Resistivity: the material's own grip on the current
You have spent this whole ladder asking how a material's structure sets a property — stiffness from the bonding-energy well, strength from how you tangle its dislocations. Electrical conduction is that same story told with electrons. Start with the everyday quantity, resistance R: push a voltage V across a wire and a current I flows, and R = V / I. But resistance is not a material property — a long thin wire resists more than a short fat one cut from the very same copper. Strip out the geometry and what is left is the material's own contribution: R = ρ x L / A, where L is length and A is cross-sectional area. That ρ is the electrical resistivity, measured in ohm-metres, and it is a true material property — a fixed number for a given material at a given temperature, no matter what shape you cut it into.
Flip it over and you get the twin quantity, electrical conductivity σ = 1 / ρ, in units of (ohm-metre)^-1, or siemens per metre — how freely charge flows rather than how hard it is held back. Engineers use whichever is handier: resistivity for insulators (big numbers), conductivity for metals (also big numbers). A useful picture: resistance is the whole plumbing job — a narrow pipe or a long run costs you flow — while resistivity is how thick the mud is inside, a property of the stuff itself. Copper's resistivity is about 1.7 x 10^-8 ohm-metres, so its conductivity is about 6 x 10^7 siemens per metre. Hold those two numbers; almost everything in this rung is a story about why some other material sits many powers of ten away from copper.
One property, twenty-four orders of magnitude
Here is what makes electrical resistivity unlike any property you have met so far. Young's modulus runs from soft rubber to stiff diamond over maybe four orders of magnitude; density spans about two. Resistivity spans about twenty-four. From copper at 10^-8 ohm-metres to polyethylene near 10^16, that is a factor of a million-billion-billion — the single widest range of any engineering property. Nature does not fill that range smoothly; it clumps materials into three great camps, and the classification you already know from the materials-classes rung turns out to be, at heart, an electrical one. Conductors are the metals: copper, aluminium, silver, iron, all near 10^-8. Insulators are most ceramics and polymers: alumina, glass, rubber, plastic, up at 10^10 to 10^16. And marooned in the vast gap between them sit the semiconductors: silicon and germanium, whose entire industry lives in that middle ground.
RESISTIVITY LADDER (ohm-metre, ~room temperature; conductivity = 1 / this) 10^-8 | silver 1.6e-8, copper 1.7e-8, aluminium 2.7e-8 CONDUCTORS 10^-7 | iron ~1e-7 (metals: an ocean 10^-6 | nichrome, stainless ~1e-6 of free electrons) ... | 10^0 | germanium ~0.5 SEMICONDUCTORS 10^3 | silicon ~2.5e3 (pure / intrinsic) (small band gap: ... | few carriers) 10^10 | 10^12 | alumina, borosilicate glass ~1e12 INSULATORS 10^14 | polystyrene, nylon ~1e14 (wide band gap: 10^16 | polyethylene, fused silica ~1e16 no free carriers) copper -> polyethylene: about 24 orders of magnitude, the widest of any property
Two honest caveats before we explain the ladder. First, the three camps are labels on a continuum, not walls — nichrome and stainless steel are 'poor conductors' at 10^-6, sitting a hundred times above copper yet still unmistakably metallic. Second, and more subtly, the label really tracks the size of a material's band gap, not its resistivity at one fixed temperature — because a semiconductor can be coaxed to conduct almost like a metal by warming it or doping it, while it insulates like glass when cold and pure. So do not memorise 'silicon = 2500 ohm-metres' as a fact of silicon; that is the resistivity of pure silicon at room temperature, and guide 4 will move it by a factor of a million just by dissolving in a pinch of phosphorus. The classes are best defined by why they conduct — which is where energy bands come in.
Where conduction comes from: a first look at bands
Why should copper hand electrons around so freely while diamond — also made of nothing but atoms and bonds — refuses? The answer is the energy band picture, and guide 2 builds it properly; here is just the orientation you need to read the rest of this guide. In a single atom, electrons may sit only at certain sharp energy levels. Pack 10^23 atoms into a solid and each of those levels smears into a near-continuous band of allowed energies. Two bands matter: the valence band, the lower one that the bonding electrons fill, and the conduction band above it, normally empty. Between them may lie a forbidden band gap — a range of energies no electron is allowed to have at all.
Now the key rule: an electron can carry current only if there is an empty energy state right beside it to move into. A completely full band is gridlocked — every seat taken, no one can shuffle — so it carries nothing. That single rule sorts the three camps. In a metal the valence band is only partly filled (or overlaps the conduction band), so there are empty seats right at the top of the electron sea — the electrons at the Fermi level, the energy up to which states are filled — and they drift the instant you apply a field. In an insulator the valence band is full and the conduction band is empty, with a wide gap between (diamond's is about 5.5 electron-volts): to conduct, an electron must jump the gap, and 5.5 eV is a cliff nothing can climb at room temperature. A semiconductor is the same picture with a small gap — silicon's is about 1.1 eV — a step low enough that a few electrons, kicked by ordinary heat, do make the jump, leaving conduction electrons above and empty seats called holes below. That is the band gap as a step the electron must climb to conduct, and its height alone separates a diamond from a silicon chip.
