Nernst Effect: Thermomagnetic Transverse Voltage Generation

Defining the effect and its geometry

The standard Nernst configuration involves three mutually orthogonal directions: a temperature gradient -∇T applied along, say, the x-axis; a magnetic field B applied along z, perpendicular to the sample plane; and the resulting Nernst electric field E_y measured along y, transverse to both. The Nernst coefficient is typically defined as N = E_y / (|∇T| B) under the condition of zero electric current flow (open-circuit measurement), analogous to how the Hall coefficient is defined under zero magnetic-field-induced current redistribution. This transverse thermomagnetic response is one entry in a broader matrix of thermogalvanomagnetic effects that also includes the longitudinal Seebeck effect (voltage from a temperature gradient with no field), the Righi-Leduc effect (a thermal analogue of the Hall effect, producing a transverse temperature gradient from a longitudinal heat current in a magnetic field), and the ordinary Hall effect (transverse voltage from a longitudinal current in a field). All of these effects share the same underlying physical origin: the Lorentz force couples the drift or diffusion of charge carriers to the applied magnetic field, and whichever driving force is present (electric field, temperature gradient) gets partially redirected into the perpendicular direction. Experimentally, isolating the pure Nernst signal requires care, since a real sample in a temperature gradient and magnetic field can simultaneously develop a longitudinal Seebeck voltage, and stray misalignment between contacts can mix a small longitudinal component into a transverse measurement. Standard practice is to measure the transverse voltage for both positive and negative field polarity and antisymmetrize the result, since the Nernst signal is odd under field reversal while any residual Seebeck contamination from contact misalignment is even, allowing the two contributions to be cleanly separated in the raw data before further analysis.

Microscopic origin: why an ordinary metal has a tiny Nernst signal

A naive picture would suggest that hot carriers diffusing down a temperature gradient should always produce a sizable transverse deflection in a field, yet the Nernst signal in a simple single-band metal with an energy-independent relaxation time is anomalously small, a fact sometimes called Sondheimer cancellation. The reason is that the Nernst effect is fundamentally sensitive to the energy dependence of the electronic relaxation time and density of states near the Fermi level, not merely to the existence of a Hall angle. In the Mott formula framework, thermoelectric transport coefficients are proportional to the energy derivative of the corresponding electrical transport coefficient evaluated at the Fermi energy. For the ordinary Hall conductivity, which itself typically varies smoothly and weakly with energy in a simple parabolic band, this derivative is small, so the Nernst signal it produces is correspondingly weak. This is why the Nernst effect becomes a particularly sensitive probe in systems where the electronic structure changes rapidly near the Fermi level, such as multiband semimetals, materials near a Lifshitz transition, or systems with strong energy-dependent scattering, where the naive cancellation is broken and a large ordinary Nernst signal emerges even without any exotic physics. This same sensitivity makes the Nernst effect a valuable complement to angle-resolved photoemission and quantum oscillation measurements when characterizing a new material's electronic structure, since a Nernst signal that is anomalously large or that changes sign as a function of temperature, field, or carrier doping is often the first transport signature that a naive single-band picture is inadequate, prompting more detailed spectroscopic or thermodynamic investigation of the underlying Fermi surface. In practice, calculating the Nernst coefficient from first principles requires going beyond a simple relaxation-time approximation to properly capture the energy dependence of both the scattering rate and the group velocity across the relevant band, which is why theoretical Nernst calculations are considerably more demanding than analogous calculations of the ordinary resistivity or Hall coefficient, and why careful comparison between measured and calculated Nernst signals has become a stringent test of proposed band-structure models for many correlated and topological materials.

Carrier type, sign, and multiband compensation

One of the most diagnostically useful features of the Nernst effect is its sign dependence on carrier type combined with its behavior in multiband materials. For a simple single-carrier-type metal or semiconductor, the sign of the ordinary Nernst signal reflects the sign of the dominant carrier's charge in combination with the curvature of its band, but unlike the Seebeck coefficient the relationship is not always a simple direct readout of electron versus hole character, because the Nernst response also depends on the mobility and its energy dependence. In compensated semimetals containing comparable densities of electrons and holes, such as bismuth or graphite, the electron and hole contributions to the ordinary Hall effect tend to cancel, but their contributions to the Nernst effect can instead add constructively, because the two carrier types drift in opposite directions under the same temperature gradient yet are deflected by the Lorentz force toward the same transverse edge when their charges are opposite. This is why semimetals like bismuth display some of the largest Nernst coefficients of any material, and why the Nernst effect has historically been prized as an especially sensitive probe of compensated, low-carrier-density semimetals where conventional Hall measurements are comparatively uninformative due to cancellation between electron and hole contributions. Graphite and more recently graphene-based devices provide a particularly clean testing ground for this compensation physics, since gate-tunable carrier density allows the relative populations of electrons and holes to be swept continuously through charge neutrality, and the Nernst signal has been shown to track this crossover with a distinctive sign change and peak structure that directly maps onto the changing balance of electron and hole contributions, providing an independent transport-based confirmation of band structure predictions for these Dirac-like materials.

