A general study of [(η5-Cp′)2Ti(η 2-Me3SiC2SiMe3)]-catalyzed hydroamination of terminal alkynes: Regioselective formation of Markovnikov and anti-Markovnikov products and mechanistic explanation (Cp′=C (5)H(5), C(5)H(4)Et, C(5)Me(5))

Article abstract of DOI:10.1002/chem.200305674

A general study of the regioselective hydroamination of terminal alkynes in the presence of [(η⁵-Cp)₂Ti(η²-Me ₃SiC₂SiMe₃)] (1), [(η⁵-CpEt) ₂Ti(η²-Me₃SiC₂SiMe₃)] (CpEt= ethylcyclopentadienyl) (2), and [(η⁵-Cp*) ₂Ti(η²-Me₃SiC₂SiMe₃)] (Cp=pentamethylcyclopentadienyl) (3) is presented. While aliphatic amines give mainly the anti-Markovnikov products, anilines and aryl hydrazines yield the Markovnikov isomer as main products. Interestingly, using aliphatic amines such as n-butylamine and benzylamine the different catalysts lead to a significant change in the observed regioselectivity. Here, for the first time a highly selective switch from the Markovnikov to the anti-Markovnikov product is observed simply by changing the catalyst. Detailed theoretical calculations for the reaction of propyne with different substituted anilines and tert-butylamine in the presence of [(η⁵-C₅H₅)Ti(= NR)(NHR)] (R=4-C₆H₄X; X=H, F, Cl, CH₃, 2,6-dimethylphenyl) reveal that the experimentally observed regioselectivity is determined by the relative stability of the corresponding π-complexes 10. While electrostatic stabilization favors the Markovnikov performance for aniline, the steric repulsive destabilization disfavors the Markovnikov performance for tert-butylamine.

Full text of DOI:10.1002/chem.200305674

FULL PAPER
A General Study of [(h5-Cp’)2Ti(h2-Me3SiC2SiMe3)]-Catalyzed
Hydroamination of Terminal Alkynes: Regioselective Formation of
Markovnikov and Anti-Markovnikov Productsand Mechanistic Explanation
(Cp’=C5H5, C5H4Et, C5Me5)
Annegret Tillack, Haijun Jiao, Ivette Garcia Castro, Christian G. Hartung, and
[a]
MatthiasBeller*
Abstract: A general study of the regio-
selective hydroamination of terminal
alkynes in the presence of [(h5-
Cp)2Ti(h2-Me3SiC2SiMe3)] (1), [(h5-
CpEt)2Ti(h2-Me3SiC2SiMe3)] (CpEt=
ethylcyclopentadienyl) (2), and [(h5-
Cp*)2Ti(h2-Me3SiC2SiMe3)] (Cp*=pen-
tamethylcyclopentadienyl) (3) is pre-
sented. While aliphatic amines give
mainly the anti-Markovnikov products,
anilines and aryl hydrazines yield the
Markovnikov isomer as main products.
Interestingly, using aliphatic amines
such as n-butylamine and benzylamine
the different catalysts lead to a signifi-
cant change in the observed regioselec-
tivity. Here, for the first time a highly
selective switch from the Markovnikov
to the anti-Markovnikov product is ob-
served simply by changing the catalyst.
Detailed theoretical calculations for
the reaction of propyne with different
substituted anilines and tert-butylamine
in the presence of [(h5-C5H5)Ti(=
NR)(NHR)] (R=4-C6H4X; X=H, F,
Cl, CH3, 2,6-dimethylphenyl) reveal
that the experimentally observed regio-
selectivity is determined by the relative
stability of the corresponding p-com-
plexes 10. While electrostatic stabiliza-
tion favors the Markovnikov perform-
ance for aniline, the steric repulsive de-
stabilization disfavors the Markovnikov
performance for tert-butylamine.
Keywords: density functional calcu-
lations ¥ hydroamination ¥ regiose-
lectivity ¥ terminal alkynes ¥ titano-
cene
Introduction
methods for the synthesis of imines and amines
(Scheme 1).[2]
Amines and their derivatives are of importance as natural
products, pharmacological agents, fine chemicals, and dye-
stuffs.[1] In general, a number of well-established ™classic∫
organic methods, for example, nucleophilic substitution, re-
duction of amides, nitro compounds, azides, exist for their
syntheses. However, apart from the reductive amination of
Due to the 100% atom economy, that is, each atom from
the starting material is present in the product, no by-prod-
ucts are formed. Easily available terminal olefins or alkynes
provide, in principle, two regioisomeric amines or imines
(Scheme 1, R’=H). For simplicity we use the traditional
Markovnikov and anti-Markovnikov terminology through-
out the text to distinguish the regioisomeric products. In
general, in polar hydroamination reactions the Markovnikov
regioisomer is favored due to the higher stability of the sec-
ondary carbocation.
The negative entropy balance of the hydroamination reac-
tion makes it necessary to use catalysts at lower tempera-
tures. In the past strong bases,[3] both liquid and solid acids[4]
and different types of transition-metal complexes[5,6] have
been employed as catalysts for olefin hydroaminations. De-
spite considerable progress in recent years,[7] a general pro-
tocol for aliphatic intermolecular olefin hydroamination still
needs to be developed. In contrast to olefins, alkynes are
more reactive in hydroaminations, which is shown by the
more exothermic (~70 kJmolÀ1) NH3 addition to acetylene
in comparison to ethylene.[8] Intramolecular cyclizations of
À
carbonyl compounds, the catalytic formation of carbon ni-
trogen bonds is rare. Clearly, there still exists considerable
interest in the development of improved methodologies for
À
the construction of carbon nitrogen bonds. From an envi-
ronmental point of view, transition metal catalyzed hydroa-
mination of olefins and alkynes are particularly attractive
[a] Dr. A. Tillack, Dr. H. Jiao, Dr. I. Garcia Castro, Dr. C. G. Hartung,
Prof. Dr. M. Beller
Leibniz-Institut f¸r Organische Katalyse (IfOK)
an der Universit‰t Rostock e.V.
Buchbinderstrasse 5 6, 18055 Rostock (Germany)
E-mail: matthias.beller@ifok.uni-rostock.de
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. Total electronic
energies and bond parameters as well as natural charges are listed.
2409
Chem. Eur. J. 2004, 10, 2409 2420
DOI: 10.1002/chem.200305674
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPER
Results and Discussion
Hydroamination of aliphatic al-
kyneswith aliphatic, benzylic
and aromatic amines: Recently,
we described the use of com-
plex 1 (Rosenthal catalyst),[21]
which is easily synthesized by
Scheme 1. Hydroamination of alkenes and alkynes.
the reaction of titanocene di-
chloride with the corresponding
aminoalkynes are easier to perform and provide different ni-
trogen-containing heterocycles. These reactions have been
achieved using a broad variety of catalysts.[9] On the other
hand, intermolecular aminations with alkynes have been re-
alized in the presence of strong bases (Cs),[10] lanthanides
(Nd),[11] actinides (U, Th),[12] and late transition metals (Ru,
silylated alkyne,[22] for the hydroamination of aliphatic al-
kynes with aliphatic amines. An advantage of this catalyst[21a]
is the relatively high stability at room temperature, however
upon heating it looses easily the alkyne, thereby generating
an active ™Cp2Ti∫-intermediate within the proposed catalytic
cycle. While reactions with the sterically hindered tert-butyl-
amine proceeded in excellent yield and selectivity, the less
crowded sec-alkylamines gave lower regioselectivities.
Despite the recent advancements in transition metal-cata-
lyzed hydroamination of alkynes until to date, no detailed
studies and general mechanistic explanation for the ob-
served regioselectivity in hydroamination reactions of termi-
nal alkynes are known. However, understanding and con-
trolling of Markovnikov or anti-Markovnikov selectivity are
of significant importance for further applications of this
chemistry and also other refinement reactions of alkynes.[23]
In order to study this chemistry more closely we investigat-
ed the aminations of 1-hexyne and 1-octyne with aliphatic
and mainly aromatic amines further on (Tables 1 and 2). Ini-
tially, we thought that sterically more hindered titanocenes
such as [(h5-CpEt)2Ti(h2-Me3SiC2SiMe3)] (CpEt=ethylcyclo-
pentadienyl) (2) and [(h5-Cp*)2Ti(h2-Me3SiC2SiMe3)] (Cp*=
pentamethylcyclopentadienyl) (3) should lead to improved
regioselectivity applying less hindered primary amines.
Here, we tested especially n-butylamine and benzylamine.
The new titanocene complex 2 and the known complex 3[22]
were synthesized by straightforward reduction from the cor-
responding titanocene dichloride in the presence of magne-
sium and Me3SiCꢀCSiMe3 in 73 and 77% yield, respectively.
In all catalytic reactions shown in Table 1 a slight excess
of amine (1.2 1.5 equiv) was employed in order to suppress
the oligomerization and polymerization of the alkynes. Nev-
ertheless some dimerization, oligomerization and polymeri-
zation of the alkyne were observed. Competition experi-
ments with isolated imines showed that decomposition of
the imine to the corresponding aldehyde is no major side-re-
action under the applied reaction conditions.
g]
Pd, Rh, Au)[13a as well as Hg, Tl[13h j]. Based on the pio-
neering work of Bergman et al.[14] metallocenes have
become popular as active catalysts in these reactions. More
recently, the groups of Bergman,[15] Doye,[16] Odom[17] and
Richeson[18] made significant contributions to the further de-
velopment of titanocene and titanium amide catalysts with
respect to hydroaminations. In most of these reactions aro-
matic, terminal alkynes or internal alkynes are treated with
anilines as substrates. During our studies on the hydroami-
nation of olefins[19] we also became interested in the selec-
tive amination of non-activated aliphatic alkynes. In a pre-
liminary communication we reported the first anti-Markov-
nikov functionalization of terminal alkynes with aliphatic
amines in the presence of [(h5-Cp)2Ti(h2-Me3SiC2SiMe3)] (1)
as catalyst (Scheme 2).[20]
Scheme 2. Titanocene-catalyzed anti-Markovnikov hydroamination of
terminal alkynes.
Outlined herein are new applications of hydroaminations
in the presence of 1 and other titanocene catalysts. For the
first time a systematic investigation of the effects of catalysts
and substrates on the regioselectivity of the hydroamination
of terminal alkynes is presented. The steric nature of the
catalyst reveals a significant influence on the regioselectivity.
