Sources and References

Primary textbook — S.O. Kasap, Principles of Electronic Materials and Devices, 4th ed., McGraw-Hill, 2017.

Online resources:

• MIT OCW 3.091SC Introduction to Solid State Chemistry (Fall 2010) — Prof. Donald Sadoway; lecture videos and summary notes covering bonding, band theory, and crystal structure.
• MIT OCW 3.024 Electronic, Optical and Magnetic Properties of Materials (Spring 2013) — full lecture notes on quantum mechanics, band structure, and optical/dielectric properties.
• M.S. Dresselhaus, Solid State Physics Part II: Optical Properties of Solids, MIT 6.732 (2001) — open PDF covering optical constants, dielectric functions, and band-to-band transitions.
• Engineering LibreTexts — Electronic Properties of Materials — modular open textbook covering density of states, Fermi level, and semiconductor band theory.
• Engineering LibreTexts — Introduction to Semiconductors (TLP Library II) — Fermi–Dirac distribution, carrier statistics, and energy bands.
• Wikibooks: Electronic Properties of Materials / Quantum Mechanics for Engineers — particle-in-a-box derivation and extensions.

---

Chapter 1: Quantum Mechanics Foundations

The whole edifice of modern electronic materials rests on quantum mechanics. Classical physics predicted that electrons orbiting a nucleus would spiral inward, radiating away their energy continuously — clearly at odds with the stable atoms we observe. The resolution required abandoning the idea that energy is exchanged continuously and accepting instead that it comes in discrete packets, or quanta. This chapter builds the quantum mechanical toolkit used throughout the course.

1.1 The Photon and Wave–Particle Duality of Light

1.1.1 Blackbody Radiation and Planck’s Hypothesis

By the late nineteenth century, classical thermodynamics predicted that a hot object should radiate an infinite amount of energy at short wavelengths — the “ultraviolet catastrophe.” Max Planck resolved this in 1900 by postulating that oscillators in the cavity wall can only exchange energy with the electromagnetic field in integer multiples of a fundamental quantum:

\[ E = h\nu \]

where \(h = 6.626 \times 10^{-34}\) J·s is Planck’s constant and \(\nu\) is the frequency of radiation. This was the first crack in the foundation of classical physics.

1.1.2 The Photoelectric Effect and Einstein’s Photon

In 1905 Einstein extended Planck’s idea: light itself consists of discrete particles — photons — each carrying energy \(E = h\nu\). When a photon strikes a metal surface, it can eject an electron only if its energy exceeds the metal’s work function \(\Phi\):

\[ KE_{\max} = h\nu - \Phi \]

The kinetic energy of the ejected electron depends on frequency, not intensity. Intensity only controls how many photons arrive per second, which governs the number of electrons emitted, not their individual energies. This result is impossible to explain with classical wave optics and was direct evidence for the quantization of light.

The momentum of a photon follows from Einstein’s relativity combined with Planck’s relation:

\[ p = \frac{h\nu}{c} = \frac{h}{\lambda} \]

where \(\lambda\) is the wavelength and \(c\) is the speed of light in vacuum.

1.2 The Electron as a Wave: de Broglie’s Hypothesis

In 1924, Louis de Broglie made the audacious proposal that matter, like light, has a dual wave–particle nature. Any particle with momentum \(p\) has an associated de Broglie wavelength:

\[ \lambda = \frac{h}{p} = \frac{h}{mv} \]

For a macroscopic object — say, a baseball — this wavelength is so astronomically small as to be undetectable. For an electron with kinetic energy of a few electron-volts, however, \(\lambda\) is on the order of an ångström (0.1 nm), which is comparable to atomic spacings in crystals. This is why electrons can diffract off crystal planes, exactly as X-rays do — a phenomenon exploited in electron diffraction and confirmed experimentally by Davisson and Germer in 1927.

1.3 The Schrödinger Equation

To describe the behaviour of a quantum particle, we need an equation that governs how the associated wave evolves in space and time. Erwin Schrödinger provided this in 1926. The time-independent Schrödinger equation, appropriate for particles in a static potential \(V(x)\), is:

\[ -\frac{h^2}{8\pi^2 m} \frac{d^2\psi}{dx^2} + V(x)\psi = E\psi \]

where \(\psi(x)\) is the wavefunction, \(m\) is the particle mass, and \(E\) is the total energy. It is more compactly written using \(\hbar = h/(2\pi)\):

\[ -\frac{\hbar^2}{2m} \frac{d^2\psi}{dx^2} + V(x)\psi = E\psi \]

The wavefunction \(\psi\) itself has no direct physical meaning, but \(|\psi(x)|^2\) is the probability density — the probability per unit length of finding the particle near position \(x\). The wavefunction must be continuous, its derivative must be continuous (wherever \(V\) is finite), and it must be normalizable:

\[ \int_{-\infty}^{\infty} |\psi(x)|^2 \, dx = 1 \]

These boundary conditions are what force energy quantization.

1.4 The Infinite Potential Well

1.4.1 Setup and Solution

The infinite potential well (or “particle in a box”) is the simplest non-trivial quantum system and yet captures the essential physics of quantum confinement. Consider a particle of mass \(m\) trapped in a one-dimensional box of width \(L\), with infinitely hard walls: \(V = 0\) inside (\(0 < x < L\)) and \(V = \infty\) outside.

Outside the box \(\psi = 0\). Inside, the Schrödinger equation reduces to:

\[ \frac{d^2\psi}{dx^2} = -k^2 \psi, \qquad k^2 = \frac{2mE}{\hbar^2} \]

The general solution is \(\psi = A\sin(kx) + B\cos(kx)\). Applying the boundary condition \(\psi(0) = 0\) forces \(B = 0\). Applying \(\psi(L) = 0\) then requires \(kL = n\pi\) for positive integers \(n = 1, 2, 3, \ldots\)

This quantization condition gives the allowed wavefunctions:

\[ \psi_n(x) = \sqrt{\frac{2}{L}} \sin\!\left(\frac{n\pi x}{L}\right) \]

and the allowed energy levels:

\[ E_n = \frac{n^2 \pi^2 \hbar^2}{2mL^2} = \frac{n^2 h^2}{8mL^2} \]

1.4.2 Physical Consequences

Energy levels scale as \(n^2\) and as \(1/L^2\). Shrinking the box dramatically increases the level spacing. This is why electrons in semiconductor quantum wells (nanometre-scale layers) have widely spaced energy levels that can be engineered by controlling the layer thickness — the basis of quantum well lasers and photodetectors.

