Electronic Structure and Spin Multiplicity of Iron Tetraphenylporphyrins in Their Reduced States as Determined by a Combination of Resonance Raman Spectroscopy and Quantum Chemistry

Article abstract of DOI:10.1021/acs.inorgchem.7b03018

Iron tetraphenylporphyrins are prime candidates as catalysts for CO₂ reduction. Yet, even after 40 years of research, fundamental questions about the electronic structure of their reduced states remain, in particular as to whether the reducing equivalents are stored at the iron center or at the porphyrin ligand. In this contribution, we address this question by a combination of resonance Raman spectroscopy and quantum chemistry. Analysis of the data allows for an unequivocal identification of the porphyrin as the redox active moiety. Additionally, determination of the spin state of iron is possible by comparing the characteristic shifts of spin and oxidation-state-sensitive marker bands in the Raman spectrum with calculations of planar porphyrin model structures.

Full text of DOI:10.1021/acs.inorgchem.7b03018

Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
Electronic Structure and Spin Multiplicity of Iron
Tetraphenylporphyrins in Their Reduced States as Determined
by a Combination of Resonance Raman Spectroscopy and Quantum
Chemistry
Christina Romelt,† Shengfa Ye,† Eckhard Bill,† Thomas Weyhermuller,† Maurice van Gastel,†
̈
̈
and Frank Neese*,†
†Max Planck Institute for Chemical Energy Conversion, Stiftstrasse 34-36, 45470 Mulheim an der Ruhr, Germany
̈
S
* Supporting Information
ABSTRACT: Iron tetraphenylporphyrins are prime candidates as catalysts for CO2 reduction. Yet, even after 40 years of
research, fundamental questions about the electronic structure of their reduced states remain, in particular as to whether the
reducing equivalents are stored at the iron center or at the porphyrin ligand. In this contribution, we address this question by a
combination of resonance Raman spectroscopy and quantum chemistry. Analysis of the data allows for an unequivocal
identification of the porphyrin as the redox active moiety. Additionally, determination of the spin state of iron is possible by
comparing the characteristic shifts of spin and oxidation-state-sensitive marker bands in the Raman spectrum with calculations of
planar porphyrin model structures.
Scheme 1. Schematic Structure of FeTPP Including
Nomenclature for the Carbon Atoms
INTRODUCTION
■
Over the last 100 years, carbon-based fuels have found tremen-
dousapplicationincombustionprocesses, havingledto an increase
in the CO2 concentration in the atmosphere by 100 ppm (roughly
40%). At present, the CO2 concentration is expected to further
increase. Owing to its abundance and contribution to the green-
house effect, the reconversion of CO2 into building blocks
(formaldehyde, formic acid, oxalic acid, methanol, CO) which can
be used in further processes is of great importance. On the basis of
its thermodynamic stability [ΔH(C + O2 → CO2) = −394 kJ/mol;
ΔH(CO2 + C → 2CO) = +173 kJ/mol], the reductive activation
of carbon dioxide has been one of the most challenging and most
relevant topics in catalysis for the last 40 years. The electrochem-
ical one-electron reduction of CO2 to the CO2•− radical anion, which
is accompanied by a bending of the CO2 molecule, is an energetically
unfavorable step that requires a large potential (E0′= −1.90 V) but
High-valent heme systems for oxidative chemistry have been
investigated extensively.8−10 Here, we focus on the reductive
chemistry of FeTPP relevant for CO2 reduction and for which
the low-valent states are relevant. Because reduction of [FeTPP]
can be greatly facilitated by coupling reduction to proton transfer.1,2
Iron porphyrins in their reduced, formal Fe(0) state, are known
to act as catalysts in the reduction of CO2. In fact, in the presence of
proton sources, iron tetraphenylporphyrins (FeTPPs, Scheme 1)
exhibit one of the largest catalytic activities and turnover frequencies
observed so far for homogeneous CO2 reduction catalysts.3−7
Received: November 29, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
for a related porphyrin system showed that the saddled and ruffled dis-
tortions are associated with energy differences on the order of about
0.5 eV only.44 As such, the potential energy surface is thus quite flat
toward these distortions. In turn, the flatness of the potential energy
surface means that mainly the low-lying “breathing” vibrations will be
sensitive to these distortions. The vibrations of relevance for this study
are not expected to significantly change upon ruffling or saddling of the
porphyrin moiety unless, perhaps, a different electronic configuration or
spin coupling becomes lower in energy. This is also not likely to be the
case because the line width of the bands in the recorded Raman spectra
are of normal width and indicate a well-defined electronic ground state
and spin coupling. Thus, we believe that planar structures suffice as mod-
els, and to not overcomplicate analysis, degrees of ruffling and saddling
of the porphyrin were not considered.
The procedure to calculate resonance Raman spectra is based on a
theory formulated by Lee and Heller,45,46 and its implementation in
ORCA has been described in detail previously.46,47 In essence, the method
encompasses four steps: (1) calculation of the ground state normal
modes through diagonalization of the Hessian matrix; (2) the calcula-
tion of electronic transitions relevant for the resonance Raman exci-
tation, e.g., by time-dependent density functional theory (TD-DFT)
calculations. Both transition energies and excited state gradients are cal-
culated, (3) estimation of excited state displacements using the ground
state Hessian and the excited state gradients, (4) simulation of the reso-
nance Raman spectra via integration of Heller’s time dependent equa-
tions. The latter step scales linearly with the number of modes and is
therefore strongly preferable to the direct calculation of multidimen-
sional Franck−Condon factors. In practice, however, it is mostly the
well-known limited accuracy of the TD-DFT method in step (2) that
can give rise to significant discrepancies between experiment and theory,
in particular for charge-transfer transitions, which includes the Soret
band of TPP. It is important to adjust the laser wavelength used in the
simulations such that it is in resonance with the calculated state, even if
the latter is significantly off the experimental band position. An advan-
tage of this approach to resonance Raman spectroscopy is that the
excited state displacements can be treated as adjustable variables and
fitted to the experimental spectra. Hence, in this case quantum chem-
istry only serves as an initial guess for the fitting procedure, which is
usually good enough to allow unambiguous assignment of the exper-
imental spectra.
can occur either at the metal or at the ligand, the exact nature of
the electronic ground state of the reduced iron species [FeTPP]−
and [FeTPP]2− has long been controversial. A variety of spec-
troscopic studies, including Mossbauer,11−16 electron paramagnetic
̈
resonance (EPR),16−19 nuclear magnetic resonance (NMR),17,20−22
X-ray,12,15,16 UV/vis,18,20,23,24 and resonance Raman24−33 experi-
ments have been published, partially supporting different elec-
tronic ground state formulations.8−10 Additionally, so far, reso-
nance Raman experiments have been performed in solution,
giving rise to solvent dependency of the electronic ground states.
