Where Ceramics Earn Their Keep: Heat
By now you can build a ceramic from the ground up: atoms held in a rigid cage by strong ionic-covalent bonds, powder pressed and fired into a dense, grainy solid. That same rigid, strongly bonded cage is exactly why we ask ceramics to do the jobs no metal or plastic can survive — line a furnace at 1600 degrees C, armour a rocket nozzle, tile the belly of a spacecraft for re-entry. To design for those jobs you have to know how a ceramic deals with heat, and 'dealing with heat' is really several different questions wearing one coat.
This rung takes those questions one at a time, and this first guide lays the two foundations named in its title: how a ceramic stores heat, measured by its heat capacity, and how it conducts heat, measured by its thermal conductivity. The next four guides build on these: guide 2 on thermal expansion and the microcracking its anisotropy can trigger, guide 3 on conduction by phonons in depth, guide 4 on the villain of the rung, thermal shock, and guide 5 on designing against it. Store and conduct first; everything else hangs off them.
Guard against one tempting shortcut right away: there is no single number for 'heatproof'. Heat capacity (how much energy a degree of warming costs) is a different property from thermal conductivity (how fast heat flows through), which is different again from thermal expansion (how much the body swells), from refractoriness (how high a temperature it tolerates without softening), and from thermal-shock resistance (whether it survives a sudden temperature change). A firebrick that happily sits red-hot for years can shatter the instant you splash cold water on it. Keep these separate and the whole rung stays clear.
Storing Heat: The Heat Capacity
Heat capacity is the price of warmth: the energy you must pour in to raise the temperature by one degree. Where does that energy go? In a metal some of it goes into speeding up the sea of free electrons — but a ceramic is an electrical insulator with almost no free electrons, so essentially all of it goes into making the atoms vibrate harder on their bond-springs. Picture every atom as a ball tethered to its neighbours by stiff springs, jiggling in three directions; heating the crystal is winding up all those little oscillators. This is the very same lattice vibration we will meet again as the carrier of heat flow — first it is where stored energy lives, then it is how energy travels.
How much can those oscillators hold? Classical physics gives a beautifully simple answer, the Dulong–Petit rule: each vibrating atom, jiggling in three directions, stores about 3 times k_B times T of energy, so a mole of atoms takes about 3R = 3 times 8.31 = 25 J per degree to warm. The key phrase is per mole of atoms. Take alumina, Al2O3: five atoms per formula unit, so its high-temperature molar heat capacity climbs to about 5 times 25 = 125 J/(mol K). Its molar mass is about 102 g, so that is roughly 1.2 J to warm one gram by one degree — and a light-atom oxide like MgO stores even more per gram, because you are paying 25 J per mole of atoms and light atoms pack more moles into every kilogram.
- Count the atoms in one formula unit. For Al2O3 that is 2 + 3 = 5 atoms; for SiC it is 2; for ZrO2 it is 3.
- Multiply by the classical per-mole-of-atoms value: molar heat capacity is about (atoms per formula) times 25 J/(mol K). For Al2O3 that is 5 times 25 = 125 J/(mol K).
- Add up the molar mass in grams from the atomic weights: Al2O3 is 2 times 27 + 3 times 16 = 102 g/mol.
- Divide to get the specific (per-gram) heat: 125 / 102 is about 1.2 J/(g K). That is the ceiling the material approaches at high temperature.
- Remember the fine print: this classical ceiling only holds well above the material's Debye temperature. At room temperature many stiff, light ceramics have not reached it — measured alumina is nearer 0.8 J/(g K) at 25 degrees C — because some vibration modes are still quantum-frozen and cannot yet soak up energy. This is what 'where the heat capacity saturates' really means.
