Two jobs, same week, same equipment, same operative. A run of first-floor offices above the shops on the Euston Road side of Warren Street, and a detached house on Fortis Green Road in N10. The Euston Road glass had been cleaned nine weeks earlier and looked, from the pavement, as though nobody had touched it since the spring. The Muswell Hill glass had gone four months and needed the work far less. Clients see that difference and assume somebody skimped. What they’re looking at is a film with a chemistry all of its own, laid down at a rate that has almost nothing to do with how well the last clean was done.

What is traffic film made of?

Soot, but soot is the smallest part of it now. Diesel exhaust particulate has fallen steadily across London since the low emission zones went London-wide, and the greying has not fallen with it. The material coating the glass on any arterial route is mostly non-exhaust: brake wear, tyre wear, road surface abrasion and the constant resuspension of everything that has already landed and been kicked back into the air by the next bus.

Brake dust is metallic. Iron, copper, antimony, ground off the pads and discs at every set of lights, which on the Euston Road means roughly every 200 metres between Marylebone and King’s Cross. Tyre wear contributes polymers, fillers and silica. Road abrasion adds mineral dust. Blend the lot with hydrocarbon residue and the diesel soot that’s still there, and you have a film that behaves less like dust and more like a thin, greasy varnish.

Why the binder matters more than the dirt

Dust falls off. Dust rinses. The reason traffic film survives London rain on a vertical pane is the oily fraction holding it to the surface – unburnt hydrocarbons and organics from tyre polymer that bind the particulate into a continuous layer rather than a scatter of separate specks.

That continuity is what the eye reads as grey. Individual particles at that size are invisible from three metres. A bound film across the whole pane scatters light evenly and drops the apparent clarity of the glass before anyone can point to a single mark on it. Customers describe it as the window looking “tired” or “misted”, and half of them are convinced the double glazing has failed.

Why does Euston Road grey glass faster than Fortis Green Road?

Volume, stop-start braking and the shape of the street. Muswell Hill sits high, has no through route of any consequence, and the traffic that does pass moves at a steady speed with light braking. Fewer vehicles, less brake material, better dispersion.

Euston Road is a canyon. Buildings both sides, six lanes, and a queue that stops and starts from seven in the morning until well after seven at night. Vertical mixing is poor between the facades, so the material rises, hangs and settles on whatever’s nearest. And it shows.

Why the first floor suffers worse than the fourth

The gradient is steep and most people underestimate it. Coarse non-exhaust particulate is heavy and doesn’t travel far vertically before it comes back down, which loads the lower two storeys of any roadside building far harder than the upper ones. On a seven-storey block on Talgarth Road in W14 I cleaned for years, the second-floor glass needed monthly attention to hold its appearance and the sixth floor held up for a quarter with no argument from anyone.

Splash matters too. Wheel spray off a wet carriageway throws a slurry of road dust and rubber directly onto ground-floor shopfronts, which is why the bottom 600mm of a retail frontage on any red route goes streaky within days. Nobody has ever thanked me for explaining this to them mid-invoice.

Can pure water alone shift traffic film?

Here I’ll part company with a good deal of the trade. Deionised water through a water-fed pole is superb at what it’s designed for – loosening soluble soiling, carrying it away, drying spotless because there’s nothing dissolved in it to leave behind. Against a bound hydrocarbon film it underperforms, and every operator working the A501, the A4 or the Old Kent Road knows it even when their marketing says otherwise.

Pure water is aggressive towards dissolved solids. It has no meaningful action on grease. Brushing harder doesn’t fix that; it burnishes the film, moves some of it, and leaves a pane that looks acceptable wet and grey again forty minutes later when it dries. The operative gets blamed. The chemistry was never on his side.

