The Shirt Needs a Stomach

Colgate-Palmolive put this one in the open, which means we can work on it publicly.

They want something that stops musty odor on natural and synthetic clothes, delivered as a softener or a spray. It has to be cost effective. Approaches with safety concerns are out, and nanoparticles are the example they give. And the brief does something most odor briefs never do. It names the molecules. Geosmin. 2-methylisoborneol. 2,4,6-trichloroanisole.

That is the whole lock.

Most odor work hides behind words like stale and damp. This one names the prisoners.

What I think they actually asked

Read the two halves of the brief against each other.

The description wants the microorganism killed and kept from growing back, with the certification to say so. The success criteria want the source of the odor fought, and name the three molecules.

Colgate's own criterion already calls the molecules the source. The description has not caught up. Those are two different products. Only one of them needs a pesticide registration.

For one of the three, the organism framing is not weak. It is aimed at the wrong room.

Trichloroanisole is usually not made where you smell it. Call it the taint.

Drug companies spent a decade recalling product for a musty smell. It came off the wooden pallets. Mold had worked on a fungicide in the timber. The taint arrived from outside. It never grew in the bottle.

Swedish researchers found the same family across hundreds of building materials, made when microbes methylate the chlorophenols left in old wood preservatives. In one study the smell came off treated wood where the mold could only be found under a microscope.

It happens to chickens too. Fungi worked a preservative in the floor litter. The birds took the smell up into their meat.

Damp, microbes, one chlorophenol. That is the whole recipe, and a house supplies all three. A shirt on a drying rack is a big absorbent surface in a room that may be making the molecule. Killing bacteria on the cloth does nothing about that.

The taint is stranger than a smell. Work at Osaka showed it suppresses olfactory signal transduction instead of triggering it. It shuts the channels a receptor cell uses to fire, at concentrations almost too small to write down.

So perfume is not fighting the contaminant. The contaminant may be turning down the sense that detects the perfume. That last step is my inference from a lab paper, not a result on fabric. If it holds, a softener carrying a trace of this is having its own scent switched off.

The real ask is not an antimicrobial. It is a way to take three molecules apart on a damp shirt without putting anything on that shirt that can attack the shirt. Softener or spray gets decided by what goes airborne, not by the chemistry.

How I go looking

Strip the laundry off it.

Destroy three trace molecules sitting on a wet surface. Use chemistry rough enough to break them. The surface is more fragile than the target.

That is not an odor problem. It is a containment problem.

Biology settled that one long ago, in an organ that holds acid strong enough to digest meat inside a bag made of meat.

The obvious candidates fail

An antimicrobial stops new production and leaves the stored inventory alone. And for the taint, the producer may be in the crawlspace rather than the closet.

Fragrance is worse than outgunned. It is aimed at a receptor the contaminant has already closed.

An adsorbent gets closer. Cyclodextrin pulls the molecule out of the air and into a cavity. But capture is not destruction. Humidity changes what a full cavity gives back.

Bulk oxidant fails twice. Chlorine, chlorine dioxide and permanganate cannot crack the hindered alcohols in geosmin and its cousin. That is why water utilities reach past them for the hydroxyl radical, which is the least fussy oxidant there is. And the fabric-care patents already refuse oxidative routes, because they bleach colored cloth. The oxidants strong enough to do it are the ones nobody wants near a colored shirt.

A flat catalytic coating solves regeneration but sets the catalyst straight against the dye.

The animal

A stomach gets you as far as containment. That is not enough, because there is no single chemistry to contain.

A neutrophil does not build one multi-tool either.

It carries at least three kinds of granule, and only one of them holds the peroxidase. That split is the classic one. The cell keeps its most dangerous enzyme in a bag of its own.

Two more details are worth stealing.

The bags come out in order. The mild ones first. The bag with the bleach in it last. The cell does not just separate its worst chemistry. It makes that chemistry the hardest to reach for.

And the cargo is set by timing. A granule takes its contents from the shift it was built on, not from labels on the proteins inside. Batches, not addresses.

The part that made me sit up

The white-rot fungus does not start by burning the ring. It undoes the reaction that made the taint.

Thirteen species were tested, and the strongest, Phlebia radiata, cleared most of it in ten days. The lignin enzymes everyone would reach for did nothing. The first move was an enzyme pulling the methyl group back off. That gives a chlorophenol. The chlorophenol got capped, then broken down, shedding chloride as the ring came apart.

A synthetic mimic of that heme chemistry is a metalloporphyrin. Pulling alkyl groups off oxygen is one of the reactions it runs.

Which is where it stops being comfortable. The proven textile use of one of these catalysts, immobilized, with peroxide at neutral pH, is bleaching industrial dye. It bleached every dye tested.

So the catalyst that undoes the taint is a catalyst built to take color out of cloth.

The catalyst needs the box for itself, too. Loose in solution these complexes wreck each other, pairing off or mistaking a neighbor's ring for something to oxidize. The fix the field landed on is holding them apart inside a porous solid. That is a description of the enzyme pocket they were copied out of.

