Bis(alkynyl), metallacyclopentadiene, and diphenylbutadiyne complexes of ruthenium

Article abstract of DOI:10.1021/om060888l

Heating diphenylbutadiyne with [Ru(CO)₂(PPh₃) ₃] or [Ru(CO)₃(PPh₃)₂] in toluene under reflux provides respectively the ruthenacyclopentadiene [Ru{κ²-CR=CPhCPh=CR}(CO)₂(PPh₃) ₂] (R = C≡CPh) or the cyclopentadienone complex [Ru{η⁴-O=CC₄Ph₂R₂}(CO) ₂(PPh₃)], the latter via [2 + 2 + 1] alkyne and CO cyclization. The bis(alkynyl) complex cis,cis,trans-[Ru(C≡CPh) ₂(CO)₂(PPh₃)₂] is not formed in either of these reactions but is the product of the reaction of [RuCl ₂(CO)₂(PPh₃)₂] with LiC≡CPh or of cis,-mer-[Ru(C≡CPh)₂(CO)(PPh₃)₃] with CO. Although the bis(alkynyl) complex does not undergo reductive elimination to provide the diyne complex, thermolysis of cis,cis,trans-[Ru(C≡CPh) (HgC≡CPh)(CO)₂-(PPh₃)₂] (obtained from [Ru(CO)₂(PPh₃)₃] and [Hg(C≡CPh) ₂]) provides a noninterconvertible 1:1 mixture of cis,cis,trans-[Ru(C≡CPh)₂(CO)₂(PPh₃) ₂] and [Ru(η-PhC≡CC≡CPh)(CO)₂(PPh ₃)₂].

Full text of DOI:10.1021/om060888l

Organometallics 2007, 26, 1325-1338
1325
Bis(alkynyl), Metallacyclopentadiene, and Diphenylbutadiyne
Complexes of Ruthenium
Anthony F. Hill,* A. David Rae,† Madeleine Schultz,‡ and Anthony C. Willis
Research School of Chemistry, Institute of AdVanced Studies, Australian National UniVersity,
Canberra, Australian Capital Territory, Australia
ReceiVed September 28, 2006
Heating diphenylbutadiyne with [Ru(CO)2(PPh3)3] or [Ru(CO)3(PPh3)2] in toluene under reflux provides
respectively the ruthenacyclopentadiene [Ru{κ2-CRdCPhCPhdCR}(CO)2(PPh3)2] (R ) CtCPh) or the
cyclopentadienone complex [Ru{η4-OdCC4Ph2R2}(CO)2(PPh3)], the latter via [2 + 2 + 1] alkyne and
CO cyclization. The bis(alkynyl) complex cis,cis,trans-[Ru(CtCPh)2(CO)2(PPh3)2] is not formed in either
of these reactions but is the product of the reaction of [RuCl2(CO)2(PPh3)2] with LiCtCPh or of cis,-
mer-[Ru(CtCPh)2(CO)(PPh3)3] with CO. Although the bis(alkynyl) complex does not undergo reductive
elimination to provide the diyne complex, thermolysis of cis,cis,trans-[Ru(CtCPh)(HgCtCPh)(CO)2-
(PPh3)2] (obtained from [Ru(CO)2(PPh3)3] and [Hg(CtCPh)2]) provides a noninterconvertible 1:1 mixture
of cis,cis,trans-[Ru(CtCPh)2(CO)2(PPh3)2] and [Ru(η-PhCtCCtCPh)(CO)2(PPh3)2].
Introduction
We have recently reported a remarkably facile cleavage of
the Csp-Csp bonds of dimetallaoctatetraynes [L(CO)2WtCCt
CCtCCtW(CO)2L] (L ) HB(pz)3, HB(pzMe2)3; pz ) pyra-
zolyl) upon reaction with the complex [Ru(CO)2(PPh3)3] (1) to
form [Ru{CtCCtW(CO)2L}2(CO)2(PPh3)2].1 The same prod-
the cyclopentadienone complex [Ru{κ1-S:η4-OdC(CMedCSC2-
H4)2S}(CO)(PPh3)] (6) is obtained from the reaction of 1 with
4,7,10-trithiatrideca-2,11-diyne.5 We have previously reported
uct is obtained from the reaction of 1 with the bis(tricarbido)
mercurials Hg{CtCCtW(CO)2L}2, via simple π adducts of
the dimetallaoctatetraynes with mercury extrusion.1 These
somewhat surprising results have led us to now revisit the
reaction of 1 with the simpler reagent PhCtCCtCPh2 and to
investigate the reactions of 1 with Hg(CtCPh)2 in the hope of
shedding light upon the curious mechanistic aspects. The
reactions of [Ru(CO)2(PPh3)2(L)] (L ) PPh3 (1), CO (2)) with
diphenylacetylene in refluxing toluene are markedly dependent
on the ligand “L”.3 Thus for 1, which has one labile phosphine,
acomplexof2-phenylindenoneresults,[Ru(η4-OdCCPhCHC6H4)-
(CO)(PPh3)2] (3), via the intermediacy of the previously
reported4 alkyne complex [Ru(η-PhCtCPh)(CO)2(PPh3)2] (4a).
In contrast, for complex 2, the tetraphenylcyclopentadienone
complex [Ru(η4-OdCC4Ph4)(CO)2(PPh3)] (5) is obtained, while
the synthesis of the diyne complex [Ru(η-PhCtCCtCPh)-
(CO)2(PPh3)2] (4b) via phosphine substitution with 1.2 In the
interim it has been reported that 2 does not react with alkynes
in refluxing toluene. It was claimed, however, that 2 could be
activated toward alkyne coupling processes by carbon dioxide
via the proposed intermediate salts cis- and trans-[Ru(CO)4-
(PPh3)2][O2CO] (7) and isolated carbonato complex [Ru(O2-
CO)(CO)2(PPh3)2] (8)6 (reported previously from the reaction
of 2 with air).8 This complex was, however, not formed from 2
and CO2 in the absence of alkynes.6 Regarding the existence of
the proposed carbonate salt 7, we note that Hieber has reported
* To whom correspondence should be addresed: E-mail: a.hill@
anu.edu.au.
† E-mail for crystallographic enquiries: rae@rsc.anu.edu.au.
‡ Current address: School of Molecular and Microbial Sciences,
University of Queensland, St Lucia, Australia 4072.
(1) Dewhurst, R. D.; Hill, A. F.; Rae, A. D.; Willis, A. C. Organome-
tallics 2005, 24, 4703.
(2) Alcock, N. W.; Hill, A. F.; Melling, R. P.; Thompsett, A. R.
Organometallics 1993, 12, 641.
(3) Hill, A. F.; Schultz, M.; Willis, A. C. Organometallics 2004, 23,
5729.
(4) Cavit, B. E.; Grundy, K. R.; Roper, W. R. J. Chem. Soc., Chem.
Commun. 1972, 60.
(5) Hill, A. F.; Rae, A. D.; Schultz, M.; Willis, A. C. Organometallics
2004, 23, 81.
(6) Yamazaki, H.; Taira, Z. J. Organomet. Chem. 1999, 578, 61.
(7) Hieber, W.; Frey, V.; John, P. Chem. Ber. 1967, 100, 1961.
(8) Valentine, J.; Valentine, D. Jr.; Collman, J. P. Inorg. Chem. 1971,
10, 219.
10.1021/om060888l CCC: $37.00 © 2007 American Chemical Society
Publication on Web 02/10/2007

1326 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
the synthesis of the salt [Os(CO)4(PPh3)2][AlCl4]2 from Al2-
Cl6, CO (300 atm, 100 °C) and [OsCl2(CO)2(PPh3)2] but that
the reaction stops for ruthenium at [RuCl(CO)3(PPh3)2][AlCl4].
Both species are exceedingly reactive toward oxygen nucleo-
philes, making a carbonate salt implausible. We have since
shown that CO2 is not required for the thermal reaction of 2
(or 1) with diphenylacetylene,3 and since the diyne complex
4b seemed a plausible mechanistic candidate for an intermediate
in the reactions of 2 with PhCtCCdCPh, we have now
investigated these reactions in detail and can clarify, at least to
a degree, the apparently disparate results in the literature.
Figure 1. Molecular geometry of 4b in the crystal of 4b‚C6H6
(50% displacement ellipsoids, phosphine phenyl groups and
hydrogen atoms omitted). Selected bond lengths (Å) and angles
(deg): Ru1-P1 ) 2.375(2), Ru1-P2 ) 2.388(2), Ru1-C9 )
2.119(7), Ru1-C10 ) 2.145(6), C3-C9 ) 1.442(9), C9-C10 )
1.305(9), C10-C11 ) 1.383(9), C11-C12 ) 1.226(9), C12-C13
) 1.415(9); P1-Ru1-P2 ) 177.9(1), C1-Ru1-C2 ) 106.9(3),
C9-Ru1-C10 ) 35.6(2), Ru1-C1-O1 ) 178.7(5), Ru1-C2-
O2 ) 178.2(6), Ru1-C9-C3 ) 147.3(4), Ru1-C9-C10 ) 73.3-
(4), Ru1-C10-C9 ) 71.1(4), Ru1-C10-C11 ) 145.5(5), C3-
C9-C10 ) 139.4(6), C9-C10-C11 ) 143.3(7), C10-C11-C12
) 177.4(7), C11-C12-C13 ) 170.0(5). The diagram depicts one
of two disordered orientations for the alkynyl phenyl ring based
on C13′.
Results and Discussion
Given the centrality of diyne coordination to the results that
follow and the scarcity of structural data for mononuclear diyne
complexes of group 8 metals (one previous example on iron),
we have now structurally characterized the key complex 4b as
a benzene monosolvate. The gross results of this study are
summarized in Figure 1; however, it should be noted that the
solution involved a considerable amount of nonstandard model-
ing due to the 50:50 positional disorder of the majority of the
phenyl rings in the molecule (see Experimental Section). The
general coordination and structural chemistry of di- and polyynes
has been reviewed recently,9 and crystallographic data9-11 for
20 mononuclear diyne complexes in addition to those for 4b
have been tabulated (Table 1), from which it can be seen that
the ratios of bond lengths for the coordinated and free CtC
bonds fall in the range 1.051-1.151: i.e., a coordinative
lengthening of 5-15%.
The anomalous example in this list is Berke’s complex [Fe-
(η-Me3SiCtCCtCSiMe3)(CO)2(PEt3)2] (1.031),11r which of all
the complexes is most akin to 4b. The modest lengthening of
the coordinated CtC bond length relative to the uncoordinated
alkyne moietry seen for this iron complex is noteworthy in that
one might expect significant retrodonation from the π-basic iron
center. In a similar manner the 6% lengthening observed for
Table 1. Structural Data for Mononuclear 1,3-Diyne
Complexes10,11
complex
a, Å
b, Å a/b, Å R, deg
[Fe(Me3SiC4SiMe3)(CO)2(PEt3)2]
[Rh(PhC4Ph)Br(PiPr3)2]
1.251 1.213 1.031
1.260 1.199 1.051
1.261 1.192 1.058
1.289 1.212 1.064
1.305 1.226 1.064
1.268 1.186 1.069
1.278 1.193 1.071
1.286 1.197 1.074
1.280 1.191 1.075
1.279 1.189 1.076
1.312 1.214 1.081
1.305 1.200 1.088
1.316 1.210 1.088
1.288 1.184 1.088
1.307 1.201 1.089
1.313 1.200 1.094
1.293 1.176 1.099
1.336 1.210 1.104
1.341 1.192 1.125
1.326 1.161 1.142
150.0
158.5
156
[Rh(Me3SiC4SiMe3)Cl(PiPr3)2]
[Ni(Me3SiC4SiMe3)(PPh3)2]
[Ru(PhC4Ph)(CO)2(PPh3)2]
148.0
143.3
144.5
144.2
151.3
146.2
149.4
141.9
146.7
140.5
152.3
142.1
140.0
147.0
127.8
138.1
149.6
130.8
t
[Ni(tBuC4 Bu){P(o-tolyl)3}2]
[Nb(PhC4Ph)Cl(C5H4SiMe3)2]a
[Ni(PhC4Ph)(PPh3)2]
[Ni(HC4CH){(iPr2PCH2)2CH2}]
t
[Ni(tBuC4 Bu)(PPh3)2]
[Ti(PhC4Ph)(C5Me5)2]
[Pt(PhC4Ph)(PPh3)2]
[Ti(PhC4Ph)(TTP)]c
t
[Ni(tBuC4 Bu)(bipy)]
[W(PhC4SiMe3)F5Na([16]aneO5)
(9) Low, P. J.; Bruce, M. I. AdV. Organomet. Chem. 2002, 48, 71.
(10) Conquest; Cambridge Crystallographic Data Centre, Release May
2006.
t
[Co(tBuC4 Bu){Ph2PCH2)CMe]PF6
[Pt(PhC4Ph){Ph2PC5H4)2Fe}]
[Zr(PhC4Ph)(CCPh)(C5H5)2Li(thf)2]
[Nb(MeC4Me)I(CO)2(PEt3)2]
[W(PhC4Ph)Cl5][PPh4]
(11) (a) Pelling, P.-M.; Kirchbauer, F. G.; Burlakov, U. U.; Baumann,
W.; Spannenburg, A. J. Am. Chem. Soc. 1999, 121, 8313. (b) Stahl, K.;
Weller, F.; Dehnicke, K. Z. Anorg. Allg. Chem. 1984, 518, 175. (c) Rupp,
R.; Huttner, G.; Lang, H.; Heinze, K.; Buchner, M.; Hoverstreydt, E. R.
Eur. J. Inorg. Chem. 2000, 1953. (d) Bonrath, W.; Porschke, K. R.; Wilke,
G.; Angermund, K.; Kruger, C. Angew. Chem., Int. Ed. Engl. 1988, 27,
833. (e) Castellano, B.; Solani, E.; Floriani, C.; Re, N.; Chiesi-Villa, A.;
Rizzoli, C. Chem., Eur. J. 1999, 5, 722. (f) Rapport, T.; Burnberg, O.;
Werner, H. Organometallics 1993, 12, 1359. (g) Garcia-Yebra, C.; Carrero,
F.; Lopez-Mardomingo, C.; Fajardo, M.; Rodriguez, A.; Antinolo, A.; Otero,
A.; Lucas, D.; Mugnier, Y, Organometallics 1999, 18, 1287 (h) Yamazaki,
S.; Deeming, A. J.; Speel, D. S. Organometallics 1998, 17, 775. (i)
Choukroun, R.; Zhao, J.; lrober, C.; Cassoux, P.; Donnadieu, B. Chem.
Commun. 2000, 1151. (j) Ursini, C. V.; Dias, G. H. M.; Horner, M.;
Bortoluzzi, A. J.; Morigaki, M. K. Polyhedron 2000, 19, 2261. (k) Gauss,
C.; Veghini, D.; Berke, H. Chem. Ber. 1997, 130, 183. (l) Werth, A.;
Dehnicke, K.; Fenske, D.; Baum, G. Z. Anorg. Allg. Chem. 1990, 591, 125.
(m) Werner, H.; Gevert, O.; Haquette, P. Organometallics 1997, 16, 803.
(n) Chen, J.; Guzei, I. A.; Woo, L. K. Inorg. Chem. 2000, 39, 3715. (o)
Rosenthal, U.; Pulst, S.; Arndt, P.; Baumann, W.; Tillack, A.; Kempe, R.
Z. Naturforsch. 1995, 50b, 368. (p) Rosenthal, U.; Pulst, S.; Arndt, P.;
Baumann, W.; Tillack, A.; Kempe, R. Z. Naturforsch. 1995, 50b, 377. (q)
Rodewald, D.; Schulzke, C.; Rehder, D. J. Organomet. Chem. 1995, 498,
29l. (r) Gauss, C.; Veghini, D.; Berke, H. Chem. Ber. 1997, 130, 183.
[Ta(PhC4Ph)(CCPh)(calix)LiOEt2]a,b 1.386 1.204 1.151
a Average of two crystallographically independent molecules. b calix )
calix[4]arene tetranion. c TTP ) tetraphenylporphyrinate dianion.
4b is also relatively small (C9-C10 ) 1.305(9) Å), although
somewhat greater than that observed for the simple alkyne
complex [Ru(η-PhCtCPh)(CO)2(PPh3)2] (1.274(3) Å).3 The
equatorial sites of a d8 trigonal bipyramid are generally
considered to be the more strongly π basic; however, in the
case of 4b and the other anomalous complex discussed above,
the two carbonyl ligands compete for the metal’s retrodative
capacity, an interpretation that would account for the modest
lengthening of the alkyne upon coordination. Notably, for potent
π-acids coordinated to the “Ru(CO)2(PPh3)2” fragment, an
alternative geometry with cis-equatorial phosphines has been
observed: e.g., in the tetrafluoroethene complex [Ru(η-C2F4)-

