The kinetic instability of σ-bound aryloxide in coordinatively unsaturated or labile complexes of ruthenium

Article abstract of DOI:10.1016/S0020-1693(02)01347-6

Reaction of RuCl₂(PPh₃)₃ (1) or RuHCl(PPh₃)₃ (2) with KOAr (Ar=4-^tBuC₆H₄) in non-alcohol solvents affords π-aryloxide derivatives Ru(η⁵-ArO)(o-C₆H₄PPh₂) (PPh₃) (3a) or RuH(η⁵-ArO)(PPh₃) ₂ (6a), respectively. The phenoxide analogues 3b and 6b are obtained on use of KOPh or TlOPh. Treatment of 1 with 1 equiv. KOAr in the presence of isopropanol liberates the phenol and acetone, affording clean 2 in quantitative yields. In 3:1 methanol-CH₂Cl₂, RuHCl(CO)(PPh₃)₃ (4) is also formed in small amounts. Reaction of 1 with 2 KOAr in 20% MeOH-CH₂Cl₂ affords a mixture of 6a and RuH₂(CO)(PPh₃)₃ (5). In the corresponding reaction of 2 with 1 KOAr, σ-π isomerization of the σ-aryloxide ligand dominates, affording 6a·MeOH as the principal product. Treatment of 6a with ethereal HCl gives [RuH(η⁶-ArOH)(PPh₃)₂]Cl (7a); the corresponding reaction of 6b yields RuCl(η⁵-PhO)(PPh₃)₂ (8b). The crystal structures of 3a, 3b, 4, 5, 7a, and 8b are reported.

Full text of DOI:10.1016/S0020-1693(02)01347-6

Inorganica Chimica Acta 345 (2003) 268Á278
/
www.elsevier.com/locate/ica
The kinetic instability of s-bound aryloxide in coordinatively
unsaturated or labile complexes of ruthenium
J.L. Snelgrove, J.C. Conrad, G.P.A. Yap, D.E. Fogg *
Department of Chemistry, Center for Catalysis Innovation and Research, University of Ottawa, Ottawa, ON, Canada K1N 6N5
Received 11 June 2002; accepted 20 September 2002
Dedicated to Dick Schrock, in honour of his groundbreaking contributions to organometallic chemistry and catalysis
Abstract
Reaction of RuCl2(PPh3)3 (1) or RuHCl(PPh3)3 (2) with KOAr (Arꢀ
4-tBuC6H4) in non-alcohol solvents affords p-aryloxide
/
derivatives Ru(h5-ArO)(o-C6H4PPh2)(PPh3) (3a) or RuH(h5-ArO)(PPh3)2 (6a), respectively. The phenoxide analogues 3b and 6b
are obtained on use of KOPh or TlOPh. Treatment of 1 with 1 equiv. KOAr in the presence of isopropanol liberates the phenol and
acetone, affording clean 2 in quantitative yields. In 3:1 methanolÁ
Reaction of 1 with 2 KOAr in 20% MeOHÁCH2Cl2 affords a mixture of 6a and RuH2(CO)(PPh3)3 (5). In the corresponding
reaction of 2 with 1 KOAr, sÁ MeOH as the principal product.
/
CH2Cl2, RuHCl(CO)(PPh3)3 (4) is also formed in small amounts.
/
/
p isomerization of the s-aryloxide ligand dominates, affording 6a×
/
Treatment of 6a with ethereal HCl gives [RuH(h6-ArOH)(PPh3)2]Cl (7a); the corresponding reaction of 6b yields RuCl(h5-
PhO)(PPh3)2 (8b). The crystal structures of 3a, 3b, 4, 5, 7a, and 8b are reported.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Ruthenium; Aryloxides; p-Phenoxide; Decarbonylation; ortho-Metallation
1. Introduction
to greater tolerance for functional groups as well as less
stringent requirements for reagent purity [5]. The
enhanced catalyst lifetime promised by use of a less
oxophilic metal is limited, however, by the accessibility
of catalyst deactivation pathways that limit lifetime,
thus increasing catalyst loading requirements, as well as
heavy-metal contamination in the organic products.
We recently identified an important catalyst deactiva-
tion pathway for catalysts of the type RuCl2L2(CHR)
Schrock’s fundamental contributions to organometal-
lic chemistry include the development of the first well-
defined catalysts for olefin metathesis, in which rigorous
consideration of mechanistic and design principles led to
systems integrating high activity with remarkable selec-
tivity [1]. Important advances have accrued in, for
example, asymmetric ring-closing and ring-opening
metathesis reactions [2] and stereoselective ring-opening
metathesis polymerization (ROMP) [3] by Group 6
alkylidene complexes. Living ROMP techniques were
also achieved, with their attendant precision of control
over complex block architectures. Issues of selectivity
are only now beginning to be explored in the later-
developed, more robust ruthenium systems, few exam-
ples of which contain a chiral ligand set [4]. Such
systems are attractive for their robustness, which leads
(L2ꢀ
phino)butane), involving extrusion of alkylidene and
formation of face-bridged dimers
RuClL2(m-
/
2 PPh3 or PP; PPꢀ/dcypb, bis(dicyclohexylphos-
Cl)3Ru(CHR)L2 [6,7]. The stability of the Ru2(m-Cl)3
structural motif results in very low metathesis activity,
even toward such reactive monomers as norbornene.
Such pathways are prohibited in the Schrock systems by
judicious incorporation of steric bulk into the ligand set,
reinforced by use of alkoxides or chelating phenoxides in
place of the chloride ligands ubiquitous in the Ru
chemistry. This design strategy addresses the problem
of ensuring access to the metal center while preventing
decomposition through bimolecular coupling reactions.
* Corresponding author. Tel.: ꢁ
562 5170.
E-mail address: dfogg@science.uottawa.ca (D.E. Fogg).
/
1-613-562 5800x6057; fax: ꢁ1-613-
/
0020-1693/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 0 2 ) 0 1 3 4 7 - 6

J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á
/
278
269
With the intention of developing long-lived, highly
active Ru metathesis catalysts, we undertook an ex-
ploration of the potential of phenoxide ligands as bulky,
tunable pseudohalide ligands. Caulton has demon-
strated the susceptibility of the Ru-benzylidene entity
to net deprotonation by alkali phenoxides [8]; we find
that this tendency is not repressed even on use of
electron deficient C6F5Oꢂ salts [9]. In view of this
incompatibility, we chose to focus on exploration of the
phenoxide chemistry of precursors of the type
RuCl2(PPh3)3 (1) and RuHCl(PPh3)3 (2).
under N2, MeOH from Mg(OCH3)2, and C6H6 from
NaÁbenzophenonone ketyl. NMR solvents (Cambridge
Isotopes) were dried over activated molecular sieves
/
˚
(Linde 4A), degassed by consecutive freeze/pump/thaw
cycles, and stored in gas-tight solvent bulbs.
RuCl2(PPh3)3 (1) [19] and RuHCl(PPh3)3 (2) [6,20]
were synthesized according to literature procedures,
potassium or thallium aryloxides (MOPh, MOAr, where
Arꢀ
4-tBuC6H4) by reaction of the appropriate phenol
/
(Aldrich) with KH or thallium ethoxide. NMR spectra
were recorded on a Bruker Avance-300 spectrometer
(121.4 MHz for 31P, 75.4 MHz for 13C, 300 MHz for
1H), IR spectra on a Bomem MB100 IR spectrometer.
Microanalyses were carried out by Guelph Chemical
Laboratories Ltd., Guelph, ON.
