Structural determination of ruthenium-porphyrin complexes relevant to catalytic epoxidation of olefins

Article abstract of DOI:10.1021/ic048587w

A reproducible synthesis of a competent epoxidation catalyst, [Ru ^VI(TPP)(O)₂] (TPP = tetraphenylporphyrin dianion), starting from [Ru^II(TPP)(CO)L] (L = none or CH3OH), is described. The molecular structure of the complex

Full text of DOI:10.1021/ic048587w

Inorg. Chem. 2005, 44, 2039−2049
Structural Determination of Ruthenium
to Catalytic Epoxidation of Olefins
−Porphyrin Complexes Relevant
Emma Gallo,† Alessandro Caselli,† Fabio Ragaini,† Simone Fantauzzi,† Norberto Masciocchi,‡
Angelo Sironi,§ and Sergio Cenini*,†
Dipartimento di Chimica Inorganica, Metallorganica e Analitica, UniVersita` di Milano,
and ISTM-CNR, Via Venezian 21, 20133 Milano, Italy, Dipartimento di Scienze Chimiche ed
Ambientali, UniVersita` dell’Insubria, Via Valleggio 11, 22100 Como, Italy,
and Dipartimento di Chimica Strutturale e Stereochimica Inorganica, UniVersita` di Milano,
and ISTM-CNR, Via Venezian 21, 20133 Milano, Italy
Received October 8, 2004
A reproducible synthesis of a competent epoxidation catalyst, [RuVI(TPP)(O)2] (TPP
starting from [RuII(TPP)(CO)L] (L
none or CH3OH), is described. The molecular structure of the complex was
) tetraphenylporphyrin dianion),
)
determined by using ab initio X-ray powder diffraction (XRPD) methods, and its solution behavior was in detail
investigated by NMR techniques such as PGSE (pulsed field gradient spin-echo) measurements. [RuIV(TPP)-
(OH)]2O, a reported byproduct in the synthesis of [RuVI(TPP)(O)2], was synthesized in a pure form by oxidation of
[RuII(TPP)(CO)L] or by a coproportionation reaction of [RuVI(TPP)(O)2] and [RuII(TPP)(CO)L], and its molecular
structure was then determined by XRPD analysis. [RuVI(TPP)(O)2] can be reduced by dimethyl sulfoxide or by
carbon monoxide to yield [RuII(TPP)(S-DMSO)2] or [RuII(TPP)(CO)(H2O)], respectively. These two species were
characterized by conventional single-crystal X-ray diffraction analysis.
Introduction
been achieved in the development of new generations of
synthetic metalloporphyrins which are able to reproduce and
mimic all the different heme-enzyme-mediated reactions
such as oxygenation, oxidation, oxidative chlorination, and
dismutation.1c,2 Among oxidation reactions catalyzed by
transition metal complexes, epoxidation of olefins is still one
of the most attractive and is largely used in organic
synthesis:4 in a single step, two C-O bonds are created with
high regio- and/or stereoselectivity control.5 The transition
metal-catalyzed epoxidation of olefins by alkyl hydroper-
oxides is currently performed on an industrial scale, i.e., the
propene epoxidation with tBuOOH and a molybdenum-based
catalyst (the Halcon process)6 or the production of fine
chemicals by the Sharpless reaction.5b
Oxidation chemistry at metalloporphyrin centers occupies
a prominent role in current research in the field of chemical
and biological catalysis.1 Among the many known species,
interest in ruthenium porphyrins2 originally stemmed, at least
in part, from attempts to model certain aspects of cytochrome
P450 systems.3 In the past two decades, much progress has
* To whom correspondence should be addressed. E-mail:
Sergio.Cenini@unimi.it. Fax: (+39) 02-50314405.
† Dipartimento di Chimica Inorganica, Metallorganica e Analitica,
Universita` di Milano, and ISTM-CNR.
‡ Universita` dell’Insubria.
§ Dipartimento di Chimica Strutturale e Stereochimica Inorganica,
Universita` di Milano, and ISTM-CNR.
(1) (a) Meunier, B. Chem. ReV. 1992, 92, 1411-1456. (b) Montanari, F.,
Casella, L., Eds. Metalloporphyrins Catalysed Oxidations; Kluwer
Academic: Dordrecht, The Netherlands, 1994. (c) Sono, M.; Roach,
M. P.; Coulter, E. D.; Dawson, J. H. Chem. ReV. 1996, 96, 2841-
2887. (d) McLain, J. L.; Lee, J.; Groves, J. T. In Biomimetic Oxidations
Catalyzed by Transition Metal Complexes; Meunier, B., Ed.; Imperial
College Press: London, 2000; pp 91-169. (e) Groves, J. T.; Shalayev,
K.; Lee, J. Biochemistry and Binding: Activation of Small Molecules.
In The Porphyrin Handbook; Kadish, K. M., Smith, K. M., Guilard,
R., Eds.; Academic Press: New York, 2000; Vol. 4, pp 17-40.
(2) Ezhova, M. B.; James, B. R. In AdVances in Catalytic ActiVation of
Dioxygen by Metal Complexes; Simandi, L., Ed.; Kluwer Academic:
Dordrecht, The Netherlands, 2002; pp 1-77.
Metalloporphyrin-catalyzed alkene epoxidation reactions
are often characterized by remarkable regio- and stereose-
(3) Ortiz de Montellano, P. R., Ed. Cytochrome P450: Structures,
Mechanism and Biochemistry; Plenum: New York, 1995.
(4) (a) Jorgensen, K. A. Chem. ReV. 1989, 89, 431-458. (b) Arends, I.
W. C. E.; Sheldon, R. A. Top. Catal. 2002, 19, 133-141.
(5) (a) Hanson, R. M. Chem. ReV. 1991, 91, 437-476. (b) Johnson, R.
A.; Sharpless, K. B. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.;
VCH: Weinheim, Germany, 1993; pp 103-158.
(6) Sheldon, R. Bull. Soc. Chim. Belg. 1985, 94, 651-670.
10.1021/ic048587w CCC: $30.25
Published on Web 02/15/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 6, 2005 2039

Gallo et al.
Scheme 1. Reported Mechanism for the Reaction of Molecular Oxygen with Ruthenium Porphyrins14
lectivities. A breakthrough in the field is the set of results
achieved in 1985 by Groves and Quinn in the aerobic
epoxidation of olefins catalyzed by [RuVI(TMP)(O)2] [TMP
) tetrakis(2,4,6-trimethylphenyl)porphyrin dianion] at room
temperature and ambient pressure.7 Since then, several high
valent dioxoruthenium(VI) porphyrin complexes have been
isolated and found to be excellent catalysts for hydrocarbon
oxidation.8 Only in one case has a Ru(porphyrin)(O)2
complex been structurally characterized.8c Systematic struc-
tural variation of the porphyrin ligand has proved to be a
useful strategy for achieving electronic and steric tuning of
the catalyst. Nevertheless, the structure-reactivity relation-
ship of oxometalloporphyrin complexes remains less under-
stood,2 and little is known about possible deactivation routes
of the catalyst.9
In the proposed and generally accepted mechanism, [RuVI-
(porphyrin)(O)2] transfers one oxygen atom to the olefinic
substrate and the resulting monooxoruthenium complex,
[RuIV(porphyrin)(O)], disproportionates into the epoxidiz-
ing agent, [RuVI(porphyrin)(O)2], and [RuII(porphyrin)]; the
latter is in turn reoxidized and re-enters the catalytic cycle.7b
The elusive [RuIV(TMP)(O)] has been spectroscopically
observed for the first time during the oxidation reaction of
[RuII(TMP)(CH3CN)2],10 and since then, a few [(L)RuIV-
(porphyrin)(O)] species have been isolated and/or character-
ized (L ) EtOH, THF, or OPPh3).11 Thus, the chemistry of
ruthenium dioxoporphyrin complexes continues to attract
much attention, despite the fact that it has been shown in
recent years that when 2,6-dichloropyridine N-oxide is
employed as the oxidant in place of dioxygen, the catalyti-
cally active species are ruthenium(V) porphyrin oxo com-
plexes, instead of ruthenium(VI) species.12,13 In this con-
text, a distinction must be made between complexes of
sterically hindered porphyrin dianions such as TMP and
those of nonsterically demanding ones such as octaethylpor-
phyrin dianion (OEP) and tetraphenylporphyrin dianion
(TPP). Collman et al.14 proposed a mechanism for the
reaction of molecular oxygen with ruthenium porphyrins,
which is outlined in Scheme 1. After reaction of dioxygen
with a RuII(porphyrin) molecule, the dioxygen adduct so
formed reacts with a second RuII(porphyrin) molecule lead-
ing to a µ-peroxo dimer. Homolytic cleavage of the O-O
bond then leads to the formation of two unstable [RuIV-
(10) Groves, J. T.; Ahn, K.-H. Inorg. Chem. 1987, 26, 3831-3833.