Comparison of various anilines, arylhydrazines and aliphatic
amines as the nitrogen source shows a general switch from
anti-Markovnikov to Markovnikov products going from ali-
phatic to aromatic amines. This unusual behavior is ex-
plained by detailed theoretical investigations, which led to a
new mechanistic rationale for the determination of the re-
giochemistry in titanocene-catalyzed hydroaminations of al-
kynes.
In the presence of 1 hydroaminations proceed at 85
1208C to give the anti-Markovnikov imines regioselectively
(Table 1, entries 1, 6, 7 and 10). Even with simple non-hin-
dered amines (n-butylamine and benzylamine) the anti-Mar-
kovnikov product dominates. However, the product yields
were lower (up to 48%) as compared to hindered amines.
In the presence of 2 and 3 higher yields of imines were
obtained both with n-butylamine and benzylamine (up to 62
and 66%, respectively). These results represent one of the
few cases of titanium-catalyzed aminations with non-hin-
dered amines.[17c,24] In the case of tert-butylamine 1 and 2
lead to excellent yields, while 3 is not active at all. Appa-
2410
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420

Hydroamination of Terminal Alkynes
2409 2420
Table 1. Hydroamination of 1-octyne with aliphatic and benzylic amines in the presence of 1 3.
conditions for a specific product
was not done. With the excep-
tion of 3,4,5-trimethoxyaniline
all other anilines gave the cor-
responding imines in the pres-
ence of 1 in good to very good
conversion (61 94%). Surpris-
ingly, the influence of steric ef-
fects is opposite for anilines
and aliphatic amines. In the
case of aliphatic alkynes in-
creased steric bulk leads to a
higher selectivity in favor of the
anti-Markovnikov products. On
the other hand, an increased
steric bulk of the anilines gave
a higher selectivity of the Mar-
kovnikov product!
For example, the reaction of
1-hexyne with 2,6-dimethylani-
line gave the Markovnikov
product with 99:1 selectivity,
while aniline gave only 3:1
(Table 2; entries 1 and 2). The
highest regioselectivities for the
Markovnikov product (>99:1)
were observed with 2,6-dime-
thylaniline (Table 2, entry 4)
Entry
Amine
Catalyst
[mol%]
T
[8C]
Amine/Alkyne
ratio
Yield [%][a]
(anti-M:M)
1
2
3
4
5
1 (2.5)
2 (2.5)
2 (2.5)
2 (2.5)
3 (2.5)
85
85
1.5:1
1.5:1
1.5:1
1.5:1
1.2:1
97 (99:1)[b] 4a
5 (98:2) 4a
97 (98:2) 4a
71 (98:2)[b,d] 4a
0[c] 4a
100
100
100
6
7
8
9
1 (10)
1 (10)
2 (10)
3 (10)
100
120
100
120
1.2:1
1.2:1
1.2:1
1.2:1
19 (79:21)[e] 4b
48 (72:28) 4b
62 (44:56) 4b
51 (22:78) 4b
10
11
12
13
1 (10)
2 (10)
2 (10)
3 (10)
120
120
100
120
1.2:1
1.2:1
1.2:1
1.2:1
46 (82:18) 4c
65 (45:55) 4c
66 (57:43) 4c
51 (19:81) 4c
and
3,4,5-trimethoxyaniline
(Table 2, entry 17). Apart from
the steric bulk, the electronic
factors of aniline also influence
[a] Reaction in toluene, reaction time 24 h. Yield and mol% catalyst refer to the alkyne. Yield of imine was
determined by GC analysis with an internal standard (hexadecane or dodecane). [b] 2 h reaction time. [c] 48 h
reaction time. [d] 73% conversion. [e] 78% conversion.
rently the steric hindrance of both the catalyst and the
amine precludes any reaction. Due to the stronger binding
of alkynes to the titanium center with increased substitution
of the cyclopentadienyl ring,[25] a higher reaction tempera-
ture (100 1208C) has to be employed in the presence of 2
or 3 compared with 1. Despite the increased binding from 1
Scheme 3. Switch of regioselectivity.
to 3, we observed in the GC/MS of the different catalytic re-
actions only Me3SiCꢀCSiMe3 as the dissociation product.
We have no proof for a protonolysis of this alkyne.
the regioselectivity. This effect is nicely demonstrated by
comparing the reactions of 4-chloroaniline, 4-fluoroaniline,
4-methylaniline, aniline, and 4-methoxyaniline with 1-octyne
(Table 2, entries 11 17). Obviously, electron donating sub-
stituents favor the anti-Markovnikov products. Nevertheless,
the steric bulk of the substrate seems to be the main deter-
mining effect as demonstrated by the reaction of 3,4,5-trime-
thoxyaniline. A comparison of catalysts 1, 2 and 3 showed
similar trends to the reactions of aliphatic amines. The more
substituted complex 2 lead to higher Markovnikov selectivi-
ty, namely, the reaction of 1-octyne with aniline gave a se-
lectivity of 75:25 and 83:17 in the presence of 1 and 2, re-
spectively (Table 1, entries 9 10).
Interestingly, the different catalysts 1 3 introduce a signif-
icant change in the observed regioselectivity. For n-butyla-
mine and benzylamine, catalyst 1 favors the anti-Markovni-
kov functionalization of 1-octyne as the main reaction path-
way (anti-M:M 4 2.5:1). On the other hand, catalyst 3
favors the Markovnikov isomer being the main product
(Scheme 3). Clearly, this simple control of regioselectivity in
hydroaminations of terminal alkynes is interesting.
Next, we were interested in the amination of terminal ali-
phatic alkynes with aromatic amines. As shown in Table 2,
ten different substituted anilines were treated with 1-hexyne
and 1-octyne. In general, anilines react slower than aliphatic
amines. Therefore, in some cases a slightly higher reaction
temperature (up to 1008C) and higher catalyst amount
(5 mol%) were necessary, in order to achieve full conver-
sion and high yield. Nevertheless, optimization of reaction
Apart from 1-hexyne and 1-octyne also other terminal al-
kynes react with 2,6-dimethylaniline. In Table 3 the hydroa-
mination of 3-phenyl-1-propyne, 3-cyclopentyl-1-propyne,
1,7-octadiyne, and phenylacetylene with 2,6-dimethylaniline
2411
Chem. Eur. J. 2004, 10, 2409 2420
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

M. Beller et al.
FULL PAPER
Table 2. Hydroamination of 1-hexyne and 1-octyne with aromatic amines in the presence of 1 3.
Hydroamination of terminal al-
kyneswith hydrazines : Recent-
ly, Odom et al. described for
the first time the titanium
amide-catalyzed reaction of ar-
ylhydrazines with alkynes.[27]
This reaction step in combina-
tion with the Fischer indole
synthesis[28] allows for an ele-
gant two-step (one-pot) synthe-
Entry
Amine
Alkyne
Catalyst
[mol%]
T
[8C]
Amine/Alkyne
ratio
Yield [%][a]
(anti-M:M)
1
2
3
1 (3.0)
1 (5.0)
1 (5.0)
85
85
85
1.5:1
1.2:1
1.5:1
94 (1:99)[b] 4d
94 (25:75)[b] 4e sis of substituted indoles. Still
today the development of new
indole syntheses is subject of
considerable efforts due to the
great variety of indole units in
68 (2:98)
natural products and pharma-
ceutical compounds. Apart
from the synthetic challenge, it
was of special interest for us to
compare the regioselective
4
5
2 (5.0)
3 (5.0)
100
100
1.5:1
1.2:1
80 (1:>99)
0[c]
attack of arylhydrazines on ter-
minal alkynes with the above
mentioned reactions of aliphat-
ic amines and anilines. As a
model system N-methyl-N-phe-
nylhydrazine was used for ami-
6
7
8
1 (5.0)
2 (2.5)
1 (5.0)
85
100
100
1.2:1
1.2:1
1.2:1
93 (12:88)
96 (6:94)
nation reactions of different al-
kynes. In general, reactions
80 (9:91)
were performed for 24 h at 85
1008C in the presence of 2.5
10 mol% of 1.
Although no systematic opti-
mization had been done, some
screening reactions using 1-
octyne demonstrated full con-
version in the presence of
2.5 mol% catalyst after 2 h. As
shown in Table 4 reaction of 1-
9
1 (5.0)
2 (2.5)
1 (2.5)
1 (2.5)
1 (2.5)
1 (2.5)
1 (2.5)
100
100
85
1.2:1
1.2:1
1.2:1
1.2:1
1.2:1
1.2:1
1.2:1
80 (25:75)
10
11
12
13
14
15
90 (17:83)
72 (25:75)
85
75 (20:80)
67 (25:75)
61 (33:67)
61 (50:50)
85
85
octyne,
3-cyclopentyl-1-pro-
pyne, 3-phenyl-1-propyne and
5-chloro-1-pentyne with N-
85
methyl-N-phenylhydrazine
in
16
17
1 (2.5)
1 (2.5)
85
85
1.2:1
1.2:1
71 (3:97)
the presence of 1 and subse-
quent treatment of the reaction
mixture with an excess of ZnCl2
gave directly the corresponding
indoles in 52 90% yield.
Except for phenylacetylene all
reactions occurred with high
Markovnikov selectivity leading
to the 2-methyl-3-alkylsubstitut-
ed indoles. However, in the
43 (1:>99)
[a] 24 h reaction time in toluene. Yield and mol% catalyst refer to the limiting alkyne. Yield was determined
by GC analysis with an internal standard (hexadecane or dodecane) after hydrolysis with 5% HCl and for 2-
octanone/n-octanale. [b] Yield refers to the imine. [c] 48 h reaction time.
is shown. Except for phenylacetylene (Table 3, entry 5),
which oligomerizes fast, reasonable to good yields (43
73%) and excellent regioselectivities for the Markovnikov
products (M:anti-M >96:4) are obtained. Notably the
double hydroamination of 1,7-octadiyne proceeds in 73%
yield with excellent selectivity giving after hydrolysis 2,7-oc-
tadione, which is otherwise difficult to access.[26]
case of phenylacetylene both indole isomers were isolated in
a ratio M:anti-M 4:1. 3-(2-Aminoethyl)-substituted indoles
are of special interest to organic synthesis because of the bi-
ological activity of the tissue hormone melatonin and the
neurotransmitter serotonin. Obviously, the reaction of com-
mercially available 5-chloro-1-pentyne with arylhydrazines
allows for a straightforward two-step preparation of this
2412
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420

Hydroamination of Terminal Alkynes
2409 2420
Table 3. Hydroamination of terminal alkynes with 2,6-dimethylaniline in the presence of 1.
out high level density functional
theory computations. The calcu-
lated bond lengths, bond angles
and natural charges as well as
the energetic data for the p-
Entry
Alkyne
T
Amine/Alkyne
ratio
Yield [%][a]
(anti-M:M)
[8C]
complexes
(10),
transition
states (11) and intermediates
(12) of the [2+2]-cycloaddition
are summarized in the Support-
ing Information.