Vortex Nernst effect in superconductors

Beyond conventional metals, the Nernst effect plays a starring role in the physics of type-II superconductors above their transition temperature. In the mixed state, magnetic flux penetrates the superconductor as quantized vortices, each carrying a normal core. A temperature gradient exerts a thermal force on these vortices, driving them to drift along the direction of heat flow, and this vortex motion generates a phase slip that produces a transverse electric field exactly analogous to the classic flux-flow voltage seen in current-driven vortex motion. Because vortices are essentially absent in a purely normal metal but persist as a measurable Nernst signal even above the zero-resistance transition temperature in underdoped cuprate superconductors, the vortex Nernst effect became one of the key pieces of experimental evidence, developed extensively by Ong and collaborators, for the existence of a broad regime of superconducting phase fluctuations above the resistive transition temperature in high-Tc cuprates. In this regime, Cooper pairs and vortex-like excitations persist without global phase coherence, a discovery that reshaped understanding of the cuprate phase diagram and the so-called pseudogap region. The characteristic signature distinguishing the vortex-Nernst contribution from the quasiparticle background is its strong nonlinearity in magnetic field at low field, reflecting the field-dependent density of thermally created vortex-antivortex pairs, together with an onset temperature that tracks the loss of long-range phase coherence rather than the loss of Cooper pairing itself. Subsequent work extended vortex-Nernst measurements to iron-based superconductors and other unconventional pairing systems, establishing the technique as a general-purpose probe for detecting fluctuating superconducting correlations wherever a resistive transition alone might otherwise be mistaken for the complete disappearance of pairing. Because the vortex contribution and any normal-state quasiparticle Nernst background can in principle overlap in temperature and field range, careful experiments typically combine Nernst measurements with independent probes such as torque magnetometry, diamagnetic susceptibility, and specific heat to build a consistent picture of where genuine superconducting fluctuations begin and end, an interpretive challenge that has itself driven significant methodological refinement in the field over the past two decades.

Applications: thermomagnetic energy conversion and characterization

Because the Nernst effect converts a temperature gradient directly into a transverse voltage without requiring a p-n junction or a thermocouple pair, it has attracted attention as the basis for solid-state Nernst-effect-based thermomagnetic generators and heat-flux sensors, particularly the anomalous Nernst effect in magnetic materials, which produces a transverse voltage even at zero external field once a material is spontaneously magnetized. Anomalous Nernst devices built from magnetic thin films offer a simplified lateral geometry compared to conventional Seebeck-based thermoelectric modules, because the output voltage can be collected along the length of a single strip of magnetic material rather than requiring alternating legs of p-type and n-type material, potentially simplifying large-area or flexible heat-harvesting applications, though efficiencies achieved so far remain modest compared to established Seebeck-based technologies. Separately, the ordinary Nernst effect has become a routine characterization tool for identifying changes in Fermi surface topology, detecting the onset of density-wave order, and probing electronic nematicity in correlated-electron materials such as iron-based superconductors and cuprates, precisely because its unusually strong sensitivity to the energy derivative of transport coefficients makes it reveal band-structure changes that are nearly invisible in resistivity or ordinary Hall measurements. In iron-based superconductor parent and underdoped compounds, for instance, the onset of an electronic nematic phase, in which the electronic structure spontaneously breaks the fourfold rotational symmetry of the underlying lattice down to twofold, produces a pronounced anisotropy between Nernst signals measured along the two originally equivalent crystal axes, an anisotropy that appears well before, and sometimes helps predict, the structural transition detected by diffraction. This makes angle-resolved Nernst measurements one of the more sensitive available electronic-symmetry-breaking detectors in the experimental toolkit of correlated-electron materials research. Nernst measurements have similarly been used to track density-wave transitions in materials such as underdoped cuprates and certain rare-earth intermetallics, where the sudden reconstruction of the Fermi surface as small electron or hole pockets form can produce a dramatic, sign-changing enhancement of the Nernst signal right at the transition temperature, providing a transport-based complement to X-ray and neutron diffraction evidence for the onset of charge or spin density-wave order.

Frequently asked questions

Try it live

Everything above runs in your browser — open Nernst Effect: Thermomagnetic Transverse Voltage Generation and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.