LiF Doping of C60 Studied with X-ray Photoemission

M3116

ECS Journal of Solid State Science and Technology, 6 (6) M3116-M3121 (2017)

JSS FOCUS ISSUE ON NANOCARBONS—IN MEMORY OF SIR HARRY KROTO

LiF Doping of C60 Studied with X-ray Photoemission Shake-Up Analysis Ayse Turak,a,z Marek Z. Zgierski,b and M. W. C Dharma-Wardanab a Department of Engineering Physics, McMaster University, W. Hamilton, b National Research Council of Canada, Ottawa, ON K1A 0R6, Canada

ON L8S 4L7, Canada

We report our investigation of the chemical doping mechanism induced by LiF interaction with fullerene thin films. High resolution Xray photoelectron spectroscopy of the C1s shake-up satellites and F1s main core level, supported by density functional calculations, suggest the formation of a charge transfer complex between covalent LiF monomers and dimers and C60 . This interaction was observed in both LiF/C60 and C60 /LiF depositions, suggesting that some charge transfer complexation can occur in these systems even without dissociation. © The Author(s) 2017. Published by ECS. This is an open access article distributed under the terms of the Creative Commons Attribution Non-Commercial No Derivatives 4.0 License (CC BY-NC-ND, http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial reuse, distribution, and reproduction in any medium, provided the original work is not changed in any way and is properly cited. For permission for commercial reuse, please email: [email protected]. [DOI: 10.1149/2.021706jss] All rights reserved. Manuscript submitted December 12, 2016; revised manuscript received February 1, 2017. Published February 10, 2017. This paper is part of the JSS Focus Issue on Nanocarbons – In Memory of Sir Harry Kroto.

Buckminsterfullerene (C60 ) and its derivatives, such as [6,6]phenyl-C61-butyric acid methyl ester (PCBM), are among the most promising and widely used materials in a variety of nanotechnology related applications. Since its discovery by Harry Kroto and colleagues in 19851 , it has been explored for a wide variety of applications including lightweight high strength nanocomposites,2,3 optical filters,4 hydrogen storage materials,5 and even antibacterial treatments.6,7 Due to the rich electronic properties in the ground and excited states, C60 is most widely used as an organic semiconductor. With the ability to act as both an electron acceptor and donor, it is a standard material in organic light emitting diodes, solar cells, transistors and Schottky diodes. The interface with the metal contacts play a crucial role in the effectiveness of organic semiconductors in device performance and stability.8 Introducing a thin buffer layer of LiF between an organic film and an Al cathode has been seen to dramatically enhance the energy-level alignment and stability of the interface.9–13 Additionally, alternating stacks of LiF and C60 14 and LiF doped C60 nanocomposite buffer layers15,16 have been shown to have high conductivities and stability in OLEDs and OPVs. Several mechanisms have been proposed for these behaviors including chemical reactions, dipole alignment, tunneling, and LiF dissociation requiring the additional presence of Al.17 In this contribution, we present results showing that in the absence of Al, there is a charge transfer interaction between LiF and C60 , leading to the formation of LiF••C60 complexes without dissociation. Using the well-developed and described X-ray photoelectron shakeup structure of C60 and the strong feature from F1s XPS core level, we see evidence of a direction independent interaction. The spectroscopic features can be described by the semi-covalent interaction of LiF monomers or dimers with C60 using density functional theory. Coupled with XPS core level shift calculations, using approximate Madelung energy, the predicted Millikan charges and bond lengths give values that approximate the observed core level shifts. Additionally, the small transfer of charge to C60 predicted by the modelling is consistent with the minimal impact on the shake-up satellites of the C1s of C60 . Experimental The samples were produced using a Kurt J. Lesker OLED cluster attached to an X-ray photoelectron spectroscopy tool. For deposition of C60 , ∼5Å of LiF on 100 nm sputter deposited Au/Si was used z

E-mail: [email protected]

