186
T. Fujita
7.6 Concluding Remarks
In this chapter, we have presented our first-principle studies on the electronically
excited states of organic semiconductor materials. In particular, we have focused on
the effects of intermolecular interactions and molecular aggregation on the electronic
states. The development of the fragment-based GW method has enabled the accurate
determinations of energy levels of charged and electronically excited states and the
elucidation of the effects of polarization and delocalization effects. By characterizing
the electronic states of the PEN clusters in detail, we have illustrated the roles of
polarization and delocalization. Moreover, we have investigated the essential role
of the IP effect in the CT states in the PEN/C 60 interface structure. Further studies
on other D/A interface systems are expected to provide a deeper understanding of
excited states in an OSC.
We have presented the computation results within the Born-Oppenheimer approximation, i.e., electronic states obtained from the electronic Schrödinger equation
with a fixed nuclear geometry. However, interactions between electronic states
and a nuclear degree of freedom strongly influence optical and transport properties. In particular, organic semiconductor materials are characterized by a relatively
strong electron–vibration (phonon) interaction compared with inorganic materials.
For example, nuclear motion has a strong effect on charge-transport mechanics [110,
111]. The role of nuclear vibration in optical spectra has been intensively studied
within the Frenkel–Holstein model [102]. In the time-resolved spectroscopy studies
on OSCs [83], charge-separation dynamics have been detected as the formation of
polarons. However, the microscopic mechanism of the polaron formation, i.e., how
an electron or a hole is dressed by molecular vibration (phonon) with increasing the
e–h separation, remains unclear. Electron–vibration interactions result in considerable non-radiative decay to the ground state [9], limiting the open-circuit voltage.
From a computational standpoint, ab initio computations of electron-vibration [91] or
electron-phonon [49] interactions in molecular aggregates are more challenging than
single-point calculations, in which the electronic Schrödinger equation is solved with
a fixed nuclear position. Future theoretical developments in this direction are necessary to elucidate the roles of molecular vibration and phonons in the excited-state
properties.
Finally, the dynamical aspects of charge photogeneration should also be addressed.
Accordingly, an FMO calculation for an aggregate provides a model excited-state
Hamiltonian, which can be used to simulate real-time dynamics in combination with
quantum dynamic theories [37, 40–43, 59, 96]. We have already applied this strategy
to excited-state dynamics in the organic semiconductor thin film [42] and the D/A
interface [37]. Future developments of electronic structure methods and quantum
dynamic theories would provide further microscopic insights into the photophysical
processes in organic optoelectronic materials.
Acknowledgements T.F. thanks Prof. Yuji Mochizuki, Dr. Tatsuya Nakano, and Dr. Yoshio
Okiyama for their collaborations on the FMO-based excited-state calculations and their implementation in the ABINIT-MP program. T.F. also thanks Prof. Yoshifumi Noguchi at Shizuoka
T. Fujita
7.6 Concluding Remarks
In this chapter, we have presented our first-principle studies on the electronically
excited states of organic semiconductor materials. In particular, we have focused on
the effects of intermolecular interactions and molecular aggregation on the electronic
states. The development of the fragment-based GW method has enabled the accurate
determinations of energy levels of charged and electronically excited states and the
elucidation of the effects of polarization and delocalization effects. By characterizing
the electronic states of the PEN clusters in detail, we have illustrated the roles of
polarization and delocalization. Moreover, we have investigated the essential role
of the IP effect in the CT states in the PEN/C 60 interface structure. Further studies
on other D/A interface systems are expected to provide a deeper understanding of
excited states in an OSC.
We have presented the computation results within the Born-Oppenheimer approximation, i.e., electronic states obtained from the electronic Schrödinger equation
with a fixed nuclear geometry. However, interactions between electronic states
and a nuclear degree of freedom strongly influence optical and transport properties. In particular, organic semiconductor materials are characterized by a relatively
strong electron–vibration (phonon) interaction compared with inorganic materials.
For example, nuclear motion has a strong effect on charge-transport mechanics [110,
111]. The role of nuclear vibration in optical spectra has been intensively studied
within the Frenkel–Holstein model [102]. In the time-resolved spectroscopy studies
on OSCs [83], charge-separation dynamics have been detected as the formation of
polarons. However, the microscopic mechanism of the polaron formation, i.e., how
an electron or a hole is dressed by molecular vibration (phonon) with increasing the
e–h separation, remains unclear. Electron–vibration interactions result in considerable non-radiative decay to the ground state [9], limiting the open-circuit voltage.
From a computational standpoint, ab initio computations of electron-vibration [91] or
electron-phonon [49] interactions in molecular aggregates are more challenging than
single-point calculations, in which the electronic Schrödinger equation is solved with
a fixed nuclear position. Future theoretical developments in this direction are necessary to elucidate the roles of molecular vibration and phonons in the excited-state
properties.
Finally, the dynamical aspects of charge photogeneration should also be addressed.
Accordingly, an FMO calculation for an aggregate provides a model excited-state
Hamiltonian, which can be used to simulate real-time dynamics in combination with
quantum dynamic theories [37, 40–43, 59, 96]. We have already applied this strategy
to excited-state dynamics in the organic semiconductor thin film [42] and the D/A
interface [37]. Future developments of electronic structure methods and quantum
dynamic theories would provide further microscopic insights into the photophysical
processes in organic optoelectronic materials.
Acknowledgements T.F. thanks Prof. Yuji Mochizuki, Dr. Tatsuya Nakano, and Dr. Yoshio
Okiyama for their collaborations on the FMO-based excited-state calculations and their implementation in the ABINIT-MP program. T.F. also thanks Prof. Yoshifumi Noguchi at Shizuoka