Coulomb Blockade

Coulomb blockade is the suppression of low-order charge transfer through a small, weakly contacted conductor because changing its integer charge costs electrostatic free energy. It is not merely a small conductance, a spectral gap, or a classical contact resistance. The diagnostic content lies in the dependence on gate charge, bias, temperature, and junction coupling.

The effect is simplest when an island is joined to source and drain by opaque tunnel barriers and coupled capacitively to a gate. The island can then remain in a definite charge sector long enough for the energy cost of one extra electron to control transport. Gate voltage shifts the relative energies of those sectors; source–drain voltage opens a transport window; tunnel events connect neighboring sectors.

This page is the canonical home for the orthodox capacitance-network ledger, free-energy changes of individual junction events, Coulomb-diamond slopes, and the operating principle of a single-electron transistor. Quantum Dots owns confinement spectra, orbital filling, and the constant-interaction addition spectrum. Mesoscopic Transport owns master equations, current noise, and counting statistics. Single-Electron Devices owns device engineering, charge sensing, pumps, turnstiles, and metrological error budgets.

Regime and Convention Ledger

Two conventions must be declared before comparing formulas.

First, this page defines

$$E_C \equiv \frac{e^2}{2C_\Sigma},$$

so that the electrostatic energy is $E_C(n-n_g)^2$. The electrostatic contribution to the spacing between successive addition thresholds is

$$\frac{e^2}{C_\Sigma} = 2E_C.$$

Some literature calls $e^2/C_\Sigma$ the charging energy. Neither convention is wrong, but silently switching between them creates a factor-of-two error.

Second, $V_j$ denotes the electrical potential of electrode $j$, whereas an electron reservoir has electrochemical potential

$$\mu_j = -eV_j.$$

Raising an electrode’s electrical potential therefore lowers its electron electrochemical potential. Coulomb-diamond slopes change sign if a different voltage or current convention is used.

The strict orthodox limit assumes:

$$k_BT \ll \frac{e^2}{C_\Sigma},$$ $$\hbar\gamma_j \ll \frac{e^2}{C_\Sigma},$$

and, for a normal tunnel junction,

$$R_j \gg R_K, \qquad R_K \equiv \frac{h}{e^2} \simeq 25.8\ \mathrm{k\Omega}.$$

Here $\gamma_j$ is a tunneling rate, so $\hbar\gamma_j$ is an energy width. These inequalities identify a controlled regime rather than sharp phase boundaries. Charge quantization is progressively renormalized and broadened as a contact opens.

Scale or parameter	Role	Failure when it is too large or too small
$e^2/C_\Sigma$	separates neighboring charge thresholds	thermal and lifetime broadening wash out blockade when unresolved
$k_BT$	thermally populates uphill events	blockade valleys acquire activated current
$\hbar\gamma_j$	lifetime broadening from contact $j$	neighboring charge sectors hybridize when broadening is strong
$R_K/R_j$	dimensionless tunnel-coupling scale	higher-order tunneling and charge fluctuations grow as it approaches unity
$e\lvert V_{\mathrm{SD}}\rvert$	opens the transport window	sequential channels appear at diamond boundaries
background charge $Q_0$	offsets the gate-induced charge	drift moves the entire pattern without changing the ideal period

Three equivalent descriptions in the simplified three-capacitor model. Source, drain, and gate induce the offset charge of an island with integer excess-electron number $n$. Neighboring electrostatic parabolas are degenerate at half-integer $n_g$. With the drain grounded and source biased at $V_{\mathrm{SD}}$, sequential-tunneling thresholds form diamond edges whose slopes are $-C_g/C_S$ and $C_g/(C_D+C_g)$. Additional stray capacitance changes the second denominator to $C_\Sigma-C_S$.