What I use on arterial frontages

A detergent stage first, then the pure water. Traffic film remover of the sort the vehicle trade has used for decades is alkaline and surfactant-heavy, and it does exactly what the name says, though the strengths sold for lorry washing are too harsh for use near anodised frames and powder-coated aluminium. I dilute well below the drum’s recommendation, apply through a low-pressure sprayer or an applicator on a shorter pole, give it two or three minutes of dwell without letting it dry, agitate, then rinse the entire elevation down with deionised water and finish as normal.

The dwell is the part everyone skips. Surfactants need contact time to lift an oily binder off silica, and a contractor working to a fifteen-minute-per-elevation rate will spray and scrub straight away, which achieves very little beyond wetting the dirt. On a bank frontage near Jamaica Road in SE16 I timed the difference across two visits: same product, same dilution, three minutes of dwell against thirty seconds, and the longer dwell held its finish for close to double the interval. It costs a few minutes. It saves a return visit.

Order matters as much as product. Detergent first and rinse second, never the reverse, because pure water applied over a partially dissolved film picks up the surfactant and drags a grey slurry down the pane and across the frame below. Work top down through the elevation, keep the rinse generous, and accept that an arterial frontage uses roughly half again as much water as a residential job of the same glass area.

The other thing worth changing on these buildings is the brush. A softer flocked head holds more solution against the glass during dwell, and it drags less of the abrasive grit across the surface, which on a road throwing off mineral dust and brake particulate is a real consideration rather than a theoretical one. Stiff bristle plus silica plus pressure is how fine scratching starts.

There’s a caveat about frames. Alkaline products and untreated aluminium have a poor relationship, and I’ve seen a spotter panel go milky where somebody used neat TFR on a shop front. Test somewhere unimportant, rinse thoroughly, and don’t let the stuff dry on anything.

When does the grey stop coming off?

At some point the film stops being a film. Iron particles sitting in a damp layer against glass corrode, and corroding iron produces compounds that bond to the silica surface rather than resting on it. Give that a few years on an unwashed roadside pane and the greying is in the glass.

You can see it once you know the sign – the pane stays faintly hazy when wet, which is the opposite of what dirt does. Wet dirt goes dark and transparent. Bonded staining stays dull under water.

The conversation nobody wants to have

At that stage the honest answer is that cleaning won’t recover it. Cerium oxide polishing brings a lot of it back, though it’s slow, it’s wet, and it needs care near the edges of toughened panes where uneven heat and pressure cause problems. On badly neglected commercial glass the polishing cost gets close enough to replacement that the client should be pricing both.

I lost a contract on a parade near Bounds Green Road in N11 for telling a landlord precisely that. He’d had the frontage cleaned twice a year for a decade, believed the greyness was a cleaning failure, and wanted somebody who would promise to fix it with a squeegee. Someone did promise. And it shows.

What cleaning frequency does an arterial frontage need?

Monthly on the lower two floors of anything on a red route. Fortnightly if it’s a retail frontage where appearance is the point. Six-weekly to quarterly once you’re above the third floor or a street back from the traffic.

Those intervals sound like a contractor talking his own book, so here’s the against-interest version: most of my Muswell Hill and Highgate domestic customers are on a schedule more frequent than they need, and about a third of them would notice nothing if they moved from six-weekly to eight. I’ve told several. Two took the saving and stayed with me for years afterwards.

Why frequency beats chemistry on a red route

A film that’s four weeks old comes off with detergent and water. A film that’s eight months old has begun the corrosion process and takes polishing. The window that gets cleaned twice a year on the Euston Road is being managed towards permanent damage, and it’s a false economy nobody notices until the glass is grey in January sunshine and everyone’s arguing about who let it happen.

Frequency is also the only variable a building manager controls. You can’t move the building off the A501. You can’t stop the buses braking outside. And it shows, on every pane below the third floor, in a way that no amount of clever chemistry undoes after the fact.

The Fortis Green house went eighteen weeks before its next clean and looked fine on week seventeen. The Warren Street offices were booked back in for four weeks and I’ve never once been early.