Electronic Effects of Group Fragments on the XPS of Fe 2 p and 3 p Photoelectron Lines of Ferrocenyl-containing Chalcones

A series of ferrocenyl-containing chalcones, Fc-CO-CH=CH-C6H4R, with the R-group on the para-position on the phenyl ring and R = OCH3 (1), CH3 (2), C6H5 (3), tBu (4), H (5), Br (6) and CF3 (7) were subjected to an X-ray photoelectron spectroscopy (XPS) study. The linear relationships obtained between the Gordy-scale group electronegativity of molecular fragment R, ÷R, and the maximum binding energies of the Fe 2p3/2and the Fe 3p3/2 photoelectron lines, confirmed communication between the iron atom of the ferrocene moiety and the molecular fragments, R, of 1–7. These relationships illustrated that the influence of the electronic properties of the molecular fragments are more pronounced in the Fe 3p photoelectron lines than in the Fe 2p photoelectron lines.


Introduction
Chalcones are organic compounds with an enone backbone, containing aromatic end groups, and are one of the primary building blocks of flavonoids. 1 In ferrocenyl-containing chalcones, one (or more) of the aromatic rings are replaced with a ferrocenyl group. 2 The catalytic, 3-5 electrochemical [6][7][8][9][10] and structural properties, 6,11,12 of ferrocene and its derivatives, like ferrocenyl-containing chalcones, have been studied with a variety of different techniques.However, characterization by means of X-ray photoelectron Spectroscopy (XPS) is unvisited for these complexes.
XPS is a very useful tool to determine elements present in a sample, the oxidation state of the elements and even the chemical environment which surrounds the element.4][15][16][17][18][19][20] Herewith we want to investigate whether the group electronegativity of the R-group's influence can be detected by the position of the binding energy of the iron in the ferrocenyl group.
In this study a variety of ferrocenyl-containing chalcones with the structure Fc-CO-CH=CH-C 6 H 4 R, where R is in the para-position on the phenyl ring and R = OCH 3 (1), CH 3 (2), C 6 H 5 (3), tBu (4), H (5), Br (6) and CF 3 (7), see Fig. 1, was subjected to X-ray photoelectron spectroscopy.The Fe 2p and Fe 3p photoelectron lines were measured and its binding energy positions were correlated against an assortment of other physical and calculated properties.

Chemicals and Instruments
The ferrocenyl-containing chalcone derivatives 1-7 were synthesized and characterized according to published methods. 6l computational and electrochemical results presented here were taken from Ref. 6. XPS were recorded of neat powered samples which were held in place on the sample holder by means of carbon tape.To study the powered samples, the X-ray photoelectron spectroscopy (XPS) was conducted on a PHI 5000 versa probe spectrometer using monochromatic Al K a radiation (hn =1486.6 eV) generated by a 25 W, 15 kV electron beam.For high-resolution spectra, the Hemispherical Analyzer pass energy was maintained at 93.90 eV.Measurements were performed using a 1 eV/step binding energies for survey scans from 0 to 1400 eV, while a 0.1 eV/step binding energy was used for the high-resolution spectra.The X-ray beam size used for the XPS measurements was 10 µm.The pressure during acquisition was less than 1.3 × 10 -9 Torr.All the absolute binding energies of the photoelectron spectra were corrected with C 1 s signal at 284.6 eV (the lowest binding energy of the simulated adventitious C 1 s photoelectron line). 21The XPS data was analyzed utilizing Multipak version 8.2c computer software, 22 and applying Gaussian-Lorentz fits (the Gaussian/Lorentz ratios were always >95%).

