Highly selective stereodivergent synthesis of separable amide rotamers, by using Pd chemistry, and their thermodynamic behavior

Article abstract of DOI:10.1002/chem.200802627

By using Pd chemistry, a highly selective stereodivergent synthesis of separable amide rotamers was achieved. Allylation of 2,4,6-tri-tert-bu- tylanilides using a π-allyl-Pd catalyst gave N-allylated anilides with moderate-to-excellent Z-rotamer selectivi

Full text of DOI:10.1002/chem.200802627

DOI: 10.1002/chem.200802627
Highly Selective Stereodivergent Synthesis of Separable Amide Rotamers, by
Using Pd Chemistry, and Their Thermodynamic Behavior
Nobutaka Ototake,[b] Masashi Nakamura,[a] Yasuo Dobashi,[b] Haruhiko Fukaya,[b] and
Osamu Kitagawa*[a]
Abstract: By using Pd chemistry, a
highly selective stereodivergent synthe-
sis of separable amide rotamers was
achieved. Allylation of 2,4,6-tri-tert-bu-
tylanilides using a p-allyl–Pd catalyst
gave N-allylated anilides with moder-
ate-to-excellent Z-rotamer selectivity
(Z/E=3 to >50). The Z-rotamer selec-
tivity depends considerably on the sub-
stituent on the anilide substrates. On
the other hand, the E rotamers of N-al-
lylated 2,4,6-tri-tert-butylanilides were
obtained with almost complete selectiv-
ity (E/Z>50) through O-allylation of
2,4,6-tri-tert-butylanilides and the sub-
sequent PdII-catalyzed Claisen rear-
rangement of the resulting O-allyl imi-
date. The prepared anilide rotamers
changed to equilibrium mixtures in
which the E rotamer was the major
isomer when heated in toluene.
Keywords: allylation
palladium
·
amides
·
·
·
rotamers
stereodivergence
Introduction
of N-methyl-2,6-di-tert-butylaniline, requires a high reaction
temperature (100–1408C) because of the low reactivity of
the aniline [Eq. (2)], a mixture of E and Z rotamers ((E)-I
and (Z)-I) was obtained in a near-equilibrium ratio (E/Z=
5.6–9). Thus, the stereoselective synthesis of the thermody-
namically unstable (Z)-I and highly selective synthesis of
(E)-I have not yet been established. Since Chupp et al. have
also found that (Z)-I has potent biological activity (inhibi-
tory activity against anthocyanin synthesis) in comparison
with (E)-I,[2b] the stereoselective synthesis of the Z rotamer
should be especially crucial.
ꢀ
Rotational isomers based on an amide C(O) N bond play
an important role in the regulation of actions of biologically
active peptides and functional molecules that have amide
skeletons.[1] Although amide rotamers are often observed by
means of NMR spectra, in general, they are difficult to iso-
late because the interconversion between two rotamers
occurs easily at RT.
In 1967, Chupp et al. reported that an amide bond of 2,6-
di-tert-butylanilide derivatives has a high rotational barrier
(27–28 kcalmolꢀ1) and the rotamers can be isolated at RT
[Eq. (1)].[2] Although this work should be noted as a rare ex-
ample of separable amide rotamers,[3] a systematic investiga-
tion using anilide substrates other than a-haloacetanilides I
(X=Cl, Br, I) has not been performed [Eq. (2)]. Moreover,
since the synthetic method of I, through the haloacetylation
[a] M. Nakamura, Prof. O. Kitagawa
Department of Applied Chemistry
Shibaura Institute of Technology
3-7-5 Toyosu, Koto-ku, Tokyo 135-8548 (Japan)
Fax : (+81)3-5859-8161
E-mail: kitagawa@shibaura-it.ac.jp
[b] N. Ototake, Prof. Y. Dobashi, H. Fukaya
School of Pharmacy
Tokyo University of Pharmacy and Life Sciences
1432-1 Horinouchi, Hachioji, Tokyo 192-0392 (Japan)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802627.
5090
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5090 – 5095

FULL PAPER
In connection with our study on the conformational analy-
sis of atropisomeric o-tert-butylanilides,[4,5] we had an inter-
est in 2,6-di-tert-butylanilide derivatives. In this paper, we
report on the kinetically controlled stereodivergent synthesis
of separable 2,6-di-tert-butylanilide rotamers.[6] The present
synthetic methods using p-allyl–Pd chemistry and PdII-cata-
lyzed Claisen chemistry give Z and E rotamers of N-allylat-
ed 2,6-di-tert-butylanilides with high stereoselectivity, re-
spectively [Eq. (3)]. An investigation of the isomerization of
these anilide rotamers under thermal conditions is also de-
scribed.
