Enantioselective halogenative semi-pinacol rearrangement: A stereodivergent reaction on a racemic mixture

Article abstract of DOI:10.1039/c4cc05923a

An efficient, quantitative deracemization strategy for optically inactive allylic cycloalkanols has been achieved using the biphasic halogenative semi-pinacol reaction protocol. The resultant β-halo spiroketones, containing three contiguous stereogenic ce

Full text of DOI:10.1039/c4cc05923a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enantioselective halogenative semi-pinacol
rearrangement: a stereodivergent reaction
on a racemic mixture†
Cite this: Chem. Commun., 2014,
50, 13461
Received 31st July 2014,
Accepted 4th September 2014
Fedor Romanov-Michailidis,a Marion Pupier,a Laure Gueneeb and
´ ´
Alexandre Alexakis*a
DOI: 10.1039/c4cc05923a
www.rsc.org/chemcomm
An efficient, quantitative deracemization strategy for optically phosphate anion-mediated phase-transfer procedure reprted by
inactive allylic cycloalkanols has been achieved using the biphasic Toste et al.8 constituted a remarkably general solution to the
halogenative semi-pinacol reaction protocol. The resultant b-halo above problem.4 As a subsequent challenge, we were interested
spiroketones, containing three contiguous stereogenic centers, in studying the influence of a pre-existing stereogenic center on
were easily recovered with high diastereomeric and enantiomeric the stereochemical outcome of the asymmetric halogenation/
purities following conventional silica gel chromatography. The semi-pinacol reaction. In our opinion, if the level of stereo-
optically active products could be further manipulated chemically, induction by the chiral, enantiopure phosphate anion would be
affording synthetically interesting scaffolds with complete preser- strong enough then the initial, optically inactive material (rac-Ax)
vation of stereoisomeric integrity.
could be quantitatively deracemized and transformed into a
mixture of diastereomeric, optically active b-halo spiroketones
R
S
Kinetic resolution (KR) describes a means of differentiating two (Bx and Bx ) (Scheme 1).
enantiomeric substrates in a racemic mixture. As opposed to
Indeed, such a process would follow a stereochemical scenario
chiral resolution, which relies on different physical properties where each enantiomer of the starting racemic mixture is trans-
of diastereomeric products, kinetic resolution relies upon formed using an enantiopure catalyst into two products that do not
differences in reaction rates between the two enantiomers.1 share an enantiomorphic relationship. Moreover, if it happens that
Although numerous methods that are highly efficient in terms the two products would be formed in high enantiomeric excesses
of enantiomer discrimination have been reported in the recent and would be easily separable, then the latter reaction would
years,2 the obligation that only half of the starting material constitute a rare example of a stereodivergent process.5
participates in the reaction compromises the productivity of the
To our great delight, upon subjecting the chiral, racemic
process. Furthermore, due to the fact that the two enantiomeric allylic cyclobutanol rac-A1 to the biphasic fluorinative reaction
species react simultaneously and at different rates, the relative conditions (Selectfluor, Na3PO4, chiral phosphoric acid L1,9 and
concentration of residual substrate enantiomers changes as the toluene) at room temperature, full consumption of the starting
reaction proceeds. As a consequence, the enantiomeric excesses of material was observed after two days of stirring (Table 1, entry 1).
the substrate and the product become a function of the conversion.3 Thin-layer chromatography of the crude reaction mixture
Driven by the demand of highly productive chemical processes, indicated the presence of two new, non-enantiomeric products
quantitative transformations of racemates into stereoisomerically with a DRf of ca. 0.4. Facile separation of the two products by
pure products are of high value to the synthetic community.7
Our group has been long interested in the development
of catalytic, enantioselective protocols for halogenative
semi-pinacol rearrangements of strained allylic cycloalkanols.6
More specifically, we have recently found out that the chiral
a Department of Organic Chemistry, University of Geneva, Quai Ernest Ansermet 30,
CH-1211 Geneva 4, Switzerland. E-mail: Alexandre.Alexakis@unige.ch
b Laboratory of Crystallography, University of Geneva, Quai Ernest Ansermet 24,
CH-1211 Geneva 4, Switzerland
† Electronic supplementary information (ESI) available. CCDC 1015663–1015666.
Scheme 1 Proposed asymmetric halogenation/semi-pinacol reaction of
chiral, racemic cycloalkanols rac-Ax.
For ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c4cc05923a
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13461--13464 | 13461

View Article Online
Communication
ChemComm
Table 1 Optimization of the conditions for the stereodivergent fluorination/
semi-pinacol reactiona
Ratiob
e.r.c
S
S
Entry Lx
Solvent
T (1C) t (h) A1/B1R/B1
B1R/B1
1
2
3
4
5
6
7e
L1 Toluene
L2 Toluene
L1 PhF–Hex
L1 PhF–Hex À12
L3 PhF–Hex À12
L4 PhF–Hex À12
L4 PhF–Hex À12
25
25
25
48
48
48
48
96
96
96
0 : 1 : 1
0 : 1 : 1
0 : 1 : 1
1 : 5.3 : 4.7
0 : 1 : 1
0 : 1 : 1
0 : 1 : 1
93 : 7/93 : 7
88 : 12/82 : 18
94 : 6/92 : 8
95 : 5/94 : 6d
95.5 : 4.5/95 : 5
96 : 4/96 : 4
96 : 4/96 : 4
Scheme 2 Substrate scope of the fluorination/semi-pinacol reaction,
performed on racemic allylic cycloalkanols rac-Ax.
a
b
Reaction conditions: see ESI. Determined by 1H NMR analysis of
ratios of both diastereomers were invariably high (ca. 95: 5 e.r. in
most cases), thus avoiding complications inherent to matched/
c
unpurified reaction mixtures. Determined by chiral HPLC analysis.
d
e
The enantiomeric excess of unreacted A1 was 7%. 5 mol% catalyst
mismatched combinations.
loading was used.
R
Single crystals of chiral, enantiopure b-fluoro spiroketones B2
,
B3 and B4 were grown for X-ray diffraction analysis.10,11 It is
S
S
conventional silica gel chromatography, followed by 1H/13C/19
F
fascinating to note that, as predicted from solution-phase
NMR analysis of the resultant pure fractions indicated that the 1H/13C/19F NMR spectral analyses,11 both enantiomers of chiral,
two products were in fact diastereomers (referred to as B1R and racemic allylic alcohols underwent the semi-pinacol rearrange-
S
B1 , depending on the absolute configuration of the derace- ment through Re-face fluorination. This result is consistent with
mized benzylic stereogenic center). The exceptionally large Rf reactivity trends of the ‘‘normal’’ prochiral allylic alcohols.4a
difference between the two diastereomers allowed the recovery Presumably, the pre-existing stereogenic center in substrates
of very pure compounds, with isolated yields close to the rac-Ax has little-to-zero influence on the reaction stereochemistry,
theoretical maximum of 50% for each. Most importantly, the which is in turn entirely dominated by the absolute configuration
S
enantiomer ratios of product b-fluoro spiroketones B1R and B1 , of the chiral, enantiopure phosphate anion.
containing three contiguous stereogenic centers, were high (93 : 7
The irrelevance of the relative orientation of the benzylic methyl
e.r. for both). The level of asymmetric induction could be group for the sense of absolute induction was demonstrated by
improved by further optimizing the reaction conditions (Table 1). subjecting the gem-dimethyl-substituted allylic alcohol A10 to the
The stereodivergent reaction displayed higher levels of fluorination/semi-pinacol sequence (Scheme 3). The expected
asymmetric induction when carried out in the fluorobenzene– b-fluoro spiroketone B10 was isolated with a high enantiomeric
n-hexane mixture instead of toluene (compare entries 1 and 3). excess value (94.5 : 5.5 e.r.), providing evidence that both the ‘‘up’’
Lowering the temperature led to a slower albeit more enantio- and ‘‘down’’ orientations of the benzylic C1-methyl group were
selective reaction, which in turn required longer reaction times readily accommodated into the chiral pocket of the catalyst, leading
and the use of the more lipophilic phosphoric acid catalysts to similar reaction rates and high levels of asymmetric induction
(L3 and L4). It is interesting to note that no kinetic resolution for both orientations.
occurred during the present fluorination/semi-pinacol reaction.
Subsequently, the stereodivergent process was extended to the
As evidenced from entry 4, when stopped at incomplete conver- bromination-initiated semi-pinacol rearrangement (Table 2). To
sion, the recovered unreacted allylic alcohol A1 was quasi-racemic this end the chiral, racemic cyclobutanol rac-A1 was reacted with
(7% ee). This indicates that the two enantiomers of the starting DABCO-derived triply charged cations T1–T3 and chiral phosphoric
material reacted with the enantiopure catalyst at similar rates.
Having established the appropriate experimental conditions for
the stereodivergent fluorination/semi-pinacol reaction, the substrate
scope of this new methodology was investigated next (Scheme 2).
The above-described fluorinative deracemization strategy worked
equally well with substituted chiral, racemic cyclobutanols (A1–A5)
and cyclopropanols (A6–A9), both based on the C1-methylated
dihydronaphthalene scaffold. In each case, both diastereomers of
S
the product b-fluoro spiroketones (BxR and Bx ) could be isolated in
Scheme 3 Probing the influence of the relative orientation of the
benzylic methyl group on the level of asymmetric induction.
high yields due to very large DRf separations. The enantiomer
13462 | Chem. Commun., 2014, 50, 13461--13464
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Table 2 Optimization of the conditions for the stereodivergent bromination/
semi-pinacol reactiona
Ratiob
e.r.c
S
S
Entry
Tx
Solvent
t (h)
A1/C1R/C1
C1R/C1
Scheme 5 Investigation of the influence of the size of the benzylic substituent
on the enantioselectivity of the stereodivergent halogenation/semi-pinacol
reaction.
