Probes, Devices, and Experiments

A probe selects a coupling channel, a region of the sample and an observation window. Choosing a measurement is therefore part of specifying the physical question. A spectral gap, a resistance anomaly and a thermodynamic transition can constrain the same proposed state, but they do not measure the same object or share the same inference limits.

Start with what the apparatus records, then ask which physical quantity can be inferred from it. The measurement guide explains calibration and this conversion; the interpretation guide helps compare the mechanisms that could produce the result. The individual probe pages show how the coupling, sample preparation and measurement conditions enter that reasoning.

Helpful background. From Models to Measured Observables helps specify the system, preparation, scale and observable. The individual probe pages state the additional theory and experimental background their reductions use.

Enter this chapter

If the measured quantity itself is unclear, begin with How Quantum Matter Is Measured. If a calibrated feature is established but its explanation is uncertain, begin with Data Interpretation and Pitfalls. Otherwise choose the relevant channel below.

Your measurement or question	Start here	What you will learn
What observable can this detector constrain?	How Quantum Matter Is Measured	Calibration, coupling, resolution, locality and the forward model.
What does a terminal voltage or current mean?	Transport Measurements	Contacts, geometry, excitation and the conversion to a material coefficient.
What produces the transverse voltage?	Hall Measurements	Terminal polarity, reversals, tensor reduction and channel assumptions.
What does an inverse-field oscillation constrain?	Quantum Oscillations	Orbit areas, masses, damping and the limits of phase inference.
Which occupied electronic states contribute to photoemission?	Angle-Resolved Photoemission Spectroscopy	Energy and momentum calibration, matrix elements and surface sensitivity.
What does a spatial tunneling spectrum resolve?	Scanning Tunneling Microscopy and Spectroscopy	Junction conditions, feedback, local spectra and scattering modulations.
Which nuclear or magnetic fluctuations scatter neutrons?	Neutron Scattering	Cross sections, polarization, normalization and resolution.
Which structural or electronic modulation scatters x-rays?	X-Ray Scattering	Diffraction, resonant channels, absorption and energy-loss interpretation.
Which symmetry channel produces an optical feature?	Raman and Optical Spectroscopy	Polarization, vertices and vibrational, magnetic or electronic assignments.
What complex response follows from an optical field or power?	Terahertz and Infrared Probes	Reference measurements, sample geometry and optical inversion.
How does excitation change the measured state?	Pump–Probe Spectroscopy	Absorbed energy, depth, overlap, delay and recovery.
What bulk energy or entropy constraint is measured?	Heat Capacity and Thermodynamics	Thermal coupling, addenda, amount convention and component separation.
What magnetic response belongs to the sample?	Magnetic Susceptibility	Backgrounds, demagnetization, normalization and history.
How did processing change the quantum system?	Device Fabrication Concepts	How interfaces, contacts, gates and fabrication history change the measured sample.
Which explanation survives the controls?	Data Interpretation and Pitfalls	Competing mechanisms, nuisance parameters and a discriminating next test.

Choose a channel that tests the proposed distinction

For charge flow, begin with Transport Measurements, then Hall Measurements or Quantum Oscillations according to the record. If the response framework itself is uncertain, first use Transport, Response, and Optics to select bulk, kinetic, optical or terminal theory. An accurately measured resistance need not equal an intrinsic resistivity, and a measured oscillation frequency does not independently fix carrier multiplicity.

For electronic spectra, ARPES selects occupied momentum-resolved photoemission intensity, while STM/STS selects a junction-weighted local signal. Their sensitivity to matrix elements, occupation, surfaces and instrumental resolution is part of the comparison. Tersoff and Hamann’s tunneling model illustrates why a relation between tunneling current and local electronic structure requires a specified junction approximation.

For structure and collective excitations, choose X-Ray Scattering, Neutron Scattering or Raman according to coupling, momentum, polarization and energy window. Van Hove’s correlation-function formulation connects scattering to dynamical structure under stated scattering assumptions; Structure Factors develops that connection. A broad continuum has several possible origins, so identify the sampled channel and plausible alternatives before assigning a phase.

For electrodynamics, THz and Infrared Probes handles the conversion from fields or powers to a complex response. Choose Pump–Probe when excitation changes preparation, occupation or time-translation invariance. A time-domain detector by itself does not make the material nonequilibrium: an equilibrium linear response can also be measured in the time domain.

