Novel transformation of α,β-unsaturated aldehydes and ketones into γ-amino alcohols or 1,3-oxazines via a 4 or 5 step, one-pot sequence

Article abstract of DOI:10.1039/c2cc36129a

An efficient, 4-step, one-pot, highly stereoselective route to γ-amino alcohols has been developed via an in situ α,β- unsaturated imine formation, β-boration, reduction (CN) and oxidation (C-B) sequence and especially for certain water-soluble γ-amino alcohols, a further step can be added to directly access the corresponding 1,3-oxazine derivatives.

Full text of DOI:10.1039/c2cc36129a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 11401–11403
www.rsc.org/chemcomm
COMMUNICATION
Novel transformation of a,b-unsaturated aldehydes and ketones into
c-amino alcohols or 1,3-oxazines via a 4 or 5 step, one-pot sequencew
a
a
b
b
´ ´
Adam D. J. Calow, Andrei S. Batsanov, Elena Fernandez,* Cristina Sole and
Andrew Whiting*a
Received 23rd August 2012, Accepted 20th September 2012
DOI: 10.1039/c2cc36129a
An efficient, 4-step, one-pot, highly stereoselective route to c-amino
alcohols has been developed via an in situ a,b-unsaturated imine
Q
formation, b-boration, reduction (C N) and oxidation (C–B)
sequence and especially for certain water-soluble c-amino alcohols,
a further step can be added to directly access the corresponding
1,3-oxazine derivatives.
Scheme 1 Substrate controlled asymmetric reduction. (a) i, NaBH4,
MeOH; ii, NaOH, H2O2; (b) i, BH3ÁTHF; ii, NaOH, H2O2.
b-boration of activated olefins has received considerable
attention,1,2 involving a Michael-like addition of a diboron reagent
to a conjugated, electron deficient alkene (e.g. 1 or 2), to give
b-borylation (e.g. 4 from imine 2). Asymmetric b-boration,
coupled with known methods for C–B functionalisation, is an
attractive concept for the control of stereochemistry. Herien,
we report an efficient 4-step, one-pot route to g-amino alcohols
5 via an in situ a,b-unsaturated imine formation, b-boration,
reduction (CQN) and oxidation (C–B) sequence which can be
extended to give the corresponding oxazines.
The ease of formation and stability of a,b-unsaturated imines is
surprisingly underexplored compared to non-conjugated imines,9
though the synthesis of dihydropyridines and pyridines from less-
substituted a,b-unsaturated imines11 has been reported. We
investigated the formation of less-substituted a,b-unsaturated
imines by ReactIR12 to understand the relative rates and selec-
tivity of a,b-unsaturated imine 2 formation vs. Michael addition
(Table 1). Indeed, ReactIR proved to be an ideal tool for
monitoring this reaction (Table 1) for both selectivity and rate.
Most imine formations were complete within 3 h; the exception
being methyl vinyl ketone (Entry 7, Table 1). Facile imine
formation is exemplified (Fig. 1) by loss of the CQO stretch
Asymmetric routes to g-amino alcohols 5 are limited,
despite their use in the pharma industry and as ligands,3 due
largely to the challenge of controlling up to 3 contiguous
stereocenters. Despite this, useful progress has been made, but
there remains scope for improved, efficient new methods.4,5
We demonstrated a highly enantio- and diastereo-selective
route to g-amino alcohols 5 via a three-step route involving
a,b-unsaturated imines 2 (R1 = Ar and R3 = Me).7,8 Asymmetric
b-boration resulted in b-boryl imines 4, which could undergo
substrate-controlled asymmetric CQN reduction and C–B oxida-
tion to give g-amino alcohols 5 (Scheme 1). Although this was a
powerful route to systems 5, the application was severely limited by
the range of a,b-unsaturated imines 2 that could be isolated.
Normally, imines are prepared by 1,2-addition of an amine
to the analogous carbonyl compound, however, competitive
1,4-addition and instability of the imines meant that this protocol
was suitable solely for stable, chalcone derived-imines.