as a substrate. For deposition of LiF, the substrate was 350Å of C60 thermally deposited onto cleaned Si with native oxide. C60 and LiF were thermally evaporated from home built crucible sources at an average rate of 1 Å/min and 2–3 Å/hr as measured by oscillating quartz crystal microbalance (Inficon XTM/2) respectively onto previously prepared substrates. Base pressure in the system was 3 × 10−9 Torr. LiF and organic deposition occurred sequentially without breaking vacuum during analysis. The growth of C60 is described by the formation of monolayers (ML) rather than in Å since the growth appears to be layer-by-layer. For C60 , 1 ML is defined as 1.15 × 1014 molecules/cm2 for a close packed FCC structure, assuming a C60 diameter of 7 Å18 and a density of 1.65 g/cm3 .19 One sandwich sample consisting of glass/2000 nm SiO/60 nm Al/5 Å LiF/60 nm C60 /5 Å LiF/ 100 nm Al/300 nm SiO was produced by vacuum thermal evaporation as described previously20 and the buried interface examined using the peeling method.21 X-ray spectra were generated with monochromated Al Kα (1486.7 eV) radiation in a 90◦ geometry (Phi 5500 ESCA). The photoelectrons were analyzed by a hemispherical analyzer using 23.35 eV pass energy, with a nominal analysis area of 800 μm2 and sampling depth ∼5 nm. XPS spectra were calibrated using the Au 4f core level from the sputtered Au film, and aligned at 285 eV to account for charging shifts. Density functional calculations were performed on various possible arrangements of LiF with C60 with the Li or the F unit placed close to the hexagonal faces, pentagonal faces, or one of the C-C bonds in the C60 molecule. In each case, the atomic positions were geometry optimized by total energy minimization to ensure that they represent realistic structures. Structures were visualized using VESTA 3.22 Using the Gaussian-98 code,23 the exchange-correlation effects were treated using the BP86 functionals of Becke et al.24 where Gaussian basis functions of the 6–31G∗ type were used.24,25 Results and Discussion In the high-resolution X-ray photoelectron spectroscopy of atoms and molecules, excitation of photoelectrons with X-ray excitation occur on the scale of 10−17 s.26 Under such conditions, there is a finite probability that the irradiated material is simultaneously ionized and left in an excited state during photo-excitation due to the change in the potential of the molecule with the creation of a 1s core-hole.27,28 The XPS spectrum, therefore, is made up of an intense peak from the original ejected photoelectron, and a number of peaks on the high binding energy side of the main peak due to the energy loss for excitation of

Downloaded on 2017-03-10 to IP 181.215.123.19 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).


M3117

Figure 1. (a) High resolution spectrum of C 1s high binding energy satellite structure for C60 . Drop down lines represent the theoretically determined orbital transitions from Enkvist et al.32 all visible in the spectrum. (b) Molecular energy levels of C60 (neglecting core hole ionization) (after Refs. 50 and 32). The transitions that correspond to the observed features in the spectrum are the HOMO-LUMO transition between 5hu and 5t1u ∗ at 1.9eV, and the dipole transition from 6hg to the LUMO at 6.0 eV. The features at 3.8 and 4.8 eV cannot be assigned to a single transition, but represent the (5hu , 7hg , and 4gg ) → (5t1u ∗ , 5t2u ∗ , 8hg ∗ , and 5gu ∗ ) monopole and dipole transitions.

electrons to higher bound orbitals, referred to as shake-up satellites. Shake-up satellites are reflective of the valence band characteristics, since they are based on elastic loss processes within the orbital structure of the atom. The energies of these excited states after the loss of a photoelectron can be approximated by the ground-state energies of the molecule with promotion of a valence electron to an available empty state. If the wave function of the frozen core hole state (with no valence electron excitation) and that of the relaxed ionic ground state overlap, most of the intensity of photoemission will go into the main core level line, and the shake-up satellites will not be resolved. However, in some systems, such as conjugated organic molecules and transition metals oxides with partially filled d-orbitals, significant charge redistribution is possible to “screen” the core hole and stabilize the ion.29,30 Satellite features have been used for polymeric systems to identify changes in the conjugation when there is little obvious difference in the C1s core levels.26 Fullerenes are particularly well suited to such an examination, as the electronic structure of C60 can accommodate significant charge redistribution.31 Therefore, C60 has a particularly well differentiated and well described19,32,33 set of satellites, which aids in spectral interpretation. A high resolution scan of the shake-up features for a layer (∼17 ML) of C60 is shown in Figure 1. Lunell and Enkvist et al.32,34 were able to assign the observed peaks in an experimental spectrum to particular transitions in the Huckel molecular orbital structure, shown in Figure 1b, and the theoretically predicted peak center of mass are shown in red in Figure 1a. Almost all of the shake-up structures correspond to some transition to the 5t1u ∗ (LUMO) level, giving a clear picture of the valence band characteristics of C60 . The shake-up feature at 1.9 eV from the mainline of the core level corresponds to electron promotion from HOMO to LUMO.32,33 The sharp features at 3.8 and 4.8 eV correspond to a number of monopole and dipole transitions, and finally the last sharp feature, at 6.0 eV, corresponds to a superposition of a π-π∗ dipole transition from a low lying orbital (6hg ) to the LUMO and a broad π plasmon.32 These features can be used to probe the nature of organic interactions with other materials as purely electrostatic interactions will result in satellite structures that replicate the molecular gas phase analog.28 With LiF-C60 interaction, there is no observed change in the main C 1s spectrum, showing a strong photoemission peak from the main photoelectron at 285 eV.19 The high binding energy satellites did, however, show some indication of a change in the local distribution of delocalised electrons due to the presence of LiF. In particular, the feature at 291 eV, + 6.0 eV above the main peak, is attenuated when either LiF is deposited on C60 or C60 is deposited on LiF. When the