Heat Capacity and Magnetic Susceptibility add bulk thermodynamic and magnetic constraints. Their extensive normalization, sample geometry and background models must accompany the reported result. Device Fabrication Concepts belongs in every comparison where processing, contacts, gates or layer history could explain a difference. It supplies the physical preparation questions rather than a cleanroom recipe.

Compare probes of the same proposed state

Before combining measurements, check that they address compatible specimens, gate settings, temperatures and histories. Determine whether they sample the same layer or depth and whether the features are stable during acquisition. Two analyses of the same raw dataset may share calibration, extrapolation and model errors even when their plots look different. Taylor and Kuyatt’s NIST guidance provides the uncertainty framework; Data Interpretation and Pitfalls supplies the material-model comparison.

One orbit does not count all carriers

Consider synthetic observations from the same stationary specimen and gate setting: an oscillation frequency $F=200\,\mathrm T$, an approximately circular ARPES contour with $k_F=0.078\,\text{Å}^{-1}$, and a low-field sheet Hall slope $R_H=-32.3\,\Omega/\mathrm T$. Suppose calibration uncertainties are much smaller than a factor of two. Use the orientation in Hall Measurements: sheet current density $j_x$ along $+x$, field along $+z$, and transverse sheet resistance $E_y/j_x$ with bottom-minus-top voltage. An ordinary single electron channel then has a negative Hall slope.

Test a definite hypothesis: one circular two-dimensional electron pocket, spin degeneracy two, no additional valley or layer channel, an isotropic one-carrier weak-field Hall model, and no surface/bulk mismatch between the probes. The orbit-area relation associated with Onsager gives

$$\begin{aligned} A_F&=\frac{2\pi eF}{\hbar} \simeq0.01909\,\text{Å}^{-2},\\ k_F&=\sqrt{A_F/\pi} \simeq0.07796\,\text{Å}^{-1}. \end{aligned}$$

The ARPES radius and the oscillation area agree to the displayed precision. But the assumed degeneracy gives

$$\begin{aligned} n_{2\mathrm D}&=\frac{2A_F}{(2\pi)^2} \simeq9.672\times10^{12}\,\mathrm{cm}^{-2},\\ R_H^{(1)}&=-\frac{1}{en_{2\mathrm D}} \simeq-64.53\,\Omega/\mathrm T. \end{aligned}$$

That Hall prediction fails. Interpreting the measured slope with the same one-carrier formula would instead give $n_H\simeq1.932\times10^{13}\,\mathrm{cm}^{-2}$. Averaging these inferred densities would hide the failed hypothesis.

Two equal-area, spin-degenerate electron channels with equal mobilities are one sufficient alternative. Their oscillation frequencies coincide while their conductivities add, giving about $-32.27\,\Omega/\mathrm T$. This countermodel is not a unique diagnosis. Unequal-mobility channels, unresolved pockets, depth mismatch or a calibration error must be tested through their own forward predictions.

To distinguish these explanations, vary angle in Quantum Oscillations, photon energy and polarization in ARPES, or gate voltage in a joint Hall and longitudinal-conductivity fit. Check whether fabrication could have left a parallel conducting layer. Fermi Surface and Hall Effect derive the area and response formulas used here. Agreement of orbit areas alone does not assign a topological phase.

Connect the measurement to a physical model

Once the probe reduction is justified, use the physical question to choose the next theory page. A quasiparticle width belongs with the spectral-function framework; a measured optical coefficient belongs with Optical Conductivity; a proposed spin-liquid identification belongs with Quantum Spin Liquids. Computational Quantum Matter supplies numerical convergence and model-to-probe comparisons.

The basic quantum measurement formalism explains outcome probabilities and conditional state updates. For current experimental results and unsettled claims about materials, follow the Quantum Matter frontiers guide. A calibrated feature remains useful even when several mechanisms are still viable.

Exercises

1. Choose a probe and a complementary control

For each observation, choose the relevant probe guide, one useful complementary control and a stronger conclusion the observation alone cannot establish: a spatially varying tunneling gap; a broad magnetic neutron continuum; and a pump-induced low-frequency inductive response.

Solution

For the tunneling gap, begin with STM/STS and check tip, setpoint and junction stability. Choose ARPES for a momentum-resolved spectral comparison or calorimetry for a bulk transition constraint, according to the proposed interpretation. A tunneling gap alone does not establish bulk superconductivity or its pairing symmetry.

