Chloro(η5-dihydropentalenyl)bis(triphenylphosphine) ruthenium(II): Synthesis, structural characterization and catalytic activity in the dimerization of phenylacetylene

Article abstract of DOI:10.1016/j.jorganchem.2003.09.052

Reactions between HRuCl(PPh₃)₃ and 1,3- or 1,5-cyclooctadiene yield the 1,2-dihydropentalenyl complex (η⁵-C₈H₉)Ru(PPh₃) ₂Cl through a series of steps including olefin insertion a

Full text of DOI:10.1016/j.jorganchem.2003.09.052

Journal of Organometallic Chemistry 689 (2004) 419–428
www.elsevier.com/locate/jorganchem
Chloro(g5-dihydropentalenyl)bis(triphenylphosphine)ruthenium(II):
synthesis, structural characterization and catalytic activity in
the dimerization of phenylacetylene
a,
b
b
Rein U. Kirss *, Richard D. Ernst , Atta M. Arif
a
Department of Chemistry, Northeastern University, 102 Hurtig Hall, Boston MA, 02115, USA
Department of Chemistry, University of Utah, 315 S. 1400 E., Salt Lake City, UT 84112, USA
b
Received 27 June 2003; accepted 19 September 2003
Abstract
Reactions between HRuCl(PPh3)3 and 1,3- or 1,5-cyclooctadiene yield the 1,2-dihydropentalenyl complex (g5-
C8H9)Ru(PPh3)2Cl through a series of steps including olefin insertion and electrocyclization. The reaction is accompanied by the
5
ꢀ
loss of two equivalents of hydrogen. The product crystallizes in the monoclinic space group P1 (No. 2). (g -C8H9)Ru(PPh3)2Cl
catalyzes the dimerization of phenylacetylene to a ꢀ2:1 mixture of Z:E 1,4-diphenyl-1-buten-3-yne. Comparison of the catalytic
activity of (g5-C8H9)Ru(PPh3)2Cl with (g5-C5H5)Ru(PPh3)2Cl, (g5-C5Me5)Ru(PPh3)H3 and {g5-HB(pz)3}Ru(PPh3)2Cl suggests
that the more electron-rich g5 ligands favor formation of the Z isomer.
Ó 2003 Elsevier B.V. All rights reserved.
Keywords: Electrocyclic reactions; Ruthenium compounds; Dimerization of alkynes
1. Introduction
Flash vacuum pyrolysis (FVP) of (g5-pentamethylcyclopentadienyl)-(g5-cyclooctadienyl)ruthenium, ({g5-C5Me5}
{g5-C8H11}Ru) results in electrocyclic ring closure of the C8H13 ligand, forming (g5-pentamethylcyclopentadienyl)
(g5-1,2-dihydropentalenyl)ruthenium ({g5-C5Me5}{g5-C8H9}Ru). The reaction is accompanied by the loss of one
equivalent of H2 (Eq. (1)) [1]. The potential synthetic utility of ruthenium-mediated electrocyclic reactions in the
construction of [3.n.0] fused ring systems can be better realized if conditions can be identified under which the reaction
occurs in solution at lower temperature.
350˚C
ð1Þ
Ru
Ru
H
2
The relatively mild conditions for the electrocyclic reaction of 1,4-pentadiene [2] (Eq. (2)) suggested that
HRuCl(PPh3)3 might be a versatile reagent for the construction of fused [3.n.0] carbocyclic ring systems from cyclic
* Corresponding author. Tel.: +1-617-373-4513; fax: +1-617-373-8795.
E-mail address: rkirss@neu.edu (R.U. Kirss).
0022-328X/$ - see front matter Ó 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2003.09.052

420
R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
dienes. In this paper we report on the reaction between HRuCl(PPh3)3 and cyclooctadienes yielding (g5-1,2-dihydr-
opentalenyl)Ru(PPh3)2Cl (Eq. (3)). We also report on the activity of (g5-C8H9)Ru(PPh3)2Cl in the catalytic dimer-
ization of phenylacetylene to isomeric 1,4-diphenyl-1-buten-3-ynes.
H
O
PPh3
∆ /
Cl
Ru
PPh3
+ xs
ð2Þ
Ru
PPh3
-PPh3, H2
Cl
PPh3
PPh3
O
H
∆ /
PPh
3
Cl
Ru
+ xs
ð3Þ
-PPh , H
3
2
PPh
Ru
3
PPh
3
Cl
PPh
3
PPh
3
2. Experimental
All compounds described in this work were handled using Schlenk techniques or a M.I. Braun glove-box under
purified argon or nitrogen atmospheres [3]. RuCl3 ꢁ H2O was purchased from Alfa Inorganics, Inc. 1,3-Cyclooctadiene
(Fluka) and 1,5-cyclooctadiene (Aldrich) were used as received. Phenylacetylene was distilled under vacuum prior to
use. Solvents were purified by refluxing over Na/benzophenone (toluene, benzene, hexane, pentane), P2O5 (2-butanone,
dichloromethane) or MgSO4 (ethanol) and distilled prior to use. HRuCl(PPh3)3 was prepared by literature procedures
[4]. Elemental analyses (C, H) were performed by Desert Analytics, Inc.
