Remarkable ligand effect in Ni- And Pd-catalyzed bisthiolation and bisselenation of terminal alkynes: Solving the problem of stereoselective dialkyldichalcogenide addition to the C≡C bond

Article abstract of DOI:10.1002/chem.200701278

We have developed two new catalytic systems based on Ni and Pd complexes to solve the challenging problem of dialkyldichalcogenide (Alk₂E ₂; E = S, Se) addition to alkynes. A comparative study of two catalytic systems - Ni/PMe₂Ph and Pd/PCy₂Ph - has revealed that the Ni catalyst is superior with respect to high catalytic activity and more general scope relative to the Pd system. A novel synthetic methodology was developed for the preparation of (Z)-bis(alkylthio)alkenes and (Z)-bis(alkylseleno)alkenes from terminal alkynes with excellent stereoselectivity and high yields.

Full text of DOI:10.1002/chem.200701278

DOI: 10.1002/chem.200701278
Remarkable Ligand Effect in Ni- and Pd-Catalyzed Bisthiolation and
Bisselenation of Terminal Alkynes: Solving the Problem of Stereoselective
ꢀ
Dialkyldichalcogenide Addition to the C C Bond
[a]
[b]
Valentine P. Ananikov,* Konstantin A. Gayduk,[a] Irina P. Beletskaya,*
Victor N. Khrustalev,[c, d] and Mikhail Yu. Antipin[c, d]
Abstract: We have developed two new catalytic systems based on Ni and Pd com-
plexes to solve the challenging problem of dialkyldichalcogenide (Alk2E2; E=S,
Se) addition to alkynes. A comparative study of two catalytic systems—Ni/PMe2Ph
and Pd/PCy2Ph—has revealed that the Ni catalyst is superior with respect to high
catalytic activity and more general scope relative to the Pd system. A novel syn-
thetic methodologywas developed for the preparation of ( Z)-bis(alkylthio)alkenes
and (Z)-bis(alkylseleno)alkenes from terminal alkynes with excellent stereoselec-
tivityand high yields.
Keywords: alkynes · chalcogens ·
heterogeneous catalysis · nickel ·
palladium
Introduction
In the pioneering work, A. Ogawa et al.[1] described the ste-
reoselective addition of Ar2S2 and Ar2Se2 to terminal
ACHTREUNG
alkynes by using a palladium catalyst (path a; Scheme 1).[2]
We have shown that an excess of the phosphane ligand is re-
quired to achieve a high yield in this reaction.[3] Byutilizing
Pd complexes and an excess of the phosphane ligand, new
synthetic protocols have been developed to perform the
Ar2E2 addition to alkynes in a molten state (solvent free
conditions) and with a polymer-supported catalyst.[4] An un-
Scheme 1. Addition of diaryldichalcogenides and dialkyldichalcogenides
to alkynes.
[a] Prof. V. P. Ananikov, K. A. Gayduk
ZelinskyInstitute of Organic Chemistry
Russian Academyof Sciences, LeninskyProspect 47
Moscow, 119991 (Russia)
selective non-catalytic side reaction (path b; Scheme 1) was
totallysuppressed in the case of regular alknyes and the
contribution of the side reaction was significantlyreduced in
the case of activated alkynes.[3] However, the Pd catalyst
showed rather poor efficiencyin the case of the dialklydi-
chalcogenide (Alk2E2; E=S, Se) addition to alkynes (path
c; Scheme 1), this reaction required a more active Rh cata-
lyst.[5]
Fax : (+7)495-135-5328
E-mail: val@ioc.ac.ru
[b] Prof. I. P. Beletskaya
Lomonosov Moscow State University, Chemistry Department
Vorobꢀevygory, Moscow, 119899 (Russia)
Fax : (+7)495-939-3618
E-mail: beletska@org.chem.msu.ru
A similar reactivitypattern was observed in another reac-
tion, that is, the transition-metal-catalyzed addition of thiols
and selenols (REH; R=Ar, Alk; E=S, Se) to alkynes. This
reaction provides a convenient synthetic route to a-substi-
tuted vinylchalcogenides with high regioselectivity (Markov-
nikov-type addition).[2,3] Pd catalysts were utilized for the
addition of arylthiols and arylselenols to alkynes;[6] however,
[c] Dr. V. N. Khrustalev, Prof. M. Yu. Antipin
Nesmeyanov Institute of Organoelement Compounds
Russian Academyof Sciences, Moscow (Russia)
[d] Dr. V. N. Khrustalev, Prof. M. Yu. Antipin
Department of Natural Sciences
New Mexico Highlands University
Las Vegas, NM 87701 (USA)
2420
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

FULL PAPER
addition of alkane thiols was carried out onlywith a Rh cat-
alytic system.[7]
Recently, we have reported on the heterogeneous catalyt-
lytic systems were also developed and the scope of both the
Ni- and Pd-catalyzed transformations has been investigated
in a comparative manner.
ic system, with [{NiACHTRE(UGN EAr)2}n] as a polymeric catalyst, that
lead to the addition of ArEH to alkynes with excellent se-
lectivity and yields under mild conditions (80–99% yields,
Results
>95:5 selectivity).[8] On the other hand, the [{Pd
ACHTREU(NG EAlk)2}n]
polymer was the material of choice for constructing a net-
work of 1D nanobelts, which was a remarkable catalyst for
the reaction between alkane thiols and alkynes, resulting in
Ni-catalyzed bisthiolation and bis-selenation of alkynes: The
performance of the catalytic system was studied by using di-
butyldisulfide addition to 1-hexyne (1a) as a model reaction.
Initial investigation of the ligand effect was carried out in
benzene at 808C for 4 h. Under these conditions traditional
Pd0/PPh3 catalytic system catalyzes Ar2S2 addition to 1a, but
failed to catalyze Bu2S2 addition (onlytraces of 2a were de-
tected). Simple replacement of Pd complex byNi did not
improve the catalytic activity of the studied system (entry 1,
Table 1). Utilization of various arylphosphanes with elec-
[9]
70–99% yield and >95:5 selectivity. It was the first suc-
cessful example of the replacement of Rh complexes byPd
in the catalytic addition reaction involving AlkE groups.
These fascinating results gave us a hope to overcome the
problem of Alk2E2 addition to alkynes and encouraged us to
carryout a comparative studyof the catalytic activityof Ni
and Pd complexes.
The products of Alk2E2 addition to alkynes, (Z)-bis-
ACHTREUNG(alkylthio)alkenes and (Z)-bis(alkylseleno)alkenes (2), are
Table 1. Ligand effect on the yield of 2a in Ni-catalyzed reaction of
in demand in various synthetic transformations and in mate-
rials science.[10,11] Undoubtedly, development of simple and
efficient synthetic procedure is required to further explore
the potential of such compounds.
Bu2S2 and 1-hexyne.[a]
To understand the mechanistic nature of the problem, it is
worth considering the difference in bond dissociation ener-
gies between the diaryldisulfides and dialkyldisulfides. The
Ligand (L)
PPh3
Yield [%] Ligand (L) Yield [%]
1
2
3
4
5
6
7
8
9
0
0
0
0
0
0
0
0
7
3
11
12
13
14
15
16
17
18
19
20
dppe
dppb
PCy3
PCy2Ph
PCyPh2
0
9
0
0
0
2
4
10
30
38
P
A
ꢁ
S S bond in diphenyldisulfide is considerably weaker than
P(o-MeOC6H4)3
P(p-MeOC6H4)3
ꢁ
the C S bond, as reflected bythe values of bond dissocia-
P
P
P
P
P
C
(MeO)2C6H3]3
tion energy(Scheme 2). The opposite relationship is
PPh
2A
PEtPh2
PMePh2
PBu3
10 dppm
PMe2Ph
[a] Bu2S2 (1 mmol) and 1-hexyne (1 mmol) in benzene (0.5 mL).
[12]
ꢁ
ꢁ
Scheme 2. C S and S S bonds dissociation energy.
tron-donor and electron-acceptor substituents did not lead
to desired product (entry2–6, Table 1). The first favorable
evidence was obtained when using phosphite ligands (en-
tries 9, Table 1). The catalytic system based on the tri(iso-
propyl)phosphite resulted in 7% of 2a (entries 9, Table 1),
while tributylphosphite and triphenylphosphite were inac-
tive (entries 7, 8, Table 1). A minor increase in the yield up
to 9% was obtained with bidentante ligand bis(diphenyl-
phosphino)butane (dppb) (entry12, Table 1), bis(diphenyl-
phosphino)methane (dppm) gave a trace amount of the
product (3%) and bis(diphenylphosphino)ethane (dppe)
was totallyinactive (entries 10, 11; Table 1). Introducing
bulky alkyl group (Cy) into the ligand did not lead to 2a
(entries 13–15, Table 1), but the presence of less bulkyalkyl
groups in the phosphane ligand increased the yield of 2a to
10% (entries 16–18, Table 1). Surprisingly, good yield of the
product (30%) was observed with PBu3 ligand (entry19,
Table 1), and even better yield of 38% was found in the
case of PMe2Ph (entry20, Table 1). Further optimization of
the reaction conditions were carried out with the most
promising PMe2Ph ligand. The catalytic reaction with this
ꢁ
ACHTREUNGobserved in the case dialkyldisulfide: the S S bond is stron-
ꢁ1
ꢁ
ger than the C S bond by ꢂ20 kcalmol . Therefore, a
more efficient catalyst should be designed to break the
stronger S S bond of dialkyldisulfides as opposed to the
easier case of diaryldisulfides. In addition to high efficiency,
the catalyst should be very selective for breaking the strong
ꢁ
ꢁ
ꢁ
S S bond and not to have an effect on the weak C S bond.
The difference between the substrates outlines the principal
difficulties in carrying out the desired catalytic transforma-
tion (compare paths a and c, Scheme 1). The same consider-
ations are also applicable for the Se analogues and will not
be repeated here.
In the present article we describe a novel catalytic system
for carrying out the stereoselective addition of dialkyldisul-
fides and dialkyldiselenides to alkynes in good yields. Cheap
and easilyavailable Ni compounds were utilized as catalysts.
The keyfindings concerning the ligand of choice and relia-
ble reaction conditions will be discussed in view of a plausi-
ble reaction mechanism. Wherever possible Pd-based cata-
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2421

I. P. Beletskaya, V. P. Ananikov et al.
Table 3. The yield of 2a in Ni-catalyzed reaction with different amounts
ligand was highlystereoselective, with onlythe formation of
the Z isomer (syn addition) being observed.
of the PMe2Ph ligand.[a]
Ligand [mol%]
Yield [%]
Prolonging the reaction time to 8 and 12 h increased the
yield of 2a to 55 and 60%, respectively(entries 1,2;
Table 2). Unfortunately, further heating at 808C did not
solvent free
in benzene[b]
1
2
3
4
15
30
45
60
67
87
78
75
58
65
63
60
Table 2. The yield of 2a at different temperatures in solvent and under
solvent free conditions.[a]
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol), [NiACHTRE(UNG acac)2] (3 mol%) and given
amount of PMe2Ph ligand (1008C, 2 h). [b] In benzene (0.5 mL).
