Deracemization of axially chiral α,β-unsaturated aldehydes through an amino-catalyzed symmetry-making-symmetry-breaking cascade

Article abstract of DOI:10.1002/chem.201003477

Highly enantioenriched products have been prepared in good yields from racemic enal starting materials through a new deracemization strategy that employs an enantioconvergent amino-catalyzed symmetry-making-symmetry-breaking cascade. The reaction design features the tandem domination of product development and catalyst control in two reaction steps (see scheme).

Full text of DOI:10.1002/chem.201003477

DOI: 10.1002/chem.201003477
Deracemization of Axially Chiral a,b-Unsaturated Aldehydes through an
Amino-Catalyzed Symmetry-Making–Symmetry-Breaking Cascade
Hao Jiang, Kim Søholm Halskov, Tore Kiilerich Johansen, and Karl Anker Jørgensen*[a]
Deracemization and desymmetrization processes repre-
sent elegant and atom-efficient routes in asymmetric organic
synthesis.[1] Whereas desymmetrization reactions have been
extensively studied, there exist only a handful of distinct
strategies that allow complete deracemization to occur with
full conversion of starting substrate, hence, reaching a theo-
retical yield of 100% as opposed to 50% in the traditional
kinetic resolution processes. In this respect, great advances
have been achieved for transition-metal-catalyzed or com-
bined metal/enzyme promoted dynamic kinetic resolution
(DKR)[2] and dynamic kinetic asymmetric transformation
(DYKAT) reactions.[3] On the contrary, deracemization
methods based on metal-free promoters are scarce and
often limited to isolated examples.[4] The obvious challenge
lies in the identification of suitable substrates that permit
either a rapid interconversion of the enantiomers or a
simple way to introduce symmetry in the molecule. Herein,
we demonstrate a novel DYKAT reaction, in which racemic
substituted 2-cyclohexylidene acetaldehydes carrying axial
chirality can be efficiently deracemized by a symmetry-
making–symmetry-breaking cascade that utilizes amino-cat-
alyzed functionalization strategies.[5] Moreover, a mechanis-
tic rationale with experimental support for tandem product-
development and catalyst control of selectivity is provided.
The design of the presented deracemization concept finds
its inspiration in the 1,2-addition of nucleophiles to confor-
mationally locked cyclohexanones (Scheme 1). In such addi-
tion reactions, either equatorial or axial attack may occur
leading to the formation of the axial- and equatorial alco-
hols, respectively. Experimentally, it has been shown that
nucleophiles, in which sterical hindrance is negligible
(NucS), prefer to add to the carbonyl from an axial trajecto-
ry, while larger and sterically more demanding nucleophiles
(NucL) preferentially add from the equatorial side (product-
development control).[6] We questioned whether the sterical-
ly controlled diastereofacial selectivity could be extended to
other “flattened” cyclic structures such as racemic enals rac-
1 (Scheme 1). Provided the necessary activation by an
Scheme 1. Design of the symmetry-making–symmetry-breaking cascade.
Elec=electrophile, NucL =large nucleophile, NucS =small nucleophile.
amino catalyst 2, a sterically demanding nucleophile should
attack the electrophilic double bond preferably from the
equatorial side, hence exhibiting product-development con-
trol.[7] The result is that symmetry is introduced in the mole-
cule with the formation of a diastereomeric pair of meso-3,
in which the axial aldehyde is favored. In the presence of an
electrophilic species, an enantioselective a-functionalization
of meso-3 is promoted by the optically active catalyst 2, thus
furnishing the desired enantioenriched products 4, which
contain three stereogenic centers. To have an efficient enan-
tioconvergent process, product-development control must
dominate in the symmetry-making step, while catalyst shall
control the symmetry-breaking step. Notably, if full catalyst
control applies in the nucleophilic addition step, a 1:1 mix-
ture of 3 will be formed, thus minimizing the enantioconver-
gency of the designed strategy. Similarly, the ring-substitu-
ents may influence the enantiocontrol of the catalyst by
means of mismatching sterical interactions in the subsequent
step. Ideally, a tandem control provided first by the sub-
[a] H. Jiang, K. S. Halskov, T. K. Johansen, Prof. Dr. K. A. Jørgensen
Center for Catalysis, Department of Chemistry
Aarhus University, 8000 Aarhus C (Denmark)
Fax : (+45)8919-6199
E-mail: kaj@chem.au.dk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201003477.
3842
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3842 – 3846

COMMUNICATION
strate and later by the catalyst, with minimum mutual inter-
ference, should give the optimum DYKAT process.