1
NMR spectra were recorded at 300 MHz for H, 75.4 MHz for 13C{1H} and 121.4 MHz for 31P{1H} on a Varian
XL300 spectrometer. Proton chemical shifts are reported relative to residual protons in the solvent (C6D5H at d 7.15
ppm or CDCl3 at d 7.24 ppm relative to TMS at 0.00 ppm). Carbon chemical shifts are reported relative to solvent
(C6D6 t at d 128.0 ppm). Phosphorus chemical shifts are reported relative to external 85% H3PO4 at 0.0 ppm.
2.1. Synthesis of (g5-C8H9)Ru(PPh3)2Cl
(a) A mixture of 938 mg (1.0 mmol) HRuCl(PPh3)3 and 0.5 mL (4 mmol) 1,5-cyclooctadiene in 40 mL 2-butanone
was refluxed under nitrogen for 48 h. Solvent was evaporated from the red-orange solution under vacuum. The crude
product was dissolved in 5 mL of toluene and passed through a 1 cm thick pad of Celite. The Celite was washed with
2ꢂ2 mL of toluene and the washings combined. Hexane (10 mL) was carefully layered on top of the toluene. An
orange brown solid precipitated upon cooling to ꢃ20 °C for several days. Single crystals were obtained by re-crys-
tallization from 5 mL CH2Cl2 layered with 5 mL of pentane at ꢃ20 °C for several weeks. Red-orange crystals of (g5-
C8H9)Ru(PPh3)2Cl (257 mg, 33% yield) were isolated. The compound crystallizes with 0.5 equivalents of pentane from
1
integration of the H NMR spectrum. The pentane can be removed by heating a finely powdered sample of (g5-
C8H9)Ru(PPh3)2Cl2 at 100 °C for 8 h under vacuum.
m.p. 135–137 °C Analysis: calculated for C44H39ClP2Ru: 68.79% C, 5.38% H; found: 68.97% C, 5.13% H.
1H (C6D6) d 1.98 d (3H), 2.16 pseudo t (2H, J ¼ 9.6, 10.5 Hz), 2.48 s (1H), 2.76 br m (1H), 4.07 s (2H), 6.88 m (18H),
7.62 m (12H), 1.25 m (2H, pentane), 0.87 t (2H, pentane) (CDCl3) d 1.62 br m (4H), 2.09 br m (1H), 2.37 s (1H), 2.49 br
m (1H), 3.94 s (2H), 7.12 t (12H), 7.22 d (6H), 7.38 br m (12H), 1.3 m (3 H, pentane), 0.90 t (3H, pentane) 13C (C6D6): d
139.47, 134.3, 108.92, 77.45, 71.73, 67.80, 29.87, 23.83. 31P{1H} (C6D6): 41.4 s.
(b) A mixture of 1.02 g (1.1 mmol) HRuCl(PPh3)3 and 1.5 mL (12 mmol) 1,5-cyclooctadiene was refluxed in 40 mL
2-butanone for 5 days. Solvent was evaporated under vacuum and the crude solid washed with 50 mL hexane followed
by 25 mL ether. The remaining solid was dissolved in 25 mL CH2Cl2 and the solution passed through a 1 cm thick pad
of alumina. The volume of the red-orange filtrate was reduced to ꢀ5–10 mL under vacuum and the remaining solution

R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
421
carefully layered with an equal volume of pentane. Cooling to ꢃ20 °C precipitated 340 mg (40% yield) of (g5-
C8H9)Ru(PPh3)2Cl. The product was spectroscopically identical to the sample above.
(c) A solution of 435 mg (0.47 mmol) HRuCl(PPh3)3 in 10 mL (82 mmol) 1,5-cyclooctadiene was heated for 48 h at
110–120 °C. Chromatography on alumina (3 ꢂ 8 cm) with 250 mL hexane followed by 300 mL CH2Cl2 eluted a yellow
band. Evaporation of the solvent under vacuum yielded 55 mg (15%) of (g5-C8H9)Ru(PPh3)2Cl.
2.2. Crystal structure determination for (g5-C8H9)Ru(PPh3)2Cl ꢁ 1/2 pentane
A red prismatic shaped crystal 0.23 ꢂ 0.23 ꢂ 0.18 mm in size was mounted on a glass fiber with traces of Paratone oil
ꢁ
and then transferred to a Nonius KappaCCD diffractometer equipped with Mo Ka radiation (k ¼ 0:71073 A). Ten
frames of data were collected at 200(0.1) K with an oscillation range of 1°/frame and an exposure time of 20 s/frame [5],
Indexing and cell refinement were based on all observed reflections ðhmax ¼ 33:77°Þ which had been indexed, integrated
and corrected for Lorentz, polarization and absorption effects using DENZO-SMN and SCALEPAC [6]. Post re-
ꢁ
ꢁ
ꢁ
finement of the unit cell gave a ¼ 11.4380(2) A, b ¼ 12.8395(3) A, c ¼ 14.1171(3) A, a ¼ 69.3640(12)°, b ¼ 87.4640(12)°,
3
ꢁ
c ¼ 84.0325(10)°, and V ¼ 1929.61(7) A . The lack of symmetry in axial photographs and the lack of systematic ab-
ꢀ
sences were consistent with the compound having crystallized in the triclinic space group P1 (No. 2). Crystallographic
data are listed in Table 1.
