Stoichiometric O2 oxidation of bis(Thioether)(Octaethylporphyrinato)ruthenium(II) complexes to the corresponding sulfoxide species in acidic media. Structural confirmation of S-bonded sulfoxides

Article abstract of DOI:10.1021/ic9908219

Exposure to O₂ (or air) of a CH₂Cl₂, benzene, or toluene solution containing PhCO₂H and Ru(OEP)(RR′S)₂ (where OEP = the dianion of 2,3,7,8,12,13,17,18-octaethylporphyrin, R = methyl, ethyl, or decyl, and R′ = methyl or ethyl), at ambient conditions, results in the selective oxidation of the axial ligand(s) on the metalloporphyrin complex to the corresponding sulfoxide(s). For example, a CD₂Cl₂ solution of Ru(OEP)(dms)₂ (dms = dimethyl sulfide) and PhCO₂H, exposed to 1 atm of O₂ at ～20 °C for 35 h, is oxidized to Ru(OEP)(dmso)₂, and the intermediates Ru(OEP)(dms)(dmso), [Ru(OEP)(dms)₂][PhCO₂], and Ru(OEP)(dms)(PhCO₂) are identified (s implies sulfur-bonded). Mechanisms invoking in situ formation of H₂O₂, disproportionation of Ruln species, and Ru^IV=O intermediates are proposed for the O₂ oxidation of the thioether ligands. X-ray analysis of Ru(OEP)(Et₂SO)₂ confirms that the sulfoxides are S-bonded.

Full text of DOI:10.1021/ic9908219

Inorg. Chem. 1999, 38, 5579-5587
5579
Stoichiometric O2 Oxidation of Bis(Thioether)(Octaethylporphyrinato)ruthenium(II)
Complexes to the Corresponding Sulfoxide Species in Acidic Media. Structural
Confirmation of S-Bonded Sulfoxides
Andrew Pacheco, Brian R. James,* and Steven J. Rettig†
Department of Chemistry, University of British Columbia, Vancouver, BC, Canada V6T 1Z1
ReceiVed July 9, 1999
Exposure to O2 (or air) of a CH2Cl2, benzene, or toluene solution containing PhCO2H and Ru(OEP)(RR′S)2
(where OEP ) the dianion of 2,3,7,8,12,13,17,18-octaethylporphyrin, R ) methyl, ethyl, or decyl, and R′ )
methyl or ethyl), at ambient conditions, results in the selective oxidation of the axial ligand(s) on the
metalloporphyrin complex to the corresponding sulfoxide(s). For example, a CD2Cl2 solution of Ru(OEP)(dms)2
(dms ) dimethyl sulfide) and PhCO2H, exposed to 1 atm of O2 at ∼20 °C for 35 h, is oxidized to Ru(OEP)-
(dmso)2, and the intermediates Ru(OEP)(dms)(dmso), [Ru(OEP)(dms)2][PhCO2], and Ru(OEP)(dms)(PhCO2) are
identified (s implies sulfur-bonded). Mechanisms invoking in situ formation of H2O2, disproportionation of RuIII
species, and RuIVdO intermediates are proposed for the O2 oxidation of the thioether ligands. X-ray analysis of
Ru(OEP)(Et2SO)2 confirms that the sulfoxides are S-bonded.
Introduction
air oxidation of an acidic solution of Ru(OEP)(RR′S)2 showed
reproducible generation of the corresponding sulfoxide com-
plexes either in benzene, toluene, or CH2Cl2, and several
intermediates were observed. In an earlier paper,6 we described
the syntheses of the Ru(OEP)(RR′S)2, Ru(OEP)(RR′SO)2,
[Ru(OEP)(RR′S)2]+, and [Ru(OEP)(PhCO2)2]- species to be
discussed in the present paper.
Studies from this group have reported previously on the
selective O2 oxidation of thioethers to sulfoxides using the
sterically hindered trans-dioxo species Ru(TMP(O)2 and Ru-
(OCP)(O)2;1 these reactions occur via O-atom transfer processes
via the RuVI(dioxo) species with involvement of RuIV(oxo)
species.2-4 We have also noted earlier that complexes such as
Ru(OEP)(RR′S)2, with a non sterically hindered porphyrin, also
catalyze the autoxidations of thioethers under certain condi-
tions.5,6 Such thioether oxidations are important commercially.2,7
We report here on the mechanism of the stoichiometric O2
oxidation of Ru(OEP)(RR′S)2 complexes, which provides a
framework for understanding the catalytic autoxidation of
thioethers to sulfoxides using such nonhindered species.8
Our previous studies have shown that when solutions of Ru-
(OEP)(RR′S)2 were exposed to air for weeks, the complexes
underwent slow ligand oxidation to give mainly Ru(OEP)-
(RR′S)(RR′SO) and Ru(OEP)(RR′SO)2, along with other minor
products.5,9 The degree of reactivity of the complex and the
product distribution were variable and depended on the thioether,
the solvent, and particularly the dryness of the solvent. However,
Experimental Section
The instrumentation, materials, and methods used for the experiments
are generally described in ref 6, which also details the syntheses of
Ru(OEP)L2 (L ) dms, Et2S, decMS, or the corresponding sulfoxides),
[Ru(OEP)L2][BF4] (L ) dms, Et2S, or decMS), and [Me4N][Ru(OEP)-
(PhCO2)2]. All solvents were thoroughly predried; CD2Cl2 was dried
over 3 Å molecular sieves, while hydrocarbon solvents were dried over
sodium-benzophenone. Unless exposure to O2 was expressly desired,
all manipulations of Ru(OEP) complexes were performed under Ar or
in vacuo.
A typical O2-oxidation experiment was initiated by breaking open
an NMR tube, previously sealed in vacuo and containing the desired
reaction mixture, in a stream of O2 (Union Carbide of Canada, USP
grade, used without further purification). For the experiment described
in most detail below, a CD2Cl2 solution (∼0.5 mL) containing ∼10
mM Ru(OEP)(dms)2 and ∼10 mM PhCO2H was prepared in vacuo, in
an NMR tube fitted with a coaxial Teflon valve (Wilmad Roto Tite).
