Synthesis, characterization, and molecular structures of nitrosyl nitrito complexes of osmium porphyrins: Disproportionation of nitric oxide in its reaction with Os(P)(CO) (P = porphyrinato dianion)

Article abstract of DOI:10.1139/v03-091

The Os(P)(NO)(ONO) compounds (P = TTP, TMP, OEP, TmTP; TTP = 5,10,15,20-tetra-p-tolylporphyrinato dianion, TMP = 5,10,15,20- tetramesitylporphyrinato dianion, OEP = octaethylporphyrinato dianion, TmTP = tetra(m-tolyl)porphyrinato dianion) have been prepar

Full text of DOI:10.1139/v03-091

872
Synthesis, characterization, and molecular
structures of nitrosyl nitrito complexes of osmium
porphyrins: Disproportionation of nitric oxide in its
reaction with Os(P)(CO) (P = porphyrinato dianion)
Felipe A. Leal, Ivan M. Lorkovic, Peter C. Ford, Jonghyuk Lee, Li Chen,
Lindsey Torres, Masood A. Khan, and George B. Richter-Addo
Abstract: The Os(P)(NO)(ONO) compounds (P = TTP, TMP, OEP, TmTP; TTP = 5,10,15,20-tetra-p-tolylporphyrinato
dianion, TMP = 5,10,15,20-tetramesitylporphyrinato dianion, OEP = octaethylporphyrinato dianion, TmTP = tetra(m-
tolyl)porphyrinato dianion) have been prepared from the reaction of the precursor carbonyl complexes Os(P)(CO) with
excess nitric oxide. Nitrous oxide was detected as a by-product of the reaction. The IR spectra of the Os(P)(NO)(ONO)
compounds (as KBr pellets) reveal bands in the 1790–1804 cm–1 range that are assigned to υNO. The IR spectra also
reveal two new bands for each complex in the 1495–1531 and 913–962 cm–1 ranges indicative of O-bound nitrito lig-
ands. The linearity of the bound NO groups and the O-binding of the trans nitrito ligands in the Os(P)(NO)(ONO)
complexes are evident in the single-crystal X-ray crystal structures of the TTP and TMP derivatives. The kinetics of
the reaction were studied by stopped-flow mixing techniques. Spectroscopic analysis of rapidly mixed solutions of
Os(P)CO and NO in toluene showed a biphasic approach to the Os(P)(NO)(ONO) and N2O products, owing to the
starting material Os(P)CO scavenging CO formed during the reaction to give Os(P)(CO)2 (KCO = 106 M–1). The
dicarbonyl was the only transient species observed. It is proposed that the rate-determining step of the reaction leading
to Os(P)(NO)(ONO) is NO displacement of CO from Os(P)(CO) via initial formation of an unstable 19 electron
Os(P)(NO)(CO) intermediate.
Key words: osmium, nitric oxide, X-ray, nitrosyl, porphyrin, kinetics.
Résumé : Les composés du type Os(P)(NO)(ONO), où P = TTP, TMP, OEP, TmTP; TTP= le dianion 5,10,15,20-tétra-
p-tolylporphyrinato, TMP = le dianion 5,10,15,20-tétramésitylporphyrinato, OEP = le dianion octaéthylporphyrinato,
TmTP = le dianion tétra(m-tolyl)porphyrinato, ont été préparés à partir de la réaction des complexes carbonylés précur-
seurs OS(P)(CO) en présence d’un excès d’oxyde nitrique. On a détecté l’oxyde nitreux comme produit secondaire de
la réaction. Les spectres IR des composés Os(P)(NO)(ONO) (sous forme de pastilles de KBr) révèlent des bandes dans
la région de 1790–1804 cm–1 que l’on attribue à la vibration υNO. Les spectres IR révèlent également, pour chaque
complexe, deux nouvelles bandes dans les régions : 1495–1531 et 913–962 cm–1 caractéristiques des ligands nitrito liés
par l’oxygène. La linéarité des liens des groupes NO et le lien impliquant l’oxygène des ligands nitrito trans dans les
complexes Os(P)(NO)(ONO) apparaissent clairement dans les structures des dérivés TTP et TMP déterminées par cris-
tallographie de rayons X sur un monocristal. On a étudié la cinétique de la réaction par les techniques de mélange à
flux stoppé. L’analyse spectroscopique des solutions OS(P)CO et de NO rapidement mélangées dans le toluène mon-
trent une approche biphasique aux produits Os(P)(NO)(ONO) et N2O à cause du produit de départ OS(P)CO qui piège
le CO formé au cours de la réaction pour donner le OS(P)(CO)2 (KCO = 106 M–1). Le composé dicarbonylé est la seule
espèce transitoire observée. On suggère que l’étape déterminante de la réaction conduisant au Os(P)(NO)(ONO) est le
déplacement du CO par NO à partir du OS(P)(CO) via la formation initiale d’un intermédiaire instable Os(P)(NO)(CO)
à 19 électrons.
Mots clés : osmium, oxyde nitrique, rayons X, nitrosyle, porphyrine, cinétique.
[Traduit par la Rédaction] Leal et al. 881
Received 14 November 2002. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 14 July 2003.
F.A. Leal, I.M. Lorkovic, and P.C. Ford.1 Department of Chemistry and Biochemistry, University of California, Santa Barbara,
California 93106, U.S.A.
J. Lee, L. Chen, L. Torres, M.A. Khan, and G.B. Richter-Addo.2 Department of Chemistry and Biochemistry, University of
Oklahoma, 620 Parrington Oval, Norman, Oklahoma 73019, U.S.A.
1Corresponding author (e-mail: ford@chem.ucsb.edu).
2Corresponding author (e-mail: grichteraddo@ou.edu).
Can. J. Chem. 81: 872–881 (2003)
doi: 10.1139/V03-091
© 2003 NRC Canada

Leal et al.
873
with Os(P)(CO) is scavenged by the remaining Os(P)(CO) to
give the dicarbonyl complex Os(P)(CO)2 in a “dead-end”
equilibrium. The latter eventually is also converted to
Os(P)(NO)(ONO).
Introduction
Metalloporphyrin-induced NO→N2O conversions are very
important biologically in the global nitrogen cycle (1). For
example, nitric oxide reductase cytochrome P450nor from
the fungus Fusarium oxysporum reduces NO to N2O, and
the enzyme contains heme at the active site (2, 3). The bac-
terial nitric oxide reductase from Paracoccus denitrificans
also contains heme at the dinuclear active site and catalyzes
the reduction of NO to N2O (4, 5). The subject of biological
NO reduction has been recently reviewed (1).