amount of deposited C60 was increased, the suppressed feature begins to emerge as seen in Figure 2. Enkvist et al.32,34 attributed this shakeup feature to the π bond on the six membered rings of C60 , specifically π-π∗ shake-up and π plasmon. The interaction therefore distrupts the shake-up of electrons involved in van der Waals bonding between C60 molecules. Such behavior has been observed in other systems that show frontier orbital mixing and slight charge transfer. Chemisorbed C60 shows a similar attenuation and loss of differentiation of the satellite structure with strongly interacting materials such as Au and Cr; whereas there is no loss of differentiation with C60 physisorption onto GaAs.27 This loss of the satellites during chemisorption is generally attributed to the electronic coupling of an absorbate hole and substrate excitations via relaxation processes.28 The attenuation of the 6 eV feature in particular can be explained using the theoretical model of Schönhammer and Gunnarsson for satellites in chemisorbed absorbate structures.35,36

Figure 2. C 1s high binding energy satellites. The dropdown lines indicate the theoretical position of the shake-up features after Enkvist et al.32 The spectra have been normalized to the main C1s photoemission peak, and offset for clarity.


M3118


Figure 4. Geometry optimized molecular structures for the interaction of LiF and C60 from density functional calculations (a) two LiF monomers with one with F pointed toward one hexagonal face of the C60 , and the other with Li pointed toward the opposite face. This represents the configuration with a charge re-distribution most closely corresponding to the binding energy shift that observed experimentally. (b) two LiF monomers with Li pointed toward adjacent C-C bonds between pentagons and hexagons. This represents the lowest energy configuration for a dimer structure interacting with C60 . The predicted binding energy shift is not visible within the broad F1s peak.

Figure 3. F 1s core level spectrum with high energy shoulder for (a) deposition of ∼5Å LiF on ∼17 ML C60 on Si (b) deposition of 2 ML C60 on ∼5Å LiF on Au, and (c) the as peeled surface of a single layer device structure glass/SiO/Al/LiF/C60 /LiF/Al/SiO. Removal of the substrate and organic layers in vacuum left behind the cathode material (SiO/Al/LiF) and approximately 50Å of C60 . Reprinted with permission from Ref. 48. The solid line represents crystalline LiF, except for (b) where it is the LiF substrate prior to C60 deposition.

In that description, a shake-up channel for the absorbate is blocked due to a transfer of charge from the substrate to a previously unoccupied absorbate valence level. In strong bonding, the probability of this charge transfer screening taking place is high and the satellite feature will be greatly attenuated. As the π-π∗ transition represents a bonding-antibonding band excitation into the LUMO, chemical interaction between the C60 and LiF is shown by the suppression of this feature. The strength of the bonding cannot be quantitatively established from the amount of attenuation, however, because the bulk π plasmon superimposed on the dipole transition prevents the feature from completely disappearing from the spectrum. The features at 3.8 eV and 4.8 eV represent multiple transitions, some to higher levels, which are less likely to be filled during charge transfer; therefore, these excitations are still expected and appear largely unchanged. However, though there is some interaction, there does not appear to be the strong formation of a C-F bond. C-F covalent bonds are expected at or above 290 eV, particularly for polyfluorinated C60 .37 If both C60 and C-F bonds were contributing to the spectrum, the intensity of the 290 eV feature might be expected to be higher than that of C60 alone. In our study, the opposite was observed, with the intensity smaller than that expected, suggesting that F is not covalently bonded to C. This interaction is also observed in the F1s core level, where a high binding energy shoulder appears when LiF and C60 interact, ∼3.2 eV ± 0.4 eV above that of ionic LiF. The appearance of this shoulder is also deposition direction independent, as it was observed both for LiF deposition on C60 and C60 deposition on LiF (Figure 3). This feature appears to always accompany the attenuated satellite in the C1s core level, disappearing for the 6 ML deposition of C60 on LiF surfaces, when the satellite was recovered. In a device like sandwich structure, with LiF on both sides of a 100 nm C60 film, this shoulder feature dominates the F1s spectrum for the 50Å region close to the top electrode, exposed by an in vacuum peeling

