TiCl4-Mediated Baylis-Hillman and aldol reactions without the direct use of a Lewis base

Article abstract of DOI:10.1016/S0040-4039(99)01992-9

TiCl₄-Mediated Baylis-Hillman and aldol reactions were developed without the direct use of a Lewis base. These processes involve the conjugate addition of TiCl₄ to α,β-unsaturated substrates followed by carbonyl coupling. Baylis-Hill

Full text of DOI:10.1016/S0040-4039(99)01992-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1–5
TiCl4-Mediated Baylis–Hillman and aldol reactions without the
direct use of a Lewis base
Guigen Li,∗ Han-Xun Wei, Joe J. Gao and Thomas D. Caputo
Department of Chemistry and Biochemistry, Texas Tech University, Lubbock, TX 79409-1061, USA
Received 17 August 1999; accepted 21 October 1999
Abstract
TiCl4-Mediated Baylis–Hillman and aldol reactions were developed without the direct use of a Lewis base. These
processes involve the conjugate addition of TiCl4 to α,β-unsaturated substrates followed by carbonyl coupling.
Baylis–Hillman olefins were obtained when α,β-unsaturated ketones were employed as the substrates, whereas
β-halogenated aldol products were generated with an α,β-unsaturated N-acyl benzoxalinone as the Michael-type
acceptor. © 1999 Elsevier Science Ltd. All rights reserved.
Keywords: titanium(IV) chloride; Baylis–Hillman reaction; aldol reaction.
The Baylis–Hillman reaction and related processes have become increasingly important in synthetic
organic chemistry because the resulting adducts have an array of multifunctional groups which can be
subjected to numerous transformations.1–5 This carbon–carbon bond formation was typically catalyzed
by DABCO or tertiary phosphines. The major drawbacks of the Baylis–Hillman reaction are shown
by its slow reaction rate and limited scope of substrates. To overcome its shortcomings, much effort
has been made on using Lewis acids or adding additives to the reaction system to activate carbonyl
electrophiles.6–10 Among various Lewis acids, TiCl4 has been successfully utilized to promote the Bay-
lis–Hillman reaction in the presence of Lewis base catalysts.11 During our study on the Baylis–Hillman
process, we found that this reaction can proceed smoothly in the presence of TiCl4 with no need of a
Lewis base. Concurrently, we also found that this C–C formation can be controlled to give β-chloro aldol
adducts simply by choosing an α,β-unsaturated N-benzoxazolinone as the substrate. In this paper, we
report our preliminary results of these new systems.
The new processes are represented in Scheme 1 with the results summarized in Tables 1 and 2. The
Baylis–Hillman reaction was realized by simply mixing three components, aldehyde, α,β-unsaturated
cycloketone and titanium(VI) chloride in dichloromethane at 0°C then to room temperature. Any
convenient vessel of appropriate size can be used for the reaction without the protection of inert gases.
An excess amount of cycloketone (ketone/aldehyde, 2/1, mol/mol) was found necessary for the complete
consumption of aldehyde.
∗
Corresponding author. Fax: 806-742-1289; e-mail: qeggl@ttu.etu
0040-4039/00/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)01992-9

2
Scheme 1.
Scheme 2.
Table 1
Results of TiCl4-mediated halogen-Baylis–Hillman reaction12
Examination of the results collected in Table 1 reveals that both aliphatic and aromatic aldehydes can
be employed as electrophilic acceptors among which aliphatic aldehydes gave better yields. For aromatic
aldehydes, electron withdrawing group attachment is necessary to achieve the modest reaction rate and
yield. Benzaldehyde resulted in only a trace amount of desired product. About 8% of diene side-products
were found in aliphatic cases 1–3, which were generated from the dehydration of Baylis–Hillman