After surveying several reactions, we found that the N-al-
lylation of 1a efficiently proceeded by using a p-allyl–Pd
complex.[8] That is, when an amide anion prepared from 1a
and NaH (2.3 equiv) was treated with allyl acetate in the
presence of a rac-BINAP/[{Pd
ACHTUNGRTENN(UNG allyl)Cl}2] (BINAP=2,2’-bis-
AHCTUNGTRENNUNG
RT, anilide 2a was obtained in a good yield (80%) without
the formation of imidate 3a (Scheme 1). Furthermore, in
the present reaction, (Z)-2a was obtained as the major rota-
mer (Z/E=4).
Results and Discussion
Stereoselective synthesis of anilide Z rotamers: In contrast
to N-alkylated anilide derivatives, 2,6-di- and 2,4,6-tri-tert-
butylanilides are known to exist as an equilibrium mixture
in which Z is the major isomer (Z/Effi8) at RT in solution.[7]
Accordingly, we expected that the Z rotamer may be ob-
tained as a major isomer through the N-alkylation of 2,6-di-
or 2,4,6-tri-tert-butylanilide. N-Allylation with anilide 1a
prepared from commercially available 2,4,6-tri-tert-butylani-
line was initially examined. However, the reaction of the
amide anion from 1a and NaH with allyl bromide in DMF
gave O-allylated imidate 3a in a good yield in place of the
desired N-allylated anilide 2a [Eq. (4); in THF, the reaction
did not proceed to result in the quantitative recovery of 1a].
This unusual O-allylation should be specific to 2,6-di-tert-
butyl derivatives, because allylation with o-tert-butylanilide
under the same conditions gave an N-allylated product in a
good yield (93%).
The formation of such an O-alkylation product was also ob-
served in the reaction of 1a with iodomethane. In this case,
a mixture of the N- and O-methylation products 4a and 5a
was obtained in 55 and 43% yields, respectively [Eq. (4)].
The steric crowding around the amide nitrogen atom by two
o-tert-butyl groups should cause the unusual O-alkylation of
1a.
Scheme 1. The N-allylation of 2,4,6-tri-tert-butylanilide using a p-allyl–Pd
catalyst and its possible mechanism.
By means of careful monitoring using TLC analyses, it
was found that in the present reaction, imidate 3a is initially
Chem. Eur. J. 2009, 15, 5090 – 5095
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5091

O. Kitagawa et al.
Table 2. Z-Selective N-allylation of various 2,4,6-tri-tert-butyl anilides 1.
formed and then the resulting 3a changes to anilide 2a via
reproduction of the p-allyl–Pd complex and anilide anion
(Scheme 1). As far as we know, in allylations of amide deriv-
atives using a p-allyl–Pd complex, there have been no re-
ports on a reaction that proceeds through such an O-allyla-
tion and subsequent O,N-allylic rearrangement.[9]
The yield of 2a and the Z-rotamer selectivity remarkably
depended on the reaction solvent and stoichiometry of NaH
(Table 1). For example, although the reaction in N,N-di-
Entry
1
R
2
Yield [%][a]
Z/E[b]
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1 f
1g
1h
Et
Me
Me2CH
cyclohexyl
MeO2C
p-Br-C6H4
(E)-MeCH=CH
(E)-MeCH=CMe
2a
2b
2c
2d
2e
2 f
2g
2h
99
91
99
81
85
38
70
75
1
4.9
3.2
>50.0
>50.0
>50.0
>50.0
3.0
Table 1. The effect of the solvent in the allylation of 2,4,6-tri-tert-butyl-
ACHTUNGTRENNUNGanilide 1a.
>50.0
[a] Isolated yield. [b] The ratio was estimated by using H NMR spectros-
copy (300 MHz).
Entry
Solvent
2a yield
(Z)-2a/
3a yield
[%][a]
(E)-2a[b]
[%][a]
pends considerably on the R substituent of anilide substrates
1. The N-allylation reactions of 1c–1 f and 1h that possess a
relatively bulky R substituent, such as isopropyl, cyclohexyl,
ester, phenyl, and methylpropenyl groups, gave (Z)-2 with
excellent selectivities ((Z)-2/(E)-2>50; entries 3–6 and 8).
