Titanium(IV) chloride, zirconium(IV) chloride or boron trichloride and phosphine-promoted Baylis-Hillman reaction of aldehydes with α,β-unsaturated ketone

Article abstract of DOI:10.1039/b008105l

In the Baylis-Hillman reaction of aldehydes with an α,β-unsaturated ketone, the chlorinated compound 1 was obtained as the major product using tributylphosphine as a Lewis base in the presence of titanium(IV) chloride, zirconium(IV) chloride or boron tric

Full text of DOI:10.1039/b008105l

View Article Online / Journal Homepage / Table of Contents for this issue
Titanium(IV) chloride, zirconium(IV) chloride or boron trichloride
and phosphine-promoted Baylis–Hillman reaction of aldehydes
with ꢀ,ꢁ-unsaturated ketone
1
Min Shi,* Jian-Kang Jiang, Shi-Cong Cui and Yan-Shu Feng
Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry,
Chinese Academy of Sciences, 354 Fenglin Lu, Shanghai 200032, China.
E-mail: mshi@pub.sioc.ac.cn
Received (in Cambridge, UK) 6th October 2000, Accepted 22nd December 2000
First published as an Advance Article on the web 29th January 2001
In the Baylis–Hillman reaction of aldehydes with an α,β-unsaturated ketone, the chlorinated compound 1 was
obtained as the major product using tributylphosphine as a Lewis base in the presence of titanium() chloride,
zirconium() chloride or boron trichloride in dichloromethane at <Ϫ20 ЊC. A plausible reaction mechanism has
been proposed. Using (R)-(ϩ)-2,2Ј-bis(diphenylphosphino)-1,1Ј-binaphthyl (BINAP) as a chiral Lewis base, 10%
ee could be achieved. We also found that, if the reaction was carried out at room temperature, the dehydrated
compound 3 was obtained as the major product in the Z-configuration.
Introduction
Recently, the Baylis–Hillman reaction has become a very hot
1–11
field for synthetic chemists. The combination of a Lewis base
such as SMe or NEt with the Lewis acid TiCl can signifi-
2
3
4
cantly speed up this reaction and give the corresponding chlor-
12
inated products. Based on those previous results, we further
attempt to explore new Lewis bases or Lewis acids and disclose
the reaction mechanism of this interesting reaction. Herein we
wish to report the full details of the TiCl , ZrCl or BCl and
4
4
3
phosphine-promoted Baylis–Hillman reaction, along with a
plausible mechanism based on the previous findings and our
own results.
Fig. 1 The crystal structure of 1a.
(
Table 1, entry 9), but the reaction is relatively slow. In all cases,
Results and discussion
Using procedures reported previously,
the reaction of p-nitrobenzaldehyde with methyl vinyl ketone in
only one isomer was obtained and their relative configurations
were confirmed as the syn form by comparison of their spectral
data with those reported previously.
configuration by X-ray analysis (Fig. 1). Thus, the Lewis base
did not affect the stereochemistry of the chlorinated products
1
2b,c
we initially attempted
12b,c
For 1a, we confirmed its
13
the presence of TiCl (1.4 equiv.) and tributylphosphine (PBu )
4
3
(
0.2 equiv.) at Ϫ78 ЊC and found that the chlorinated product
1
. This is another novel system for producing syn form
was also formed and the results were very similar to those using
12b,c
chlorinated products in the Baylis–Hillman reaction.
SMe or NMe as a Lewis base.
For example, we carried out
2
3
On the other hand, we examined many metal halides, such as
PdCl , RhCl , Cp ZrCl , ZrCl , AlCl , TMSCl, SiCl , BF , BCl ,
as Lewis acids and found that BCl and ZrCl also worked for
this reaction, although they are not as effective as TiCl
example, we carried out the Baylis–Hillman reaction using BCl
3
as a Lewis acid and PBu as a Lewis base under the same
the reaction of p-nitrobenzaldehyde with methyl vinyl ketone in
2
3
2
2
4
3
4
3
3
the presence of TiCl at Ϫ78 ЊC. No reactions occurred (Table
4
3
4
1
, entry 1). After adding 20 mol% of tributylphosphine (PBu )
3
12b,c
. For
4
as a Lewis base,
the reaction took place smoothly to give the
12
chlorinated product 1a as the major product (Scheme 1)
3
reaction conditions as those shown in Scheme 1. The syn
chlorinated products can be also obtained (Scheme 2, Table 2).