A recent resurgence in interest in the nature, stability,
and strength of the late transition metalÃ/oxygen bond
stems from the potential of aryloxides as substrates (in,
for example, nucleophilic phenol functionalization [10]
or aryloxide carbonylation [11,12]) as well as ancillary
ligands. Several important studies have led to reevalua-
2.2. Ru(h5-ArO)(o-C6H4PPh2)(PPh3)×
ArOH)
/
ArOH (3a×
/
tion of the assumption that late transition metalÃ
/oxygen
bond energies are inherently low, as the Pearson ‘hard-
soft’ formalism would suggest [13,14]. Calorimetric data
are consistent with a dependence of formation enthal-
pies on phenol acidity [15], though a suggested 1:1
A suspension of 1 (52 mg, 0.055 mmol) and KOAr
(20.6 mg, 0.109 mmol) in 2 ml C6H5CH3 was stirred for
12 h, until no 31P NMR peaks for 1 were evident. The
solution was filtered through Celite to remove an
insoluble orange precipitate, concentrated, and treated
with cold hexanes to precipitate the bright yellow
correlation between MÃ
/
X bond energies and HÃ/X bond
strengths [13,16] has recently undergone some modifica-
tion to take into account ground-state polarization
effects, in which electron-withdrawing substituents can
promote binding to the metal center [17].
product. Yield after reprecipitation from Et2OÁhexanes:
/
39 mg (66%). X-ray quality crystals were grown by slow
concentration of a CHCl3 solution of the filtrate.
Within late transition metal complexes, the aryloxide
ligand can function as a terminal (or, less commonly,
bridging) O-bound ligand, or as a p-aryl donor.
Examples of both s- and p-aryloxide derivatives are
now known for Groups 6 through 10 [18]. While general
trends associate p-bound aryloxide with PPh3, COD,
Cp*, and C2H4 ligands, and s-bound aryloxide with CO
2
31P{1H} NMR (CDCl3): d 49.3 (d, JPP
ꢀ 39 Hz, 1P),
/
2
1
ꢂ
/
27.5 (d, JPP
(br s, 1H, ArOH), 8.04 (m, 1H, Ph), 7.5Á
Ph), 4.60 (d, JHH
4.6 Hz, 1H, h5-4-tBuC6H4O), 4.53
(d, JHH
ꢀ
/
39 Hz, 1P). H NMR (CD2Cl2): d 9.2
/6.7 (m, 32H,
ꢀ
/
ꢀ
/
4.6 Hz, 1H, h5-4-tBuC6H4O), 4.32 (d, JHH
ꢀ
/
4.6 Hz, 1H, h5-4-tBuC6H4O), 4.22 (d, JHH
ꢀ4.6 Hz, 1H,
/
and PR3 (Rꢀalkyl) this pattern of behaviour is not
/
h5-4-tBuC6H4O), 1.34 (s, 9H, Bu of h5-ArO), 1.01 (s,
9H, tBu of ArOH). IR (Nujol, cmꢂ1) 3100 (m, br, nOH),
1539 (m, nCÄO), 1264 (m, nCÃO). Anal. Calc. for
C56H56O2P2Ru: C, 72.79; H, 6.11. Found: C, 72.69; H,
6.05%.
t
invariable, and in some cases the s- and p-derivatives
are obtained by markedly similar routes. The need for
reliable predictors of synthetic outcomes prompted us to
undertake a systematic investigation of the reactions of
chlororuthenium phosphine complexes with simple
aryloxide ligands. We find a pattern of high reactivity
within coordinatively unsaturated complexes containing
labile ligands, and demonstrate the thermodynamic
preference for p-bound aryloxide within such species.
2.3. Ru(h5-PhO)(o-C6H4PPh2)(PPh3)×
PhOH)
/
PhOH (3b×
/
Addition of KOPh to 1 was carried out as for 3a,
using THF as solvent. Complex 3b was obtained in 54%
yield, as a phenol solvate. Crystals were grown by slow
2. Experimental
diffusion of hexanes into a C6H6 solution. 31P{1H}
2
2.1. General procedures
NMR (CDCl3): d 49.3 (d, JPP
ꢀ
/
38 Hz, 1P), ꢂ27.6 (d,
/
2JPP
PhOH), 7.63Á
2H, m-h5-C6H5O), 4.68 (t, JHH
C6H5O), 4.37 (d, JHH
ꢀ
/
38 Hz, 1P). 1H NMR (CDCl3): d 9.0 (s, 1H,
6.55 (m, 34H, Ph), 5.06 (t, JHH 5.5 Hz,
5.5 Hz, 1H, p-h5-
All reactions were carried out under N2 using
standard Schlenk or drybox techniques. Dry, oxygen-
free C6H5CH3, and THF were obtained using an
Anhydrous Engineering solvent purification system,
˚
and stored over Linde 4A molecular sieves. Methylene
chloride and isopropanol were distilled from CaH2
/
ꢀ
/
ꢀ
/
ꢀ
5.6 Hz, 2H, o-h5-C6H5O). IR
/
(Nujol, cmꢂ1): 3100 (m, br, nOH), 1530 (s, nCÄO), 1275 (s,
nCÃO). Anal. Calc. for C48H40O2P2Ru: C, 71.01; H, 4.97.
Found: C, 71.45; H, 4.97%.

270
J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á278
/
2.4. RuHCl(PPh3)3 (2)
ml CH2Cl2 caused a slow colour change to yellow as the
salt dissolved. The mixture was stirred for 2 h at r.t.,
then filtered through Celite, concentrated, and treated
with cold hexanes to precipitate the bright yellow
A solution of KOAr (230 mg, 1.22 mmol) in 10 ml
isopropanol was added to 1 (1.17 g, 1.22 mmol) in 40 ml
C6H6. The mixture was refluxed for 8 h, then concen-
trated to approximately 2 ml and diluted with 10 ml
hexanes. Clean 2 was filtered off and washed with
product. Reprecipitation from C6H6Áhexanes gave 73
/
mg of 6a (88%). 31P{1H} NMR (C6D6): d 59.3 (br s);
2
partially resolved at 180 K (CD2Cl2): d 60.9 (d, JPP
ꢀ
/
hexanes and MeOH (4ꢃ
NMR data agree with the values reported [20].
/
3 ml each). Yield 1.08 g (96%).
48 Hz), 58.6 (br s). 1H NMR (C6D6): d 8.5Á
6.8 (m, 30H,
/
Ph), 5.12 (d, JHH
(d, JHH
ꢀ
6.5 Hz, 2H, h5-4-tBuC6H4O), 3.85
/
ꢀ
/
6.5 Hz, 2H, h5-4-tBuC6H4O), 1.10 (s, 9H, tBu
2
11.6 (t, JPP
2.5. Reactions of 1 with KOAr in 3:1 MeOHÁ
/CH2Cl2
of h5-ArO), ꢂ
/
ꢀ 35 Hz, 1H, RuH). IR
/
(Nujol, cmꢂ1): 1977 (m, nRuÃH); 1558 (m, nCÄO). Anal.
Calc. for C46H44OP2Ru: C, 71.21, H, 5.72. Found: C,
Addition of a solution of KOAr (37.4 mg, 0.199
mmol) in 6 ml MeOH to a stirred solution of 1 (200 mg,
0.209 mmol) in 2 ml CH2Cl2 caused an immediate
colour change from brown to purple. The solution was
stirred for 8 h at room temperature (r.t.), then reduced
in volume to 1 ml and filtered to obtain purple 2 as a
mixture with 4 (identified by comparison to literature
NMR data; see Section 6). Yield after washing with
71.75; H, 6.29%. (b) 6a×
predissolve the KOAr salt (1:1 MeOHÁ
much more rapid reaction, with near-quantitative con-
version to 6a×MeOH within 10 min at r.t. NMR
/
HOMe. Use of MeOH to
/
CH2Cl2) gave
/
parameters agreed with those for 6a, with the addition
of peaks for MeOH. A small amount of 5 (ratio 1:8 vs.
