Classification of materials

Pick up a copper wire, a glass rod and a sliver of silicon and hold them side by side. All three are ordinary solids, packed with roughly the same jaw-dropping number of atoms — about $10^{23}$ of them per cubic centimetre — yet one conducts electricity almost perfectly, one refuses to conduct at all, and the third sits in between, quietly waiting for a nudge of heat or an added impurity to wake it up. Copper lights your room the instant a switch closes; glass would leave the room dark for centuries even under full mains voltage; and silicon, the strange in-between material, is the reason a phone in your pocket can think at all. The difference is not about which atoms are "good conductors" and which are not — carbon in diamond and carbon in graphite behave completely differently, so it cannot just be the element. The real answer is about what happens to the outermost electrons when roughly $10^{23}$ atoms are packed shoulder to shoulder into a solid. Their once-sharp, individual energy levels smear into continuous bands of allowed energy, with a possible forbidden zone — the energy gap — between them. How wide that gap is, and whether the bands overlap at all, is the single qualitative fact from which the entire three-way classification of solids falls out. This article builds that band picture from scratch: what a band is, why a gap can exist, and how comparing the gap size to the tiny thermal energy available at room temperature ( $k_BT \approx 0.026\ \text{eV}$ ) sorts every solid you will ever touch into a conductor, an insulator, or a semiconductor. It is also the doorway into the whole "Semiconductor Electronics" chapter — the moment you can explain why silicon, not diamond, sits inside every digital device, you are ready for how impurities turn a semiconductor into something even more useful.

Key takeaways

Atoms in a solid share their outer energy levels, which spread out into continuous energy bands — the valence band (bound outer electrons) and the conduction band (electrons free to carry current).
The energy gap $E_g$ is the forbidden zone between these bands where no electron state can exist; its size is the single number that decides a solid's electrical identity.
Conductors have overlapping or partly filled bands — effectively $E_g = 0$ — so free electrons exist even near absolute zero.
Insulators have a large gap (typically greater than about $3\ \text{eV}$ , e.g. diamond $\approx 5.4\ \text{eV}$ ), far beyond the reach of room-temperature thermal energy $k_BT \approx 0.026\ \text{eV}$ .
Semiconductors have a small gap (roughly $0.2$ to $3\ \text{eV}$ ; silicon $\approx 1.1\ \text{eV}$ , germanium $\approx 0.7\ \text{eV}$ ) — an insulator at $0\ \text{K}$ , but weakly conducting at room temperature as a few electrons cross the gap.
Conductivity trends run opposite ways with temperature: a semiconductor's conductivity rises sharply as $T$ increases (more electrons cross the gap), while a metal's conductivity falls slightly (lattice vibrations scatter the already-free electrons more).
This qualitative band picture — no derivation required by the syllabus — is exactly why silicon and germanium, not copper or diamond, are the materials chosen to build diodes and every other semiconductor device.
Key idea: the size of the energy gap $E_g$ between the valence and conduction bands, compared with the thermal energy $k_BT$ available at room temperature, is what separates conductors (no gap), insulators (huge gap), and semiconductors (small gap that a few electrons can jump).

The core insight

Whether a solid conducts, insulates, or sits in between as a semiconductor is decided by one qualitative fact — how large the forbidden energy gap $E_g$ is between its filled valence band and its empty conduction band, judged against the tiny thermal energy $k_BT \approx 0.026\ \text{eV}$ available at room temperature.

Where it fits in the syllabus

This article opens Unit IX, "Semiconductor Electronics: Materials, Devices and Simple Circuits," which the CBSE 2025-26 prescription keeps at a strictly qualitative level for band theory — no derivation of how bands arise from quantum mechanics is expected, only the picture and its consequences. The chapter narrows sharply after this point: only intrinsic and extrinsic semiconductors, the p-n junction, and the diode as a rectifier remain in the current syllabus, while Zener diodes, LEDs, photodiodes, solar cells, transistors, and digital logic gates have all been rationalised out for 2025-26 even though older textbook copies still print them. Everything you read here about energy bands is the essential foundation the rest of the (now-shorter) chapter is built on.

The physical picture

In a single, isolated atom, an electron can only occupy certain sharp, discrete energy levels — think of the neat spectral lines you may have seen for hydrogen. Bring two atoms close together and their electron wavefunctions start to overlap; each original sharp level splits into two closely spaced levels, one slightly higher and one slightly lower in energy, because the electrons of one atom now "feel" the presence of the other. Now imagine not two atoms but roughly $10^{23}$ of them, all locked into a rigid crystal lattice. Each atomic level splits not into two but into $10^{23}$ sub-levels, packed so closely together that they merge into what looks, for all practical purposes, like a continuous smear of allowed energy — an energy band. Two bands matter for electrical behaviour. The valence band is the (usually highest) band that is occupied by the electrons still tied to individual atoms — the outer, bonding electrons. The conduction band is the next band up, the range of energies in which an electron is no longer bound to any one atom and is free to drift through the crystal and carry current. Between the top of the valence band and the bottom of the conduction band there may — or may not — be a forbidden range of energies where no electron state exists at all, called the energy gap or band gap, $E_g$ .