(11) (a) Leung, W.-H.; Che, C.-M. J. Am. Chem. Soc. 1989, 111, 8812-
8818. (b) Groves, J. T.; Roman, J. S. J. Am. Chem. Soc. 1995, 117,
5594-5595. (c) Bailey, A. J.; James, B. R. Chem. Commun. 1996,
2343-2344. (d) Cheng, S. Y. S.; James, B. R. J. Mol. Catal. A: Chem.
1997, 117, 91-102.
(12) Sharma, P. K.; de Visser, S. P.; Ogliaro, F.; Shaik, S. J. Am. Chem.
Soc. 2003, 125, 2291-2300.
(7) (a) Groves, J. T.; Quinn, R. Inorg. Chem. 1984, 23, 3844-3846.
(b) Groves, J. T.; Quinn, R. J. Am. Chem. Soc. 1985, 107, 5790-
5792.
(13) (a) Ohtake, H.; Higushi, T.; Hirobe, M. Tetrahedron Lett. 1992, 33,
2521-2524. (b) Ohtake, H.; Higushi, T.; Hirobe, M. J. Am. Chem.
Soc. 1992, 114, 10660-10662. (c) Ohtake, H.; Higuchi, T.; Hirobe,
M. Heterocycles 1995, 40, 867-903. (d) Higuchi, T.; Hirobe, M.
J. Mol. Catal. A: Chem. 1996, 113, 403-422. (e) Groves, J. T.;
Bonchio, M.; Carofiglio, T.; Shalyaev, K. J. Am. Chem. Soc. 1996,
118, 8961-8962. (f) Gross, Z.; Ini, S.; Kapon, M.; Cohen, S.
Tetrahedron Lett. 1996, 37, 7325-7328. (g) Gross, Z.; Ini, S. J.
Org. Chem. 1997, 62, 5514-5521. (h) Gross, Z.; Ini, S. Inorg. Chem.
1999, 38, 1446-1449. (i) Zhang, R.; Yu, W.-Y.; Wong, K.-Y.;
Che, C.-M. J. Org. Chem. 2001, 66, 8145-8153. (j) Le Maux, P.;
Lukas, M.; Simonneaux, G. J. Mol. Catal. A: Chem. 2003, 206, 95-
103.
(8) Selected examples: (a) Leung, W.-H.; Che, C.-M.; Yeung, C.-H.; Poon,
C.-K. Polyhedron 1993, 12, 2331-2334. (b) Liu, C.-J.; Yu, W.-Y.;
Peng, S.-M.; Mark, T. C. W.; Che, C.-M. J. Chem. Soc., Dalton Trans.
1998, 805-812. (c) Lai, T.-S.; Zhang, R.; Cheung, K.-K.; Kwong,
H.-L.; Che, C.-M. Chem. Commun. 1998, 1583-1584. (d) Liu, C.-J.;
Yu, W.-Y.; Che, C.-M.; Yeung, C.-H. J. Org. Chem. 1999, 64, 7365-
7374. (e) Zhang, R.; Yu, W.-Y.; Lai, T.-S.; Che, C.-M. Chem.
Commun. 1999, 409-410. (f) Zhang, R.; Yu, W.-Y.; Sun, H.-Z.; Liu,
W.-S.; Che, C.-M. Chem.sEur. J. 2002, 8, 2495-2507. (g) Tavare`s,
M.; Ramasseul, R.; Marchon, J.-C.; Valle´e-Goyet, D.; Gramain, J.-C.
J. Chem. Res., Synop. 1994, 74.
(9) (a)Scharbert, B.; Zeisberger, E.; Paulus, E. J. Organomet. Chem. 1995,
493, 143-147. (b) Cunningham, I. D.; Danks, T. N.; Hay, J. N.;
Hamerton, I.; Gunathilagan, S.; Janczak, C. J. Mol. Catal. A: Chem.
2002, 185, 25-31.
(14) (a) Collman, J. P.; Barnes, C. E.; Brothers, P. J.; Collins, T. J.; Ozawa,
T.; Gallucci, J. C.; Ibers, J. A. J. Am. Chem. Soc. 1984, 106, 5151-
5163. (b) Collman, J. P.; Brauman, J. I.; Fitgerald, J. P.; Sparanpany,
J. W.; Ibers, J. A. J. Am. Chem. Soc. 1988, 110, 3486-3495.
2040 Inorganic Chemistry, Vol. 44, No. 6, 2005

Structure of Ruthenium-Porphyrin Complexes
Chart 1. Ruthenium Tetraphenylporphyrinate Complexes
(porphyrin)(O)] complexes. Depending on the steric proper-
ties of the porphyrin, either a trans-dioxo-RuVI(porphyrin)
complex (through disproportionation) or a µ-oxo-RuIV-
(porphyrin) dimer is formed, by formal addition of a water
molecule (Scheme 1).
In addition to the steric properties of the porphyrin, the
solvent also plays a key role in driving the reaction in one
direction or the other. Indeed, it has been shown that it is
possible to obtain trans-dioxo complexes of nonsterically
hindered porphyrins such as TPPH2 and OEPH2 by chemical
oxidation with m-CPBA (m-chloroperoxybenzoic acid) of the
starting carbonyls when working in the presence of weak
coordinating solvents such as alcohols.11a,15
Since the early work of Masuda and co-authors,16 several
independent syntheses of [RuIV(porphyrin)(OH)]2(O) com-
plexes have been reported.17 Despite these extensive inves-
tigations, very little is known about the possible pathways
leading to these catalytically inert µ-oxo-RuIV(porphyrin)
dimers under the conditions employed in catalysis or about
the exact role played by the solvent or by the oxidant (in its
pristine or reduced forms).
Our interest has been focused on the synthesis and
structural characterization of the highly reactive [RuVI(TPP)-
(O)2] species. Several previously published papers deal with
the synthesis of trans-dioxo complexes of nonsterically
hindered porphyrins, mainly employing [RuII(porphyrin)-
(CO)(MeOH)] as the starting material and carrying out the
oxidation with oxidants such as m-CPBA, TBHP (tert-
butylhydroperoxide), or PhIO.11a,15,18 In our hands and in the
case of TPP as the porphyrin, however, none of the published
methods showed a good reproducibility in the yield or purity
of the desired product, [RuVI(TPP)(O)2].
The aim of this work was to provide a reproducible
synthesis of [RuVI(TPP)(O)2], and to achieve a deeper
understanding of its structure and reactivity. Key to the
success of this goal was the use X-ray powder diffraction
(XRPD) methods, both to solve ab initio the structure of
complexes for which single crystals could not be grown
and to assess the purity of the same compounds. XRPD, as
a highly accurate structural and analytical tool, was in-
deed employed in the characterization of, among others,
the [RuVI(TPP)(O)2] species, which, as later discussed,
exhibits nontrivial behavior in solution and is unstable in
many solvents. The numbering scheme for the relevant
ruthenium tetraphenylporphyrinate derivatives is given in
Chart 1.
Figure 1. Schematic drawing of the molecular structure of the [RuVI
(TPP)(O)2] species (3) showing the two oxo fragments trans to each other.
Selected distances for 3: 2.14(6) Å for the Ru-N distance and 1.74(1) Å
for the Ru-O distance.