For the neutron continuum, begin with Neutron Scattering and test polarization, momentum dependence and background subtraction. Thermodynamic and magnetic measurements can constrain competing states. A broad continuum alone does not establish deconfined fractional excitations; compare the possible mechanisms in Quantum Spin Liquids.

For the pump-induced signal, use Pump–Probe for preparation and recovery and THz/IR for the field-to-response inversion. Add heating and penetration-depth controls and an independent order-sensitive response. A transient inductive feature alone does not prove an equilibrium superconducting phase. The useful control is the one that distinguishes the proposed alternatives, rather than merely repeating the original measurement.

2. Two reductions of one record

Two unbiased estimates of the same dimensionless parameter each have standard uncertainty $s=0.04$ and correlation coefficient $\rho=0.90$ because they share calibration. Find the standard uncertainty of their equally weighted mean. Does agreement between the estimates provide two independent confirmations of a mechanism?

Solution

Including the covariance gives

$$\operatorname{Var}\!\left(\frac{x_1+x_2}{2}\right) =\frac{s^2}{2}(1+\rho).$$

The standard uncertainty is $s\sqrt{(1+\rho)/2}\simeq0.03899$. Incorrectly assuming independence would give $s/\sqrt2\simeq0.02828$. At $\rho=1$, the uncertainty remains $s$: a fully shared contribution does not average away.

This calculation concerns a specified common measurand and uncertainty model, not the discrepancy between competing physical mechanisms. Two reductions of one record are not two independent probes. How Quantum Matter Is Measured supplies calibration and covariance questions; Data Interpretation supplies the mechanism comparison. Even measurements with different instruments can share specimen or calibration uncertainties.

3. Optical weight and static screening

An equilibrium sample has a regular-background-subtracted optical response $\sigma_2\simeq D/\omega$, with $D=1.0\times10^{18}\,\mathrm{S\,m^{-1}\,s^{-1}}$. An analysis of the same dataset reports missing-area consistency. If the pole is the local London superfluid term, what penetration depth follows? State the remaining assumptions and an independent screening test.

Solution

Under the stated London interpretation, $D=1/(\mu_0\lambda^2)$. With $\mu_0\simeq1.2566\times10^{-6}\,\mathrm{H\,m^{-1}}$, the conditional result is $\lambda=(\mu_0D)^{-1/2}\simeq0.892\,\mu\mathrm m$.

The missing-area argument associated with Ferrell and Glover requires its response-accounting assumptions. For the usual positive-frequency finite-frequency comparison, check that the normal reference has no zero-frequency delta function, the full-response contact terms are matched, the integration cutoff is converged, and the inferred pole is the local London term without an additional ballistic contribution. The generalized optical accounting states these limits. Two reductions of one optical dataset can share calibration and extrapolation errors.

A clean normal Drude response can have zero-frequency optical weight without static Meissner screening. Use Magnetic Susceptibility with geometry, demagnetization, backgrounds and history controlled for a separate screening measurement; Heat Capacity can add a bulk transition and entropy constraint. London Theory derives the stiffness-to-screening relation. If excitation drives the sample out of equilibrium, return to Pump–Probe before applying an equilibrium phase interpretation.

References

• R. A. Ferrell and R. E. Glover III, “Conductivity of Superconducting Films: A Sum Rule,” Physical Review 109, 1398–1399 (1958), doi:10.1103/PhysRev.109.1398. Original optical missing-area relation; the linked Optical Conductivity page explains its assumptions.
• L. Onsager, “Interpretation of the de Haas–van Alphen Effect,” Philosophical Magazine 43, 1006–1008 (1952), doi:10.1080/14786440908521019. The connection between oscillation frequency and orbit area.
• B. N. Taylor and C. E. Kuyatt, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results, NIST Technical Note 1297, 1994 edition, official text. Combined standard uncertainty and correlated inputs, especially section 5 and Appendix A.
• J. Tersoff and D. R. Hamann, “Theory of the scanning tunneling microscope,” Physical Review B 31, 805–813 (1985), doi:10.1103/PhysRevB.31.805. A specified approximation connecting tunneling and local electronic structure.
• L. Van Hove, “Correlations in Space and Time and Born Approximation Scattering in Systems of Interacting Particles,” Physical Review 95, 249–262 (1954), doi:10.1103/PhysRev.95.249. Scattering and time-dependent correlations under stated approximations.