1.5 The Heisenberg Uncertainty Principle

Quantum mechanics places fundamental limits on how precisely two complementary observables can be known simultaneously. The most important pair is position and momentum:

\[ \Delta x \cdot \Delta p_x \geq \frac{\hbar}{2} \]

A similar relation holds for energy and time: \(\Delta E \cdot \Delta t \geq \hbar/2\). These are not limitations of measurement apparatus — they reflect a fundamental feature of nature. A particle confined to a region of size \(\Delta x\) must have a spread in momentum of at least \(\hbar/(2\Delta x)\), and therefore a minimum kinetic energy. This is precisely the zero-point energy of the infinite well.

1.6 The Finite Potential Well and Quantum Tunneling

1.6.1 Finite Potential Well

When the walls of the potential well have finite height \(V_0\), the wavefunction does not abruptly vanish at the boundaries. Instead, it exponentially decays into the classically forbidden region:

\[ \psi \propto e^{-\kappa x}, \qquad \kappa = \sqrt{\frac{2m(V_0 - E)}{\hbar^2}} \]

This “evanescent” penetration into the barrier means that the allowed energies are slightly lower than in the infinite-well case (the effective box is a little wider), and there are only a finite number of bound states. For an electron in a semiconductor quantum well of finite depth, these are the real energy levels observed experimentally.

1.6.2 Quantum Tunneling

When a particle encounters a potential barrier of height \(V_0 > E\) and finite width \(d\), classical mechanics says it must be reflected. Quantum mechanics predicts a nonzero transmission probability:

\[ T \approx e^{-2\kappa d}, \qquad \kappa = \sqrt{\frac{2m(V_0-E)}{\hbar^2}} \]

Tunneling is exponentially sensitive to barrier width and height. It is not a curiosity — it underpins the operation of tunnel diodes, scanning tunnelling microscopes (STM), alpha-particle radioactive decay, and the gate oxide leakage that limits transistor scaling. In STM, the tunneling current varies by roughly an order of magnitude for every 0.1 nm change in tip-surface separation, giving atomic-scale resolution.

1.7 The 3D Potential Box and Three Quantum Numbers

Extending the particle-in-a-box to three dimensions — a rectangular box with sides \(L_x, L_y, L_z\) — requires three independent quantum numbers \(n_x, n_y, n_z\):

\[ E_{n_x n_y n_z} = \frac{h^2}{8m}\!\left(\frac{n_x^2}{L_x^2} + \frac{n_y^2}{L_y^2} + \frac{n_z^2}{L_z^2}\right) \]

For a cubic box (\(L_x = L_y = L_z = L\)), distinct combinations of quantum numbers can give the same total energy — a phenomenon called degeneracy. This degeneracy will reappear when we count electronic states in metals and semiconductors.

1.8 The Hydrogen Atom

1.8.1 Quantum Numbers

The hydrogen atom — a single electron in the Coulomb potential of a proton — is the paradigmatic quantum system with an exact analytical solution. In spherical coordinates, separation of variables yields three quantum numbers:

• Principal quantum number \(n = 1, 2, 3, \ldots\) — sets the energy: \(E_n = -13.6\text{ eV}/n^2\)
• Angular momentum quantum number \(\ell = 0, 1, \ldots, n-1\) — sets the magnitude of orbital angular momentum: \(L = \hbar\sqrt{\ell(\ell+1)}\)
• Magnetic quantum number \(m_\ell = -\ell, \ldots, +\ell\) — sets the z-component of angular momentum

A fourth quantum number, the spin quantum number \(m_s = \pm 1/2\), arises from relativistic quantum mechanics (Dirac equation) and gives the electron its intrinsic angular momentum.

1.8.2 Orbitals and Energy Levels

The \(\ell = 0\) states are called \(s\)-orbitals (spherically symmetric), \(\ell = 1\) are \(p\)-orbitals (dumbbell-shaped), \(\ell = 2\) are \(d\)-orbitals, and \(\ell = 3\) are \(f\)-orbitals. Each orbital can hold 2 electrons (one for each spin state). The notation \(2p\) means \(n = 2, \ell = 1\); there are three such states (\(m_\ell = -1, 0, +1\)), so the \(2p\) subshell can hold 6 electrons.

1.9 Multi-Electron Atoms and the Periodic Table

For atoms with more than one electron, electron–electron repulsion prevents an exact solution, but the hydrogenlike orbital picture remains a good starting point. Two principles govern how electrons fill orbitals:

Pauli Exclusion Principle: No two electrons in the same atom can have the same set of four quantum numbers \((n, \ell, m_\ell, m_s)\). Equivalently, each orbital holds at most two electrons, with opposite spins.

Hund’s Rule: When filling orbitals of equal energy (degenerate subshell), electrons occupy separate orbitals with parallel spins before pairing up. This minimises electron–electron repulsion.

These rules, combined with the aufbau (building-up) principle, generate the electronic configurations that explain the structure of the periodic table and the chemical properties of elements — including why carbon’s four valence electrons make it such a versatile bonding partner in both organic molecules and semiconductor devices.

---

Chapter 2: Bonding, Crystallography, and Defects

Understanding how atoms join together and arrange themselves in space is the foundation for all material properties. The same silicon atom that forms a covalent network in a semiconductor can also appear in amorphous glass; the difference is purely in atomic arrangement, and that difference controls all electrical, optical, and mechanical properties.