After the t ) 0 spectrum was obtained in vacuo, the experiment was
initiated by opening the valve in a darkened room and exposing the
reaction mixture to 1 atm of O2 (US Airweld), thoroughly dried by a
column of Sicapent (a P2O5-based drying agent). After the Teflon valve
was reclosed, the NMR tube was vigorously shaken to dissolve the O2
in the solution and then was inserted into the 300 MHz NMR
instrument, thermostated at 23.0 °C. Spectral changes were monitored
for ∼12 h before the tube was removed from the instrument, shaken,
and stored in the dark at 23 °C; further spectra were collected after
∼22 and 40 h. Of note, the results obtained in this specific experiment
did not differ significantly from those obtained using the somewhat
less rigorous method described above, or when concentrations of Ru
and acid were varied. The experiments do not constitute a rigorous
kinetic study, primarily because of the difficulty in maintaining a
constant [O2] in solution within the NMR tube.
* To whom correspondence should be addressed.
† Deceased on October 27, 1998.
(1) Abbreviations used: TMP, dianion of meso-tetramesitylporphyrin;
OCP, dianion of meso-tetra(2,6-dichlorophenyl)porphyrin; OEP, di-
anion of 2,3,7,8,12,13,17,18-octaethylporphyrin; Por, a generic por-
phyrin; dms, dimethyl sulfide; dmso, dimethyl sulfoxide (s or o within
an abbreviation for any sulfoxide ligand implies S- or O-bonded,
respectively); decMS, n-decyl methyl sulfide; decMSO, n-decyl methyl
sulfoxide; R, methyl, ethyl, or decyl; R′, methyl or ethyl; CV, cyclic
voltammetry.
(2) Rajapakse, N.; James, B. R.; Dolphin, D. Catal. Lett. 1989, 2, 219;
Stud. Surf. Sci. Catal. 1990, 55, 109.
(3) Mlodnicka, T.; James, B. R. In Metalloporpyrin-Catalyzed Oxidations;
Montanari, F., Casella, L., Eds.; Kluwer: Dordrecht, The Netherlands,
1994; p 121.
(4) Cheng, S. Y. S.; James, B. R. J. Mol. Catal. A 1997, 117, 91.
(5) James, B. R.; Pacheco, A.; Rettig, S. J.; Ibers, J. A. Inorg. Chem.
1988, 27, 2414.
(6) Pacheco, A.; James, B. R.; Rettig, S. J. Inorg. Chem. 1995, 34, 3477.
(7) Hill, C. L.; Gall, R. D. J. Mol. Catal. A 1996, 114, 103.
(8) (a) Pacheco, A. A. Ph.D. Dissertation, The University of British
Columbia, Vancouver, BC, 1992. (b) Pacheco, A.; James, B. R. To
be published.
(9) Pacheco, A. A. M.Sc. Dissertation, The University of British Columbia,
Vancouver, BC, 1986, and unpublished data.
10.1021/ic9908219 CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/10/1999

5580 Inorganic Chemistry, Vol. 38, No. 24, 1999
Pacheco et al.
1
Figure 1. Selected H NMR spectral changes over time (min) after an acidic CD2Cl2 solution of Ru(OEP)(dms)2 (1) is exposed to 1 atm of O2 at
23.0 °C; 2 is Ru(OEP)(dms)(dmso), 3 is Ru(OEP)(dmso)2, 4 is [Ru(OEP)(dms)2]+, and 5 is Ru(OEP)(dms)(PhCO2). The complete spectra are
provided as Supporting Information, and the corresponding peak assignments for each species are summarized in Table 2.
Table 1. Crystallographic Dataa
ated with the 0.17-occupancy components of the disordered ethyl
groups) were fixed in calculated positions (C-H ) 0.98 Å and BH
)
compound
formula
fw
cryst syst
space group
a, Å
Ru(OEP)(Et2SO)2
C44H64N4O2S2Ru
846.21
monoclinic
P21/n
10.235(2)
10.224(1)
20.796(1)
95.454(8)
2166.2(4)
2
1.2Bbonded atom). A correction for secondary extinction (Zacharaisen type)
was applied, the final value of the extinction coefficient being 2.29(9)
× 10-6. Neutral atom scattering factors for all atoms11 and anomalous
dispersion corrections for the non-hydrogen atoms12 were taken from
standard sources. A complete table of crystallographic data, atomic
coordinates and equivalent isotropic thermal parameters, hydrogen atom
parameters, anisotropic thermal parameters, bond lengths, bond angles,
torsion angles, intermolecular contacts, and least-squares planes are
included as Supporting Information (Tables S1-S8). The crystal
structure confirms that the Ru(OEP)(RR′SO)2 complexes contain
S-bonded ligands as suggested earlier by IR and NMR data.6
b, Å
c, Å
â, deg
V, Å3
Z
F
calc, g/cm3
1.297
21
Cu
T, °C
radiation
λ, Å
1.541 78
41.34
Results and Discussion
µ, cm-1
trans factors (rel)
R(F)
0.61-1.00
Reaction of Ru(OEP)(dms)2 (1) with O2 and PhCO2H in
0.042
1
Rw(F)
0.046
CH2Cl2. Figure 1 shows the H NMR spectral changes over
time after a solution containing ∼10 mM 1 and 10 mM PhCO2H
in CD2Cl2 is exposed to 1 atm of O2 at room temperature (23.0
°C). For simplicity only part of the spectrum is shown, but
corresponding changes are seen throughout the spectrum, from
δ -3 to 25 (see Table 2 and Supporting Information, Figure
S1). When the Et2S or decMS bis-thioether complexes are
exposed to the same conditions, spectral changes analogous to
those illustrated in Figure 1 are observed, at least in the δ 5-25
region; at δ < 5, the spectra of the reaction mixtures are too
complicated to interpret readily.
The sharp singlet at δ 9.60 is due to the meso proton of Ru-
(OEP)(dms)(dmso) (2), and that at δ 9.78 is similarly assigned
to Ru(OEP)(dmso)2 (3).6 Although 2 could not be obtained pure,
in titrations of 1 with dmso, or of 3 with dms, the mixed species
could be unequivocally identified by 1H NMR (Figure 2).
Separate signals are observed at δ -2.07 and -2.87 for
coordinated dmso and dms, respectively, and the presence of
the multiplet for the OEP methylenes shows that the complex
is not symmetrical about the porphyrin plane.5,13 Also, for the
2
a R ) ∑||Fo| - |Fc||/∑|Fo|, Rw ) (∑w(|Fo| - |Fc|)2/∑w|Fo| )1/2
.
