Internal alkyne isomerization to vinylidene versus stable π-Alkyne: Theoretical and experimental study on the divergence of analogous Cp*Ru and TpRu systems

ARTICLE

pubs.acs.org/Organometallics

Internal Alkyne Isomerization to Vinylidene versus Stable π-Alkyne:

Theoretical and Experimental Study on the Divergence of Analogous

Cp*Ru and TpRu Systems

Vinay K. Singh,^†Emilio Bustelo,^†Isaac de los Ríos,^†Ignacio Macías-Arce,^†M. Carmen Puerta,*^,†

,†

‡

,‡

ꢀ

Pedro Valerga,* Manuel Angel Ortu~no, Gregori Ujaque, and Agustí Lledꢀos*

^†Departamento de Ciencia de los Materiales e Ingeniería Metalꢀurgica y Química Inorgꢀanica, Facultad de Ciencias, Universidad de Cꢀadiz,

11510 Puerto Real (Cꢀadiz), Spain

^‡Departament de Química, Universitat Autꢁonoma de Barcelona, 08193 Bellaterra, Barcelona, Spain

S Supporting Information

b

ABSTRACT: The activation of internal alkynes by Cp*Ru and

TpRu complexes gives respectively π-alkyne and disubstituted

vinylidene as stable species, even though both systems bear

identical pyridylphosphine ligands (k²P,N- ⁱPr₂PXPy, X = NH,

CH₂, S). The activation of the alkynones PhCtCCOR (R =

Me, Ph) by [TpRuCl(ⁱPr₂PXPy)] complexes allowed us to

isolate and characterize metastable η¹-OdC(R)CtCPh ad-

ducts. These complexes isomerize spontaneously to vinylidene both in solution and in the solid state. Kinetic studies have been

carried out in solution by ³¹P{¹H} NMR and in the solid state by IR spectroscopy, providing the Eyring and AvramiꢀErofeev

parameters, respectively. The activation of internal alkynes without ketone groups provided vinylidene species as well, but without

isolable intermediates. In contrast with the TpRu system, the activation of alkynones by [Cp*RuCl(ⁱPr₂PXPy)] always results in

stable π-alkyne species. Representatives of both Cp*Ruꢀπ-alkyne and TpRuꢀvinylidene compounds have been characterized by

X-ray diﬀraction. DFT calculations have been carried out with the actual experimental complexes, including solvent eﬀects, in order

to analyze the mechanism of the π-alkyne to vinylidene isomerization of internal alkynes and to explain the divergent results

obtained for Tp and Cp*.

’ INTRODUCTION

π-acceptor capability with respect to the π-alkyne isomer. As a

general rule, the relative vinylidene stability will increase with the

electron density on the metal center, particularly with the

presence of stronger (and bulkier) donor ligands in the metal co-

ordination sphere. This is the case of the numerous examples of

stable vinylidene complexes of the type [Cp*Ru(dCdCHR)-

(PR₃)₂]⁺.⁶In contrast, electron-poor dicarbonyl systems such as

[CpFe(CO)₂(dCdCRR⁰)]⁺(R, R⁰= Me, Ph) spontaneously

undergo isomerization to π-alkyne complexes.⁷Intermediate

situations have led to the observation of vinylidene/π-alkyne

equilibria^4eand to reversible alkyne/vinylidene isomerization,

which is essential for several catalytic reactions via vinylidene

intermediates.¹

The conversion of internal alkynes to vinylidenes (RCt

CR⁰f :CdCRR⁰) is a quite uncommon process that has just

emerged during the last few years as a new promising route for

the transformation of internal alkynes via vinylidene inter-

mediates.²Earlier, this process had only been reported for a

series of heteroatom-substituted alkynes, namely alkynylsilanes,⁸

tin acetylides,⁹1-iodo-1-alkynes,¹⁰and mercaptoacetylene.¹¹In

It is well-known that the relative stability of alkyne and

vinylidene isomers is reversed upon coordination to a variety

of transition metals. This process (HCtCR f :CdCHR) is the

main synthetic route to vinylidene species^1,2and is a key step for

several catalytic alkyne transformations.³The simplest role of the

metal is to stabilize the vinylidene lone pair of electrons by means

of a dative carbonfmetal bond, throughout a concerted 1,2-H

rearrangement. A stronger participation of the metal involves the

oxidative addition of the alkyne to yield alkynylꢀhydride inter-

mediates, followed by a formal 1,3-H migration. The nature and

pathway of the migrating hydrogen have been the objects of

numerous theoretical studies.⁴The concerted 1,2-H shift is

predominant in ruthenium chemistry, with the remarkable ex-

ception of some electron-rich complexes such as [Cp*Ru(L)-

(PEt₃)₂][BAr^F], for which the full sequence of intermediates

4

was isolated and characterized by X-ray diﬀraction (π-alkyne/

alkynylꢀhydride/vinylidene).⁵

The π-coordination of the alkyne to the metal is generally

accepted as the ﬁrst mechanistic step, but both electronic and

steric factors contribute to favor the spontaneous rearrangement

to the linear vinylidene chain. The higher thermodynamic sta-

bility of the vinylidene ligand is mainly explained by its better

Received: March 28, 2011

Published: July 05, 2011

r

2011 American Chemical Society

4014

dx.doi.org/10.1021/om200273v Organometallics 2011, 30, 4014–4031

|

Organometallics

ARTICLE

Scheme 1. Synthesis of TpRuꢀη¹-Ketone Adducts

Scheme 2. Synthesis of Cp*Ruꢀπ-Alkynone Complexes

Table 1. Selected IR (Nujol, cm^ꢀ1) and ³¹P{¹H} and ¹³C-

{¹H} NMR Data (δ) for Compounds 5ꢀ8 and the Free

Alkynyl Ketones

¹³C{¹H}

our ﬁrst contribution to this ﬁeld, we showed that TpRu

complexes can also eﬀect the alkyne to vinylidene isomerization

in internal alkynones.¹⁸In agreement with Shaw’s results, we

were able to isolate the η¹-ketone-bound alkynone complexes as

the kinetic products, which then undergo isomerization to the

vinylidene form, both in solution and in the solid state. The

kinetic studies on this transformation support the existence of an

intramolecular process, and a concerted 1,2-acyl migration is

proposed as the key step. Although the presence of π-alkyne

species during the isomerization process can be reasonably

proposed on the basis of the ³¹P{¹H} NMR and IR monitoring

of the reaction, their role as actual intermediates of the process is

not yet clear.

compd

ν(CtC)

³¹P{¹H}

CtC

CdO

4-phenyl-3-butyn-2-one 2203, 2126^a

88.9, 90.0

87.2, 95.7

184.1

178.5

1,3-diphenylpropynone

2200

2204, 2160^a

5

6

7

8

71.7

71.5

87.4, 106.7 198.3

85.9, 106.3 189.7

87.4, 107.8 198.5

86.0, 107.5 189.7

2196

2203, 2159^a

2188

139.5

^aA weak, broad shoulder is observed beside the main peak.

recent times, the migration of carbon substituents has been

shown to be possible, opening a novel access to disubstituted

vinylidenes from internal alkynes.

This paper is aimed at comparing the behavior of Tpꢀ

and Cp*ꢀruthenium complexes to facilitate the conversion of

internal alkynes into their vinylidene isomers. We extend our

previous results on [TpRu(PⁱPr₂Me)(MeCN)]⁺and [TpRu(k²P,

N-ⁱPr₂PNHPy)]⁺to the related [TpRu(k²P,N-ⁱPr₂PCH₂Py)]⁺

and [TpRu(k²P,N-ⁱPr₂PSPy)]⁺fragments, in order to assess the

eﬀect of the pyridylphosphine spacer group (NH, CH₂, S) on the

isomerization process, supported by kinetic studies providing the

Eyring plots and activation parameters. The use of P,N-donor ligands

adds additional interest to this research because of their potential

hemilabile character,¹⁹which in the future might allow the develop-

ment of stoichiometric and catalytic transformation of internal

alkynes. Disubstituted vinylidenes have also been obtained starting

from internal alkynes not bearing a ketone group: RCtCR⁰(R, R⁰:

Ph, COOMe; Me, COOEt; Ph, Ph).

The ﬁrst example reported the formation of bridging vinyli-

dene species by activation of acetylenedicarboxylate within a

binuclear ruthenium complex.¹²More recently, Shaw et al. have

described that internal alkynones, RCtCCOR⁰, are also trans-

formed into their isomeric disubstituted vinylidenes by

[CpRuCl(PPh₃)₂].¹³A ruthenium cyclophosphate complex is

able to reproduce this behavior with aryl-, alkyl-, and carboxy-

substituted internal alkynes, establishing their relative migratory

aptitude and providing the ﬁrst mechanistic insight.¹⁴Aryl- and

alkyl-disubstituted vinylidenes have also been obtained by reac-

tion of [CpRu(dppe)Cl] and [CpFe(dppe)Cl] with PhCtCAr

(Ar = p-C₆H₄X; X = OMe, Me, H, Cl, COOEt) and 2-pentyne.¹⁵

Finally, the reverse process (vinylidene to π-alkyne iso-

merization) has been reported for the aforementioned electro-

philic iron complexes.⁷From these studies the migratory aptitude

of the alkyne substituents have been disclosed, electron-with-

drawing groups increasing the migratory aptitude in the order

CO₂Et ≈ C₆H₄-CO₂Et-p > Me > Ph > C₆H₄-Me-p > C₆H₄-

OMe-p.^2,14,15

An important issue regarding the scope of the conversion of

internal alkynes into their corresponding vinylidene ligands and

its potential exploitation in synthetic chemistry is assessment

of the inﬂuence of the transition-metal ligands in the transfor-

mation. Most of the complexes able to perform such processes

are half-sandwich ruthenium complexes with Cp-related ligands,

and a natural extension is to check the behavior of tris(pyrazolyl)-

borate (Tp) ligands. Comparisons between Cp- and Tp-type

ligands can be found regarding H₂¹⁶and CꢀH¹⁷activation. In

Finally, the related complexes [Cp*RuCl(k²P,N-ⁱPr₂PXPy)]

(X = NH, CH₂) have been tested in the activation of internal

alkynes, isolating and characterizing π-alkyne complexes as stable

products, which resist isomerization to vinylidene even at high

temperatures and long reaction times. X-ray structures have been

determined for the compounds [Cp*RuCl(k²P,N-ⁱPr₂PSPy)],

[Cp*Ru(η²-PhCtCCOPh)(k²P,N-ⁱPr₂PNHPy)], [TpRu{dCd

C(COMe)Ph}(k²P,N-ⁱPr₂PSPy)][BAr^F], and [TpRu{dCd

4

C(COOMe)Ph}(k²P,N-ⁱPr₂PCH₂Py)][BAr^F].

4

The two related ruthenium complexes diﬀer only in the

presence of a Tp or Cp* ligand yet exhibit contrasting behavior

regarding the isomerization of the same internal alkyne to

vinylidene. This ﬁnding makes these complexes very appealing

for analysis of the factors which control this process. With the

4015

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Table 2. Selected IR (Nujol, cm^ꢀ1) and ³¹P{¹H} and ¹³C-

{¹H} NMR Data (δ) for 9ꢀ12, 19, and 20

¹³C{¹H}

ν(CtC)

³¹P{¹H}

CtC

CdO

190.9 (s)^a

9

1828

66.4 (s, 15%)

86.10 (d, 9.6 Hz)^a

69.8 (s, 85%)^a116.8 (d, 4.8 Hz)^a

66.8 (s, 22%)

85.02 (d, 8.8 Hz)^a

68.3 (s, 78%)^a113.4 (d, 5.0 Hz)^a

128.0 (s, 7%)

84.59 (d, 9.2 Hz)^a

128.3 (s, 93%)^a115.7 (br s)^a

126.9 (s, 86%)^a85.15 (d, 8.3 Hz)^a

127.3 (s, 14%) 114.2 (d, 4.5 Hz)^a

67.2 (s, 40%)^b87.02 (s)/94.19 (s)^b159.0 (d, 3.8 Hz)^b

69.8 (s, 60%)^b82.23 (s)/92.81 (s)^b158.6 (d, 3.8 Hz)^b

127.7 (s, 45%)^c76.90 (d, 9.7 Hz)^c

127.9 (s, 55%)^c105.9 (d, 4.6 Hz)^c

78.33 (d, 5.6 Hz)^c

10

11

12

19

20

1800

1842

1871

1842

1856

185.3 (d, 3.8 Hz)^a

190.2 (s)^a

185.4 (d, 3.8 Hz)^a

162.3 (d, 4.8 Hz)^c

163.0 (s)^c

Figure 1. ORTEP view of the cation of [Cp*Ru(η²-PhCtCCOPh)-

(k²P,N-ⁱPr₂PNHPy)][BPh₄] (12). Selected bond lengths (Å) and

angles (deg): Ru1ꢀC16 = 2.125(3), Ru1ꢀC17 = 2.178(2), Ru1ꢀ

P1= 2.3057(9), Ru1ꢀN1=2.121(2), C16ꢀC17 = 1.260(4), C16ꢀC24 =

1.460(4), C17ꢀC18 = 1.461(4); C16ꢀRu1ꢀC17 = 34.01(10),

C16ꢀC17ꢀC18 = 152.8(3), C17ꢀC16ꢀC24 = 150.3(3), P1ꢀRu1ꢀ

N1 = 79.62(6).

102.1 (d, 9.8 Hz)^c

^{a 13}C{¹H} NMR data correspond to the major rotamer. ^{b 13}C{¹H} NMR

signals are clearly identiﬁable and assignable to each rotamer. ^{c 13}C{¹H}

NMR signal assignment to rotamers is ambiguous and exchangeable.

goal of unraveling the mechanistic features of the isomerization

of internal alkynes and justifying the contradictory results ob-

tained for Tp and Cp* ruthenium complexes, we have performed

DFT calculations that are also collected in this article.

to vinylidene (particularly those derived from the 4-phenyl-3-

butyn-2-one activation) and very labile when treated with acet-

one, yielding the corresponding acetone adducts. In absence of

better ligands, these compounds are isolable as dark solids (brown

to violet), which should be stored at low temperature to prevent

isomerization to vinylidene even in the solid state.

