Two distinct mechanisms of alkyne insertion into the metal-sulfur bond: Combined experimental and theoretical study and application in catalysis

Article abstract of DOI:10.1002/chem.200902928

The present study reports the evidence for the multiple carboncarbon bond insertion into the metalheteroatom bond via a five-coordinate metal complex. Detailed analysis of the model catalytic reaction of the carbonsulfur (C-S) bond formation unveiled the mechanism of metal-mediated alkyne insertion: A new pathway of CS bond formation without preliminary ligand dissociation was revealed based on experimental and theoretical investigations. According to this pathway alkyne insertion into the metal-sulfur bond led to the formation of intermediate metal complex capable of direct CS reductive elimination. In contrast, an intermediate metal complex formed through alkyne insertion through the traditional pathway involving prelimi-nary ligand dissociation suffered from "improper" geometry configuration, which may block the whole catalytic cycle. A new catalytic system was developed to solve the problem of stereoselective S-S bond addition to internal alkynes and a cost-efficient Ni-catalyzed synthetic procedure is reported to furnish formation of target vinyl sulfides with high yields (up to 99%) and excellent Z/E selectivity (>99:1).

Full text of DOI:10.1002/chem.200902928

FULL PAPER
DOI: 10.1002/chem.200902928
Two Distinct Mechanisms of Alkyne Insertion into the Metal–Sulfur Bond:
Combined Experimental and Theoretical Study and Application in Catalysis
Valentine P. Ananikov,*[a] Konstantin A. Gayduk,[a] Nikolay V. Orlov,[a]
Irina P. Beletskaya,*[b] Victor N. Khrustalev,[c] and Mikhail Yu. Antipin[c]
Abstract: The present study reports the
evidence for the multiple carbon–
carbon bond insertion into the metal–
heteroatom bond via a five-coordinate
metal complex. Detailed analysis of the
model catalytic reaction of the carbon–
alkyne insertion into the metal–sulfur
bond led to the formation of intermedi-
nary ligand dissociation suffered from
“improper” geometry configuration,
which may block the whole catalytic
cycle. A new catalytic system was de-
veloped to solve the problem of stereo-
ꢀ
ate metal complex capable of direct C
S reductive elimination. In contrast, an
intermediate metal complex formed
through alkyne insertion through the
traditional pathway involving prelimi-
ꢀ
selective S S bond addition to internal
ꢀ
sulfur (C S) bond formation unveiled
alkynes and a cost-efficient Ni-cata-
lyzed synthetic procedure is reported
to furnish formation of target vinyl sul-
fides with high yields (up to 99%) and
excellent Z/E selectivity (>99:1).
the mechanism of metal-mediated
ꢀ
alkyne insertion: a new pathway of C
Keywords: heterogeneous catalysis ·
homogeneous catalysis · insertion ·
S bond formation without preliminary
ligand dissociation was revealed based
on experimental and theoretical inves-
tigations. According to this pathway
nickel
· stereoselectivity · vinyl
sulfides
Introduction
involves a preliminary ligand dissociation and coordination
of the CC multiple bond to the metal (“activation of the
multiple bond”). The necessity of preliminary activation of
the unsaturated substrate introduces some prerequisites into
catalyst design and excludes several classes of metal com-
plexes from consideration as possible catalysts.[1] Of course,
a catalytic system without these limitations would be a fasci-
nating and novel opportunity for constructing carbon–heter-
Insertion of the carbon–carbon (CC) multiple bonds into a
metal–heteroatom bond is a key-step of transition-metal-cat-
alyzed heterofunctionalization of alkynes, alkenes, allenes,
and dienes.[1] It is generally accepted that the insertion step
[a] Prof. V. P. Ananikov, K. A. Gayduk, Dr. N. V. Orlov
Zelinsky Institute of Organic Chemistry
Russian Academy of Sciences, Leninsky Prospect 47
Moscow, 119991 (Russia)
ꢀ
oatom (C E) bonds.
An excellent model for the study of catalytic C E bond
ꢀ
formation is alkyne insertion into the metal–sulfur bond.
The field remains less studied, since sulfur species are well-
known catalyst poisons,[2] but a solution to this problem was
rationalized not long ago through the utilization of an
excess of the phosphine ligand to suppress catalyst deactiva-
Fax : (+7)499-1355328
E-mail: val@ioc.ac.ru
[b] Prof. I. P. Beletskaya
Chemistry Department
Lomonosov Moscow State University, Vorob’evy gory
Moscow, 119899 (Russia)
tion.[3] Nowadays transition-metal-catalyzed S S and S H
bond addition to alkynes is a versatile approach to vinyl sul-
fides, which combines 100% atom efficiency with excellent
stereoselectivity and high yields. Convenient synthetic meth-
ods were developed to carry out stereoselective addition of
ꢀ
ꢀ
Fax : (+7)495-9393618
E-mail: beletska@org.chem.msu.ru
[c] Dr. V. N. Khrustalev, Prof. M. Yu. Antipin
Nesmeyanov Institute of Organoelement Compounds
Russian Academy of Sciences, Vavilova str. 28
Moscow, 119991 (Russia)
ꢀ
S S bond to terminal alkynes (3) and regio- and stereoselec-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902928. It contains: 1) the
structures optimized at the PBE/TZP level; 2) molecular structures of
2d and 2e and details of X-ray crystal structure determination; and
3) tables with geometry parameters of 2d and 2e.
ꢀ
tive addition of S H bond to terminal and internal alkynes
(5 and 4, respectively; Scheme 1).[3] However, none of the
ꢀ
known catalytic systems were succeeded to carry out S S
bond addition to internal alkynes leading to 2 (Scheme 1).
Chem. Eur. J. 2010, 16, 2063 – 2071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2063

to alkynes is now recognized as a well-established synthetic
method.
The first study of [PdACTHNUTRGNEUNG(PPh3)4]-catalyzed diaryldichalcoge-
nide addition to terminal alkynes leading to formation of
product 3 with high selectivity and yields was reported by
Ogawa, Sonoda et al.[11] A further mechanistic study re-
vealed the di- and polynuclear nature of intermediate transi-
tion-metal complexes in the catalytic reaction and the im-
portance of the excess of phosphine ligands to achieve high
performance of the catalytic system.[12] An important step of
ꢀ
Scheme 1. Transition-metal-catalyzed S-S and S H bonds addition to al-
kynes.
