Paper folding limit

Why can a sheet of paper not be folded in half more than about seven times, no matter how large it is?

The claim circulated for decades as settled trivia: no sheet of paper can be folded in half more than seven times, and the size of the sheet makes no difference. It sounds like a law rather than an observation, which is much of its appeal.

But "no matter how large" is a very strong thing to assert. Fold a sheet and you halve its extent while doubling its thickness — so a bigger sheet starts further from the wall. Why would starting further away buy you nothing at all? Either the claim is wrong, or something about the folding consumes the extra size as fast as you add it.

Set up the bookkeeping first. Take a strip of paper, length L, thickness t, and fold it in half repeatedly in the same direction. After n folds you have 2 to the power n layers, so the stack is t times 2 to the n thick, while the strip is L divided by 2 to the n long.

Both quantities move exponentially, in opposite directions, which is already the heart of it. After ten folds the stack is over a thousand sheets thick and the strip a thousandth of its length. Doubling the starting length buys exactly one more fold — so exponential growth does not care much about your head start, which should make you suspicious of "no matter how large", but equally suspicious of the number seven, since enough doublings buy any number you like.

Now ask what actually stops the fold, because thickness by itself does not. The obstruction is at the crease. Paper cannot bend to a zero radius, and the layers on the outside of that bend must travel further than the layers on the inside. That extra distance comes out of the flat length available. So each fold does not merely halve what remains — it also spends a slice of material on the rounded edge, and that slice is proportional to the thickness of the stack being bent.

The stack thickness is growing exponentially, so the material spent on creases is growing exponentially too, while the length remaining shrinks exponentially. Two exponentials racing towards each other from opposite sides is not a gentle limit. It is a wall.

That is where the folklore got its plausibility. For ordinary paper about a tenth of a millimetre thick, the length needed for seven single-direction folds is under a metre — so a printer sheet really does stop at six or seven, and the claim looked like a law because everybody testing it was testing the same size of sheet.

The proper ending came in 2002, when Britney Gallivan, then a high-school student in California, took the problem apart rather than repeating the experiment. She derived the loss function — the exact relationship between thickness, number of folds and minimum length — and then used it to buy the right material. Folding a single roll of tissue paper about 1.2 kilometres long, in one direction, she reached twelve folds, and published the derivation. In 2012 students at a school in Massachusetts used the same reasoning and a longer roll to reach thirteen.

The honest correction, then, is not that the limit was wrong but that it was misdescribed. There is no fixed maximum number of folds — there is an equation relating folds to the length required, and it demands length so ferociously that the practical ceiling for anything you can lay hands on is low.

A few refinements make the picture accurate rather than merely vivid.

The direction of folding matters, and the two cases have different equations. Folding always the same way, along a long strip, the required length scales with about the square of the number of layers. Alternating the direction on a square sheet is more efficient, and for that case the required width grows as two raised to roughly one-and-a-half times the number of folds. Alternating helps, but it is still exponential, and Gallivan's twelve-fold record used the single-direction method precisely because the arithmetic there is cleanest to plan against.

Thickness is the other lever, and it is the one people reach for first. Thinner paper folds more times — but only logarithmically more, since halving the thickness is worth about one extra fold. Gold leaf can be folded many times not because foil is special but because it is extraordinarily thin.

Finally, the wall is a material one and not purely geometric. Gallivan's equation gives the length below which folding is impossible; well before that, the force needed to crease a stack of thousands of layers exceeds what a person can apply, and the sheet tears or the crease refuses to take. The equation is a lower bound on what you would need, not a promise that having it is enough.

• How the same loss-function reasoning governs the minimum bend radius allowed in sheet metal and in fibre-optic cable.
• Why paper creases permanently at all — what fails in the fibre structure at the fold.

Every term the collection defines is gathered in the glossary.