The structure was solved by a combination of direct and heavy atom methods using SIR 97 [7]. All of the non-
hydrogen atoms were refined with anisotropic displacement coefficients U(H) ¼ 1.2U(C) or 1.5U(Cmethyl), and their
coordinates were allowed to ride on their respective carbons. There is a disordered solvent molecule of pentane in the
2
cell sitting on an inversion center. The weighting scheme employed was w ¼ 1=½r2ðFo2Þ þ ð0:0254PÞ þ 1:3619Pꢄ where
P ¼ ðFo2 þ 2Fc2Þ=3. Scattering factors were taken from the International Tables for Crystallography [8–10].
2.3. Isomerization of 1,5-cyclooctadiene to 1,3-cyclooctadiene catalyzed by HRuCl(PPh3)3
A solution of 34 mg (0.037 mmols, 0.2 mol% Ru) HRuCl(PPh3)3 in 2 mL (16 mmols) 1,5-cyclooctadiene was heated
for 20 h at 100 °C. Complete isomerization to 1,3-cyclooctadiene was verified by NMR spectroscopy.
13C(CDCl3): d 130.78, 125.78, 27.87, 23.00.
Table 1
Crystal data and structure refinement for (g5-C8H9)Ru(PPh3)2Cl
Empirical formula
Formula weight
Temperature
C46:50H45ClP2Ru
802.29
200(0.1) K
ꢁ
0.71073 A
Wavelength
Crystal system
Space group
Unit cell dimensions
Triclinic
ꢀ
P1 (No. 2)
ꢁ
a ¼ 11.4380(2) A a ¼ 69.3640(12)°
ꢁ
b ¼ 12.8395(3) A b ¼ 87.4640(12)°
ꢁ
c ¼ 14.1171(3) A c ¼ 84.0325(10)°
3
ꢁ
1929.61(7) A
Volume
Z
2
Density (calculated)
Absorption coefficient
F ð000Þ
1.381 M g/m3
0.591 mmꢃ1
830
Crystal size
0.23 ꢂ 0.23 ꢂ 0.18 mm
h range for data collection
Index ranges
3.21–33.77°.
0 6 h 6 17, ꢃ19 6 k 6 20, ꢃ21 6 l 6 22
14,291
92.6%
Reflections collected
Completeness to h ¼ 33.77°
Absorption correction
Refinement method
Data/restraints/parameters
Goodness-of-fit on F 2
Final R indices [I > 2rðIÞ]
R indices (all data)
Largest diff. peak and hole
Scalepack
Full-matrix least-squares on F 2
12,173/0/485
1.030
R1 ¼ 0:0365, wR2 ¼ 0:0841
R1 ¼ 0.0478, wR2ðF 2Þ ¼ 0:0902
ꢃ3
ꢁ
0.763 and )0.700 e A

422
R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
2.4. Dimerization of phenylacetylene catalyzed by (g5-C8H9)Ru(PPh3)2Cl and (g5-C5H5)Ru(PPh3)2Cl
Following the procedure for the (g5-C5Me5)Ru(PPh3)H3-catalyzed dimerization of terminal alkynes [11], a 5 mm
NMR tube equipped with a Teflon valve (J. Youngs, Inc.) was charged with the ruthenium(II) catalyst, 0.017 mmol
(g5-C5H5)Ru(PPh3)2Cl or 0.022 mmol (g5-C8H9)Ru(PPh3)2Cl, in an argon atmosphere. About 0.5 mL C6D6 and 30
equivalents of phenyl acetylene were added. The tubes were sealed and heated to 80 °C in a thermostated sand bath for
1
24 h. The progress of the reaction was monitored by H NMR spectroscopy. Products were identified by comparison
of the observed chemical shifts with literature data [11]. The percent conversion and product ratios were determined
from integration of the singlet at 2.74 ppm (PhCCH) and the resonances for the vinyl protons of E- and Z-1,4-di-
phenyl-1-buten-3-yne (d 6.31 d, J ¼ 16.2 Hz and 5.79 d, J ¼ 11.7 Hz, respectively).
3. Results
Reaction of HRuCl(PPh3)3 with either 1,3- or 1,5-cyclooctadiene for 2–5 days leads to a change from the purple-red
color of the starting material to the yellow-orange color of (g5-C8H9)Ru(PPh3)2Cl. The product can be routinely
isolated in 30–40% yield after crystallization. The reaction takes place in 2-butanone, benzene and neat cyclooctadiene
although the highest yields are obtained from reactions in 2-butanone. An excess of cyclooctadiene is essential. We
believe that o-metallation of intermediates in the reaction may be contributing to the consistently low yield of (g5-
C8H9)Ru(PPh3)2Cl [12]. Heating HRuCl(PPh3)3 in neat 1,5-cyclooctadiene is accompanied by complete, catalytic
isomerization of the excess 1,5-C8H12 to the conjugated 1,3-C8H12 isomer. No attempts were made to detect the
formation of 1,4-cyclooctadiene nor to optimize the conditions for isomerization of 1,5-cyclooctadiene. The complex
(g5-1,2-dihydropentalenyl)ruthenium bis(triphenylphosphine)chloride was characterized by elemental analysis and by
NMR spectroscopy.