On the other hand, with anilides 1a, 1b, and 1g that possess
a less-bulky R substituent, such as ethyl, methyl, and pro-
penyl groups, a decrease in the Z selectivity was observed
((Z)-2/(E)-2=3.0–4.9; entries 1, 2, and 7).
1
2
3
4
5
6
DMF
DMA
CH3CN
toluene
THF
80
79
7
5
90
15
4.0
3.2
6.1
2.1
1.0
4.0
0
0
0
0
0
DMF
(1.2 equiv NaH)
39
1
[a] Isolated yield. [b] The ratio was estimated by using H NMR spectros-
copy (300 MHz).
The Z selectivity observed in the present reaction may be
rationalized in accordance with the transition-state model
shown in Scheme 2. The transition state (TS-E) for (E)-2
may be disfavored in comparison with TS-Z for (Z)-2 be-
methylACHTUNGTRENNUNGacetamide (DMA) gave a similar result to that in
DMF (entry 2), when CH3CN and toluene were used as a
solvent, the reaction hardly proceeded (entries 3 and 4). In
THF, 2a was obtained in a good yield (90%) and Z-rotamer
selectivity was not observed (E/Z=1, entry 5).
On the other hand, the use of 1.2 equiv of NaH resulted
in a remarkable decrease in the chemical yield of 2a (15%)
due to the formation of imidate 3a (39%) and the recovery
of starting material 1a (entry 6). Thus, to get 2a in a good
yield, the use of more than 2 equiv of NaH is required. NaH
may play an important role in the conversion of 3a to 2a as
well as in the formation of the anilide anion. Indeed, treat-
ment of isolated 3a with the BINAP/Pd catalyst in DMF did
not produce anilide 2a, whereas smooth conversion to 2a
was observed by the addition of NaH (1.0 equiv) to the re-
action mixture (Scheme 1).
The use of 1,1’-bis(diphenylphosphino)ferrocene (dppf) as
a phosphine ligand led to a further increase in the chemical
yield. In this case, 2a was obtained with an excellent yield
(99%) at a ratio of Z/E=4.9 (Table 2, entry 1). Under opti-
mized conditions (NaH (2.3 equiv), allyl acetate (1.5 equiv),
Scheme 2. The transition-state model for Z-selective N-allylation.
dppf (5.0 mol%), and [{Pd
(allyl)Cl}2] (2.2 mol%) in DMF at
cause of the steric repulsion between the R substituent and
the o-tert-butyl group in the iminoalcoholate intermediate.
In the reactions of 1c–1 f and 1h, which possess a bulky R
substituent, excellent Z selectivity would be observed by fur-
ther destabilization of TS-E.
RT), N-allylations of various 2,4,6-tri-tert-butylanilides 1
were examined further. As shown in Table 2, the reactions
proceeded with good-to-excellent yields (70–99%; entries 1–
5, 7, and 8) except for that of benzamide 1 f (38%; entry 6).
Furthermore, in all reactions, the preferential formation of
the Z rotamer was observed. The Z-rotamer selectivity de-
5092
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5090 – 5095

Separable Amide Rotamers
FULL PAPER
Table 3. Synthesis of O-allyl imidates 3.
Stereoselective synthesis of anilide E rotamers: Thermo-
ACHTUNGTRENNUNGdynamically stable (E)-2 can be obtained as a major rotamer
by thermal isomerization.[2] However, the equilibrium ratio
under thermal conditions was not so high (see Table 5
below). In particular, in anilides 1c and 1d that have a
bulky R substituent, such as isopropyl and cyclohexyl
groups, the preference of E rotamers considerably decreased
(E/Z=1.4). Thus, to selectively get an E rotamer of various
anilides, the development of a kinetically controlled stereo-
selective synthetic method is required. As mentioned above,
N-allylation of anilides 1 using a p-allyl–Pd catalyst proceed-
ed via the formation of O-allyl imidate 3 followed by the re-
action of the p-allyl–Pd complex and anilide anion repro-
duced from the resulting 3 (Scheme 1). On the other hand,
it has been well known that in the presence of a PdII catalyst
such as PdCl2, the O-allyl imidate is converted to an N-allyl
amide through the Claisen mechanism but not the p-allyl–
Pd mechanism.[10] Accordingly, we expected that the highly
selective synthesis of the E rotamer may be achieved by a
Entry
1
R
3
Yield [%][a]
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1 f
1g
1i
Et
Me
3a
3b
3c
3d
3e
3 f
3g
3i
94
96
93
82
70
73
86
95
Me2CH
cyclohexyl
MeO2C
p-Br-C6H4
(E)-MeCH=CH
2-furanyl
[a] Isolated yield.