[FeTPP] and [FeTPP]− are even known to change their spin
state upon axial ligation of solvent molecules.24 Because an under-
standing of the electronic nature of catalytically active species is a
crucial step toward understanding the catalytic mechanism and
designing novel, even more effective and stable catalysts, we recently
reported new spectroscopic insights into the nature of the elec-
tronic ground states of the reduced species using Mossbauer and
̈
XAS techniques combined with quantum chemical calculations of
their spectroscopic properties.34 For [FeTPP]−, an Fe(II) d6 S = 1
spin state with an antiferromagnetically coupled TPP3−• radical
was formulated.34 Upon further reduction, an electronic ground
state best described as Fe(II) d6 S = 1 intermediate-spin antifer-
romagnetically coupled TPP4−•• diradical was found. In fact, in a
recently studied related Fe-NO porphyrin system, repression of
the Soret band was observed upon reduction, while EPR data
showed that the ferrous character of iron was retained, thus indic-
ative of a porphyrin reduction.35
Because the equilibrium bond distances are largely determined
by the electronic structure, they are sensitive to whether reduc-
tion is metal- or ligand-centered. For example, the bond distances
of the porphyrin are expected to increase more if reduction
occurs into a ligand-centered π* orbital than into an iron 3d
orbital. Bond distances are even sensitive to the spin coupling of
unpaired electrons. As such, vibrational spectroscopy in com-
bination with quantum chemistry provides an ideal tool for inves-
tigating the electronic structure of the reduced intermediates
with focus on the porphyrin, whereas the previous Mossbauer
̈
RESULTS
investigation focused on Fe, and for unequivocal confirmation of
the above assignments for [FeTPP]− and [FeTPP]2−. We do so by
performing resonance Raman spectroscopy of the series Fe(III)-
ClTPP, [Fe(II)TPP(MeTHF)2], [Fe(II)TPP], [FeTPP]−, and
[FeTPP]2− as solids, where spin- and oxidation-state-sensitive
marker bands were identified and assigned with the assistance of
computational resonance Raman spectroscopy. Also, we examine
and discuss agreements and discrepancies of our assignment with
literature data, in particular the in-depth experimental and theo-
retical study of FeClTPP by Lehnert et al.25
■
Fe(III)ClTPP. Experimental and calculated Raman and reso-
nance Raman spectra for FeClTPP were previously reported by
Lehnert et al.25 in 2006. The authors presented a complete assign-
ment of all observed vibrations. In particular, for planar iron
tetraphenylporphyrins, four marker bands were identified, which
are known to show characteristic behavior upon either metal cen-
ter or ligand reduction (vide infra).
The laser wavelengths for calculating our resonance Raman
profiles were determined by comparing experimental and calcu-
lated absorption spectra, as indicated by arrows in Figure 1. The
most intense band of the calculated spectrum corresponds to the
Soret band, the weaker bands are traditionally called Q-bands.48
The calculated spectrum was shifted by 2300 cm−1 such that it
has optimal overlap with the bands in the experimental spectrum.
Because the experimentally observed intensities for the Soret
band and Q bands are dominated by π → π* transitions,48 the
calculated electronic states with energy close to the experimental
excitation energies and with π → π* character as identified by visu-
alizing their difference density plots (see Supporting Information)
were selected. However, the difference densities also identified a
transition with chloride-to-iron charge transfer character that
contributes to the Soret band, whose Raman spectrum resembles
the experimental data and which has been assigned before by
magnetic circular dichroism (MCD) spectroscopy by the Lehnert
group.49 The excitation energies giving rise to calculated spectra
EXPERIMENTAL SECTION
■
All synthetic procedures, sample preparation, and detailed descriptions
of resonance Raman experiments, calculations, and model geometries
are given in the Supporting Information.
All calculations were performed with the ORCA program.36 Geom-
etry optimizations were carried out on model structures derived from
experimental X-ray data13,16,37,38 and employ the BP86 functional39
together with the def2-TZVP basis set40,41 and the resolution of identity
(RI) approximation. Counterions were not included in the calculation.
No scaling factors are employed; the BP86 functional was chosen
because it provides good optimized geometries, and a typical accuracy of
the calculated vibrational frequencies amounts to 25 cm−1.42,43
A few words are necessary with respect to using planar model struc-
tures. Porphyrins are known to harbor varying degrees of ruffled and
saddled distortions dependent on their substituents. A recent evaluation
by density functional theory (DFT) methods of the energetic landscape
B
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
band (Lehnert: exp. 1364 cm−1, calcd. 1379 cm−1; here: calcd.
1347 cm−1), which was assigned to a symmetric νsym(pyrrole half-
ring) vibration (see Supporting Information, paragraph 3, for a
detailed visualization).25,52 Marker band D occurs at 1553 cm−1
(Lehnert: exp. 1557 cm−1, calcd. 1598 cm−1; here: calcd. 1547 cm−1)
and corresponds to a combined ν(Cβ−Cβ) + νsym(Cα−Cm) +
δsym(Cβ−H) vibration. Regrettably, the weak spin-state-sensitive
band C of b1g symmetry observed by Lehnert, which was assigned
to a mixed νasym(Cα−Cm) + νasym(Cα−Cβ) + δasym(Cβ−H) por-
phyrin core vibration, has negligible intensity in our resonance
Raman spectra at all employed wavelengths. Therefore, only
bands E, A, and D are further considered.
Lehnert’s calculated bands mostly have a Raman shift (−4 to
+41 cm−1) larger than that of the experimental bands due to the
neglect of anharmonicity effects. Applying our combination of
functional and basis set, we find that our calculated bands are
mostly shifted to lower wavenumbers, the deviation from exper-
imental values ranging from −5 to −18 wavenumbers. Assign-
ment of our calculated bands to the experimentally observed
marker bands can also be rationalized by their intensity pattern.
Upon Soret excitation, only vibrational modes with A1g sym-
metry should be significantly enhanced. The marker bands E, A,
and D are such totally symmetric modes, whereas band C trans-
forms as B1g. Moreover, the intensity is largest for those normal
modes that give rise to large changes to the excited-state geom-
etry with respect to the ground state geometry, for which dif-
ference density plots are given for Soret excitation and Q-band
excitation in the Supporting Information.
