Titanium/palladium-mediated regioselective propargylation of ketones using propargylic carbonates as pronucleophiles

Article abstract of DOI:10.1002/adsc.201000655

An efficient protocol for the synthesis of homopropargylic alcohols using propargylic carbonates as pronucleophiles is reported. The reaction is based on a combination of transition metal (palladium) and radical chemistry (titanium). The reaction takes pl

Full text of DOI:10.1002/adsc.201000655

COMMUNICATIONS
DOI: 10.1002/adsc.201000655
Titanium/Palladium-Mediated Regioselective Propargylation of
Ketones using Propargylic Carbonates as Pronucleophiles
Alba Millꢀn,a Luis ꢁlvarez de Cienfuegos,a,* Ana Martꢂn-Lasanta,a
Araceli G. CampaÇa,a and Juan M. Cuervaa,*
a
Department of Organic Chemistry, Faculty of Sciences, University of Granada, Campus Fuentenueva s/n,
E-18071 Granada, Spain
Fax : (+34)-958-248437; e-mail: lac@ugr.es or jmcuerva@ugr.es
Received: August 22, 2010; Published online: December 22, 2010
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.201000655.
Abstract: An efficient protocol for the synthesis of
homopropargylic alcohols using propargylic carbo-
nates as pronucleophiles is reported. The reaction is
based on a combination of transition metal (palla-
dium) and radical chemistry (titanium). The reac-
tion takes place with an excellent regioselectivity
and tolerates a great degree of substitution of the
starting propargylic carbonate, thus being an inter-
esting tool in the context of synthetic organic
chemistry.
species some Barbier-type protocols based on Cr,[4]
Sm,[5] Ce,[6] Mg,[7] Zn,[8] Cd,[9] Sn,[10] or In[11] and acti-
vated starting materials, such as propargylic halides,
have been developed. It is worth noting that the re-
ported regioselectivity of these protocols is generally
poor. In this context, we have recently described a
Cp2TiCl-catalyzed regioselective propargylation of al-
dehydes and ketones using propargyl bromide and
chloride as pronucleophiles, yielding exclusively the
homopropagylic alcohol.[12] In spite of its interest, this
titanoceneACTHNUGRTNEUNG(III)-based procedure also requires activat-
Keywords: homogeneous catalysis; propargylation;
radical reactions; synthetic methods; titanium
ed starting materials, which often are cumbersome in
their introduction and manipulation, especially during
the final steps of a synthetic sequence. Therefore, the
extension of this reaction to other more stable pronu-
cleophiles while maintaining the exceptional regiose-
lectivity would be highly valuable.
In recent years alkynes have become versatile build-
Propargylic carbonates are readily available com-
ing blocks in organic synthesis. Thus, for example, pounds, which might be used instead of the corre-
many interesting transformations have been devel- sponding propargylic halides. Nevertheless, they are
oped based on the reactivity of alkynes.[1] Moreover, inert in the classical Barbier conditions, although
in the context of the synthesis of complex natural some remarkable examples have been described, their
products their presence is required in the key steps.[2] scope being usually limited to aldehydes as electro-
Therefore, reactions able to efficiently introduce this philes.[13,14] In this case, the main limitation is the in-
functional group in a given structure are indispensa- ability of Cp2TiCl to promote the homolytic breakage
[15]
À
ble. In this sense, a straightforward procedure would of C O bonds. To avoid this fact we have recently
be the direct propargylation of aldehydes and ketones developed a new procedure for the versatile allylation
from suitable and stable starting materials. Although of carbonyl compounds based on a combination of
these transformations are apparently simple, a careful Ti/Pd complexes.[16] Under these reaction conditions
inspection of the described protocols shows that they readily available and stable allylic carbonates could
have been scarcely reported.
be used as starting materials.
Methods for the propargylation of carbonyl com-
Herein, we want to communicate a novel and mild
pounds using propargylic organometallic species exist method of propargylation mediated by a combination
in literature.[3] Nevertheless, most of them have some of Ti and Pd complexes (Scheme 1). This method uses
drawbacks derived from their high reactivity, complex propargylic ethyl carbonates as starting materials and
preparation and/or poor regioselectivity, usually yield- works with terminal and internal alkynes with differ-
ing mixtures of homopropargylic and allenylic alco- ent substitution patterns, giving in moderate to good
hols. To circumvent the prior preparation of these yields the corresponding coupling products with com-
Adv. Synth. Catal. 2011, 353, 73 – 78
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
73

Alba Millꢀn et al.