2.1 Atomic Structure and Atomic Mass

The mass of an atom is dominated by its nucleus (protons + neutrons), while its volume and chemistry are dominated by its electron cloud. The atomic mass unit (amu or u) is defined as one-twelfth the mass of a carbon-12 atom: 1 u = \(1.66054 \times 10^{-27}\) kg. The molar mass in g/mol numerically equals the atomic mass in amu, and one mole contains \(N_A = 6.022 \times 10^{23}\) atoms (Avogadro’s number).

2.2 Types of Chemical Bonding

2.2.1 Covalent Bonding

In covalent bonds, neighbouring atoms share electrons, with the shared electron density concentrated between the nuclei. The bond is directional and strong — bond energies typically range from 1 to 7 eV. Silicon, germanium, and diamond all form tetrahedral sp³-hybridized covalent networks: each atom bonds to four neighbours in a structure that repeats throughout the crystal. The directionality of covalent bonds is why these materials have fixed crystal structures and why their band gaps arise from discrete bonding/anti-bonding energy splitting (see Chapter 3).

2.2.2 Metallic Bonding

In metals, valence electrons are delocalized — they belong to the crystal as a whole rather than to individual bonds. This electron sea model explains both high electrical conductivity (electrons are free to move under an applied field) and high thermal conductivity, as well as the characteristic ductility of metals (the electron sea accommodates slip between planes). Bond energies per atom are moderate (1–5 eV).

2.2.3 Ionic Bonding

Ionic bonds form between electropositive and electronegative species through electron transfer. The resulting Coulomb attraction between oppositely charged ions is strong and non-directional, giving high melting points and brittle mechanical behaviour. The Madelung constant \(\mathcal{M}\) characterizes the geometry-dependent sum of Coulomb interactions in a particular crystal structure:

\[ E_{\text{Coulomb}} = -\mathcal{M} \frac{e^2}{4\pi\epsilon_0 r} \]

where \(r\) is the nearest-neighbour distance. Ionic solids are typically electrical insulators at low temperature (ions are not mobile and electrons are localized), though they can conduct via ion migration at elevated temperatures.

2.2.4 Secondary (van der Waals) Bonding

Even charge-neutral atoms and molecules attract each other through induced-dipole interactions arising from quantum fluctuations in the electron distribution. These London dispersion forces are much weaker than primary bonds (0.01–0.1 eV) and fall off rapidly with distance (\(\sim r^{-6}\)). They are nonetheless critical in layered materials like graphite and MoS₂, where covalent bonding within layers and van der Waals bonding between layers give these materials their distinctive anisotropic properties.

Hydrogen bonds — a special secondary bond involving a hydrogen atom between two electronegative atoms — are stronger (0.1–0.5 eV) and highly directional; they govern the structure of ice and many biological molecules.

2.2.5 Mixed Bonding

Most real materials exhibit mixed bonding character. III-V semiconductors (GaAs, InP) have bonds intermediate between covalent and ionic; the ionicity parameter \(f_i\) quantifies this mixture. SiO₂ (silica) bonds are roughly 50% ionic and 50% covalent. The bonding character strongly influences the band gap, dielectric constant, and optical properties.

2.3 The Crystalline State

2.3.1 Crystal Lattices and Unit Cells

A crystal is a solid in which atoms are arranged in a pattern that repeats periodically in three dimensions. The repeating unit — the unit cell — is described by three lattice vectors \(\mathbf{a}, \mathbf{b}, \mathbf{c}\) and the angles between them. The fourteen Bravais lattices enumerate all distinct ways of filling space periodically in three dimensions. For electronic materials, the most important are:

• Simple cubic (SC) — rarely occurs in elemental form (Po is the only example)
• Face-centred cubic (FCC) — copper, aluminium, gold, nickel; also the structure of the diamond cubic lattice (which is two interlocked FCC lattices)
• Body-centred cubic (BCC) — iron, tungsten, chromium
• Hexagonal close-packed (HCP) — titanium, zinc, magnesium

Silicon and germanium adopt the diamond cubic structure: an FCC lattice with a two-atom basis, giving each atom four tetrahedral nearest neighbours consistent with sp³ hybridization.

2.3.2 Crystal Directions and Planes (Miller Indices)

Directions in a crystal are specified by the vector \([uvw]\) — the smallest integer set proportional to the direction components in the lattice basis. Equivalent directions form a family denoted \(\langle uvw \rangle\).

Crystal planes are specified by Miller indices \((hkl)\), found by:

1. Taking the reciprocals of the fractional intercepts the plane makes with the crystallographic axes.
2. Reducing to the smallest integers.
3. Enclosing in parentheses.

Negative intercepts are written with a bar over the index: \((\bar{1}10)\). The family of equivalent planes is denoted \(\{hkl\}.\) In cubic systems, the direction \([hkl]\) is perpendicular to the plane \((hkl)\) — a convenient property that does not hold in lower-symmetry systems.

2.3.3 Bragg Diffraction and X-ray Diffraction

When X-rays (or electrons or neutrons) are incident on a crystal, they scatter coherently from successive planes of atoms. Constructive interference occurs only when the Bragg condition is satisfied:

\[ 2d_{hkl}\sin\theta = n\lambda \]

where \(d_{hkl}\) is the interplanar spacing for the \((hkl)\) family, \(\theta\) is the glancing angle (complement of the angle of incidence), and \(\lambda\) is the wavelength. X-ray diffraction (XRD) is the primary tool for crystal structure determination: the pattern of diffraction peaks uniquely identifies the crystal structure, lattice parameter, and can reveal strain, texture, and phase composition.