ꢀ
the catalytic reaction—alkyne insertion into the ArS M
bond—was studied by Kuniyasu, Kambe et al.[13] Catalytic
activity of Rh complexes in this reaction was established by
Yamaguchi et al.;[14] however, Pt complexes were found in-
active and the difference between Pd and Pt was rational-
ized.[15] For synthetic purposes microwave-assisted synthe-
sis,[16] solvent-free reactions,[17] and polymer-supported cata-
lysts[18] were developed for the addition reactions involving
terminal alkynes. Recently this fascinating synthetic method-
ology was extended to include dialkyldichalcogenides by uti-
lizing an Ni catalyst,[19] cyanothiolation of terminal alkynes
on Pd,[20] and preparation of dienes from terminal alkynes
on Pt and Ni.[21,22] The mechanism of the catalytic reaction
At first sight the question may seem very special and less
important. However, the lack of success in carrying out the
reaction for several years and the failure to identify a clear
reason in spite of continuous efforts, highlighted a stumbling
block in our understanding of mechanistic picture of this
field.
To understand the nature of this problem first we should
consider briefly the scope and mechanism of both catalytic
ꢀ
ꢀ
reactions, that is, S S and S H bond addition. The first
study reported for the arylthiol addition to alkynes utilized
a Pd catalyst and was successfully carried out for terminal
and internal alkynes.[4] Catalytic activities of Pt,[5] Rh and
Ir,[4,6] and Ni[5] complexes in this reaction aimed at the for-
mation of 4 and 5 are now established. A high-performance
catalytic system for gram-scale preparation of vinyl sulfides
from terminal and internal alkynes has been developed by
using nano-structured Ni complexes as the catalytic spe-
cies.[7] A self-organized nano-sized catalytic system based on
Pd complexes solved the problem of stereoselective alkane-
thiol addition to the triple bond of alkynes.[8] The mecha-
ꢀ
of S S bond addition to terminal alkynes was shown to in-
ꢀ
clude the following steps: 1) oxidative addition of the S S
bond, 2) dissociation of the ligand L and alkyne coordina-
ꢀ
ꢀ
tion, 3) alkyne insertion into the M S bond, and 4) C S re-
ductive elimination from the metal center (Scheme 3).
ꢀ
nism of S H bond addition to alkynes involving phosphine
complexes of transition metals as a catalyst was shown to in-
clude the following steps: 1) oxidative addition, 2) dissocia-
tion of the ligand L and alkyne coordination, 3) alkyne in-
ꢀ
ꢀ
sertion into the M S bond, and 4) C H reductive elimina-
tion (Scheme 2).[9] No differences in the mechanism of the
catalytic cycle have been reported for terminal and internal
alkynes, and similar synthetic procedures have been used for
the preparation of both types of vinyl sulfides 4 and 5.[3–8]
Thus, preparation of vinyl sulfides through the thiol addition
ꢀ
Scheme 3. The mechanism of the catalytic S S bond addition to terminal
alkynes.[10]
In spite of intrinsic similarity between these reactions (cf.
Schemes 2 and 3), only terminal alkynes gave the final prod-
ꢀ
uct in the S S bond addition reaction (Scheme 1). To sum-
marize, the origins of such dramatic difference in reactivity
between the terminal and internal alkynes in the catalytic
ꢀ
S S bond addition remains unclear and the results published
in the literature give no hints to overcome the problem.
The fact not only breaks consistency of an overall mecha-
nistic picture of this field, but also limits synthetic applica-
tion of this catalytic methodology. Vinyl sulfides 2 are in
demand in organic synthesis, catalysis, and material sci-
ence;[23] thus the development of a cost-efficient and eco-
ꢀ
Scheme 2. The mechanism of the catalytic S H bond addition to terminal
and internal alkynes.[9,10]
2064
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2063 – 2071

ꢀ
C S Bond Formation
FULL PAPER
Table 1. Ligand effect on the yield of 2a in Ni-catalyzed reaction of
Ph2S2 and 3-hexyne (1a).[a]
friendly synthetic procedures to access them would be
useful.
In the present study, we have solved this intriguing prob-
lem and revealed the mechanistic reasons of different activi-
Entry
Ligand (L)
Yield [%]
Entry
Ligand (L)
Yield [%]
1
2
3
4
5
PPh3
PCy3
PBu3
0
0
0
0
0
6
7
8
9
10
PCyPh2
PCy2Ph
PMe2Ph
PTh3
PMePh2
0
0
0
0
ꢀ
ty between terminal and internal alkynes in the catalytic S
S bond addition reaction. A novel catalytic system was de-
veloped to carry out stereoselective addition of diaryldisul-
fides to terminal alkynes with good yields and excellent se-
lectivity.
[b]
P
P
N
E
99
[a] 1 mmol of Ph2S2, 1.5 mmol of 3-hexyne, 3 mol% of [NiACHTUNGTRENNUNG(acac)2],
30 mol% of the ligand, 1008C, 8 h, solvent free. [b] PTh3 =tris(2-thienyl)-
phosphine.
Results and Discussion
(99%) and PMe2Ph (0%) ligands highlight the extraordina-
ry nature of the catalytic system studied (Table 1).
The performance of the catalytic system was investigated
using diphenyl disulfide addition to 3-hexyne (1a) as a
model reaction (Scheme 4). The reaction was carried out
Optimization of the reaction conditions has shown that
the addition reaction can also be carried out in the presence
of solvent (toluene, 0.5 mL, 8 h), but resulted in a lower
yield—80% (cf. 99% solvent free; entry 10, Table 1). The
temperature of 1008C was found optimal, since the reaction
was not complete after 8 h at 808C. A further increase of re-
action temperature was impractical, since it diminished ste-
reoselectivity of the reaction due to Z/E isomerization (ob-
1
served with H NMR spectroscopy).
Scheme 4. The model catalytic reaction.
The scope of the developed catalytic system was investi-
gated for a variety of internal alkynes and diaryldisulfides
with different substituents (Table 2). The excellent stereose-
lectivity was observed in all studied cases (Z/E> 99:1).
under solvent-free conditions. Since Pd and Pt catalytic sys-
tems have already been studied
and showed poor performance
in the reaction of interest, we
investigated Ni complexes as
possible catalysts.
First, we studied the catalytic
system with triphenylphosphine
ligand, since PPh3 was most fre-
quently used as a ligand in cata-
Table 2. The scope of the Ni-catalyzed addition of diaryldisulfides to internal alkynes.[a]
Entry
Alkyne
Product[b]
Yield[c] [%]
99 (85)
ꢀ
RS SR
1
Ph2S2
Ph2S2
2
3
93 (82)
99 (91)
ꢀ
lytic reactions of S S bond ad-
dition to terminal alkynes.[3]
Carrying out the model reac-
tion at 1008C for 8 h failed to
produce the desired product 2a
(entry 1, Table 1). Next, we uti-
lized various phosphine and
phosphite ligands with alkyl
and aryl groups and found
them inactive in the formation
of 2a (entries 2–9, Table 1). Sur-
prisingly, using the PMePh2
ligand resulted in 99% yield of
product 2a and excellent ste-
Ph2S2
4
5
6
Ph2S2
Ph2S2
Ph2S2
99 (89)
91 (80)
60 (55)
7
1b
N
75 (67)
reoselectivity,
(entry 10, Table 1).