Results and Discussion
The systematic variation of the molecular fragment, R, in the para-position on the phenyl ring of the series of ferrocenylcontaining chalcones, Fc-CO-CH=CH-C  OCH 3 (1), CH 3 (2), C 6 H 5 (3), tBu (4), H (5), Br (6) and CF 3 (7), allows for the investigation of the influence of the electronic properties of R on the binding energy of the photoelectron lines as detected by X-ray photoelectron spectroscopy (XPS).The differences in binding energies detected of the photoelectron lines will be related to the alterations in the electronic structure of the ferrocenyl-containing chalcones as well as DFT calculated parameter and other physical properties.
Firstly, the influence of the electronic properties of the different molecular fragments (electron-withdrawing and donating) on XPS measured binding energies of the Fe 2p peaks of the ferrocenyl-containing chalcones will be described and this concept will then be expanded to the Fe 3p peaks.
The Fe 2p 1/2 and Fe 2p 3/2 photoelectron lines, as measured by XPS of the Fe 2+ ions of the ferrocenyl-containing chalcones (1-7) gave sharp well-defined single peaks, with some showing a small unsymmetrical tail toward the high energy side (Fig. 2, left).This tail is normal for metals of the first row of the transition metals. 23,24,26aumkin et al. reported the deconvolution of Fe(II) complexes with multiplet splitting peaks which was attributed to different non-equivalent states of the Fe(II). 25However, since the main purpose of this article is the comparison of the maximum binding energy of the Fe 2p envelopes with different physical properties of the R-groups, the photoelectron lines will not be deconvoluted into multiplet peaks.During the fitting of these Fe 2p 1/2 and Fe 2p 3/2 photoelectron lines, a single asymmetric Gaussian peak with a full width at half maximum (FWHM) of ca.1.8 eV was utilized giving chi-square values between 0.5 and 1.1.The goodness of the fit as well as the sharpness of the peaks verifies that only one iron species is present in the ferrocenylcontaining chalcone. 26The sharpness of the peaks also indicates that the chalcones are stable under irradiation with X-ray and did not decompose during the experiment.The asymmetric index, a, (which is the ratio of the half width at half maximum on the high energy side to the half width at half maximum on the low energy side) of the Fe 2p 1/2 and Fe 2p 3/2 photoelectron peaks being 1.25 and 1.3, respectively.These values are typical for transition metal ions of the first row. 27he iron in the ferrocenyl groups have equal 3d subshell population (3d 6 ) and therefore the iron atoms is a low-spin Fe(II) species.Therefore, during the fitting of the Fe 2p and Fe 3p peaks there is no need to take any final-state effects like multiplet splitting and shake-up peaks into account.
RESEARCH ARTICLE E. Erasmus, 95 S. Afr.J. Chem., 2017, 70, 94-99, <http://journals.sabinet.co.za/sajchem/>.The maximum binding energy of the Fe 2p 3/2 photoelectron lines of 1-7 was located between 707.53 to 708.05 eV (see Table 1).The change in position of the binding energy is caused by the inductive electronic effects of the molecular fragments, R.This 0.52 eV binding energy range, is considered large since a 1 eV binding energy change is associated with a full iron redox state change from Fe 2+ to Fe 3+ .Also, this range of binding energies are larger than the binding energy range of 0.34 eV which was found for the Fe 2p 3/2 photoelectron lines of Mn(B-diketonato) 3 complexes. 28,29he binding energies obtained for the Fe 2p 3/2 photoelectron lines of 1-7 correlates very well with other Fe(II) species, iron disphosphide, 30 bis(cyclopentadenyl)iron, 31 and the ferrocenylgroup in Mn(B-diketonato) 3 , 28,29 all located at ca. 707.7 eV.However, the Fe 2p of the ferrocenyl-containing chalcones is located at lower binding energy compared to a ferrocenyl linked via an aminoalkyl silane onto silicon (709.7 eV). 32 spin-orbit splitting of ca.13.5 eV was obtained between the Fe 2p 3/2 and Fe 2p 1/2 photoelectron lines, depending on the molecular fragment, R (see Fig. 3 middle and Table 1).
The only other photoelectron lines which could be detected in the XPS of 1-7 are carbon (with a maximum binding energy at ca. 284.6 eV) and oxygen (with a maximum binding energy at ca. 528.8 eV).The ferrocenyl-containing chalcones, 6-7 also displayed bromine at 69.6 eV and fluorine at 687.8 eV, respectively.The correct atomic percentage ratio of 1:1 between the Fe 2p peak and the Br 3d peak, within experimental error, was obtained for 6 (experimental ratio is 1:1.1).For 7 the correct atomic percentage ratio of 1:3.2 was experimentally obtained between the Fe 2p and the F 2s (theoretical value 1:3).All the detected photoelectron lines were charge corrected against C 1 s at 284.6 eV (the lowest binding energy of the simulated adventitious C 1 s photoelectron line).
The substructure of the carbon photoelectron line was fitted with separate peaks located at 292.8 eV (representing the C-F), 286.7 eV (representing the C=O) and 284.6 eV (representing the all the other carbon involved in C-C bonding) allocated to the ferrocenyl-chalcone, 7 was used as an example, in the expected ratio of 1:1:18, while the additional peak detected at 288.4 eV (representing the O=C=O), 286.6 eV (representing the C=O) and 284.6 eV (representing the all the other carbon involved in C-C bonding) comes from the adventitious carbon always present on all samples (see Fig. S1 for the C 1 s are of 7 in the Supplementary Information).
4][15][16][17][18][19][20] An increase in the Gordy-scale group electronegativity, c R , of the molecular fragment, R, causes an increase in the binding energy of the Fe 2p 3/2 photoelectron line, as well as an increase in the spin-orbit splitting (DBE 2p1/2-2p3/2 ) between the binding energy of the Fe 2p 1/2 and Fe 2p 3/2 photoelectron lines (see Fig. 3 left and middle as well as Table 1).As the Gordy-scale group electronegativity, c R , increases, more electron density is pulled towards the stronger electron-withdrawing molecular fragments.This causes the Fe 2+ ions to have less electron density around it.Therefore, the Fe 2+ ion binds stronger to its own electrons, hence the increased binding energy.The linear relationships obtained between the binding energy (BE) of the iron Fe 2p  The Fe 3p 1/2 and Fe 3p 3/2 photoelectron lines of these ferrocenyl-containing chalcone 1-7, is represented by a single broad peak, with no definite shape showing the separation between the 3p 1/2 and 3p 3/2 lines (see Fig. 2).This, however, is normally the case for Fe 3p peaks since the spin-orbit splitting between the Fe 3p 1/2 and Fe 3p 3/2 photoelectron lines is less than 2 eV.
