Fire Damage Assessment in Process Plants: How In-Situ Metallography Decides What Runs Again

Paresh Haribhakti

Sep 4

10 min read

Updated: 5 days ago

After the flames are out, the hardest question begins: which equipment is safe to return to service? The metal remembers every temperature it saw, and the microstructure is where it writes the record down.

AT A GLANCE
The problem	After a fire in a refinery, petrochemical, fertiliser or chemical plant, hundreds of equipment items sit under one question: reuse, repair or replace. Fire damage assessment is the discipline that answers it with evidence. Replace everything and the outage runs into months and crores. Return one heat-damaged vessel and the next failure happens at operating pressure.
The framework	API 579-1/ASME FFS-1 Part 11, Assessment of Fire Damage: map the plant into heat exposure zones, evaluate material condition, then assess each component's fitness for continued service.
The evidence	Steel records its thermal history in the microstructure: pearlite degradation and spheroidisation, re-arranged ferrite and pearlite after crossing the transformation temperature, untempered martensite and quench cracks where firewater hit hot steel, sensitisation and liquid metal embrittlement in stainless steels. In-situ metallography reads all of it on the standing equipment, paired with hardness surveys.
The decision	Per component: return to service as-is, re-rate, repair, or replace, each verdict carried by microstructural and hardness evidence that stands up to insurers, regulators and the plant's own engineers.
The bench	TCR Advanced has executed major post-fire damage assessments across India's refining, petrochemical, fertiliser and chemical sectors, reading fire-exposed microstructures against a database of 1,50,000+ micrographs and 10,000+ investigations.

Why it matters

A process plant fire ends twice. The first ending is the flames going out. The second, slower ending is the reusability decision: every vessel, column, exchanger, pipe rack, storage tank and structural member in the affected area now carries an unknown thermal history, and someone must certify, in writing, which of them can safely hold pressure again. Fire damage assessment is that certification, and it is where fortunes swing. Scrapping sound equipment wastes crores and stretches the outage by the delivery time of replacement vessels, often 6–18 months for critical items. Returning damaged equipment gambles the plant on steel whose strength, toughness or corrosion resistance the fire has quietly taken away.

The difficulty is that the fire leaves no instrument record. Thermocouples do not cover pipe racks, flames move, and radiant exposure varies over metres from severe to trivial. The only witness present at every location, for the full duration, at the exact metal temperature that matters, is the steel itself. Steel is a recording medium: once carbon steel or alloy steel passes through the critical temperature regimes, the exposure leaves signatures in the microstructure, and those signatures survive after the fire is out. In-situ metallography is how the record is read on standing equipment, without cutting anything, and it is the reason a fire damage assessment can be precise instead of precautionary.

The framework: API 579-1/ASME FFS-1 Part 11 and heat exposure zones

API 579-1/ASME FFS-1, the fitness-for-service standard, dedicates Part 11 to the assessment of fire damage, and it names metallographic examination among the material evaluation methods on which the reusability judgement rests. The logic runs in three steps. First, zone the plant: using physical evidence, every location is assigned a heat exposure zone from I to VI, an indicative scale of the temperature the area saw. Second, evaluate material condition in the higher zones, principally by field hardness surveys and in-situ metallography. Third, carry components with evidence of degradation into a fitness-for-service assessment, dimensional checks for distortion, and a documented run, re-rate, repair or replace verdict.

Zone	Indicative exposure	Typical evidence and material concern
I	No heat exposure	Outside the affected area; no assessment beyond documentation
II	Ambient to about 65°C	Smoke and water exposure only; cleanliness and contamination checks, not metallurgy
III	About 65–205°C	Light heat: scorched paint, softened plastics; metals unaffected, non-metallics (gaskets, seals, instruments) suspect
IV	About 205–425°C	Moderate heat: charred paint, melted lead and solder; carbon steel largely unaffected metallurgically, cold-worked and heat-treated items (bolting, springs) suspect
V	About 425–730°C	Heavy heat: melted zinc (~420°C) and aluminium (~660°C); spheroidisation and tempering effects in carbon steel, sensitisation range for austenitic stainless steel, strength loss and distortion under load
VI	Above about 730°C	Severe heat: softened glass, melted copper alloys higher still; carbon steel crosses the transformation temperature, microstructure re-forms, quench effects where firewater struck hot steel

The zone boundaries are anchored by temperature indicators the fire leaves behind: lead and solder melt around 327°C and below, zinc coatings on galvanised gratings and cable trays melt near 420°C, aluminium instrument housings and ladders near 660°C, glass softens around 700–750°C. A melted zinc handrail beside an unblistered painted vessel is a boundary drawn in physical evidence. Zoning by indicators is fast and covers the whole plant, but it bounds the exposure of an area, not the condition of a component. The component-level verdict belongs to the metal, and that is where the microstructure takes over.