’ RESULTS AND DISCUSSION

The IR and NMR spectra, elemental analysis, and the aforemen-

tioned reactivity support the presence of the η¹-O-ketone adducts

(Scheme 1) [TpRu{η¹-OdC(R)CtCPh}(k²P,N-ⁱPr₂PXPy)]-

Experimental Results on the Activation of Internal Al-

kynes. Hydrotris(pyrazolyl)borate (Tp) and pentamethylcyclo-

pentadienyl (Cp*) are isoelectronic ligands widely employed in

organometallic chemistry and catalysis. Both polydentate ligands

are able to block three coordination sites on a metal center, giving

rise to a large family of compounds with three-legged piano-stool

geometries. In spite of their obvious similarities, the Cp* deriv-

atives are usually considered to be more electron rich than the

Tp analogues, and the increased donation of the Cp* ligand

to the metal results in a greater π back-bonding when suitable

π-acceptor ligands are present.

[BAr^F] (X=CH₂, R=Me(5), Ph(6);X=S, R=Me(7), Ph(8)).

4

Most of the spectroscopic features (IR and NMR) of com-

plexes 5ꢀ8 are comparable to those of the free organic species

(Table 1), which are in good agreement with the existence of a

weak metalꢀligand interaction, and thus the electronic distribu-

tion along the ligand remains almost unchanged. The sharp and

intense IR ν(CtC) absorptions in the range 2159ꢀ2204 cm^ꢀ1

clearly rule out the presence of π-alkyne species, the bands

for which are typically found between 1700 and 2000 cm^ꢀ1

.

Consequently, the chloride abstraction process is much harder

for the TpRu starting complexes [TpRuCl(k²P,N-PⁱPr₂XPy)]

(X = CH₂(1), S (2)); the process is only possible in the presence

The ¹³C{¹H} NMR spectra show two signals at around 86 and

107 ppm for the quaternary alkynyl carbon atoms. Whereas the

ﬁrst signal is only slightly shifted 1ꢀ2 ppm upﬁeld compared to

that for the free alkynyl ketone, the second signal is somewhat

more aﬀected by the coordination to the metal (12ꢀ17 ppm

downﬁeld). In all cases, no CꢀP coupling is observed, and the

¹³C{¹H} CO chemical shift (δ 189ꢀ190 ppm) is in the normal

range for organic carbonyl groups. These data also rule out the

possibility of a η²-ketone complex, for which the CO signal

should clearly be shifted upﬁeld.²⁰Therefore, it seems reasonable

to propose the presence of a σ-bonded alkynyl ketone ligand

coordinated through the oxygen atom in compounds 5ꢀ8.

Recently, Shaw et al. have reported the formation of σ-keto

alkynyl complexes with the [CpRu(PPh₃)₂]⁺fragment.¹³

of NaBAr^F(Ar^F= 3,5-bis(triﬂuoromethyl)phenyl). On the

4

other hand, the Cp*Ru derivatives [Cp*RuCl(k²P,N-ⁱPr₂PXPy)]

(X = CH₂(3), NH (4)) can release the halide spontaneously in

methanol solution. Thus, NaBPh₄has been employed instead of

NaBAr^Ffor practical reasons. The use of NaBAr^Fin Cp*Ru

4

complexes has also been tested, and the substitution did not show

noticeable diﬀerences in the reactivity described in this paper.

Synthesis and Characterization of η¹-Alkynone Complexes.

The treatment of fluorobenzene solutions of complexes 1 and 2

with alkynyl ketones, PhCtCCOR (R = Ph, Me), in the pre-

sence of NaBAr^F₄produces a slow color change from the initial

light yellow to darker colors ranging from green to violet. The re-

action is complete in 30 min at 0 °C for the 4-phenyl-3-butyn-2-

one derivatives, whereas the slower reaction with diphenylpro-

pynone requires stirring the mixture for 2 h at room temperature.

The reaction products are metastable with regard to isomerization

Synthesis and Characterization of π-Alkyne Complexes. The

treatment of the Cp*Ru complexes 3 and 4 with the alkynyl

ketones PhCtCCOR (R = Ph, Me), in methanol at room tem-

perature and in the presence of NaBPh₄, causes the immediate

4016

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Table 3. Selected Bond Lengths and Angles for 12 and Related RuꢀAlkyne Complexes

compd

RuꢀC (Å)

CꢀC (Å)

bent-back angle (deg)

[Cp*Ru(ⁱPr₂PNHPy)(η²-PhCtCCOPh)]⁺(12)

2.125

2.178

2.182

2.180

2.204

2.209

2.222

2.326

2.135

2.133

2.302

2.308

2.183

2.211

2.203

2.130

2.232

2.202

1.260

27.2

29.7

[Cp*Ru(PEt₃)₂(η²-HCtCH)]^{+ 5}

[CpRu(PMe₂Ph)₂(η²-HCtCH)]^{+ 25}

[Cp*Ru(η⁶-dienyne)]^{+ 26}

1.220

1.224

1.237

1.261

1.228

1.272

1.169

1.228

[Cp*Ru(η³-allyl)(η²-PhCtCPh)]²⁷

[(η⁵-C₂B₉H₁₁)Ru(CO)₂(η²-MeCtCPh)]²⁸

[(Tripod)Ru(η²-PhCtCPh)]^{+ a,29}

[Ru(P₃O₉)(MeCtCCO₂Et)(dppe)][PPN]^b,14

[CpRu(dppe)(η²-EtCtCMe)]^{+ 15}

29.3

29.2

23.4

18.4

31.9

32.2

19.45

49.23

35.94

26.42

^aTripod = η⁵:η¹:η¹-MeC(CH₂Cp)(CH₂PPh₂)₂. ^bPPN = (Ph₃P)₂N⁺.

precipitation of yellow solids, isolated and characterized as the

π-alkyne complexes (Scheme 2) [Cp*Ru(η²-PhCtCCOR)-

(k²P,N-ⁱPr₂PXPy)][BPh₄] (X = CH₂, R = Me (9), Ph (10);

X = NH, R = Me (11), Ph (12)). These compounds are not labile

against weak nucleophiles such as acetone, and they do not iso-

merize to vinylidene even at high temperatures and long reaction

times. Compounds 9ꢀ12 were heated at 80 °C for 5 days with-

out noticeable changes. The use of higher temperatures (DMF,

140 °C, 24 h) leads to a mixture of decomposition products, in-

cluding the two-step release of the pyridylphosphine and the

formation of the sandwich complex [Cp*Ru(η⁶-C₆H₅BPh₃)].²¹

The [B(Ar^F)₄] analogue of compound 11 has also been heated at

120 °C in 1,1,2,2-tetrachloroethane-d₂and the process moni-

tored by ³¹P{¹H} NMR. None of the detected species corre-

spond to vinylidene, as shown in the ¹³C{¹H} NMR spectrum by

the lack of the characteristic low-field resonance. Instead, pro-

cesses involving alkynone rotation, alkynone release, and the

aforementioned decomposition pathways are detected.

The IR spectra show very broad absorptions centered at 1828,

1800, 1842, and 1871 cm^ꢀ1, respectively, for complexes 9ꢀ12

(Table 2). Although the elemental analyses match well with a

pure substance of the expected empirical formula, the NMR

spectra display two sets of signals. The ratio of the major to minor

product is >78% based on the ³¹P{¹H} signal. This observation is

attributed to the existence of two alkyne rotamers due to the

restricted rotation about the Ruꢀalkyne bond.²²The ¹³C{¹H}

NMR of the major product shows doublet signals for the alkyne

quaternary carbon atoms (Table 2).

as one site, and the pyridylphosphine occupying the other two

coordination positions. The alkyne is asymmetrically bonded to

ruthenium (2.125(3), 2.178(2) Å), which is consistent with the

presence of an asymmetric ruthenium center and two diﬀerent

substituents on the alkyne. The alkyne carbon atom bearing the

phenyl group is farther from ruthenium than that bonded to the

acyl group. Although the NMR spectra show the presence of two

alkyne rotamers (86:14 ratio for 12), the crystal structure only

corresponds to one of them, likely the major rotamer, with the

acyl group in a cisoid disposition to the phosphorus atom and

transoid to the pyridyl nitrogen atom. This orientation seems to

minimize the steric hindrance by pointing the CO group toward

the phosphine isopropyl groups.

The strength of the metalꢀalkyne interaction can be readily

assessed by the metalꢀC(alkyne) bond distance and the degree

to which the substituents on the alkyne ligand are bent back from

the CꢀC axis. Table 3 gives a list of these bond lengths and

angles for various π-alkyne complexes related to 12. The Ruꢀ

C16/C17 distances (2.152 Å in average) are at the lower end of

the range. Consistent with the alkyne acting as only a two-elec-

tron donor, the C16ꢀC17 distance is lengthened to 1.260(4) Å,

which is comparable tothe mean value (1.269 Å) foundinalkyneꢀ

metal complexes,²³and considerably longer than the mean CtC

bond distance for organic disubstituted alkynes (1.192 Å).²⁴The

alkyne substituents are bent back (C16ꢀC17ꢀC18 = 152.8(3)°,

C17ꢀC16ꢀC24 = 150.3(3)°), with a deviation from linearity of

27.2 and 29.7°, respectively, conﬁrming a signiﬁcant π back-

bonding interaction from the metal to the alkyne.

All these features are quite characteristic of the presence of a

η²-alkyne ligand, and this has been deﬁnitively conﬁrmed by

X-ray diﬀraction analysis of suitable monocrystals of 12, obtained

by slow diﬀusion of Et₂O into an acetone solution of the com-

plex. An ORTEP view of the cation complex 12 is shown in

Figure 1 with selected bond lengths and angles.

The geometry around the ruthenium atom in 12 can be

described as a three-legged piano stool, with the cyclopentadie-

nyl ligand occupying three sites, the alkyne midpoint considered

It is interesting to compare the structure of complex 12 (which

does not isomerize to vinylidene) with the two structurally

characterized π-alkyne complexes reported by Ishii et al. (last

two entries in Table 3), which do isomerize to vinylidene. In both

cases, the rutheniumꢀalkyne interaction seems slightly weaker

than in 12, exhibiting longer RuꢀC distances and shorter CꢀC

bond lengths. However, the bent-back angle is larger at one side

of the alkyne in Ishii’s complexes (i.e., the COOEt bent-back

angle is 49.23°, whereas that for the Me group is 19.45°).

4017

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Scheme 3. Synthesis of TpRuꢀVinylidene Complexes

Table 4. Selected ³¹P{¹H} and ¹³C{¹H} NMR Data (δ) for

Compounds 13ꢀ18

¹³C{¹H}

compd

RudCdC

²J_CP(Hz)

CdO

RudCdC

³¹P{¹H}

13

14

15

16

17

18

357.3

356.5

363.1

357.5

361.3

364.8

18.2

19.5

18.3

18.8

18.5

18.7

196.6

189.9

196.3

166.0

166.9

131.8

131.5

136.2

123.8

113.9

130.8

64.3

63.6

123.6

64.7

65.7

65.9

derivatives (13 and 15) and for 48 h at 50 °C in the case of the

phenyl derivative 14, in order to complete the reaction. The

complex with X = S and R = Ph undergoes decomposition during

the long reaction time or on heating at higher temperatures.

Although the vinylidene complex can be detected, it could not be

isolated and characterized properly.

Compounds 13ꢀ15 have been characterized by spectroscopy

and by the X-ray diﬀraction analysis of monocrystals of 15 (X = S,

R = Me), grown by slow diﬀusion of petroleum ether into a

diethyl ether solution of the complex. Figure 2 gives an ORTEP

view of the cationic complex with selected bond lengths and

angles.

The X-ray structure of complex 15 shows a distorted-octahe-

dral coordination around the ruthenium center, with three facial

coordination sites occupied by the N atoms of the Tp ligand, two

other positions by the P and N atoms of the 2-(diisopro-

pylphosphinothio)pyridine ligand, and the sixth position by a

linear carbon chain. This ligand exhibits a short RuꢀC1 distance

of 1.805(2) Å, which reﬂects the strong back-bonding from the

metal characteristic of the vinylidene ligands. The C1ꢀC2 bond

length of 1.329(3) Å, and the C1ꢀC2ꢀC3/C1ꢀC2ꢀC5 bond

angles (117.60(18) and 118.60(18)°, respectively) ﬁt well with

a vinylidene ligand with sp²hybridation at C2. The carbon chain

is almost linearly assembled to ruthenium (Ru1ꢀC1ꢀC2 =

169.94(17)°). These data compare well with the structural infor-

mation available for related complexes such as [CpRu{dCd

Figure 2. ORTEP view of the cation of [TpRu{dCdC(COMe)Ph}-

(k²P,N-PⁱPr₂SPy)][BAr^F] (15). Selected bond lengths (Å) and angles

4

(deg): Ru1ꢀC1 = 1.805(2), C1ꢀC2 = 1.329(3), C2ꢀC3 = 1.489(3),

C2ꢀC5 = 1.496(3), Ru1ꢀP1 = 2.2916(7), Ru1ꢀN7 = 2.1326(17);

Ru1ꢀC1ꢀC2 = 169.94(17), C1ꢀC2ꢀC3 = 117.60(18), C1ꢀC2ꢀ

C5 = 118.60(18), P1ꢀRu1ꢀN7 = 83.93(5).

Synthesis and Characterization of Disubstituted Keto Vinyl-

idene Complexes. The activation of alkynones by Cp*Ru and

TpRu complexes have yielded substantially distinct compounds,

namely the π-alkyne complexes 9ꢀ12 and the η¹-O-keto alkyne

complexes 5ꢀ8, easily identifiable by their characteristic spectro-

scopic features. They are also remarkably different in stability:

whereas the π-alkyne compounds are stable under an inert atmo-

sphere, the alkynyl ketone adducts undergo spontaneous iso-

merization to vinylidene both in solution and in the solid state.

It is noteworthy that π-alkyne complexes are generally described

as reasonable intermediates in the first step of the alkyneꢀ

vinylidene isomerization, which has been shown to be the case in

the isomerization of terminal alkynes.^1ꢀ6,30

C(COPh)Ph}(PPh₃)₂][BAr^F] (1.818 and 1.335 Å; 120.06,

4

119.74, and 170.73°),¹³[CpRu{dCdC(Ph)(p-C₆H₄OMe)}-

(dppe)][BAr^F] (1.838 and 1.327 Å; 122.5, 118.4, and

4

173.3°),¹⁵and [TpRu{dCdC(COPh)Ph}(PⁱPr₂Me)(MeCN)-

[BAr^F] (1.814 and 1.322 Å; 118.7, 124.1, and 177.7°).¹⁸The

4

enone moiety (CdCCOMe) is almost planar (dihedral angles:

C1ꢀC2ꢀC3ꢀC4 = 13.15°, C1ꢀC2ꢀC3ꢀO1 = 12.17°), adopt-

ing an s-cis conformation with the CO group pointing to the

pyridine side.