Charging Energy

Capacitance network

Consider an island coupled capacitively to source, drain, gate, and any other conductors. Define

$$C_\Sigma = C_S+C_D+C_g+C_0,$$

where $C_0$ collects stray and deliberately added capacitances not shown separately. After the linear capacitor degrees of freedom and the work done by ideal voltage sources are accounted for, the charge-dependent free energy can be written, up to an $n$-independent term, as

$$F_n = E_C \left( n-n_g \right)^2,$$

with

$$n_g \equiv \frac{ C_SV_S+C_DV_D+C_gV_g+Q_0 }{e}.$$

$n$ is the integer number of excess electrons and $Q_0$ is a static offset charge. A shift $Q_0\mapsto Q_0+e$ simply relabels the charge sectors. Slowly varying fluctuators instead produce charge-offset drift and low-frequency noise.

The free-energy cost to add the $(n+1)$th excess electron is

$$\begin{aligned} \mu_{n+1}^{\mathrm{isl}} &\equiv F_{n+1}-F_n \\ &= 2E_C \left( n+\frac12-n_g \right). \end{aligned}$$

Neighboring sectors are degenerate when

$$n_g = n+\frac12.$$

At zero source–drain bias, changing the gate voltage by

$$\Delta V_g = \frac{e}{C_g}$$

advances $n_g$ by one and returns the electrostatic problem to the same point in the next charge sector. This is the ideal gate period of Coulomb oscillations.

At the center of the $n$th stability interval, $n_g=n$, the costs to add or remove an electron are both

$$E_C = \frac{e^2}{2C_\Sigma}.$$

The second difference of the island energy is instead

$$\begin{aligned} F_{n+1} -2F_n +F_{n-1} &= \frac{e^2}{C_\Sigma} \\ &= 2E_C. \end{aligned}$$

That second difference is the electrostatic contribution to the addition-energy spacing measured between neighboring degeneracy points.

What the capacitance model omits

A metallic island can have a level spacing much smaller than $k_BT$ and still show charging effects. A few-electron semiconductor dot usually requires a discrete-spectrum term as well:

$$\mathcal E_N = \sum_{i\in\mathrm{occupied}} \epsilon_i + E_C \left( N-n_g \right)^2 + \delta E_{\mathrm{int}}.$$

Then

$$\mu_N = \mathcal E_N-\mathcal E_{N-1}$$

contains orbital, exchange, valley, spin, and residual-interaction contributions. The capacitance model predicts the common electrostatic backbone; it does not prove equal peak spacing or identify an excited state. The full addition-spectrum interpretation is developed in Quantum Dots.

Charging-scale example

Suppose the coefficient in the declared convention is

$$E_C = 1.00\ \mathrm{meV}.$$

Then

$$\begin{aligned} C_\Sigma &= \frac{e^2}{2E_C} \\ &\simeq 80.1\ \mathrm{aF}. \end{aligned}$$

The corresponding electrostatic addition increment is $2E_C=2.00\ \mathrm{meV}$. A paper using $E_C=e^2/C_\Sigma$ would quote that latter number as its charging energy for the same device.

Tunnel Barriers

Charging energy alone does not produce an observable transport blockade. The island must exchange charge slowly enough with its reservoirs that integer charge sectors remain meaningful, yet rapidly enough for a current to be measured.

Opaque-junction criterion

For a many-channel normal tunnel junction,

$$G_T \simeq \frac{e^2}{h} \sum_a T_a,$$

where every transmission eigenvalue satisfies $T_a\ll1$. The dimensionless conductance is

$$g_T \equiv \frac{G_T}{e^2/h} = \frac{R_K}{R_T}.$$

The orthodox expansion is controlled when $g_T\ll1$, equivalently $R_T\gg R_K$. This resistance criterion concerns quantum charge fluctuations. It is distinct from the dynamical time constant $R_TC$ and from the requirement that the measured current exceed an instrument’s noise floor.

An experimentally useful barrier therefore obeys competing demands:

$$\hbar\gamma \ll k_BT,\frac{e^2}{C_\Sigma}$$

for thermally broadened sequential tunneling, while

$$\gamma \gtrsim \frac{1}{t_{\mathrm{meas}}}$$

is needed to observe enough events in a measurement interval $t_{\mathrm{meas}}$. Closing a barrier improves charge isolation but eventually makes transport impractically slow.