Figure 1. (a) Experimental (solvent: 2-MeTHF; inset: concentrated
sample) and (b) calculated absorption spectra of Fe(III)ClTPP. The
calculated spectrum (BP86 functional) was red-shifted by 2300 cm−1 to
match the experimental spectrum. Laser energies (24 096 and 17 825 cm−1)
and corresponding values used in the calculation (see text) are high-
lighted with arrows.
best resembling the experimental spectra are indicated with
arrows in Figure 1b. The excitation energy for the Soret band
amounts to 22 020 cm−1 (vertical excitation; 0,0 transition:
21 485 cm−1) that for the Q-band 14 900 cm−1 (vertical excita-
tion; 0,0 transition: 12 085 cm−1).
Figure 2 shows the experimental and calculated resonance
Raman spectra of FeClTPP for Soret and Q-band excitations.
Reduced Iron Tetraphenylporphyrins. After successfully
validating experiment and theory with literature data for the
markers bands E, A, and D in the oxidized state of Fe(III)ClTPP,
these bands can now be used for elucidation of the electronic
structure and spin state of the reduced FeTPP. The procedure of
computation of the resonance Raman spectra is identical to that
for the oxidized state (for comparison between experimental and
calculated UV−vis spectra of the reduced state, see Supporting
Information). Assignment of the calculated marker bands to the
experimentally observed bands was aided by the fact that the
intensity pattern changes when going from Soret to Q-band exci-
tation. Figure 3 shows the experimental and calculated resonance
Raman spectra for Soret excitation.
[Fe(II)TPP(MeTHF)2]. The resonance Raman spectrum of
[FeTPP(MeTHF) ] (Mossbauer parameters: δ = 0.99 mm/s,
̈
2
ΔEQ = 2.73 mm/s, indicative of high-spin, S = 2, Fe(II)) was
recorded in frozen-solution, which was chosen here to ensure
ligation of the axial solvent molecules because these weakly
bound ligands might dissociate upon laser irradiation. As can be
seen, the experimental and the calculated resonance Raman spec-
tra are in excellent agreement (Figure 3a), especially in the region
between 1200 and 1700 cm−1, regarding both the shift as well as
the relative intensity (high intensity modes have A1g symmetry).
The calculated frequency of marker band E (νbreathing(Fe−N)) shifts
from 381 cm−1 for FeClTPP to 380 cm−1 for [FeTPP(MeTHF)2],
thus staying almost constant despite the change in oxidation state
of the metal center. The experimental Fe−N distances decrease
from 2.071 Å for Fe(III)ClTPP to 2.056 Å for [Fe(II)TPP-
(THF)2].13,37 The calculated bond lengths follow this trend
(2.096 Å for Fe(III)ClTPP to 2.07 Å for [Fe(II)TPP(THF)2]).
In the experimental spectrum, two bands at 367 and 397 cm−1 can
be seen. Also, a band seems to be present in between those two
bands at roughly 382 cm−1. In the calculated spectrum, a weak
band can also be seen next to marker band E at 398 cm−1, which
corresponds to a pyrrole τpyr twist vibration. The normal mode
Figure 2. (a) Calculated and (b) experimental resonance Raman spectra
upon Q-band excitation (561 nm, excitation energy in calculation:
12 085 cm−1) of Fe(III)ClTPP. (c and d) Raman spectra with excitation
in the Soret band (415 nm, excitation energy in calculation: 21 485 cm−1).
Spin-state-sensitive marker bands are labeled E, A, and D according to
Lehnert.25
The oxidation state sensitive marker band E (in the nomen-
clature of Lehnert25) at 391 cm−1 (Lehnert:25 exp. 392 cm−1, calcd.
388 cm−1; here: calcd. 381 cm−1) was previously assigned to a por-
phyrin core deformation26,50,51 but was reassignedbyLehnertetal.
to the totally symmetric νbreathing(Fe−N) vibration25 (for visuali-
zation of displacements, see Supporting Information). The
experimental and calculated shifts for this band as well as its
assignment are consistent with those reported by Lehnert.
The experimentally observed vibrational band A at 1361 cm−1
corresponds to an oxidation- and spin-state sensitive marker
C
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
[Fe(II)TPP]. Resonance Raman experiments and calculations
were further carried out for the planar, intermediate spin
[Fe(II)TPP] (Figure 3b). Marker band E was calculated to lie at
380 cm−1. For Fe(III)ClTPP, it was found that the calculated
intensities of marker bands A and D are in very good agreement
with experiment, while marker band E shows significant devi-
ations in intensity. This is due to the fact that in the experimental
setup, the absorption bands are dominated by π → π* exci-
tations, while LMCT excitations play only a minor role. When
calculating the electronic states belonging to the Soret band,
however, it was observed that the individual transitions consist of
mixtures of LMCT and π → π* excitations. Thus, bands corre-
sponding to Fe−N stretching vibrations (including marker bandE)
are calculated to have higher intensity than they would have upon
pure π → π* excitation. For Fe(III) and Fe(II) high spin, the
calculated Raman shifts of all bands were found to be lower than
the experimentally observed shifts. Therefore, in the experimen-
tal spectrum of [FeTPP], the band with a shift of 394 cm−1 was
here assigned to the symmetric Fe−N breathing vibration.
Marker band A is characterized by a calculated shift of 1353 cm−1.
Upon analysis, this band was found to be comprised of four sepa-
rate vibrations with verysimilarRaman shifts. These different vibra-
tions can be partially resolved in the experimental Soret excited
spectrum and are seen well-resolved in the Q-band excited spec-
trum (see Supporting Information). On the basis of the calcu-
lated intensity patterns for these vibrations for both excitations
and comparison with the experimentally found intensity pattern,
the band 1360 cm−1 was assigned to marker band A. Marker band
D exhibits a calculated frequency of 1559 cm−1. Here, the exper-
imentally observed band at 1566 cm−1 is assigned due to its
change in intensity when comparing Soret excitation to Q-band
excitation.
Figure 3. Experimental and calculated resonance Raman spectra for
(a) [Fe(II)TPP(MeTHF)2], (b) [Fe(II)TPP], (c) [FeTPP][Na(THF)3],
and (d) [FeTPP][Na(THF)3]2 upon Soret excitation. The band marked
with an asterisk is from the reducing agent (Na-anthracenide). Spectra
upon Q-band excitation are included in the Supporting Information.
structure seems to be such that the Fe−N vibration mixes with
other modes so that all bands in this region contain Fe−N
character.
