α-Hydroxyalkyl heterocycles via chiral allylic boronates: Pd-catalyzed borylation leading to a formal enantioselective isomerization of allylic ether and amine

Article abstract of DOI:10.1021/ja903946f

(Chemical Equation Presented) An efficient catalytic enantioselective preparation of synthetically useful pyranyl and piperidinyl allylic boronates was achieved via a palladium-catalyzed borylation/isomerization reaction on the corresponding alkenyl triflates. The influence of the base and solvent was found to be crucial on the regio- and enantioselectivity of this reaction. The overall borylation process constitutes a successful example of formal asymmetric isomerization of allylic ether/amine. The resulting allylic boronate reagents add to various aldehydes in a one-pot process to give synthetically useful alpha-hydroxyalkyl derivatives in high stereoselectivity.

Full text of DOI:10.1021/ja903946f

Published on Web 06/24/2009
r-Hydroxyalkyl Heterocycles via Chiral Allylic Boronates:
Pd-Catalyzed Borylation Leading to a Formal Enantioselective Isomerization
of Allylic Ether and Amine
Ste´phanie Lessard, Feng Peng, and Dennis G. Hall*
Department of Chemistry, UniVersity of Alberta, Edmonton, AB, T6G 2G2, Canada
Received May 14, 2009; E-mail: dennis.hall@ualberta.ca
Several natural products and synthetic drugs are made of oxygen
or nitrogen containing rings flanked by a hydroxyalkyl substituent next
to the heteroatom. Notable examples include higher order sugars (e.g.,
ꢀ-KDO, a key component of the lipopolysaccharide cell wall of Gram-
negative bacteria), numerous piperidine alkaloids, marine oligopyrans,
and the antihypertension drug Nebivolol.1 A powerful and highly
stereoselective approach to these important heterocycles exploits the
allylboration of aldehydes with chiral cyclic allylic boronates.2 Despite
its many attributes, this strategy is limited by a dearth of enantiocon-
trolled preparative methods for the required allylic boronates. One
method, reported by our group, exploits a catalytic enantioselective
hetero [4+2] cycloaddition approach to a pyranyl allylic boronate,
which can react with high diastereoselectivity with aldehydes to provide
the desired R-hydroxyalkyl dihydropyrans (eq 1).3
temperature, and other parameters. Indeed, as shown in Table 1,
we found that the nature of the base has a determining influence
on the course and enantioselectivity of this borylation reaction.
Table 1. Optimization of Base and Temperaturea
conversion %b
entry
base (equiv)
temp (°C)
3
2
ee of 3 (%)c
1
2
3
4
5
6
7
8
Et3N (3.0)
(i-Pr)MeNt-Bu (3.0)
DTBMP (3.0)
s-collidine (3.0)
proton sponge (1.5)
PhNMe2 (3.0)
PhN(i-Pr)2 (3.0)
4-MeOC6H4NMe2 (3.0)
4-FC6H4NMe2 (3.0)
PhNMe2 (3.0)
80
80
80
80
80
80
80
80
80
25
25
25
25
25
100
100
57
35
65
84
75
87
-
84
17
86
88
88
-
-
56
49
84
53
74
87
78
79
-
86
-
89
91
92
28
<5
35
16
25
13
15
16
62
14
12
12
9
As highlighted in our total synthesis of thiomarinol,4 the residual
ethoxy substituent on the product can be extraneous toward several
applications, and additional operations are required for its removal.
With a view to expand the breadth of this approach, we longed for
developing a general method that would circumvent the formation of
an acetal and that could be suitable to the preparation of other
heterocycles beyond just pyrans. In this regard, our attention was
attracted to a report by Masuda and co-workers on the borylation of
alkenyl triflate 1.5 Instead of the expected alkenylboronate 2, the
authors observed allylic boronate 3 as the major product (eq 2).
10
11
12d
PhNMe2 (0.75)
PhNMe2 (1.1)
13d e PhNMe2 (1.1)
,
14d f PhNMe2 (1.1)
,
a Reaction conditions: 0.25 mmol of 1 at 0.0625 M. Except in entries
12–14, 1.5 equiv of pinBH were used. See Supporting Information (SI)
for detailed procedures. b Measured by 1H NMR (300 MHz in CDCl3)
of the crude reaction mixture. The remainder of 100% is unreacted 1.
c Measured by chiral HPLC (see SI for details) on the hydrocin-
namaldehyde allylboration product 4a (cf. eq 3a). Absolute configuration
assigned by chemical correlation (see SI). d 1.1 equiv of pinBH.
e Solvent is dioxane. f Solvent is dioxane, reaction time: 4 h.