1
2
85
85
1.5:1
1.5:1
68 (2:98)
72 (3:97)
3
4
5
85
100
85
1.2:1
4:1
43 (2:98)
73 (4:96)
28 (50:50)
The detailed mechanism of
the catalytic hydroamination of
alkynes has been computed
previously by Straub and Berg-
man using the simplified [(h5-
C5H5)Ti(=NH)(NH2)] and HCꢀ
CH models.[29b] Along the reac-
tion path on the potential
1.5:1
[a] 5 mol% catalyst 1, 24 h reaction time in toluene. Yield and mol% catalyst refer to the alkyne. Yield was
determined by GC analysis with an internal standard (hexadecane or dodecane) after hydrolysis with 5% HCl
and for ketone/aldehyde.
class of indoles. Surprisingly, when we performed the reac-
tion of N-methyl-N-phenylhydrazine with 5-chloro-1-pen-
tyne in the presence of 10 mol% of 1 directly, the corre-
sponding hydrochloride of N-methyl-3-(2-aminoethyl)-2-
methylindole was obtained in good yield (Table 4, entry 6).
After addition of NaOH the free indole (9c) was easily iso-
lated and characterized. Keeping the commercial availability
of different substituted aryl hydrazines in mind obviously
this reaction can be extended to various other substituted
serotonin products.
Mechanism and theoretical calculations of the hydroamina-
tion of terminal alkynes: Based on the original work by
Bergman[15] the mechanism of the titanocene-catalyzed hy-
droamination has been recently defined more precisely by
Doye et al.[29a] and Bergman et al.[29b] As shown in Scheme 4,
the active catalyst is believed to be a titanium imido species,
which is in equilibrium with the corresponding bisamidotita-
nium complexand dinuclear titanium complexes. A formal
[2+2]-cycloaddition of the titanium imido species and the
alkyne gives a titanaazacyclobutene derivative. Subsequent
protonation by excess amine and tautomerization of the cor-
responding enamine leads to the imine product and the
active catalyst is recovered.
Although the proposed catalytic cycle is in agreement
with most previously reported observations, some questions
remained considering our experimental results. For example,
Bergman et al.[29b] concluded that the regioselectivity of the
reaction is determined in the [2+2]-cycloaddition step,
which should also be the rate-determining step. However,
with regard to our results it is not clear why alkylamines
favor the anti-Markovnikov products, while anilines and ar-
ylhydrazines favor the Markovnikov products. Which factors
determine the regioselectivity of the process and how is it
controlled? Not surprisingly, known mechanistic investiga-
tions and calculations used symmetrical alkynes as model
systems in order to circumvent the problem of regioselectiv-
ity. In order to get insight into the origin of the unusual dif-
ferences in regioselectivity reported above, we have carried
Scheme 4. Proposed mechanism of the titanocene-catalyzed amination of
alkynes.
energy surface (PES), they found the formation of a p-com-
plexbetween the metal complexand ethyne being the first
step which is also endergonic, followed by a [2+2]-cycload-
dition as the rate-determining step from the calculated acti-
vation parameters. They also postulated that the regioselec-
tivity is controlled by the cycloaddition, and the related
points on the PES of the first part of the ethyne hydroami-
nation path is shown in Scheme 5. However, these simplified
models do not explain the observed regioselectivity in case
of terminal alkynes.
On the basis of Bergman×s finding,[29b] we used the ™real∫
complexes [(h5-C5H5)Ti(=NR)(NHR)][30] to model the differ-
ence in regioselectivity of hydroamination between substi-
tuted anilines (R=4-C6H4X (X=H, F, Cl, CH3), and 2,6-di-
methylphenyl) and tert-butylamine (R=C(CH3)3). Instead
of HCꢀCH, we used H3C-CꢀCH as a model for the em-
ployed terminal aliphatic alkynes. The related Markovnikov
2413
Chem. Eur. J. 2004, 10, 2409 2420
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

M. Beller et al.
FULL PAPER
Table 4. Hydroamination of terminal alkyne with N-methyl-N-phenylhydrazine in the presence of 1.
activation and reaction. The re-
sults clearly show that Markov-
nikov (denoted as M) and anti-
Markovnikov (denoted as anti-
M) reactions have approximate-
ly the same Gibbs free energies
for activation (DG°), and
Entry
1
Alkyne
Indole
Catalyst
[mol%]
T
[8C]
Yield[a]
[%]
their
negligible
differences
(<0.3 kcalmolÀ1) could not be
responsible for the observed
difference in the regioselectivi-
ty. For example, on the basis of
the difference in the activation
free energy (considering these
two transition states in an equi-
librium, and their equilibrium
constant (K) is determined by
their relative Gibbs free energy,
DDG° = ÀRTlnK), the predi-
cated product ratio would be
roughly 60 to 40 in favor of the
5.0
2.5
100
100
90
2
3
88[b] (70)
3.0
5.0
100
85
82 (62)
84 (67)
anti-Markovnikov
pathway.
This result gives the right trend
for tert-butylamine, but the op-
posite trend for aniline. On the
other hand, the predicated
product ratios do not agree
with the experimental finding
in both cases (Tables 1 and 2).
Therefore, the transition state
of the [2+2]-cycloaddition step
(11) does not control the regio-
selectivity.
4
5
5.0
100
42 (25)
Alternatively,
one
might
10 (5)
(64)[c]
expect that the regioselectivity
can be explained by the ther-
modynamic effect or the differ-
ence of reaction free energies
(DDG). As given in Table 5, the
anti-Markovnikov products (12)
of both tert-butylamine and ani-
line are more favored energeti-
cally than the Markovnikov iso-
6
10.0
100
[a] Reaction in toluene, 24 h reaction time, alkyne/hydrazine ratio 1:1.2. 3 4 equiv ZnCl2, reaction time 24 h.
Yield and mol% catalyst refer to the alkyne. Yield was determined by GC analysis with an internal standard
(hexadecane), isolated yield are given in parentheses [b] 2 h reaction time to hydrazone. [c] The indole was
formed without ZnCl2 and isolated as the hydrochloride (isolated yield).
Scheme 5. [2+2]-Cycloaddition: 10 (p-complex); 11 (transition state) and
12 (intermediate).
Scheme 6. [2+2]-Cycloaddition products with CH3C=CH.
and anti-Markovnikov products of the [2+2]-addition are
shown in Scheme 6.
Initially, the basic thermodynamic data for the reaction of
tert-butylamine and aniline with propyne were calculated.
Table 5 summarizes the computed Gibbs free energies for
mers by 1.5 and 2.0 kcalmolÀ1, and the predicated product
ratio would be 93 to 7 for tert-butylamine and 97 to 3 for
aniline. This is in good agreement with the experimental
finding for tert-butylamine, but the result for aniline is total-
ly wrong. Therefore, the regioselectivity also can not be ex-
2414
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420

Hydroamination of Terminal Alkynes
2409 2420
[a]
Table 5. Gibbs free energies [kcalmolÀ1
]
of activation (DG°, transition
kovnikov products with a percentage ratio of 76 to 24, again
in perfect agreement with the experimental result (75 to 25,
Table 1). As for aniline, para-substituted anilines also favor
Markovnikov over anti-Markovnikov products, and it is very
interesting to see that the ratio of Markovnikov to anti-Mar-
kovnikov depends on the electronegativity of the substitu-
ent, as also found experimentally. For example, the most
electronegative F substituent results in a higher Markovni-
kov to anti-Markovnikov ratio (10 f:10e 93:7) than the less
electronegative Cl (10h:10g 84:16) or H (10d:10c 74:26).
Therefore, the energetic difference between the p-com-
plexes reproduces the regioselectively not only qualitatively,
but also quantitatively.
Apart from this perfect agreement between theory and
experiment, it is interesting and also unavoidable to look for
the driving force and insight of these differences. Are these
results due to steric or electronic effects? To answer this
question, we analyzed the structural parameters and natural
charge distributions. The optimized structures and natural
charges (bold and italics) of tert-butylamine and aniline as
substrates are shown in Figure 1, and selected bond parame-
ters and the atomic natural charges from NBO analyses are
summarized in the Supporting Information.
Initially, we were interested in the distances between the
CꢀC triple bond of propyne and the formal Ti=N double
bond of [(h5-C5H5)Ti(=NR)(NHR)]. For tert-butylamine, the
Ti C1 and Ti C2 distances (2.621 and 2.711 ä) of the anti-
Markovnikov p-complex 10a are shorter than those (2.644
and 2.753 ä) of the Markovnikov isomer (10b). This differ-
ence indicates a stronger interaction between metal center
and substrate in 10a than in 10b. Therefore, the former
should be favored energetically, and this is confirmed by the
calculated energy difference of 2.45 kcalmolÀ1. For aniline,
however, the Ti C1 distance (2.563 ä) of the anti-Markovni-
kov p-complex( 10c) is longer than that (2.412 ä) of the
Markovnikov isomer (10d), while the Ti C2 distance of the
former (2.596 ä) is shorter than that (2.685 ä) of the latter.
This mixed behavior agrees with their rather small energetic
order by 0.62 kcalmolÀ1.
Secondly, we were interested in the orientation of pro-
pyne with respect to the metal complexes [(h5-C5H5)Ti(=
NR)(NHR)]. In both anti-Markovnikov p-complexes (10a
and 10c) (R = C(CH3)3 and C6H5), propyne has nearly the
same orientation as indicated by the calculated N1TiC1C2
torsion angles (À161.5 vs À165.18), and the formal four-
membered rings have very close bond angles (see Support-
ing Information). Therefore, there is no significant differ-
ence in conformation between 10a and 10c.
states, 11) and reaction (DG, intermediates, 12) relative to p-complexes
(10).