Also of note is that the ground state of [Fe(II)TPP] is calcu-
lated to be of Eg symmetry with a (dxy)2(dxzdyz)3(dz2)1 electronic
configuration whereas [Fe(II)TPP(MeTHF)2] is of high spin
state. Therefore, the differences in frequency of the marker bands
A and D in these two systems (see Table 1), which are repro-
duced by the calculations, are additionally related to the change
in spin state upon presence/absence of the axial ligands.
[FeTPP]−. Owing to the method of preparation, the anionic
complex [FeTPP]− is obtained as the sodium solvate complex
[FeTPP][Na(THF)3]. This species is known to exist in an over-
Marker band A (νsym[pyrrole half-ring] vibration, normal mode
corresponds to a superposition of Fe−N, Cα−Cβ, and Cα−N
stretching frequencies) changes from 1347 cm−1 for Fe(III)-
ClTPP to 1331 cm−1 in the calculated spectrum. On the basis of
the similarities in intensities of experimental and calculated spec-
tra, this vibration is assigned to the experimentally observed band
at 1342 cm−1. As compared to the oxidized state, this would mean
a downshift of 16 cm−1 in the calculated spectra vs a downshift of
19 cm−1 for marker band A, which is in very good agreement.
Marker band D is easily identified in the experimental spec-
trum. Its frequency is sensitive to C−C bond lengths of the por-
phyrin core; in particular, elongated Cβ−Cβ and Cα−Cm bonds
are expected to lead to lower frequencies. The calculated value of
D of 1531 cm−1 is lower by 16 cm−1 as compared to the oxidized
state. In comparison, the experimental band at 1535 cm−1 shows
a similar large intensity and a downshift of 18 cm−1. A complete
overview over all experimental and calculated marker bands for
all investigated species is given in Table 1.
16−18
1
all spin state of S = /2,
giving three possible spin state
formulations:
Calculations were carried out for the pure planar [FeTPP]− as
well as for the sodium solvate complex [FeTPP][Na(THF)3].
In all of the previously described theoretical investigations,
unrestricted Kohn−Sham calculations were carried out. For
[FeTPP]−, however, this leads to convergence of both a and b to
the broken symmetry solution of calculation of c. A restricted
open-shell Kohn−Sham calculation, on the other hand, leads to
a
Table 1. Experimental and Calculated Shifts of the Marker Bands of All Investigated Iron Tetraphenylporphyrin Species
band E [cm−1
]
band A [cm−1
]
band D [cm−1
]
species
FeClTPP
exp
391
calcd
381
exp
1361
calcd
1347
exp
1553
calcd
1547
[FeTPP(MeTHF)2]
[FeTPP] S = 1
[FeTPP]− d6 TPP•
[FeTPP]2− d6 TPP••
[FeTPP]2− d8 S = 0
380(−1)
390(9)
1342(−19)
1360(−1)
1358(−3)
1357(−4)
1331(−16)
1353(6)
1535(−18)
1566(13)
1531(−16)
1559(12)
1549(2)
394(3)
391(0)
382(1)
1351(4)
1538(−15)
1531(−22)
390(−1)
372(−9)
386(5)
1353(6)
1523(−24)
1553(6)
1335(−12)
a
Shifts with respect to FeClTPP are indicated in parentheses.
D
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the iron d7 low-spin configuration a. By comparing the electronic
energies of c and a, the difference in energies is found to be
311 kcal/mol, thus rendering the d7 low spin configuration (a) far
less stable than the broken symmetry solution c, as well as explaining
the unanimous convergence to c. Convergence to b by rotating
orbital energies of the case a restricted open-shell Kohn−Sham
calculation was attempted but not successful. For the broken
symmetry calculation c, the Raman and resonance Raman spec-
tra were then computed and compared to the experimental
results (Figure 3c).
calculated absorption spectrum of case c better resembles the
experimentally observed spectrum than that of case a.
Therefore, resonance Raman calculations were performed for c.
Excitation energies of 457 nm (Soret) and 532 nm (Q-band)
were used in the calculation to simulate experimental excitation
at 415 nm (Soret) and 561 nm (Q-band). Marker band E is cal-
culated to exhibit a shiftof 386 cm−1 for a and 372cm−1 for c. In the
experimental spectrum, a band at 390 cm−1 is present, which is
assigned here to band E.
When calculating the resonance Raman spectrum for a, the
shift of marker band A is calculated to be 1335 cm−1 as compared
to 1353 cm−1 for c. Two experimentally observed bands are pres-
ent with frequencies of 1342 and 1357 cm−1, respectively. Upon
closer inspection of the calculated marker bands, the observed
vibrational bands of both electronic configurations actually con-
sists of two vibrations with marker band A being the one with the
larger shift. Accordingly, in the experimental spectrum, the band
with the larger shift of 1357 cm−1 is here assigned to marker
band A. Because the calculated shifts for all other iron species are
observed to be only slightly lower than the experimental values
(−14 to −7 wavenumbers), the calculated shift for case c is in
much better agreement with experiment than that for case a.
Marker band D is computed to occur at 1553 cm−1 for case a
and 1523 cm−1 for case c. In the experimental spectra, a very broad
band, stretching from 1518 to 1570 cm−1 can be observed. In the
calculations of both possible electronic configurations, the inten-
sity of marker band D is fairly low (see Figure 3, 4). Therefore, in
the experimental spectrum, this marker band can be assigned to
the left-hand side shoulder of the observed vibrational band
instead of its maximum.
Marker band E is calculated to lie at 382 cm−1. Upon excitation
in the Soret band, a broad band has been observed with two max-
ima at around 377 and 391 cm−1. Upon Q-band excitation, only
one vibration is observed experimentally, exhibiting a shift of
391 cm−1. Upon comparison with the calculated intensity pat-
terns, it is here assigned to the corresponding marker band E.
Compared to the experimental spectrum of Fe(III), this band
shows identical shifts and follows the calculated trends. Marker
band A is calculated to exhibit a shift of 1351 cm−1. In the calcu-
lation, this band does not seem to be resonance enhanced upon
Soret excitation and only weakly enhanced upon Q-band exci-
tation. On the basis of the changes in intensity when going from
Soret to Q-band excitation and the intensity pattern of the neigh-
boring peaks, it is here assigned to the experimentally observed
band at 1358 cm−1. Marker band D was identified by its charac-
teristic ν(Cβ−Cβ) + νsym(Cα−Cm) + δsym(Cβ−H) vibration and
exhibits a frequency of 1549 cm−1. On the basis of the intensity
pattern in this region and changes in intensity when going from
Soret to Q-band excitation, it is here assigned to the experi-
mentally observed vibration at 1538 cm−1.
