The first successful intermolecular Heck reaction of Baylis-Hillman adducts: synthesis of β-aryl substituted Baylis-Hillman adducts

Article abstract of DOI:10.1016/j.tetlet.2008.03.102

The first successful intermolecular Heck reaction between Baylis-Hillman adducts and aryl iodides was achieved under the conditions comprising Pd(OAc)₂/n-Bu₄NBr/KOAc in CH₃CN.

Full text of DOI:10.1016/j.tetlet.2008.03.102

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 3248–3251
The first successful intermolecular Heck reaction of Baylis–Hillman
adducts: synthesis of b-aryl substituted Baylis–Hillman adducts
Jeong Mi Kim a, Ko Hoon Kim a, Taek Hyeon Kim b, Jae Nyoung Kim a,*
a Department of Chemistry and Institute of Basic Science, Chonnam National University, Gwangju 500-757, Republic of Korea
b Department of Applied Chemistry and Center for Functional Nano Fine Chemicals, College of Engineering,
Chonnam National University, Gwangju 500-757, Republic of Korea
Received 14 February 2008; revised 17 March 2008; accepted 18 March 2008
Available online 25 March 2008
Abstract
The first successful intermolecular Heck reaction between Baylis–Hillman adducts and aryl iodides was achieved under the conditions
comprising Pd(OAc)2/n-Bu4NBr/KOAc in CH3CN.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Baylis–Hillman adducts; Heck reaction; Pd(OAc)2
During the last two decades, notable improvements in
Baylis–Hillman chemistry have been achieved in view of
the reaction rate and synthetic applications of Baylis–Hill-
man adducts.1,2 However, the general and efficient syn-
thesis of b-branched Baylis–Hillman adducts has
remained unsolved although a few approaches have been
reported.3 Thus, the development of a new method of these
compounds would complement other previous by reported
methods.3
The most simple and convenient method for the prepa-
ration of b-aryl substituted Baylis–Hillman adducts could
be the palladium-mediated Heck reaction with aryl halides.
Actually intermolecular Heck type arylation of Baylis–Hill-
man adducts has been examined by some research groups.4
However, the reactions gave benzyl-substituted b-keto ester
4 as the major product instead of b-aryl substituted Baylis–
Hillman adduct 3 (Scheme 1).4 Compound 4 was generated
via the syn-elimination of HaPdOAc from the intermediate
(I) and the following keto–enol tautomerization.4c,d This
unfavorable result might be the principal reason for the
lack of reports on the synthesis of b-aryl substituted
Baylis–Hillman adducts via the Heck type arylation
strategy.
However, the results of most of the Pd-mediated reac-
tions can be altered by changing the reaction conditions,5
thus we decided to find a suitable reaction condition for
the intermolecular Heck arylation. Although the acidity
of Ha must be different from that of Hb/Hc, but while there
is only one Ha hydrogen on one side, there are two Hb and
Hc hydrogens on the carbon holding palladium. Thus, we
imagined that we could control the reaction pathway by
using excess amounts of relatively strong base. We
examined the reaction of Baylis–Hillman adduct 1a and
iodobenzene (2a) under various conditions in Table 1 in
these respects, and found an efficient condition compris-
ing Pd(OAc)2 (15 mol %)/TBAB (1.0 equiv)/KOAc (3.0
equiv)/CH3CN/reflux (Table 1, entry 7). From the experi-
ments, we found that the use of relatively larger amounts
of Pd(OAc)2 and KOAc was crucial for the improvement
of the yield of Heck product 3. The use of bromobenzene
(2d) was less effective (entry 10).
The reaction of 1a and 2a under the optimized condi-
tions afforded b-keto ester 4a (25%) and the desired Heck
product 3a (62%) as cleanly separable E/Z mixture
(26:36, entry 1 in Table 2).6,7 The reactions between
Baylis–Hillman adducts (1a–c) and aryl iodides (2a–c) gave
*
Corresponding author. Tel.: +82 62 530 3381; fax: +82 62 530 3389.