6a) were also evident. (c) 6a×
/
ArOH. The ArOH solvate
MeOH (2ꢃ/10 ml) 110 mg (59%; contains 10% 4).
was obtained by addition of 1 equiv. of 4-tBuC6H4OH
to the product in (a). NMR parameters agreed with
those for 6a, with the addition of peaks for the phenol.
2.6. Reactions of 1 with KOAr in 20% MeOHÁ
/CH2Cl2
Addition of KOAr (7.9 mg, 0.042 mmol) in 320 ml
MeOH to a solution of 1 (20 mg, 0.021 mmol) in 1.60 ml
CH2Cl2 gave complete conversion to RuH2(CO)(PPh3)3
2.9. RuH(h5-PhO)(PPh3)2 (6b)
(5) and RuH(h5-ArO)(PPh3)2 (6a) (ratio ꢀ
/3:1) within
Solid TlOPh (35 mg, 0.11 mmol) was added to a
stirred purple solution of 2 (100 mg, 0.11 mmol) in 4 ml
THF. After 48 h at r.t., the orange solution was filtered
through Celite, concentrated, and treated with hexanes.
The resulting orange precipitate was filtered off, washed
with hexanes, and dried under vacuum. Yield: 74 mg
(70%). NMR data agree with the values reported [21,22].
15 min. Over 24 h in solution, a new singlet due to
[RuH(h6-ArOH)(PPh3)2]Cl (7a) appears (approximately
24% integrated intensity); control experiments with
isolated 6a (0.016 M) indicate approximately 20%
chlorination in neat CH2Cl2 over this period, proceeding
to completion over 5 days at r.t. Characterization data
for known 5 are given in Section 6; for 6a and 7a, see
below.
2.7. Labelling studies
2.10. [RuH(h6-ArOH)(PPh3)2]Cl×
/
ArOH (7a×/ArOH)
(a) A solution of KOAr (10.0 mg, 0.0532 mmol) in
0.35 ml CH3OD was added to a solution of 1 (25.5 mg,
(a) Addition of ethereal HCl (1 M, 1.0 ml, 1.0 mmol)
ArOH (65 mg, 0.084 mmol) in 2 ml
CH2Cl2 caused a colour change from yellow to orange.
to a solution of 6a×
/
2
0.0266 mmol) in 0.35 ml CH2Cl2. After 24 h at r.t., H
NMR showed incorporation of deuterium into the
The solution was stirred for 1 h, then stripped of solvent.
1
aryloxide as ArOD; H NMR (C6D6) showed hydride
Reprecipitation from CH2Cl2Á
pale orange 7a as an HOAr solvate (74%). (b) A solution
of 6a×ArOH (10 mg) in CDCl3 (1 ml) underwent gradual
/hexanes gave 50 mg of
peaks for 5 and 6a. Some 4 was also evident. A control
experiment carried out in the absence of 1 showed no
deuteration of KOAr. (b) The corresponding experiment
with CD3OD gave ArOD, RuDCl(CO)(PPh3)3,
RuD2(CO)(PPh3)3, and RuD(h5-ArO)(PPh3)2. Some
deuteration of the aromatic rings was observed. No
/
conversion to 7a×
/
ArOH over 5 days at r.t. Pale orange,
X-ray quality crystals deposited from solution. 31P{1H}
NMR (C6D6): d 52.5 (s). 1H NMR (C6D6): d 8.87 (br s,
1H, ArOH), 7.4Á
/
7.0 (m, 33H, Ph and ArOH), 6.93Á
/
1
hydride peaks were evident by H NMR.
6.82 (m, 2H, Ph), 5.63 (d, JHH
4-tBuC6H4O), 3.94 (d, JHH
ꢀ
/
6.4 Hz, 2H, h6-
ꢀ
6.4 Hz, 2H, h6-4-tBu-
/
2.8. RuH(h5-ArO)(PPh3)2 (6a)
C6H4O), 1.25 (s, 9H, tBu of h6-ArO), 1.19 (s, 9H, tBu of
2
10.8 (t, JPP
ArOH), ꢂ
/
ꢀ36 Hz, RuH). Anal. Calc. for
/
(a) Addition of solid KOAr (20.2 mg, 0.107 mmol) to
a stirred purple solution of 2 (100 mg, 0.107 mmol) in 5
C56H59ClO2P2Ru: C, 69.88, H, 6.18. Found: C, 70.27;
H, 6.38%.

J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á
/
278
271
2.11. RuCl(h5-PhO)(PPh3)2 (8b)
3. Results and discussion
(a) Addition of KOPh (29 mg, 0.22 mmol) to a stirred
purple solution of 2 (200 mg, 0.21 mmol) in 4 ml CH2Cl2
caused a colour change from deep purple to bright
orange. After 2 h, the solution was filtered through
Celite, concentrated, and treated with cold hexanes to
precipitate an orange powder. This was filtered off,
The room-temperature reaction of RuCl2(PPh3)3 (1)
with 2 equiv. of KOAr (Arꢀ
4-tBuC6H4) in toluene
ArOH, in approximately 70%
isolated yield (Eq. (1)) [25]. Treatment of 1 with
potassium phenoxide in THF yields the corresponding
/
yields ortho-metallated 3a×
/
phenoxide derivative 3b×
/
PhOH.
washed with hexanes (3ꢃ1 ml) then dried under
/
vacuum. Yield: 132 mg (83%). Hydride 6b was not
1
detected. 31P{1H} NMR (CDCl3): d 29.1. H (CDCl3):
d 7.44 (m, 12H, Ph), 7.26 (m, 6H, Ph), 7.13 (m, 12H,
ð1Þ
Ph), 5.34 (t, JHH
JHH
ꢀ
6.0 Hz, 2H, m-h5-PhO), 4.10 (t,
/
/
ꢀ ꢀ 6.1 Hz,
5.9 Hz, 1H, p-h5-PhO), 3.77 (d, JHH
/
2H, o-h5-PhO). IR (Nujol, cmꢂ1): 1589 (s, nCÃO). Anal.
Calc. for C42H35ClOP2Ru: C, 65.16, H, 4.56. Found: C,
65.20; H, 4.75% (b) 8b×
equiv. to solutions of 8b formed as in (a) in CD2Cl2Á
hexanes resulted in slow crystallization to yield X-ray
quality redÁorange crystals with a molecule of cocrys-
tallized phenol. NMR parameters were essentially
identical to those in (a), with the addition of peaks for
the phenol.
/
PhOH. Addition of phenol (2
31P{1H} NMR analysis is diagnostic [26] for formation
of a four-membered ring by o-metallation of PPh3: two
doublets are observed, with a chemical shift difference of
nearly 80 ppm. p-Coordination of one aryloxide ligand is
indicated by the distinctive upfield shifts of its aromatic
/
/
1
protons in the H NMR spectrum, the resonances for
which are found between 4.6 and 4.2 ppm for 3a, or 5.1Á
/
1
4.3 ppm for 3b. The H NMR values for 3b correspond
2.12. X-ray structural determinations
well with data reported for the h5-PhO ligand [21,22]; see
6b below. Also evident are signals for a ‘normal’,
solvating, hydrogen-bonded phenol in the aromatic
region (3a, 3b); in the case of 3a, two distinct tert-butyl
singlets are also apparent. A sharp singlet at approxi-
mately 9.0 ppm, which exchanges with D2O, is assigned
to the phenolic proton. Infrared analysis reveals a
medium-intensity band at approximately 1530 cmꢂ1
(absent from the spectrum of the potassium phenolates),
suggesting h5-coordination of aryloxide to give an
oxocyclohexadienyl ligand with partial double-bond
Suitable crystals of 3a×
/
(ArOH)2(CHCl3), 3b×
/
PhOH,
7a×ArOH and 8b×PhOH were selected, mounted on thin
/
/
glass fibers using paraffin oil and cooled to the data
collection temperature (203(2) K). Data were collected
on a Bruker AX SMART 1k CCD diffractometer using
graphite-monochromated Mo Ka radiation (wavelength
˚
0.71073 A), with 0.38 v-scans at 0, 90, and 1808 in f.
Unit-cell parameters were determined from 60 data
frames collected at different sections of the Ewald
sphere. Semi-empirical absorption corrections based on
equivalent reflections were applied [23].