The 1H NMR spectrum of (g5-C8H9)Ru(PPh3)2Cl in benzene or chloroform is unusual in that the chemical shift of
the unique proton on the g5 ring is shifted upfield to 2.48 ppm. The other chemically equivalent protons on the g5-ring
are observed at 4.07 ppm, well within the typical range for cyclopentadienyl ruthenium compounds [13]. The 1H NMR
spectrum also contains resonances assigned to 0.5 equivalent of pentane that co-crystallizes with (g5-
C8H9)Ru(PPh3)2Cl. Some pentane remains even after heating under vacuum.
Three resonances at 67.80, 71.73 and 77.45 ppm in the 13C{1H} spectrum are assigned to the carbon atoms of the
unsaturated g5 ring. Two resonances at 29.87 and 23.83 ppm are assigned to the carbon atoms on the saturated
portion of the g5-C8H9 ring. Resonances assigned to the aryl carbons of the two PPh3 ligands are observed at 108.92,
1
134.30 and 139.47 ppm. In the H coupled 13C spectrum, the 71.73 and 77.45 ppm resonances become doublets with
JCH ¼ 175.6 and 178.5 Hz, respectively. The resonance at 67.80 ppm remains a singlet consistent with its assignment to
the bridgehead carbons in the g5-C8H9 ring. A singlet is observed in the 31P NMR spectrum at 41.4 ppm, typical of
ruthenium(II) triphenylphosphine complexes [13].
The identity of the product was confirmed by single crystal X-ray diffraction. The solid state structure of (g5-
C8H9)Ru(PPh3)2Cl is shown in Fig. 1. There is a disordered solvent molecule of pentane in the cell sitting on an in-
version center. The bond distances and bond angles (Table 2) are similar to those reported for (g5-C5H5)Ru(PPh3)2Cl
[14], (g5-C5Me5)Ru(PPh3)2Cl [15] and (g5-indenyl)Ru(PPh3)2Cl [16]. Comparisons with the latter compound are
perhaps the most appropriate from the perspective of the fused hydrocarbon ligand. The average Ru–C bond distance in
5
5
ꢁ
ꢁ
(g -C8H9)Ru(PPh3)2Cl, 2.221 A, is about 2% shorter than in (g -C9H7)Ru(PPh3)2Cl, 2.272 A. A key difference,
however, is the absence of a ring slippage toward an g3-coordination of the g5-C8H9 ligand in (g5-C8H9)Ru(PPh3)2Cl.
The slip parameter for (g5-C8H9)Ru(PPh3)2Cl is only 0.0421 compared to 0.2053 in (g5-C9H7)Ru(PPh3)2Cl [16]. The
Ru–Cl and Ru–P distances are nearly identical for (g5-C8H9)Ru(PPh3)2Cl and (g5-C9H7)Ru(PPh3)2Cl, and the P–Ru–
P angle in (g5-C8H9)Ru(PPh3)2Cl is only about 2° smaller than in (g5-C9H7)Ru(PPh3)2Cl. Both compounds adopt a
conformation in which the Ru–Cl bond bisects the bridgehead C–C bond. Finally, the saturated ring in (g5-
C8H9)Ru(PPh3)2Cl has an envelope configuration in which C7 is bent down from the plane of the other four atoms with
an envelope angle (angle between the C(1,5,6,8) and C(6,7,8) planes) of 26.45°.
The (g5-C8H9)RuCl(PPh3)2 complex (3 mol%) catalyzes the dimerization of phenylacetylene to a mixture of Z- and
E-1,4-diphenyl-1-buten-3-yne in benzene solution (Table 3). The reaction is slow, with only 44% conversion of phe-
nylacetylene after 24 h at 80 °C. The major products in the dimerization reaction are E- and Z-1,3-diphenyl-1-buten-3-
1
yne which are readily distinguished by comparison of their H NMR spectra with literature data. The Z-isomer has a
pair of doublets at 5.79 and 6.41 ppm (Jcis ¼ 11.7 Hz) while the E-isomer has a much larger coupling constant
(Jtrans ¼ 16.2 Hz) for its two doublets at 6.30 and 7.04 ppm [11]. The ratio of the Z:E isomers is found to be 2:1 from
integration of the spectra recorded after 24 h of reaction. While these isomers account for 90% of the observed

R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
423
Fig. 1. Solid state structure of (g5-C8H9)Ru(PPh3)2Cl.
