Charge transport and morphology in soft semiconductors
Semiconducting polymers have become an increasingly interesting class of materials for use in organic electronic devices. By combining favourable electronic properties with an ease of processing that arises due to their self-organization, they are especially attractive for low-cost, large-area applications such as displays, lighting, or photovoltaics. The interest in the use of polymer semiconductors in practical applications is mirrored by efforts of scientists from a range of different disciplines to understand the fundamental properties of these materials.
In light of this, it is surprising that currently the relationship between the chemical composition of the material and the charge mobility is not known and that even the qualitative description of the charge transport is actively debated1. The lack of a clear structure-property relationship is especially penalizing in organic electronics - whose main advantage is the ability of synthetic organic and polymer chemists to tailor-make materials for applications. Compared to solid state semiconductors such as silicon, the theoretical description of charge transport in soft semiconductors is complicated as the charge carrier motion is coupled to the nuclear dynamics on a number of timescales. This complicates the description of charge carrier mobility and invalidates the normal assumptions that charge carrier dynamics is either much faster or much slower than the nuclear motions.
Microstructure and charge transport in polymer semiconductors
The charge carrier motion in organic semiconductors is strongly influenced by nuclear degrees of freedom. In order to investigatethe link between the nuclear and electronic degrees of freedom I have recently combined atomistic molecular dynamics and quantum chemical to study archetypal polymer semiconductor, poly(3-hexylthiophene) (P3HT). Molecular dynamics simulations2 were to characterise the microstructure of P3HT in the microcrystalline state and to reveal the hidden disorder not resolved by experimental measurements. It is shown that the charge carriers become localized at long-lived traps. The existence of activated transport for very ordered polymer phases is explained, and the trapped states, postulated by many phenomenological models, are described with chemical detail3. We also studied the charge transfer integrals, which determine the charge carrier transfer rates and are among the key parameters for charge transport models were also calculated2. This work is currently being extended to study the effect of polymer steroregularity on the microstructure and charge transport and to a range of different polymer semiconductors.
Phase behaviour and morphology of polymer semiconductors
The morphology in polymer semiconductors over mesoscopic lengthscales (100-1000 nm) is now known to have a strong influence on the experimentally realisable charge mobility - with grain boundaries and interfaces in particular expected to be important in transistor and photovoltaic applications, but the effect of these is currently unknown. The effects of morphology and interfaces are especially for polymer blends and composites, which are of potential importance in solar cells. Due to the length and timescales involved, atomistic simulations are untenable; instead corse-grained models, more commonly used to study the rheological and viscoelastic properties of polymers, may be employed. Recently using such a model I have studied the relationship between molecular architecture and phase behaviour of so-called Hairy-Rod polymers (consisting of a rigid backbone and flexible sidechains), which is a common structural motif in polymer semiconductors. A range of phases were observed - lamellar, hexagonal, and inverted hexagonal - depending on the length and density of the flexible side-chains4.
- David L. Cheung and Alessandro Troisi, 'Modelling charge transport in organic semiconductors: from quantum dynamics to soft matter', Phys. Chem. Chem. Phys., 10, 38, 5941-5952 (2008) doi:10.1039/b807750a
- David L. Cheung, David P. McMahon, and Alessandro Troisi, `Computational study of structure and charge transfer parameters of low-molecular mass P3HT', J. Phys. Chem. B, 113, 28, 9393-9401 (2009) doi:10.1021/jp904057m
- David L. Cheung, David P. McMahon, and Alessandro Troisi, `A realistic description of the charge carrier wavefunction in microcrystalline polymer semiconductors', J. Am. Chem. Soc., accepted (2009) doi:10.1021/ja903843c
- David L. Cheung and Alessandro Troisi, `Molecular structure and phase behaviour of hairy-rod polymers', Phys. Chem. Chem. Phys., 11, 12, 2105-2112 (2009) doi:10.1039/b818428c
Nanoparticles at soft interfaces
The adhesion of solid particles to soft interfaces, such as the oil-water interface, has been long known (first observed over 100 years ago). Recently the potential of such interfaces to provide a template for the formation of dense nanoparticle structures, such as nanoparticle monolayers and membranes, has stimulated much experimental interest. The adhesion of nanoparticles to soft-interfaces may also be used to stabilised the formation of large-scale supramolecular structures, such as nanoparticle stabilised fluid droplets or so-called bijels, nanoparticle stabilised foams and gels. Despite this interest, there is still much unclear about the behaviour of nanoparticles at soft interfaces - due to the small length and timescales, such systems are experimentally challenging to study, which hampers the development of an accurate microscopic understanding.
One particularly important aspect of this microscopic picture is the stability of nanoparticles at interfaces and the nanoparticle-interface interaction. The stability of micron-sized particles is well described by macroscopic, continuum theories, in which the free energy is determined by changes in the interfacial and aurface areas. Despite being grounded in macroscopic properties, such as interfacial and line tensions, these are routinely used to estimate the stability of nanoparticles at interfaces. In order to test their applicability I used molecualr simulation to calcualte the nanoparticle-interface interaction for a nanoparticle at a model liquid interface1. The interaction found from simulation was both stronger and longer-ranged than the theoretical prediction - the difference arises due to the neglcet of capillary waves (fluctuations in the interface position) in the macroscopic theory. These capillary waves broaden the interface, giving an interfacial width comparable to the size of the nanoparticle; capillary waves may also give rise to bridging between the nanoparticle and interface, giving a much longer range than expected.
This has then been extended to Janus particles - nanoparticles with two faces (e.g. one hydrophobic face and one hydrophilic face). This was found to increase the stability of nanoparticles at the interface - with the stability increasing as the difference between the surface tensions on the two faces increases. The increase in stability is smaller than predicted by theory, which is due to the ability of the nanoparticles to rotate (which is neglected theoretically). Simulation movies (shown below) show that for low surface tension differences the nanoparticle rotates almost freely - on increasing the surface tension differences ,the rotation slows but the nanoparticle still has a lot of orientational freedom.
- David L. Cheung and Stefan A. F. Bon,`Interaction of nanoparticles with ideal liquid-liquid interfaces', Phys. Rev. Lett., 102, 6, 066103/1-4 (2009) doi:10.1103/PhysRevLett.102.066103
- David L. Cheung and Stefan A. F. Bon, `Stability of Janus nanoparticles at fluid interfaces', Soft Matter, 5, 20, 3969-3976 (2009)