E_g = E_C - E_V \qquad (1)

where: $E_C$ is the energy at the bottom edge of the conduction band, $E_V$ is the energy at the top edge of the valence band, and $E_g$ (measured in electron-volts, $\text{eV}$ ) is simply the size of the forbidden window between them. A physical reading of equation (1): $E_g$ is not a place an electron can sit — it is the minimum energy a valence electron must be handed, from somewhere, before it can join the conduction band and start carrying current. If $E_g$ happens to be zero or negative (the bands touch or overlap), there is no obstacle at all, and free electrons are available no matter how cold the solid is. The three-way classification of solids follows directly from comparing $E_g$ with the thermal energy an electron typically has available at a given temperature, $k_BT$ , where $k_B = 1.38\times10^{-23}\ \text{J/K}$ is the Boltzmann constant (or, in the friendlier electron-volt units used throughout this chapter, $k_BT \approx 0.026\ \text{eV}$ at room temperature, $T \approx 300\ \text{K}$ ):

Conductors (metals): the conduction and valence bands overlap, or the conduction band is only partly filled to begin with, so mobile electrons exist even at very low temperature. Effectively $E_g \approx 0$ .
Insulators: $E_g$ is large — typically greater than about $3\ \text{eV}$ (diamond $\approx 5.4\ \text{eV}$ ). The available thermal energy, $k_BT \approx 0.026\ \text{eV}$ , is utterly dwarfed by this gap, so essentially no valence electron can ever cross it at room temperature; the valence band stays full and the conduction band stays empty.
Semiconductors: $E_g$ is small — roughly $0.2$ to $3\ \text{eV}$ (silicon $\approx 1.1\ \text{eV}$ , germanium $\approx 0.7\ \text{eV}$ ). At $0\ \text{K}$ a pure semiconductor behaves exactly like an insulator — a full valence band, an empty conduction band — but at room temperature a small yet significant fraction of electrons acquire enough thermal energy to cross the narrow gap, giving a small conductivity that climbs steeply as $T$ rises.

Figure

Three side by side band diagrams for a conductor an insulator and a semiconductor, each showing a shaded valence band and an empty conduction band with a gap of different size between them, labelled with approximate energy gap values

This single comparison — $E_g$ against $k_BT$ — is why semiconductors are so useful in the first place: unlike a metal, whose conductivity is essentially fixed, a semiconductor's conductivity can be controlled, either by raising the temperature or, far more usefully for building devices, by deliberately adding tiny amounts of the right impurity atoms — the subject of the very next article in this chapter.

Derivation AComparing the energy gap with thermal energy to decide whether a solid conducts

The goal here is not a proof from first principles — the syllabus explicitly treats band theory qualitatively — but a careful, honest argument for why comparing $E_g$ to $k_BT$ is the right test, and how that single ratio explains both why silicon conducts a little at room temperature while diamond conducts essentially nothing, and why that same silicon is a perfect insulator at absolute zero. The reasoning uses only the two numbers already introduced, $E_g$ and $k_BT \approx 0.026\ \text{eV}$ at $T = 300\ \text{K}$ , and the physical fact that the fraction of electrons able to acquire a given amount of thermal energy falls off extremely fast — roughly exponentially — as that amount grows relative to $k_BT$ .

Step 1. Set up the physical question. At any temperature above absolute zero, the atoms of a crystal are vibrating, and this thermal agitation constantly hands out small, random packets of energy to the electrons, of a typical size $k_BT$ . For an electron in the valence band to become a mobile, conduction-band electron, it must receive at least $E_g$ of energy in one such kick — nothing smaller will do, because there are no allowed states inside the gap for it to sit in partway across.

Step 2. Form the ratio $E_g/k_BT$ for each class of solid. For silicon, $E_g/k_BT = 1.1/0.026 \approx 42$ . For diamond, $E_g/k_BT = 5.4/0.026 \approx 208$ . For a conductor the ratio is effectively zero or undefined, since there is no gap to cross at all — free electrons are already sitting in the (partly filled or overlapping) conduction band.

Step 3. Interpret what a large ratio means physically. The number of electrons energetic enough to have picked up an amount $E_g$ or more of thermal energy shrinks very rapidly — roughly like $e^{-E_g/k_BT}$ — as $E_g/k_BT$ grows. A ratio near $40$ – $45$ , as for silicon, still leaves a small but genuinely measurable population of electrons crossing the gap at room temperature. A ratio above $200$ , as for diamond, drives that population down to a number so vanishingly small that no practical current flows; diamond behaves, for every everyday purpose, like a perfect insulator at $300\ \text{K}$ .