Free-energy ledger for one event

For any proposed tunneling event, define

$$\Delta F \equiv F_{\mathrm{final}}-F_{\mathrm{initial}},$$

including the island and voltage-source work appropriate to that event. In the normal-metal orthodox approximation, a junction of resistance $R_T$ has rate

$$\gamma(\Delta F) = \frac{1}{e^2R_T} \frac{-\Delta F}{ 1-\exp\left( \Delta F/k_BT \right) }.$$

The zero-temperature limits expose the physics:

$$\gamma(\Delta F) \longrightarrow \begin{cases} -\Delta F/(e^2R_T), & \Delta F<0, \\[4pt] 0, & \Delta F>0. \end{cases}$$

Downhill events occur; uphill events freeze out. At finite temperature,

$$\frac{ \gamma(\Delta F) }{ \gamma(-\Delta F) } = \exp \left( -\frac{\Delta F}{k_BT} \right),$$

which is local detailed balance.

As a scale estimate, take $R_T=1.0\ \mathrm{M\Omega}$ and a favorable event with $\Delta F=-0.20\ \mathrm{meV}$ at sufficiently low temperature. Then

$$\gamma \simeq 1.25\times10^9\ \mathrm{s^{-1}},$$

and the associated energy width is only

$$\hbar\gamma \simeq 0.82\ \mu\mathrm{eV}.$$

A gigahertz event rate can therefore coexist with sub-microelectronvolt lifetime broadening. Rate and energy width must not be denoted by one unqualified symbol.

Sequential tunneling is not the whole current

Inside an ideal diamond, every first-order route contains an uphill step, so sequential current is thermally suppressed. The measured current need not vanish:

• elastic cotunneling transfers charge through a virtual island state without exciting the island;
• inelastic cotunneling leaves an excitation behind once sufficient bias is available;
• Kondo correlations can restore a zero-bias resonance in an odd-occupation valley;
• photon- or phonon-assisted tunneling exchanges energy with another mode;
• leakage paths or unintended islands can mimic a residual blockade current.

These mechanisms have distinct temperature, bias, magnetic-field, and barrier-coupling dependences. A dark region in a conductance map establishes suppressed conductance, not by itself the microscopic reason.

An electromagnetic environment can also absorb or supply energy during tunneling. This dynamical Coulomb blockade is described by an environmental energy-exchange distribution such as $P(E)$. It is related historically and mathematically to charging physics, but it is not identical to the static integer-charge blockade of a double-junction island.

Coulomb Diamonds

A Coulomb diamond is a low-conductance region in a map of differential conductance,

$$\frac{dI}{dV_{\mathrm{SD}}},$$

versus gate voltage and source–drain bias. Its boundaries are alignments between an island charge-transition electrochemical potential and a reservoir electrochemical potential.

Deriving the edge slopes

Choose the explicit convention

$$V_D=0, \qquad V_S=V_{\mathrm{SD}},$$

so

$$\mu_D=0, \qquad \mu_S=-eV_{\mathrm{SD}}.$$

For one transition, absorb the charge index, offset charge, and voltage-independent microscopic energies into $\mu^0$. The island transition is

$$\mu^{\mathrm{isl}} = \mu^0 - e\frac{C_g}{C_\Sigma}V_g - e\frac{C_S}{C_\Sigma}V_{\mathrm{SD}}.$$

The drain-alignment edge satisfies

$$\mu^{\mathrm{isl}} = \mu_D = 0.$$

Differentiating along that line gives

$$\left. \frac{dV_{\mathrm{SD}}}{dV_g} \right|_D = -\frac{C_g}{C_S}.$$

The source-alignment edge satisfies

$$\mu^{\mathrm{isl}} = \mu_S = -eV_{\mathrm{SD}},$$

and therefore

$$\left. \frac{dV_{\mathrm{SD}}}{dV_g} \right|_S = \frac{C_g}{C_\Sigma-C_S}.$$

For the three-capacitor model with $C_0=0$,

$$C_\Sigma-C_S = C_D+C_g.$$

The slope labels swap if the opposite lead is biased, and both signs can flip if the plotted bias is defined oppositely. Published slope formulas are meaningful only with their bias convention attached.