Table 4. Stereoselective synthesis of E rotamer 2 through a PdII-catalyzed
Claisen rearrangement.
PdII-catalyzed Claisen rearrangement of imidate
(Scheme 3).
3
Entry
3
R
2
Yield [%][a]
E/Z[b]
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3 f
3g
3i
Et
Me
2a
2b
2c
2d
2e
2 f
2g
2i
95
94
94
95
86
96
88
95
1
>50
>50
>50
>50
>50
>50
>50
>50
Me2CH
cyclohexyl
MeO2C
p-Br-C6H4
(E)-MeCH=CH
2-furanyl
[a] Isolated yield. [b] The ratio was estimated by using H NMR spectros-
copy (300 MHz).
Scheme 3. The highly selective synthesis of the E rotamer through a PdII-
catalyzed Claisen mechanism.
in Table 4, high chemical yields (86–96%) and almost com-
plete stereoselectivities (E/Z>50) were observed. The ex-
cellent E-rotamer selectivity of the present reaction, which
does not depend on substituent R, should be especially
noted.
Thus, we succeeded in the highly selective stereodivergent
synthesis of Z and E rotamers of 2,4,6-tri-tert-butylanilide
derivatives by using p-allyl–Pd chemistry and Pd-catalyzed
Claisen chemistry. Although the stereoselective synthesis of
separable amide rotamers has so far been reported by sever-
al groups,[11,3e–f] to the best of our knowledge, there has been
no report on kinetically controlled stereodivergent synthesis.
Various O-allyl imidates 3 were prepared in good yields
through allylation of anilides 1 as shown in Equation (4)
(see above). That is, the reaction of the amide anion from
anilides 1a–1d, 1g, and 1i with allyl bromide proceeded
with high O-allylation selectivity to give imidates 3 in good
yields (Table 3, entries 1–4, 7, and 8), whereas in the reac-
tions of 1e and 1 f, a slight decrease in chemical yields, due
to the formation of N-allylation product 2, was observed
(entries 5 and 6).
Subsequently, a Claisen rearrangement with O-allyl imi-
date 3 was investigated. It was found that when imidate 3a
was treated with [PdCl2ACHTNUTRGNENUG(PhCN)2] (5 mol%) at RT in
CH2Cl2, N-allylanilide 2a was obtained in an excellent yield
(95%; Table 4, entry 1). Furthermore, the reaction proceed-
ed with almost complete stereoselectivity to give (E)-2a as a
single rotamer. The present reaction can be applied to vari-
ous imidate substrates 3a–3i. That is, in all reactions shown
Thermodynamic behavior and stereochemistry of anilide ro-
tamers: Next, the thermodynamic behavior of the various
N-allylanilides 2a–2h was investigated. Anilides 2a–2h ex-
isted without isomerization between the E and Z rotamers
for several days at RT. However, anilides 2a–2h with a
major Z isomer changed to an equilibrium mixture in which
the E isomer predominated when heated for 10–30 h at
Chem. Eur. J. 2009, 15, 5090 – 5095
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5093

O. Kitagawa et al.
1008C in toluene (Table 5).[12] The equilibrium ratios (E/Z=
1.4–10.1) depended considerably on substituent R. The ther-
mal stabilities of (E)-2 may be rationalized on the basis of
Table 5. Thermal stabilities of anilide rotamers.
Scheme 4. The X-ray crystal structure of (E)-2d.
Entry
2
R
Heating time [h]
E/Z[a]
the allylic hydrogen atoms and the a-hydrogen atoms or
ester Me group in the Z rotamers was stronger than that in
the E rotamers.