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1327
(CO)2(PPh3)2].12 That the coordinated CtC bond of 4b is longer
than that of 4a might be traced to the electron-withdrawing
nature of the alkynyl substituent, increasing retrodonation. This
is supported by the increase in the frequency and force constant
of the carbonyl associated infrared absorptions (4a, νCO 1962,
Scheme 1. Reactions of 1 and 2 with Diphenylbutadiyne
(L ) PPh3)
1892 cm-1, kCO ) 15.00 N cm-1; 4b, νCO 1971, 1908 cm-1
kCO ) 15.19 N cm-1).
,
Heating 1 with excess diphenylbutadiyne (toluene, reflux)
leads to immediate formation of 4b, followed by the slow (12-
14 h) formation of a new compound, 9, which gives rise to two
new νCO bands (2017, 1960 cm-1), in addition to an alkynyl
band at 2151 cm-1 (toluene solution). A further structured
absorption also develops in the range 1600-1595 cm-1
;
however, this could not be unequivocally assigned to the
coordinated alkyne, due to the aryl bands also appearing in this
region. The change in color from yellow to dark orange is
accompanied by the replacement of the original 31P{1H} NMR
singlet due to 1 (δP 50.9) with that of 4b (δP 43.7) and finally
with a new signal at δP 36.4. In refluxing xylene, the reaction
is complete in 10 min, giving the same product. The same
spectroscopic changes are seen when preisolated 4b is heated
in the presence of excess PhCtCCtCPh. When 4b was heated
in d8-toluene in the absence of additional diyne, after overnight
reflux the resonance due to 4b had decreased, being replaced
by a major resonance due to [Ru(CO)3(PPh3)2] (2; δP 57.1),
which may be accounted for by dissociation of the alkyne ligand
from 4b. Smaller resonances were observed at δP 51.3, 51.2,
50.2, 46.7, 38.2, 36.4 (9, vide infra), 25.5 (OPPh3), and -4.1
(PPh3). After 2 days at reflux, an additional resonance at δP
45.2 was observed; the components were all present in ap-
proximately equal amounts, although slightly more of 4b and
the component at δP 38.2 were present. Further heating did not
change the composition of the mixture, and no tractable product
could be isolated. Reflux of 4b in the lower boiling C6D6
resulted in a similar mixture, with an additional resonance at
δP 33.7. Thus, it appears that further diyne is required for the
clean formation of 9 and that the small amount of 9 formed in
the absence of added diyne can be accounted for by ligand
scrambling in 4b. Within this manifold of reagents, 2 appears
to be particularly stable and generally accumulates in such ligand
scrambling reactions.
The presence of a cis-/trans-“RuII(CO)2(PPh3)2” group strad-
dling a molecular plane of symmetry in the complex 9 was
deduced from a combination of spectroscopic data (two ν(CO)
absorptions, one 13C{1H} carbonyl resonance, a single 31P{1H}
NMR resonance, and virtual triplet structure for PC6H5 13C-
{1H} NMR signals). The nature of the organic ligand(s)
occupying the remaining two sites did not, however, follow
unambiguously from spectroscopic data, due in part to the
complexity of the aromatic regions of the 13C{1H} and 1H NMR
spectra. The identity of the product was eventually established
(12) Burrell, A. K.; Clark, G. R.; Rickard, C. E. F.; Roper, W. R.; Ware,
D. C., J. Organomet. Chem. 1990, 398, 133.

1328 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
Table 2. Structural Data (Å) for Ruthenacyclopentadienes
and -trienes10,12-15
complex
a, Å
b, Å
c, Å
ra
9
2.178
2.187
2.121
2.10
2.129
2.093
2.114
2.088
2.109
2.118
1.942
1.367
1.371
1.36
1.459
1.066
11
1.46
1.066
1.036
1.067
1.090
1.019
Figure 2. Molecular geometry of 9 in the crystal (50% displace-
ment ellipsoids, phosphine phenyl groups omitted, remaining phenyl
groups simplified, hydrogen atoms omitted, one of two crystallo-
graphically independent molecules shown). Selected bond lengths
(Å) and angles (deg): Ru1-P1 ) 2.388(6), Ru1-P2 ) 2.393(6),
Ru1-C1 ) 2.00(2), Ru1-C2 ) 1.99(2), Ru1-C5 ) 2.19(2), Ru1-
C8 ) 2.18(2), O1-C1 ) 1.09(2), O2-C2 ) 1.10(2), C3-C4 )
1.19(3), C3-C31 ) 1.42(3), C4-C5 ) 1.43(3), C5-C6 ) 1.37-
(3), C6-C7 ) 1.46(3), C7-C8 ) 1.37(3), C8-C9 ) 1.42(3), C9-
C10 ) 1.14(3); P1-Ru1-P2 ) 174.4(3), C1-Ru1-C2 ) 99.5(12),
C1-Ru1-C8 ) 87.3(11), C2-Ru1-C5 ) 96.4(11), C5-Ru1-
C8 ) 76.8(10), Ru1-C1-O1 ) 173(3), Ru1-C2-O2 ) 176(3),
C4-C3-C3 ) 178(3), C3-C4-C5 ) 172(3), Ru1-C5-C4 )
116(2), Ru1-C5-C6 ) 115(2), C4-C5-C6 ) 129(2), C5-C6-
C7 ) 116(2), C5-C6-C61 ) 118(2), C7-C6-C61 ) 126(2),
C6-C7-C8 ) 118(2), C6-C7-C71 ) 122(2), C8-C7-C71 )
120(2), Ru1-C8-C7 ) 114(2), Ru1-C8-C9 ) 123(2), C7-C8-
C9 ) 123(2), C8-C9-C10 ) 174(3), C9-C10-C101 ) 174(3).
1.38
12
benzo
13
benzo
14
1.351
1.422
1.359
1.400
1.347
1.352
1.403
1.437
1.472
1.471
1.377
10
a r ) 2c/(b + b′) provides a singular indication of the degree of multiple
bond localization.
CtCPh),14 which has been obtained from the reaction of [Ru3-
(NCMe)2(CO)10] with diphenylbutadiyne, the amine ligand
arising from reduction of Me3NO required for the in situ
synthesis of [Ru3(NCMe)2(CO)10] from [Ru3(CO)12]. The
complex 11 was only obtained in 1.5% yield (3 mg), precluding
conjecture as to its mechanistic significance, with the major
products being tri- and tetrametallic diyne complexes, in addition
to a bimetallic ruthenacyclopentadiene and a diruthenatropolone.
Each of these feature alkyne coupling with regiochemistries
distinct from that required for the formation of either 9 or 11.
crystallographically and shown to be the ruthenacyclopentadiene
(“ruthenole”) complex [Ru(κ2-CRdCPhCPhdCR)(CO)2(PPh3)2]
(9, R ) CtCPh; Scheme 1). The results of this study are
summarized in Figure 2, which depicts one of two crystallo-
graphically independent but similar molecules of the metalla-
cycle in the asymmetric unit. The geometric parameters of the
“Ru(CO)2(PPh3)2” fragment are unremarkable for this metal-
ligand combination in a C2V octahedral arrangement. Bond
distances around the RuC4 ring establish that 9 is best described
as a ruthenacyclopentadiene, with an essentially localized Ru-
CR, CRdCâ, and Câ-Câ′ bonding pattern. A ruthenacycle with
the same connectivity, the complex [RuBr(κ2-C4Ph4)(η-C5H5)]
(10), has been alternatively described, with some justification,
as a ruthencyclopentatriene.13 The dichotomy arises from the
number of valence electrons required by the ruthenium to attain
an 18-electron configuration. Table 2 summarizes structural
parameters for the mononuclear ruthenacyclopentadienes
9-14,13-16 including Singleton’s ruthenacyclopentatriene 10.13
Although not strictly mononuclear, the complex {Ru{κ2-C4(CO2-
Et)4}(CO)3}2 (14) is included, since the metallacyclopentadiene
is not intimately involved in bridging the two metals; rather,
the dimer is held together by weak coordination of the ester
substituents.16 This is in contrast to the large number of bi- and
polynuclear ruthenacyclopentadiene compounds wherein the
prevalent “flyover” arrangement of the metallacyclopentadiene
(“metallaole”) supports a metal-metal bond.17
(17) (a) Inagaki, A.; Takao, T.; Moriya, M.; Suzuki, H. Organometallics
2003, 22, 2196. (b) Inagaki, A.; Musaev, D. G.; Toshifumi, T.; Suzuki, H.;
Morokuma, K. Organometallics 2003, 22, 1718. (c) Bruce, M. I.; Pyke, S.
M.; Zaitseva, N. N.; Skelton, B. W.; White, A. H. HelV. Chim. Acta 2001,
84, 3197. (d) Suzuki, H.; Inagaki, A.; Matsubara, K.; Takemori, T. Pure
Appl. Chem. 2001, 73, 315. (e) Matsuzaka, H.; Ichikawa, K.; Ishioka, T.;
Sato, H.; Okubo, T.; Ishii, T.; Yamashita, M.; Kondo, M.; Kitagawa, S. J.
Organomet. Chem. 2000, 596, 121. (f) Inagaki, A.; Takenori, T.; Tanaka,
M.; Suzuki, H. Angew. Chem., Int. Ed. 2000, 39, 404. (g) Zdanovich, V. I.;
Lagunovoa, V. Yu.; Dolgushin, F. M.; Yanovsky, A. I.; Ezernitskaya, M.
G.; Petrovskii, P. V.; Koridze, A. A. Russ. Chem. Bull. 1998, 47, 1789. (h)
Koridze, A. A.; Zdanovich, V. I.; Lagunova, V. Yu.; Ezernitskaya, M. G.;
Petrovskii, P. V.; Dolgushin, F. M.; Yanovsky, A. I. Russ. Chem. Bull.
1998, 47, 988. (i) Bruce, M. I.; Zaitseva, N. N.; Skelton, B. W.; White, A.
H. Russ. Chem. Bull. 1998, 47, 983. (j) Bruce, M. I.; Hinchliffe, J. R.;
Humphrey, P. A.; Surynt, R. J.; Skelton, B. W.; White, A. H. J. Organomet.
Chem. 1998, 552, 109. (k) Bruce, M. I.; Zaitseva, N. N.; Skelton, B. W.;
White, A. H. J. Organomet. Chem. 1997, 536/537, 93. (l) Tunik, S. P.;
Grachova, E. V.; Denisov, V. R.; Starova, G. L.; Nikol’skii, A. B.;
Dolgushin, F. M.; Yanovsky, A. I.; Struchkov, Yu. T. J. Organomet. Chem.
1997, 536/537, 339. (m) Blake, A. J.; Haggitt, J. L.; Johnson, B. F. G.;
Parsons, S. J. Chem. Soc., Dalton Trans. 1997, 991. (n) Blenkiron, P.;
Enright, G. D.; Carty, A. J. Chem. Commun. 1997, 483. (o) Koridze, A.
A.; Zdanovich, V. I.; Andrievskaya, N. V.; Siromakhova, Yu.; Petrovskii,
P. V.; Ezernitskaya, M. G.; Dolgushin, F. M.; Yanovskii, A. I.; Struchkov,
Yu. T. IzV. Akad. Nauk, Ser. Khim. 1996, 1261; (p) Nishio, M.; Matsuzaka,
H.; Mizobe, Y.; Tanase, T.; Hidai, M. Organometallics 1994, 13, 4214.
(q) Suzuki, H.; Omori, H.; Lee, D.-H.; Yoshida, Y.; Fukushima, M.; Tanaka,
M.; Moro-oka, Y. Organometallics 1994, 13, 1129. (r) Koridze, A. A.;
Yanovsky, A. I.; Struchkov, Yu. T. J. Organomet. Chem. 1992, 441, 277.
(s) Boroni, E.; Costa, M.; Predieri, G.; Sappa, E.; Tiripicchio, A. J. Chem.
Soc., Dalton Trans. 1992, 2585. (s) Bruce, M. I.; Humphrey, P. A.;
Miyamae, H.; Skelton, B. W.; White, A. H. J. Organomet. Chem. 1992,
429, 187. (t) Osella, D.; Gobetto, R.; Milone, L.; Zanello, P.; Mangani, S.
Organometallics 1990, 9, 2167. (u) Campion, B. K.; Heyn, R. H.; Tilley,
T. D. Organometallics 1990, 9, 1106. (v) Omori, H.; Suzuki, H.; Morooka,
Y. Organometallics 1989, 8, 1576. (w) Han, S. H.; Geoffroy, G. L.;
Rheingold, A. L. Organometallics 1987, 6, 2380. (x) Busetti, V.; Granozzi,
G.; Aime, S.; Gobetto, R.; Osella, D. Organometallics 1984, 3, 1510. (y)
Bruce, M. I.; Matisons, J. G.; Skelton, B. W.; White, A. H. J. Organomet.
Chem. 1983, 251, 249. (z) Noda, I.; Yasuda, H.; Nakamura, A. J.
Organomet. Chem. 1983, 250, 447; (aa) Aime, S.; Deeming, A. J. J. Chem.
Soc., Dalton Trans. 1981, 828. (ab) Low, P. J.; Udachin, K. A.; Enright,
G. D.; Carty, A. J. J. Organomet. Chem. 1999, 578, 103.
Structural data for 9, contextualized by Table 2, clearly
indicate that the ruthenacyclopentadiene description is apt,
consistent with the 16-valence-electron nature of the “Ru(CO)2-
(PPh3)2” fragment. Most relevant to the structure of 9 is the
complex [Ru(κ2-CRdCPhCPhdCR)(CO)3(NMe3)] (11, R )
(13) Albers, M. O.; De Waal, D. J. A.; Liles, D. C.; Robinson, D. J.;
Singleton, E.; Wiege, M. B. J. Chem. Soc., Chem. Commun. 1986, 1680-
1682.
(14) Bruce, M. I.; Zaitseva, N. N.; Skelton, B. W.; White, A. H. Inorg.
Chim. Acta 1996, 250, 129.
(15) Burns, R. M.; Hubbard, J. L. J. Am. Chem. Soc. 1994, 116, 9514.
(16) Lindner, E.; Jansen, R.-M.; Mayer, H. A.; Hiller, W.; Fawzi, R.
Organometallics 1989, 8, 2355.