Solvent
T [oC]
t [h]
Yield [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
benzene
benzene
benzene
benzene
benzene
benzene
acetone
toluene
80
80
8
12
1.5
2
2.5
4
4
4
4
4
55
60
61
65
66
60
58
54
52
23
46
57
87
50
67
lyst did not lead to further improvements, probably due to
solubilityreasons.
Finally, we studied the influence of the Bu2S2:alkyne ratio
on the yield of product 2a (Table 4). Use of an excess of the
100
100
100
100
100
100
100
70
Table 4. [Ni
ratios.[a]
ACHTRE(UGN acac)2]/PMe2Ph catalyzed reaction with different Bu2S2:1a
THF
solvent free
solvent free
solvent free
solvent free
solvent free
solvent free
Bu2S2:1a
Yield[b] [%]
80
4
1
2
4
solvent free
in benzene[c]
100
100
100
110
1
2
3
4
2:1
1:1
1:2
1:3
66
87
90
95
58
73
80
94
2
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol), [Ni
(30 mol%) in solvent (0.5 mL) or solvent free.
ACHTRE(UNG acac)2] (3 mol%), PMe2Ph
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol), [NiACHTRE(UGN acac)2] (3 mol%), and
PMe2Ph (30 mol%) (1008C, 2 h). [b] For entry1 the yield was calculated
based on initial amount of 1a; for entries 2–4 based on initial amount of
Bu2S2. [c] In benzene (0.5 mL).
ACHTREUNGimprove the yield. Slightly better yields of 61–66% were ob-
served bycarrying out the reaction at 100 8C (entries 3–5,
Table 2). It is veryimportant to point out that longer heat-
ing caused a decrease in the yield of 2a (entry6, Table 2).
Trying different media for the reaction did not solve the
problem, leading to <60% yield of the product (entries 7–9,
Table 2). Much better results were found byapplying a sol-
vent-free reaction under optimized conditions (entries 10–
13, Table 2). Carrying out the catalytic reaction at 1008C for
2 h led to 87% of 2a. At this point (2 h) the alkyne was
completelyconsumed, while some Bu 2S2 (ꢂ5–10%) re-
mained unreacted. This suggests that the alkyne was in-
volved in a side reaction. Longer reaction time again result-
ed in a decrease of the yield of 2a (entry14, Table 2). The
yield of 2a observed at 1108C (entry15, Table 2) was small-
er than that observed at 1008C (entry13, Table 2) indicating
that further increase of the temperature is not feasible.
Therefore, 1008C and 2 h were found to be a good estimate
of reliable reaction conditions. The reasons of decreasing
product yield at higher temperature or longer time as well
as the nature of the side reactions will be analyzed further
on.
Bu2S2 over 1a decreased the yield of 2a compared to equi-
molar ratio of the reagents (entries 1, 2; Table 4). Use of an
excess of the alkyne over Bu2S2 led to noticeable improve-
ment, reaching a 95% yield at a Bu2S2:1a=1:3 ratio (en-
tries 3, 4; Table 4).
The scope of the developed catalytic system was investi-
[13]
gated for a varietyof terminal alkynes
and dialkyldisul-
fides with different substituents (Table 5). The scope of the
catalytic system will be discussed for the substrates ratio
Bu2S2:alkyne=1:3, which results in better yields. Despite
the lower yields, the data on the equimolar Bu2S2:alkyne
ratio is also presented in Table 5, since this protocol is eco-
nomic in terms of consumed alkyne and could be also useful
in some cases.
Non-activated (1a) and activated (1b) alkynes reacted
with Bu2S2 in good yields and selectivity (entries 1, 2;
Table 5); this result is an important advantage of the devel-
oped catalytic system. For instance, known synthetic proce-
dures of Ar2S2 addition to activated alkynes usually result in
poor yields and stereoselectivity.[14]
Our catalytic system was tolerant to functional groups in
alkynes leading to high yields and excellent selectivity for
1c–1e (entries 3–5, Table 5). The onlyexception was ob-
served for alkynes bearing an OH group (entries 6, 7;
Table 5). In these cases we were unable to improve the yield
of desired addition product byapplying a 1:3 reagent ratio,
due to quick polymerization of the alkynes and tar forma-
tion. To avoid polymerization in the case of 1 f, the reaction
We have found that the amount of phosphane ligand
plays an important role in the studied catalytic system
(Table 3). In both cases—reaction in benzene and under sol-
vent-free conditions—the best results were achieved with
the ratio Ni:L=1:10 (entry2, Table 3). Using smaller
(entry1, Table 3) or greater (entries 3, 4; Table 3) ratios di-
minished the performance of the catalytic reaction.
In all reactions throughout the article 3 mol% of Ni com-
pound was used; an attempt to increase amount of the cata-
2422
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
Table 5. The scope of the Ni-catalyzed addition of dialkyldisulfides to alkynes.[a]
same excellent stereoselectivity
and with a slightlybetter yield
with respect to the 1 mmol
ꢁ
Alkyne
RS SR
Product[h]
Yield[b] [%]
R2S2:1=1:1[c]
R2S2:1=1:3[d]
scale
Table 5).
reaction
(entry1,
87 (75) 90 (78)[i]
67 (60)
95 (80)
ꢁ
1
2
1a
1b
BuS SBu
2a
In short, the developed cata-
lytic system allows the addition
dialkyldisulfides to alkynes in
good isolated yields (58–87%)
and excellent stereoselectivity
(>99:1). Rapid, drycolumn
flash chromatography[16] was
successfullyutilized for isola-
tion and purification of all
ꢁ
BuS SBu
2b
95 (81)
ꢁ
3
4
5
1c
1d
1e
BuS SBu
2c
2d
2e
60 (55)
54 (48)
82 (75)
85 (75)
77 (71)[e]
85 (80)
ꢁ
BuS SBu
products
study.
throughout
the
ꢁ
BuS SBu
[17]
The synthetic procedure op-
ꢁ
timized for the addition of S S
ꢁ
6
7
1 f
1g
BuS SBu
2 f
2g
63 (58)[f]
74 (61)
–
–
bond to alkynes showed excel-
lent results in the addition of
ꢁ
the Se Se bond (Table 6). The
ꢁ
BuS SBu
Ni-catalyzed
addition
of
Bu2Se2 to 1a and 1e was car-
ried out with high isolated
yields of 87–88% (entries 1, 2;
Table 6). Addition of Me2Se2
to 1e was performed with 59%
ꢁ
8
9
1a
1g
1a
1g
MeS SMe
2h
2i
65 (55)
85 (76)
72 (60)
68 (64)
70 (60)
92 (81)
76 (62)
95 (84)
ꢁ
MeS SMe
isolated
yield
(entry 3,
Table 6). All the studied bis-se-
lenation reactions were carried
out with high stereoselectivity
(Z:E>99:1). Compared with
the Ni-catalyzed bisthiolation
reaction discussed earlier, Ni-
ꢁ
10
11
iPrS SiPr
2j
ꢁ
iPrS SiPr
2k
12
13
1a
1g
CyS SCy
2 L
2m
67 (60)[g]
75 (65)[g]
95 (81)[e]
94 (87)[e]
ꢁ
ꢁ
catalyzed Se Se bond addition
to alkynes suffers less from
side reactions and high yields
were obtained even at 1:1 ratio
of the substrates (Table 6).
ꢁ
CyS SCy
[a] See experimental part for complete description of synthesis and isolation details. [b] NMR yield after com-
pleting the reaction and isolated yield after separation and purification (in parenthesis). [c] 1008, 2 h, solvent
free (Method A; see Experimental Section). [d] 1008C, 1 h, solvent free (Method B; see Experimental Sec-
tion). [e] Reaction time 2 h. [f] 808C, 4 h, 0.5 mL of benzene (Method C; see Experimental Section). [g] Re-
action time 4 h. [h] Stereoselectivity >99:1. [i] The reaction on 10 mmol scale.
Table 6. Ni-catalyzed addition of Alk2Se2 to alkynes.[a]
was carried out at 808C in benzene (entry6, Table 5). Com-
pounds 2 f and 2g were obtained in good isolated yields of
58–61%.
Alkyne
RSe SeR
Product
Yield[b] [%]
95 (87)
ꢁ
ꢁ
1
1a
BuSe SeBu
2n
Other dialkyldisulfides, for example, Me2S2, iPr2S2, and
Cy2S2, were successfullyused in the addition reaction (en-
tries 8–13, Table 5). In the case of Cy2S2, a longer reaction
time (4 h) was required to compensate for the lower reactiv-
ꢁ
2
3
1e
1e
BuSe SeBu
2o
2p
95 (88)
71 (59)
[15]
ityof this bulkysubstrate.
ꢁ
MeSe SeMe
An important advantage of the developed synthetic pro-
cedure is the abilityto scale up to greater amount of the
product. Carrying out the catalytic reaction of 1a and Bu2S2
at 10 mmol scale resulted in 90% NMR yield (78% isolated
yield, >2 g of 2a). The reaction was carried out with the
[a] Conditions: 1008C, 2 h, 3 mol% of [NiCAHTRE(UNG acac)2], solvent free (Method
A; see Experimental Section). [b] NMR spectrscopic yield after complet-
ing the reaction and isolated yield after separation and purification (in
parenthesis).
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2423

I. P. Beletskaya, V. P. Ananikov et al.
Table 9. The yield of 2a in Pd-catalyzed reaction with different amounts
Pd-catalyzed bisthiolation and bis-selenation of alkynes:
The good results obtained for the Ni complexes led us to re-
investigate the possibility of creating a catalytic system
based on Pd. Most of the studied ligands did not result in
good yields of 2a in the model reaction of Bu2S2 addition to
1a; selected representative examples are listed in Table 7
(entries 1–7). Surprisingly, introducing the bulky Cy group
of the PCy2Ph ligand.[a]
Ligand [mol%]
Yield [%]
1
2
3
4
15
30
45
60
49
76
95
90
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol), [Pd2ACHTRE(UNG dba)3] (1.5 mol%), and a
given amount of PCy2Ph ligand (1408C, 6 h).
Table 7. Ligand effect on the yield of 2a in Pd-catalyzed reaction of
Bu2S2 and 1-hexyne.[a]
However, taking into account that the Pd system does not
suffer from high temperatures and longer reaction times, uti-
lization of 30 mol% is also acceptable.
A practical and convenient synthetic procedure was devel-
oped for the Pd-catalyzed reaction at 1408C[18] for 12 h with
30 mol% of the ligand or at 1408C for 6 h with 45 mol% of
the ligand (Table 10). Excellent isolated yields of 83–89%
and excellent stereoselectivity( >99:1) were observed at the
ratio Bu2S2:alkyne=1:1 as shown for the selected examples
of 2a, 2e, and 2d (entries 1–3, Table 10).
Ligand (L)
PPh3
PACHTREUNG(OiPr)3
dppb
PEtPh2
PBu3
PMePh2
PMe2Ph
PCy3
Yield [%]
solvent free
in benzene[b]
1
2
3
4
5
6
7
8
9
6
4
3
3
2
3
2
3
8
4
2
3
3
2
3
2
3
6
Table 10. Pd-catalyzed Alk2S2 addition to alkynes.[a]
PCyPh2
PCy2Ph
10
16
10
Yield[b] [%]
ꢁ
RS SR
Alkyne
Product
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol). [b] In benzene (0.5 mL).