For p-nitrobenzaldehyde, 80% yield of 1a could be obtained,
but needed a longer reaction time (40 h) at Ϫ78 ЊC. The results
are summarized in Table 2.
Scheme 1
(
conditions B). At Ϫ78 ЊC, the reaction proceeded very well for
aryl aldehydes having strong electron-withdrawing groups. The
other aldehydes needed a higher temperature (Ϫ20 ЊC) for
completion of the reaction. The results are summarized in
Table 1. Triphenylphosphine can catalyze the reaction as well
Scheme 2
3
90
J. Chem. Soc., Perkin Trans. 1, 2001, 390–393
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b008105l

View Article Online
Table 1 Baylis–Hillman reaction of aldehydes with methyl vinyl ketone in the presence of TiCl and Lewis base
4
a
Entry
R
Lewis base
Conditions
Temp./ЊC
Time/h
Yield of 1 (%)
b
1
2
3
4
5
6
7
8
9
p-NO Ph
None
PBu3
PBu3
PBu3
PBu3
PBu3
PBu3
PBu3
PBu3
A
B
Ϫ78
12
12
12
48
72
72
72
72
24
—
80
80
81
51
57
52
40
40
2
c
p-NO Ph
Ϫ78
Ϫ78
Ϫ78
Ϫ20
Ϫ20
Ϫ20
Ϫ20
Ϫ78
2
m-NO Ph
B
B
B
B
B
B
B
2
p-CF Ph
3
p-EtPh
Ph
p-ClPh
C H
4
9
p-NO Ph
2
a
b
c
Isolated yields. Aldehyde:TiCl = 1:1.4. Aldehyde:TiCl :Lewis base = 1:1.4:0.2.
4
4
Table 2 Baylis–Hillman reaction of aldehydes with methyl vinyl ketone in the presence of ZrCl , BCl and PBu
4
3
3
a
Entry
R
Lewis acid
Conditions
Temp./ЊC
Time/h
Yield of 1 (%)
b
1
2
3
4
5
6
7
8
9
p-NO Ph
ZrCl4
ZrCl4
ZrCl4
BCl3
BCl3
BCl3
BCl3
BCl3
BCl3
B
Ϫ78
40
40
48
40
40
40
80
80
80
50
50
41
80
60
55
45
25
26
2
m-NO Ph
B
B
B
B
B
B
B
B
Ϫ78
Ϫ78
Ϫ78
Ϫ78
Ϫ20
Ϫ20
Ϫ20
Ϫ20
2
p-CF Ph
3
p-NO Ph
2
m-NO Ph
2
p-CF Ph
3
Ph
p-EtPh
p-ClPh
a
b
Isolated yields. Aldehyde:BCl :Lewis base = 1:1.4:0.2.
3
Concerning the reaction mechanism, Kataoka has reported a
plausible reaction mechanism for the Baylis–Hillman reaction
using SMe as a Lewis base and TiCl as a Lewis acid and
2
4
explained why the syn-form chlorinated products could be
12a
formed predominantly. During our own investigation on the
Baylis–Hillman reactions mentioned above, we always observed
that, when TiCl4 was combined with Lewis bases such as
amines, phosphines or SMe in dichloromethane, the solution
2
immediately became dark red with some precipitation at low
temperature. This phenomenon suggests that coordination of
the Lewis base to TiCl takes place and consequently changes
4
the oxidation state of the Ti metal center. In fact, the previous
studies on the reactions of phosphines with TiCl have shown
4
that many phosphines can coordinate to the Ti metal center.
For example, the matrix-isolation technique and an infrared
spectroscopy study have proved and characterized the 1:1
14
complexes of TiCl with PH or AsH . In addition, in the
4
3
3
normal reaction of TiCl with phosphines, the formation of
4
face-sharing ionic Ti() or Ti() clusters has been reported in
which the molecular structures of [PPh ] [Cl Ti(µ-Cl) (µ-Cl) -
4
3
3
3
3
TiCl ]ؒ2CH Cl and [PPh ] [Cl Ti(µ-Cl) TiCl ]ؒ4CH Cl ؒ0.5Ph-
Me have been determined by X-ray analysis. Furthermore, in
3
2
2
4 3
3
3
3
2
2
1
5
Scheme 3
the reactions of amines with TiCl , the formation of similar
4
1
6
ionic Ti complexes having a chloride ion has been reported1.