method21 (vis. Figure 3c). This core level, generally around 688.5 eV, has been observed in fluorinated fullerene structures, and has been attributed to a semi-covalent F attachment to the molecule through the π-bonds.38 To examine if there were possible structures that could give rise to these spectroscopic features, we performed density functional calculations to determine the geometry optimized interaction between LiF and C60 . Many possible arrangements can arise as the individual interacting molecules may have Li or the F unit close to the hexagonal faces, pentagonal faces, or one of the C-C bonds in the C60 molecule. These various arrangements for the interaction between LiF and C60 were geometry optimized by total energy minimization to ensure that they represent realistic structures. The Mullikan charges and binding energies were determined for an ionic LiF molecule, for an isolated LiF molecule, and for fullerenes interacting with LiF monomers in three configurations: (1) Li close to a hexagonal face with the Li-F bond normal to the face (2) similarly, except with F close to the hexagonal face and (3) the Li-F bond slanted to have the F atom above a pentagonal face. The interaction between C60 and one or two LiF monomers in these configurations were examined. The geometry optimized bond lengths, and Mullikan charges are given in Table I. Figure 4 shows the two energetically favorable configurations for two monomers interacting with the C60 molecule. The Li-F bond distance (i.e. nearest neighbor distance) in FCC-solid LiF is 2.014Å while the bond length R0 in the isolated molecule is 1.586 Å. During thermal evaporation, LiF vapor is known to consist of monomers, dimers and trimers.39–41 The experimentally determined monomer length, 1.51Å42 is very similar to the predicted value for isolated LiF molecules. This calculated reduction in bond length is accompanied by a reduction of the energy gap by more than a factor of two. Thus the HOMO-LUMO gap in the monomer is predicted as 4.78 eV. Similarly, the largely ionic (92%)Li+F- bond in the solid phase becomes significantly covalent in the isolated monomer, with Mullikan charges of +0.5 on Li and -0.5 on F. As seen from Table I, the Li-F bond distance when in proximity of a C60 molecule is also very close to R0 . That is, upon interaction, there will be no free Li+ cations and F- anions; it contains essentially covalent LiF molecules. For LiF deposition on the organic surface, therefore, some of the LiF could be interacting with C60 , without relaxing to the length expected for crystalline LiF. That the bond distance upon interaction is very close to R0 for the isolated monomer also implies that the interaction between LiF and C60 is also not strong. In fact, our calculations indicate that there



M3119

Table I. Theoretical binding energy shifts for model structure of LiF-C60 interaction. Structure LiFcryst LiFmonomer LiF C60 -FLi1 C60 -LiF2 C60 -LiF3 LiF-C60 -LiF4 LiF-C60 -FLi5 FLi-C60 -LiF6

qF −1 −1 −0.5 −0.491 −0.5 −0.499 −0.493 −0.504 −0.490 −0.489 −0.480 −0.489

q Li +1 +1 +0.5 +0.523 +0.365 +0.348 +0.535 +0.347 +0.524 +0.521 +0.374 +0.355

qC60

r F Li (Å)

r FC60 (Å)

– – – −0.033 +0.135 +0.151 −0.042 +0.157 −0.034 −0.032 +0.106 +0.134

2.01∗

– – – 4.5‡ 6.6‡ 6.89‡ 6.09‡ 7.22‡ 6.22‡ 6.22‡ 6.67‡ 6.62‡

1.51§ 1.586 1.573 1.574 1.577 1.573 1.574 1.573 1.573 1.572 1.573

VF

(eV)

12.52 9.53 4.54 4.69 3.63 3.49 5.47 3.60 4.69 4.67 3.52 3.48

E B (F) Li F – −2.9 1.9 2.3 1.0 1.1 3.0 0.9 2.3 2.3 1.3 1.1

∗ From

ao for crystalline LiF after Euwema et al.49 § From monomer size during vapor deposition.42 ‡ From C60 diameter after Troullier et al.,18 taking the LiF molecule at a minimized distance away from the surface of C60 . 1 F near a hexagonal face, the LiF bond is normal to the face. 2 Li near a hexagonal face, the LiF bond is normal to the face. 3 Li is on a bond between a hexagonal and a pentagonal face. The LiF bond is slanted so that the F atom is above the pentagonal face. 4 F near a hexagonal face, Li near the opposite hexagonal face. LiF bonds normal to the hexagonal faces. 5 F adjacent to a hexagonal face. The F-C distances are around 3.05Å to all six C atoms in the hexagon. LiF bonds normal to opposing hexagonal faces. 6 Li on adjacent C-C bonds between hexagons and pentagons.