COMMUNICATIONS
tries 1-6, 11, 12, and 16, 17) alkynes are suitable sub-
strates for this transformation. Phenyl (entry 11) and
trimethylsilyl (entry 12) groups are also allowed in
this reaction. It is also especially remarkable that re-
tention of the regiochemistry of the starting alkyne is
achieved in entries 13–17. It is also worth noting that
the addition of a-substituted propargyl derivatives to
tetralone gives only one stereoisomer (entries 14 and
16).[21] The only exception to the regioselectivity was
observed when the reaction was carried out intramo-
lecularly (entry 21). In that case, the conformational
restrictions favor the formation of the corresponding
six-membered carbocycle. Aldehydes are substrates of
Scheme 1. Ti/Pd-mediated regioselective propargylation of
ketones.
plete regioselectivity toward the homopropargylic al-
cohol. This work represents an improvement of our
original procedure, especially when ketones are used
as electrophiles. The success of the multimetallic pro-
tocol seems to be based on the in situ synthesis of al-
lenyl titanocene complexes II via transient propargyl-
ic radicals of type I.
Initially, we carried out the addition of 2-decynyl
ethyl carbonate to 2-decanone in the presence of dif-
ferent palladium complexes (0.1 mol%) and in situ
generated Cp2TiCl.[17,18] Despite the number of pal-
A
pre-catalysts
(COD)Cl2) and phosphines/phos-
(o-tol)3, (C6F5)3,
[Pd
A
PdACTHNGUTRENU(NG dba)2,
A
ACHTUNGTRENNUNG
PACHTUNGTRENNUNG
P
G
P
E
P
N
a
simple 1:2 mixture of palladium dichloride and PPh3
gave the best result (71%, Table 1, entry 1). The ho-
mopropargylic alcohol was the only isolated product,
pointing out the remarkable regioselectivity of the
process. Moreover, the amount of Cp2TiCl could be
diminished until only 40 mol% obtaining a similar
yield (67%, Table 1, entry 1).[19] In this case a mixture
2,4,6-collidine (7 mmol), trimethylsilyl chloride
(4 mmol) and Mn dust (8 mmol) is required to regen-
erate the titaniumACTHNUTRGNEUNG
(III) species.[16,20] With these encour-
aging results we submitted other carbonyl compounds
and other propargyl derivatives to these reaction con-
ditions to establish the scope of the reaction. The re-
sults are summarized in Table 1.
This reaction allows the efficient regioselective
propargylation of different ketones, having different
electronic and steric profiles, at room temperature
with moderate to good yields. Interestingly, the yields
using substoichiometric amounts of titanium catalyst
are similar or even better than those corresponding to
the stoichiometric conditions. Thus, for example, the
reaction takes place with aromatic (entries 3, 4, 8, 14
and 16), cyclic (entries 5, 6, 9, 10, 15, and 17) and acy-
clic (entries 1, 2, 7, 11, 12 and 13) ketones. Moreover,
different substitution is also tolerated in the pronu-
Scheme 2. Proposed mechanism for the Ti/Pd-mediated
cleophile. Terminal (entries 7-10) and internal (en- propargylation of carbonyl compounds.
74
asc.wiley-vch.de
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 73 – 78

Titanium/Palladium-Mediated Regioselective Propargylation of Ketones
Table 1. Ti/Pd-mediated propargylation of aldehydes and ketones.
Entry
Propargyl carbonate
Ketone
Product[a]
(Protocol,[b] Yield [%])
1
2
3
R3 =heptyl, R4 =H
R3 =heptyl, R4 =H
R3 =heptyl, R4 =H
2-decanone
3-pentanone
2-tetralone
A, 71; B, 67
A, 88; B, 54
A,75
4
5
R3 =heptyl, R4 =H
R3 =heptyl, R4 =H
cyclopropyl phenyl ketone
B, 62
tert-butylcyclohexanone
A, 65;[c] B, 94[c]
6
7
8
R3 =heptyl, R4 =H
R3 =H, R4 =H
R3 =H, R4 =H
cyclobutanone
2-decanone
2-tetralone
A 64; B, 85
A, 53;[d] B, 66
A, 53; B, 68
9
R3 =H, R4 =H
R3 =H, R4 =H
tert-butylcyclohexanone
A, 66;[e] B, 92[e]
B, 77
10
adamantanone
11
12
13
R3 =Ph, R4 =H
R3 =SiMe3, R4 =H
R3 =H, R4 =Me
2-decanone
3-pentanone
2-decanone
A, 54
A, 57
B, 93[f]
14
15
16
17
R3 =H, R4 =Me
R3 =H, R4 =Me
R3 =Et, R4 =Me
R3 =Et, R4 =Me
2-tetralone
A, 44
tert-butylcyclohexanone
2-tetralone
A, 72;[e] B, 61[e]
A, 71; B 76
A, 99;[e] B, 90[e]
tert-butylcyclohexanone
Adv. Synth. Catal. 2011, 353, 73 – 78
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
75

Alba Millꢀn et al.