Systematic absences in the diffraction data and unit-
cell parameters were uniquely consistent with the
reported space groups P21/n (No. 14), Pbcn (No. 60),
character in the CÃ
/
O bond [21]. A broad, intense
H) stretching
band at 3100 cmꢂ1 is assigned to the n(OÃ
/
vibration of the phenol solvate. The strong propensity of
p-phenoxide species for H-bonding, a manifestation of
the basicity of the phenoxide oxygen, has been repeatedly
P21/n (No. 14), and P21/c (No. 14), for 3a×
/
noted [17,27Á30]. X-ray analysis supports the proposed
/
(HOAr)2(CHCl3), 3b×PhOH, 7a×PhOH and 8b×PhOH,
/
/
/
structures, with the exception of the observation of a
double phenol solvate in the case of 3a. We attribute this
to a further, slow reaction that takes place during
crystallization, as microanalytical and 1H NMR integra-
tion data for the bulk product clearly indicate the
presence of a single solvating phenol. Discussion of
other details of the structures is deferred until presenta-
tion of XRD data for related products (vide infra).
We propose that 3 is generated via initially-formed
bis-phenoxide intermediate A (Scheme 1). The alterna-
tive possibility of consecutive chloride metathesis-b-
elimination sequences via RuHCl(PPh3)3 (2) [31] is less
likely, given the difference in reactivity between 1 and 2
respectively. The structures were solved by direct
methods, completed with difference Fourier syntheses
and refined with full-matrix least-squares procedures
based on F2. The hydride ligand in 7a, H(1), was located
from the difference map and refined with a riding model.
A molecule of chloroform and two molecules of 4-tert-
butylphenol in 3a, a molecule of phenol disordered with
site occupancy distribution of 70/30 in 3b, a molecule of
4-tert-butylphenol in 7a, and a molecule of phenol in 8b
were located in the respective asymmetric units. All non-
hydrogen atoms were refined with anisotropic displace-
ment parameters. All non-hydride hydrogen atoms were
treated as idealized contributions. All scattering factors
are contained in the SHEXTL-6.12 program library [24].
in CH2Cl2ÁMeOH; vide infra. The availability of three
/

272
J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á278
/
phenoxide species would require loss of a strongly
bound alkylphosphine ligand. The higher lability of
PPh3, versus alkylphosphine ligands, has recently been
the subject of renewed attention [44].
3.1. Reaction in the presence of methanol
Use of methanol to predissolve the aryloxide salt,
with the intention of improving stoichiometric control in
reactions of 1 with 1 equiv. aryloxide, instead resulted in
formation of hydride derivatives. Thus, reaction of 1
with 1 equiv. of aryloxide in 3:1 MeOHÁCH2Cl2 affords
/
RuHCl(PPh3)3 (2). The low solubility of 2 in this solvent
system limits further reaction, though approximately
10% RuHCl(CO)(PPh3)3 4 is also formed (eq. (2)). Both
products are readily identified by their characteristic
NMR patterns (see Section 6, including XRD data for
4).
Scheme 1.
coordination sites, arising from the coordinative un-
saturation of A in conjunction with the lability of bound
PPh3, permits sÁp isomerization of one phenoxide
/
ligand to give mixed s/p-aryloxide species B. In turn,
the basicity of the phenoxide oxygen within B permits
abstraction of an ortho-proton from the phenyl group of
a remaining PPh3 ligand, resulting in ortho-metallation
to give 3, elimination of phenol, and formation of
solvated 3. Elimination of alkane provides a driving
force for ortho-metallation of PPh3 in related Ru alkyl
complexes [32]. In the present systems, steric interac-
tions within the piano-stool structure B may also be
important, given the absence of o-metallation (via loss
of H2) in the reactions of 2 with aryloxide discussed
below. Consistent with the intermediacy of a highly
reactive s-phenoxide species, we have recently isolated
stable Ru(s-OAr) species by using electron-deficient
perfluororophenoxide ligands, in which both the basi-
city of the oxygen and the p-donor capabilities of the
ring are sharply attenuated [33].
ð2Þ
We propose that the reaction proceeds via methano-
lysis of the O-bound aryloxide in RuCl(s-OAr)(PPh3)3
C, liberating the phenol (detected by 1H NMR analysis)
and forming a methoxide intermediate (Scheme 2).1
Exchange equilibria of free phenols with bound meth-
oxide have previously been exploited to design efficient
routes to phenoxo complexes [29,38,39], as noted above.
The p-aryloxide binding motif is common in late
transition metal complexes. We suggest that the com-
mon feature is the availability of three coordination sites
in the precursors, whether directly or in the form of
labile donor ligands such as PPh3, dihydrogen, and
dative chloride or oxygen bonds. Examples in Ru
chemistry include the synthesis of Ru(p-ArO)(H)(PR3)2
by reaction of phenols with RuH2(PPh3)4 [21,34],
Scheme 2.
RuH2(H2)2(PCy3)2 [35], and RuH(C6H4PPh2)(P-
nꢁ
1
Ph3)2(Et2O) [32], and of Ru(p-ArO)L3
(nꢀ
/
1, L3ꢀ
/
While we cannot rule out some contribution from reaction of 1
with small amounts of methoxide generated directly via exchange with
KOAr, co-production of decarbonylation and p-aryloxide products (5,
arene [36]; nꢀ0, L3ꢀCp* [29,37Á39]) by reaction of
/
/
/
phenols or phenoxides with [RuCl(p-arene)]2(m-Cl)2,
[Cp*Ru(OMe)]2, or [Cp*Ru(m-Cl)]4. The high reactivity
of the putative s-phenoxide intermediates in this
chemistry contrasts with their stability in complexes
such as RuH(s-OAr)(PMe3)4 or Ru(s-OPh)2(PMe3)4
[17,40,41], and coordinatively unsaturated species
6a; vide infra) in 20% MeOHÁCH2Cl2 requires that the rates of
/
reaction with KOAr and putative KOMe are at least comparable
under these conditions (under which the Ru(s-OAr) intermediate is
available to participate in methanolysis as well as isomerization).
Furthermore, reaction of
2 with KOAr at higher MeOH
concentrations (50%) yields a much lower proportions of 5 versus 6a
(ratio 1:8). This observation, which is difficult to reconcile with
reaction via free methoxide, is consistent with a RuH(OAr)(PPh3)3
intermediate in which isomerization competes with alcoholysis.
RuH(s-OPh)(CO)L2 (Lꢀ
/
PiPr3 [42], PMetBu2 [43]). In
each of the latter complexes, isomerization to a p-

J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á
/
278
273
Scheme 4.
CH3OD confirm that the hydride ligands of 5 and 6a
originate only in the methyl group of methanol, in
keeping with the general mechanism shown in Scheme 2.