is only a small amount of charge transfer. The DFT calculations of the charge redistribution on the interaction of C60 and LiF show two general cases, depending on the relative position of the Li atom which is a strong electron acceptor. When the F atom is pointed toward C60 , the average positive charge on the Li varies between 0.535 to 0.521, while when the F is pointed away from the molecule, the Li charge varies from 0.374 to 0.347. As the positive charge on the Li can easily engage with the mobile π-electrons from the C60 molecule, there is much stronger transfer of charge when the Li is in closer proximity. For either case, because F is a relatively weak donor, the average negative charge on the F only varies slightly between 0.4800.504. Thus, compared to nominal LiF, the average F negative charge has been lowered by 0.009, while the Li charge has been raised on average by about 0.028 or 0.124, depending on the LiF molecule orientation. This net change is accommodated by charge transfer to the C60 molecule. This can result in at most ∼0.15 electrons transfer between the Li0.5+ part of the LiF molecule and the C60 units, in the most stable configurations. This small amount of charge transfer also provides some insight into the satellite reorganization with LiF interaction. The small amount of charge transfer is not sufficient to produce a strong bond, but some charge transfer disrupts the π−π∗ interaction between adjacent C60 molecules. The dominant mechanism appears to be the strong polarizability of the LiF and C60 unit. The charge transfer effects as well as bondpolarization effects influence the binding energies (energy eigenvalues) of the electronic states of the LiF-C60 molecular complexes. The presence of LiF produces a slight decrease in the C60 energy gap.20 Here the distance between LiF and C60 center, as well as the other inter-nuclear distances have been energy optimized, and hence this is the upper-bound to the gap reduction that may arise from interaction with LiF. As stated above, the contraction in the bond length also results in a significant change in the LiF bandgap. The largest effect in the spectrum, therefore, could be expected on the F 1s core level. As the Mullikan charges and bonds lengths represent geometry optimized structures, it is possible to describe the chemical shift, or the change in the expected binding energy versus that of crystalline LiF. The theoretical minimum energy configurations for a LiF molecule interacting with a C60 molecule indicates that the charge distribution over the LiF changes, with a subsequent shortening of the LiF bond. The charge potential theory of Siegbahn et al.43 can therefore be used with this model to explain the appearance of the shoulder in the XPS

spectra. A chemical shift due to a change in the local environment can be expressed by E b (F1s) = k j q F + V mad + E FR , where EB is the change of the core binding energy versus a reference compound, kj is the interaction coefficient between core electrons and valence electrons, qF is the difference in the effective local charge on the atom of interest, VMad is the difference in the Madelung R potential due to the surrounding atoms, andE ea A is the difference in the relaxation energy term due to photoelectron emission. To correspond to the LiF monomer interaction described by the model, an ionic LiF molecule can be taken as the reference compound, so that the effective local charge difference can be estimated from the ideal ionic crystal and the predicted Mullikan charges. The interaction coefficient, k, was taken as 19.6 for F 1s, from the empirical formula determined by Sleigh et al.44 The Madelung potential using the shortened bond lengths can be approximately described by a Coulombic interaction assuming each atom as a point charge in space (in this treatment, C60 is taken as a single point charge of molecular radius). The point-charge model for the potential in eV on a given atom, i, can be described by qj V Mad = 14.4α Mad r i= j i j where qj is the charge on the other atoms, and rij the interatomic spacing to the atom of interest in Å. For a LiF crystal, the Madelung constant, αLiF , is 1.7545 and the interatomic interactions can be estimated from a single molecule and the crystal lattice constant, ao . For all other structures used in the model, the Madelung constant is simply taken as 1 in the absence of long range crystal structure, and all of the nearest neighbor interactions calculated. Assuming that equal shifts in binding energy in all core levels occur with a change in the chemical environment, the relaxation energy can be estimated from the change in the modified Auger parameter,46,47 as defined by α = 2E FR = E b (F1s) + E k (FK L L ) where Ek (FKLL) is the kinetic energy of the first peak of the Auger transition associated with excitation in the F 1s core level. Using the modified Auger parameters for ∼5-10Å LiF deposited on Si and on C60 ,48 we can estimate the relaxation energy to be ∼0.125 eV.