Amine
Model
DG°
DG
(H3C)3CNH2
anti-M
M
6.2 (11a)
6.5 (11b)
3.2 (11c)
3.5 (11d)
2.1 (11e)
3.3 (11 f)
3.2 (11g)
4.0 (11h)
2.5 (11i)
3.0 (11j)
2.9 (11k)
5.6 (11l)
À17.8 (12a)
À16.3 (12b)
À15.7 (12c)
À13.7 (12d)
À16.1 (12e)
À12.7 (12 f)
À15.7 (12g)
À13.0 (12h)
À16.2 (12i)
À14.4 (12j)
À13.8 (12k)
À10.8 (12l)
C6H5NH2
anti-M
M
4-F-C6H4NH2
anti-M
M
4-Cl-C6H4NH2
4-CH3-C6H4NH2
2,6-dimethyl-C6H3NH2
anti-M
M
anti-M
M
anti-M
M
[a] At B3LYP/LANL2DZp//B3LYP/LANL2DZ including the correction
of thermal energy and entropy contribution at T=298.15 K.
plained by the thermodynamic effect. However, what are
the driving forces for these differences?
As the first step along the reaction path, we were wonder-
ing whether the relative stability of the Markovnikov and
anti-Markovnikov p-complexes (10) should be responsible
for the observed difference in the regioselectivity. One
might consider that the two p-complexes should be in equili-
brium during their formation, and their equilibrium constant
(K) is determined by their relative Gibbs free energy (DG=
ÀRTlnK), which is also the difference of the two competing
reaction energies on the basis of the active catalyst [(h5-
C5H5)Ti(=NR)(NHR)]. The consequence is that the more
stable the p-complex, the more dominant the product.
These results are summarized in Table 6.
[a]
Table 6. Gibbs free energy [kcalmolÀ1
]
difference between the anti-
Markovnikov and Markovnikov p-complexes (10) and the related ratio
(anti-M:M).
Amine
anti-M
M
Ratio[b]
(H3C)3CNH2
C6H5NH2
4-F-C6H4NH2
4-Cl-C6H4NH2
4-CH3-C6H4NH2
2,6-dimethyl-C6H3NH2
0.00 (10a)
0.00 (10c)
0.00 (10e)
0.00 (10g)
0.00 (10i)
0.00 (10k)
2.45 (10b)
98:2 (99:1))
26:74 (25:75)
7:93 (25:75)
16:84 (20:80)
23:77 (33:67)
1:99 (2:98)
À0.62 (10d)
À1.56 (10 f)
À0.97 (10h)
À0.73 (10j)
À2.56 (10l)
[a] At B3LYP/LANL2DZp//B3LYP/LANL2DZ including the correction
of thermal energy and entropy contribution at T=298.15 K. [b] The ob-
served ratio is given in parenthesis.
For the reaction with tert-butylamine, the anti-Markovni-
kov p-complex( 10a) is computed to be lower in energy
than the Markovnikov one (10b) by 2.45 kcalmolÀ1, and,
therefore, 10a should be more dominant over 10b, and this
should also be the case for the subsequent products. This
free energy difference gives a percentage ratio of 98:2 for
anti-Markovnikov to Markovnikov products, and this ratio
matches the experimental finding (99:1) perfectly (Table 1)!
In contrast to tert-butylamine, the anti-Markovnikov p-
complex( 10c) of aniline is computed to be higher in energy
than the Markovnikov one (10d) by 0.62 kcalmolÀ1, and this
energy difference favors the Markovnikov over anti-Mar-
However, the propyne unit has different orientations in
the Markovnikov p-complexes (10b and 10d) as indicated
by the calculated N1TiC1C2 torsion angles. For example, the
N1TiC1C2 torsion angle for 10d (R=C6H5) of only À0.28
shows that the formal four-membered ring is nearly planar,
and this also indicates that there is no steric interaction be-
tween the methyl group and the phenyl ring. In contrast, the
N1TiC1C2 torsion angle for 10b (R=C(CH3)3) of À38.48
shows the non-planarity of the four-membered ring and re-
veals at the same time the steric repulsive interaction be-
tween the methyl groups of propyne and the tert-butyl
2415
Chem. Eur. J. 2004, 10, 2409 2420
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

M. Beller et al.
FULL PAPER
Apart from the orientation of propyne, the calculated nat-
ural charge distributions are informative for understanding
the energetic difference and, therefore, the difference in the
regioselectivity between tert-butylamine and aniline as sub-
strates. As shown in Figure 1, the p-complexes 10a d are
electrostatic in nature as indicated by the calculated natural
charges from NBO analysis. In 10a d, the formal Ti=N
double is highly polarized with positively charged titanium
and negatively charged nitrogen centers. The terminal ꢀCH
(C1) carbon is negative charged in all these p-complexes,
while ꢀC2 is nearly neutral.
For tert-butylamine as substrate, the electrostatic stabiliz-
ing interaction in the Markovnikov p-complex( 10b) should
be weaker than in the anti-Markovnikov isomer (10a) due
to the steric repulsive interaction between the methyl
groups in propyne and tert-butyl unit and the resulting
longer Ti C1 and Ti C2 distances. This is indicated by the
N1TiC1C2 dihedral angle of À38.4 and the rather long N1 C2
distance of 3.160 ä in 10b, and, therefore, is responsible for
the reduced electronic stabilization.
For aniline as substrate, no such steric repulsive interac-
tion in the p-complexes (10c and 10d) is observed, and the
alternating positive and negative charge interaction of the
Markovnikov p-complex( 10d) in the nearly planar four-
membered ring is the decisive factor for the enhanced stabil-
ity over the anti-Markovnikov isomer (10c).
It is, therefore, to conclude that the experimentally ob-
served difference in the regioselectivity between tert-butyla-
mine and aniline as substrates is determined by their rela-
tive stability of their p-complexes (10) with terminal aliphat-
ic alkynes, that is, electrostatic stabilization favors the Mar-
kovnikov performance for aniline, while the steric repulsive
destabilization disfavors the Markovnikov performance for
tert-butylamine.
In addition to this insight comparison, we have also calcu-
lated the effect of substituted anilines. As given in Table 5,
all substituted anilines favor Markovnikov p-complexes
(10 f, h, j and l), and, therefore, the Markovnikov hydroami-
nation products (the related structural parameters and natu-
ral charges are given in the Supporting Information). This
agrees with the experimental results. As shown in Table 5,
the Gibbs free energy difference between the anti-Markov-
nikov (10k) and Markovnikov (10l) p-complexes for 2,6-di-
methyl aniline (2.65 kcalmolÀ1) is much larger than that for
aniline (10c and 10d; 0.62 kcalmolÀ1). This large energetic
difference can be ascribed to the steric repulsive interaction
among the methyl groups of propyne, 2,6-dimethyl phenyl
ring and the (h5-C5H5) ligand, and this is indicated by the
Figure 1. The computed N1TiC1C2 torsion angles and natural charges
(bold and italics) for the p-complexes of tert-butylamine (top) and aniline
(bottom) as substrates.
group. Therefore, the orientation of propyne in the p-com-
plexes 10b and 10d might be indicative for their stability.
That tert-butyl is more bulky than phenyl is also shown by
shortest H¥¥¥H distances among these ligands (Hsub¥¥¥HCp =
2.300, Hsub¥¥¥HMe =2.286 and HMe¥¥¥HCp =2.365 ä; see Sup-
porting Information) in 10k. The direct consequence is the
elongation of Ti C1 and Ti C2 distances (2.594 and 2.702 ä),
as compared to corresponding values (2.563 and 2.596 ä) of
anilines, as shown in Figure 2. In contrast, the Ti C1 and Ti
C2 distances are very close for aniline and 2,6-dimethylani-
line in the Markovnikov p-complexes.
These results also give a nice explanation about the reac-
tivity and regioselectivity of catalyst 3 with the permethylat-
ed cyclopentadienyl ring as ligand. For example, there is no
À
À
the longer Ti C1 and Ti C2 distances in 10a (2.621 and
2.711 ä) and 10b (2.644 and 2.753 ä) than in 10c (2.563 and
2.596 ä) and 10d (2.412 and 2.685 ä), respectively. This is
consistent with the difference of the calculated Gibbs free
activation energies for both substrates (6.2 and 6.5 kcal-
molÀ1 for R=C(CH)3)3 vs 3.2 and 3.5 kcalmolÀ1 for R=
C6H5, Table 5), that is propyne is stronger activated in the
p-complexes with R=C6H5 than with R=C(CH3)3.
2416
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420

Hydroamination of Terminal Alkynes
2409 2420
amines, for example, n-butylamine and benzylamine favor
the anti-Markovnikov functionalization of 1-octyne in the
presence of catalyst 1, while catalyst 3 favors the Markovni-
kov isomer being the main product. Clearly, such an easy
control of regioselectivity in hydroaminations of terminal al-
kynes has not been reported previously.
Experimental Section
Computation: All calculations were carried out by using the Gaussian 98
program.[31] All structures were first optimized at the Hartree Fock (HF)
level of theory with the LANL2DZ[32] basis set, and the nature of the op-
timized structures on the potential energy surface (PES) was character-
ized by the calculated number of imaginary frequency (NImag) at the
same level of theory (HF/LANL2DZ), i.e., minimum structures without
(NImag=0), and transition states with only one imaginary frequency
(NImag=1).[33] This moderate theoretical method was used to make a
systematic comparison for large systems possible. The related frequency
calculations provided at the same time zero-point energies (ZPE) and
thermal energies as well as entropies at given temperature (T=
298.15 K), and all these data were scaled by an empirical factor of 0.8929
and used for the calculations of the thermodynamic parameters.[33] The
HF structures obtained were further refined at the electron correlated
B3LYP DFT level of theory with the LANL2DZ basis set (B3LYP/
LANL2DZ), and the final energies were the single-point energies at the
B3LYP level with the B3LYP/LANL2DZ geometries and the LANL2DZ
basis set by adding a set of polarization functions (LANL2DZp[32c]). This
combination (B3LYP/LANL2DZp) is found to be appropriate for the ac-
curate determination of the structural isomer ratios on the basis of the
computed relative energies in titanium and ziconium complexes.[34] The
energies for discussion and interpretation are the Gibbs free energies
(DG = DHÀTDS). Natural charges were obtained with the method of
Natural Bond Orbital (NBO) analyses.[35] The calculated total electronic
energies, ZPE, NImag, thermal energies and entropies of the p-com-
plexes, transition states and products of [2+2] cycloadditions are sum-
marized in the Supporting Information.