ꢁ
1
2
3
1a
1e
1d
BuS SBu
2a
2e
2d
95 (85)
95 (89)
92 (83)[c]
ꢁ
BuS SBu
ꢁ
BuS SBu
to the phosphane ligand improved the performance of the
catalytic reaction (entries 8–10, Table 7). The best result was
achieved with the PCy2Ph ligand—16% of 2a after 2 h at
1008C (entry10, Table 7). Although in the case of Pd a
much smaller difference was observed between the solvent-
free reaction and that in solvent (Table 7), we have chosen
solvent-free methodologyfor green chemistryreasons.
[a] Conditions: [Pd2ACHTRE(UNG dba)3] (1.5 mol%), PCy2Ph (45 mol%), solvent free,
1408C, 6 h (or PCy2Ph (30 mol%), 1408C, 12 h). [b] NMR spectroscopic
yield after completing the reaction and isolated yield after separation
and purification (in parenthesis). [c] Reaction time: 9 h (45 mol% of
PCy2Ph) or 18 h (30 mol% of PCy2Ph).
Pd-catalyzed transformation was much slower (16% of
2a; entry10, Table 7) compared to the Ni-catalyzed trans-
formation (87% of 2a; entry13, Table 2) under the same
conditions (1008C, 2 h). However, in contrast to Ni-cata-
lyzed reaction, longer heating and higher temperatures ap-
plied in the Pd catalytic system did not cause noticeable de-
crease in the product yield (Table 8). It was possible to carry
out Pd-catalyzed reaction in the temperature range 100–
2008C with 16–80% yield of 2a after 2 h (entries 1–5,
Table 8). Further heating increased the yield upon complete
conversion of the substrates.
An attempt to carryout catalytic Bu 2Se2 addition to 1a in
Pd0/PCy2Ph system resulted only in 60% yield of 2n (55%
isolated yield) and about 40% of 1a remained unreacted
1
(Scheme 3). Two-dimensional H-77Se HMQC NMR analysis
of the reaction crude confirmed the presence of Bu2Se
(ꢂ30%).
In the case of Pd-catalyzed reaction the best yield was
achieved with 45 mol% of the ligand (entries 1–4; Table 9).
Scheme 3. Pd-catalyzed reaction of Bu2Se2 and 1a.
Table 8. The yield of 2a at different temperatures in Pd-catalyzed reac-
tion.[a]
31P{1H} examination of the crude product indicated the
formation of Se=PCy2Ph, presumablyfrom the following re-
action: PCy2Ph+Bu2Se2!Se=PCy2Ph+Bu2Se, which also
takes place under Pd-catalyzed conditions. To confirm this
assumption we carried out the reaction with equimolar ratio
of PCy2Ph and Bu2Se2 in the absence of alkyne. After 6 h at
1408C nearlycomplete conversion of the phosphane to Se =
PCy2Ph (ꢂ95%) was detected byNMR spectroscopy.
T [8C]
Yield [%]
1
2
3
4
5
100
120
140
160
200
16
26
44
64
80
[a] Bu2S2 (1 mmol), 1-hexyne (1 mmol), [Pd
PCy2Ph (30 mol%), 2 h, solvent free.
2ACHTRE(UNG dba)3] (1.5 mol%), and
2424
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
X-ray structure determination of 2e, 2i, 2k, and 2m: The Z
configuration of most of the products was determined by2D
NOESY NMR experiments and some representative cases
were a subject of X-raystructure determination. For this
purpose we have prepared oxalic salts of 2i, 2e, 2k, and 2m
(crystallized from MeOH). The molecular structures are
shown in Figure 1, along with the atom numbering schemes.
Selected bond lengths and angles are listed in Table 11. The
X-rayanalysis confirmed the Z configuration of the double
bonds and revealed several unusual structural and crystal
packing features.
The salts comprise the ammonium sulfur-containing
[(RS)HC=C(SR)CH2NHMe2]+ ions and the oxalic acid
[HOOC-COO]ꢁ ions forming tight ionic pairs bybifurcate
hydrogen bonds between the amino-H atom of the cation
and two oxygen atoms of the anion (Figure 1, Table 12). It is
interesting to note that different oxygen atoms of the anion
take part in the hydrogen-bonding interactions, namely,
either the carbonyl and hydroxyl oxygen atoms (in the case
of 2i and 2m) or the two carbonyl oxygen atoms (in the
case of 2e, 2k, and 2m). Remarkably, both types of hydro-
gen-bonding interaction were present simultaneouslyin the
crystal of compound 2m for the two independent ionic
pairs. Furthermore, the bifurcate hydrogen bonds are mainly
characterized bythe asmymetrical arrangement of the
amino-H atom relative to the two oxygen atoms, except for
one of the two independent ionic pairs in compound 2m, in
which the amino-H atom is almost symmetrically located
relative to the two oxygen atoms (Table 12). We have denot-
ed this unusual ionic pair of compounds as 2m and 2m*, be-
cause it possesses a number of the additional interesting
properties discussed below. The asymmetrical arrangement
of the amino-H atom relative to the two oxygen atoms in
the bifurcate hydrogen bonds was also observed in the pre-
[4b,8c,9]
viouslystudied related compounds.
Moreover, the OH
group of one anion is bound up bythe strong hdyrogen
bond with the negativelycharged oxygen atom of another
anion (Table 12).
The conformations of the cations and anions are worth
mentioning. The cations of 2i, 2e, 2k, and 2m have a Z con-
figuration,[19] which is responsible for the structural features
of the ionic pair 2m*. It turned out that in the ionic pair
2m*, the anion lies over the =C(H)SR fragment, whereas
that in the other ionic pairs resides on the opposite side of
ꢁ
the cation, over the SR fragment (Figure 1). The arrange-
ment of the terminal substituents at the two sulfur atoms
relative to the C=C double bond in the cations is of particu-
lar interest. So, the substituent at the sulfur atom from the
more crowded side of the cation is located gauche to the
Figure 1. Molecular structures of oxalic salts of 2e, 2i (the two independ-
ent ionic pairs are shown; the alternative positions of the disordered
butyl groups with less occupancies are not shown), 2m, and 2k (the two
independent ionic pairs are shown; the alternative position of the disor-
dered cyclohexyl group with less occupancy is not shown) depicting ther-
mal ellipsoids at the 50% probabilitylevel; most hdyrogen atoms are
omitted for clarity; dash lines indicate the bifurcate hydrogen bonds.
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2425

I. P. Beletskaya, V. P. Ananikov et al.
Table 11. Selected bond lengths [Š] and angles [8] for 2i, 2e, 2k, and
2m.
Table 12. Hydrogen bonds for 2i, 2e, 2k, and 2m [lengths in Š and
angles in 8].[a]
ꢁ
ꢁ
compound 2i
D H···A
d
(D H)
d
U
d
(D···A)
aACHTREUNG(DHA)
ꢁ
S1 C2
1.767(3)
1.815(3)
1.735(3)
1.805(3)
1.341(4)
C2-S1-C4
C3-S2-C5
C3-C2-S1
C2-C3-S2
C4-S1-C2-C3
C5-S2-C3-C2
101.45(15)
99.27(15)
119.8(2)
compound 2i
ꢁ
S1 C4
ꢁ
N1 H1N···O1
0.92
0.92
0.97
1.93
2.26
1.50
2.764(3)
2.944(3)
2.468(2)
149
131
178
ꢁ
S2 C3
ꢁ
N1 H1N···O3
ꢁ
S2 C5
126.2(2)
i
ꢁ
O3 H3O···O2
ꢁ
C2 C3
ꢁ135.1(3)
compound 2e
169.7(3)
ii
ii
ꢁ
N1 H1N···O1
0.90
0.90
0.89
0.91
0.91
0.88
2.29
2.03
1.62
2.22
2.07
1.60
3.048(3)
2.790(3)
2.509(3)
3.010(3)
2.791(3)
2.485(3)
141
142
179
144
135
176
compound 2e
ꢁ
N1 H1N···O3
ꢁ
S1 C2
1.761(3)
1.809(5)
1.734(3)
1.811(4)
1.343(4)
1.764(3)
1.811(3)
1.734(3)
1.8217
C2-S1-C4
C3-S2-C8
C3-C2-S1
C2-C3-S2
103.15(18)
99.93(17)
119.2(2)
124.5(3)
103.11(14)
102.1(2)
93.7(3)
119.9(2)
125.0(2)
125.4(3)
ꢁ162.03
132.1(3)
ꢁ154.4(3)
168.6(5)
iii
ꢁ
O2 H2O···O3
ꢁ
S1 C4
iv
iv
v
ꢁ
N2 H2N···O5
ꢁ
S2 C3
ꢁ
N2 H2N···O7
ꢁ
S2 C8
ꢁ
O6 H6O···O7
ꢁ
C2 C3
C17-S3-C19
C18-S4-C23
C18-S4-C23’
C18-C17-S3
C17-C18-S4
C4-S1-C2-C3
C8-S2-C3-C2
C19-S3-C17-C18
C23-S4-C18-C17
C23’-S4-C18-C17
compound 2k
ꢁ
S3 C17
ꢁ
N1 H1N···O1
0.88
0.88
0.99
2.29
2.01
1.52
3.0020(11)
2.7619(11)
2.5104(9)
138
142
176
ꢁ
S3 C19
ꢁ
N1 H1N···O3
ꢁ
S4 C18
vi
ꢁ
O2 H2O···O3
ꢁ
S4 C23
compound 2m
ꢁ
S4 C23’
1.857(8)
1.334(4)
ꢁ
N1 H1N···O1
0.93
0.93
0.99
0.93
0.93
0.91
2.17
2.11
1.53
2.26
1.99
1.60
2.901(7)
2.879(8)
2.480(8)
3.004(7)
2.776(7)
2.486(7)
135
139
160
137
141
167
ꢁ
C17 C18
ꢁ
N1 H1N···O3
vii
ꢁ
O1 H1O···O4
ꢁ
N2 H2N···O5
ꢁ
N2 H2N···O7
compound 2k
viii
ꢁ
O6 H6O···O7
ꢁ
S1 C2
1.7665(9)
1.8329(10)
1.7390(10)
1.8326(11)
1.3429(13)
C2-S1-C4
C3-S2-C7
C3-C2-S1
C2-C3-S2
C4-S1-C2-C3
C7-S2-C3-C2
102.64(4)
99.20(5)
120.57(7)
125.90(8)
ꢁ136.15(8)
176.10(9)
[a] Symmetry transformations used to generate equivalent atoms: i: x,
yꢁ1, z; ii: xꢁ1, ꢁy+1, zꢁ1/2; iii: x, ꢁy+1, z+1/2; iv: xꢁ1, y, z; v: x,
ꢁy, z+1/2; vi: x, ꢁy+3/2, zꢁ1/2; vii: ꢁx+3/2, y, z+1/2; viii: ꢁx+3/2, y,
zꢁ1/2.