The interplanar spacing for a cubic crystal with lattice parameter \(a\) is:

\[ d_{hkl} = \frac{a}{\sqrt{h^2 + k^2 + l^2}} \]

2.4 Crystalline Defects

2.4.1 Point Defects

No real crystal is perfect. Point defects are zero-dimensional imperfections:

• Vacancy: a missing atom. Vacancy concentration follows Arrhenius behaviour: \(n_v/N = \exp(-Q_v/k_BT)\), where \(Q_v\) is the vacancy formation energy. At room temperature, vacancy concentrations are tiny, but at processing temperatures (800–1200°C for silicon), they become significant.
• Interstitial: an extra atom squeezed into a non-lattice site. Common in radiation-damaged materials.
• Substitutional impurity: a foreign atom occupying a lattice site. This is the mechanism of deliberate doping in semiconductors — substituting a phosphorus atom (5 valence electrons) for silicon (4 valence electrons) introduces a loosely bound extra electron that can freely conduct.
• Frenkel defect: a vacancy–interstitial pair. Common in ionic crystals.
• Schottky defect: paired cation–anion vacancies maintaining charge neutrality in an ionic crystal.

2.4.2 Line Defects: Dislocations

Dislocations are one-dimensional defects characterized by the Burgers vector \(\mathbf{b}\), which quantifies the magnitude and direction of lattice distortion. The two fundamental types are:

• Edge dislocation: an extra half-plane of atoms terminated inside the crystal. The Burgers vector is perpendicular to the dislocation line.
• Screw dislocation: a helical distortion; the Burgers vector is parallel to the dislocation line.

Dislocations govern plastic deformation (they are why metals yield at stresses far below the theoretical shear strength) and can also act as traps for charge carriers, reducing carrier lifetime in semiconductors.

2.4.3 Planar and Surface Defects

Grain boundaries separate regions of different crystallographic orientation in polycrystalline materials. They scatter electrons and phonons, raising electrical resistivity and lowering thermal conductivity. Stacking faults are planar errors in the stacking sequence (e.g., ABCABABCABC instead of the perfect FCC sequence ABCABCABC). Surfaces are the ultimate planar defect — they break the periodicity, create dangling bonds, and often reconstruct into new surface structures to minimize energy.

2.4.4 Stoichiometry Defects

In compound semiconductors and oxides, the composition can deviate from exact stoichiometry, creating an excess of one species. In ZnO, for example, an excess of zinc atoms (or equivalently, oxygen vacancies) makes the material intrinsically n-type. These non-stoichiometry defects can dramatically alter electrical properties.

2.5 Glasses and Amorphous Semiconductors

Silica glass (SiO₂) has the same short-range order as crystalline quartz — each silicon atom bonds tetrahedrally to four oxygens — but lacks the long-range periodic order of a crystal. The random network of SiO₄ tetrahedra gives glass its isotropic optical properties, its absence of a sharp melting point (it softens over a range of temperatures), and its characteristic conchoidal fracture.

Amorphous silicon (a-Si) is of enormous practical importance as the absorber layer in thin-film solar cells and TFT-LCD backplane transistors. Its disordered structure creates a distribution of localized “tail states” at the band edges rather than sharp band edges, reducing carrier mobility compared to crystalline silicon but enabling large-area deposition on glass substrates at low temperatures.

---

Chapter 3: Band Theory and Electronic Properties

The quantum mechanical treatment of electrons in a crystal reveals why some materials conduct electricity, some insulate, and some are semiconductors. This is perhaps the most powerful result of quantum mechanics applied to solids.

3.1 From Hydrogen Molecule to Molecular Orbital Theory

When two hydrogen atoms approach each other, their atomic 1s wavefunctions overlap and combine into two molecular orbitals. The bonding orbital (symmetric combination) has lower energy than the isolated atoms and corresponds to electron density concentrated between the nuclei. The antibonding orbital (antisymmetric combination, denoted \(\sigma^*\)) has a node between the nuclei and higher energy.

This picture — linear combination of atomic orbitals (LCAO) — generalizes to any number of atoms. When \(N\) identical atoms form a crystal, each atomic energy level splits into a band of \(N\) closely spaced levels. Because \(N \sim 10^{23}\), the levels are effectively continuous: they form a quasi-continuous energy band.

3.2 Band Formation and Electron Properties

As the interatomic spacing decreases from infinity to the equilibrium lattice constant, the atomic energy levels broaden into bands. The width of the resulting band depends on how strongly the wavefunctions overlap — valence band states with large orbital overlap (e.g., 2p orbitals pointing along the bond direction) form wide bands; more tightly bound core states with little overlap remain as narrow bands close to their atomic energies.

The resulting electronic band structure — energy \(E\) as a function of wavevector \(\mathbf{k}\) — determines all electronic properties. Each \(\mathbf{k}\)-state in a band can hold 2 electrons (spin up + spin down). The number of \(\mathbf{k}\)-states per unit volume per unit energy is the density of states \(g(E)\).

3.3 Semiconductors: Band Gap and Carrier Types

Silicon has \(E_g = 1.12\) eV; germanium has \(E_g = 0.67\) eV; GaAs has \(E_g = 1.42\) eV. Insulators have \(E_g \gtrsim 5\) eV (diamond: 5.5 eV; SiO₂: ~9 eV).

Intrinsic carrier concentration: Both thermally generated electrons and holes conduct. Their concentration is:

\[ n_i = \sqrt{N_c N_v}\, \exp\!\left(-\frac{E_g}{2k_B T}\right) \]

where \(N_c\) and \(N_v\) are the effective densities of states at the CB and VB edges. For silicon at 300 K, \(n_i \approx 1.5 \times 10^{10}\) cm\(^{-3}\) — tiny compared to the \(\sim 5 \times 10^{22}\) cm\(^{-3}\) atoms in the lattice.

3.4 Effective Mass

In a crystal, an electron in a band does not move as a free particle with mass \(m_e\). The periodic potential of the lattice modifies its response to applied forces. Near the bottom of a band (where the dispersion \(E(k)\) is parabolic), the electron behaves like a free particle with an effective mass:

\[ m^* = \frac{\hbar^2}{\left(\frac{d^2E}{dk^2}\right)} \]

The effective mass can be much smaller than \(m_e\) (e.g., in GaAs, \(m^*_e \approx 0.067\, m_e\)) or can even be negative near the top of a band — leading to the concept of holes. A hole is the absence of an electron near the top of a nearly full valence band, and it behaves as a positively charged particle with positive effective mass \(m_h^*\).

