Variable Range Hopping in Disordered Materials

• Variable Range Hopping is a conduction mechanism in disordered materials where charge carriers use phonon-assisted tunneling between localized states, optimized over spatial and energy differences.
• It employs Mott's law and its Efros–Shklovskii extension to relate experimental parameters like localization length and density of states to observable temperature-dependent conductivity.
• Recent advances include integral modeling and memory effects analyses that refine predictions and clarify transitions in transport behavior across amorphous, molecular, and low-dimensional systems.

Variable Range Hopping (VRH) describes electronic transport in disordered materials where carriers are localized and conduction proceeds via phonon-assisted tunneling between spatially separated states. Unlike nearest-neighbor hopping, VRH involves a trade-off between tunneling distance and energy difference, resulting in temperature-dependent conductivity characterized by stretched exponential laws. The VRH framework provides a unifying description of transport in a wide range of amorphous, molecular, granular, and low-dimensional systems, and elucidates effects due to disorder, electron-electron interactions, and dimensionality.

1. Fundamental Principles and Universal Law

VRH is distinguished from activated hopping by the variable distance over which carriers hop. The rate for an electron (or hole) hopping from site $i$ to $j$, separated by $r_{ij}$ with energy mismatch $\epsilon_{ij}$, is

$$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$

where $\xi$ is the localization length. The DC conductivity is dominated by the "optimal" hops that minimize the total exponent, subject to the condition that available final states exist within the hop range and energy window.

The resulting Mott law for conductivity in $d$ dimensions is

$$\sigma(T) \;\simeq\; B(T) \exp\left[ - (T_0/T)^{1/(d+1)} \right]$$

where $T_0 = \beta_d / [\rho \xi^d ]$, $\rho$ is the single-particle density of states (DOS) at the Fermi level, and $j$0 is a numerical constant. The exponent $j$1 reflects optimization over both spatial and energy variables (Agam et al., 2014, Wu et al., 2014).

Parameters and Scaling

• $j$2: Localization length (nm scale in strong localization)
• $j$3: Density of localized states at $j$4
• $j$5: Characteristic temperature, inversely proportional to $j$6 and $j$7
• $j$8: Weakly temperature-dependent prefactor, often sub-leading (power-law)

2. Extensions: Coulomb Gap and Efros–Shklovskii VRH

In the presence of long-range unscreened Coulomb interactions, a soft gap opens in the single-particle DOS at $j$9 ("Coulomb gap"), leading to the Efros–Shklovskii (ES) law: $r_{ij}$0 where $r_{ij}$1 and $r_{ij}$2 (Huang et al., 2017, Joung et al., 2012, Shukla et al., 2021). This $r_{ij}$3 exponent is universal in any dimension under ES conditions. The VRH regime often exhibits a crossover from Mott ($r_{ij}$4) to ES ($r_{ij}$5) behavior as temperature decreases and interaction effects become dominant (Bennaceur et al., 2010, Huang et al., 2017).

3. Influence of Disorder, Dimensionality, and Energy-Dependent DOS

The VRH exponent $r_{ij}$6 can differ from canonical values if the DOS near $r_{ij}$7 is energy-dependent or if the transport is constrained (e.g., quasi-1D structures). For a DOS $r_{ij}$8, the hopping exponent generalizes to

$r_{ij}$9

in the VRH expression $\epsilon_{ij}$0 (Tsebro et al., 2022). As the dimensionality $\epsilon_{ij}$1 is reduced, or as $\epsilon_{ij}$2 increases (stronger pseudogap), the hopping law smoothly evolves towards nearest-neighbor hopping ($\epsilon_{ij}$3). In disordered carbon aerogels, variation from $\epsilon_{ij}$4 to $\epsilon_{ij}$5 is observed as the material transitions from 3D to quasi-1D hopping (Tsebro et al., 2022).

In graphene with covalently attached impurities, the impurity wavefunction decays as a power-law rather than exponentially. There, VRH conductivity follows a power law of temperature, $\epsilon_{ij}$6, with $\epsilon_{ij}$7 and $\epsilon_{ij}$8 the localization exponent (Liang et al., 2012).

4. Advanced Theoretical Refinements: Memory Effects and Percolation

A rigorous treatment reveals that occupation-number "memory" effects strongly suppress cross-Fermi-energy hops: transitions that create an electron-hole pair rarely contribute to DC transport due to near-immediate recombination. This splits the VRH percolation network into two weakly coupled subnetworks ("two-color percolation"). The main transport proceeds via (a) electron-like hops between normally empty sites and (b) hole-like hops between normally full sites; type (c) cross-chemical-potential hops are exponentially suppressed (Agam et al., 2014). The result is an additional subleading exponential correction to Mott's law,

$\epsilon_{ij}$9

where $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$0 is a universal exponent $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$1. This refinement can explain unexpectedly large extracted prefactors in experimental data (Agam et al., 2014).