1
2
3
4
5
6
7
8
2a
2b
2c
2d
2e
2 f
2g
2h
Et
Me
Me2CH
cyclohexyl
MeO2C
p-Br-C6H4
(E)-MeCH=CH
(E)-MeCH=CMe
12
20
11
30
30
20
10
10
6.0
10.1
1.4
1.4
4.0
3.5
4.9
2.2
Conclusion
In conclusion, we developed a highly selective stereodiver-
gent synthesis of separable rotamers of 2,4,6-tri-tert-butyl-
anilide derivatives through use of Pd chemistry. The present
work should be noted as a first example of kinetically con-
trolled stereodivergent synthesis of separable amide rotam-
ers. In addition, the interesting mechanism of the Z-selective
N-allylation, which proceeded through O-allylation and sub-
sequent O,N-allylic rearrangement, and the thermodynamic
stabilities of the various amide rotamers, were clarified.
Thus, the present work should provide broad interest from
the viewpoint of structural organic chemistry and palladium
chemistry as well as being a unique stereoselective reaction
(rotamer-selective reaction).
[a] The ratio was estimated by using 1H NMR spectroscopy (300 MHz).
n–p repulsion between the lone pairs on the carbonyl
oxygen and aromatic ring, and steric repulsion between the
R and allyl groups (see the scheme in Table 5).[5a,13] Namely,
the destabilization of the Z rotamer due to both n–p repul-
sion and steric repulsion may bring about the E-rotamer
preference.[4,14] In the case of anilide substrates that have a
bulky R substituent, a decrease in the E/Z ratio of 2 should
be observed because of the strong steric repulsion between
R and the tert-butylphenyl group in (E)-2.
The equilibrium ratios ((E)-2b/(Z)-2b=10.1; (E)-2c/(Z)-
2c=1.4) of anilide rotamers 2b and 2c shown in Table 5
(entries 2 and 3) agree well with the values based on E/Z
energy differences (DGE–Z =DGEꢀDGZ) estimated by MO
calculations. These ab initio calculations with the HF/6-
31G* basis set gave E/Z energy differences (DGE–Z =ꢀ1.706
and ꢀ0.130 kcalmolꢀ1 for 2b and 2c, respectively) that cor-
responded to the values (E)-2b/(Z)-2b=10.0 and (E)-2c/
(Z)-2c=1.2 at 373 K.[15] Thus, the thermal stabilities of the
anilide rotamers were also supported by MO calculations.
The stereochemistries of the anilide rotamers were deter-
Experimental Section
1
The melting points given are uncorrected. H and 13C NMR spectra were
recorded on 400 and 100 MHz spectrometers, respectively. The chemical
shifts in the 1H and 13C NMR spectra are expressed downfield from
CHCl3 (d=7.26 ppm) and CDCl3 (d=77.0 ppm), respectively. Mass spec-
tra were recorded by electron impact or chemical ionization. Column
chromatography was performed on silica gel (75–150 mm). Medium-pres-
sure liquid chromatography (MPLC) was performed on a 30ꢁ4 cm (i.d.)
prepacked column (silica gel, 50 mm) with a UV detector.
1
mined based on chemical shifts in H NMR spectra. In gen-
eral, hydrogen atoms on substituent R in (E)-2 appeared at
a higher field than those of (Z)-2 because of the anisotropy
effect caused by the tert-butylphenyl group having a large
twist angle.[2] For example, in a-hydrogen atoms of (E)-2a–
2d and 2g, ester methyl hydrogen atoms of (E)-2e, and o-
hydrogen atoms of (E)-2 f, high field shifts of d=0.15–
0.53 ppm were observed. For further confirmation of the
General procedure for the Pd-catalyzed N-allylation of 1: Under an Ar
atmosphere, NaH (228 mg, 5.7 mmol, 60% assay) was added to 1b
(752 mg, 2.48 mmol) in DMF (10 mL). After being stirred for 5 min at
RT, a suspension of [{PdACTHNUTRGNEUNG(allyl)Cl}2] (19.8 mg, 0.054 mmol), dppf (68.7 mg,
0.124 mmol), and allyl acetate (0.4 mL, 3.72 mmol) in DMF (3.0 mL) was
added to the mixture, and then the reaction mixture was stirred for 15 h
at RT. The mixture was poured into a 2% aqueous HCl solution and ex-
tracted with AcOEt. The AcOEt extracts were washed with brine, dried
over MgSO4, and evaporated to dryness. Purification of the residue by
column chromatography (hexane/AcOEt=12) gave a mixture of (Z)-2b
and (E)-2b (787 mg, 92%, E/Z=3.2). (Z)-2b (more polar) and (E)-2b
(less polar) were separated by means of MPLC (hexane/AcOEt=7).
