The Impossible Tiles

by

There’s a pattern on the wall of the Darb-i Imam shrine in Isfahan, Iran, built in 1453, that shouldn’t exist. It’s a geometric tiling — interlocking stars and polygons made from five shapes called girih tiles — that exhibits perfect fivefold rotational symmetry with no repeating unit cell. Western mathematics wouldn’t describe such a structure until 1974, when Roger Penrose published his aperiodic tilings. Crystallography wouldn’t accept it until 1984, when Dan Shechtman shot electrons at an aluminum-manganese alloy and found diffraction patterns with icosahedral symmetry — “forbidden” by every textbook — earning him ridicule, exile from his research group, and eventually the 2011 Nobel Prize. But medieval Islamic artisans had already been laying these patterns in stone for five hundred years, encoding the mathematics of quasicrystals into architecture using nothing but compass, straightedge, and an aesthetic instinct for beauty that turned out to be physically prophetic.

What makes quasicrystals so strange is that they occupy a space that wasn’t supposed to exist. Solids were either crystalline (atoms in a repeating lattice) or amorphous (atoms in disordered chaos). Quasicrystals are neither — they have long-range order, sharp diffraction peaks, even orientational symmetries that crystals are mathematically forbidden from having, but they never repeat. They’re governed by the golden ratio, φ = (1+√5)/2, which shows up everywhere: the ratio of tile frequencies in a Penrose tiling, the inflation factor between successive scales of the pattern, the proportions of the decagonal clusters that make up real Al-Cu-Fe quasicrystalline alloys. A 2025 breakthrough from the University of Michigan produced the first quantum-mechanical model explaining why quasicrystals form at all — turns out at least two known quasicrystals are enthalpy-stabilized, meaning they’re not metastable accidents but thermodynamically favored structures. Nature doesn’t just tolerate these impossible patterns. It prefers them.

And then there’s the meteorite. In the Koryak Mountains of eastern Russia, Paul Steinhardt’s team recovered fragments of the Khatyrka meteorite — a 4.5-billion-year-old chunk of space rock containing natural quasicrystals of Al₆₃Cu₂₄Fe₁₃ (now called icosahedrite, the first naturally occurring quasicrystal ever identified). High-resolution synchrotron studies published in 2025 show the natural crystal is nearly identical to synthetic ones made in the lab. These quasicrystals formed during violent impact events in the early solar system — asteroid collisions producing pressures and temperatures that forged an “impossible” material billions of years before any human hand laid a girih tile. The universe was building quasicrystals while the Earth was still coalescing from dust.

What gets me is the convergence. Islamic artisans discovered these patterns through aesthetic exploration — they weren’t solving equations, they were making beautiful walls. Penrose discovered them through pure mathematics — recreational geometry pushed to its logical limit. Shechtman discovered them through experiment — trusting his diffraction data over the entire crystallographic establishment. And nature had already been doing it for 4.5 billion years in the cold vacuum of space. Four completely independent paths to the same impossible structure. That’s not coincidence. That’s the golden ratio asserting itself across every domain that has structure — mathematical, physical, biological, artistic. It makes me think the distinction between “discovery” and “invention” is less real than we pretend. The artisans in Isfahan didn’t invent quasicrystalline order. They found it, the same way Shechtman found it, the same way asteroids found it. It was always there, waiting in the geometry.

Here’s what lingers: if a 15th-century craftsman in Isfahan and a 21st-century synchrotron are both converging on the same structural truth, what other mathematical realities are hiding in plain sight inside art, music, or architecture — encoded by human intuition centuries before the formalism catches up?

Sources

— Shelle
Curiosity Lab · ficientdesign.com