Step 4. Extend the same logic to $T \to 0\ \text{K}$ . As $T \to 0$ , $k_BT \to 0$ , so $E_g/k_BT \to \infty$ for any solid with a nonzero gap — meaning that at absolute zero, even silicon and germanium behave exactly like insulators, with a completely full valence band and a completely empty conduction band. This is exactly the qualitative fact stated in the syllabus: a pure semiconductor is a perfect insulator at $0\ \text{K}$ , and only becomes weakly conducting once the temperature is raised enough for the ratio $E_g/k_BT$ to fall to a workable size again.

Step 5. Read off the classification rule this yields, without needing the exact exponential formula (which lies beyond this qualitative treatment). Small $E_g/k_BT$ at the temperature you care about means significant conduction (a conductor, or a "turned on" semiconductor); very large $E_g/k_BT$ means essentially no conduction (an insulator, or a semiconductor left too cold to conduct). The size of $E_g$ relative to $k_BT$ , not the size of $E_g$ alone, is therefore the physically correct comparison — the same $1.1\ \text{eV}$ gap that makes silicon a workable semiconductor at $300\ \text{K}$ would make it behave like an insulator if you could somehow cool it to a few kelvin.

Worked examples

Board level - comparing silicon and diamond at room temperature

The energy gap of silicon is about $1.1\ \text{eV}$ and that of diamond is about $5.4\ \text{eV}$ . At room temperature ( $T \approx 300\ \text{K}$ ), the average thermal energy available to an electron is roughly $k_BT \approx 0.026\ \text{eV}$ . Using these numbers, explain why silicon shows measurable conductivity at room temperature but diamond does not.

Step 1. Compare the available thermal energy with each gap by forming the ratio $E_g/k_BT$ . For silicon: $E_g/k_BT = 1.1/0.026 \approx 42$ . For diamond: $E_g/k_BT = 5.4/0.026 \approx 208$ .

Step 2. Interpret the ratio physically. The fraction of electrons able to jump the gap falls off very rapidly as this ratio increases. A ratio of about $42$ for silicon still allows a small but significant number of valence electrons into the conduction band at $300\ \text{K}$ .

Step 3. State the conclusion. For diamond, the ratio (about $208$ ) is so large that essentially no electrons can cross the gap at room temperature — diamond remains an excellent insulator. Silicon's smaller gap allows just enough electrons across to make it a working semiconductor, which is why silicon (and germanium), not diamond, are used in semiconductor devices.

Competitive level - identifying an unknown material from its temperature behaviour

A material has its valence band completely full and its conduction band completely empty at $0\ \text{K}$ , with a definite gap between the two bands. As temperature is raised from $0\ \text{K}$ to room temperature, its conductivity increases sharply. Classify this material and contrast its behaviour with that of a typical metal as temperature rises.

Step 1. Identify the class. A filled valence band and empty conduction band at $0\ \text{K}$ describes both insulators and semiconductors — the size of the gap decides which. Since the conductivity increases sharply with only a modest rise in temperature, the gap must be small enough for thermal excitation to matter over that range, identifying the material as a semiconductor.

Step 2. Explain the mechanism behind the sharp rise. As $T$ increases, a growing number of valence electrons acquire enough thermal energy $k_BT$ to cross the small gap $E_g$ into the conduction band. Every electron that crosses becomes a mobile charge carrier, so the material's conductivity rises with temperature.

Step 3. Contrast this with a metal. In a metal, the conduction band is already partly filled or overlaps the valence band, so free electrons exist even at very low temperature — there is no gap to wait for. As temperature rises in a metal, conductivity actually decreases slightly, because increased lattice vibrations scatter the already-available free electrons more often, hindering their motion rather than adding new ones. Answer: the material is a semiconductor; its conductivity rises with temperature (opposite to a metal, whose conductivity falls slightly) because new carriers are being thermally generated across a small gap rather than merely being scattered. Timing note: this kind of "classify and contrast" question should take under two minutes once you know the rule — check whether conductivity rises (semiconductor: new carriers) or falls (metal: same carriers, more scattering) with rising temperature, and let that single trend fix the classification before you write a word of explanation.

