Interdot Tunnel Coupling in Quantum Dot Devices

• Interdot tunnel coupling is the coherent quantum-mechanical interaction between localized electronic states in adjacent quantum dots, defined by the matrix element t_c.
• Experimental techniques such as spectroscopic analysis, time-resolved charge sensing, and microwave-assisted tunneling enable precise extraction and control of t_c.
• Optimizing tunnel coupling underpins quantum device performance by facilitating charge hybridization, spin exchange, and scalable quantum information processing.

Interdot tunnel coupling refers to the coherent quantum-mechanical coupling between localized electronic states in adjacent quantum dots, typically denoted as $t_c$ (or $t_{ij}$ for dots $i$ and $j$). In a double quantum dot (DQD) system, $t_c$ reflects the amplitude for an electron to tunnel directly from one dot to the other, and serves as a fundamental parameter governing charge hybridization, spin exchange, and the formation of molecular-like bonding/antibonding orbitals. The precise control and measurement of $t_c$ is a central aspect of quantum dot device physics, with implications for quantum information processing, charge/spin transport, and engineered quantum many-body systems.

1. Theoretical Framework: Two-Level Hamiltonian and Molecular States

The interdot tunnel coupling arises from an effective two-level Hamiltonian describing the subspace with $N$ electrons, in which a single excess electron can reside either in the left or right dot. The generic Hamiltonian is

$$H = \begin{pmatrix} E_L & t_c \ t_c & E_R \end{pmatrix}$$

where $E_L$, $E_R$ are the local electrochemical potentials of the left/right dot, and $t_{ij}$0 is the (real, positive) matrix element for tunneling. The detuning $t_{ij}$1 sets the energy bias between localized charge states. Diagonalization yields eigenenergies

$t_{ij}$2

with an anticrossing gap of $t_{ij}$3 at resonance ($t_{ij}$4) (Wang et al., 2010). The eigenstates are coherent molecular states: $t_{ij}$5 with mixing angle $t_{ij}$6.

This framework is directly applicable to a wide variety of quantum dot devices, including etched graphene dots (Wang et al., 2010), AlGaAs/GaAs heterostructures (Maisi et al., 2015), Si donor chains (Gorman et al., 2016), and InAsP nanowire quantum dot molecules (Phoenix et al., 2021); and extends naturally to triple/quadruple dot arrays with generalizations to higher-dimensional tight-binding Hamiltonians (Takakura et al., 2014).

2. Experimental Realization and Gate-Control Techniques

The tunnel coupling $t_{ij}$7 is engineered by the electrostatic profile of the interdot barrier, which is controlled via local gates situated between the dots. The dependence is typically exponential,

$t_{ij}$8

where $t_{ij}$9 is the voltage applied to the interdot barrier gate, and $i$0 encodes the sensitivity set by device geometry and screening (Lim et al., 2024, Hsiao et al., 2020, Nurizzo et al., 2022). In far-advanced MOS Si 2D arrays, tuning rates up to 30 decades/V have been demonstrated, enabling $i$1 to be swept over many orders of magnitude with <100 mV voltage changes (Lim et al., 2024). In graphene and GaAs devices, similar exponential behavior is observed, with $i$2 tuned from sub-$i$3eV to $i$4 meV (Wang et al., 2010, Nurizzo et al., 2022).

Gate architectures offering independent control—so-called "virtual barrier gates"—are obtained by calibrating and inverting the crosstalk matrix, allowing orthogonal adjustment of each $i$5 in an array (Hsiao et al., 2020). Modern approaches automate the calibration and feedback (Diepen et al., 2018), which is essential for scaling to large quantum dot registers.