While trialkylamine bases effectively suppressed the formation of
alkenylboronate 2, the enantiomeric excess (ee) of desired product 3
was relatively low (entries 1-2). Aromatic amines such as 2-6-di-
tert-butylmethylpyridine led to a much higher ee, but a substantial
proportion of undesired 2 was observed (entry 3). Aniline derivatives
were examined next because of their similar basicity to that of
pyridines. To our delight, the use of N,N-dimethylaniline (DMA) led
to a clean and reproducible reaction with an optimal enantioselectivity
of 87% ee for 3 and only a small proportion of alkenylboronate (2)
(entry 6). Larger N-alkyl substituents led to a lower ee (entry 7) and
so did electronic modifications (entries 8-9). With DMA as the optimal
base, the effect of its stoichiometry and that of pinBH, as well as the
reaction temperature, were examined.10 The temperature was found
to have essentially no effect on the ee of 3 above 25 °C (compare
entries 6 and 10). Whereas the use of a single equivalent of base is
sufficient, a substoichiometric amount is highly detrimental to the 3:2
selectivity (entry 11). The use of a slight excess of pinBH was found
Intrigued by this process and encouraged by its potential as a
preparative method for heterocyclic allylic boronates, we sought to
optimize this formal isomerization toward a catalytic enantioselective
formation of allylboronate 3. This project presented a challenge because
the asymmetric isomerization of allylic ethers6 has been much less
successful compared to the analogous isomerization of allylic amines,7
which is even applied industrially in the ton-scale synthesis of menthol.8
A first round of optimization with 1 and pinacolborane (pinBH)
was aimed at perfecting the enantioselectivity of allylboronate
product 3 while suppressing the formation of alkenylboronate 2.
An initial screen of different chiral ligands and reaction conditions
(source of palladium, solvent) revealed that the use of TANIA-
PHOS9 (see structure in Figure 1) and Pd(OAc)2 in toluene
constituted a good starting point from where to optimize base,
9
9612 J. AM. CHEM. SOC. 2009, 131, 9612–9613
10.1021/ja903946f CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
to be suitable, leading to the conditions of entry 12.11 One last round
of fine-tuning10 examined the influence of solvent, ligand, and reaction
time under the provisionally optimal conditions (1.1 equiv of DMA
and pinBH, 25 °C, 16 h). Although no diphosphine ligand bettered
TANIAPHOS,10 use of dioxane as solvent led to an improved ee of
91% (entry 13). Furthermore, the reaction is complete within 4 h (entry
14; 92% ee).
Realizing the sensitive nature of allylic boronates, we sought to
develop a one-pot borylation/isomerization/allylation protocol that
would circumvent the isolation of 3. This was achieved by simple
addition of the aldehyde and increasing the temperature to 80 °C
(eq 3). Considering it is a two-stage process, products of representa-
tive aldehydes were obtained in acceptable isolated yields and with
high stereoselectivity.12
This novel catalytic enantioselective process is proving suitable to
prepare other classes of heterocyclic allylic boronates. When subjected
to similar conditions, the parent N-Boc piperidine 5 provided the
allylboration products 7 in good yields with 86% ee (eq 4).12 Compared
to the pyran analogue, the intermediate allylboronate 6 was formed in
a lower 4:1 ratio with the corresponding alkenylboronate.10 Aldehyde
allylation with 6 proceeded cleanly under microwave heating. Further
optimization could focus on varying the carbamate O-substituent.
In conclusion, an efficient catalytic enantioselective preparation of
heterocyclic allylic boronates was described. The overall borylation
constitutes a successful example of formal asymmetric isomerization
of allylic ether/amine. Reagents 3 and 6 add to various aldehydes to
give useful R-hydroxyalkyl pyrans and piperidines in high stereose-
lectivity. Applications and extensions of this method are underway.
Acknowledgment. This work was funded by the Natural Sciences
and Engineering Research Council (NSERC) of Canada and the
University of Alberta. S.L. thanks the U of A for a Queen Elizabeth
II Graduate Scholarship. The authors thank Solvias AG (Dr. H. Steiner
and Dr. H.-U. Blaser) for a generous gift of ligands and Michelle
Morrow for assistance with preliminary experiments.