3
adducts. However, there was difficulty in identifying the major side-products (>10%) for cases 4–7.
Attempts to improve the yields under various conditions proved to be unsuccessful. A trace amount
of desired product was detected in several other solvents, such as benzene, THF, etc. Among various
metal promoters (AlCl3, ZnCl2, etc.) we examined, only TiCl4 showed good efficiency for this new
process. So far, only 2-cyclohexen-1-one and 2-cyclopenten-1-one have been employed as Michael-
type acceptors. The cyclic ketones with larger cyclic rings and other similar derivatives, such as α,β-
unsaturated lactones and lactams, will be subjected to this reaction in future. Acyclic methyl vinyl ketone
was also utilized for this system and gave complex products. The modification of the present system by
using Bu4NI:TiCl4 (1:1) mixture instead of TiCl4 resulted in the expected Baylis–Hillman adducts with
modest yields (51–67%) (Scheme 2).
As mentioned before, TiCl4 has been previously utilized in the Baylis–Hillman system as the promoter
to activate aldehydes or ketones via forming coordination complexes. Recently, Kataoka and co-workers
reported the chalcogeno-Baylis–Hillman reaction using a series of sulfides and selenides as Lewis base
catalysts, and TiCl4 to activate α,β-unsaturated cycloketones (for Michael-type addition) and aldehydes
(for the C–C coupling).9,10 This reaction proceeded to completion using 3 equiv. of α,β-unsaturated
cycloketones in the presence of 10 mol% of chalcogenides and 1.0 equiv. of TiCl4 in dichloromethane.
In our conditions (2.0 equiv. of α,β-unsaturated cycloketones and 1.2 equiv. of TiCl4), there is no Lewis
base catalyst specially added into the reaction mixture. However, the reaction took a little longer time to
go to completion.
During the reaction process, we believe that the chlorine anion released from TiCl4 acted as the actual
Lewis base to participate in the conjugate addition to form the halogenated titanium enolate. This enolate
will then react with the aldehyde to give the coupling adduct. The release of chlorine anion could be
assisted by the coordination of aldehyde or cycloketone oxygen onto TiCl4 metal center as shown in
Scheme 3.
Scheme 3.
To confirm this hypothesis, we made the effort to isolate the chlorinated aldol adducts generated
from the aldol addition without much success. Interestingly, the chlorinated aldol adducts became the
major products when an α,β-unsaturated N-acyl benzoxalinone was utilized as the Michael-type acceptor
(Scheme 4 and Table 2). The reaction was performed to completion under similar conditions (CH2Cl2,
room temperature, 1 h). This process should also be synthetically useful because the resulting chlorinated
aldol adducts can be readily transformed to Baylis–Hillman olefins by treating with DBU, or to many
other compounds via SN2 substitution of β-chlorine by various nucleophiles.
In conclusion, the new TiCl4-based Baylis–Hillman reaction and aldol addition have been developed
for the synthesis of Baylis–Hillman olefins and chlorinated aldol adducts. The reactions can be performed
at room temperature in any convenient vial without the inert gase protection. These new procedures could
provide novel strategies in order to achieve their asymmetric versions in the future.

4
Scheme 4.
Table 2
Results of TiCl4-mediated halogen aldol reaction
Acknowledgements
We would like to dedicate this paper to Professor Richard A. Bartsch on the occasion of his 60th
birthday. We gratefully acknowledge generous support from the Robert A. Welch Foundation (grant
no. D-1361) and the South Plains Foundation. We also would like to thank Ms. Sun Hee Kim for her
assistance.
References
1. For reviews, see: (a) Ciganek, E. Org. React. 1997, 51, 201.1; (b) Basavaiah, D.; Rao, P. D.; Hyma, R. S. Tetrahedron
1996, 52, 8001; (c) Drewes, S. E.; Roos, G. H. P. Tetrahedron 1988, 44, 4653.
2. Brzezinski, L. J.; Rafel, S.; Leahy, J. M. J. Am. Chem. Soc. 1997, 119, 4317.
3. Marko, I. E.; Giles, P. G.; Hindley, N. J. Tetrahedron 1997, 53, 1015.
4. Richter, H.; Jung, G. Tetrahedron Lett. 1998, 39, 2729.
5. Barrett, A. G. M.; Cook, A. S.; Kamimura, A. Chem. Commun. 1999, 2533.
6. Kundig, E. P.; Xu, L. H.; Romanens, P; Bernardinelli, G. Tetrahedron Lett. 1993, 34, 7049.
7. Aggarwal, V.; Mereu, A.; Tarver, G. J.; McCague, R. J. Org. Chem. 1998, 63, 7183.
8. Kawamura, M.; Kobayashi, S. Tetrahedron Lett. 1999, 40, 1539.
9. Kataoka, T.; Iwama, T.; Tsujiyama, S.-i.; Iwamura, T.; Watanable, S.-i. Tetrahedron 1998, 54, 11813.
10. Ono, M.; Nishimura, K.; Nagaoka, Y.; Tomioka, K. Tetrahedron Lett. 1999, 40, 1509.
11. (a) Nagaoka, Y.; Yomioka, K. J. Org. Chem. 1998, 63, 6428; (b) Harrowven, D.; Hannam, J. C. Tetrahedron 1999, 55,
9341.