3. Measurement Protocols and Extraction of $i$6

There are several complementary methods for extracting interdot tunnel coupling parameters:

• Spectroscopic Analysis of Charge Stability Diagrams: In honeycomb charge-stability maps, $i$7 is revealed by the anticrossing gap between resonance lines as gate voltages tune through charge degeneracy (Wang et al., 2010). The minimum splitting at triple points or vertices yields $i$8.
• Finite-Bias Resonance Fitting: In the sequential tunneling regime ($i$9; $j$0 is the dot-lead tunnel rate), the current as a function of detuning $j$1 follows a Lorentzian profile (Stoof-Nazarov model): $j$2 Fit parameters provide $j$3 in energy units (Fringes et al., 2011, Banszerus et al., 2019, Banszerus et al., 2020).
• Time-Resolved Charge Sensing: Using an SET or QPC as a charge sensor, one monitors single-electron tunneling events stochastically and extracts rates directly. In the regime $j$4, the width of the transition in unloading rate $j$5 is set by $j$6, even if $j$7 (Gorman et al., 2016).
• Photon- or Microwave-Assisted Tunneling: Under resonant drive, sidebands in charge detection or direct microwave spectroscopy reveal the energy difference between bonding/antibonding states, allowing direct extraction of $j$8 (Diepen et al., 2018, Borjans et al., 2021).
• Pulsed-Gate Lock-In: The decay rate of response as a function of pulse frequency (measured via an SET signal) yields $j$9; analysis of the frequency dependence enables precise quantification, particularly in MOS arrays (Lim et al., 2024).
• Optical Spectroscopy in Nanowires: For vertical quantum dot molecules, photoluminescence at cryogenic temperatures directly measures the energy gap $t_c$0 via s-shell transitions (Phoenix et al., 2021).

A summary table (values for $t_c$1 or its extraction in typical platforms):

System & Measured $t_c$2 range	Measurement Method
Graphene DQD (Wang et al., 2010)	$t_c$3 up to 0.72 meV
Bilayer graphene DQD (Fringes et al., 2011, Banszerus et al., 2020, Banszerus et al., 2019)	$t_c$4 from $t_c$51.5 $t_c$6eV (etched) to several GHz (AFM-gated)
SiMOS 2x2 Quan. Dot (Lim et al., 2024)	up to 30 decades/V tuning
Si donor chain (Gorman et al., 2016)	$t_c$7 = 2.2–5.5 GHz
GaAs triple dot (Nurizzo et al., 2022)	$t_c$8 Hz to $t_c$9 Hz
InP nanowire DQD (Phoenix et al., 2021)	$t_c$0 up to 24 meV

4. Physical Consequences: Molecular Regimes, Exchange, and Hybridization

The tunneling matrix element $t_c$1 defines the energy scale for hybridization between localized charge states. The following physical phenomena directly arise from its value or tuning:

• Formation of Molecular States: At finite $t_c$2, the system supports delocalized bonding/antibonding states, with their composition and splitting set by $t_c$3 and detuning $t_c$4 (Wang et al., 2010, Baines et al., 2012).
• Coherent Charge Oscillations and Exchange Coupling: In two-electron regimes, $t_c$5 mediates spin exchange, with exchange energy $t_c$6 for Hubbard-like onsite repulsion $t_c$7 (Lim et al., 2024). Fast two-qubit gating requires $t_c$8 in the GHz regime, while idling/quiescent states benefit from very small $t_c$9.
• Tunneling-Induced Transparency and Interference: In coupled dot molecules, tunnel coupling establishes coherent interference paths, observable as transparency windows in optical absorption when spin-conserving and spin-flip tunneling are both present (Borges et al., 2014).
• Spin-Related Effects: In materials with significant spin-orbit coupling, tunneling not only hybridizes charge but also allows spin-flip processes. The branching ratio $N$0 sets a quantitative limit for spin-preserving tunneling, controlled by the ratio $N$1 where $N$2 is the dot separation and $N$3 is the spin-orbit length (Maisi et al., 2015).
• Charge and Spin Pumping: $N$4 sets both the level splitting and, via quantum charge fluctuations, the regime in which charge or spin can be pumped adiabatically through the DQD (Riwar et al., 2010).
• Thermoelectric and Magneto-Transport Phenomena: The splitting of hybridized levels sets the resonant energies for transport; spin-dependent tunnel couplings can be used to engineer spin selectivity and optimize thermoelectric efficiency (Wang et al., 2012).

