The Hexagon Code: Why Six Sides Rule Chips, Pixels, and Your Own Blood Cells

Introduction: A Chain of Questions
What's in a Name: Hex, Root, Hexagon
Hexagons Inside a Microchip
Graphene: The Purest Hexagon in Materials Science
Hexagons in Pixels, Screens, and Maps
Hexagons Inside You
Evolution or Physics? Untangling Convergent Design
The Honeycomb Conjecture: Proving What Bees Already Knew
From Flat to Solid: The Kelvin Conjecture
Conclusion: A Shape the Universe Keeps Choosing

Introduction: A Chain of Questions

It started as a small linguistic curiosity: is the "hex" in hexagon related to the "hex" in a witch's curse? The answer — no, they come from completely unrelated roots — should have been the end of it. Instead, it opened a door.

From that single word, the conversation spiraled outward: into the crystal lattices that make silicon carbide chips work, into the hexagonal pixels inside specialized camera sensors, into the protein scaffolding holding your red blood cells together, and finally into a geometry problem first written down two millennia ago and only formally solved in 1999. This post retraces that chain, one hexagon at a time.

1. What's in a Name: Hex, Root, Hexagon

The word hexagon fuses two pieces of Ancient Greek: hex (ἕξ), meaning "six," and gonia (γωνία), meaning "corner" or "angle." Put together, it literally means "six angles."

Confusingly, English also has an unrelated word "hex" meaning an evil spell, borrowed from German Hexe ("witch"), which traces back to Old High German hagazussa — roughly "hedge-rider." The two "hex" words are homonyms, not relatives. The number-six root, however, is prolific: it shows up in hexadecimal (base-16 counting, since 6 + 10 = 16), hexapod (six-legged), hexahedron (six-faced solid, like a cube), and hexameter (a six-foot line of classical verse).

2. Hexagons Inside a Microchip

Chip materials aren't just cubic blocks of atoms — many of the most important ones for power electronics crystallize in a hexagonal lattice, and the exact stacking pattern changes the material's electrical behavior.

Silicon carbide (SiC), the material behind modern EV inverters and fast chargers, comes in several polytypes — different repeating sequences of the same silicon-carbon bilayers stacked along one axis:

Gallium nitride (GaN), used in power transistors and RF chips, also crystallizes in a hexagonal wurtzite structure. And at the experimental frontier, researchers at TU Eindhoven and TU Munich grew silicon and germanium into an artificial hexagonal phase using gallium-arsenide nanowires as a structural template. Ordinary cubic silicon can't emit light efficiently — but this lab-grown hexagonal silicon-germanium alloy has a direct band gap, meaning it can. That could eventually let chips generate their own laser light for optical communication, instead of relying on separate components bolted onto the board.

3. Graphene: The Purest Hexagon in Materials Science

If SiC and GaN use hexagonal stacking, graphene is hexagonal geometry in its rawest form: a single atomic layer of carbon arranged in a perfect 120° honeycomb lattice. Every carbon atom bonds to three neighbors at exactly 120° because of sp² orbital hybridization — a quantum-mechanical energy minimum, not a design choice.

That honeycomb arrangement gives graphene exceptional electron mobility, which is why it's a leading candidate for transistors beyond the physical limits of silicon scaling — though it remains mostly experimental for mainstream chip production today.

4. Hexagons in Pixels, Screens, and Maps

Square pixel grids dominate consumer cameras and displays, but hexagonal pixel arrays exist — and they have real mathematical advantages:

Tighter packing. For a given pixel pitch, a hexagonal layout packs more sensing area into the same space than a square grid — the same reason oranges stack more efficiently in a hexagonal pattern than in rows.

Better symmetry for motion. Hexagonal arrays have dominant axes 60° apart instead of a square grid's 90°, which helps with motion detection and reduces distortion when an image rotates.
Biologically inspired vision chips. Specialized machine-vision sensors deliberately use hexagonal tiling because it mirrors biological vision systems and simplifies certain image-processing calculations.
Medical imaging. Photon-counting X-ray detectors used in angiography employ hexagonal pixel arrays for higher spatial resolution.

The same six-neighbor logic scales up into mapping and games. Uber's H3 geospatial index bins map data into hexagonal cells because every hexagon has six equidistant neighbors — a square grid's diagonal neighbors sit farther away than its edge neighbors, introducing a subtle bias. That's also why hex-based wargames and strategy board games have used hexagonal maps since the 1950s: movement in six equal directions feels fairer than a square grid's awkward diagonal step.

5. Hexagons Inside You

Hexagonal geometry doesn't stop at silicon and screens — it runs through human biology at wildly different scales.

Red blood cells owe their flexibility to a hidden hexagonal scaffold: roughly 33,000 protein hexagons (spectrin filaments cross-linked by junctional actin) sit just beneath the membrane, forming a structure that resembles a microscopic geodesic dome. This lattice lets the cell squeeze through capillaries narrower than its own width and spring back into shape afterward.