Failure to observe chloride complex 8a in the experi-
ments utilizing 1 equiv. aryloxide suggests that the rate
of methanolysis of RuCl(s-OAr)(PPh3)3 is significantly
greater than that of isomerization, despite the thermo-
dynamic acidity of the phenol product. Given that
neither Ru(p-ArO)(s-OMe)(PPh3)2 nor ortho-metal-
lated 3a is observed on treatment of 1 with 2 equiv. of
aryloxide in the presence of MeOH, we propose that this
reaction likewise involves rapid methanolysis of one s-
bound aryloxide ligand, proceeding via Ru(OMe)(s-
OAr)(PPh3)3 (D, Scheme 4), rather than a s/p-(OR)2
species of type B (see Scheme 1). The ratio of decarbo-
nylation product 5 to p-aryloxide 6a is probably
Scheme 3.
The thermodynamically uphill exchange in the present
system is driven by facile b-elimination of formaldehyde.
Ample precedents exist for the decarbonylation of
bound methoxide via b-elimination and subsequent
reaction with formaldehyde [45]. Such a reaction
between formaldehyde and 2 generates coordinatively
saturated 4. We have also employed isopropanol in
place of methanol [6], this reaction terminating with b-
elimination of acetone, which resists decarbonylation
[31]. Quantitative yields of clean 2 are then obtained.
Reaction of 1 with 2 equiv. of aryloxide were carried
out at lower methanol concentrations, to limit compli-
cations arising from the insolubility of the products. In
governed by the relative rates of sÁp isomerization
/
within intermediate RuH(s-OAr)(PPh3)3 (E, Path A),
vs. reaction with formaldehyde. Contribution from Path
B is improbable given the coordinative and electronic
saturation of Ru(p-ArO)(s-OMe)(PPh3)2.
Methanolysis of E itself is disfavoured, as indicated
by experiments involving reaction of 2 with KOAr. As
expected, solely 6a is observed for reactions in neat
CH2Cl2 over 2 h at r.t. (Scheme 5). More significantly,
only traces of 5 are found on use of methanol to
predissolve the salt. Instead, this protocol effects im-
20% MeOHÁCH2Cl2, 1 is completely transformed into
/
RuH2(CO)(PPh3)3 (5) and RuH(h5-ArO)(PPh3)2 (6a)
(ratio 3:1) within 15 min at r.t. (Scheme 3). Complex 5 is
identified on the basis of its reported NMR parameters
[46] (see Section 6 for NMR and XRD data), while the
identity of 6a was confirmed following its independent
synthesis from 2, as discussed below. Over 24 h in this
solvent system, in situ 31P{1H} NMR analysis also
reveals formation of approximately 24% [RuH(h6-
ArOH)(PPh3)2]Cl (7a) (identified on the basis of NMR
and XRD analysis; vide infra). Control experiments
with isolated RuH(h5-ArO)(PPh3)2 in neat CH2Cl2
indicate that 7a is formed via reaction of 6a with this
solvent. No evidence was observed for the neutral
chloride complex RuCl(h5-ArO)(PPh3)2 8a.
mediate, near-quantitative conversion of 2 to to 6a×
MeOH. This implies that (in contrast to the behaviour
of RuCl(s-OAr)(PPh3)3 C), the rate of sÁp isomeriza-
/
/
tion of RuH(s-OAr)(PPh3)3 E significantly exceeds that
of methanolysis, possibly owing to the presence of two
strong trans-labilizing ligands. The reactivity of com-
plexes of type A or D toward isomerization and/or
proton abstraction may also be enhanced by repulsion
between the filled p-symmetry orbitals on the two
oxygen ligands and a filled metal d orbital [43].
Formation of carbonyl complexes 4 and 5 requires the
presence of both aryloxide and methanol in the reaction
medium. No evidence of decarbonylation has been
reported on refluxing 1 in the presence of methanol,
conditions which permit isolation of 1 in near-quantita-
tive yield [19]. At higher temperatures (150 8C),
Ru2Cl4(PPh3)4 is reportedly formed, with methyl for-
mate, hydrogen and dimethoxymethane as byproducts
[47]. The presence of aryloxide thus activates 1 toward
decarbonylation of methanol under unprecedentedly
mild conditions. Labelling studies with CD3OD and
Scheme 5.

274
J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á278
/
ꢂ90 8C, on which a doublet and a broad singlet emerge,
/
the latter of which is assumed to conceal a second
doublet. This behaviour suggests freezing of the aryloxy
ring to give two distinct conformers, one containing
equivalent, the other inequivalent phosphines. The IR
spectrum includes a strong band at 1558 cmꢂ1, con-
sistent with h5-oxocyclohexadienyl character, as noted
above for 3. A band for n(RuÃ
/
H) appears at 1977
cmꢂ1. The identity of the complex is supported by
microanalytical data, as well as by comparison to NMR
data for the reported phenol analogue 6b [21,22,32].
In contrast to 6a, phenoxide derivative 6b was not
accessible via reaction of 2 with KOPh in CH2Cl2, this
affording only chloride analogue 8b. Instead, 6b was
prepared by treating 2 with thallium phenoxide in
benzene. The identity of the complex was confirmed
by comparison to literature NMR values, 6b having
originally been prepared via reaction of phenol with
RuH2(PPh3)4 [21,32]. Isolated 6b proved highly reactive
toward sources of chloride, and treatment with CDCl3
or ethereal HCl (reactions which in the case of 6a
terminate at ion pair 7a) proceeded to neutral, p-phenol
complex 8b with no sign of any intermediates (Scheme
Scheme 6.
1H NMR analysis of 6a is aided by the absence of
solvating phenol. p-Coordination of the aryloxide ligand
is again indicated by the upfield shift of its aromatic
protons (d 5.12, 3.85), which integrate to 4H. Observa-
tion of a hydride triplet at ꢂ
11.6 ppm, and a 31P NMR
/
singlet (d 59.3), indicates the presence of 2 equiv. PPh3
groups, suggesting rapid rotation of the aryloxy ring in
solution. Partial resolution occurs on cooling to
Table 1
Details of crystal data and structure refinement
3a
3b
7a
8b
Empirical formula
Formula weight
Crystal system
C67H71Cl3O3P2Ru
1193.60
C48H40O2P2Ru
811.81
C56H59ClO2P2Ru
962.49
C48H41ClO2P2Ru
848.27
monoclinic
P21/n
orthorhombic
Pbcn
monoclinic
P21/n
monoclinic
P21/c
Space group
Unit cell dimensions
˚
a (A)
13.721(2)
25.215(3)
17.372(2)
90
35.692(3)
12.157(1)
17.925(2)
90
14.0684
23.3999(19)
14.8621(12)
90
12.4189(4)
10.4770(4)
29.9400(11)
90
˚
b (A)
˚
c (A)
a (8)
b (8)
g (8)
97.471(2)
90
90
90
107.7270(10)
90
101.8240(10)
90
3
˚
Volume (A )
Z
5959(1)
4
1.330
7778(1)
8
4660.3(6)
4
1.372
3812.9(2)
4
1.478
Dcalc (g cmꢂ1
Absorption coefficient
)
1.387
0.525
0.497
0.505
0.606
(mmꢂ1
F(000)
)
2488
3344
2008
1744
Crystal size (mm)
u Range for data col-
lection (8)
0.20ꢃ
/
0.20ꢃ
/0.20
0.10ꢃ
/
0.10ꢃ
/
0.02
0.1ꢃ
/
0.1ꢃ
28.78
/0.1
0.2ꢃ
/
0.2ꢃ
28.74
/
0.2
1.70Á24.71
/
2.10Á24.71
/
1.68Á
/
1.39Á
/
Limiting indices
ꢂ
/
165
05l 5
47 562/10 154 [R(int)
/
h 5
20
/
16, 05
/
k 5
/
29,
05
/
h 5
/
41, 05
/
k 5
/
14,
ꢂ
/
185
05l 5
41 617/11 188 [R(int)
/
h 5
20
/
17, 05
/
k 5
/
31,
ꢂ
/
165
05l 5
0.0755] 11 273/7714 [R(int)
/
h 5
40
/
15, 05
/
k 5
/
13,
/
/
05
/
l 521
/
/
/
/
/
Reflections collected/
unique
ꢀ
/
0.1332] 61 494/6631 [R(int)
0.1395]
ꢀ
/
ꢀ/
ꢀ
/
0.0119]
Max/min transmission
Data/restraints/para-
meters
0.928079 and 0.786416
10154/0/685
0.928078 and 0.725343
6631/2/482
0.928076 and 0.771026
11188/0/562
0.928075 and 0.780626
7714/0/487
a
R
0.0470
0.1010
1.010
0.0371
0.0778
1.017
0.0428
0.0879
1.006
0.0336
0.0391
1.048
b
Rw
Goodness-of-fit on F2
a
Rꢀ
Rwꢀ
/
a jjFojꢂ
/
jFcjj/ajFoj.