ꢁ
S1 C4
ꢁ
S2 C3
ꢁ
S2 C7
ꢁ
C2 C3
Compound 2m
ꢁ
S1 C2
ꢁ
S1 C4
1.766(6)
1.799(7)
1.743(7)
1.806(7)
1.354(8)
1.754(6)
1.794(7)
1.731(7)
1.823(7)
1.338(8)
C2-S1-C4
C3-S2-C10
C3-C2-S1
C2-C3-S2
C21-S3-C23
C22-S4-C29
C22-C21-S3
C21-C22-S4
C4-S1-C2-C3
C10-S2-C3-C2
C23-S3-C21-C22
C29-S4-C22-C21
103.3(3)
100.2(3)
118.0(5)
124.1(5)
102.4(3)
100.3(3)
119.7(6)
126.1(5)
ꢁ147.9(7)
174.4(7)
142.4(6)
172.9(6)
ꢁ
groups of the cations separate the anion chains from each
other within the sheets. The cation sheets are formed bysul-
fide parts of the cations and are located between the anion
sheets. However, the alternation of the anion and cation
layers in 2i occurs in pairs,[20] that is, two adjusted anion
layers are located between two adjusted cation layers,
whereas in 2e, 2k, 2m the structures are built up bythe al-
ternation of one anion layer and two cation layers.[19]
S2 C3
ꢁ
S2 C10
ꢁ
C2 C3
ꢁ
S3 C21
ꢁ
S3 C23
ꢁ
S4 C22
ꢁ
S4 C29
ꢁ
C21 C22
Apparently, both the unusual structure of the ionic pair
2m* and the different crystal packing are determined by the
ꢁ
spatial size of the terminal SR substituents at the C=C
double bond. Indeed, the pairwise alternation of the anion
C=C double bond, but the substituent at the sulfur atom
from the less crowded side of the cation is located trans to
this bond (Figure 1, Table 11).
and cation layers in 2i can be explained bythe small spatial
size of the SMe substituents; in the literature such struc-
ꢁ
It is known that the oxalic acid anion often has a twisted
geometryin salts including different hydrogen bonds. Simi-
lar conformation of the anions is also found in 2i (the twist
angle is 21.08), 2e (the twist angles are 19.6 and 23.08 for
the two independent anions), 2k (the twist angle is 23.38)
and 2m (the twist angle is 26.68), except for the unusual
ionic pair 2m*, in which the anion has a nearlyplanar con-
formation with the twist angle of only6.3 8. It should be
noted that the planar oxalic acid anion was also revealed
earlier in one of the cases,[9] whereas in the other related
structures[4b,8c] it has the expected twisted conformation.
The crystal packing of compounds 2i, 2e, 2k, and 2m de-
scribed in this paper (Figure 2) is built up bythe alternate
anion and cation layers. The anions bound in the chains by
tures have onlybeen observed in the absence of the second
-SR substituent at the C=C double bond.[20] On the other
ꢁ
hand, the presence of the two bulky SCysubstituents at
the C=C double bond in the structure of 2m gives rise to
the two independent ionic pairs of the different configura-
tions to form the crystal of high density.
Crystal structures of vinylchacogenides with carboxylic
group provide an important information concerning non-
bonded X···O interactions (X=S, Se). Such interactions are
of great importance for understanding of the mechanism of
biological action of chalcogen-containing species[21] as well
as for structural investigations.[22] The present compounds
possess unusual structure and bonding properties compared
to the molecules studied earlier and provide important infor-
mation for future studies of this phenomenon.
ꢁ
the O H···O hydrogen bonds form sheets, while the amino
2426
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
Scheme 4. Plausible mechanism of the catalytic reaction.
According to experimental studies, oxidative addition of
ꢁ
an E E bond to a low-valent-metal complex mayresult in a
diverse range of product depending on metal and ligand
type.[3,23] Oxidative addition of diaryldichalcogenides to Pd0
led to a mixture of trans and cis mono- and dinuclear com-
[4c]
plexes detected by 31P NMR spectroscopy. It was shown
that an excess of phosphane ligand is required to suppress
polymerization of Pd complexes and avoid formation of cat-
alytically inactive [{Pd(SR)2}n] species.[4c,3] Probably, a simi-
lar effect was observed in the studied Ni system, which also
required an excess of the ligand to achieve good product
yields. Unfortunately, for the Ni-based catalytic system we
were unable to obtain high-quality 31P NMR spectra due to
[24]
verybroad lines.
ꢁ
Theoretical studies of the oxidative addition of AlkS
0
ꢁ
SAlk and AlkSe SeAlk to Pd have been performed earli-
er;[25] however, no data is available concerning the reactivity
of the Ni complexes. To reveal relative reactivityof Pd and
ꢁ
ꢁ
Ni complexes in E E and C E bond oxidative addition, we
have carried out a series of theoretical calculations on the
model reactions (Figure 3).
Starting from the initial point ML2 + R2E2 (I) the first
step is coordination of the chalcogen atom to the metal
center. For M=Pd this step is exothermic by DH=ꢁ2.3 and
[26]
ꢁ3.6 kcalmolꢁ1 for E=S and Se, respectively(Table 13).
For M =Ni much stronger binding of R2E2 was calculated
DH=ꢁ13.1 and ꢁ13.5 kcalmolꢁ1 for E=S and Se, respec-
ꢁ
tively. Optimized Pd E1 bond lengths in II are 2.433 and
ꢁ
2.519 Š; optimized Ni E1 bond lengths in II are 2.143 and
2.262 Š for E=S and Se, respectively(Table 14). It is impor-
tant to note that stronger binding in the case of Ni caused
ꢁ
considerable elongation of the E1 E2 bond in II: 2.174 and
2.442 Š, compared to 2.128 and 2.398 Š for Pd (E=S and
Se, respectively). Thus, R2E2 coordination to the NiL2 com-
plex results in significant elongation (and most likelyweak-
Figure 2. Crystal packing of oxalic salts of 2e, 2i, 2m, and 2k.
ꢁ
Discussion
ening) of the E E bond compared to PdL2.
Starting from the complex II the reaction mayproceed
ꢁ
A plausible catalytic reaction mechanism involves the fol-
either byan E E oxidative addition pathway II!III-TS!
ꢁ
ꢁ
lowing steps (Scheme 4): 1) oxidative addition of E E bond
IV or a C E oxidative addition pathway II!V-TS !VI
to M0, 2) alkyne coordination to the metal center, 3) alkyne
(Figure 3). Let us consider the former pathwayfirst.
ꢁ
ꢁ
insertion into the M E bond, and 4) C E reductive elimina-
In the case of Pd symmetric three-centered transition
state was located, movement along the reaction coordinate
tion leading to the product 2 and regeneration of M0.
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2427

I. P. Beletskaya, V. P. Ananikov et al.
Figure 3. B3LYP/SDD 6–311G(d) optimized molecular structures of I–VI with atom numbering and converged symmetry point groups (in bold).
ꢁ1
ꢁ
ꢁ
(Figure 3). The calculated acti-
Table 13. Calculated DE, DH, and DG energysurfaces [in kcalmol ] of the E E and E C oxidative addition
reaction (E=S, Se) to M
ACHTREUNG
vation barriers for M=Pd are
DH° =28.1
and
23.3 kcal
M=Pd
molꢁ1; for the same step in-
volving Ni complexes the cal-
culated activation barriers are
DH° =15.6 and 11.9 kcalmolꢁ1
(for E=S and Se, respective-
ly). According to the length of
E=S
E=Se
E=S
E=Se
I
0.0/0.0/0.0
0.0/0.0/0.0
0.0/0.0/0.0
0.0/0.0/0.0
II
III-TS
IV
V-TS
VI
ꢁ14.2/ꢁ13.1/ꢁ2.4
ꢁ14.6/ꢁ13.5/ꢁ2.6
ꢁ3.1/ꢁ2.3/6.5
ꢁ4.5/ꢁ3.6/5.6
–/–/–
–/–/–
9.9/9.9/20.8
3.6/3.8/15.5
ꢁ34.3/ꢁ32.4/ꢁ19.2
3.3/2.5/14.8
ꢁ16.5/ꢁ15.7/ꢁ3.2
ꢁ29.1/ꢁ27.1/ꢁ13.8
ꢁ1.1/ꢁ1.6/11.1
ꢁ18.1/ꢁ17.0/ꢁ4.2
ꢁ14.5/ꢁ12.9/ꢁ0.1
26.7/25.8/36.4
0.7/1.3/13.4
ꢁ12.6/ꢁ10.9/2.2
20.3/19.7/30.5
ꢁ2.3/ꢁ1.4/11.0
ꢁ
M E bond in V-TS and III-TS,
[a] Calculated data on energy, enthalpy, and free energy is given in the orderE/H/G for each case.
the former transition state pos-
sesses more late character (see
ꢁ
involves elongation and breakage of the E E bond and for-
Table 14) in agreement with higher activation barriers.
From a kinetic point of view Ni complexes should be
ꢁ
mation of two new M E bonds (Figure 3). The activation
barriers for the reaction are DH° =12.2 and 7.4 kcalmolꢁ1
for E=S and Se, respectively(Table 13). The oxidative addi-
tion step is exothermic by DH=ꢁ6.6 and ꢁ3.4 kcalmolꢁ1
more reactive with respect to Pd in C E bond oxidative ad-
ꢁ
dition leading to VI. However, according to thermodynamic
properties the energydifference between the complexes IV
and VI is more favorable towards formation IV in the case
ꢁ
mol for E=S and Se, respectively. A rather long M E bond
length in III-TS (Table 14) confirms the earlier character of
the transition state in agreement with energetic data
(Table 13). On the DG surface the activation barriers are
DG° =20.8 and 15.5 kcalmolꢁ1[27] and the oxidative addition
step was nearlythermo neutral or endoergic, DG=ꢁ0.1 and
2.2 kcalmolꢁ1 for E=S and Se, respectively.
of Ni (on DH surface IV is more stable than VI by16.7 and
ꢁ1
10.1 kcalmolꢁ1 for M=Ni and by14.2 and 9.5 kcalmol
M=Pd; see Table 13).
for
The following conclusions concerning the relative reactivi-
tyof the studied transition metal complexes can be empha-
sized:
ꢁ
In the case of Ni complex oxidative addition of the E E
bond took place as a barrierless reaction[28] and was signifi-
1) NiL2 complexes are more reactive in E E bond oxida-
ꢁ
cantlymore exothermic DH=ꢁ16.8 and ꢁ11.2 kcalmolꢁ1
tive addition than PdL2.
ꢁ
for E=S and Se, respectively. On the DG surface the reac-
2) For both metals E E bond oxidative addition is energeti-
callymore favorable than C E bond oxidative addition.