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1329
The regioselectivity observed in the exclusive formation of
the R,R′-bis(alkynyl)ruthenacyclopentadiene geometry of 9 is
noteworthy; since phenyl substituents exert a greater steric
influence than do alkynyls, it might be presumed that the
regioselectivity is electronic in origin. Thus, while the location
of the phenyl groups â to the ruthenium and remote from the
bulky phosphines in the final product may be the least sterically
congested, this would involve greater steric pressures en route
to the transition state for alkyne coupling (Scheme 1). The
regiochemistry of diyne coupling is of interest in that the R,R′
(i.e., 2,5) geometry may result in interesting electronic, nonlinear
optical, and luminescent properties, as demonstrated by Marder
for the luminescent rhodacyclopentadiene complex mer-[Rh-
(CtCSiMe3)(PMe3)3{C(CtCR)CRCRC(CtCR)}] (15, R )
C6H4Me-4).18 The complex 15 may be considered isoelectronic
butadiyne have so far been unsuccessful. Heating 1 with excess
Me3SiCtCCtCSiMe3 in toluene under reflux provides an
inseparable mixture of numerous phosphine complexes (31P-
{1H} NMR: inter alia δP 33.6, 35.7, 49.1, 49.4, 50.6, 51.3, 51.9,
54.1, 56.6), the predominant species being [Ru(CO)3(PPh3)2]
(2; δP 57.1). The reaction of 1 with Me3SiCtCCtCSiMe3
under ambient conditions leads to a 5:1 equilibrium mixture of
1 and what we presume to be the simple but very labile diyne
complex [Ru(η-Me3SiCtCCtCSiMe3)(CO)2(PPh3)2] (4c) on
the basis of a 31P{1H} resonance at δP 45.5, in a region
comparable to that for 4b (δP 43.7). Attempts to isolate 4c,
however, only led to recovery of the sparingly soluble 1 from
a presumed crystallization-perturbed equilibrium, indicating that
the liberated phosphine competes effectively with the diyne. In
a similar manner, the reaction of [Ru(η-C2H4)(CO)2(PPh3)2]
(20)4 with Me3SiCtCCtCSiMe3 proceeds to an approximate
2:1 equilibrium mixture of 20 (δP 58.0) and 4c when carried
out in a sealed vessel. However, in contrast to the reaction of
1 and Me3SiCtCCtCSiMe3, the volatility of ethylene allows
the equilibrium to be drawn to the formation of 4c by either a
continuous nitrogen purge of the system or, alternatively (and
more effectively), by removing the ethylene by periodic
evacuation of the reaction vessel. The increased lability of Me3-
SiCtCCtCSiMe3 in 4c relative to that of PhCtCCtCPh in
4b has compromised the acquisition of satisfactory elemental
microanalytical data, with samples being invariably contami-
nated with [Ru(η-O2)(CO)2(PPh3)2] (21;4 δP 34.9),20 the forma-
tion of which is essentially irreversible. Ruthenacyclopentadienes
have come to be recognized as mechanistically significant
species in a range of alkyne transformations mediated by low-
valent ruthenium complexes. This has included the synthesis
of pyridines from nitriles and R,ω-diynes,21 the cycloisomer-
ization of diynols to 2-vinyl-1-acylcycloalkenes,22 and the
catalytic linear coupling of alkynes with alkenes.23 Ruthenacy-
clopentadienes might also appear as attractive mechanistic
intermediates in the formation of cyclopentadienone complexes,
for which there are many ruthenium examples.3,5,6,24 However,
the stability of 9 (recovered unchanged from refluxing xylene)
would appear to add to the evidence against such routes.
Furthermore, we note that 9 fails to react with either carbon
monoxide or ethyne (2 atm, d6-benzene, 80 °C). These reagents
might be expected to intercept any coordinatively unsaturated
ruthenacyclohexadienone tautomer if formed, given that [Os-
(SCCHdCHCHdCH)(CO)(PPh3)2] reacts rapidly with CO.25
with 9, although the coligand sets are quite disparate, the former
being devoid of effective π acidic ligands while the latter has
strong π acids trans to the metallacycle. Unfortunately, Marder’s
complex 15 is a unique example (now augmented by 9) where,
within detectable limits, the desired R,R′ isomer is exclusively
formed, in high yield. Thus, while other mononuclear bis-
(alkynyl)metallacyclopentadienes have been obtained via diyne
coupling (16-19),19 the desirable R,R′ isomer is either not
formed, or is the minor component.
(18) Rourke, J. P.; Batsanov, A. S.; Howard, J. A. K.; Marder, T. B.
Chem. Commun. 2001, 2626.
(19) (a) Shimura, T.; Ohkubo, A.; Aramaki, K.; Uekusa, H.; Fujita, T.;
Ohba, S.; Nishihara, H. Inorg. Chim. Acta 1995, 230, 215. (b) Fujita, T.;
Uekusa, H.; Ohkubo, A.; Shimura, T.; Aramaki, K.; Nishihara, H.; Ohba,
S. Acta Crystallogr. 1995, C51, 2265. (c) Hsu, D. P.; Davis, W. M.;
Buchwald, S. L. J. Am. Chem. Soc. 1993, 115, 10394. (d) Burlakov, V. V.;
Ohff, A.; Lefeber, C.; Tillack, A.; Baumann, W.; Kempe, R.; Rosenthal,
U. Chem. Ber. 1995, 128, 967. (e) Rosenthal, U.; Pellny, P.-M;, Kirchbauer,
F. G.; Burlakov, V. V. Acc. Chem. Res. 2000, 33, 119.
(20) The complex was also crystallographically characterized and refined
to a more precise model than that previously reported (R ) 0.066) in: Hiraki,
K.; Kira, S.; Kawano, H. Bull. Chem. Soc. Jpn. 1997, 70, 1583.
(21) Varela, J. A.; Castedo, L.; Saa, C. J. Org. Chem. 2003, 68, 8595.
(22) Trost, B. M.; Rudd, M. T. J. Am. Chem. Soc. 2003, 125, 11516.
(23) Yi, C. S.; Liu, N. Synlett 1999, 281.
We have not spectroscopically detected alternative regioiso-
mers. Our attempts to extend this reaction to bis(trimethylsilyl)-

1330 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
Scheme 2. Systematic Preparation of cct-22 (L ) PPh3)
Figure 3. (a, top) Molecular geometry of 22 in a crystal of 220.896
-
Although they were eventually obtained, crystallographic
grade crystals of 9 were not immediately forthcoming, a situation
that led us to explore possible formulations for 9 via the less
fashionable technique of unequivocal independent synthesis,
summarized in Scheme 2. Our initial suspicions focused on the
previously unknown complex [Ru(CtCPh)2(CO)2(PPh3)2] (22),
(24a‚CHCl3)0.104 (50% displacement ellipsoids, phosphine phenyl
groups and hydrogen atoms omitted). Selected bond lengths (Å)
and angles (deg): Ru1-P1 ) 2.392(1), Ru1-P2 ) 2.380(1), Ru1-
C1 ) 1.921(5), Ru1-C2 ) 1.916(5), Ru1-C3 ) 2.068(4), Ru1-
C11 ) 2.088(5), O1-C1 ) 1.133(5), O2-C2 ) 1.137(5), C3-
C4 ) 1.194(6), C4-C5 ) 1.448(5), C11-C12 ) 1.177(7); P1-
Ru1-P2 ) 175.3(1), C1-Ru1-C2 ) 95.7(2), C1-Ru1-C11 )
82.1(2), C2-Ru1-C3 ) 89.2(2), C3-Ru1-C11 ) 93.1(2), Ru1-
C1-O1 ) 178.0(4), Ru1-C2-O2 ) 178.4(4), Ru1-C3-C4 )
171.8(5), C3-C4-C5 ) 170.6(6), Ru1-C11-C12 ) 171.7(5),
C11-C12-C13 ) 177.9(6). (b, bottom) Molecular geometry of
24a in a crystal of 220.896(24a‚CHCl3)0.104 (50% displacement
ellipsoids, phosphine phenyl groups and hydrogen atoms omitted).
Selected bond lengths (Å) and angles (deg): Ru1-Cl1 ) 2.572-
(12), Cl1‚‚‚H1C19 ) 2.66(1); Cl1‚‚‚H1C19-C19 ) 156(3)°. Other
metric parameters are as for 22.
perhaps arising from C-C bond cleavage by analogy with 4d
or 4e, which readily oxidatively add one C-C bond.1 The
complexes [Ru(CtCPh)2(CO)2(PR3)2] (R ) Et, iPr) arise from
the reaction of [RuCl2(CO)2(PEt3)2] with LiCtCPh26 and
carbonylation of [Ru(CtCPh)2(CO)(PiPr3)2],27 respectively, and
it is noteworthy that they have distinct geometries; the PEt3
derivative assumes the all-trans geometry, in contrast to
cis,cis,trans-[Ru(CtCPh)2(CO)2(PiPr3)2]. Applying the former
strategy26 to the synthesis of the PPh3 derivative 22, we find
that it is in fact the cct-22 isomer that is obtained upon treating
cct-[RuCl2(CO)2(PPh3)2] (23; 31P{1H} NMR δ 17.7, IR 2056,
1998 cm-1) with freshly prepared lithium phenylacetylide (from
nBuLi and HCtCPh). This reaction proceeds via the mono-
substitution product [RuCl(CtCPh)(CO)2(PPh3)2] (24a) (ob-
served by 31P{1H} NMR spectroscopy, δ 22.5, but not isolated)
to the desired compound cct-22. The intermediate complex 24a
is analogous to [Ru(CtCC6H4Me-4)Cl(CO)2(PPh3)2] (24b; δP
23.1), which was has been previously obtained via an alternative
route that ensures exclusive formation of the monosubstituted
derivative.28 Although the stereochemistry of 22 follows from
spectroscopic data (two ν(CO) IR absorptions, one 31P reso-
nance, and virtual triplet C1,2,3,5,6(PC6H5) 13C resonances), the
complex was also structurally characterized. The molecular
geometry of 22 is summarized in Figure 3. Notably, detailed
analysis of the diffraction data indicated that the small twinned
needle chosen for study was a solid solution of 22 and the
intermediate 24a. In retrospect, it should be noted that the model
used for the structural solution of 24b included an anomalously
short CtC bond length that may well have arisen from a similar
superposition of chloride and alkynyl groups. In the present
(24) (a) Smith, T. P.; Kwan, K. S.; Taube, H.; Bino, A.; Cohen, S. Inorg.
Chem. 1984, 23, 1943. (b) Koridze, A. A.; Zdanovich, V. I.; Lagunova, V.
Y.; Petukhova, I. I.; Dolgushin, F. M.; Starikova, Z. A.; Ezernitskaya, M.
G.; Petrovskii, P. V. IzV. Akad. Nauk SSSR, Ser. Khim. 2002, 807. (c)
Yamazaki, Y.; Goto, M.; Kageyama, Y.; Tomohiro, T.; Okuno, H. Z.
Naturforsch. 1996, 51b, 301. (d) Schneider, B.; Goldberg, I.; Reshef, D.;
Stein, Z.; Shvo, Y. J. Organomet. Chem. 1999, 588, 92. (e) Shvo, Y.;
Czarkie, D.; Rahamim, Y.; Chodosh, D. F. J. Am. Chem. Soc. 1986, 108,
7400. (f) Akabori, S.; Kumagai, T.; Shirahige, T.; Sato, S.; Kawazoe, K.;
Tamura, C.; Sato, M. Organometallics 1987, 6, 2105. (g) Casey, C. P.;
Singer, S. W.; Powell, D. R.; Hayashi, R. K. Kavana, M. J. Am. Chem.
Soc. 2001, 123, 1090. (h) Hill, A. F.; Schultz, M.; Willis, A. C.
Organometallics 2005, 24, 2027. (i) Meijer, R. H.; Ligthart, G. B. W. L.;
Meuldijk, J.; Vekemans, J. A. J. M.; Hulshof. L. A.; Mills, A. M.; Kooijman,
H.; Spek, A. L. Tetrahedron 2004, 60, 1065.
(25) Elliott, G. P.; Roper, W. R.; Waters, J. M. J. Chem. Soc., Chem.
Commun. 1982, 811.
(26) Sun, Y.; Taylor, N. J.; Carty, A. J. Organometallics 1992, 11, 4293.
(b) Sun, Y.; Taylor, N. J.; Carty, A. J. J. Organomet. Chem. 1992, 423,
C43.
(27) (a) Werner, H.; Esteruelas, M. A.; Otto, H. Organometallics 1986,
5, 2295. (b) Werner, H.; Meyer, U.; Esteruelas, M. A.; Sola, E.; Oro, L. A.
J. Organomet. Chem. 1989, 366, 187. (c) Esteruelas, M. A.; Lahoz, F. J.;
Lopez, A. M.; On˜ate, E.; Oro, L. A. Organometallics 1994, 13, 1669.
(28) Bedford, R. B.; Hill, A. F.; Thompsett, A. R.; White, A. J. P.;
Williams, D. J. Chem. Commun. 1996, 1059.

