Synthesis of 1-aryl-1-phenylpropenes using
an alkylation-rearrangement-methylation-isomerization
one-pot reaction sequence
(d, J = 2.1, 3H), 6.71 (d, J = 2.1, 1H), 7.25 (m, 5H); 13
C
derived from a double [3,3]-sigmatropic rearrangement
NMR δ 14.0, 16.0, 24.8, 38.1, 55.9, 60.5, 111.9, 123.3, [9]). Indeed, adding Me4NCl and K2CO3 to the reaction
124.8, 126.6, 126.7, 128.2, 133.7, 138.4, 139.2, 142.5, mixture of para-methoxyphenyl allyl ether in PEG
145.2, 152.9. Selected 1H NMR data for the (E)-isomer: resultedinthesimultaneousO-methylationoftheClaisen
δ 0.95 (t, J = 7.3, 3H), 1.84 (d, J = 7.1, 3H), 3.30 (s, 3H), rearrangement product when heating for 3 h at 180°C,
3.80 (s, 3H), 5.90 (q, J = 7.1, 1H).
but the presence of the base partially caused a double-
(Z/E)-2,6-Dimethoxy-1-(1-phenylprop-1-en-1-yl) bond migration in the resulting ortho-allylanisole (e.g.
naphthalene (6c). The (E/Z) mixture was obtained as about half of the product isomerized to the E/Z mixture
an amorphous solid that partially crystallized to a white of the ortho-propenylanisoles). Such a rearrangement-
powder over several days (1.70 g, 56%). Mp = 56–66°C; methylation-isomerization sequence could not be
IR (KBr) 1627, 1597, 1507, 1494, 1460, 1422, 1375, optimized to give the ortho-methoxy propenylbenzenes
1341, 1253 cm-1; MS (ESI) m/z 305 (MH+); HRMS (ESI) as the sole products, because the isomerization step
calcd for C21H21O2 [M+H]+ 305.1542, found 305.1548. was inefficient and furthermore led to an inseparable
1
NMR data for the major (Z)-isomer: H NMR δ 1.51 (d, mixture of diastereomers.
J = 6.9, 3H), 3.78 (s, 3H), 3.88 (s, 3H), 6.57 (q, J = 6.9,
Such a portion of double-bond migration is surprising
1H), 7.00–7.35 (m, 8H), 7.60 (d, J = 9.1, 1H), 7.75 (d, considering the relatively low basicity of the K2CO3. In
J = 9.1, 1H); 13C NMR δ 15.7, 55.5, 57.1, 77.2, 106.2, our previous work [8], a stronger base like NaOH had
115.0, 119.6, 123.4, 126.0, 126.1, 126.8, 127.0, 127.8, to be used at 150°C in PEG for the O-methylation of
128.4, 128.6, 129.0, 130.3, 136.5, 141.8, 152.9, 156.3. allylphenols like eugenol and 2-allylphenol, when such
1
Selected H NMR data for the (E)-isomer: δ 2.02 (d, J isomerization was desirable (K2CO3 was still ineffective
= 7.1, 3H), 3.71 (s, 3H), 3.93 (s, 3H), 5.80 (q, J = 7.1, at this temperature). However, we rationalized that if
1H).
using phenyl cinnamyl ethers as substrates, instead
(Z)-6-Bromo-2-methoxy-1-(1-phenylprop-1-en-1- of the plain allyl ethers, the reaction could be more
yl)naphthalene (6d). The (E/Z) mixture was obtained efficient and the isomerization in the last step would be
as an off-white powder (1.90 g, 54%), which upon complete: (a) the intermediate ether 3a rearranges faster
recrystallization from n-hexane gave the pure (Z)-isomer. [9]; (b) the double-bond isomerization of 5a should also
Mp = 124–127°C; 1H NMR δ 1.49 (d, J = 6.9, 3H), 3.80 be considerably easier due to the higher acidity of the
(s, 3H), 6.57 (q, J = 6.9, 1H), 7.17 (m, 5H), 7.35 (d, J benzhydrylic hydrogen in comparison with the benzylic
= 9.0, 1H), 7.39 (dd, J = 2.0, 9.0, 1H), 7.56 (d, J = 9.0, one; (c) the E : Z ratio could be more biased toward one
1H), 7.76 (d, J = 9.0, 1H), 7.96 (d, J = 2.0, 1H); 13C NMR isomer; (d) and therefore, the reaction should proceed
δ 15.5, 56.6, 114.8, 117.3, 119.7, 122.8, 125.7, 126.3, faster toward the end product and thus, overall, be less
126.7, 127.0, 128.1, 128.2, 129.8, 130.2, 131.5, 135.7, prone to side reactions.
141.2, 154.3; IR (KBr) 1584, 1494, 1459, 1437, 1333,
To further expand the scope of such a one-pot
1270 cm-1; MS (EI) m/z: 354 (M+, 76), 352 (M+, 76), 273 reaction we included the aryl cinnamyl ether formation
(100), 258 (57), 195 (31), 91 (46), 74 (56); Anal. Calcd as the first step of the reaction sequence. Allyl
for C20H17BrO: C 68.00, H 4.85. Found: C 67.99, H 4.90. halides like 2 are quite reactive toward phenolates at
Selected 1H NMR data for the (E)-isomer: δ 2.02 (d, J = temperatures much lower than approximately 150°C,
7.1, 3H), 3.73 (s, 3H), 5.79 (q, J = 7.1, 1H).
as required for the Me4NCl-based methylation. In this
way a complementary use of the two alkylating reagents
together in the same reaction mixture is made possible
duetotheirtemperature-dependentreactivity.Anattempt
was thus made by slowly heating a mixture of para-
methoxyphenol (1a), cinnamyl chloride (2), Me4NCl and
K2CO3 in PEG, from room temperature to 180°C, and
keeping it at this temperature for one hour (Scheme 2).
The reaction was much cleaner and chromatographic
separation gave (Z/E)-6a as a viscous oil. The ratio of
diastereomers was 10 : 1, but their configuration could
not be unambiguously assigned based on the NMR
analysis of the chromatographically inseparable mixture.
This same problem was encountered by another group
of researchers who obtained an inseparable 85 : 15
mixture of (Z/E)-6a by the FeCl3-catalyzed alkenylation
3. Results and discussion
Polar and protic solvents were found to enhance the
reaction rate of the aromatic Claisen rearrangement
[15]. Ethyleneglycol, mono- and diethyl ethers of
diethyleneglycol, which are solvents related to PEG,
were already found to be efficient, but we could find no
examples using PEG for this purpose. Our preliminary
experiments with para-methyl or para-methoxyphenyl
allyl ethers confirmed that PEG is an efficient solvent for
the Claisen rearrangement when heating the reaction
mixture for a minute at 230°C (para substitution was
preferred in order to avoid the common side products
906