Table 1 Monitoring imine formation by ReactIR
Substrate 1
Addition product 2/3a
Entry
Timeb (min)
1
2
3
1a
1b
1c
2ac
2b
2c
12
15
50
4
5
1d
1e
2d
2e
90
a Centre for Sustainable Chemical Processes, Dept. of Chemistry,
Durham University, South Road, Durham DH1 3LE, UK.
E-mail: andy.whiting@durham.ac.uk
100
b Dept. Quı´mica Fı´sica i Inorga`nica, Universitat Rovira I Virgili,
43007 Tarragona, Spain
6
7
1f
2f
360
30
1g
3g
w Electronic supplementary information (ESI) available: Experimental
details, characterisation data, ReactIR, 1H and 13C NMR and crystallo-
graphic data in CIF format. CCDC 896882. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/
c2cc36129a
a
1,2-, 1,4-addition. Conditions: THF (7 mL), 3 A MS (2.5 g), 1 (2.8 mmol)
b
and BnNH2 (2.8 mmol). Reaction completion. 2a – unstable.
c
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 11401–11403 11401

View Article Online
(entry 6, Table 2) and without oxazine 6h. Deliberate for-
mation of oxazine 6h could also be achieved by increasing the
MeOH and oxidant concentrations in the final step leading to
the formation of 6h in 51% yield (entry 9, Table 2).
Following the optimisation (Table 2), the conversion of
further unsaturated aldehydes and ketones to the corresponding
amino alcohols 5 was investigated (Table 3). This proved to be
successful, resulting in yields varying from 20–90% of the
amino alcohols 5 (Table 3). The structure of 5f was confirmed
by single crystal X-ray structure determination (Fig. 2). For
most substrates, the more nucleophilic phosphine, PnBu3 (i.e.
see entries 1–2, 5–10, Table 3), was required to facilitate efficient
b-boration. However, purification of some g-amino alcohols
(i.e. 5c, d and j) proved difficult due to high affinity to silica gel.
In these cases, and as investigated on 1h (entry 9, Table 2 and
entry 7, Table 3), rather than isolation of the g-amino alcohol,
conversion to the oxazine 5 was achieved by addition of CH2O
to the crude amino alcohols, i.e. 5c, d and j, resulting in 42–75%
yields over 5 steps of the corresponding oxazines 6c, d and j
(Table 3, entries 2, 3 and 9).
Fig. 1 ReactIR plots over time for the formation of 2d via 1,2-addition
of benzylamine to methacrolein 1d, rt, THF, 3 A MS.
(1698 cmÀ1) and concomitant gain of the CQN (asym + sym)
stretches (1640 & 1621 cmÀ1).
The in situ-generated a,b-unsaturated imines 2 formation was
exploited by trapping with a borylnucleophile, i.e. developing a pro-
cess not dissimilar to reductive amination where imine trapping is
achieved using a hydride nucleophile.13 In theory, it should be
possible for in situ-formed imines to be utilised in a one-pot route
to g-amino alcohols 5 without isolation. Hence, initially chalcone
1h (Table 2) was investigated using various Cu–ligand–base com-
binations and imine formation strategies (i.e. comparing the pre-
formed imine with that formed in situ). Normal6 catalyst loadings
needed to be increased from 1 to 5% to improve conversion of 1h
to 5h (Table 2, entries 1–3), and PnBu3 rather than PPh3 also
facilitated the formation of 5h while minimising side products
(Table 2, entries 4–6). Intriguingly, the formation of 5h appeared to
plateau to only 62–63% (see entries 3–6, Table 2), however, on
further investigation, it was found that oxazine 6h was also formed
(e.g. entry 1, Table 2). This was unexpected since there was no
obvious source of a formaldehyde equivalent in the reaction to
explain the transformation of 4h into 5h.12 MeOH solvent used
for the C–B oxidation step was most likely to act as a precursor
of formaldehyde (or equivalents thereof). This was circum-
vented by removing the MeOH (by evaporation) after the
reduction step, which resulted in the formation of 5h in 90% yield
ReactIR studies on the formation of the a,b-unsaturated
imines derived from 1i and j revealed that these substrates
reacted significantly slower than 1a–g with BnNH2 and in fact
were difficult to follow by ReactIR. Therefore, 1i and j and
BnNH2 were added simultaneously to the borylation reaction
in the presence of 3 A molecular sieves (see ESIw) with the
expectation that it might be possible to trap the more reactive
unsaturated imine by borylation. Indeed, this strategy worked
(see entries 8 and 9, Table 3) resulting in the formation of 5i
and 6j in reasonable overall yields.