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1331
study, a model based on a (22)0.894(24a‚CHCl3)0.106 stoichiom-
etry was refined satisfactorily, and information in Figure 3a is
based on the major contributor 22, while Figure 3b depicts the
minor component 24a‚CHCl3. Although the structural study
confirms the expected identity and connectivity, the nonstandard
nature of the solution makes it inappropriate to discuss geo-
metrical features in detail, other than to note that the stereo-
chemistry agrees with that predicted on the basis of spectro-
scopic data.
Our initial attempt to prepare 22 involved the synthesis of
the precursor cct-23 via carbonylation (1 atm, CH2Cl2) of
[RuCl2(PPh3)3], which provided a mixture of ttt-23 (IR 2000
cm-1) and cct-23 isomers (IR 2063, 1999 cm-1). Heating this
mixture under reflux followed by crystallization from dichlo-
romethane and ethanol appeared to give the pure cct-23 isomer,
as described previously.29,30 However, during attempts to obtain
crystals of 22 from an apparently spectroscopically pure solution
of that compound, a few crystals of a second compound were
obtained and structurally characterized as the sparingly soluble
complex cis,mer-[Ru(CtCPh)2(CO)(PPh3)3] (25a). This com-
pound has been briefly mentioned in a previous report,31 in
which it was obtained via the reaction of [Ru(CtCCMe3)2(CO)-
(PPh3)3] (25b) with excess phenylacetylene and characterized
on the basis of melting point, infrared (ν(CtC) 2075; ν(CO)
1965 cm-1) and elemental data (for the 1.5C6H6 solvate).35
Given that our structural characterization arose inadvertently,
more complete characterization thus seemed desirable (see
Experimental Section). The complex was therefore obtained on
a preparative scale via a two-step sequence: the complex [RuH-
(CtCPh)(CO)(PPh3)3] (26) is available via either the reaction
Figure 4. Molecular geometry of 25a in a crystal of 25a‚CH2Cl2
(50% displacement ellipsoids, phenyl groups simplified, hydrogen
atoms omitted). Selected bond lengths (Å) and angles (deg): Ru1-
P1 ) 2.377(3), Ru1-P2 ) 2.419(3), Ru1-P3 ) 2.478(3), Ru1-
C1 ) 1.910(10), Ru1-C2 ) 2.058(9), Ru1-C10 ) 2.034(9), O1-
C1 ) 1.139(9), C2-C3 ) 1.192(10), C3-C4 ) 1.475(10), C10-
C11 ) 1.186(10), C11-C12 ) 1.459(10); P1-Ru1-P2 ) 161.5(1),
P1-Ru1-P3 ) 97.8(1), P1-Ru1-C1 ) 91.1(3), P1-Ru1-C2 )
85.5(2), P1-Ru1-C10 ) 82.8(2), P2-Ru1-P3 ) 97.6(1), P2-
Ru1-C1 ) 99.9(3), P2-Ru1-C2 ) 82.2(2), P2-Ru1-C10 )
84.1(2), P3-Ru1-C1 ) 86.9(3), P3-Ru1-C2 ) 98.3(2), C1-
Ru1-C10 ) 81.4(4), C2-Ru1-C10 ) 93.4(3), Ru1-C1-O1 )
174.6(9), Ru1-C2-C3 ) 178.6(8), C2-C3-C4 ) 176.7(9), Ru1-
C10-C11 ) 169.7(8), C10-C11-C12 ) 173.6(9).
The molecular geometry of 25a is summarized in Figure 4,
which reveals a distorted-octahedral geometry about the six-
coordinate ruthenium center. The primary distortions may be
attributed to the steric pressures associated with the meridional
coordination of three bulky phosphine ligands. Somewhat
surprisingly, there are very few structural data presently available
for the mer-M(PPh3)3 geometry at a six-coordinate metal that
do not include at least one hydride coligand (and attendant
deformations). Indeed, these are limited to the rhenium salts
36
[ReF(CO)(NO)(PPh3)3]ClO4 and (17-electron) [Re(O2CMe)-
(CO)(PPh3)3]PF6.37 Unit cell data are available for the complex
25b;10,38 however, geometric data have yet to appear, precluding
useful comment. The Ru(PPh3)3 fragment in 25a is deformed
from ideal octahedral geometry such that the angle between trans
phosphorus donors is reduced to 161.5(1)° with cis angles
between these and P2 being 97.8(1) and 97.6(1)°. Similar angles
(161.9, 94.7, and 96.8°) are observed for [ReF(CO)(NO)(PPh3)3]-
ClO4.36 The alkynyl ligands in 25a show statistically identical
CtC bond lengths (1.187(13), 1.192(11) Å) and comparable
alkynyl Ru-C bond lengths (2.034(9), 2.057(8) Å), with these
values falling within the range for six-coordinate ruthenium
alkynyls.10 The two mutually trans phosphine ligands are
arranged so as to allow one phenyl group from each to nestle
between the two alkynyl ligands, their planes lying almost
orthogonal to the bis(alkynyl) coordination plane.
The origin of the initially obtained crystals of 25a remains a
mystery, but these could in principle arise from two possible
complications. First, the carbonylation of [RuCl2(PPh3)3] pro-
vides a solvent-dependent mixture of cct-23 and ttt-23.29,30
Heating or recrystallization is presumed to result in complete
conversion to the cct-23 isomer. However, given the overlap
of infrared bands for both isomers, there exists the possibility
that residual amounts of the ttt-23 isomer remained and that
subsequent reaction provided ttt-22, the mutually trans carbonyl
of [RuH(κ2-C6H4PPh2)(CO)(PPh3)2]32 with HCtCPh33 or of
[RuHCl(CO)(PPh3)3] with LiCtCPh.31 The former route avoids
contamination with [Ru(CHdCHPh)Cl(CO)(PPh3)2]34 but is less
convenient for large-scale preparations, due to the requisite
involvement of organomercurials in the synthesis of [RuH(κ2-
C6H4PPh2)(CO)(PPh3)2]. The sample of 26 we obtained via the
latter route has spectroscopic data (δH -7.68 (2JPH ) 86, 25
Hz); IR (Nujol) 2088 (νCtC), 2000 (νRuH), 1938 (νCO)) somewhat
different from those reported (δH ) -8.13 (2JPH ) 85, 25 Hz);
IR (KBr) 2012 (νCtC), 1993 (νRuH), 1923 (νCO)).28,32 However,
with additional 31P{1H} NMR data (δP 45.03 (d), 23.95 (t),2JPP
) 16 Hz) we are confident of the authenticity of our product.
Treating 26 with excess HCtCPh in toluene provides 25a over
48 h, which in turn is cleanly carbonylated (26 °C, CH2Cl2, 1
atm) to provide cct-22 (Scheme 2).
(29) James, B. R.; Markham, L. D.; Hui, B. C.; Rempel, G. L. J. Chem.
Soc., Dalton Trans. 1973, 2247.
(30) Stephenson, T. A.; Wilkinson, G. J. Inorg. Nucl. Chem. 1966, 28,
945.
(31) Wakatsuki, Y.; Yamazaki, H.; Kumegawa, N.; Satoh, T.; Satoh, J.
Y. J. Am. Chem. Soc. 1991, 113, 9604.
(32) Roper, W. R.; Wright, L. J. J. Organomet. Chem. 1982, 234, C5.
(33) Wright, L. J. Ph.D. Thesis, University of Auckland, 1980.
(34) Torres, M. R.; Vegas, A.; Santos, A.; Ros, J. J. Organomet. Chem.
1986, 309, 169.
(36) Cameron, T. S.; Grundy, K. R.; Robertson, K. N. Inorg. Chem. 1982,
21, 4149.
(37) (a) Cameron, C. J.; Fanwick, P. E.; Leeaphon, M.; Waltron, R. A.
Inorg. Chem. 1989, 18, 1101. N.B.: when one considers the curious void
in the coordination sphere of this complex and its coordinative unsaturation
(17 electrons), it is tempting to consider an alternative formulation as the
18-electron hydride complex [ReH(O2CMe)(CO)(PPh3)3]+.
(38) Wakatsuki, Y.; Satoh, M.; Yamazaki, H. Chem. Lett. 1989, 1585.
(35) Elemental microanalytical data reported in ref 31 were erroneously
calculated for C69H56OP3Ru rather than the correct formula C63H51OP3Ru.