-
Results and Discussion
Synthesis of [RuVI(TPP)(O)2] (3) and [RuIV(TPP)(OH)]2-
(O) (4). Che and co-workers reported that 3 can be obtained
by oxidation of [Ru(TPP)(CO)(MeOH)] (2) by m-CPBA in
15,18
EtOH11a or in a EtOH/CH2Cl2
mixture as solvents, but
in the latter case, some quantitative details of the experi-
mental procedures (such as solvent amounts) have never been
published. We have performed an extensive investigation of
the effect of the solvent composition and the oxidant amount.
Under the best experimental conditions (5:1 EtOH/CH2Cl2
mixture and 20:1 m-CPBA/2), complex 3 was formed
selectively and was isolated in a 78% yield by simple filtra-
tion of the resulting suspension without the need for any
further purification. Single crystals of 3 could not be grown,
at least in part because it decomposes on long standing in
most solvents which dissolve it. However, the structure of
this complex could be determined by ab initio XRPD on the
crude (polycrystalline) solid (Figure 1).
The perfect coincidence between the experimental and
simulated XRPD patterns also allows to state that the material
obtained under the best experimental conditions has a very
high purity, no amorphous halo, and no extra peaks being
observed (see the Experimental Section).
The outcome of some of the other reactions performed in
this optimization study is worth mentioning.
(a) If ethanol alone is employed as solvent, a viscous
suspension of a very fine material is obtained, from which
the product could not be efficiently separated by either
filtration or centrifugation. Addition of CH2Cl2 to this mixture
after the reaction was completed improved filterability, but
the product was found to contain only a small amount of 3,
accompanied by several byproducts.
(b) If CH2Cl2 alone is employed as a solvent, the product
is not 3, but the µ-oxo dimeric complex, [RuIV(TPP)(OH)]2-
(O) (4). The complex was isolated, as a pure polycrystalline
phase, in ∼60% yield and was, again, structurally character-
ized by ab initio XRPD (Figure 2). Several complexes of
the general [Ru(porphyrin)OR]2O type (R ) H, alkyl, or aryl)
have been described in the literature, and the X-ray structures
of two of them have been reported.14a,16 However, despite
the fact that 4 has been said to be a byproduct in the synthesis
of 3, it has never been isolated in a pure form, nor was it
fully characterized before.
(15) Ho, C.; Leung, W.-H.; Che, C.-M. J. Chem. Soc., Dalton Trans. 1991,
2933-2939.
(16) Masuda, H.; Taga, T.; Osaki, K.; Sugimoto, H.; Mori, M.; Ogoshi, H.
J. Am. Chem. Soc. 1981, 103, 2199-2203.
(17) Mosseri, S.; Neta, P.; Hambright, P. J. Phys. Chem. 1989, 93, 2358-
2362 and reference therein.
(18) Che, C.-M.; Yu, W.-Y. Pure Appl. Chem. 1999, 71, 281-288.
Inorganic Chemistry, Vol. 44, No. 6, 2005 2041

Gallo et al.
Scheme 2. Alternative Pathways Leading to the Formation of
[RuIV(TPP)(OH)]2O (4)
Figure 2. Schematic drawing of the molecular structure of the [RuIV
-
is shifted on the side of the hydroxo complex even in the
presence of an excess of ethanol.
(TPP)(OH)]2O species (4) showing the two hydroxo and the µ-oxo fragments
trans to each other. Selected distances for 4: 2.13(1) Å for the Ru-N
distance, 1.90(6) Å for the Ru-O distance, and 2.11(3) Å for the Ru-OH
distance.
Moreover, to better understand the reaction mechanism,
we also investigated the reaction of 2 with 3. When
equimolar amounts of 2 and 3 are dissolved in CH2Cl2 at
room temperature, no reaction occurs, at least for several
hours. However, if a small amount of hydrochloric or
m-chlorobenzoic acid is added, a reaction yielding the µ-oxo
dimer 4 is observed (path b in Scheme 2). The promotional
effect of acids on the reactivity of related porphyrin
complexes has previously been noted,13b-d,i but to the best
of our knowledge, the coproportionation reaction of 2 and 3
to give 4 has never been reported.
The aforementioned data, taken together, allow us to shed
some light on the mechanism of the formation of 3 and 4 by
reaction of 1 or 2 with m-CPBA. In the literature, the reaction
has been proposed to start with the oxidation of 1 or 2 to a
mono-oxo-RuIV intermediate. In the absence of coordinating
ligands, this complex dimerizes to afford 4 (path a in Scheme
2), also in agreement with the mechanism proposed for the
reaction of 1 with dioxygen (Scheme 1). However, THF is
a coordinating solvent as ethanol, yet it only partly inhibits
the formation of 4. Therefore, there must be a second role
for ethanol which explains why in its presence the reaction
can become completely selective. During the synthesis, acids
(m-CPBA and its deoxygenated analogue m-chlorobenzoic
acid) are present in large amounts, and our results imply that
the formed 3 should immediately react with the remaining
1 or 2 if either is present in solution (path b in Scheme 2).
The second role of ethanol is now easily identified when
we consider that 1 and 2 are poorly soluble in this solvent
and, worth noting, 3 is virtually insoluble in it. Thus, ethanol
acts as a precipitating solvent for the product and prevents
it from reacting with the remaining starting material. THF,
which is not a precipitating agent for 3, is, therefore, less
effective in preventing the formation of 4.
(c) Provided the correct solvent mixture is employed, 3
could equally well be synthesized starting from [Ru(TPP)-
(CO)] (1) instead of its methanol adduct [Ru(TPP)(CO)(CH3-
OH)] (2). Together with the result of the experiment
described just above, this implies that the presence of a sixth
ligand in the starting complex is not influential and that the
solvent composition is the important parameter in driving
the reaction toward the formation of 3 or 4.
(d) If THF is employed in a mixture with CH2Cl2 in place
of ethanol under the optimized conditions, 3 is still the major
product, but a large amount of 4 is also formed.
(e) In general, a large excess of oxidant is required to
consume all of the starting 1 or 2. The minimum m-CPBA:
Ru molar ratio also depends on the solvent composition and
is 15 in neat ethanol, but is reduced to ∼10 in a 1:10 CH2-
Cl2/EtOH mixture. In neat CH2Cl2, a 7:1 molar excess of
m-CPBA with respect to ruthenium is sufficient to consume
all of the starting complex, but as previously mentioned, the
only isolated product was 4. In CH2Cl2, 4 remained the only
product even if a larger excess (20:1) of m-CPBA was
employed, indicating that the outcome of the reaction is not
dictated by a low concentration of the acid in solution. Under
our experimental conditions (m-CPBA as the oxidizing
agent), 4 could not be further oxidized to 3.
(f) When nonideal solvents or oxidant amounts were
employed, the isolated solid contained a mixture of 3 and 4.
It should be recalled that [Ru(TPP)(OR)]2O complexes (R
) Me or Et) have been reported in the literature and found
to exchange the terminal alkoxy group with another alcohol
or with water very easily in the presence of acids.14a The
crystallographically characterized species 4 never contacted
any alcohol, so no doubt can be raised about its identity. On
the other hand, we cannot exclude the possibility that the
compound identified by 1H NMR as 4 in mixtures obtained
from EtOH and CH2Cl2 is at least in part the ethoxy [Ru-
(TPP)(OEt)]2O derivative. However, we never detected in
Solution Behavior and 1H NMR Spectra of [Ru(TPP)-
(O)2] (3). When a high-purity (XRPD evidence) sample of
3 was dissolved in CDCl3, two sets of signals were observed.