Figure 2. The conformation for the p-complexes of 2,6-dimethyl amine
substrate [ä].
reaction for catalyst 3 with tert-butylamine as substrate, and
this is apparently due to the bulky property of the active
catalyst,
[(h5 C5(CH3)5)Ti(=N(C(CH3)3)(NH(C(CH3)3)],
General: Chemicals were obtained from Aldrich, Fluka, Acros and
Strem and unless otherwise noted were used without further purification.
Amines were distilled from CaH2. Alkynes were degassed, flushed with
argon and stored over molecular sieves (4 ä). Absolute solvents were
purchased from Fluka. All operations were carried out under an argon
atmosphere.
which hinders the effective coordination between the titani-
um center and the alkyne. Using n-butylamine and benzyla-
mine as substrates, on the other hand, the less bulky catalyst
1 favors the anti-Markovnikov products, while the more
bulky catalyst 3 favors the Markovnikov pathway.
Catalysts 1 and 3 were synthesized according to a literature procedure.[22]
Imines 4a d were isolated after distillation of the crude hydroamination
mixtures, indoles 5a 8a and 8b were isolated by column chromatogra-
phy, compound 9a was isolated as hydrochloride adduct and all products
Conclusion
1
were characterized by NMR, MS, IR and elemental analyses. The H and
13C NMR chemical shifts are reported relative to the center of solvent
peak (CDCl3: 7.25 (1H), 77.0 (13C); [D8]THF: 1.73 (1H), 25.2 (13C)). Iden-
tification of all other products was performed via comparison with au-
thentic products. 4e was synthesized according to ref. [13c]. Not commer-
cially available ketones, such as octan-2,7-dione and cyclopentylacetone,
were isolated after amination and hydrolysis. The analytical data is in
agreement with literature data.[26,36]
In conclusion we have presented the first general study of
the regioselective hydroamination of terminal alkynes. As
nitrogen source aliphatic amines, anilines and arylhydrazines
were used. Depending on the amine the Markovnikov or
the anti-Markovnikov regioisomer is formed preferentially.
The experimentally observed isomer distribution is ex-
plained perfectly by detailed theoretical investigations
which demonstrate, that the regioselectivity is determined
by the relative stability of the corresponding p-complexes
10. This leads to a general understanding of titanocene-cata-
lyzed hydroaminations of unsymmetrical alkynes. Interest-
ingly, electrostatic stabilization favors the Markovnikov per-
formance for aromatic amines, while steric repulsive destabi-
lization disfavors the Markovnikov performance for sterical-
ly hindered aliphatic amines.
Synthesis of complex 2: [(CpEt)2TiCl2] (1.0 g, 3.3 mmol), finely shaved
magnesium (95 mg, 3.9 mmol) and bis(trimethylsilyl)acetylene (0.61 g,
3.6 mmol) were stirred in THF (10 mL) at room temperature under
argon for 2 h. The resulting dark solution was filtered and evaporated in
vacuo to dryness at room temperature. Then the residue was dissolved in
pentane (10 mL) and the solution again filtered. Cooling the pentane fil-
trate to À788C gave yellow-green crystals of complex 2 which were sepa-
rated from the mother liquor by decanting and drying in vacuo at room
temperature to yield the title compound (0.98 g, 73%). 1H NMR
([D8]THF, 400 MHz): d = 6.58 (t, J=2.6 Hz, 4H), 5.75 (t, J=2.6 Hz,
4H), 2.26 (q, J=7.5 Hz, 4H), 1.30 (t, J=7.7 Hz, 6H), À0.34 (s, 18H); 13C
NMR ([D8]THF, 100 MHz): d = 245.1, 137.7, 115.2, 114.9, 25.1, 15.3, 1.1;
29Si NMR ([D8]THF, 79 MHz): d = À13.9; MS (EI, 70 eV): m/z: 404 (1)
[M +], 234 (100) [M +ÀMe3SiC2SiMe3], 170 (5) [Me3SiC2SiMe3+], 155
Significant changes in the regioselectivity are also ob-
served using different titanocene catalysts 1 3 with aliphatic
2417
Chem. Eur. J. 2004, 10, 2409 2420
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

M. Beller et al.
FULL PAPER
+
(50) [Me3SiC2SiMe3 ÀMe]; FT IR (nujol): n˜ = 1677, 1639, 1243, 860,
1,2-Dimethyl-3-pentylindole (5a): According to the general procedure 1-
octyne (0.8 mL, 5.4 mmol) and N-methyl-N-phenylhydrazine (0.76 mL,
832, 787, 750 cmÀ1; elemental analysis calcd (%) for C22H36Si2Ti: C 65.31,
H 8.97; found: C 65.60, H 9.02.
6.5 mmol) were treated in the presence of 2.5 mol%
1 (45 mg,
0.13 mmol) in toluene (4 mL) at 1008C for 2 h. Then ZnCl2 (2.9 g,
21.6 mmol) was added. The residue was purified by column chromatogra-
phy (n-hexane/ethyl acetate 5:1) to afford 5a as a pale yellow oil. GC
yield: 88% (isolated yield: 814 mg (70%)). 1H NMR (CDCl3, 400 MHz):
d = 7.59 (d, J=7.7 Hz, 1H), 7.30 (d, J=8.1 Hz, 1H), 7.24 7.18 (m, 1H),
7.16 7.11 (m, 1H), 3.69 (s, 3H), 2.78 (t, J=7.7 Hz, 2H), 2.41 (s, 3H), 1.68
(m, 2H), 1.46 1.39 (m, 4H), 0.97 (t, J=7.1 Hz, 3H); 13C NMR (CDCl3,
100 MHz): d = 136.5, 132.5, 127.8, 120.3, 118.4, 118.0, 111.7, 108.4, 31.8,
30.8, 29.4, 24.4, 22.6, 14.1, 10.2; MS (EI, 70 eV): m/z (%): 215 (12) [M +],
158 (100) [M +ÀC4H9]; FT IR (neat): n˜ = 736 cmÀ1 (ArH o-disubst.); ele-
mental analysis calcd (%) for C15H21N: C 83.67, H 9.83, N 6.50; found: C
84.03, H 10.14, N 6.60.
General procedure for the reaction of alkyneswith amines : In an Ace-
pressure tube under an argon atmosphere a solution of the catalyst in tol-
uene was added to a mixture of the alkyne and the amine. This mixture
was heated at the given temperature for the specified time (see Tables 1
3). Isolation of the product was done by fractional distillation in vacuo.
N-tert-Butyl-octylidene-amine (4a): According to the general procedure
1-octyne (3.2 mL, 21.5 mmol) and tert-butylamine (3.5 mL, 32.2 mmol)
were treated in the presence of 2.5 mol% 1 (188 mg, 0.54 mmol) in tolu-
ene (8 mL) at 858C for 2 h. Fractional distillation afforded 4a as a color-
less oil. GC yield: 97% (isolated yield: 2.8 g (71%)); b.p. 48 498C/
0.1 mbar; 1H NMR (CDCl3, 400 MHz): d = 7.56 (t, J=5.3 Hz, 1H), 2.20
(m, 2H), 1.46 (m, 2H), 1.34 1.21 (m, 8H), 1.14 (s, 9H), 0.85 (t, J=
7.1 Hz, 3H); 13C NMR (CDCl3, 100 MHz): d = 159.3, 56.4, 36.4, 31.7,
29.6, 29.2, 29.1, 26.4, 22.6, 14.0; MS (EI, 70 eV): m/z (%): 184 (15) [M +
+H], 183 (2) [M +], 168 (30) [M +ÀCH3], 112 (14), 99 (100) [C7H15+], 84
(86) [C4H9NCH+], 57 (91) [C4H9+], 43 (38), 41 (36); FT IR (neat): n˜ =
1671 cmÀ1 (C=N); elemental analysis calcd (%) for C12H25N: C 78.62, H
13.74, N 7.64; found: C 78.19, H 13.99, N 7.42.
1,2-Dimethyl-3-cyclopentylindole (6a): According to the general proce-
dure 3-cyclopentyl-1-propyne (0.76 mL, 5.7 mmol) and N-methyl-N-phe-
nylhydrazine (0.82 mL, 6.9 mmol) were treated in the presence of 3.0
mol% 1 (60 mg, 0.17 mmol) in toluene (4 mL) at 1008C for 24 h. There-
after ZnCl2 (2.3 g, 17.2 mmol) was added. The residue was purified by
column chromatography (n-hexane/ethyl acetate 5:1) to afford 6a as a
colorless solid. M.p. 102 1048C; GC yield: 82% (isolated yield: 754 mg
(62%)). 1H NMR (CDCl3, 400 MHz): d = 7.68 (d, J=7.9 Hz, 1H), 7.31
(d, J=8.1 Hz, 1H), 7.20 (m, 1H), 7.10 (m, 1H), 3.68 (s, 3H), 3.35 3.25
(m, 1H), 2.43 (s, 3H), 2.11 1.96 (m, 6H), 1.86 1.77 (m, 2H); 13C NMR
(CDCl3, 100 MHz): d = 136.9, 132.0, 126.1, 120.2, 119.1, 118.2, 114.1,
108.7, 37.4, 32.9, 29.4, 26.4, 10.4; MS (EI, 70 eV): m/z (%): 213 (67) [M +
], 198 (21) [M +ÀCH3], 184 (100) [M +ÀC2H5], 170 (19) [M +ÀC3H7], 158
(29); FT IR (KBr): n˜ =738 cmÀ1 (ArH o-disubst.); elemental analysis
calcd (%) for C15H19N: C 84.46, H 8.98, N 6.57; found: C 84.81, H 9.27,
N 6.45.
N-n-Butyl-octylidene-amine/N-n-butyl-2-octylidene-2-amine (ratio 2.6:1)
(4b): According to the general procedure 1-octyne (2.4 mL, 16.0 mmol)
and n-butylamine (1.9 mL, 19.2 mmol) were treated in the presence of
10.0 mol% 1 (558 mg, 1.60 mmol) in toluene (6 mL) at 1208C for 24 h.
Fractional distillation afforded 4b as a colorless oil. GC yield: 48%; b.p.