As a proof of concept, a nucleophile carrying an internal
electrophilic site was selected to evaluate our designed
DYKAT reaction. As such, the bulky N-centered aziridina-
tion reagent BocNHOTs suited this purpose (Table 1).[8]
relation suggests that in addition to the bulkiness of the nu-
cleophile, the conformational restrain of the cyclic enal is
also crucial for the diastereoselectivity. Therefore, the ability
of rac-1 to perform ringflip should directly correlate to the
selectivity in the symmetry-making step (Scheme 2). The ob-
served increase in the d.r. going from R=Me to tBu is thus
a direct reflection on the substrateꢁs ability to lock the con-
formation of the ring, as the equatorial attack to the ring-
flipped compound would lead to the build-up of the unde-
sired isomer of 3. Additionally, it is also well-known that O-
substituents introduce more flexibility in the cyclic structure,
which transmits to the observed tendency of a drop in the
d.r. value.[9] Moreover, to the best of our knowledge this
represents the first general study of asymmetric amino-cata-
lyzed aziridination reaction of b,b-disubstituted enals.[8d] The
relative configuration of 4e (major) was established by X-
ray analysis and confirmed the proposed structure, whereas
the absolute configuration has been assigned by correlation
and an independent synthesis (see the Supporting Informa-
tion). The catalyst 2a has proven to be highly effective in
steric face-shielding because it blocked the Re-face of the
enamine intermediate while favoring the Si-face electrophil-
ic attack in many a-functionalization reactions, including
aziridination reactions.[5,8,10] The absolute configuration for 4
is determined to be (2S,3R,6S), as illustrated in Table 1 (see
the Supporting Information).
Table 1. Deracemization by an enantioconvergent amino-catalyzed aziri-
dination reaction.[a]
1 (R)
Yield [%][b]
d.r.[c]
ee [%][d]
1
2
3
4
5
6
1a (Me)
1b (Et)
1c (Pr)
1d (Ph)
1e (tBu)
1 f (OTBDMS)
4a: 85
4b: 93
4c: 93
4d: 95
4e: 99
4 f: 99
3.9:1
3.7:1
4:1
4.6:1
9:1
96 (97)[e]
95
97
95 (95)[e]
95
1.5:1
89 (88)[e]
[a] Reactions performed with 1 (0.12 mmol), BocNHOTs (0.10 mmol),
NaOAc (0.3 mmol), 2a (0.0025 mmol) in CH2Cl2 (0.5 mL) at RT for 20 h.
[b] Yield of the isolated diastereomeric mixture. [c] Determined by
1H NMR spectroscopic analysis. [d] Determined by CSP-HPLC after de-
rivatizations. [e] Value in parenthesis is of the minor diastereomer. Boc=
tert-butoxycarbonyl, TMS=trimethylsilyl, Ts=4-toluenesulfonyl.
To include other amino-catalyzed functionalization reac-
tions and to further study the role of the nucleophile and of
the catalyst, we attempted the amino-catalyzed epoxida-
tion[10] in the designed system (Table 2).
Gratifyingly, by performing
the aziridination reaction of
rac-1a containing a methyl sub-
stituent on the ring with just
2.5 mol% (S)-2-[bis(bis-trifluor-
omethylphenyl)trimethylsilylox-
ymethyl]pyrroli-dine 2a as cata-
lyst, the desired product 4a is
formed in 85% yield and
96% ee. More importantly, the
diastereomeric ratio of the
Scheme 2. Effect of ringflip and of the R group on the diastereoselectivity. TBS= tert-butyldimethylsilyl.
product reached 3.9:1, thus sug-
gesting that there is indeed sub-
strate control in the symmetry-
making 1,4 addition step (Table 1, entry 1). Using other
weak conformation-locking groups at the ring, such as ethyl-
or propyl groups, similar results regarding yield, diastereo-
and enantioselectivity were obtained (Table 1, entries 2 and
3). Larger groups like Ph- or tBu proved better in control-
ling the diastereoselectivity and provided values of the dia-
stereomeric ratio (d.r.) up to 9:1, whereas the excellent
yields and enantioselectivities were maintained (Table 1, en-
tries 4 and 5). Interestingly, for enals with a heteroatom sub-
stituent on the ring (-OTBDMS, 1 f) a poor 1.5:1 diastereo-
selectivity was obtained; however, the yield and enantiose-
lectivity remained high (Table 1, entry 6). This observed cor-
Treating the conformationally locked enal 1e with aque-
ous H2O2 and catalyst 2a, the desired epoxide product 5e
was formed in 62% yield, 4.6:1 d.r., and 92% ee (Table 2,
entry 1). The lower d.r. obtained compared with the aziridi-
nation reaction is presumably due to the reduced size of the
nucleophile, whereas the volatile nature of the resulting ep-
oxides also diminishes the isolated yields. Other sterically
more demanding oxidants result in a noticeable increase in
d.r., thus furnishing 5 as a single diastereomer, albeit with
lower yield or enantiomeric excess (Table 2, entries 2–4).