Table 2
5
ꢁ
Selected bond distances (A) and bond angles (°) for (g -C8H9)Ru(PPh3)2Cl
ꢁ
Bond distances (A)
Ru–C(3)
Ru–C(2)
Ru–C(4)
Ru–C(1)
Ru–C(5)
Ru–P(2)
Ru–P(1)
Ru–Cl
2.1719(16)
2.1903(17)
2.2339(16)
2.2505(17)
2.2579(16)
2.3044(4)
2.3220(5)
2.4462(4)
C(1)–C(2)
C(1)–C(5)
C(1)–C(8)
C(2)–C(3)
C(3)–C(4)
C(4)–C(5)
C(5)–C(6)
C(6)–C(7)
C(7)–C(8)
1.420(3)
1.428(2)
1.500(3)
1.438(3)
1.448(2)
1.406(3)
1.499(3)
1.545(3)
1.545(3)
Bond angles (°)
P(2)–Ru–Cl
P(1)–Ru–Cl
90.074(15)
90.792(15)
99.963(15)
107.48(16)
140.95(18)
110.77(16)
108.03(16)
107.74(16)
C(5)–C(4)–C(3)
C(4)–C(5)–C(1)
C(4)–C(5)–C(6)
C(1)–C(5)–C(6)
C(5)–C(6)–C(7)
C(6)–C(7)–C(8)
C(1)–C(8)–C(7)
106.99(16)
109.76(15)
139.60(17)
110.29(16)
102.95(16)
106.81(17)
102.20(17)
P(2)–Ru–P(1)
C(2)–C(1)–C(5)
C(2)–C(1)–C(8)
C(5)–C(1)–C(8)
C(1)–C(2)–C(3)
C(2)–C(3)–C(4)
products, other resonances are observed in the region characteristic of vinyl protons. An additional pair of signals is
observed at 5.68 (d, J ¼ 0.9 Hz) and 5.75 (d, J ¼ 0.9 Hz) with another pair of doublets at 5.28 (J ¼ 1.8 Hz) and 5.38
(J ¼ 1.8 Hz) ppm. These resonances account for 9% and 1% of the products, respectively. The identity of the two minor
products has not been firmly established. The chemical shifts and coupling constants of the more prominent of the two
minor constituents are similar to the spectroscopic data reported for the vinyl protons of the head-to-tail isomer, 2,4-
diphenyl-1-buten-3-yne in CDCl3 (5.79 d and 6.00 d, J ¼ 0.9 Hz) [17]. The 2,4-diphenyl-1-buten-3-yne isomer, however,
decomposes upon attempts at purification [16]. The compound responsible for the resonances at 5.28 and 5.38 ppm

424
R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
Table 3
Product ratios from the catalytic dimerization of phenylacetylene
Catalysta
% Conversion
Z
E
{g5-HB(pz)3}Ru(PPh3)2Cl
(g5-C5H5)Ru(PPh3)2Cl
(g5-C8H9)Ru(PPh3)2Cl
(g5-C5Me5)Ru(PPh3)H3
80
23
44
85
6
91
43
30
33
45
60
67
a Catalysts (2–5 mol%) in 0.3–0.5 M PhCCH in benzene at 80° for 24 h.
remains unknown although both styrene and 1,4-diphenylbutatriene [18] can be excluded based on comparison with
literature data or authentic samples.
Spectra recorded at shorter reaction times (and lower conversions) indicate that the Z:E ratio is nearly time-in-
dependent. However, the relative amounts of the suspected head-to-tail dimer decrease as reaction time increases,
consistent with the reported thermal instability of 2,4-diphenyl-1-buten-3-yne.
For comparison, (g5-C5H5)Ru(PPh3)2Cl was also investigated as a catalyst for the dimerization of phenylacetylene.
After 24 h of reaction under identical conditions, only 23% of the phenylacetylene was consumed yielding the same
mixture of products but with a different ratio of Z:E isomers (1:1), 6% each of 2,4-diphenyl-1-buten-3-yne and the same
unidentified compound responsible for the doublets at 5.28 and 5.38 ppm.
4. Discussion
By analogy to the reaction between HRu(PPh3)3Cl and 1,4-pentadiene [2] (Eq. (2)) the formation of (g5-
C8H9)Ru(PPh3)2Cl most likely follows a sequence of steps shown in Scheme 1. The reaction probably starts with the
loss of one triphenylphosphine ligand followed by coordination and insertion of cyclooctadiene into the Ru–H bond
forming an g3-cyclooctenyl species. The facile, quantitative isomerization of 1,5-cyclooctadiene to the conjugated 1,3-
isomer by only 0.2 mol% HRu(PPh3)3Cl suggests that the insertion process is both fast and reversible. The olefin
insertion product, (g3-C8H13)Ru(PPh3)2Cl, then loses hydrogen forming the g5-cyclooctadienyl ligand in (g5-
C8H11)Ru(PPh3)2Cl. While it is known that transition metal centers can quite easily bring about activation of C–H
bonds in the saturated segments of cyclo-diene or cyclo-dienyl ligands [19], the fate of the hydrogen that is lost is still in
question [20]. In the reaction between HRu(PPh3)3Cl and 1,4-pentadiene (or cycloheptadiene), a transfer of hydrogen
to a second equivalent of diene was observed, producing pentene (or cycloheptene) [2,19]. A similar process appears
H
PPh3
Ru
Ru
Cl
Cl
PPh3
Cl
Ru
PPh3
PPh3
PPh3
-PPh3
PPh3
PPh3
Ru
Cl
PPh3
Ru
Ru
PPh3
Cl
PPh3
PPh3
Cl
PPh3
PPh3
H2
Scheme 1. Proposed mechanism for formation of (g5-C8H9)Ru(PP3)2Cl cyclooctadiene and HRuCl(PPh3)3.

R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
425
likely here, as no product was isolated when a stoichiometric amount of cyclooctadiene underwent reaction with
HRu(PPh3)3Cl.
