Table of Contents (click to expand)
- Are We Technically Touching Anything?
- Does A Knife Really Slide Between Atoms?
- Why Does A Sharp Knife Cut Better Than A Blunt One?
- How Do Cuts Happen If Atoms Don't Touch?
- Why Does Slicing Cut Better Than Pushing?
- Why Are Tomatoes And Bread So Hard To Cut?
- Why Do Razor Blades Go Dull If Hair Is 50 Times Softer Than Steel?
- Which Metals Can Be Cut With A Knife?
- Can A Knife Split An Atom?
- What Does Sharpening Actually Do?
- So, What Actually Happens At The Atomic Level When A Knife Cuts Something?
A knife never slips between atoms: even a fresh razor's edge is rounded to a radius of about 40 nanometers, some 260 atom-spacings, so it presses on chemical bonds, not gaps. Cutting works by piling force onto a tiny area until the bonds at one point break, and the sharp tip of the crack that starts there multiplies the stress so the next bonds along the line give way one after another, like a zipper opening. Sliding the blade as you press cuts the downward force needed by half or more, which is why sawing through a tomato beats pushing, and why a hair 50 times softer than steel can still chip a razor by loading it from the side.
Ask a room of people how a knife cuts and most will draw the same picture. The blade is so thin that it slips between the atoms, and the two halves fall apart. A tidy picture, and the one most people carry.
It is also wrong, because a knife edge is a few hundred atoms wide at best. There is no gap for it to find.
So what does the edge do instead? It leans on chemical bonds until they break, and it does so in a clever order. That order explains why sharp beats blunt, why sawing beats pushing, and why a hair can chip a steel razor.
Are We Technically Touching Anything?
Press a fingertip on the table and nothing passes through, yet an atom is mostly empty space. We measured how empty in an earlier article. The nucleus holds nearly all the mass, yet it is around 100,000 times smaller across than the atom.
The rest is the electron cloud, and a cloud has no hard rim. That is why "do atoms touch?" has no one-word answer. The University of Illinois physics department calls an atom's electrons "a spread-out wave." Two such waves, it adds, "can and often do overlap." Two atoms never sit in the same spot, but their clouds meet, mix, and push back on each other all the time.
When two clouds overlap and share electrons, the atoms are held together, and that link is a chemical bond. Pulling one apart takes real energy. A single carbon-to-carbon bond, the kind in a tomato's cell walls, needs about 356 kilojoules per mole to break. It is 0.154 nanometers long, and a nanometer is a millionth of a millimeter.
Cutting a solid means breaking some of those bonds along a line, and nothing else.

Does A Knife Really Slide Between Atoms?
Two numbers settle it, and the first is the gap. The space a blade would need to slip into is about one bond length, 0.154 nanometers for carbon to carbon.
The second number is the blade. The sharpest everyday edge is a safety razor. In 2020, Gianluca Roscioli and colleagues at MIT put one under an electron microscope. The study appeared in the journal Science. Each blade is "honed to a wedge geometry with an angle of 17° and a tip radius of 40 nm." That radius is 40 nanometers. Divide it by 0.154 and the rounded tip is about 260 bond lengths from its center to its surface. The whole tip is roughly 500 atoms wide.
A kitchen knife is far blunter than that. Materials scientist Helmut Föll says sharpness starts at "a few micrometers" of tip radius. A micrometer is a thousand nanometers, so a "sharp" kitchen edge is tens of thousands of atoms wide. Next to a 0.154-nanometer gap, it is a freight train trying to fit through a letterbox.
The edge never goes between anything. It sits on top of thousands of atoms at once and pushes.