Heme-dependent NO oxidations are gaining widespread
recognition in the chemistry of NO. For example, some
flavohemoglobins have recently been described as NO
dioxygenases (6, 7). Metal-assisted NO→NO2 conversions
are also important in the biological and urban environment.
In particular, knowledge of the efficiency of metal-assisted
conversions of NO → NxOy is an essential component in the
design of catalytic converters for automobile engines (8).
While the mechanisms of such conversions are not known
with certainty, synthetic metal complexes that catalyze the
formation of NO2 (bound or free) from NO and vice versa,
as well as the structures of metal–NO2 coordination com-
plexes, are of interest in this regard. In some instances, the
addition of oxygen (or air) to the metal–NO precursor is
necessary for the production of the metal–NO2 complex. We
recently demonstrated such a metal–NO to metal–NO2 con-
version using an iron nitrosyl porphyrin (9). Other metallo-
porphyrin–NO complexes of Co and Rh have been shown to
behave similarly (10, 11).
There are many examples of non-heme transition metal
complexes that promote NO disproportionation to give metal
nitrite derivatives and N2O (12). However, we have found
that ferrous porphyrins do not promote NO disproportiona-
tion in the absence of oxygen; they react with NO to pro-
duce only mononitrosyl or weakly bound dinitrosyl
complexes (13). The apparently contradictory results con-
cerning the ability of iron porphyrins to promote the dispro-
portionation of NO (in the presence or absence of trace air)
are discussed in ref. 14. Iron porphyrin nitrite complexes
containing the N-bound nitrite ligand are known, and their
solid-state structures have been reviewed (15). We reported
that ruthenium porphyrins react with NO in the absence of
oxygen to produce the nitrosyl nitrito complexes
Ru(P)(NO)(ONO) and N2O (16–18), and we spectroscopi-
cally characterized a trans-dinitrosyl intermediate Ru(P)(NO)2
during this NO disproportionation reaction (19, 20). The oc-
currence of NO disproportionation with ruthenium, but not
with iron, led us to extend these studies to the osmium
porphyrin congeners.
Experimental section
All reactions were performed under an atmosphere of
prepurified nitrogen using standard Schlenk glassware and
(or) in an Innovative Technology Labmaster 100 Dry Box.
Solutions for spectral studies were also prepared under a ni-
trogen atmosphere. Solvents were distilled from appropriate
drying agents under nitrogen just prior to use: benzene (Na),
toluene (Na or CaH2), hexane, chloroform, and cyclohexane
(CaH2).
Chemicals
The Os(P)(CO) compounds were prepared by literature
methods (P = TTP, TMP, OEP, TmTP; TTP = 5,10,15,20-
tetra-p-tolylporphyrinato dianion, TMP = 5,10,15,20-tetra-
mesitylporphyrinato dianion, OEP = octaethylporphyrinato
dianion, TmTP = tetra(m-tolyl)porphyrinato dianion) (21).
Os(OEP)(CO) was also purchased from Midcentury Chemi-
cals. Chloroform-d (99.8%) was obtained from Cambridge
Isotope Laboratories. Nitric oxide (98%, Matheson Gas) for
the synthesis work was passed through KOH pellets and two
cold traps (dry ice–acetone, –78°C) to remove higher nitro-
gen oxides. For experiments involving detection of evolved
N2O, NO was purified as described elsewhere (14).
Instrumentation
Infrared spectra were recorded on a Bio-Rad FT-155 FT-
1
IR spectrometer. H NMR spectra were obtained on Varian
300 MHz or 400 MHz spectrometers and the signals refer-
enced to the residual signal of the solvent employed. All
coupling constants are in Hz. FAB mass spectra were ob-
tained on a VG-ZAB-E mass spectrometer.
Preparation of Os(TTP)(NO)(ONO)
A Schlenk flask was charged with Os(TTP)(CO) (0.055 g,
0.062 mmol) and benzene (30 mL). The mixture was stirred
to generate an orange-red solution, and NO gas was then
bubbled through the solution for ~15 min. The solvent was
removed in vacuo, and the residue was redissolved in ben-
zene (10 mL) and filtered through a neutral alumina column
(1.5 × 15 cm). The red band was collected and dried in
vacuo. The product was further purified by crystallization
In this paper, we show that osmium porphyrins react with
NO to give nitrosyl nitrito products with the release of N2O.
Solid-state molecular structures for two of these nitrosyl
nitrito products have been obtained and, to the best of our
knowledge, represent the first published osmium nitrito X-
ray crystal structures to be reported. While the products are
analogous to those found for the reaction of NO with similar
ruthenium porphyrins Ru(P)(CO) (19, 20), unlike the ruthe-
nium analogues, reactions of NO with Os(P)(CO) proceed
without the generation of observable nitrosyl-containing in-
termediates. Instead, the CO released from reaction of NO
from
a
toluene–hexane mixture at –20°C to give
Os(TTP)(NO)(ONO) (0.025 g, 0.027 mmol, 43% isolated
yield). IR (KBr) (cm–1): υNO = 1804 (s), υONO = 1528 (m),
1
921 (m). H NMR (CDCl3) δ: 9.00 (s, 8H, pyrrole-H of
TTP), 8.15 (d, J = 8 Hz, 4H, o-H of TTP), 8.05 (d, J = 8 Hz,
4H, o′-H of TTP), 7.56 (app t (overlapping d’s), 8H, m/m′-H
of TTP), 2.70 (s, 12H, CH3 of TTP). Low-resolution FAB-
MS m/z (%): 906 ([Os(TTP)(ONO)]+, 15%), 890
([Os(TTP)(NO)]+, 100%), 860 ([Os(TTP)]+, 18%). This
compound was prepared previously in low yield from the re-
action of Os(TTP)(NO)(S-i-C5H11) with excess NO (22).