In any case, (g5-C8H11)Ru(PPh3)2Cl subsequently underwent a thermally allowed, disrotatory electrocyclization [1]
to yield (g5-C8H9)Ru(PPh3)2Cl, accompanied by loss of a second equivalent of H2. Others [20] have suggested that this
loss of hydrogen does not involve hydrogenation of excess diene but rather direct loss of H2 gas. In this respect, the
synthesis of (g5-C8H9)Ru(PPh3)2Cl differs from the gas phase chemistry of (g5-C5Me5)(g5-C8H11)Ru for which the
absence of excess diene requires that 2 equivalents of hydrogen are lost as H2.
As noted above, the analogous (g5-C5H7)Ru(PPh3)2Cl [21] compound cyclizes to (g5-C5H5)Ru(PPh3)2Cl [2] while
(g5-C7H9)Ru(PPh3)2Cl [21] does not, possibly as formation of the [3.2.0] bicyclic product is thermodynamically uphill
(Eq. (4)) [18]. It has been reported that at near-ambient temperatures (283 K) reaction between HRuCl(PPh3)3 and 1,5-
cyclooctadiene does not result in olefin insertion but stops at a diene complex, (g4-C8H12)Ru(H)(PPh3)2Cl [18]. The
much longer reaction times needed to produce (g5-C8H9)Ru(PPh3)2Cl (>48 h) than required for the conversion of
HRuCl(PPh3)3 and 1,4-pentadiene to (g5-C5H5)Ru(PPh3)2Cl (ꢀ16 h) [21] reveal a dramatically reduced reactivity for
cyclooctadienes.
Ru
Cl
PPh3
-PPh3
PPh3
ð4Þ
CH2Cl2
r.t.
HRuCl(PPh3)3 + xs
-H2
-PPh3
Ru
Cl
PPh3
PPh3
Curiously, spectroscopic evidence from the reaction between HRuCl(PPh3)3 and 1,3,5,7-cyclooctatetraene suggests
the formation of (g5-bicyclo-[5.0.1]octadienyl)Ru(PPh3)2Cl [22] rather than (g5-C8H7)Ru(PPh3)2Cl, which would
contain the g5-pentalenyl ligand (Eq. (5)) [23]. It may be that the strain induced by ring fusion at sp2 rather than sp3
carbon centers (i.e., bottom product of Eq. (5) vs. the top product of (5)) becomes prohibitively unfavorable.
-PPh
3
Ru
Cl
PPh
3
PPh
3
ð5Þ
CH Cl
2
2
HRuCl(PPh ) + xs
3 3
r.t.
-H
-PPh
2
Ru
3
Cl
PPh
PPh
3
3

426
R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
The dimerization of terminal alkynes is catalyzed by a variety of ruthenium(II) compounds including (g5-
C5Me5)Ru(PPh3)H3 (a mixed hydrogen/hydride complex) [11], and {HB(pz)3}Ru(PPh3)2Cl [24,25]. The initial steps of
the reaction mechanism vary depending on the complex; however, in each case, the key C–C bond forming step in-
volves migration of an acetylide to the a-carbon of the ruthenium-coordinated vinylidene ligand (Scheme 2). On
comparison of the structure of the proposed (acetylide)(vinylidene) intermediates derived from (g5-C8H7)
Ru(PPh3)2Cl, (g5-C5H5)Ru(PPh3)2Cl, (g5-C5Me5)Ru(PPh3)H3, and {HB(pz)3}Ru(PPh3)2Cl, one finds that the sole
distinguishing feature between them is the identity of the g5 ligand. In each case the coordination sphere of ruthenium
contains triphenylphosphine, a phenylacetylide and a phenyl-substituted vinylidene ligand. The ‘‘common’’ interme-
diate suggests that the product distribution (Z:E ratio) from dimerization of phenylacetylene in the presence of these
four Ru(II) compounds may provide clues to the electronic and steric effects of the g5 ligand on the selectivity of the
alkyne dimerization.
The g5-indenyl complexes (g5-C9H7)Ru(PPh3)2H and (g5-C9H7)Ru(PPh3)2(CCPh) are also active in the catalytic
dimerization of phenylacetylene [26]. Unlike the other four compounds cited above, however, ring slip of the indenyl
group from g5 to g3 coordination plays an important role in the reaction mechanism. This difference in mechanism
precludes further meaningful comparisons between (g5-indenyl)ruthenium(II) complexes and (g5-C8H9)Ru(PPh3)2Cl,
(g5-C5H5)Ru(PPh3)2Cl, (g5-C5Me5)Ru(PPh3)H3, or {HB(pz)3}Ru(PPh3)2Cl.
The results of our studies and literature data for the dimerization of phenylacetylene are listed in Table 3. While
there are differences in solvent and reaction temperature it is clear that Z- and E-1,4-diphenyl-1-buten-3-yne are the
major products in each case. The ratio of the Z to E isomer, however, ranges from ꢀ1:15 (for HB(pz)3) to ꢀ2:1 in favor
of the Z isomer (for g5-C5Me5 and g5-C8H9). The Z:E ratio from reactions catalyzed by (g5-C5H5)Ru(PPh3)2Cl lies in
between, providing a nearly equal distribution between the Z and E isomers. A systematic study of the three ligands,
C5Me5, C5H5, and HB(pz)3, found that the C5Me5 ligand is the most electron-rich of the three g5 ligands while the
Scheme 2. Access to a common intermediate in the dimerization of phenylacetylene from (g5-C8H9)Ru(PPh3)2Cl, (g5-C5H5)Ru(PPh3)2Cl, (g5-
C5Me5)Ru(PPh3)H3, And (g5-HB{pz}3)Ru(PPh3)2Cl.