The gate lever arm is

$$\alpha_g \equiv \frac{C_g}{C_\Sigma},$$

so a small horizontal gate displacement at fixed lead voltages changes the transition energy by

$$\delta\mu^{\mathrm{isl}} = -e\alpha_g\,\delta V_g.$$

This conversion is often more reliable than treating the full vertical diamond height as simply an energy divided by $e$. Bias voltage is partitioned capacitively, and the partition depends on which electrode is driven.

Extracting capacitances from a diamond

Suppose a stability diagram made with the convention above has:

$$\Delta V_g = 20.0\ \mathrm{mV},$$ $$m_- = -0.080, \qquad m_+ = +0.100.$$

The gate period gives

$$\begin{aligned} C_g &= \frac{e}{\Delta V_g} \\ &\simeq 8.01\ \mathrm{aF}. \end{aligned}$$

The negative edge gives

$$\begin{aligned} C_S &= -\frac{C_g}{m_-} \\ &\simeq 100.1\ \mathrm{aF}, \end{aligned}$$

and the positive edge gives

$$\begin{aligned} C_\Sigma-C_S &= \frac{C_g}{m_+} \\ &\simeq 80.1\ \mathrm{aF}. \end{aligned}$$

If stray capacitance is negligible, then

$$C_D = 80.1\ \mathrm{aF}-C_g \simeq 72.1\ \mathrm{aF},$$

so

$$C_\Sigma \simeq 180.2\ \mathrm{aF}, \qquad \alpha_g \simeq 0.0444.$$

The declared charging coefficient is then

$$E_C = \frac{e^2}{2C_\Sigma} \simeq 0.445\ \mathrm{meV}.$$

If $C_0$ is appreciable, the two slopes determine $C_S$ and the combination $C_D+C_g+C_0$, not $C_D$ alone. A three-capacitor fit would then be overinterpreted.

Reading structure beyond the edges

Feature in a stability diagram	Leading interpretation	Corroborating checks
repeated closed diamonds	stable integer charge sectors	one-electron gate periodicity, reproducible edge slopes, thermal activation
lines parallel to an edge outside blockade	additional island transition or excitation	magnetic-field evolution, temperature dependence, repetition across charge sectors
threshold inside a blockade region	inelastic cotunneling onset	weak gate dependence within one valley, barrier-order scaling
zero-bias ridge in an odd valley	possible Kondo resonance	odd–even pattern, logarithmic temperature evolution, Zeeman splitting
shifted duplicate diamonds	second island or charge rearrangement	abrupt offset jumps, multiple slope families, hysteresis
rounded or absent vertices	thermal broadening, lifetime broadening, noise, or open contacts	compare temperature and barrier-gate sweeps

Diamond edges establish energetic alignment. They do not by themselves identify spin, valley, phonon, or orbital character. That assignment needs a controlled perturbation and a model whose parameters remain consistent across the map.

Single-Electron Transistor

A normal-state single-electron transistor consists of one island, two tunnel junctions, and at least one gate capacitor. The gate modulates a source–drain current without ideally drawing a dc gate current.

At zero bias, adjacent charge states are degenerate whenever

$$n_g = n+\frac12.$$

Near such a point, both directions of a sequential cycle can become energetically accessible, producing a conductance peak. Near integer $n_g$, the island lies in the middle of a stable charge sector and sequential conductance is suppressed. Repeating this pattern gives Coulomb oscillations with ideal period

$$\Delta V_g = \frac{e}{C_g}.$$

The useful transfer quantity is the transconductance

$$g_m \equiv \frac{\partial I}{\partial V_g}.$$

It is largest on the side of a conductance peak rather than at the current maximum. A nearby charge changes the effective offset charge and moves the operating point, which is why a single-electron transistor can act as an electrometer.

The word transistor should not conceal the differences from a field-effect transistor:

Conventional field-effect transistor	Single-electron transistor
gate controls a continuous carrier-density barrier	gate shifts discrete charge degeneracies
transfer curve usually has one main turn-on	ideal transfer curve is periodic
operation need not resolve $e$-scale charge	operation relies on charge quantization
room-temperature operation is common	temperature must remain below the relevant charging scale
offset charge mainly shifts a threshold	one-electron offset shifts an entire oscillation period

The ideal periodic picture is modified by background-charge motion, cotunneling, barrier asymmetry, finite level spacing, superconductivity, nonequilibrium heating, and the impedance of the external circuit. Charge sensitivity, radio-frequency embedding, pumps, turnstiles, and metrological accuracy require a device-level noise and error ledger; they are not implied by the existence of a Coulomb peak.