3) The activation of C E bond bytransition metal com-
ꢁ1
ꢁ
tion is favorable in energyby19.2 and 13.8 kcalmol
pared to I and by16.8 and 11.2 kcalmol
com-
compared to II
ꢁ1
ꢁ
for E=S and Se, respectively.
plexes should be more easyfor E =Se compared to E=
The second pathway—C E oxidative addition—proceeds
via the V-TS transition state, starting from complex II
S.[29]
ꢁ
2428
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
Table 14. Optimized geometryparameters [bond lengths in Š, angles in 8] of structures I–VI corresponding to
known in the framework of
catalytic heterofunctionalyza-
tion reactions and correspond-
ing discussion will not be re-
peated here.[2]
ꢁ
ꢁ
the E E and E C oxidative addition reaction (E=S, Se) to M
ACHTREU(NG PH3)2 calculated at the B3LYP/SDD 6–
311G(d) level for M=Pd and Ni (see Figure 3 for atom numbering).[a]
ꢁ
M P1
ꢁ
M P2
ꢁ
M E1
ꢁ
M C1
ꢁ
E1 E2
ꢁ
E1 C1
ꢁ
E2 C2
M
E
P1-M-P2
I
Ni
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
S
Se
2.115
2.115
2.289
2.115
2.136
2.139
2.305
2.311
2.387
2.388
2.218
2.214
2.337
2.340
2.204
2.202
2.376
2.383
2.181
2.182
2.317
2.323
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
2.094
2.355
2.094
2.355
2.174
2.442
2.128
2.398
2.454
2.669
–
1.832
1.982
1.832
1.982
1.841
1.989
1.837
1.985
1.834
1.984
1.837
1.983
1.837
1.982
2.215
2.328
2.354
2.474
–
–
–
–
–
180.0
180.0
180.0
180.0
117.6
117.0
121.1
120.6
112.0
112.0
100.6
99.6
104.0
103.3
118.2
117.5
118.4
119.0
97.8
On the basis of our studywe
can suggest a plausible mecha-
nism of the catalytic reaction.
However, at the moment we
cannot clearlysaywhich step
of the catalytic cycle is respon-
sible for the observed ligand
effect. In addition to intrinsic
stabilization (or destabiliza-
tion) of transition-metal inter-
mediates bythe ligand, the
second important factor is
ligand dissociation (or associa-
tion) in each step of the cata-
lytic cycles, which significantly
complicates the overall picture.
Further studies are required to
reveal mechanistic details of
the developed catalytic system.
An important question is the
nature of the side reactions
Pd
Ni
Pd
Pd
Ni
Pd
Ni
Pd
Ni
Pd
II
2.155
2.162
2.330
2.336
–
–
–
–
–
2.143
2.262
2.433
2.519
2.419
2.518
2.246
2.367
2.388
2.504
2.102
2.224
2.312
2.412
2.227
2.348
2.379
2.494
1.831
1.980
1.832
1.981
–
–
–
–
–
III-TS
IV
–
–
–
–
–
V-TS
VI
2.216
2.219
2.410
2.413
2.271
2.260
2.415
2.406
2.069
2.067
2.290
2.283
1.962
1.966
2.093
2.099
2.134
2.399
2.106
2.373
2.099
2.365
2.090
2.357
1.832
1.981
1.832
1.982
1.830
1.980
1.830
1.981
–
–
–
97.1
99.6
99.6
[a] Imaginaryfrequencies for the transition states: III-TS (Pd, S)=171i cmꢁ1, III-TS (Pd, Se)=77i cmꢁ1, V-TS
(Ni, S)=299i cmꢁ1, V-TS (Ni, Se)=228i cmꢁ1, V-TS (Pd, S)=281i cmꢁ1, V-TS (Pd, Se)=226i cmꢁ1
.
The results of the theoretical studyare in excellent agree-
ment with experimental findings and explain several impor-
tant points. Particularly, relative reactivity of Pd and Ni
that take place in the studied system. To find an answer we
have carried out the reactions of pure product 2a under dif-
ferent conditions. These reactions model possible transfor-
mations of the product under catalytic conditions.
No reaction took place between the 2a and Bu2S2 in the
Ni/PMe2Ph system after 6 h at 1008C (Scheme 5), but if 2a
was heated in the presence of Ni/PMe2Ph (without Bu2S2)
broad signals in 1H NMR spectrum appear at d=6.0 and
6.5 ppm, suggesting that polymerization of 2a took place.
Therefore, under catalytic conditions at the end of reaction
(when Bu2S2 is consumed) the product can be converted to
a polymer.
ꢁ
complexes in E E oxidative reaction is in agreement with
observed reactivityof these complexes in the catalytic reac-
tion. Smaller activation barriers and more stable products of
ꢁ
ꢁ
the E E bond oxidative addition compared to C E bond
explains high selectivityof transition-metal-catalyzed trans-
formation towards the R2E2 substrate (see Scheme 2 and
corresponding discussion). Relative reactivityof sulfur and
ꢁ
selenium compounds in C E bond activation provides nec-
essarybasis for understanding whythe side reaction leading
to R2E was observed onlyin the case of E =Se (see
Scheme 3 and corresponding discussion). According to the
Heating of 1a in Ni/PMe2Ph system resulted in the forma-
tion of 1,3,5-tributylbenzene (1H dAr =7.2 ppm) and oligo-
meric species (several signals in the range d=5.5–4.5 ppm).
ꢁ
ꢁ
theoretical study, the oxidative addition of S S and Se Se
bonds to PdL2 and NiL2 complexes is characterized byrela-
tivelysmall activation barriers, therefore, it maynot be con-
sidered as a rate-determining step.
The next step after completing the oxidative addition is
ꢁ
alkyne insertion into the M E bond (Scheme 4). Byexperi-
mental[30] and theoretical studies[31] it was shown that alkyne
ꢁ
insertion into the M E bond takes place with high stereose-
lectivityand is responsible for overall stereoselectivityof
the catalytic cycle (providing that Z/E isomerization of the
product double bond and noncatalytic addition reactions are
avoided). In this case, only Z configuration of the double
bond should be expected as a result of syn-addition. High
stereoselectivityin the studied ssytem ( Z:E >99:1) con-
firmed byNMR spectroscopyand X-raystudies is in agree-
ment with proposed reaction mechanism. The last step of
the catalytic cycle, reductive elimination (Scheme 4), is
Scheme 5.
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2429

I. P. Beletskaya, V. P. Ananikov et al.
1H, 31P, 13C, and 77Se nuclei, respectively. The spectra were processed on a
PC Linux workstation using XWINMR software package. All 2D spectra
were recorded byusing an inverse triple resonance probe head with
active shielded Z-gradient coil. 1H and 13C chemical shifts are reported
relative to the corresponding solvent signals used as internal reference,
external Ph2Se2/CDCl3 (d=463.0 ppm) and H3PO4/H2O was used for 77Se
and 31P, respectively. 2D 1H-77Se HMQC and NOESY NMR experiments
Triple-bond trimerization and oligomerization catalyzed by
Ni complexes are well-known reactions,[32] which also took
place under the studied reaction conditions.
If Pd catalytic system was utilized instead of Ni neither
decomposition of 2a nor oligomerization of the alkyne were
observed.
These results are in excellent agreement with the findings
concerning catalytic system described earlier. Thus, it
became clear whyan excess of alkyne might be required to
complete conversion of R2E2 in Ni-catalyzed transformation
(but not for Pd case) and the reason for a decrease of the
product yield if reaction time is overestimated.
[4a,34]
were carried out as described previously.
General procedure for the Ni-catalyzed synthesis of 2a–2p
Method A: [NiACHTREUNG
(acac)2] (3.0”10ꢁ5 mol, 7.7 mg), R2E2 (1.0”10ꢁ3 mol) and
PMe2Ph (3.0”10ꢁ4 mol, 41.5 mg) were placed in a reaction vessel and
stirred at room temperature until a homogeneous brown solution was
formed (ca. 5–10 min). The alkyne (1.0”10ꢁ3 mol) was added to the solu-
tion and the reaction was carried out at 1008C under stirring until com-
plete conversion of the R2E2 (2 h for most substrates; see Tables 5 and
6).[35]
Method B: The same as Method A except 3.0”10ꢁ3 mol of the alkyne
Conclusions
was added (1 h for most substrates; see Tables 5 and 6).[35]
Method C: The same as Method A except that reaction was carried out
in benzene (0.5 mL), which was placed into the reaction vessel first, fol-
We have developed two novel catalytic systems based on Pd
and Ni complexes, which solved the problem of stereoselec-
tive addition of dialkyldichalcogenides to alkynes (60–90%
yields and Z:E >99:1 selectivity). The choice of the ligand
was the keyfinding to carryout the reaction: PMe 2Ph was
the optimal ligand for the Ni catalytic system and PCy2Ph
was the optimal ligand for the Pd catalytic system. In both
Ni and Pd catalytic systems an excess of the ligand was re-
quired to achieve high product yields.
lowed bycompounds in the order: R 2E2, PMe2Ph, [Ni
ACHTRE(UNG acac)2], alkyne
(see Table 5).[35]
General procedure for the Pd-catalyzed synthesis of 2a, 2e, 2d: [Pd2-
(dba)3] (1.5”10ꢁ5 mol, 15.5 mg), R2E2 (1.0”10ꢁ3 mol) and PCy2Ph (3.0”
AHCTREUNG
10ꢁ4 mol, 82.3 mg) were placed in a reaction vessel and stirred at room
temperature until a homogeneous dark-brown solution was formed (ca.
5–10 min). The alkyne (1.0”10ꢁ3 mol) was added to the solution and the
stirring was continued for additional 10 min. The reaction was carried out
at 1408C under stirring until complete conversion of the R2E2 (12–18 h,
see Table 10).
Ni catalyst has several important advantages: 1) superior
catalytic activity, 2) high stereoselectivity, 3) low cost, 4) rel-
ativelymild reaction conditions (100 8C, 2 h), and 5) good
yields for both E=S and Se. The disadvantages of the Ni
system include side reactions involving alkyne (for E=S,
but not for E=Se) and decomposition of the product at the
end of reaction. Careful optimizations of reaction conditions
minimized the influence of these drawbacks on the efficien-
cyof the synthetic procedure.
The Pd catalyst was significantly less active and required
higher temperature and longer time (1408C, 12 h). As a
result, Pd/PCy2Ph catalytic system has a narrow scope and
was applicable onlyfor the addition of dialkyldisulfides to
alkynes. The low catalytic activity of Pd required more
harsh reaction conditions leading to decomposition of dia-
lkyldiselenide. However, an important advantage of the Pd
system is an absence of the side reactions involving the
alkyne (for both E=S and Se).
Isolation and characterization: After completion of the reaction, the
products were purified bydrycolumn flash chromatographyon silica.
[16]
For the compounds 2e, 2i, 2k, 2m, 2o, and 2p hexane/ethylacetate gradi-
ent elution was applied (prior to chromatographythe silica was washed
bya solution of 5–6 drops of Et 3N in 20 mL of hexane). For the other
compounds hexane/toluene gradient elution was used. After drying in
vacuum the pure products were obtained. The isolated yields given below
were calculated based on initial amount of the R2E2.
AHCTRE(GNU 1Z)-1,2-Bis(butylsulfanyl)-1-hexene (2a): Yellow oil; method A: 75%
(0.200 g), method B: 80% (0.216 g); 1H NMR (500 MHz, CDCl3): d=
0.91 (m, 9H), 1.32 (m, 2H), 1.40–1.65 (m, 10H), 2.25 (t, J=7.6 Hz, 2H),
2.68 (t, J=7.5 Hz, 4H), 6.05 ppm (s, 1H); 13C{1H} NMR (126 MHz,
CDCl3): d=13.56, 13.59, 13.81, 21.66, 21.83, 21.97, 30.68. 30.97, 31.91,
32.35, 33.71, 36.55, 126.74, 131.98 ppm; MS (EI): m/z (%): 260 (18) [M]+;
elemental analysis calcd (%) for C14H28S2: C 64.55, H 10.83, S 24.62;
found: C 64.48, H 10.94, S 24.41.