R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
R=H, Me
427
R5
R5
Ru
C
Ru
Ph3P
C
Ph3P
C
C
C
C
C
Ph
Ph
C
Ph
H
H
Ph
Ph
Ph
C
C
C
H
Ph
C
H
C
C
C
C
H
H
Ph
E
Z
Scheme 3. Proposed geometric isomers contributing to Z- and E-1, 4-diphenyl-1-buten-3-yne in g5-LRu(PPh3)(CBCPh){@C@C(H)Ph} (L ¼ C5H5,
C5Me5, C8H9, HB{pz}3).
tris(pyrazolyl)borate ligand is actually the least electron rich member of the series [27]. Hence, one sees a possible
correlation between the isomer ratio and the electronic properties of the ligands.
While the Z:E ratio remains fairly constant in reactions of phenylacetylene catalyzed by (g5-C8H9)Ru(PPh3)2Cl,
when (g5-C5H5)Ru(PPh3)2Cl is used as a catalyst, the relative amount of Z-1,4-diphenyl-1,3-butenyne increases with
time. A similar observation was reported for the dimerization of phenylacetylene catalyzed by (g5-C9H7)Ru(PPh3)2Cl
and attributed to faster oligomerization of the E-1,4-diphenyl-1,3-butenyne product compared to the Z isomer [25]. A
Z:E ratio that favors the E isomer at short reaction times using (g5-C5H5)Ru(PPh3)2Cl is consistent with the proposed
electronic effect of the g5-ligands. Data at shorter reaction times for (g5-C5Me5)Ru(PPh3)H3, or {g5-
HB(pz)3}Ru(PPh3)2Cl are unfortunately not available for further comparison of the Z:E ratio as a function of reaction
time.
The steric effect of these g5-coordinated ligands, however, falls in the order HB(pz)3 ꢀC5Me5 > C5H5 [28]. With its
fused saturated five-membered ring, the g5-C8H9 ligand in (g5-C8H9)Ru(PPh3)2Cl is expected to be more electron rich
than the C5H5 ligand but less than the C5Me5 group. The fused bicyclic ligand in (g5-C8H9)Ru(PPh3)2Cl may also
exert a significant steric effect along one edge of the g5-group; however, rotation about the Ru–C8H9 axis makes it
unlikely that the dihydropentalenyl ligand would have a greater steric effect than either the C5Me5 or HB(pz)3 ligand.
For a series of (g5-C5Me5)RuLH3 compounds [11] it has been shown that as the size of L increases (e.g., PCy3 vs
PPh3), the Z isomer is increasingly favored over the E isomer despite the greater electron-donating ability of PCy3 vs.
PPh3. The selectivity is proposed to arise from the two orientations for the phenyl-substituted vinylidene group shown
in Scheme 3. Were the steric effects of the g5-ligand to play a similar role, one would expect the Z isomer to also be
favored in reactions catalyzed by {HB(pz)3}Ru(PPh3)2Cl. The observed trend toward increasing Z isomer in Table 3
follows the order C5Me5 ꢀ C8H9 > C5H5 > HB(pz)3, an order which clearly does not reflect the presumed steric effect
of the four g5-ligands. Thus the data in Table 3 suggest that the dimerization of phenylacetylene to the Z isomer is
favored as the electron density on the g5 ring increases.
5. Conclusions
Reaction of HRuCl(PPh3)3 with cyclooctadiene allows the synthesis of a ruthenium(II) complex, (g5-
C8H9)Ru(PPh3)2Cl, containing a fused [3.3.0] ring at moderate temperature through a series of steps including a
disrotatory electrocyclic reaction. Thus we have achieved our goal of moderating the conditions needed to effect
electrocyclic reactions at a ruthenium center. Relatively few examples of electrocyclic reactions of transition metal

428
R.U. Kirss et al. / Journal of Organometallic Chemistry 689 (2004) 419–428
coordinated ligands have been reported [1,28]. While studies to determine the scope of the reaction are in progress, we
find that (g5-C8H9)Ru(PPh3)2Cl catalyzes the dimerization of phenylacetylene to a mixture of Z- and E-1,4-diphenyl-
1-buten-3-yne. Comparison of the Z:E ratio of products using (g5-C8H9)Ru(PPh3)2Cl, (g5-C5H5)Ru(PPh3)2Cl, (g5-
C5Me5)Ru(PPh3)2Cl and {HB(pz)3}Ru(PPh3)2Cl demonstrates the presence of an electronic effect of the g5-ligand on
the selectivity of the reaction.
5.1. Supplementary material
Crystallographic data for the structural analysis of (g5-C8H9)Ru(PPh3)2Cl has been deposited with Cambridge
Crystallographic Data Centre, CCDC, No. 211774. Copies of this information may be obtained free of charge from
The Director, CCDC, 12 Union Road, Cambridge CB2 1EZ [fax: +44(1223)336-033] or http://www.ccdc.cam.ac.uk.
Acknowledgements
R.U.K. acknowledges the Northeastern University Research and Scholarship Development Fund for support of
this work. R.D.E. is grateful to the University of Utah and the National Science Foundation for partial support of this
work.