b
2 1/2
] .
/
[a wd2/a wFo

J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á
/
278
275
Table 2
Selected intramolecular bond distances for 3a, 3b, 7a and 8b
Table 3
Selected bond angles for 3a, 3b, 7a and 8b
Bond
3a
3b
7a
8b
Angle
3a
3b
7a
8b
RuÃ
/
X
Xꢀ
/
C1;
Xꢀ
/
C1;
Xꢀ
/
H1;
Xꢀ
/
Cl;
P1Ã
P1Ã
/
RuÃ
RuÃ
/
P2
X
95.18(4)
96.03(3)
97.90(3)
98.44(10)
XꢀCl;
88.70(9)
XꢀCl;
2.074(4)
2.3258(11)
2.3313(12)
2.083(4)
2.3273(10)
2.3060(10)
2.279(3)
2.233(3)
2.270(3)
2.220(3)
2.270(3)
2.478(4)
1.260(4)
1.396(5)
1.401(5)
1.394(5)
1.412(5)
1.435(5)
1.445(5)
1.402(5)
1.799(3)
1.404(3)
1.383(5)
1.377(5)
1.390(5)
1.394(5)
1.53(0.03)
2.3289(8)
2.3319(8)
2.329(3)
2.263(3)
2.302(3)
2.247(3)
2.255(3)
2.438(3)
1.338(3)
1.417(4)
1.406(4)
1.429(4)
1.405(4)
1.423(4)
1.394(4)
2.428(3)
2.355(3)
2.353(3)
/
/
Xꢀ
66.60(11)
XꢀC1;
/C1;
Xꢀ
66.68(10)
XꢀC1;
/
C1;
Xꢀ
84(1)
XꢀH1;
/
H1;
/
RuÃ
RuÃ
RuÃ
RuÃ
RuÃ
RuÃ
RuÃ
RuÃ
C42Ã
C37Ã
C38Ã
C39Ã
C40Ã
C41Ã
C42Ã
/
P1
/
P2
P2Ã
/
RuÃ
/
X
/
/
/
/
/
C37 2.273(4)
C38 2.253(4)
C39 2.357(4)
C40 2.230(4)
C41 2.260(4)
C42 2.452(3)
2.293(10)
2.210(10)
2.192(10)
2.235(10)
2.319(10)
2.536(10)
1.248(11)
1.392(14)
1.415(14)
1.376(14)
1.425(14)
1.417(14)
1.453(15)
87.13(11)
108.7(3)
97.8(3)
90.32(9)
108.6(2)
98.1(2)
76(1)
93.81(9)
/
RuÃ
C1ÃC6Ã
C6P1ÃRu
C1ÃC2Ã
C2ÃC3Ã
C3ÃC4Ã
C4ÃC5Ã
C5ÃC6Ã
C6ÃC1Ã
C37Ã
C38Ã
C39Ã
C40Ã
C41Ã
C42Ã
/
C1Ã
/
C6
/
/
/
P1
/
/
86.81(13)
119.3(4)
121.7(4)
102.2(4)
118.2(4)
123.3(4)
117.3(4)
86.64(11)
120.1(4)
121.9(4)
120.4(4)
117.0(3)
124.2(3)
116.4(3)
121.6(4)
117.2(4)
121.7(4)
122.1(3)
113.1(3)
122.6(4)
/
/
/
C3
C4
C5
C6
C1
C2
/
/
/
/
O1 1.286(5)
C38 1.418(5)
C39 1.415(5)
C40 1.409(5)
C41 1.405(5)
C42 1.428(5)
C37 1.425(5)
/
/
/
/
/
/
/
/
/
/
/
/
/
C38Ã
C39Ã
C40Ã
C41Ã
C42Ã
C37Ã
/
C39 122.3(4)
C40 114.4(4)
C41 123.6(4)
C42 121.2(4)
C37 114.3(4)
C38 121.8(4)
122.2(3)
116.8(3)
120.4(3)
121.6(3)
117.1(3)
120.6(3)
120.1(11)
119.7(11)
120.2(11)
121.7(11)
114.4(10)
121.5(11)
/
/
/
/
/
/
C1Ã
C6Ã
C1Ã
C2Ã
C3Ã
C4Ã
C5Ã
/
C6
P1
C2
C3
C4
C5
C6
1.408(5)
1.792(4)
1.406(5)
1.390(5)
1.393(5)
1.375(5)
1.389(5)
/
/
/
/
/
/
/
/
/
/
/
which a chloroform solvate is also present), and to a
single phenol molecule in both 3b and 8b. Retention of
the phenol solvate of the starting material is character-
istic in late-metal h5-aryloxide systems [17,29,38], in
which robust hydrogen bonding is thought to alleviate
/
6). Conversion of h5-phenoxide to h6-phenol derivatives
has been accomplished previously by treatment with
mineral acids [38]. The identity of 8b is supported by 1H
NMR and IR data, which reveal the disappearance of
(respectively) the hydride signal and the n(RuÃ
/H)
stretch. These changes are accompanied by a nearly 30
ppm upfield shift of the 31P{1H} NMR singlet, from
approximately 57 to 29 ppm. The change from 6a to 7a
is less spectroscopically dramatic, with a approximately
4 ppm upfield shift in the 31P{1H} NMR singlet being
the only indication of change. While the higher affinity
of the phenoxide species for chloride may be accounted
for by the greater electron-deficiency of the metal center,
versus that in 6a, we considered it prudent to confirm
this unexpected difference in reactivity by undertaking
X-ray crystallographic analysis of both 7a and 8b.
3.2. X-ray structural analyses of 3a×
/(ArOH)2(CHCl3),
3b×PhOH, 7a×ArOH and 8b×PhOH
/
/
/
Crystal and data collection parameters are provided
in Table 1, intramolecular bond distances and angles in
Tables 2 and 3. ORTEP representations are shown in Fig.
1 (3a), Fig. 2 (3b), Fig. 3 (7a) and Fig. 4 (8b). All four
complexes are distorted three-legged piano stools con-
taining p-bound aryloxide and two PPh3 ligands. In 3a
and 3b, one phosphine is h1-bound, one ortho-metal-
lated. ORTEP models for 3a, 3b, and 8b reveal that the
oxygen atom of the p-bonded aryloxide is hydrogen-
bonded, to two 4-tert-butylphenol molecules in 3a (in
Fig. 1. ORTEP diagram of 3a×
/(ArOH)2(CHCl3); hydrogen atoms
omitted.