The first set of signals, marked with an asterisk in Figure 3,
can be attributed to the highly symmetric monomeric [RuVI-
(TPP)(O)2] (3a). The D4h symmetry is retained in solution,
and one signal was observed at 9.10 ppm corresponding to
eight pyrrolic hydrogen atoms (Hâ) along with two signals
in the aromatic region, one for the eight ortho hydrogen
1
the H NMR spectrum of these mixtures run in CDCl3 the
signals at -3.86 and -4.03 ppm, which are diagnostic of
the ethoxy group in the axial position of the porphyrin
complex.14a Since the solvents employed for the synthesis
of 4 have never been dried, it appears that the equilibrium
2042 Inorganic Chemistry, Vol. 44, No. 6, 2005

Structure of Ruthenium-Porphyrin Complexes
The relative intensity of the signals in our spectra in
solution would imply that the 3a:3b ratio = 3:1, which is
inconsistent with the data obtained in the solid state by
XRPD, which indicate that only one species is present. The
second group of signals did not disappear when the solution
was filtered through Celite, indicating that it is not due to
incompletely dissolved material. The relative ratio of the two
species remained almost unchanged even using strictly
anhydrous CDCl3 (freshly distilled over CaH2) or if H2O or
D2O was added to the solution, at least on the minute time
scale. Investigations over long period of times of solutions
of 3 are hampered by the formation of variable amounts of
4 upon prolonged standing, especially when water had been
added. Of note is the fact that the signal at -2.70 ppm did
not disappear after addition of D2O. Moreover, samples of
3 always contained some water, even if heated in vacuo for
several hours. The only observable phenomenon, when
samples containing smaller amounts of water are examined,
is a shift to higher fields (from 1.55 to 0.5 ppm) in the
position of the water signal itself. In the case of 3, two
cavities (per unit cell) with a volume of ∼45 Å3 each, i.e.,
comparable with that of water molecules, are present. The
shortest O‚‚‚H distance would lie at ∼2.8 Å; thus, we cannot
exclude the partial occupancy of these cavities by occluded
water molecules, not visible by XRPD. To exclude the
possibility that 3 reacts with CDCl3, a solution of the product
Figure 3. 1H NMR spectrum of 3 in CDCl3.
1
for which an H NMR spectrum had been recorded was
evaporated in a vacuum and analyzed by XRPD. Its
diffraction pattern was found to be that of the starting [Ru-
(TPP)(O)2] species, 3. Identical results were obtained when
3 was treated in C6H6.
Figure 4. 1H NMR spectrum of 4 in CDCl3.
To investigate the nature of the second species, we
performed PGSE (pulsed field gradient spin-echo) NMR
measurements20 on CDCl3 solutions of 3 using the standard
stimulated echo pulse sequence at room temperature. These
measurements can afford specific information about the size
of the species present in solution, in that the hydrodynamic
radii and the volumes of the diffusing particles can be
estimated.21 According to the Stokes-Einstein equation (eq
1), the diffusion coefficient (Dt) is inversely proportional to
the hydrodynamic radius (rH) of the diffusing particle.
atoms (8.38 ppm) and one for the eight meta hydrogen atoms
which overlap with the four para hydrogen atoms (7.85
ppm).19
The second pattern observed, marked with an empty circle
in Figure 3, belongs to a species with apparently lower sym-
metry (3b) in which the signals for the eight ortho hydrogen
atoms and for the eight meta hydrogen atoms split into two
sets of signals each and the pyrrolic protons resonate at higher
fields (8.61 ppm). This pattern indicates that the plane of
symmetry determined by the porphyrin ring is lost, as usually
happens for dimeric porphyrin complexes. The species pro-
ducing this last set of signals is always present in any spec-
trum of 3 we recorded in this solvent. It should not be con-
fused with 4, whose spectrum in the same solvent is reported
in Figure 4. For the latter, the pyrrolic protons resonate at
8.67 ppm and can indeed be observed when samples obtained
under nonoptimized conditions are analyzed. In the 1H NMR
spectrum of 3, a singlet was also present at -2.70 ppm,
indicating that the hydrogen atoms that produce it are in the
shielding cone of the porphyrin. The same situation was
observed when CD2Cl2 or C6D6 was employed in place of
CDCl3, although in C6D6 a poorly defined spectrum was
obtained, due to the low solubility of 3 in this solvent.
kT
6πηrH
Dt )
(1)
where k is the Boltzmann constant, T is the absolute
temperature, and η is the viscosity of the medium. Conse-
quently, the volume (V) of the diffusing particle that in the
Stokes-Einstein model is assumed to be spherical is
inversely proportional to Dt3.
(20) Johnson, C. S., Jr. Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34,
203-256 and references therein.
(21) For some recent references, see: (a) Babushkin, D. E.; Brintzinger,
H.-H. J. Am. Chem. Soc. 2002, 124, 12869-12873. (b) Burini, A.;
Fackler, J. P., Jr.; Galassi, R.; Macchioni, A.; Omary, M. A.;
Rawashdeh-Omary, M. A.; Pietroni, B. R.; Sabatini, S.; Zuccaccia,
C. J. Am. Chem. Soc. 2002, 124, 4570-4571. (c) Drago, D.; Pregosin,
P. S.; Pfaltz, A. Chem. Commun. 2002, 286-287. (d) Macchioni, A.;
Romeni, A.; Zuccaccia, C.; Guglielmetti, G.; Querci, C. Organome-
tallics 2003, 22, 1526-1533. (e) Pregosin, P. S.; Mart´ınez-Viviente,
E.; Kumar, P. G. A. Dalton Trans. 2003, 4007-4014.
(19) The spectrum described here correlates well with the one reported by
Che and co-workers in Figure 3 in ref 11a for the same product.
However, the latter does not correspond exactly to the one described
in the Experimental Section of the same paper.
Inorganic Chemistry, Vol. 44, No. 6, 2005 2043

Gallo et al.
Figure 7. ORTEP drawing of the molecular structure of the [RuII(TPP)-
(S-DMSO)2] species (5) showing the two DMSO molecules trans to each
other. Only one conformation of the disordered phenyl fragments and DMSO
molecules has been reported for clarity. Selected distances for 5: 2.043-
(2)-2.046(2) Å for the Ru-N distance and 2.3140(7) Å for the Ru-S
distance.
The position of the bridging molecule explains both why
it does not exchange with external D2O and why it resonates
1
at negative fields in the H NMR spectra. The stability of
3b suggests that even the RudO groups not involved in
dimer formation should form strong hydrogen bonds. This
is in full agreement with the observed shift in the position
of the water signal. This can now be easily explained by a
time-mediated exchange of free water molecules with the
ones hydrogen bonded to the “terminal” RudO groups.
Clearly, even chloroform itself can form a hydrogen bond
to the RudO group, but because it is present in a large
excess, no shift is detectable.
Figure 5. DOSY display of 3 at 500 MHz.
Reaction of [Ru(TPP)(O)2] with Dimethyl Sulfoxide and
Synthesis of [Ru(TPP)(S-DMSO)2] (5). When a sample of
3 was dispersed in (CD3)2SO, a heterogeneous mixture was
obtained, which showed in the 1H NMR spectrum many ill-
resolved signals (as expected for a dispersion of small
particles). When the temperature was increased to 110 °C,
the solid material completely dissolved and the complex
1
pattern disappeared, leading to a much simpler H NMR
Figure 6. Proposed solution structures for 3a and 3b.
spectrum (see the Experimental Section). This spectrum
belongs to a species with apparent D4h symmetry, and shows
a singlet at 8.42 ppm (assigned to eight pyrrolic protons)
and multiplets centered at 8.08 and 7.76 ppm, for the eight
ortho and twelve meta and para aromatic protons, respec-
tively. Interestingly, this pattern was conserved when the
sample was cooled to room temperature. Despite the similar-
ity of the spectra of 3 and 5, the two species turned out to
be different. Indeed, when this dissolution process was
repeated on a preparative scale and the product isolated by
precipitation and redissolved in CDCl3, a spectrum different
from the one of 3, but still showing D4h symmetry, was
observed.
The new species was unequivocally characterized as [Ru-
(TPP)(S-DMSO)2] (5) by conventional single-crystal X-ray
diffraction analysis (Figure 7).
The synthesis of 5 has already been reported in the
literature.23 [Ru(TPP)(S-DMSO)2] was prepared by photo-
lyzing [Ru(TPP)(CO)(ROH)] in DMSO solution. Other [Ru-
(porphyrin)(S-DMSO)2] complexes have been obtained using
the same procedure described above24 or by reacting the [Ru-
(porphyrin)]2 dimer, which has been prepared by photolysis,
The V3b:V3a ratio was calculated as (D3a/D3b)3, obtaining
a volume ratio of 2.0:1, indicating that 3b is a dimer with
respect to 3a. Noteworthy, the signal at -2.70 ppm also
belongs to the dimer and must be associated with it (Figure
5).
