Resonant Photoemission Spectroscopy as a Probe for Ultrafast Electron Transfer in Organic Semiconductors

Electron transfer (ET) is a fundamental process in chemical and physical systems. This thesis is concerned with ET in organic semiconductor systems, which have prominent applications as low cost solar cells and miniaturized, next generation electronics. Within these contexts, one of the emerging research questions concerns ET on the femtosecond (informally, the ultrafast) timescale. While prior results have connected the ultrafast timescales to the majority of delocalized electrons in a system are generated on these femtosecond timescales, an accurate understanding of these dynamics has proven elusive. This is because the popular probe of of electron dynamics, optical pump-probe experiments, can struggle with such short timescales.However, Resonant Photoemission Spectroscopy (RPES), a variant of X-ray absorption and photoemission spectroscopy, can probe ET rates at such ultrafast timescales (0.5 fs to 50 fs) with the added benefit of elemental selectivity and surface sensitivity. This technique is a stead-state experiment based on comparing the electron transfer rate to the core-hole decay rate, which is well known and generally unperturbed in low-Z atoms prevalent in organic semiconductors. The applications of this work concern boththe fundamental quantum mechanics at interfaces and the nature of the core-hole in the X-ray excited state as well as more applied work concerning the rational design of future organic semiconductors. This thesis leverages RPES to explore the question of ET at interfaces in the context of organic semiconductors.The first study involves between Copper(II) phthalocyanine (CuPc), a model planar organometallic compound, and its fluorinated analog Copper(II) hexadecafluorophthalocyanine (F 16 ) and examines the role of local electronic environment on ET rates. Fluorine was chosen as the fluorine atom is highly electronegative and thus a large perturbation to the electronic structure while simultaneously being very small and thus having a minor impact on the atomic structure. Additionally, fluorination of organic semiconductors is a common tool in modifying the electronic structure by lowering the HOMO and LUMO energies, reducing the degradative oxidation processes, and often yields p-type semiconductors. Experimental RPES results show electron transfer on the nitrogen K-edge to be faster for CuPc over F 16 by a factor of two while electron transfer on the carbon K-edge showed F 16 to be faster than CuPc. DFT results show a almost no modification of the electronic structure but a large shift in the energy levels of the fluorine bound carbons. An examination of the X-ray diffraction pattern and the corresponding literature results in two distinct crystal structures. This implies the difference in ET is connected to the in-column stacking and neighboring column stacking. The second study explores the size dependence of a molecule by examining a series of 4-,5-, and 6- molecule long thiophene molecules. The role of molecular size has been throughly explored before, though typically in the context of polymer systems and not in the context of ultrafast dynamics. The thiophene monomer is a well-known baseline in the field with sexithiophene (6 monomers) being one of the most studied small molecules and the functionalized polymer molecule, poly(3-hexylthiophene-2,5-diyl), as one of the most studied polymers. The initial hypothesis argued the larger molecules would have more delocalized electrons which would, in turn, screen the core-hole from the excited electrons. This would have implied easier delocalization (faster rates) as the molecule grew in size. The alternative hypothesis reasoned that the core-hole was a strong localizing potential and the ET time would be constant as a function molecule length. Instead, the results indicated that 5-thiophene molecule had the fastest delocalization time, beyond the range of the RPES technique. The rational for this result is inconclusive but may be related to the odd number of monomer units, thus implyingsome type of odd-even effect in conjugation length.The third study involves a series of functionalized thiophenol molecules bound to a gold surface. The molecular system of choice is a fluorinated thiophenol and the experiment involves excitations on the fluorine K-edge. The premise of the experiment involves the electrons excited from the fluorine 1s orbital and, possibly, delocalizing across the benzene structure through the sulfur bond and into the gold substrate/continuum. These experiments are strongly reminiscent of break-junction STM experiments popularized in the molecular electronics community. In the first experiment, the location of the fluorine atom was moved from being meta-oriented with respect to the sulfur atom to being para-oriented relative to the sulfur atom. This experiment probed the destructive interference properties of meta-oriented electron transfer to contrast with the constructive interference properties of the para-oriented molecule. The results show a comparable electron transfer rate between the two molecule, implying an inability to probe the constructive/destructive properties via the RPE process. The second experiment examined a series of para-oriented thiophenol molecules which were functionalized with a chlorine atom or a methyl group at the meta-position. The purpose of this experiment was to understand the role of functionalization on ET from the excited molecule. In this experiment, the ET from all three molecules were roughly comparable ( 5 fs), implying a weak perturbation on the total electronic structure of the molecule/substrate system. In consultation with prior literature, this result like points to a strong Fermi-pinning effect, whereby the electronic structure of the (much larger) gold continuum forces the relative energy level of the molecule to be relatively constant across functionalization.Last, an incomplete study concerning the role of functionalization in various para-oriented on strongly coupled interface is presented. This project strongly mirrors the third and has strong relevance to molecular electronics. By systematically varying the identify of the para-oriented group (a fluoro, hydroxyl, amine, nitro and cyano groups) and exciting on the K-edge of the respective atom in the para-oriented group, one can examine the role of the side-chain on ET rates. Presented first is a compilation of experimental XAS for some of the molecules. Of particular interest is the compilation of DFT generated spectra and molecular orbital contour plots for all the molecules. The nitro group is shown to be extremely localizing and is likely not a good candidate for excitation on the side-chain K-edge. The hydroxyl group shows a strong resonant enhancement in a preliminary RPES spectra. Prior literature and theoretical results argue that the cyano group should also be a strong candidate, although binding may be a minor issue. Theory data supports the amine group as another strong candidate but may pose experimental difficulties as the amine group also can bond to the gold substrate competing with the sulfur bond