Why Does A Sharp Knife Cut Better Than A Blunt One?
The school answer is pressure, and it is right as far as it goes. OpenStax's College Physics defines pressure as "the force divided by the area" it acts on, which as a formula is:
P = F ÷ A
Here P is pressure, F is the force you push with, and A is the area that push is spread over. Shrink the area and the pressure climbs. The textbook's example is a needle, whose "sharp end" turns an ordinary push into one "great enough to break the skin."
Run the numbers for a knife. Push down with 20 newtons, about the weight of a 2 kg (4.4 lb) bag of flour, on a blade 10 cm long:
- A sharp edge meets the food along a strip about 1 micrometer wide. Over a 10 cm blade, that is an area of 0.0000001 square meters.
- Divide 20 newtons by that area and the pressure is 200 megapascals (about 29,000 psi).
- Let the edge wear to 5 micrometers wide, and that push now gives 40 megapascals (about 5,800 psi).
The blunt edge gets one-fifth the pressure from the same push, so the food squashes instead of splitting. Engineers use force-over-area inside a solid too, where they call it stress, and that is the word from here on.
Pressure explains why sharp beats blunt. It does not explain how a cut starts, or why it keeps going.
How Do Cuts Happen If Atoms Don't Touch?
Zip up a jacket, then open it. You do not rip all the teeth apart at once. The slider parts them one pair at a time, and each pair needs only a small tug. A cut works like that. The stress under the edge only has to break the bonds at one point, and after that, the point does the work.
In 1913, C. E. Inglis studied a plate with an elliptical hole. The stress was highest at the ends of the ellipse, and it grew as the hole got "longer and thinner." A crack is a long, thin ellipse, so its tip is where the stress piles up. We walked through the equation in our article on tiny cracks:
stress at the tip = stress far away × (1 + 2√(a ÷ ρ))
Here a is half the length of the flaw, and ρ (the Greek letter rho) is the radius of its tip. Take a nick 0.1 mm deep with a tip as sharp as a fresh razor. The multiplier comes out at about 100. The bonds at that tip feel a hundred times the stress of the bonds a millimeter away, so they are the ones that break.
A. A. Griffith added the energy accounting in the 1920s. A crack can only grow, he showed, if there is enough stored energy "to create the new surface area of the crack." Every broken bond makes a little new surface, and the energy in the squeezed material pays for it. As long as the blade feeds stress to the tip, the tip keeps moving.
That is the whole mechanism. The knife starts a crack under its edge, then rides the crack tip through the food, breaking bonds one row at a time.

Why Does Slicing Cut Better Than Pushing?
Press a knife straight down on a tomato and the skin dents. Draw it back an inch as you press and the skin parts. Tony Atkins, an engineer, spent a career on the mechanics of cutting. In 2016 he wrote that "however sharp a knife, cutting is easier" with a slide added to the push.
Atkins gave the effect a number. The slice–push ratio, written ξ (the Greek letter xi), compares two distances. It is how far the blade slides along its edge, divided by how far it pushes across. For a thin blade with no friction in a floppy material, the downward push you need, V, is:
V = Rw ÷ (1 + ξ²)
Here R is the toughness of the food, the energy it takes to make each square meter of new cut surface, and w is the width of the cut. With ξ = 0, pure pushing, you need the full force, Rw. Slide as far as you push and the push halves. Slide three times as far as you push and it falls to one-tenth.
The blade is not cheating: the work of breaking the bonds does not change. Sliding spreads that work over a longer path and loads the crack tip in a kinder direction. Atkins adds that the effect "does not depend on there being teeth on a blade." The sawing is what cuts.
This physics has a darker side. Atkins again: paper cuts, "cutting the tongue on licking the flap of an envelope," come from this effect. The envelope is not sharp. Your tongue slid along it.

Why Are Tomatoes And Bread So Hard To Cut?
A tomato is soft, so it ought to be easy. The trouble is that soft things move out of the way. A crack needs stress at its tip, and a floppy material spends the push on stretching and denting instead. The skin dips, the flesh below it squashes, and the first bonds never see the stress they need.
The fix is in Atkins' formula. You cannot raise the toughness R, and you do not want a wider cut, so you raise ξ with a light press and a long slide. Bread adds air to the story. A tough crust sits over a crumb that collapses, so the knife cuts a stiff shell resting on a cushion.

Why Do Razor Blades Go Dull If Hair Is 50 Times Softer Than Steel?
Now for the paradox everyone has met at the bathroom sink. The steel in a razor has a measured hardness of 8.7 gigapascals in the MIT study. The outer layer of a human hair is about 170 megapascals, a gap the paper calls "substantial (~50 times)." Yet a razor loses its edge after a few shaves. A hair, which is not known for its toughness, wins.
The MIT team put a blade inside an electron microscope and watched it cut single hairs. MIT's summary of the paper reports little rounding of the edge. Instead, there were "chips forming along certain regions of the razor's edge."
The chips did not appear at random. When the blade cut straight down through a hair, with no sideways force, nothing chipped. Tilt the blade 21° to the cutting direction, as it is in a real shave, and the hair bends and shoves one side of the edge. The chips "most commonly appear in correspondence to the edges of the hair." One hair can leave two of them, one from each side.
The last ingredient is the steel itself. Fine steel is a patchwork of grains and hard particles, some bigger than the 40 nm tip. Where a hair's edge lands on a weak patch at an angle, a microcrack starts and a chip flakes off. All of that has to line up at once, so the paper notes that "hair-induced chipping is not common." Typical blades "fail only after multiple uses." The lesson follows from the physics: a blade dies from sideways loads.