X-ray Crystallographic Analysis of Ru(OEP)(Et2SO)2. Crystals
of Ru(OEP)(Et2SO)2 suitable for X-ray analysis were obtained by slow
evaporation of a solution initially containing ∼5 mM Ru(OEP)(Et2SO)2
and ∼1 M Et2SO in benzene. Selected crystallographic data appear in
Table 1. The final unit-cell parameters were obtained by least-squares
on the setting angles for 25 reflections with 2θ ) 71.3-89.5°. The
intensities of three standard reflections, measured every 200 reflections
throughout the data collection, showed only small random fluctuations.
The data were processed10 and corrected for Lorentz and polarization
effects and absorption (empirical, based on azimuthal scans).
The structure was solved by conventional heavy-atom Patterson
methods and was refined by full-matrix least-squares procedures to R
) 0.042 (Rw ) 0.046) for 3332 reflections with I g 3σ(I). The Ru
atom lies on a crystallographic inversion center. The Et2SO ligand was
modeled as 2-fold disordered (83:17) with respect to rotation about
the Ru-S bond. In addition the â carbons of the ethyl groups were
further disordered. The S atom itself may also be disordered, but the
components could not be resolved. The internal geometry of the
Et2SO ligand is unreliable as a result of the disorder. The 0.17-
occupancy O and C atoms were refined with isotropic thermal
parameters, while the remaining non-hydrogen atoms were refined with
anisotropic thermal parameters. Hydrogen atoms (except those associ-
(11) International Tables for X-Ray Crystallography; Kynoch Press:
Birmingham, England, 1974; Vol. IV, pp 99-102.
(12) International Tables for Crystallography; Kluwer Academic Publish-
ers: Boston, MA, 1992; Vol. C, pp 200-206.
(10) teXsan: Crystal Structure Analysis Package; Molecular Structure
(13) James, B. R.; Dolphin, D.; Leung, T. W.; Einstein, F. W. B.; Willis,
A. C. Can. J. Chem. 1984, 62, 1238.
Corporation: The Woodlands, TX, 1995.

Structural Confirmation of S-Bonded Sulfoxides
Inorganic Chemistry, Vol. 38, No. 24, 1999 5581
1
Table 2. The H NMR Shifts in CD2Cl2 for the Ru(OEP) Complexes Shown in Figure 1a
axial ligand signals, δ
OEP signals, δ
PhCO2
CH3
CH2
Hmeso
dms
dmso
Ho
Hm
Hp
Ru(OEP)(dms)2 (1)
1.81, t
1.83, t
1.87, t
1.52c
3.85, q
9.32, s
9.60, s
9.78, s
1.73
-2.66, s
-2.87, s
Ru(OEP)(dms)(dmso) (2)
Ru(OEP)(dmso)2 (3)
3.92, m
3.98, q
23.85
-2.07, s
-2.18, s
[Ru(OEP)(dms)2]+ (4)b
Ru(OEP)(dms)(PhCO2) (5)
[Me4N][Ru(OEP)(PhCO2)2] (6)d
-0.17
-0.49
0.46
16.75, 12.87
8.08
4.07
15.34
17.86
9.84
10.74
8.75
9.35
-0.72
2.72
a The assignments for species 1, 3, 4, and 6 are discussed in ref 6. b The signals are independent of the counterion. c All the signals attributed
to RuIII complexes are broad and lacking in fine structure. d The Me4N+ signal is at δ 5.64.
Figure 2. 1H NMR spectrum of a CD2Cl2 solution of 1 (∼5 mM), dms (∼30 mM), and dmso (∼10 mM). Under these conditions, the major
Ru(OEP) species in solution is 2, which has a characteristic UV/vis λmax at 403 nm.8 Some 3 is also present.
Table 3. 1H NMR Signals (δobs), Assigned to the (1)/(4) Exchange
for the Data of Figure 1 (1 ) RuII(OEP)(dms)2, 4 )
[RuIII(OEP)(dms)2]+), and the Calculated Remaining Mole Fraction
of 1 (NII)a
of RuIII/RuII, confirming that electron transfer between 1 and 4
at 20 °C is very rapid (see also below).14 Of note, the time-
averaged OEP methylene signal for 1 and 4 shifts over time
towards the RuIII position, showing that the ratio of RuIII/RuII
for the bis-dms species is increasing with time. Table 3 lists
the calculated value of the mole fraction, NII ([1]/([1] + [4])),
for each spectrum collected. Rapid electron transfer between
metal centers of metalloporphyrins of this type is well docu-
mented and almost certainly occurs via an outer-sphere process
mediated by porphyrin ligand cation radicals.15 Crystallographic
data for Ru(OEP)(decMS)2 and [Ru(OEP)(decMS)2]+ show no
significant difference in corresponding bond lengths or bond
angles,5,6 and thus within these low-spin systems the minimal
bond reorganizational energies required are expected to give
rapid electron exchange.16,17
δobs (NII)
time (min)
CH3
CH2
Hmeso
SCH3
25
50
1.69 (0.59) 12.1 (0.59)
1.67 (0.52) 13.2 (0.53)
1.65 (0.45) 14.4 (0.47)
1.64 (0.41) 15.4 (0.42)
1.63 (0.38) 16.1 (0.39)
1.62 (0.34) 17.1 (0.34)
1.61 (0.31) 17.4 (0.32)
6.16 (0.58) -1.63 (0.59)
5.74 (0.53) -1.49 (0.53)
5.28 (0.47) -1.33 (0.47)
4.90 (0.42) -1.21 (0.42)
4.64 (0.38) -1.12 (0.38)
4.33 (0.34) -1.01 (0.34)
4.12 (0.32) -0.93 (0.31)
90
145
210
340
460
660
795
1290
2340c
1.59 (0.24) 18.2 (0.28) ∼3.9 (0.3)
-0.83 (0.27)
-0.79 (0.25)
1.59 (0.24) 18.0 (0.28) ∼3.9 (0.3)
1.54 (0.07) 22.2b
22.6b
2.34 (0.08) -0.38 (0.08)
The shapes of the time-averaged OEP methylene signals of
1 and 4 require some comment. In the first two or three spectra
collected, this signal appears noticeably flattened on top,
compared to a typical Gaussian or Lorentzian curve. This is
attributed to the fact that the signal moved significantly
downfield while the spectrum in question was being acquired
(the line shape was monitored during acquisition, and became
broader as more transients were collected). In the spectra
collected between 60 and 300 min, the methylene signal appears
normal for a paramagnetic species, indicating a decrease in the
a Calculated by the formula NII ) (δobs - δIII)/(δIII - δII), where δII
and δIII are the δ values for 1 and 4, respectively. b These methylene
signals are no longer in the fast-exchange range. c Only the methylene
signal is still detectable after 2340 min.
analogous system involving Et2S and Et2SO, the equilibrium
constants as well as the rate constants for the substitution
processes have been determined by stopped-flow spectropho-
tometry.8
The broad signal in Figure 1 that shifts over time is one of
four, attributable to a time-averaged spectrum of [Ru(OEP)-
(dms)2]+ (4) and 1 rapidly exchanging via electron transfer; all
four peak positions for the spectra collected are listed in Table
3. Figure 3 shows the spectrum of pure 4 within the BF4- salt.
When this complex was mixed with 1 in CD2Cl2, only time-
(14) Drago, R. S. Physical Methods for Chemists, 2nd ed.; W. B.