In addition to the X-ray data, the most characteristic vinylidene

feature can be found in the ¹³C{¹H} NMR spectra. The presence

of a carbene carbon atom is revealed by characteristic low-ﬁeld

resonances observed at 357.3, 356.5, and 363.1 ppm, respec-

tively, for compounds 13ꢀ15, which appear as doublets with

CꢀP coupling constants of 18ꢀ19 Hz. A summary of the main

spectroscopic features of all the vinylidene complexes reported in

this paper is shown in Table 4.

Activation of Internal Alkynes Not Bearing a Ketone Group.

In order to test the synthesis of vinylidene complexes from

internal alkynes not bearing a ketone group, we have tried the

reaction of the starting complex 1 with a series of disubstituted

alkynes RCtCR⁰(R, R⁰: Ph, COOMe; Me, COOEt; Ph, Ph).

Therefore, the synthesis of disubstituted vinylidenes can be

accomplished either by heating a 1,2-dichloroethane solution of

complexes 5ꢀ7 or directly from the starting complexes 1 and 2 in

ﬂuorobenzene with heating (Scheme 3). As previously observed,

the isomerization rate reveals a strong dependence on the ketone

substituent (Me . Ph), and thus the preparation of the vinyli-

dene complexes [TpRu{dCdC(COR)Ph}(k²P,N-ⁱPr₂PXPy)]-

[BAr^F] (X = CH₂, R = Me (13), Ph (14); X = S, R = Me (15))

4

requires heating for 1 h at 50 °C in the case of the methyl

4018

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Scheme 4. Synthesis of TpRuꢀVinylidene Complexes with-

Scheme 5. Synthesis of Cp*Ruꢀπ-Alkyne Complexes with-

out a Ketone Group

group (torsion angles: C1ꢀC2ꢀC9ꢀO1 = 6.49°, C1ꢀC2ꢀ

C9ꢀO2 = 6.09°).

On the other hand, the activation of the internal alkyne

PhCtCCOOMe by the starting complexes 3 and 4 yields the

expected π-alkyne derivatives [Cp*Ru(η²-PhCtCCOOMe)-

(PⁱPr₂XPy)][BPh₄] (X = CH₂(19), NH (20)), which do not

isomerize to vinylidene (Scheme 5). The spectra of complexes 19

and 20 exhibit the same characteristics observed for the related

compounds 9ꢀ12 (selected NMR and IR data are shown in

Table 2), which are consistent with the existence of a π-alkyne

ligand: the presence of rotamers (40:60 ratio for 19, 45:55 ratio

for 20, based on the integrals of the ³¹P{¹H} NMR spectra),

CꢀP coupling for the alkyne quaternary carbon ¹³C{¹H} NMR

signals, and broad IR absorptions at 1842 and 1856 cm^ꢀ1

respectively.

,

These results conﬁrm that Cp*Ru complexes stabilize the π-

alkyne as a ﬁnal product which is reluctant to isomerize to

vinylidene regardless of the substituents on the alkyne. On the

other hand, TpRu complexes seem to avoid π-alkyne formation

by stabilizing the η¹-ketone adduct as a metastable intermediate

to vinylidene. Despite the formal similarities between Tp and

Cp/Cp* ligands, the obvious diﬀerences in size and electronic

properties result in a diﬀerent chemistry.³¹In particular, the steric

bulk and the rigidity of the Tp ligand appear to disfavor higher

coordination numbers of the metal center, which may favor the

η¹-keto alkyne with regard to the η²-alkynyl ketone coordination.

Kinetic Studies of the Alkynone to Vinylidene Isomerization

in Solution. A qualitative analysis of the experimental conditions

required for the η¹-keto alkyne to vinylidene isomerization

shows that the presence of a ketone group significantly increases

the reaction rate (COMe . COOMe; see synthesis of com-

pound 13 vs 16) and that the ketone substituent also plays an

important role in the isomerization rate (Me . Ph; see synthesis

of 13 vs 14).

Figure 3. ORTEP view of the cation of [TpRu{dCdC(COOMe)-

Ph}(k²P,N-PⁱPr₂CH₂Py)][BAr^F] (16). Selected bond lengths (Å) and

4

angles (deg): Ru1ꢀC1 = 1.803(2), C1ꢀC2 = 1.330(3), C2ꢀC3 =

1.490(4), C2ꢀC9 = 1.480(3), Ru1ꢀP1 = 2.3086(8), Ru1ꢀN7 =

2.105(2); Ru1ꢀC1ꢀC2 = 171.1(2), C1ꢀC2ꢀC3 = 119.1(2),

C1ꢀC2ꢀC9 = 119.7(2), P1ꢀRu1ꢀN7 = 81.95(6).

The corresponding vinylidenes [TpRu(dCdCRR⁰)(k²P,N-

PⁱPr₂CH₂Py)][BAr^F] (R, R⁰: Ph, COOMe (16); Me, COOEt

4

(17); Ph, Ph (18)) have been isolated as orange-brown solids

after heating the reaction mixture for at least 20 h at 75 °C

(Scheme 4).

Attempts to isolate any intermediate failed and yielded only

a complex mixture of decomposition products. However, the

direct preparation from 1 gave complexes 16ꢀ18 in pure form in

moderate to good yields. These complexes have been character-

ized by spectroscopy (showing the typical carbenic resonance as

doublets at 357.5, 361.3, and 364.8 ppm, respectively, in their

¹³C{¹H} NMR spectra; see Table 4), and particularly by the

X-ray structure of complex 16. An ORTEP view of the cation

complex is shown in Figure 3 with selected bond lengths and

angles.

The structure of 16 resembles that previously described for

complex 15, showing a distorted-octahedral geometry around

the ruthenium center and the presence of the vinylidene ligand

with almost identical bond lengths and angles. However, the

ligand conformation is s-trans, showing a planar CdCCOOMe

Quantitative kinetic measurements have been obtained by

monitoring the evolution of the ³¹P{¹H} NMR signals of com-

pounds 5ꢀ7 in 1,1,2,2-tetrachloroethane-d₂solution at variable

temperature. In all cases, the isomerization processes obey a ﬁrst-

order rate law. Thus, the ﬁrst-order rate constants have been

obtained at diﬀerent temperatures and the corresponding Eyring

plots provide the activation parameters. ln(I_t/I₀) vs time and

Eyring plots can be found in the Supporting Information. Table 5

gives the rate constants at each temperature, the temperature

ranges at which the isomerization rates were measurable, and the

Eyring activation parameters. The results of our previous kinetic

study on the [TpRu(k²P,N-ⁱPr₂PNHPy)]⁺fragment are shown

for comparative purposes.¹⁸

In all cases ΔG^q

is around 23ꢀ25 kcal mol^ꢀ1, with

298

activation enthalpies of 17ꢀ22 kcal mol^ꢀ1and slightly negative

4019

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Table 5. Rate Constants and Activation Parameters for the η¹-Keto Alkyne f Vinylidene Isomerization

10^ꢀ4k, s^ꢀ1

ligand, alkyne

T, K ⁱPr₂PNHPy, PhCtCCOMe^{18 i}Pr₂PCH₂Py, PhCtCCOMe (5 f 13) ⁱPr₂PSPy, PhCtCCOMe (7 f 15) ⁱPr₂PCH₂Py, PhCtCCOPh (6 f 14)

313

318

323

328

333

338

343

348

353

358

363

2.36(4)

3.91(5)

6.16(12)

11.1(2)

13.1(3)

21.1(3)

1.80(2)

4.40(8)

6.80(9)

10.3(2)

17.8(5)

2.40(4)

3.80(7)

5.90(14)

11.68(23)

14.2(4)

2.47(6)

3.97(9)

5.10(12)

8.22(19)

12.5(4)

ligand, alkyne

ⁱPr₂PNHPy,

ⁱPr₂PCH₂Py,

ⁱPr₂PSPy,

ⁱPr₂PCH₂Py,

PhCtCCOMe¹⁸

PhCtCCOMe (5 f 13)

PhCtCCOMe (7 f 15)

PhCtCCOPh (6 f 14)

ΔH^q, kcal mol^ꢀ1

ΔS^q, cal mol^ꢀ1K ^ꢀ1

ΔG^q₂₉₈, kcal mol^ꢀ1

17.6 ( 1.0

ꢀ19 ( 3

21.9 ( 1.9

ꢀ5 ( 6

19.9 ( 1.6

ꢀ13 ( 5

18.9 ( 1.0

ꢀ20 ( 3

23.2 ( 1.3

23.5 ( 2.5

24.0 ( 2.1

24.9 ( 1.4

values for the activation entropies of the process. The ﬁrst-order

kinetic data and the activation parameters point to a simple con-

certed process. The negative activation entropy may be under-

stood as a decrease of translational, rotational, and vibrational

degrees of freedom on the route to the transition state, suggesting

a concerted 1,2-sigmatropic shift of the alkyne substituent for the

rate-determining step of the reaction.

the η¹-keto alkyne and the ﬁnal vinylidene complexes. In

addition to the fact that π-alkynes in the described Cp*Ru

systems do not rearrange to vinylidene, this opens two questions:

what is the role of π-alkyne species in the isomerization process,

and how can these two related ruthenium complexes, diﬀering

only in the presence of a Tp or a Cp* ligand, exhibit such a

contrasting behavior regarding the isomerization of the same

internal alkyne to vinylidene?

These results are in good agreement with the literature for

similar alkyne/vinylidene isomerization processes, such as the

concerted 1,2-SiMe₃shift in trans-[RhCl(L)(PPrⁱ₃)₂] (ΔH^q=

19.8 kcal mol^ꢀ1, ΔS^q= ꢀ4.8 cal mol^ꢀ1K^ꢀ1).^8cKinetic studies

on terminal alkyne isomerization via a direct 1,2-H shift have been

reported with comparable activation parameters.³²However, these

are the ﬁrst complete kinetic studies about the migration of carbon

functional groups in metal-mediated alkyne/vinylidene isomeriza-

tion. The migration of acetyl, formyl, or carbomethoxy groups in

pure organic systems, such as the migration of an acetyl group in

1,3-cyclohexadiene (ΔS^q= ꢀ7.8 cal mol^ꢀ1K^ꢀ1), usually requires

higher activation energies and similar negative values of ΔS^q.³³

The rate constants and the temperature range at which they have

been measured clearly indicate that the eﬀect of the spacer group X

in the pyridylphosphine ligand (PⁱPr₂XPy) is very small. The

process is only slightly faster with X = NH, CH₂than with X = S

as linking group. The eﬀect of the ketone group (COR) is much

stronger, where the derivatives with R = Ph isomerize too slowly to

be measured, except for the 6 f 14 isomerization (X = CH₂, R =

Ph). This process requires temperatures 30 K higher than that for

the isomerization of 5(X=CH₂, R = Me), and 20 K higher than that

for 7 (X = S, R = Me) to reach similar reaction rates. The large eﬀect

of the R group (PhCtCCOR, R = Me, Ph) on the reaction rate

shows that the migrating group is predominantly the acyl group.

The ³¹P{¹H} NMR monitoring of the reaction during the kinetic

studies did not show any signals other than those corresponding to

For the ﬁrst question, it is noteworthy that Gladysz et al. have

reported that CdO/σ type bonds are kinetically favored with

respect to CdC and CtC/π bonds, which are the thermo-

dynamically preferred binding sites for the coordination of R,

β-unsaturated aldehydes and ketones to the rhenium fragment

[CpRe(NO)(PPh₃)]⁺.³⁴On the basis of our experimental results

and the previous literature, a mechanistic proposal was already

outlined in our last publication on this subject. However, to the

best of our knowledge, there is no precedent of a thorough

theoretical study on the mechanistic features of the isomerization

of internal alkynes, which can eventually justify the contradictory

results obtained for Tp and Cp* ruthenium complexes.

Kinetic Studies of the Alkynone to Vinylidene Isomerization

in the Solid State. An additional support for a concerted mech-

anism is the existence of solid-state isomerization, which was

described in detail in our last paper.¹⁸In this paper, we have also

followed the solid-state isomerization 5 f 13 by IR spectrosco-

py. The starting η¹-ketone complex displays a strong ν(CtC)

vibration at 2204 cm^ꢀ1(with a shoulder at 2160 cm^ꢀ1char-

acteristic of the 4-phenylbutyn-2-one derivatives). During the

isomerization to vinylidene, the ν(CtC) IR band gradually

disappears. Monitoring the integrated intensity of the ν(CtC)

IR band as a function of time has allowed us to evaluate

the advance of isomerization at different temperatures (37, 45,

and 50 °C).

4020

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Scheme 6. Ligands and Substrate Used in the Calculations

Scheme 7. AlkynoneꢀVinylidene Isomerization for Tp Complex 1^a

^aRelative ΔG_FBZvalues are given in kcal mol^ꢀ1with respect to the separated species.

The corresponding R (fraction of transformed solid) vs time

curves ln(I_t/I₀) for the 5 f 13 isomerization in the solid state at

37, 45, and 50 °C can be found in the Supporting Information, as

well as the graphics corresponding to a ﬁt to the AvramiꢀErofeev

equation.³⁵In contrast to the reaction in solution, the solid-state

isomerization does not ﬁt with a ﬁrst-order rate law, as the

process considerably slows down at values of R around 0.5ꢀ0.6,

requiring 3ꢀ4 days for completion. The n values measured at 37,

45, and 50 °C were 0.91(9), 0.51(4), and 0.59(7), respectively.

The mechanism controlling the reaction should be nucleation

and nuclei growth, rather than the chemical reaction itself. Values

of the Avrami exponent around 0.5 correspond to the thickening

of large plates after lateral collision.³⁶Thus, at a certain value of

R (around R = 0.5 for us), the product crystal nuclei get into

contact with each other, the overlapping of the product nuclei

takes place, and a decrease of the reaction rate is observed.

