The Drude model is a classical theory of electrical and thermal transport in materials, introduced by Paul Drude in 1900. It treats conduction electrons as mobile particles that accelerate under applied fields and lose their organized motion through collisions. Despite its simplified assumptions, the model explains the basic form of electrical conduction, gives a useful account of the frequency-dependent response of metals, and provides a starting point for more advanced transport theories. Its central parameter is a relaxation time, which characterizes how quickly the average carrier momentum decays.
Physical assumptions
Drude adapted ideas from classical mechanics and the kinetic theory of an ideal gas to electrons in metals. In its simplest form, the model assumes that:
- A material contains a number density of mobile electrons, each with charge and mass , where is the elementary charge.
- Between collisions, electrons move according to classical equations of motion.
- Collisions randomize the electron motion, removing its average directed component.
- Collisions occur at a constant average rate , where is the relaxation time.
- The positive ionic background provides charge neutrality but does not itself carry the electronic current.
In the simplest stochastic interpretation, the probability of an electron experiencing a collision during a short interval is . The probability of remaining collision-free for a time is therefore .
The electrons have random motion even without an applied field. An applied electric field produces a small additional average velocity, called the drift velocity. Electrical current depends on this drift, not on the magnitude of the random velocities.
The original picture did not give a microscopic quantum account of collisions. Modern applications often retain the Drude equations while interpreting as a phenomenological momentum-relaxation time. Such a transport time need not equal the time between all microscopic scattering events, because small-angle scattering relaxes directed momentum less efficiently than large-angle scattering.
Equation of motion and direct-current conductivity
For a spatially uniform material without a magnetic field, the average drift velocity obeys
The first term on the right is the electric force; the second represents the average loss of directed momentum. This damping term describes an ensemble average rather than a continuous friction force acting on each electron between collisions.
The current density is
For a constant field, after transients have decayed,
so
This is the local form of Ohm’s law. The corresponding resistivity is
The model thus relates macroscopic electrical resistance to carrier density, carrier mass, and momentum relaxation. It does not, by itself, determine how varies with temperature, impurities, or crystal structure.
The positive magnitude of the carrier mobility is
giving . For electrons, the drift velocity points opposite to the electric field, while conventional current points along it.
When the field is removed, the drift velocity decays exponentially:
In a steady conducting state, the power dissipated per unit volume is
This energy transfer underlies Joule heating. The fixed-temperature Drude description implicitly assumes that energy delivered to the carriers can ultimately be transferred to their surroundings.
Magnetic fields and the Hall effect
In a magnetic field, the average equation of motion includes the Lorentz force:
The magnetic force deflects moving carriers sideways. In a sample carrying current along , with a magnetic field along , charge accumulates at the transverse boundaries until a Hall electric field develops.
For one type of electron carrier, with no transverse current,
The negative sign identifies negatively charged carriers under this convention. This result makes the Hall effect a way to estimate carrier density when a single-carrier description is valid.
However, some metals have positive Hall coefficients, which the original free-electron picture cannot explain. Modern transport theory accounts for such behavior through electronic band structure and, in suitable cases, positively charged holes. In multiband materials, the Hall coefficient depends on the densities and mobilities of different carriers and generally cannot be interpreted as simply .
The simplest isotropic, single-carrier Drude model with a constant relaxation time also predicts no longitudinal magnetoresistance when the transverse current is constrained to vanish. Observed magnetoresistance therefore often requires additional physics, such as multiple carrier populations or anisotropic electronic structure.
Frequency-dependent conductivity
For an oscillating field, write
Solving the equation of motion gives
The real and imaginary parts are
These signs correspond to the stated convention.
When , the response approaches the direct-current result, and current is nearly in phase with the field. When , carrier inertia becomes important: the dissipative conductivity falls, and the response becomes predominantly reactive.
The real conductivity has a peak centered at zero frequency, often called the Drude peak. Its characteristic width in angular frequency is . Measurements of this response in spectroscopy can constrain carrier scattering rates and the ratio of carrier density to mass.
Dielectric response and plasma frequency
The same model describes the free-carrier contribution to a material’s permittivity. With the same time convention, the relative permittivity is
where is the vacuum permittivity and represents a background contribution from other electronic processes over the frequency range of interest.
Defining the unscreened plasma frequency by
and the damping rate by , gives
This expression connects electrical transport to optical properties. In a weakly damped frequency range where the real part of the permittivity is negative, a bulk material generally does not support ordinary propagating transverse electromagnetic waves and can be strongly reflective.
If is approximately real and constant, the zero of the real permittivity occurs near in the weak-damping limit. This screened plasma edge need not coincide with the unscreened . Interband transitions and losses can substantially alter the observed optical response.
Practical optical descriptions often combine a Drude free-carrier term with bound-electron oscillator terms. These Drude–Lorentz models separate, approximately, intraband conduction from resonant absorption processes.
Thermal transport and the classical-statistical failure
Drude also treated electrons as carriers of heat. A kinetic estimate of the electronic thermal conductivity is
where is the electronic heat capacity per unit volume and is a characteristic carrier speed.
The model qualitatively explains why good electrical conductors are often good thermal conductors: the same mobile particles transport charge and energy. It also motivates the approximate proportionality expressed by the Wiedemann–Franz law,
where is the Lorenz number.
Its classical statistics are nevertheless incorrect for ordinary metals. The equipartition theorem predicts an electronic heat capacity of
where is the Boltzmann constant. Measured electronic heat capacities at temperatures well below the Fermi temperature are much smaller and approximately proportional to temperature.
The resolution requires Fermi–Dirac statistics and the Pauli exclusion principle. Only states near the Fermi energy can readily change their occupations under weak thermal excitation. In the low-temperature, elastic-scattering limit, quantum transport theory yields
rather than the value obtained from a classical electron gas. Deviations from this limit can occur when heat and charge currents relax differently.
Quantum extensions and modern use
Arnold Sommerfeld’s quantum free-electron treatment, developed in the late 1920s, replaced classical electron statistics with Fermi–Dirac statistics. The resulting Sommerfeld model corrected major thermodynamic failures while preserving much of the simple transport framework.
In crystalline materials, further improvements incorporate band structure. For a simple isotropic parabolic band, the Drude conductivity can retain its form with the electron mass replaced by an effective mass:
More general bands require direction-dependent masses or transport integrals rather than a single scalar .
The Drude framework remains useful for interpreting electrical and optical measurements in metals, doped semiconductors, and other conducting systems. Modern fits often treat and as effective parameters. A good fit to a Drude spectrum establishes a useful response function, but does not by itself demonstrate that the carriers obey classical statistics or identify the microscopic scattering mechanism.
Limits of validity
The model’s principal limitations follow from its independent-carrier, single-relaxation-time description:
- Microscopic scattering is unspecified. Temperature-dependent resistance requires a theory of processes such as electron–phonon scattering, impurity scattering, and other momentum-relaxing mechanisms.
- Carrier populations may be complex. Multiple bands and anisotropic velocities can invalidate single-density and single-mass interpretations.
- Transport may be nonlocal or ballistic. When sample dimensions or field-variation scales become comparable to the mean free path, a local conductivity may be insufficient.
- Strong interactions and quantum interference are omitted. Correlated-electron behavior, localization, and coherent quantum corrections can lie outside the model.
- Linear response can fail. Strong fields may substantially alter the carrier distribution or temperature, making constant parameters inappropriate.
The mean free path is often estimated as . In a degenerate metal, the relevant characteristic speed is generally the Fermi velocity, not the classical thermal speed. This distinction illustrates why the Drude transport equations can remain useful even when the original classical interpretation of electron motion is no longer valid.