The cracks run in straight lines that follow the steel inside the concrete. That is not settlement, and it gets worse by itself

There is a particular pattern of damage on older concrete buildings, and once you can recognise it you will see it everywhere in coastal and humid Indian cities. Cracks that run in long straight lines rather than wandering. Rust-coloured staining bleeding out of them. Patches where the surface has come away entirely, exposing a bar behind, and areas that sound hollow when you tap them.

Those straight lines are straight because they are following the reinforcing steel buried a few centimetres inside. This is not settlement, it is not shrinkage, and it is not the building shifting. It is the steel rusting and the rust having nowhere to go — and the reason it deserves an article of its own is that it is the one common building failure that accelerates by itself once it starts.

Concrete does not keep water out. It does something better

The intuition is that concrete protects steel by being a barrier — keeping air and water away from the bar. That is a small part of it and not the important part. Concrete is porous, and water and air do get in.

The real protection is chemical. The water inside concrete's pores is strongly alkaline, around pH 13, because of the calcium hydroxide produced when cement hydrates. At that alkalinity steel forms a passive film — an extremely thin, stable oxide layer that keeps the metal from dissolving. It is the same phenomenon that makes stainless steel stainless, arrived at by a completely different route: not by alloying the metal, but by keeping it in an environment where the film is stable.

So a reinforcing bar in sound concrete is not merely sheltered. It is chemically prevented from corroding, and it will sit there indefinitely, wet or not, as long as the chemistry holds.

Everything that goes wrong is a story about losing that alkalinity. There are exactly two ways.

Route one: the air quietly neutralises the concrete

Carbon dioxide from the atmosphere diffuses into concrete and reacts with the calcium hydroxide, converting it to calcium carbonate. This is carbonation, and it is not damage in itself — carbonated concrete is often slightly harder. What it does is drop the pore water's pH from about 13 to around 8 or 9, and below roughly pH 10 the passive film is no longer stable.

Carbonation advances as a front, moving inwards from the surface, and its depth grows roughly with the square root of time. That relationship is the single most useful fact in this subject, because of what it implies about design.

Doubling the concrete cover over the steel does not double the time before carbonation reaches it. It roughly quadruples it. Cover depth is therefore a far more powerful lever on service life than most people assume, and it is also the variable most often lost on site — a bar tied slightly out of position, a spacer omitted, a slab poured a little thin, and thirty years of design life quietly becomes fifteen.

Indian practice recognises this: IS 456 specifies nominal cover by exposure category, increasing from 20 mm in mild conditions up to 75 mm in extreme ones. Those numbers are not padding. They are a durability calculation, and they are the part of a drawing that a rushed site is most likely to treat as approximate.

Carbonation is the dominant mechanism inland, in urban and industrial air. It typically produces general corrosion along a length of bar, and therefore long, straight cracks tracking the reinforcement.

Route two: chloride gets to the bar and attacks it locally

Chloride behaves differently and is considerably nastier.

Chloride ions reaching the steel break down the passive film locally, in spots, even while the surrounding concrete is still perfectly alkaline. The result is pitting: intense, concentrated attack on small areas of the bar while the rest looks fine. A bar can lose a serious fraction of its cross-section at a pit with very little external warning, because the corrosion is not spread out enough to crack the cover early.

The chloride comes from four places, and three of them are avoidable:

• Marine atmosphere — airborne salt, which is why coastal buildings age differently from inland ones at the same age and grade.
• Sea sand or unwashed aggregate, which puts chloride into the concrete at the mixer. This is a real and persistent problem in Indian construction wherever river sand is scarce or expensive, and it is the worst version of the problem because the contamination is uniform and present from day one.
• Contaminated mixing water.
• Groundwater and de-icing salts, the last being largely irrelevant in most of India.

Codes limit the chloride content of concrete constituents for exactly this reason, and testing sand for chloride before it goes into a structural pour is cheap relative to what it prevents.

Why it accelerates

Here is the part that makes reinforcement corrosion different from most building defects.

Iron oxides occupy several times the volume of the steel they came from. A bar that corrodes inside concrete is therefore generating a product that does not fit in the space available, and it exerts real pressure on the surrounding cover — enough to crack it.