Z/E>99:1
8
9
1b
1b
G
95 (82)
99 (85)
It is evident that such dra-
matic difference cannot be ex-
plained by simple consideration
of the ligand effect. Pronounced
variation in the reaction yield
[a] See Experimental Section for complete description of synthesis and isolation details. [b] Stereoselectivity>
99:1. [c] NMR yield after completing the reaction and isolated yield after separation and purification (in pa-
with the PPh3 (0%), PMePh2 renthesis).
Chem. Eur. J. 2010, 16, 2063 – 2071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2065

I. P. Beletskaya, V. P. Ananikov et al.
High product yields (82–91% isolated) were found in the
catalytic Ph2S2 addition to 3-hexyne, 2-hexyne, and 4-octyne
(entries 1–3, Table 2). High yields and selectivity were also
obtained in the case of activated alkyne 1d (entry 4,
Table 2), in contrast to the previous studies of Ph2S2 addition
to terminal alkynes, in which the presence of the Ph groups
decreased the observed selectivity.[3] The catalytic system
was tolerant to functional groups in alkynes leading to good
yields for 1e, 1 f (entries 5, 6, Table 2); somewhat lower
yields compared to 1a–1d were caused by by-product for-
mation due to alkyne polymerization. Although polymeri-
zation of alkynes on Ni is a facile and well-known reac-
tion,[24] in the developed catalytic system we observed it
only with some selected alkynes to a limited extent. Elec-
tron donor and acceptor substituents in the phenyl ring of
Ar2S2 did not affect the scope of the developed catalytic
system and resulted in good yields of products (entries 7–9,
Table 2).
After exploring the synthetic potential of the developed
catalytic system, it is of interest to understand the mecha-
nism of this reaction and to reveal the origins of the crucial
difference in reactivity of internal alkynes depending on the
ligand L. Considering a commonly accepted mechanism of
ꢀ
S S bond addition, we may assume that failure of internal
alkynes to proceed in the catalytic reaction with PPh3 and
PMe2Ph ligands may originate either due to inability to un-
dergo coordination and insertion steps or due to loss of re-
ꢀ
activity on the C S reductive elimination step (Scheme 3).
To distinguish between these two possibilities we have de-
termined catalytic activity of particular metal/ligand combi-
ꢀ
nations in the S H bond addition reaction (this catalytic
Figure 1. X-ray structures of 2d (top) and 2e (bottom; only one of the
two crystallographically independent cations is shown).
ꢀ
cycle involves alkyne insertion, but does not involve C S re-
ductive elimination; see Scheme 2). We have found very low
activity in the Ni/PMe2Ph catalytic system: addition of
PhSH to 3-hexyne (1008C, 8 h) resulted only in 30% yield
of 4a. This result has shown that for the PMe2Ph ligand
was in total agreement with proposed mechanism confirm-
ꢀ
ing the involvement of an alkyne insertion step into the Ni
ꢀ
alkyne coordination and insertion into the Ni S bond can
S bond in the catalytic cycle and ruling out the possibilities
[26]
ꢁ
be a limiting factor. Carrying out this reaction with the Ni/
PMePh2 and Ni/PPh3 systems led to 95 and 73% yields of
4a,[25] respectively. Therefore, for the both ligands L=
PMePh2 and PPh3 alkyne coordination and insertion pro-
ceeded smoothly, since these stages are involved in the
mechanism of catalytic formation of product 4a (Scheme 2).
of other routes of C C bond activation. Remarkably, that
catalytic procedure provides access to thermodynamically
less stable Z isomers that cannot be reached in high selectiv-
ity by other synthetic routes.
Our study has shown that alkyne insertion took place
with both PMePh2 and PPh3 ligands. Therefore, it could be
that the reductive elimination stage is responsible for the
absence of catalytic activity with PPh3 ligand. We decided to
carry out a theoretical study to verify this assumption and to
ꢁ
To confirm the proposed route of C C bond insertion in
the case of catalytic systems with PMePh2 and PPh3 ligands,
we carried out a structural study to determine configuration
of the double bond of the addition product 2.[26] In all cases
in the present study only a single isomer was detected on
the product formation step with excellent selectivity
(Table 2). The syn fashion of the addition reaction and Z
configuration of the final product were confirmed by X-ray
structure analysis for 2d and 2e, the S-C-C-S dihedral
angles are equal to 18.6(1) and ꢀ0.5(4), 2.3(5)8 (for the two
crystallographically independent cations), respectively
(Figure 1).[27] The 2D NOESY spectroscopy was utilized to
determine geometry of products 2b and 2e–2i; a Z configu-
ration was found in all cases. The structure of the products
ꢀ
find out to what extent C S reductive elimination is sensi-
tive to the substitution of the carbon atom in a-position to
the metal. The model system used in the study is shown on
Scheme 5, the theoretical calculations were carried out at
the PBE/TZP level. Full geometry optimization was carried
for the PPh3 ligands and SPh groups without any simplifying
approximations (100 and 103 atoms in structures I–IV and
V–VIII, respectively).
ꢀ
Theoretical study was carried out for the C S reductive
elimination stages modeling both reactions (Scheme 5):
vinyl ligand resulted from the terminal (R2 =H) and internal
2066
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2063 – 2071

ꢀ
C S Bond Formation
FULL PAPER
for the internal alkynes (R2 ¼
H) complex 9 is a resting state
lying off the catalytic cycle. In
this context it is worth mention-
ꢀ
ing that a recent study of C C
reductive elimination from Ir
complexes has shown that
ꢀ
ligand reorientation in the Ir
C
A
to DG ꢂ17 kcalmolꢀ1 to the
activation barrier, therefore de-
creasing reaction rate by many
orders of magnitude.[31]
¼
ꢀ
Scheme 5. The model system for theoretical study of C S reductive elimination reaction involving a-unsubsti-
tuted vinyl group (top) and a-Me-substituted vinyl group (bottom); C-H and C-Me groups in a-position are
highlighted with green color.[28]
alkynes (R2 =Me). Starting from initial complexes I and V
both transition stages were successfully located (II-TS and
VI-TS, respectively). Overcoming transition states led to in-
termediate formation of p-complexes III and VII, followed
by dissociation of the organic products (IV and VIII).[29]
Representative optimized structures of complex V and tran-
sition state VI-TS together with selected structural parame-
ters are shown in Figure 2. Comparing both reductive elimi-
nation reactions I!IV and V!VIII we did not notice a sig-
nificant difference in the key structure units.[28] This observa-
tion was in total agreement with calculated energetic param-
¼
eters. The activation energies were DE =10.0 and
13.8 kcalmolꢀ1 for the I!II-TS and V!VI-TS barriers, re-
spectively. Indeed, enlarged steric strain due to the presence
of the additional Me group increased the activation energy
by 3.8 kcalmolꢀ1. This could make the latter reductive elimi-
nation reaction less favorable, but clearly this cannot ac-
count for observed dramatic difference in reactivity. There-
ꢀ
fore, difficulties on the C S reductive elimination step
Figure 2. The structures of V and VI-TS optimized at PBE/TZP level
(some atoms are omitted for clarity[28]); selected bond lengths are given
in ꢁ.
cannot be assumed as a reason for complete loss of catalytic
activity in the Ni/PPh3 system. The calculated reaction free
energies were DG=ꢀ5.8 and ꢀ7.9 kcalmolꢀ1 for the I!III
and V!VII reactions, respectively, thus, providing the nec-
essary driving force for both transformations.
