucose | 0.40–0.50 | Aerobic, 37°C, μ < 0.3 h−1 | Shiloach & Fass 2005 |
| E. coli (K-12, BL21) | Glycerol | 0.45–0.55 | Aerobic, 37°C, no acetate overflow | Korz et al. 1995 |
| S. cerevisiae | Glucose (aerobic) | 0.45–0.50 | Aerobic, 30°C, below Crabtree threshold | Verduyn et al. 1991 |
| S. cerevisiae | Glucose (anaerobic) | 0.05–0.10 | Anaerobic, 30°C, ethanol production | Verduyn et al. 1990 |
| Pichia pastoris | Glycerol | 0.40–0.50 | Aerobic, 30°C, batch phase | Cos et al. 2006 |
| Pichia pastoris | Methanol | 0.30–0.40 | Aerobic, 30°C, induction phase | Cos et al. 2006 |
| CHO cells | Glucose | 0.15–0.25 | 37°C, chemically defined media | Xing et al. 2011 |
| Bacillus subtilis | Glucose | 0.35–0.45 | Aerobic, 37°C, minimal media | Dauner & Sauer 2001 |
| Lactobacillus spp. | Glucose | 0.10–0.20 | Microaerobic/anaerobic, lactic acid production | Hofvendahl & Hahn-Hägerdal 2000 |
| Corynebacterium glutamicum | Glucose | 0.40–0.50 | Aerobic, 30°C, non-producing strain | Wendisch et al. 2000 |
Mammalian cells are metabolically inefficient compared to microbes. CHO cells consume glucose primarily through aerobic glycolysis (the Warburg effect), converting much of it to lactate rather than routing it through the TCA cycle. This results in Yx/s values 2–4 times lower than bacteria or yeast on the same substrate.
E. coli produces acetate when growth rate exceeds ~0.3 h−1 on glucose (acetate overflow). This reduces Yx/s from 0.5 to as low as 0.25. Similarly, S. cerevisiae produces ethanol above its critical growth rate (~0.28 h−1 on glucose), dropping Yx/s dramatically. The key to maintaining high Yx/s is controlling the feed rate to keep μ below the overflow threshold.
3. Oxygen Yield (Yo/s)
Oxygen yield tells you how much oxygen is consumed per mole of substrate metabolized. This is critical for sizing your aeration system and determining whether your bioreactor's kLa can support the planned feed rate.
| Organism | Substrate | Yo/s (mol O2 / mol substrate) | Notes |
|---|---|---|---|
| E. coli | Glucose | 3.5–5.5 | Fully aerobic; lower at overflow conditions |
| E. coli | Glycerol | 3.0–4.5 | More reduced substrate, higher O2 per carbon |
| S. cerevisiae | Glucose (aerobic) | 4.0–6.0 | Below Crabtree threshold |
| Pichia pastoris | Glycerol | 3.5–5.0 | Batch/fed-batch growth phase |
| Pichia pastoris | Methanol | 1.0–1.5 | High O2 demand per gram substrate; heat generation is significant |
| CHO cells | Glucose | 1.5–3.0 | Partial oxidation; much glucose to lactate |
| Bacillus subtilis | Glucose | 3.5–5.0 | Fully aerobic |
| Corynebacterium glutamicum | Glucose | 3.0–4.5 | Lysine-producing strain consumes more |
Pichia on methanol deserves special attention: although Yo/s in mol/mol is lower than glucose, methanol has a much lower molecular weight (32 vs. 180 g/mol), so the O2 demand per gram of substrate is actually much higher. Methanol-fed Pichia fermentations are among the most oxygen-demanding processes in industrial biotechnology, often requiring O2 enrichment and/or elevated back-pressure.
4. Product Yield Examples (Yp/s)
| Product | Organism | Yp/s or qP | Typical Titer | Notes |
|---|---|---|---|---|
| Recombinant insulin | E. coli BL21(DE3) | qP = 0.05–0.15 g/g/h | 3–8 g/L (inclusion bodies) | IPTG induction; requires refolding |
| Monoclonal antibody (mAb) | CHO-K1 / CHO-DG44 | qP = 20–80 pg/cell/day | 3–10 g/L | 14-day fed-batch; modern high-producing clones |
| Ethanol | S. cerevisiae | Yp/s = 0.46–0.48 g/g | 80–120 g/L | Near theoretical max (0.51 g/g); anaerobic |
| L-Lysine | C. glutamicum | Yp/s = 0.20–0.30 g/g | 120–170 g/L | Fed-batch, 48–72 h; industrial strains |
| Citric acid | Aspergillus niger | Yp/s = 0.80–0.95 g/g | 150–200 g/L | Very high conversion; surface or submerged culture |
| Lactic acid | Lactobacillus spp. | Yp/s = 0.85–0.95 g/g | 100–180 g/L | Homofermentative strains; near theoretical |
| Recombinant protein (secreted) | Pichia pastoris | qP = 0.01–0.10 g/g/h | 1–12 g/L | Methanol induction; protein-dependent |
For growth-associated products (ethanol, primary metabolites), Yp/s is relatively constant and you can predict titer from substrate consumption. For non-growth-associated products (mAbs, secondary metabolites), production rate depends on viable cell density and time, making qP (specific productivity) the more useful parameter.
5. How to Use Yx/s in Feed Calculations
The exponential feed equation for a fed-batch fermentation is derived directly from Yx/s:
where:
F(t) = feed rate at time t (L/h)
μset = desired specific growth rate (h−1)
X0 = cell density at start of feeding (g/L)
V0 = volume at start of feeding (L)
Yx/s = biomass yield on substrate (g/g)
Sf = substrate concentration in feed (g/L)
If Yx/s is overestimated, the feed rate will be too low and growth will be nutrient-limited. If Yx/s is underestimated, substrate will accumulate—potentially triggering overflow metabolism (acetate in E. coli, ethanol in yeast) or inhibiting growth.
In practice, it is safer to underestimate Yx/s slightly (feed a bit more) and monitor dissolved oxygen and off-gas CO2 to detect overflow. Many processes use a DO-stat or pH-stat strategy that adjusts feed rate in real-time based on metabolic feedback rather than relying solely on a fixed Yx/s value.
Design Your Feed Profile
Enter your organism, Yx/s, target growth rate, and starting conditions. Get the complete exponential feed profile with time-stamped flow rates.
Fed-Batch Calculator →For more on feeding strategy design and how to handle overflow metabolism, see:
- Fed-Batch Feeding Strategies: Exponential vs. Linear
- Fermentation Economics Calculator — see how yield coefficients directly impact cost of goods.