COMMUNICATIONS
Table 1. (Continued)
Entry
Propargyl carbonate
Ketone
decanal
Product[a]
(Protocol,[b] Yield [%])
A, 59; B, 56
18
R3 =H, R4 =H
19
20
R3 =H, R4 =H
R3 =H, R4 =H
citronelal
A, 51;[f] B, 67[f]
phenylpropanal
B, 54
21
A, 70; B 60
[a]
All products are racemic and only relative stereochemistry is shown in some cases.
[b]
Protocol A: carbonyl compound (1 mmol), propargyl carbonate (4 mmol), Cp2TiCl2 (2 mmol), PdCl2 (0.1 mmol), PPh3
(0.2 mmol), Mn (8 mmol) in THF at room temperature for 48 h. Protocol B: carbonyl compound (1 mmol), propargyl car-
bonate (4 mmol), Cp2TiCl2 (0.4 mmol), PdCl2 (0.2 mol), PPh3 (0.4 mmol), Mn (8 mmol), Me3SiCl (4 mmol), and 2,4,6-col-
lidine (7 mmol) in THF at room temperature for 48 h.
55:45 mixture of axial:equatorial alcohols.
PCy3 was used instead of PPh3.
[c]
[d]
[e]
[f]
A 45:55 mixture of axial:equatorial alcohols.
A 1:1 mixture of syn:anti stereoisomers.
this reaction (entries 18–20). Nevertheless, moderate carbonate, yielding an initial palladium(II) transient
yields are usually obtained owing to a side reaction complex. A subsequent single-electron transfer (SET)
that we have commonly observed. Cp2TiCl is able to reaction mediated by Cp2TiCl provokes a fragmenta-
transfer slowly a cyclopentadienyl group to the alde- tion of the generated palladium(I) intermediate,
hyde yielding the corresponding Grignard-type alco- yielding propargylic radicals of type I and a Pd(0)
hol. Therefore, the titanium catalyst is decomposed in complex, closing the first catalytic cycle.[23] These radi-
the reaction and/or is consuming the starting alde- cals are trapped in the reaction media yielding the
hyde.
corresponding allenyltitanocene(IV) intermediates
Concerning the mechanism, ketyl radicals do not (II), the presence of which is hypothesized to explain
seem to be involved in this transformation based on the complete regiochemistry of the addition reac-
the retention of the cyclopropyl group in entry 4. On tion.[24] Finally, the corresponding homopropargylic al-
the other hand, it is known that allenyltitanocene de- cohols are obtained after a subsequent nucleophilic
rivatives add to carbonyl compounds to yield homo- addition and an acidic work-up. In protocol B a
propargylic alcohols with good regioselectivities titanoceneACTHUNRTGNE(GNU III)-regenerating agent, the combination
toward the homopropargylic alcohol.[3,14,22] Moreover, 2,4,6-collidine, trimethylsilyl chloride and Mn dust,
the regioselectivity observed in the present work is was added to close the second catalytic cycle.
similar to those obtained in our previously reported
In summary, we have developed a general Ti/Pd-
protocol based only in Cp2TiCl. These two data sets mediated regioselective propargylation of aldehydes
strongly suggest that titanocene intermediates are in- and ketones using stable propargylic carbonates as
volved in this transformation, probably allenyltitano- pronucleophiles, which takes place at room tempera-
cenes. This hypothesis is compatible with the fact that ture under mild conditions. Regardless of its interest
no reaction is observed in the absence of titanocene. for organic synthesis, this reaction has scarce prece-
The role of palladium catalyst seems to be restricted dents in literature. Interestingly, the reaction can be
to the initial activation of the starting propargyl car- carried out using substoichimetric amounts of both Ti
bonate by a known oxidative addition. Therefore, we and Pd precatalysts, retaining in most of the cases the
propose the following mechanism for the transforma- yields and the regio- and stereochemical profile.
tion (Scheme 2).