M3120


Table I also lists the predicted simplified Madelung energy and binding energy shifts from the Mullikan charges and bond lengths predicted by the theoretical model. The LiF-C60 interaction appears to be consistent with the observed binding energy shifts. It should be noted that the relative position of the LiF molecule did have an effect on the charge distribution. Two LiF monomers interacting with C60 normal to the hexagonal faces, one F near a hexagonal face and one Li near the opposite hexagonal face, LiF-C60 -LiF shown in Figure 4a, best predicts the shift in the F 1s spectrum, even though it is not the lowest energy structure predicted by the DFT calculations (Figure 4b for two LiF interaction FLi-C60 -LiF). Though the structure with two F facing the C60 molecule (LiF-C60 -FLi) suggests a slightly lower but similar chemical shift in Table I, this structure is energetically highly unlikely. In these samples, prepared at room temperature with no annealing, the predicted structures other than the last mentioned are energetically possible, as thermodynamic equilibrium may not have been reached during evaporation.

Summary In an analysis of the interaction of LiF and C60 , it appears that LiF forms a charge transfer complex with C60 , which can be observed in high resolution XPS scans of both the F1s and C 1s shake-up satellites. DFT calculations suggest that the interaction is between monomer or dimer type LiF molecules in the C60 matrix. This interaction shows that even in the absence of dissociation, charge transfer complexes exist in such systems and can explain some of the observed behavior of LiF doping in C60 based devices.

Acknowledgments The author acknowledge ZhengHong Lu at the University of Toronto for facilitating these experiments and giving guidance, and Lars Jeurgens at the Max Planck Institute for Metal Research for fruitful discussions. A.T. acknowledges financial support from 4361002013 RGPIN.

References 1. H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl, and R. E. Smalley, C60: Buckminsterfullerene, Nature, 318(6042), 162 (1985). 2. S. E. Shin, H. J. Choi, J. Y. Hwang, and D. H. Bae, Strengthening Behavior of Carbon/Metal Nanocomposites, Sci. Rep., 5, 16114 (2015). 3. L. A. Chernozatonskii, V. P. Val’chuk, N. A. Kiselev, O. I. Lebedev, A. B. Ormont, and D. N. Zakharov, Synthesis and Structure Investigations of Alloys with Fullerene and Nanotube Inclusions, Carbon, 35(6), 749 (1997). 4. H. Peng, F. S. M. Leung, A. X. Wu, Y. Dong, Y. Dong, N.-T. Yu, X. Feng, and B. Z. Tang, Using Buckyballs To Cut Off Light! Novel Fullerene Materials with Unique Optical Transmission Characteristics, Chem. Mater., 16(23), 4790 (2004). 5. Y. Gao, X. Wu, and X. C. Zeng, Designs of Fullerene-Based Frameworks for Hydrogen Storage, J. Mater. Chem. A, , 2(16), 5910 (2014). 6. Q. Liu, Q. Cui, X. J. Li, and L. Jin, The Applications of Buckminsterfullerene C60 and Derivatives in Orthopaedic Research, Connect. Tissue Res., 55(2), 71 (2014). 7. D. Y. Lyon and P. J. J. Alvarez, Fullerene Water Suspension (nC60) Exerts Antibacterial Effects via ROS-Independent Protein Oxidation, Environ. Sci. Technol., 42(21), 8127 (2008). 8. A. Turak, Interfacial Degradation in Organic Optoelectronics, RSC Adv., 3(18), 6188 (2013). 9. C. J. Huang, D. Grozea, A. Turak, and Z. H. Lu, Passivation Effect of Al/LiF Electrode on C-60 Diodes, Appl. Phys. Lett., 86(3), 33107 (2005). 10. A. Turak, C. J. Huang, D. Grozea, and Z. H. Lu, Oxidation of LiF-Coated Metal Surfaces, J. Electrochem. Soc., 154(8), H691 (2007). 11. D. Grozea, A. Turak, Y. Yuan, S. Han, Z. H. Lu, and W. Kim, Enhanced Thermal Stability in Organic Light-Emitting Diodes through Nanocomposite Buffer Layers at the Anode/Organic Interface, J. Appl. Phys., 101(3), 33522 (2007). 12. L. Hung, C. W. Tang, Mason, and M. G. Enhanced Electron Injection in Organic Electroluminescence Devices Using an Al/LiF Electrode, Appl. Phys. Lett., 70(2), 152 (1997). 13. C. J. Brabec, S. E. Shaheen, C. Winder, N. S. Sariciftci, P. Denk, and S. N. Sariciftci, Effect of LiF/Metal Electrodes on the Performance of Plastic Solar Cells, Appl. Phys. Lett., 80(7), 1288 (2002).