5. Tunability, Temporal Control, and Device Segmentation

A key enabling feature in modern devices is dynamic tunability of $N$5 over many orders of magnitude, both statically via gate voltages and dynamically via fast pulsed-gate protocols:

• Full Range: Experimental results demonstrate control from sub-Hz (for array segmentation and long-term isolation) up to several GHz (coherent evolution and fast exchange) with exponential sensitivity (Nurizzo et al., 2022).
• Segmentation and Modularity: By lowering $N$6 to the sub-Hz regime, quantum dot arrays can be segmented into independent submodules for readout or manipulation without disturbing neighboring elements (Nurizzo et al., 2022).
• Fast Switching Capability: High-bandwidth gating enables nanosecond switching between isolated and strongly coupled regimes, a critical capability for spin- and charge-transfer protocols.
• Orthogonal Control in Arrays: Calibration of "virtual barrier gate" axes based on measured crosstalk matrices allows selective control of each $N$7 in a larger array without mutual interference, an essential step in scaling quantum dot qubit platforms (Hsiao et al., 2020, Lim et al., 2024).

6. Material, Geometry, and Environmental Dependencies

Interdot tunnel coupling is not a universal constant but depends on detailed device parameters:

• Geometric Scaling: $N$8 decays rapidly with increasing interdot spacing, typically following an empirical inverse-cube law for nanowire vertical dots (Phoenix et al., 2021) or exponential with lithographic interdot gap in planar devices.
• Material Dependence: GaAs devices often support larger $N$9 due to higher dielectric constant and lighter mass, while Si and graphene devices can accommodate highly flexible control but may face additional valley structure complications (Borjans et al., 2021). In strained Si/SiGe wells, atomic-scale disorder induces spatial variation in both the magnitude and phase of intra- and inter-valley tunnel couplings (Borjans et al., 2021).
• External Fields: Magnetic fields can renormalize $$H = \begin{pmatrix} E_L & t_c \ t_c & E_R \end{pmatrix}$$0 via orbital effects, particularly in vertical stacks, providing an in-situ tuning knob albeit with slow and limited range (Phoenix et al., 2021). AC fields can be engineered to modulate $$H = \begin{pmatrix} E_L & t_c \ t_c & E_R \end{pmatrix}$$1 selectively for spectroscopic or dynamical purposes (Giavaras et al., 2020).

7. Relevance for Quantum Technologies and Future Directions

Precise, rapid, and high-fidelity control of interdot tunnel coupling is foundational to the operation of quantum dot-based quantum information processors and quantum simulators:

• Qubit Operation: Fast exchange gates, idling with negligible residual coupling, and robust transfer between modules all rely on controlled $$H = \begin{pmatrix} E_L & t_c \ t_c & E_R \end{pmatrix}$$2 (Lim et al., 2024).
• Array Scale-Up: Tunable $$H = \begin{pmatrix} E_L & t_c \ t_c & E_R \end{pmatrix}$$3 enables error correction via surface code tilings, design of modular quantum processors, and realization of analog simulation regimes.
• Noise and Fidelity Considerations: The need to minimize leakage and decoherence stemming from inappropriate hybridization or environmental fluctuations directly motivates both precise static tuning and dynamic control methodologies (Nurizzo et al., 2022).
• Spin and Valley Structure: In Si and graphene systems, control over both magnitude and character (including valley and spin selection) of tunneling is critical to avoid unintentional leakage, loss of spin/valley polarization, or forbidden transitions (Borjans et al., 2021, Maisi et al., 2015).

Comprehensive understanding and optimization of interdot tunnel coupling represent an intersection of quantum device engineering, condensed matter theory, and advanced measurement science, underpinned by a robust theoretical framework and increasingly sophisticated experimental strategies.