The microstructure as a thermal witness: carbon and low-alloy steels

Carbon steel's service microstructure, ferrite grains and lamellar pearlite colonies, is a known starting state. Fire exposure moves it through a sequence of changes, each tied to a temperature regime, so the observed state read from a replica places the metal temperature within a band the way a thermometer would have, had one been installed:

The carbon steel temperature ladder in a fire. The microstructural state read from an in-situ replica places the metal temperature within a band, location by location.

Below about 425°C, plain carbon steel is metallurgically untouched. The replica shows the original ferrite and pearlite, and the component's case closes quickly, which is exactly the point: a defensible clean bill is as valuable as a defensible condemnation, because it is what lets the undamaged majority of the plant restart.

Between about 425°C and the A1 transformation temperature near 727°C, the pearlite begins to degrade. The lamellar cementite breaks down and coarsens into spheroids, the same spheroidisation that takes decades at service temperatures, compressed into hours at fire temperatures. Hardness falls with it. The degree of spheroidisation on the replica, from partly broken lamellae to fully spheroidised carbides, indicates both the temperature reached and the time spent there, and the paired hardness reading quantifies the strength penalty for the fitness-for-service check.

Above A1, the steel begins transforming to austenite, and the original microstructure is erased. What forms on cooling depends entirely on the cooling path. Slow cooling in still air after the fire dies leaves a re-arranged, re-nucleated ferrite and pearlite structure, often with grain growth from the austenite excursion; faster cooling from well above A3 writes a Widmanstätten structure of acicular ferrite plates. Either way, the re-formed structure no longer matches the original heat treatment certificate, and properties must be re-established by evidence, not assumed from the mill certificate.

Where firewater struck steel that was above the transformation regime, the cooling path is a quench. Carbon and alloy steel transform to untempered martensite, hard, brittle and often carrying quench cracks from the transformation stresses. A hardness survey finds these patches immediately, readings above 400 HV against a base of under 200 HV for ordinary carbon steel, and the replica confirms the acicular martensitic structure and reveals any cracking. These local hard zones are the most dangerous product of a fire, because they sit in a component that may look completely undamaged.

Fire damage signatures from the TCR Advanced field replication archive, all readable on standing equipment by in-situ metallography.

Stainless steels and the special threats a fire brings

Sensitisation. Austenitic stainless steel held anywhere in the 425–815°C band, which a Zone V fire delivers over large areas, precipitates chromium carbides on its grain boundaries. The vessel afterwards looks perfect: dimensions unchanged, surface bright under the soot, hardness barely moved. But the grain boundaries are now chromium-depleted, the intent of the alloy's corrosion resistance is lost along them, and the first wet service exposure can open intergranular corrosion or intergranular stress corrosion cracking. This is the fire damage that inspection by eye cannot find at all; on a replica, the carbide-decorated boundaries are unmistakable.

Liquid metal embrittlement. Process plants are full of galvanised steel: gratings, cable trays, handrails, ladders. Above about 420°C the zinc melts, drips and runs onto whatever sits below, and molten zinc in contact with hot austenitic stainless steel penetrates the grain boundaries and cracks it, the liquid metal embrittlement mechanism recognised in API RP 571. Molten copper from cabling attacks similarly at higher temperatures. The assessment therefore walks the fire scene looking up as well as down: stainless lines beneath melted galvanised structures are sampled for replication whether or not they look damaged, because zinc penetration is a metallographic finding, not a visual one.

Quench cracking and hard zones. Wherever firewater met hot steel, carbon, low-alloy or martensitic stainless, the hardness survey and replica hunt for untempered martensite and its cracks, as described above. Air-hardening alloy grades are the most susceptible; a fire event followed by deluge is exactly the overheat-and-quench cycle these steels must never see.

The quietly weakened. Cold-drawn bolting, springs, strain-hardened stainless and precipitation-hardened components lose their designed strength at temperatures far below anything that changes a pressure vessel's plate, from about 205°C upward. They rarely look damaged. Fire damage assessment scope includes them as a class, and hardness testing screens them in minutes apiece.

From replica to reusability: how the assessment runs

The field programme is built for speed, because every day of assessment is a day of outage. The walkdown and temperature-indicator survey produce the zone map, usually within days even for a large fire area. Hardness surveys then grid the Zone V and VI equipment, a fast, quantitative screen that flags both the soft (spheroidised, over-tempered) and the hard (quenched) anomalies against unexposed reference locations on the same component. In-situ metallography follows at the flagged locations and at mandatory points, welds and heat-affected zones, stainless steel under melted zinc, the hottest-zone pressure boundaries, reading the actual microstructural state that the hardness numbers imply. Replication is non-destructive, needs no post-repair, and multiple windows run in parallel, so a component's material verdict is typically available within days of access, not weeks.