© 2003 NRC Canada

874
Can. J. Chem. Vol. 81, 2003
instrument. Stopped-flow mixing experiments were per-
formed on an Applied Photophysics SX17-MV instrument
with custom made tonometers (20) for NO solution prepara-
tion and transfer to drive syringes. Flow IR experiments
were performed with a home-built, manually actuated
stopped-flow mixing apparatus utilizing gas tight syringes
(Hamilton 5 mL) and PEEK tubing, valves, and fittings
(Upchurch), connected after mixing to a standard amalgam-
sealed CaF2 cell inserted into the FT-IR instrument. To
avoid back-pressure-induced IR cell rupture, the stop syringe
was disconnected and the output directed to a waste con-
tainer. FT-IR spectra showing intermediate species in the re-
action of Os(P)(CO) with NO were obtained by a single
interferometric scan during constant flow of freshly mixed
(aging time ~0.2 s) solutions through the cell.
Preparation of Os(TMP)(NO)(ONO)
A
Schlenk flask was charged with Os(TMP)(CO)
(0.082 g, 0.082 mmol) and benzene (30 mL). The mixture
was stirred to generate a dark orange-red solution, and NO
gas was then bubbled through the solution for 30 min. Dur-
ing this time, the color of the reaction mixture turned bright
red. The solvent was removed in vacuo, and the residue was
redissolved in benzene (15 mL) and filtered through a neu-
tral alumina column (2 × 20 cm). The red band was col-
lected, and the filtrate was dried in vacuo to give
Os(TMP)(NO)(ONO) (0.077 g, 0.073 mmol, 90% isolated
yield). IR (KBr) (cm–1): υNO = 1799 (s), υONO = 1531 (m),
1
919 (m). H NMR (CDCl3) δ: 8.77 (s, 8H, pyrrole-H of
TMP), 7.29 (s, 4H, m-H of TMP), 7.26 (s, 4H, m′-H of
TMP), 2.62 (s, 12H, p-CH3 of TMP), 1.95 (s, 12H, o-CH3 of
TMP), 1.69 (s, 12H, o′-CH3 of TMP). Low-resolution FAB-
MS m/z (%): 1048 ([Os(TMP)(NO)(ONO)]+, 6%), 1019
([Os(TMP)(ONO) + H]+, 43%), 1002 ([Os(TMP)(NO)]+,
100%), 972 ([Os(TMP)]+, 13%).
CO binding equilibrium constant measurements
Carbon monoxide (Praxair, 99.5%) was added by gas-tight
syringe to toluene solutions (~5 mL) of Os(OEP)(CO) and
Os(TmTP)(CO) (~8 µM) under Ar (PT = 760 torr (1 torr =
133.322 Pa)) within special cuvette cells (~70 mL total vol-
ume) at 21 ± 1°C. [CO] was calculated from the known sol-
ubility of CO in toluene (25) and the known volume of
solution and headspace. Under these conditions the [CO]
may be less than [Os] within the solution, but the total
amount of CO within the closed system of the cell is always
at least ten times greater than the amount of Os in the solu-
tion, and therefore [CO] may be considered to be constant,
independent of the extent of reaction.
The following two compounds were also prepared simi-
larly.
Os(OEP)(NO)(ONO)
43% isolated yield. IR (KBr) (cm–1): υNO = 1790 (s),
υONO = 1495 (m), 962 (s br (overlap with porphyrin band)).
1H NMR (CDCl3) δ: 10.41 (s, 4H, meso-H of OEP), 4.16 (q,
J = 8 Hz, 16H, CH2CH3 of OEP), 2.00 (t, J = 8 Hz, 24H,
CH2CH3 of OEP). This compound has similar spectral prop-
erties to a compound formulated as Os(OEP)(NO)2 (23).
Os(TmTP)(NO)(ONO)
This compound was obtained in 95% yield (by H NMR
spectroscopy) when the reaction was performed in dry tolu-
ene or methylcyclohexane and in 25% yield (by H NMR
Structural determinations by X-ray crystallography
Crystal data were collected on a Siemens (Bruker) P4
diffractometer using monochromated Mo Kα radiation (λ =
0.71073 Å).3 The data were corrected for Lorentz and polar-
ization effects, and empirical absorption corrections based
on ψ-scans were applied (26). The structures were solved by
the heavy atom method using the SHELXTL 5.03 (Bruker)
system and refined by full-matrix least squares on F2 using
all reflections (SHELXL-93). The thermal ellipsoids in
Figs. 1 and 2 are drawn at the 35% probability level. Details
of the crystal data and refinement are given in Table 1.
1
1
spectroscopy) when performed in CH2Cl2. The compound
was further purified by chromatography over silica gel using
a pentane–CH2Cl2 (2:1) solvent mixture. IR (KBr) (cm–1):
υNO = 1803 (s), υONO = 1528 (m), 913 (m). IR (cyclohexane)
(cm–1): υNO = 1805 (s), υONO = 1540 (m). H NMR (CDCl3)
1
δ: 9.02 (s, 8H, pyrrole-H of TmTP), 8.10 (m, 4H, o-H of
TmTP), 7.99 (m, 4H, o′-H of TmTP), 7.50–7.60 (m, 8H,
m/p-H of TmTP), 2.66 (s, 6H, CH3 of TmTP), 2.62 (s, 6H,
CH3 of TmTP).
Os(TTP)(NO)(ONO).toluene
A suitable crystal for structure determination was grown
by recrystallization of the compound from toluene–hexane at
–20°C. All the non-hydrogen atoms were refined anisotro-
pically except the atoms (C25–C33) belonging to the disor-
dered toluene molecules, which were refined isotropically.
Hydrogen atoms were included in the refinement with ideal-
ized parameters; the hydrogen atoms for the disordered sol-
vent molecules were not included.
Considerable difficulty was encountered during the refine-
ment because of the disorder of the Os atom and the atoms
belonging to the axial groups and the solvent molecules. The
structure was solved initially in the non-centric space group
Cc and gave very poor refinement with unreasonable N—O
Kinetics experiments
NO was plumbed via stainless steel lines through a stain-
less steel column of Ascarite (Thomas Scientific) to scrub
out NO2 and (or) N2O3. When desired, the N2O impurity
was removed by passage of NO through a cold (–78°C) acti-
vated silica column (6′ × 1/4" i.d. stainless steel, Alltech).
NO was quantified manometrically using known solubilities
in toluene and cyclohexane (24) and transferred cryogeni-
cally to desired reaction flasks of known total volume and
solution volume fitted with Teflon high vacuum stopcocks.
