Becker Double-Layer Capacitor

Everything in this device happens on surfaces that are inside the carbon, and a pore full of air has no surface at all as far as the electrolyte is concerned. Wetting is therefore the whole build, and skipping it is why most first attempts measure microfarads instead of farads. Dissolve 30 g of salt in 100 ml of warm distilled water - a strong brine, roughly as salty as it will go. Stir 6 g of activated charcoal powder into about 40 ml of it and keep stirring until the floating layer is gone and you have a black slurry with no dry lumps. Then leave it. Overnight is right; an hour is not. If you have a syringe you can speed it up: draw the slurry into a syringe, block the nozzle with a finger, pull the plunger back hard, and watch the air come out of the pores as bubbles. Release and repeat five or six times. Distilled water and pure salt matter more than they look. Tap water brings calcium and magnesium that will slowly clog exactly the pores you are trying to open.

Cut two 30 x 30 mm squares of graphite foil as current collectors, each with a 20 mm tab left on for the connection. Graphite rather than metal: this is salt water with a voltage across it, and steel or copper will corrode into it and ruin the measurement within minutes. Spread half the wet carbon slurry onto each square as an even layer 2-3 mm thick, out to about 25 mm and no further - leave a dry margin so the two layers cannot meet at the edges. Wet a piece of filter paper in the brine and lay it over one electrode, then press the second electrode face-down onto it so the sandwich reads: foil, carbon, paper, carbon, foil, with the tabs sticking out on opposite sides. That paper is the only thing standing between you and a short circuit, and it is one sheet thick. Two sheets is a perfectly reasonable choice for a first build; it costs you a little capacitance and removes the most common failure.

Sandwich the stack between two rigid plastic plates and clamp it just tight enough to hold everything in contact - firm hand pressure, not a vice. Squeezing hard drives the electrolyte out of the carbon and undoes step 1. Stand the whole clamped assembly in a shallow dish with a few millimetres of brine in the bottom so the edges stay wet. A double-layer cell that dries out does not fail dramatically; it simply loses capacitance, quietly, and you will blame your measurement. Check with a multimeter on resistance across the two tabs before charging: you want a reading in the tens of ohms, not near zero. Near zero means the two carbon layers have found each other and the separator has to be redone.

This is the measurement, and it needs no instrument more exotic than a watch. Charge the cell to 0.9 V from a bench supply, current-limited to 50 mA, and stop there. **Do not go above about 1.2 V** - water splits at 1.23 V, and gas generated inside the pores blocks the surface you spent last night wetting. Disconnect the supply, connect a known resistor - 100 ohms is a good start - across the tabs, start the stopwatch, and write down the voltage every ten seconds until it has fallen to about a third of where it began. A capacitor discharging through a resistor reaches 36.8 per cent of its starting voltage after exactly one time constant, and that time constant is R times C. So if 0.9 V falls to 0.33 V in 300 seconds through 100 ohms, then C = 300 / 100 = **3 farads**. From six grams of charcoal, a jam jar of brine and a piece of paper. Do it twice with different resistors. If the two answers agree you have measured a capacitance; if they do not, something is leaking and the flow step will tell you what.

**Where the "double" comes from.** Put a voltage on the carbon and two sheets of charge form facing each other: electrons in the carbon surface, and solvated ions in the liquid drawn up against it. Between them sits a single layer of solvent molecules. That is the separation - a few tenths of a nanometre - and since capacitance goes as 1 over the separation, it is where the absurd numbers come from. **Why the surface area is a lie, usefully.** A gram of activated carbon has 1,000 to 3,000 square metres of surface by gas adsorption, which should give hundreds of farads. Measured electrodes give about 100 F/g. The difference is that a nitrogen molecule fits into pores that a solvated ion dragging its water shell cannot enter. Both numbers are correct; they are answers to different questions. **Why it is not a battery.** Energy is one half C V squared, so this cell holds tens of joules where an AA holds thousands. What it has instead is power and endurance: it will take that charge in and give it back in seconds, and do it hundreds of thousands of times, which is why supercapacitors sit in regenerative brakes, camera flashes and memory backup rather than in phones. **Why yours is worse than a bought one, and that is fine.** Commercial cells use an organic electrolyte at 2.5-2.7 V instead of water at under 1, and energy goes as voltage squared - a factor of seven before any other difference. They also use carbon with a pore-size distribution matched to the ion, not the general-purpose powder sold for fish tanks. Your device is the real mechanism at a fifth of the performance, which is exactly what a demonstration should be. **The historical footnote worth keeping.** Becker patented an effect he could not explain, and said so in the specification. The explanation existed and had for a century, in electrochemistry rather than electronics. Nothing about the device needed inventing after 1957; what it needed was somebody reading both literatures.