Common traps

Thinking insulators have no conduction band at all. Every solid has one; in an insulator it is simply empty because the gap $E_g$ is too large for any practical number of electrons to cross at ordinary temperatures.
Confusing "small gap" with "no gap." Semiconductors have a genuine forbidden gap and are not conductors in disguise — at $0\ \text{K}$ a pure semiconductor is a perfect insulator, full stop.
Assuming semiconductor conductivity behaves like metallic conductivity with temperature. It is the opposite trend: semiconductors conduct better as $T$ rises, metals conduct slightly worse.
Believing the valence band is where "free" conduction electrons live. Mobile, current-carrying electrons belong to the conduction band; the valence band holds the more tightly bound outer electrons of the atoms.
Quoting $E_g$ as if it alone decides everything, without reference to $k_BT$ . The same numerical gap can behave like an insulator's gap at low temperature and a semiconductor's gap at room temperature — the comparison, not the raw value, is what carries the physics.

Common trapA small energy gap does not make a semiconductor a "weak conductor" in the metallic sense

It is tempting to picture a semiconductor as simply a "not very good" metal — a conductor with slightly fewer free electrons. This is physically wrong and will cost marks. A metal's free electrons exist by default, at any temperature, because there is no gap standing between the valence and conduction bands; more heat only scatters those already-free electrons and makes conduction marginally worse. A semiconductor's free electrons, by contrast, exist only because thermal energy has actively lifted them across a genuine forbidden gap $E_g$ ; more heat creates additional carriers and makes conduction better. These are two different mechanisms producing opposite temperature trends, not two points on the same "how good a conductor is it" scale.

You should be able to…

Define valence band, conduction band, and energy gap $E_g$ in your own words, and state the relation $E_g = E_C - E_V$ .
State the approximate range of $E_g$ for conductors, insulators, and semiconductors, quoting silicon ( $\approx 1.1\ \text{eV}$ ), germanium ( $\approx 0.7\ \text{eV}$ ), and diamond ( $\approx 5.4\ \text{eV}$ ) as examples.
Explain, using the size of $E_g$ relative to thermal energy $k_BT \approx 0.026\ \text{eV}$ , why insulators do not conduct at room temperature while semiconductors do so weakly.
Explain why semiconductor conductivity increases with temperature while metallic conductivity decreases, and give the physical reason for each trend.
Distinguish the band picture of a metal (overlapping or partly filled bands) from that of an insulator or semiconductor (full valence band, gap, empty conduction band at $0\ \text{K}$ ).
State that a pure semiconductor behaves exactly like an insulator at $0\ \text{K}$ , and explain what changes as temperature rises.

Ask the tutor: Still fuzzy on why a "small" energy gap makes such a big practical difference between silicon and diamond — ask your AI tutor to walk through the exponential falloff of crossing electrons with $E_g/k_BT$ using a few more numeric comparisons.

Why does adding heat help a semiconductor but hurt a metal, if heat is just heat either way?

Heat does the same basic thing in both cases — it makes the crystal lattice vibrate more and gives electrons small random kicks of energy. In a metal, free electrons already exist regardless of temperature, so the extra lattice vibration only gets in their way, scattering them more often and slightly reducing conductivity. In a semiconductor, there are no free conduction electrons to begin with unless one is handed enough energy, at least $E_g$ , to jump the gap from the valence band. Extra heat creates brand-new mobile electrons faster than it can scatter the existing ones, so the net effect is a sharp rise in conductivity, not a fall.

If a semiconductor is basically an insulator at 0 K, why do we not just call it an insulator?

The difference is entirely about the size of the gap relative to ordinary, everyday temperatures. Both insulators and semiconductors have a full valence band, an empty conduction band, and a genuine gap at absolute zero — in that narrow sense they look identical. But an insulator's gap (several eV) stays effectively uncrossable even at room temperature, while a semiconductor's gap (roughly $0.2$ to $3\ \text{eV}$ ) is small enough that ordinary room-temperature thermal energy lifts a small but real population of electrons across it. So "semiconductor" describes a material whose behaviour genuinely changes character between very cold and room temperature, while an insulator stays an insulator throughout the temperatures you would ever encounter it at.

Why silicon and germanium specifically, and not some metal with a tiny bit of resistance?

Because the mechanism matters, not just the amount of current that flows. Silicon and germanium are chosen because their gap sits in exactly the useful window — small enough that thermal energy or, far more controllably, added impurity atoms can switch on real conduction, yet large enough that the material is not simply always-on like a metal. That controllability is the entire point: engineers can dial a semiconductor's conductivity up or down by design (as the next article on n-type and p-type doping shows), which is impossible with a metal whose free-electron count is essentially fixed by its crystal structure.

Does this energy-band idea come up again later in the chapter, or is it just background?

It is the foundation the rest of the chapter stands on, even though later articles rarely draw the band diagram explicitly. When you learn how doping creates n-type and p-type semiconductors, or how a p-n junction forms a depletion region, the underlying reason those tricks work at all is that silicon's gap is small enough to manipulate. Examiners do sometimes ask you to justify, in words, why silicon rather than diamond or copper is used in a diode — and the honest answer is exactly this energy-gap comparison, so it is worth being able to state fluently rather than just memorising the eV values.