44 458C/0.1 mbar. 1H NMR for main product (CDCl3, 400 MHz): d =
7.59 (t, J=5.0 Hz, 1H), 3.32 (t, J=6.9 Hz, 2H), 2.24 1.99 (m, 2H), 1.62
1.20 (m, 14H), 0.95 0.80 (m, 6H); 13C NMR for main product (CDCl3,
100 MHz): d = 164.8, 61.1, 35.8, 32.8, 31.7, 29.2, 29.0, 26.1, 22.6, 20.3,
14.0, 13.8; MS (EI, 70 eV): m/z (%): 183 (0.7) [M +], 182 (1.4) [M +ÀH],
168 (0.9) [M +ÀCH3], 154 (3) [M +ÀC2H5], 140 (12) [M +ÀC3H7], 112
(30), 99 (26), 84 (100) [CH=NC4H9+], 57 (40) [C4H9+], 56 (23), 41 (17)
[C3H5+]; FT IR (neat): n˜ = 1662 cmÀ1 (C=N); elemental analysis calcd
(%) for C12H25N: C 78.62, H 13.74, N 7.64; found: C 78.28, H 14.03, N
7.47.
1,2-Dimethyl-3-phenylindole (7a): According to the general procedure 3-
phenyl-1-propyne (0.43 mL, 3.5 mmol) and N-methyl-N-phenylhydrazine
(0.49 mL, 4.2 mmol) were treated in the presence of 5.0 mol% 1 (61 mg,
0.17 mmol) in toluene (2.5 mL) at 858C for 24 h. Then ZnCl2 (1.9 g,
14.0 mmol) was added. The residue was purified by column chromatogra-
phy (n-hexane/ethyl acetate 10:1) to afford 7a as colorless needles. M.p.
1078C (lit.[37] 113 1148C); GC yield: 84% (isolated yield: 519 mg
N-Benzyl-octylidene-amine/N-benzyl-2-octylidene-amine (ratio 4.6:1)
(4c): According to the general procedure 1-octyne (2.4 mL, 15.8 mmol)
and benzylamine (2.1 mL, 19.0 mmol) were treated in the presence of
10.0 mol% 1 (544 mg, 1.56 mmol) in toluene (6 mL) at 1208C for 24 h.
Fractional distillation afforded 4c as a colorless oil. GC yield: 46%; b.p.
97 988C/0.1 mbar. 1H NMR for main product (CDCl3, 400 MHz): d =
7.78 (t, J=5.0 Hz, 1H), 7.30 7.14 (m, 5H), 4.56 (s, 2H), 2.30 2.20 (m,
2H), 1.60 1.40 (m, 2H), 1.25 (m, 8H), 0.88 (t, 3H); 13C NMR for main
product (CDCl3, 100 MHz): d=166.3, 139.3, 128.4, 127.8, 126.8, 65.1,
35.9, 31.5, 29.4, 29.2, 26.0, 22.5, 14.0; MS (EI, 70 eV): m/z (%): 217 (1)
[M +], 202 (0.4) [M +ÀCH3], 188 (2) [M +ÀC2H5], 174 (3) [M +ÀC3H7],
160 (3) [M +ÀC4H9], 146 (20), 133 (83) [M +ÀC6H12], 132 (49) [M +
ÀC6H13], 91 (100) [PhCH2+]; FT IR (neat): n˜ = 1663 cmÀ1 (C=N).
N-(2-Hexylidene)-2,6-dimethylaniline (4d): According to the general
procedure 1-hexyne (1.5 mL, 12.9 mmol) and 2,6-dimethylaniline
(2.4 mL, 19.4 mmol) were treated in the presence of 3.0 mol% 1 (140 mg,
0.39 mmol) in toluene (5 mL) at 858C for 24 h. Fractional distillation af-
forded 4d as a colorless oil. GC yield: 94% (isolated yield: 1.4 g (54%));
b.p. 65 668C/0.1 mbar; 1H NMR (CDCl3, 400 MHz): d = 6.97 (d, J=
7.5 Hz, 2H), 6.83 (t, J=7.5 Hz, 1H), 2.44 (t, J=7.7 Hz, 2H), 1.97 (s, 6H),
1.69 (m, 2H), 1.59 (s, 3H), 1.44 (m, 2H), 0.97 (t, J=7.3 Hz, 3H); 13C
NMR (CDCl3, 100 MHz): d = 172.2, 149.2, 128.2, 126.3, 122.9, 41.2, 29.2,
23.1, 20.1, 18.3, 14.4; MS (EI, 70 eV): m/z (%): 203 (14) [M +], 188 (7)
[M +ÀCH3], 161 (14), 146 (100) [M +ÀC4H9], 121 (20), 105 (21), 77 (16)
[Ph+]; FT IR (neat): n˜ = 1664 cmÀ1 (C=N); elemental analysis calcd (%)
for C14H21N: C 82.70, H 10.41, N 6.89; found: C 82.62, H 10.59, N 6.85.
1
(67%)). H NMR (CDCl3, 400 MHz): d = 7.76 (d, J=7.9 Hz, 1H), 7.60
7.51 (m, 4H), 7.40 7.35 (m, 2H), 7.29 (m, 1H), 7.20 (m, 1H), 3.76 (s,
3H), 2.54 (s, 3H); 13C NMR (CDCl3, 100 MHz): d = 136.6, 135.8, 133.3,
129.6, 128.4, 126.9, 125.6, 121.1, 119.6, 118.6, 113.9, 108.7, 29.5, 11.0; MS
(EI, 70 eV): m/z (%): 221 (100) [M +], 204 (11), 144 (10); FT IR (KBr):
n˜ = 742 (ArH o-disubst.), 704, 772 cmÀ1 (ArH monosubst.).
1-Methyl-3-phenylindole (8a):[38] According to the general procedure
phenylacetylene (0.8 mL, 7.3 mmol) and N-methyl-N-phenylhydrazine
(1.0 mL, 8.7 mmol) were reacted in the presence of 5.0 mol% 1 (125 mg,
0.36 mmol) in toluene (5.5 mL) at 1008C for 24 h. Then ZnCl2 (2.9 g,
21.6 mmol) was added. The residue was purified by column chromatogra-
phy (n-hexane/ethyl acetate 10:1) to afford 8a as a green yellow oil. GC
yield: 42% (isolated yield: 378 mg, 25%). 1H NMR (CDCl3, 400 MHz):
d = 8.05 (d, J=7.9 Hz, 1H), 7.74 (m, 2H), 7.52 (m, 2H), 7.43 7.28 (m,
4H), 7.27 (s, 1H), 3.84 (s, 3H); 13C NMR (CDCl3, 100 MHz): d = 137.4,
135.6, 128.7, 127.2, 126.5, 126.1, 125.6, 121.9, 119.9, 119.8, 116.6, 109.5,
32.7; MS (EI, 70 eV): m/z (%): 207 (100) [M +], 192 (15) [M +ÀCH3],
165 (29), 104 (13); FT IR (KBr): n˜ = 744 (ArH o-disubst.), 698, 766 cmÀ1
(ArH monosubst.).
Together with 8a another product was isolated, which after comparison
with previously reported NMR data[39] could be characterized as 1-
methyl-2-phenylindole (8b): GC yield: 10% (isolated yield: 76 mg
(5%)). 1H NMR (CDCl3, 400 MHz): d = 7.74 (d, J=7.9 Hz, 1H), 7.63
7.22 (m, 8H), 6.67 (s, 1H), 3.81 (s, 3H); 13C NMR (CDCl3, 100 MHz): d
= 141.5, 138.3, 132.8, 129.3, 128.4, 127.9, 127.8, 121.6, 120.4, 119.8, 109.6,
101.6, 31.1; MS (EI, 70 eV): m/z (%): 207 (100) [M +], 165 (12), 104 (11),
102 (14).
2-(1,2-Dimethyl-indole-3-yl)ethylamine hydrochloride (9a)[40] and 2-(1,2-
dimethyl-indole-3-yl)ethylamine (9c):[41] According to the general proce-
dure 5-chloro-1-pentyne (0.15 mL, 1.4 mmol) and N-methyl-N-phenylhy-
drazine (0.21 mL, 1.7 mmol) were reacted in the presence of 10.0 mol%
1 (51 mg, 0.14 mmol) in toluene (4 mL) at 1008C for 24 h. During this
General procedure for the reaction of alkyneswith hydrazines : In an
Ace-pressure tube under an argon atmosphere a solution of the catalyst
in toluene was added to a mixture of alkyne and hydrazine. This mixture
was heated at the given temperature for the specified time (see Table 4).
Then 3 4 equiv of ZnCl2 were added. The pressure tube was heated
again at 1008C for 24 h. After filtration and removal of the solvent in
vacuo, the product was isolated by column chromatography.
2418
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420

Hydroamination of Terminal Alkynes
2409 2420
time the corresponding hydrochloride 9a precipitated. The mixture was
diluted with hexane (5 mL) and the precipitate was filtered. M.p. 224
2268C (lit.[40] 239 2408C). Yield 210 mg (64%). For the isolation of
amine 9c, the corresponding hydrochloride was dissolved in water
(20 mL). This solution was treated with CH2Cl2 (20 mL) and subsequent-
ly with NaOH (to pH 9). Both layers were separated and the aqueous
phase was washed twice with CH2Cl2 (10 mL). All organic phases were
combined and dried over MgSO4. After evaporation of the solvent the
free amine 9c was obtained as an colorless oil (159 mg, 58%). 1H NMR
(CDCl3, 400 MHz): d = 7.54 (d, J=7.7 Hz, 1H), 7.26 (d, J=8.1 Hz, 1H),
7.17 (m, 1H), 7.09 (m, 1H), 3.65 (s, 3H), 2.96 (m, 2H), 2.88 (m, 2H),
2.38 (s, 3H), 1.39 (brs, 2H); 13C NMR (CDCl3, 100 MHz): d = 136.5,
133.5, 127.7, 120.4, 118.6, 117.8, 108.4, 108.3, 42.7, 29.4, 28.6, 10.2; MS
(EI, 70 eV): m/z (%): 188 (22) [M +], 158 (100) [M +ÀCH2NH2], 143 (8);
FT IR (neat): n˜ = 3359, 3290, 1653 (NH2), 739 cmÀ1 (ArH o-disubst.).
Horino, J. Am. Chem. Soc. 2003, 125, 9560; c) P. N. O×Shaughnessy,
P. Scott, Tetrahedron: Asymmetry 2003, 14, 1979; d) G. A. Molander,
J. A. C. Romero, Chem. Rev. 2002, 102, 2161; e) S. Hong, T. J.
Marks, J. Am. Chem. Soc. 2002, 124, 7886; f) M. R. Douglass, M.