Once the cover cracks, carbon dioxide, water, oxygen and chloride all reach the bar far more easily than they did through sound concrete. Corrosion speeds up. More rust, more pressure, wider cracks, then spalling — the cover breaking away in pieces. And now the bar is exposed to the atmosphere directly.

That is a positive feedback loop. Most building defects are static or progress slowly and linearly; this one compounds. It is the reason the gap between "a few rust stains" and "structural repair" can be shorter than owners expect, and the reason early intervention is worth so much more than it appears.

Corrosion damage in concrete is the rare defect that pays for delay with interest. The cracking is not the disease appearing — it is the disease acquiring a faster route in.

Reading the damage

You can get quite far without instruments.

Cracks that run in long straight lines, parallel to each other and spaced regularly, are following bars. Settlement and shrinkage cracks wander, branch and change direction; corrosion cracks are as straight as the steel is.

Rust staining bleeding from a crack or a joint is direct evidence, though it can also come from tying wire or a chair left too close to the surface.

Hollowness on tapping. Tap suspect areas with a light hammer or the handle of a screwdriver. Sound concrete rings; concrete that has delaminated from a corroding bar underneath sounds dull and hollow. This is the single most useful field check available to a non-specialist, and it finds damage before anything is visible.

Location tells you a lot. Underside of balconies and chajjas, the soffit of slabs over bathrooms, parapets, columns near ground level, and anywhere water sits or runs. Corrosion needs moisture, so it appears first where the building is wettest.

For anything beyond that, the professional tools are a cover meter to find the bars and measure the cover actually achieved, a half-cell potential survey to map where corrosion is probable across a large area without breaking anything, carbonation depth testing with phenolphthalein on a fresh break, and chloride sampling by drilling dust at successive depths. Any competent structural consultant will offer these; a contractor who proposes to start chipping without them is guessing.

Why patch repairs so often fail

This is the most valuable thing in this article for anyone about to spend money on repairs.

The intuitive fix is to break out the damaged concrete, clean the bar, and patch it with fresh mortar. Done that way alone, it frequently causes a new failure within a couple of years, in a ring immediately around the patch. The mechanism has a name — the incipient anode or halo effect — and it follows from the electrochemistry.

The fresh patch is highly alkaline, so the steel inside it is repassivated and stops corroding. But that steel is electrically continuous with the bar in the surrounding old concrete, which is carbonated or chloride-contaminated. The repaired section becomes a cathode and drives corrosion in the adjacent unrepaired steel, which becomes the anode. Having removed the corrosion at one spot, the repair has intensified it just outside the patch. The area ratio problem again: a large cathode feeding a small anode.

This is why serious repair specifications do more than patch. They break out to a defined distance beyond visibly damaged concrete and behind the bar; they may add corrosion inhibitors or sacrificial anodes at the patch edges; and for chloride-contaminated structures they consider electrochemical treatments — realkalisation, chloride extraction, or impressed-current cathodic protection, which is the one method that reliably arrests corrosion regardless of how much chloride is in the concrete.

None of that is exotic, and skipping it is why so many buildings are repaired repeatedly.

What a coating can and cannot do here

Coatings have a genuine role, and it is narrower and more specific than the marketing suggests.

An anti-carbonation coating applied to a concrete surface is designed to resist carbon dioxide diffusion, slowing the advance of the carbonation front and buying service life on a structure that is not yet damaged. What makes it a real engineering product rather than paint is a tension in the specification: it must be a high barrier to carbon dioxide and simultaneously a low barrier to water vapour, so that moisture already inside the concrete can still escape. Trap the water in and you have made the problem worse. The performance is measured as an equivalent air-layer thickness for each gas, under the European standards for masonry and concrete coatings, and a supplier who cannot quote both numbers is selling paint.

Two honest limits. A surface coating slows carbonation; it does very little about chloride already at the bar, and nothing about steel that is already corroding. And it is preventive maintenance — most useful applied to a sound structure, largely pointless applied over active damage as cosmetic cover, which is exactly how it is often sold.

The same logic that ran through surface preparation applies with more force here: the coating is the last step and the smallest part of the job. Fixing where water is getting in — terrace waterproofing, blocked drainage, a leaking parapet detail — usually does more for a reinforced concrete building than anything applied to its surface.