Upon comparison of the calculated shifts of pure [FeTPP]−
and the sodium solvate complex [FeTPP][Na(THF)3], the fre-
quencies of marker bands E, A, and D for the solvate complex
were calculated to be larger by 4 cm−1. The relative shifts do not
change. Furthermore, this small upshift does not affect the observed
general trends when comparing calculated to experimental spec-
tra and reduction trends.
DISCUSSION
■
Several resonance Raman studies of reduced iron tetraphenyl-
porphyrins have been published. Because the reduced species
were prepared in situ by electrochemical reduction, all of these
spectra have been recorded in solution, employing solvents such
as DMF,24,28 butyronitrile, or DMSO.27 Especially [FeTPP] and
[FeTPP]−, however, can change spin state upon axial ligation of
solvent molecules.24 Thus, the frequencies of the marker bands
reported in literature may not always be accurate, in particular
when the nature of the electronic ground state has not been spec-
ified. Still, experimentally, several decades of resonance Raman
spectroscopy of metal porphyrins24,26−31,33 have established
trends for the different marker bands upon reduction provided
that the axial ligand does not change and no high-spin to low-
spin/intermediate-spin crossover, or vice versa, occurs: (1) The
frequency of marker band A is largely unaffected upon reduction
of the macrocycle; (2) marker band A downshifts upon metal
centered reduction; (3) marker band D significantly downshifts
upon macrocycle reduction; and (4) marker band D is largely
unaffected upon metal center reduction, provided the axial ligand
does not change,
To verify these trends and compare them with the observed
trends for our measurements in the solid state, we decided to
calculate the shifts of marker bands A and D for the ligand itself
(H2TPP) upon one and two electron reduction, respectively
(Table 2).
The calculations confirm that both bands A and D follow the
postulated trends upon macrocycle reduction. We then com-
pared these shifts to our calculated as well as experimentally
observed shifts for the [FeTPP] reduction series. First, it can be
seen that the calculated trends upon macrocycle reduction to give
TPP radicals are in very good agreement with the experimentally
[FeTPP]2−. [FeTPP]2−/[FeTPP][Na(THF)3]2 is known to
exist in an overall spin state of S = 0,16 giving three possible spin
state formulations (see Scheme 3):
For a, a spin-restricted Kohn−Sham calculation was per-
formed, converging to the desired Fe(0) d8 low spin state. As has
been already observed for the [FeTPP]− species, b converges to
the broken symmetry scenario of c upon performing unrestricted
Kohn−Sham calculations, as could perhaps be predicted by exam-
ining the spin-coupling scheme in Scheme 3b. Hence, convergence
to b was not achieved. When the single point energies of c and a
are compared, the difference in energy is found to be 6.6 kcal/mol,
thus indicating that the iron d8 low-spin configuration a slightly
less stable than that of c.
The experimental absorption spectrum (see Supporting
Information) shows a very distinctive splitting of the Soret
band into two maxima at 27 905 cm−1 (360 nm) and 22 230 cm−1
(450 nm), which is in agreement with literature.18,20,23,24 The
calculated absorption spectrum for configuration a (Fe(0) d8 low
spin) shows only one strong band, which is assigned to the Soret
band here. Also, the Q-band feature next to the Soret band is
captured but exhibits very low intensity. The calculated absorp-
tion spectrum of c (Fe(II) d6 intermediate spin TPP diradical),
shows two strong bands whose relative positions are in very good
agreement with experiment. Furthermore, the Q-band next to
the Soret band is well-reproduced by the calculation. Thus, the
E
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 2. Calculated Frequencies (cm−1) of Marker Bands A
and D upon H2TPP Reduction and Comparison to Calculated
and Experimental Marker Bands for [FeTPP] Reduction
with iron in the low-spin state. Unfortunately, calculation of res-
onance Raman excitation profiles was not possible for the
restricted-open-shell Fe(I) d7 low-spin calculation. However,
Raman and resonance Raman excitation profiles were obtain-
able for the [Fe(0)TPP]2− species with an Fe(0) d8 low spin
configuration. The calculated frequencies of marker bands A and
D (Table 1) show significant deviations from the experimentally
observed frequencies and trends upon reduction, thus further
supporting ligand reduction rather than reduction of the central
iron.
a
species
A (calcd)
A (exp)
D (calcd)
D (exp)
H2TPP
1355
1357
1352
1353
1351
1353
1543
1526
1500
1559
1549
1523
H2TPP−
H2TPP2−
[FeTPP]
[FeTPP]−
[FeTPP]2−
1360
1358
1357
1566
1538
1531
To provide further confirmation for ligand reduction, we next
compared the results to the shifts which were obtained and
reported in literature for ZnTPP reduction.28,33 ZnTPP,
exhibiting an Zn(II) d10 configuration, is known to be ligand
reduced to give the Zn(II) TPP radical anion. Table 3 shows a
series of vibrational mode shifts upon ZnTPP reduction, among
them the shifts of marker bands A and D. Next to these, our
observed shifts for [FeTPP] reduction are shown. Assignment of
the vibrations apart from the previously presented marker bands
was performed with the aid of our calculations and visualization
of the vibrational modes. For the majority of the investigated
vibrations, the trends in shifts for the FeTPP species are in quite
good agreement with the shifts observed for ZnTPP reduction
and the formation of ligand radicals. Certain deviations in the
shifts can be rationalized because the exact composition of the
vibrational modes might differ upon going from Zn to Fe.
Besides marker bands A and D, the spin-state-sensitive band E
was identified for our [FeTPP] series. Band E was assigned to the
symmetric νbreathing(Fe−N) vibration and is therefore sensitive to
the Fe−N bond lengths of the complex. Lehnert et al.25 pos-
tulated (by comparing different metal porphyrins with each
other) that with decreasing M−N distance, the frequencies of
ν(M−N) are expected to increase. Here, the experimental Fe−N
distances, as obtained by X-ray crystallographic analysis, decrease
from 2.071 Å for Fe(III)ClTPP to 1.966 Å for Fe(II)TPP S = 1.