3.5 Density of States

The number of allowed quantum states per unit volume per unit energy interval is the density of states \(g(E)\). For a three-dimensional free-electron gas (a model valid near band edges):

\[ g(E) = \frac{4\pi (2m^*)^{3/2}}{h^3} \sqrt{E - E_c} \]

for the conduction band (taking \(E_c\) as the band edge reference). This square-root dependence on energy reflects the geometry of counting states in \(k\)-space: the volume of a spherical shell in \(k\)-space scales as \(k^2\, dk \propto \sqrt{E}\, dE\).

In lower-dimensional systems (quantum wells, wires, dots), the density of states changes shape dramatically — becoming step-like in 2D and spike-like (delta function) in 0D. This engineering of the DOS is the basis for high-performance semiconductor lasers.

3.6 Boltzmann and Fermi–Dirac Statistics

3.6.1 The Fermi–Dirac Distribution

Electrons are fermions — they obey the Pauli exclusion principle — so their occupation of energy states follows the Fermi–Dirac distribution:

\[ f(E) = \frac{1}{\exp\!\left(\dfrac{E - E_F}{k_B T}\right) + 1} \]

At \(T = 0\), \(f(E) = 1\) for \(E < E_F\) and \(f(E) = 0\) for \(E > E_F\) — a sharp step. At finite temperature, the step broadens over an energy range of order \(k_B T\) around the Fermi level \(E_F\). In a metal, \(E_F\) lies inside a band, and the electrons near \(E_F\) are the ones that respond to applied electric fields (those deep in the filled states cancel out). In an intrinsic semiconductor, \(E_F\) lies near mid-gap.

3.6.2 Boltzmann Approximation

When \(E - E_F \gg k_B T\) (i.e., in a non-degenerate semiconductor where the Fermi level is well inside the gap), the \(+1\) in the denominator is negligible and the Fermi–Dirac distribution reduces to the simpler Boltzmann distribution:

\[ f(E) \approx \exp\!\left(-\frac{E - E_F}{k_B T}\right) \]

This approximation, valid for most semiconductor doping levels at room temperature, greatly simplifies carrier concentration calculations.

3.7 The Free Electron Model of Metals

In 1928, Arnold Sommerfeld treated the conduction electrons in a metal as a quantum gas of free fermions — the Sommerfeld free electron model. Electrons are confined in a cubic box of side \(L\), and the allowed \(k\)-states are filled from the bottom up according to the Pauli principle. The resulting Fermi sphere in \(k\)-space has radius \(k_F\) determined by the electron density \(n\):

\[ k_F = \left(3\pi^2 n\right)^{1/3} \]

The energy at the Fermi surface — the Fermi energy — is:

\[ E_F = \frac{\hbar^2 k_F^2}{2m_e} = \frac{\hbar^2}{2m_e}\left(3\pi^2 n\right)^{2/3} \]

For copper (\(n \approx 8.5 \times 10^{28}\) m\(^{-3}\)), \(E_F \approx 7\) eV — corresponding to a Fermi velocity \(v_F \approx 1.6 \times 10^6\) m/s, much larger than any classical thermal velocity. This explains why the electronic heat capacity of metals is far smaller than the classical prediction (only electrons within \(\sim k_B T\) of \(E_F\) can be thermally excited).

3.8 Electrical Conductivity and the Drude–Sommerfeld Picture

The dc electrical conductivity of a metal is:

\[ \sigma = \frac{ne^2\tau}{m^*} \]

where \(n\) is the free carrier density, \(e\) the electron charge, \(\tau\) the mean scattering time (mean free time between collisions), and \(m^*\) the effective mass. In this picture, electrons are scattered by:

• Phonons (lattice vibrations): scattering rate increases with temperature, giving the characteristic \(\rho \propto T\) behaviour of pure metals at room temperature.
• Impurities and defects: temperature-independent contribution, dominant at low temperature (Matthiessen’s rule: \(\rho = \rho_{\text{phonon}}(T) + \rho_{\text{defect}}\)).

The mean free path \(\ell = v_F \tau\) in a pure metal at room temperature is typically 10–50 nm. As device features shrink below this length, classical conductivity assumptions break down.

3.9 Metal Contacts and the Seebeck Effect

3.9.1 Contact Potential

When two dissimilar metals are joined, electrons flow from the metal with lower work function to the one with higher work function until the Fermi levels equalize. This creates a built-in contact potential at the interface:

\[ V_{\text{contact}} = \frac{\Phi_2 - \Phi_1}{e} \]

where \(\Phi_1, \Phi_2\) are the work functions. The net contact potential around a closed circuit of dissimilar metals is zero at uniform temperature — no perpetual-motion machine is possible.

3.9.2 Seebeck Effect and Thermocouples

If a temperature gradient exists across a metal, electrons on the hot side have higher average energy and diffuse toward the cold side, building up a charge imbalance and an electric potential — the Seebeck effect. The open-circuit voltage per unit temperature difference is the Seebeck coefficient (thermopower) \(S\):

\[ V_{\text{Seebeck}} = S \cdot \Delta T \]

For metals, \(S\) is typically a few \(\mu\)V/K. For semiconductors it can be hundreds of \(\mu\)V/K. A junction of two materials with different \(S\) — a thermocouple — produces a measurable voltage proportional to temperature difference and is used ubiquitously for temperature measurement and, in arrays, for thermoelectric power generation.

3.10 Band Theory of Metals

In the free electron model the periodic potential of the lattice is ignored. In the real nearly-free electron model, the periodic potential opens band gaps at the Brillouin zone boundaries (where Bragg reflection causes standing waves). A metal is any material where the Fermi level lies within a band — partial filling means electrons near \(E_F\) can easily be excited and conduct. Monovalent metals (Na, Cu, Ag, Au) have roughly spherical Fermi surfaces; divalent and higher metals have more complex Fermi surface topologies that produce the distinctive optical and transport properties of each metal.

