Microwave-assisted allylation of acetals with allyltrimethylsilane in the presence of CuBr

Article abstract of DOI:10.1021/jo049015w

We describe the first synthesis of homoallyl ethers from acetals and allyltrimethylsilane using microwave heating and CuBr as a promoter. This method works best for aromatic acetals, giving the corresponding homoallyl ethers in good to quantitative yield.

Full text of DOI:10.1021/jo049015w

SCHEME 1
Micr ow a ve-Assisted Allyla tion of Aceta ls
w ith Allyltr im eth ylsila n e in th e P r esen ce
of Cu Br
Michael E. J ung* and Andreas Maderna
We first studied the allylation of benzaldehyde dimeth-
yl acetal with allyltrimethylsilane as a model reaction.
Among the various metal salts screened, we found that
stoichiometric amounts of CuBr promoted the allylation
when the starting materials were heated in a conven-
tional microwave reactor in anhydrous 1,2-dichloroethane
for 60 min at 100 °C (see Experimental Section for a
detailed procedure). Different solvents were examined for
this reaction, but only dichloromethane and 1,2-di-
chloroethane could be successfully used, with the latter
being preferred due to its higher boiling point. In diethyl
ether or THF, the allylations did not take place, presum-
ably due to competitive binding of the solvent to the CuBr
during the reaction. With catalytic amounts of CuBr, the
product yields were low, and microwave heating of the
starting materials in the absence of CuBr was unsuc-
cessful. Interestingly, benzaldehyde itself did not give the
allylation under these conditions, and only starting
materials could be recovered. We also tested allyltribut-
ylstannane, which is known to be an excellent allylation
agent for various carbonyl compounds.1b However, in the
allylation of both benzaldehyde and its dimethyl acetal
with allyltributylstannane, no product formation was
detected.6 Table 1 summarizes the results of the allyla-
tion of various acetals with allyltrimethylsilane.7 Aro-
matic acetals 1a -c and 1e gave the best product forma-
tion, with isolated yields of the corresponding homoallyl
ethers ranging from 82 to 100%. From a comparison of
the yields of 3a -e, it is apparent that the electron-
withdrawing nitro substituent on the aromatic ring
disfavors the reaction. With the chloro and bromo sub-
stituents in 1b and 1c, the corresponding homoallyl
ethers 3b and 3c are formed in slightly decreased yields
compared to the unsubstituted 1a . Nitro derivative 3d
is formed in only moderate yield, reflecting the strong
electron-withdrawing effect of the nitro substituent present
in 1d . In contrast, acetal 1e, having a propyl substituent
in the para position on the aromatic ring, gave 3e in
quantitative yield, which can be attributed to the electron-
donating character of the alkyl substituent.
Department of Chemistry and Biochemistry, University of
California, Los Angeles, California 90095-1569
jung@chem.ucla.edu
Received J une 10, 2004
Abstr a ct: We describe the first synthesis of homoallyl
ethers from acetals and allyltrimethylsilane using micro-
wave heating and CuBr as a promoter. This method works
best for aromatic acetals, giving the corresponding homoallyl
ethers in good to quantitative yield.
The allylation of carbonyl compounds and their deriva-
tives with allyltrimethylsilane is an important method
for the synthesis of homoallyl alcohols and ethers.1 For
aldehydes, this reaction can be catalyzed by fluoride ions
or Lewis acids, whereas for acetals and ketals, only Lewis
acids, such as TiCl4,2 TMSOTf,3 and others,4 are known
to catalyze or promote this reaction. We recently de-
scribed a mixed Lewis acid system comprised of AlBr3
and CuBr, which acts as a potent catalyst for the
allylation of aromatic and aliphatic acetals and ketals
with allyltrimethylsilane.5 To scavenge unwanted HBr
that is formed or present in the reaction, a small amount
of AlMe3 was added to the reaction mixtures. Despite the
efficiency of this and other previously described Lewis
acids for this process, a method not requiring the use of
highly reactive and acidic reagents would be desirable
since it might help to preserve sensitive functional groups
present in the starting materials. In this paper, we
describe the first allylation of acetals 1 with allyltrim-
ethylsilane 2 in the absence of strong Lewis acids, a
reaction promoted by microwave heating in the presence
of CuBr, to generate the desired homoallyl ethers 3
(Scheme 1).
(1) For reviews, see: (a) Mukaiyama, T.; Murakami, M. Synthesis
1987, 1043-1054. (b) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93,
2207.