1332 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
Scheme 3. Reactions of 1 with Hg(CtCPh)2 (L ) PPh3)
ligands of which might be expected to be more labile. Since
the species ttt-23 (νCO 2005 cm-1) and cct-23 (νCO 2064, 2001
cm-1) have a coincident infrared absorption, the presence of
the former could well have been masked by the latter.
Alternatively, under different conditions, monocarbonylation of
[RuCl2(PPh3)3] in dimethylacetamide (dma) has been reported
to provide [RuCl2(CO)(dma)(PPh3)2], recrystallization of which
from CH2Cl2 and MeOH provides the 16-electron species
[RuCl2(CO)(PPh3)2].29,30 Were this species to be present, then
reaction with LiCtCPh and adventitious phosphine would be
expected to provide 25a. The complex [Ru(CtCPh)2(CO)(Pi-
Pr3)2] is isolable as a stable 16-electron species27 due to the
steric bulk of the PiPr3 ligands, while three of the more sterically
modest PPh3 ligands can be accommodated in 25a.
An alternative approach to the synthesis of 22 was investi-
gated via the oxidative addition of bis(phenylethynyl)mercury
(27)40,41 to 1. Precedent for such a reaction comes from a variety
of sources. (i) Collmann has previously shown that the reaction
of 27 with Vaska’s complex provides [IrCl(CtCPh)(HgCt
CPh)(CO)(PPh3)2].42 (ii) We have implicated similar oxidative
addition processes in the catalytic demercuration of bis(alkynyl)-
mercurials by [RhCl(CO)(PPh3)2]28 and [Rh(PPh3)2([9]aneS3)]-
PF6 ([9]aneS3 ) 1,4,7-trithiacyclononane).43 (iii) Redox trans-
metalation by bis(alkynyl) mercurials has proven to be a versatile
entry into organolanthanide chemistry.44,45 (iv) Roper has shown
that the Hg-C bonds of Hg(CF3)2 and Hg{C(N2)CO2Et}2 add
to 246 and [OsCl(NO)(PPh3)3],47 respectively. (v) The bis-
(tricarbido) complexes [Ru(CtCCtW(CO)2L)2(CO)2(PPh3)2]
(L ) HB(pz)3, HB(pzMe2)3) are obtained from the reaction of
1 with Hg{CtCCtW(CO)2L}2.1
a
Table 3. IR and 31P NMR Data for RuR(HgR)(CO)2(PPh3)2
R
ν
CO, cm-1
δP, ppm
2JPHg, Hz
CF3 (28b)46
2018, 1963
2017,1 971
2027, 1974
2025, 1966
CCPh (28a)
36.4
33.7
34.3
392
382
360
C3W(CO)2Tp (28c)1
C3W(CO)2Tp* (28d)1
a Tp ) HB(pz)3, Tp* ) HB(pzMe2)3.
Treating a solution of 1 in toluene with 1 equiv of Hg(Ct
CPh)2 (27) results in the formation of an essentially colorless
complex formulated as cct-[Ru(CtCPh)(HgCtCPh)(CO)2-
(PPh3)2] (28a) in 67% yield (Scheme 3). Infrared data for 28a
include carbonyl absorptions at 2017 and 1971 cm-1 and a νCC
absorption at 2101 cm-1. The former may be compared with
those for the closely related complex [Ru(CF3)(HgCF3)(CO)2-
(PPh3)2] (28b; IR (Nujol) 2018, 1963 cm-1) (Table 3),46 from
which it may be concluded that the alkynyl is similar in field
strength to the CF3 ligand. A single, albeit broad and unresolved,
carbonyl resonance in the 13C{1H} NMR spectrum suggests
overlap of what should be two unique (triplet) resonances for
the chemically distinct CO ligands. The corresponding reso-
nances in the complex [Ru(CtCR)(HgCtCR)(CO)2(PPh3)2]
(28c, R ) CtW(CO)2{HB(pzMe2)3})1 are observed at δC 200.7
and 201.5, and hence overlap of the resonances in 28a would
appear plausible. The appearance of a single resonance in the
31P{1H} NMR spectrum further confirms the cct-28a geometry,
featuring satellites arising from coupling to mercury (2JHgP(cis)
) 392 Hz). A molecular ion is not observed in the APCI mass
spectrum; however, abundant peaks due to [M - HgCCPh]+
and [M - HgCCPh + MeCN]+ could be identified by isotopic
simulation, suggesting some fragility in the Ru-Hg bond.
The complex 28a slowly extrudes mercury in solution at
ambient temperature to provide almost exclusively the bis-
(alkynyl) complex 22. However, heating 28a in refluxing d6-
benzene results in the deposition of elemental mercury (in
contrast to the case for 28b, which is prepared in refluxing
toluene, 45 min) and the formation of an approximately 1:1
mixture of 4b and 22 (by 31P{1H} NMR) (see Scheme 3). As
noted above, a range of rhodium complexes will catalyze the
demercuration of bis(alkynyl) mercurials to the corresponding
diyne, which is liberated from the various rhodium(I) centers.
Similar reductive elimination processes have been suggested
in the stoichiometric homocoupling of alkynyllithiums by
[NiCl2(PPh3)2].48 For such a process to occur, the most plausible
intermediate, in the case of [RhCl(CO)(PPh3)2] would be the
d6-octahedral cis-bis(alkynyl) complex [Rh(CtCPh)2Cl(CO)-
(PPh3)2], which would be isoelectronic with 22. However,
heating 22 in toluene under reflux does not lead to the reductive
elimination of diphenylbutadiyne and neither does it lead to the
formation of its structural isomer, the butadiyne complex 4b.
Thus, 22 is not an intermediate in the conversion of 28a to 4b.
Furthermore, although heating 4b alone in refluxing toluene does
produce a plethora of ruthenium phosphine complexes (vide
supra), 22 could not be spectroscopically identified among these
(39) Given that dissolution of this complex in dichloromethane leads to
replacement of the initial infrared absorption (1940 cm-1) by a new one at
1970 cm-1 and that addition of methanol precipitates the original complex,
it might be better formulated as [RuCl2(CO)(MeOH)(PPh3)2] or [RuCl2-
(CO)(OH2)(PPh3)2].
(40) Methods of Elemento-Organic Chemistry; Nesmeyanov, A. N.,
Kocheshkov, K. A., Eds.; p 67. North-Holland: Amsterdam, 1967; Vol. 4
(Mercury).
(41) For a discussion of the safety hazards associated with the use of
bis(alkynyl) mercurials, see: Dewhurst, R. D.; Hill, A. F.; Smith, M. K.
Organometallics 2006, 25, 2388.
(42) Collman, J. P.; Kang, J. W. J. Am. Chem. Soc. 1967, 89, 844.
(43) Hill, A. F.; Wilton-Ely, J. D. E. T. Organometallics 1997, 16, 4517.
(44) (a) Deacon, G. B.; Wilkinson, D. L. Inorg. Chim. Acta 1988, 142,
155. (b) Deacon, G. B.; Koplick, A. J. J. Organomet. Chem. 1978, 146,
C43. (c) Deacon, G. B.; Koplick, A. J.; Raverty, W. D.; Vince, D. G. J.
Organomet. Chem. 1979, 182, 121. (d) Deacon, G. B.; Koplick, A. J.; Tuong,
T. D. Aust. J. Chem. 1982, 35, 941. (e) Deacon, G. B.; Newham, R. H.
Aust. J. Chem. 1985, 38, 1757. (f) Deacon, G. B.; Nickel, S.; MacKinnon,
P.; Tiekink, E. R. T. Aust. J. Chem. 1990, 43, 1245.
(45) Lin, G.; McDonald, R.; Takats, J. Organometallics 2000, 19, 1814.
(46) Clark, G. R.; Hoskins, S. V.; Roper, W. R. J. Organomet. Chem.
1982, 234, C9.
(47) Gallop, M. A.; Jones, T. C.; Rickard, C. E. F.; Roper, W. R. J.
Chem. Soc., Chem. Commun. 1984, 1002.
(48) Smith, E. H.; Whittal, J. Organometallics 1994, 13, 5169.

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1333
(31P{1H} NMR). Thus, 4b is not an intermediate in the
conversion of 28a to 22 and an alternative common intermediate
needs to be considered. The possibility that the oxidative
addition of 27 is reversible such that mercury extrusion occurs
remote from metal followed by PhCtCCtCPh recoordination
can be excluded: heating a solution of 27 in toluene under reflux
does not lead to any spectroscopically observable butadiyne or
any visible mercury deposition over a period of 24 h. Neither
is there any indication that mercury extrusion from 27 is
photolytic (24 h, d6-benzene, domestic sunlamp); although a
pale yellow color develops, there is no spectroscopically
observable diyne formation. Furthermore, repeated cycling of
a sample of pure41 27 through its melting point does not lead
to appreciable decomposition or variation in melting point (mp
121-124 °C). Thus, we are convinced that the extrusion of
mercury from 28a is an intramolecular process. These observa-
tions taken together are consistent with two distinct mechanisms
for the extrusion of mercury from 28a. The first involves direct
extrusion from the linear Ru-Hg-C linkage of 28a to provide
22, a process that operates slowly at room temperature. Such
mercury extrusions are implicit in the formation of [Ru(C6H4-
Me-4)Cl(CO)(PPh3)2] from [RuHCl(CO)(PPh3)3] and Hg(C6H4-
Me-4)2 via the putative intermediate [Ru(HgC6H4Me-4)Cl(CO)-
(PPh3)2].49 The second mechanism, which only operates
significantly at elevated temperatures, is suggested to involve
the inner-sphere reductive elimination of Hg(CtCPh)2, which
nevertheless remains bound to the ruthenium center. Two
variants on how the mercurial might bind are as a π adduct
through one CtC bond (A in Scheme 3) or as a σ Lewis acid
adduct through mercury (B). Either of these would involve
geometries that place the alkynyl groups in closer mutual
proximity than in the precursor 28a. In support of the somewhat
curious intermediate involving a RufHg interaction, we note
that (i) Dixneuf has reported the isolation of a stable HgCl2
adduct of a zerovalent iron carbene complex,50 (ii) Lewis base
adducts of bis(alkynyl) mercurials have been previously ob-
served,51 and (iii) the ability of zerovalent ruthenium and
osmium centers to act as Lewis bases has been demonstrated.52,53
We suggest that the trigonal coordination geometry may
predispose the mercury toward reductive elimination of diyne.
We concede that this remains conjecture, with which the
observations are nevertheless consistent.
tions (2018, 1966, and 1619 cm-1); however, the curiously
coincident strong and weak stretches reported at 1928 cm-1 are
absent from our spectrum. The 31P{1H} NMR spectrum of the
redissolved crystals of complex 29 used for the crystallographic
study comprises a single resonance at δP 41.9, while the original
report claims two resonances at δP 36.4 and 33.1 due to an
inseparable mixture of what was claimed to be two regioisomers
of 29 (N.B.: our ruthenacyclopentadiene complex 9 has a single
resonance at δP 36.4). The published 13C data for the “second
isomer” include without explanation two ketonic resonances (δC
175.0, 166.3).
The regioselectivity of the phenylacetylide units in 29 is
different from that which would be obtained were 9 to be an
intermediate in its formation via carbonyl insertion and reductive
elimination, and as noted above, 9 is stable with respect to either
CO insertion or reductive elimination under the conditions of
the formation of 29. Were reductive elimination to occur from
9 (without prior CO insertion), the product would be the cis-
bis(phenylethynyl)cyclobutadiene complex [Ru{η4-C4Ph2(Ct
CPh)2}(CO)x(PPh3)3-x] (“30”, x ) 1, 2). Inseparable isomers
of such a cyclobutadiene complex have been claimed as further
products in the reaction of 2 with PhCtCCtCPh;6 however,
we note that infrared data reported for the complexes “30” (x
) 1) are identical with those we obtained for 9.55
The ruthenium(0) compound 4b does not react with bis-
(trimethylsilyl)butadiyne in refluxing toluene over 3 days,
presumably a reflection of the weak binding of this diyne that
was demonstrated for the complex 4c. Thus, the same spectro-
scopic (31P NMR) fingerprint for the plethora of products is
obtained as when Me3SiCtCCtCSiMe3 was absent.
Finally, returning to the reported role of CO2 in the “activa-
tion” of 2, we considered that adventitious oxygen might have
been a factor. A recently established mechanism for the
conversion of coordinated CO to CO2 involves the initial
conversion of the carbonyl ligand of [OsCl(NO)(CO)(PPh3)2]
to a peroxycarbonyl by reaction with O2.56 When it is heated,
the peroxycarbonyl complex behaves differently, depending on
whether it is in solution or in the solid state. Thus, as a
suspension in refluxing heptane, rearrangement of the peroxy-
carbonyl to a conventional carbonato complex is observed.
However, when the complex is heated in solution in refluxing
benzene in the presence of additional phosphine, CO2 and OPPh3
The behavior of 2 toward diphenylbutadiyne contrasts mark-
edly with that of 1. We find that the cyclopentadienone complex
[Ru{η4-O)CC4Ph2(CtCPh)2}(CO)2(PPh3)] (29) is indeed formed
when 2 and PhCtCCtCPh are heated in refluxing toluene but
that in contrast to previous claims6 (i) CO2 is not required for
the reaction to proceed and (ii) small amounts of 9 (ca. 10-
20%) are also obtained. This parallels our findings for the
reactions of 1 and 2 with diphenylacetylene.3 Although our
measurements of the unit cell of a crystal of 29 agree with those
reported,6 we find some inconsistencies in the spectroscopic
data.55 Our infrared spectrum has three of the reported absorp-
(49) (a) Roper, W. R.; Wright, L. J. J. Organomet. Chem., 1977, 142,
C1. (b) Rickard, C. E. F.; Roper, W. R.; Taylor, G. E.; Waters, J. M.; Wright,
L. J. J. Organomet. Chem. 1990, 389, 375.
(50) Le Bozec, H.; Dixneuf, P. H.; Adams, R. D. Organometallics 1984,
3, 1919.
(51) Cano, E. M.; Santos, M. A.; Ballester, R. L. Anal. Quim. 1977, 73,
1051.
(54) Crossley, I. R.; Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.;
Williams, D. J. Chem. Commun. 2005, 221.
(52) Davis, H. B.; Einstein, F. W. B.; Glavina, P. G.; Jones, T.; Pomeroy,
R. K.; Rushman, P. Organometallics 1989, 8, 1030.
(53) (a) Hill, A. F.; Owen, G. R.; White, A. J. P.; Williams, D. J. Angew.
Chem., Int. Ed. 1999, 38, 2759. (b) Foreman, M. R. St.-J.; Hill, A. F.; Owen,
G. R.; White, A. J. P.; Williams, D. J. Organometallics 2003, 22, 4446. (c)
Foreman, M. R. St.-J.; Hill, A. F.; White, A. J. P.; Williams, D. J.
Organometallics 2004, 23, 913.
(55) No explanation was provided for why two different isomers of a
monocarbonyl complex give rise to two νCO IR absorptions (reported, 2011.5
vs, 1955.6 cm-1; cf. 2012, 1956 cm-1 for 9). The [M - CO]+ base peak
reported in the FAB-MS spectrum thus corresponds to [M - 2CO]+.
(56) Clark, G. R.; Laing, K. R.; Roper, W. R.; Wright, A. H. Inorg.
Chim. Acta, 2004, 357, 1767.