Future work will be directed to the development of
The process begins with the reaction between the an enantioselective version of this reaction using
palladium catalyst and the corresponding propargyl chiral titanocene complexes.
76
asc.wiley-vch.de
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 73 – 78

Titanium/Palladium-Mediated Regioselective Propargylation of Ketones
Soc. 2010, 132, 6638–6639; d) D. R. Fandrick, K. R.
Experimental Section
Fandrick, J. T. Reeves, Z. Tan, W. Tang, A. G. Capacci,
S. Rodriguez, J. J. Song, H. Lee, N. K. Yee, C. H. Sen-
ayake, J. Am. Chem. Soc. 2010, 132, 7600–7601.
[4] a) A. Fürstner, Chem. Rev. 1999, 99, 991–1046; b) P.
Place, C. Verniere, J. Gore, Tetrahedron 1981, 37,
1359–1367; c) K. Namba, Y. Kishi, Org. Lett. 2004, 6,
5031–5033; d) M. Inoue, M. Nakada, Org. Lett. 2004,
6, 2977–2980.
[5] P. Girard, J. L. Namy, H. B. Kagan, J. Am. Chem. Soc.
1980, 102, 2693–2698.
[6] T. Imamoto, T. Kusumoto, Y. Tawarayama, Y. Sugiura,
T. Mita, Y. Hatanaka, M. Yokohama, J. Org. Chem.
1984, 49, 3904–3912.
[7] R. Baker, M. A. Brimble, Tetrahedron Lett. 1986, 27,
3311–3314.
[8] a) C. Chen, D. J. Crich, J. Chem. Soc. Chem. Commun.
1991, 1289–1290; b) L. W. Bieber, M. F. da Silva, R. C.
da Costa, L. O. S. Silva, Tetrahedron Lett. 1998, 39,
3655–3658.
[9] B. Alcaide, P. Almendros, C. Aragoncillo, R. Rodrꢂ-
guez-Acebes, J. Org. Chem. 2001, 66, 5208–5216.
[10] M. Banarjee, S. Roy, Chem. Commun. 2003, 534–535.
[11] a) S. Singh, S. Kumar, S. S. Chimni, Tetrahedron: Asym-
metry 2002, 13, 2679–2687; b) T.-P. Loh, M.-J. Lin, K.-
L. Tan, Tetrahedron Lett. 2003, 44, 507–509; c) W.
Miao, W. Lu, T. H. Chan, J. Am. Chem. Soc. 2003, 125,
2412–2413; d) J. Park, S. Kim, P. H. Lee, Org. Lett.
2008, 10, 5067–5070.
General Procedure for Pd(0)–TiACTHNUTRGNE(NUG III)-Mediated
Propargylation of Carbonyl Compounds with
Propargyl Carbonates
Protocol A: Rigorously deoxygenated THF[26] (20 mL) was
added to
a mixture of Cp2TiCl2 (2.0 mmol), PdCl2
(0.1 mmol), PPh3 (0.2 mmol) and Mn dust (8.0 mmol) under
an argon atmosphere and the suspension was stirred at
room temperature until it turned dark green (about 15 min).
A solution of carbonyl compound (1.0 mmol) and propargyl
carbonate (4.0 mmol) in THF (2 mL) was then added. The
mixture was stirred for 48 h and then diluted with EtOAc,
washed with brine, dried over anhydrous NaSO4 and the sol-
vent removed. The residue was submitted to flash chroma-
tography (EtOAc/hexane mixtures) to give the correspond-
ing products.
Protocol B: Rigorously deoxygenated THF[26] (20 mL)
was added to a mixture of Cp2TiCl2 (0.4 mmol), PdCl2
(0.2 mmol), PPh3 (0.4 mmol) and Mn dust (8.0 mmol) under
an argon atmosphere, and the suspension was stirred at
room temperature until it turned dark green (about 15 min).
A solution of carbonyl compound (1.0 mmol), propargyl car-
bonate (4.0 mmol), 2,4,6-collidine (7.0 mmol) and Me3SiCl
(4.0 mmol) in THF (2 mL) was then added. The mixture
was stirred for 48 h and then diluted with EtOAc, washed
with 10% aqueous HCl solution and washed with brine,
dried over anhydrous NaSO4 and the solvent removed. The
residue was submitted to flash chromatography (EtOAc/
hexane mixtures) to give the corresponding products.