Quantum Dots and Qubits

Coulomb blockade is a transport regime, not a material class. Its microscopic content depends on the island.

Island platform	What is quantized	Important additions to the orthodox model
normal-metal grain	total excess charge	dense quasiparticle spectrum, environmental impedance
gate-defined semiconductor dot	charge plus resolvable orbitals	spin, valley, exchange, spin–orbit coupling
nanowire, nanotube, or molecular dot	charge and confined longitudinal states	shells, vibrations, contact-dependent hybridization
superconducting island	charge parity and condensate phase	gap, quasiparticle poisoning, Josephson coupling, possible $2e$ periodicity
unintended disorder island	localized charge	unstable capacitances and multiple overlapping diamonds

For a spin qubit, blockade can stabilize a chosen electron number and provide the charge-sector map used for initialization and readout. It does not establish coherent control. Demonstrating a qubit additionally requires a two-level encoding, calibrated manipulation, state-selective measurement, and coherence or fidelity evidence.

Pauli spin blockade is also distinct. Coulomb blockade forbids a charge transition because no energetically allowed charge state lies in the transport window. Pauli blockade forbids or slows a transition because spin symmetry excludes the accessible final state even when the charge configuration is energetically allowed. Spin Qubits develops that distinction in the context of preparation, control, and readout.

Superconducting circuits provide another important boundary. A superconducting island can exhibit charging effects, but coherent Cooper-pair tunneling introduces a Josephson energy $E_J$ that competes with $E_C$. The spectrum then depends on both charge and phase; ordinary normal-state sequential tunneling is no longer a complete description.

Nearby phenomena that are not interchangeable

Name	What is suppressed	Primary mechanism
Coulomb blockade	low-order transfer onto an integer-charge island	electrostatic cost plus weak contacts
Pauli blockade	a spin-selective transition	fermionic symmetry and selection rules
Coulomb gap	low-energy single-particle density of states in a disordered interacting system	long-range interactions and localization
dynamical Coulomb blockade	low-energy tunneling through a junction in a circuit	energy exchange with electromagnetic modes
superconducting gap blockade	quasiparticle current below a gap threshold	pairing spectrum
photon blockade	occupation of a nonlinear optical mode	anharmonic level spacing

Similar current suppression does not make these mechanisms equivalent. Gate periodicity, bias thresholds, symmetry dependence, environmental impedance, and spectroscopy distinguish them.

Common Mistakes

1. Calling every low-current region Coulomb blockade. Contact depletion, a band gap, localization, a superconducting gap, and instrument floor can all suppress current.
2. Using “charging energy” without declaring the factor of two. State whether it means $e^2/(2C_\Sigma)$ or $e^2/C_\Sigma$.
3. Equating diamond height with addition energy by inspection. First declare which lead is biased and include capacitive division.
4. Confusing tunneling rate and energy broadening. A rate $\gamma$ becomes an energy only after multiplication by $\hbar$.
5. Treating $R_T\gg R_K$ as a thermal criterion. It controls quantum charge fluctuations; $k_BT\ll e^2/C_\Sigma$ is the separate thermal condition.
6. Assigning every parallel line to an excited orbital. Lead resonances, phonons, and unintended dots can produce similar slopes.
7. Claiming perfect isolation from a dark diamond. Cotunneling and other higher-order processes remain possible.
8. Equating charge stability with qubit coherence. A well-defined charge sector is useful infrastructure, not a coherence measurement.

Exercises

1. Degeneracy of neighboring charge sectors

Starting from

$$F_n = E_C(n-n_g)^2,$$

show that $F_n=F_{n+1}$ at $n_g=n+1/2$. Find the cost to add an electron at the center $n_g=n$ of the $n$th stability interval.