Ogasawara, S. Hong, M. V. Metz, T. J. Marks, Organometallics 2002,
21, 283; g) Y. K. Kim, T. Livinghouse, Angew. Chem. 2002, 114,
3797; Angew. Chem. Int. Ed. 2002, 41, 3645; h) Y. K. Kim, T. Living-
house, J. E. Bercaw, Tetrahedron Lett. 2001, 42, 2933; i) S. Tian,
V. M. Arredondo, C. L. Stern, T. J. Marks, Organometallics 1999, 18,
2568; j) E. L. Eliel, S. H. Wilen, L. N. Mander, Stereochemistry of
Organic Compounds, Wiley, New York, 1994, 682; k) M. A. Giar-
dell, V. P. Conticello, L. Brard, M. R. Gagnÿ, T. J. Marks, J. Am.
Chem. Soc. 1994, 116, 10241; l) M. R. Gagnÿ, C. L. Stern, T. J.
Marks, J. Am. Chem. Soc. 1992, 114, 275; m) M. R. Gagnÿ, T. J.
Marks, J. Am. Chem. Soc. 1989, 111, 4108.
[6] a) For hydroamination catalyzed by late transition metals, see R.
Dorta, P. Egli, F. Z¸rcher, A. Togni, J. Am. Chem. Soc. 1997, 119,
10857; b) M. Beller, M. Eichberger, H. Trauthwein, Angew. Chem.
1997, 109, 2306; Angew. Chem. Int. Ed. Engl. 1997, 36, 2225; c) J.-J.
Brunet, G. Commenges, D. Neibecker, K. Philippot, J. Organomet.
Chem. 1994, 469, 221; d) J.-J. Brunet, D. Neibecker, K. Philippot,
Tetrahedron Lett. 1993, 34, 3877; e) J.-J. Brunet, D. Neibecker, K.
Philippot, J. Chem. Soc. Chem. Commun. 1992, 1215; f) D. Selent,
D. Scharfenberg-Pfeiffer, G. Reck, R. Taube, J. Organomet. Chem.
1991, 415, 417; g) A. L. Casalnuovo, J. L. Calabrese, D. Milstein, J.
Am. Chem. Soc. 1988, 110, 6738; h) D. Steinborn, R. Taube, Z.
Chem. 1986, 26, 349; i) L. S. Hegedus, J. M. McKearin, J. Am. Chem.
Soc. 1982, 104, 2444; j) B. Pugin, L. M. Venanzi, J. Organomet.
Chem. 1981, 214, 125, and references therein; k) D. R. Coulson, Tet-
rahedron Lett. 1971, 12, 429.
Acknowledgements
This work has been supported by the State of Mecklenburg-Western
Pommerania, the ™Fonds der Chemischen Industrie∫ and the ™Bundesmi-
nisterium f¸r Bildung und Forschung∫ (BMBF). Mrs. C. Mewes (IfOK) is
particularly thanked for her excellent technical support. We thank Prof.
Dr. M. Michalik, Dr. W. Baumann, Dr. C. Fischer, Mrs. H. Baudisch and
Mrs. S. Buchholz (all IfOK) for analytical support. We also thank Prof.
Dr. U. Rosenthal (IfOK) for general discussions.
[1] For general references, see: a) J. March, Advanced Organic Chemis-
try, 4th ed., Wiley, New York, 1992, pp. 768, and references therein;
b) J. P. Collman, B. M. Trost, T. R. Veroeven, in Comprehensive Or-
ganometallic Chemistry, Vol. 8 (Eds.: G. Wilkinson, F. G. A. Stone),
Pergamon Press, Oxford, 1982, pp. 892, and references therein;
c) M. S. Gibson, in The Chemistry of Amino Group (Ed.: S. Patai),
Wiley, New York, 1968, pp. 61.
[2] For leading reviews on hydroamination, see: a) F. Pohlki, S. Doye,
Chem. Soc. Rev. 2003, 32, 104; b) I. Bytschkov, S. Doye, Eur. J. Org.
Chem. 2003, 935; c) M. Beller, C. Breindl, M. Eichberger, C. G. Har-
tung, J. Seayad, O. R. Thiel, A. Tillack, H. Trauthwein, Synlett 2002,
1579; d) M. Nobis, B. Drie˚en-Hˆlscher, Angew. Chem. 2001, 113,
4105; Angew. Chem. Int. Ed. 2001, 40, 3983; e) J.-J. Brunet, D. Nei-
becker, in Catalytic Heterofunctionalization, (Eds.: A. Togni, H.
Gr¸tzmacher), Wiley-VCH, Weinheim, 2001, p. 91; f) T. E. M¸ller,
M. Beller, Chem. Rev. 1998, 98, 675; g) J. J. Brunet, D. Neibecker, F.
Niedercorn, J. Mol. Catal. 1989, 49, 235.
[3] For alkalimetal and base catalyzed hydroamination, see: a) A. Ates,
C. Quinet, Eur. J. Org. Chem. 2003, 1623; b) B. M. Trost, W. Tang, J.
Am. Chem. Soc. 2003, 125, 8744; c) J. Seayad, A. Tillack, C. G. Har-
tung, M. Beller, Adv. Synth. Catal. 2002, 344, 795; d) M. Beller, C.
Breindl, Chemosphere 2001, 43, 21; e) C. G. Hartung, C. Breindl, A.
Tillack, M. Beller, Tetrahedron 2000, 56, 5157; f) M. Beller, C.
Breindl, T. H. Riermeier, M. Eichberger, H. Trauthwein, Angew.
Chem. 1998, 110, 3571; Angew. Chem. Int. Ed. 1998, 37, 3389; g) M.
Beller, C. Breindl, Tetrahedron 1998, 54, 6359; h) D. Steinborn, B.
Dies, I. Wagner, R. Taube, Z. Chem. 1989, 29, 333; i) G. P. Pez, J. E.
Galle, Pure Appl. Chem. 1985, 57, 1917; j) J. Wollensak, R. D. Clos-
son, Org. Synth. 1963, 43, 45; k) R. Stroh, J. Ebersberger, H. Haber-
land, W. Hahn, Angew. Chem. 1957, 69, 124; l) B. W. Howk, E. L.
Little, S. L. Scott, G. M. Whitman, J. Am. Chem. Soc. 1954, 76, 1899.
[4] For acid-catalyzed hydroamination, see: a) M. Lequitte, F. Figueras,
C. Moreau, S. Hub, J. Catal. 1996, 163, 255; b) A. Cermona, A.
Corma, M. Iglesias, A. San Josÿ, F. Sµnchez, J. Organomet. Chem.
1995, 492, 11; c) N. Mizuno, M. Tabata, T. Uematsu, M. Iwamoto, J.
Catal. 1994, 146, 249; d) N. Mizuno, M. Tabata, T. Uematsu, M. Iwa-
moto, J. Chem. Soc. Faraday Trans. 1 1993, 89, 3513; e) P. Fink, J.
Datka, J. Chem. Soc. Faraday Trans. 1 1989, 85, 3079; f) M. Deeba,
M. E. Ford, T. A. Johnson, J. Chem. Soc. Chem. Commun. 1987, 562.
[5] For hydroamination mediated by early transition metals and lantha-
noid complexes, see: a) D. V. Gribkov, K. C. Hultzsch, F. Hampel,
Chem. Eur. J. 2003, 9, 4796; b) Y. K. Kim, T. Livinghouse, Y.
[7] a) M. Utsunomiya, R. Kuwano, M. Kawatsura, J. F. Hartwig, J. Am.
Chem. Soc. 2003, 125, 5608; b) J.-J. Brunet, N. C. Chu, O. Diallo, E.
Mothes, J. Mol. Catal. A 2003, 198, 107; c) K. Li, P. N. Horton, M. B.
Hursthouse, K. K. Hii, J. Organomet. Chem. 2003, 665, 250; d) U.
Nettekoven, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124, 1166; e) X.
Cheng, K. K. Hii, Tetrahedron 2001, 57, 5445; f) M. Kawatsura, J. F.
Hartwig, J. Am. Chem. Soc. 2000, 122, 9546; g) H. M. Senn, P. E.
Blˆchl, A. Togni, J. Am. Chem. Soc. 2000, 122, 4098; h) A. Tillack,
H. Trauthwein, C. G. Hartung, M. Eichberger, S. Pitter, A. Jansen,
M. Beller, Monatsh. Chem. 2000, 131, 1327; i) M. Beller, O. R. Thiel,
H. Trauthwein, C. G. Hartung, Chem. Eur. J. 2000, 6, 2513.
[8] a) Y. Li, T. J. Marks, J. Am. Chem. Soc. 1996, 118, 9295; b) J. B.
Pedley, R. D. Naylor, S. P. Kirby, Thermochemical Data of Organic
Compounds, Chapman and Hall, London, 1986, AppendixTables 1,
3.
[9] a) S. Burling, L. D. Field, H. L. Li, B. A. Messerle, P. Turner, Eur. J.
Inorg. Chem. 2003, 3179; b) C. Li, R. K. Thomson, B. Gillon, B. O.
Patrick, L. L. Schafer, Chem. Commun. 2003, 2462; c) J. Barluenga,
M. Trincado, E. Rubio, J. M. Gonzµlez, Angew. Chem. 2003, 115,
2508; Angew. Chem. Int. Ed. 2003, 42, 2406; d) C. Koradin, W.
Dohle, A. L. Rodriguez, B. Schmid, P. Knochel, Tetrahedron 2003,
59, 1571; e) L. Ackermann, R. G. Bergman, Org. Lett. 2002, 4, 1475;
f) I. Bytschkov, S. Doye, Tetrahedron Lett. 2002, 43, 3715; g) V. Neff,
T. E. M¸ller, J. A. Lercher, Chem. Commun. 2002, 906; h) J. Pen-
zien, R. Q. Su, T. S. M¸ller, J. Mol. Catal. A 2002, 182 183, 489; i) T.
Kondo, T. Okada, T. Suzuki, T. Mitsudo, J. Organomet. Chem. 2001,
622, 149, and references therein; j) M. Tokunaga, M. Ota, M.-A.
Haga, Y. Wakatsuki, Tetrahedron Lett. 2001, 42, 3865; k) T. E.
M¸ller, M. Grosche, E. Herdtweck, A.-K. Pleier, E. Walter, Y.-K.
Yan, Organometallics 2000, 19, 170; l) J. Penzien, T. E. M¸ller, J. A.
Lercher, Chem. Commun. 2000, 1753; m) S. Burling, L. D. Field,
B. A. Messerle, Organometallics 2000, 19, 87; n) T. E. M¸ller, A.-K.
Pleier, J. Chem. Soc. Dalton Trans. 1999, 583; o) D. Duncan, T. Liv-
inghouse, Organometallics 1999, 18, 4421; p) M. R. B¸rgstein, H.