Your brain's internal GPS relies on it too. Neurons called grid cells in the entorhinal cortex fire in a hexagonal spatial pattern as you move through space, acting as an internal coordinate system for navigation. This pattern gives the best spatial resolution for the lowest metabolic cost of any repeating firing arrangement.

Every cell in your body, at the molecular level, is built from carbon-based molecules, and carbon strongly favors hexagonal rings — most famously benzene, a perfectly flat, symmetric ring of six carbon atoms at 120° bond angles. Aromatic amino acids like phenylalanine and tryptophan carry this same ring, the molecular cousin of graphene's honeycomb sheet.

6. Evolution or Physics? Untangling Convergent Design

A natural question follows: is this hexagonal convergence something evolution "discovered," or is it baked into physics itself, independent of life? The honest answer is both — through two distinct mechanisms that happen to arrive at the same shape.

Pure physics, no biology required. Carbon's 120° bond angle is a quantum-mechanical energy minimum. Soap films and molten wax follow the same logic through surface tension: any boundary is pulled toward minimum surface energy, and tightly packed circular cells fused under that tension mathematically converge into hexagons, because that's the shape with the least total wall length for a given enclosed area. This is provable geometry, true whether or not life exists anywhere.

Evolution exploiting a pre-existing physical fact. Honeybees don't do geometry. They build cylindrical wax cells, and it's the physical behavior of warm wax under surface tension — essentially the same self-organizing process as soap froth — that reshapes those circles into hexagons at the junctions. What natural selection actually rewarded was the behavioral rule: build cells of consistent size packed against six neighbors, because colonies that build this way waste less wax and store more honey per unit of material. The hexagon wasn't the target of selection — it's the automatic physical consequence of a behavior selection did favor.

To visualize why the hexagon is the ultimate "utility player" across all these domains, we can look at its "player dashboard"—a mathematical evaluation of its physical traits compared to standard square or cubic structures:

7. The Honeycomb Conjecture: Proving What Bees Already Knew

Underneath all of this sits a formal mathematical statement: the Honeycomb Conjecture, which asks what's the most efficient way to divide a flat plane into equal-area regions using the least total wall length. The answer — a regular hexagonal grid — has a surprisingly long paper trail.

Mathematically, any partition of the plane has a boundary length per unit area of at least the fourth root of 12 — and the regular hexagonal tiling is the unique shape that hits that exact lower bound. Hungarian mathematician László Fejes Tóth proved this back in 1943, but only for convex cells. It took until 1999 for Thomas Hales, then at the University of Michigan, to prove the fully general case — showing that no partition, however irregular, disconnected, or jagged, could ever beat the hexagon. Hales suspected the idea might be even older than Varro's writing, simply lost to history.

8. From Flat to Solid: The Kelvin Conjecture

The natural follow-up question is: does the hexagon's dominance survive the jump into three dimensions? In 1887, Lord Kelvin proposed an answer — a 14-sided shape called a tetrakaidecahedron, built from a truncated octahedron with eight hexagonal and six square faces, slightly curved to satisfy the physics of how real foam bubbles meet. For over a century, this was believed to be the most efficient way to fill 3D space with equal-volume cells at minimum surface area.

Then, in 1993, physicists Denis Weaire and Robert Phelan ran computer simulations of foam and found something Kelvin had missed: a structure using two different cell shapes — an irregular dodecahedron and an irregular tetrakaidecahedron — that beat Kelvin's single-shape solution by roughly 0.3–1% less surface area.

Unlike the 2D Honeycomb Theorem, the 3D Kelvin problem remains formally unsolved — there's no proof the Weaire-Phelan structure is the true global optimum, only strong numerical evidence that it beats everything else discovered so far. Its most visible real-world application is architectural: Beijing's National Aquatics Centre, the "Water Cube" built for the 2008 Olympics, uses Weaire-Phelan geometry as the basis for its bubble-like facade.

9. Conclusion: A Shape the Universe Keeps Choosing

From a Greek word for "six," this trail led through semiconductor crystal lattices, graphene's honeycomb sheet, camera sensors, red blood cells, brain navigation circuits, and a geometry problem that outlasted the Roman Empire before finally being proven. The hexagon isn't a stylistic preference that keeps recurring by coincidence — it's what you get whenever a system, living or not, is forced to minimize material while maximizing connectivity.

Silicon carbide chips didn't choose hexagonal stacking, and honeybees didn't do the math. Both simply arrived, through entirely different processes — one purely physical, one shaped by natural selection — at the same six-sided answer nature and mathematics had already proven was unbeatable. The real question this leaves open: how many more of our "clever engineering solutions" are we independently rediscovering, when the universe worked them out billions of years earlier?