The influence of water on the diastereoselectivity was ruled
out by a control experiment (Table 2, entry 5), whereas less
Chem. Eur. J. 2011, 17, 3842 – 3846
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3843

K.-A. Jørgensen et al.
Table 2. Influence of the nucleophile and catalyst on the amino-catalyzed
enantioconvegent epoxidation reaction.[a]
ity of the epoxidation reaction in a positive manner. Inter-
estingly, by employing a protic solvent such as EtOH, prod-
uct 5 was furnished with improved selectivities (4:1 d.r. and
89% ee), albeit only traces of product could be isolated
(Table 2, entry 11), presumably because of the competitive
oxa-Michael reaction of the alcoholic solvent.
The observed correlation between diastereoselectivity and
the size of catalyst and nucleophile is presumably a result of
“product-development control” (Scheme 3).[7] The relative
stability of the formed products favors the positioning of the
larger group (L) equatorially, whereas the smaller substitu-
ent (S) is placed axially. By increasing the size of the cata-
lyst or decreasing the bulkiness of the nucleophile, the rela-
tive stability of the major isomer of meso-3 is reduced, thus
disfavoring its formation and diminishing the facial selectivi-
ty of the 1,4-addition reaction. Consequently, to obtain opti-
mal diastereoselectivity, a large nucleophile in combination
with a small catalyst should be used, and vice versa.
Having studied the influence of the catalyst, solvent, and
nucleophile on amino-catalyzed epoxidation, the scope of
the deracemization reaction is presented (Table 3). Employ-
ing H2O2 as oxidant, the desired epoxides 5 were formed in
80–92% ee, albeit in only moderate d.r. (up to 4.6:1 d.r.,
Table 3, entries 1–5). Superior diastereoselectivities could
also be achieved, on the expense of the enantioselectivity,
by the application of a more bulky peroxide source that led
to a marked improvement of the d.r. for all substrates (up to
20:1; Table 3, entries 6–10). It is noteworthy that in both
methods, the same correlation between the size of the R
group and diastereoselectivity (as that of the aziridination
process) is observed, thus confirming the previously given
rationale on the influence of ring-flexibility.
R
Cat. Peroxide
Conv. [%][b]
d.r.[c] ee [%][d]
1
2
3
4
1e (tBu) 2a
1e (tBu) 2a
1e (tBu) 2a
1e (tBu) 2a
H2O2 (33%)
UHP[e]
tBuO2H
cumene
>95 (62)
>95 (15)
>95 (38)
>95 (traces)
4.6:1 92 (86)
20:1 96
20:1 70
20:1 74 (60)
peroxide
5
1e (tBu) 2a
tBuO2H+H2O >95
(2 equiv)
20:1 72
AHCTUNGTRENNUNG
6
7
8
9
10
1b (Et)
1b (Et)
1c (Pr)
1c (Pr)
1c (Pr)
2a
2b
2a
2c
2d
2a
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
>95 (51)
>95
>95
<25
>95
3:1 88 (91)
20:1
–
3:1 87 (89)
1.5:1 84 (84)
10:1 63
11[f] 1c (Pr)
>95 (traces)
4:1 89
[a] Reactions were performed with peroxide (0.13 mmol), 1 (0.1 mmol), 2
(0.01 mmol) in 0.2 mL CH2Cl2. [b] Determined by H NMR spectroscopic
1
analysis. Yield of isolated product in parentheses. [c] Determined by
1H NMR spectroscopic analysis or GC of the crude mixture. [d] Deter-
mined by GC or HPLC analysis with a chiral stationary phase. Results in
parentheses are for the minor diastereomer. [e] Urea hydrogen peroxide.
[f] EtOH (96%) was used as solvent. TES=triethylsilyl.
structurally rigid enals, such as 1b,c resulted, as expected, in
inferior diastereoselectivities (Table 2, entries 6 and 8). The
use of the achiral pyrrolidine as
A summary of the reaction concept is outlined in
Scheme 4. Chiral racemic enals rac-1 are introduced into the
catalyst in combination with
H2O2 resulted in improved dia-
stereoselectivity compared with
the application of 2a and H2O2
(Table 2, entry 7), thus confirm-
ing an interference by the chiral
catalyst in the symmetry-
making step. Increasing the
steric bulk of the catalyst 2 by
employing the O-TES prolinol
Scheme 3. Product-development controlled facial selectivity influenced by the relative size of catalyst and nu-
analogue 2c led to a further cleophile.
drop in the d.r., whereas the
less sterically demanding cata-
lyst 2d had the opposite effect
(Table 2, entries 9 and 10).