Saunders: Philadelphia, PA, 1992; Chapter 8.
(15) Castro, C. E. In The Porphyrins; Dolphin, D., Ed.; Academic Press:
New York, NY, 1978; Vol. V, Chapter 1.
(16) Wilkins, R. G. Kinetics and Mechanisms of Reactions of Transition
Metal Complexes; VCH: Weinheim, 1991; Chapter 5.
1
averaged H NMR signals could be seen, and the location of
all four signals depended exclusively on the concentration ratio
(17) Stynes, H. C.; Ibers, J. A. Inorg. Chem. 1971, 10, 2304.

5582 Inorganic Chemistry, Vol. 38, No. 24, 1999
Pacheco et al.
Figure 3. 1H NMR spectrum of [Ru(OEP)(dms)2][BF4] (4[BF4]); 20.0 °C in CD2Cl2; S ) solvent.
rate of downfield motion. However, the signal again appears
noticeably broadened in the spectra collected between ∼350
and 800 min, before sharpening up again in the last two spectra.
This second episode of broadening is not observed for the other
signals attributable to the 1 and 4 exchange system. The
necessary condition for detecting separate resonances for the
proton in each of the RuII and RuIII environments is given by τ
> 1/(20.5π∆ν0), where τ is the lifetime of the proton at each
site, and ∆ν0 is the separation of the peaks (in hertz) in the
absence of exchange.14 For the OEP methylene protons ∆ν0 is
20 ppm (see Table 2), i.e., 6000 Hz for the 300 MHz machine
used, and thus τ must be less than 3.8 × 10-5 s in the range
where separate resonances are not seen. When two proton
environments are being exchanged by a second-order process,
such as is the case here, the additional condition k[x] ) τ-1 (k
is the second-order rate constant for the process, and [x] is the
concentration of the less abundant of the exchanging species)
must hold if separate resonances are not seen.14 As discussed
further below, both [1] and [4] decrease with time, and the
broadening observed between ∼350 and 800 min appears to be
due to a decrease of [1] to a point where the methylene signal
begins to slip out of the fast-exchange region. In the range in
which the methylene signal is noticeably broadened, [1] ∼0.5-1
mM, and [4] ∼ 1.5 mM (see below), which yields a rough
estimate of k ∼5 × 107 M-1 s-1 for the second-order rate
constant. None of the other signals attributable to [1] and [4] is
broadened, because ∆ν0 is much smaller for these signals, so
that they remain in the fast-exchange region throughout the time
investigated. Finally, the sharper methylene signals observed
after 800 min, now in the slow-exchange regime, must cor-
respond essentially to 4 only.
2.58, assigned to free dms and dmso, respectively, a cluster of
signals centered at δ ∼7.5 attributable to free benzoic acid and
benzoate (see below), and a broad signal around δ ∼6.5 which
shifts over time and is discussed further below. On the basis of
mass balance and an integral analysis of the corresponding
spectra, the product distributions of 1-5, and of free dms, dmso,
and benzoate, can be determined as a function of time (Figure
4). By the time the first spectrum was collected, [1] had dropped
to ∼20% of its initial concentration; at this stage the RuIII species
4 and 5 account for ∼70% of the total [Ru(OEP)], while 2
accounts for the remaining 10%. Significant amounts of 3 are
not seen until ∼800 min after exposure to O2.
The initial rapid production of 4 and 5 is rationalized by the
following reaction sequence [Ru ) Ru(OEP)]:
2RuII(dms)2 (1) + O2 + 2PhCO2H f
2[RuIII(dms)2]+ (4) + 2PhCO2- + H2O2 (1)
4 + PhCO2- h RuIII(dms)(PhCO2) (5) + dms
(2)
Step one involves electron transfer from RuII to O2 to form RuIII
and superoxide stabilized by protons as HO2, followed by its
disproportionation to give H2O2 and O2. We have previously
established such a mechanism by detection of HO2 within related
bis(phosphine) complexes of Ru(II).13,18 The disproportionation
is fast and irreversible19 and would drive reaction 1 to
completion. The amounts of RuIII species initially generated after
exposure to O2 are almost certainly determined by the amount
of O2 immediately available in solution, as the initial amount
of O2 in the CH2Cl2 (∼8 mM20) is in the range of [1]. Under 1
atm of pressure the NMR tube (approximately 4 mL capacity)
contains ∼30 times more O2 than is required to eventually
oxidize all of 1 to 3.
If a 10-fold excess of dms is added to a CH2Cl2/PhCO2H
solution of 1, the thermal reaction with O2 is completely
inhibited (catalytic O2 oxidation of excess dms under corre-
sponding conditions can occur with a visible light source8). This
implies that an inner-sphere process, with O2 replacing a bonded
dms, occurs prior to electron transfer in reaction 1. There is
precedence for such an inner-sphere mechanism as well as for
The signals at δ 12.95 and 16.80 (but shifted slightly
downfield in the final two spectra) in Figure 1 are attributed to
the OEP methylene protons of the paramagnetic RuIII complex
Ru(OEP)(dms)(PhCO2) (5) (see Table 2 and below). Two
signals are observed because the two different axial ligands make
the anisochronous methylene protons magnetically inequiva-
lent.13 The signals at δ 9.85 and 15.30 in Figure 1 are also
attributed to 5 (Hm and Ho of the coordinated benzoate,
respectively).
Complexes 1-5 account for all of the metalloporphyrin-
1
related H NMR signals observed after the exposure of 1 to
(18) James, B. R.; Mikkelsen, S. R.; Leung, T. W.; Williams. G. M.; Wong,
R. Inorg. Chim. Acta 1984, 85, 209.
(19) Sawyer, D. T.; Valentine, J. S. Acc. Chem. Res. 1981, 14, 393.