Based on our experimental and theoretical study we pro-
pose another mechanism of the catalytic reaction in the Ni/
PMePh2 system (Scheme 7). The more electron-donating
PMePh2 ligand does not dissociate easily in contrast to
labile PPh3 ligand; this retards formation of the p-complex 7
A more detailed inspection of the transition-state VI-TS
and initial complex V has shown that C=C unit of the vinyl
ligand should be out-of-plane with P-Ni-P unit in order to
ꢀ
undergo C S bond formation by reductive elimination
(Figure 2).[30] For the considered
pathway it means that complex
9 has to undergo ligand reorien-
tation before the actual reduc-
tive elimination could take
place (Scheme 6). However, this
reorientation is retarded in case
of internal alkyne (R2 ¼H) due
to steric hindrance between the
R2 group and PR’3 ligand. Thus,
for the terminal alkynes (R2 =
H) complex 9 is an intermediate
of the catalytic reaction, while
ꢀ
Scheme 6. C S reductive elimination by means of the ligand dissociative pathway (L=PPh3).
Chem. Eur. J. 2010, 16, 2063 – 2071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2067

I. P. Beletskaya, V. P. Ananikov et al.
Table 3. The yields of 2a in Ni-catalyzed reaction of Ph2S2 and 3-hexyne
(1a) at different L/Ni ratios.[a]
Entry
MePh2P [mol%]
L/Ni ratio
Yield [%] after
3h
1h
8h
1
2
3
4
5
6
2:1[b]
5:1[b]
10:1[c]
15:1[c]
20:1[c]
0
5
26
43
55
0
8
55
79
92
0
12
99
99
99
15
30
45
60
[a] Reaction conditions: [NiACTHNUGTRNE(GNU acac)2] (3 mol%), Ph2S2 (1 mmol), 3-hexyne
(1.5 mmol), 1008C, solvent free. [b] Heterogeneous reaction mixture with
metal species partially precipitated. [c] Homogeneous reaction mixture.
ꢀ
Scheme 7. C S reductive elimination through the apical site of the metal
without ligand dissociation (L=PMePh2).
best result within the studied range of ratios was observed
for the 20-fold excess of the ligand (entry 5, Table 3).[33] It is
worth noting that such a large excess of the ligand did not
block the catalytic activity.
For economic reasons we have chosen L/Ni=10:1 for the
synthetic procedure (Table 2), since it is a minimal excess of
the phosphine ligand that ensures a homogeneous system
(Table 3) and high performance of the catalytic reaction.
and the resting state 9. On the other hand, the PMePh2
ligand is also less sterically crowded compared to PPh3 and
allows alkyne molecule to coordinate to the apical site of
the metal complex 6 leading to complex 10. Insertion of the
alkyne into the Ni S in such a case would provide complex
11 with the vinyl ligand in out-of-plane orientation to under-
ꢀ
go reductive elimination. Therefore, vinyl ligand reorienta-
ꢀ
tion is not required and C S bond formation may proceed
Conclusion
directly from the complex 11. It should be pointed out that
complex 8 may also represent a resting state in case of high
thermodynamic stability of vinyl ligand coordination in che-
To summarize, the present experimental and theoretical
study has not only solved a long standing problem of selec-
ꢀ
ꢀ
late fashion (taking into account that direct C S reductive
elimination from complex 8 is impossible due to geometry
reasons).
tive S S bond addition to internal alkynes, but also resulted
in development of a new look at the mechanistic picture of
these reactions.
To scrutinize the reliability of this mechanism we carried
out two sets of experiments. First, within this mechanism
not only internal, but also terminal alkynes should be suc-
cessfully involved in the Ni/PMePh2 catalytic system. To
check this possibility we carried out Ph2S2 addition to 1-
hexyne. Indeed, we observed almost quantitative conversion
of the alkyne after 3 h at 708C in excellent agreement with
proposed mechanism.[32]
Second, within the proposed mechanism an unusual de-
pendence of the performance of the catalytic system should
be expected upon changing ligand/metal ratio. An excess of
the ligand would decrease the possibility of ligand dissocia-
tion pathway (leading to 7 and resting state 9), therefore, in-
creasing the contribution of the pathway involving the
apical metal site (10) and enhancing the product formation.
If phosphine ligand dissociation precedes alkyne coordina-
tion, the opposite dependence should be observed—an
excess of the ligand would retard alkyne coordination and
suppress the reaction. We have carried out the NMR moni-
toring of the catalytic reactions at different L/Ni ratios
(Table 3). No reaction was observed with the 1:2 ratio
(entry 1, Table 3) and only 12% of the product was formed
with the 1:5 ratio (entry 2, Table 3). With the higher ligand/
metal ratios, particularly 10:1, 15:1, and 20:1, complete con-
version was observed after 8 h at 1008C (entries 3–5,
Table 3), while comparison of the yields at shorter reaction
time has clearly indicated reaction rate enhancement. The
With a more labile ligand (PPh3), the dissociative pathway
ꢀ
of C S bond formation is facilitated, which strongly depends
on the ability of the vinyl ligand to undergo reorientation in
order to adopt proper conformation required for reductive
elimination. Hindered reorientation of the vinyl ligand
caused by substitution of the a-carbon atom blocked the
catalytic reaction with internal alkyne. With a more strongly
bound ligand (PMePh2), the alkyne insertion took place
through another pathway and did not require preliminary
ligand dissociation. This pathway led to intermediate metal
complex, which may undergo direct reductive elimination
without the need of ligand reorientation. Thus, we have
ꢀ
found that two pathways of C S reductive elimination are
possible and the nature of the ligand is the key controlling
factor. The first pathway selectively involves only the termi-
nal alkynes, while the second pathway is suitable for both
the terminal and internal alkynes.
Based on discovered catalytic system we have developed
ꢀ
the first synthetic approach to accomplish stereoselective S
S bond addition to internal alkynes. The addition reaction
was carried out with high yields and excellent selectivity for
various alkynes and diaryldisulfides. The readily available
nickel complex [NiACTHNUTRGNEUNG(acac)2] (acac=acetylacetonate) was uti-
lized as a catalyst precursor.
