A straightforward route to homoallyl-homocrotylamines promoted by a titanium complex

Article abstract of DOI:10.1002/ejoc.201201262

π-Allyltitanium complexes, generated in situ from 1,3-dienes and [Cp₂TiH], react with benzotriazole derivatives to give homoallylic amines in good yields. Under similar conditions, triple cascade reactions (allyltitanation followed by cationic 2-aza-Cope rearrangement followed by a second allyltitanation) occur from bis(benzotriazolyl) compounds affording a straightforward route to homoallyl-(E)-homocrotylamines. A theoretical study provides further insight into the factors that govern the selectivity of this sequence of reactions. The titanium-promoted reductive coupling of 1,3-dienes with bis(benzotriazolyl) compounds as substrates led selectively to homoallyl-homocrotylamines through a triple cascade reaction (allyltitanation - cationic 2-aza-Cope rearrangement - allyltitanation). Copyright

Full text of DOI:10.1002/ejoc.201201262

FULL PAPER
DOI: 10.1002/ejoc.201201262
A Straightforward Route to Homoallyl-Homocrotylamines Promoted by a
Titanium Complex
Stéphanie Toulot,[a] Quentin Bonnin,[a] Virginie Comte,[a] Louis Adriaenssens,[a]
Philippe Richard,[a] and Pierre Le Gendre*[a]
Keywords: Nitrogen heterocycles / Sigmatropic rearrangement / Allylation / Titanium / Allenes
π-Allyltitanium complexes, generated in situ from 1,3-dienes
and [Cp2TiH], react with benzotriazole derivatives to give
homoallylic amines in good yields. Under similar conditions,
triple cascade reactions (allyltitanation followed by cationic
2-aza-Cope rearrangement followed by a second allyltitan-
ation) occur from bis(benzotriazolyl) compounds affording a
straightforward route to homoallyl-(E)-homocrotylamines. A
theoretical study provides further insight into the factors that
govern the selectivity of this sequence of reactions.
Introduction
Efficient syntheses of homoallylamines represent a major
field of investigation in organic chemistry. This interest is
mainly due to the important role that homoallylamines play
as intermediates in the production of a large number of
products[1] including amino acids,[2] drugs,[3] alkaloids,[4]
and β-lactam derivatives.[5] Different synthetic methodolo-
gies have been developed, most frequently involving imines
Scheme 1. Equilibrium of benzotriazole adducts in solution.
Results and Discussion
We first synthesised several benzotriazole adducts 1–9 ac-
and allylating agents such as allylsilanes,[6] allylstannanes,[7] cording to the procedures described in the literature (Fig-
and allylboron species.[8] However, these syntheses often ure 1).[15–17] The reaction of formaldehyde with N-methyl-
suffer from the poor electrophilic character of imines, which aniline, dibenzylamine, N-methylbenzylamine, or dieth-
needs to be augmented by, for example, the presence of an ylamine and one equivalent of benzotriazole produced sub-
electron-withdrawing group[9] or Lewis acid activation.[10] strates 1–4, respectively.[15] Similarly, isobutyraldehyde and
Compounds with a benzotriazolyl group (Bt) α to an amino pyrrolidine gave product 5 in high yield.[15] Finally, the reac-
functionality, developed mainly by Katritzky and collabora- tion of benzylamine, phenylethylamine, isopropylamine, or
tors,[11] represent a class of very interesting alternative sub- cyclohexylamine with two equivalents of formaldehyde and
strates for this reaction. In solution, these adducts exist in two equivalents of benzotriazole afforded bis(benzotriazol-
equilibrium with a highly electrophilic imminium salt yl) compounds 6–9. These are of special interest because
(Scheme 1). In addition, these compounds are readily ac- they represent substrates containing two possible sites of
cessible, inexpensive, and stable towards moisture. As part reaction.[17]
of our ongoing research on the potential of butadiene deriv-
atives,[12] we recently started to explore the titanium-pro-
moted reductive coupling of 1,3-dienes with benzotriazole
compounds as an alternative approach to homallylic
amines.[13,14]
[a] Institut de Chimie Moléculaire de l’Université de Bourgogne,
ICMUB-UMR CNRS 6302, Université de Bourgogne,
9 avenue Alain Savary, BP 47870, 21078 Dijon Cedex, France
Fax: +33-3-80396098
Figure 1. Benzotriazole derivatives synthesised in this study.
E-mail: pierre.le-gendre@u-bourgogne.fr
With the benzotriazole derivatives in hand, we explored
the titanium-promoted reductive coupling of isoprene with
substrate 1 (Scheme 2). The π-allyltitanium complex was
Homepage: www.icmub.fr/335-equipe-ombc3
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201201262.
736
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 736–741

Homoallyl-Homocrotylamines
readily prepared by hydrotitanation of isoprene using Table 1. Reductive coupling of benzotriazole derivative 1–9 with
1,3-dienes promoted by [Cp2TiH].[a]
[Cp2TiH] generated in situ by the reaction of [Cp2TiCl2]
with isopropylmagnesium chloride.[18] The resulting purple
π-allyltitanium complex was added at 10 °C over three
hours by means of a syringe pump to a solution of benzotri-
azole compound 1 in tetrahydrofuran (THF).[19] After the
addition time, the solution was warmed to room tempera-
ture and stirred for one hour. Under these optimised condi-
tions, allylation occurred smoothly, giving rise to the de-
sired homoallylamine 10 in 86% yield. Notably, the analo-
gous reaction with an unactivated imine required a much
longer reaction time (12 h) at room temperature to go to
completion.[20] The substrate scope was then evaluated by
using benzotriazole compounds 2–9; the results are summa-
rised in Table 1. The reductive coupling of isoprene and
benzotriazole adducts 2 and 3 gave 11 and 12, respectively,
in good yields (entries 2 and 3). We also demonstrated the
feasibility of conducting the reaction with the more com-
plex diene myrcene (entries 4 and 5).