Solution

Set the two parabolas equal:

$$(n-n_g)^2 = (n+1-n_g)^2.$$

Subtracting the left side from the right gives

$$2(n-n_g)+1 = 0,$$

so

$$n_g = n+\frac12.$$

At $n_g=n$,

$$\begin{aligned} F_{n+1}-F_n &= E_C \left[ (n+1-n)^2-(n-n)^2 \right] \\ &= E_C. \end{aligned}$$

The central barrier to either neighboring sector is therefore $E_C=e^2/(2C_\Sigma)$ in the convention used here.

2. Capacitance from a charging scale

An experiment quotes $E_C=0.60\ \mathrm{meV}$ using $E_C=e^2/(2C_\Sigma)$. Find $C_\Sigma$ and the electrostatic addition increment.

Solution

The capacitance is

$$\begin{aligned} C_\Sigma &= \frac{e^2}{2E_C} \\ &\simeq 1.34\times10^{-16}\ \mathrm F \\ &= 134\ \mathrm{aF}. \end{aligned}$$

The electrostatic second difference is

$$\frac{e^2}{C_\Sigma} = 2E_C = 1.20\ \mathrm{meV}.$$

A source using $e^2/C_\Sigma$ as its definition would quote $1.20\ \mathrm{meV}$ as the charging energy.

3. Gate period and offset charge

A device has $C_g=4.00\ \mathrm{aF}$. Find the ideal Coulomb-oscillation period. If a fluctuator changes $Q_0$ by $0.15e$, by what fraction of one period does the pattern shift?

Solution

The period is

$$\begin{aligned} \Delta V_g &= \frac{e}{C_g} \\ &\simeq 40.1\ \mathrm{mV}. \end{aligned}$$

Because $n_g=(C_gV_g+Q_0+\cdots)/e$, keeping $n_g$ fixed requires

$$C_g\delta V_g+\delta Q_0 = 0.$$

Thus

$$\frac{\delta V_g}{\Delta V_g} = -\frac{\delta Q_0}{e} = -0.15.$$

The pattern shifts by $15\%$ of a period, in the direction opposite to the offset-charge change.

4. Diamond slopes

The drain is grounded and the source is biased. A diamond has slopes $m_-=-0.050$ and $m_+=+0.125$, and its gate period gives $C_g=6.0\ \mathrm{aF}$. Find $C_S$ and $C_\Sigma-C_S$. If $C_0=0$, find $C_D$ and $C_\Sigma$.

Solution

Using the declared convention,

$$m_- = -\frac{C_g}{C_S},$$

so

$$C_S = -\frac{C_g}{m_-} = 120\ \mathrm{aF}.$$

The positive edge gives

$$m_+ = \frac{C_g}{C_\Sigma-C_S},$$

hence

$$C_\Sigma-C_S = \frac{C_g}{m_+} = 48\ \mathrm{aF}.$$

If $C_0=0$, then

$$C_D = 48\ \mathrm{aF}-C_g = 42\ \mathrm{aF},$$

and

$$C_\Sigma = 120\ \mathrm{aF}+48\ \mathrm{aF} = 168\ \mathrm{aF}.$$

Without the $C_0=0$ assumption, the data determine $C_D+C_0=42\ \mathrm{aF}$ rather than $C_D$ separately.

5. Detailed balance of the orthodox rate

For

$$\gamma(\Delta F) = \frac{1}{e^2R_T} \frac{-\Delta F}{ 1-\exp(\Delta F/k_BT) },$$

show that $\gamma(\Delta F)/\gamma(-\Delta F)=\exp(-\Delta F/k_BT)$. Interpret the result for $\Delta F>0$.

Solution

Write $x=\Delta F/k_BT$. Apart from the common prefactor,

$$\gamma(\Delta F) \propto \frac{-\Delta F}{1-e^x},$$

whereas

$$\begin{aligned} \gamma(-\Delta F) &\propto \frac{\Delta F}{1-e^{-x}} \\ &= \frac{\Delta F e^x}{e^x-1}. \end{aligned}$$

Therefore

$$\frac{\gamma(\Delta F)}{\gamma(-\Delta F)} = e^{-x} = \exp \left( -\frac{\Delta F}{k_BT} \right).$$

For $\Delta F>0$, the forward event raises free energy and is Boltzmann-suppressed relative to its downhill reverse.

