Charge transport mechanisms

Charge transport mechanisms are theoretical models that aim to quantitatively describe the electric current flow through a given medium.

Theory

Crystalline solids and molecular solids are two opposite extreme cases of materials that exhibit substantially different transport mechanisms. While in atomic solids transport is intra-molecular, also known as band transport, in molecular solids the transport is inter-molecular, also known as hopping transport. The two different mechanisms result in different charge mobilities.

In disordered solids, disordered potentials result in weak localization effects (traps), which reduce the mean free path, and hence the mobility, of mobile charges. Carrier recombination also decreases mobility.

Comparison between band transport and hopping transport

Parameter	Band transport (ballistic transport)	Hopping transport
Examples	crystalline semiconductors	disordered solids, polycrystalline and amorphous semiconductors
Underlying mechanism	Delocalized molecular wavefunctions over the entire volume	Transition between localized sites via tunnelling (electrons) or overcoming potential barriers (ions)
Inter-site distance	Bond length (less than 1 nm)	Typically more than 1 nm
Mean free path	Larger than the inter-site distance	Inter-site distance
Mobility	Typically larger than 1 cm^2/Vs; independent of electric field; decreases with increasing temperature	Typically smaller than 0.01 cm^2/Vs; depends on electric field; increases with increasing temperature

Starting with Ohm's law and using the definition of conductivity, it is possible to derive the following common expression for current as a function of carrier mobility μ and applied electric field E:

${\displaystyle I=GV=\sigma {\frac {A}{\ell }}V=\sigma AE=en\mu AE}$

The relationship ${\displaystyle \sigma =en\mu }$ holds when the concentration of localized states is significantly higher than the concentration of charge carriers, and assuming that hopping events are independent from each other.

Generally, the carrier mobility μ depends on temperature T, on the applied electric field E, and the concentration of localized states N. Depending on the model, increased temperature may either increase or decrease carrier mobility, applied electric field can increase mobility by contributing to thermal ionization of trapped charges, and increased concentration of localized states increases the mobility as well. Charge transport in the same material may have to be described by different models, depending on the applied field and temperature.[1]

Concentration of localized states

Carrier mobility strongly depends on the concentration of localized states in a non-linear fashion.[2] In the case of nearest-neighbour hopping, which is the limit of low concentrations, the following expression can be fitted to the experimental results:[3]

${\displaystyle \mu \propto \exp \left(-{\frac {2}{\alpha N_{0}^{1/3}}}\right)}$

where ${\displaystyle N_{0}}$ is the concentration and ${\displaystyle \alpha }$ is the localization length of the localized states. This equation is characteristic of incoherent hopping transport, which takes place at low concentrations, where the limiting factor is the exponential decay of hopping probability with inter-site distance.[4]

Sometimes this relation is expressed for conductivity, rather than mobility:

${\displaystyle \sigma =\sigma _{0}\exp \left(-{\frac {\gamma }{\alpha N_{0}^{1/3}}}\right)}$

where ${\displaystyle N_{0}}$ is the concentration of randomly distributed sites, ${\displaystyle \sigma _{0}}$ is concentration independent, ${\displaystyle \alpha }$ is the localization radius, and ${\displaystyle \gamma }$ is a numerical coefficient.[4]

At high concentrations, a deviation from the nearest-neighbour model is observed, and variable-range hopping is used instead to describe transport. Variable range hopping can be used to describe disordered systems such as molecularly-doped polymers, low molecular weight glasses and conjugated polymers.[3] In the limit of very dilute systems, the nearest-neighbour dependence ${\displaystyle \ln \sigma \propto -\gamma \alpha ^{-1}N_{0}^{-1/3}}$ is valid, but only with ${\displaystyle \gamma \simeq 1.73}$.[3]

Temperature dependence

At low carrier densities, the Mott formula for temperature-dependent conductivity is used to describe hopping transport.[3] In variable hopping it is given by:

${\displaystyle \sigma =\sigma _{0}\exp \left[-\left({\frac {T_{0}}{T}}\right)^{\frac {1}{4}}\right]}$

where ${\displaystyle T_{0}}$ is a parameter signifying a characteristic temperature. For low temperatures, assuming a parabolic shape of the density of states near the Fermi level, the conductivity is given by:

${\displaystyle \sigma =\sigma _{0}\exp \left[-\left({\frac {{\tilde {T}}_{0}}{T}}\right)^{\frac {1}{2}}\right]}$