References
[1] R.U. Kirss, A. Quazi, C.H. Lake, M.R. Churchill, Organometallics 12 (1993) 4145.
[2] B.E. Mann, P. Manning, C.M. Spencer, J. Organometal. Chem. 312 (1986) C64.
[3] D.F. Shriver, Manipulation of air Sensitive Compounds, McGraw Hill, New York, 1969.
[4] R.A. Schunn, E.R. Wonchoba, Inorg. Synth. 13 (1972) 131.
[5] COLLECT Data Collection Software Nonius B.V. 1998.
[6] Z. Otwinowski, W. Minor, Methods Enzymol. 276 (1997) 307.
[7] SIR 97(Rlease 1.02) A. Altomare, M.C. Burla, M. Camalli, G. Carcarano, C. Giacovazzo, A. Guagliardi, A.G.G. Moliteni, G. Polidori, R.
Spagna.
P
P
P
P
P
2
2
1=2
2
1=2
[8] RðF Þ ¼ R1 ¼ ðjFoj ꢃ jFcjÞ= jFoj, wRðF 2Þ ¼ wR2 ¼ ½ ðwFo2 ꢃ Fc2Þ = ðF 2Þ ꢄ , and S ¼ goodness-of-fit on F 2 ¼ ½ ðwFo2 ꢃ Fc2Þ =ðn ꢃ pÞꢄ
where n is the number of reflections and p is the number of parameters refined.
,
[9] E.N. Maslen, A.G. Fox, M.A. OÕKeefe, in: A.J.C. Wilson, (Ed.), International Tables for Crystallography: Mathematical, Physical and
Chemical tables, vol. C, Dordrecht, The Netherlands, 1992, p. 476 (Chapter 6).
[10] D.C. Creagh, W.J. McAuley, in: A.J.C. Wilson, (Ed.), International Tables for Crystallography: Mathematical, Physical and Chemical tables,
vol. C, Dordrecht, The Netherlands 1992, p. 206 (Chapter 4).
[11] C.S. Yi, N. Liu, Organometallics 15 (1996) 3968.
[12] B.R. James, L.D. Markham, D.K.W. Wong, J. Chem. Soc. Chem Commun. (1974) 439.
[13] (a) M.I. Bruce, N.J. Windsor, Aust. J. Chem. 30 (1977) 1601;
(b) R.J. Haines, A.L. Du Preez, J. Organometal. Chem. 84 (1975) 357.
[14] M.I. Bruce, F.S. Wong, B.W. Skelton, A.H. White, J. Chem. Soc. Dalton Trans. (1981) 1398.
[15] I.A. Guzei, M.A. Paz-Sandoval, R. Torres-Lubian, P. Juarez-Saavedra, Acta Crystallogr. Sect. C C55 (1999) 1090.
[16] M. Kamigaito, Y. Watanabe, T. Ando, M. Sawamoto, J. Am. Chem. Soc. 124 (2002) 9994.
[17] J. Ohshita, K. Furumori, A. Matsuguchi, M. Ishikawa, J. Org. Chem 55 (1990) 3277.
[18] M.A. Esteruelas, J. Herrero, A.M. Lopez, M. Olivan, Organometallics 20 (2001) 3203.
[19] R. Basta, R.D. Ernst, A.M. Arif, Organometallics 22 (2003) 812.
[20] (a) M. Grassi, B.E. Mann, P. Manning, C.M.J. Spencer, Organometal. Chem. 307 (1986) C55;
(b) G. Alibrandi, B.E. Mann, J. Chem. Soc. Dalton Trans. (1994) 951.
[21] g5-C5H7 and g5-C7H9 refer to the pentadienyl and cycloheptadienyl ligands, respectively.
[22] G. Alibrandi, B.E. Mann, J. Chem. Soc. Dalton Trans. (1992) 1439.
[23] S. Barlow, D. OÕHare, Chem. Rev. 97 (1997) 637.
[24] g5-HB(pz)3 refers to the tris(pyrazolyl)borate ligand.
[25] C. Slugovc, K. Mereiter, E. Zobetz, R. Schmid, K. Kirchner, Organometallics 15 (1996) 5275.
[26] M. Bassetti, S. Marini, F. Tortorella, V. Cadierno, J. Diez, M.P. Gamasa, J. Gimeno, J. Organometal. Chem. 593–594 (2000) 292.
[27] D.M. Tellers, S.J. Skoog, R.G. Bergman, T.B. Gunnoe, W.D. Harman, Organometallics 19 (2000) 2428.
[28] (a) N. Kitajima, B.W. Tolman, Prog. Inorg. Chem. 43 (1995) 419;
(b) R.U. Kirss, Organometallics 11 (1992) 497;
(c) H. Lehmkuhl, W. Leuchte, E. Janssen, J. Organomet. Chem. 30 (1971) 407;
(d) K.K. Joshi, R.H.B. Mais, F. Nyman, P.G. Owston, A.M. Wood, J. Chem. Soc. A (1968) 318;
(e) J. Mueller, W. Holzinger, F.H. Kohler, Chem. Ber. 109 (1976) 1222;
(f) R. Basta, R.D. Ernst, Reaction of FeCl2 with the cyclononadienyl anion yields bis(indenyl)diiron, unpublished results.