Photoemission yield spectroscopy investigations of the electronic properties of copper phtalocyanine thin films UHV annealed and exposed to oxygen

Abstract

Introduction Considerable efforts have been made during the last two decades in order to understand the electronic properties of phthalocyanine (Pc) compounds, well-known organic materials with interesting semiconducting [ 1 to 31 and catalytic [4] properties. Because of both high temperature and high chemical stability [5], they became widely used in thin film technology for specific optoelectronic devices [6], solar cells [7], and gas sensors [8].As for the most popular copper phthalocyanine (CuPc) thin films, up to now only UPS [9 to 131, SPS [14, 151, and EELS [16] methods have been used to study their electronic properties. In the earliest UPS studies of Vilesov et al. [9], reexamined by Schechtman and Spicer [lo], only the optical density of states in the upper part of the valence band up to -7 eV has been determined. It was characteristic of the conjugated inner ring of the main molecule and found to be in rather weak agreement with the free-electron molecular orbital calculations [ll]. Thereafter, Pong and Smith [12], and Hoechst et al. [13] have studied the copper phthalocyanine (CuPc) photoemission spectra for the photon energy range of 7 to 23 eV, but only for the higher photon energies they observed the structure characteristic of the optical density of states. In addition, the electron mean free path for energies larger than 1.5 eV above the valence band was estimated to be about 1 nm [12]. The electronic states near the top of the valence band have also been observed in the surface photovoltage spectroscopy (SPS) experiments of Dahlberg and Musser [14, 151 as well as in electron loss spectroscopy (EELS) studies of Tada et al. [16].Because of the limited energy range and sensitivities of the above-mentioned conventional methods, many questions concerning the electronic properties of the copper phthalocyanine (CuPc) thin films still remain unanswered. Investigations using other techniques are expected to give new and complementary information.One of the best and suitable methods for this purpose seems to be photoemission yield spectroscopy (PYS), developed by the group of Sebenne in Paris [17]. It consists in measuring the total number of photoemitted electrons per incident photon at a given light energy E (usually up to 6.2 eV). Because of the large dynamic range (up to eight orders of magnitude) and high energy resolution (up to 0.02 eV), PYS allows the determination, among others, of the work function cp and the ionization energy @. Moreover, with the debatable, but

Chemistry and electronic properties of ferromagnetic metal‐organic semiconductor interfaces: Fe on CuPc

“…3(d). Four special points of curve in this figure – Φ Au , Φ CuPc , Φ FePc , and Φ Fe are typical characteristic work functions of the corresponding materials 37–39. From the fact that the work function curve between pristine CuPc and thick iron overlayer has some minimum we can suppose that, at low coverage of iron on the CuPc surface, the formation of some intermediate phase with work function Φ = 3.9 eV takes place, which is very close to the work function value Φ = 3.8 eV, measured for FePc thin films 28.…”

Orbital Mediated Tunneling in Vanadyl Phthalocyanine Observed in both Tunnel Diode and STM Environments

“…It is equally likely that this feature is associated with the density of states of the tip/substrate system since a similar (although less clear) rise in intensity occurs in the d I /d V of the VOPc free region of the surface (Figure C). At this time we have no assignment for the sharp increase in intensity at large negative bias, other than to note that it is probably due to some oxidative process and that it is almost certainly not the HOMO which should occur at least 0.5−1.0 eV higher in energy. − Clearly, however, this current enhancement involves some deeper occupied orbitals of the VOPc.…”

“…At this time we have no assignment for the sharp increase in intensity at large negative bias, other than to note that it is probably due to some oxidative process and that it is almost certainly not the HOMO which should occur at least 0.5-1.0 eV higher in energy. [24][25][26] Clearly, however, this current enhancement involves some deeper occupied orbitals of the VOPc. In comparing Figures 1 and 4, one may ask about the absence of any of the inelastic vibrational structure.…”

A Scanning Tunneling Microscopy and Spectroscopy Study of Vanadyl Phthalocyanine on Au(111): the Effect of Oxygen Binding and Orbital Mediated Tunneling on the Apparent Corrugation

“…31 The peak in dI/dV at -1.8 V is more difficult to assign. The literature abounds with assignments for the Pc ring band gap in phthalocyanines of the order of 2 eV, [32][33][34] placing the HOMO (2a 1u ) at about -0.6 V bias. Thus, on the basis of conventional assignments, the peak at -1.8 V must be due to a lower occupied orbital.…”

“…It should also be noted that in the O-down picture, the molecule does not see the full (2 eV bias scan. Instead, the scan range would be only about (1 V. This effective reduction by 2 in the potential would place the HOMO (about 2 eV below the LUMO [32][33][34] ) at the extreme edge, or perhaps outside, of the bias scan window. A final caution is that the analysis presented here is based on classical electrostatics and is certainly flawed.…”