14. K. Kawano and C. Adachi, Reduced Initial Degradation of Bulk Heterojunction Organic Solar Cells by Incorporation of Stacked Fullerene and Lithium Fluoride Interlayers, Appl. Phys. Lett., 96(5), 53307 (2010). 15. D. Gao, M. G. Helander, Z. Wang, D. P. Puzzo, M. T. Greiner, and Z. Lu, C60: LiF Blocking Layer for Environmentally Stable Bulk Heterojunction Solar Cells, Adv. Mater., 22(47), 5404 (2010). 16. Y. Q. Zhao, C. J. Huang, T. Ogundimu, and Z. H. Lu, Transparent Conducting C60:LiF Nanocomposite Thin Films for Organic Light-Emitting Diodes, Appl. Phys. Lett., 91(10), 103109 (2007). 17. E. D. Głowacki and K. L. Marshall,; C. W. Tang and N. S. Sariciftci, Doping of Organic Semiconductors Induced by Lithium Fluoride/Aluminum Electrodes Studied by Electron Spin Resonance and Infrared Reflection-Absorption Spectroscopy, Appl. Phys. Lett., 99(4), 43305 (2011). 18. N. Troullier and J. L. Martins, Structural and Electronic Properties of C60, Phys. Rev. B, 46(3), 1754 (1992). 19. J. A. Leiro, M. H. Heinonen, T. Laiho, and I. G. Batirev, Core-Level XPS Spectra of Fullerene, Highly Oriented Pyrolitic Graphite, and Glassy Carbon, J. Electron Spectrosc. Relat. Phenom., 128 (2), 205 (2003). 20. Z. H. Lu, C. C. Lo, C. J. Huang, Y. Y. Yuan, M. W. C. Dharma-wardana, and M. Z. Zgierski, Quasimetallic Behavior of Carrier-Polarized C60 Molecular Layers: Experiment and Theory, Phys. Rev. B, 72(15), 155440 (2005). 21. A. Turak, D. Grozea, X. D. Feng, Z. H. Lu, H. Aziz, and A. M. Hor, Metal/AlQ(3) Interface Structures, Appl. Phys. Lett., 81(4), 766 (2002). 22. K. Momma and F. Izumi, VESTA 3 for Three-Dimensional Visualization of Crystal, Volumetric and Morphology Data, J. Appl. Crystallogr., 44(6), 1272 (2011). 23. M. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, V. Zakrzewski, J. Montgomery, R. Stratmann, and J. Burant, others. Gaussian 98, Revision A. 9. Gaussian Inc Pittsburgh PA 1998. 24. A. D. Becke, Density-functional Thermochemistry. III. The Role of Exact Exchange, J. Chem. Phys., 98(7), 5648 (1993). 25. C. Lee, W. Yang, and R. G. Parr, Development of the Colle-Salvetti CorrelationEnergy Formula into a Functional of the Electron Density, Phys. Rev. B, 37(2), 785 (1988). 26. T. L. Barr, Modern ESCA The Principles and Practice of X-Ray Photoelectron Spectroscopy, CRC Press, 1994. 27. T. R. Ohno, Y. Chen, S. E. Harvey, G. H. Kroll, J. H. Weaver, R. E. Haufler, and R. E. Smalley, C60 Bonding and Energy-Level Alignment on Metal and Semiconductor Surfaces, Phys. Rev. B, 44(24), 13747 (1991). 28. E. Umbach, Satellite Structures in Photoemission Spectra from Different Types of Absorbates, Surf. Sci., 117(1), 482 (1982). 29. S. Hüfner, Unfilled Inner Shells: Transition Metals and Compounds. In Photoemission in Solids II; D. L. Ley and P. D. M. Cardona, Eds.; Topics in Applied Physics; Springer Berlin Heidelberg, 1979; pp 173. 30. A. Schöll, Y. Zou, M. Jung, T. Schmidt, R. Fink, and E. Umbach, Line Shapes and Satellites in High-Resolution X-Ray Photoelectron Spectra of Large π-Conjugated Organic Molecules, J. Chem. Phys., 121(20), 10260 (2004). 31. A. Nakajima and M. Uchino, Charge Redistribution in a Charge Storage Layer Containing C60 Molecules and Organic Polymers for Long Electron Retention, Appl. Phys. Lett., 101(21), 213301 (2012). 32. C. Enkvist, S. Lunell, B. Sjögren, S. Svensson, P. A. Brühwiler, A. Nilsson, A. J. Maxwell, and N. M˚artensson, C1s Shakeup Spectrum of C60: Global ChargeTransfer Satellites and Their Relation to the X-Ray Threshold Singularities in Macroscopic Systems, Phys. Rev. B, 48(19), 14629 (1993). 33. J. H. Weaver, J. L. Martins, T. Komeda, Y. Chen, T. R. Ohno, G. H. Kroll, N. Troullier, R. E. Haufler, and R. E. Smalley, Electronic Structure of Solid C60: Experiment and Theory, Phys. Rev. Lett., 66(13), 1741 (1991). 34. S. Lunell, C. Enkvist, M. Agback, S. Svensson, and P. A. Brühwiler, Core Photoionization Satellites in Fullerence and Related Model Systems, Int. J. Quantum Chem., 52(1), 135 (1994). 35. O. Gunnarsson and K. Schönhammer, CO on Cu(100) - Explanation of the ThreePeak Structure in the X-Ray-Photoemission-Spectroscopy Core Spectrum, Phys. Rev. Lett., 41(23), 1608 (1978). 36. K. Schönhammer and O. Gunnarsson, Correlation Effects on Core Level Spectra of Adsorbates, Solid State Commun. , 26(6), 399 (1978). 37. D. M. Cox, S. D. Cameron, A. Tuinman, A. Gakh, J. L. Adcock, R. N. Compton, E. W. Hagaman, K. Kniaz, and J. E. Fischer, X-Ray Photoelectron and NMR Studies of Polyfluorinated C60: Evidence That C-C Bonds Are Broken, J. Am. Chem. Soc., 116(3), 1115 (1994). 38. D. Claves, J. Giraudet, A. Hamwi, and R. Benoit, Structural, Bonding, and Electrochemical Properties of Perfluorinated Fullerene C70, J. Phys. Chem. B, 105(9), 1739 (2001). 39. G. M. Rothberg, M. Eisenstadt, and P. Kusch, Free Evaporation of Alkali Halide Crystals, J. Chem. Phys., 30(2), 517 (1959). 40. M. Eisenstadt, G. M. Rothberg, and P. Kusch, Molecular Composition of Alkali Fluoride Vapors, J. Chem. Phys., 29(4), 797 (1958). 41. M. F. Butman, A. A. Smirnov, L. S. Kudin, and Z. A. Munir, Mass-Spectrometric Study of the Temperature Variation in the Dimer-to-Monomer Ratio in the FreeSurface Vaporization Fluxes from Alkali Halide Single Crystals, J. Mater. Synth. Process., 8(2), 93 (2000). 42. C. P. Baskin, C. F. Bender, and P. A. Kollman, Dimers of Lithium Fluoride and Sodium Hydride, J. Am. Chem. Soc., 95(18), 5868 (1973). 43. K. Siegbahn, C. Nordling, A. Fahlman, R. Nordberg, K. Hamrin, J. Hedman, G. Johansson, T. Bergmark, S.-E. Karlsson, and I. Lindgren, ESCA – - Atomic, Molecular and Solid State Structure Studied by Means of Electron Spectroscopy; Nova Acta Regiae Soc. Sci. Ups, 1967, Vol. 20.