(20) IUPAC Solubility Data Series; Battino, R., Kertes, A. S., Eds.;
Pergamon Press: Elmsford, NY, 1981; Vol. 7, p 452.
O2, with the possible exception of some very minor signals at
δ 0.95 and 6.08 seen only in the final spectrum (see Table 2
1
and Figure S1). In addition to these signals, the complete H
NMR spectra (Figure S1) also exhibit singlets at δ 2.09 and

Structural Confirmation of S-Bonded Sulfoxides
Inorganic Chemistry, Vol. 38, No. 24, 1999 5583
case outer-sphere electron transfer from RuII to O2 must also
be kinetically slow.
Experimentally the quotient ([5][dms])/([4][PhCO2-]) is found
to be reasonably constant (average value 8 ( 2; Figure S2), as
required by the equilibrium depicted in eq 2. This implies that
4 and PhCOO- are sufficiently large and hydrophobic to exist
as independent ions (rather than as an ion pair) in CH2Cl2 (ꢀ )
8.93);22 of note, the BF4- salt of 4 has a molar conductivity of
66 Ω-1 cm2 mol-1, compared to 22 Ω-1 cm2 mol-1 for [n-Bu4N]-
[BF4].6
Hydrogen peroxide is known to react with thioethers,
especially in acidic non-hydroxylic solvents,23 and thus the H2O2
produced in reaction 1 could react with free dms (produced in
reaction 2) to give water and dmso:
H2O2 + dms f H2O + dmso
(3)
As previously mentioned, apart from the spectra collected at t
) 0 and t ) 2340 min, all the rest exhibit a broad signal in the
range δ ∼6.2-7.1. This signal initially shifts upfield from δ
6.6 to 6.25 and then shifts downfield while getting broader and
less intense; after 1290 min it is just visible at δ ∼7.1. The
signal is tentatively assigned to H2O, H2O2, and the CO2H proton
of PhCO2H, all rapidly exchanging with each other. The relative
concentrations of these species would then determine the exact
position of the resulting signal at a given time. Hydrogen
peroxide could also react, but presumably less readily, with
coordinated dms (other fates for H2O2, such as reaction with
benzoic acid to form a peracid intermediate, cannot be ruled
out, but there is no evidence to invoke such reactions). Typically
reaction of H2O2 with thioethers results in coproduction of
sulfone;23 however, no sulfone is detected in the current case
(a small quantity of sulfone is produced in the photoactivated
8
O2 oxidation of excess thioether catalyzed by Ru(OEP)(RR′S)2 ).
Ligand exchange between the species 1, 2, and 3 is fast
relative to the observed rate of dms oxidation,8 and so the
concentrations of these species are determined by the relative
affinity of free dms and dmso for Ru. Indeed, the quotient ([2]-
[dms])/([1][dmso]) is essentially constant over the reaction time,
with an average value of 70 ( 15 (Figure S2), which is
comparable to the equilibrium constant of 16.7 ( 0.6 determined
directly for the substitution of Et2S by Et2SO on RuII(Et2S)2.8
For the equilibrium constant ([3][dms])/([2][dmso]), an estimate
of 0.4 is obtained from the data set collected at t ) 1290 min,
close to the value of 0.37 ( 0.01 obtained independently for
the analogous Et2S/Et2SO system.8
Figure 4 shows that, after the initial rapid oxidation step, the
concentrations of the RuIII species 4 and 5 gradually decrease
(Figure 4b), while those of the RuII species 2 and 3 increase
(Figure 4a). In the final spectrum of Figure 1, species 3
comprises about 90% of the product distribution (cf. Figure 4a).
These data are not explained by eqs 1-3, which give no
indication of how a net reduction of the RuIII species is achieved
after their initial production. Considerable insight into the
mechanism of this gradual reduction of the Ru center came from
unsuccessful attempts to prepare pure 5. Figure 5a shows the
1H NMR spectrum obtained for an approximately 1:1 mixture
of 4 and [Me4N][Ru(PhCO2)2] (6) in CD2Cl2, while Figure 5b
shows the spectrum of pure 6 for comparison.6 The major signals
in Figure 5a are attributable to the desired complex 5, and some
Figure 4. (a, b) Product distributions of 1-5, as a function of time,
for the experiment illustrated in Figure 1. (c) Concentrations of free
dms and dmso, as a function of time, for the same experiment. At any
-
given time the concentration of free PhCO2 should equal that of 4.
The concentrations were determined from mass balance and an integral
analysis of the spectra partially shown in Figure 1 and fully provided
as Supporting Information (Figure S1). The data are not considered to
constitute a rigorous kinetic study (see Experimental Section).
direct outer-sphere electron transfer,15,21 although the latter is
extremely unfavorable in the present system:
O2 + e- h O2 , E° ) -0.8 V vs Ag/AgCl (ref 19)
-
4 + e- h 1, E° ) 0.22 V vs Ag/AgCl (ref 6)
These data translate to an equilibrium constant value of about
10-17 for the outer-sphere process RuII + O2 h RuIII + O2
;
-
protonation and removal of the superoxide can nullify the back
reaction of this equilibrium, but the findings imply that in this
(22) Sawyer, D. T.; Roberts, J. L., Jr. Experimental Electrochemistry for
Chemists; John Wiley and Sons: New York, 1974; Chapter 4.
(23) Barnard, D.; Bateman, L.; Cuneen, J. I. In Organic Sulfur Compounds;
Kharasch, N., Ed.; Pergamon Press: New York, NY 1961; Vol. I,
Chapter 21.
(21) Chu, M. M. L.; Castro, C. E.; Hathaway, G. M. Biochemistry 1978,
17, 481.

5584 Inorganic Chemistry, Vol. 38, No. 24, 1999
Pacheco et al.
Figure 5. 1H NMR spectra of (a) a mixture (approximately 1:1) of [Ru(OEP)(dms)2][BF4] (4) and [Me4N][Ru(OEP)(PhCO2)2] (6) ∼1 h after
mixing and (b) pure 6. Both spectra in CD2Cl2 at 20.0 °C, with samples sealed under vacuum; S ) solvent.
presence of BF4 counterion).6 The RuIV/RuIII potential for 5,
containing a single benzoate, would certainly be higher than
0.23 V, but it must be in the range where the equilibrium (4)
can be realized, although lying well to the left-hand side in CH2-
Cl2 solution. Such disproportionation of RuIII into RuII and RuIV
species is known.24
-
signals due to residual 6 are also present because the original
mixture was not exactly 1:1. The key feature is that Figure 5a
also shows small approximately equal amounts of 1 and 2.