[12] J. Justicia, I. Sancho-Sanz, E. ꢁlvarez-Manzaneda, J. E.
Oltra, J. M. Cuerva, Adv. Synth. Catal. 2009, 351, 2295–
2300.
[13] Some remarkable examples based on palladium cata-
lysts and (over)stoichiometric amounts of ZnEt2, or sa-
marium(II) complexes, have been described. Neverthe-
less, these protocols have some limitations derived
from the reaction conditions or the reactivity of the
corresponding organometallic species, as well as they
provide mixtures of regioisomers: a) Y. Tamaru, S.
Goto, A. Tanaka, M. Shimizu, M. Kimura, Angew.
Chem. 1996, 108, 962–963; Angew. Chem. Int. Ed.
Engl. 1996, 35, 878–880; b) T. Tabuchi, J. Inanaga, M.
Yamaguchi, Chem. Lett. 1987, 2275–2278; c) J. M. Aur-
recoechea, R. FaÇanas, M. Arrate, J. M. Gorgojo, N.
Aurrekoetxea, J. Org. Chem. 1999, 64, 1893–1901;
d) Y. Makioka, K. Koyama, T. Nishiyama, K. Takaki,
Y. Taniguchi, Y. Fujiwara, Tetrahedron Lett. 1995, 36,
6283–6286.
[14] Methods based on Ti(II) have a better profile but they
are limited by the use of (over)stoichiometric amounts
of titanium(II) complexes: a) T. Nakawawa, A. Kasat-
kin, F. Sato, Tetrahedron Lett. 1995, 36, 3207–3210;
b) S. Okamoto, D. K. An, F. Sato, Tetrahedron Lett.
1998, 39, 4551–4554; c) S. Okamoto, D. K. An, F. Sato,
Tetrahedron Lett. 1998, 39, 4861–4864; d) T. Hanazawa,
S. Okamoto, F. Sato, Org. Lett. 2000, 2, 2369–2371;
e) F. Yang, G. Zhao, Y. Ding, Tetrahedron Lett. 2001,
42, 2839–2841; f) C. Delas, S. Okamoto, F. Sato, Tetra-
hedron Lett. 2002, 43, 4373–4375; g) F. Yang, G. Zhao,
Y. Ding, Z. Zhao, Y. Zheng, Tetrahedron Lett. 2002, 43,
1289–1293; h) Y. Yatsumonji, T. Sugita, A. Tsubouchi,
T. Takeda, Org. Lett. 2010, 12, 1968–1971.
Acknowledgements
We thank Junta de Andalucꢀa (Project P09-FQM-4571),
MICINN (Grant CTQ2008-4091) for financial support. AM,
AML and AGC thank MICINN for their FPU fellowships.
LAdC thanks UGR for his research contract. Finally we
would like to thank Dr. A. M. Echavarren for his helpful dis-
cussions during the preparation of the manuscript.
References
[1] For recent reviews, see: a) S. T. Diver, A. J. Giessert,
Chem. Rev. 2004, 104, 1317–1382; b) E. Jimꢄnez-
NfflÇez, A. M. Echavarren, Chem. Rev. 2008, 108, 3326–
3350.
[2] For recent syntheses, see: a) J. D. Trenkle, T. F. Jami-
son, Angew. Chem. 2009, 121, 5470–5472; Angew.
Chem. Int. Ed. 2009, 48, 5366–5368; b) K. Molawi, N.
Delpont, A. M. Echavarren, Angew. Chem. Int. Ed.
2010, 49, 3517–3519.
[3] a) H. Yamamoto, in: Comprehensive Organic Synthesis,
(Eds.: B. M. Trost, I. Fleming), Pergamon Press:
Oxford, 1991, Vol. 2, pp. 81–98. For recent references,
see: b) D. R. Fandrick, K. R. Fandrick, J. T. Reeves, Z.
Tan, C. S. Johnson, H. Lee, J. J. Song, N. K. Yee, C. H.
Senayake, Org. Lett. 2010, 12, 88–91; c) S.-L. Shi, L.-X.
Xu, K. Oisaki, M. Kanai, M. Shibasaki, J. Am. Chem.
Adv. Synth. Catal. 2011, 353, 73 – 78
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
77

Alba Millꢀn et al.