Berberich, P. W. Roesky, Organometallics 1998, 17, 1452; q) L. S.
Hegedus, Angew. Chem. 1988, 100, 1147; Angew. Chem. Int. Ed.
Engl. 1988, 27, 1113.
[10] D. Tzalis, C. Koradin, P. Knochel, Tetrahedron Lett. 1999, 40, 6193.
[11] a) Y. Li, T. J. Marks, Organometallics 1996, 15, 3370; b) Y. Li, T. J.
Marks, J. Am. Chem. Soc. 1998, 120, 1757.
[12] a) J. Wang, A. K. Dash, M. Kapon, J.-C. Berthet, M. Ephritikhine,
M. S. Eisen, Chem. Eur. J. 2002, 8, 5384; b) T. Straub, A. Haskel,
2419
Chem. Eur. J. 2004, 10, 2409 2420
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

M. Beller et al.
FULL PAPER
T. G. Neyroud, M. Kapon, M. Botoshansky, M. S. Eisen, Organome-
tallics 2001, 20, 5017; c) M. S. Eisen, T. Straub, Haskel, A. J.
Alloys Compd. 1998, 271 273, 116; d) A. Haskel, T. Straub, M. S.
Eisen, Organometallics 1996, 15, 3773.
[29] a) F. Pohlki, S. Doye, Angew. Chem. 2001, 113, 2361; Angew. Chem.
Int. Ed. 2001, 40, 2305; b) B. F. Straub, R. G. Bergman, Angew.
Chem. 2001, 113, 4768; Angew. Chem. Int. Ed. 2001, 40, 4632.
[30] In addition to [(h5-C5H5)Ti(=NR)(NHR)], we have calculated the
reactions with (h5-C5H5)2Ti(=NR) as the active catalyst (R=
C(CH3)3, C6H5). However, no energetic difference is found between
the Markovnikov and anti-Markovnikov p-complexes using [(h5-
C5H5)2Ti(=NR)(HCꢀCCH3)]. In addition, the ligand exchange reac-
tion between [(h5-C5H5)2Ti(=NR)(HCꢀCCH3)] and NH2R to give
[(h5-C5H5)Ti(=NR)(NHR)(HCꢀCCH3)] and C5H6 is exothermic by
15.4 kcalmolÀ1 for R=C6H5, and by 12.8 kcalmolÀ1 for R=
C(CH3)3, and therefore the release of C5H6 is favored energetically.
This result agrees with the experimental and theoretical finding by
Bergman (ref. [15]) for a release of C5H6 under catalytic conditions.
Hence, the active catalyst under our conditions should be [(h5-
C5H5)Ti(=NR)(NHR)] rather than [(h5-C5H5)2Ti(=NR)].
[31] Gaussian 98, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuse-
ria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Mont-
gomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam,
A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Mo-
rokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W.
Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian, Inc., Pittsburgh PA, 1998.
[32] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299; b) T. H. Dun-
ning, Jr., P. J. Hay, Modern Theoretical Chemistry (Ed.: H. F. Schae-
fer, III), Plenum, New York, 1976, pp. 1; c) for polarization functions
see: S. Huzinaga, J. Anzelm, M. Klobukowski, E. Radzio-Andzelm,
Y. Sakai, H. Tatewaki, Gaussian Basis Sets for Molecular Calcula-
tions, Elsevier, Amsterdam, 1984.
[33] J. B. Foresman, ·. Frisch, Exploring Chemistry with Electronic
Structure Methods: A Guide to Using Gaussian, 2nd ed., Gaussian,
Inc., Pittsburgh PA, 1996.
[34] E. D. Jemmis, A. K. Phukan, J. Jiao, U. Rosenthal, Organometallics
2003, 22, 1958.
[35] a) NBO Program, 3.1, E. D. Glendening, A. E. Reed, J. E. Carpen-
ter, F. Weinhold; b) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem.
Rev. 1988, 88, 899.
[13] a) E. Mizushima, T. Hayashi, M. Tanaka, Org. Lett. 2003, 5, 3349;
b) T. Shimada, Y. Yamamoto, J. Am. Chem. Soc. 2002, 124, 12670;
c) C. G Hartung, H. Trauthwein, A. Tillack, M. Beller, J. Org.
Chem. 2001, 66, 6339; d) M. Tokunaga, Y. Wakatsuki, J. Synth. Org.
Chem. Jpn. 2000, 58, 587; e) M. Tokunaga, M. Eckert, Y. Wakatsuki,
Angew. Chem. 1999, 111, 3416; Angew. Chem. Int. Ed. 1999, 38,
3222; f) Y. Uchimaru, Chem. Commun. 1999, 1133; g) I. Kadota, A.
Shibuya, L. M. Lutete, Y. Yamamoto, J. Org. Chem. 1999, 64, 4570;
h) J. Barluenga, A. Aznar, R. Liz, R. Rodes, J. Chem. Soc. Perkin
Trans. 1 1980, 2732; i) J. Barluenga, A. Aznar, Synthesis 1977, 195;
j) J. Barluenga, A. Aznar, Synthesis 1975, 704.
[14] a) A. M. Baranger, P. J. Walsh, R. G. Bergman, J. Am. Chem. Soc.
1993, 115, 2753; b) P. J. Walsh, A. M. Baranger, R. G. Bergman, J.
Am. Chem. Soc. 1992, 114, 1708.
[15] J. S. Johnson, R. G. Bergman, J. Am. Chem. Soc. 2001, 123, 2923.
[16] a) F. Pohlki, A. Heutling, I. Bytschkov, T. Hotopp, S. Doye, Synlett
2002, 799; b) I. Bytschkov, S. Doye, Eur. J. Org. Chem. 2001, 4411;
c) E. Haak, H. Siebeneicher, S. Doye, Org. Lett. 2000, 2, 1935; d) E.
Haak, I. Bytschkov, S. Doye, Angew. Chem. 1999, 111, 3584; Angew.
Chem. Int. Ed. 1999, 38, 3389.
[17] a) Y. Shi, C. Hall, J. T. Ciszewski, C. Cao, A. L. Odom, Chem.
Commun. 2003, 586; b) Y. Shi, J. T. Ciszewski, A. L. Odom, Organo-
metallics 2001, 20, 3967; c) C. Cao, J. T. Ciszewski, A. L. Odom, Or-
ganometallics 2001, 20, 5011.
[18] T.-G. Ong, G. P. A. Yap, D. S. Richeson, Organometallics 2002, 21,
2839.
[19] a) H. Trauthwein, A. Tillack, M. Beller, Chem. Commun. 1999,
2029; b) M. Beller, O. R. Thiel, H. Trauthwein, Synlett 1999, 243;
c) M. Beller, H. Trauthwein, M. Eichberger, C. Breindl, T. E.
M¸ller, Eur. J. Inorg. Chem. 1999, 1121; d) M. Beller, H. Trauth-
wein, M. Eichberger, C. Breindl, J. Herwig, T. E. M¸ller, O. R.
Thiel, Chem. Eur. J. 1999, 5, 1306; e) M. Beller, H. Trauthwein, M.
Eichberger, C. Breindl, A. Zapf, T. E. M¸ller, J. Organomet. Chem.
1998, 566, 277.
[20] A. Tillack, I. Garcia Castro, C. G. Hartung, M. Beller, Angew.
Chem. 2002, 114, 2646; Angew. Chem. Int. Ed. 2002, 41, 2541.
[21] For reviews, see: a) U. Rosenthal, V. V. Burlakov, P. Arndt, W. Bau-
mann, A. Spannenberg, Organometallics 2003, 22, 884; b) U. Rosen-
thal, V. V. Burlakov, in Titanium and Zirconium in Organic Synthesis
(Ed.: I. Marek), Wiley-VCH, Weinheim, 2002, p. 355. The Rosenthal
catalysts are now commercially available from Fluka.
[22] V. V. Burlakov, A. V. Polyakov, A. I. Yanovsky, Y. T. Struchkov,
V. B. Shur, M. E. Vol×pin, U. Rosenthal, H. Gˆrls, J. Organomet.
Chem. 1994, 476, 197.
[23] For a recent review on selective Markovnikov and anti-Markovni-
kov functionalization reactions, see: J. Seayad, A. Tillack, H. Jiao,
M. Beller, Angew. Chem. 2004, 116, in press; Angew. Chem. Int. Ed.
2004, 43, in press.
[36] R. Jaouhari, C. Filliatre, B. Maillard, J. J. Villenave, Tetrahedron
1982, 38, 3137.
[37] a) P. E. Verkade, E. F. Janetzky, Recl. Trav. Chim. Pays-Bas 1943, 62,
775; b) C. Berti, L. Greci, M. Poloni, J. Chem. Soc. Perkin Trans. 1
1981, 1610.
[38] a) J. Rodriguez, A. Lafuente, P. Garcia-Almaraz, J. Heterocycl.
Chem. 2000, 37, 1281; b) A. Couture, E. Deniau, Y. Gimbert, P.
Grandclaudon, J. Chem. Soc. Perkin Trans. 1 1993, 2463; c) T.
Nishio, J. Org. Chem. 1988, 53, 1323.
[39] M. Ishikura, I. Agata, N. Katagiri, J. Heterocycl. Chem. 1999, 36,
873.
[24] A. Heutling, S. Doye, J. Org. Chem. 2002, 67, 1961.
[40] R. V. Heinzelman, J. Szmuszkovicz, Prog. Drug Res. 1963, 75, 6;
Chem. Abstr. 1964, 60, 13215b.
[41] I. I. Grandberg, N. I. Afonina, T. I. Zuyanova, Khim. Geterosikl.
Soedin. 1968, 1038.
œ
[25] V. Varga, K. Mach, M. Polasek, P. Sedmera, J. Hiller, U. Thewalt,
S. I. Troyanov, J. Organomet. Chem. 1996, 506, 241.
[26] J.-L. Jiang, Z. Shi, Synth. Commun. 1998, 28, 4137.
[27] C. Cao, Y. Shi, A. L. Odom, Org. Lett. 2002, 4, 2853.
[28] a) B. Robinson, The Fischer Indole Synthesis, Wiley, Chichester,
1982; b) D. L. Hughes, Org. Prep. Proced. Int. 1993, 25, 607.
Received: October 10, 2003 [F5674]
Published online: March 25, 2004
2420
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 2409 2420