---

Chapter 4: Optical Properties of Materials

Light–matter interactions are governed by how the electromagnetic field couples to the electrons and ions in a material. Understanding this coupling allows us to design optical fibres, anti-reflection coatings, photodetectors, and laser gain media.

4.1 Light Waves in a Homogeneous Medium

Maxwell’s equations in a linear, homogeneous, isotropic dielectric with no free charges give a wave equation for the electric field \(\mathbf{E}\):

\[ \nabla^2 \mathbf{E} = \mu_0 \epsilon_0 \epsilon_r \frac{\partial^2 \mathbf{E}}{\partial t^2} \]

The wave propagates at speed \(v = c/n\), where \(c\) is the speed of light in vacuum and \(n = \sqrt{\epsilon_r}\) (for non-magnetic materials) is the refractive index. In an absorbing medium, the dielectric constant becomes complex:

\[ \tilde{\epsilon} = \epsilon_1 + i\epsilon_2 \]

and so does the refractive index: \(\tilde{n} = n + i\kappa\), where \(\kappa\) is the extinction coefficient. A plane wave propagating in the \(z\)-direction:

\[ E(z,t) = E_0 \exp\!\left(i\left(\frac{2\pi \tilde{n} z}{\lambda} - \omega t\right)\right) = E_0 e^{-\frac{2\pi\kappa z}{\lambda}} \exp\!\left(i\left(\frac{2\pi n z}{\lambda} - \omega t\right)\right) \]

The imaginary part causes exponential attenuation; the amplitude decays by factor \(1/e\) in the skin depth \(\delta = \lambda/(2\pi\kappa)\). The optical intensity decays with the absorption coefficient \(\alpha = 4\pi\kappa/\lambda\) following Beer–Lambert:

\[ I(z) = I_0 e^{-\alpha z} \]

4.2 Refractive Index and Dispersion

4.2.1 Refractive Index

The refractive index of a medium is determined by how polarizable its constituent atoms or electrons are at the frequency of the light. For frequencies far from any resonance of the material:

\[ n^2 - 1 = \frac{Ne^2}{\epsilon_0 m_e} \sum_j \frac{f_j}{\omega_j^2 - \omega^2} \]

where \(N\) is the density of oscillators, \(f_j\) is the oscillator strength, and \(\omega_j\) is the resonance frequency. This Lorentz oscillator model explains why the refractive index is frequency-dependent (dispersion).

4.2.2 Dispersion and the Sellmeier Equation

For transparent optical materials (where absorption is negligible), an empirical Sellmeier equation fits the dispersion accurately:

\[ n^2(\lambda) = 1 + \sum_i \frac{B_i \lambda^2}{\lambda^2 - C_i} \]

The Sellmeier coefficients \(B_i, C_i\) are tabulated for common optical glasses and crystals. Dispersion is responsible for chromatic aberration in lenses, pulse broadening in optical fibres, and the separation of white light by a prism.

4.3 Group Velocity and Group Index

A short pulse of light is a superposition of many frequencies. The phase velocity \(v_p = c/n\) describes the speed of the carrier wave, while the group velocity describes the speed of the envelope (i.e., the energy transport):

\[ v_g = \frac{d\omega}{dk} = \frac{c}{n + \omega\,(dn/d\omega)} = \frac{c}{n_g} \]

where \(n_g = n - \lambda\,(dn/d\lambda)\) is the group index. In most optical glasses \(n_g > n\), so pulses travel slower than the phase velocity. Near a strong absorption resonance, \(dn/d\omega\) can be large and anomalous dispersion (\(v_g > c\)) can appear — though the signal velocity never exceeds \(c\).

4.4 Fresnel Equations and Reflectance

At an interface between two media with refractive indices \(n_1\) and \(n_2\), the fraction of power reflected (for normal incidence) is:

\[ R = \left(\frac{n_1 - n_2}{n_1 + n_2}\right)^2 \]

For an air–glass interface (\(n_1 = 1, n_2 = 1.5\)), \(R \approx 4\%\). Anti-reflection coatings exploit destructive interference between reflections from the front and back surfaces of a thin film to reduce \(R\) to near zero at a target wavelength — essential for camera lenses, solar cells, and laser facets.

The general Fresnel equations for oblique incidence distinguish between s-polarized (transverse electric, TE) and p-polarized (transverse magnetic, TM) components. At Brewster’s angle \(\theta_B = \arctan(n_2/n_1)\), the p-polarized reflected amplitude vanishes completely, leaving only the s-component. This is the principle behind polarising sunglasses and Brewster windows in laser cavities.

4.5 Lattice Absorption

Ionic crystals such as NaCl and GaAs absorb infrared radiation strongly in the Reststrahlen band — a frequency range near the transverse optical (TO) phonon frequency. When light drives the relative motion of cation and anion sublattices resonantly, energy is absorbed and transferred to the lattice. This sets an infrared transparency cutoff: below the Reststrahlen wavelength (long wavelength side) and above it (short wavelength, visible) the crystal is transparent; within the band it is highly reflecting and absorbing.

4.6 Band-to-Band Absorption

For photon energies exceeding the band gap (\(h\nu > E_g\)), a photon can excite an electron from the valence band to the conduction band, creating an electron–hole pair. The absorption coefficient rises sharply above \(E_g\):

• Direct gap semiconductors (GaAs, InP, GaN): the conduction band minimum and valence band maximum occur at the same \(\mathbf{k}\)-value. A vertical transition in \(k\)-space is allowed, giving a sharp absorption edge: \(\alpha \propto \sqrt{h\nu - E_g}\).
• Indirect gap semiconductors (Si, Ge): the transition requires a phonon to conserve momentum. The onset is more gradual and at lower energy than the direct gap, with \(\alpha \propto (h\nu - E_g)^2\) at low temperatures.

The sharp absorption edge enables optical determination of the band gap. It also means that GaAs is far better than silicon for LEDs and lasers (radiative recombination is momentum-allowed), while silicon’s indirect gap makes optical emission very inefficient.

