Electronic structures of CuTPP and CuTPP(F) complexes. A combined experimental and theoretical study I

Abstract

Copper complexes of tetraphenylporphyrin (H2TPP) and tetrakis(pentafluorophenyl)porphyrin (H2TPP(F)) deposited as thin films on Au(111) have been studied experimentally and theoretically. Core level emissions from C 1s, N 1s, F 1s and Cu 2p as well as valence states of CuTPP and CuTPP(F) have been investigated using surface photoelectron spectroscopy. The interpretation of experimental results has been guided by theoretical calculations carried out on isolated species in the habit of the density functional theory. Reference to experimental and theoretical outcomes pertaining to H2TPP and H2TPP(F) allowed a confident and detailed assignment of the title molecules' X-ray and ultraviolet photoemission data. With specific reference to the latter, similar to copper phthalocyanine (CuPc), whose coordinative pocket mirrors the CuTPP/CuTPP(F) ones, the lowest ionization energy of the title compounds implies electron ejection from a ring orbital rather than from the Cu 3d-based singly occupied molecular orbital. Moreover, analogous to CuPc, the ionic contribution appears to play an important role in the Cu-N bonding. Nevertheless, differences in the number, symmetry, nature and relative position of CuTPP/CuTPP(F) occupied frontier orbitals compared to CuPc may be stated only by considering in great detail the Cu-ligand covalent interactions.

“Pigments of Life”, Molecules Well Suited to Investigate Metal–Ligand Symmetry‐Restricted Covalency

“…Experimental and theoretical studies[16b], [16c], [16d], carried out in our group on several Cu II complexes confirmed the legitimacy of exploiting the Cu II L 3 ‐edge position to estimate the ligand‐field strength exerted on the Cu II center. On the other hand, TD‐DFT SO‐ZORA TDA calculations on Cu II complexes having ligands with [PcCu, TPPCu, TPP(F)Cu, (acac) 2 Cu[16d]] or without ([CuCl 4 ] 2– ,[16d] the oxidized blue copper site in plastocyanin[16d]) low‐lying π* orbitals revealed the possibility of a Cu–L covalency underestimation if the Cu II L 3 ‐edge intensity of the former complexes is used as a gauge.…”

“…Cu II ions have a 3d 9 electronic configuration and, in the D 4 h Cu II complexes considered herein, the fivefold degeneracy of the 3d AOs is lifted by the square planar ligand field (see Figure , panels a and b); four low‐lying, double occupied MOs (DOMOs) of symmetry a 1g ( z 2 ), b 2g ( xy ) and e g ( xz, yz ), as well as a singly occupied MO (SOMO) of symmetry b 1g ( x 2 – y 2 ) are then generated , . All the Cu II 3d‐based MOs are Cu–N Py[55] antibonding in nature (see Figure );[16b], , the rather small and very similar Cu–N Py NM I in PcCu (0.26, see Table ) and TPPCu/TPP(F)Cu (0.27, see Table ) are therefore not particularly surprising. If we add that the Cu and N Py Hirshfeld gross atomic charges ( H Q Cu and H Q N , respectively) in PcCu, TPPCu and TPP(F)Cu amounts to 0.42, 0.40, 0.42 and –0.15 respectively (see Table ), it can be assessed with some confidence that ionic contributions play an important role in the Cu–N Py bonding.…”

“…If we add that the Cu and N Py Hirshfeld gross atomic charges ( H Q Cu and H Q N , respectively) in PcCu, TPPCu and TPP(F)Cu amounts to 0.42, 0.40, 0.42 and –0.15 respectively (see Table ), it can be assessed with some confidence that ionic contributions play an important role in the Cu–N Py bonding. [16b], …”

“…X‐ray absorption spectroscopy (XAS) is solidly recognized as a tool able to probe, element‐selectively, the empty frontier molecular orbitals (MOs), the M coordinative environment, as well as the nature and the strength of the M–ligand (M–L) interaction in M complexes . In this microreview, we focus on the comparison of homogeneous theoretical results, some of them recently published by our group,[16f] pertaining to the modelling of XAS outcomes recorded at the M L 2,3 ‐edge of a series of PcM‐based complexes (M = Cu, V;[16f], see Figure , panels a, c and d) as well as of TPPCu/TPP(F)Cu, (see Figure , panel b) deposited as films on coinage metals. Even though the Cu II L 2,3 ‐edge spectra of CuPc,[16c], [16d] TPPCu and TPP(F)Cu have been modelled in the past by exploiting a relativistic time‐dependent density functional theory based on the two‐component zeroth‐order regular approximation within the Tamm–Dancoff approximation (hereafter, TD‐DFT SO‐ZORA TDA), their ab initio simulation has been reconsidered herein to ensure a comparison between totally homogeneous theoretical outcomes (Supporting Information).…”

“…In fact, besides ligand‐field and M–L covalency effects, the spin–orbit (SO) coupling between the possible many final‐state multiplets needs to be taken into account . The M L 2,3 ‐edge spectra of PcCu, PcV, PcVO,, TPPCu[30b] and TPP(F)Cu[30b] reviewed herein have been simulated by using the current Restricted Open Shell Configuration Interaction with Singles (DFT/ROCIS) method,, , which has been widely employed to model M L 2,3 ‐edge spectra of mononuclear complexes,, [33j], 2D complex systems,[16e] clusters[33i] and protoporphyrins , . Moreover, according to a well‐tested procedure,, , the outcomes of L 2,3 ‐edge XAS measurements considered herein have been rationalized by assuming that PcM‐based and TPPCu/TPP(F)Cu deposits consist of weakly interacting molecules and by completely neglecting the adsorbate/substrate interaction.…”