Each component then closes into one of four verdicts under the API 579-1/ASME FFS-1 framework: return to service as-is, where microstructure and hardness match the unexposed state; re-rate or run with an FFS justification, where degradation is real but bounded, with the measured hardness anchoring the strength used in the calculation; repair, where damage is local and removable; or replace, where transformation, embrittlement or cracking has consumed the component's integrity. The evidence file behind each verdict, zone map, indicator photographs, hardness grids, micrographs, is precisely what insurers, jurisdictional inspectors and the plant's own management need to release the restart, and it is why the metallurgical record matters as much as the verdict itself. In TCR Advanced's AiOM deployments, the fire assessment record also becomes each item's new baseline, because a fire-exposed plant restarts with a different metallurgical history than it shut down with.

The first days after the fire: protect the evidence

Do not scrap, sandblast or repaint yet. Melted zinc runs, blistered paint, charred insulation and distorted brackets are the temperature indicators the zoning depends on. Cleaning the scene before it is mapped erases the fire's thermometer.

Photograph and map early. Record what melted and what did not, location by location, and where firewater and deluge actually fell, because the quench-crack hunt follows the water, not the flames.

Hold heat-treated and cold-worked items as a class. Bolting in Zone IV and above is cheap to replace and expensive to trust; segregate it until screened.

Bring the metallurgy in at the start, not the end. An assessment that begins with the zone map and the microstructure shortens the outage twice over: the undamaged majority restarts on evidence instead of waiting on doubt, and the truly damaged minority is found before it is bolted back into a live plant.

"After a fire, every painted surface tells you what the paint saw. Only the microstructure tells you what the steel saw. We have signed reusability decisions on thousands of fire-exposed items, and the rule has never changed: zone by the indicators, judge by the metal."

Paresh Haribhakti, Managing Director, TCR Advanced Engineering

Frequently asked questions

What is fire damage assessment under API 579-1/ASME FFS-1?

Part 11 of API 579-1/ASME FFS-1 is the fitness-for-service methodology for equipment exposed to fire. The plant is mapped into heat exposure zones I to VI using physical temperature indicators, material condition in the hotter zones is evaluated by hardness testing and in-situ metallography, and each component receives a documented verdict: return to service, re-rate, repair or replace.

Why is metallography central to deciding equipment reusability after a fire?

Because the microstructure is the only temperature record present at every location for the whole fire. Carbon and alloy steels leave signatures of their exposure, pearlite degradation and spheroidisation from 425°C up, a fully re-formed ferrite and pearlite arrangement once the transformation temperature near 727°C is crossed, and untempered martensite where firewater quenched hot steel. Reading these on a replica bounds the metal temperature and the property loss, component by component.

Can fire-damaged equipment look completely undamaged?

Yes, and those are the dangerous cases. Austenitic stainless steel sensitised in the 425 to 815°C band looks perfect while its intergranular corrosion resistance is gone. Quenched hard zones sit inside visually sound components. Cold-worked bolting loses strength from about 205°C with no visible change. All three are found by hardness surveys and in-situ metallography, not by visual inspection.

What is liquid metal embrittlement in a plant fire?

Galvanised gratings, cable trays and handrails shed molten zinc once the fire exceeds about 420°C. Molten zinc contacting hot austenitic stainless steel penetrates its grain boundaries and cracks it, a mechanism recognised in API RP 571; molten copper from cabling behaves similarly at higher temperatures. Stainless lines below melted galvanised structures are replicated during assessment even when they look sound.

Is the assessment destructive? How fast can it run?

In-situ metallography removes less than about 0.5 mm from a small prepared window and needs no repair, so pressure boundaries are assessed without cutting or welding. Zoning takes days, hardness surveys screen equipment in minutes per location, and replication runs in parallel across the plant, so material verdicts on critical items typically follow within days of access.

Which industries use fire damage assessment of this kind?

Refineries, petrochemical complexes, fertiliser plants and chemical process industries are the principal users, wherever a fire touches pressure equipment, piping, storage tanks or structural steel. The same API 579-1/ASME FFS-1 Part 11 framework, hardness surveys and in-situ metallography apply across all of them, and TCR Advanced has executed such assessments across each of these sectors in India.

About the author

Paresh Haribhakti is Managing Director of TCR Advanced Engineering Pvt. Ltd., Vadodara. A metallurgical engineer from M.S. University of Baroda, he is co-author of an ASM International monograph on boiler tube failure analysis (2018), a contributor to ASM Handbook Volume 11A (2021), and the architect of the AiOM asset integrity platform. He has led 9,000+ industrial failure and integrity investigations, including major post-fire damage assessments across India's process industries.