276
J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á278
/
the development of significant anionic charge on the
oxygen atom [27,29]. The ORTEP model for 7a reveals an
associated chloride counterion and a molecule of co-
crystallized tert-butylphenol, which is not hydrogen-
bonded to the oxygen of the phenol ligand.
The four-membered ortho-metallated ring in 3a and
3b is highly strained: angles of approximately 67, 109,
98, and 878 for P1Ã
/
RuÃ
/
C1, RuÃ
/
C1Ã
/
C6, C1Ã
/
C6Ã
/
P1,
and C6ÃP1ÃRu, respectively, are significantly distorted
/
/
from the idealized geometry about ruthenium (908),
carbon (1208), and phosphorus (1098). Cyclometallated
angles and bond lengths for 3a and 3b correspond
closely with those reported for K[RuH2(o-
C6H4PPh2)(PPh3)2] [48]. While substitution of the aryl
ring does not affect these angles, the phenol derivatives
3b and 8b reveal displacement of the Ru atom from the
Fig. 2. ORTEP diagram of 3b×PhOH; hydrogen atoms omitted.
/
˚
centroid of the ring toward C39, by 0.3 or 0.15 A,
C37 distances
normalized for displacement from the ring).
respectively (Fig. 5; RuÃ
/
C42 and RuÃ
/
Relief of ring strain does not significantly affect the
P2 angles, which are approximately 988 for 7a
RuÃX’ angle (XꢀCl or H, corre-
P1Ã
/
RuÃ
/
and 8b. The ‘P1Ã
/
/
/
sponding to C6 in 3a/b) is predictably increased, by
nearly 228 in the case of 8b. To the steric pressure of the
phenolic tert-butyl substituent and the low steric
demand exerted by hydride we attribute the narrow
splay of the ‘legs’ of the piano stool in 7a (sum of angles
around Ruꢀ2578), vs. 8b (2818). The corresponding
/
value for [Ru(h6-C6H5CH3)(PPh3)2Cl][BF4] (9) [49] is
approximately 78 smaller than for 8b, revealing the
smaller but still significant effect of a methyl substituent
on the ring (its effect being amplified by the planarity of
the aromatic ring in 9). The ClÃ
/
RuÃ
/
P2 bond angle in 9
(85.3(1)8) is also approximately 98 smaller than the
corresponding value of 93.81(9)8 in 8b. This may again
be due to the planarity of the aromatic ring in 9, which
promotes interaction between the methyl group and the
Fig. 3. ORTEP diagram of 7a×
/
ArOH; chloride counter-ion, hydrogen
chloride ligand, compressing the ClÃ
The aryloxy rings in 3a/b and 8b possess partial h5-
O bond dis-
/
RuÃ
/
P2 angle.
atoms (except hydride) and solvate molecule omitted.
oxocyclohexadienyl, character, with CÃ
/
tances ranging from 1.248(11) for 8b to 1.286(5) for
Fig. 4. ORTEP diagram of 8b×
/PhOH; hydrogen atoms and solvate
molecule omitted.
Fig. 5. Displacement of Ru from centroid of arene ring (8b).

J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á
/
278
277
Table 4
Least-squares deviation from plane for 3a, 3b, 7a and 8b (relative to
ortho and meta carbons (C37, C38, C40, C41)
repulsion between the filled p-symmetry orbitals on
oxygen and a filled metal d orbital. Pursuit of the
development of phenoxides as ancillary ligands in late
transition-metal chemistry may thus be advantaged by
use of electron-deficient aryloxides in which both h5-
donation and oxygen basicity are disfavoured, and this
is our current focus.
˚
3a (A)
˚
3b (A)
˚
7a (A)
˚
8b (A)
Atom
C39
C42
0.094
0.168
0.049
0.164
0.047
0.133
0.0142
0.2115
3a. Comparable values were reported for Cp*Ru(h5-
t
5
˚
5. Supporting information available
2,6- Bu2C6H3O) 10 (1.256(4) A) [39] and Cp*Ru(h -
C6H5O)(C6H5OH)2 (1.284(7) A) [38]. All of these values
˚
˚
O bond distance of 1.22 A cited for
the carbonyl group of benzoquinone than to the
NMR spectra showing reaction of 1 with KOAr;
spectroscopic data for literature compounds (2, 4, 5, 6b);
are closer to the CÃ
/
˚
summary of NMR data for 2Á
/
8; XRD data for 3a×
/
phenolic CÃ
/
O bond distance of approximately 1.36 A
O bond distance in 7a is
1.338(3) A. For the h -oxocyclohexadienyl derivatives,
carbon distances are likewise generally
(ArOH)2(CHCl3), 3b×PhOH, 4, 5, 7a×
/
/
ArOH, 8b×PhOH.
/
[50]. In comparison, the CÃ
/
5
˚
the carbonÃ
/
shorter in the ‘pentadienyl’ fragment of the ring than
in the bonds to C42, and puckering of the ring is evident
6. Supplementary material
in the rather long RuÃ
/
C42 bond distance (reaching a
˚
maximum of 2.536(10) A in the case of 8b). These effects
The material is available from the authors on request.
are much less marked for 7a, though this complex also
shows some degree of non-planarity in the p-ArOH ring,
Acknowledgements
as well as variation in the internal CÃ
116.8(3) to 122.2(3)8. In comparison, the range for 3a is
114.3(4)Á123.6(4)8. Elevation of C42 and C39 (the
/C angles from
This work was supported by the Natural Sciences and
Engineering Research Council of Canada, the Canada
Foundation for Innovation, the Ontario Innovation
Trust, and the Ontario Research and Development
Corporation.
/
oxygen-functionalized and para-carbons, respectively)
above the plane of the ortho and meta carbons is noted
in all four complexes (Table 4), though these values are
˚
less dramatic than the elevations of 0.409 and 0.185 A
originally suggested [21] for 6b, or the values found for
˚
10, in which the oxygen-bearing carbon was 0.26 A
above the plane of the ring.
References
[1] (a) R.R. Schrock, Chem. Rev. 102 (2002) 145;
(b) R.R. Schrock, J. Chem. Soc., Dalton Trans. (2001) 2541;
(c) R.R. Schrock, Topics Organomet. Chem. 1 (1998) 1.
[2] (a) A.F. Kiely, J.A. Jernelius, R.R. Schrock, A.H. Hoveyda, J.
Am. Chem. Soc. 124 (2002) 2868;
4. Conclusions
Within this chemistry, a pattern emerges of heigh-
tened reactivity of s-aryloxide species containing direct
or incipient coordinative unsaturation. The strength of
(b) A.H. Hoveyda, R.R. Schrock, Chem. Eur. J. 7 (2001) 945;
(c) D.S. La, E.S. Sattely, J.G. Ford, R.R. Schrock, A.H.
Hoveyda, J. Am. Chem. Soc. 123 (2001) 7767;
(d) J.B. Alexander, D.S. La, D.R. Cefalo, A.H. Hoveyda, R.R.
Schrock, J. Am. Chem. Soc. 120 (1998) 4041.
the RuÃ/O bond in such complexes is undermined by the
accessibility of low-energy reaction pathways involving
bond cleavage and leading to 18-electron products.
These pathways derive from the basicity of the aryloxide
oxygen, or from the availability of p electrons in the
aromatic system. Thus, while 1 is itself stable to
refluxing methanol, reaction with aryloxide in methanol
[3] K.M. Totland, T.J. Boyd, G.G. LaVoie, W.M. Davis, R.R.
Schrock, Macromolecules 29 (1996) 6114.
[4] (a) T. Weskamp, W.C. Schattenmann, M. Spiegler, W.A.