A tentative mechanistic cycle must reconcile the unlikeliness of 2
being an intermediate. The 3:2 ratio remains essentially unchanged
with temperature and time under the optimal reaction conditions.10
Moreover, a control using a 50:50 mixture of 1 and 2 as substrates
produced 3 and 2 in a 45:55 ratio.13 Alkenylboronate 2 is the sole
product when B2pin2 is employed, which suggests the involvement of
a Pd hydride species when pinBH is used.14 Thus, a stereodetermining
hydropalladation (1 to II via I) may be key to the isomerization to III
(via ꢀ-H elimination) (Figure 1). It would be followed by a stereospe-
cific and regioselective allylic borylation of the reactive intermediate
III into 3, via IV and V. The deuteration pattern of 6 prepared from
pinBD and 5 is consistent with this proposed cycle (see box, Figure
1).10 The role of the base, however, is unclear at this preliminary stage.
Supporting Information Available: Full experimental details,
additional tables of optimization studies, chiral HPLC chromatograms,
and NMR spectral reproductions for new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Prisant, L. M. J. Clin. Pharmacol. 2008, 48, 225–239.
(2) (a) Tailor, J.; Hall, D. G. Org. Lett. 2000, 2, 3715–3718. (b) Deligny, M.;
Carreaux, F.; Carboni, B.; Toupet, L.; Dujardin, G. Chem. Commun. 2003,
276–277.
(3) (a) Gao, X.; Hall, D. G. J. Am. Chem. Soc. 2003, 125, 9308–9309. (b)
Gao, X.; Hall, D. G.; Deligny, M.; Favre, A.; Carreaux, F.; Carboni, B.
Chem.sEur. J. 2006, 13, 3132–3142.
(4) Gao, X.; Hall, D. G. J. Am. Chem. Soc. 2005, 127, 1628–1629.
(5) Murata, M.; Oyama, T.; Watanabe, S.; Masuda, Y. Synthesis 2000, 778–
780.
(6) For rare examples, see: (a) Hiroya, K.; Kurihara, Y.; Ogasawara, K. Angew.
Chem., Int. Ed. Engl. 1995, 34, 2287–2289. (b) Faitg, T.; Soulie´, J.;
Lallemand, J.-Y.; Mercier, F.; Mathey, F. Tetrahedron 2000, 56, 101–104.
(7) Krompiec, C.; Krompiec, M.; Penczek, R.; Ignasiak, H. Coord. Chem. ReV.
2008, 252, 1819–1841.
(8) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis; John Wiley &
Sons: New York, 1994; Chapter 2, pp 95-121. (b) Otsuka, S.; Tani, K.
Synthesis 1991, 665. (c) Inoue, S.-i.; Takaya, H.; Tani, K; Otsuka, S.; Sato,
T.; Noyori, R. J. Am. Chem. Soc. 1990, 112, 4897–4905, and references
cited.
(9) Type SL-T001-1 was used. Ireland, T.; Tappe, K.; Grossheimann, G.;
Knochel, P. Chem.sEur. J. 2002, 8, 843–852. Structure corrigendum:
Chem.sEur. J. 2008, 14, 3509.
(10) See details and additional results in the Supporting Information.
(11) At this stage, one last screen of several Pd sources confirmed that Pd(OAc)2
is the most suitable precatalyst although Pd(O2CCF3)2, Pd(PPh3)4, and PdCl2
were equally efficient.
(12) Unoptimized isolated yields from a reaction scale of 1.0 mmol of 1 or 5.
The ee’s of products 4a, 4c-e, and 7a-b were measured by chiral HPLC.10
The ee of 4b was measured by 19F NMR of its Mosher ester.10 Values for
all products are similar within analytical error: it is known that pyranyl
allylic boronates similar to 3 (cf. eq 1) do not suffer erosion of optical
purity in the allylboration process (ref 3b). As in the allylboration of eq 1
(ref 3b), only one diastereomer was identified.
(13) Product 2 was prepared independently according to: Takagi, J.; Takahashi,
K.; Ishiyama, T.; Miyaura, N. J. Am. Chem. Soc. 2002, 124, 8001–8006.
(14) Murata, M.; Oyama, T.; Watanabe, S.; Masuda, Y. J. Org. Chem. 2000,
65, 164–168.
Figure 1. Postulated mechanism for the borylation/isomerization of 1 to
give 3. (Note: the chiral diphosphine ligand was simplified.)
JA903946F
9
J. AM. CHEM. SOC. VOL. 131, NO. 28, 2009 9613