6. Rate, broadening, and thermal resolution

A junction produces a tunneling rate $\gamma=5.0\times10^8\ \mathrm{s^{-1}}$ at $T=100\ \mathrm{mK}$. Compare $\hbar\gamma$ and $k_BT$ in microelectronvolts. Is the resonance predominantly lifetime broadened in this comparison?

Solution

The lifetime width is

$$\begin{aligned} \hbar\gamma &\simeq 5.27\times10^{-26}\ \mathrm J \\ &\simeq 0.329\ \mu\mathrm{eV}. \end{aligned}$$

The thermal scale is

$$\begin{aligned} k_BT &\simeq 1.38\times10^{-24}\ \mathrm J \\ &\simeq 8.62\ \mu\mathrm{eV}. \end{aligned}$$

Thus

$$\hbar\gamma \ll k_BT.$$

Within this comparison the line is thermally, not lifetime, broadened. A complete conclusion would also check voltage noise and the intrinsic excitation spacing.

7. Classifying a blockade

For each observation, name the leading mechanism and one discriminating test:

1. A conductance dip through one tunnel junction deepens when the series environmental impedance is increased.
2. A double dot carries current for one bias polarity but becomes trapped in a triplet configuration for the opposite polarity.
3. A disordered insulator develops a soft suppression of tunneling density of states near the Fermi level.
4. A gated island shows periodic peaks and closed diamonds.

Solution

1. The leading interpretation is dynamical Coulomb blockade. Varying the impedance spectrum or measuring environmental sidebands tests energy exchange with the circuit.
2. This is Pauli spin blockade. Magnetic-field dependence, spin relaxation, or spin-resonance lifting distinguishes the spin selection rule.
3. This is a Coulomb gap. Its density-of-states scaling and dependence on disorder and dimensionality distinguish it from an island charging diagram.
4. This is ordinary island Coulomb blockade. The one-electron gate period, capacitance-consistent slopes, and thermal activation provide quantitative checks.

The shared word blockade describes suppressed response, not a shared microscopic Hamiltonian.

8. What blockade proves for a qubit

A semiconductor device displays a clean one-electron stability region and a charge-sensor signal for loading one electron. Which qubit claims are supported, and what remains to be demonstrated?

Solution

The data support controlled charge occupation, an estimate of addition energies, and a working charge-sensitive measurement channel. They may also identify a gate region suitable for isolating a spin.

They do not by themselves establish a qubit. One must still identify the encoded two-level subspace, demonstrate state preparation and calibrated control, measure state-selective readout, quantify relaxation and coherence, and report fidelity or error evidence appropriate to the claimed operation. Coulomb blockade supplies charge-sector infrastructure; it is not a coherence experiment.

Connections

• Quantum Dots develops confinement, shell filling, orbital contributions to addition energies, and spectroscopy beyond the metallic-island model.
• Single-Electron Devices turns the blockade ledger into electrometers, event-resolved detectors, pumps, turnstiles, and uncertainty-aware current standards.
• Mesoscopic Transport derives sequential master equations, stationary current, shot noise, and full counting statistics.
• Quantum Point Contacts treats a nearby constriction as a charge detector, including responsivity, bandwidth, and backaction.
• Conductance Quantization gives the transmission-eigenvalue language needed as a tunnel contact becomes open.
• Quantum Coherence in Conductors separates phase coherence, escape, thermal averaging, and energy relaxation.
• Anderson Insulators distinguishes the distributed soft Coulomb gap and hopping network from discrete-island charging thresholds.
• Kondo Effect owns the correlated zero-bias resonance that can appear in an odd Coulomb valley.
• Anderson Impurity Model supplies the interacting-level model connecting Coulomb valleys, mixed valence, and local moments.
• Chemical Potential distinguishes reservoir electrochemical potentials from finite-system addition thresholds.
• Spin Qubits covers spin-to-charge conversion, Pauli blockade, coherent control, and readout evidence.
• Condensed Matter Roadmap places charging phenomena after quantum statistics and transport and before device-specific qubit engineering.