stereoACHTUNGTRENNUNGchemistry, X-ray crystal analysis of (E)-2d and a
NOESY experiment of 2a and 2e were performed. The X-
ray crystal structure of (E)-2d showed the E conformational
structure and large twisting of the tert-butylphenyl group
(twist angle of tert-butylphenyl group=878; Scheme 4).[16] In
the NOESY experiment of 2a and 2e, the NOE between
(Z)-N-Allyl-N-(2,4,6-tri-tert-butylphenyl)acetamide ((Z)-2b): M.p. 141–
1438C; 1H NMR (CDCl3): d=7.39 (s, 2H), 5.41 (tdd, J=6.7, 10.2,
17.0 Hz, 1H), 5.16 (qd, J=1.3, 17.0 Hz, 1H), 5.02 (qd, J=1.3, 10.2 Hz,
5094
www.chemeurj.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5090 – 5095

Separable Amide Rotamers
FULL PAPER
1H), 4.23 (td, J=1.3, 6.7 Hz, 2H), 2.11 (s, 3H), 1.37 (s, 18H), 1.28 ppm
(s, 9H); 13C NMR (CDCl3): d=172.9, 148.3, 146.4, 132.1, 131.7, 125.4,
118.5, 56.2, 37.3, 34.6, 33.1, 31.3, 23.7 ppm; IR (KBr): n˜ =1649 cmꢀ1; MS:
m/z: 344 [M ++H]; elemental analysis calcd (%) for C23H37NO: C 80.41,
H 10.86, N 4.08; found: C 80.33, H 10.58, N 3.96.
Chim. Acta 1978, 61, 512–516; f) D. Seebach, W. Wykypiel, W. Lu-
bosch, H. O. Kalinowski, Helv. Chim. Acta 1978, 61, 3100–3102;
g) K. C. Rice, A. Brossi, J. Org. Chem. 1980, 45, 592–601; h) L. T.
Tan, R. T. Williamson, W. H. Gerwick, J. Org. Chem. 2000, 65, 419–
425.
[4] O. Kitagawa, M. Yoshikawa, H. Tanabe, T. Morita, M. Takahashi, Y.
Dobashi, T. Taguchi, J. Am. Chem. Soc. 2006, 128, 12923–12931.
[5] For papers on conformational analysis of o-tert-butylanilide deriva-
tives by other groups, see: a) D. P. Curran, G. R. Hale, S. J. Geib, A.
Balog, Q. B. Cass, A. L. G. Degani, M. Z. Hemandes, L. C. G. Frei-
tas, Tetrahedron: Asymmetry 1997, 8, 3955–3975; b) A. D. Hughes,
D. A. Price, N. S. Simpkins, J. Chem. Soc. Perkin Trans. 1 1999,
1295–1304; c) K. Kondo, T. Iida, H. Fujita, T. Suzuki, K. Yamagu-
chi, Y. Murakami, Tetrahedron 2000, 56, 8883–8891; d) S. Dantale,
V. Reboul, P. Metzner, C. Philouze, Chem. Eur. J. 2002, 8, 632–640;
e) J. Clayden, L. Lemiꢂgre, M. Pickworth, L. Jones, Org. Biomol.
Chem. 2008, 6, 2908–2913.
[6] For a preliminary communication of this work, see: N. Ototake, T.
Taguchi, O. Kitagawa, Tetrahedron Lett. 2008, 49, 5458–5460.
[7] a) H. Kessler, A. Rieker, Liebigs Ann. Chem. 1967, 708, 57–68;
b) G. V. Zyryanov, E. M. Hampe, D. M. Rudkevich, Angew. Chem.
2002, 114, 4010–4013; Angew. Chem. Int. Ed. 2002, 41, 3854–3857.
It is well known that in secondary amides, Z rotamers are thermody-
namically more stable than E rotamers.