Infrared solution spectra were recorded in amalgam-sealed
cells with CaF2 windows (ICL) on a Bio-Rad FTS-60 FT-IR
3 Supplementary data (drawings and tables of crystallographic data for Os(TTP)(NO)(ONO) and Os(TMP)(NO)(ONO)) may be purchased
from the Depository of Unpublished Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0S2, Canada
(http://www.nrc.ca/cisti/irm/unpub_e.shtml for information on ordering electronically). CCDC 214514 and 214515 contain the supplemen-
tary data for this paper. These data can be obtained, free of charge, via www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge, U.K.; fax +44 1223 336033; or deposit@ccdc.cam.ac.uk).
© 2003 NRC Canada

Leal et al.
875
Fig. 2. (a) Molecular structure of Os(TMP)(NO)(ONO). Hydro-
gen atoms have been omitted for clarity. Only one of the two
disordered positions of the Os atom and the NO and ONO
groups is shown; (b) View of the orientation of the nitrito ligand
relative to the porphyrin skeleton. The porphyrin mesityl sub-
stituents and the trans NO ligand have been omitted for clarity.
Fig. 1. (a) Molecular structure of Os(TTP)(NO)(ONO). Hydro-
gen atoms have been omitted for clarity. Only one of the two
disordered positions of the Os atom and the NO and ONO
groups is shown; (b) View of the orientation of the nitrito ligand
relative to the porphyrin skeleton. The porphyrin tolyl substitu-
ents and the trans NO ligand have been omitted for clarity.
0.0583 for 3651 “observed reflections” [I > 2σ(I)] and wR2 =
0.1665 for all reflections (4421 unique data) were obtained.
and O—N—O bond lengths and several non-positive defi-
nite temperature factors. The structure was then solved and
refined in the centric space group C2/c, where the molecule
lies on the inversion center and only half of the molecule is
unique. Thus, the axial NO and ONO groups are disordered
over both sides of the porphyrin plane. The initial refine-
ment used a model in which the Os atom was placed on a
special position at the inversion center. This model gave
elongated thermal ellipsoids for the Os atom, and the Os—N3
and Os—O2 distances were unreasonable. Finally, the Os
atom was allowed to refine out of the porphyrin plane to-
wards the NO ligand (0.241(3) Å out of the 24-atom mean
porphyrin plane). In the final model, the Os atom and the
NO and ONO groups are completely disordered over two
sites. In addition to the disorder of the molecule, the crystal
lattice contains two sites with partially disordered toluene
solvent molecules. One of these toluene solvent sites is 50%
populated, and the other site is also 50% populated; disorder
at these sites results in the two sites having only 25% occu-
pancy. The toluene solvent molecules were refined isotro-
pically because of this disorder, and hydrogen atoms for
these molecules were not included in the refinement.
SHELXTL restraints (ISOR, DELU, SIMU, and DFIX) were
needed to stabilize the refinement. The final values of R1 =
Os(TMP)(NO)(ONO)·3(benzene)
Suitable crystals for X-ray crystallography were grown by
slow evaporation of a benzene solution of the compound at
room temperature under inert atmosphere. All the non-
hydrogen atoms were refined anisotropically, and the hydro-
gen atoms were included in the refinement with idealized
parameters. The asymmetric unit contains one molecule of
Os(TMP)(NO)(ONO) and three molecules of benzene. The
Os atom and the NO and ONO groups are disordered over
two sites along the axial direction. In addition, one of the
benzene solvent molecules is disordered over two sites. At-
tempts to refine the two fragments did not yield a stable re-
finement. Thus, the geometry of the two fragments was
idealized in the final cycles of refinement, and the thermal
parameters were fixed at the values obtained from the earlier
refinement cycles. The final values of R1 = 0.0374 for 8660
“observed reflections” [I > 2σ(I)] and wR2 = 0.1036 for all
reflections (10 961 unique data) were obtained.
Os(OEP)(NO)(ONO)
Both laboratories have independently crystallized this
compound. Although we have not been able to grow high-
© 2003 NRC Canada

876
Can. J. Chem. Vol. 81, 2003
Table 1. Crystal data and structure refinement.
Compound
Formula (fw)
Os(TTP)(NO)(ONO)·toluene
C55H44N6O3Os (1027.16)
Os(TMP)(NO)(ONO)·3(C6H6)
C74H70N6O3Os (1281.56)
T (K)
Crystal system
Space group
153(2)
Monoclinic
C2/c
173(2)
Triclinic
P1
Unit cell dimensions
a (Å), α (°)
b (Å), β (°)
26.861(5), 90
9.947(2), 119.19(3)
21.933(4), 90
5116.1(18), 4
1.334
12.1823(12), 81.362(9)
13.031(2), 80.014(10)
20.732(3), 76.857(10)
3135.2(7), 2
c (Å), γ (°)
V (Å3), Z
Dcalcd (g cm–3)
1.358
Absorption coefficient (mm–1)
F(000)
2.539
2064
2.087
1312
Crystal size (mm)
θ range for data collection (°)
Index ranges
0.48 × 0.36 × 0.28
2.13–24.99
0 ≤ h ≤ 31, 0 ≤ k ≤ 11, –26 ≤ l ≤ 22
4522
4421 (Rint = 0.0649)
Empirical
0.8876 and 0.5998
4421/14/327
0.38 × 0.36 × 0.26
1.80–25.00
–13 ≤ h ≤ 14, –14 ≤ k ≤ 15, –24 ≤ l ≤ 24
11 526
10 966 (Rint = 0.0337)
Semi-empirical
0.4872 and 0.3951
10 961/6/841
Reflections collected
Independent reflections
Absorption correction
Max. and min. transmission
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2σ (I)]
R indices (all data)
Largest diff. peak and hole (e Å–3)
1.172
1.068
R1 = 0.0583, wR2 = 0.1553
R1 = 0.0732, wR2 = 0.1665
1.573 and –0.706
R1 = 0.0374, wR2 = 0.0858
R1 = 0.0568, wR2 = 0.1036
0.794 and –1.357
quality, good-sized crystals, the results of separate prelimi-
nary X-ray crystallographic analyses (by both laboratories)
on the crystals obtained to date unambiguously confirm the
nitrosyl nitrito Os(OEP)(NO)(ONO) formulation.
porphyrin complexes of the form M(P)(X)(Y) (29, 30). The
FAB mass spectrum of Os(TTP)(NO)(ONO) reveals peaks
due to loss of the NO and (or) ONO ligands.