Compared to the Fe 2p photoelectron lines, also only one Gaussian peak was fitted for the Fe 3p photoelectron lines of 1-7, with a FWHM ranging between 2.2 and 2.9 eV.The asymmetric index, a, of the Fe 3p 1/2 and Fe 3p 3/2 photoelectron peaks are both 1.1.The maximum binding energy for the Fe 3p 3/2 photoelectron lines was detected at ca. 53.9 eV (see Table 1), with a spin-orbit splitting of ca.1.6 eV between the Fe 3p 3/2 and Fe 3p 1/2 photoelectron lines.
Similar to what was found for the Fe 2p 3/2 photoelectron lines, an increase in molecular fragment's Gordy-scale group electronegativity led to an increase in the maximum binding energy of the Fe 3p 3/2 photoelectron lines (Fig. 3, right and Table 1).The maximum binding energy of the Fe 3p 3/2 photoelectron lines is located in the range 53.6-54.5 eV.This is a 0.9 eV span, obtained for Fe 2+ of the ferrocenyl fragment within the chalcone.The binding energy span for the Fe 3p 3/2 photoelectron lines is almost equivalent to an entire oxidation state change, which is normally 1 eV.Compared to the Fe 2p 3/2 photoelectron lines which had a binding energy span of 0.53 eV, the Fe 3p 3/2 photoelectron lines has a 0.9 eV binding energy span, almost double that of the Fe 3p 3/2 photoelectron lines.Also, the slope of the equation that fits the relationships between the binding energy of phototelectron lines and the Gordy-scale group electronegativity of the Fe 3p 3/2 which is 1.02 (Equation 3) is much steeper than the Fe 3p 3/2 which is 0.62 (Equation 1).This clearly shows that the electrons located in higher energy level orbitals experience the influence of the electronic properties of the changing molecular fragments more intense.
The linear relationships obtained between the binding energy (BE) of the iron Fe 3p 3/2 photoelectron lines and the Gordy-scale group electronegativity of the various molecular fragments, c R (Fig. 3, right) is described by the equation: The linearity of the relationships between Gordy-scale group electronegativities of the molecular fragment, c R , and the different binding energies of both the Fe 2p 3/2 and Fe 3p 3/2 photoelectron lines, shows that the electronic communication through the bonds from the molecular fragment, R, to the Fe 2+ in the ferrocenyl-containing chalcones, 1-7, is very good.
The position of the molecular fragment on the phenyl was varied (to be either ortho, meta or para, see Fig. 4) to detect if the position also makes a different in the measured binding energy of the Fe 2p and Fe 3p photoelectron lines.
The comparative XPS spectra of the Fe 2p and Fe 3p areas of the ferrocenyl-containing chalcones, 7-9, with CF 3 containing chalcone at different position on the phenyl ring are shown in Fig. 5.As the CF 3 fragment is moved from ortho to meta to para, thus moving further away from the chalcone backbone, the binding energy of the Fe 2p 3/2 and Fe 3p 3/2 photoelectron lines increases (see Table 1).A binding energy difference of 0.35 eV for the Fe 2p 3/2 and 0.59 eV for the Fe 3p 3/2 photoelectron lines was found between the ortho and the para positions, respectively.This clearly shows that the position on the phenyl ring also has a pronounced effect on the electronic structure of the ferrocenylcontaining chalcones.
The linear relationships obtained between the binding energy (BE) of the iron Fe 3p 3/2 photoelectron lines and the Hammett constants of the R-fragments (Fig. 6) are described by the equations: For the halogen-containing chalcones (1, 6-7) For the non-halogen containing chalcones (1-5) BE = -1.04s R + 707.51;R 2 = 0.88 The V-shape correlation observed is similar to the correlation observed between the reduction potential and the Gordy group electronegativity, c R , previously reported. 6 linear correlation was obtained between the Gordy group electronegativity, c R and the binding energies of the Fe 2p and Fe 3p photoelectron lines while a V-shape correlation was obtained when using the Hammett constants.
5][36][37] On account of the V-shaped curve found between the Gordy group electronegativity, c R , and the Hammett constants of different R-groups, 6 the V-shaped curve found between the Hammett constants and the binding energies of the Fe 2p and Fe 3p photoelectron lines was not unexpected.][40][41] The Hammett constant is a value composing of a mixture of electronic effects including polar and resonance effects. 42Since polar effects which is incorporated in the Hammett constant represents a similar effects as electronegativity of the Gordyscale group electronegativity, it is proposed that the V-shape presumably comes from the resonance effect incorporated in the Hammett constant.This could possibly explain the difference between the halogen-containing chalones (electron-withdrawing R-groups) and the non-halogen-containing chalcones (electron-donating).When a core electron is emitted due to photoionization, the chalcone will try to stabilize this newly formed positive species.The type of stabilization will be dependent on the different R-groups.The halogen-containing chalcones (6-7), which has an electron-withdrawing R-group has can form a stabilized zwitterion through an inductive effect.While the non-halogen-containing chalcones (1-5), which has an electrondonating R-groups can form stabilized zwitterions through a resonance effect (see Ref. 6 for the proposed structures).
The V-shape correlation that was obtained highlights the changeover in stabilization mechanism for the chalcones having either electron-donating or electron-withdrawing R-groups and it distinguished between polar (electronegativity) and resonance effects.

