The equation has no linear segments. It is exponential.

Digging in a little further, it seems the standard practice is to plot your data on a sq. rt time (t

^{1/2}) axis. In this case, if there is a single dominating interparticle diffusion step, the data should be linear. If there are multiple rate-limiting steps, you will have more than one linear region in your plot.

See, for an example, Hwang et al. Phys. Chem. Chem. Phys., 2015,17, 21974-21981

https://pubs.rsc.org/en/content/articlelanding/2015/cp/c5cp03416gI suppose if your rate constants are very close together, it may be hard to deconvolute these even on a (t

^{1/2}) axis. Nevertheless, it is common to plot diffusion curves on a t

^{1/2} axis due to the frequency with which this shows up in solutions to the diffusion equation.