The one diagram every engineer knows
By now you can read a phase diagram fluently. You know a diagram names the components and the phases, that solidus and liquidus fence off where liquid and solid live, that a horizontal tie line plus the see-saw of the lever rule hands you the amount of each phase, and that the invariant reactions — eutectic, peritectic, eutectoid — each let one phase split or merge at a single fixed point. This last guide spends all of that skill on one diagram: iron and carbon, the alloy system that quite literally built the modern world. Steel is roughly nine of every ten tonnes of metal humanity makes, and the iron-carbon diagram is the metallurgist's roadmap for every last one of them.
The magic is the sheer leverage of carbon. Pure iron is soft and unremarkable; add a fraction of one percent of carbon and you can make anything from a bendable paperclip to a razor edge to a bridge girder. Even pure iron already does something surprising on its own: it is polymorphic (it shows allotropy), meaning it swaps crystal structures as it heats. Room-temperature iron is body-centered cubic; heat it past 912 degrees C and it rearranges into face-centered cubic; heat it further and it flips back. That built-in shape change, combined with carbon, is the whole engine of the diagram.
Meet the phases: ferrite, austenite, cementite
Because iron changes its own crystal structure with temperature, it offers carbon two very different homes. At room temperature and up to 912 degrees C, iron is body-centered cubic, and iron in this form is called ferrite (the alpha phase). Between 912 and 1394 degrees C iron is face-centered cubic, and this form is austenite (the gamma phase). Same element, two crystal packings, two personalities — and carbon lives in each of them by squeezing into the gaps between the iron atoms, an interstitial solid solution rather than a swap of one atom for another.
Here is the twist that catches everyone. Ferrite (BCC) dissolves almost no carbon — at most 0.022 weight percent, and essentially nothing at room temperature. Austenite (FCC) dissolves a great deal — up to 2.14 weight percent, a hundred times more. Yet FCC is the more densely packed structure (packing factor 0.74 versus BCC's 0.68). How can the tighter-packed crystal swallow more carbon? Because it is not about total empty space but about the size of the individual holes: FCC's largest interstitial gaps (the octahedral holes) are bigger than BCC's, so a carbon atom fits into one without straining the lattice as much. This counterintuitive fact — the denser crystal holds more carbon — is the single most important thing on the whole diagram, because it means cooling austenite forces its dissolved carbon back out.
That expelled carbon has to go somewhere, and it goes into the third phase: cementite, Fe3C, iron carbide. Cementite is not a solid solution but an intermetallic compound with a fixed composition of 6.70 weight percent carbon — it sits as a single vertical line on the far right of the diagram. It is hard and brittle, almost ceramic in character: wonderful for resisting scratches and wear, hopeless on its own because it shatters. So a steel is really a partnership — soft, tough, forgiving ferrite reinforced by hard, strong, brittle cementite. Neither is much use alone; together they are the most versatile structural material we have.
The eutectoid reaction and the birth of pearlite
Now the pieces snap together. From guide 4 you know the eutectoid reaction is the solid-state twin of the eutectic: on cooling, a single solid phase splits into two different solids at one fixed point. In the iron-carbon system that point sits at 727 degrees C and 0.76 weight percent carbon, and the reaction is austenite going to ferrite plus cementite. It has to happen this way: as carbon-rich austenite cools past 727 degrees, its FCC lattice reverts toward BCC ferrite, which cannot hold the carbon, so the rejected carbon piles up into cementite right alongside. Three phases meet at that instant, the Gibbs phase rule pins the degrees of freedom at zero, and so — like every invariant reaction — it proceeds at a single temperature and a single composition.
Iron-carbon (Fe-Fe3C) diagram -- schematic, NOT to scale
T/C L (liquid)
^ L+gamma __4.30%__ L+Fe3C
1147 +=============[EUTECTIC]=============+ L -> gamma + Fe3C
| gamma (AUSTENITE, FCC) |
| dissolves C up to 2.14% |
| |
727 +=======[EUTECTOID]=================+ | gamma -> alpha + Fe3C
| 0.76% | | ( = PEARLITE )
| alpha + Fe3C | |
| (ferrite + cementite) Fe3C = CEMENTITE
+----+------+---------------+----------+---> wt% C
0.022 0.76 2.14 6.70
ferrite | | cementite
max C | <-- STEEL --> | <- CAST IRON ->Because both products crystallize together out of the same parent grain, they grow side by side as alternating flat plates — thin layers of ferrite and cementite stacked like plywood, or the rings of a tree, or the leaves of a flaky pastry. This layered two-phase structure is pearlite, named because the fine stripes scatter light with a pearly sheen under the microscope. Run the lever rule across the tie line from ferrite (0.022%) to cementite (6.70%) at just below 727 degrees and you find pearlite is about 89 percent ferrite and 11 percent cementite by mass — so the soft ferrite layers come out roughly eight times thicker than the hard cementite layers between them.