Herrmann, Angew. Chem., Int. Ed. 37 (1998) 2490;
(b) D. Amoroso, D.E. Fogg, Macromolecules 33 (2000) 2815;
(c) D. Amoroso, G.P.A. Yap, D.E. Fogg, Can. J. Chem. 79 (2001)
958;
triggers
decarbonylation
chemistry,
affording
RuH2(CO)(PPh3)3 or RuHCl(CO)(PPh3)3. On reaction
(d) T.J. Seiders, D.W. Ward, R.H. Grubbs, Org. Lett. 3 (2001)
3225.
of either 1 or RuHCl(PPh3)3 with aryloxide in the
[5] Recent reviews: (a) T.M. Trnka, R.H. Grubbs, Acc. Chem. Res.
34 (2001)18;
(b) A. Furstner, Angew. Chem., Int. Ed. 39 (2000) 3012.
¨
absence of methanol, sÁ
p isomerization leads to h5-
/
aryloxide products, typically with hydrogen bonds off-
setting the anionic charge on the oxygen. The high
reactivity of the putative bis(phenoxide) or mixed
phenoxide/methoxide intermediates may arise from
[6] D. Amoroso, J.L. Snelgrove, J. Conrad, S.D. Drouin, G.P.A.
Yap, D.E. Fogg, Adv. Syn. Catal. 344 (2002) 757.
[7] D. Amoroso, G.P.A. Yap, D.E. Fogg, Organometallics 21 (2002)
3335.

278
J.L. Snelgrove et al. / Inorganica Chimica Acta 345 (2003) 268Á278
/
[8] J.N. Coalter, J.C. Bollinger, O. Eisenstein, K.G. Caulton, New J.
[30] C. Di Bugno, M. Pasquali, P. Leoni, P. Sabatino, D. Braga,
Inorg. Chem. 28 (1989) 1390.
[31] A. Aranyos, G. Csjernyik, K.J. Szabo´, J.-E. Backvall, Chem.
Chem. 24 (2000) 925.
[9] D. Amoroso, S.D. Drouin, G.P.A. Yap, D.E. Fogg, to be
submitted (Organometallics).
¨
Commun. (1999) 351.
[32] D.J. Cole-Hamilton, G. Wilkinson, New J. Chem. 1 (1977) 141.
[33] J. Conrad, G.P.A. Yap, D.E. Fogg, to be submitted (Organome-
tallics).
[10] J. Le Bras, H. Amouri, J. Vaissermann, Inorg. Chem. 37 (1998)
5056 (and references therein).
[11] D.W. Dockter, P.E. Fanwick, C.P. Kubiak, J. Am. Chem. Soc.
118 (1996) 4846.
[34] R.O. Rosete, D.J. Cole-Hamilton, G. Wilkinson, J. Chem. Soc.,
Dalton Trans. (1979) 1618.
[12] W.M. Rees, M.R. Churchill, J.C. Fettinger, J.D. Atwood,
Organometallics 4 (1985) 2179.
[35] M.L. Christ, S. Sabo-Etienne, G. Chung, B. Chaudret, Inorg.
Chem. 33 (1994) 5316.
[13] H.E. Bryndza, L.K. Fong, R.A. Paciello, W. Tam, J.E. Bercaw, J.
Am. Chem. Soc. 109 (1987) 1444.
[36] M.A. Bennett, T.W. Matheson, J. Organomet. Chem. 175 (1979)
87.
[14] P.L. Holland, R.A. Andersen, R.G. Bergman, J. Huang, S.P.
Nolan, J. Am. Chem. Soc. 119 (1997) 12800.
[37] B. Chaudret, H. Xiaodong, Y. Huang, J. Chem. Soc., Chem.
Commun. (1989) 1844.
[15] J. Huang, C. Li, S.P. Nolan, J.L. Petersen, Organometallics 17
(1998) 3516.
[38] U. Koelle, M.H. Wang, G. Raabe, Organometallics 10 (1991)
2573.
[16] H.E. Bryndza, W. Tam, Chem. Rev. 88 (1988) 1163.
[17] J.R. Fulton, A.W. Holland, D.J. Fox, R.G. Bergman, Acc. Chem.
Res. 35 (2002) 44.
[39] S.D. Loren, B.K. Campion, R.H. Heyn, T.D. Tilley, B.E.
Bursten, K.W. Luth, J. Am. Chem. Soc. 111 (1989) 4712.
[40] K. Osakada, K. Ohshiro, A. Yamamoto, Organometallics 10
(1991) 404.
[18] J. Ca´mpora, M.L. Reyes, T. Hackl, A. Monge, C. Ruiz,
Organometallics 19 (2000) 2950 (and references therein).
[19] P.S. Hallman, T.A. Stephenson, G. Wilkinson, Inorg. Synth. 12
(1970) 237.
[41] J.F. Hartwig, R.G. Bergman, R.A. Andersen, J. Organomet.
Chem. 394 (1990) 417.
[42] M.A. Tena, O. Nurnberg, H. Werner, Chem. Ber. 126 (1993)
¨
[20] P.R. Hoffman, K.G. Caulton, J. Am. Chem. Soc. 97 (1975) 4221.
[21] D.J. Cole-Hamilton, R.J. Young, G. Wilkinson, J. Chem. Soc.,
Dalton Trans. (1976) 1995.
1597.
[43] J.T. Poulton, K. Folting, W.E. Streib, K.G. Caulton, Inorg.
Chem. 31 (1992) 3190.
[22] T. Yamamoto, S. Miyashita, Y. Naito, S. Komiya, T. Ito, A.
Yamamoto, Organometallics 1 (1982) 808.
[44] M.S. Sanford, J.A. Love, R.H. Grubbs, J. Am. Chem. Soc. 123
(2001) 6543.
[23] R. Blessing, Acta Crystallogr., A 51 (1995) 33.
[24] G.M. Sheldrick, SHEXTL-6.12, Bruker AXS, Madison, WI, 2001.
[25] Also formed are variable amounts of a fine orange precipitate, the
extreme insolubility of which (in benzene, methylene chloride,
dmso) precluded characterization.
[45] (a) B.N. Chaudret, D.J. Cole-Hamilton, R.S. Nohr, G. Wilk-
inson, J. Chem. Soc., Dalton Trans. (1977) 1546;
(b) L.S. Van Der Sluys, G.J. Kubas, K.G. Caulton, Organome-
tallics 10 (1991) 1033.
[46] P.S. Hallman, B.R. McGarvey, G. Wilkinson, J. Chem. Soc., Sect.
A (1968) 3143.
[26] P.E. Garrou, Chem. Rev. 81 (1981) 229.
[27] S.E. Kegley, C.J. Schaverien, J.H. Freudenberger, R.G. Bergman,
S.P. Nolan, C.D. Hoff, J. Am. Chem. Soc. 109 (1987) 6563.
[28] Y.-J. Kim, K. Osakada, A. Takenaka, A. Yamamoto, J. Am.
Chem. Soc. 112 (1990) 1096.
[47] T.A. Smith, R.P. Aplin, P.M. Maitlis, J. Organometal. Chem. 291
(1985) C13.
[48] G.P. Pez, R.A. Grey, J. Corsi, J. Am. Chem. Soc. 103 (1981) 7528.
[49] J.R. Polam, L.C. Porter, Inorg. Chim. Acta 205 (1993) 119.
[50] F.H. Allen, O. Kennard, D.G. Watson, L. Brammer, A.G. Orpen,
R. Taylor, J. Chem. Soc., Perkin Trans. 2 (1987) S1.
[29] K. Bucken, U. Koelle, R. Pasch, B. Ganter, Organometallics 15
¨
(1996) 3095.