A paper wallpaper in the lab feeds on humidity in the room

On a bench at Binghamton University in New York State lies a sheet of laboratory paper. It does not look like store-bought wallpaper. It measures two by two centimetres and has three chemical zones. When the air is humid, a current flows from graphite electrodes at the edges. It is small — microwatts per square centimetre. In the lab it is still enough to power a wireless keyboard and, along the way, to dry the air a little.

Guangya Yuan, Yang Gao and Seokheun Choi described the design on 16 September 2026 in Advanced Energy Materials. They call it the first wallpaper-type moist-electric generator: a device that turns ambient humidity into electricity and at the same time acts as a thin drying membrane. It is not a substitute for solar panels. It is an attempt to turn a wall into a quiet power source for sensors and small building electronics.

Why indoor humidity, not the outdoors

Moist-electric generators are not a new idea. For years labs have built structures in which water from the air splits into ions and a concentration gradient yields a voltage. The practical problem is outdoors. Humidity, temperature and wind jump around. Power is tiny — microwatts per square centimetre — and conditions are unstable. Choi puts it plainly: outdoors there is plenty of moisture, but the weather spoils the deal.

Indoors, relative humidity usually sits between 30 and 60 percent. Breathing, cooking and bathing add vapour. For the microwatts and milliwatts that environmental sensors, radio modules and small Internet-of-Things gadgets need, that is enough. A wall also offers a huge surface that nobody wants covered in batteries or cables.

What was missing was an architecture that can spread across a large plane, keep moisture transport directional, and not look like a circuit board glued to the paint. The Binghamton team solved it like paper-electronics engineers: three zones on one sheet, wiring hidden on the back.

Three zones on a piece of Whatman

The carrier is Whatman 3 MM filter paper, 340 micrometres thick. That is lab paper, not a commercial roll of wallpaper. Each tile is about two centimetres on a side.

Glycerol sits on the periphery. The syrupy liquid grabs water vapour from the air and turns it into liquid water in the paper pores. The middle zone holds polyvinylpyrrolidone, or PVP. The polymer binds water more tightly and has smaller pores, so liquid is drawn inward — Laplace pressure and a chemical-potential difference do the work. The centre is wax-treated. Wax blocks liquid but leaves micropores through which vapour can escape. It behaves like a breathable rain jacket: water stays in the paper instead of soaking the wall, and vapour still leaves.

Water therefore enters at the edge and leaves as vapour in the centre. As long as that flow lasts, uneven wetting and an ion gradient persist. Chemical groups on glycerol, cellulose and PVP help water release protons. There are more of them at the edge than in the centre. Graphite electrodes collect that difference.

This “one-way” moisture transport is what the authors treat as the key. In earlier generators, moisture often absorbed and evaporated in the same place. The gradient blurred quickly. Here inlet and outlet sit apart on a flat surface, so a tile can lie flush on a wall.

Current without cables on the face

A homeowner does not want wallpaper laced with silver traces. So the connections run on the back. A laser drills one-millimetre holes. The holes are filled with silver paste that joins both faces of the sheet. The process mixes wax printing, airbrushing, screen printing and laser work. No cleanroom is required. Choi stresses that everything should be printable so it can scale.

Tiles are wired in series and in parallel. In tests both topologies gave similar results. What matters is area and the through-hole trick that hides the wiring.

What the lab delivered

A single tile at 80 percent relative humidity yields about 0.34 volts and about 2.2 microwatts per square centimetre. The signal holds for roughly 90 minutes. Ten tiles in series reach about 2.9 volts. An array of 35 units powered a humidity sensor for about a quarter of an hour.

The largest setup — 1,596 units — at about 38 percent humidity gave about 3.5 volts. With a capacitor that stores charge and releases it in bursts, that was enough for a wireless keyboard. The paper does not report a total wall power figure, so there is no need to invent one. The point is different: at typical room humidity, not only in an 80-percent climate chamber, a real gadget can move.

The same wall also dries the air a little. In a sealed 30-by-30-by-15-centimetre chamber, 28 tiles pulled humidity from 75 to 50 percent in about four minutes. When the paper was already wet and the air dry, it raised humidity from 15 to 20 percent in about 16 minutes — so it acts as a buffer. In a room the large array lowered humidity from 38 to 32 percent in fifteen minutes. That is an early result, not proof that wallpaper will replace air conditioning.

