The Journal of Organic Chemistry
NOTE
Table 2. Scope of the Palladium-Catalyzed Synthesis of 2a
Table 3. Synthesis of Chiral 1,5-Amino Alcohols 2a
% ee 1b,c
product
% ee 2c
yield (%)
entry
R
R1
R2
ArX
yield (%) product
entry
1
2
p-MeOC6H4 Bn
H
PhI
PhI
75
90
70
72
70
54
73
74
79
84
90
80
2a
2b
2c
2d
2e
2f
1
2
3
4
5
6
96
96
97
97
97
95
2a
2d
2h
2i
96
96
97
97
97
95
73
84
79
80
86
90
p-MeOC6H4 n-Bu
H
3
p-MeOC6H4 n-Bu n-Bu PhI
4
p-MeOC6H4 Bn
p-MeOC6H4 Bn
p-MeOC6H4 Bn
p-MeOC6H4 Bn
H
2-naphthylI
p-MeC6H4I
m-BrC6H4I
p-NO2C6H4Br
PhI
5
H
2j
6
H
2k
a Reaction conditions: 1 (1 equiv), ArX (1.25 equiv), amine (1.5 equiv),
Pd(PPh3)4 (2.5 mol %), Et3N (2 equiv), CH2Cl2, reflux, 16 h. b Optically
enriched homoallenols 1 were synthesized following our procedure, see
ref 7. c The ee value was determined by chiral HPLC.
7
H
2g
2h
2i
8
p-ClC6H4
p-ClC6H4
p-ClC6H4
Bn
Bn
Bn
H
9
H
2-naphthylI
PhI
10
11
12
Bn
H
2j
PhCH2CH2 Bn
PhCH2CH2 Bn
PhI
2k
2l
Bn
PhI
Scheme 2. Proposed Catalytic Cycle
a Reaction conditions: 1 (1 equiv), ArX (1.25 equiv), amine (1.5 equiv),
Et3N (2 equiv), Pd(PPh3)4 (2.5 mol %), CH2Cl2, reflux, 16 h.
attempts at changing solvent and additive yielded a mixture of
isomeric products 2a and 3a (entries 1 and 3ꢀ5). We subse-
quently identified dichloromethane and 2 equiv of triethylamine
as particularly effective for this reaction as, under these condi-
tions, the desired compound 2a was obtained in 75% yield as a
pure diastereomer without any trace of regioisomer 3a (entry
6).11,12 Furthermore, cycloetherification of 1a was not observed
under these conditions.13
With these optimized catalytic reaction conditions in hand, we
then explored the scope of this three-component transformation
by reacting a range of homoallenols 1 with a focused selection of
amines and aryl halides (Table 2). Regardless of the coupling
partners, the desired products 2 were obtained as pure diaster-
eomers in 54ꢀ90% yields without any trace of the 1,3-amino
alcohols 3. Both primary and secondary amines proved to be
suitable nucleophiles for this transformation. With respect
to aryl iodide substitution, para- and meta-substituents were
tolerated, even the presence of a m-bromo substituent (entry 6).
No products were obtained when ortho-substituted aryl iodides
were employed as substrates. Apart from aryl iodides,
only a highly electrophilic aryl bromide, p-nitrobromobenzene,
turned out to be a suitable substrate for this transformation
(entry 7).
We next focused our attention on the synthetic potential of
this three-component transformation for creating chiral building
blocks 2 from enantioenriched homoallenols 1 (Table 3). No
epimerization at the stereogenic center was observed during the
transformations and consequently, the desired 1,5-amino alco-
hols (Z)-2 were obtained as optically enriched compounds from
the corresponding homoallenols 1.
In summary, we have disclosed a regio- and stereoselective
palladium-catalyzed three-component transformation allowing a
simple access to optically enriched 1,5-amino alcohols 2 from
readily available homoallenols 1. This reaction possesses the
ability to generate valuable chiral building blocks 2, merging an
allene, an amine, and an alcohol into the same molecule in an
atom-economic fashion. We anticipate that this transformation
will find many applications in different areas of organic chemistry.
’ EXPERIMENTAL SECTION
General Procedure for the Palladium-Catalyzed Synthesis
of 2. A flame-dried Schlenk tube flushed with nitrogen was charged with
homoallenol 1 (0.52 mmol), the corresponding electrophile (0.65
mmol), and amine (0.78 mmol). To this mixture was added Et3N
(1.04 mmol), anhydrous CH2Cl2 (2.5 mL), and Pd(PPh3)4 (13 μmol).
The tube was allowed to reflux under nitrogen for the appropriate length
of time. The solvent was removed in vacuo. The crude mixture was then
purified by flash column chromatography on silica gel with pentane:
EtOAc as the eluent to give the corresponding 1,5-amino alcohol 2.
Characterization data for 2a: TLC: Rf 0.58 (EtOAc:pentane 3:2). [R]20
D
þ34.0 (c 0.2, CHCl3) for 96% ee (R). 1H NMR (400 MHz, CDCl3): δ
7.35ꢀ7.27 (m, 12H), 6.87 (d, J = 8.5 Hz, 2H), 5.93 (t, J = 8.5 Hz, 1H),
4.80 (app dd, J = 7.5, 3.8 Hz, 1H), 3.81 (s, 2H), 3.79 (s, 3H), 3.64 (d, J =
12.0 Hz, 1H), 3.56 (d, J = 12.0 Hz, 1H), 2.70ꢀ2.56 (m, 2H). 13C NMR
(100 MHz, CDCl3): δ 158.6 (1C), 141.8 (1C), 140.0 (1C), 138.9 (1C),
137.5 (1C), 129.0 (1C), 128.5 (4C), 128.4 (2C), 127.2 (1C), 127.1
(1C), 126.7 (2C), 126.0 (2C), 113.6 (2C), 71.9 (1C), 55.2 (1C), 53.8
(1C), 47.1 (1C), 39.6 (1C). IR (neat): ν 3414, 2092, 1644, 1512,
1443 cmꢀ1. HRMS (ESI) calcd for C25H28NO2 [M þ H] 374.2120,
found 374.2120. Enantiomeric excess has been determined by HPLC
analysis, using an IA column (heptane/ethanol 90/10, 1.0 mL/min), tr =
23.7 min for (S) and tr = 25.4 min for (R).
A proposed catalytic cycle helps to explain the regio- and
stereochemical course of this transformation (Scheme 2). First,
oxidative addition of the aryl iodide to palladium(0), followed by
insertion of this species to allene 1 would afford the thermo-
dynamically more stable η3-allyl complex anti-A.14 In this inter-
mediate, the allene substituent (RCH(OH)CH2) would be in
the anti-position relative to the aryl group in order to minimize
steric interactions, even if the presence in solution of syn-A
cannot be ruled out at this stage. Then, anti-A would undergo a
nucleophilic attack by the amine preferably at the terminal
carbon for steric reasons to provide (Z)-2 as the unique regio-
and stereoisomer.15
3537
dx.doi.org/10.1021/jo1025628 |J. Org. Chem. 2011, 76, 3536–3538