Both 3a and 3b undergo extremely fast reactions in the
presence of a suitable reductant. This has been shown by
adding a 4-fold excess of Ph3P to a solution of 3 in CDCl3.11d
An instantaneous reaction occurred; no intermediate could
be detected at room temperature, and the only observed
complex was [RuII(TPP)(PPh3)2].22 The 31P NMR spectrum
of the reaction mixture showed the presence of Ph3PO also.
In conclusion, even if an unequivocal characterization of
the second species cannot be given, all available experimental
evidence points to 3a and 3b being two different solvates of
3. In particular, 3b appears to be a dimeric species held
together by hydrogen bonding to a water molecule (Figure
6). Note that at this stage we cannot exclude the presence of
two bridging water molecules.
(22) Ariel, S.; Dolphin, D.; Domazetis, G.; James, B. R.; Leung, T. W.;
Rettig, S. J.; Trotter, J.; Williams, G. M. Can. J. Chem. 1984, 62,
755-762.
(23) Levine, L. M. A.; Holten, D. J. Phys. Chem. 1988, 92, 714-720.
2044 Inorganic Chemistry, Vol. 44, No. 6, 2005

Structure of Ruthenium-Porphyrin Complexes
Scheme 3. Reaction of [Ru(TPP)(O)2] (3) with Carbon Monoxide
complexes with the [Ru(porphyrin)(CO)(H2O)] general for-
mula have been isolated as single crystals in a few
cases,9a,13f,28 but apparently, no bulk material having this
composition has ever been obtained. It has been suggested
that the water molecule in these complexes is very labile
and that its coordination must be stabilized by hydrogen
bonding.
The UV-visible spectrum of 6 is indistinguishable from
those of 1 and 2. Moreover, the compound is diamagnetic
at room temperature, and no EPR signal was observed to
100 K. All these data indicate that complex 6 has formulation
B above (Scheme 3).
This assignment was further confirmed by an independent
synthesis of 6 from 2 in a boiling ethyl acetate/water mixture
(see the Experimental Section). Finally, reaction of 6 with
tert-butylpyridine afforded the previously reported [Ru(TPP)-
(CO)(4-tBu-Py)].29 The water included in 6 during its
preparation from 3 must originate from the CO gas, of which
it represents the main contaminant.
Characterization of “[Ru(TPP)(CO)]”. The [Ru(TPP)-
(CO)] complex (1) is the entry point for most ruthenium-
tetraphenylporphyrin chemistry and is even commercially
available, but has been characterized less thoroughly than
may be expected. It is generally synthesized by reaction of
TPPH2 with Ru3(CO)12 in refluxing Decaline or toluene, and
the product is purified by column chromatography on silica.
However, different preparations of 1 with the synthetic
method described above afford materials that show slightly
different νCO stretching frequencies (in our hands, in the range
of 1950-1956 cm-1) even when measured on the same
instrument. By comparison, both [Ru(TPP)(CO)(H2O)] (6)
and [Ru(TPP)(CO)(MeOH)] (2) reproducibly exhibit an IR
absorption at 1949 cm-1.
Figure 8. ORTEP drawing of the molecular structure of the [RuII(TPP)-
(CO)(H2O)] species (6) showing the carbonyl (CO) and the water molecule
(Ow) trans to each other. Water hydrogen atoms are not shown. Selected
distances for 6: 1.75(7) Å for the Ru-C distance, 2.045(3) Å for the Ru-N
distance, and 2.22(4) Å for the Ru-O distance.
in the presence of DMSO.25 Our synthesis does not require
any specialized equipment, such as medium-pressure mercury
lamps or lasers, at any stage of the reaction and appears to
be more suited for the preparation of large amounts of 5,
which can be regarded as an interesting starting material for
the synthesis of other ruthenium-porphyrin complexes. It
should also be noted that related [Ru(porphyrin)L2] com-
plexes (L ) arylamine or alkylamine) have been obtained
by reaction of [Ru(porphyrin)(O)2] with amines or hydroxyl-
amines.11c,26
Reduction of [Ru(TPP)(O2)] (3) by Carbon Monoxide.
During recent work in our laboratories on the reactivity of
aryl azides with 1 and 2, we isolated a [RuIV(TPP)(CO)-
(ArN)] complex having an imido group coordinated to a
ruthenium(IV) center.27 With this result in mind, we decided
to test if an analogous complex could be isolated having an
oxo ligand in place of the imido one. Toward this aim, we
subjected 3 to a CO atmosphere in CH2Cl2. No reaction was
observed at atmospheric CO pressure, but when the reaction
was conducted under 60 bar of CO in an autoclave, a
crystalline product, 6, precipitated out of the solution. Some
of the crystals were of sufficient quality for determining the
single-crystal X-ray structure (Figure 8). The remaining
polycrystalline material was analyzed by XRPD and found
to be identical with the single crystal. Thus, the whole
obtained material is composed by the same complex, and
the single crystals are highly representative of the bulk
composition.
We originally attributed the variability of the observed CO
stretching frequency in the solid to the presence of different
crystal packing forms (i.e., polymorphs), but XRPD mea-
surements on several samples allowed us to discard this
possibility. Indeed, the [Ru(TPP)(CO)] complex crystallizes
in only one form, a tetragonal phase isomorphous with 6
and 2. This should not be a surprise, since the crystal packing
is mostly dictated by the mutual interlocking of the phenyl
rings of the TPP moieties. However, we found that a small,
but non-negligible, variability existed in the refined lattice
parameters of [Ru(TPP)(CO)] from different preparations
(although fully maintaining the I4/m crystal symmetry). In
addition, a slight improvement of the agreement factors of
The obtained complex (Figure 8) has a coordinated CO
ligand, and the opposite position in the coordination sphere
is occupied by an oxygen atom. However, the refined Ru-O
distance (2.24 Å) is inconsistent with the formulation of this
group as an oxo ligand (the expected RudO distance lies in
the 1.66-1.82 Å range) and indicates a single-bond character
for this interaction. The isolated complex may thus be [RuIII-
(TPP)(CO)(OH)] (A) or [RuII(TPP)(CO)(H2O)] (B). Aquo
(24) (a) Hopf, F. R.; O’Brien, T. P.; Scheidt, W. R.; Whitten, D. G. J. Am.
Chem. Soc. 1975, 97, 277-281. (b) Ikonen, M.; Guez, D.; Marvaud,
V.; Markovitsi, D. Chem. Phys. Lett. 1994, 231, 93-97.
(25) (a) Pacheco, A.; James, B. R.; Rettig, S. J. Inorg. Chem. 1995, 34,
3477-3484. (b) Pacheco, A.; James, B. R.; Rettig, S. J. Inorg. Chem.
1999, 38, 5579-5587.
(26) (a) Huang, J.-S.; Sun, X.-R.; Leung, S. K.-Y.; Cheung, K.-K.; Che,
C.-M. Chem.sEur. J. 2000, 6, 334-344. (b) Liang, J.-L.; Huang, J.-
S.; Zhou, Z.-Y.; Cheung, K.-K.; Che, C.-M. Chem.sEur. J. 2001, 7,
2306-2317. (c) Leung, S. K.-Y.; Huang, J.-S.; Liang, J.-L.; Che, C.-
M.; Zhou, Z.-Y. Angew. Chem., Int. Ed. 2003, 42, 340-343.
(27) S. Cenini et al., unpublished results.
(28) (a) Kadish, K. M.; Hu, Y.; Mu, X. H. J. Heterocycl. Chem. 1991, 28,
1821-1824. (b) Slebodnick, C.; Kim, K.; Ibers, J. A. Inorg. Chem.
1993, 32, 5338-5342. (c) Birnbaum, R. A.; Schaefer, W. P.; Labinger,
J. A.; Bercaw, J. E.; Gray, H. B. Inorg. Chem. 1995, 34, 1751-1755.
(29) Eaton, S. S.; Eaton, G. R.; Holm, R. H. J. Organomet. Chem. 1972,
39, 179-195.
Inorganic Chemistry, Vol. 44, No. 6, 2005 2045

Gallo et al.
the pertinent structural refinements could be achieved by
partially occupying the “empty” coordination site with a
water molecule. The following observations finally solve the
problem.