Counting the carriers: sigma = n e mu
How much a material conducts boils down to three factors, tied by one tidy equation: σ = n x |e| x μ. Here n is the number of mobile charge carriers per cubic metre, |e| is the charge each one carries (1.6 x 10^-19 coulombs, the same for every electron), and μ is the electron mobility — how fast a carrier drifts for a given push, in metres-per-second per volt-per-metre. Conductivity is just carriers, times their charge, times how nimble they are. This little formula is the workhorse of the whole rung: it says a material can conduct well either by having many carriers or by letting each one move freely, and it lets us see exactly where a metal and a semiconductor part ways.
Put in real numbers and the answer is startling. Copper carries roughly one free electron per atom, so n is about 8.5 x 10^28 carriers per cubic metre — an ocean. Pure silicon at room temperature has only about 1.5 x 10^16 carriers per cubic metre, the few that thermal kicks have knocked across the gap. That is thirteen orders of magnitude fewer carriers, and it is essentially the entire reason silicon insulates while copper conducts. Here is the twist that surprises everyone: silicon's carriers are actually more mobile than copper's — silicon's electron mobility is about 0.14 versus copper's roughly 0.003 in the same units. The metal wins conduction not because its electrons are nimble but because it owns an unthinkable crowd of them. Carrier count, not mobility, writes the headline.
Why heat and dirt do opposite things to metals and semiconductors
Guide 3 is devoted to this reversal, but it is too neat to leave unpreviewed. In a metal, mobility is limited by scattering: an electron drifting through a perfect, still lattice would sail on forever, but anything that breaks the perfect periodicity knocks it off course and adds resistance. Two things break it. Heat sets the atoms vibrating (those vibrations are called phonons), and the hotter it gets the more the electron collides — so a metal's resistivity rises with temperature, climbing almost linearly, which is why an incandescent filament resists far more when glowing than when cold. Impurities and defects break it too: a foreign solute atom, a grain boundary, or the dislocation tangle from cold-work each scatter electrons. The two contributions simply add (Matthiessen's rule), so pure annealed copper conducts best and every alloying atom you dissolve in costs you conductivity.
You have seen that scattering picture before, wearing different clothes. Every trick that strengthened a metal earlier in this ladder — dissolving in solute atoms, refining the grains, cold-working to pile up dislocations — works by littering a dislocation's path with obstacles. The very same obstacles sit in an electron's path, which is why strengthening a conductor almost always makes it a worse conductor: pure copper is soft but superbly conductive, while the harder copper alloys, brass and bronze, conduct noticeably less. Power-line engineers feel this tradeoff daily, choosing between pure-but-weak and strong-but-resistive metal. Now the semiconductor does the opposite. Warm a semiconductor and yes, its scattering worsens too — but heat also flings vastly more electrons across that small gap, and the flood of new carriers overwhelms the extra scattering. So a semiconductor's resistivity falls as it heats up, the exact reverse of a metal, and that reversal is the whole principle behind a thermistor.
The rest of the electrical family: doping, junctions, and dielectrics
The band picture and σ = n e μ open two more doors that the remaining guides walk through. The first is the whole of semiconductor engineering. Because a pure (intrinsic) semiconductor has so few carriers, its conductivity is exquisitely tunable: dissolve in a trace of an element with one spare electron (phosphorus in silicon) and you flood the conduction band with negative carriers — an n-type semiconductor; dissolve in one that is short an electron (boron) and you create positive holes — a p-type semiconductor. Join a p-type piece to an n-type piece and you get a p-n junction, a one-way valve for current that is the seed of every diode, transistor, solar cell, and LED. Guide 4 makes all of this concrete; for now just note that it is nothing more than deliberately setting n, atom by atom.
The second door leads the other way — to materials that do not conduct at all, and are prized for it. In an insulator no charge crosses the material, but an applied field can still nudge the bound charges a whisker within each atom, a response called polarization. That is what lets a dielectric material store energy in a capacitor: slip such a material between two charged plates and it holds far more charge at the same voltage, its dielectric constant telling you how many times more. Push the field too hard, though, and even an insulator gives way — its dielectric strength is the field at which it breaks down and arcs through, the reason a spark eventually jumps a gap and a cable carries a voltage rating. Guide 5 covers dielectrics in full.
One family sits at the crossroads of the mechanical rung and this one, and it is worth meeting before we part. In a few crystals whose structure lacks a centre of symmetry, squeezing the crystal shifts its internal charges and produces a voltage — and applying a voltage in turn makes it change shape. These are the piezoelectric materials: charge from stress, and strain from charge, the same effect running both ways. Quartz keeps time in a watch by vibrating this way; the ceramic in a barbecue igniter throws a spark when you squeeze it; ultrasound scanners and precision actuators live on the effect. A cousin group, the ferroelectrics, even hold a switchable built-in polarization that traces a hysteresis loop just like a magnet's. Those close out the rung. So the map ahead: guide 2 builds the bands you just glimpsed, guide 3 explains the temperature reversal, guide 4 dopes silicon into devices, and guide 5 turns to dielectrics and these charge-from-stress materials. The single thread through all five is the one you started with — a material's electrical behaviour is written by what its electrons are allowed to do.