In most labs, OD600 is the reading used to determine harvest timing, the induction point, growth rates, which medium moves forward, when to start the feed, and whether the 100 L run behaved like the 5 L run. Every one of those decisions is read off how the culture grows over time. This article covers what OD600 reports, where it holds up, and where it misleads.
What is OD600
Optical density at 600 nanometres (OD600) is the most common method for tracking bacterial cell concentration in liquid culture. A spectrophotometer passes light through a cuvette of culture broth and measures the fraction that reaches the detector. The number reported is called absorbance, but in a turbid bacterial suspension the signal is dominated by light scattering, not absorption.[1][2] More cells, larger cells, and more particles in the path all increase the reading.
The wavelength of 600 nm is a practical choice rather than a fundamental one. It sits in a window where most bacterial cultures scatter light strongly while pigments and common medium components absorb relatively little.[1] That tradeoff between sensitivity and chemical specificity is why 600 nm became the de facto standard, even though wavelengths from 540 nm to 660 nm are sometimes used.
OD600 is fast, cheap, and ubiquitous. Every cultivation lab has the equipment, every protocol quotes it, and every microbiologist knows the readout. What it gives you is a turbidity-based proxy for biomass — not a direct count of cells.
Applications of OD600
OD600 has two dominant uses in cultivation work.
- Growth curves. Repeated readings through a culture, taken every few minutes to every few hours depending on the organism, are used to build a growth curve. The curve is then read for the lag, exponential, stationary, and decline phases of bacterial growth.
- Culture standardisation. OD600 is used to dose bacteria into downstream experiments at a target density, ensuring that experimental replicates start with comparable cell concentrations. The same logic applies when seeding fermenters at a defined inoculum.
The reason OD600 is so widely used is operational: a measurement takes under a minute and costs nothing per sample beyond the cuvette. The downside is what the number actually represents. OD600 also suits a first pass across many candidates, such as a 96-well plate where the job is to sort out the worst offenders.
Considerations and limitations
OD600 is a biomass-related signal, not a cell count, and the gap between the two has practical consequences.
OD depends on cell size and morphology. Two cultures with identical cell numbers but different cell sizes will give different OD values. Stress, nutrient limitation, and approaching stationary phase all shift cell size, so OD can keep changing while the cell count is steady.[3][4]
OD600 adds cell number and cell size together, and nothing in the reading lets you separate them afterwards. A measurement that reports the two separately, from one reading, gives you both. Impedance flow cytometry works this way. Measuring bacterial cell size covers what the size signal carries and how to read it.
OD does not distinguish cells from other particles. Insoluble medium components, antifoam droplets, precipitates after pH or temperature shifts, and cell debris from lysis all add to the reading.[5] OD cannot tell them apart from cells. Some media defeat optical density entirely, which we cover in OD600-incompatible matrices.
OD does not distinguish cells by physiological state. Microbiology recognises a continuum of states between active growth and full lysis: cells that are dividing, cells that have stopped dividing but remain metabolically active, viable but non-culturable cells, cells with damaged membranes, and lysed debris.[6] OD600 measures turbidity, and any cell, particle or fragment contributes to that signal regardless of which state it is in. After lysis, the released debris continues to contribute to some extent until it settles or is degraded.
OD600 reports a relative number. Without a calibration curve specific to the organism and the instrument — and revalidated when the medium or process conditions change — OD600 is not convertible to cells/mL or any standard biological unit.[4][7] We walk through the practicalities in building an OD600-to-CFU calibration curve.
OD600 has a narrow linear range. Above an absorbance of roughly half a unit in a standard 1 cm cuvette, the relationship between signal and concentration becomes non-linear, and dilution is required to stay in range.[3]
OD readings are instrument-dependent. Spectrophotometer geometry, detector design, and bandpass differ between models. Two instruments will give different OD values for the same sample unless they are cross-calibrated.[7]
OD600 cannot compare cultivations against each other
OD600 works as a signal within one unchanged cultivation. It does not survive a comparison between cultivations.
Even after blanking the medium, OD600 still depends on cell size and intracellular composition. Both properties are shaped by the medium, the strain, and the cultivation conditions a cell grows under. A medium that produces cells with denser cytoplasm or more storage compounds reads higher than a medium producing the same cell count with leaner cells. Two strains with different cell sizes will give different OD curves at the same true cell concentration. Blanking corrects for the optics of the broth itself. It does not correct for what the broth has done to the cells. Screen growth media works through a media comparison where the OD ranking and the cell-count ranking disagree.
There is no single calibration factor for OD600 that survives a change in medium or strain, because the OD-to-cell ratio depends on cell properties that the new condition has changed.
Most importantly, the error is silent. The OD curve looks clean, the ranking is internally consistent, the results of a single run are reproducible, and the ranking can still be wrong, with no diagnostic from inside OD that tells you when.
Alternatives
Several methods coexist with OD600 in cultivation labs. Each answers a slightly different question.