(2) Hosomi, A.; Endo, M.; Sakurai, H. Chem. Lett. 1976, 941-942.
(3) (a) Tsunoda, T.; Suzuki, M.; Noyori, R. Tetrahedron Lett. 1980,
21, 71-74. (b) Murata, S.; Suzuki, M.; Noyori, R. J . Am. Chem. Soc.
1980, 102, 3248. (c) Zerth, H. M.; Leonard, N. M.; Mohan, R. S. Org.
Lett. 2003, 5, 55-57.
Next, we examined aliphatic acetals 1f and 1g and
obtained the expected products 3f and 3g in yields of 58
and 65%, respectively. Longer reaction times or larger
amounts of CuBr did not improve the product yields. The
lowest yield was with cyclic ketal 1h , which provided 3h
in only 21% yield. Interestingly, the only other product
detected in the crude reaction mixture was cyclohexanone
(79% by GC), resulting from cleavage of the ketal function
in 1h .
(4) (a) For AlCl3 and BF3-Et2O: Hosomi, A.; Endo, M.; Sakurai, H.
Chem. Lett. 1978, 499. (b) For trityl perchlorate and Ph2BOTf:
Mukaiyama, T.; Nagaoka, H.; Murakami, M.; Ohshima, M. Chem. Lett.
1985, 977-980. (c) For montmorillonite: Kawai, M.; Onaka, M.; Izumi,
Y. Chem. Lett. 1986, 381-384. (d) For TMSI: Sakurai, H.; Sasaki, K.;
Hosomi, A. Tetrahedron Lett. 1981, 22, 745-748. (e) For TiCp2(OTf)2:
Hollis, T. H.; Robinson, N. P.; Whelan, J .; Bosnich, B. Tetrahedron
Lett. 1993, 34, 4309-4312. (f) For BiBr3: Komatsu, N.; Domae, T.;
Wada, M. Tetrahedron Lett. 1997, 38, 7215-7218. (g) For Bi(OTf)3:
Wieland, L. C.; Zerth, H. M.; Mohan, R. S. Tetrahedron Lett. 2002, 43,
4597-4600. (h) For TMSNTf2: Ishii, A.; Kotera, O.; Saeki, T.; Mikami,
K. Synlett 1997, 1145-1146. (i) For Sc(OTf)3: Yadav, J . S.; Subba
Reddy, B. V.; Srihari, P. Synlett 2001, 673-675.
(6) We have no good explanation for why the allylstannane does not
undergo the same allyl transfer as does the allylsilane, although we
observed similar results in our earlier Lewis acid-catalyzed allylations.5
(7) Compounds 3d and 3e are new compounds; the other products
are known in the literature. For 3a , see ref 3c. For 3b, see ref 4g. For
3c, see ref 4i. For 3f, see ref 4f. For 3g, see: Barentsen, H. M.; Sieval,
A. B.; Cornelisse, J . Tetrahedron 1995, 51, 7495. For 3h , see ref 2.
(5) J ung, M. E.; Maderna, A. Tetrahedron Lett. 2004, 45, 5301-04.
10.1021/jo049015w CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/05/2004
J . Org. Chem. 2004, 69, 7755-7757
7755

TABLE 1. Allyla tion of Aceta ls 1a -h w ith
Allyltr im eth ylsila n e 2 Usin g Micr ow a ve Hea tin g in th e
P r esen ce of Cu Br
strong Lewis acids using microwave heating and CuBr
as a promoter to produce the homoallyl ether 3. The
reaction works best for aromatic acetals in the absence
of strong electron-withdrawing substituents on the aro-
matic ring.
Exp er im en ta l Section
Gen er a l In for m a tion . All reactions were carried out using
a CEM Corporation Focused Microwave System, Model Discover.
1,2-Dichloroethane was dried over phosphorus pentoxide and
distilled. All commercial reagents were used directly. NMR data
were obtained on a 500 MHz spectrometer and IR data on an
FT-IR spectrometer.