selectivity. As a matter of fact, we utilized a famous chiral
ligand, (R)-(ϩ)-2,2Ј-bis(diphenylphosphino)-1,1Ј-binaphthyl
2c
Based on these previous findings and our own investigation,
1
8
we believe that, in the above Baylis–Hillman reactions, Lewis
(BINAP), as a chiral ligand for this purpose. Using 20 mol%
of BINAP as a chiral Lewis base and p- or m-nitrobenzalde-
hyde as a substrate, Baylis–Hillman olefin 2a or 2b could be
obtained in only 10% and 9% ee, respectively, after treating the
chlorinated product 1a or 1b with triethylamine or DBU
(Scheme 4). This result also partially supports the reaction
mechanism shown in Scheme 3.
bases such as amines or phosphines first react with TiCl to give
4
an ionic Ti metal cluster having a chloride anion which then
undergoes Michael attack on the α,β-unsaturated ketone to ini-
tiate the reaction (Scheme 3). Using BCl or ZrCl as a Lewis
3
4
acid, the reaction proceeds by the same mechanism. If the
Lewis bases can coordinate to Lewis acids such as TiCl , ZrCl4
4
or BCl and kick out the chloride ion, the reaction can take
On the other hand, by carrying out this reaction at room
temperature, we confirmed that the elimination product 3 was
the only product (Scheme 5). This reaction was first disclosed
by Li and co-workers. They reported that, in the presence of
3
place to give the corresponding chlorinated product at low
temperature. This finding obviously opens the way for achieving
catalytic enantioselective Baylis–Hillman reactions by means of
17
chiral phosphine ligands in a catalytic process. However, as
can be seen from Scheme 3, the chiral Lewis base obviously
only forms an ion pair with the enolate and it does not partici-
pate in the attack on aldehydes in the last step. Thus it is very
difficult to use this reaction process to achieve high enantio-
stoichiometric or nonstoichiometric TiX , compound 3 could
4
19
be formed in the Z-configuration. Later we also reported that
3 could be exclusively obtained in the TiCl and chalcogenide-
4
promoted Baylis–Hillman reaction at room temperature
12b
(20 ЊC). The Z-configuration has been disclosed by X-ray
J. Chem. Soc., Perkin Trans. 1, 2001, 390–393 391

View Article Online
Experimental
General
Mps were obtained with a Yanagimoto micro melting point
1
apparatus and are uncorrected. H NMR spectra were recorded
on a Bruker AM-300 spectrometer for solutions in CDCl with
3
tetramethylsilane (TMS) as internal standard; J-values are in
Hz. Mass spectra were recorded with an HP-5989 instrument
and HRMS was measured by a Finnigan MAϩ mass spectro-
meter. Organic solvents were dried by standard methods when
necessary. Some of the solid compounds reported in this paper
gave satisfactory CHN microanalyses with a Carlo-Erba 1106
analyzer. Commercially obtained reagents were used without
further purification. All reactions were monitored by TLC with
Huanghai 60F254 silica gel-coated plates. Flash column chrom-
atography was carried out using 200–300 mesh silica gel at
increased pressure.
Scheme 4
The physical data and spectral data of 1a–g are the same as
12
those reported previously.
Scheme 5
Typical reaction procedure for the preparation of syn-3-(chloro-
methyl)-4-hydroxy-4-(4Ј-nitrophenyl)butan-2-one 1a
1
2b
19
analysis. Based on Li’s report, compound 3 could be formed
even in the absence of any Lewis base. In order to clarify the
formation of 3, we treated 1a and 2a directly with TiCl in
dichloromethane at room temperature. We found that 1a can be
transformed to 3a within 3 h, whereas the reaction of 2a was
much slower (Scheme 6). These results strongly suggest that 3a
To a solution of tributylphosphine (5.05 mg, 0.025 mmol) in
dichloromethane (1.3 mL) was added 1.4 M titanium tetra-
chloride in dichloromethane (0.7 mL, 0.7 mmol) at Ϫ78 ЊC.