[(Z)-1,2-Bis(butylsulfanyl)ethenyl]benzene (2b): Light yellow oil;
method A: 60% (0.164 g), method B: 81% (0.240 g). Identified accord-
ing to the published data.[5]
AHCTRE(GNU 1Z)-1,2-Bis(butylsulfanyl)-3-methoxy-1-propene (2c): Light yellow oil;
method A: 55% (0.133 g), method B: 75% (0.198); 1H NMR (500 MHz,
CDCl3): d=0.84 (m, 6H), 1.36 (m, 4H), 1.49 (quintet, J=7.2, 7.7 Hz,
2H), 1.56 (quintet, J=7.2, 7.7 Hz, 2H), 2.66 (t, J=7.4 Hz, 2H), 2.70 (t,
J=7.4 Hz, 2H), 3.25 (s, 3H), 3.94 (s, 2H), 6.36 ppm (s, 1H); 13C{1H}
NMR (126 MHz, CDCl3): d=13.36, 13.40, 21.42, 21.62, 31.03, 31.85,
32.27, 33.50, 57.18, 75.18, 126.41, 131.69 ppm; MS (EI): m/z (%): 248 (39)
[M]+; elemental analysis calcd (%) for C12H24OS2: C 58.01, H 9.74, S
25.81; found: C 58.25, H 9.91, S 25.49.
Experimental Section
General: Unless otherwise noted, the synthetic work was carried out
under argon atmosphere. [Pd2ACHTRE(UGN dba)3] (dba=dibenzylideneacetone) was
prepared according to a published procedure.[33] Bu2Se2 was prepared ac-
cording to a modified literature procedure (see below). Other reagents
were obtained from Acros and Lancaster and used as supplied (checked
[(Z)-1,2-Bis(butylsulfanyl)ethenyl](trimethyl)silane (2d): Colorless oil;
method A: 48% (0.120 g), method B: 71% (0.195 g). Identified accord-
ing to the published data.[5]
byNMR spectroscopybefore use). [Ni ACHTRE(UNG acac)2] was dried under vacuum
N-[(2Z)-2,3-Bis(butylsulfanyl)-2-propenyl]-N,N-dimethylamine
(2e):
(0.1–0.05 Torr, 608C, 30 min) before use. Solvents were purified accord-
ing to published methods. The reaction was carried out in PTFE screw
capped tubes or flasks.
Light yellow oil; method A: 75% (0.183 g), method B: 80% (0.208 g);
1H NMR (500 MHz, CDCl3): d=0.88 (m, 6H), 1.39 (m, 4H), 1.53 (quin-
tet, J=7.4 Hz, 2H), 1.59 (quintet, J=7.4 Hz, 2H), 2.18 (s, 6H), 2.69 (t,
J=7.4 Hz, 2H), 2.76 (t, J=7.4 Hz, 2H), 2.97 (s, 2H), 6.26 ppm (s, 1H);
13C{1H} NMR (126 MHz, CDCl3): d=13.60, 13.65, 21.66, 21.93, 30.96,
All NMR measurements were performed using a three channel Bruker
DRX-500 spectrometer operating at 500.1, 202.5, 125.8 and 95.4 MHz for
2430
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
32.09, 32.49, 33.77, 44.99, 65.24, 128.29, 130.32 ppm; MS (EI): m/z (%):
261 (5) [M]+; elemental analysis calcd (%) for C13H27NS2: C 59.71, H
10.41, N 5.36, S 24.52; found: C 59.60, H 10.53, N 5.57, S24.33.
45.58, 66.19, 126.73, 131.47 ppm; MS (EI): m/z (%): 313 (11) [M]+; ele-
mental analysis calcd (%) for C17H31NS2: C 65.12, H 9.96, N 4.47, S
20.45; found: C 64.89, H 10.00, N 4.59, S 20.24.
A
ACHTRE(UNG 1Z)-1,2-Bis(butylselanyl)-1-hexene (2n): Light yellow oil; 87%
1
method C: 58% (0.151 g); H NMR (500 MHz, CDCl3): d=0.93 (m, 6H),
1.41 (s, 6H), 1.41–1.46 (m, 4H), 1.59 (quintet, J=7.1, 7.5 Hz, 2H), 1.64
(quintet, J=7.1, 7.5 Hz, 2H), 2.74 (t, J=7.3 Hz, 2H), 2.79 (t, J=7.3 Hz,
2H), 6.73 ppm (s, 1H); 13C{1H} NMR (126 MHz, CDCl3): d=13.40, 13.47,
21.49, 21.80, 29.05, 31.68, 32.35, 33.36, 33.61, 74.27, 133.50, 136.73 ppm;
MS (EI): m/z (%): 262 (48) [M]+; elemental analysis calcd (%) for
C13H26OS2: C 59.49, H 9.98, S 24.43; found: C 59.39, H 10.08, S 24.43.
(0.300 g); 1H NMR (500 MHz, CDCl3): d=0.92 (m, 9H), 1.31 (m, 2H),
1.42 (m, 4H), 1.51 (m, 2H), 1.62–1.74 (m, 4H), 2.31 (t, J=7.1 Hz, 2H),
2.71 (t, J=7.3 Hz, 2H), 2.75 (t, J=7.4 Hz, 2H), 6.55 ppm (s, 1H); 13C{1H}
NMR (126 MHz, CDCl3): d=13.53, 13.56, 13.86, 21.95, 22.84, 22.97,
25.20, 26.44, 31.13, 32.64, 33.00, 39.67, 124.41, 133.71 ppm; MS (EI): m/z
(%): 356 (7) [M]+; elemental analysis calcd (%) for C14H28Se2: C 47.46,
H 7.97, Se 44.57; found: C 47.38, H 7.99, Se 44.64.
1-[(Z)-1,2-Bis(butylsulfanyl)ethenyl]cyclohexanol (2g): Light yellow oil;
method B: 61% (0.160 g); H NMR (500 MHz, CDCl3): d=0.93 (m, 6H),
N-[(2Z)-2,3-Bis(butylselanyl)-2-propenyl]-N,N-dimethylamine
(2o):
1
Light yellow oil; 88% (0.313 g); 1H NMR (500 MHz, CDCl3): d=0.81
(m, 6H), 1.33 (m, 4H), 1.52–1.64 (m, 4H), 2.12 (s, 6H), 2.63 (t, J=
7.3 Hz, 2H), 2.74 (t, J=7.3 Hz, 2H), 2.94 (s, 2H), 6.73 ppm (s, 1H);
13C{1H} NMR (126 MHz, CDCl3): d=13.24, 13.27, 22.46, 22.66, 24.81,
26.15, 32.44, 32.76, 44.68, 67.62, 127.84, 129.48 ppm; MS (EI): m/z (%):
357 (16) [M]+; elemental analysis calcd (%) for C13H27NSe2: C 43.95, H
7.66, N 3.94, Se 44.45; found: C 44.09, H 7.51, N 4.20, Se 44.50.
1.20 (m, 1H), 1.43 (m, 4H), 1.53–1.74 (m, 13H), 2.07 (s, 1H), 2.74 (t, J=
7.5 Hz, 2H), 2.78 (t, J=7.5 Hz, 2H), 6.72 ppm (s, 1H); 13C{1H} NMR
(126 MHz, CDCl3): d=13.53, 13.61, 21.65, 21.99, 22.02, 25.43, 31.89,
32.52, 33.56, 33.95, 36.48, 75.00, 134.19, 137.51 ppm; MS (EI): m/z (%):
284 (18) [MꢁH2O]+; elemental analysis calcd (%) for C16H30OS2: C
63.52, H 9.99, S 21.20; found: C 63.80, H 9.99, S 21.03.
ACHTREUNG(1Z)-1,2-Bis(methylsulfanyl)-1-hexene (2h): Light yellow oil; method A:
N-[(2Z)-2,3-Bis(methylselanyl)-2-propenyl]-N,N-dimethylamine
(2p):
55% (0.100 g), method B: 60% (0.110 g); 1H NMR (500 MHz, CDCl3):
d=0.92 (t, J=7.2 Hz, 3H), 1.33 (m, 2H), 1.50 (quintet, J=7.4, 7.9 Hz,
2H), 2.25 (s, 3H), 2.28 (t, J=7.2 Hz, 2H), 2.30 (s, 3H), 5.95 ppm (s, 1H);
13C{1H} NMR (126 MHz, CDCl3): d=13.83, 14.51, 17.26, 21.98, 30.64,
35.73, 125.71, 133.25 ppm; elemental analysis calcd (%) for C8H16S2: C
54.49, H 9.15, S 36.37; found: C 54.56, H 9.48, S 36.45; MS (EI): m/z
(%): 176 (26) [M]+.
Light yellow oil; 59% (0.159 g); 1H NMR (500 MHz, CDCl3): d=2.10 (s,
3H), 2.14 (s, 9H), 2.96 (s, 2H), 6.66 ppm (s, 1H); 13C{1H} NMR
(126 MHz, CDCl3): d=5.02, 6.54, 44.68, 66.66, 126.78, 130.28 ppm; MS
(EI): m/z (%): 273 (11) [M]+; elemental analysis calcd (%) for
C7H15NSe2: C 31.01, H 5.58, N 5.17, Se 58.25; found: C 30.83, H 5.65, N
5.28, Se 58.11.
(acac)2] (3.0”10ꢁ4 mol, 77.0 mg), Bu2S2
Scaling synthetic procedure: [NiACHTREUNG
(1.0”10ꢁ2 mol, 1.784 g) and PMe2Ph (3.0”10ꢁ3 mol, 0.415 g) were placed
in a reaction vessel and stirred at room temperature until a homogeneous
brown solution was formed (ca. 5–10 min). The alkyne 1a (1.0”10ꢁ2 mol,
0.821 g) was added to the solution and the stirring was continued for ad-
ditional 10 min. The reaction was carried out for 2 h at 1008C under stir-
ring. Isolated yield after purification on silica: 78% (2.120 g).
N-[(2Z)-2,3-Bis(methylsulfanyl)-2-propenyl]-N,N-dimethylamine
(2i):
Light yellow oil; method A: 76% (0.140 g), method B: 81% (0.148 g);
1H NMR (500 MHz, CDCl3): d=2.22 (s, 6H), 2.33 (s, 3H), 2.34 (s, 3H),
3.02 (s, 2H), 6.17 ppm (s, 1H); 13C{1H} NMR (126 MHz, CDCl3): d=
13.96, 16.92, 44.57, 64.15, 128.63, 129.28 ppm; MS (EI): m/z (%): 177 (13)
[M]+; elemental analysis calcd (%) for C7H15NS2: C 47.41, H 8.53, N
7.90, S 36.16; found: C 47.54, H 8.37, N 7.81, S 35.93.