5
12. Typical procedure (Table 1, entry 1): Into a dry vial was added isovaleraldehyde (86.1 mg, 1.0 mmol), 2-cyclohexen-1-one
(192.3 mg, 2.0 mmol) and freshly distilled dichloromethane (1.5 mL). The reaction vial was immersed in a 0°C bath, and
TiCl4 (1.2 mL, 1.0 M in CH2Cl2, 1.2 mmol) was then added dropwise. The resulting solution was stirred at 0°C for 10
minutes and at room temperature for 2 hours. The reaction was finally quenched by saturated aqueous NaHCO3 solution
(3.0 mL). The phases were separated, and the aqueous phase was extracted with ethyl acetate (3×10 mL). The combined
organic layers were washed with water and brine, dried over anhydrous magnesium sulfate, and concentrated to dryness.
Purification by flash chromatography (EtOAc/hexane, 3/50, v/v) provided product 1 (124 mg, 68% yield) as colorless oil.
1H NMR (200 MHz, CDCl3): δ 7.17 (t, J=4.25 Hz, 1H), 5.05 (dd, J=8.79, 4.41 Hz, 1H), 2.42–2.51 (m, 4H), 1.54–2.04
(m, 5H), 0.90–0.96 (m, 6H); 13C NMR (75 MHz, CDCl3): δ 196.7, 147.1, 140.4, 54.8, 46.8, 38.3, 25.9, 25.6, 22.9, 22.4,
1
21.2. Compound 2: colorless oil (107 mg, 64% yield); H NMR (200 MHz, CDCl3): δ 7.17 (t, J=4.33 Hz, 1H), 4.97 (t,
J=6.38 Hz, 1H), 2.42–2.50 (m, 4H), 1.36–2.07 (m, 7H), 0.92 (t, J=7.39 Hz, 3H); 13C NMR (75 MHz, CDCl3): δ 196.9,
1
147.2, 140.2, 56.5, 40.0, 38.3, 26.0, 22.5, 20.0, 13.4. Compound 3: colorless oil (143 mg, 57% yield); H NMR (200
MHz, CDCl3): δ 7.16 (t, J=4.23 Hz, 1H), 4.95 (t, J=6.96 Hz, 1H), 2.44–2.50 (m, 4H), 1.96–2.04 (m, 2H), 1.67–1.82 (m,
2H), 1.25 (s, 16H), 0.88 (t, J=4.88 Hz, 3H); 13C NMR (75 MHz, CDCl3): δ 196.8, 147.2, 140.2, 56.8, 38.3, 37.9, 31.8,
29.5, 29.4, 29.3, 28.9, 26.8, 26.0, 22.6, 22.5, 14.0. Compound 4: colorless oil (113 mg, 49% yield); 1H NMR (200 MHz,
CDCl3): δ 8.21 (dd, J=6.95, 1.99 Hz, 2H), 7.59 (dd, J=7.30, 1.80 Hz, 2H), 7.32–7.34 (m, 1H), 5.67 (m, 1H), 3.76 (d,
J=5.41 Hz, 1H), 2.61–2.67 (m, 2H), 2.46–2.51 (m, 2H); 13C NMR (75 MHz, CDCl3): δ 209.2, 159.9, 148.5, 147.5, 146.6,
127.0, 123.7, 68.9, 35.1, 26.8. Compound 5: colorless oil (139 mg, 56% yield); 1H NMR (200 MHz, CDCl3): δ 8.20 (dd,
J=7.02, 1.89 Hz, 2H), 7.56 (dd, J=6.43, 1.51 Hz, 2H), 6.83 (t, J=3.79 Hz, 1H), 5.61 (d, J=5.99 Hz, 1H), 3.58 (d, J=5.95
Hz, 1H), 2.40–2.50 (m, 4H), 1.95–2.08 (m, 2H); 13C NMR (75 MHz, CDCl3): δ 200.0, 149.3, 148.1, 147.2, 140.2, 127.1,
1
123.5, 72.1, 38.4, 25.7, 22.4. Compound 6: colorless oil (121 mg, 47% yield); H NMR (200 MHz, CDCl3): δ 7.62 (d,
J=9.40 Hz, 2H), 7.52 (d, J=8.30 Hz, 2H), 7.26–7.29 (m, 1H), 5.63 (s, 1H), 3.54 (b, 1H), 2.59–2.64 (m, 2H), 2.45–2.50 (m,
2H); 13C NMR (75 MHz, CDCl3): δ 209.5, 159.6, 147.1, 145.2, 126.6, 125.4, 125.5 (q), 69.3, 35.2, 26.7. Compound 7:
colorless oil (143 mg, 53% yield); 1H NMR (200 MHz, CDCl3): δ 7.60 (d, J=9.35 Hz, 2H), 7.48 (d, J=8.28 Hz, 2H), 6.77
(t, J=4.08 Hz, 1H), 5.58 (d, J=5.61 Hz, 1H), 3.57 (d, J=5.78 Hz, 1H), 2.38–2.49 (m, 4H), 1.97–2.07 (m, 2H); 13C NMR
(75 MHz, CDCl3): δ 200.2, 147.8, 145.8, 140.5, 126.7, 125.2 (q), 72.1, 38.5, 25.7, 22.4.