Scheme 2. Reductive coupling of benzotriazole derivative 1 with
isoprene.
The reductive coupling of benzotriazole compound de-
rived from aliphatic aldehyde 5 with isoprene led to forma-
tion of homoallylamine 15 in moderate yield and low dia-
stereoselectivity (Table 1, entry 6).[21] We then investigated
the double allyltitanation of bis(benzotriazolyl) compound
6. Surprisingly, a tertiary amine 16 with both homoallyl-
and homocrotyl fragments was formed as the major prod-
uct (entry 7) and the expected bis(homoallyl)amine 21
could only be detected in trace amounts (Ͻ5%; Figure 2).
Compound 16 was obtained as a single diastereoisomer and
the configuration of the crotyl fragment was determined to
be (E) according to NOE measurements. This reaction
could be extended to other bis(benzotriazolyl) derivatives
7–9 and could be also performed with myrcene in place
of isoprene (entries 8–11). In all cases, the homoallyl- (E)-
homocrotylamines were isolated as single products. The for-
mation of these products might be explained by two con-
comitant allylations, each with different yet defined regiose-
lectivity, on the iminium and CH2Bt groups (Figure 2).
However, this mechanism seems unlikely and, thus, we
prefer to explain the reactivity by invoking a mechanism
[a] Reaction conditions: THF (20 mL), benzotriazole derivative
(1.2 mmol), diene (2.4 mmol), [Cp2TiH] (1.2 mmol). [b] syn/anti ra-
tio 40:60. [c] [Cp2TiH] (2 equiv.), diene (4 equiv.) and bis(benzo-
triazolyl) derivative (1 equiv.) were used.
involving
a triple cascade reaction as illustrated in
Scheme 3. The allyltitanation of the iminium benzotriazole
intermediate B leads first to the expected homoallylamine
C. Due to the presence of the second benzotriazolyl frag-
ment, a second iminium D is generated directly after allyl-
ation. This creates a species perfectly set up for a cationic
2-aza-Cope rearrangement. This [3+3] sigmatropic reaction,
originally described by Geissman in 1950,[22] and beauti-
Figure 2. Left: Bis(homoallyl)amine 21, expected product from the
reductive coupling of isoprene and the bis(benzotriazolyl) com-
pound 6. Right: Iminium benzotriazole intermediate with two dif-
fully applied by Overman to natural products synthe- ferent potentially reactive sites toward allyltitanium complexes.
Eur. J. Org. Chem. 2013, 736–741
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
737

P. Le Gendre et al.
FULL PAPER
sis,[23,24] leads, in our case, to homocrotyl iminium salt E.
Finally, a second allyltitanation occurs affording, after hy-
drolysis, the homoallyl- homocrotylamine F.
Scheme 4. Calculated mechanism for the cationic 2-aza-Cope re-
arrangement (I and VI match compounds D and E, respectively, in
Scheme 3).
Table 2. DFT free-energies relative to V-a, in kcalmol–1 (RM1 free-
energies are reported in parentheses).
Scheme 3. Proposed mechanism for the formation of the homo-
allyl-homocrotylamines.
I
II
III
IV
V
VI
a
b
2.7
8.8
7.8
8.0
0
–0.9
(2.7)
5.1
(14.2)
12.0
(11.9)
11.2
(13.2)
11.4
(0)
1.3
(0.3)
(–1.6)
0.6
(3.0)
(15.9)
(14.9)
(16.9)
(–0.6)
The proposed mechanism would seem to suggest a vari-
ety of possible products and it is, therefore, notable that we
detect mainly one. Because the 2-aza-Cope rearrangement
is reversible, species E and D will exist in equilibrium. Allyl-
titanation should be possible on either species, however, we
detected mainly the product corresponding to allylation of
species E, implying that allyltitanation takes place after re-
arrangement. Furthermore, we observed only the (E)-
homocrotylamine, suggesting that the 2-aza-Cope re-
arrangement is highly selective. The detection of a single
product, despite the multitude of pathways available, in-
trigued us and we therefore performed a theoretical study
to gain insight into the factors that govern the selectivity of
this sequence of reactions.
We first investigated the potential energy surface with the
semiempirical RM1 Hamiltonian to find the most likely re-
action pathways. These pathways were then investigated in
detail by performing DFT B3LYP/6-31G** level calcula-
tions.[25–28] The model chosen for this study was the N-
benzyliminium cation (species D in Scheme 3 and species I
in Scheme 4). The results of the calculations suggest the
most stable conformation is the chair-like structure (I-a) in
which the methyl group in the β-position with respect to
the N-atom is located in a pseudoequatorial position. The
conformation with the same methyl group in the pseudoax-
ial position (I-b) is 2.4 kcalmol–1 higher in energy
(Table 2).[29] The orientation of this methyl group is indeed
crucial because it will define the final configuration (E or
Z) of the homocrotylamine. Both reaction paths (“a” and
“b”) exhibit the same features, proceeding via a short-lived
intermediate carbocation (III-a, III-b), which is probably
As shown in Table 2, path “a” is consistently favoured
over path “b”. This is in agreement with the fact that only
the (E) isomer of the homocrotyl fragment is experimentally
observed and is consistent with chair-like conformers gen-
erally preferring substituents to be in the pseudoequatorial
position.