The evidence for the insertion step through a five-coordi-
nate metal complex reported in the present study may be
also useful in taking a new look at (and possibly reevaluate)
2068
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2063 – 2071

ꢀ
C S Bond Formation
FULL PAPER
[(Z)-2-Phenyl-1,2-bis(phenylsulfanyl)ethenyl]benzene (2d): 1H NMR
(500 MHz; CDCl3): d=6.80–6.89 (m, 6H), 6.98 (t, J=6.74 Hz, 2H), 7.01–
7.07 (m, 8H), 7.22 ppm (d, J=8.20 Hz, 4H); 13C{1H} NMR (126 MHz;
CDCl3): d=126.57, 126.68, 127.26, 128.37, 130.45, 131.36, 134.15, 138.02,
138.57 ppm; MS (EI): m/z (%): 396 (7) [M+]; elemental analysis calcd
(%) for C26H20S2: C 78.75, H 5.08, S 16.17; found: C 78.68, H 5.37, S
16.09.
reaction mechanisms of some other catalytic cycles involving
insertion of carbon–carbon multiple bonds, like alkynes and
alkenes heterofunctionalization, Heck reaction, and so
forth.
A
(2e):
Experimental Section
1H NMR (500 MHz; CDCl3): d=1.02 (t, J=7.33 Hz, 3H), 2.19 (s, 6H),
2.41 (q, J=7.33 Hz, 2H), 3.07 (s, 2H), 7.17 (t, J=7.33 Hz, 1H), 7.22–7.31
(m, 5H), 7.34 (d, J=7.33 Hz, 2H), 7.40 ppm (d, J=7.33 Hz, 2H); 13C{1H}
NMR (126 MHz; CDCl3): d=13.85, 25.78, 45.02, 60.23, 126.05, 127.42,
128.69, 128.83, 129.57, 129.81, 132.61, 133.82, 135.48, 148.37 ppm; MS
(EI): m/z (%): 329 (17) [M+]; elemental analysis for the salt calcd (%)
for C21H25NO4S2: C 60.12, H 6.01, N 3.34; found: C 59.89, H 6.09, N 3.20.
General procedures: Unless otherwise noted, the synthetic work was car-
ried out under argon atmosphere. [NiACHTUNRGTNEUNG(acac)2] was dried under vacuum
(0.1–0.05 Torr, 608C, 30 min) before use. Other reagents were obtained
from Acros and Lancaster and used as supplied (checked by NMR spec-
troscopy before use). Solvents were purified according to published
methods. The reaction was carried out in PTFE screw capped tubes or
flasks.
Methyl (2Z)-3-phenyl-2,3-bis(phenylsulfanyl)-2-propenyl ether (2 f):
1H NMR (500 MHz; CDCl3): d=3.01 (s, 3H), 3.78 (s, 2H), 7.03-7.18 (m,
8H), 7.20 (d, J=6.60 Hz, 2H), 7.28 (t, J=7.23 Hz, 1H), 7.35 (t, J=
7.60 Hz, 2H), 7.55 ppm (d, J=7.16 Hz, 2H); 13C{1H} NMR (126 MHz;
CDCl3): d=57.57, 71.14, 127.21, 127.55, 127.61, 128.26, 128.33, 128.82,
129.64, 131.61, 131.76, 132.94, 133.02, 133.88, 137.34, 144.48 ppm; MS
(EI): m/z (%): 364 (7) [M+]; elemental analysis calcd (%) for C22H20OS2:
C 72.49, H 5.53, S, 17.59; found: C 72.17, H 5.40, S 17.21.
All NMR measurements were performed by using
a three-channel
Bruker AVANCE 500 spectrometer operating at 500.1, 202.5, and
125.8 MHz for H, 31P, and 13C nuclei, respectively. The spectra were pro-
1
cessed on a Linux workstation by using TOPSPIN software package. All
2D spectra were recorded using inverse triple resonance probehead with
1
active shielded Z-gradient coil. The H and 13C chemical shifts are report-
ed relative to the corresponding solvent signals used as internal refer-
1
1-Methyl-4-({(1Z)-2-[(4-methylphenyl)sulfanyl]-1-propyl-1-propenyl}sul-
fanyl)benzene (2g): 1H NMR (500 MHz; CDCl3): d=0.83 (t, J=7.52 Hz,
3H), 1.47–1.51 (m, 2H), 1.92 (s, 3H), 2.24 (t, J=7.60 Hz, 2H), 2.32 (s,
3H), 2.33 (s, 3H), 7.09–7.14 (m, 4H), 7.25 (d, J=7.26 Hz, 2H), 7.29 ppm
(d, J=7.26 Hz, 2H); 13C{1H} NMR (126 MHz; CDCl3): d=13.62, 20.23,
21.05, 21.10, 21.94, 35.53, 128.36, 129.63, 130.64, 130.87, 131.67, 132.58,
133.74, 134.51, 136.42, 137.30 ppm; MS (EI): m/z (%): 328 (42) [M+]; ele-
mental analysis calcd (%) for C20H24S2: C 73.12, H 7.36, S 19.52; found:
C 72.92, H 7.19, S 19.27.
ence. Estimated errors in the yields determined by H NMR spectroscopy
were <2%.
Developed synthetic procedure: [NiACTHNUTRGNEUNG
(acac)2] (3.0ꢂ10ꢀ5 mol, 7.8 mg), Ar2S2
(1.0ꢂ10ꢀ3 mol) and PMePh2 (3.0ꢂ10ꢀ4 mol, 60.0 mg) were placed into re-
action vessel and stirred at room temperature until homogeneous brown
solution was formed (ca. 1–2 min). Alkyne (1.5ꢂ10ꢀ3 mol) was added to
the solution and the reaction was carried out at 1008C for 8 h under stir-
ring. In case of the compound 2c 13 h were required to achieve full con-
version of Ph2S2.
1-Chloro-4-({(1Z)-2-[(4-chlorophenyl)sulfanyl]-1-propyl-1-propenyl}sulfa-
nyl)benzene (2h): 1H NMR (500 MHz; CDCl3): d=0.85 (t, J=7.42 Hz,
3H), 1.46–1.52 (m, 2H), 1.96 (s, 3H), 2.26 (t, J=7.53 Hz, 2H), 7.24 (d,
J=8.78 Hz, 2H), 7.27 (d, J=8.78 Hz, 2H), 7.28–7.30 ppm (m, 4H);
13C{1H} NMR (126 MHz; CDCl3): d=13.61, 20.42, 21.90, 35.72, 129.05,
129.10, 131.21, 132.46, 132.73, 133.34, 133.49, 133.70, 134.22, 135.28 ppm;
MS (EI): m/z (%): 368 (32) [M+]; elemental analysis calcd (%) for
C18H18Cl2S2: C 58.53, H 4.91, S 17.36; found: C 58.74, H 5.01, S 17.19.