Attempts to crystallize out the desired product (5) from the
mixture using hydrocarbon solvents resulted in a dramatic
increase in the concentrations of the RuII species 1 and 2 in the
isolated solid (this time with a predominance of 1) and in the
generation of other unidentified diamagnetic products (not
shown).8a To account for the generation of 1 we speculate that
a “disproportionation” such as the following takes place:
The oxidizing ability of RuIV species via RuIVdO intermedi-
ates as O-atom donors is well established within N4-donor ligand
systems,25-27 and thus speculative pathways, such as Scheme
1, for the subsequent generation of the observed (2) product in
Figure 5a are readily formulated. Here, the RuIVdO species
presumably arises by deprotonation of coordinated OH; subse-
[RuIII(dms)2]+ (4) + RuIII(dms)(PhCO2) (5) h
RuII(dms)2 (1) + [RuIV(dms)(PhCO2)]+ (4)
(24) Farrer, B. T.; Thorp, H. H. Inorg. Chem. 1999, 38, 2497.
(25) Griffith, W. P. Chem. Soc. ReV. 1992, 21, 179.
(26) Cheng, W.-C.; Yu, W.-Y.; Cheung, K.-K.; Che, C.-M. J. Chem. Soc.,
Dalton Trans. 1994, 57 and references therein.
(27) (a) Binstead, R. A.; Stultz, L. K.; Meyer, T. J. Inorg. Chem. 1995,
34, 546 and references therein. (b) Stultz, L. K.; Binstead, R. A.;
Reynolds, M. S.; Meyer, T. J. J. Am. Chem. Soc. 1995, 117, 2520
and references therein.
Earlier CV studies6 have shown that species 6, with two
coordinated benzoates, has a RuIV/RuIII potential of 0.23 V,
which almost exactly matches the RuIII/RuII potentials of
[Ru(OEP)(RR′S)2]+/0 complexes (0.22 V for the species in the

Structural Confirmation of S-Bonded Sulfoxides
Inorganic Chemistry, Vol. 38, No. 24, 1999 5585
Scheme 1. Plausible Route for Oxidation of dms by RuIV
a sulfoxide to RuIII makes the metal much more reducible than
its thioether counterpart (e.g., the reduction potential for the
[RuIII(dmso)2]+/RuII(dmso)2 couple is 0.74 V6); thus, if a species
such as [RuIII(dms)(dmso)]+ is formed during the RuII(dms)2
oxidation sequence, it will be rapidly and preferentially reduced
to the RuII form, either in a “disproportionation” step analogous
to eq 4, or via electron transfer from RuII(OEP)(dms)2. However,
CV data reveal a strong preference of RuIII for thioether rather
than sulfoxide coordination, implying that [RuIII(dms)(dmso)]+
species probably do not play an important role in the overall
reaction. Figure 7 shows the CV of a solution containing
primarily RuII(decMS)(decMSO) prepared by mixing 0.78 mM
RuII(decMS)2, 71 mM decMS, and 25 mM decMSO (cf. Figure
2; also, RuII(RR′S)(RR′SO) complexes have a λmax at 403 nm,
and this can be used to investigate the composition of the above
mixture8). Initially, as the potential is scanned in the positive
direction, a large signal attributed to the oxidation of RuII-
(decMS)(decMSO) is found at 0.59 V, and a minor signal
attributed to RuII(decMS)2 oxidation is found at 0.28 V. As the
potential is scanned back in the negative direction, the 0.21 V
peak due to [RuIII(decMS)2]+ reduction is now the major one,
while that at 0.50 V due to [RuIII(decMS)(decMSO)]+ reduction
is comparatively minor. The observations are explained by the
sequence given in Scheme 2, and show that thioether coordina-
tion to RuIII is preferred over sulfoxide coordination. From
Figure 7 ∆E° for reduction of [RuIII(decMS)(decMSO)]+ by RuII-
(decMS)2 is found to be 0.3 V, which translates to an equilibrium
constant of 105. As mentioned earlier (cf. also Figure S2), the
equilibrium constants for replacement of aliphatic thioethers by
sulfoxides in RuII species have magnitudes of ∼20-70. Com-
bining the equilibrium constants for the two processes yields
an expected value of (2-7) × 10-4 for the replacement of a
quent O-atom transfer to the thioether is well documented.2,3
The conversion from O- to S-bonded dmso (linkage isomer-
ization) is thought to occur in this system via stepwise
dissociation/association of dmso,8 although an intramolecular
process via a π-bonded SdO moiety has been proposed for [Ru-
(NH3)5(dmso)]2+ 28
. In Figure 5a, trace H2O in the solvent could
be sufficient to generate the small quantities of 1 and 2 present.
Presumably in the presence of hydrocarbons, additional mech-
anisms are operative in generating the large amounts of RuII
products observed during the attempted crystallization proce-
dures. The presence of such mechanisms suggests that as the
solvent system becomes progressively less polar, the force
driving RuIII species towards RuII must become substantial.
According to eqs 1-3, the reaction being followed in Figure
1 will ultimately generate in situ 1 equiv of H2O for every 2
equiv of RuII oxidized to RuIII. Thus for this system, eq 4 and
Scheme 1 combined provide a plausible route by which all of
the RuIII species could be reduced back to RuII. The presence
of RuIV via equilibrium 4 could well explain the slight shifts
observed in the signals of 5 toward the end of the stoichiometric
oxidation reaction (cf. Figure 1), when [1] in particular has
decreased to minimal levels. Furthermore, eq 4 and Scheme 1
provide a second route for dms oxidation to dmso, so that the
sequence of reactions 1-4, followed by the reactions of Scheme
1, and dms/dmso ligand exchange processes, together give
eventual formation of Ru(OEP)(dmso)2, via the overall stoichi-
ometry of eq 5.
+
thioether by a sulfoxide in RuIII(RR′S)2 species.