The results of the calculations predict the product V-a is
more stable than the reactant I-a (ΔG = –2.7 kcalmol–1). A
further simple reorientation of the benzyl group leads to a
more stable conformation VI-a, further increasing the pre-
dicted ΔG between products and reactants to –3.6
kcalmol–1. This would lead to an equilibrium ratio of about
1000:2 in favour of VI (species E, Scheme 3) at the reaction
temperature (10 °C). The stabilisation obtained through the
rearrangement is likely due to the formation of a more sub-
stituted alkene. Furthermore the low activation energy of
6.1 kcalmol–1 is in the range of other cationic 2-aza-Cope
rearrangements previously reported[30] and suggests a fast
process. Thus, once species D is formed it rapidly enters
into an equilibrium heavily favouring species E before the
second equivalent of π-allyltitanium complex has the
chance to attack. Once it does attack, the π-allyltitanium
complex meets a 1000:2 mixture of E/D, producing the
homoallyl-homocrotylamine F as the predominant species.
Conclusions
We have described the titanium-promoted synthesis of
stabilised by hyperconjugation, and two transition states tertiary homoallylamines from 1,3-dienes and benzotriazole
(II-a, II-b and IV-a, IV-b) with geometries close to that of derivatives. The reaction occurred within three hours at
the intermediate. Similar energy profiles have previously 10 °C and led to the target homoallylamines in moderate to
been proposed for aza-Cope rearrangements involving good yields. The addition of two equivalents of allyltitan-
vinylsilanes.[30]
ium complexes to bis(benzotriazolyl) compounds led se-
738
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 736–741

Homoallyl-Homocrotylamines
N,N-Dibenzyl-2,3-dimethyl-3-butenylamine (11): Pale-yellow oil
(308 mg). 1H NMR (300 MHz, CDCl3): δ = 7.40–7.24 (m, 10 H,
Ar), 4.71 (pseudo s, 1 H, =CH2), 4.68 (pseudo s, 1 H, =CH2), 3.56
lectively to homoallyl-homocrotylamines. This unpredicted
reactivity was interpreted as a triple cascade reaction (allyl-
titanation – cationic 2-aza-Cope rearrangement – allyltitan-
ation). A theoretical study provided insight into the
thermodynamic and kinetic factors that govern the cationic
2-aza-Cope rearrangement and accounted for the exclusive
formation of the homoallyl-(E)-homocrotylamines.
2
2
(d, JH,H = 13.5 Hz, 2 H, CH2N), 3.48 (d, JH,H = 13.5 Hz, 2 H,
2
CH2N), 2.54–2.39 (m, 2 H, CH, CH2N), 2.23 (dd, JH,H = 12.3,
3
3JH,H = 7.2 Hz, 1 H, CH2N), 1.52 (s, 3 H, CH3), 0.97 (d, JH,H
=
6.9 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 149.3
(=C), 140.0 (2 C, Ar), 129.1 (4 C, Ar), 128.2 (4 C, Ar), 126.9 (2 C,
Ar), 110.1 (=CH2), 58.8 (2 C, CH2N), 58.7 (CH2N), 39.5 (CH),
19.1 (CH3), 18.0 (CH3) ppm. MS (ESI): m/z calcd. for C20H25N
[MH]+ 280.2021; found 280.2023. MS: m/z = 279 [M]+, 210, 91.
C20H25N (279.42): calcd. C 85.97, H 9.02, N 5.01; found C 85.45,
H 8.53, N 5.07.
Experimental Section
General: Column chromatography was performed using alumina
1
gel (Merck, aluminium oxide 90 standardised). H and 13C NMR
N-Benzyl-N,2,3-trimethyl-3-butenylamine (12): Yellow oil (182 mg).
1H NMR (300 MHz, CDCl3): δ = 7.33–7.23 (m, 5 H, Ar), 4.74
(pseudo s, 1 H, =CH2), 4.72 (pseudo s, 1 H, =CH2), 3.50 (br. s, 2
H, CH2N), 2.48–2.40 (m, 2 H, CH, CH2N), 2.22 (m, 1 H, CH2N),
spectra were recorded with a Bruker DRX 300 spectrometer. 1H
NMR chemical shifts (δ) are reported in parts per million (ppm)
downfield of TMS and are referenced relative to residual CHCl3 (δ
= 7.26 ppm). Chemical shifts for 13C are reported in parts per mil-
lion downfield of TMS and are referenced to the carbon resonance
of the solvent (CDCl3: δ = 77.0 ppm). Data are represented as fol-
lows: chemical shift, multiplicity (br = broad, s = singlet, d = doub-
let, t = triplet, m = multiplet), coupling constants in Hertz (Hz),
and integration. Gas chromatography mass spectroscopy (GC–MS)
spectra were collected with a Finnigan Trace GC ultra gas chroma-
3
2.19 (s, 3 H, CH3N), 1.67 (s, 3 H, CH3), 1.04 (d, JH,H = 7 Hz, 3
H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 149.4 (=C), 139.8
(1 C, Ar), 129.2 (2 C, Ar), 128.3 (2 C, Ar), 127.0 (1 C, Ar), 110.0
(=CH2), 62.9 (CH2N), 62.7 (CH2N), 42.6 (CH3N), 39.5 (CH), 19.4
(CH3), 18.2 (CH3) ppm. MS (ESI): m/z calcd. for C14H21N [MH+]
204.17468; found 204.17429. MS: m/z = 203.2 [MH+], 134.04 [M –
isoprene]. C14H21N (203.3): calcd. C 82.70, H 10.41, N 6.89; found
C 82.89, H 10.30, N 6.88.