4021

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Scheme 8. Transition States of the 1,2-Acyl Shift in the Tp

Complexes

for the TpRu starting complexes is only possible in the presence

of NaBAr^F. The relative stabilities of the η¹-O-ketone and

4

π-alkyne adducts also agree with the experimental evidence. Cal-

culations show that the η¹-ketone species 5 are favored com-

pared to the π-alkyne intermediates Tp-π-alk; indeed, the latter

have Gibbs energies above those of the separated species 0-Tp

(4.3 kcal mol^ꢀ1for Tp-π-alk_a and 1.2 kcal mol^ꢀ1for Tp-π-

alk_b) and, therefore, they should be hardly detected. The

calculated ν(CtC) absorptions for the two conformational iso-

mers of 5 (5a, 2201 cm^ꢀ1; 5b, 2180 cm^ꢀ1) are in good agree-

ment with the experimental IR ν(CtC) absorptions of complex

5 (2159ꢀ2204 cm^ꢀ1).³⁷

Despite the η¹-O-keto alkyne being the most stable adduct

resulting from the initial interaction of 0-Tp and the alkynone, it

is reasonable to assume that a π-alkyne intermediate is required

in order to make the isomerization proceed. The existence of a

weak metalꢀligand interaction in 5 (ΔG_FBZvalues of binding of

only 1.9 kcal mol^ꢀ1in 5a and 2.9 kcal mol^ꢀ1in 5b) makes such

intermediates accessible by a dissociationꢀassociation process.

In this way the nondetected π-alkyne rotamers Tp-π-alk_a

and Tp-π-alk_b can be formed.³⁸From these intermediates a

1,2-acyl migration leads to the vinylidene isomers 13a,b after

crossing transition states TS(Tp-π-alk_af13a) and TS(Tp-π-

Figure 4. Optimized structures of selected species of the alkynoneꢀ

vinylidene isomerization in the TpRu complex. Bond distances are given

in Å and angles in deg. Hydrogen atoms on Tp and P,N ligands have

been removed for clarity.

ꢀ

alk_bf13b) depicted in Scheme 8 and placed 20.8 and 26.1 kcal

Computational Study of the Alkyne to Vinylidene Isomer-

ization. In order to analyze the mechanism of the π-alkyne f

vinylidene isomerization of internal alkynes and to explain

the contrasting results obtained for Tp and Cp* ruthenium

complexes, we have undertaken a theoretical study of the

formation of disubstituted vinylidenes from the internal alky-

none PhCtCCOMe (A-alk) on the coordination sphere of

ruthenium complexes [LRu(k²P,N-ⁱPr₂PCH₂Py)Cl] (L = Tp,

Cp*). No model has been used in the computational study,

calculations being performed using the actual experimental

complexes 1 and 3 including solvent effects with a continuum

description of the solvent (fluorobenzene in the case of Tp

complexes and methanol in the case of Cp* complexes). All the

values collected in the energy profiles are relative Gibbs energies

in solution (ΔG_solv). Scheme 6 summarizes the systems used in

the calculations.

Isomerization in the TpRu Complex. The isomerization has

been studied in fluorobenzene (FBZ), the solvent used in the

experiments. Scheme 7 shows all the calculated species and their

relative Gibbs energies, taking as zero energy the separated

[TpRu(k²P,N-PⁱPr₂CH₂Py)]⁺(0-Tp), Cl^ꢀ, and PhCtCCOMe

(A-alk). The optimized structures of selected Tp species relevant

for the alkynoneꢀvinylidene isomerization are shown in Figure 4.

The chloride precursor [TpRuCl(k²P,N-PⁱPr₂CH₂Py)] (1) is

much more stable that the separated species, indicating the

diﬃculty of releasing the halide. Indeed, the chloride abstraction

mol^ꢀ1above the separated reactants, respectively (routes A-Tp

and B-Tp, Scheme 7). Vinylidene complexes are much more

stable than the alkynones, giving the thermodynamic driving

force for the isomerization.

Structural and electronic changes along the 1,2-acyl migration

in the Tp complex are collected in Table 6. In comparison with

the migration in the metal-free system, the transition states in the

TpRu complex show more synchronous CꢀC bond-making and

bond-breaking processes, the C(O)ꢀC_Rdistance being only

slightly shorter than the C(O)ꢀC_βdistance. The C_RꢀC_βbond

distances of transition states are practically the same as in related

π-alkyne geometries.

The electrophilic nature of the 1,2-shift is also evident from the

charges shown in Table 6: the acyl migrates with a substantial

amount of positive charge. The presence of the metal allows a

better distribution of the negative charge created in C_β.

Starting from the most stable η¹-O-ketone rotamer 5b, the

overall activation barriers are 23.7 and 29.0 kcal mol^ꢀ1for routes

A-Tp and B-Tp, respectively. The Gibbs activation energy of

the A-Tp pathway is very close to the experimentally determined

ΔG^qfor the 5 f 13 isomerization (23.5 ( 2.5 kcal mol^ꢀ1

,

298

Table 5), stressing the reliability of the proposed mechanism.

Pathway A-Tp leads directly to the vinylidene rotamer 13a, in

which the carbonyl group of the acyl substituent is placed s-trans

with respect to the CdC bond of the vinylidene moiety and anti

4022

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Table 6. Structural and Electronic Changes along the 1,2-Acyl Migration in Complex 1

Tp-π-alk_a

TS(Tp-π-alk_af13a)

Tp-π-alk_b

TS(Tp-π-alk_af13b⁰)

d(C(O)ꢀC_R), Å

d(C(O)ꢀC_β), Å

d(C_RꢀC_β), Å

q(COMe)

1.475

2.634

1.259

+0.07

ꢀ0.06

+0.05

1.800

1.925

1.263

+0.24

+0.01

ꢀ0.00

1.463

2.606

1.269

+0.00

ꢀ0.11

+0.12

1.788

1.904

1.263

+0.20

+0.02

ꢀ0.01

q(C_R)

q(C_β)

Table 7. NBO Analysis of Intermediate Tp-agost

We have analyzed the electronic structure of intermediate Tp-

agost with the NBO method (Table 7). The σ(C(O)ꢀC_R)

orbital exhibits low occupancy, and the σ(C(O)ꢀC_R) to n*(C_β)

delocalization energy is quite signiﬁcant: 35.4 kcal mol^ꢀ1. Elec-

tron delocalization from this orbital to metal empty orbitals is

also important (entries 3ꢀ6 in Table 7).

a

NBO label

localized orbital

occ

energy, au

E_deloc

b

BD(1)

LP*(1)

BD*(1)

LP*(4)

LP*(5)

LP*(6)

σ(C(O)ꢀC_R)

n*(C_β)

1.779

0.683

0.320

0.438

0.208

0.191

ꢀ0.685 52

ꢀ0.189 25

+0.129 57

+0.733 07

+0.216 34

+0.238 35

35.4

7.6

5.0

2.6

4.4

σ*(RuꢀC_R)

n*(Ru)

To assess the role that agostic CꢀC intermediates could play

in the isomerization of internal alkynes and to test the stability of

this particular structure, we have performed optimizations chang-

ing the acyl group COMe to H, Me, and Ph groups (PhCtCR

alkynes, R = H, Me, Ph). When the group is H (terminal alkyne),

the structure evolves to the well-known η²-(CꢀH) agostic inter-

action. The optimization containing a Me group leads to the

π-alkyne structure, and that containing Ph group yields the

vinylidene species. It is not possible to ﬁnd such η²-(CꢀC)

agostic intermediates when the phenyl group is replaced by a

methyl (MeCtCCOMe alkynone). This observation points out

that electron delocalization toward the phenyl is also playing a

role in the stabilization of the agostic structure. This means that

Tp-agost-like species are in general not feasible intermediates;

the structure Tp-agost seems to be only reliable for alkynones with a

phenyl group. In any event, it is not a crucial intermediate, because

the reaction can also take place through π-alkyne complexes.

To sum up, the relative stability between η¹-ketone and

π-alkyne species in TpRuꢀalkynone complexes has been de-

monstrated, justifying the experimental detection of the η¹-

O-ketone adducts 5ꢀ8. The geometry of the most stable vinyl-

idene (13b) agrees with the experimental X-ray structure of 15. A

feasible pathway for the isomerization of the internal alkyne to

the vinylidene has been found (route A-Tp), involving a 1,2-acyl

direct shift from nondetected π-alkyne complexes and a Gibbs

energy barrier (23.7 kcal mol^ꢀ1) very similar to the experimental

barrier (23.5 kcal mol^ꢀ1). An alternative route from a η²-(CꢀC)

agostic intermediate (C-Tp) which does not involve π-alkynes

has also been described and cannot be discarded.

n*(Ru)

^aE_delocin kcal mol^ꢀ1

σ(C(O)ꢀC_R) orbital.

.

^bE_delocvalues are given with respect to the

with respect to the pyridine ligand. This is not the most stable

rotamer of vinylidene. The most stable orientation of the co-

ordinated vinylidene is found in 13b (CdO s-cis with respect to

the CdC bond and syn with respect to the pyridine), which also

corresponds to the arrangement found in the X-ray structure of

15 (Figure 2). By means of selected rotation processes, 13a can

easily evolve to 13b. We have calculated the rotational barriers for

the pathway connecting 13a with 13b, showing an overall barrier

of 8.5 kcal mol^ꢀ1, which is considerably lower than the Gibbs

energy of activation for the 1,2-acyl shift.

For the alkyneꢀvinylidene transformation of terminal alkynes

on ruthenium half-sandwich complexes, in addition to the direct

1,2-hydrogen shift, it has been proposed that the process can go

through a preliminary slippage to an η²-(CꢀH) agostic inter-

mediate which then undergoes the 1,2-hydrogen shift.^30b,39For

the isomerization of an internal alkyne the equivalent agostic

intermediate should be a η²-(CꢀC) agostic species. Agostic CꢀC

structures have been described,⁴⁰including with acyl ligands.^40a

We have been able to ﬁnd such a structure (Tp-agost, Figure 4).⁴¹

The intermediate Tp-agost lies 20.8 kcal mol^ꢀ1above 0-Tp and

19.6 kcal mol^ꢀ1above the most stable π-alkyne rotamer (Tp-π-

alk_b). The structural features of Tp-agost place it at an advanced

stage of the acyl shif: the C(O)ꢀC_Rbond distance is elongated to

1.583 Å, i.e., it is 0.167, 0.120, and 0.138 Å larger than in η¹-ketone

8b, π-alkyne 8b, and free alkynone A-alk, respectively, and in

addition the C(O)ꢀC_RꢀC_βangle has closed to 101°. To as-

certain whether this structure is an artifact of the calculation, we

have reoptimized it using another functional (M06). Indeed, it

remains as a minimum with similar geometrical parameters

(C(O)ꢀC_Rbond distance of 1.578 Å and C(O)ꢀC_RꢀC_βangle

of 101.2°) and relative Gibbs energy (17.7 kcal mol^ꢀ1above Tp-

π-alk_b). From this intermediate the isomerization process (route

C-Tp) easily evolves through TS(Tp-agostf13b), yielding di-

rectly the most stable vinylidene rotamer 13b. This process has an

overallactivationbarrier (from5btoTS(Tp-agostf13b)) of 25.6

kcal mol^ꢀ1. This barrier is only 1.9 kcal mol^ꢀ1higher than that of

route A-Tp involving the π-alkyne intermediate and cannot be

discarded as a feasible pathway.

Isomerization in the Cp*Ru Complex. For the Cp*Ru com-

plex, we have studied the three pathways described for the TpRu

complex. The isomerization has been studied in methanol, the

solvent used in the experiments. Scheme 9 shows all the cal-

culated species and their relative Gibbs energies, taking as zero

energy the separated [Cp*Ru(k²P,N-PⁱPr₂CH₂Py)]⁺(0-Cp*),

Cl^ꢀ, and PhCtCCOMe (A-alk). The optimized structures of

selected Cp* species relevant for the alkynoneꢀvinylidene iso-

merization are shown in Figure 5. Table 8 collects structural data

as well as charge distributions using NBO calculations.

According to theoretical calculations, the energy of chloride

precursor 3 is quite similar to that of the separated species (0-Cp*

+ Cl^ꢀ+ A-alk), in agreement with the spontaneous release of the

halide in methanol solution. The π-alkyne intermediates (9a,b)

are more stable than the separated species, both rotamers having

4023

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Scheme 9. AlkynoneꢀVinylidene Isomerization for Cp* Complexes 3^a

^aΔG_methanolvalues given in kcal mol^ꢀ1with respect to the separated species.

similar energies, in agreement with the experimental detection of

two rotamers in solution. Moreover, the energy of η¹-keto alkyne

intermediates (Cp*-η¹-O) is too high to allow their detection

in solution. The isomerization process takes place from the

π-alkyne through a sigmatropic 1,2-shift of the acyl group.

Two transition states similar to those of the TpRu system

(Scheme 8) were located, TS(9afCp*-vinyl_a) and TS-

(9bfCp*-vinyl_b) (routes A-Cp* and B-Cp*, respectively),

depending on the π-alkyne rotamer. Starting from the corre-

sponding π-alkyne species, 9a,b, the activation barriers for each

process are 32.7 and 28.8 kcal mol^ꢀ1, respectively. Both barriers

for the isomerization in the Cp*Ru complex with a direct 1,2-acyl

migration mechanism are notably higher than those for the TpRu

complex (23.7 kcal mol^ꢀ1). An alternative route (C-Cp*) involv-

ing intermediate Cp*-agost, similar to Tp-agost, was investi-

gated, but no such intermediate was located in the Cp*Ru

complex. It appears that the possibility of a η²-(CꢀC) inter-

mediate is limited to the TpRu complex with the PhCtCCOMe

alkynone.

Tpꢀπ-alkyne complexes (1.259 and 1.269 Å), suggesting a

stronger π-bond/metal interaction in the Cp* complexes, as

the π-alkyne binding energies to the unsaturated metal fragments

0-Tp and 0-Cp* conﬁrm.

As for the Tp complex, in the Cp* system the increase of

positive charge in the migratory group COMe (from reactant to

transition state) substantiates the electrophilic nature of the

process. The charge increase in C_Rand the decrease in C_β

support this statement.

In conclusion, calculations conﬁrm the capability of the Cp*Ru

derivatives [Cp*RuCl(k²P,N-ⁱPr₂PXPy)] (X = CH₂(3), NH

(4)) to release the halide spontaneously and the calculations

justify the nondetection of η¹-O-ketone adducts. Both isomer-

ization routes A-Cp* and B-Cp* demand too much energy, and

they should not occur under standard reaction conditions. The

π-alkyne complexes are the only accessible species in this system,

in agreement with the experimental detection of complexes

9ꢀ11 and the nondetection of vinylidenes Cp*-vinyl.