Where the voltage comes from

In short: water in the paper pores is not electrically inert. Near the surface of cellulose, glycerol and PVP some molecules give up protons. Positive and negative ions spread unevenly along the tile because there is more moisture at the edge than in the waxed centre. Graphite electrodes collect that potential difference. There is no spinning turbine and no bottle of liquid electrolyte with two metals. There is a humidity gradient kept alive by zone geometry.

That is why architecture matters more than “adding more moisture-hungry chemistry”. If absorption and evaporation sit in the same place, the gradient flattens and the current dies. Separating inlet and outlet on the wall plane keeps the flow. The authors also note that they do not need a hard sealed gasket in the centre: wax pores that are phase-selective are enough to keep liquid in the paper while vapour escapes.

What one may promise — and what not

Lab numbers are easy to inflate. Microwatts per square centimetre sound weak next to panels counted in watts. In a building, though, many devices sleep. A sensor wakes every few minutes, sends a short radio frame and sleeps again. A capacitor fed from wall area can cover that cycle. The keyboard in the experiment is a demonstration that “something familiar can move”, not the target product.

Humidity is the second promise track. A drop from 38 to 32 percent in fifteen minutes on a large array shows that the paper takes part in the vapour balance. How long capacity lasts, how it behaves across seasons, and whether it starts to smell or discolour are still open. So the text sticks to “buffer” and “light regulation”, not “air-conditioning replacement”.

What this means for a building

Air conditioning and dehumidifiers burn power to pull moisture out. Here moisture is fuel and, at the same time, a parameter the device gently nudges. Choi talks about cutting HVAC load and powering sensors without batteries. That is a sensible horizon. There is no claim of replacing the grid or a photovoltaic roof.

For the smart-building market the logic is simple. A CO₂, humidity or occupancy sensor needs little power, but batteries must be swapped and cables pulled. If a wall itself yields microwatts and slightly evens out humidity, denser sensing without half-yearly service becomes more plausible.

A Poland angle need not be forced. We have an ageing building stock, indoor moisture after insulation upgrades, and a rising number of office sensors. Binghamton’s technology is not a shelf product. It does show that paper electronics can aim at a problem visible in every building handover: moisture and no spare socket in the wall.

Limits and open questions

This is still a proof of concept. The carrier is Whatman, not commercial wallpaper that survives washing and years of exposure. Durability in a real living room — dust, seasonal swings, paint, adhesive — is not shown yet. Power is small; without a capacitor the keyboard would not “kick”. The authors do not claim that 1,596 tiles are a store-ready product.

Scaling has traps too. Each tile must keep three chemical zones, repeatable printing and rear connections without visible veins. Choi wants printability precisely to avoid hand jewellery work. Until there are months of run data and a price per square metre, it is safer to call this a research platform than a finishes revolution.

Competition in MEGs takes many roads: other polymers, oxides, textiles. Binghamton’s edge is not a microwatt record but the “absorb at the edge, evaporate in the centre” geometry aimed at a wall, plus hidden wiring. That is an engineering choice, not a marketing one.

What next

The next step is more power per centimetre, longer stability and a material closer to wallpaper you can paste and wipe. In parallel sits the receiving electronics: sensors that sleep most of the time and wake on microjoules from a capacitor. If those two tracks meet, a wall could become both a humidity buffer and a local power plant for a sensor mesh.

For now we have papertronics with concrete numbers: 0.34 volts and 2.2 microwatts per square centimetre from one tile, a keyboard on 1,596 units, humidity down from 38 to 32 percent. That is enough to take humidity wallpaper seriously — as a building tool, not as a science-fiction carpet.

Bibliography

1. Yuan G., Gao Y., Choi S. Moist-Electric Wallpaper With Engineered Unidirectional Moisture Transport for Indoor Energy Harvesting and Humidity Management. Advanced Energy Materials (2026). DOI: 10.1002/aenm.71603
2. Kocher C. Researchers develop wallpaper that generates power. Binghamton News (2026). https://www.binghamton.edu/news/story/6491/wallpaper-that-generates-power-binghamton-university-researchers-have-developed-it
3. Kardoudi O. Smart wallpaper turns your wall into a small indoor powerplant. New Atlas (2026-10-04). https://newatlas.com/materials/binghamton-wallpaper-humidity-energy/