tograph
(column
Thermo
TR
5MS;
capillary
30 mϫ0.25 IDϫ0.25 μm film) and Trace DSC mass detector.
Electrospray ionisation (ESI) mass spectrometry experiments were
performed with a MicrOTOF Q. Elementary analysis were ob-
tained with an EA 1108 CHNS-O FISONS detector.
N,N-Diethyl-2,7-dimethyl-3-methylene-6-octenylamine (13): Colour-
less oil (182 mg). H NMR (300 MHz, CDCl3): δ = 5.12 (t, JH,H
= 6.9 Hz, 1 H, =CH), 4.76 (pseudo s, 1 H, =CH2), 4.74 (pseudo s,
1
3
1 H, =CH2), 2.59–2.34 (m, 5 H, CH2N), 2.32–1.85 (m, 6 H, CH2N,
3
Materials: All solvents were distilled before use from sodium/
benzophenone complex under an atmosphere of argon. Deuterated
solvents were purchased from euriso-top and were used without
further purification. Benzotriazole compounds 1–9 were prepared
according to previously reported procedures.[15–17]
CH, CH2), 1.68 (s, 3 H, CH3), 1.61 (s, 3 H, CH3), 1.04 (d, JH,H
=
6.6 Hz, 3 H, CH3), 0.98 (t, JH,H = 7.2 Hz, 6 H, CH3) ppm. 13C
NMR (75 MHz, CDCl3): δ = 154.0 (=C), 131.6 (=C), 124.5 (=CH),
108.0 (=CH2), 59.0 (CH2N), 47.6 (2 C, CH2N), 38.6 (CH), 34.5
(CH2), 26.8 (CH2), 25.8 (CH3), 18.8 (CH3), 17.8 (CH3), 11.8 (2 C,
CH3) ppm. MS (ESI): m/z calcd. for C15H29N [MH]+ 224.2334;
found 224.2379. MS: m/z = 223 [M]+, 86, 58.
3
General Procedure for the Synthesis of Amines 10–15: Diene
(2.4 mmol) was added to a solution of [Cp2TiCl2] (1.2 mmol) in
THF (20 mL). The reaction was stirred at room temperature for
15 min and cooled to 0 °C and then a solution of iPrMgCl (2 m in
THF, 0.6 mL, 1.2 mmol) was added dropwise by using a syringe.
The resulting purple solution was stirred for 5 min at 0 °C followed
by 30 min at room temperature. The allyltitanocene generated in
situ was added over 3 h by using a syringe pump to a solution of
benzotriazole compound (1.2 mmol) in THF (4 mL) at 10 °C. After
the addition, the resulting green solution was warmed and was
stirred at room temperature for 1 h. After hydrolysis by the ad-
dition of 2 m NaOH (20 mL) and extraction with Et2O (4ϫ30 mL),
the combined organic layers were washed with water (3ϫ50 mL)
until neutral pH. The solution was dried with MgSO4, filtered, and
concentrated in vacuo to yield an oil. The compounds were purified
by alumina column chromatography.
N,N-Dibenzyl-2,7-dimethyl-3-methylene-6-octenylamine (14): Pale-
yellow oil (291 mg). 1H NMR (300 MHz, CDCl3): δ = 7.42–7.27
(m, 10 H, Ar), 5.13 (m, 1 H, =CH), 4.77 (pseudo s, 1 H, =CH2),
4.74 (pseudo s, 1 H, =CH2), 3.62 (d, 2JH,H = 13.5 Hz, 2 H, CH2N),
2
3.52 (d, JH,H = 13.5 Hz, 2 H, CH2N), 2.48 (m, 2 H, CH2N), 2.30
(m, 1 H, CH), 2.21–1.85 (m, 4 H, CH2), 1.74 (s, 3 H, CH3), 1.65
3
(s, 3 H, CH3), 1.06 (d, JH,H = 6.9 Hz, 3 H, CH3) ppm. 13C NMR
(75 MHz, CDCl3): δ = 153.3 (=C), 140.0 (Ar), 131.5 (=C), 129.1
(Ar), 128.9 (Ar), 128.3 (Ar), 126.9 (Ar), 124.6 (=CH), 108.3
(=CH2), 59.5 (CH2N), 58.9 (CH2N), 58.4 (CH2N), 38.5 (CH), 33.7
(CH2), 26.6 (CH2), 25.8 (CH3), 18.5 (CH3), 17.8 (CH3) ppm. MS
(ESI): m/z calcd. for C25H33N [MH]+ 348.2685; found 348.2685.
MS: m/z = 347 [M]+, 209, 99. C25H33N (347.54): calcd. C 86.40, H
9.57, N 4.03; found C 85.45, H 8.53, N 5.07.