The asymmetric potential of this 4-step (or 5), one-pot method
was also investigated using 1h reacting with BnNH2 with a
Josiphos-type chiral diphosphine 7 (Scheme 2). Scheme 2b
shows the formation of the b-boryl imine 4h in up to 92% ee
(+)-(R). Importantly, the asymmetric induction was almost
identical to that obtained when the enantioselective b-boration
took place from the isolated the a,b-unsaturated imine (Scheme 2),
Table 2 Optimisation of the four-step, one-pot methodology on chalcone 1 h
Conversiond (%)
Isolated yield (%)
dej of the isolated productd (%)
Entry
CuCl (%)
L (%)
Base (%)
Timec (h)
5h
6ha
5h
6h
5hj
6h
1
2
3
4
5
6
1
3
5
5
5
5
PPh3 (2)
PPh3 (6)
KOt-Bu (20)
NaOt-Bu (9)
KOt-Bu (18)
KOt-Bu (18)
KOt-Bu (18)
NaOt-Bu (18)
KOt-Bu (36)
NaOt-Bu (15)
NaOt-Bu (15)
24
24
24
37
42
63
62
30
29
27
36
34
0
27
34
54
17
40
62
56
63
90
25
—
—
32
—
—
—
—
—
—
30
51
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
anti >99%
PPh3 (10)
PPh3 (10)
PnBu3 (10)
PnBu3 (10)
PPh3 (20)
PPh3 (10)
PPh3 (10)
48
18e
18f
24
63
>95
40
52
7
10i
5
8g
9h
18e
18e
5
44
a
b
6h was confirmed by the reaction of 5h with CH2O (1.1 equiv.) in THF, rt, 4.5 h (74% yield). NaBH4 (4.2 mmol), MeOH (3 mL). Time for
c
reaction of 1h, benzylamine and Cu–B cat. in one-pot. Determined by 1H NMR. Imines were formed in situ (1 : 1 amine:a,b-unsaturated
carbonyl, 3 A MS, THF, 6 h) and transferred to Cu–B cat (18 h). The same as entry 5 except MeOH removed prior to oxidation. [O] NaOH,
d
e
f
g
h
H2O2 (1 : 1, 20 equiv.), MeOH (10 ml), 4 h reflux. [O] NaOH, H2O2 (1 : 1, 40 equiv.), MeOH (15 ml), 4 h reflux. High catalyst loadings favour
i
boration of the a,b-unsaturated carbonyl without formation of imine. See previous work.6a
j
c
11402 Chem. Commun., 2012, 48, 11401–11403
This journal is The Royal Society of Chemistry 2012

View Article Online
Table 3 Four-step, one-pot route to g-amino-alcohols
Conversione (isolated yield)
(%)
dee (%)
Entry Substrate
5
6
5
6
1h
—
—
2
—
—
—
—
—
—
Scheme 2 Asymmetric borylation by in situ imine formation followed
by b-boration.
3h
boration of a,b-unsaturated ketone 1h followed by imine
formation of the resulting b-boryl ketone 7h (see ESIw).
In summary, a stereoselective 4-step, one-pot protocol for the
synthesis of g-amino alcohols in 20–90% yields has been developed.
A 5-step version to 1,3-oxazines has also been demonstrated which
exhibits impressive efficiency (42–75%) considering the number of
steps. The asymmetric potential has been demonstrated and this
methodology is being developed for the control of multiple
stereogenic centres. In addition, although a,b-unsaturated imines
are little used or studied compared to their carbonyl analogues,
their formation can be followed by in situ IR, and subsequent
trapping by borylation is an ideal way to demonstrate their
formation. Further applications will be reported in due course.