Preparation of Bu2Se2:[36] The synthesis was carried out at 208C under
argon atmosphere under rigorous stirring. Se powder (4.0 g, 0.05 mol;
200 mesh) was placed into two-necked flask followed byaddition of
EtOH (80%, 150 mL). NaBH4 (3.8 g, 0.1 mol) was slowlyadded to the
solution (Caution: gas evolution at this stage) until complete dissolution
of Se and formation of colorless suspension with white-graysolid (a
slight excess of NaBH4 might be required). DMF (100 mL) was added to
the solution and it was stirred until the color was turned to red-brown
(ca. 10 min). EtOH (80%, 50 mL) was added to the solution and stirring
was continued until termination of gas evolution (traces of unreacted hy-
drides should be removed). Se powder (4.0 g, 0.05 mol; 200 mesh) was
added to the solution and stirred until complete dissolution and forma-
tion of clear dark-red solution. BuBr (13.7 g, 0.1 mol) was slowlyadded
to the solution during 30 min, changing its color to yellow. The reaction
was quenched byaddition of water (300 mL) and extracted byhexanes
(3”200 mL). The combined organic phase was washed with water (3”
500 mL) and dried over Na2SO4. The solvent was removed under reduced
pressure and pure product was obtained after distillation (55–578C,
0.06 Torr), yellow-orange oil, yield 12.33 g (90%).
ACHTREUNG(1Z)-1,2-Bis(isopropylsulfanyl)-1-hexene (2j): Light yellow oil; method
A: 60% (0.143 g), method B: 62% (0.146 g); 1H NMR (500 MHz,
CDCl3): d=0.92 (t, J=7.0 Hz, 3H), 1.26 (d, J=6.7 Hz, 6H), 1.31 (d, J=
6.7 Hz, 6H), 1.32 (m, 2H), 1.52 (quintet, J=7.2, 7.6 Hz, 2H), 2.27 (t, J=
7.3 Hz, 2H), 3.09 (m, 1H), 3.31 (m, 1H), 6.22 ppm (s, 1H); 13C{1H} NMR
(126 MHz, CDCl3): d=13.75, 21.94, 23.23, 23.54, 30.65, 35.09, 37.26,
37.66, 127.52, 132.11 ppm; MS (EI): m/z (%): 232 (67) [M]+; elemental
analysis calcd (%) for C12H24S2: C 62.00, H 10.41, S 27.59; found: C
62.11, H 10.57, S 27.54.
N-[(2Z)-2,3-Bis(isopropylsulfanyl)-2-propenyl]-N,N-dimethylamine (2k):
Light yellow oil; method A: 64% (0.154 g), method B: 84% (0.205 g);
1H NMR (500 MHz, CDCl3): d=1.26 (d, J=6.5 Hz, 6H), 1.33 (d, J=
6.5 Hz, 6H), 2.23 (s, 6H), 3.02 (s, 2H), 3.14 (m, 1H), 3.45 (m, 1H),
6.51 ppm (s, 1H); 13C{1H} NMR (126 MHz, CDCl3): d=23.33, 23.67,
35.33, 37.36, 45.07, 66.24, 127.90, 131.53 ppm; MS (EI): m/z (%): 233 (48)
[M]+; elemental analysis calcd (%) for C11H23NS2: C 56.60, H 9.93, N
6.00, S 27.47; found: C 56.27, H 9.79, N 6.17, S 27.23.
{[(Z)-1-Butyl-2-(cyclohexylsulfanyl)ethenyl]sulfanyl}cyclohexane
(2l):
Theoretical calculations: Geometryand energyof the reactants, inter-
mediates, transition states, and products of the reactions were calculated
using the B3LYP hybrid density functional method[37] in conjunction with
the standard 6–311G(d) basis set[38] for H, C, P, S, and Se and the triple-z
basis set with the Stuttgart/Dresden effective core potentials[39] (SDD)
for Pd and Ni (denoted as B3LYP/SDD 6–311G(d) level). In previous
studies it was established that this level of theoryreasonablydescribes
the energyand geometryparameters of the systems involving transition
metal complexes.[40]
Yellow oil; method A: 60% (0.192 g), method B: 81% (0.269 g);
1H NMR (500 MHz, CDCl3): d=0.92 (t, J=7.4 Hz, 3H), 1.20–1.44 (m,
12H), 1.51 (quintet, J=7.4 Hz, 2H), 1.61 (m, 2H), 1.77 (m, 4H), 1.92 (m,
2H), 1.99 (m, 2H), 2.25 (t, J=7.7 Hz, 2H), 2.82 (m, 1H), 3.02 (m, 1H),
6.22 ppm (s, 1H); 13C{1H} NMR (126 MHz, CDCl3): d=13.84, 22.01,
25.59, 25.73, 25.98, 26.00, 30.74, 33.56, 33.77, 37.73, 43.50, 45.76, 127.50,
131.12 ppm; MS (EI): m/z (%): 312 (37) [M]+; elemental analysis calcd
(%) for C18H32S2: C 69.16, H 10.32, S 20.52; found: C 69.07, H 10.58, S
20.38.
For all studied structures, normal coordinate analysis was performed to
characterize the nature of the stationarypoints and to calculate thermo-
dynamic properties (298.15 K and 1 atm). Transition states were con-
firmed with IRC (intrinsic reaction coordinate) calculations byusing the
standard method.[41] All calculations were performed without anysymme-
N-[(2Z)-2,3-Bis(cyclohexylsulfanyl)-2-propenyl]-N,N-dimethylamine
(2m): Light yellow oil; method A: 65% (0.211 g), method B: 87%
(0.288 g); 1H NMR (500 MHz, CDCl3): d=1.13–1.41 (m, 10H), 1.54 (m,
2H), 1.70 (m, 4H), 1.86 (m, 2H), 1.93 (m, 2H), 2.15 (s, 6H), 2.80 (m,
1H), 2.93 (s, 2H), 3.10 (m, 1H), 6.43 ppm (s, 1H); 13C{1H} NMR
(126 MHz, CDCl3): d=25.38, 25.58, 25.78, 33.41, 33.67, 43.44, 44.93,
[42]
tryconstraints with the Gaussian 03 program.
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2431

I. P. Beletskaya, V. P. Ananikov et al.
Table 15. Crystallographic data for 2i, 2e, 2k, and 2m.
2i·HOOC-COOH
2e·HOOC-COOH
2k·HOOC-COOH
2m·HOOC-COOH
formula
Mr
C9H17NO4S2
267.36
C15H29NO4S2
351.51
C13H25NO4S2
323.46
C19H33NO4S2
403.58
T [K]
100
100
100
100
cryst size [mm]
crystal system
space group
a [Š]
b [Š]
c [Š]
0.50”0.05”0.05
monoclinic
C2/c
28.823(6)
5.5553(13)
16.177(4)
90
0.30”0.30”0.06
monoclinic
Pc
8.3017(7)
21.9390(18)
11.0390(9)
90
0.30”0.30”0.20
monoclinic
P21/c
8.2680(5)
20.6710(12)
11.0938(7)
90
0.40”0.05”0.05
orthorhombic
Pca21
25.195(5)
15.550(3)
11.045(2)
90
a [8]
b [8]
92.681(6)
90
2587.5(10)
8
107.725(5)
90
1915.1(3)
4
110.925(5)
90
1770.97(19)
4
90
90
g [8]
V [Š3]
Z
4327.2(15)
8
1.239
1calcd [gcmꢁ3
]
1.373
1.219
1.213
F
A
1136
760
696
1744
]
0.410
0.293
0.311
0.269
q range [8]
index range
1.41 to 27.99
ꢁ37ꢃhꢃ37
ꢁ7ꢃkꢃ7
ꢁ21ꢃlꢃ21
13686
1.86 to 30.04
ꢁ11ꢃhꢃ11
ꢁ30ꢃkꢃ30
ꢁ15ꢃlꢃ15
23430
1.97 to 30.04
ꢁ11ꢃhꢃ11
ꢁ29ꢃkꢃ29
ꢁ15ꢃlꢃ15
22765
1.62 to 26.14
ꢁ31ꢃhꢃ31
ꢁ19ꢃkꢃ18
ꢁ13ꢃlꢃ13
32015
reflns collected
unique rfelns
3092 (Rint =0.0606)
1492
0.0569/0.0561
0.1673/0.0723
3092/0/149
0.932
0.420/ꢁ0.443
0.979/0.821
–
10962 (Rint =0.0318)
8624
0.0619/0.1432
0.0828/0.1583
10962/22/369
1.002
0.838/ꢁ0.812
0.989/0.919
0.05(7)
5123 (Rint =0.0271)
4477
0.0281/0.0687
0.0346/0.0727
5123/0/187
1.004
0.447/ꢁ0.285
0.942/0.912
–
8500 (Rint =0.0707)
3831
0.0772/0.1367
0.2046/0.1756
8500/26/437
1.014
0.610/ꢁ0.460
0.989/0.909
0.00(12)
reflns with I>2s(I)
R1/wR2 [I>2s(I)]
R1/wR2 (all data)
data/restraints/parameters
GOF on F2
largest diff peak/hole (eŠ
max/min transmission
absolute structure parameter
ꢁ3
)
X-ray crystal structure determination: Data were collected on a Bruker
[1] a) H. Kuniyasu, A. Ogawa, S. Miyazaki, I. Ryu, N. Kambe, N.
Sonoda, J. Am. Chem. Soc. 1991, 113, 9796; b) A. Ogawa, J. Organo-
met. Chem. 2000, 611, 463.
three-circle diffractometer equipped with a SMART APEX II CCD de-
tector and corrected for absorption using the SADABS program.[43] Data
[44]
reduction was performed byusing APEX2
and SAINTPlus[45] pro-
[2] Related reviews on the field: a) Catalytic Heterofunctionalization
(Eds.: A. Togni, H. Grützmacher) Wiley-VCH, Weinheim, 2001;
b) M. Beller, J. Seayad, A. Tillack, H. Jiao, Angew. Chem. 2004, 116,
3448; Angew. Chem. Int. Ed. 2004, 43, 3368; c) T. Kondo, T. Mitsu-
do, 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.
[3] I. P. Beletskaya, V. P. Ananikov, Eur. J. Org. Chem. 2007, 3431.
[4] 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, G. G. Aleksandrov, I. L. Eremenko, J. Organo-
met. Chem. 2003, 687, 451; c) V. P. Ananikov, M. A. Kabeshov, I. P.
Beletskaya, V. N. Khrustalev, M. Yu. Antipin, Organometallics 2005,
24, 1275; d) V. P. Ananikov, I. P. Beletskaya, Org. Biomol. Chem.
2004, 2, 284; e) V. P. Ananikov, M. A. Kabeshov, I. P. Beletskaya,
Synlett 2005, 1015.
[5] a) M. Arisawa, M. Yamaguchi, Org. Lett. 2001, 3, 763; b) M. Arisa-
wa, Y. Kozuki, M. Yamaguchi, J. Org. Chem. 2003, 68, 8964.
[6] a) H. Kuniyasu, A. Ogawa, K. Sato, I. Ryu, N. Kambe, N. Sonoda, J.
Am. Chem. Soc. 1992, 114, 5902; b) H. Kuniyasu, A. Ogawa, K.-I.
Sato, I. Ryu, N. Sonoda, Tetrahedron Lett. 1992, 33, 5525; c) I.
Kamiya, E. Nishinaka, A. Ogawa, J. Org. Chem. 2005, 70, 696.