N-(2,3-Dimethyl-3-butenyl)-N-methylaniline (10): Yellow oil 1-(2,4,5-Trimethyl-5-hexen-3-yl)pyrrolidine (15): Orange oil
(195 mg). 1H NMR (300 MHz, CDCl3): δ = 7.24–7.30 (m, 2 H,
Ar), 6.69–6.76 (m, 3 H, Ar), 4.80 (pseudo s, 2 H, =CH2), 3.44 (dd,
(136 mg). 1H NMR (300 MHz, CDCl3): δ (major isomer) = 4.74
(pseudo s, 1 H, =CH2), 4.68 (pseudo s, 1 H, =CH2), 2.84 (m, 4 H,
2
3
3JH,H = 6.6, JH,H = 14.7 Hz, 1 H, CH2N), 3.21 (dd, JH,H = 6.6,
CH2), 2.53 (m, 1 H, CH), 2.46 (m, 1 H, NCH), 1.77 (m, 1 H, CH),
2JH,H = 14.7 Hz, 1 H, CH2N), 2.99 (s, 3 H, CH3N), 2.69 (m, 1 H, 1.73 (m, 4 H, CH2), 1.68 (s, 3 H, CH3), 0.99 (d, JH,H = 6.6 Hz, 3
3
3
CH), 1.80 (s, 3 H, CH3), 1.08 (d, JH,H = 6.9 Hz, 3 H, CH3) ppm. H, CH3), 0.93 (d, 3JHH = 6.6 Hz, 3 H, CH3), 0.83 (d, 3JHH = 6.9 Hz,
13C NMR (75 MHz, CDCl3): δ = 149.4 (=C), 148.4 (1 C, Ar), 129.2
(2 C, Ar), 115.8 (1 C, Ar), 112.0 (2 C, Ar), 110.8 (=CH2), 58.2
3 H, CH3); δ (minor isomer) = 4.71 (pseudo s, 2 H, =CH2), 2.67
(m, 4 H, CH2), 2.51 (m, 1 H, CH), 2.46 (m, 1 H, NCH), 1.89 (m,
3
(CH2N), 39.7 (CH3N), 39.5 (CH), 20.1 (CH3), 17.5 (CH3) ppm. MS 1 H, CH), 1.68 (s, 3 H, CH3), 1.65 (m, 4 H, CH2), 1.01 (d, JH,H
(ESI): m/z calcd. for C13H19N [MH]+ 190.1590; found 190.1591.
MS: m/z = 189 [M]+, 120, 77. C13H19N (189.3): calcd. C 82.48, H
10.12, N 7.40; found C 82.43, H 9.93, N 7.46.
= 6.9 Hz, 3 H, CH3), 0.95 (d, JH,H = 6.9 Hz, 3 H, CH3), 0.92 (d,
3JH,H = 6.6 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ
(major isomer) = 149.0 (=C), 109.4 (=CH2), 65.2 (CH), 50.0 (CH2),
3
Eur. J. Org. Chem. 2013, 736–741
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
739

P. Le Gendre et al.
FULL PAPER
40.6 (CH), 29.9 (CH), 23.7 (CH2), 19.7 (CH3), 18.5 (CH3), 17.6
29.8 (CH2, 1 C, Cy), 28.5 (CH2, 1 C, Cy), 26.5 (CH2, 3 C, Cy),
(CH3), 16.7 (CH3); δ (minor isomer) = 149.2 (=C), 108.8 (=CH2), 20.2 (CH3), 18.0 (CH3), 16.2 (CH3), 13.5 (CH3) ppm. MS (ESI):
65.2 (CH), 49.5 (CH2), 41.4 (CH), 27.4 (CH), 22.9 (CH2), 21.3 m/z calcd. for C18H33N [MH+] 264.2686; found 264.2701. MS: m/z
(CH3), 20.2 (CH3), 18.2 (CH3), 15.6 (CH3) ppm. MS (ESI): m/z = 263 [M]+, 194, 111, 55. C18H33N (263.46): calcd. C 82.06, H
calcd. for C13H25N [MH]+ 196.2060; found 196.2074. MS: m/z =
195 [M]+, 152, 126,70. C13H25N (195.34): calcd. 79.93, H 12.90, N
7.17; found C 80.25, H12.76, N 7.32.
12.63, N 5.32; found C 82.33, H 12.67, N 5.14.
(E)-N-(2,7-Dimethyl-3-methylene-6-octenyl)-N-(3-ethylidene-7-
methyl-6-octenyl)cyclohexylamine (20): Yellow oil (402 mg). 1H
NMR (300 MHz, CDCl3): δ = 5.23 (q, 3JH,H = 6.6 Hz, 1 H, =CH),
5.14 (m, 2 H, =CH), 4.74 (m, 2 H, =CH2), 2.54–2.34 (m, 4 H, Cy,
CH, CH2N), 2.16–2.02 (m, 12 H, CH2N, CH2), 1.82–1.72 (m, 4 H,
General Procedure for the Synthesis of Homoallyl-(E)-homocrotyl-
amines 16–20: Produced as described for the synthesis of 10–15
except that two equivalents of allyltitanium complex were used.
