20-09-2013, 03:13 PM
Charge carrier dynamics in polymer solar cells
carrier dynamics in polymer.pdf (Size: 1.47 MB / Downloads: 49)
Introduction
An introduction is given to the different types of polymer solar cells. The basic
operation of polymer bulk-heterojunction solar cells is described. The bulk-
heterojunction solar cells studied in this thesis are introduced and a short overview is
given of the charge recombination processes in bulk-heterojunction polymer solar
cells as studied by different techniques. The main issues concerning the
recombination processes addressed in this thesis are presented.
Polymer solar cells
Polymer solar cells are promising alternatives for conventional energy sources. Due to
the lower cost of manufacturing, easy processing and flexibility of polymers, solar
cells made of organic materials may compete with current inorganic solar cells.
Early polymer solar cells showed rather low solar power conversion efficiencies.1-
In organic materials, photoexcitation results in a strong Coulombically bound charge
pair (referred to as exciton). Without addition of electron donating/accepting dopants
or functional groups, the efficiency of charge carrier generation in polymers is low.
The energy needed for exciton dissociation comes available by relaxation from higher
excited singlet states to the lowest excited singlet state.5 However, this energy is
rapidly dissipated.
Polymer solar cells based on a combination of donor and acceptor materials,
showed a dramatic increase in efficiency.4,6-8 In such devices photogeneration of the
charge carriers occurs mainly by charge transfer at a donor–acceptor interface. Solar
cells consisting of a bilayer of donor and acceptor materials have a single donor–
acceptor interface across which the charge transfer step takes place. Provided that the
exciton diffusion length is larger than the distance to the donor–acceptor interface,
charge generation takes place with near unit efficiency.9 The disadvantage of this
architecture is that the optimal thicknesses of the donor and acceptor layer are equal to
the exciton diffusion length. Measured exciton diffusion lengths of various conjugated
polymers range from 5 to 14 nm.10 If the exciton decays in a time shorter than
necessary to diffuse to the interface, it will not contribute to the photogeneration of
charges.
Exciton generation and charge transfer
To achieve a high efficiency of charge generation, the active material in the solar cells
should absorb as large a part of the solar light as possible. Upon absorption of a
photon, an exciton is created in the active layer. The exciton diffuses to the interface
between the donor and acceptor materials where it is dissociated into charges by a
charge transfer reaction.
Organic molecular materials are characterized by low values for the relative
dielectric constant, typically εr = 3 to 4.30 The disordered arrangement of molecules or
chain segments in the donor and acceptor domains will limit the delocalization length
of the charge carriers to almost molecular dimensions. Therefore, one expects a strong
Coulombic binding between a photogenerated electron and hole at the donor–acceptor
interface. The electron and the hole may either undergo internal recombination
(geminate recombination) or dissociate into free charge carriers when the carriers
escape their mutual Coulombic attraction.
Charge transport and collection
In order to collect the photogenerated charges, the carriers have to migrate through the
active materials to the electrodes. The active layer in polymer solar cells is usually
deposited by spin-coating. In such a spin-coated film, the polymer chains are arranged
in a disordered fashion. Conformational and chemical defects in the polymer chains
and molecules will restrict the charge carriers to small segments, often referred to as
sites. As a result, the delocalization length of the charge carriers is limited to almost
molecular dimensions.
Recombination processes
After charge transfer, the charges may either escape their mutual Coulombic attraction
and can be collected, or they recombine with a charge carrier of opposite sign to the
ground state at the donor–acceptor interface. At the interface, the charges can either
recombine geminately or non-geminately.
In case of geminate recombination, the charge carriers within a single
photogenerated charge pair recombine. In this case, the recombination rate is expected
to be independent on excitation density, and the concentration of the generated charge
pairs depends linearly on excitation density (monomolecular decay).
In case of non-geminate recombination, the charge carriers from multiple charge
pairs recombine and the recombination kinetics is therefore strongly dependent on
excitation density. Non-geminate recombination is expected at high excitation
densities. In the absence of trap sites, non-geminate recombination is a bimolecular
reaction and occurs at a single reaction rate (non-dispersive).
Photoinduced absorption spectroscopy
The recombination dynamics of the photogenerated charges can be studied by
photoinduced absorption (PIA) spectroscopy. In a PIA measurement, the material
under investigation is excited via an allowed optical transition from the neutral ground
state. The change in transmission of a probe beam is monitored after excitation at the
wavelength at which the photoexcitations show an intense absorption band. In
polymer–fullerene blends, usually the bleaching or the absorption of the positively
charged photoexcitations (referred to as polarons or radical cations or holes) in the
polymer are probed, because of the weak absorption coefficients associated with the
charged states in the fullerene.
Early transient absorption measurements on the picosecond time scale on
polymer–fullerene blends showed the fast subpicosecond charge transfer 11,53-55 and a
much slower decay of the charge separated state.12,60 More recently, transient
nanosecond absorption measurements have been carried out,39,51,56,58 showing that the
dynamics of charge recombination extends into the millisecond time scale.
Transient optical spectroscopy on films of MDMO-PPVCBM at room
temperature show a strongly excitation intensity dependent decay of the positively
charged MDMO-PPV polarons, at time scales <20 ns, consistent with non-geminate
recombination. On time scales >300 ns a slow power-law type of decay of the polaron
concentration n following laser excitation according to n(t) = t–0.4 was found, with
decay and amplitude independent on incident laser intensity.56 This power-law type
decay is indicative of dispersive recombination dynamics of localized charge
carriers.57 Intensity independent decay is expected if the recombination is geminate.
However, at higher excitation densities one would expect non-geminate
recombination to be dominant. The intensity independent decay and amplitude cannot
be explained in simple terms of a non-geminate bimolecular recombination reaction.
Instead, the decay has been attributed to recombination, in which the dynamics is
governed by thermal activation of the photogenerated holes from a limited number of
traps and therefore shows an apparent monomolecular, intensity independent decay.56
At very low temperature measurements suggest that the recombination is mainly due
to tunelling.
Aim of the thesis
Studies on charge separation and recombination processes in polymer bulk-
heterojunction photovoltaic devices under actual working conditions have started only
recently, and are of major interest to the understanding of the principles of operation
of the devices. At present it is not fully understood why the separation of the
photogenerated charges at the bulk-heterojunction interface is so efficient. The charge
separation is likely to be affected by several factors, such as temperature, the electric
field present in the device, and disorder in the active materials. For many
combinations of donor and acceptor materials used in photovoltaic devices, the
mobilities of the electron in the accepting component and the hole in the donating
component are different. The influence of this difference in mobility on the
dissociation of geminate charge pairs at the interface has not yet been systematically
investigated. Because of the low mobility of the charges in disordered polymeric
materials, charge collection is expected to compete with charge recombination.
However, if the charge recombination dynamics is sufficiently slow, charge
recombination may not limit charge collection.