At high carrier densities, an Arrhenius dependence is observed:[3]

${\displaystyle \sigma =\sigma _{0}\exp \left(-{\frac {E_{a}}{k_{\text{B}}T}}\right)}$

In fact, the electrical conductivity of disordered materials under DC bias has a similar form for a large temperature range, also known as activated conduction:

${\displaystyle \sigma =\sigma _{0}\exp \left[-\left({\frac {E_{a}}{k_{\text{B}}T}}\right)^{\beta }\right]}$

Applied electric field

High electric fields cause an increase in the observed mobility:

${\displaystyle \mu \propto \exp \left({\sqrt {E}}\right)}$

It was shown that this relationship holds for a large range of field strengths.[5]

AC conductivity

The real and imaginary parts of the AC conductivity for a large range of disordered semiconductors has the following form:[6][7]

${\displaystyle \Re \sigma (\omega )=C\omega ^{s}}$

${\displaystyle \Im \sigma (\omega )=C\tan {\frac {\pi s}{2}}\omega ^{s}}$

where C is a constant and s is usually smaller than unity.[4]

In its original version[8][9] the random barrier model (RBM) for AC conductivity in disordered solids predicted

${\displaystyle \sigma (\omega )=\sigma _{0}{\frac {i\omega \tau }{\ln(1+i\omega \tau )}}\,.}$

Here ${\displaystyle \sigma _{0}}$is the DC conductivity and ${\displaystyle \tau }$ is the characteristic time (inverse frequency) of onset of AC conductivity. Based on the almost exact Alexander-Orbach conjecture for the harmonic dimension of the percolation cluster,[10] the following more accurate representation of the RBM AC conductivity was given in 2008[11]

${\displaystyle \ln {\tilde {\sigma }}=({i{\tilde {\omega }}}/{\tilde {\sigma }})^{2/3}}$

in which ${\displaystyle {\tilde {\sigma }}=\sigma (\omega )/\sigma _{0}}$and ${\displaystyle {\tilde {\omega }}}$ is a scaled frequency.

Ionic conduction

Similar to electron conduction, the electrical resistance of thin-film electrolytes depends on the applied electric field, such that when the thickness of the sample is reduced, the conductivity improves due to both the reduced thickness and the field-induced conductivity enhancement. The field dependence of the current density j through an ionic conductor, assuming a random walk model with independent ions under a periodic potential is given by:[12]

${\displaystyle j\propto \sinh \left({\frac {\alpha eE}{2k_{\text{B}}T}}\right)}$

where α is the inter-site separation.

Experimental determination of transport mechanisms

Characterization of transport properties requires fabricating a device and measuring its current-voltage characteristics. Devices for transport studies are typically fabricated by thin film deposition or break junctions. The dominant transport mechanism in a measured device can be determined by differential conductance analysis. In the differential form, the transport mechanism can be distinguished based on the voltage and temperature dependence of the current through the device.[13]

Electronic transport mechanisms[13]

Transport mechanism	Effect of electric field	Functional form	Differential form
Fowler-Nordheim tunneling (field emission)^a		${\displaystyle I=A_{\text{eff}}{\frac {e^{3}m}{8\pi hm\phi _{\text{B}}}}E^{2}\exp \left(-{\frac {8\pi {\sqrt {2m}}}{3he}}{\frac {\phi _{\text{B}}^{3/2}}{E}}\right)}$	${\displaystyle {\frac {\mathrm {d} \ln I}{\mathrm {d} V}}={\frac {1}{V}}+{\frac {4\pi }{3he}}{\sqrt {2m}}{\frac {\phi _{\text{B}}^{3/2}}{V^{2}}}}$
Thermionic emission^b	Lowers barrier height	${\displaystyle I=A_{\text{eff}}A^{\star }T^{2}\exp \left[-{\frac {e}{k_{\text{B}}T}}\left(\phi _{\text{B}}-{\sqrt {\frac {eE}{4\pi \epsilon _{0}\epsilon _{r}}}}\right)\right]}$	${\displaystyle {\frac {\mathrm {d} \ln I}{\mathrm {d} V}}={\frac {e}{4k_{\text{B}}T}}\left({\frac {e}{\pi \epsilon _{0}\epsilon _{r}}}\right)^{\frac {1}{2}}V^{-{\frac {1}{2}}}}$
Arrhenius equation^c		${\displaystyle I=e\mu nE\exp \left(-{\frac {E_{a}}{k_{\text{B}}T}}\right)}$	${\displaystyle {\frac {\mathrm {d} \ln I}{\mathrm {d} E}}={\frac {1}{E}}}$
Poole–Frenkel hopping	Assists thermal ionization of trapped charges	${\displaystyle I=e\mu nE\exp \left[-{\frac {e}{k_{\text{B}}T}}\left(\phi _{\text{B}}-{\sqrt {\frac {eE}{4\pi \epsilon _{0}\epsilon _{r}}}}\right)\right]}$	${\displaystyle {\frac {\mathrm {d} \ln I}{\mathrm {d} E}}={\frac {1}{E}}+{\frac {e}{4k_{\text{B}}T}}\left({\frac {e}{\pi \epsilon _{0}\epsilon _{r}}}\right)^{\frac {1}{2}}E^{-{\frac {1}{2}}}}$
Thermally-assisted tunneling^d		${\displaystyle I={\frac {V}{2\pi }}\left({\frac {k_{\text{B}}Tt}{2\pi }}\right)^{\frac {1}{2}}\exp \left(-{\frac {\phi }{k_{\text{B}}T}}+{\frac {V^{2}\Theta }{24\left(k_{\text{B}}T\right)^{3}}}\right)}$	${\displaystyle {\frac {\mathrm {d} \ln I}{\mathrm {d} V}}={\frac {1}{V}}+{\frac {V\Theta }{12\left(k_{\text{B}}T\right)^{3}}}}$