3
Cy), 1.68 (s, 6 H, CH3), 1.62 (s, 6 H, CH3), 1.58 (d, JH,H = 7 Hz,
(E)-N-Benzyl-N-(2,3-dimethyl-3-butenyl)-3-methyl-3-pentenyl-
amine (16): Orange oil (215 mg). 1H NMR (300 MHz, CDCl3): δ
3 H, CH3), 1.30–1.20 (m, 6 H, Cy), 1.04 (d, J = 6.6 Hz, 3 H,
CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 154.3 (=C), 139.1
(=C), 131.6 (=C), 131.5 (=C), 124.6 (2 C, =CH), 124.3 (=CH),
107.7 (=CH2), 60.3 (CHN), 57.1 (CH2N), 50.9, 39.7, 39.6, 37.9,
35.0, 30.4, 30.2, 27.2, 27.0, 26.9, 26.6 (CH3), 25.8 (CH3), 18.4
(CH3), 17.8 (CH3), 13.4 (CH3) ppm. MS (ESI): m/z calcd. for
C28H49N [MH]+ 400.3943; found 400.4008. MS: m/z = 399 [M]+,
262. C28H49N (399.69): calcd. C 84.14, H 12.36, N 3.50; found C
84.32, H 12.12, N 3.68.
3
= 7.34–7.21 (m, 5 H, Ar), 5.19 (q, JH,H = 7 Hz, 1 H, =CH), 4.71
2
(pseudo s, 1 H, =CH2), 4.69 (pseudo s, 1 H, =CH2), 3.59 (d, JH,H
2
= 13.8 Hz, 1 H, CH2N), 3.53 (d, JH,H = 13.8 Hz, 1 H, CH2N),
2.55–2.10 (m, 7 H, CH2N, CH2, CH), 1.64 (s, 3 H, CH3), 1.56 (d,
3JH,H = 7 Hz, 3 H, CH3), 1.53 (s, 3 H, CH3), 1.01 (d, 3JH,H = 7 Hz,
3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 149.6 (=C),
140.4 (=C), 134.5 (1 C, Ar), 129.1 (2 C, Ar), 128.1 (2 C, Ar), 126.8
(1 C, Ar), 119.5 (=CH), 109.9 (=CH2), 59.2 (CH2N), 58.9 (CH2N),
53.0 (CH2N), 39.6 (CH), 36.9 (CH2), 19.6 (CH3), 18.1 (CH3), 16.0
(CH3), 13.5 (CH3) ppm. MS (ESI): m/z calcd. for C19H29N [MH]+
272.2372; found 272.2361. MS: m/z = 271 [M]+, 209, 91. C19H29N
(271.44): calcd. C 84.07, H 10.77, N 5.16; found C 83.73, H 10.67,
N 5.07.
Supporting Information (see footnote on the first page of this arti-
1
cle): Copies of H and 13C NMR spectra of 10–20.
Acknowledgments
The authors thank the Conseil Régional de Bourgogne (PARI
SMT 08 IME - Région Bourgogne), the Ministère de l’Enseigne-
ment Supérieur et de la Recherche and the Centre National de la
Recherche Scientifique (CNRS) for financial support. The authors
wish to thank Prof. F. Denat and Dr. F. Boschetti for helpful dis-
cussions and CheMaTech Company for the generous supply of
benzotriazole.
(E)-N-(2,3-Dimethyl-3-butenyl)-3-methyl-N-phenylethyl-3-penten-
1
ylamine (17): Yellow oil (205 mg). H NMR (300 MHz, CDCl3): δ
3
= 7.28–7.16 (m, 5 H, Ar), 5.21 (q, JH,H = 7 Hz, 1 H, =CH), 4.71
(pseudo s, 2 H, =CH2), 2.70 (s, 4 H, CH2), 2.56 (m, 2 H, CH2),
2.49 (m, 2 H, CH2), 2.31 (m, 1 H, CH), 2.11 (m, 2 H, CH2), 1.69
(s, 3 H, CH3), 1.60 (s, 3 H, CH3), 1.55 (d, 3JH,H = 7 Hz, 3 H, CH3),
1.00 (pseudo s, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ =
149.4 (=C), 128.9 (Ar), 128.5 (Ar), 126.1 (Ar), 119.9 (br., =CH),
110.2 (=CH2), 59.1 (CH2N), 56.3 (CH2N), 53.4 (CH2N), 39.6 (CH),
36.7 (CH2), 33.3 (CH2), 19.7 (CH3), 18.3 (CH3), 16.1 (CH3), 13.5
(CH3) ppm. MS (ESI): m/z calcd. for C20H31N [MH+] 286.2529;
found 286.2543. MS: m/z = 286 [MH]+, 216.12 [M – isoprene].
C20H31N (285.4): calcd. C 84.15, H 10.95, N 4.91; found C 81.74,
H 11.12, N 5.08.
[1] a) J. S. Yadav, B. V. Subba Reddy, K. Ramesh, G. G. K. S. Na-
rayana Kumar, R. Grée, Tetrahedron Lett. 2010, 51, 818–821;
b) D. L. Wright, J. P. Schulte II, M. A. Page, Org. Lett. 2000,
2, 1847–1850; c) Y.-F. Wang, T. Izawa, S. Kobayashi, M. Ohno,
J. Am. Chem. Soc. 1982, 104, 6465–6466.
[2] a) P. Veeraraghavan Ramachandran, D. Biswas, Org. Lett.
2007, 9, 3025–3027; b) T. P. Loh, D. S. C. Ho, K. Xu, K. Sim,
Tetrahedron Lett. 1997, 38, 865–868.