(a) The IR spectrum of the crude material before the
chromatographic separation exhibits an absorption at 1956
cm-1, although other bands can also be attributed to Ru3-
(CO)12 used as a starting material.
(b) In one experiment, a sample of [Ru(TPP)(CO)] was
obtained by precipitating the compound at room temperature
from a chloroform/ethyl acetate fraction of the chromato-
graphic column. The sample was briefly dried in a dinitrogen
stream without either heating it or placing it in vacuo. The
IR spectrum of the material so obtained exhibited a νCO of
1950 cm-1.
(c) If the sample of [Ru(TPP)(CO)] precipitated from the
solution is heated in vacuo at 120 °C, the νCO stretching
gradually moves from 1950 to 1956 cm-1 over a few hours.
(d) The thermogravimetric curves of different [Ru(TPP)-
(CO)] samples exhibited rather erratic behavior, with weight
losses, at 250 °C, of 3-5%.
Figure 9. Selected portion (16.5° < 2θ < 30°) of different XRPD patterns
of species of nominal [Ru(TPP)(CO)] formula (from four different prepara-
tions), showing the small, though significant, variability of the peak positions
(see arrows). Whole-profile (structureless) refinements by the Le Bail
method and corrections for sample displacement errors allowed us to assess
the variability of the lattice parameters, which ranged from 13.538 to 13.613
Å and from 9.683 to 9.688 Å for a and c, respectively (the length of the
tetragonal axis being nearly constant), in agreement with a solid solution
behavior.
(99%, Aldrich) were used as received. Ruthenium dodecacarbonyl,30
tetraphenylporhyrin,31 [Ru(TPP)(4-tBu-Py)(CO)],29 and [Ru(TPP)-
(CH3OH)(CO)] (2)32 were prepared according to literature proce-
dures. 1H NMR and 31P NMR spectra were recorded on Advanced
300-DRX Bruker instruments. PGSE measurements were performed
on an Advanced 500-DRX Bruker spectrometer equipped with a
QNP probe with a Z-gradient coil. Infrared spectra were obtained
as Nujol mulls on a Bio-Rad FTS-7 spectrophotometer. UV-visible
spectra were recorded on a Hewlett-Packard 8452A spectropho-
tometer. Thermogravimetric experiments were performed on a
Perkin-Elmer TGA 7 instrument. Elemental analyses and mass
spectra were recorded in the analytical laboratories of Milan
University.
All this experimental evidence leads us to the following
conclusions.
(1) The compound usually formulated as [Ru(TPP)(CO)]
is actually a solid-state solution in which variable proportions
of pristine [Ru(TPP)(CO)] and [Ru(TPP)(CO)(H2O)] can be
present. The νCO absorptions of the two complexes are too
close to be separated even when working at a 1 cm-1
resolution. Therefore, only one band is observed, the position
of which depends on the relative proportion of the two
complexes in the mixture.
(2) In the synthesis, authentic [Ru(TPP)(CO)] is initially
formed, which is quantitatively transformed in [Ru(TPP)-
(CO)(H2O)] during the chromatographic purification, water
clearly originating from silica itself or from moist solvents.
(3) When the solutions obtained by the chromatographic
purification are evaporated and the resulting solid is heated
in vacuo to remove the last traces of solvent, part of the
coordinated water is released (in a poorly reproducible way),
causing the observed variation of the IR spectrum.
(4) The shifts of the peak positions of the measured XRPD
patterns (well above the instrumental reproducibility) are,
thus, a manifestation of the different compositions of the
differently prepared samples (see Figure 9).
(5) Accordingly, the TG traces, showing different slopes
in the 100-400 °C range, speak for (at least) three nearly
overlapping events {i.e., water and CO elimination, as well
as sublimation of the “[Ru(TPP)]n” species}; indeed, dis-
sociation of 1 mol of CO or H2O implies a 3.8 or 2.4%
weight loss, respectively.
Preparation of [Ru(TPP)(CO)] (1). A modification of the
procedure of Rillema et al.32 was used. Ru3(CO)12 (0.575 g, 0.899
mmol) and TPPH2 (1.24 g, 2.03 mmol) were suspended in dry
Decalin (50 mL) under a dinitrogen atmosphere. The reaction
mixture was refluxed for 4 h, and the resulting solid was collected
in a filter and washed with n-hexane (3 × 10 mL) to remove
Decalin. The violet solid was then purified by flash chromatography
on silica. Toluene was used to elute unreacted Ru3(CO)12, a 7:3
CH2Cl2/n-hexane mixture to elute TPPH2, and finally CHCl3 to yield
1 (1.03 g, 69%): 1H NMR (300 MHz, CDCl3, 300 K) δ 8.71 (s,
8H, Hâ), 8.26 (m, 4H, Ho), 8.15 (m, 4H, Ho), 7.75 (m, 12H, Hm-p);
IR (Nujol) ν ) 1950-1956 cm-1 (CO, see Characterization of [Ru-
(TPP)(CO)] in the Results and Discussion), 1008 cm-1 (oxidation
marker); UV-vis (CH2Cl2) λmax (log ꢀM) ) 588 (3.51), 528 (4.29),
412 nm (5.38). Anal. Calcd for C45H28N4ORu (741.81): C, 72.86%;
H, 3.80%; N, 7.55%. Found: C, 72.38%; H, 4.05%; N, 7.22%.
Preparation of [Ru(TPP)(O)2] (3). A modification of the
procedure of Che15 was used. In the optimized procedure, m-CPBA
(1.67 g, 7.43 mmol) was dissolved in EtOH (40 mL) and the
resulting solution was added in one portion to an orange suspension
of 2 (0.298 g, 0.385 mmol) in CH2Cl2 (10 mL) and EtOH (10 mL).
The resulting dark suspension was stirred for 1 h, and the dark
Experimental Section
(30) Eady, C. R.; Jackson, P. F.; Johnson, B. F. G.; Lewis, J.; Malatesta,
M. C.; McPartlin, M.; Nelson, W. J. H. J. Chem. Soc., Dalton Trans.
1980, 383-392.
Unless otherwise specified, all reactions and manipulations were
performed in air using magnetic stirring. The solvents for the
syntheses were of analytical grade, except for Decalin (sodium)
which was dried, distilled, and stored under dinitrogen. m-
Chloroperoxybenzoic acid (77%, Aldrich) and triphenylphosphine
(31) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.; Assour,
J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476.
(32) Rillema, D. P.; Nagle, J. K.; Barringer, L. F., Jr.; Meyer, T. J. J. Am.
Chem. Soc. 1981, 103, 56-62.
2046 Inorganic Chemistry, Vol. 44, No. 6, 2005

Structure of Ruthenium-Porphyrin Complexes
a
Table 1
Synoptic Collection of Crystal Data for the Four Structurally Characterized Species
[Ru(TPP)(O)2] (3)
[Ru(TPP)(OH)]2(O) (4)
[Ru(TPP)(S-DMSO)2] (5)
[Ru(TPP)(CO)(H2O)] (6)
formula
fw (amu)
cryst syst
space group
a (Å)
C44H28N4O2Ru
745.77
tetragonal
I4/m
13.3992(4)
13.3992(4)
9.7116(3)
90
90
90
1743.6(1)
2
1.420
C88H56N8O3Ru2
1477.56
tetragonal
P4/nnc
13.2334(4)
13.2334(4)
19.4274(7)
90
90
90
3402.2(2)
2
1.442
C48H40N4O2RuS2
870.03
monoclinic
I2/m
14.629(2)
9.356(1)
14.465(2)
90
91.12(1)
90
C45H30N4O2Ru
759.80
tetragonal
I4/m
13.558(2)
13.558(2)
9.685(1)
90
90
90
1780.2(2)
2
1.417
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
1979.4(2)
2
1.460
D
calc (g/cm3)
method
powder XRD
powder XRD
single-crystal XRD
single-crystal XRD
Details of Single-Crystal Analyses
[Ru(TPP)(S-DMSO)2]
[Ru(TPP)(CO)(H2O)]
0.485
776
absorption coefficient (mm-1
F(000)
)
0.548
896
cryst size
0.10 mm × 0.15 mm
0.15 mm × 0.15 mm
× 0.25 mm
× 0.30 mm
limiting indices
-19 e h e 19,
-12 e k e 12,
-19 e l e 19
13697/2635
(Rint ) 0.029)
Sadabs
-16 e h e 16,
-16 e k e 16,
-11 e l e 11
9948/842
(Rint ) 0.080)
Sadabs
no. of reflections collected/unique
absorption correction
refinement method
full-matrix least-squares
full-matrix least-squares
on F2
on F2
data/restraints/parameters
goodness of fit on F2
2635/0/205
1.099
842/0/76
1.158
final R indices [I > 2σ(I)]
R1 ) 0.036, wR2 ) 0.094
R1 ) 0.036, wR2 ) 0.094
a Rint ) ∑|Fo2 - Fo (mean)|/∑Fo . Rσ ) ∑σ(Fo )/∑Fo . R1 ) ∑||Fo| - |Fc||/∑|Fo|. wR2 ) {∑[w(Fo2 - Fc )2]/∑[w(Fo )2]}1/2. Goodness of fit ) [S/(n -
2
2
2
2
2
2
2
2
p)]1/2 ) {∑[w(Fo - Fc )2]/(n - p)}1/2
.