^a ${\displaystyle I}$ is the measured current, ${\displaystyle V}$ is the applied voltage, ${\displaystyle A_{\text{eff}}}$ is the effective contact area, ${\displaystyle h}$ is Planck's constant, ${\displaystyle \phi _{\text{B}}}$ is the barrier height, ${\displaystyle E}$ is the applied electric field, ${\displaystyle m}$ is the effective mass.

^b ${\displaystyle A^{\star }}$ is Richardson's constant, ${\displaystyle T}$ is the temperature, ${\displaystyle k_{\text{B}}}$ is Boltzmann's constant, ${\displaystyle \epsilon _{0}}$ and ${\displaystyle \epsilon _{r}}$ are the vacuum the relative permittivity, respectively.

^c ${\displaystyle E_{a}}$ is the activation energy.

^d ${\displaystyle t=t(y)}$ is an elliptical function; ${\displaystyle \Theta }$ is a function of ${\displaystyle t}$, the applied field and the barrier height.

It is common to express the mobility as a product of two terms, a field-independent term and a field-dependent term:

${\displaystyle \mu =\mu _{0}\exp \left(-{\frac {\phi }{k_{\text{B}}T}}\right)\exp \left({\frac {\beta {\sqrt {E}}}{k_{\text{B}}T}}\right)}$

where ${\displaystyle \phi }$ is the activation energy and β is model-dependent. For Poole–Frenkel hopping, for example,

${\displaystyle \beta _{\text{PF}}={\sqrt {\frac {e^{3}}{\pi \epsilon _{0}\epsilon _{r}}}}}$

Tunneling and thermionic emission are typically observed when the barrier height is low. Thermally-assisted tunneling is a "hybrid" mechanism that attempts to describe a range of simultaneous behaviours, from tunneling to thermionic emission.[14][15]

See also

• Electron transfer

Further reading

• Nevill Francis Mott; Edward A Davis (2 February 2012). Electronic Processes in Non-Crystalline Materials (2nd ed.). OUP Oxford. ISBN 978-0-19-102328-6.
• Sergei Baranovski, ed. (22 September 2006). Charge Transport in Disordered Solids with Applications in Electronics. Wiley. ISBN 978-0-470-09504-1.
• B.I. Shklovskii; A.L. Efros (9 November 2013). Electronic Properties of Doped Semiconductors. Solid-State Sciences. 45. Springer Science & Business Media. ISBN 978-3-662-02403-4.
• Harald Overhof; Peter Thomas (11 April 2006). Electronic Transport in Hydrogenated Amorphous Semiconductors. Springer Tracts in Modern Physics. 114. Springer Berlin Heidelberg. ISBN 978-3-540-45948-4.
• Martin Pope; Charles E. Swenberg (1999). Electronic Processes in Organic Crystals and Polymers. Oxford University Press. ISBN 978-0-19-512963-2.