[3] a) J. E. Kropft, I. C. Meigh, W. P. Magnus, M. W. P. Bebbing-
ton, S. M. J. Weinreb, J. Org. Chem. 2006, 71, 2046–2055; b)
F. D. Suvire, M. Sortino, V. V. Kouznetsov, M. L. Y. Vargas,
S. A. Zacchino, U. M. Cruz, R. D. Enriz, Bioorg. Med. Chem.
2006, 14, 1851–1862.
[4] M. Atobe, N. Yamakasi, C. Kibayashi, Tetrahedron Lett. 2005,
46, 2669–2673.
(E)-N-(2,3-Dimethyl-3-butenyl)-N-isopropyl-3-methyl-3-penten-
1
ylamine (18): Yellow oil (160 mg). H NMR (300 MHz, CDCl3): δ
3
= 5.21 (q, JH,H = 6.7 Hz, 1 H, =CH), 4.70 (pseudo s, 1 H, =CH2),
3
4.69 (pseudo s, 1 H, =CH2), 2.91 (sept, JH,H = 6.7 Hz, CH), 2.42
(m, 2 H, CH2N), 2.37 (m, 1 H, CH2N), 2.21 (m, 1 H, CH), 2.17
(m, 1 H, CH2N), 2.07 (m, 2 H, CH2), 1.70 (s, 3 H, CH3), 1.61 (br.
3
3
s, 3 H, CH3), 1.56 (d, JH,H = 6.7 Hz, 3 H, CH3), 1.00 (d, JH,H
=
3
6.7 Hz, 3 H, CH3), 0.96 (d, JH,H = 6.7 Hz, 3 H, CH3), 0.92 (d,
3JH,H = 6.7 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ =
150.1 (=C), 134.9 (=C), 119.2 (=CH), 109.4 (=CH2), 55.5 (CH2N),
50.3 (CHN), 49.8 (CH2N), 40.4 (CH), 39.7 (CH2), 20.2 (CH3), 19.0
(CH3), 17.9 (CH3), 17.4 (CH3), 16.2 (CH3), 13.5 (CH3) ppm. MS
(ESI): m/z calcd. for C15H29N [MH+] 224.2372; found 224.2364.
MS: m/z = 224.4 [MH+], 154.12, 112.06.
[5] H. Lu, C. Li, Org. Lett. 2006, 8, 5365–5367.
[6] a) P. M. A. Rabbat, S. C. Valdez, J. L. Leighton, Org. Lett.
2006, 8, 6119–6121; b) M. Billet, P. Klotz, A. Mann, Tetrahe-
dron Lett. 2001, 42, 631–634.
[7] a) P. K. Kalita, P. Phukan, Tetrahedron Lett. 2008, 49, 5495–
5497; b) H. Nakamura, H. Iwama, Y. Yamamoto, J. Am. Chem.
Soc. 1996, 118, 6641–6647; c) G. E. Keck, E. J. Enholm, J. Am.
Chem. Soc. 1985, 50, 146–147.
[8] F. Watanabe, K. Ito, S. Itsuno, Tetrahedron: Asymmetry 1995,
(E)-N-(2,3-Dimethyl-3-butenyl)-N-(3-methyl-3-pentenyl)cyclo-
hexylamine (19): Yellow oil (252 mg). 1H NMR (300 MHz, CDCl3):
δ = 5.22 (q, 3JH,H = 6.5 Hz, 1 H, =CH), 4.70 (pseudo s, 2 H, =CH2),
2.50–2.47 (m, 4 H, Cy, CH, CH2N), 2.25 (m, 2 H, CH2N), 2.09 (m,
2 H, CH2), 1.76–1.72 (m, 4 H, Cy), 1.70 (s, 3 H, CH3), 1.61 (s, 3
6, 1531–1534.
[9] T. Akiyama, M. Sugano, H. Kagoshima, Tetrahedron Lett.
2001, 42, 3889–3892.
[10] a) T. Ollevier, T. Ba, Tetrahedron Lett. 2003, 44, 9003–9005; b)
C. Kleber, Z. Andrade, N. R. Azevedo, G. R. Oliveira, Synthe-
sis 2002, 928–936.
[11] a) A. R. Katritzky, B. R. Rogovoy, Chem. Eur. J. 2003, 9, 4586–
4593; b) A. R. Katritzky, X. Lan, J. Z. Yang, O. V. Denisko,
Chem. Rev. 1998, 98, 409–548.
3
H, CH3), 1.57 (d, JH,H = 7 Hz, 3 H, CH3), 1.30–1.20 (m, 6 H,
Cy), 1.02 (d, 3JH,H = 6.5 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz,
CDCl3): δ = 150.1 (=C), 134.9 (=C), 119.2 (=CH), 109.4 (=CH2),
60.3 (CH, Cy), 56.1 (CH2N), 50.3 (CH2N), 40.5 (CH), 39.7 (CH2),
740
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 736–741

Homoallyl-Homocrotylamines
[12] a) L. Bareille, P. Le Gendre, C. Moïse, Chem. Commun. 2005,
775–777; b) L. Bareille, S. Becht, J. L. Cui, P. Le Gendre, C.
Moïse, Organometallics 2005, 24, 5802–5806; c) A. Perrier, V.