Ligand Effects in the Isomerization of Internal Alkynes: Tp vs

Cp*. The Gibbs energy profiles depicted in Figure 6 describe the

most favorable pathways for the isomerization of the internal

alkynone PhCtCCOMe in the Tp and Cp* complexes in the

solvent used in the experiments (fluorebenzene and methanol,

respectively). Additional ΔG profiles comparing Cp* and Tp

complexes using the same solvent have been included in the

The calculated ν(CtC) absorptions for the two rotamers of

9 (9a, 1858 cm^ꢀ1; 9b, 1839 cm^{ꢀ1 37}

are in good agreement

)

with the experimental IR ν(CtC) absorptions of complex

12 (1828 cm^ꢀ1) and are consistent with a strong π-bond/metal

interaction. The CtC distances in the Cp*ꢀπ-alkyne com-

ꢀ

plexes (1.279 and 1.282 Å) are more elongated than in the

4024

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

the Cp* ligand, an energy decomposition analysis (EDA) was

carried out to compare both coordination modes (Tables 9 and

10). The most stable isomers were selected for the study:

Cp*-η¹-O_a and 9a as η¹-O-ketone and π-alkyne complexes,

respectively, in the Cp* system and 5b (η¹-O-ketone) and Tp-

π-alk_b (π-alkyne) in the Tp system.

In the EDA the bonding energy (ΔE_b) between the unsatu-

rated metal fragment [LRu(k²P,N-ⁱPr₂PCH₂Py)]⁺(L = Tp,

Cp*) and the alkynone in each coordination mode is decom-

posed into two main terms: one accounts for the distortion from

their free geometries to those they adopt in the complexes

(ΔE_dist), the other one concerns the interaction energy between

the distorted fragments to form the complex structure (ΔE_inter).

The latter in turn can be split into several contributions: ΔE_Pauli

comprises the destabilizing interactions between occupied orbi-

tals and is responsible for any steric repulsion; ΔE_elstcorresponds

to the classical electrostatic interaction; ΔE_oiaccounts for

electron pair bonding, charge transfer, and polarization.

ΔE_b¼ ΔE_dist+ ΔE_int

where

Figure 5. Optimized structures of selected species of the alkynoneꢀ

vinylidene isomerization in the Cp*Ru complex. Bond distances are

given in Å and angles in deg. Hydrogen atoms on Tp and P,N ligands

have been removed for clarity.

ΔE_int¼ ΔE_Pauli+ ΔE_elst+ ΔE_oi

Table 9 collects the diﬀerent energy contributions of Cp*-

related species. The distortion energy ΔΔE_distfavors the

η¹-ketone Cp*-η¹-O_a over π-alkyne 9a by 28.0 kcal mol^ꢀ1

.

Table 8. Structural and Electronic Changes along the 1,2-

Acyl Migration in the Cp* Complex

According to ΔΔE_inter, ΔΔE_Paulialso encourages coordination

of the alkynone in a η¹-ketone mode by 133.7 kcal mol^ꢀ1

but ΔΔE_elstand ΔΔE_oiterms support π-alkyne 9a by 88.1 and

82.6 kcal mol^ꢀ1. The overall energy ΔΔE_bfavors the π-alkyne

coordination mode by 9.0 kcal mol^ꢀ1due to strong electrostatic

and orbital contributions.

TS

9a (9afCp*-vinyl_a) 9b (9bfCp*-vinyl_b)

d(C(O)ꢀC_R), Å 1.456

d(C(O)ꢀC_β), Å 2.614

1.787

1.897

1.265

+0.18

ꢀ0.01

ꢀ0.00

1.458

2.604

1.282

ꢀ0.02

ꢀ0.13

+0.08

1.752

1.871

1.265

+0.12

ꢀ0.02

+0.01

For Tp complexes (Table 10), ΔΔE_distshows the same trend

as for Cp* analogues: i.e., η¹-ketone 5b is preferred over π-alkyne

Tp-π-alk_b by 27.1 kcal mol^ꢀ1. ΔΔE_Paulifavors η¹-ketone 5b by

106.4 kcal mol^ꢀ1, whereas ΔΔE_elstand ΔΔE_oiterms encourage

the π-alkyne bonding mode Tp-π-alk_b by 64.3 kcal mol^ꢀ1. The

overall energy ΔE_bslightly favors η¹-ketone coordination mode

d(C_RꢀC_β), Å

q(COMe)

q(C_R)

1.279

ꢀ0.00

ꢀ0.10

+0.06

q(C_β)

over the π-alkyne by 4.9 kcal mol^ꢀ1

.

Supporting Information and show essentially the same behavior.

The thermodynamics for the reaction from the separated species

(0-Tp and 0-Cp*) are very similar in both systems, the vinylidene

being in both cases about 12 kcal mol^ꢀ1more stable. The most

apparent features of the plots in Figure 6 are the different sta-

bilities of η¹-ketone and π-alkyne species in both systems and the

differences in the barrier height for the 1,2-acyl migration. As the

migratory aptitude of the substituent is the same in both systems

because it is always a COMe acyl, the ligand (Tp or Cp*) decides

the fate of the alkyne in the coordination sphere of the ruthe-

nium. The increased stability of the π-alkyne combined with a

higher barrier for its isomerization to vinylidene avoids the trans-

formation in Cp*Ru complexes. In contrast, destabilization of the

π-alkyne and a lower barrier for the isomerization of the internal

alkyne allow vinylidene formation in TpRu complexes even for

alkyne substituents with lower migration aptitude than acyls. The

relative energy of the 1,2-acyl migration transition state with

respect to the separated reactants is about 5 kcal mol^ꢀ1lower for

the TpRu complex, indicating better suitability of the TpRu

fragment for stabilizing this TS.

According to the aforementioned data, the driving force which

decides the relative stability is the interaction contribution. For

the Cp* system, ΔE_intercan counterbalance ΔE_distto form a

π-alkyne structure; for the Tp system the interaction is substan-

tially reduced in the π-alkyne mode and increased in the

η¹-ketone mode.

When each coordination mode is compared for Tp and Cp*

systems (Table 11), the distortion trend ΔΔE_distis similar

between both of them; Tp complexes are more stabilized than

Cp* species by approximately 2ꢀ3 kcal mol^ꢀ1. Relating the

interaction contribution in η¹-ketone complexes, ΔΔE_interfavors

Tp-5b by 6.8 kcal mol^ꢀ1. Although ΔΔE_Pauliprefers Cp*-η¹-

O_a, ΔΔE_elstand ΔΔE_oifavor Tp-5b. On the other hand,

π-alkyne complexes show the opposite trend: Cp*-9a is preferred

by 8.0 kcal mol^ꢀ1. ΔΔE_Pauliencourages Tp-π-alk_b, but this

stabilization is canceled by electrostatic and orbital contributions.

These results show the inﬂuence of the ligands according to

the coordination mode. Electrostatic and orbital contributions

favor η¹-ketone interaction in Tp complexes and π-alkyne

interaction in Cp* complexes. The opposite trend occurs for

the Pauli term. As an overall result, ΔΔE_bencourages η¹-ketone

formation in Tp complexes by 8.6 kcal mol^ꢀ1and favors π-alkyne

To further analyze why the η¹-O-ketone coordination mode is

preferred with the Tp ligand and the π-alkyne is preferred with

4025

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Figure 6. Gibbs energy proﬁles (kcal mol^ꢀ1) for the most favorable pathways for the isomerization of the internal alkynone PhCtCCOMe in the Tp

(ﬂuorebenzene solvent) and Cp* complexes (methanol solvent). Results for the Cp complex in methanol are also included for comparison.

Table 9. Energy Decomposition Analysis for Cp* Complexes^a

Table 11. Energy Decomposition Analysis Comparison be-

tween Tp and Cp* Complexes^a

ΔE_η1-ketone(Cp*-η¹-O_a) ΔE_π-alkyne(9a) ΔΔE^b

ΔΔE

ΔE_dist[Cp*(PN)Ru]⁺

5.4

1.0

11.6

22.8

ꢀ6.2

ꢀ21.8

ꢀ28.0

ꢀ133.7

88.1

η¹-ketone

π-alkyne

TS TS(9b-Cp*-vinyl_b)ꢀ

TS(Tp-π-alk_a-13a)

ΔE_dist(PhCCCOMe)

(Cp*-η¹-O_a)ꢀ5b 9aꢀ(Tp-π-alk_b)

ΔE_dist

ΔE_Pauli

ΔE_elst

ΔE_oi

6.4

34.4

62.0

195.7

ΔE_dist

ΔE_Pauli

ΔE_elst

ΔE_oi

1.8

ꢀ6.8

8.0

2.7

20.5

1.7

13.4

ꢀ3.3

ꢀ6.2

3.9

ꢀ53.6

ꢀ32.9

ꢀ24.5

ꢀ18.1

ꢀ141.7

ꢀ115.5

ꢀ61.5

ꢀ27.1

82.6

ꢀ15.8

ꢀ12.7

ꢀ8.0

ꢀ5.3

ΔE_inter

ΔE_b

37.0

5.6

9.0

ΔE_inter

6.8

^aGas-phase energies are given in kcal mol^ꢀ1. ^bΔΔE = ΔE(η¹-ketone) ꢀ

ΔE_b

8.6

5.7

ΔE(π-alkyne).

^aGas-phase energies are given in kcal mol^ꢀ1

.

Table 10. Energy Decomposition Analysis for Tp

Complexes^a

both transition states (Table 11). In the Cp*Ru complex the larger

steric repulsion (ΔΔE_Pauli= 13.4 kcal mol^ꢀ1) cannot be counter-

balanced by the electrostatic and orbital interactions, placing this

transition state about 5 kcal mol^ꢀ1above that of the TpRu complex.

As previously reported complexes able to achieve the vinyli-

dene formation bear Cp ligands,^13ꢀ15we wondered about the

behavior that the hypothetical Cp complex [CpRuCl(k²P,

N-ⁱPr₂PXPy)]⁺would exhibit. We have studied the isomerization

of PhCtCCOMe in this complex, the main results being shown

in Figure 6 (optimized structures collected in the Supporting

Information). As expected, the relative energy of the transition

state with respect to separated reactants is the lowest in the Cp

complex (17.7 kcal mol^ꢀ1, 8 kcal mol^ꢀ1lower than for the Cp*

complex). However, the π-alkyne intermediate is stabilized by the

ΔE_η1-ketone(5b) ΔE_π-alkyne(Tp-π-alk_b) ΔΔE^b

ΔE_dist[Tp(PN)Ru]⁺

2.2

2.4

11.8

19.9

ꢀ9.6

ꢀ17.5

ꢀ27.1

ꢀ106.4

64.3

ΔE_dist(PhCCCOMe)

ΔE_dist

ΔE_Pauli

ΔE_elst

ΔE_oi

4.6

31.7

68.8

175.2

ꢀ61.6

ꢀ38.5

ꢀ31.3

ꢀ26.7

ꢀ125.9

ꢀ102.8

ꢀ53.5

ꢀ21.8

64.3

ΔE_inter

22.2

ΔE_b

ꢀ4.9

^aGas-phase energies are given in kcal mol^ꢀ1. ^bΔΔE = ΔE(η¹-ketone) ꢀ

ΔE(π-alkyne).

same amount, leading to almost the same barrier (28 kcal mol^ꢀ1

)

in both systems and thus hampering the isomerization also in this

CpRu complex. Summing up, the isomerization of internal alkynes

requires both stabilization of the transition state of the 1,2-mig-

ration and destabilization of the π-alkyne intermediate. The

nature of the migrating group inﬂuences the transition state, but

the nature of the metal ligands aﬀects both the transition state

and π-alkyne stabilities.

in Cp* complexes by 5.3 kcal mol^ꢀ1. This means that the donor

ability of the Cp* ligand increases the electronic density on the

metal, encouraging back-bonding in π-acceptor alkyne and,

therefore, destabilizing σ-donor η¹-ketone.

The reasons for the lower relative energy of the transition state

in the Tp system are also apparent when the EDA is applied to

4026

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

i

’ CONCLUSIONS

ligands Pr₂PCH₂Py,^{19 i}Pr₂PNHPy, and Pr₂PSPy⁴³were respectively

prepared as described in previous papers by our group, following suitable

adaptations of published procedures.⁴⁴The complexes [Cp*RuCl(k²P,

N-ⁱPr₂PXPy)] (X = CH₂(3),¹⁹NH (4)⁴³) and the NaB(Ar^F)₄salt

(Ar^F= 3,5-CF₃C₆H₃)⁴⁵were synthesized according to reported meth-

ods. All internal alkynes were purchased from Aldrich and directly

employed without further puriﬁcation. IR spectra were recorded from

Nujol mulls on a Perkin-Elmer FTIR Spectrum 1000 spectropho-

tometer. NMR spectra were recorded on Varian Inova 400 and 600 MHz

and Varian Gemini 300 MHz equipment. Chemical shifts are given

in parts per million from SiMe₄(¹H and ¹³C{¹H}) or 85% H₃PO₄

(³¹P{¹H}). ¹H and ¹³C{¹H} NMR signal assignments were conﬁrmed

The participation of vinylidenes in the catalytic transformation

of internal alkynes has been generally ruled out due to the

consensus that internal alkynes did not isomerize to vinylidene.

Reports by Shaw¹³and Ishii et al.^14,15recently called our atten-

tion to this new ﬁeld involving CꢀC activation by mononuclear

ruthenium complexes.¹⁸

In this paper, we have completed the initial kinetic studies in

solution and in the solid state. From an experimental point of view

wehave analyzedhow the diﬀerentmodiﬁcations on the transition-

metal ligands aﬀect this transformation, namely (a) the pyridyl-

phosphine linking group (X = NH, S, CH₂), (b) the alkyne

substituents with and without ketone groups, and (c) the classical

comparison between Cp- and Tp-type ligands (in this case be-

tween Cp*Ru and TpRu systems), in addition to the minor eﬀect

of the counterion (BPh₄and B(Ar^F)₄) in Cp*Ru complexes.

In addition to the relatively important eﬀects of modiﬁcations

a and b on the isomerization rate of the TpRuꢀη¹-Od

C(R)CtCPh intermediate into TpRudCdC(Ph)COR, the

most intriguing feature is the divergent behavior observed for

the analogous Cp*Ru system, which yields stable π-alkyne

complexes, not observed in most of the TpRu complexes (only

as very minor products during the isomerization of some of

them).¹⁸The kinetic measurements, both in solution and in the

solid state, are consistent with an intramolecular concerted

mechanism via a sigmatropic 1,2-C shift.