1
2
3
T1
T2
T3
T3
T3
PhEt
PhEt
PhEt
PhEt
PhEt
24
24
24
72
72
1 : 1.5 : 1.7
1 : 1.9 : 2.3
1 : 4.6 : 4.9
0 : 1 : 1
93 : 7/93 : 7
92.5 : 7.5/93.5 : 6.5
95 : 5/95:5d
4
95 : 5/95 : 5
89 : 11/93 : 7
5e
1 : 2.2 : 3.9
Reaction conditions: see ESI. b Determined by 1H NMR analysis of unpur-
ified reaction mixtures. c Determined by chiral HPLC analysis. d The enan-
tiomeric excess of unreacted A1 was 5%. e 5 mol% catalyst loading was used.
a
Single crystals of the chiral, enantiopure b-bromo spiroketone
C2S were grown for X-ray diffraction analysis.10,11 This allowed the
assignment of the relative and absolute configurations of products.
It is interesting to note that electrophilic bromination takes
place through the Si-face of the starting allylic alcohol rac-A2, as
opposed to the Re-face fluorination discussed above. This
switch in the sense of absolute induction when moving from
fluorine to the heavier halogens, while keeping the absolute
configuration of the phosphate anion intact, is consistent with
our earlier observations made by studying the iodination of
‘‘normal’’ prochiral allylic alcohols.4b
The importance of the size of the benzylic substituent in the
substrate allylic alcohols was addressed next. To this end, a set
of dihydronaphthalene-based cyclobutanols containing either
an ethyl (A10), a n-butyl (A11), or an isopropyl (A12) substituent at
the benzylic position were synthesized and subjected to the
stereodivergent halogenative reaction, under our previously
described optimized conditions (Scheme 5).
As can be seen from Scheme 5 above, the steric size of the
benzylic substituent had only an insignificant effect on the yield
and enantioselectivity of the stereodivergent halogenation/
semi-pinacol reaction. A small increase in the level of asym-
metric induction for the more sterically congested congeners
was nevertheless noted. Thus, upon comparing the ethyl-(B11)
and isopropyl-substituted (B13) b-fluoro spiroketones, a slightly
higher enantiomer ratio (about 2%) was observed for the latter. A
similar trend was also noted for the brominative reaction as well
(compare b-bromo spiroketones C10 and C12).
acid (Ra)-H8-TRIP (L5), under reaction conditions similar to those
reported recently for the iodination/semi-pinacol reaction.4b
As can be seen from Table 2, upon employing the brominating
reagent T3 in ethylbenzene at room temperature (entry 4), the
bromination/semi-pinacol reaction provided a successful platform
for performing the stereodivergent reaction. As already observed for
the fluorination case, the moderately large DRf separations (ca. 0.2–0.3)
between the two diastereomeric products (C1R and C1S) allowed the
recovery of very pure materials. Furthermore, the products were formed
with very high levels of asymmetric induction (up to 97 : 3 e.r.), and with
isolated yields close to the theoretical maximum of 50%. Once again,
no kinetic resolution was observed during the present bromination/
semi-pinacol rearrangement sequence (entry 3).
The optimal conditions being established, the substrate
scope of the stereodivergent, brominative semi-pinacol reaction
was probed next (Scheme 4).
Furthermore, the synthetic utility of b-halo spiroketones
was proven by carrying out a set of derivatization reactions.
Synthetic manipulation of products involved a stereospecific
Baeyer–Villiger oxidation of spiro-cyclobutanone B8R, as well
as diastereoselective reduction of spiro-cyclopentanones
R
B2R, B5 and C4S. Details of these transformations are given
in the ESI.†11
In the context of the halogenation/semi-pinacol rearrange-
ment sequence, the interaction between a chiral, enantiopure
Scheme 4 Substrate scope of the bromination/semi-pinacol reaction,
performed on racemic allylic cycloalkanols rac-Ax.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 13461--13464 | 13463