Compound purification and characterization: After completion of the re-
action the products were purified by dry column flash chromatography
on silica.[34] Dry column flash chromatography has several practical ad-
vantages: 1) only a small amount of silica required, 2) quick elution, and
3) economy of solvents. However, slightly better isolated yields (by ꢂ5–
10%) may be achieved using conventional column chromatography.
Hexanes/dichloromethane (2 f, 2h, 2i), hexanes/benzene (2a, 2b, 2c, 2d,
2g) and hexanes/ethyl acetate (2e) gradient elution was applied. Silica
was washed with a solution of Et3N (5–6 drops) in hexanes (20 mL) prior
to chromatography of product 2e. After drying in vacuum the pure prod-
ucts were obtained. The isolated yields were calculated based on initial
amount of Ar2S2.
1-Methoxy-4-({(1Z)-2-[(4-methoxyphenyl)sulfanyl]-1-propyl-1-propenyl}-
sulfanyl) benzene (2i): 1H NMR (500 MHz; CDCl3): d=0.81 (t, J=
7.35 Hz, 3H), 1.48–1.54> (m, 2H), 1.86 (s, 3H), 2.17 (t, J=7.52 Hz, 2H),
3.80 (s, 3H), 3.81 (s, 3H), 6.85 (d, J=3.07 Hz, 2H), 6.87 (d, J=3.07 Hz,
2H), 7.34 (d, J=8.82 Hz, 2H), 7.37 ppm (d, J=8.82 Hz, 2H); 13C{1H}
NMR (126 MHz; CDCl3): d=13.55, 19.98, 21.89, 35.22, 55.30, 114.53,
114.55, 125.17, 125.82, 133.26, 133.51, 133.55, 133.86, 134.79, 159.13,
159.55 ppm; MS (EI): m/z (%): 360 (8) [M+]; elemental analysis calcd
(%) for C20H24O2S2: C 66.63, H 6.71, S 17.79; found: C 66.68, H 6.58, S
17.93.
A
(2a):
1H NMR (500 MHz; CDCl3): d=1.06 (t, J=7.39 Hz, 6H), 2.34 (q, J=
7.39 Hz, 4H), 7.21 (t, J=7.20 Hz, 2H), 7.28 (t, J=7.67 Hz, 4H), 7.35 ppm
(d, J=7.14 Hz, 4H); 13C{1H} NMR (126 MHz; CDCl3): d=14.02, 26.46,
126.60, 128.83, 130.76, 135.06, 139.13 ppm; MS (EI): m/z (%): 300 (52)
[M+]; elemental analysis calcd (%) for C18H20S2: C 71.95, H 6.71, S
21.34; found: C 72.01, H 7.06, S 21.29.
PhSH addition to 3-hexyne catalyzed by Ni/PPh3, Ni/PMePh2 and Ni/
PMe2Ph systems: [NiACTHNUTRGNEUNG
(acac)2] (1.5ꢂ10ꢀ5 mol, 3.9 mg), PhSH (5.0ꢂ
A
(2b):
10ꢀ4 mol, 55.1 mg), and PR3 (1.5ꢂ10ꢀ4 mol) were placed into reaction
vessel and stirred at room temperature for about 1–2 min. 3-Hexyne
(5.0ꢂ10ꢀ4 mol, 41.1 mg) was added to the solution and the reaction was
carried out at 1008C for 8 h under stirring.
1H NMR (500 MHz; CDCl3): d=0.84 (t, J=7.37 Hz, 3H), 1.48–1.57 (m,
2H), 1.97 (s, 3H), 2.28 (t, J=7.60 Hz, 2H), 7.19 (t, J=7.40 Hz, 1H), 7.23
(t, J=7.40 Hz, 1H), 7.25–7.31 (m, 4H), 7.33 (d, J=7.15 Hz, 2H),
7.38 ppm (d, J=7.01 Hz, 2H); 13C{1H} NMR (126 MHz; CDCl3): d=
13.60, 20.39, 21.91, 35.69, 126.33, 127.12, 128.30, 128.83, 130.07, 132.03,
134.28, 134.42, 134.97, 135.28 ppm; MS (EI): m/z (%): 300 (90) [M+]; ele-
mental analysis calcd (%) for C18H20S2: C 71.95, H 6.71, S 21.34; found:
C 71.99, 6.80, S 21.23.
Catalytic reaction with various Ni/PMePh2 ratios: [NiACTHNUTRGNEUNG
(acac)2] (3.0ꢂ10ꢀ5
mol, 7.8 mg), Ph2S2 (1.0ꢂ10ꢀ3 mol, 218.4 mg) and appropriate amount of
PMePh2 were placed into reaction vessel and stirred at room temperature
until homogeneous brown solution was formed (ca. 1–2 min). 3-Hexyne
(1.5ꢂ10ꢀ3 mol, 123.2 mg) was added to the solution and the reaction was
carried out at 1008C for 1 or 3 h under stirring.
A
(2c):
1H NMR (500 MHz; CDCl3): d=0.82 (t, J=7.33 Hz, 6H), 1.47–1.53 (m,
4H), 2.28 (t, J=7.65 Hz, 4H), 7.19 (t, J=7.07 Hz, 2H), 7.27 (t, J=
7.55 Hz, 4H), 7.34 ppm (d, J=7.14 Hz, 4H); 13C{1H} NMR (126 MHz;
CDCl3): d=13.71, 22.44, 35.15, 126.53, 128.77, 130.71, 135.10,
138.09 ppm; MS (EI): m/z (%): 328 (40) [M+]; elemental analysis calcd
(%) for C20H24S2: C 73.12, H 7.36, S 19.52; found: 72.94, 7.42, S 19.20.
Theoretical calculations: The calculations were performed using the PBE
exchange-correlation functional,[35] the scalar-relativistic Hamiltonian and
large all-electron basis sets of triple-z quality with polarization functions
as implemented in Priroda program.[36] Priroda makes use of the “resolu-
tion-of-identity” approach to solving the SCF equations and other effi-
Chem. Eur. J. 2010, 16, 2063 – 2071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2069

I. P. Beletskaya, V. P. Ananikov et al.
ciency-enhancing techniques,[37] which make the performance of the code
very efficient. Good accuracy in geometry optimization and energy calcu-
lations with Priroda program was confirmed for various systems involving
transition metal compounds.[38] Our extensive testing over the known
structures of metal chalcogenides confirmed the reliability of these calcu-
lations for the systems similar to those considered in the present study.
[14] a) M. Arisawa, M. Yamaguchi, Org. Lett. 2001, 3, 763; b) M. Arisa-
wa, Y. Kozuki, M. Yamaguchi, J. Org. Chem. 2003, 68, 8964.