A detailed computational study of the process, which explains

the contrasting behavior of Cp*Ru, has been performed with the

actual experimental complexes in order to analyze the mecha-

nism, showing (a) the electrophilic nature of the acyl 1,2-shift,

(b) the weaker metalꢀπ-alkyne interaction in TpRu complexes

destabilizing such intermediates, (c) the lower barrier for the iso-

merization in the TpRu complex, and (d) the good agreement

between the calculated Gibbs activation energy and that deter-

mined experimentally by the kinetic measurements. Calculations

on Cp*Ru show that π-alkyne complexes are the only accessible

species, being more stable than η¹-O-ketone adducts, due to the

increased electrostatic and orbital interactions in the Cp*Ru sys-

tem. The routes to vinylidene demand too much energy due to

the large steric repulsion in the transition state of the migration.

In summary, the computational analysis of the internal alkyne

to vinylidene isomerization in the coordination sphere of TpRu

and Cp*Ru systems provides a sound explanation of a series of

divergent experimental facts. The isomerization of internal

alkynes requires both low-energy transition states for the 1,2-

migration and destabilization of the π-alkyne intermediates. The

nature of the metal ligands has impact on both factors. A better

understanding of the factors controlling the process should lead

to the design of suitable transition-metal systems able to repro-

duce this behavior in a reversible fashion.⁴²This would constitute

a great step into the future catalytic transformations of internal

alkynes.

by ¹H-gCOSY, 135-DEPT, and gHSQC(¹Hꢀ C) experiments. ¹H and

13

¹³C{¹H} NMR signals corresponding to BPh₄and B(Ar^F)₄anions are

omitted for clarity. Microanalysis was performed on a LECO CHNS-932

elemental analyzer at the Servicio Central de Ciencia y Tecnología,

Universidad de Cꢀadiz.

TpRuCl(j²P,N-PⁱPr₂CH₂Py)] (1). The starting complex [TpRuCl-

(PPh₃)₂] (874 mg, 1 mmol) was suspended in 20 mL of toluene. After

addition of 209 mg (1 mmol) of PⁱPr₂CH₂Py, the mixture was stirred at

65 °C over 16 h. Then, the suspension was warmed to room temperature

and a solid was formed. The solvent was partially removed under

vacuum, and petroleum ether (20 mL) was added to complete the preci-

pitation, and the resulting yellow microcrystalline solid was filtered,

washed with petroleum ether, and dried under vacuum.

Yield: 414 mg (74%). Anal. Calcd for C₂₁H₃₀BClN₇PRu: C, 45.1; H,

5.41. Found: C, 45.1; H, 5.43. IR (Nujol, cm^ꢀ1): ν(BH) 2479,

ν(CdC)/ν(CdN) 1599. ¹H NMR (400 MHz, CD₂Cl₂, 298 K):

δ 0.68, 1.24, 1.31, and 1.51 (m, 3H each, PCH(CH₃)₂), 1.83 and 3.48

(m, 1H each, PCH(CH₃)₂), 3.66 (dd, 1H, ²J_HH= 15.6 Hz, ²J_HP= 10.7

Hz, PCH^aH^b), 4.19 (dd, 1H, ²J_HH= 15.6 Hz, ²J_HP= 10.0 Hz, PCH^aH^b),

6.03 and 6.20 (t, 1H each, ³J_HH= 2.0 Hz, Tp), 6.28 (m, 1H, Tp), 6.39 (d,

1H, ³J_HH= 2.0 Hz, Tp), 6.83 (t, 1H, ³J_HH= 6.0 Hz, Py), 7.50 (m, 2H,

Py), 7.75, 7.79, and 7.81 (d, 1H each, ³J_HH= 1.6 Hz, Tp), 7.77 (m, 2H,

Tp), 7.92 (d, ³J_HH= 5.6 Hz, Py). ³¹P{¹H} NMR (161.89 MHz, CD₂Cl₂,

298 K): δ 77.4. ¹³C{¹H} NMR (100.58 MHz, CD₂Cl₂, 298 K): δ 18.53,

19.27, 19.71, and 19.94 (d, ²J_CP= 4.7 Hz, PCH(CH₃)₂), 25.24 (d, ¹J_CP

=

15.5 Hz, PCH(CH₃)₂), 26.12 (d, ¹J_CP= 17.9 Hz, PCH(CH₃)₂), 38.03

(d, ¹J_CP= 22.0 Hz, PCH₂), 105.2, 105.7, and 105.9 (s, Tp), 121.6 (s, C⁵

Py), 122.5 (d, ³J_CP= 8.7 Hz, C³Py), 142.5 and 145.8 (s, Tp), 143.9 (s, C⁴

Py), 155.3 (s, C⁶Py), 166.2 (d, ²J_CP= 5.5 Hz, C²Py).

Synthesis of η¹-O-Alkynyl-Ketone Complexes. In a typical

preparation, a solution of complex 1 (0.15 g, 0.27 mmol) or 2 (0.15 g,

0.26 mmol) in 10 mL of fluorobenzene was treated with a slight excess of

the corresponding internal alkyne (0.30 mmol: 46 μL of 4-phenyl-3-

butyn-2-one, 62 mg of diphenylpropynone), and NaBAr^F₄(0.30 mmol,

266 mg). The reaction mixture was stirred in ice bath for 30 min in the

case of the 4-phenyl-3-butyn-2-one derivatives 5 and 7, and for 2 h at

room temperature for the diphenylpropynone complexes 6 and 8. The

initial yellow solution slowly turns to darker colors ranging from green to

violet. The mixture was then filtered though Celite to remove the NaCl

salt, and the volatiles were removed under vacuum. The oily residue was

washed with petroleum ether (2 ꢁ 10 mL) and taken to dryness to yield

the corresponding product as solids.

[TpRu{η¹-OdC(Me)CtCPh}(j²P,N-PⁱPr₂CH₂Py)][BAr^F] (5).

4

Yield: 310 mg (75%). Anal. Calcd for C₆₃H₅₀B₂F₂₄N₇OPRu: C 49.4, H

3.29; Found: C 49.4, H 3.28. IR (Nujol, cm^ꢀ1): ν(BH) 2488, ν(CꢂC)

2204, 2160, ν(CdN)/ν(CdC)/ν(CdO) 1664, 1610, 1576. ¹H NMR

(400 MHz, CDCl₃, 298 K): δ 0.81, 0.97, 1.28, and 1.37 (m, 3H each,

PCH(CH₃)₂), 2.02 and 2.56 (m, 1H each, PCH(CH₃)₂), 2.27 (s, 3H,

COCH₃), 3.80 (m, 2H, PCH₂), 6.07 and 6.22 (t, 1H each, ³J_HH= 2.0 Hz,

Tp), 6.34 (br s, 1H, Tp), 6.44 (d, 1H, ³J_HH= 2.0 Hz, Tp), 7.03 (m, 1H,

’ EXPERIMENTAL SECTION

All synthetic operations were performed under dry dinitrogen or

argon using conventional Schlenk techniques. Tetrahydrofuran, diethyl

ether, and petroleum ether (boiling point range 40ꢀ60 °C) were ob-

tained oxygen- and water-free from an Innovative Technology, Inc.,

solvent puriﬁcation apparatus. Fluorobenzene, methanol, 1,2-dichloro-

ethane, and other solvents were of anhydrous quality and were used as

received. All solvents were deoxygenated immediately before use. The

Py), 7.05 (d, 2H, ³J_HH= 7.6 Hz, Ph), 7.35 ꢀ 8.00 (m, 22H, Tp + Ph +

3

BAr^F+ Py), 8.17 (d, 1H, J_HH= 5.8 Hz, Py). ³¹P{¹H} NMR

4

4027

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

(161.89 MHz, CDCl₃, 298 K): δ 71.7 (s). ¹³C{¹H} NMR (150.81 MHz,

CDCl₃, 283 K): δ 17.11, 18.04, 18.78, and 18.92 (s, PCH(CH₃)₂), 24.13

(d, ¹J_CP= 19.8 Hz, PCH(CH₃)₂), 26.33 (d, ¹J_CP= 21.3 Hz, PCH(CH₃)₂),

34.38 (s, CH₃CO), 35.10 (d, ¹J_CP= 21.4 Hz, PCH₂), 87.42 and 106.7 (s,

CꢂCPh), 106.3, 106.8, and 106.9 (s, Tp), 118.3 (s, C⁵Py), 120.0 ꢀ 138.0

(s, Ph + Tp + BAr^F₄+ Py), 141.7 and 146.2 (s, Tp), 144.3 (s, C⁴ꢀPy),

154.8 (s, C⁶Py), 164.8 (d, ²J_CP= 4.3 Hz, C²Py), 198.3 (s, COCH₃).

Synthesis of η²-Alkyne Complexes. In a typical preparation,

0.15 g of complex 3 or 4 (0.31 mmol) was dissolved in 10 mL of meth-

anol, in addition to a slight excess of the corresponding internal alkyne

(0.35 mmol: 54 μL of 4-phenyl-3-butyn-2-one, 72 mg of diphenylpropy-

none, 53 μL of methyl phenylpropiolate, 42 μL of ethyl 2-butynoate), and

NaBPh₄(0.35 mmol, 120 mg). A yellow precipitate was immediately

formed. The suspension was stirred for 10 min at room temperature. The

mixture was filtered and the solid washed with small amounts of ethanol

(<5 mL) and petroleum ether (2 ꢁ 10 mL) and then taken to dryness to

yield the corresponding products as orange-yellow microcrystalline solids.

[Cp*Ru(η²-PhCtCCOMe)(k²P,N-PⁱPr₂CH₂Py)][BPh₄] (9). Yield: 262

mg (93%). Anal. Calcd for C₅₆H₆₃BNOPRu: C 74.0, H 6.99; Found: C

74.0, H 6.98. IR (Nujol, cm^ꢀ1): ν(η²-CtC) 1828, ν(CO) 1644, ν(Ph)

1579 cm^ꢀ1. ¹H NMR (400 MHz, CD₂Cl₂, 298 K): Major product δ 0.51,

1.03, 1.45, and 1.76 (m, 3H each, PCH(CH₃)₂), 1.49 (d, ⁴J_HP= 1.3 Hz,

15H, C₅(CH₃)₅), 2.59 (s, 3H, COCH₃), 1.93 and 2.81 (m, 2H, PCH-

(CH₃)₂), 3.02 (dd, 1H, ²J_HH= 17 Hz, ²J_HP= 7.2 Hz, PCH^aH^b), 3.43 (dd,

1H, ²J_HH= 17 Hz, ²J_HP= 4.8 Hz, PCH^aH^b), 6.90 ꢀ 7.50 (m, 27 H, Ph +

BPh₄+ Py), 7.57 (t, 1H, ³J_HH= 7.7 Hz, Ph), 8.02 (d, 1H, ³J_HH= 6.0 Hz,

Py). ³¹P{¹H} NMR (161.9 MHz, CD₂Cl₂, 298 K): δ 66.4 (s, 15%) and

69.8 (s, 85%). ¹³C{¹H} NMR (75.4 MHz, CD₂Cl₂, 298 K): Major

product δ 9.40 (s, C₅(CH₃)₅), 18.21 ꢀ 19.90 (m, PCH(CH₃)₂), 26.42

2-tetrachloroethane-d₂were prepared and immediately cooled to

213 K. The sample was warmed to room temperature just before being

inserted into the probe of the Varian UNITY-400 spectrometer at 298 K.

Once the shims were adjusted, the probe was heated to the desired

temperature. The NMR temperature controller was previously cali-

brated against a methanol sample, the reproducibility being (1 °C.

³¹P{¹H} NMR spectra were recorded for at least 3 half-lives at regular

intervals using the spectrometer software for accurate time control. Peak

intensities were analyzed from stacked plots of the ³¹P{¹H} NMR

spectra. First-order rate constants were derived from the least-squares best-

fit lines of the ln(I₀/I_t) vs time plots (see the Supporting Information). The

uncertainty in the isomerization rate constants represents the standard

deviation derived from the slope of the best-fit line. Uncertainties in the

activation enthalpies and entropies were calculated from the uncertainties

in the slope and intercept of the best-fit lines of the Eyring plots and the

error propagation formulas derived from the Eyring equation.

Kinetic Studies of the Alkynyl Ketone to Vinylidene Solid-

State Isomerization. Solid samples of the alkynyl ketone 5 were

prepared as Nujol mulls between two NaCl crystal windows. The

samples were kept in an oven at the desired temperature (37, 45, and

50 °C). As each sample was taken out, the IR spectrum was recorded at

room temperature, and then the sample was returned to the oven. The

R (fraction of transformed solid) vs time curves were obtained by moni-

toring the decrease of the integrated intensity of the ν(CtC) IR band

for each compound as a function of time. The value R at a given t instant

was calculated from: R(t) = (I₀ꢀ I_t)/I₀. Rate constants (k) and the

values of n were derived from least-squares fits of ln(ln(1/(1 ꢀ R))) vs

ln t plots (AvramiꢀErofeev equation). The uncertainty in the value of n

and in the isomerization rate constants represents 1 standard deviation,

derived from the slope and the intercept of the best-fit lines.

Crystal Structure Analysis. Crystals of 2, 12, 15, and 16 suitable

for X-ray structural determination were mounted on glass fibers and

then transferred to the cold nitrogen gas stream of a Bruker Smart APEX

CCD three-circle diffractometer (T = 100 K) with a sealed-tube source

and graphite-monochromated Mo KR radiation (λ = 0.710 73 Å) at the

Servicio Central de Ciencia y Tecnología de la Universidad de Cꢀadiz. In

each case, four sets of frames were recorded over a hemisphere of the

reciprocal space by ω scans with δ(ω) = 0.30° and an exposure of 10 s

per frame. Correction for absorption was applied by scans of equivalents

using the SADABS program.⁴⁶An insignificant crystal decay correction

was also applied. The structures of 2, 12, and 16 were solved by direct

methods. The structure of 15 was solved by vectorial methods. All the

structures were refined on F²by full-matrix least squares (SHELX97)⁴⁷

by using all unique data. All non-hydrogen atoms were refined aniso-

tropically with hydrogen atoms included in calculated positions (riding

model). Three disordered CF₃groups in compound 16 were refined

split in two complementary orientations using displacement parameter

restraints. The program ORTEP-3 was used for plotting.⁴⁸

1

(d, J_CP= 16.7 Hz, PCH(CH₃)₂), 29.51 (d, J_CP= 26.0 Hz, PCH-

(CH₃)₂), 33.15 (s, CH₃CO), 37.92 (d, ¹J_CP= 22.5 Hz, PCH₂), 86.10 (d,

2

²J_CP= 9.6 Hz, CtC), 98.80 (d, J_CP= 2.2 Hz, C₅(CH₃)₅), 116.8 (d,

²J_CP= 4.8 Hz, CtC), 121.8 ꢀ 136.0 (s, Ph + BPh₄+ Py), 138.9 (s, ipso-

Ph), 139.4 (s, C⁴-py), 157.8 (s, C⁶-py), 163.6 (d, ²J_CP= 5.8 Hz, C²-py),

190.9 (s, COCH₃).