Which Metals Can Be Cut With A Knife?
A blade can only start a crack in something whose bonds give way before its own do. That one rule sorts the metals.
At the soft end sit the alkali metals: lithium, sodium and potassium. A Royal Society of Chemistry lab guide cuts them with "a scalpel or sharp knife." Of the three, "lithium is the hardest to cut." All three are "shiny when freshly cut and tarnish rapidly in air," which is why they live in bottles of oil. Sodium cuts like cold cheddar. Chemists find this less charming than it sounds, because of what it does next in water.
Mercury is off the list for a different reason. PubChem calls it a "mobile liquid metal, at room temperature." You do not cut a liquid.
At the hard end, the knife loses. On the Mohs scratch scale, a fingernail rates about 2.2, a copper penny 3.1, a steel nail 5 and glass 5.5, with diamond at 10. Anything harder than the blade's steel scratches the blade instead. So diamond is a cutting tool and not a thing you cut.

Can A Knife Split An Atom?
No, and the numbers say why. A knife breaks the bonds between atoms, which cost a few electron volts each. Splitting an atom means breaking up its nucleus. HyperPhysics puts nuclear binding energies at "millions of electron volts." The electrons that make chemical bonds are held with "tens of eV." Iron-56, the stuff of the blade, holds each particle in its nucleus with 8.8 million electron volts.
A million-fold gap is not something a sharper edge closes. We covered the paper version in are you splitting atoms when you tear paper, and the answer for a knife is the same. The atoms are pushed apart, whole and unharmed.
What Does Sharpening Actually Do?
A dull edge has lost its geometry. Either the tip has rounded, so ρ has grown and the stress multiplier has shrunk, or it has chipped. Sharpening grinds the two faces of the wedge back until they meet in a fresh, tight tip. Every stroke on a stone removes steel; nothing is "realigned."
The tighter the tip, the higher the stress at the crack it starts, and the less force you need. There is a catch in that formula. A tip a few hundred atoms wide is also a tip a bent hair can chip. Sharpness and fragility are the same property seen from two sides.

So, What Actually Happens At The Atomic Level When A Knife Cuts Something?
Put the pieces together. The edge, a few hundred atoms wide at best, lands on top of thousands of atoms and pushes. Force over that tiny area drives the stress under the edge past what the bonds there can take, and the first few break. That makes a crack, and the crack has a tip. Inglis' geometry piles the stress onto the tip, so the next bonds in line break next. Griffith's energy balance keeps the tip moving as long as the blade feeds it. The zipper opens.
Slide the blade and you load the crack tip more kindly, so half the push, or a tenth of it, does the job. And because the edge is a fine wedge of patchy steel, a sideways load from something as soft as a hair can flake a piece off it.
Nothing in that story goes between atoms, and nothing splits one. A knife is a machine for breaking bonds a row at a time. The whole art of sharpness is getting the first row to break with the least effort. Watch the skin of the next tomato you cut. It dents, and then, the moment you draw the blade back, it parts.

References (click to expand)
- How hair deforms steel — Roscioli, Taheri-Mousavi & Tasan, Science 369, 689–694 (2020)
- Why shaving dulls even the sharpest of razors — MIT News, 6 August 2020
- Slice–push, formation of grooves and the scale effect in cutting — A. G. Atkins, Interface Focus 6, 20160019 (2016), PMC4843631
- 11.3 Pressure — College Physics 2e, OpenStax
- Cracking Dams: History of fracture mechanics (Inglis 1913, Griffith 1920s) — Sethna group, Cornell University
- Can Objects Touch? — Ask the Van, University of Illinois Physics
- Bond Lengths and Energies — UCLA Molecular Biology Institute lecture notes
- 12.2.4 Sharpness — H. Föll, Iron, Steel and Swords, University of Kiel
- Reactivity trends of the alkali metals — Royal Society of Chemistry, Education
- Mercury — PubChem, National Library of Medicine
- Mohs Scale of Hardness — Arkansas Geological Survey
- Nuclear Binding Energy — HyperPhysics, Georgia State University







