Role of Initial and Final States in Molecular Spectroscopies: Example of Tetraphenyldibenzoperiflanthene (DBP) on Graphite

Abstract

Electronic devices based on organic semiconductors are a fast-growing field of technology. A detailed understanding of the interplay between optical and electronic properties of organic molecular thin films as well as the physical structure is highly beneficial for the optimization of such devices. This requires investigations by means of different yet complementary spectroscopic techniques, where the comparability between the different data sets can become confusing. Consequently, the core aspect of this work is the consistent unification of spectroscopic data obtained by a diversity of experimental techniques within the complementary pictures of molecular states and energy levels. The confusion is obvious in the example of epitaxial films of tetraphenyldibenzoperiflanthene (DBP) on graphite(0001) surfaces being discussed here. One puzzling issue is the widening of the transport gap by around 0.6 eV during film growth, whereas the optical gap remains virtually constant. Furthermore, two-photon photoemission spectroscopy (2PPE) yields several features related to the same unoccupied orbital, and it is a nontrivial task to correlate them with the results of other spectroscopies. These issues can be resolved when all spectroscopic data are consistently interpreted in the framework of a theoretical model which was recently introduced by us, taking the initial and final states of the underlying probe processes into account. Applying that model, all peak shifts are explained here consistently by means of polarization and charging energies and further physical effects in context with the microscopic structure of the DBP thin films.

Recent Progress in the Consistent Interpretation of Complementary Spectroscopic Results Obtained on Molecular Systems

“…Moreover, all calculations yielded transport gaps (and an optical gap in the case of TD-DFT) somewhat smaller than the respective measured gaps. Yet, the difference between both gaps, E B = E trans − E opt , in our calculations for gas-phase pentacene [24] agrees almost perfectly with the experimental value, presumably due to a cancellation of computational errors.…”

“…Kirchhuebel et al [24] Gas-phase PEN f) [24] PEN in a polarizable medium g) 2.77 -----0.95 1.03 -Neaton et al [118] Face In Figure 12 , with L 0 and H 0 computed via DFT, and E opt computed via TD-DFT [17] ; c) For E opt , we used the experimental data of PEN/α-Al 2 O 3 (0001) in Figure 12; d) In Figure 12, we assumed P + = P − ; e) In Figure 12, we assumed w to have the same value for all levels. It can be obtained via w = EA(adsorbed) − EA(gas phase) − P − = −(IE(gas phase) − IE(adsorbed) − P + ); f) In our previous report, [24] only some of the quantities listed here were explicitly provided. The calculation result for the transition with E opt = 1.90 eV yielded a much smaller oscillator strength than for another transition at 4.24 eV.…”

“…Since is expected to be rather small, it is sometimes neglected for the sake of simplicity. [ 17,24 ] In addition, the terms w , P + , and P − apply to molecular films, but they are negligible for molecules in the gas phase.…”

“…[ 17 ] Further details are found in our previous work. [ 24 ] The computational results are summarized in Table 3 along with the main experimental results depicted in Figure 12 and additional theoretical studies.…”

“…In this section we lay out the concepts of drawing accurate energy level diagrams brought forward in the literature. [17,24] This model was developed for the often encountered case of weak molecule-substrate interaction, and excludes hybridization, or, in other words, fractional charge transfer.…”