We were also able to prepare other nitrosyl nitrito com-
plexes of the type Os(P)(NO)(ONO), where P = TMP, OEP,
and TmTP. The IR spectra of the complexes (as KBr pellets)
reveal their nitrosyl υNO bands at 1803 cm–1 (TmTP),
1799 cm–1 (TMP), and 1790 cm–1 (OEP), reflecting the in-
creasing electron donor properties of the porphyrin macro-
Results and discussion
We reported that the Ru(P)(CO) compounds (P = OEP,
TPP, TmTP) react with NO in solution to give the nitrosyl
nitrito complexes Ru(P)(NO)(ONO) and N2O as the final
products (16, 17, 19, 20). Using similar methodologies, we
have prepared the nitrosyl nitrito complexes of osmium por-
phyrins. Thus, the reaction of Os(TTP)(CO) with NO in ben-
zene at ambient temperature for ~15 min gives, after
workup, the nitrosyl nitrito product Os(TTP)(NO)(ONO) in
43% isolated yield. The product is soluble in dichloro-
methane, benzene, and toluene, but is insoluble in hexane.
The IR spectrum of Os(TTP)(NO)(ONO) (as a KBr pellet)
reveals a band at 1804 cm–1 assigned to υNO. This υNO band
is of higher frequency than the corresponding bands for the
alkoxide complex Os(TTP)(NO)(O-i-C5H11) (1770 cm–1)
(22), the thiolate derivative Os(TTP)(NO)(S-i-C5H11)
(1760 cm–1) (22), and the organometallic compound
Os(TTP)(NO)Me (1732 cm–1) (27). In addition, the IR spec-
trum indicates new bands at 1528 and 921 cm–1 due to the
O-bound nitrito ligand (28).
cycles along the series TTP
TmTP < TMP < OEP, as
expected. The TMP, OEP, and TmTP derivatives also show
medium intensity bands assignable to nitrito υN-O. The N2O
gaseous by-product formed during the reaction was identi-
fied by IR spectroscopy as described previously (19).
Solid-state molecular structures
As mentioned in the Introduction, no X-ray crystal struc-
tures of osmium nitrito complexes had been reported prior to
our study. Interestingly, there is only a single report of an os-
mium nitro (i.e., η1-N bound) structure in the literature,
namely that for Os(H2L)(PPh3)2(CO)(η1-NO2)·H2O (L = 3-
ethyliminio-5-methyl-2-oxidophenyl-C1,O) (31), although
some bridging nitro moieties [Os-N(O)O-Os] have been
characterized by crystallography (32–34).
We were successful in obtaining suitable crystals of two
of the Os(P)(NO)(ONO) compounds for single-crystal X-ray
diffraction studies. The solid-state molecular structure of
Os(TTP)(NO)(ONO) is shown in Fig. 1a, and the axial
nitrito ligand conformation is shown in Fig. 1b. As is evident
in Fig. 1, the ONO group is bound to the Os center through
an O-atom, and the ONO group essentially bisects adjacent
porphyrin nitrogen atoms, with an N2-Os-O2-N4 torsion an-
gle of 33° (Fig. 1b). The molecular structure of the related
The 1H NMR spectrum of Os(TTP)(NO)(ONO) (in
CDCl3) reveals peaks due to the pyrrole protons of the por-
phyrin macrocycle (at δ 9.00) and the tolyl substituents. The
inequivalence of the ortho protons of the tolyl substituents
suggests restricted rotation of the tolyl groups in this unsym-
metrical Os(P)(NO)X compound; such a feature of restricted
rotation is not uncommon for p-substituted tetraphenyl-
© 2003 NRC Canada

Leal et al.
877
Fig. 3. Structural data for Os(TTP)(NO)(ONO) and Os(TMP)(NO)(ONO). Selected bond lengths and angles are shown in the top
sketches. Perpendicular atom displacements from the 24-atom porphyrin planes (in 0.01 Å units) are shown in the bottom sketches.
Os(TMP)(NO)(ONO) is shown in Fig. 2. In this compound,
the nitrito ligand is also bound to osmium through an O-
atom, and the ONO ligand essentially bisects adjacent
porphyrin nitrogens, with an N1-Os-O1-N6 torsion angle of
46°. Selected bond lengths and angles for both compounds
and atom displacements from the 24-atom mean porphyrin
planes are displayed in Fig. 3.
There are several features to note from these solid-state
structures. First, the nitrosyl Os-NO groups are essentially
linear. Second, the Os atoms are displaced from the 24-atom
mean porphyrin plane towards the axial NO ligand; 0.24 Å
for the TTP compound and 0.22 Å (or 0.15 Å for the second
disordered component) for the TMP compound. Third, the
bond lengths within the nitrito ligand are consistent with the
Os-O-N=O formulation; thus, the O—N bond lengths are
longer than the nitrito N=O bond lengths. Surprisingly, this
is not always the case for metal nitrito crystallographic data.
Selected structural data for M(NO)(ONO)-containing com-
plexes reported to date are collected in Table 2. However,
there is little consistency in the reported bond lengths in-
volving the ONO ligand, and this may be partly because of
the fact that many of these structures suffer from ligand dis-
order problems.
Proposed reaction pathway for the formation of nitrosyl
nitrito species
UV–vis spectroscopic monitoring of the reaction of a di-
lute solution of Os(OEP)(CO) in toluene with NO (Fig. 4,
top) reveals an apparent clean conversion of Os(OEP)(CO)
to Os(OEP)(NO)(ONO). Stopped-flow experiments moni-
tored at several single-wavelengths suggest that the reaction
is biphasic, involving the formation of a transient species
with spectral features similar but not identical to that of
Os(OEP)(CO) (Fig. 4, bottom and inset). Within a few sec-
onds, this intermediate spectrum gives way to that of the fi-
nal Os(OEP)(NO)(ONO) product. Isosbestic points for the
slower and faster processes are observed at 536 and 553 nm
and therefore the faster and slower processes may be indi-
vidually followed at these monitoring wavelengths. Under
these conditions, the reactions appeared to be first order with
the respective rate constants kobs(536) = 7 ± 1 s–1 (fast) and
k
obs(553) = 0.7 ± 0.1 s–1 (slow). Analogous UV–vis spectral
features are observed for the TmTP derivative as well.