The calculated bond lengths follow this trend (2.096 Å for
Fe(III) vs 1.982 Å for Fe(II) S = 1). Thus, with decreasing bond
length, the frequency of Fe−N is expected to increase. Indeed, an
increase in shift can be observed, although being quite small
(391 cm−1 for Fe(III) to 394 cm−1 for Fe(II) S = 1). In the series
[FeTPP] → [FeTPP]− → [FeTPP]2−, the changes in Fe−N
bond lengths determined by X-ray crystallographic analyses are
almost negligible (experimental: 1.966 → 1.980 → 1.969 Å; calcu-
lated: 1.982 → 1.998 → 2.015 Å). The experimentally observed
shifts in marker band E (394 → 391 → 390 cm−1), also being
extremely small, thus agree with the similarity in bond lengths of
the reduced species. Thecalculatedshifts of band E (387 → 382 →
372 cm−1) show some deviations from the experimental ones, a
fact that can be rationalized because the calculated Fe−N bond
lengths are slightly higher than the experimentally observed bond
lengths. It is expected that reduction of the iron center would lead
a
The calculations for [FeTPP] reduction are shown for the electronic
configurations with reduced macrocycle.
observed shifts and therefore strongly support macrocycle rather
than metal-centered reduction.
As mentioned before, three conceivable spin states for [FeTPP]−
and [FeTPP]2− were considered (cf. Schemes 2 and 3). For both
Scheme 2. Considered Spin-Coupling Formulations for
[FeTPP]− with Formal Labeling of the Frontier Iron 3d
Orbitals and the TPP Gouterman Orbitals ex and ey
Scheme 3. Considered Schematic Spin-Coupling
Formulations for [FeTPP]2− with Formal Labeling of the
Frontier Iron 3d Orbitals and the TPP Gouterman Orbitals ex
and ey
[FeTPP]− and [FeTPP]2−, calculations reached convergence for
only two spin states. Of the two spin states, the energy of the
Fe(II) d6 intermediate-spin state with a ligand radical (Scheme 2)
or ligand diradical species (Scheme 3) was calculated to be
significantly lower than the energy of the metal-reduced species
Table 3. Comparison of Selected Vibrational Modes (cm−1) of ZnTPP and FeTPP upon Reduction
mode
ZnTPP
[ZnTPP]−
Δν
[FeTPP]
[FeTPP]−
[FeTPP]2−
Δν
+7/0
ϕ(phenyl)
1598
1547
1548
1360
1352
1272
1237
1595
1528
1532
1345
1351
1257
1231
−4
−19
−16
−15
−1
1587
1562
1566
1367
1360
1288
1234
1594
1552
1538
1342
1358
1268
1230
1594
1542
1531
1342
1357
1265
1231
ν(Cα−Cm)
D(ν(Cα−Cm))
ν(Cα−Cβ)
A(ν(Cα−N))
ν(Cm−Ph)
ν(Cm−Ph)
−10/−10
−28/−7
−25/0
−2/−1
−20/−3
−4/+1
−15
−4
F
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
to somewhat elongated bond lengths and therefore more
pronounced downshifts for marker band E.
REFERENCES
■
(1) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M.
Electrocatalytic and Homogeneous Approaches to Conversion of Co2
to Liquid Fuels. Chem. Soc. Rev. 2009, 38, 89−99.
(2) Mondal, B.; Song, J. S.; Neese, F.; Ye, S. F. Bio-Inspired
Mechanistic Insights into Co2 Reduction. Curr. Opin. Chem. Biol. 2015,
25, 103−109.
(3) Bhugun, I.; Lexa, D.; Saveant, J. M. Catalysis of the Electrochemical
Reduction of Carbon Dioxide by Iron(0) Porphyrins: Synergystic Effect
In summary, assignment of the spin- and oxidation-state sensitive
marker bands E, A, and D of planar iron tetraphenylporphyrins in
their reduced forms [FeTPP], [FeTPP]−, and [FeTPP]2− was
accomplished by the aid of resonance Raman spectroscopy with
excitation in the Soret and Q bands. The interpretation of the
data was aided by quantum chemical calculations. On the basis of
the observed trends upon reduction and comparison of our exper-
imental and calculated frequencies with those of H2TPP and ZnTPP,
macrocycle reduction was confirmed for all anionic [FeTPP]
of Weak Bronsted Acids. J. Am. Chem. Soc. 1996, 118, 1769−1776.
̈
(4) Hammouche, M.; Lexa, D.; Saveant, J. M.; Momenteau, M.
Catalysis of the Electrochemical Reduction of Carbon Dioxide by
Iron(0) Porphyrins. J. Electroanal. Chem. Interfacial Electrochem. 1988,
249, 347−51.
species. Whereas our previous Mossbauer study focused on Fe,
̈
Raman spectroscopy employed here allowed clearly the eluci-
dation of the electronic ground state of the catalytically highly
relevant [FeTPP]2− species. For [FeTPP]−, an Fe(II) d6, inter-
mediate spin, iron center with an antiferromagnetically coupled
TPP radical was found. For [FeTPP]2−, the electronic ground
state concerns an Fe(II) d6 intermediate spin iron center antiferro-
magnetically coupled to a TPP diradical tetra-anion (see Scheme 4).
(5) Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Saveant, J. M.
Proton-Coupled Electron Transfer Cleavage of Heavy-Atom Bonds in
Electrocatalytic Processes. Cleavage of a C-O Bond in the Catalyzed
Electrochemical Reduction of Co2. J. Am. Chem. Soc. 2013, 135, 9023−
9031.
(6) Costentin, C.; Passard, G.; Robert, M.; Saveant, J. M. Ultraefficient
Homogeneous Catalyst for the Co2-to-Co Electrochemical Conversion.
Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 14990−14994.
(7) Costentin, C.; Passard, G.; Robert, M.; Saveant, J. M. Pendant Acid-
Base Groups in Molecular Catalysts: H-Bond Promoters or Proton
Relays? Mechanisms of the Conversion of Co2 to Co by Electro-
generated Iron(0)Porphyrins Bearing Prepositioned Phenol Function-
alities. J. Am. Chem. Soc. 2014, 136, 11821−11829.
Scheme 4. Experimental and Calculated Shifts of Markers
Bands A and D upon Reduction from [FeTPP] to [FeTPP]2−,
Exclusively Compatible with an [Fe(II)TPP••]2− Reduction
(8) Mack, J.; Stillman, M. J.; Kobayashi, N. Application of Mcd
Spectroscopy to Porphyrinoids. Coord. Chem. Rev. 2007, 251, 429−453.
(9) Nam, W. High-Valent Iron(Iv)-Oxo Complexes of Heme and
Non-Heme Ligands in Oxygenation Reactions. Acc. Chem. Res. 2007, 40,
522−531.
(10) Groves, J. T. High-Valent Iron in Chemical and Biological
Oxidations. J. Inorg. Biochem. 2006, 100, 434−447.