Synthesis of Vinylidene Complexes. These complexes were

prepared by the same procedure employed for their isomeric complexes

5 ꢀ 8, but stirring the solution mixture during 1 h at 50 °C for the

4-phenyl-3-butyn-2-one derivatives 13 and 15, and for 48 h at 50 °C in

the case of the diphenypropynone complex 14. Compounds 16 ꢀ 18

were heated for 20 h at 75 °C. The workup yields the corresponding

complexes as dark brown solids. Suitable X-ray single crystals of 14 have

been grown by slow diffusion of petroleum ether into a diethylether

solution of the complex at 4 °C.

[TpRu{dCdC(COMe)Ph}(k²P,N-PⁱPr₂CH₂Py)][BAr^F] (13). Yield: 303

4

mg (73%). Anal. Calcd for C₆₃H₅₀B₂F₂₄N₇OPRu: C, 49.4, H, 3.29. Found:

C, 49.2, H, 3.25. IR (Nujol, cm^ꢀ1): ν(BH) 2497, ν(CdN)/ν(CdC)/

ν(CdO) 1666, 1609, 1582, 1566. ¹H NMR (400 MHz, CDCl₃, 298 K):

δ 0.62, 1.00, 1.27, and 1.37 (m, 3H each, PCHCH₃), 1.90 and 2.80 (m, 1H

each, PCHCH₃), 1.84 (s, 3H, COCH₃), 3.72 (m, 2H, PCH₂), 6.15 and

6.19 (t, 1H each, ³J_HH= 2.0 Hz, Tp), 6.33 and 6.57 (br s, 1H each, Tp),

6.72 (d, 2H, ³J_HH= 7.2 Hz, Ph), 7.00ꢀ7.91 (m, 23H, Tp + Ph + B(Ar^F)₄+

Py), 8.51 (d, 1H, ³J_HH=6.0Hz, Py). ³¹P{¹H} NMR (161.89 MHz, CDCl₃,

298 K): δ 64.3 (s). ¹³C{¹H} NMR (75.45 MHz, CDCl₃, 298 K): δ 17.11

and 19.18 (both d, ²J_CP= 5.6 Hz, PCH(CH₃)₂), 17.49 and 18.63 (both s,

Table 12 summarizes the crystal data and data collection and reﬁne-

ment details for 2, 12, 15, and 16. An ORTEP diagram of 2 can be found

in the Supporting Information. CCDC 815989ꢀ815992 contain sup-

plementary crystallographic data for this paper. These data can be

obtained free of charge from The Cambridge Crystallographic Data

Centre via www.ccdc.cam.ac.uk/data_request/cif.

Computational Details. All calculations were performed at the

DFT level by means of the PBE0 functional⁴⁹as implemented in Gaussian

09.⁵⁰This functional correctly described the uniform electron gas (UEG)

limit, which accurately reproduces agostic bonding situations.⁵¹Test

calculations were performed with the M06 functional.⁵²The Ru atom

was described using the scalar-relativistic StuttgartꢀDresden SDD

pseudopotential⁵³and its associated double-ζ basis set complemented

with a set of f-polarization functions.⁵⁴The 6-31G** basis set was used

for the H⁵⁵and B, C, N, O, P, and Cl atoms.⁵⁶Diffuse functions were

added for O and Cl atoms⁵⁷in all calculations. Solvent effects were

PCH(CH₃)₂), 25.23 (d, ¹J_CP= 20.3 Hz, PCH(CH₃)₂), 28.56 (d, ¹J_CP

=

1

26.9 Hz, PCH(CH₃)₂), 29.92 (s, CH₃CO), 36.38 (d, J_CP= 27.4 Hz,

PCH₂), 106.9, 107.3, and 107.4 (br s, Tp), 124.3 (s, C⁵Py), 124.6 (d,

³J_CP= 8.0 Hz, C³Py), 128.0 ꢀ 130.5 (s, Ph + B(Ar^F)₄), 131.8 (s,

RudCdC), 136.4, 137.4, and 137.8 (s, Tp), 140.5 and 142.5 (s, Tp), 141.0

(s, C⁴Py), 147.1 (s, Tp), 153.3 (s, C⁶Py), 161.7 (d, ²J_CP= 3.8 Hz, C²Py),

196.6 (s, COCH₃), 357.3 (d, ²J_CP= 18.2 Hz, RudC).

Kinetics Studies of the Alkynyl Ketone to Vinylidene Iso-

merization. NMR samples of the alkynyl ketones 5ꢀ7 in 1,1,2,

4028

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

Table 12. Crystal Data and Details of the Structure Determination of Compounds 2, 12, 15, and 16

2

12

15

16

Crystal Data

formula

C

₂₀H₂₈BClN₇PRuS

C₃₆H₄₄N₂OPRu,

C

₃₀H₃₆BN₇OPRuS,

C₃₀H₃₇BN₈O₂PRu,

C₃₂H₁₂BF₂₄

1547.75

C₂₄H₂₀

971.98

triclinic

B

C₃₂H₁₂BF₂₄

formula wt

cryst syst

space group

a, Å

576.86

1548.79

triclinic

monoclinic

P2₁/c (No. 14)

30.334(6)

10.568(2)

15.019(3)

90.00

triclinic

P1 (No. 2)

13.599(3)

13.869(3)

13.899(3)

70.76(3)

84.98(3)

89.92(3)

2464.5(10)

2

P1 (No. 2)

11.765(2)

16.274(3)

18.114(4)

87.45(3)

71.52(3)

79.53(3)

3234.2(13)

2

P1 (No. 2)

12.484(3)

13.946(3)

20.184(4)

74.11(3)

77.51(3)

85.73(3)

3299.4(14)

2

b, Å

c, Å

R, deg

β, deg

92.76(3)

90.00

γ, deg

V, Å³

4809.1(16)

8

Z

D_calcd, g/cm³

μ(Mo KR), mm^ꢀ1

F(000)

1.594

1.310

1.590

1.557

0.940

0.394

0.417

0.380

2352

1020

1556

dryst size, mm

0.04 ꢁ 0.52 ꢁ 0.56

0.13 ꢁ 0.16 ꢁ 0.35

0.36 ꢁ 0.54 ꢁ 0.62

0.35 ꢁ 0.42 ꢁ 0.54

Data Collection

100

temp, K

100

radiation (λ, Å)

min, max θ, deg

data set

Mo KR (0.710 73)

0.7, 25.00

Mo KR (0.710 73)

1.8, 25.0

Mo KR (0.710 73)

1.8, 27.5

Mo KR (0.710 73)

1.1, 27.5

ꢀ36 to +33; ꢀ12

to +12; ꢀ17 to +17

29 934, 8374; 0.076

6835

ꢀ16: to +13; ꢀ16

to +16; ꢀ16 to +16

17 485, 8603; 0.023

8237

ꢀ15 to +15; ꢀ21

to +20; ꢀ21 to +23

27 453, 14 466; 0.017

13 661

ꢀ16 to +14; ꢀ18

to +18; ꢀ26 to +26

27 367, 14 711; 0.014

13 867

total, unique no. of data; R(int)

no. of obsd data (I > 2.0 σ_I)

Reﬁnement

8603, 604

0.0381, 0.0854,^b1.04

N_ref, N_param

8374, 585

0.0625, 0.1295,^a1.07

14 466, 897

0.0371, 0.0922,^c1.04

14 711, 936

0.0428, 0.1094,^d1.02

R1, wR2, S

max, av shift/error

0.00, 0.00

min, max resd dens, e Å^ꢀ3

ꢀ0.48, 1.00

0.45, 0.65

ꢀ0.58, 0.81

ꢀ0.91, 1.46

2

^aw = 1/[σ²(F_o) + (0.0492P)²+ 14.3646P], where P = (F_o+ 2F_c²)/3. ^bw = 1/[σ²(F_o) + (0.0312P)²+ 3.1323P], where P = (F_o+ 2F_c²)/3.

^cw = 1/[σ²(F_o) + (0.0435P)²+ 2.929P], where P = (F_o²+ 2F_c²)/3. ^dw = 1/[σ²(F_o) + (0.0537P)²+ 5.2586P], where P = (F_o²+ 2F_c²)/3.

2

introduced through CPCM⁵⁸single-point calculations on gas-phase

optimized geometries. The solvents employed in the experiments

(fluorobenzene, ε = 5.42; methanol, ε = 32.61) were described with

this continuum method. The structures of the reactants, intermediates,

transition states, and products were fully optimized without any sym-

metry restriction. Transition states were identified by having one imagi-

nary frequency in the Hessian matrix. It was confirmed that transition

states connect with the corresponding intermediates by means of appli-

cation of an eigenvector corresponding to the imaginary frequency and

subsequent optimization of the resulting structures. Gibbs energies in

solution at 298 K (ΔG_solv) were obtained from adding the gas-phase

Gibbs energy corrections of the solute to ΔE_solvas indicated below,⁵⁹

where “gp” means gas-phase calculations:

were considered using the ZORA formalism⁶³and TZP (core double-ζ,

valence triple-ζ, polarized) basis set⁶⁴was used for all atoms.

’ ASSOCIATED CONTENT

S

Supporting Information. Text giving synthesis and char-

b

acterization details of η¹-O-alkynyl ketone, η²-alkyne, and vinyl-

idene complexes, CIF ﬁles, ﬁgures giving ln(I₀/I_t) vs time and

Eyring plots for isomerizations in solution, R vs time and ln(ln-

(1/(1 ꢀ R))) vs ln t plots for the isomerizations in the solid state,

and an ORTEP view of the structure of compound 2, tables giving

Cartesian coordinates, absolute energies, and Gibbs energies

(hartrees) of all the optimized structures, and ﬁgures giving

optimized structures of selected species of the alkynoneꢀ

vinylidene isomerization in the CpRu complex. This material is

available free of charge via the Internet at http://pubs.acs.org.

ΔG_solv¼ ΔE_solv+ ðΔG_gpꢀ E_gpÞ

Natural bond orbital (NBO) analysis was performed with NBO

Version 3.1⁶⁰incorporated in the Gaussian 09 package. Binding inter-

actions in selected complexes were analyzed using the energy-decom-

position analysis⁶¹with PBE0 functional as implemented in ADF-

2009.⁶²According to this set of calculations, scalar relativistic eﬀects

’ AUTHOR INFORMATION

Corresponding Author

*E-mail: pedro.valerga@uca.es (P.V.); agusti@klingon.uab.es (A.L.).

4029

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

’ ACKNOWLEDGMENT

(14) Ikeda, Y.; Yamaguchi, T.; Kanao, K.; Kimura, K.; Kamimura, S.;

Mutoh, Y.; Tanabe, Y.; Ishii, Y. J. Am. Chem. Soc. 2008, 130,

16856–16857.

(15) Mutoh, Y.; Ikeda, Y.; Kimura, K.; Ishii, Y. Chem. Lett. 2009,

38, 534–535.

We thank the Spanish MICINN (Projects CTQ2010-15390,

V.K.S. sabbatical leave SB2005-0124, CTQ2008-06866-C02-01,

and ORFEO Consolider-Ingenio 2010 CSD2007-00006) and

“Junta de Andalucía” (PAI - FQM188 and Project of Excellence

PAI05 FQM094 - Project of Excellence P08-FQM-03538) for

ﬁnancial support and for a postdoc contract to E.B. and Johnson

Matthey plc for generous loans of ruthenium trichloride. M.A.O.

acknowledges the Spanish MICINN for an FPU fellowship.

(16) (a) Gelabert, R.; Moreno, M.; Lluch, J. M.; Lledꢀos, A. Organo-

metallics 1997, 16, 3805–3814. (b) Jia, G.; Lau, C. P. Coord. Chem. Rev.

1999, 190ꢀ192, 83–108. (c) Man, M. L.; Zhu, J.; Ng, S. M.; Zhou, Z. Y.;

Yin, C. Q.; Lin, Z. Y.; Lau, C. P. Organometallics 2004, 23, 6214–6220.

(17) Besora, M.; Vyboishchikov, S. F.; Lledꢀos, A.; Maseras, F.;

Carmona, E.; Poveda, M. L. Organometallics 2010, 29, 2040–2045.

(18) de los Ríos, I.; Bustelo, E.; Puerta, M. C.; Valerga, P. Organo-

metallics 2010, 29, 1740–1749.

(19) Jimꢀenez-Tenorio, M.; Puerta, M. C.; Valerga, P.; Moncho, S.;

Ujaque, G.; Lledꢀos, A. Inorg. Chem. 2010, 49, 6035–6057.

(20) Huang, Y.-H.; Gladysz, J. A. J. Chem. Educ. 1988, 65, 298–303.

(21) Gemel, C.; LaPensꢀee, A.; Mauthner, K.; Mereiter, K.; Schmid,

R.; Kirchner, K. Monatsh. Chem. 1997, 128, 1189–1199.

(22) (a) Reger, D. L.; Klaeren, S. A.; Lebioda, L. Organometallics

1988, 7, 189–193. (b) Ciardi, C.; Reginato, G.; Gonsalvi, L.; de los Ríos,

I.; Romerosa, A.; Peruzzini, M. Organometallics 2004, 23, 2020–2026.

(23) Orpen, A. G.; Brammer, L.; Allen, F. H.; Kennard, O.; Watson,

D. G.; Taylor, R. J. Chem. Soc., Dalton Trans. 1989, S1.

(24) Manna, J.; John, K. D.; Hopkins, M. D. Adv. Organomet. Chem.

1995, 38, 79–154.

’ REFERENCES

(1) (a) Selegue, J. P. Coord. Chem. Rev. 2004, 248, 1543–1563.

