DURABLE integrates gel-droplet bacterial isolation, rapid carbon-steel bead screening, genome sequencing, metabolite profiling, and electrochemical validation to connect fuel-tank microorganisms with corrosion mechanisms. In laboratory experiments, 84 isolates were screened in approximately two days, with 42 corrosive and 42 noncorrosive isolates; selected chemicals accelerated corrosion in R2A medium or inhibited it in 0.5 M hydrochloric acid. Cultivable diversity was not assessed comprehensively, and validation of microbial consortia and field mitigation remains future work.

Some microbial chemicals help metal break down. Others slow the damage. Follow the clues from a rust signal to the chemistry behind it.

A tiny bead. A revealing change.

Corrosion is the chemical breakdown of metal. In this test, microbes grow beside tiny carbon-steel beads. A brown liquid signals rust. Choose a strain, then let the experiment unfold.

Illustrative animation of Figure 3b’s observations. Steel beads: 500 μm across. Time between observed days is schematic.

F32-44 is ready. Advance to day 2 to look for the rust signal.

to reveal corrosive activity, versus a conventional month-long screen

bacterial isolates screened from two fuel-tank biofilms

corrosive / noncorrosive isolates in the main Results

The signal is a beginning.

DURABLE connects a quick screen to the microbes’ identities and the chemicals they make. Its screening step is about 15 times faster; confirmation follows.

Porous gel droplets grow individual microbes while allowing chemical exchange.

Steel beads in small wells reveal a rust signal in about two days.

Selected strains meet larger steel pieces in a 34-day coupon experiment.

Genome sequencing and chemical profiling connect strains with their capabilities.

Purified compounds are tested directly for their effects on steel.

Change the chemical. Change the outcome.

Microbes make small chemicals called metabolites. The researchers identified candidates, then tested purified chemicals on steel. Explore the actual measured results, one compound and dose at a time.

Methanesulfonic acid

R2A nutrient medium

Numbers: supplementary Table S2, page 3. Surface and moving particles are illustrative, not a prediction of pit size or growth.

No added compound. Move the dose control to a tested concentration.

Corrosion current is an electrical signal of metal dissolution. A lower current means less corrosion in that test. Each chemical has its own matched control.

mM means millimoles per litre: how much chemical is dissolved. Every position is a dose the researchers tested.

The first entry is each compound’s matched control. Nicotinic acid and hypoxanthine have their lowest reported current at 5 mM. Values are transcribed from Table S2; no intermediate doses are estimated.

Look at the steel itself.

Electron microscopy makes the outcome visible. At 10 mM, the two accelerator tests show roughened steel. The four inhibitor tests show reduced damage compared with their acid control.

A microbe can make both kinds of chemistry.

The protective compound hypoxanthine also accumulated in corrosive F32-44, at lower levels than in the noncorrosive strains. The paper points to the overall chemical profile as the key to the outcome.

Connecting microbes with their chemistry could guide targeted corrosion control in fuel infrastructure. The next steps described by the researchers include testing communities of bacteria and fungi together.

DURABLE

DURABLE: A Workflow for Determining Corrosion-Driving and Protective Microbial Mechanisms

Aifen Zhou, Yuwen Li, Han Zhang, Evelyn Callaway, Song-I Han, Jason S. Lee, Kyongbum Lee, Arul Jayaraman, Paul de Figueiredo and Arum Han. Environmental Science & Technology 60(33), 23319–23332 (2026).

Source figures © 2026 The Authors, published by American Chemical Society, licensed under CC BY 4.0. Figures 1 and 3 are reproduced unchanged; Figure 6e is cropped and compressed as indicated. The graphical abstract is an AI-generated conceptual illustration based on source Figures 1, 3 and 6. The animated surfaces are teaching illustrations; displayed chemical currents are reported laboratory measurements.

The microbes came from petroleum-based Navy F76 diesel and hydroprocessed renewable-diesel storage tanks. The work connects isolated strains, laboratory steel tests and chemical analysis; the paper’s proposed mitigation strategies build on those experiments.

Source figure