Doubling Time from Specific Growth Rate

Two ways of saying the same thing, and both are needed. The specific growth rate $\mu$ is a reciprocal time: it is the fractional increase in biomass per unit time, and it is the form that goes into equations. The doubling time is the interval in which the population doubles, and it is the form people say out loud. The bridge is $t_d = \ln 2/\mu$, and the constant is 0.693.

Why $\mu$ rather than $t_d$ in the mathematics? Because $\mu$ is additive and $t_d$ is not. A culture growing at $\mu$ while dying at a specific death rate $k_d$ has a net rate of $\mu - k_d$; a chemostat washing out at dilution rate $D$ has a net biomass rate of $\mu - D$. Subtract, add, put it in an exponent, and the algebra stays simple. Doing the same thing with doubling times requires reciprocals at every step and produces expressions nobody can read. Every mass balance in bioprocess engineering is therefore written in $\mu$.

Why $t_d$ when talking to another person? Because a doubling is something you can picture and a reciprocal hour is not. "This culture doubles every twenty minutes" lands; "this culture has a specific growth rate of 2.08 per hour" does not, even though it is the identical fact. Both numbers belong on the same page of a notebook.

A short table for scale, because the range is enormous and it explains a great deal about how bioreactors differ. E. coli in rich medium at 37 °C doubles in about 20 minutes, which is $\mu \approx 2.1$ h⁻¹. A brewing yeast on glucose is nearer 90 minutes, $\mu \approx 0.46$ h⁻¹. A filamentous fungus might take 4 to 6 hours. A CHO cell line used for protein production takes around 24 hours, $\mu \approx 0.029$ h⁻¹. And an autotrophic nitrifier in a wastewater plant can take a day or more, which is exactly why nitrification is the first thing lost when a plant is overloaded or gets cold — the organisms responsible cannot grow fast enough to replace what washes out. Nearly three orders of magnitude separate the fastest bacterium from the mammalian cell, and every factor of ten is a reason their vessels, their sterility requirements and their economics look nothing alike.

Two errors to watch. The first is dividing by 2 instead of by $\ln 2$, which is off by 44% and has appeared in more student reports than anyone would like. The second is unit slippage: $\mu$ in reciprocal hours gives a doubling time in hours, $\mu$ in reciprocal days gives days, and the two get mixed whenever a wastewater number meets a fermentation one.

One conceptual point, since it affects how the number should be read. This is a MEAN generation time for a population, not the life history of any individual cell. Cells in a culture divide over a broad distribution of intervals — some take half the mean, some twice — and the smooth exponential curve is the aggregate of a great many staggered divisions. That is why growth looks continuous at all, and it is why a synchronised culture, in which the cells have been forced to divide together, shows a visible staircase instead of a smooth line and drifts back to a smooth line within a few generations.