[15] I. P. Beletskaya, V. P. Ananikov, Pure. Appl. Chem. 2007, 79, 1041.
[16] a) V. P. Ananikov, I. P. Beletskaya, Russ. Chem. Bull. 2004, 53, 561;
b) V. P. Ananikov, N. V. Orlov, I. P. Beletskaya, Russ. Chem. Bull.
2005, 54, 576.
[17] V. P. Ananikov, I. P. Beletskaya, Org. Biomol. Chem. 2004, 2, 284.
[18] V. P. Ananikov, M. A. Kabeshov, I. P. Beletskaya, Synlett 2005, 1015.
[19] V. P. Ananikov, K. A. Gayduk, I. P. Beletskaya, V. N. Khrustalev, M.
Yu. Antipin, Chem. Eur. J. 2008, 14, 2420.
[20] a) I. Kamiya, J. Kawakami, S. Yano, A. Nomoto, A. Ogawa, Orga-
nometallics 2006, 25, 3562; b) W. Zheng, A. Ariafard, Z. Lin, Orga-
nometallics 2008, 27, 246.
[21] H. Kuniyasu, K. Takekawa, F. Yamashita, K. Miyafuji, S. Asano, Y.
Takai, A. Ohtaka, A. Tanaka, K. Sugoh, H. Kurosawa, N. Kambe,
Organometallics 2008, 27, 4788.
Full geometry optimization was performed without any symmetry con-
straints. Normal coordinate analysis was performed for all stationary
points to verify the transition states (one imaginary frequency) and equi-
librium structures (no imaginary frequencies) and to calculate zero point
energy correction and Gibbs free energies. Utilizing intrinsic reaction co-
ordinate calculations, reactants, transition states, and corresponding prod-
ucts were unambiguously proven to be connected by a single minimal-
energy reaction path.
[22] V. P. Ananikov, N. V. Orlov, M. A. Kabeshov, I. P. Beletskaya, Z. A.
Starikova, Organometallics 2008, 27, 4056.
Acknowledgements
[23] Selected references on the application of the vinyl sulfides as build-
ing blocks in organic synthesis, as ligands in catalysis and structural
motifs in material science: a) G. Matsubayashi, M. Nakano, H.
Tamura, Coord. Chem. Rev. 2002, 226,143; b) N. Robertson, L.
Cronin, Coord. Chem. Rev. 2002, 227, 93; c) M. Fourmiguꢄ, Acc.
Chem. Res. 2004, 37, 179; d) A. Kobayashi, E. Fujiwara, H Kobaya-
shi, Chem. Rev. 2004, 104, 5243; e) R. Kato, Chem. Rev. 2004, 104,
5319; f) U. T. Mueller-Westerhoff, B. Vance, D. I. Yoon, Tetrahedron
1991, 47, 909; f) S. D. P. Baugh, Z. Yang, D. K. Leung, D. M. Wilson,
R. Breslow, J. Am. Chem. Soc. 2001, 123, 12488; g) B. J. Vesper, K.
Salaita, H. Zong, C. A. Mirkin, A. G. M. Barrett, B. M. Hoffman, J.
Am. Chem. Soc. 2004, 126, 16653; h) S. Belviso, G. Ricciardi, F. Lelj,
J. Mater. Chem. 2000, 10, 297; i) S. H. Eichhorn, D. W. Bruce, D.
Guillon, J.-L. Gallani, T. Fischer, J. Stumpe, T. Geue, J. Mater.
Chem. 2001, 11, 1576; j) F. Bonosi, G. Ricciardi, F. Lelj, G. Martini,
J. Phys. Chem. 1994, 98, 10613; k) S. B. Sesalan, A. Gul, Monatsh.
Chem. 2000, 131, 1191; l) H. Braun, A. Amann, Angew. Chem. 1975,
87, 773; Angew. Chem. Int. Ed. Engl. 1975, 14, 755; m) M. E. Garst,
P. Arrhenius, J. Org. Chem. 1983, 48, 16.
The research was supported by the Russian Foundation for Basic Re-
search (Project No. 10-03-00370), grant MK-1085.2008.3, and Program
No 1 of Division of Chemistry and Material Sciences of RAS.
[1] a) Catalytic Heterofunctionalization (Eds.: A. Togni, H. Grꢃtzmach-
er), Wiley-VCH, Weinheim, 2001; b) M. Beller, J. Seayad, A. Till-
ack, H. Jiao, Angew. Chem. 2004, 116, 3448; Angew. Chem. Int. Ed.
2004, 43, 3368; c) T. Kondo, T. Mitsudo, Chem. Rev. 2000, 100, 3205;
d) I. Beletskaya, C. Moberg, Chem. Rev. 2006, 106, 2320; e) F.
Alonso, I. P. Beletskaya, M. Yus, Chem. Rev. 2004, 104, 3079; f) M.
Suginome, Y. Ito, J. Organomet. Chem. 2003, 685, 218; g) M. Sugi-
nome, Y. Ito, J. Organomet. Chem. 2003, 680, 43; h) A. Ogawa, J.
Organomet. Chem. 2000, 611, 463; i) D. M. Ide, M. P. Eastlund, C. L.
Jupe, R. A. Stockland, Jr., Curr. Org. Chem. 2008, 12, 1258.
[2] L. L. Hegedus, R. W. McCabe, Catalyst Poisoning, Marcel Dekker,
New York, 1984.
[3] I. P. Beletskaya, V. P. Ananikov, Eur. J. Org. Chem. 2007, 3431.
[4] a) H. Kuniyasu, A. Ogawa, K. Sato, I. Ryu, N. Kambe, N. Sonoda, J.
Am. Chem. Soc. 1992, 114, 5902; b) A. Ogawa, T. Ikeda, K. Kimura,
T. Hirao, J. Am. Chem. Soc. 1999, 121, 5108.
[5] V. P. Ananikov, D. A. Malyshev, I. P. Beletskaya, G. G. Aleksandrov,
I. L. Eremenko, Adv. Synth. Catal. 2005, 347, 1993.
[24] a) P. W. Jolly, G. Wilke, The Organic Chemistry of Nickel, Academic
Press, New York, 1974; b) “Polymerization of Acetylenes”: T.
Masuda, F. Sanda, M. Shiotsuki in Comprehensive Organometallic
Chemistry III, Vol. 11 (Eds.: D. M. P. Mingos, R. H. Crabtree),
Elsevier, Oxford, 2007, pp. 557–593.
[6] a) C. Cao, L. R. Fraser, J. A. Love, J. Am. Chem. Soc. 2005, 127,
17614; b) Y. Misumi, H. Seino, Y. Mizobe, J. Organomet. Chem.
2006, 691, 3157; c) S. Shoai, P. Bichler, B. Kang, H. Buckley, J. A.