1334 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
are liberated to provide [OsCl(NO)(PPh3)3], thereby converting
the substitution-inert complex [OsCl(NO)(CO)(PPh3)2] into the
labile [OsCl(NO)(PPh3)3].56,57 This pair of complexes in some
ways reflect the properties of substitution inert 2 and labile 1,
respectively (the unique properties of nitrosyl ligands notwith-
standing). Furthermore, Caulton has also proposed a peroxy-
carbonyl complex as an intermediate in the oxidation of
[Ru(CO)2(PEt3)3] to provide [Ru(κ2-O2CO)(CO)(PEt3)3].58 We
therefore considered whether oxygen might be serving the same
purpose in activating 2.
to remove carbonyl ligands from ruthenium carbonyl clusters
through oxidative conversion to CO2 (e.g., in the synthesis of
1114). The reaction of 2 with PhCtCCtCPh under ambient
conditions was next investigated and found to proceed extremely
slowly at room temperature. Over a period of 7 days ca. 16%
conversion to 4b was observed in addition to 2-3% of an
unidentified compound with δP 40.7. Gentle photolysis of the
same mixture (domestic sunlamp) resulted in a complete
consumption of 2 and formation of a 1:1 mixture of 4b and the
unknown compound. Remarkably, when 2 and PhCtCCtCPh
were combined in d6-benzene under an atmosphere of oxygen,
there was a dramatic acceleration in the consumption of 2 such
that after 6 days phosphine oxide accounted for 90% of
phosphorus-containing material, with a further 6% being due
to the dioxygen adduct 21. We have not characterized the
phosphine-free ruthenium-containing products; the ruthenium
cluster chemistry of 1,3-diynes is complex14,17k,64 and is beyond
the scope of this discussion but has been reviewed recently.9
What needs to be accounted for is the synergic effect of these
two reagents in stripping phosphine from a complex that is
generally considered to be inert toward ligand dissociation. We
have attempted to rationalize these observations by considering
the mechanistic manifold presented in Scheme 1. This assumes
that distinct equilibria exist between 2 and the coordinatively
unsaturated species “Ru(CO)3(PPh3)” and, to a lesser extent,
“Ru(CO)2(PPh3)2”. The latter would explain the, albeit exceed-
ingly slow, conversion of 2 to 4b in the presence of diyne and
absence of oxygen. The former species, which is more prevalent
at higher temperatures, would account for the formation of the
cyclopentadienone complexes 5 and 29, in reactions that parallel
the observations of Takats and Caulton concerning the lability
of complexes of the form [M(η-alkyne)(CO)3(L)] (M ) Fe, Ru,
Os; alkyne ) HCtCH, F3CCtCCF3; L ) CO, PMe3, PPh3,
PCy3, P(OPh)3).65 The following borrows heavily from their
conclusions; however, the principles are worth restating. A key
inference from these studies is that stabilizing “four-electron
donor” alkyne behavior, which underpins the wealth of d4-
octahedral alkyne chemistry,66 can destabilize complexes with
higher d occupancies. Thus for dn (n g 6), all “t2g-type” orbitals
are fully occupied, resulting in a four-electron conflict with any
potentially π basic ligands. An implication of this is that
“unwelcome” four-electron donation by an alkyne in such
systems will be ameliorated by π acidic coligands, able to
withdraw excessive electron density: i.e., phosphines become
labilized while CO ligands become more tightly bound.
As noted above, Collman had long since reported that the
solid-state reaction of 2 with air provided the carbonato complex
8 with infrared absorptions at 2045, 1950, 1890, 1675, and 1635
cm-1.8 Infrared data subsequently reported for 8 arising from
the reaction of 2 with alkynes and CO2 only displayed
absorptions at 2046, 1982, and 1626 cm-1,6 clearly at variance
with the earlier report. The hydrolysis of [Ru(O2CNiPr2)2(CO)2-
(PPh3)2] has since been shown to provide 8, which was
crystallographically characterized as a monohydrate with IR
absorptions at 2046, 1982, 1651, and 1626 cm-1 59
Thus, the
.
product obtained from the reaction of 2, PhCtCCtCPh, and
CO2 would indeed appear to be the authentic carbonato complex.
However, in apparent contrast to this, Calderazzo has shown59
that 2 does not react with CO2 in the presence of HNiPr2 (5
atm, 60 °C, 10 h). Furthermore, hydrolysis of [Ru(O2CNiPr2)2-
(CO)2(PPh3)2] in the presence of CO results in reduction to 2
via 8.59 Thus, the implication is that Collman’s solid-state
preparation also included a second component (νCO 1950, 1890,
1675 cm-1) that may well correspond to the peroxycarbonyl
complex [Ru(κ2-OOCO)(CO)2(PPh3)2]. The ketonic νCO IR
absorption would not appear to be sufficiently diagnostic to
distinguish such possibilities, given that the conversion of [OsCl-
(κ2-OOCO)(NO)(PPh3)2] (νCO 1710 cm-1) to the carbonato
complex [OsCl(κ2-O2CO)(NO)(PPh3)2] (νCO 1705 cm-1) is only
attended by a very modest shift in frequency.56
We have therefore investigated the reaction of 2 with oxygen
by in situ 31P NMR spectroscopy. At room temperature in d6-
benzene solution under 1 atm of oxygen an exceedingly slow
reaction ensues, whereby after 15 days a mixture of 2 (δP 56.3),
the dioxygen adduct 21 (δP 35.2), and phosphine oxide (δP 24.9)
is obtained in a ratio of 10:3:2. During this period a minor
resonance at δP 27.5 grows and decays, reaching a maximum
after 2 days but representing no more than 4-5% of phosphorus
present. A resonance due to free PPh3 also grows and decays
at a similar rate. This may well correspond to the carbonato
complex 8,60 in which case it may be assumed that liberated
phosphine is capable of reducing this to “Ru(CO)2(PPh3)2”,
OPPh3, and CO2 (cf. ref 56), the first species being rapidly
trapped as 21. Carbon dioxide is a poor ligand for zerovalent
ruthenium,59-63 as evidenced by the widespread use of Me3NO
We may assume that the initial stages in the disparate
reactions of 1 and 2 with alkynes involve the formation of the
complexes [Ru(η2-RCtCR)(CO)2(PPh3)2] (R * H67) (4) and
[Ru(η2-RCtCR)(CO)3(PPh3)], respectively, and that in each
case incipient π donation will destabilize phosphine coordination
relative to the carbonyl ligands. Thus, phosphine dissociation
(57) Hill, A. F.; Roper, W. R.; Waters, J. M.; Wright, A. H. J. Am. Chem.
Soc. 1983, 105, 5939.
(58) Ogasawara, M.; Maseras, F.; Gallego-Planas, N.; Kawamura, K.;
Ito, K.; Toyota, K.; Streib, W. E.; Komiya, S.; Eisenstein, O.; Caulton, K.
G. Organometallics 1997, 16, 1979.
(64) (a) Aime, S.; Betancello, R.; Busetti, V.; Gobetto, R.; Granozzi,
G.; Osella, D. Inorg. Chem. 1986, 25, 4004. (b) Rivomanana, S.; Lavigne,
G.; Lugan, N.; Bonnet. J.-J. Inorg. Chem. 1991, 30, 4110. (c) Rivomanana,
S.; Lavigne, G.; Lugan, N;, Bonnet. J.-J. Organometallics 1991, 10, 2285.
(d) Rivomanana, S.; Lavigne, G.; Lugan, N;, Bonnet. J.-J.; Yanez, R.;
Mathieu, R. J. Am. Chem. Soc. 1989, 111, 8959. (e) Deeming, A. J.; Felix,
M. S. B.; Bates, P. A.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans.
1987, 461.
(59) Dell’Amico, D. B.; F. Calderazzo, F.; Labella, L.; Marchetti, F. J.
Organomet. Chem. 2000, 596, 144.
(60) Complex 8 has been reported59 to have a single 31P resonance at δP
28.0 in methanol.
(61) CS2, however, reacts with 1 to provide the simple π adduct [Ru-
(SCS)(CO)2(PPh3)2],62 and a similar adduct [Ru(SCO)(CO)2(PPh3)2] is
obtained with COS; however, excess COS converts this to the dithiocar-
bonato-S,S′ complex [Ru(S2CO)(CO)2(PPh3)2].63
(62) (a) Grundy, K. R.; Harris, R. O.; Roper, W. R. J. Organomet. Chem.
1975, 90, C34. (b) Clark, G. R.; Collins, T. J.; James, S. M.; Roper, W. R.
J. Organomet. Chem. 1977, 125, C23. (c) Boniface, S. M.; Clark, G. R. J.
Organomet. Chem. 1980, 184, 125.
(65) (a) Pearson, J.; Cooke, J.; Takats, J.; Jordan, R. B. J. Am. Chem.
Soc. 1998, 120, 1434. (b) Mao, T.; Zhang, Z.; Washington, J.; Takats, J.;
Jordan, R. B. Organometallics 1999, 18, 2331. (c) Gagne´, M. R.; Takats,
J. Organometallics 1988, 7, 6850. (d) Burn, M. J.; Kiel, G.-Y.; Seils, F.;
Takats, J.; Washington, J. J. Am. Chem. Soc. 1989, 111, 6850. (e) Marinelli,
G.; Streib, W. E.; Huffman, J. C.; Caulton, K. G.; Gagne´, M. R.; Takats,
J.; Dartiguenave, M.; Chardon, C.; Jackson, S. A.; Eisenstein, O. Polyhedron
1990, 9, 1867-1881.
(63) Gaffney, T. R.; Ibers, J. A. Inorg. Chem. 1982, 21, 2851.
(66) Templeton, J. L. AdV. Organomet. Chem. 1989, 29, 1.