COMMUNICATIONS
[15] There is a report of a Ti
G
place when a mixture of a Pd(0) complex, such as Pd-
AHCTUNGTRENNUNG
(PPh3)4, and pre-generated (Cp2TiCl)2 is used.[25] Never-
À
breakage of C O bonds but it requires high tempera-
ture, which is incompatible with the presence of the
carbonyl group owing to the tendency of Cp2TiCl to
transfer a Cp group by Grignard-type reactions: R.
Horacio, A. L. Dieguez, D. Victoriano, J. F. Arteaga,
J. A. Dobado, M. M. Herrador, J. F. Quꢂlez del Moral,
A. F. Barrero, J. Am. Chem. Soc. 2010, 132, 254–259.
[16] A. G. CampaÇa, B. Bazdi, N. Fuentes, R. Robles, J. M.
Cuerva, J. E. Oltra, S. Porcel, A. M. Echavarren,
Angew. Chem. 2008, 120, 7625–7629; Angew. Chem.
Int. Ed. 2008, 47, 7515–7519.
theless, the reaction rate is highly accelerated in the
presence of Mn dust. It is also worth noting that in ab-
sence of Cp2TiCl the reaction does not take place. One
possible explanation for this fact is that Cp2TiCl is not
only acting as an SET reagent, being also a pre-activa-
tor for the reduction of the initial transient palladium
complexes. A coordination with the ethyl carbonate
anionic ligand could favor an Mn(0)-mediated SET.
This hypothesis is supported by the previously reported
Lewis acidic character of Cp2TiCl: M. Paradas, A. G.
CampaÇa, R. E.; Estꢄvez, L. ꢁlvarez de Cienfuegos, T.
Jimꢄnez, R. Robles, J. M. Cuerva, J. E. Oltra, J. Org.
Chem. 2009, 74, 3616–3619.
[17] The
single-electron
bis(cyclopentadienyl)titaniumAHCUTNGTREG(NNUN III) chloride can be gen-
erated in situ by stirring comercial Cp2TiCl2 with Mn
dust in THF, where it exists as a mixture of the mono-
nuclear Cp2TiCl and the dinuclear (Cp2TiCl)2 species;
see: R. J. Enemærke, J. Larsen, T. Skrydstrup, K.
Daasbjerg, J. Am. Chem. Soc. 2004, 126, 7853–7864.
For clarity reasons, in this report we represent this
complex generated in situ as Cp2TiCl.
[24] Nevertheless, alternative mechanisms based on a direct
addition of propargylic radicals to the carbonyl com-
pound in the presence of titanoceneACTHNUGRTENUNG(III) cannot be
ruled out: a) A. Fernꢀndez-Mateos, S. Encinas Madra-
zo, P. Herrero Teijón, R. Rubio Gonzꢀlez, J. Org.
Chem. 2009, 74, 3913–3918. This mechanism can also
explain the excellent regiochemistry taking into ac-
count that in substituted propargylic radicals the
alkyne form prevails: b) C. L. Parker, A. L. Cooksy, J.
Phys. Chem. A 1999, 103, 2160. For a related mecha-
nism using the Nozaki–Hiyama–Kishi reaction see:
c) M. Bandini, P. G. Cozzi, A. Umani-Ronchi, Poly-
hedron 2000, 19, 537–539.
[18] Manganese is usually used in excess (800 mol%) but it
can be filtered and reused in subsequent experiments.
[19] The Cp2TiCl and Pd catalyst loadings can be smaller
but in some cases incomplete conversions are obtained.
The reported loadings have been optimized for com-
plete conversions in all the experiments.
[20] 2,4,6-Collidine can be recovered at the end of the reac-
tion by a simple acid-base extraction.
[25] The synthesis of highly air-sensitive (Cp2TiCl)2 has
been described: R. S. P. Coutts, P. C. Wailes, R. L.
Martin, J. Organomet. Chem. 1973, 47, 375–382.
[26] Taking into account that Cp2TiCl is an air-sensitive
compound, oxygen must be removed from the solution.
Therefore, argon or nitrogen gas must be bubbled
through the solvent (THF) during 10 min prior to use.
[21] The stereochemistry was assigned by chemical correla-
tion with the corresponding alkene derivative: D. See-
bach, L. Widler, Helv. Chim. Acta 1982, 65, 1972–1981.
[22] M. Ishiguro, N. Ikeda, H. Yamamoto, J. Org. Chem.
1982, 47, 2225–2227.
[23] Nevertheless, control experiments suggest that this step
of the mechanism is not simple. The reaction takes
78
asc.wiley-vch.de
ꢃ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 73 – 78