References

1. Bof Bufon, Carlos C.; Vervacke, Céline; Thurmer, Dominic J.; Fronk, Michael; Salvan, Georgeta; Lindner, Susi; Knupfer, Martin; Zahn, Dietrich R. T.; Schmidt, Oliver G. (2014). "Determination of the Charge Transport Mechanisms in Ultrathin Copper Phthalocyanine Vertical Heterojunctions". The Journal of Physical Chemistry C. 118 (14): 7272–7279. doi:10.1021/jp409617r. ISSN 1932-7447.
2. Gill, W. D. (1972). "Drift mobilities in amorphous charge‐transfer complexes of trinitrofluorenone and poly‐n‐vinylcarbazole". Journal of Applied Physics. 43 (12): 5033–5040. doi:10.1063/1.1661065. ISSN 0021-8979.
3. Sergei Baranovski; Oleg Rubel (14 August 2006). "Description of Charge Transport in Disordered Organic Materials". In Sergei Baranovski (ed.). Charge Transport in Disordered Solids with Applications in Electronics. Materials for Electronic & Optoelectronic Applications. John Wiley & Sons. pp. 221–266. ISBN 978-0-470-09505-8.
4. Sergei Baranovski; Oleg Rubel (14 August 2006). "Description of Charge Transport in Amorphous Semiconductors". In Sergei Baranovski (ed.). Charge Transport in Disordered Solids with Applications in Electronics. Materials for Electronic & Optoelectronic Applications. John Wiley & Sons. pp. 49–96. ISBN 978-0-470-09505-8.
5. Van der Auweraer, Mark; De Schryver, Frans C.; Borsenberger, Paul M.; Bässler, Heinz (1994). "Disorder in Charge Transport in doped polymers". Advanced Materials. 6 (3): 199–213. doi:10.1002/adma.19940060304. ISSN 0935-9648.
6. Jonscher, A. K. (June 1977). "The 'universal' dielectric response". Nature. 267 (5613): 673–679. doi:10.1038/267673a0. ISSN 0028-0836.
7. Igor Zvyagin (14 August 2006). "AC Hopping Transport in Disordered Materials". In Sergei Baranovski (ed.). Charge Transport in Disordered Solids with Applications in Electronics. Materials for Electronic & Optoelectronic Applications. John Wiley & Sons. pp. 339–377. ISBN 978-0-470-09505-8.
8. Dyre, Jeppe C. (1988). "The random free‐energy barrier model for ac conduction in disordered solids". Journal of Applied Physics. 64 (5): 2456–2468. doi:10.1063/1.341681. ISSN 0021-8979.
9. Dyre, Jeppe C.; Schrøder, Thomas B. (2000). "Universality of ac conduction in disordered solids". Reviews of Modern Physics. 72 (3): 873–892. doi:10.1103/RevModPhys.72.873. ISSN 0034-6861.
10. Alexander, S.; Orbach, R. (1982). "Density of states on fractals : " fractons "". Journal de Physique Lettres. 43 (17): 625–631. doi:10.1051/jphyslet:019820043017062500. ISSN 0302-072X.
11. Schrøder, Thomas B.; Dyre, Jeppe C. (2008). "ac Hopping Conduction at Extreme Disorder Takes Place on the Percolating Cluster". Physical Review Letters. 101 (2): 025901. doi:10.1103/PhysRevLett.101.025901.
12. Bernhard Roling (14 August 2006). "Mechanisms of Ion Transport in Amorphous and Nanostructured Materials". In Sergei Baranovski (ed.). Charge Transport in Disordered Solids with Applications in Electronics. Materials for Electronic & Optoelectronic Applications. John Wiley & Sons. pp. 379–401. ISBN 978-0-470-09505-8.
13. Conklin, David; Nanayakkara, Sanjini; Park, Tae-Hong; Lagadec, Marie F.; Stecher, Joshua T.; Therien, Michael J.; Bonnell, Dawn A. (2012). "Electronic Transport in Porphyrin Supermolecule-Gold Nanoparticle Assemblies". Nano Letters. 12 (5): 2414–2419. doi:10.1021/nl300400a. ISSN 1530-6984. PMID 22545580.
14. Murphy, E. L.; Good, R. H. (1956). "Thermionic Emission, Field Emission, and the Transition Region". Physical Review. 102 (6): 1464–1473. doi:10.1103/PhysRev.102.1464. ISSN 0031-899X.
15. Polanco, J. I.; Roberts, G. G. (1972). "Thermally assisted tunnelling in dielectric films (II)". Physica Status Solidi A. 13 (2): 603–606. doi:10.1002/pssa.2210130231. ISSN 0031-8965.