E-mail address: kimjn@chonnam.ac.kr (J. N. Kim).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.03.102

J. M. Kim et al. / Tetrahedron Letters 49 (2008) 3248–3251
3249
OH
Ha
O
OH
EWG
PdOAc
Ar
- HaPdOAc
EWG
Ar
ArI (2)
EWG
R
Hb
Hc
R
R
Pd(OAc)2
4
1
(I)
1a: R = Ph, EWG = COOMe
1b: R = Ph, EWG = COOEt
1c: R = 4-MeC6H4, EWG = COOMe
1d: R = 2-ClC6H4, EWG = COOMe
1e: R = 2-MeOC6H4, EWG = COOMe
1f: R = 2,6-Cl2C6H3, EWG = COOMe
- HbPdOAc or
- HcPdOAc
OH
OH
1g: R =
n
-pentyl, EWG = COOMe
EWG
EWG
R
R
2a: PhI
2b: 4-MeC6H4I
2c: 2-MeC6H4I
Ar
Ar
3a-i (
3a-i (Z)
E)
Scheme 1.
Table 1
The reaction of 1a and iodobenzene (2a) under various conditions
OH
O
O
COOMe
Ph
COOMe
Ph
COOMe
Ph
Ph
Ph
Ph
conditions
1a
2a
+
3a (E/Z
)
4a
6a
Entry
Conditionsa
Productsb (%)
1
2
3
4
5
6
Pd(OAc)2 (5 mol %)/PPh3 (10 mol %)/TBAB (1.0 equiv)/KOAc (3.0 equiv)/CH3CN/reflux/24 h
Pd(OAc)2 (5 mol %)/PPh3 (10 mol %)/KOAc (3.0 equiv)/CH3CN/reflux/7 h
Pd(OAc)2 (5 mol %)/TBAB (1.0 equiv)/KOAc (3.0 equiv)/CH3CN/reflux/12 h
Pd(OAc)2 (5 mol %)/TBAB (1.0 equiv)/K2CO3 (3.0 equiv)/CH3CN/reflux/4 h
Pd(OAc)2 (10 mol %)/TBAB (1.0 equiv)/Et3N (3.0 equiv)/CH3CN/reflux/1 h
Pd(OAc)2 (10 mol %)/TBAB (2.0 equiv)/KOAc (3.0 equiv)/CH3CN/reflux/24 h
1a (5)/3a (45)/4a (42)
1a (30) /3a (15)/4a (50)
1a (14)/3a (50)/4a (30)
*
*
No 1a/3a (36)/4a (34)/6a (18)
*
*
*
1a (20) /3a (15) /4a (60)
*
*
*
No 1a/3a (50) /4a (15) /6a (15)
8
9
10
a
Pd(OAc)2 (15 mol %)/TBAB (1.0 equiv)/KOAc (0.3 equiv)/CH3CN/reflux/24 h
Pd(OAc)2 (15 mol %)/TBAB (1.0 equiv)/KOAc (3.0 equiv)/DMF/80 °C/24 h
Pd(OAc)2 (15 mol %)/TBAB (1.0 equiv)/KOAc (3.0 equiv)/CH3CN/reflux/24 h
1a (21)/3a (42)/4a (24)
*
*
No 1a/3a (50) /4a (25)
*
*
*
1a (30) /3a (35) /4a (10)
1a (1.0 mmol) and 2a (2.0 mmol) were used and bromobenzene (2d) was used in entry 10.
Isolated yield and the yield marked with asterisk were estimated on TLC and decarboxylated compound 5a (Table 3) was observed in some cases in
b
trace amounts.
Table 2
similar results (entries 1–5). In all the cases, desired Heck
products 3a–e were isolated in 61–68% yields and b-keto
esters 4a–e in 21–27% yields. It is interesting to note that
the reaction of 1d and 2a under the same conditions pro-
duced 3f in 82% yield (entry 6). We could not observe the
formation of the corresponding b-keto ester 4f. The results
could be explained well by the increased steric hindrance
around the proton Ha during the last elimination stage in
intermediate (I) to increase the formation of Heck product.
Similar tendency was observed in 2-methoxy and 2,6-
dichloro derivatives (entries 7 and 8). Moreover, Z-isomer
was the predominant one (43–76:3–19) for the Baylis–
Hillman adducts having ortho-substitutent, presumably
due to the steric hindrance between the two aryl moieties.