Conducting Heat: A Rumour Passed Down the Lattice
Now let heat flow rather than just sit. Fourier's law says the heat flux is proportional to the temperature gradient, q = -k times (dT/dx), and the constant k is the thermal conductivity — big k, heat pours through; small k, it dribbles. In a metal the messengers are those same free electrons, sprinting down the gradient. A ceramic has no such runners, so it carries heat a slower way: by phonons, quantized packets of lattice vibration that hand their jiggle from atom to atom like a rumour whispered down a line of people. A hot atom vibrates hard, jostles its neighbour, and the disturbance ripples down the bonds.
Why is a phonon a slow, fragile messenger? Because it keeps getting scattered — knocked off course so the whisper has to start over. Heat the crystal and you fill it with more phonons that collide with one another, so in a clean crystal k falls roughly as 1 over T as temperature rises. Worse, anything that breaks the perfect repeating lattice scatters phonons too: foreign atoms dissolved in solid solution, grain boundaries, and above all pores all chop the distance a phonon travels between mishaps. Each obstacle shortens the phonon mean free path, and phonon scattering is the master lever behind almost every conductivity number you will meet in guide 3.
The Ceramic Conductivity Spectrum
Room-temperature thermal conductivity k [ W/(m K) ] ( dense unless noted; ordered high -> low ) diamond .............................. ~2000 SiC (dense) ........................ ~120 AlN .................................. ~180 alumina Al2O3 (dense) ............... ~30 quartz crystal (SiO2) ............... ~6-10 YSZ zirconia (dense) ............... ~2 fused silica glass (SiO2) ........... ~1.4 porous insulating firebrick .......... ~0.1 still air (for comparison) .......... ~0.025 metals for scale: copper ~400, steel ~50
Read that list and one myth dies: 'electrical insulator' does not mean 'poor heat conductor'. Diamond, SiC, and AlN are all electrical insulators, yet they carry heat as well as or better than steel, purely on phonons. The recipe for a phonon superhighway is light atoms, stiff bonds, and a clean, simple, well-ordered lattice — light and stiff make fast waves, order lets them travel far. Ruin any ingredient and conductivity collapses: zirconia is heavy, and its lattice is deliberately disordered by the stabilizing dopants that hold its cubic form, so it sits near the bottom at about 2. And porosity is the master switch of all — foam a ceramic full of air pockets and its conductivity plunges toward that of the trapped air itself.
This huge range is a gift, because ceramics get asked for both extremes. Sometimes you want the highest conductivity you can buy: a silicon-carbide or AlN substrate that whisks heat out of a power chip, or a heat-spreader that resists thermal shock by never letting a hot spot build. Other times you want the lowest: a porous, disordered zirconia thermal-barrier coating sprayed on a turbine blade, a thin insulating skin that lets the metal underneath run in a gas stream far hotter than it could otherwise bear. Same physics of phonons and scattering, aimed in opposite directions — and this is a reminder that porosity is often engineered on purpose, not just a defect. Guide 5 returns to that coating in detail.
From Storing and Conducting to Surviving
Storing and conducting are not really separate in service; the quantity that governs how a ceramic responds to a changing temperature blends them into one: the thermal diffusivity, a = k / (rho times c_p), conductivity divided by density times heat capacity. It answers a different question from k. Conductivity asks how much heat flows in steady state; diffusivity asks how fast a temperature change spreads inward — and a high heat capacity actually slows the spread, because every layer soaks up energy before passing it on. For dense alumina a is around 10 mm^2/s; for zirconia, with its low k, it is nearer 1 — meaning a zirconia part equalizes temperature about ten times more sluggishly, which as we will see is a double-edged trait.
The same strong bonds that make ceramics stingy heat conductors also make them extraordinarily hard to melt. Undoing that rigid ionic-covalent cage takes enormous thermal energy, so ceramics boast the highest melting points known: alumina melts near 2054 degrees C, MgO near 2852, and the refractory carbides such as HfC push past 3900 — which is exactly why we line furnaces and rocket throats with them. But 'does not melt' is not 'does not move': held hot and loaded for long enough, ceramics slowly deform by high-temperature creep, grains sliding past one another and any glassy grain-boundary film softening and flowing. Refractoriness sets the ceiling; creep sets how close to it you dare run for years.