Rep r esen ta tive P r oced u r e. In a typical experiment, 1.5
mmol of CuBr was placed in a 10 mL glass pressure vial
equipped with a stir bar. The pressure vial was closed using a
PTFE silicon septum. Anhydrous 1,2-dichloroethane was added
into the vial followed by 1.5 mmol of acetal 1 and 2.25 mmol of
allyltrimethylsilane 2. The suspension was heated to 100 °C
while being stirred for 60 min in the CEM microwave reactor
described above. For acetals 1f-h , 1.5 mmol of the acetal, 2.2
mmol of CuBr, and 3 mmol of silane 2 were used. After the
completion of the reaction, the suspension was filtered, the
volatiles were removed in vacuo, and the crude oils were purified
by flash chromatography. The isolated yields of homoallyl ethers
3 are given in Table 1. We have not carried out the experiment
on any scale larger than the 1.5 mmol scale described above due
to the size limitations of the commercial microwave equipment
we are using, although we would expect that with a larger
instrument, one could carry out the experiment on a much larger
scale.
4-Meth oxy-4-p h en yl-1-bu ten e (3a ): 1H NMR (500 MHz,
CDCl3) δ 7.4-7.3 (m, 5H, Ar), 5.8 (m, 1H), 5.1 (m, 2H), 4.2 (t,
1H, J ) 5.9 Hz), 3.3 (s, 3H), 2.6 (m, 1H), 2.5 (m, 1H); 13C NMR
(126 MHz, CDCl3) δ 141.7, 134.8, 128.3, 127.6, 126.7, 116.9, 83.6,
56.6, 42.5; IR (neat) ν 3028, 2980, 2936, 2821, 1641, 1454, 1357,
1100, 915, 700 cm-1
.
4-(4-Ch lor op h en yl)-4-m eth oxy-1-bu ten e (3b): 1H NMR
(500 MHz, CDCl3) δ 7.3 (d, 2H, J ) 8.4 Hz), 7.2 (d, 2H, J ) 8.3
Hz), 5.7 (m, 1H), 5.0 (m, 2H), 4.1 (t, 1H, J ) 6.7 Hz), 3.2 (s, 3H),
2.5 (m, 1H), 2.3 (m, 1H); 13C NMR (126 MHz, CDCl3) δ 140.0,
134.2, 133.1, 128.4, 128.2, 117.1, 82.8, 56.6, 42.3; IR (neat) ν
Under normal conditions, CuBr is not generally rec-
ognized as a Lewis acid. For example, in most Lewis
acidity scales,8 it is not included. Indeed, in one of the
few reports determining the Lewis acidity of Cu(I) salts,9
it is listed as the weakest of all salts examined. Thus, it
is fair to say that under usual conditions, it behaves as,
at best, a very weak Lewis acid. However, under our
microwave conditions, it appears to act as a very mild,
but efficient, Lewis acid, promoting the formation of the
carbocation (oxocarbenium ion) by complexation with the
oxygen atom of the acetal. The lack of reaction in strongly
coordinating solvents, where it would be completely
complexed, and the fact that stoichiometric amounts of
CuBr are required are also in agreement with this
hypothesis. Also, normal thermal heating of the reaction
mixture does not afford the products observed under
microwave irradiation.10The precise mechanism for its
activation under microwave irradiation remains to be
determined.
2982, 2934, 2823, 1598, 1598, 1489, 1090, 1015 cm-1
.
4-(4-Br om op h en yl)-4-m eth oxy-1-bu ten e (3c): 1H NMR
(500 MHz, CDCl3) δ 7.4 (d, 2H, J ) 8.3 Hz), 7.1 (d, 2H, J ) 8.3
Hz), 5.7 (m, 1H), 5.0 (m, 2H), 4.1 (t, 1H, J ) 6.5 Hz), 3.2 (s, 3H),
2.5 (m, 1H), 2.3 (m, 1H); 13C NMR (126 MHz, CDCl3) δ 140.7,
134.2, 131.4, 128.4, 121.3, 117.3, 83.0, 56.7, 42.3; IR (neat) ν
3077, 2980, 2931, 2821, 1641, 1591, 1485, 1404, 1344, 1104,
1010, 917 cm-1
.
4-(3-Nitr op h en yl)-4-m eth oxy-1-bu ten e (3d ): 1H NMR (500
MHz, CDCl3) δ 8.2 (s, 1H), 8.1 (m, 1H), 7.6 (d, 1H, J ) 7.6 Hz),
7.5 (m, 1H), 5.7 (m, 1H), 5.0 (m, 2H), 4.3 (t, 1H, J ) 6.6 Hz), 3.2
(s, 3H), 2.5 (m, 1H), 2.4 (m, 1H); 13C NMR (126 MHz, CDCl3) δ
148.4, 140.3, 133.4, 132.9, 129.3, 123.4, 122.0, 117.9, 82.7, 57.0,
42.2; IR (neat) ν 2932, 2830, 1530, 1349, 1108, 1058 cm-1
EI-LRES m/z ) 207 [M+] (100).