5. Thermoelectric Response and Seebeck Coefficient in the VRH Regime

Standard Mott VRH predicts a Seebeck coefficient (thermopower) scaling with temperature as

$$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$2

from the Mott-Cutler formula. However, when localization length is energy-dependent, the refined scaling is

$$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$3

as predicted by Kubo-Luttinger linear response with Anderson localization scaling (Yamamoto et al., 2022). Experimental data for thiospinel CuCrTiS$$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$4 confirm this $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$5 behavior for $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$6 (Yamamoto et al., 2022). In CVD single-layer MoS$$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$7, simultaneous measurement of $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$8 and $$\Gamma_{ij}\;\sim\;\exp\!\bigl[ -2 r_{ij}/\xi\;-\;|\epsilon_{ij}|/T \bigr]$$9 verifies $\xi$0 for $\xi$1, with room-temperature $\xi$2 up to $\xi$3 mV/K (Wu et al., 2014).

6. Field-Driven and Nonlinear Regimes

In sufficiently large electric fields, the VRH process may become field-driven. The conductance crosses over from thermally-activated (Pollak–Riess) to field-driven (Shklovskii-type): $\xi$4

$\xi$5

The crossover field $\xi$6 separating these regimes is analytically determined from the two characteristic field scales $\xi$7 and $\xi$8; e.g., in 2D, $\xi$9 (Cheah et al., 2013).

Non-Ohmic (field-driven) VRH is also directly observed in reduced graphene oxide sheets, with field scaling consistent with ES physics: $d$0 (Joung et al., 2012).

7. Generalizations: Integral VRH and Transport Modeling in Disordered Systems

Conventional VRH theories rely on "optimal-hop" approximations, but integral VRH (IVRH) models provide a physically motivated formula for the full conductivity: $d$1 where $d$2 encodes system geometry (2D, 3D, multi-layer), and $d$3 is a dimensionless factor obtained from Monte Carlo validation (Qin et al., 15 Jan 2026). IVRH unifies the low-$d$4 (Mott) and high-$d$5 (Arrhenius) regimes without ad hoc regime separation, reproducing a smooth crossover and reducing fitting ambiguity in experimental data, with robust application demonstrated for monolayer MoS$d$6 and WS$d$7 (Qin et al., 15 Jan 2026).

8. Materials, Experimental Validation, and Defect Physics

VRH is ubiquitous in a wide range of systems, including:

• Polycrystalline ZnO, with a clear Mott$d$8ES crossover at cryogenic temperatures controlled by oxygen defects (Huang et al., 2017).
• Disordered organic semiconductors: energetically disordered systems require VRH models accounting for non-nearest-neighbor transitions; a critical ratio of inter-site distance to localization radius, $d$9, determines the regime (Upreti et al., 2021).
• Graphene and reduced graphene oxide: ES VRH dominates due to structural disorder and interaction-driven Coulomb gaps; measured localization lengths and extracted bandgaps validate theory (Joung et al., 2012).
• Pyrite (FeS$$\sigma(T) \;\simeq\; B(T) \exp\left[ - (T_0/T)^{1/(d+1)} \right]$$0) thin films: p-type conduction proceeds via defect-induced VRH associated with sulfur vacancy clusters (Shukla et al., 2021).
• Semiconductor nanocrystal arrays: quantum-confinement-induced donor-number fluctuations generate the disordered Coulomb landscape essential for ES-VRH, with marked crossover to activated transport as disorder is reduced (Skinner et al., 2012).

Negative magnetoresistance and weak-localization signatures co-exist with a variable $$\sigma(T) \;\simeq\; B(T) \exp\left[ - (T_0/T)^{1/(d+1)} \right]$$1 as the system transitions from VRH to nearest-neighbor hopping in quasi-one-dimensional disordered carbon aerogels (Tsebro et al., 2022).

9. Non-Standard Mechanisms and Environmental Coupling

Environmental effects, such as coupling to marginally localized phonons or proximity to metallic gates, modify VRH transport:

• In 1D, phonons with only a delocalized zero mode ("marginally localized phonon bath") suppress low-order phonon processes; VRH is dominated by exponentially rare many-phonon processes, resulting in a singular prefactor to the Mott law (Banerjee et al., 2015).
• Proximity to a metallic plate statically screens the Coulomb gap (restoring a flat DOS at low energy) and dynamically induces polaronic suppression of phonon-assisted hopping via coupling to an Ohmic bath of metal electrons. This introduces a multiplicative suppression factor in $$\sigma(T) \;\simeq\; B(T) \exp\left[ - (T_0/T)^{1/(d+1)} \right]$$2, notable over a wide intermediate temperature regime, and can reduce VRH conductivity by more than an order of magnitude without altering the underlying exponent (Asban et al., 2020).

---

Variable Range Hopping remains a powerful conceptual and quantitative framework for understanding electronic conduction in disordered and low-dimensional systems. Recent theoretical advances, rigorous experimental validation, and development of integral models continue to extend its applicability and resolve longstanding questions pertaining to interaction effects, scaling of thermoelectric properties, and the roles of geometry, dimensionality, and disorder.