4.7 Scattering

Light propagating through a medium can be scattered by fluctuations in refractive index caused by impurities, density fluctuations, or structural inhomogeneities. Rayleigh scattering from objects much smaller than \(\lambda\) gives a cross-section proportional to \(\lambda^{-4}\) — shorter wavelengths scatter much more strongly, explaining the blue colour of the sky. In optical fibres, Rayleigh scattering from frozen density fluctuations sets the fundamental transmission loss limit (approximately 0.2 dB/km at 1550 nm for silica fibre) and determines the optimal wavelength for long-distance fibre-optic communication.

---

Chapter 5: Dielectric Materials and Their Properties

Dielectrics are electrical insulators whose most important property is their response to an electric field — they polarize. This polarization stores energy (making them useful as capacitor dielectrics), shifts resonance frequencies (enabling frequency-selective filters), and can generate voltages under mechanical stress (piezoelectricity).

5.1 Polarization and Relative Permittivity

When an electric field \(\mathbf{E}\) is applied to a dielectric, positive and negative charges within each atom or molecule are displaced relative to each other, creating electric dipoles. The macroscopic polarization \(\mathbf{P}\) (dipole moment per unit volume) is related to the field by:

\[ \mathbf{P} = \epsilon_0 \chi_e \mathbf{E} \]

where \(\chi_e\) is the electric susceptibility. The relative permittivity (dielectric constant) is:

\[ \epsilon_r = 1 + \chi_e \]

The displacement field is \(\mathbf{D} = \epsilon_0 \epsilon_r \mathbf{E}\). A large \(\epsilon_r\) means the material can store more charge at a given voltage; it arises from large and easily polarizable charge distributions.

5.2 Electronic Polarization

Electronic polarization is the displacement of the electron cloud relative to the nucleus under an applied field. It occurs in all materials and responds extremely rapidly — up to optical frequencies (\(\sim 10^{15}\) Hz). The contribution to the dielectric constant is related to the refractive index by \(\epsilon_r(\text{optical}) = n^2\), which is typically 2–4 for common dielectrics.

The atomic polarizability \(\alpha_e\) relates the induced dipole moment \(\mathbf{p}\) to the local field: \(\mathbf{p} = \epsilon_0 \alpha_e \mathbf{E}_{\text{local}}\). The Clausius–Mossotti relation connects the microscopic polarizability to the macroscopic permittivity:

\[ \frac{\epsilon_r - 1}{\epsilon_r + 2} = \frac{N\alpha_e}{3} \]

where \(N\) is the number density of polarizable units.

5.3 Polarization Mechanisms

Several physical mechanisms contribute to \(\epsilon_r\), each active up to a characteristic frequency:

As the frequency increases past each mechanism’s cutoff, that mechanism can no longer follow the field and its contribution to \(\epsilon_r\) drops. A material that appears as an excellent capacitor dielectric at kHz may behave very differently at GHz.

5.4 Frequency Dependence: Dielectric Constant and Loss

In the presence of damping (collisions, viscosity of dipole rotation), the permittivity becomes complex:

\[ \tilde{\epsilon}_r(\omega) = \epsilon_r'(\omega) - i\epsilon_r''(\omega) \]

The imaginary part \(\epsilon_r''\) represents dielectric loss — energy dissipated as heat per cycle. The loss tangent is:

\[ \tan\delta = \frac{\epsilon_r''}{\epsilon_r'} \]

A low loss tangent is essential for microwave substrates (PCB materials, antenna radomes) and optical fibres. Near a resonance frequency, the Debye equations (for dipolar relaxation) or Lorentz equations (for resonance) describe how \(\epsilon_r'\) and \(\epsilon_r''\) vary, with characteristic dips and peaks linked by the Kramers–Kronig relations.

5.5 Gauss’s Law and Boundary Conditions

At an interface between two dielectrics, Maxwell’s boundary conditions require:

• Normal component of \(\mathbf{D}\) is continuous across the interface (in the absence of free surface charge): \(D_{1n} = D_{2n}\), i.e., \(\epsilon_{r1} E_{1n} = \epsilon_{r2} E_{2n}\)
• Tangential component of \(\mathbf{E}\) is continuous: \(E_{1t} = E_{2t}\)

These boundary conditions determine the electric field distribution in layered capacitor structures and explain why the tangential electric field component (parallel to the interface) is not enhanced at the interface, while the normal component is.

5.6 Capacitor Dielectrics

A parallel-plate capacitor with dielectric filling has capacitance:

\[ C = \frac{\epsilon_0 \epsilon_r A}{d} \]

where \(A\) is the plate area and \(d\) the plate separation. A high \(\epsilon_r\) allows either a smaller device for the same capacitance or a larger stored charge at the same voltage. Key dielectric materials:

• SiO₂ (\(\epsilon_r \approx 3.9\)): traditional gate oxide in MOSFETs; being replaced by high-\(\kappa\) dielectrics (HfO₂, \(\epsilon_r \approx 25\)) as gate thicknesses scale below 2 nm.
• BaTiO₃ and related ferroelectrics (\(\epsilon_r > 1000\)): used in multilayer ceramic capacitors (MLCCs).
• Al₂O₃ and Ta₂O₅: common in electrolytic and DRAM capacitors.

The breakdown field \(E_{\text{bd}}\) sets the maximum voltage the dielectric can sustain before avalanche ionization destroys the insulator. For SiO₂, \(E_{\text{bd}} \approx 10\) MV/cm. The figure of merit for energy storage is \(\epsilon_r E_{\text{bd}}^2\).

5.7 Piezoelectricity

In non-centrosymmetric crystals, applying a mechanical stress shifts the relative positions of positive and negative charge centres, producing a net polarization — the direct piezoelectric effect. Conversely, applying an electric field produces a mechanical strain — the converse piezoelectric effect. The constitutive relations are:

\[ P_i = d_{ijk} \sigma_{jk} \qquad \text{(direct)} \]\[ \varepsilon_{ij} = d_{kij} E_k \qquad \text{(converse)} \]

where \(d_{ijk}\) is the piezoelectric tensor. Of 32 crystal classes, 20 are non-centrosymmetric and potentially piezoelectric. Quartz (\(\alpha\)-SiO₂) and PZT (lead zirconate titanate) are the most commercially important piezoelectrics.