However, none of the other
chiral amines (2c,d) could pro-
vide the satisfactory enantiose-
lectivity with H2O2 as the per-
oxide source. Evaluation of
other polar and nonpolar sol-
vents did not affect the selectiv- Scheme 4. Reaction concept.
3844
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3842 – 3846

Deracemization of Axially Chiral a,b-Unsaturated Aldehydes
COMMUNICATION
Table 3. Deracemization by an enantioconvergent amino-catalyzed epox-
idation reaction.[a]
Acknowledgements
This work was made possible by grants from Carlsberg Foundation and
the OChemSchool. Thanks are expressed to Drs. Steen Uttrup Petersen
and Frank Jensen for the assignment of the configurations.
Yield [%][b]
d.r.[c]
3:1
3:1
2.5:1
4.6:1
1.5:1
6:1
ee [%][d]
Keywords: deracemization · dynamic kinetic asymmetric
transformation · enantioconvergent synthesis · nucleophilic
addition · organocatalysis
1 (R)
Peroxide
1
2
3
4
5
6
7
8
9
1a (Me)
1b (Et)
1d (Ph)
1e (tBu)
1 f (OTBDMS)
1b (Et)
1c (Pr)
1d (Ph)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
H2O2 (33%)
tBuOOH
tBuOOH
tBuOOH
tBuOOH
tBuOOH
5a: 84
5b: 51
5d: 61
5e: 62
5 f: 88
5b: 78
5c: 46
5d: 74
5e: 38
5 f: 58
91
88 (91)
87 (91)
92 (86)
80 (94)
82 (76)
86 (73)
70
[1] For recent reviews, see: a) C. C. Gruber, I. Lavandera, K. Faber, W.
Kroutil, Adv. Synth. Catal. 2006, 348, 1789; b) E. Garcꢂa-Urdiales, I.
Alfonso, V. Gotor, Chem. Rev. 2005, 105, 313; c) M. C. Willis, J.
Chem. Soc. Perkin Trans. 1 1999, 1765.
5.8:1
11.5:1
20:1
1e (tBu)
1 f (OTBDMS)
70
83 (91)
[2] For recent reviews and seminal work, see: a) H. Pellissier, Tetrahe-
dron 2003, 59, 8291; b) F. F. Huerta, A. B. E. Minidis, J.-E. Bꢃckvall,
Chem. Soc. Rev. 2001, 30, 321; c) M. T. Reetz, K. Schimossek,
Chimia 1996, 50, 668; for reviews and seminal work on spontaneous
resolution via chiral symmetry-breaking in crystallizations, see: d) J.
Crusats, S. Veintemillas-Verdaguer, J. M. Ribo, Chem. Eur. J. 2006,
12, 7776; e) D. G. Blackmond, Chem. Eur. J. 2007, 13, 3290, and the
references therein; f) W. L. Noorduin, E. Vlieg, R. M. Kellogg, B.
Kaptein, Angew. Chem. 2009, 121, 9778; Angew. Chem. Int. Ed.
2009, 48, 9600; g) C. Viedma, Astrobiology 2007, 7, 312.
10
1.7:1
[a] Reactions performed with 1 (0.1 mmol), peroxide (0.13 mmol), 2a
(0.01 mmol) in CH2Cl2 (0.2 mL) at RT for 20 h. [b] Yield of the isolated
diastereomeric mixture. [c] Determined by 1H NMR spectroscopic analy-
sis. [d] Determined by CSP-GC or CSP-HPLC after derivatizations.
TBDMS=tert-butyldimethylsilyl.
catalytic cycle by condensation with the amino catalyst 2.
The resulting iminium species is, preferentially, subjected to
equatorial attack by the heteroatom-centered nucleophile
(HXLg) because of the higher relative stability of the
formed product. Importantly, the normal facial-shielding
control of the amino catalyst is, in the conjugate addition
step, overruled by product-development control. Upon the
nucleophilic attack, symmetry is introduced in the chiral
molecule (rac-1), thus generating the meso-compound 3.
Final cyclization from the Si-face of the catalyst-bound en-
amine species, now following catalyst-control, completes the
catalytic cycle with the formation of optically active epox-
ides 5 and aziridines 4 in concert with the liberation of the
amino catalyst.