In addition, we obtained Heck product 3i as the sole prod-
uct in 54% yield when we use the Baylis–Hillman adduct 1g
Heck reaction of Baylis–Hillman adducts to prepare b-aryl Baylis–
Hillman adductsa
Entry
Substrates
Products (%)
1
2
3
4
5
6
7
8
9
a
1a + 2a
1a + 2b
1a + 2c
1b + 2a
1c + 2a
1d + 2a
1e + 2a
1f + 2a
1g + 2a
3a-E (26)
3b-E (28)
3c-E (35)
3d-E (23)
3e-E (24)
3f-E (10)b
3g-E (19)
3h-E (3)c
3i-E (30)
3a-Z (36)
3b-Z (38)
3c-Z (33)
3d-Z (38)3a
3e-Z (38)
3f-Z (72)
3g-Z (43)
3h-Z (76)
3i-Z (24)
4a (25)4a
4b (24)4a
4c (21)
4d (24)
4e (27)4a
4f (not)
4g (15)4a
4h (not)
4i (not)
Conditions: compound
1
(1.0 equiv), compound
2
(2.0 equiv),
Pd(OAc)2 (15 mol %), TBAB (1.0 equiv), KOAc (3.0 equiv), CH3CN,
reflux, 24 h.
b
Contaminated with some 3f-Z.
Contaminated wih some 1f.
c

3250
J. M. Kim et al. / Tetrahedron Letters 49 (2008) 3248–3251
Table 3
The reaction of 1a and iodobenzene (2a)/bromobenzene (2d) under reported conditions
O
OH
O
O
COOMe
Ph
COOMe
Ph
COOMe
Ph
Ph
Ph
Ph
Ph
Ph
3a (E/Z
)
4a
5a
6a
Entry
Conditions
Productsa (%)
1
10
1a/2d (2.0 equiv)/Pd(OAc)2 (2 mol %)/TBAB (1.0 equiv)/NaHCO3 (2.5 equiv)/THF/reflux/7 h
1a/2d (2.0 equiv)/Pd(OAc)2 (2 mol %)/TBAB (1.0 equiv)/NaHCO3 (2.5 equiv)/THF/reflux/17 h
4a (81)4a
1a (15) /3a (5) /4a (65)
*
*
2
20
1a/2d (not mentioned)/Pd(OAc)2 (1 mol %)/PPh3 (2 mol %)/Et3N (1.25 equiv)/sealed tube/100 °C/20 h
1a/2d (2.0 equiv)/Pd(OAc)2 (1 mol %)/PPh3 (2 mol %)/Et3N (1.25 equiv)/sealed tube/100 °C/20 h
4a (48)/5a (28)4b
1a (5) /3a (5) /4a (28)/5a (6)
*
*
3
1a/2a (2.0 equiv)/Pd(OAc)2 (2.5 mol %)/TBAB (0.5 equiv)/KOAc (0.3 equiv)/DMF/70 °C/3 h
1a/2a (2.0 equiv)/Pd(OAc)2 (2.5 mol %)/TBAB (0.5 equiv)/KOAc (0.3 equiv)/DMF/70 °C/24 h
1a/2a (2.0 equiv)/Pd(OAc)2 (10 mol %)/TBAB (0.5 equiv)/KOAc (0.3 equiv)/DMF/70 °C/24 h
4a (79)4c
1a (50) /4a (22)
*
30
300
a
1a (28)/3a (39)/no 4a/no 5a/6a (13)
Isolated yield and the yield marked with asterisk are estimated yield on TLC.
derived from hexanal. The results might be due to relatively
lower acidity of the corresponding Ha to make the elimina-
tion of HaPdOAc difficult to form b-keto ester 4i. The reac-
tion of methyl (2-hydroxymethyl)cinnamate (1h) and 2a
was not effective and we obtained small amounts of the
corresponding acetate (8%), and 1h was recovered in 54%
yield.
From the whole results we could draw some trends for
the Heck reaction of Baylis–Hillman adducts: (i) the yield
of Heck product was increased when we use relatively
larger amounts of Pd(OAc)2, (ii) larger R (Scheme 1) made
the elimination of HaPdOAc difficult and increase the Heck
product, (iii) excess amounts of KOAc made the reaction
more effective, and (iv) Heck product could be the sole
product for the substrate having less acidic Ha.
(ca. 5%), and Heck product 3a (ca. 5%) (entry 20).
However, when we repeated the third condition (entry
3),4c we observed somewhat different results. Under the
exactly same conditions (entry 30), we obtained b-keto ester
4a in only 22% and 1a was remained in about 50% yield.
When we increased the amount of Pd(OAc)2 to 10 mol %
(entry 300), we observed the formation of Heck product 3a
(20%) and an oxidized compound 6a (20%), and we could
not observe the formation of b-keto ester 4a at all.
In summary, we prepared some b-aryl Baylis–Hillman
adducts via the Heck type reaction of Baylis–Hillman
adduct and aryl iodide under the influence of Pd(OAc)2
(15 mol %)/TBAB (1.0 equiv)/KOAc (3.0 equiv) in reflux-
ing CH3CN in a moderate yield as a cleanly separable
E/Z mixture.
