Microbial bioremediation

Almost every organic contaminant of concern has a known microbial degradation pathway. Petroleum hydrocarbons, chlorinated solvents, nitroaromatics, phenols, many pesticides: the enzymes have been characterised, the organisms are in culture collections, and in a shaken flask with the compound dissolved in water the reaction runs to completion. Field remediation nonetheless stalls, and it stalls for reasons that have nothing to do with degradative capability.

Bioavailability is the rate law

A bacterium takes up dissolved substrate. In soil, most of the contaminant is not dissolved. Hydrophobic compounds partition into soil organic matter, and the partition coefficient scales with hydrophobicity — above roughly log Kow 5, the aqueous fraction becomes small enough that the observed degradation rate is set by how fast the molecule leaves the organic phase, not by how fast a cell can process it. Contaminants also age: over months to years they diffuse into intraparticle nanopores and condensed organic domains from which desorption is slow and biphasic. This is why a site fits first-order desorption kinetics rather than Monod kinetics, why measured half-lives lengthen as remediation proceeds, and why bulk soil concentration overstates what is available. Surfactants and solvent flushing attack exactly this constraint, at the cost of mobilising the plume.

Redox zonation decides what is possible where

Microbial communities use electron acceptors in order of energy yield: oxygen, then nitrate, manganese(IV), iron(III), sulfate, and finally carbon dioxide. A hydrocarbon plume therefore organises itself into concentric zones, aerobic at the fringe and methanogenic at the core. The consequence is that a compound’s treatability depends on where it sits. Petroleum hydrocarbons oxidise readily but consume the oxygen supply, which in saturated soil is the limiting reagent. Highly chlorinated solvents such as PCE are the opposite case: they are already oxidised, resist aerobic attack, and are dechlorinated only reductively, which requires strongly reducing conditions and a hydrogen-releasing donor. That reaction commonly stalls at cis-1,2-dichloroethene and vinyl chloride — a more toxic and more mobile intermediate than the parent — unless organisms of the genus Dehalococcoides carrying the appropriate reductive dehalogenase genes are present.

Co-metabolism has no reward

Some transformations give the cell nothing. Methane monooxygenase and toluene dioxygenase will oxidise trichloroethene fortuitously, but the organism gains no carbon or energy from it, the epoxide product is damaging, and the contaminant competes with the growth substrate for the same enzyme. Co-metabolic treatment must therefore feed a primary substrate continuously, and increasing the substrate to grow biomass also suppresses the reaction being paid for.

Why bioaugmentation usually fails

Adding a cultured degrader to a site normally does nothing measurable. The introduced population meets protozoan grazing, phage, competition from a community already selected by that site’s chemistry, and physical conditions unlike those it was grown in; its numbers fall by orders of magnitude within weeks. Biostimulation succeeds where augmentation does not because it removes the actual constraint — an electron acceptor, a donor, nitrogen or phosphorus — and lets the resident community expand. The clear exception proves the rule: where a required functional guild is genuinely absent, as Dehalococcoides often is, bioaugmentation with a defined consortium works, and is standard practice.

The residual limit is thermodynamic. At very low concentrations the energy from degradation no longer covers maintenance, and the curve flattens above the target — often above a drinking-water standard.