ECS Journal of Solid State Science and Technology, 6 (6) M3116-M3121 (2017) 44. C. Sleigh, A. P. Pijpers, A. Jaspers, B. Coussens, and R. J. Meier, On the Determination of Atomic Charge via ESCA Including Application to Organometallics, J. Electron Spectrosc. Relat. Phenom., 77(1), 41 (1996). 45. D. B. Sirdeshmukh, L. Sirdeshmukh, and K. G. Subhadra, Alkali Halides: A Handbook of Physical Properties; Springer series in materials science, Springer: Berlin, 2001. 46. G. Moretti, Auger Parameter and Wagner Plot in the Characterization of Chemical States by X-Ray Photoelectron Spectroscopy: A Review, J. Electron Spectrosc. Relat. Phenom., 95 (2), 95 (1998).

M3121

47. M. Satta and G. Moretti, Auger Parameters and Wagner Plots, J. Electron Spectrosc. Relat. Phenom., 178–179, 123 (2010). 48. A. Turak, Cathode Interface Structure in Organic Semiconductor Devices. PhD, University of Toronto: Toronto, Canada, 2006. 49. R. N. Euwema, G. G. Wepfer, G. T. Surratt, and D. L. Wilhite, HartreeFock Calculations for Crystalline Ne and LiF, Phys. Rev. B, 9(12), 5249 (1974). 50. R. C. Haddon, L. E. Brus, and K. Raghavachari, Electronic Structure and Bonding in Icosahedral C60, Chem. Phys. Lett., 125(5), 459 (1986).