Comte, C. Moïse, P. Le Gendre, Chem. Eur. J. 2010, 16, 1562–
1568; d) A. Perrier, V. Comte, C. Moïse, P. Richard, P. Le Gen-
dre, Eur. J. Org. Chem. 2010, 1562–1568.
[13] For other transition-metal-promoted reductive coupling of π-
unsaturated compounds with imines derivatives, see, for exam-
ple: a) F. Shibahara, J. F. Bower, M. J. Krische, J. Am. Chem.
Soc. 2008, 130, 6338–6339; b) M. Kimura, M. Mori, N. Mukai,
K. Kojima, Y. Tamaru, Chem. Commun. 2006, 2813–2815.
[14] The synthesis of homoallylamines with the aid of titanium
complexes from imines and allyl halides, allyl carbonates, allyl
ethers and allylic alcohols has been fully described, see, for
example: a) Y. Gao, F. Sato, J. Org. Chem. 1995, 60, 8136–
8137; b) I. L. Lysenko, H. Goo Lee, J. K. Cha, Org. Lett. 2009,
11, 3132–3134; c) M. Z. Chen, M. McLaughlin, M. Takahashi,
M. A. Tarselli, D. Yang, S. Umemura, G. C. Micalizio, J. Org.
Chem. 2010, 75, 8048–8059.
utes led to the target homoallylamine 15 with slightly better
diastereoselectivity but in lower yield (syn/anti ratio 30:70; yield
48%).
R. M. Horowitz, T. A. Geissman, J. Am. Chem. Soc. 1950, 72,
1518–1522.
[22]
[23]
a) Z. D. Aron, L. E. Overman, Org. Lett. 2005, 7, 913–916; b)
L. E. Overman, L. T. Mendelson, E. J. Jacobsen, J. Am. Chem.
Soc. 1983, 105, 6629–6637.
For total syntheses of alkaloids, see: a) C. L. Martin, L. E.
Overman, J. M. Rohde, J. Am. Chem. Soc. 2008, 130, 7568–
7569; b) J. M. Fevig, R. W. Marquis Jr., L. E. Overman, J. Am.
Chem. Soc. 1991, 113, 5085–5086. For syntheses of other natu-
ral products, see: c) D. J. Bennett, N. M. Hamilton, Tetrahedron
Lett. 2000, 41, 7961–7964; d) S. Deloisy, H. Kunz, Tetrahedron
Lett. 1998, 39, 791–794.
Theoritical calculations: performed by using Hyperchem
v. 8.0.3[26] for the RM1[27] semiempirical method and Jaguar
v. 5.5[28] for the DFT B3LYP/6-31G** method. The conforma-
tional searches (RM1) were carried out by minimisation of
starting conformations generated by random variations of se-
lected dihedral angles. The transition structures were located
(in both methods, RM1 and DFT) by quadratic synchronous
transit method (QST), the frequencies were checked and the
visualisation of the mode with negative eigenvalue clearly cor-
responds to the expected C–C bond formation or bond break-
ing. The RM1 calculations were run on a PC, whereas DFT
calculations were performed by using HPC resources from
DSI-CCUB (Université de Bourgogne).
[24]
[25]
[15] A. R. Katritzky, K. Yannakopoulou, P. Lue, D. Rasala, L. Ur-
ogdi, J. Chem. Soc. Perkin Trans. 1 1989, 225–233.
[16] A. R. Katritzky, S. Rachwal, B. Rachwal, J. Org. Chem. 1995,
60, 3993–4001.
[17] A. R. Katritzky, S. Rachwal, B. Rachwal, J. Chem. Soc. Perkin
Trans. 1 1987, 799–804.
[18] H. A. Martin, F. Jellinek, J. Org. Chem. 1966, 6, 293–296.
[19] The order of addition of the substrates, the reaction tempera-
ture, and the addition rate of the allyltitanium complex to the
benzotriazole derivative are crucial for obtaining the product
in good yield. A change in the order of addition of the sub-
strates gave the product in very poor yield. The addition rate of
the THF solution of allyltitanium complex to the benzotriazole
derivative can be accelerated (0.3 mmolmin–1 instead of
0.5 mmolh–1) without drastically decreasing the yield (76% in-
stead of 86%) if care is taken to maintain the temperature of
the reaction mixture at –12 °C. These data can be explained by
the low stability of the allyltitanium reagent in the reaction
mixture.
[26]
HyperChem, v. 8.0.3, Hypercube, Inc., 1115 NW 4th Street,
Gainesville, Florida 32601, USA.
[27] G. B. Rocha, R. O. Freire, A. M. Simas, J. P. Stewart, J. Com-
put. Chem. 2006, 27, 1101–1111.
[28] Jaguar, v. 5.5, Schrodinger, LLC, Portland, Oregon, 2003.
[29] For clarity, we have represented the conformation I-b as the
opposite enantiomer of I-a. Clearly the energy paths for two
enantiomers with the same conformation are identical.
[30] L. Kvaernø, P.-O. Norrby, D. Tanner, Org. Biomol. Chem. 2003,
1, 1041–1048.
[20] B. Klei, J. H. Teuben, H. J. de Liefde Meijer, J. Chem. Soc.,
Chem. Commun. 1981, 342–344.
[21] The analogous reaction conducted at lower temperature
(–12 °C) by adding the allyltitanium complex within a few min-
Received: September 21, 2012
Published Online: December 14, 2012
Eur. J. Org. Chem. 2013, 736–741
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
741