(E)-N-Allyl-N-(2,4,6-tri-tert-butylphenyl)acetamide ((E)-2b): M.p. 70–
738C; H NMR (CDCl3): d=7.42 (s, 2H), 5.38 (tdd, J=6.8, 10.2, 17.1 Hz,
1
1H), 5.19 (qd, J=1.5, 17.1 Hz, 1H), 5.01 (qd, J=1.5, 10.2 Hz, 1H), 4.27
(brd, J=6.8 Hz, 2H), 1.85 (s, 3H), 1.34 (s, 18H), 1.31 ppm (s, 9H);
13C NMR (CDCl3): d=171.7, 149.4, 146.7, 132.9, 132.1, 125.9, 118.8, 53.5,
37.6, 34.7, 33.4, 31.3, 24.4 ppm; IR (KBr): n˜ =1655 cmꢀ1; MS: m/z: 344
[M ++H]; elemental analysis calcd (%) for C23H37NO: C 80.41, H 10.86,
N 4.08; found: C 80.34, H 10.75, N 3.98.
General procedure for the synthesis of O-allyl imidate 3: Under an Ar
atmosphere, NaH (28 mg, 0.69 mmol, 60% assay) was added to 1a
(95 mg, 0.3 mmol) in DMF (3 mL). After being stirred for 5 min at RT,
allyl bromide (39 mL, 0.45 mmol) was added to the mixture, and then the
reaction mixture was stirred for 3 h at RT. The mixture was poured into a
solution of NH4Cl and was extracted with AcOEt. The AcOEt extracts
were washed with brine, dried over MgSO4, and evaporated to dryness.
Purification of the residue by column chromatography (hexane only and
then hexane/AcOEt=50) gave 3a as a colorless oil (101 mg, 94%).
Allyl
N-(2,4,6-tri-tert-butylphenyl)propanimidate
(3a):
1H NMR
(CDCl3): d=7.24 (s, 2H), 6.11 (tdd, J=5.4, 10.6, 17.2 Hz, 1H), 5.43 (qd,
J=1.5, 17.2 Hz, 1H), 5.27 (qd, J=1.5, 10.6 Hz, 1H), 4.91 (td, J=1.5,
5.4 Hz, 2H), 1.84 (q, J=7.6 Hz, 2H), 1.33 (s, 18H), 1.32 (s, 9H),
1.00 ppm (t, J=7.6 Hz, 3H); 13C NMR (CDCl3): d=160.8, 144.3, 142.9,
138.5, 134.0, 121.6, 116.7, 65.7, 36.0, 34.6, 31.7, 31.4, 24.9, 8.6 ppm; IR
(neat): n˜ =1668 cmꢀ1; MS: m/z: 358 [M ++H]; elemental analysis calcd
(%) for C24H39NO: C 80.61, H 10.99, N 3.92; found: C 80.77, H 10.68, N
3.82.
[8] a) J. Tsuji, J. Synth. Org. Chem. Jpn. 1999, 57, 1036–1050; b) B. M.
Trost, Acc. Chem. Res. 1980, 13, 385–393.
[9] The allylation of o-tert-butylanilide using a p-allyl–Pd catalyst di-
rectly gave an N-allylated product without the formation of an O-al-
lylation product. a) O. Kitagawa, M. Kohriyama, T. Taguchi, J. Org.
Chem. 2002, 67, 8682–8684; b) J. Terauchi, D. P. Curran, Tetrahe-
dron: Asymmetry 2003, 14, 587–592.
[10] For typical examples, see: a) L. E. Overman, Acc. Chem. Res. 1980,
13, 218–224; b) T. Ikariya, Y. Ishikawa, K. Hirai, S. Yoshikawa,
Chem. Lett. 1982, 1815–1818; c) T. G. Schenck, B. Bosnich, J. Am.
Chem. Soc. 1985, 107, 2058–2066; d) M. Calter, T. K. Hollis, L. E.
Overman, J. Ziller, C. G. Zipp, J. Org. Chem. 1997, 62, 1449–1456.
[11] a) P. Beak, W. J. Zajdel, J. Am. Chem. Soc. 1984, 106, 1010–1018;
b) D. R. Hay, Z. Song, S. G. Smith, P. Beak, J. Am. Chem. Soc. 1988,
110, 8145–8153; c) F. Yokokawa, H. Sameshima, T. Shioiri, Synlett
2001, 0986–0988; d) S. Deng, J. Taunton, J. Am. Chem. Soc. 2002,
124, 916–917.
General procedure for the Pd-catalyzed aza-Claisen rearrangement of
imidate 3: Under an Ar atmosphere, [PdCl2ACTHNUTRGNEUNG(PhCN)2] (7.2 mg, 0.02 mmol)
was added to 3a (135 mg, 0.38 mmol) in CH2Cl2 (3 mL), and then the re-
action mixture was stirred for 2 h at RT. The CH2Cl2 solvent was evapo-
rated to dryness. Purification of the residue by column chromatography
(hexane/AcOEt=10) gave (E)-2a (128.3 mg, 95%).