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1335
from either intermediate and coordination of a second alkyne
would provide [Ru(RCtCR)2(CO)2(PPh3)] and [Ru(RCtCR)2-
(CO)3], respectively. In general, the presence of strongly π acidic
coligands promotes migratory insertion, such that the latter
would be more prone than the former to migratory insertion of
carbonyl and alkyne ligands. Precedent for this type of coupling
and its encouragement by π acid coordination is provided by
the isolation of the osmacyclobutenethione [Os{κ2-C(dS)CPhd
CPh}(CO)2(PPh3)2] upon carbonylation of the alkyne complex
[Os(η-PhCtCPh)(CS)(PPh3)2].68 The former ([Ru(η-RCtCR)2-
(CO)2(PPh3)]), in contrast, enters into alkyne coupling to form
a ruthenacyclopenta-1,3,5-triene that may be stabilized by
readdition of phosphine to relocalize the metallacycle bonding
to a ruthenacyclopenta-2,4-diene. The possibility that the
R-alkynyl substituent might also serve this role is suggested
(vide infra), supported by the observation of such an interaction
in the zirconacyclopentadiene 1919b and the demonstrated
hemilability of R-alkynyl-substituted vinyl ligands.69
The regioselectivity observed in formation of the R,R′-bis-
(alkynyl) isomer of the ruthenacyclopentadiene 9 and the
alternative regiochemistry of the 3,4-bis(ethynyl) isomer of the
cyclopentadienone complex 29 may be rationalized as follows.
The first step in the construction of the cyclopentadienone most
likely involves carbonyl/alkyne coupling. This may be viewed
as a migratory insertion of one of the Ru-Cδ- bonds of a
“metallacyclopropene” type of alkyne onto one coordinated
δ+CO. In this case of the two available Ru-C carbons, that
bearing the phenyl substituent will be more electron-rich than
that with the electron-withdrawing alkynyl substituents, such
that the latter remains remote from the carbonyl. The insertion
of the second alkyne into the resulting vinyl ligand simply
follows the steric distinction between Ph and CCPh substituents.
The origins of the regiochemistry associated with 9 are less clear.
However, one possibility is that the alkynyl groups are not
entirely innocent but perhaps coordinate weakly to the ruthenium
center during the approach to the C-C bond-forming transition
state, thereby lowering the energy associated with development
of coordinative unsaturationsa role not as effectively served
by the phenyl substituents. We have previously discussed the
operation of hemilabile R-alkynyl coordination in the chemistry
of the cationic complex [Ru{η3-C(CtCPh))CHPh}(CO)2-
(PPh3)2]+ obtained upon protonation of 4b.2,69
or existence of the complexes previously formulated as cyclo-
butadiene adducts 1,2- and 1,3-[Ru{η4-C4Ph2(CtCPh)2}(CO)-
(PPh3)2], noting that some of the published spectroscopic data6
correspond to the ruthenacyclopentadiene complex 9. While 9
might appear to be a plausible intermediate en route to the
proposed cyclobutadiene complexes, this transformation does
not ensue under the reported reaction conditions.
Experimental Section
General Procedures. All air-sensitive manipulations were
carried out under a dry, oxygen-free nitrogen atmosphere using
standard Schlenk and vacuum manifold techniques or in an argon
filled drybox, using dried and degassed solvents. NMR spectra were
recorded on a Varian Inova 300 (1H at 300.75 MHz, 13C at 75.4
MHz, 31P at 121.4 MHz) instrument. The chemical shifts (δ) for
1H and 13C{1H} spectra are given in ppm relative to residual signals
of the solvent and for 31P{1H} spectra relative to an external 85%
H3PO4 reference. The coupling constants (J) are given in Hz with
an estimated error of (0.5 Hz. Virtual triplet 13C resonances for
the trans-Ru(PPh3)2 phenyl groups are indicated by “vt” with
apparent JPC couplings given. Mass spectra of the complexes were
obtained on a Micromass ZMD spectrometer using the APCI
technique in acetonitrile by the Mass Spectrometry service of the
Research School of Chemistry, Australian National University. The
microanalyses were carried out by the microanalytical service of
the Research School of Chemistry, Australian National University.
The synthesis of compounds [Ru(CO)2(PPh3)3] (1),4,70 [Ru(CO)3-
(PPh3)2] (2),71 [Ru(O2)(CO)2(PPh3)2] (21),4 [Ru(η2-PhCtCCtCPh)-
(CO)2(PPh3)2] (4b),2 [Ru(C2H4)(CO)2(PPh3)2] (20),4 [RuH(Ct
40
CPh)(CO)(PPh3)3],31 and Hg(CtCPh)2 have been described
previously, while diynes were obtained from commercial sources.
Alkynyl mercurials should be treated as potentially explosive and
appropriate precautions taken.41 N.B.: chlorinated solvents such
as CHCl3 are inappropriate for investigations involving 1 (δP
(CDCl3) 57.70), due to the rapid formation of [RuCl2(CO)2(PPh3)2]
(δP (CDCl3) 17.57). The reaction with CH2Cl2 is somewhat slower,
allowing some rapid reactions to be carried out in that solvent.
The structures of the four compounds 4b, 9, 22, and 25a each
showed interesting crystallographic features. The constrained least-
squares refinement program RAELS200072 was used for the
refinement of all four structures, as it allowed sensitive control of
the refinements using appropriate parametrizations. Extra details
of these refinements that included rigid-body TLX parametrizations
and the use of refinable local coordinates relative to refinable local
orthonormal axial systems73 may be obtained from the CIF files in
the Supporting Information.
Conclusions
Synthesis of [Ru(κ2-CRdCPhCPhdCR)(CO)2(PPh3)2] (9, R
) CCPh). A mixture of [Ru(CO)2(PPh3)3] (1; 0.40 g, 0.42 mmol)
and diphenylbutadiyne (0.26 g, 1.27 mmol) was heated anaerobi-
cally under reflux in toluene (80 mL), resulting in a color change
to dark orange. The reaction progress was followed by infrared
and 31P{1H} NMR spectroscopy, and after 3 days at reflux, no
further product was formed. The toluene was removed under
reduced pressure, and the orange-brown residue was washed with
diethyl ether. Recrystallization from toluene/hexane or THF/diethyl
ether yielded orange crystals (0.36 g, 79%). Anal. Found: C, 76.92;
H, 4.81; P, 5.65. Calcd for C70H50O2P2Ru: C, 77.41; H, 4.64; P,
The product of the thermal reaction of 1 with excess
diphenylbutadiyne is a ruthenacyclopentadiene, while the same
reaction with diphenylacetylene yields a completely unrelated
product involving a coordinated indenone ligand. In contrast,
the reactions of 2 with both alkynes proceed via [2 + 2 + 1]
cycloaddition to give coordinated cyclopentadienones. The
disparate reactivity on the part of complexes 1 and 2, which
differ only by replacement of one PPh3 with CO, has been
interpreted in terms of the relative lability of CO and PPh3. Thus,
both complexes appear to dissociate a PPh3 ligand (1 spontane-
ously and 2 on heating) to generate “Ru(CO)2(PPh3)2” and “Ru-
(CO)3(PPh3)” in solution, respectively, the distinct electronic
properties of which impinge on the reactivity of alkynes that
subsequently coordinate. We find no evidence for the formation
1
5.70. Mp: 210-212 °C. H NMR (298 K, 300 MHz) C6D6: δ
7.97-7.91 (m, 12 H), 7.30, 7.27 (s × 2, 4 H), 7.09-6.90 (m, 30
H), 6.68, 6.66 (s × 2, 4 H). 13C{1H} NMR (298 K, 75 MHz): in
(70) Hill, A. F.; Tocher, D. A.; White, A. J. P.; Williams, D. J.; Wilton-
Ely, J. D. E. T. Organometallics 2005, 24, 5342.
(71) Collman, J. P.; Roper, W. R. J. Am. Chem. Soc. 1965, 87, 4008.
(72) (a) Rae, A. D. RAELS2000; Australian National University,
Canberra, Australia, 2000. (b) Rae, A. D.; Edwards, A. J. Proceedings,
20th European Crystallography Conference; Cracow, Poland, 2001; p 147.
(73) Rae, A. D. Acta Crystallogr., Sect. A 1975, A31, 570.
(67) In the case of terminal alkynes, oxidative addition of the alkyne
occurs to provide [RuH(CtCR)(CO)2(PPh3)2]: Bartlett, M. J.; Hill, A. F.;
Smith, M. K. Organometallics 2005, 24, 5795.
(68) Elliott, G. P.; Roper, W. R. J. Organomet. Chem. 1983, 250, C5.
(69) Hill, A. F.; Melling, R. P.; Thompsett, A. R. J. Organomet. Chem.,
1991, 402, C8.

1336 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
C6D6, δC 200.1 (br, CO), 164.6 (RuCdC), 142.3 (RuCdC), 134.8
(vt, JCP ) 5.5, C2,6(PC6H5)), 134.2 (vt, JPC ) 23.2, C1(PC6H5)),
129.9 (C4(PC6H5)), 131.5, 130.6, 129.3 (C2,3,5,6(CC6H5)), remaining
C6H5 peaks obscured by C6D6, 103.7, 103.0 (CtC); in CD2Cl2, δC
199.8 (br, CO), 164.1 (RuCdC), 142.3 (RuCdC)), 134.7 (vt, JPC
) 5.5), 133.9 (vt, JPC ) 23.1, C1(PC6H5)), 128.0 (vt, JPC ) 4.8,
C3,5(PC6H5)), 130.0 (C4(PC6H5)), 131.1, 130.1, 128.3, 126.3
(C2,3,5,6(CC6H5)), 126.5, 125.1 (C4(CC6H5)), 126.8 (C1(CC6H5)),
103.5, 102.6 (CtC). N.B.: free PhCtCCtCPh has δC 73.9 and
81.5 for CtC. 31P{1H} NMR (298 K, 121 MHz): in C6D6, δP 36.4
(s); in CD2Cl2, δP 36.2. IR: in Nujol, 2144 (CtC), 2012, 1956
(CO), 1592 (CdC) cm-1; in toluene, 2151 (CtC), 2017, 1960
(CO), 1594 (CdC) cm-1. MS (APCI+): 1124.2 [M + Na + O],
1096 [M + Na + O - CO], 834 [M - PPh3 - CO + Na + O].
Crystal data for 9: C70H50O2P2Ru, Mw ) 1086.11, monoclinic,
P21/a (No. 14), a ) 11.8422(3) Å, b ) 44.1243(9) Å, c ) 21.2472-
(6) Å, â ) 91.373(1)°, V ) 11099.1(5) Å,3 Z ) 8, Fcalcd ) 1.300
Mg m-3, T ) 200(2) K, 10 359 independent measured reflections,
R1 ) 0.075, wR2 ) 0.106, 3447 absorption-corrected reflections
(I > 3σ(I), 2θ e 40°), 380 parameters. Compound 9 crystallized
in space group P21/a as a very thin needle and was 0.521(5):0.479
twinned. Only 34% of the Mo KR reflection data to 2θ ) 40° could
be reliably measured, requiring sensible constraints to simulate an
anisotropic refinement. The structure can be described as displacive
modulation of an idealized Pcab (space group No. 61) structure
with 8 molecules per unit cell. The two Ru atoms in the asymmetric
Ru)0.894(C47H36Cl4O2P2Ru)0.106, Mw ) 889.56, monoclinic, P21/n
(No. 14), a ) 10.2783(1) Å, b ) 18.7156(2) Å, c ) 22.5047(3) Å,
â ) 97.566(1)°, V ) 4291.42(8) Å,3 Z ) 4, Fcalcd ) 1.377 Mg
m-3, T ) 200(2) K, 7555 independent measured reflections, R1 )
0.041, wR2 ) 0.061, 5330 absorption-corrected reflections (I >
3σ(I), 2θ e 46°), 206 parameters. Compound 22 crystallized in
space group P21/n as a thin needle and was twinned. The twin plane
involves an a glide perpendicular to c* at z ) 1/4 and the twin ratio
refined to 0.805(3):0.195. The structure contained 0.894 of the
desired product and 0.106 of an impurity in which one ligand (atoms
C3-C10) was replaced by a coordinated Cl and a chloroform of
crystallization (24a‚CHCl3). The twin was modeled using intensities
obtained assuming there was only one twin component and adjusting
2
the model for intensities to be Y(h) ) ∑iaipj|F(hi)| where ai is the
amount of the ith twin component and pj is the fraction of the second
twin component included in the collected intensity.72 The modeling
of the twinned reflection data was as for compound 9, and values
of pj as a function of the h reflection index are detailed under
_refine_special_details in the CIF file. The overall refinement used
206 variables to define the twin overlap and 64 non-H atom
positions (CCDC 622359).
Synthesis of [Ru(CtCPh)2(CO)(PPh3)3] (25a). The compound
[RuH(CtCPh)(CO)(PPh3)3] (26) was prepared as described in ref
31, with minor variations in the spectroscopic data being ob-
served: IR (Nujol): 2088 (CtC), 2000 (Ru-H), 1938 (M-CO)
2
cm-1
.
31P{1H} NMR (298 K, 121.5 MHz): 45.03 (d, JPP ) 16,
1
2
1
unit are related by the pseudo c glide 0.38 - x, y, /2 + z. For
axial P), 23.95 (t, JPP ) 16, equatorial P). H NMR (298 K, 300
Pcab this operation would be 1/2 - x, y, 1/2 + z. Synthetic precession
patterns obtained from the CCD diffraction images showed that
the twin plane was normal to c* and may be described as a b glide
perpendicular to c* at z ) 1/2. Twin related reflections were
separated along c* in reciprocal space, and pj is a function of the
h reflection index. Refined values are detailed under _refine_
special_details in the CIF file. The overall refinement used 380
variables to define the twin overlap and 150 non-H atom positions
(CCDC 622358).
2
2
MHz): -7.68 (dt, JHP ) 86, JHP ) 25). Complex 26 (2.30 g,
2.26 mmol) was weighed into a large Schlenk flask equipped with
a magnetic stirrer bar. Toluene (200 mL) was added, followed by
ethynylbenzene (2.0 mL, 1.80 g, 18 mmol). The mixture was stirred
at room temperature, and after 2 days the infrared and 31P{1H}
NMR spectra indicated that the reaction was complete. Yield 2.20
g (87%). IR (C6H6): 2109 (CtC), 1972 (CO) cm-1 31P{1H} NMR
.
(298 K, 121 MHz, C6D6): δ 30.16 (d, 2JPP ) 22, axial P), 19.70 (t,
2JPP ) 22 Hz, equatorial P). Crystal data for 25a‚CH2Cl2: C72H57-
Cl2OP3Ru, Mw ) 1203.14, monoclinic, C2/c, a ) 22.9151(2) Å, b
) 12.9560(1) Å, c ) 40.1644(6) Å, â ) 103.6561(4)°, V )
11587.2(2) Å,3 Z ) 8, Fcalcd ) 1.379 Mg m-3, T ) 200(2) K, yellow
plate, 8056 independent measured reflections, R1 ) 0.051, wR2
) 0.061, 3318 absorption-corrected reflections (I > 3σ(I), 2θ e
46°), 204 parameters. The crystal used for compound 25a was small
and of poor quality. Sensible constraints were used to obtain an
anisotropic atom refinement using 204 variables to define 82
independent non-H atom positions in the C2/c structure, which
contained a 0.556(11):0.444 disordered CH2Cl2 solvent molecule
(CCDC 622360).
Synthesis of [Ru(CtCPh)2(CO)2(PPh3)2] (22). Lithium
phenylacetylide was prepared by adding nBuLi (1.60 mL, 1.6
moldm-3, 2.66 mmol) to a THF (20 mL) solution of HCtCPh (0.29
mL, 0.27 g, 2.66 mmol) at -78 °C and warming the solution room
temperature with stirring. The cct isomer of [RuCl2(CO)2(PPh3)2]
(23; 0.50 g, 0.66 mmol) (prepared by bubbling CO through a
solution of [RuCl2(PPh3)3] and then warming in dichloromethane)
was weighed into a separate Schlenk flask equipped with a magnetic
stirrer bar and suspended in THF (50 mL). A dark red lithium
acetylide solution was added, and the mixture was heated to reflux.
After 30 min at reflux, the 31P{1H} NMR and infrared spectra
indicated complete conversion to the product. The solvent was
removed under reduced pressure, and the product was extracted
into dichloromethane. Cooling resulted in the precipitation of a small
amount of LiCl, which was removed by filtration through diato-
maceous earth. Hexane was layered onto the solution, leading to
the precipitation of the product as a pale yellow powder. Yield:
0.28 g (48%). Mp: 218-220 °C. Anal. Found: C, 72.11; H, 4.63;
P, 6.23. Calcd for C54H40O2P2Ru·H2O (by IR and 1H NMR
Synthesis of [Ru(CtCPh)(HgCtCPh)(CO)2(PPh3)2] (28a).
[Ru(CO)2(PPh3)3] (1; 0.22 g, 0.23 mmol) and Hg(CtCPh)2 (0.10
g, 0.25 mmol) were weighed together into a Schlenk flask equipped
with a magnetic stirrer bar. Toluene (30 mL) was added via cannula,
and the bright yellow slurry was stirred at room temperature. After
15 min the reagents had dissolved and the color had changed to
pale yellow. The 31P{1H} NMR spectrum of the reaction mixture
indicated that all of the starting material had been consumed. The
solution was filtered through diatomaceous earth, the volume was
reduced under dynamic vacuum to ca. 5 mL, and the product
crystallized at -20 °C as a fine white powder, which was
recrystallized from benzene as a hemisolvate. Yield: 0.16 g (67%).
Anal. Found: C, 61.2; H, 3.95; P, 5.61. Calcd for C54H40HgO2P2-
Ru‚0.5C6H6: C, 60.9; H, 3.86; P, 5.51 (C6H6 by NMR). Mp: 110-
112 °C dec. 1H NMR (298 K, 300 MHz, C6D6): δ 8.22-8.16 (m,
12 H), 7.57-7.54 (dd, 2 H), 7.28-7.25 (dd, 2 H), 7.07-6.91 (m,
24 H). 13C{1H} NMR (C6D6, 298 K, 75 MHz): δ 201.2, 200.8
(br, CO), 138.0 (vt, JCP ) 23, C1(PC6H5)), 134.3, 134.0, 132.3,
127.2, 126.9, 126.8, 126.6, 125.6 (CC6H5; some uncertainty in
1
integration): C, 71.91; H, 4.69; P, 6.87. H NMR (298 K, 300
MHz): in CDCl3, δ 8.10-8.07 (m, 12 H, PC6H5), 7.35-7.33 (m,
18 H, PC6H5), 7.02-6.99 (m, 6 H, CC6H5), 6.72-6.69 (m, 4 H,
CC6H5). 13C{1H} NMR (298 K, 75 MHz): in CDCl3, δ 134.8 (vt,
1JCP ) 24, C1(PC6H5)), 134.7 (C1(CC6H5)), 134.2 (vt, JCP ) 5.3,
C2,6(PC6H5)), 130.3, 127.5. 124.5 (C2-6(CC6H5)), 129.80 (C4-
(PC6H5)), 127.9 (vt, JCP ) 4.9, C3,5(PC6H5)), 114.4 (CtC), 107.6
(CtC); RuCO resonance not located. 31P{1H} NMR (298 K, 121
MHz): in C6D6, δP 27.85; in CDCl3, δP 28.18. IR (THF): 2108
(CtC), 2045, 1996 (CO) cm-1. MS (APCI+): 920 [M + 2H2O],
906 [M + Na]. Crystal data for “22” (N.B. the structure was
modeled as comprising ca. 10% of 24a‚CHCl3): (C54H40O2P2-
assignments due to overlap of C6D6 absorption), 133.69 (vt, JCP
)