In summary, in CD2Cl2 containing PhCO2H, O2 initially
oxidizes RuII(dms)2 to a mixture of [RuIII(dms)2]+ and RuIII-
(dms)(PhCO2), apparently via an inner-sphere, acid-promoted
step which also produces H2O2. The latter can oxidize dms to
dmso, generating H2O in the process. Subsequently, [RuIII-
(dms)2]+ is reduced back to the corresponding RuII complex
by RuIII(dms)(PhCO2), whose oxidation potential has been
lowered upon replacement of the coordinated thioether by
benzoate. The [RuIV(dms)(PhCO2)]+ thus formed can react with
the water produced in the H2O2 oxidation of dms, to give
OdRuIV(dms) and regenerate free PhCO2H. Finally, like H2O2,
OdRuIV(dms) can also oxidize dms to dmso. The net reaction
(eq 5) is thus oxidation of both dms ligands in 1 by 1 equiv of
O2. The mechanism may be compared to that proposed by
Riley30 for the O2 oxidation of dms to dmso catalyzed by cis-
RuCl2(dmso)4 in alcohols. Here, free H2O2 was suggested to
be formed via an initial 2e O2 oxidation of RuII with direct
generation of RuIV; in our system, the detection of RuIII strongly
favors the 1e process followed by a disproportionation. Riley
also considered reaction 6 for the subsequent reduction of RuIV
RuII(OEP)(dms)2 + O2 h RuII(OEP)(dmso)2
(5)
This net reaction shows that the benzoic acid is not consumed
and acts as a catalyst, specifically to promote reaction 1;
similarly water also acts as a catalyst, being formed in reaction
3 and consumed within Scheme 1. The mechanism implies that
the oxygen of the dmso ligands comes via H2O2 (eqs 1 and 3)
and H2O (Scheme 1). Attempts to investigate the role of water
more extensively, for example, by presaturating the CH2Cl2 with
H2O (or H2O18), were thwarted by the coproduction of signifi-
cant amounts of [Ru(OEP)L]2O (L ) anionic ligand) under these
conditions.8 Figure 6a shows in detail how the benzoate proton
signal positions vary as the oxidation reaction progresses. It is
immediately clear that at 2340 min the benzoate signals are
shifted somewhat upfield from their position at t ) 0. However,
a comparable shift is observed if 1 equiv of dmso is added to
a solution containing only benzoic acid (Figure 6b), which shows
that such a shift is expected when benzoic acid interacts with
RuIV(dms) + H2O h RuII + dmso + 2H+
(6)
dmso (presumably an acid-base interaction29). Thus the H
1
NMR results show that benzoic acid is recovered intact at the
end of the oxidation reaction. At intermediate times, the Hm
and Hp signals shift upfield and come closer together, behaviors
also seen when comparing the H NMR spectrum of the salt
[Me4N][PhCO2] with that of PhCO2H.
One final point for consideration is the possible role of a
species such as [RuIII(dms)(dmso)]+, which could be generated
by ligand exchange as free dmso accumulates. Coordination of
back to RuII; this is essentially the chemistry of Scheme 1,
although we invoke the presence of a RuIVdO intermediate.
We prefer a pathway via this intermediate versus the previously
proposed nucleophilic attack of water on dms,30 because attack
of a thioether at an electrophilic oxo center has been demon-
strated2,25,27 and, in our system, this provides the most obvious
pathway to the detected mixed dms/dmso species 2.
1
(28) Scott, A. Y. N.; Taube, H. Inorg. Chem. 1982, 21, 2542.
(29) Jaswal, J. S.; Rettig, S. J.; James, B. R. Can. J. Chem. 1990, 68, 1808.
(30) (a) Riley, D. P. Inorg. Chem. 1983, 22, 1965. (b) Riley, D. P.; Shumate,
R. S. J. Am. Chem. Soc. 1984, 106, 3179.

5586 Inorganic Chemistry, Vol. 38, No. 24, 1999
Pacheco et al.
1
Figure 6. (a) Changes over time (min) in the H NMR phenyl signals due to free benzoic acid and benzoate, for the same experiment as shown
in Figure 1. (b) Comparison of the first and last spectra shown in Figure 6a with spectra of free benzoic acid in CD2Cl2, with and without an added
equivalent of free dmso, in the absence of Ru(OEP) species.
Scheme 2. Proposed Reactions for the Data of Figure 7
O2, under conditions analogous to those described for the
reactions in CH2Cl2, also results in the production of the Ru
mono- and bis-sulfoxide complexes. The general mechanisms
are probably similar, but the lower polarity of the hydrocarbon
solvents (dielectric constant ≈ 2)31 results in some minor, but
interesting, differences. The reaction is much slower in hydro-
Figure 7. Cyclic voltammogram of a solution (CH2Cl2/[n-Bu4N][BF4])
initially containing 0.78 mM Ru(OEP)(decMS)2, 71 mM decMS, and
25 mM decMSO.
carbons; e.g., after 35 h in a benzene solution otherwise
analogous to the CH2Cl2 solution previously discussed, only
Reaction of Ru(OEP)(RR′S)2 Complexes with O2 and
PhCO2H in Hydrocarbon Solvents. In benzene or toluene
containing PhCO2H, exposure of RuII(dms)2 or RuII(Et2S)2 to
(31) The Handbook for Chemistry and Physics, 62nd ed.; Weast, R. C.,
Astle, M. J., Eds.; Chemical Rubber Company: Boca Raton, FL, 1981;
pp E-53,54.

Structural Confirmation of S-Bonded Sulfoxides
Inorganic Chemistry, Vol. 38, No. 24, 1999 5587
Et2SO systems; the complex [RuBr2(NO)(Et2SO)][µ-Br] 2 con-
tains O-bound sulfoxides.33 Nevertheless, S-bonded sulfoxides
generally have geometry very similar to that of the correspond-
ing free sulfoxide.32 The geometry of the coordinated Et2SO in
7 appears similar to that of the S-bonded Et2SO in some PtII
complexes,34 which is close to tetrahedral at the S atom with
an S-O bond length of 1.48 Å; the disorder problem in the
Et2SO ligands of 7, however, precludes a more accurate
comparison. The Ru-S bond length (2.319 Å) in 7 is in the
range of 2.30-2.36 Å found generally for mutually trans-
sulfoxides in non-porphyrin RuII systems.32
Worth noting is that structural data show that the phthalo-
cyanine complex FePc(dmso)2 (a tetraazaporphyrin species) has
S-bonded sulfoxides,35 while IR data again suggest S-bonded
sulfoxides in RuPc(dmso)2.36 S-Bonded sulfoxide complexes (as
judged by IR data) of the sterically hindered Ru(TMP) system
have also been synthesized in this laboratory.2 To our know-
ledge, the only other structurally characterized bis(sulfoxide)
complexes of metalloporphyrins are of [Fe(OEP)(dmso)2][PF6]37
and bis(tetramethylene sulfoxide)(tetraphenylporphinato)iron-
(III) perchlorate,38 where all the sulfoxides are O-bonded.