(E)-N-Allyl-N-(2,4,6-tri-tert-butylphenyl)propanamide (E-2a): M.p. 74–
1
768C; H NMR (CDCl3): d=7.42 (s, 2H), 5.39 (tdd, J=6.7, 10.3, 17.0 Hz,
1H), 5.20 (dd, J=1.5, 17.0 Hz, 1H), 5.02 (dd, J=1.5, 10.3 Hz, 1H), 4.27
(d, J=6.7 Hz, 2H), 2.03 (q, J=7.3 Hz, 2H), 1.32 (s, 18H), 1.32 (s, 9H),
1.08 ppm (t, J=7.3 Hz, 3H); 13C NMR (CDCl3): d=174.4, 149.3, 146.8,
132.2, 132.2, 126.0, 118.6, 53.6, 37.5, 34.7, 33.4, 31.2, 29.2, 8.6 ppm; IR
(KBr): n˜ =1645 cmꢀ1; MS: m/z: 358 [M ++H]; elemental analysis calcd
(%) for C24H39NO: C 80.61, H 10.99, N 3.92; found: C 80.36, H 10.76, N
3.62.
[12] N-Allylanilides 2a–2h in toluene were heated until the changes of
E/Z ratios became negligible (reaction time described in Table 5).
On the other hand, although thermal isomerization of franyl deriva-
tive 2i (E rotamer) was also investigated, the exact equilibrium
ratio ((E)-2i/(Z)-2i) could not be determined, because of the intra-
molecular Diels–Alder products generated from (Z)-2i.
[13] For typical papers on the E-rotamer preference of N-alkylated ani-
lide derivatives, see: a) B. F. Pederson, B. Pederson, Tetrahedron
1956, 2995–3001; b) A. Itai, Y. Toriumi, N. Tomioka, H. Kagechika,
I. Azumaya, K. Shudo, Tetrahedron Lett. 1989, 30, 6177–6180; c) I.
Azumaya, K. Yamaguchi, I. Okamoto, H. Kagechika, K. Shudo, J.
Am. Chem. Soc. 1995, 117, 9083–9084.
[14] The E-rotamer preference of N-methylanilides has been rationalized
on the basis of such an n–p interaction: S. Saito, Y. Toriumi, N. To-
mioka, A. Itai, J. Org. Chem. 1995, 60, 4715–4720.
Acknowledgements
This work was partly supported by a Grant-in-Aid for Scientific Research
and the Ministry of Education, Science, Sports and Culture of Japan.
[15] All calculations were carried out by using the GAMESS (US) pack-
age: M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S.
Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su,
T. L. Windus, M. Dupuis, J. A. Montgomery, J. Comput. Chem. 1993,
14, 1347–1363.
[16] CCDC 713422 ((E)-2d) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
[1] a) A. Tanatani, I. Azumaya, H. Kagechika, J. Synth. Org. Chem.
Jpn. 2000, 58, 556–567; b) R. P. Cheng, S. H. Gellman, W. F. DeGra-
do, Chem. Rev. 2001, 101, 3219–3232; c) D. J. Hill, M. J. Mio, R. B.
Prince, T. S. Hughes, J. S. Moore, Chem. Rev. 2001, 101, 3893–4011.
[2] a) J. P. Chupp, J. F. Olin, J. Org. Chem. 1967, 32, 2297–2303;
b) W. T. Molin, C. A. Porter, J. P. Chupp, K. Naylor, Pespect. Chem.
Biochem. Phys. 1990, 36, 277–280.
[3] For examples of separable amide rotamers, see: a) A. Mannschreck,
Tetrahedron Lett. 1965, 6, 1341–1347; b) L. Toldy, L. Radics, Tetra-
hedron Lett. 1966, 7, 4753–4757; c) H. A. Staab, D. Lauer, Tetrahe-
dron Lett. 1966, 7, 4593–4598; d) H. A. Staab, D. Lauer, Chem. Ber.
1968, 101, 864–878; e) R. Schlecker, D. Seebach, W. Lubosch, Helv.
Received: December 13, 2008
Published online: March 30, 2009
Chem. Eur. J. 2009, 15, 5090 – 5095
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5095