At the earliest stage of this study, we examined the
reported conditions on the Pd-mediated reaction of
Baylis–Hillman adduct and aryl bromide/iodide.4a–c The
following three conditions were repeated in some cases
without much problem in our hands, but showed much dis-
crepancy to the reported results in some cases. The results
are summarized in Table 3.
Acknowledgments
This work was supported by the Korea Research
Foundation Grant funded by the Korean Government
(MOEHRD, KRF-2007-313-C00417). Spectroscopic data
were obtained from the Korea Basic Science Institute,
Gwangju branch.
(i) PhBr/Pd(OAc)2 (2 mol %)/TBAB (1.0 equiv)/NaHCO3
(2.5 equiv)/THF/reflux.4a
References and notes
(ii) PhBr/Pd(OAc)2 (1 mol %)/PPh3 (2 mol %)/Et3N
(1.25 equiv)/sealed tube/100 °C.4b
1. For the general review on Baylis–Hillman reaction, see: (a) Basavaiah,
D.; Rao, A. J.; Satyanarayana, T. Chem. Rev. 2003, 103, 811–891; (b)
Ciganek, E. In Organic Reactions; Paquette, L. A., Ed.; John Wiley &
Sons: New York, 1997; Vol. 51, pp 201–350; (c) Basavaiah, D.; Rao, P.
D.; Hyma, R. S. Tetrahedron 1996, 52, 8001–8062; (d) Kim, J. N.; Lee,
K. Y. Curr. Org. Chem. 2002, 6, 627–645; (e) Lee, K. Y.; Gowrisankar,
S.; Kim, J. N. Bull. Korean Chem. Soc. 2005, 26, 1481–1490 and
references cited therein.
2. For our recent chemical transformations involving Baylis–Hillman
adducts, see: (a) Gowrisankar, S.; Lee, H. S.; Lee, K. Y.; Lee, J.-E.;
Kim, J. N. Tetrahedron Lett. 2007, 48, 8619–8622; (b) Gowrisankar, S.;
Lee, H. S.; Kim, J. M.; Kim, J. N. Tetrahedron Lett. 2008, 49, 1670–
1673; (c) Lee, H. S.; Kim, S. H.; Kim, T. H.; Kim, J. N. Tetrahedron
Lett. 2008, 49, 1773–1776.
(iii) PhI/Pd(OAc)2 (2.5 mol %)/TBAB (0.5 equiv)/KOAc
(0.3 equiv)/DMF/70 °C.4c
When we carried out the reaction of 1a and 2a under the
first condition (entry 1),4a almost similar results were
observed (entry 10). However, the reaction was not com-
pleted even after 17 h and we recovered the remaining 1a
(15%) and observed the formation of small amounts of
Heck product 3a (5%). When we used the second condition
(entry 2),4b we observed the formation of b-keto ester 4a
(28%), decarboxylated product 5a (6%),4b remaining 1a

J. M. Kim et al. / Tetrahedron Letters 49 (2008) 3248–3251
3251
3. For the synthesis of b-branched Baylis–Hillman adducts, see: (a)
Ramachandran, P. V.; Rudd, M. T.; Burghardt, T. E.; Reddy, M. V.
R. J. Org. Chem. 2003, 68, 9310–9316; (b) Shanmugam, P.; Rajasingh,
P. Chem. Lett. 2005, 1494–1495; (c) Concellon, J. M.; Huerta, M. J.
Org. Chem. 2005, 70, 4714–4719.
4. For the intermolecular palladium-mediated Heck type reactions of
Baylis–Hillman adducts, see: (a) Basavaiah, D.; Muthukumaran, K.
Tetrahedron 1998, 54, 4943–4948; (b) Sundar, N.; Bhat, S. V. Synth.
Commun. 1998, 28, 2311–2316; (c) Kumareswaran, R.; Vankar, Y. D.
Synth. Commun. 1998, 28, 2291–2302; (d) Perez, R.; Veronese, D.;
Coelho, F.; Antunes, O. A. C. Tetrahedron Lett. 2006, 47, 1325–1328;
(e) Kabalka, G. W.; Venkataiah, B.; Dong, G. Org. Lett. 2003, 5,
3803–3805.
m/z 269 (M++H). Anal. Calcd for C17H16O3: C, 76.10; H, 6.01. Found:
C, 76.38; H, 6.27.