The methods that count cells directly have existed for a long time. They have stayed off the growth curve because they are too slow or too laborious to run as a time series, so OD600 has kept the job by default.
| Method | What it measures | Strengths | Limitations |
|---|---|---|---|
| CFU plating | Cells competent to form colonies on a defined medium | Direct, biologically meaningful, regulatory standard for QC release | Slow (24–72 h or longer), labour-intensive, misses viable but non-culturable cells, undercounts aggregates |
| Dry cell weight | Total dried biomass per volume | Calibrated mass unit, used for yield calculations | Slow (hours), destructive, insensitive at low concentrations, biomass is not cell number |
| Fluorescence flow cytometry | Per-cell scatter and fluorescence; can include viability and metabolism dyes | Rich per-cell information | Expensive; complex sample prep; primarily designed for eukaryotes; undercounts aggregates |
| BactoBox (impedance flow cytometry) | Every whole bacterial cell in the sample, counted one cell at a time and reported as cells/mL, together with the average size of those cells, from the same reading. See Understanding BactoBox cell counts and Measuring bacterial cell size | Cell number and cell size reported separately from one reading, ~2 minutes per sample, benchtop format | Electrical signal only — no versatile tags or dyes; undercounts aggregates and reads them as larger cells without proper sample preparation |
The fast method is indirect, and the direct ones have been slow. That is the trade the field has lived with.
Common behaviour of OD600
Common questions and answers.
Why does my OD600 keep rising after my cells have stopped dividing?
Cells continue to change size and refractive properties for a while after division has slowed or stopped. Depending on the organism and the limitation that triggered slowdown, cells may shrink, swell with storage granules such as polyhydroxyalkanoates or glycogen, or change shape.[8][9] Each of those changes alters how the cell scatters light at 600 nm even though no new cells have been added.[1][3] The result is an OD600 that drifts upward — or downward — even though the cell count has plateaued. The OD plateau and the moment cell division actually ends rarely line up exactly, which is the subject of identifying the harvest point.
Can OD600 tell me whether my cells are alive or dead?
No, but the more accurate question is what definition of "alive" the experiment requires, because microbiology recognises several distinct definitions and OD600 cannot answer any of them.[6] Cells can be actively dividing, non-dividing but metabolically active, viable but non-culturable, structurally intact but metabolically inactive, or partially lysed. Each of these states still contributes to turbidity to varying degrees, because turbidity reflects the bulk presence of light-scattering material in the broth, not any individual cell's state. A flat or slowly declining OD tail can therefore hide a culture that has lost culturability without losing structure, a population of whole cells that is dropping, or a culture beginning to lyse. Asking whether cells are alive requires deciding which definition matters for the decision at hand — culturable, metabolically active, structurally intact, or membrane-intact — and choosing a method that targets that specific definition.
Why does my OD600 jump after I add antifoam, inducer, or a base correction?
Many additions introduce particulates, precipitates, or refractive-index shifts that change the OD signal independently of the cell population. Antifoams form droplets that scatter light. pH corrections can precipitate medium components. Inducer stocks sometimes carry insolubles. The OD jump is real, but it is not a change in the cells.
Why is my OD600 different on a different spectrophotometer?
OD600 is not a standardised quantity. Instruments differ in optical path geometry, detector design, and spectral bandpass, so the same sample can read differently between machines. A 2020 inter-laboratory study across 244 laboratories found that calibration against serial dilutions of silica microspheres allows OD-derived cell counts to be compared across instruments, but without such calibration, OD readings cannot be meaningfully compared between instruments.[7]
Closing
OD600 is fast and convenient, and within one unchanged cultivation it will tell you whether a run is behaving like the last one. It will not tell you whether one cultivation grew better than another. The reading adds cell number and cell size together, and both of them change when the medium or the conditions change, so two runs can sit on the same OD curve with different cultures behind them.
Measured separately, and often enough to follow the culture through the run, count and size give a growth curve you can compare with another one.
References
- Myers JA, Curtis BS, Curtis WR. Improving accuracy of cell and chromophore concentration measurements using optical density. BMC Biophys. 2013;6:4. https://doi.org/10.1186/2046-1682-6-4
- Koch AL. Turbidity measurements of bacterial cultures in some available commercial instruments. Anal Biochem. 1970;38(1):252-9. https://doi.org/10.1016/0003-2697(70)90174-0
- Stevenson K, McVey AF, Clark IBN, Swain PS, Pilizota T. General calibration of microbial growth in microplate readers. Sci Rep. 2016;6:38828. https://doi.org/10.1038/srep38828
- Mira P, Yeh P, Hall BG. Estimating microbial population data from optical density. PLoS One. 2022;17(10):e0276040. https://doi.org/10.1371/journal.pone.0276040
- Routledge SJ. Beyond de-foaming: the effects of antifoams on bioprocess productivity. Comput Struct Biotechnol J. 2012;3:e201210014. https://doi.org/10.5936/csbj.201210014
- Davey HM. Life, death, and in-between: meanings and methods in microbiology. Appl Environ Microbiol. 2011;77(16):5571-6. https://doi.org/10.1128/AEM.00744-11
- Beal J, Farny NG, Haddock-Angelli T, Rai V, Davies J, Patron N, et al. Robust estimation of bacterial cell count from optical density. Commun Biol. 2020;3:512. https://doi.org/10.1038/s42003-020-01127-5
- Åkerlund T, Nordström K, Bernander R. Analysis of cell size and DNA content in exponentially growing and stationary-phase batch cultures of Escherichia coli. J Bacteriol. 1995;177(23):6791-7. https://doi.org/10.1128/jb.177.23.6791-6797.1995
- Slaninova E, Sedlacek P, Mravec F, Mullerova L, Samek O, Krzyzanek V, et al. Light scattering on PHA granules protects bacterial cells against the harmful effects of UV radiation. Appl Microbiol Biotechnol. 2018;102(4):1923-31. https://doi.org/10.1007/s00253-018-8760-8