To gain further insight into these processes, stopped-flow
mixing experiments coupled with FT-IR spectroscopic detec-
tion were performed on the more soluble TmTP and OEP
derivatives. The FT-IR difference spectrum of the solution
© 2003 NRC Canada

878
Can. J. Chem. Vol. 81, 2003
Table 2. Selected structural data for metal nitrosyl nitrito complexes.
M—ONO
cis–transa (Å)
MO—NO
(Å)
MON—O
(Å)
M-O-NO
(°)
MO-N-O
(°)
Compound
Reference
CpCr(NO)2(ONO)
[Ni(NO)(ONO)dppe]2
[Fe(L)(NO)(ONO)(NO2)]ClO4
cis
cis
cis
1.982(4)
2.123(12)
1.914(3)
[1.935(3)]
1.96(1)
2.030(8)
2.033(4)
2.08
2.011 avg.
1.953(8)
2.080(8)
2.00(2)
1.292(8)
1.097(14)
1.322(5)
[1.305(5)]
1.34(1)
1.182(9)
1.192(13)
1.216(6)
[1.213(6)]
1.13(1)
125(1)
115(2)
39
40
41
b
109.0(12)
125.1(3)
[125.7(3)]
115.0(7)
125.6d
127.4(17)
116.3(4)
[116.1(4)]
112(1)
118(1)
116.3(6)
118
116.5 avg.
108.2(8)
112.2(14)
109(5)
108.0(30)
117.3(9)
110.9(20)
c
Cr(py)3(NO)(ONO)2·py
cis
cis
cis
cis
trans
trans
trans
trans
trans
trans
trans
42
43
43
44
37
35
36
16
17
17
38
[Ru(NO)(ONO)(2,2′-bpy)2](PF6)2
[Ru(NO)(ONO)(2,2′-bpy)(py)2](PF6)2
Mn(NO)2(ONO)(PEt3)2
1.24(1)
1.324(6)
1.17(1)
1.227(7)
e
124.0d
1.19
e
123
e
Ru(sal2en)(NO)(ONO)f
e
[Mn(Pc)(NO)(ONO)](PNP)g
[Fe(TpivPP)(NO)(ONO)]–h
Ru(TPP)(NO)(ONO)i
1.189(7)
1.188(14)
0.94(5)
1.16(2)
1.214(10)
1.148(18)
1.165(7)
1.28(2)
1.33(4)
1.23(2)
1.188(9)
1.126(25)
117.9(6)
124.1(8)
137(3)
137.3(18)
122.0(6)
124.0(11)
Ru(TPP)(NO)(ONO)i
Ru(OEP)(NO)(ONO)
Ru(TTP)(NO)(ONO)j,k
1.90(2)
1.984(6)
1.998(6)
[1.136(35)] [1.447(39)] [128.0(18)] [92.2(24)]
Os(TTP)(NO)(ONO)i
Os(TMP)(NO)(ONO)i
trans
trans
2.000(6)
2.020(4)
[1.996(4)]
1.296(11)
1.240(8)
1.188(12)
1.179(8)
123.2(7)
122.2(4)
[121.6(4)]
113.8(11)
115.1(7)
This work
This work
[1.258(7)]
[1.154(8)]
[113.4(6)]
aCis or trans with respect to NO and ONO.
bdppe = Ph2PCH2CH2PPh2.
cTwo molecules in the unit cell. The values for the second molecule are in brackets. L = 1,4,7-triazacyclononane.
dData obtained from the Cambridge Structural Database.
eMetrical data not reported.
fsal2en = N,N′-ethylenebis(salicylideneiminato). There are two molecules in the unit cell.
gPNP = (Ph3P)2N+.
hTpivPP = α,α,α,α-tetrakis(o-pivalamidophenyl)porphyrinato dianion. The ONO group is present as a linkage isomer together with the N-bound NO2
group in the same molecule.
iThe axial NO and ONO ligands are disordered over the two porphyrin faces.
jThe ONO ligand is disordered. The values for the second disordered component are in brackets.
kTwo molecules in the unit cell, and only the values for molecule A are shown.
obtained from the reaction of Os(TmTP)(CO) in cyclohex-
ane with NO (Fig. 5, top) revealed, in addition to
Os(TmTP)(CO) (υCO 1924 cm–1) and the final product
Os(TmTP)(NO)(ONO) (υNO 1805 cm–1; υONO 1540 cm–1), a
new species displaying a strong band at 1974 cm–1.
Since the independent reaction of Os(TmTP)(CO) with
added CO gave a new species displaying a υCO band at
1974 cm–1 twice as intense as that of Os(TmTP)(CO)
(Fig. 5, bottom), we assign this band to the trans dicarbonyl
complex Os(TmTP)(CO)2. Therefore, its appearance as a
transient in the reaction of NO with Os(TmTP)(CO) can be
attributed to the Os(TmTP)(CO) scavenging the CO released
during the reaction.
ment described above. No other species with distinctive IR
spectral signatures, such as expected for Os–nitrosyl inter-
mediates, were detected during the stopped-flow study of the
reaction of Os(P)(CO) with NO. This is in sharp contrast to
the analogous reaction of Ru(P)(CO) with NO where the
dinitrosyl complex Ru(P)(NO)2 was a spectroscopically ob-
servable intermediate (20). Given that under the reaction
conditions [NO] is always much greater than [CO], detecting
Os(P)(CO)2 as the only transient suggests that K2 for CO
binding to Os(P)(CO) is at least four orders of magnitude
larger than the equilibrium constant for analogous binding of
NO.
Scheme 1 is a proposed explanation for these kinetics and
spectral
observations
during
the
formation
of
Formation of the dicarbonyl complexes Os(P)(CO)2
(eq. [1]) was studied independently in toluene by probing
optical spectrum changes upon addition of CO to toluene so-
lutions of Os(P)(CO) (e.g., Fig. 6). In this manner the equi-
librium constants K2 = 0.9 ± 0.2 × 106 and 1.0 ± 0.2 ×
106 M–1 were determined for P = TmTP and OEP, respec-
tively. The latter value is ~102 larger than that observed for the
corresponding Ru(OEP)(CO) complex (1.1 × 104 M–1) (20).
Os(P)(NO)(ONO) from Os(P)(CO) and NO. This is based
largely on the behavior of the ruthenium analogues for
which the dinitrosyl intermediate Ru(P)(NO)2 was observed
(19, 20). The initial step is very likely the reversible and
endergonic formation of the 19-electron complex
Os(P)(CO)(NO). Upon rate-limiting CO dissociation from
Os(P)(CO)(NO), a rapid sequence of steps would lead to the
disproportionation products. However, CO is also a product,
and it will be rapidly scavenged by Os(P)(CO) to give the
much less reactive Os(P)(CO)2 in a dead-end equilibrium.