After stirring for 5 min, a solution of p-nitrobenzaldehyde (76
mg, 0.5 mmol) in dichloromethane (1.0 mL) and methyl vinyl
ketone (105 mg, 1.5 mmol, 123 µL) were added. The reaction
mixture was kept for 24 h at Ϫ78 ЊC. The reaction was
4
quenched by addition of saturated aqueous NaHCO solution
3
(
1.0 mL). After filtration, the filtrate was extracted with
dichloromethane (5.0 mL × 2) and dried over anhydrous
MgSO . The solvent was removed under reduced pressure and
4
the residue was purified by flash silica gel chromatography to
give compound 1a (102 mg, 80%) as a colorless solid (eluent:
ethyl acetate–petroleum ether = 1:4); mp 90–91 ЊC; IR (KBr)
Ϫ1
1
ν 1720 cm (C᎐O); H NMR (CDCl , 300 MHz) δ 2.20 (3H, s,
3
Me), 2.93 (1H, br s, OH), 3.22–3.38 (1H, m), 3.67 (1H, dd,
J 11.3, 4.0 Hz), 3.89 (1H, dd, J 11.3, 9.2 Hz), 5.11 (1H, d, J 5.6
Hz), 7.56 (2H, d, J 8.6 Hz, Ar), 8.25 (2H, d, J 8.6 Hz, Ar); MS
Scheme 6
ϩ
ϩ
ϩ
(
EI) m/z 258 (MH , 0.60), 208 (M Ϫ 49, 60), 71 (M Ϫ 186,
is derived directly from 1a formed first in the reaction. Thus, we
believe that, at room temperature, the chlorinated products 1
could be formed either in the absence or in the presence of
Lewis base, but they are rapidly transformed to the elimination
product 3 exclusively.
1
11 12
00) [Found: C, 51.64; H, 4.94; N, 5.35%. C H ClNO4
requires C, 51.27; H, 4.69; N, 5.44%].
Acknowledgements
We thank the State Key Project of Basic Research (Project 973)
Conclusion
(
No. G2000048007) for financial support. We also thank the
Inoue Photochirogenesis Project (ERATO, JST) for chemical
reagents.
We found that the titanium() chloride and phosphine-
promoted Baylis–Hillman reaction is a very efficient reaction
system for producing chlorinated compound 1. Phosphine
compounds are good Lewis bases for this reaction. BCl and
3
References
ZrCl are good Lewis acids for this reaction as well. In com-
4
12b,c
1 For reviews, see: (a) E. Ciganek, Org. React. (N. Y.), 1997, 51, 201;
parison with the previous results,
the activities of Lewis
(
8
b) D. Basavaiah, P. D. Rao and R. S. Hyma, Tetrahedron, 1996, 52,
acids for this reaction are TiCl > BCl > ZrCl and the
4
3
4
001; (c) S. E. Drewes and G. H. P. Roos, Tetrahedron, 1988, 44,
activities of Lewis bases are NEt > PBu > SMe . The best
3
3
2
4653.
combination of Lewis acid and Lewis base for this reaction is
2 (a) L. J. Brzezinski, S. Rafel and J. M. Leahy, J. Am. Chem. Soc.,
1997, 119, 4317; (b) K. Morita, Z. Suzuki and H. Hirose, Bull.
Chem. Soc. Jpn., 1968, 41, 2815; (c) T. Miyakoshi and S. Saito,
Nippon Kagaku Kaishi, 1983, 1623; Chem. Abstr., 1984, 100,
TiCl (1.4 equiv.) with NEt (0.2 equiv.). The relative configur-
4
3
ation of 1 was not affected at all by the Lewis bases (SMe2,
NEt , PBu ) or Lewis acids (TiCl , BCl , ZrCl ) used. This
3
3
4
3
4
1
56191g; (d) Toyo Rayon Co., French Patent, 1,506,132 (1967);
reaction was initiated by chloride ion attacking at the α,β-
unsaturated ketone Michael acceptor. The chloride ion was
produced by coordination of Lewis bases (SMe , NEt , PBu )
Chem. Abstr., 1969, 70, 19613u.
3
I. E. Marko, P. G. Giles and N. J. Hindley, Tetrahedron, 1997, 53,
2
3
3
1015.
to Lewis acids such as TiCl , BCl , and ZrCl . In this reaction
4 H. Richter and G. Jung, Tetrahedron Lett., 1998, 39, 2729.
4
3
4
5
6
7
A. G. M. Barrett, A. S. Cook and A. Kamimura, Chem. Commun.,
999, 2533.
E. P. Kunidig, L. H. Xu, P. Romanens and G. Bernardinelli,
Tetrahedron Lett., 1993, 34, 7049.