5.8 Quartz Oscillators and Filters

When a quartz crystal is cut at a specific orientation (e.g., AT-cut for temperature stability), its resonant frequency is determined almost entirely by its dimensions and the elastic constants of the crystal — quantities that are extremely stable with temperature, time, and aging. The equivalent circuit of a quartz resonator has an extremely high quality factor \(Q \sim 10^4\)–\(10^6\), far exceeding what is achievable with LC circuits.

This exceptional frequency stability makes quartz the frequency standard in virtually all clocks, watches, GPS receivers, and communication equipment. Piezoelectric BAW (bulk acoustic wave) and SAW (surface acoustic wave) resonators are used in the RF front-end filters of every mobile phone, selecting the desired band while rejecting interferers.

---

Chapter 6: Metallic Materials and Electronic Transport

The preceding chapters developed the quantum mechanical basis for metallic conduction. This chapter consolidates the key results and connects them to device-relevant scenarios including contacts, thermocouples, and resistivity in real materials.

6.1 Quantum Theory of Metals Revisited

The Sommerfeld model (Section 3.7) predicts the correct order of magnitude for the electronic heat capacity of metals and the correct qualitative behaviour of electrical conductivity. The key insight is that only electrons within \(\sim k_B T\) of the Fermi level participate in thermal and electrical response. For copper at 300 K, \(k_B T/E_F \approx 0.004\) — only 0.4% of conduction electrons are thermally active.

The total energy of the free electron gas at \(T = 0\) is \((3/5)N E_F\), giving an average energy per electron of \((3/5)E_F\) — far above the classical \((3/2)k_B T\). The pressure of the electron gas (\(P = (2/3)(U/V)\)) is called the Fermi pressure and is what prevents the collapse of metals under their own electrostatic forces.

6.2 Conduction in Metals: Scattering Mechanisms and Resistivity

The resistivity of a real metal is the sum of contributions from all scattering mechanisms (Matthiessen’s rule):

\[ \rho = \rho_{\text{thermal}}(T) + \rho_{\text{impurity}} + \rho_{\text{dislocation}} \]

• Thermal (phonon) scattering: at room temperature, \(\rho_{\text{thermal}} \propto T\). The phonon scattering rate is proportional to the number of phonons present, which in turn is proportional to temperature for \(T \gg \Theta_D\) (Debye temperature).
• Impurity scattering: each substitutional impurity of different valence than the host creates a local perturbation in potential. The contribution is temperature-independent (Lenz’s rule) and proportional to impurity concentration.
• Grain boundary and dislocation scattering: important in thin films and nanocrystalline metals, where grain sizes can be comparable to the mean free path.

6.3 Metal–Metal Contacts and Work Functions

The work function \(\Phi\) of a metal is the minimum energy required to remove an electron from the metal surface to vacuum: it is the energy difference between the vacuum level and the Fermi level. When two metals with work functions \(\Phi_1 \ne \Phi_2\) are brought into electrical contact, charge flows until \(E_F\) is equalized, establishing the contact potential \(({\Phi_2 - \Phi_1})/e\).

This built-in potential has no effect in an isolated loop at uniform temperature (Volta’s theorem: the total EMF around a closed circuit of conductors at the same temperature is zero). However, if the junctions are at different temperatures, a net EMF appears — the Seebeck effect that drives thermocouple operation.

Metal–semiconductor contacts are governed by similar physics, with the Schottky barrier height \(\phi_B = \Phi_m - \chi_s\) (where \(\chi_s\) is the semiconductor electron affinity) controlling rectifying or ohmic behaviour.

6.4 The Seebeck Effect in Depth

The Seebeck coefficient \(S\) (thermopower) for a metal within the Sommerfeld model is:

\[ S = -\frac{\pi^2 k_B^2 T}{3eE_F} \]

This is typically of order \(-10\) \(\mu\)V/K for metals — small but measurable and reproducible. Semiconductors can have thermopower of \(\pm 100\)–\(1000\) \(\mu\)V/K because the Fermi level is near the band edge where the density of states changes rapidly.

The thermoelectric figure of merit for energy harvesting applications is:

\[ ZT = \frac{S^2 \sigma T}{\kappa} \]

where \(\kappa\) is the thermal conductivity. High \(ZT\) requires high Seebeck coefficient, high electrical conductivity, and low thermal conductivity — properties that are difficult to optimize simultaneously because they are interrelated through the electronic structure. State-of-the-art thermoelectric materials (Bi₂Te₃ and related chalcogenides) achieve \(ZT \approx 1\)–\(2\) near room temperature, sufficient for cooling electronics (Peltier coolers) and waste heat recovery.

6.5 Band Theory of Metals: Fermi Surfaces and Optical Properties

The Fermi surface of a real metal — the constant-energy surface \(E(\mathbf{k}) = E_F\) in \(\mathbf{k}\)-space — is far more complex than a sphere when the periodic potential of the lattice is included. For noble metals (Cu, Ag, Au), the Fermi sphere intersects the Brillouin zone boundary, creating “necks” along the \(\langle 111 \rangle\) directions. These topological features produce characteristic structure in the optical reflectance spectra.

Colour of metals arises from interband transitions at optical frequencies. For most metals, free-electron (intraband) absorption makes them highly reflecting across the visible. In copper and gold, however, the \(d\)-band lies close to the Fermi level, and interband transitions from \(d\)-states to the Fermi surface occur at ~2 eV (copper) and ~2.5 eV (gold), absorbing blue and violet light and giving the characteristic reddish and yellow colours, respectively. Silver’s \(d\)-band is lower, so transitions occur in the UV, and silver reflects all visible frequencies nearly equally — appearing white or grey.

---