Conclusion
The changes in the binding energy obtained from the Fe 2p and Fe 3p photoelectron lines upon substituting the group fragment R was investigated.The influence of the changing fragment R on the binding energy of the Fe 3p photoelectron lines (binding energy range = 0.90 eV) was found to be more pronounce than what was found for the binding energy of the Fe 2p photoelectron lines (binding energy range = 0.52 eV).A linear relationship was obtained between the Gordy-scale group electronegativities, c R , of the molecular fragments, R, and the binding energies of the Fe 2p 3/2 , Fe 3p 3/2 and spin-orbit splitting (DBE 2p1/2-2p3/2 ) of the binding energy of the Fe 2p 1/2 and Fe 2p 3/2 photoelectron lines.In contrast to this, a V-shape relationship was found between the binding energy of the Fe 2p 3/2 , Fe 3p 3/2 photoelectron lines and the Hammett constants of the R-groups.This V-shape correlation could possibly be attributed to the different forms of stabilization of zwitterion, for electron-withdrawing R-groups stabilization occurs through an inductive effect, while for electron-donating R-groups stabilization occurs through a resonance effect.The position of the molecular fragment R with in the phenyl ring also has a pronounce effect on the binding energy, a binding energy range of 0.35 eV for the Fe 2p 3/2 and 0.59 eV for the Fe 3p 3/2 photoelectron lines was found between the ortho and the para positions, respectively.

Figure 1
Figure 1 Structure of the ferrocenyl-containing chalcones investigated during this study.

Figure 3 1 )
Figure 3 Left: relationship between binding energy of the Fe 2p 3/2 photoelectron line and Gordy-scale group electronegativities, c R , of the molecular fragments, R. Middle: relationship between the spin-orbit splitting (DBE 2p1/2-2p3/2 ) of the binding energy of the Fe 2p 1/2 and Fe 2p 3/2 photoelectron lines and Gordy-scale group electronegativities, c R , of the molecular fragments, R. Right: relationship between binding energy of the Fe 3p 3/2 photoelectron line and Gordy-scale group electronegativities, c R , of the molecular fragments, R.

Figure 4
Figure 4 Structure of the ferrocenyl-containing chalcone with a CF 3 group on either the ortho (o), meta (m) or para (p) position on the phenyl ring.
a Data obtained from reference 18.