V. Aggarwal, A. Mereu, G. J. Tarver and R. MaCague, J. Org.
Chem., 1998, 63, 7183.
system, it is very difficult to achieve high enantioselectivity
just by using catalytic amounts of chiral Lewis bases. Efforts
are underway to elucidate the mechanistic details of this reac-
tion and to disclose its scope and limitations. Studies along
these lines are currently in progress.
1
3
92
J. Chem. Soc., Perkin Trans. 1, 2001, 390–393

View Article Online
Ϫ3
8
9
M. Kawamura and S. Kobayashi, Tetrahedron Lett., 1999, 40, 1539.
(a) T. Kataoka, T. Iwama, S.-i. Tsujiyama, T. Iwamura and
S.-i. Watanaba, Tetrahedron, 1998, 54, 11813; (b) T. Kataoka,
T. Iwama, S. Kinoshita, Y. Tsujiyama, T. Iwamura and S. Watanabe,
Synlett, 1999, 197; (c) T. Kataoka, T. Iwama, S. Tsujiyama,
K. Kanematsu, T. Iwamura and S. Watanabe, Chem. Lett., 1999,
(#19); Zvalue = 4; Dcalc = 1.441 g cm ; F000 = 536.00; µ(Mo-
Ϫ1
Kα) = 3.23 cm ; temperature of data collection: 24 ЊC; residuals: R;
R = 0.065; 0.052; residuals: R1 = 0.033; no. of reflections to calc.
w
R1: 1108.
14 J. B. Everhart and B. S. Ault, Inorg. Chem., 1996, 35, 4090.
15 (a) L. Chen and F. A. Cotton, Polyhedron, 1998, 17, 3727; (b) L.
Chen, F. A. Cotton, K. R. Dunbar, X. Feng, R. A. Heintz and
C. Uzelmeir, Inorg. Chem., 1996, 35, 7358; (c) F. A. Cotton, C. A.
Murllo and M. A. Petrukhina, J. Organomet. Chem., 1999, 573, 78.
16 (a) M. Antler and A. W. Laubengayer, J. Am. Chem. Soc., 1955, 77,
5250; (b) P. Bharathi and M. Perisaamy, Org. Lett., 1999, 1, 857.
17 Y. Iwabuchi, M. Nakatani, N. Yokoyama and S. Hatakeyama,
J. Am. Chem. Soc., 1999, 121, 10219.
2
1
57; (d) T. Kataoka, T. Iwama and S. Tsujiyama, Chem. Commun.,
998, 197.
1
1
0 M. Ono, K. Nishimura, Y. Nagaoka and K. Tomioka, Tetrahedron
Lett., 1999, 40, 1509.
1 (a) Y. Nagaoka and K. Yomioka, J. Org. Chem., 1998, 63, 6428;
(
2
b) G. Li, H.-X. Wei, J. J. Gao and T. D. Caputo, Tetrahedron Lett.,
000, 41, 1.
1
1
2 (a) T. Kataoka, H. Kinoshita, T. Iwama, S.-I. Tsujiyama, T.
Iwamura, S.-I. Watanabe, O. Muraoka and G. Tanabe, Tetrahedron,
18 (a) R. Noyori and H. Takaya, Acc. Chem. Res., 1990, 23, 345;
(b) R. Noyori, S. Suga, K. Kawai, S. Okada and M. Kitamura,
Pure Appl. Chem., 1988, 60, 1597; (c) R. Noyori, Chem. Soc. Rev.,
1989, 18, 187.
19 G. Li, J. Gao, H.-X. Wei and M. Enright, Org. Lett., 2000, 2,
617.
2
4
2
000, 56, 4725; (b) M. Shi and J.-K. Jiang, Tetrahedron, 2000, 56,
793; (c) M. Shi, J.-K. Jiang and Y.-S.; Feng, Org. Lett., 2000, 2,
397.
3 The crystal data of 1a:† empirical formula: C H ClNO ; formula
1
1
12
4
weight: 257.67; crystal color, habit: colorless, column; crystal
dimensions: 0.28 × 0.30 × 0.18 mm; crystal system: orthorhombic;
lattice type: primitive; lattice parameters: a = 10.615(1), b =
† CCDC reference number 207/505. See http://www.rsc.org/suppdata/
p1/b0/b008105l/ for crystallographic files in .cif format.
3
1
4.277(1), c = 7.838(1) Å, V = 1187.8(3) Å ; space group: P2 2 2
1
1
1
J. Chem. Soc., Perkin Trans. 1, 2001, 390–393
393