Compound 3a-E: 26%; colorless oil; IR (film) 3515, 2953, 1713, 1693
cmꢀ1; 1H NMR (CDCl3, 300 MHz) d 3.75 (s, 3H), 4.06 (d, J = 11.4 Hz,
1H), 5.88 (d, J = 11.4 Hz, 1H), 7.24–7.43 (m, 10H), 7.96 (s, 1H); 13C
NMR (CDCl3, 75 MHz) d 52.07, 69.72, 125.43, 127.26, 128.41, 128.68,
129.10, 129.22, 132.37, 134.19, 141.84, 142.66, 168.01; ESIMS m/z 269
(M++H). Anal. Calcd for C17H16O3: C, 76.10; H, 6.01. Found: C,
76.31; H, 6.39.
Compound 3i-Z: 24%; colorless oil; IR (film) 3446, 2931, 1716,
;
1225 cmꢀ1 1H NMR (CDCl3, 300 MHz) d 0.89 (t, J = 6.9 Hz, 3H),
1.26–1.52 (m, 6H), 1.67–1.74 (m, 2H), 2.36 (d, J = 6.3 Hz, 1H), 3.67 (s,
3H), 4.40 (d, J = 6.3 Hz, 1H), 6.84 (s, 1H), 7.23–7.34 (m, 5H); 13C
NMR (CDCl3, 75 MHz) d 13.99, 22.53, 25.39, 31.59, 36.16, 51.75,
74.62, 128.20 (2C), 128.24, 133.27, 135.35, 136.40, 169.44; ESIMS m/z
263 (M++H). Anal. Calcd for C16H22O3: C, 73.25; H, 8.45. Found: C,
73.33; H, 8.72.
5. Tsuji, J. Palladium Reagents and Catalysts; John Wiley & Sons:
Chichester, 2004.
6. Typical procedure for the synthesis of 3a: A stirred solution of 1a
(192 mg, 1.0 mmol), 2a (408 mg, 2.0 mmol), Pd(OAc)2 (34 mg,
15 mol %), TBAB (322 mg, 1.0 mmol), KOAc (294 mg, 3.0 mmol) in
CH3CN (3 mL) as heated to reflux for 24 h. After the usual aqueous
workup and column chromatographic purification process (hexanes/
EtOAc, 7:1), we obtained compounds 3a-Z (97 mg, 36%), 3a-E (70 mg,
26%), and 4a (67 mg, 25%) as colorless oils. Other compounds were
synthesized similarly and the structures were identified by their
spectroscopic data. Representative spectroscopic data of prepared
compounds 3a-Z, 3a-E, 3i-Z, and 3i-E are as follows.
Compound 3i-E: 30%; white solid, mp 57–58 °C; IR (film) 3527, 2954,
1697, 1250 cmꢀ1 1H NMR (CDCl3, 300 MHz) d 0.85 (t, J = 6.9 Hz,
;
3H),1.16–1.49 (m, 6H), 1.60–1.72 (m, 1H), 1.85–1.96 (m, 1H), 3.32 (d,
J = 11.4 Hz, 1H), 3.85 (s, 3H), 4.63–4.71 (m, 1H), 7.26–7.42 (m, 5H),
7.69 (s, 1H); 13C NMR (CDCl3, 75 MHz) d 13.96, 22.55, 25.73, 31.52,
36.76, 51.94, 68.94, 128.49, 128.71, 129.06, 134.15, 134.63, 140.23,
168.33; ESIMS m/z 263 (M++H). Anal. Calcd for C16H22O3: C, 73.25;
H, 8.45. Found: C, 73.56; H, 8.42.
Compound 3a-Z: 36%; colorless oil; IR (film) 3469, 3028, 1728,
;
1713 cmꢀ1 1H NMR (CDCl3, 300 MHz) d 3.00 (d, J = 5.7 Hz, 1H),
7. E and Z isomers could be separated cleanly without loss in most
cases, and the chemical shifts of the vinyl protons of E isomers
3.54 (s, 3H), 5.60 (d, J = 5.7 Hz, 1H), 6.92 (s, 1H), 7.24–7.45 (m, 10H);
13C NMR (CDCl3, 75 MHz) d 51.67, 75.60, 126.56, 127.98, 128.16,
128.34, 128.38, 128.50, 135.16, 135.26, 135.41, 140.92, 169.08; ESIMS
appeared at 7.69–7.99 ppm and Z isomers at 6.57–7.11 ppm,
which are the characteristic chemical shifts ranges of this type
compounds.1–3