;
4-(4-P r op ylp h en yl)-4-m eth oxy-1-bu ten e (3e): 1H NMR
(500 MHz, CDCl3) δ 7.2 (d, 2H, J ) 8.0 Hz), 7.1 (d, 2H, J ) 8.0
Hz), 5.8 (m, 1H), 5.0 (m, 2H), 4.2 (t, 1H, J ) 6.1 Hz), 3.4 (m,
2H), 2.6 (t, 3H, J ) 4.4 Hz), 2.4 (m, 1H), 1.6 (m, 1H), 1.2 (t, 3H,
J ) 7.0 Hz), 1.0 (t, 3H, J ) 7.4 Hz); 13C NMR (126 MHz, CDCl3)
δ 141.8, 139.5, 135.0, 128.4, 126.4, 116.4, 81.5, 63.9, 42.3, 37.6,
24.4, 15.2, 13.8; IR (neat) ν 2962, 2931, 2870, 1720, 1092, 917
cm-1. EI-HRES calcd 204.1470. Found 204.1461.
In summary, we have described the allylation of acetals
1 with allyltrimethylsilane 2 in the absence of normal
(8) (a) Branch, C. S.; Bott, S. G.; Barron, A. R. J . Organomet. Chem.
2003, 666, 23. (b) Laszlo, P.; Teston, M. J . Am. Chem. Soc. 1990, 112,
8750. (c) Brown, I. D.; Skowron, A. J . Am. Chem. Soc. 1990, 112, 3401.
(d) Childs, R. F.; Mulholland, D. L.; Nixon, A. Can. J . Chem. 1982, 60,
801. (e) Deters, J . F.; McCusker, P. A.; Pilger, R. C., J r. J . Am. Chem.
Soc. 1968, 90, 4583.
1
4-Meth oxy-1-u n d ecen e (3f): H NMR (500 MHz, CDCl3) δ
5.8 (m, 1H), 5.0 (m, 2H), 3.3 (s, 3H), 3.2 (m, 1H), 2.2 (m, 2H),
1.4-1.2 (m, 12H), 0.8 (m, 3H, J ) 6.2 Hz); 13C NMR (126 MHz,
CDCl3) δ 135.0, 116.7, 80.5, 56.5, 37.8, 33.4, 31.8, 29.7, 29.3, 25.3,
(9) Zhang, Y. Inorg. Chem. 1982, 21, 3889.
22.7, 14.1; IR (neat) ν 2955, 2927, 2856, 1462, 1098, 911 cm-1
.
(10) Although we have not investigated this thoroughly, simply
heating the mixture gives back mostly the starting materials with the
very slow formation of allylated products.
4-Meth oxy-5-p h en yl-1-p en ten e (3g): 1H NMR (500 MHz,
CDCl3) δ 7.2-7.3 (m, 5H), 5.8 (m, 1H), 5.1 (m, 2H), 3.4 (m, 1H),
7756 J . Org. Chem., Vol. 69, No. 22, 2004

3.3 (s, 3H), 2.8 (m, 1H), 2.7 (m, 1H), 2.2 (m, 2H); 13C NMR (126
MHz, CDCl3) δ 138.9, 134.7, 129.4, 128.2, 126.1, 117.1, 81.7, 57.0,
39.8, 37.5; IR (neat) ν 3027, 2928, 2823, 1641, 1495, 1454, 1359,
Ack n ow led gm en t. The authors thank the National
Science Foundation (CHE-0314591) for financial sup-
port.
1098, 914, 745, 700 cm-1
.
1-Meth oxy-1-(2-p r op en yl)cycloh exa n e (3h ): 1H NMR (500
MHz, CDCl3) δ 5.8 (m, 1H), 5.0 (m, 2H), 3.1 (s, 3H), 2.2 (d, 2H,
J ) 7.2 Hz), 1.2-1.6 (m, 10H); 13C NMR (126 MHz, CDCl3) δ
134.0, 117.1, 74.8, 48.1, 40.9, 33.8, 25.8, 21.8; IR (neat) ν 2931,
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and the 1H and 13C NMR spectra for all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
2855, 2823, 1639, 1455, 1081, 910 cm-1
.
J O049015W
J . Org. Chem, Vol. 69, No. 22, 2004 7757