violet solid that formed was collected by filtration, washed with
ethanol to remove the residual m-CPBA, and dried in a vacuum
(0.22 g, 78%): 1H NMR (300 MHz, CDCl3, 300 K) monomeric
species (3a) δ 9.10 (s, 8H, Hâ), 8.38 (m, 8H, Ho), 7.85 (m, 12H,
Hm-p), dimeric species (3b) δ 8.87 (m, 8H, Ho), 8.61 (s, 16H, Hâ),
7.98 (m, 8H, Hm), 7.84 (m, 8H, Hp), 7.60 (m, 8H, Hm), 7.50 (m,
8H, Ho), -2.70 (s); 1H NMR (300 MHz, CD2Cl2, 300 K)
monomeric species δ 9.17 (s, 8H, Hâ), 8.40 (m, 8H, Ho), 7.90 (m,
12H, Hm-p); dimeric species δ 8.92 (m, 8H, Ho), 8.66 (s, 16H, Hâ),
8.02 (m, 8H, Hm), 7.90 (m, 8H, Hp), 7.62 (m, 8H, Hm), 7.48 (m,
8H, Ho), -2.80 (s); relative monomeric/dimeric species ratio =3/
1; IR (Nujol) ν ) 1017 cm-1 (oxidation marker), 818 cm-1 (RuO2);
UV-vis (CH2Cl2) λmax (log ꢀM) ) 542 (sh) (4.03), 520 (4.18), 418
nm (5.18). Anal. Calcd for C44H28N4O2Ru (745.80): C, 70.86%;
H, 3.78%; N, 7.51%. Found: C, 70.75%; H, 4.07%; N, 7.65%.
Preparation of [Ru(TPP)(OH)]2O (4). In method a, a solution
of m-CPBA (51.4 mg, 0.228 mmol) in CH2Cl2 (25 mL) was added
to a suspension of 1 (104 mg, 0.14 mmol) in CH2Cl2 (35 mL). The
resulting red solution was stirred overnight at room temperature
and filtered through a short column of neutral alumina to yield 4
(62.2 mg, 61%). In method b, complex 3 (52.3 mg, 7.01 × 10-2
mmol) was added to a solution of 2 (53.8 mg, 6.95 × 10-2 mmol)
in CH2Cl2 (60 mL) to yield a red solution that was stirred at room
temperature for 30 min. After the addition of HCl (0.5 mL), the
reaction mixture turned green and [Ru(TPP)(OH)]2O was formed
(TLC monitoring, CH2Cl2 as the eluant). The solution was then
extracted with water (3 × 20 mL), and the organic layer was filtered
through a short column of neutral alumina to yield 4 (56.9 mg,
55%). The same reaction was observed when using m-chlorobenzoic
acid instead of hydrochloric acid: 1H NMR (300 MHz, CDCl3,
300 K) δ 8.89 (d, J ) 7.6 Hz, 8H, Ho), 8.67 (s, 16H, Hâ), 8.00 (t,
J ) 7.6 Hz, 8H, Hm), 7.86 (t, J ) 7.6 Hz, 8H, Hp), 7.59 (t, J ) 7.6
Hz, 8H, Hm,), 7.48 (d, J ) 7.6 Hz, 8H, Ho); IR (Nujol) ν ) 1014
cm-1 (oxidation marker); UV-vis (CH2Cl2) λmax (log ꢀM) ) 556
(4.30), 392 nm (5.58). Anal. Calcd for C88H58N8O3Ru2 (1475.60):
C, 71.63%; H, 3.83%; N, 7.59%. Found: C, 71.49%; H, 3.90%;
N, 7.65%.
Preparation of [Ru(TPP)(S-DMSO)2] (5). 3 (51.0 mg, 6.85 ×
10-2 mmol) was suspended in DMSO (15 mL), and the resulting
suspension was heated at 110 °C for 3 h. The total consumption of
3 was verified via TLC on alumina with dichloromethane as the
eluent. The resulting pink solid was collected by filtration, washed
with n-hexane, and dried in a vacuum (46.2 mg, 78%): 1H NMR
(300 MHz, CDCl3, 300 K) δ 8.59 (s, 8H, Hâ), 8.13 (m, 8H, Ho),
1
7.71 (m, 12H, Hm-p), -1.64 [s, 12H, (CH3)2SO]; H NMR (300
MHz, DMSO-d6, 300 K) δ 8.42 (s, 8H, Hâ), 8.08 (m, 8H, Ho),
7.76 (m, 12H, Hm-p); IR (Nujol) ν ) 1105 cm-1 (SdO), 1006 cm-1
(oxidation marker); UV-vis (CH2Cl2) λmax (log ꢀM) ) 524 (3.6),
414 nm (4.7). Anal. Calcd for C48H40N4O2RuS2 (870.06): C,
66.26%; H, 4.63%; N, 6.44%. Found: C, 66.50%; H, 4.80%; N,
6.30%. Recrystallization from DMSO gave crystals suitable for
X-ray analysis.
Preparation of [Ru(TPP)(CO)(H2O)] (6). In method a, complex
1 (91.3 mg, 1.23 mmol) was suspended in a biphasic mixture of
water (6 mL) and ethyl acetate (3 mL) and the resulting mixture
was refluxed and vigorously stirred in air for 4 h. After the mixture
had cooled, the aqueous layer was eliminated and the organic layer
evaporated to dryness. The residue was boiled in n-hexane, and
the resulting purple solid was then collected in a filter and dried in
a vacuum (84.2 mg, 90%). In method b, a red solution of 3 (56.4
Inorganic Chemistry, Vol. 44, No. 6, 2005 2047

Gallo et al.
Figure 10. Rietveld refinement plots for [RuVI(TPP)(O)2] (3) (top) and [RuIV(TPP)(OH)]2O (4) (bottom). The horizontal axis is 2θ (degrees), and the
vertical axis is counts. Difference plot and peak markers at the bottom. The sections above 30° (2θ) have been magnified in the insets.
mg, 7.56 × 10-2 mmol) in CH2Cl2 (15 mL) was placed in a glass
liner inside an autoclave. The autoclave was frozen at dry ice
temperature, evacuated, and filled with dinitrogen three times; then
CO (60 bar) was added at room temperature, and the solution was
stirred at room temperature for 5 h. The resulting purple solid was
collected in a filter and dried in a vacuum (29.8 mg, 52%): 1H
NMR (300 MHz, CDCl3, 300 K) δ 8.71 (s, 8H, Hâ), 8.26 (m, 4H,
Ho), 8.15 (m, 4H, Ho), 7.75 (m, 12H, Hm-p); IR (Nujol) ν ) 1949
cm-1 (CO), 1009 cm-1 (oxidation marker). Anal. Calcd for
C45H30N4O2Ru (759.83): C, 71.13%; H, 3.98%; N, 7.37%.
Found: C, 70.85%; H, 4.12%; N, 7.05%. Recrystallization from
CH2Cl2 gave crystals suitable for X-ray analysis.