Thus, the reaction slows, owing to the lower effective con-
centration of Os(P)(CO). Since no other transient species
[1]
Os(P)(CO) + CO ↔ Os(P)(CO)2
The high affinity of Os(P)(CO) for CO explains the detec-
tion of Os(P)(CO)2 as a transient in the stopped-flow experi-
© 2003 NRC Canada

Leal et al.
879
Fig. 5. Top: Difference FT-IR spectrum of the solution obtained
Fig. 4. Top: UV–vis spectra of the reactant Os(OEP)(CO) in to-
luene and the product of its reaction with nitric oxide,
by mixing solutions of Os(TmTP)(CO) (~100 µM in cyclohexane
after mix) with NO (~4 mM in cyclohexane after mix). The dif-
ference spectra shown are the result of subtracting the spectrum
of the final product solution. Therefore, the υNO at 1805 cm–1
and υONO at 1540 cm–1 for the product Os(TmTP)(NO)(ONO)
are seen as negative peaks. The only observable new species in
this reaction is the dicarbonyl Os(TmTP)(CO)2 identified by its
peak at 1974 cm–1 (see below). The peak marked with an aster-
isk is a cyclohexane subtraction artifact and that marked “x” is
suspended insoluble Os(TmPP)(CO). Bottom: Difference infrared
spectrum of Os(TmTP)(CO)2 in cyclohexane after CO addition to
a solution of Os(TmTP)(CO), minus the spectrum observed prior
to CO addition. The new species shows a single carbonyl band
at 1974 cm–1 that is twice as intense as that for the monocarbonyl
(at 1924 cm–1), consistent with a trans-dicarbonyl structure.
Os(OEP)(NO)(ONO). Bottom: Spectral changes upon stopped-
flow mixing of a solution of Os(OEP)(CO) (10 µM in toluene
after mix) with nitric oxide (5 mM after mix). The stacked plot
shows no obvious intermediate, but single wavelength observa-
tion shows evidence for a fast (7 s–1) and a slow (0.7 s–1) pro-
cess in this reaction. However, the two stages in the kinetics
suggest the formation of an intermediate in solution whose spec-
trum is very similar to that of the starting solution.
Scheme 1.
were observed in the stopped-flow IR experiments, we con-
clude that likely intermediates along the productive pathway
— for example, the dinitrosyl Os(P)(NO)2 — are too reac-
tive to build to detectable steady-state concentrations.
This scheme would explain the observation of the two-
stage kinetics observed under a large excess of NO. Initially
all the osmium carbonyl is in the monocarbonyl form, and
the reaction proceeds rapidly. Since the fast reaction was
monitored near the Os(P)(CO)–Os(P)(CO)2 isosbestic point
(Fig. 6), this will show (pseudo) first order kinetics for the
© 2003 NRC Canada

880
Can. J. Chem. Vol. 81, 2003
Fig. 6. Measurement of the equilibrium constant for the reaction
of CO with Os(OEP)(CO) in toluene. The inset shows a
Fe(P)(NO) analogs to promote such disproportionation in
rigorously dried and deaerated solvents (14). A greater par-
allel can be drawn to the Ru(P)(CO) analogues (20); how-
ever, the likely osmium nitrosyl intermediates suggested by
this analogy are apparently too reactive to build to detect-
able concentrations.
Lineweaver–Burke fit for the spectral change induced by addi-
tion of CO aliquots to a dry toluene solution of Os(OEP)(CO)
(~10 µM). The total amount of CO (most of which is contained
in the headspace but in equilibrium with that in solution) avail-
able for the equilibrium reaction is much greater than the total
amount of Os in the solution. Thus, [CO] in solution can be ap-
proximated as being constant, although it is much lower than
[Os]. The measured equilibrium constant is 1.0 ± 0.2 × 106 M–1.
Acknowledgments
GBRA acknowledges National Science Foundation (CHE-
0076640) for funding. Studies at the University of Califor-
nia, Santa Barbara (UCSB) were supported by grants from
the National Science Foundation (CHE-0095144) and the
ACS Petroleum Research Fund for support of this research
to PCF. FAL thanks the University of California CAMP pro-
gram for a research fellowship.
References
1. I.M. Wasser, S. de Vries, P. Moenne-Loccoz, I. Schroder, and
K.D. Karlin. Chem. Rev. 102, 1201 (2002).
2. H. Shimizu, S.-Y. Park, Y. Shiro, and S.-i. Adachi. Acta
Crystallogr. D58, 81 (2002).
3. D.L. Harris. Int. J. Quantum Chem. 88, 183 (2002).
4. S.J. Field, L. Prior, M.D. Roldan, M.R. Cheesman, A.J.
Thomson, S. Spiro, J.N. Butt, N.J. Watmough, and D.J. Rich-
ardson. J. Biol. Chem. 277, 20 146 (2002).
5. E. Pinakoulaki, S. Gemeinhardt, M. Saraste, and C. Varotsis. J.
Biol. Chem. 277, 23 407 (2002).
6. P.R. Gardner, A.M. Gardner, L.A. Martin, and A.L. Salzman.
Proc. Natl. Acad. Sci. U.S.A. 95, 10 378 (1998).
7. A.M. Gardner and P.R. Gardner. J. Biol. Chem. 277, 8166
(2002).
8. R.L. Klimisch and J.G. Larson. In The catalytic chemistry of
nitrogen oxides. Plenum Press, New York. 1975.
9. L. Cheng, D.R. Powell, M.A. Khan, and G.B. Richter-Addo.
Chem. Commun. 2301 (2000).
10. T.S. Kurtikyan. Russ. J. Inorg. Chem. 25, 28 (1999).
11. B.B. Wayland and A.R. Newman. Inorg. Chem. 20, 3093
(1981).
12. P.C. Ford and I.M. Lorkovic. Chem. Rev. 102, 993 (2002).
13. I. Lorkovic and P.C. Ford. J. Am. Chem. Soc. 122, 6516 (2000).
14. M.D. Lim, I.M. Lorkovic, K. Wedeking, A.W. Zanella, C.F.
Works, S.M. Massick, and P.C. Ford. J. Am. Chem. Soc. 124,
9737 (2002); T.S. Kurtikyan, G.G. Martirosyan, I.M. Lorkovic,
and P.C. Ford. J. Am. Chem. Soc. 124, 10 124 (2002).