On the product side, Smart Warrior Coatings, part of Reinste Nano Ventures — which, in the interest of disclosure, belongs to the same group as this publication — lists industrial corrosion clear coats as a distinct line from its surface and hygiene products, and that separation is the right shape. The question to put to any supplier here is a specification question, not a product one: which exposure condition, which diffusion figures, on what substrate condition, and what does the technical data sheet say it does not do.

Why it matters for students and researchers

Reinforcement corrosion has an unusual property for a research area: it is simultaneously well understood in principle and badly predicted in practice.

Service-life modelling is the clearest gap. The square-root-of-time carbonation model and the diffusion models for chloride ingress are decades old and calibrated largely against European exposure data. Indian conditions — higher temperatures accelerating every reaction, monsoon wet-dry cycling rather than steady exposure, and a coastal population density that puts an enormous amount of building stock in a marine atmosphere — are not what those coefficients were derived from. Recalibrating them against Indian field data is patient, unglamorous, genuinely valuable work that needs cores and time rather than expensive instruments.

Non-destructive assessment is the second. Half-cell potential mapping indicates probability, not rate; resistivity and polarisation resistance help but remain interpretive. Better field methods for estimating actual corrosion rate — and cheap embedded sensors that report on a bar's condition over decades — would change maintenance from reactive to planned across an entire building stock.

On materials, the live directions are worth knowing. Corrosion-resistant reinforcement — stainless, galvanised, fusion-bonded epoxy coated, and composite bars — each with its own trade-offs and, in the epoxy case, a well-documented history of disappointing field performance where the coating was damaged in handling. Migrating corrosion inhibitors that penetrate hardened concrete. Supplementary cementitious materials such as fly ash and slag, which densify the pore structure and slow chloride ingress while, in some cases, reducing the alkalinity reserve — a genuine trade-off between the two failure routes that deserves more attention than it gets.

And there is a straightforward field-survey question worth someone's thesis: what cover depth is actually being achieved on Indian sites, as against what is specified? A cover meter, a sampling frame and a few months would produce a number that matters enormously and, as far as published work goes, barely exists.

Frequently asked questions

How do I tell corrosion cracks from settlement cracks?

Geometry, mostly. Corrosion cracks follow the reinforcement, so they run in long straight lines, often parallel and regularly spaced, and they concentrate on the wet faces of a building — balcony soffits, chajjas, parapets, bathroom slabs. Settlement and shrinkage cracks wander, branch, change direction and often run diagonally. Rust staining and a hollow sound on tapping settle it. If in doubt, tap a wide area: delamination is audible long before it is visible.

Is a hairline crack in my slab dangerous?

Not necessarily — concrete cracks for many benign reasons, including drying shrinkage and thermal movement. What warrants attention is a crack that is straight and follows the line of the steel, one that is stained rust-coloured, one that sounds hollow around it, or one that is getting wider over months. Structural judgement on any specific crack belongs to a qualified engineer, but those four signs are the ones worth acting on rather than watching.

Does waterproofing the terrace help the columns below?

Often substantially, yes, and it is routinely underestimated. Corrosion needs moisture, and a great deal of the water reaching reinforcement in an Indian building arrives from above — a failed terrace membrane, ponding from blocked outlets, a parapet junction that leaks. Stopping the water at source slows every corrosion process happening below it, and it is usually cheaper than the structural repair it defers.

Why did the repair fail again after two years?

Most likely the incipient anode effect. Patching one damaged area repassivates the steel inside the patch, which then drives corrosion in the still-contaminated steel just outside it, producing fresh damage in a ring around the repair. Avoiding it requires breaking out well beyond the visible damage and behind the bar, and on chloride-contaminated structures usually some additional measure — inhibitors, sacrificial anodes or cathodic protection — rather than mortar alone.

Is higher-grade concrete the answer?

It helps, but less than cover does, and the two are often confused when a specification is being value-engineered. A denser, lower-permeability mix slows both carbonation and chloride ingress, and supplementary materials such as fly ash or slag improve that further. But carbonation depth grows with the square root of time, so the distance the front has to travel is the dominant term — which means correct, properly achieved cover, with spacers actually used and bars actually where the drawing puts them, buys more durability than upgrading the grade while letting the cover slip.