View Article Online
Communication
ChemComm
H. Takaya, J. Org. Chem., 1988, 53, 708–710; (g) A. H. Hoveyda and
R. R. Schrock, Chem. – Eur. J., 2001, 7, 945–950.
3 (a) G. Bredig and K. Fajans, Ber. Dtsch. Chem. Ges., 1908, 41, 752–762;
(b) V. S. Martin, S. S. Woodward, T. Katsuki, Y. Yamada, M. Ikeda and
K. B. Sharpless, J. Am. Chem. Soc., 1981, 103, 6237–6240.
phosphate anion and a chiral, racemic allylic cycloalkanol
displayed two salient features that allowed its use in a successful
stereodivergent reaction. First, both enantiomers of the starting
material reacted at similar rates. Second, the relative orientation of
the benzylic substituent in the substrate had little-to-zero influence
on the level of asymmetric induction. These ‘‘abnormalities’’,
combined with the fortuitously large DRf separations of the diastereo-
meric products for a broad range of substrate cycloalkanols, led to a
very efficient deracemization technique described in this commu-
nication. The product b-halo spiroketones were recovered with very
high stereochemical purities and could be further manipulated
chemically.
´ ´
4 (a) F. Romanov-Michailidis, L. Guenee and A. Alexakis, Angew.
Chem., Int. Ed., 2013, 52, 9266–9270; (b) F. Romanov-Michailidis,
´ ´
L. Guenee and A. Alexakis, Org. Lett., 2013, 15, 5890–5893.
5 (a) S. El Baba, J.-C. Poulin and H. B. Kagan, Tetrahedron, 1984, 40,
4275–4284; (b) H. B. Kagan, Tetrahedron, 2001, 57, 2449–2468;
(c) M. Kurosu and Y. Kishi, J. Org. Chem., 1998, 63, 6100–6101;
(d) J. R. Dehil and V. Gotor, J. Org. Chem., 2002, 67, 1716–1718;
(e) K. Tanaka and G. C. Fu, J. Am. Chem. Soc., 2003, 125, 8078–8079.
6 (a) B. M. Wang, L. Song, C. A. Fan, Y. Q. Tu and W. M. Chen, Synlett,
2003, 1497–1499; (b) Z.-M. Chen, Q.-W. Zhang, Z.-H. Chen, H. Li,
Y.-Q. Tu, F.-M. Zhang and J.-M. Tian, J. Am. Chem. Soc., 2011, 133,
8818–8821; (c) H. Li, F.-M. Zhang, Y.-Q. Tu, Q.-W. Zhang,
Z.-M. Chen, Z.-H. Chen and J. Li, Chem. Sci., 2011, 2, 1839–1841;
(d) M. Wang, B. M. Wang, L. Shi, Y. Q. Tu, C.-A. Fan, S. H. Wang,
X. D. Hu and S. Y. Zhang, Chem. Commun., 2005, 5580–5582.
7 For recent reviews on quantitative asymmetric transformations of
racemates, see: (a) K. Faber, Chem. – Eur. J., 2001, 7, 5005–5010;
(b) J. Steinreiber, K. Faber and H. Griengl, Chem. – Eur. J., 2008, 14,
8060–8072; (c) H. B. Kagan, Tetrahedron, 2001, 57, 2449–2468.
8 V. Rauniyar, A. D. Lackner, G. L. Hamilton and F. D. Toste, Science,
2011, 334, 1681–1684.
Notes and references
1 J. C. Fiaud and H. B. Kagan, in Top. Stereochem., ed. E. L. Eliel and
S. H. Wilen, vol. 18, John Wiley and Sons, Inc., New York, 1988,
pp. 249–340.
2 For a selection of efficient KR processes, see: (a) R. P. Wurz,
E. C. Lee, J. C. Ruble and G. C. Fu, Adv. Synth. Catal., 2007, 349,
2345–2352; (b) V. Martin, S. Woodard, T. Katsuki, Y. Yamada,
M. Ikeda and K. B. Sharpless, J. Am. Chem. Soc., 1981, 103,
6237–6240; (c) M. S. VanNieuwenhze and K. B. Sharpless, J. Am.
Chem. Soc., 1993, 115, 7864–7865; (d) J. F. Larrow, S. E. Schaus and
9 For the first report on the employment of the TRIP phosphoric acid,
see: M. Klussmann, L. Ratjen, S. Hoffmann, V. Wakchaure,
R. Goddard and B. List, Synlett, 2010, 2189–2192.
S
S
E. N. Jacobsen, J. Am. Chem. Soc., 1996, 118, 7420–7421; 10 The stereochemistries of B2R, B3S, B4 and C2 were confirmed by
(e) S. Hashiguchi, A. Fujii, K.-J. Haack, K. Matsumura, T. Ikariya
and R. Noyori, Angew. Chem., Int. Ed. Engl., 1997, 37, 288–290;
X-ray diffraction analysis. CCDC 1015663–1015666 contains all of
the crystallographic data for this communication.
( f ) M. Kitamura, I. Kasahara, K. Manabe, R. Noyori and 11 See the ESI† section for details.
13464 | Chem. Commun., 2014, 50, 13461--13464
This journal is ©The Royal Society of Chemistry 2014