And nobody has run the reaction I am asking for. These catalysts have been fixed to silica. They strip halogens off phenols. They pull alkyl groups off oxygen in drug work. None has been pointed at a chloroanisole.

There is a second reason the split is forced. The other two molecules want the opposite conditions. Geosmin loses water under acid and becomes argosmin, odorless at the concentrations that matter. Its cousin does the same. That reaction has been run at scale in a drinking water pipeline, where acid dosing lifted geosmin removal from around 57 percent to 83, and bench work showed the acid alone was doing it.

The acid wants less water around it. The metal wants water and peroxide moving freely. One box cannot serve both.

The glass box

So the answer is not one protected reactor. It is a lot of them. Each carries only the chemistry its own target needs.

Which means a hollow shell that already comes by the billion, in a hard material that takes surface chemistry, cheap enough to leave on a shirt.

Diatoms grow one. The frustule is a glass box with a porous wall, and glass takes grafted chemistry easily. Thalassiosira pseudonana runs four to six microns across, smaller than a red blood cell, and has been cultured and sequenced for sixty years.

The organism is not the insight. It only grows the box.

So how would I build it

Everything from here is design and theory.

What ends up on the shirt is millions of glass boxes. Most of them sour. A few of them holding something that would take the color out of the shirt if it got loose.

The many carry an acid group bonded to the inside wall, using surface chemistry glass has taken for decades. Their job is geosmin and its cousin. They protect their own container as a bonus, because acid is where glass dissolves slowest.

The few carry the metal catalyst, bonded the same way. Their job is the taint. These are the ones that need the box. A dye is stuck to the fiber and can only be wrecked by contact, and a micron of glass means the fiber never reaches the metal.

Two locks, not one. The bond stops the catalyst leaving. The box stops the cloth arriving.

The mix is lopsided, because the problems are. Geosmin needs damp and a mold. The taint needs damp, a mold, and a chlorophenol somewhere in the building. The second is a subset of the first, and it is the one that needs the rough chemistry. Put the dangerous chemistry on the rare target, at a few percent of solids. Then follow the neutrophil and make it the slowest to wake up.

Wetting is the trigger, whatever the format. You put it on a damp shirt. It works while the shirt dries. Drying shuts it off. Nothing needs a humidity gate, because the user is the gate. My guess at a first format is non-aerosol delivery or rinse deposition.

On the safety exclusion, these boxes are microns across, so the nanoparticle definition is probably not the obstacle. That settles a classification, not a safety case. The real burden is fines, and what goes airborne during application.

What would keep me up at night

The first knife-edge is that the middle step is worse than the target.

Take the methyl off the taint and you have made a chlorophenol. The fungus caps that as a sugar and comes back for it later, and an enzyme can cut the cap off again. A shirt has no later. And the chlorophenol is exactly what cork fungi methylate to make the taint in the first place, using an enzyme those same chlorophenols switch on.

Stall the second step and the device makes the smell it was sold to remove.

So the second step has to finish, and parent loss is not enough to show it. One iron catalyst with peroxide opens the ring and releases up to two chlorides per chlorophenol. Chloride release, with product analysis, is what says the ring is coming apart.

The second knife-edge is sebum.

Bare diatomaceous earth kills insects partly by pulling the grease off their shells. A treated box will not have that surface, so the mechanism does not carry over. The warning does. A shirt holds skin oil and softener residue at thousands of times the odorant. In natural water, where the ratio runs the same way, activated carbon takes up an order of magnitude less geosmin and its cousin than it does from clean water. The boxes can fill with the wrong thing long before they meet the right one.

The third is the acid itself. The evidence for that dehydration sits at pH 2 to 5 in open water. Whether a bonded acid site does the same job on a damp shirt at room temperature is unproven. It is the first number I would want.

What I would run first

One comparison, and it tests the claim rather than the concept.

Four arms, on soiled polyester and polyester-elastane, dirtied with skin oil rather than clean. Acid boxes alone. Catalyst boxes alone. The two mixed. And, as the arm I expect to fail, one box carrying both chemistries in the same cavity, because that is what a formulator tries first.

Challenge each molecule separately. Dry, rewet, dry again. Track headspace, free chloride, and color on the dye as three separate numbers.

If the mix beats the shared box, the architecture is proven in one comparison. If the shared box wins, I am wrong about all of it and the answer is simpler than I think.

The wider door

The field names the ingredient it can buy. Antimicrobial. Fragrance. Encapsulant. Oxidant. The hidden machine sits in the wall between the thing you bought and the thing it has to work beside.

Here the answer was already known and unusable. People have been taking these molecules apart for years. What nobody could do was take them apart on a shirt, because everything strong enough to break the smell is strong enough to break the color.

The chemistry was never the missing part. The wall was.

And the shirt is only where you notice it. The room is where it is made.

Once the wall is the product, the question stops being which molecule is safe enough to use. It becomes which cure got shelved because nobody built it somewhere to stand.

Somewhere, something has already solved the problem you’re looking at.

James Stephens