advanced36 min studyLesson 518 in Physics

Resolve the charging energy of an island

Use a constant-interaction model to locate charge degeneracies, gate periods, and the limits of sequential tunneling.

On this page

What you’ll understand

  • Distinguish the charging coefficient from the addition-energy spacing.
  • Find gate-controlled degeneracies between integer charge states.
  • State the temperature and weak-coupling conditions for Coulomb blockade.

A small island connected through weak tunnel barriers cannot generally accept charge continuously. Moving one electron changes its electrostatic energy by an amount that can exceed thermal fluctuations. Gate voltage shifts the relative energies of integer charge states, producing conductance peaks when adjacent occupations become accessible. This charging phenomenon is distinct from confinement-level spacing, although real quantum dots exhibit both and can reveal their combined structure in an addition spectrum.

Declare the charging-energy convention

Let the island's total capacitance to gates and leads be CΣC_\Sigma. In the constant-interaction model, the electrostatic part of its energy is a parabola EC(N−ng)2E_C(N-n_g)^2, where N is integer excess electron number and ngn_g is the dimensionless charge induced by gates and fixed offsets. We define EC=e2/(2CΣ)E_C=e^2/(2C_\Sigma). Some references instead call e2/CΣe^2/C_\Sigma the charging energy, so a quoted numerical value needs its defining formula before comparisons can be made.

Single-particle confinement energies can be added to this electrostatic term. The energy needed to add the Nth electron is a difference of total energies, not the charging coefficient itself. With constant orbital energies, consecutive addition energies differ by 2EC2E_C plus the relevant single-particle spacing. Spin multiplicity can let two electrons occupy the same orbital, but it does not remove the electrostatic cost of adding the second charge. Exchange and correlation may require a more detailed model.

E(N)=EC(N−ng)2+∑j=1Nϵj,EC=e22CΣ,μN=E(N)−E(N−1)=ϵN+EC[2(N−ng)−1]E(N)=E_C(N-n_g)^2+\sum_{j=1}^{N}\epsilon_j,\qquad E_C=\frac{e^2}{2C_\Sigma},\qquad \mu_N=E(N)-E(N-1)=\epsilon_N+E_C[2(N-n_g)-1]
N counts excess electrons relative to a chosen reference; ε_j are single-particle energies in a consistent reservoir reference. E_C is the quadratic coefficient, while 2E_C is the purely electrostatic spacing between successive addition energies. Capacitance is in farads and all displayed energies must share units.↔ Scroll wide equations horizontally.

Move between neighboring charge parabolas with a gate

First neglect single-particle spacings, as appropriate for a metallic island or a charging-dominated approximation. Neighboring states N and N plus one have equal energy at ng=N+1/2n_g=N+1/2. Away from that crossing, one integer minimizes the energy. Near a crossing, weak tunneling can alternate occupations and carry current if both lead transitions are energetically allowed. The equality is relative to the reservoir electrochemical reference; a finite bias opens a window and shifts the relevant alignment conditions.

For a gate capacitance CgC_g and otherwise fixed voltages, increasing gate voltage by e/Cge/C_g shifts induced charge by one electron. The lever arm α=Cg/CΣ\alpha=C_g/C_\Sigma converts a small gate-voltage change into an addition-energy shift of magnitude eαΔVge\alpha\Delta V_g. When orbital spacings are important, peak separations can depart from the simple metallic period. A measured finite-bias blockade threshold also depends on source and drain capacitances and voltage division, so it cannot universally be identified with EC/eE_C/e.

ng=CgVg+Q0e,ΔVg=eCg,α=CgCΣ,kBT,ℏΓ≪EC,Rt≫he2n_g=\frac{C_gV_g+Q_0}{e},\qquad \Delta V_g=\frac e{C_g},\qquad \alpha=\frac{C_g}{C_\Sigma},\qquad k_BT,\hbar\Gamma\ll E_C,\quad R_t\gg\frac h{e^2}
The period is for the charging-only model with other voltages fixed; Q₀ is an offset charge. Γ is an escape-rate scale, and R_t denotes each weak tunnel junction resistance. The inequalities express well-resolved charging and weak coupling. Resolving orbital levels additionally requires temperature and broadening below their spacings.↔ Scroll wide equations horizontally.

Identify what the blockade approximation leaves out

A sequential-tunneling description treats charge changes as separate weak transition events and uses probabilities for the island's occupations. Large tunnel resistances suppress charge uncertainty from coupling to the reservoirs, while low temperature suppresses thermal access to unfavorable states. Neither condition guarantees exactly zero current between peaks: thermally activated transport and higher-order cotunneling can remain. Strong coupling can wash out charge quantization and bring in correlations such as Kondo physics that the classical charging parabola does not describe.