(11) Corban, G. J.; Hadjikakou, S. K.; Tsipis, A. C.; Kubicki, M.; Bakas,
T.; Hadjiliadis, N. Inhibition of Peroxidase-Catalyzed Iodination by
Thioamides: Experimental and Theoretical Study of the Antithyroid
Activity of Thioamides. New J. Chem. 2011, 35, 213−224.
(12) Collman, J. P.; Hoard, J. L.; Kim, J.; Lang, G.; Reed, C. A.
Synthesis, Stereochemistry, and Structure-Related Properties of Alpha,
Beta, Gamma, Delta-Tetraphenylporphinatoiron(Ii). J. Am. Chem. Soc.
1975, 97, 2676−2681.
We hope our studies afford important requisites for a detailed
understanding of the catalytic mechanism for CO2 reduction by
these complexes.
(13) Li, N.; Su, Z.; Coppens, P.; Landrum, J. X-Ray Diffraction Study of
the Electronic Ground State of (Meso-Tetraphenylporphinato)Iron(Ii).
J. Am. Chem. Soc. 1990, 112, 7294−7298.
ASSOCIATED CONTENT
■
S
* Supporting Information
(14) Lang, G.; Spartalian, K.; Reed, C. A.; Collman, J. P. Mossbauer
̈
The Supporting Information is available free of charge on the ACS
Publications website at DOI: 10.1021/acs.inorgchem.7b03018.
Effect Study of the Magnetic Properties of S = 1 Ferrous
Tetraphenylporphyrin. J. Chem. Phys. 1978, 69, 5424.
(15) Hu, C.; Noll, B. C.; Schulz, C. E.; Scheidt, W. R. Four-Coordinate
Iron(Ii) Porphyrinates: Electronic Configuration Change by Inter-
molecular Interaction. Inorg. Chem. 2007, 46, 619−621.
(16) Mashiko, T.; Reed, C. A.; Haller, K. J.; Scheidt, W. R. Nature of
Iron(1) and Iron(0) Tetraphenylporphyrin Complexes. Synthesis and
Molecular Structure of (Dibenzo-18-Crown-6) Bis(Tetrahydrofuran)
Sodium (Meso-Tetraphenylporphinato)Ferrate and Bis[Tris-
(Tetrahydrofuran)Sodium](Meso-Tetraphenylporphinato)Ferrate.
Inorg. Chem. 1984, 23, 3192−3196.
Synthesis protocols, sample preparation, details of Raman
calculations, normal mode graphics, differences densities,
and Cartesian coordinates of the used model structures
(PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: frank.neese@cec.mpg.de; Tel.: +49 208 306 3656.
ORCID
■
(17) Hickman, D. L.; Shirazi, A.; Goff, H. M. Deuterium Nmr
Spectroscopic Studies of Low-Valent Iron Porphyrin Species. Inorg.
Chem. 1985, 24, 563−566.
Shengfa Ye: 0000-0001-9747-1412
Thomas Weyhermuller: 0000-0002-0399-7999
̈
(18) Lexa, D.; Momenteau, M.; Mispelter, J. Characterization of the
Reduction Steps of Fe(Iii) Porphyrins. Biochim. Biophys. Acta, Gen. Subj.
1974, 338, 151−163.
(19) Cohen, I. A.; Ostfeld, D.; Lichtenstein, B. Characterization of a D7
Iron System. Tetraphenylporphine Iron(I) Anion. J. Am. Chem. Soc.
1972, 94, 4522−4525.
(20) Sinyakov, G. N.; Shulga, A. M. 1h Nmr Spectra and Electronic
Structure of Reduced Iron Porphyrins: Fe(Ii), Fe(I) and Fe(0)
Porphyrins. J. Mol. Struct. 1993, 295, 1−14.
Maurice van Gastel: 0000-0002-1547-6365
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors gratefully acknowledge financial support of this
work by the Max Planck Society.
■
G
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(21) Walker, F. A. Pulsed Epr and Nmr Spectroscopy of Paramagnetic
Iron Porphyrinates and Related Iron Macrocycles: How to Understand
Patterns of Spin Delocalization and Recognize Macrocycle Radicals.
Inorg. Chem. 2003, 42, 4526−4544.
(22) Yamaguchi, K.; Morishima, I. Low-Valent Iron Porphyrins. Nmr
Evidence for Pi Anion-Radical Character in Two-Electron-Reduced
Iron(Iii) Meso- or Beta-Pyrrole-Substituted Porphyrins. Inorg. Chem.
1992, 31, 3216−3222.
(23) Reed, C. A. Iron(I) and Iron(Iv) Porphyrins. Adv. Chem. Ser.
1982, 201, 333−356.
(24) Donohoe, R. J.; Atamian, M.; Bocian, D. F. Characterization of
Singlyreduced Iron (Ii) Porphyrins. J. Am. Chem. Soc. 1987, 109, 5593−
5599.
(25) Paulat, F.; Praneeth, V. K. K.; Naether, C.; Lehnert, N. Quantum
Chemistry-Based Analysis of the Vibrational Spectra of Five-Coordinate
Metalloporphyrins [M(Tpp)Cl]. Inorg. Chem. 2006, 45, 2835−2856.
(26) Burke, J. M.; Kincaid, J. R.; Peters, S.; Gagne, R. R.; Collman, J. P.;
Spiro, T. G. Structure-Sensitive Resonance Raman Bands of
Tetraphenyl and ″Picket Fence″ Porphyrin-Iron Complexes, Including
an Oxyhemoglobin Analog. J. Am. Chem. Soc. 1978, 100, 6083−6088.
(27) Anxolabehere, E.; Chottard, G.; Lexa, D. Highly Reduced Iron
Porphyrins: Uv-Vis and Resonance Raman Spectroelectrochemical
Studies of Fetpp and Fetf5pp. New. J. Chem. 1994, 18, 889−899.
(28) De Silva, C.; Czarnecki, K.; Ryan, M. D. Visible and Rr Spectra of
Low Valent Fe Porphyrins. Inorg. Chim. Acta 1999, 287, 21−26.
(29) Parthasarathi, N.; Hansen, C.; Yamaguchi, S.; Spiro, T. G.
Metalloporphyrin Core Size Resonance Raman Marker Bands
Revisited: Implications for the Interpretation of Hemoglobin Photo-
product Raman Frequencies. J. Am. Chem. Soc. 1987, 109, 3865−3871.
(30) Srivatsa, G. S.; Sawyer, D. T.; Boldt, N. J.; Bocian, D. F. Epr and
Resonance Raman Studies of the (5,10,15,20-Tetraphenylporphinato)
Ferrate(I) Anion: Formation of a Five-Coordinate Pyridine Adduct.