(b) Bruce, M. I. In Metal Vinylidenes and Allenylidenes in Catalysis;

Bruneau, C., Dixneuf, P. H., Eds.; Wiley-VCH: Weinheim, Germany,

2008; pp 1ꢀ60.

(2) Lynam, J. M. Chem. Eur. J. 2010, 16, 8238–8247.

(3) (a) Trost, B.; McClory, A. Chem. Asian J. 2008, 3, 164–194.

(b) Odedra, A.; Liu, R.-S. In Metal Vinylidenes and Allenylidenes in

Catalysis; Bruneau, C., Dixneuf, P. H., Eds.; Wiley-VCH: Weinheim,

Germany, 2008; pp 193ꢀ214. (c) Bruneau, C.; Dixneuf, P. H. Angew.

Chem., Int. Ed. 2006, 45, 2176–2203. (d) Trost, B.; Frederiksen, M. U.;

Rudd, M. T. Angew. Chem., Int. Ed. 2005, 44, 6630–6666.

(25) Lomprey, J. R.; Selegue, J. P. J. Am. Chem. Soc. 1992,

(4) (a) García-Yebra, C.; Lꢀopez-Mardomingo, C.; Fajardo, M.;

Anti~nolo, A.; Otero, A.; Rodríguez, A.; Vallat, A.; Luca, D.; Mugnier,

Y.; Carbꢀo, J. J.; Lledꢀos, A.; Bo, C. Organometallics 2000, 19, 1749–1765.

(b) Wakatsuki, Y. J. Organomet. Chem. 2004, 4092–4109. (c) De Angelis,

F.; Sgamellotti, A.; Re, N. Dalton Trans. 2004, 3225–3230. (d) De

Angelis, F.; Sgamellotti, A.; Re, N. Organometallics 2007, 26, 5285–5288.

(e) Bassetti, M.; Cadierno, V.; Gimeno, J.; Pasquini, C. Organometallics

2008, 27, 5009–5016.(f) Zhu, J.; Lin, Z. In Metal Vinylidenes and

Allenylidenes in Catalysis; Bruneau, C., Dixneuf, P. H., Eds.; Wiley-

VCH: Weinheim, Germany, 2008; pp 129ꢀ157. (g) Cabeza, J. A.;

Pꢀerez-Carre~no, E. Organometallics 2010, 29, 3973–3978. (h) Grotjahn,

D. B.; Zeng, X.; Cooksy, A. L.; Kassel, W. S.; DiPasquale, A. G.;

Zakharov, L. N.; Rheingold, A. L. Organometallics 2007, 26,

3385–3402. (i) Vastine, B. A.; Hall, M. B. Organometallics 2008, 27,

4325–4333.

114, 5518–5523.

(26) O’Connor, J. M.; Friese, S. J.; Rodgers, B. L.; Rheingold, A. L.;

Zakharov, L. J. Am. Chem. Soc. 2005, 127, 9346–9347.

(27) Older, C. M.; Stryker, J. M. Organometallics 2000, 19, 2661–

2663.

(28) Jeﬀery, J. C.; Jelliss, P. A.; Psillakis, E.; Rudd, G. E. A.; Stone,

F. G. A. J. Organomet. Chem. 1998, 17–27.

(29) Urtel, K.; Frick, A.; Huttner, G.; Zsolnai, L.; Kircher, P.; Rutsch,

P.; Kaifer, E.; Jacobi, A. Eur. J. Inorg. Chem. 2000, 33–50.

(30) (a) Bruce, M. I. Chem. Rev. 1991, 91, 197–257. (b) Silvestre, J.;

Hoﬀmann, R. Helv. Chim. Acta 1985, 68. (c) Koga, N.; Yamazaki, H.;

Morokuma, K. J. Am. Chem. Soc. 1994, 116, 8105–8111. (d) Grotjahn,

D. B.; Miranda-Soto, V.; Kragulj, E. J.; Lev, D. A.; Erdogan, G.; Zeng, X.;

Cooksy, A. L. J. Am. Chem. Soc. 2008, 130, 20–21.

(31) Becker, E.; Pavlik, S.; Kirchner, K. Adv. Organomet. Chem. 2008,

(5) Bustelo, E.; Carbo, J. J.; Lledos, A.; Mereiter, K.; Puerta, M. C.;

Valerga, P. J. Am. Chem. Soc. 2003, 125, 3311–3321.

56, 155ꢀ 197 and references therein.

(32) (a) Alkynetovinylideneisomerization of [CpRu(η²-HCtCMe)-

(6) Puerta, M. C.; Valerga, P. Coord. Chem. Rev. 1999, 193ꢀ

(PMe₃)₂][PF₆] (ΔH^q= 23.4 kcal mol^ꢀ1, ΔS^q= 3.9 cal mol^{ꢀ1 ꢀ1}):

K

195, 977–1025.

Bullock, R. M. J. Chem. Soc., Chem. Commun. 1989, 165–167. (b)

Vinylidene to alkyne isomerization of [CpMo(dCdCH^tBu)(CO)-

(NO)] (ΔH^q= 29.0 kcal mol^ꢀ1, ΔS^q= 1.8 cal mol^ꢀ1K^ꢀ1): Ipaktschi,

J.; Mohsseni-Ala, J.; Uhlig, S. Eur. J. Inorg. Chem. 2003, 4313–4320. (c)

Release of terminal alkynes from indenylruthenium vinylidene com-

plexes (ΔH^q= 24 kcal mol^ꢀ1, ΔS^q= ꢀ3 cal mol^ꢀ1K^ꢀ1): ref 4e.

(33) Spangler, C. W. Chem. Rev. 1976, 76, 187ꢀ 217 and references

therein.

(7) Bly, R. S.; Zhong, Z.; Kane, C.; Bly, R. K. Organometallics 1994,

13, 899–905.

(8) (a) Werner, H.; Baum, M.; Schneider, D.; Windm€uller, B.

Organometallics 1994, 13, 1089–1097. (b) Connelly, N. G.; Geiger,

W. E.; Lagunas, M. C.; Metz, B.; Rieger, A. L.; Rieger, P. H.; Shaw, M. J.

J. Am. Chem. Soc. 1995, 117, 12202–12208. (c) Katayama, H.; Onitsuka,

K.; Ozawa, F. Organometallics 1996, 15, 4642–4645. (d) Werner, H.;

Lass, R. W.; Gevert, O.; Wolf, J. Organometallics 1997, 16, 4077–4088.

(e) Jimꢀenez, M. V.; Sola, E.; Lahoz, F. J.; Oro, L. A. Organometallics 2005,

24, 2722–2729. (f) Ilg, K.; Paneque, M.; Poveda, M. L.; Rendꢀon,

N.; Santos, L. L.; Carmona, E.; Mereiter, K. Organometallics 2006,

25, 2230–2236.

(9) (a) Baum, M.; Mahr, N.; Werner, H. Chem. Ber 1994, 127,

1877–1886. (b) Venkatesan, K.; Blacque, O.; Fox, T.; Alfonso, M.;

Schmalle, H. W.; Berke, H. Organometallics 2004, 23, 1183–1186.

(c) Venkatesan, K.; Blacque, O.; Fox, T.; Alfonso, M.; Schmalle,

H. W.; Kheradmandan, S.; Berke, H. Organometallics 2005, 24, 920–932.

(10) Miura, T.; Iwasawa, N. J. Am. Chem. Soc. 2002, 124, 518–519.

(11) Millert, D. C.; Angelici, R. J. Organometallics 1991, 10, 81–89.

(12) King, P. J.; Knox, S. A. R.; Legge, M. S.; Orpen, A. G.;

Wilkinson, J. N.; Hill, E. A. Dalton Trans. 2000, 1547–1548.

(34) Wang, Y.; Agbossou, F.; Dalton, D. M.; Liu, Y.; Arim, A. M.;

Gladysz, J. A. Organometallics 1993, 12, 2699–2713.

(35) Solid-state reactions can be described by equations of the type

F(R) = kt, where the function F(R) depends on the reaction-controlling

mechanism, as well as the size and shape of the reacting particles. This

type of conversion function may be described in terms of the reaction-

rate-controlling AvramiꢀErofeev model, by the general formula R = 1 ꢀ

exp(ꢀktⁿ), where n is empirically determined via a ﬁtting procedure.

The Avrami exponent is indicative of the reaction mechanism and crystal

growth dimensions: (a) Avrami, M. J. Chem. Phys. 1939, 7, 1103–1112.

(b) Avrami, M. J. Chem. Phys. 1940, 8, 212–224. (c) Avrami, M. J. Chem.

Phys. 1941, 9, 177–184.

(36) Haber, J.; Jamroz, K. J. Solid State Chem. 1982, 44, 291–297.

(37) (a) The scaling factor of 0.9536 recommended in ref 37b for the

vibrational frequencies calculated with the employed combination of

exchange-correlation functional and basis set has been applied.

(13) Shaw, M. J.; Bryant, S. W.; Rath, N. Eur. J. Inorg. Chem.

2007, 3943–3946.

4030

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Organometallics

ARTICLE

(b) Tantirungrotechai, Y.; Phanasant, K.; Roddecha, S.; Surawatana-

wong, P.; Sutthikhum, V.; Limtrakul, J. J. Mol. Struct. (THEOCHEM)

2006, 760, 189–192.

(38) We have not been able to ﬁnd a nondissociative transition state

for the η¹-O-ketone f π-alkyne interconversion despite performing an

extensive search.

(56) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon,

M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654–3665.

(57) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer,

P. v. R. J. Comput. Chem. 1983, 4, 294–301.

(58) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. J. Comput. Chem.

2003, 24, 669–681.

(39) De Angelis, F.; Sgamellotti, A.; Re, N. Organometallics 2002,

(59) Braga, A. A. C.; Ujaque, G.; Maseras, F. Organometallics 2006,

21, 5944–5950.

25, 3647–3658.

(40) (a) Ujaque, G.; Maseras, F.; Lledꢀos, A.; Contreras, L.; Pizzano,

A.; Rodewald, D.; Sꢀanchez, L.; Carmona, E.; Monge, A.; Ruiz, C.

Organometallics 1999, 18, 3294–3305. (b) Jaﬀart, J.; Etienne, M.;

Reinhold, M.; McGrady, J. E.; Maseras, F. Chem. Commun. 2003,

876–877. (c) Boulho, C.; Keys, T.; Coppel, Y.; Vendier, L.; Etienne,

M.; Locati, A.; Bessac, F.; Maseras, F.; Pantazis, D. A.; McGrady, J. E.

Organometallics 2009, 28, 940–943. (d) Etienne, M.; McGrady, J. E.;

Maseras, F. Coord. Chem. Rev. 2009, 253, 635–646.

(41) We have not been able to locate a transition state connecting

Tp-agost with the π-alkyne complex, but small distortions on the

geometry of Tp-agost lead directly to the π-alkyne, indicating that it

should be a shallow minimum.

(60) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.

NBO Version 3.1.

(61) (a) Kitaura, K.; Morokuma, K. Int. J. Quantum Chem. 1976,

10, 325–340. (b) Ziegler, T.; Rauk, A. Theor. Chim. Acta 1977, 46, 1–10.

(62) ADF2009, SCM; Theoretical Chemistry, Vrije Universiteit,

Amsterdam, The Netherlands, http://www.scm.com.

(63) (a) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.

1994, 101, 9783–9792. (b) van Lenthe, E.; Ehlers, A.; Baerends, E. J.

J. Chem. Phys. 1999, 110, 8943–8953.

(64) van Lenthe, E; Baerends, E. J. J. Comput. Chem. 2003, 24,

1142–1156.

(42) In a recent communication, the ﬁrst example of a reversible

disubstituted vinylidene-to-internal alkyne isomerization on reaction

with a monophosphine has been shown: Mutoh, Y.; Imai, K.; Kimura, Y.;

Ikeda, Y.; Ishii, Y. Organometallics 2011, 30, 204–207. We have checked

this possibility with one of our vinylidene systems, and this also

conﬁrmed the release of the alkyne upon treatment with triphenylpho-

sphine and heating.

(43) Macías-Arce, I.; Puerta, M. C.; Valerga, P. Eur. J. Inorg. Chem.

2010, 1767–1776.

(44) Jansen, A.; Pitter, S. Monatsh. Chem. 1999, 130, 783–794.

(45) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992,

11, 3920–3922.

(46) Sheldrick, G. M. SADABS, 2001 version; University of

G€ottingen, G€ottingen, Germany, 2001.

(47) (a) Sheldrick, G. M. SHELXTL version 6.10, Crystal Structure

Analysis Package; Bruker AXS, Madison, WI, 2000. (b) Sheldrick, G. M.

Acta Crystallogr. 2008, A64, 112–122.

(48) Farrugia, L. J. ORTEP-3 for Windows, version 1.076. J. Appl.

Crystallogr. 1997, 30, 565.

(49) (a) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996,

77, 3865–3868. (b) Perdew, J. P.; Ernzerhof, M.; Burke, K. J. Chem. Phys.

1996, 105, 9982–9985.

(50) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;

Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;

Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara,

M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;

Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.; , Jr.,

Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,

K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;

Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.;

Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.;

Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;

Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;

Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich,

€

S.; Daniels, A. D.; O. Farkas, Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; ,

Fox, D. J. Gaussian 09, Revision A.1; Gaussian, Inc., Wallingford, CT, 2009.

(51) Pantazis, D. A.; McGrady, J. E.; Besora, M.; Maseras, F.;

Etienne, M. Organometallics 2008, 27, 1128–1134.

(52) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215–241.

(53) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.

Theor. Chim. Acta 1990, 77, 123–141.

(54) Ehlers, A. W.; Bohme, M.; Dapprich, S.; Gobbi, A.; Hollwarth,

A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G.

Chem. Phys. Lett. 1993, 208, 111–114.

(55) Hehre, W. J.; Ditchﬁeld, R.; Pople, J. A. J. Chem. Phys. 1972,

56, 2257–2261.

4031

dx.doi.org/10.1021/om200273v |Organometallics 2011, 30, 4014–4031

Article Doi

Internal alkyne isomerization to vinylidene versus stable π-Alkyne: Theoretical and experimental study on the divergence of analogous Cp*Ru and TpRu systems

DOI: 10.1021/om200273v

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om200273v

Full text of DOI:10.1021/om200273v

Products guided by the article

R&D Labs maybe for 1314131-11-7

Relevant to this article

Hot Product