We thank the EPSRC for a grant (to ADJC), MEC for funding
(CTQ2010-16226) and a grant (to CS), and the EPSRC Mass
Spectrometry Service, Swansea.
4h
1d
rac-anti rac-anti
>99
5
>99
rac-anti
>99
6
rac-anti rac-anti
>99
7i
8
>99
40
Notes and references
1 (a) L. Mantilli and C. Mazet, ChemCatChem, 2010, 2, 501–504;
(b) K. Muther, M. Oestreich and J. A. Schiffner, Angew. Chem.,
Int. Ed., 2010, 49, 1194–1196; (c) E. Hartmann, M. Oestreich and
D. J. Vyas, Chem. Commun., 2011, 47, 7917–7932; (d) A. D. J.
Calow and A. Whiting, Org. Biomol. Chem., 2012, 29, 5485–5497;
´ ´ ´
(e) J. Cid, J. J. Carbo, E. Fernandez and H. Gulyas, Chem. Soc.
Rev., 2012, 41, 3558–3570.
2 S. Mun, J.-E. Lee and J. Yun, Org. Lett., 2006, 8, 4887–4889.
3 H.-U. Blaser, Chem. Rev., 1992, 92, 935–952.
4 See: (a) H. Geng, G. Hou, W. Wu, W. Zhang, X. Zhang, L. Zhou
and Y. Zou, Angew. Chem., Int. Ed., 2009, 48, 6052–6054;
(b) W. Gao, D. Liu, C. Wang and X. Zhang, Angew. Chem.,
Int. Ed., 2005, 44, 1687–1689.
5 J. A. Ellman, T. Kochi and T. P. Tang, J. Am. Chem. Soc., 2002,
124, 6518–6519.
¨
9
>95
a
Standard borylation conditions: CuCl (5%), PnBu3 (10%), NaOtBu
b
(15%), B2pin2 (1.1 equiv.), MeOH (2 equiv.), THF. In situ imine
formation (0–7 h), see ESI. c NaBH4 (excess), MeOH (2 h), removal of
d
MeOH under reduced pressure. NaOH, H2O2 oxidation (THF, reflux
1 hour). Determined by 1H NMR of isolated 5/6, see ESI. 5 stirred in
THF and CH2O (1.1 equiv.) overnight, 6 obtained by column
e
f
g
chromatography. 64%-inseparable impurity (see ESI). Standard
h
i
conditions, except PPh3 (10%) used as ligand. Standard conditions,
except NaOtBu (18%) used as base.
6 See: (a) E. Ferna
Catal., 2011, 353, 376–384; (b) E. Ferna
Mata, A. Tatla and A. Whiting, Chem.–Eur. J., 2011, 17, 14248–14257.
7 E. Fernandez and C. Sole, Chem.–Asian J., 2009, 4, 1790–1793.
´
ndez, H. Gulya
´
s, C. Sole
´
and A. Whiting, Adv. Synth.
´
ndez, H. Gulya
´
s, C. Sole, J. A.
´
´
´
8 S. A. Moyer, S. D. Pearce, J. W. Rigoli and J. M. Schomaker, Org.
Biomol. Chem., 2012, 10, 1746–1749.
9 R. G. Bergman, D. A. Colby and J. A. Ellman, J. Am. Chem. Soc.,
2008, 130, 3645–3651.
10 C. F. Carter, H. Lange, S. V. Ley, I. R. Baxendale, B. Wittkamp, J. G.
Goode and N. L. Gaunt, Org. Process Res. Dev., 2010, 14, 393–404.
11 S. Gomez, T. Maschmeyer and J. A. Peters, Adv. Synth. Catal.,
2002, 344, 1037–1057.
12 G. Bertoli, C. Cimarelli, E. Marcantoni, G. Palmieri and
M. Petrini, J. Org. Chem., 1994, 59, 5328–5335.
Fig. 2 Olex213 thermal ellipsoid plot (50% probability) of 5f.
as well as the absolute stereochemistry. This is consistent with
13 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard
and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
in situ imine formation followed by boration and not direct
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 11401–11403 11403