Bis(alkynyl) Complexes of Ruthenium
Organometallics, Vol. 26, No. 6, 2007 1337
5.8, C2,6(PC6H5)), 130.2 (C4(PC6H5)), 128.7 (vt, JPC ) 4.9,
C3,5(PC6H5)] 120.4, 120.0, 108.9, 103.3 (CtC × 4). 31P{1H} NMR
2
(298 K, 121 MHz, C6D6): δ 36.4 (s + d, JPHg ) 392 Hz). IR
(Nujol): 2101 (CtC), 2017, 1971 (CO) cm-1. MS (APCI+): 824
[M + MeCN - HgCtCPh], 783 [M - HgCtCPh].
Thermolysis of [Ru(CtCPh)(HgCtCPh)(CO)2(PPh3)2] (28a).
A sample of 28a (20 mg) in C6D6 was heated to reflux; after 10
min at reflux, the compounds 4b, 28a, and 22 were observed by
31P{1H} NMR spectroscopy in a 1:5:1 ratio; the infrared spectrum
was consistent with the presence of this mixture. Further reflux
(30 min total) resulted in visible deposition of elemental mercury
and the complete disappearance of the resonance for [Ru(CtCPh)-
(HgCtCPh)(CO)2(PPh3)2]; a 1:1 mixture of 4b and 22 was
observed along with several minor unidentified resonances in the
range δ 41-32.
Figure 5. Molecular geometry of 21 in the crystal (50% displace-
ment ellipsoids, phenyl groups omitted) Selected bond lengths (Å)
and angles (deg): Ru1-P1 ) 2.3707(6), Ru1-P2 ) 2.3797(7),
Ru1-O1 ) 2.0281(17), Ru1-O2 ) 2.0360(17), Ru1-C3 ) 1.890-
(3), Ru1-C4 ) 1.886(2), O1-O2 ) 1.462(3), O3-C3 ) 1.145-
(3), O4-C4 ) 1.142(3); O1-Ru1-O2 ) 42.16(8), O1-Ru1-C3
) 153.88(10), O2-Ru1-C3 ) 111.99(10), O1-Ru1-C4 )
112.16(9), O2-Ru1-C4 ) 153.97(9), C3-Ru1-C4 ) 93.89(10),
Ru1-O1-O2 ) 69.21(10), O1-O2-Ru1 ) 68.63(9), Ru1-C3-
O3 ) 178.0(2), Ru1-C4-O4 ) 177.4(2).
Synthesis of [Ru{η4-OdCC4Ph2(CtCPh)2}(CO)2(PPh3)] (29).
Pale yellow 2 (0.30 g, 0.42 mmol) and diphenylbutadiyne (0.26 g,
1.27 mmol) were weighed into a small Schlenk flask equipped with
a magnetic stirrer bar. Toluene (50 mL) was added, and the solution
was stirred and heated at reflux for 12 h, resulting in a color change
from pale yellow to orange-brown and the appearance in the infrared
spectrum of metal carbonyl bands at 2021 and 1969 cm-1 (cf. 2018,
1967 cm-1 6); the 31P{1H} NMR spectrum of the crude mixture
contained resonances at δ 41.8 (29; approximate integral 1), 39.8
(0.3), 36.4 (9; 0.4), and -4.7 (1; free PPh3). The reaction mixture
was filtered to remove a small amount of pale solid, and the solvent
was removed under reduced pressure. Diethyl ether was added to
the flask to wash the insoluble product, which was isolated by
filtration as an orange powder. The supernatant diethyl ether
contained some 9, identified by 31P{1H} NMR and infrared
spectroscopy. The unit cell of 29 was determined crystallographi-
cally and was the same as that reported (reported values6 in
parentheses): a ) 12.65 (12.66) Å, b ) 13.35 (13.30) Å, c ) 24.51
(24.55) Å, â ) 102.1 (101.9)°. Yield: 0.23 g, 64%. Anal. Found:
C, 74.80; H, 4.31; P, 3.86. Calcd for C53H35O3PRu: C, 74.72; H,
of 20 and 4c. Heating this mixture under reflux (sealed tube) for
12 h provided a mixture comprising 20 (125), 2 (75), 1 (21), 4c
(35), and various minor peaks.
(c) A mixture of the ethylene complex Ru(C2H4)(CO)2(PPh3)2
(20; 30 mg, 0.042 mmol, δP (C6D6) 58.03) and Me3SiCtCCt
CSiMe3 (8 mg, 0.05 mmol) in d6-benzene was stirred for 10 min.
The sample was then degassed by two freeze/thaw cycles in vacuo
and allowed to reach equilibrium by stirring for 10 min. The cycle
was repeated, providing a solution comprising a 1:4 mixture of 20
and 4c.
(d) Nitrogen was gently bubbled through a solution of the
ethylene complex Ru(C2H4)(CO)2(PPh3)2 (20; 30 mg, 0.042 mmol,
δP (C6D6) 58.03) and Me3SiCtCCtCSiMe3 (8 mg, 0.05 mmol)
in d6-benzene for 20 min to provide a 1:20 mixture of 20 and 4c.
The solvent was removed, and limited spectroscopic data were
acquired for the resulting pale yellow powder, which also contained
1
small amounts of Me3SiCCCCSiMe3 (δH (C6D6) 0.03). H NMR
1
4.14; P, 3.64. Mp: 195-198 °C. H NMR (298 K, 300 MHz,
(298 K, 300 MHz, C6D6): -0.04, 0.24 (s × 2, 9H × 2, SiCH3),
7.00-7.16 (m, 18H, H3-5(C6H5)), 7.77-7.85 (m, 12 H, H2,6(C6H5)).
13C{1H} NMR (298, 75 MHz, only limited data due to solution
instability, C6D6): δC 134.8 (vt, JPC ) 5.7, C2,6(PC6H5)), 134.6 (vt,
JPC ) 5.6, C3,5(PC6H5)), 132.1 (vt, C1(PC6H5)), 130.4 (C4(PC6H5)),
0.79, -0.69 (SiCH3). IR: in Nujol, 2116 (CtC), 1976, 1906 (CO)
cm-1; in C6D6, 2067 (CtC), 1974, 1911 (CO) cm-1. Satisfactory
elemental microanalytical data were not obtained.
Crystal Structure Determination of [Ru(O2)(CO)2(PPh3)2]
(21). This complex was prepared according to the method of Roper.4
The room-temperature crystal structure has been previously re-
ported;20 however, we have acquired a higher quality data set at
200 K leading to a more precise refinement model (Figure 5).
Crystal data for 21: C38H30O4P2Ru, Mw ) 713.67, triclinic, P1h (No.
2), a ) 9.7315(1) Å, b ) 9.8696(2) Å, c ) 17.2896(5) Å, a )
96.0096(11)°, â ) 91.8165(15)°, γ ) 92.1069(15)°, V ) 1649.30-
(6) Å,3 Z ) 2, Fcalcd ) 1.437 Mg m-3, T ) 200(2) K, 7547
independent measured reflections, R1 ) 0.029, wR2 ) 0.029, 5370
absorption-corrected reflections (I > 3σ(I), 2θ e 54.94°), 407
parameters (CCDC 622357).
Crystal Structure Determination of [Ru(η-PhCtCCtCPh)-
(CO)2(PPh3)2] (4b). This complex was prepared according to the
method of Alcock et al.2 Crystal data for 4b‚C6H6: C60H46O2P2-
Ru, Mw ) 962.04, triclinic, P1h (No. 2), a ) 11.1877(2) Å, b )
12.2839(2) Å, c ) 17.9333(4) Å, R ) 92.913(1)°, â ) 100.133-
(1)°, γ ) 101.607(1)°, V ) 2367.0(1) Å,3 Z ) 2, Fcalcd ) 1.350
Mg m-3, T ) 200(2) K, yellow block, 8316 independent measured
reflections, R1 ) 0.066, wR2 ) 0.102, 6003 absorption-corrected
reflections (I > 3σ(I), 2θ e 55°), 228 parameters. Compound 4b
was refined as a 1:1 disordered structure in space group P1h. Diffuse
scattering along lines parallel to b* implied the structure was ordered
C6D6): δH 7.4-6.8 (m, C6H5). 13C{1H} NMR (298 K, 75 MHz,
2
C6D6): δC 201.4 (d, JPC ) 12, RuCO), 168.3 (s, CO), 133.3 (d,
JPC ) 11.0, C2,6(PC6H5)), 131.9 (d, JPC ) 46.8, C1(PC6H5)), 130.1
4
(d, JPC ) 2.7, C4(PC6H5)), C3,5(PC6H5) obscured by C6D6
resonance, 132.1, 129.7, 128.6 (C2,3,5,6(CC6H5); fourth resonance
obscured by C6D6), 134.0, 128.9, 128.8, 126.9 (C1,4(CC6H5)). 96.0,
85.6 (CdCCO), 83.3, 82.9 (CtC). N.B.: some ambiguity in
assignments due to overlap of C6D6 resonance. 31P{1H} NMR (298
K, 121 MHz): in C6D6, δP 41.9; in CDCl3, δP 43.77. IR: in Nujol,
2202 (CtC), 2019, 1969 (RuCO), 1621 (CdO) cm-1; in CH2Cl2,
2246 (CC), 2026, 1975 (RuCO), 1612 (CO) cm-1; in toluene, 2021,
1969 (RuCO) cm-1. N.B.: previously reported data (Nujol) include
two (sic) absorptions at 1928.2 vs and 1928.2 w cm-1. These were
absent from the spectra we obtained and presumably represent a
further product. MS (ESI+): 852 [M+], 263 [HPPh3].
Attempted Synthesis of [Ru(η2-Me3SiCtCCtCSiMe3)(CO)2-
(PPh3)2] (4c). (a) The 31P NMR spectrum of a solution of [Ru-
(CO)2(PPh3)3] (1; 30 mg, 0.032 mmol, δP (C6D6) 58.03) and
Me3SiCtCCtCSiMe3 (8 mg, 0.05 mmol) in d8-toluene was
measured and found to comprise a 5:1 mixture of [Ru(CO)2(PPh3)3]
(δP 50.73) and 4c (δP 45.68). Heating this mixture to reflux for 15
min resulted in the formation of a 1:1 mixture of 1 and 2 (δP 56.69)
in addition to a broadened peak due to PPh3. The spectra remained
invariant on continued heating (2-12 h). After 3 days the spectrum
revealed primarily 2 and PPh3.
(b) A mixture of the ethylene complex Ru(C2H4)(CO)2(PPh3)2
(20; 30 mg, 0.042 mmol, δP (C6D6) 58.03) and Me3SiCtCCt
CSiMe3 (8 mg, 0.05 mmol) in d6-benzene was stirred for 10 min.
The 31P{1H} NMR spectrum of the solution comprised a 2:1 mixture

1338 Organometallics, Vol. 26, No. 6, 2007
Hill et al.
in layers perpendicular to b*. Packing considerations show that
layers take up one of two inversion-related options, which order
all atoms except those of one triphenylphosphine. The choice
between the two options for these atoms is determined by the choice
between inversion-related options for an adjacent layer. Sensible
constraints were used to define 115 independent non-H atom
positions using 228 variables and to obtain a successful anisotropic
atom refinement (CCDC 622356).
Supporting Information Available: CIF files giving crystal-
lographic details for 4b, 9, 21, 22, and 25a. This material is available
free of charge via the Internet at http://pubs.acs.org. These files
are also available from the Cambridge Crystallographic Database:
4b (CCDC 622356), 9 (CCDC 622358), 21 (CCDC 622357), 22
(CCDC 622359), and 25a (CCDC 622360).
OM060888L