[7] a) C. Cao, L. R. Fraser, J. A. Love, J. Am. Chem. Soc. 2005, 127,
17614; b) V. Circu, M. A. Fernandes, L. Carlton, Polyhedron 2003,
22, 3293; c) Y. Misumi, H. Seino, Y. Mizobe, J. Organomet. Chem.
2006, 691, 3157.
grams. For details, see Table 15. The structures were solved bydirect
methods and refined byfull-matrix least-squares techniques on F2 with
anisotropic displacement parameters for all the non-hydrogen atoms. The
two butyl groups in 2e and one cyclohexyl group in 2m were disordered
over two sites each with occupancies of 0.65:0.35, 0.65:0.35, and
0.60:0.40, respectively. The amino-hydrogen atoms of the cations and hy-
droxy-hydrogen atoms of the anions were localized in the difference
Fourier syntheses and included in the refinement with fixed positional
and isotropic displacement parameters. The other hydrogen atoms were
placed in calculated positions and refined within the riding model with
fixed isotropic displacement parameters (Uiso(H)=1.5Ueq(C) for the CH3
groups and Uiso(H)=1.2Ueq(C) for the other groups). All calculations
were carried out using the SHELXTL (PC Version 6.12) program.[46]
CCDC 656216 (2e), 656217 (2i); 656218 (2k), and 656219 (2m) contain
the supplementarycrystallographic 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.
Acknowledgements
The research work was supported bythe Russian Foundation for Basic
Research (Project No. 07-03-00851), Research grant MD-4094.2007.3,
and Program No 1 of Division of Chemistryand Material Sciences of
RAS. We thank Prof. T. V. Timofeeva for fruitful discussion of X-ray
structures and help in this work.
[8] 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; c) V. P. Ananikov, D. A. Malyshev, I. P.
2432
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434

Alkyne Chemistry
FULL PAPER
Beletskaya, G. G. Aleksandrov, I. L. Eremenko, Adv. Synth. Catal.
2005, 347, 1993.
[9] 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.
[10] Organoselenium Chemistry: Modern Developments in Organic Syn-
thesis (Ed.: T. Wirth), Springer, Berlin, 2000.
[11] Some representative examples: a) S. D. P. Baugh, Z. Yang, D. K.
Leung, D. M. Wilson, R. Breslow, J. Am. Chem. Soc. 2001, 123,
12488; b) B. J. Vesper, K. Salaita, H. Zong, C. A. Mirkin, A. G. M.
Barrett, B. M. Hoffman, J. Am. Chem. Soc. 2004, 126, 16653; c) S.
Belviso, G. Ricciardi, F. Lelj, J. Mater. Chem. 2000, 10, 297; d) S. H.
Eichhorn, D. W. Bruce, D. Guillon, J.-L. Gallani, T. Fischer, J.
Stumpe, T. Geue, J. Mater. Chem. 2001, 11, 1576; e) F. Bonosi, G.
Ricciardi, F. Lelj, G. Martini, J. Phys. Chem. 1994, 98, 10613; f) S. B.
Sesalan, A. Gul, Monatsh. Chem. 2000, 131, 1191; g) H. Braun, A.
Amann, Angew. Chem. 1975, 87, 773; Angew. Chem. Int. Ed. Engl.
1975, 14, 755; h) M. E. Garst, P. Arrhenius, J. Org. Chem. 1983, 48,
16.
[28] It was not possible to locate the corresponding transition state nei-
ther bycreating several starting geometries byhands nor byQST
(STQN) method built in Gaussian 03 package. In all cases the struc-
tures were converged to the product IV during geometryoptimiza-
tion. Most likely, this indicates that the actual activation barrier is
rather small and it was impossible to locate due to exothermic reac-
tion. To ensure that this is not a failure of B3LYP, the calculations
for this transition state were repeated at MP2 level and led to the
same result.
[29] Preliminaryestimation based on the calculations of the oxidative ad-
ꢁ
dition step. The actual mechanism of C E bond breakage could be
quite complicated, see for example the studyinvolving diaryldichal-
cogenides (ref. [25b]).
[30] a) H. Kuniyasu, F. Yamashita, J. Terao, N. Kambe, Angew. Chem.
2007, 119, 6033; Angew. Chem. Int. Ed. 2007, 46, 5929; b) T. Sagawa,
Y. Sakamoto, R. Tanaka, H. Katayama, F. Ozawa, Organometallics
2003, 22, 4433; c) K. Sugoh, H. Kuniyasu, H. Kurosawa, Chem. Lett.
2002, 106.
[12] a) S. W. Benson, Chem. Rev. 1978, 78, 23; b) S. Antonello, R. Benas-
si, G. Gavioli, F. Taddei, F. Maran, J. Am. Chem. Soc. 2002, 124,
7529.
[31] a) I. P. Beletskaya, V. P. Ananikov, Pure Appl. Chem. 2007, 79, 1041;
b) Q. Cui, D. G. Musaev, K. Morokuma, Organometallics 1997, 16,
1355; c) Q. Cui, D. G. Musaev, K. Morokuma, Organometallics 1998,
17, 742.
[13] Internal alkynes were inactive in the catalytic reaction as was found
for the 4-octyne and bis(trimethylsilyl)acetylene.
[32] 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.
[33] T. Ukai, H. Kawazura, Y. Ishii, J. J. Bonnet, J. A. Ibers, J. Organo-
met. Chem. 1974, 65, 253.
[34] V. P. Ananikov, I. P. Beletskaya, Russ. Chem. Bull. 2003, 52, 811.
[35] NMR monitoring (or GC) is the easiest wayto determine appropri-
ate reaction time.
ꢁ
[14] Competitive noncatalytic side-reaction of E E bond addition to al-
kynes (path b; Scheme 1) decreased the overall selectivity in the
case of diaryldisulfides (see ref. [3]). This noncatalytic reaction was
not observed in dialkyldisulfides addition to alkynes under studied
conditions.
[15] We were unable to carryout the catalytic reaction of 1a with tBu2S2
and diadamantyldisulfide. This finding confirms the influence of
bulkysubstituents bonded to the sulfur atom on the reactivityof di-
alkyldisulfides.
[16] a) D. S. Pedersen, C. Rosenbohm, Synthesis 2001, 2431; b) L. M.
Harwood, Aldrichimica Acta 1985, 18, 25.
[36] Original synthetic procedure: a) A. Krief, M. Derock, Synlett 2005,
1012; b) A. Krief, M. Trabelsi, W. Dumont, M. Derock, Synlett 2004,
1751.
[37] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098; b) C. Lee, W. Yang,
R. G. Parr, Phys. Rev. B 1988, 37, 785; c) A. D. Becke, J. Chem.
Phys. 1993, 98, 5648.
[38] a) R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys.
1980, 72, 650; b) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980,
72, 5639.
[39] a) P. Schwerdtfeger, M. Dolg, W. H. E. Schwarz, G. A. Bowmaker,
P. D. W. Boyd, J. Chem. Phys. 1989, 91, 1762; b) D. Andrae, U. Häu-
ßermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim. Acta 1990, 77,
123; c) A. Bergner, M. Dolg, W. Küchle, H. Stoll, H. Preuß, Mol.
Phys. 1993, 80, 1431.
[17] Drycolumn flash chromatographyhas several practical advantages:
1) small amount of silica required; 2) quick elution; 3) economyof
solvents. However, slightlybetter isolated yields (by ꢂ5–10%) may
be achieved using conventional column chromatography.
[18] Using higher temperatures and shorter reaction time is also possible.
However, it requires special attention to experimental equipment
(high pressure maydevelop in reaction vessel) and in some cases
mayfacilitate side reactions.
[19] See also published X-raystructure in reference [4b].
[20] See also published X-raystructures in references [8c,9].
[21] a) G. Mugesh, H. B. Singh, Chem. Soc. Rev. 2000, 29, 347; b) G.
Mugesh, W.-W. du Mont, H. Sies, Chem. Rev. 2001, 101, 2125.
[22] a) C. Bleiholder, R. Gleiter, D. B. Werz, H. Koppel, Inorg. Chem.
2007, 46, 2249; b) M. Tiecco, L. Testaferri, C. Santi, C. Tomassini, S.
Santoro, F. Marini, L. Bagnoli, A. Temperini, F. Costantino, Eur. J.
Org. Chem. 2006, 4867; c) M. Iwaoka, S. Takemoto, M. Okada, S.
Tomoda, Bull. Chem. Soc. Jpn. 2002, 75, 1611; d) M. Iwaoka, S.
Takemoto, M. Okada, S. Tomoda, Chem. Lett. 2001, 30, 132; e) D.
Pal, P. Chakrabarti, J. Biomol. Struct. Dyn. 2001, 19, 115; f) R. M.
Minyaev, V. I. Minkin, Can. J. Chem. 1998, 76, 776.
[23] For clarityreasons onlymononuclear metal species are shown on
Scheme 4 to illustrate the nature of the elementarysteps of the cata-
lytic cycle.
[24] It is a known fact that Ni compounds veryoften result in broad
NMR signals of unacceptable quality, see: Transition Metal Nuclear
Magnetic Resonance (Ed.: P. S. Pregosin), Elsevier, Amsterdam,
1991.
[25] a) J. M. Gonzales, D. G. Musaev, K. Morokuma, Organometallics
2005, 24, 4908; b) V. P. Ananikov, N. V. Orlov, I. P. Beletskaya, Russ.
Chem. Bull. 2005, 54, 576.
[26] The discussion for the DH surface is also valid for the DE surface
and will not be repeated in the article.
[40] a) V. P. Ananikov, D. G. Musaev, K. Morokuma, Organometallics
2005, 24, 715; b) V. P. Ananikov, D. G. Musaev, K. Morokuma, J.
Am. Chem. Soc. 2002, 124, 2839; c) V. P. Ananikov, D. G. Musaev,
K. Morokuma, Organometallics 2001, 20, 1652.
[41] a) C. Gonzalez, H. B. Schlegel, J. Chem. Phys. 1989, 90, 2154; b) C.
Gonzalez, H. B. Schlegel, J. Phys. Chem. 1990, 94, 5523.
[42] Gaussian 03 (Revision C2), M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
yJr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
[27] Since II is higher in energyon the DG surface than I the overall 1!
III-TS barrier should be taken into account.
Chem. Eur. J. 2008, 14, 2420 – 2434
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2433

I. P. Beletskaya, V. P. Ananikov et al.
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pittsburgh,
PA, 2004.
[43] G. M.Sheldrick, SADABS, v. 2.03, Bruker/Siemens Area Detector
Absorption Correction Program, Bruker AXS, Madison, Wisconsin
(USA), 2003.
[45] Bruker AXS, SAINTPlus for NT, v. 6.2, Data Reduction and Cor-
rection Program, Bruker AXS, Madison, Wisconsin (USA), 2001.
[46] G. M.Sheldrick, SHELXTL for NT, v. 6.12, Structure Determination
Software Suite, Bruker AXS, Madison, Wisconsin (USA), 2001.
[44] Bruker AXS, APEX2 software package, Bruker Molecular Analysis
Research Tool, v. 1.27, Bruker AXS, Madison, Wisconsin (USA),
2005.
Received: August 17, 2007
Published online: January21, 2008
2434
www.chemeurj.org
ꢁ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 2420 – 2434