15. G.R.A. Wyllie and W.R. Scheidt. Chem. Rev. 102, 1067
(2002).
16. K.M. Kadish, V.A. Adamian, E.V. Caemelbecke, Z. Tan, P.
Tagliatesta, P. Bianco, T. Boschi, G.-B. Yi, M.A. Khan, and
G.B. Richter-Addo. Inorg. Chem. 35, 1343 (1996).
17. K.M. Miranda, X. Bu, I. Lorkovic, and P.C. Ford. Inorg.
Chem. 36, 4838 (1997).
18. I.M. Lorkovic, K.M. Miranda, B. Lee, S. Bernhard, J.R.
Schoonover, and P.C. Ford. J. Am. Chem. Soc. 120, 11 674
(1998).
19. I.M. Lorkovic and P.C. Ford. Chem. Commun. 1225 (1999).
20. I.M. Lorkovic and P.C. Ford. Inorg. Chem. 38, 1467 (1999).
21. C.-M. Che, C.-K. Poon, W.-C. Chung, and H.B. Gray. Inorg.
Chem. 24, 1277 (1985).
absorbance changes due to reaction along the productive
pathway to give Os(P)(NO)(ONO) (reaction of Os(P)(CO)
with the released CO is not observed owing to the isosbestic
behavior at that wavelength). Once most of the remaining
osmium carbonyl is sequestered as the dicarbonyl, further
reaction will be slower, owing to the low steady-state con-
centration of the monocarbonyl. Consistent with this inter-
pretation is the qualitative observation that the overall
reaction is dramatically slowed in the presence of excess
added CO. The slow second stage observed in the absence
of added CO also appeared to follow first order kinetics;
however, this may result from a fortuitous combination of
conditions, since the continued production of CO should
make the reaction auto-inhibitory.
Conclusion
We have successfully prepared several nitrosyl nitrito
complexes of the form Os(P)(NO)(ONO) from the reaction
of the precursor carbonyl complexes Os(P)(CO) with nitric
oxide. Single-crystal X-ray crystallographic studies of two
of these compounds confirm the O-binding modes of the
nitrito ligands. To the best of our knowledge, these are the
first X-ray crystal structures of osmium nitrito moieties to be
reported.
Results from stopped-flow kinetics studies of the reactions
suggest a mechanism in which initial adduct formation be-
tween Os(P)(CO) and NO is followed by a rate-determining
replacement of CO by a second NO molecule. Subsequent
reaction with more NO leads to an effective NO
disproportionation to give the final nitrosyl nitrito product
Os(P)(NO)(ONO) and nitrous oxide. The behavior of the os-
mium complexes is in marked contrast to the failure of
22. G.-B. Yi, L. Chen, M.A. Khan, and G.B. Richter-Addo. Inorg.
Chem. 36, 3876 (1997).
23. J. Buchler and P.D. Smith. Chem. Ber. 109, 1465 (1976).
© 2003 NRC Canada

Leal et al.
881
24. C.L. Young. Oxides of nitrogen. In IUPAC solubility data se-
ries. Vol. 8. Pergamon Press, Oxford. 1981.
25. R.W. Cargill. Carbon monoxide. In IUPAC solubility data se-
ries. Vol. 43. Pergamon Press, Oxford. 1990.
26. A.T.C. North, D.C. Philips, and F.S. Mathews. Acta
Crystallogr. A24, 351 (1968).
27. L. Cheng, L. Chen, H.-S. Chung, M.A. Khan, G.B. Richter-
Addo, and V.G. Young, Jr. Organometallics, 17, 3853 (1998).
28. K. Nakamoto. Infrared and Raman spectra of inorganic and co-
ordination compounds. Part B: applications in coordination,
organometallic, and bioinorganic chemistry. John Wiley and
Sons, New York. 1997. pp 50–51.
35. M. Goldner, B. Geniffke, A. Franken, K.S. Murray, and H.
Homborg. Z. Anorg. Allg. Chem. 627, 935 (2001).
36. H. Nasri, M.K. Ellison, S. Chen, B.H. Huynh, and W.R.
Scheidt. J. Am. Chem. Soc. 119, 6274 (1997).
37. M.A.A.F.C.T. Carrondo, P.R. Rudolf, A.C. Skapski, J.R.
Thornback, and G. Wilkinson. Inorg. Chim. Acta, 24, L95
(1977).
38. D.S. Bohle, C.-H. Hung, and B.D. Smith. Inorg. Chem. 37,
5798 (1998).
39. I.L. Eremenko, A.A. Pasynskii, V.T. Kalinnikov, Y.T.
Struchkov, and G.G. Aleksandrov. Inorg. Chim. Acta, 52, 107
(1981).
29. S.S. Eaton and G.R. Eaton. J. Am. Chem. Soc. 97, 3660 (1975).
30. S.S. Eaton and G.R. Eaton. J. Am. Chem. Soc. 99, 6594 (1977).
31. K. Pramanik, P. Ghosh, and A. Chakravorty. J. Chem. Soc.
Dalton Trans. 3553 (1997).
32. B.K.-M. Hui and W.-T. Wong. J. Chem. Soc. Dalton Trans.
447 (1998).
40. J. Kriege-Simondsen and R.D. Feltham. Inorg. Chim. Acta, 71,
185 (1983).
41. C. Hauser, T. Glaser, E. Bill, T. Weyhermuller, and K.
Wieghardt. J. Am. Chem. Soc. 122, 4352 (2000).
42. C.M. Lukehart and J.M. Troup. Inorg. Chim. Acta, 22, 81
(1977).
33. B.K.-M. Hui and W.-T. Wong. J. Chem. Soc. Dalton Trans.
3977 (1998).
34. B.K.-M. Hui and W.-T. Wong. J. Chem. Soc. Dalton Trans.
2177 (1996).
43. D. Ooyama, N. Nagao, H. Nagao, Y. Miura, A. Hasegawa, K.
Ando, F.S. Howell, M. Mukaida, and K. Tanaka. Inorg. Chem.
34, 6024 (1995).
44. R.D. Wilson and R. Bau. J. Organomet. Chem. 191, 123 (1980).
© 2003 NRC Canada