Hypo, eutectoid, hyper: reading any steel
Almost no real steel sits exactly at 0.76 percent carbon, so the diagram sorts them into three families. A steel with less carbon (hypoeutectoid, below 0.76%) cools as pure austenite until it crosses into the two-phase field, where it first sheds soft proeutectoid ferrite onto the austenite grain boundaries; the remaining austenite grows richer in carbon until it hits 0.76%, then transforms wholesale to pearlite. The finished microstructure is grains of clean ferrite dotted with islands of layered pearlite. A steel with more carbon (hypereutectoid, above 0.76%) does the mirror image: it sheds proeutectoid cementite as a brittle network first, then the rest becomes pearlite — harder, but more fragile because of that cementite web.
- Take a common structural steel with 0.40 percent carbon, cooled slowly to just below 727 degrees C. It is hypoeutectoid. First split it into microstructural constituents using the tie line from ferrite (0.022%) to the eutectoid (0.76%).
- Fraction of pearlite = (0.40 - 0.022) / (0.76 - 0.022) = 0.378 / 0.738, about 0.51. So roughly 51 percent pearlite and 49 percent proeutectoid ferrite — very nearly half and half, the lever rule balancing like a see-saw.
- Now ask a different question — how much total cementite of any kind? Extend the tie line all the way to cementite (6.70%): fraction cementite = (0.40 - 0.022) / (6.70 - 0.022) = 0.378 / 6.678, about 0.057. Only about 6 percent cementite by weight; the other 94 percent is ferrite.
- Read the lesson: a tiny weight of carbon makes only a few percent of cementite, yet that small hard fraction is what does the strengthening. A little carbon buys a lot of hardness — but pays for it in ductility, exactly the trade-off you have met all the way up this ladder.
This is why a plain-carbon steel's strength rises steadily from soft, ductile low-carbon iron up toward hard, wear-resistant high-carbon steel. More carbon means more pearlite, means more cementite lamellae standing in the way of gliding dislocations, so yield strength and hardness climb while ductility and toughness fall — the same strength-ductility trade you know well by now. Past the eutectoid, the brittle proeutectoid cementite network wrapped around the grains makes the fall in toughness sharper still, which is why the hardest, most wear-resistant tool steels are also the ones most likely to chip.
The honest limits of the map
Here is the most important caveat in this whole rung, so hold it tight: a phase diagram is an equilibrium map. Every phase, every amount, every reaction on it assumes you cooled slowly enough for diffusion to keep pace. But the most useful steel structures are deliberately not at equilibrium and therefore appear nowhere on this diagram. Quench austenite fast in water and the carbon has no time to diffuse out; instead the FCC lattice snaps into a distorted, carbon-trapped structure called martensite, harder than anything the equilibrium diagram allows. Cool at an in-between rate and you get bainite, or fine pearlite finer than slow cooling ever makes. None of these are on the Fe-Fe3C diagram, because it only knows the resting destination, not the route.
There is a second, sneakier asterisk. Cementite itself is only metastable. The true thermodynamic ground state of iron-and-carbon is iron plus graphite, not iron plus Fe3C — cementite is a long-lived impostor that never quite gets around to breaking down. That is exactly why we call it the Fe-Fe3C diagram and not the Fe-C diagram. In the cast irons on the right of the map (above 2.14% carbon), the carbon often does win its way out as graphite: as flakes in gray cast iron, or as neat spheres in ductile cast iron. So even the word equilibrium here comes with a footnote, and a good metallurgist keeps both diagrams — the metastable Fe-Fe3C and the true Fe-graphite — in mind.
Step back and see how far you have come. You can now take any binary diagram, pull a tie line, run the lever rule as a see-saw, name a eutectic from a eutectoid, and read the phases and amounts straight off the map — and you have just aimed the whole toolkit at steel, the alloy that framed our buildings, laid our rails, and edged our tools. What this equilibrium map cannot answer is the question a metallurgist actually gets paid to control: what happens when you do not cool slowly. That question — nucleation, growth, the ticking clock of diffusion, and the TTT and CCT curves that chart it — is exactly where the next rung begins, and where a phase diagram finally becomes a heat-treatment recipe.