The distinction between a gate-induced continuous offset and integer electron number is central. A gate changes electrostatic preference continuously without forcing the actual isolated island to hold a fractional electron count in a particular charge eigenstate. Near degeneracy the mean occupation can be fractional because probabilities or coherent superpositions involve adjacent integer states, depending on coupling. The elementary energy diagram determines which states are favorable; transition rates and reservoir occupations are still needed to calculate a conductance peak's height and width.

A small island with a measurable gate period

Neglect orbital spacings and take CΣ=40 aFC_\Sigma=40\,\mathrm{aF}, Cg=8 aFC_g=8\,\mathrm{aF}, and e=1.6×10−19 Ce=1.6\times10^{-19}\,\mathrm C. Find ECE_C, the charging-only addition spacing, gate period, and lever arm. At ng=0.25n_g=0.25, compare charge states N = 0 and N = 1.

  1. Since one attofarad is 10−18 F10^{-18}\,\mathrm F, EC=e2/(2CΣ)=3.2×10−22 J=2.0 meVE_C=e^2/(2C_\Sigma)=3.2\times10^{-22}\,\mathrm J=2.0\,\mathrm{meV}. The spacing between successive electrostatic addition energies is twice that, 4.0 meV4.0\,\mathrm{meV}.

  2. The gate period is e/Cg=0.020 V=20 mVe/C_g=0.020\,\mathrm V=20\,\mathrm{mV}, while α=Cg/CΣ=0.20\alpha=C_g/C_\Sigma=0.20. These are consistent because moving through one period shifts an addition energy by eα(20 mV)=4 meVe\alpha(20\,\mathrm{mV})=4\,\mathrm{meV}.

  3. The two energies are E(0)=2(0.25)2=0.125 meVE(0)=2(0.25)^2=0.125\,\mathrm{meV} and E(1)=2(0.75)2=1.125 meVE(1)=2(0.75)^2=1.125\,\mathrm{meV}. Adding the first excess electron costs 1.0 meV relative to the chosen zero reservoir reference, so N = 0 is favored.

The result

The charging coefficient is 2.0 meV, addition spacing 4.0 meV, gate period 20 mV, and lever arm 0.20. At induced charge 0.25, the zero-excess-electron state is lower than the one-electron state by 1.0 meV.

Explore the model

Interactive lab

Interactive lab · function

Compare distances from a charge minimum

Reset and set both controls to coefficient a = 1 and position x = 0.25; predict y = 0.0625. Reset and set both controls to a = 1 and x = 0.75; predict y = 0.5625. Interpreting x as the magnitude of N − n_g illustrates normalized electrostatic energies. The lab displays one continuous parabola, not a set of integer charge states, gate offsets, tunneling rates, or blockade diamonds.

Compare distances from a charge minimumThis lab uses only y = ax². Playback traces the curve within the fixed integration interval from 0 to x; it does not estimate numerical error or change the chosen point. Current readings: y at chosen x: 0.608; Tangent slope · 2ax: 1.35; Signed area · ax³/3: 0.182. All parameters are adjustable with the labeled controls below.−2.4−1.201.22.4−9−4.504.59xy(x, ax²)
y = ax²TangentSigned integral
y at chosen x
0.608
Tangent slope · 2ax
1.35
Signed area · ax³/3
0.182
t = 0 s

This lab uses only y = ax². Playback traces the curve within the fixed integration interval from 0 to x; it does not estimate numerical error or change the chosen point. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forQuadratic charging energy must be evaluated at the allowed integer occupations; the continuous graph alone does not quantize charge.

Check your understanding

With E_C = e²/(2CΣ) and negligible orbital spacing, what is the purely electrostatic spacing between consecutive addition energies?

Try it yourself

For the same charging-only island, compare N = 3 and N = 4 at ng=3.5n_g=3.5, then at ng=3.6n_g=3.6. Which state is favored after the change, by how much, and what gate-voltage increment causes it when other voltages stay fixed?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Evaluate the squared distance of each integer from n_g. A change of one unit in induced charge corresponds to one gate period.

Compare with the solution

At ng=3.5n_g=3.5, both energies are 2(0.5)2=0.50 meV2(0.5)^2=0.50\,\mathrm{meV}, so the states are degenerate. At 3.6, E(3)=2(0.6)2=0.72 meVE(3)=2(0.6)^2=0.72\,\mathrm{meV} and E(4)=2(0.4)2=0.32 meVE(4)=2(0.4)^2=0.32\,\mathrm{meV}. The four-electron state is lower by 0.40 meV. Increasing induced charge by 0.1 needs ΔVg=0.1(20 mV)=2.0 mV\Delta V_g=0.1(20\,\mathrm{mV})=2.0\,\mathrm{mV}. This energy comparison alone does not specify a tunneling rate or a measured peak current.

Take it with you

  • State the charging-energy convention before comparing energy scales.
  • Gate voltage shifts the relative energies of integer charge states.
  • Resolved Coulomb blockade requires low temperature and weak coupling; leakage and higher-order processes can remain.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.