Summary
Figure 8. ORTEP view of Ru(OEP)(Et2SO)2. Disorder and hydrogen
atoms are omitted for clarity; 33% thermal ellipsoids are shown.
A mechanism is proposed for the stoichiometric O2 oxidation
of Ru(OEP)(RR′S)2 complexes to Ru(OEP)(RR′SO)2 in CH2-
Cl2 solutions containing PhCO2H. The reaction is initiated by
an inner-sphere process in which a one-electron transfer
generates RuIII and superoxide; the latter disproportionates to
give H2O2 that oxidizes free dms, displaced from the [RuIII-
(OEP)(RR′S)2]+ by substitution with benzoate. “Disproportion-
ation” of the [Ru(OEP)(RR′S)2]+ and Ru(OEP)(RR′S)(PhCO2)
species by a one-electron redox process generates a RuIV
intermediate; this with water formed from the H2O2 oxidation
of dms leads to the formation of a RuIVdO species that transfers
oxygen to a second mole of free dms. The net reaction is
catalytic in PhCO2H and water.
Table 4. Selected Bond Lengths (Å) and Angles (deg) for
Ru(OEP)(Et2SO)2
Ru(1)-S(1)
Ru(1)-N(2)
S(1)-C(19)
2.319(1)
2.051(3)
1.777(8)
Ru(1)-N(1)
S(1)-O(1)
S(1)-C(21)
2.062(3)
1.508(4)
1.776(9)
S(1)-Ru(1)-S(1)a
S(1)-Ru(1)-N(1)a
S(1)-Ru(1)-N(2)a
N(1)-Ru(1)-N(2)
N(2)-Ru(1)-N(2)a
Ru(1)-S(1)-C(19)
O(1)-S(1)-C(19)
C(19)-S(1)-C(21)
180.0
S(1)-Ru(1)-N(1)
S(1)-Ru(1)-N(2)
N(1)-Ru(1)-N(1)a
N(1)-Ru(1)-N(2)a
Ru(1)-S(1)-O(1)
Ru(1)-S(1)-C(21)
O(1)-S(1)-C(21)
91.8(1)
90.3(1)
180.0
89.7(1)
117.3(2)
112.6(3)
106.0(4)
88.2(1)
89.7(1)
90.3(1)
180.0
112.1(3)
106.3(4)
100.9(6)
A crystal structure of Ru(OEP)(Et2SO)2 confirms that the
sulfoxides are bonded via the S atoms.
a Symmetry operation: 1 - x, 1 - y, 1 - z.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for financial support
and Johnson Matthey Ltd. and Colonial Metals Inc. for the loan
of Ru.
∼65% of RuII(dms)2 (1) has reacted, and the product is mainly
RuII(dms)(dmso) (2) with traces of bis(sulfoxide) (3), and RuIII-
(dms)(PhCO2) (5). No [RuIII(dms)2]+ is seen [the 1H resonances
for 1-3 in C6D6 are very similar to those in CD2Cl2 (Table 2),
while some of those of 5 are shifted by up to 1.5 ppm8a]. The
implications are that, in the hydrocarbons, reaction 2 lies far to
the right with a resulting decreased contribution from the
“disproportionation” reaction (4). In the corresponding system
with RuII(Et2S)2, the findings are similar in that no [RuIII(Et2S)2]+
is seen, although more significant amounts of RuIII(Et2S)(PhCO2)
are observed.8
Supporting Information Available: Tables of crystallographic data
collection procedures and parameters, complete atomic coordinates and
thermal parameters, bond distances and angles, torsion angles, and least-
squares planes (Tables S1-S8); the complete 1H NMR spectra (Figure
S1) shown in part in Figure 1, and plots showing equilibrium quotients
for [5][dms]/[4][PhCO2-] and [2][dms]/[1][dmso] (Figure S2). This
material is available free of charge via the Internet at http://pubs.acs.org.
IC9908219
Crystal Structure of Ru(OEP)(Et2SO)2 (7). The synthesis
of 7, along with other Ru(OEP)(RR′SO)2 complexes, from the
[Ru(OEP)]2 precursor has been described earlier;6 the evidence
for coordination via the S-atom of the sulfoxides was solely
solid-state IR data (e.g., νSO 1105 vs 1001 cm-1 for free Et2-
(32) (a) Yapp, D. T. T.; Rettig, S. J.; James, B. R.; Skov, K. Inorg. Chem.
1997, 36, 5635. (b) Calligaris, M.; Carugo, O. Coord. Chem ReV. 1996,
153, 83.
(33) Ferguson, J. E.; Page, C. T.; Robinson, W. T. Inorg. Chem. 1976, 15,
2270.
(34) (a) Kukuskin, V. Y.; Belsky, V. K.; Konovalov, V. E.; Shifrina, R.
R.; Moiseev, A. I.; Vlasova, R. A. Inorg. Chim. Acta 1991, 183, 57.
(b) Belsky, V. K.; Konovalov, V. E.; Kukuskin, V. Y. Acta Crystal-
logr., Sect. C: Cryst. Struct. Commun. 1993, C49, 751.
(35) Calderazzo, F.; Pampaloni, G.; Vitali, D.; Collmati, I.; Dessy, G.; Fares,
J. J. Chem. Soc., Dalton Trans. 1980, 1965.
(36) Dolphin, D.; James, B. R.; Murray, A. J.; Thornback, J. R. Can. J.
Chem. 1980, 58, 1125.
(37) Muthusamy, M.; Andersson, L. A.; Sun, J.; Loehr, T. M.; Thomas,
C. S.; Sullivan, E. P., Jr.; Thomson, M. A.; Long, K. M.; Anderson,
O. P.; Strauss, S. H. Inorg. Chem. 1995, 34, 3953.
(38) Mashiko, T.; Kastner, M. E.; Spartalian, K.; Scheidt, W. R.; Reed, C.
A. J. Am. Chem. Soc. 1978, 100, 6354.
1
SO), with H NMR data subsequently being ascribed to the
species. The X-ray data for 7 confirm that the sulfoxide ligands
are S-bonded (Figure 8). Selected bond lengths and angles are
given in Table 4.
There is nothing remarkable about the structure, although it
is perhaps the first for a metalloporphyrin containing an
S-coordinated sulfoxide. The variable orientations for the ethyl
groups of the OEP are commonly seen,5 and the bond lengths
and angles for the Ru(OEP) moiety are normal.5,6 There is a
plethora of structural data for non-porphyrin RuII S-bonded
sulfoxide complexes,32 but none, to our knowledge, of S-bonded