Single-Crystal X-ray Analyses. Crystals of [Ru(TPP)(S-
DMSO)2] (5) and [Ru(TPP)(CO)(H2O)] (6) suitable for X-ray
crystallographic analysis were selected, attached to a glass fiber,
and placed on the diffractometer. All data were collected at room
temperature using a Bruker AXS diffractometer (equipped with Mo
KR radiation and a CCD area detector). Absorption corrections were
applied using SADABS.33 The WinGX suite of programs was used
for the structure solutions and refinements.34 The crystal structures
were determined by direct methods and refined by full-matrix least-
squares procedures. All non-hydrogen atoms were refined aniso-
tropically. Split atom models were used to account for disorder of
dimethyl sulfoxide and of the peripheral phenyls in 5 and of the
CO/Ow in 6. Hydrogen atoms were included in the refinement at
calculated positions (apart from the hydrogen atoms of the methyl
groups of 5), using a riding model included in ShelX.35 Hydrogen
atoms were given isotropic thermal parameters 1.2 times the thermal
parameter of the atoms to which they were attached. The water
molecule hydrogen atoms could not be detected in 6, also because
they are necessarily (crystallographically) disordered about the
4-fold axis (as well as by the statistical mirror plane owned by the
whole complex). In addition, the lack of suitable hydrophilic sites
in the TPP periphery makes it likely that a dynamic reorientation
of the H2O molecule occurs. Selected details of the data collection
and refinement are given in Table 1.
X-ray Powder Diffraction Studies. Purple powders 3 and 4
were gently ground in an agate mortar and then deposited in the
hollow (0.1 mm deep) of a zero-background quartz monocrystal
plate. Diffraction data (Cu KR, λ ) 1.5418 Å) were collected on
a vertical scan θ:2θ Bruker D8 ADVANCE diffractometer,
(33) Sheldrick, G. M. (1996) SADABS, University of Go¨ttingen, Go¨t-
tingen, Germany (freely available on line at http://shelx.uni-ac.gwdg.de/
axs/.
(35) Sheldrick, G. M. SHELX97: Programs for Crystal Structure Analysis,
release 97-2; Institu¨t fu¨r Anorganische Chemie der Universita¨t:
Go¨ttingen, Germany, 1998.
(34) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837-838.
2048 Inorganic Chemistry, Vol. 44, No. 6, 2005

Structure of Ruthenium-Porphyrin Complexes
equipped with parallel (Soller) slits, a secondary beam curved
graphite monochromator, a Na(Tl)I scintillation detector, and pulse
height amplifier discrimination. The generator was operated at 40
kV and 40 mA. The receiving slit was 0.2 mm. Nominal resolution
for the present setup is 0.07° 2θ (fwhm) for the Si(111) peak at
28.44° (2θ). Long scans were performed with 5° < 2θ < 80°, t )
30 s, and ∆2θ ) 0.02° and used for structure solution and
refinement.
structures reported in this paper have been deposited with the
Cambridge Crystallographic Data Centre (Supplementary Publica-
tion CCDC 237680-237683). Copies of the data can be obtained
free of charge on application to CCDC, 12 Union Road, Cam-
bridge CB21EZ, U.K. [fax, (+44)1223-336-033; e-mail, deposit@
ccdc.cam.ac.uk] or can be retrieved free of charge via www.
ccdc.cam.ac.uk/conts/retrieving.html.
Indexing, using TREOR,36 of the low-angle diffraction peaks
suggested tetragonal cells with the following approximate dimen-
sions: a ) 13.41 Å and c ) 9.73 Å [M(20) ) 22; F(20) ) 34
(0.011, 54)] and a ) 13.24 Å and c ) 19.44 Å [M(20) ) 21; F(20)
) 34 (0.012, 52)] for 3 and 4, respectively. Systematic absences
indicated, among others, I4/m and P4/nnc as the probable space
groups, later confirmed by successful solution and refinement.
Structure solutions were initiated by the simulated annealing
technique implemented in TOPAS,37 using, as fragments, the metal
atom and idealized pyrrolic and benzylic fragments. The final
refinements were performed by the Rietveld method with the aid
of TOPAS, with peak shapes described by the fundamental
parameter approach and an isotropic crystal size broadening factor
of Lorentzian contribution. The TPP fragments as a whole, lying
on 4 or 4/m symmetry elements, have been eventually carefully
modeled by constraining some of the rotational parameters of their
constituents. A soft constraint was also given to the Ru-OH bond
distance (2.10 Å) in 4. The background function was modeled by
a polynomial function, while further systematic errors were
corrected with the aid of a sample displacement shift and a preferred
orientation model (001 pole, in the March-Dollase formulation);
a single isotropic displacement parameter was also refined. Scat-
tering factors were taken from the internal library of TOPAS. Final
agreement factors Rp, Rwp, and RBragg equaled 0.081, 0.112, and
0.038 for 3 and 0.086, 0.117, and 0.040 for 4, respectively, for
3751 data points collected in the 5° < 2θ < 80° range. Final
Rietveld refinement plots are given in Figure 10. Eventually, species
3 and 4 were found to be isostructural with [RuIV(TPP)(Br)2]38 and
[RuIV(OEP)(OH)]2O‚(CH3OH),16 respectively. A summary of the
crystal data and refinement procedures can be found in Table 1.
The structure of a similar species, namely, Ti(TPP)Cl2, has been
refined (not determined) by powder diffraction methods (using
synchrotron radiation), taking the advantage of the already known
fractional coordinates of the bromine derivative.39
Conclusions
In this paper, we have reported a reproducible synthesis,
in yield and purity, of [RuVI(TPP)(O)2] (3). Its molecular
structure was determined by X-ray powder diffraction
(XRPD) analysis. The solution behavior of 3 was investigated
by NMR spectroscopy; PGSE measurements allowed us to
identify the presence of two different solvates of 3 in the
CDCl3 solution. The reactivity of [RuVI(TPP)(O)2] was also
investigated. It reacts in the presence of acids with [RuII-
(TPP)(CO)] to yield [RuIV(TPP)(OH)]2O, which can also be
obtained by oxidation of [RuII(TPP)(CO)] in CH2Cl2. Com-
plex [RuIV(TPP)(OH)]2O was isolated in a polycrystalline
form which allowed us to determine its molecular structure
by XRPD. The reaction of [RuVI(TPP)(O)2] with DMSO and
carbon monoxide gave [RuII(TPP)(S-DMSO)2] and [RuII-
(TPP)(CO)(H2O)], respectively. These complexes were com-
pletely characterized, including conventional single-crystal
X-ray diffraction analysis. Moreover, the commonly em-
ployed Ru(TPP)(CO) has been shown to be a solid solution
of authentic [Ru(TPP)(CO)] and [Ru(TPP)(CO)(H2O)] in
variable proportions.
Acknowledgment. We thank MIUR (Programmi di
Ricerca Scientifica di Rilevante Interesse Nazionale, PRIN
2003033857), the Fondazione Provinciale Comasca, and the
University of Insubria (Progetto di Ateneo “Sistemi Polia-
zotati”) for financial support. We are grateful to Mr. Pasquale
Illiano (Dipartimento di Chimica Inorganica Metallorganica
e Analitica, Universita` degli Studi di Milano) for performing
PGSE analysis. N.M. and A.S. thank Dr. A. Kern (Bruker
AXS, Karlsrube, Germany) for providing the â-version of
TOPAS.
Crystallographic data (excluding structure factors) for the
(36) Werner, P. E.; Eriksson, L.; Westdahl, M. J. Appl. Crystallogr. 1985,
18, 367-370.
Supporting Information Available: X-ray data for 3-6.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(37) TOPAS, version 3.1; Bruker AXS: Karlsrube, Germany, 2004.
(38) Ke, M.; Sishta, C.; James, B. R.; Dolphin, D.; Sparapany, J. W.; Ibers,
J. A. Inorg. Chem. 1991, 30, 4766-4771.
(39) Christensen, A. N.; Grand, A.; Lehman, M. S.; Cox, D. E. Acta Chem.
Scand. 1990, 44, 103-105.
IC048587W
Inorganic Chemistry, Vol. 44, No. 6, 2005 2049