References

1. D. V. Averin and K. K. Likharev, “Coulomb Blockade of Single-Electron Tunneling, and Coherent Oscillations in Small Tunnel Junctions,” Journal of Low Temperature Physics 62, 345–373 (1986), doi:10.1007/BF00683469.
2. T. A. Fulton and G. J. Dolan, “Observation of Single-Electron Charging Effects in Small Tunnel Junctions,” Physical Review Letters 59, 109–112 (1987), doi:10.1103/PhysRevLett.59.109.
3. U. Meirav, M. A. Kastner, and S. J. Wind, “Single-Electron Charging and Periodic Conductance Resonances in GaAs Nanostructures,” Physical Review Letters 65, 771–774 (1990), doi:10.1103/PhysRevLett.65.771.
4. C. W. J. Beenakker, “Theory of Coulomb-Blockade Oscillations in the Conductance of a Quantum Dot,” Physical Review B 44, 1646–1656 (1991), doi:10.1103/PhysRevB.44.1646.
5. M. A. Kastner, “The Single-Electron Transistor,” Reviews of Modern Physics 64, 849–858 (1992), doi:10.1103/RevModPhys.64.849.
6. H. Grabert and M. H. Devoret, eds., Single Charge Tunneling: Coulomb Blockade Phenomena in Nanostructures, NATO ASI Series B, vol. 294, Plenum Press, 1992, doi:10.1007/978-1-4757-2166-9.
7. G.-L. Ingold and Yu. V. Nazarov, “Charge Tunneling Rates in Ultrasmall Junctions,” in Single Charge Tunneling, pp. 21–107, Plenum Press, 1992, doi:10.1007/978-1-4757-2166-9_2.
8. H. van Houten, C. W. J. Beenakker, and A. A. M. Staring, “Coulomb-Blockade Oscillations in Semiconductor Nanostructures,” in Single Charge Tunneling, pp. 167–216, Plenum Press, 1992, doi:10.1007/978-1-4757-2166-9_5.
9. D. V. Averin and Yu. V. Nazarov, “Macroscopic Quantum Tunneling of Charge and Co-Tunneling,” in Single Charge Tunneling, pp. 217–247, Plenum Press, 1992, doi:10.1007/978-1-4757-2166-9_6.
10. K. K. Likharev, “Single-Electron Devices and Their Applications,” Proceedings of the IEEE 87, 606–632 (1999), doi:10.1109/5.752518.
11. M. H. Devoret, D. Esteve, H. Grabert, G.-L. Ingold, H. Pothier, and C. Urbina, “Effect of the Electromagnetic Environment on the Coulomb Blockade in Ultrasmall Tunnel Junctions,” Physical Review Letters 64, 1824–1827 (1990), doi:10.1103/PhysRevLett.64.1824.
12. Y. V. Nazarov, “Coulomb Blockade without Tunnel Junctions,” Physical Review Letters 82, 1245–1248 (1999), doi:10.1103/PhysRevLett.82.1245.
13. W. G. van der Wiel, S. De Franceschi, J. M. Elzerman, T. Fujisawa, S. Tarucha, and L. P. Kouwenhoven, “Electron Transport through Double Quantum Dots,” Reviews of Modern Physics 75, 1–22 (2002), doi:10.1103/RevModPhys.75.1.
14. R. Hanson, L. P. Kouwenhoven, J. R. Petta, S. Tarucha, and L. M. K. Vandersypen, “Spins in Few-Electron Quantum Dots,” Reviews of Modern Physics 79, 1217–1265 (2007), doi:10.1103/RevModPhys.79.1217.

Further reading

• C. W. J. Beenakker and H. van Houten, “Quantum Transport in Semiconductor Nanostructures,” Solid State Physics 44, 1–228 (1991), doi:10.1016/S0081-1947(08)60091-0.
• T. Ihn, Semiconductor Nanostructures: Quantum States and Electronic Transport, Oxford University Press, 2010, doi:10.1093/acprof:oso/9780199534425.001.0001.