Inorg. Chem. 1985, 24, 2123−2125.
Approximate Coulomb Potentials. Theor. Chem. Acc. 1997, 97, 119−
124.
(42) Jonas, V.; Thiel, W. Theoretical-Study of the Vibrational-Spectra
of the Transition-Metal Carbonyls M(Co)(6) M = Cr, Mo, W, M(Co)
(5) M = Fe, Ru, Os, and M(Co)(4) M = Ni, Pd, Pt. J. Chem. Phys. 1995,
102, 8474−8484.
(43) Jonas, V.; Thiel, W. Density Functional Study of the Vibrational
Spectra of Octahedral Transition-Metal Hexacarbonyls: Neutral
Molecules (M = Cr, Mo, W) and Isoelectronic Ions (M = V, Nb, Ta;
Mn, Re; Fe, Ru, Os; Co, Rh, Ir; Pt; Au). Organometallics 1998, 17, 353−
360.
(44) Conradie, J.; Ghosh, A. Energetics of Saddling Versus Ruffling in
Metalloporphyrins: Unusual Ruffled Dodecasubstituted Porphyrins.
ACS Omega 2017, 2, 6708−6714.
(45) Lee, S. Y.; Heller, E. J. Time-Dependent Theory of Raman
Scattering. J. Chem. Phys. 1979, 71, 4777−4788.
(46) Petrenko, T.; Neese, F. Analysis and Prediction of Absorption
Band Shapes, Fluorescence Band Shapes, Resonance Raman Intensities,
and Excitation Profiles Using the Time-Dependent Theory of Electronic
Spectroscopy. J. Chem. Phys. 2007, 127, 164319.
(47) Petrenko, T.; Neese, F. Efficient and Automatic Calculation of
Optical Band Shapes and Resonance Raman Spectra for Larger
Molecules within the Independent Mode Displaced Harmonic
Oscillator Model. J. Chem. Phys. 2012, 137, 234107.
(48) Gouterman, M.; Stryer, L. Fluorescence Polarization of Some
Porphyrins. J. Chem. Phys. 1962, 37, 2260−2266.
(49) Paulat, F.; Lehnert, N. Detailed Assignment of the Magnetic
Circular Dichroism and Uv-Vis Spectra of Five-Coordinate High-Spin
Ferric Fe(Tpp)(Cl). Inorg. Chem. 2008, 47, 4963−4976.
(50) Oshio, H.; Ama, T.; Watanabe, T.; Kincaid, J.; Nakamoto, K.
Structure Sensitive Bands in the Vibrational Spectra of Metal Complexes
of Tetraphenylporphine. Spectrochim. Acta, Part A 1984, 40A, 863−870.
(51) Burke, J. M.; Kincaid, J. R.; Spiro, T. G. Resonance Raman Spectra
and Vibrational Modes of Iron(Iii) Tetraphenylporphine.Mu.-Oxo
Dimer. Evidence for Phenyl Interaction and Lack of Dimer Splitting. J.
Am. Chem. Soc. 1978, 100, 6077−6083.
(31) Yamaguchi, H.; Soeta, A.; Toeda, H.; Itoh, K. Raman Scattering
Study on Electrochemical Reduction Products of Magnesium, Zinc and
Copper Tetraphenylporphines. J. Electroanal. Chem. Interfacial Electro-
chem. 1983, 159, 347−359.
(52) Li, X. Y.; Czernuszewicz, R. S.; Kincaid, J. R.; Su, Y. O.; Spiro, T. G.
Consistent Porphyrin Force Field. 1. Normal-Mode Analysis for Nickel
Porphine and Nickel Tetraphenylporphine from Resonance Raman and
Infrared Spectra and Isotope Shifts. J. Phys. Chem. 1990, 94, 31−47.
(32) Rush, T. S.; Kozlowski, P. M.; Piffat, C. A.; Kumble, R.; Zgierski,
M. Z.; Spiro, T. G. Computational Modeling of Metalloporphyrin
Structure and Vibrational Spectra: Porphyrin Ruffling in Nitpp. J. Phys.
Chem. B 2000, 104, 5020−5034.
(33) Reed, R. A.; Purrello, R.; Prendergast, K.; Spiro, T. G. Resonance
Raman Characterization of the Radical Anion and Triplet States of Zinc
Tetraphenylporphine. J. Phys. Chem. 1991, 95, 9720−9727.
(34) Romelt, C.; Song, J.; Tarrago, M.; Rees, J. A.; van Gastel, M.;
̈
Weyhermuller, T.; DeBeer, S.; Bill, E.; Neese, F.; Ye, S. Electronic
̈
Structure of a Formal Iron(0) Porphyrin Complex Relevant to Co2
Reduction. Inorg. Chem. 2017, 56, 4745−4750.
(35) Goodrich, L. E.; Roy, S.; Alp, E. E.; Zhao, J. Y.; Hu, M. Y.; Lehnert,
N. Electronic Structure and Biologically Relevant Reactivity of Low-Spin
{Feno}(8) Porphyrin Model Complexes: New Insight from a Bis-Picket
Fence Porphyrin. Inorg. Chem. 2013, 52, 7766−7780.
(36) Neese, F. The Orca Program System. Wiley Interdiscip. Rev.
Comput. Mol. Sci. 2012, 2, 73−78.
(37) Scheidt, W. R.; Finnegan, M. G. Structure of Monoclinic
Chloro(Meso-Tetraphenylporphyrinato)Iron(Iii). Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun. 1989, 45, 1214−1216.
(38) Reed, C. A.; Mashiko, T.; Scheidt, W. R.; Spartalian, K.; Lang, G.
High-Spin Iron(Ii) in the Porphyrin Plane. Structural Characterization
of (Meso-Tetraphenylporphinato)Bis(Tetrahydrofuran)Iron(Ii). J. Am.
Chem. Soc. 1980, 102, 2302−2306.
(39) Becke, A. D. Density-Functional Exchange-Energy Approxima-
tion with Correct Asymptotic Behavior. Phys. Rev. A: At., Mol., Opt. Phys.
1988, 38, 3098−3100.
̈
(40) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
̈
Auxiliary Basis Sets to Approximate Coulomb Potentials. Chem. Phys.
Lett. 1995, 240, 283−290.
(41) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Auxiliary
Basis Sets for Main Row Atoms and Transition Metals and Their Use to
H
DOI: 10.1021/acs.inorgchem.7b03018
Inorg. Chem. XXXX, XXX, XXX−XXX