Love, Organometallics 2007, 26, 5778; d) L. R. Fraser, J. Bird, Q.
Wu, C. Cao, B. O. Patrick, J. A. Love, Organometallics 2007, 26,
5602; e) S. Burling, L. D. Field, B. A. Messerle, K. Q. Vuong, P.
Turner, J. Chem. Soc. Dalton Trans. 2003, 4181.
[7] a) V. P. Ananikov, N. V. Orlov, I. P. Beletskaya, Organometallics
2006, 25, 1970; b) V. P. Ananikov, N. V. Orlov, I. P. Beletskaya, Orga-
nometallics 2007, 26, 740.
[25] In case of PPh3 ligand a slightly lower yield was observed due to by-
products formation through the known mechanism involving benze-
nethiol conversion to diphenyl disulfide (see ref. [5]).
[26] The insertion step is known to proceed with high selectivity and
leads to the formation of syn-addition products with Z configuration
of the double bond.[1,3] This rules out an alternative route involving
external attack of the sulfur group, which results in formation of
anti-addition products with E configuration of the double bond.
[27] See the Supporting Information for detailed description of X-ray
study and geometry parameters.
[28] See Figure S1 in the Supporting Information for the optimized struc-
tures I, II-TS, IV, V, VI-TS, and VIII.
[29] The calculated binding energies of the C=C double bond in the p
complexes III and VII were 15.9 and 9.1 kcalmolꢀ1 relative to IV
and VIII, respectively.
[8] V. P. Ananikov, N. V. Orlov, I. P. Beletskaya, V. N. Khrustalev, M.
Yu. Antipin, T. V. Timofeeva, J. Am. Chem. Soc. 2007, 129, 7252.
[9] In the absence of the phosphine ligand another mechanism may
take place involving protonolysis as
a product formation step
(Scheme 2). See reference [3] for discussion of various mechanistic
details. In any case, alkyne coordination and insertion remain as the
key steps in all reported reactions.
[30] For a related discussion of the influence of in-plane and out-of-
ꢀ
plane ligand orientation on C C reductive elimination see: a) V. P.
[10] Only mononuclear transition metal complexes are shown on the
scheme for clarity reasons, see reference [3,12] for more details.
[11] H. Kuniyasu, A. Ogawa, S. Miyazaki, I. Ryu, N. Kambe, N. Sonoda,
J. Am. Chem. Soc. 1991, 113, 9796.
[12] a) V. P. Ananikov, I. P. Beletskaya, G. G. Aleksandrov, I. L. Eremen-
ko, Organometallics 2003, 22, 1414; b) V. P. Ananikov, M. A. Kabe-
shov, I. P. Beletskaya, V. N. Khrustalev, M. Yu. Antipin, Organome-
tallics 2005, 24, 1275.
Ananikov, D. G. Musaev, K. Morokuma, J. Am. Chem. Soc. 2002,
124, 2839; b) V. P. Ananikov, D. G. Musaev, K. Morokuma, Organo-
metallics 2005, 24, 715; c) V. P. Ananikov, D. G. Musaev, K. Moroku-
ma, Eur. J. Inorg. Chem. 2007, 5390.
[31] R. Ghosh, X. Zhang, P. Achord, T. J. Emge, K. Krogh-Jespersen,
A. S. Goldman, J. Am. Chem. Soc. 2007, 129, 853.
[32] In addition to product 3a (ꢂ50%) the formation of 1,3-dienes was
also observed according to a known reaction. Both reactions—for-
mation of alkene 3a and dienes—involve alkyne insertion and re-
[13] H. Kuniyasu, F. Yamashita, J. Terao, N. Kambe, Angew. Chem. 2007,
119, 6033; Angew. Chem. Int. Ed. 2007, 46, 5929.
2070
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2063 – 2071

ꢀ
C S Bond Formation
FULL PAPER
ductive elimination steps (see ref. [22] for the discussion of the
mechanism of dienes formation).
Chernyshev, Dalton Trans. 2009, 2482; b) A. A. Popov, A. V. Burt-
sev, V. M. Senyavin, L. Dunsch, S. I. Troyanov, J. Phys. Chem. A
2009, 113, 263; c) J. R. Khusnutdinova, L. L. Newman, P. Y. Zavalij,
Y.-F. Lam, A. N. Vedernikov, J. Am. Chem. Soc. 2008, 130, 2174;
d) A. A. Popov, L. Dunsch, J. Am. Chem. Soc. 2008, 130, 17726;
e) A. Yahav-Levi, I. Goldberg, A. Vigalok, A. N. Vedernikov, J. Am.
Chem. Soc. 2008, 130, 724; f) G. A. Shamov, G. Schreckenbach,
R. L. Martin, P. J. Hay, Inorg. Chem. 2008, 47, 1465; g) J. J. Berard,
G. Schreckenbach, P. L. Arnold, D. Patel, J. B. Love, Inorg. Chem.
2008, 47, 11583; h) S. Sanz, L. A. Jones, F. Mohr, M. Laguna, Orga-
nometallics 2007, 26, 952; i) G. A. Shamov, G. Schreckenbach, J.
Phys. Chem. A 2006, 110, 9486; j) E. N. Esenturk, J. Fettinger, B.
Eichhorn, J. Am. Chem. Soc. 2006, 128, 9178; k) Y. Smurnyy, C.
Bibal, M. Pink, K. G. Caulton, Organometallics 2005, 24, 3849; l) C.
Bibal, Y. D. Smurnyy, M. Pink, K. G. Caulton, J. Am. Chem. Soc.
2005, 127, 8944.
[33] It should be noted that an additional role of the ligand excess is to
suppress polymerization of Ni sulfide and formation of metal-con-
taining precipitate as was reported earlier.[12,17,19] At the moment it
is impossible to separate unambiguously these two types of the
ligand effects and to estimate their contribution to the overall per-
formance enhancement of the studied catalytic reaction. In any case,
it is clear that the pathway involving alkyne coordination to the
apical Ni site without the need of phosphine ligand dissociation
better agrees with the results of the NMR monitoring.
[34] a) D. S. Pedersen, C. Rosenbohm, Synthesis 2001, 2431; b) L. M.
Harwood, Aldrichim. Acta 1985, 18, 25.
[35] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[36] Priroda, an electronic structure code program, D. N. Laikov Moscow
State University, Moscow (Russia), 2004.
[37] a) D. N. Laikov, Chem. Phys. Lett. 1997, 281, 151; b) D. N. Laikov,
Chem. Phys. Lett. 2005, 416, 116.
[38] Some selected recent examples: a) N. M. Logacheva, V. E. Baulin,
A. Yu. Tsivadze, E. N. Pyatova, I. S. Ivanova, Y. A. Velikodny, V. V.
Received: October 22, 2009
Published online: January 28, 2010
Chem. Eur. J. 2010, 16, 2063 – 2071
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2071