In conclusion, we have presented a novel deracemization
reaction of axially chiral a,b-unsaturated aldehydes that em-
ploys an enantioconvergent amino-catalyzed symmetry-
making–symmetry-breaking cascade, in which product-de-
velopment and catalyst control dominate in tandem. It is
further confirmed that the size of the nucleophile and the
conformational rigidity of the substrate can overrule the
facial differentiation of the amino catalyst in conjugate addi-
tions (the symmetry-making step). Highly enantioenriched
products were formed from chiral racemic enals, with good
to excellent yields and moderate to high diastereoselectivi-
ties using as low as 2.5 mol% catalyst. The majority of cur-
rent deracemization approaches are based on substrates in-
corporating a labile stereogenic center that equilibrates in
solution; on the contrary, reports based on symmetry-intro-
duction are very elusive. To the best of our knowledge, this
represents one of the first secondary amine catalyzed
DYKAT based on symmetry-introduction.
[3] a) R. S. Ward, Tetrahedron: Asymmetry 1995, 6, 1475; b) M.-J. Kim,
J. Ahn, J. Park, Curr. Opin. Biotechnol. 2002, 13, 578; c) B. M. Trost,
M. R. Machacek, A. Aponick, Acc. Chem. Res. 2006, 39, 747; d) O.
Pꢄmies, J.-E. Bꢃckvall, Chem. Rev. 2003, 103, 3247.
[4] For recent examples based on phosphine catalysis, see: a) S. W.
Smith, G. C. Fu, J. Am. Chem. Soc. 2009, 131, 14231; b) R. Sinini, J.
Sun, G. C. Fu, Proc. Nat. Acad. Sci. USA 2010, 107, 20652; for ex-
amples based on hydrogen-bonding catalysis, see: c) A. Berkessel, S.
Mukherjee, F. Cleemann, T. N. Muller, J. Lex, Chem. Commun.
2005, 1898; d) A.-N. R. Alba, X. Companyꢅ, G. Valero, A. Moyano,
R. Rios, Chem. Eur. J. 2010, 16, 5354; e) T. Bui, S. Syed, C. F. Barba-
s III, J. Am. Chem. Soc. 2009, 131, 8758; for recent examples based
on amino-catalysts, see: f) J. Peng, X. Huang, H.-L. Cui, Y.-C. Chen,
Org. Lett. 2010, 12, 4260; g) P. Galzerano, G. Bencivenni, F. Pes-
ciaioli, A. Mazzanti, B. Giannichi, L. Sambri, G. Bartoli, P. Mel-
chiorre, Chem. Eur. J. 2009, 15, 7846.
[5] For recent reviews, see: a) P. Melchiorre, M. Marigo, A. Carlone, G.
Bartoli, Angew. Chem. 2008, 120, 6232; Angew. Chem. Int. Ed. 2008,
47, 6138; b) A. Mielgo, C. Palomo, Chem. Asian J. 2008, 3, 922; c) S.
Bertelsen, K. A. Jørgensen, Chem. Soc. Rev. 2009, 38, 2178; d) E.
Marquꢆs-Lꢅpez, R. P. Herrera, M. Christmann, Nat. Prod. Rep.
2010, 27, 1138; e) C. Grondal, M. Jeanty, D. Enders, Nat. Chem.
2010, 2, 167; f) A. Dondoni, A. Massi, Angew. Chem. 2008, 120,
4716; Angew. Chem. Int. Ed. 2008, 47, 4638; g) B. List, Chem.
Commun. 2006, 819.
[6] a) D. H. R. Barton, J. Chem. Soc. 1953, 1027; b) B. W. Gung, Chem.
Rev. 1999, 99, 1377; and the references therein.
[7] W. C. Dauben, G. J. Fonken, D. S. Noyce, J. Am. Chem. Soc. 1956,
78, 2579.
[8] For the first amino-catalyzed enal aziridination reaction, see: a) J.
Vesely, I. Ibrahem, G.-L. Zhao, R. Rios, A. Cꢅrdova, Angew. Chem.
2007, 119, 792; Angew. Chem. Int. Ed. 2007, 46, 778; for application
of TsONHBoc, see: b) H. Arai, N. Sugaya, N. Sasaki, K. Makino, S.
Lectard, Y. Hamada, Tetrahedron Lett. 2009, 50, 3329; for improved
enal aziridination protocols, see: c) Ł. Albrecht, H. Jiang, G. Dick-
meiss, B. Gschwend, S. G. Hansen and K. A. Jørgensen, J. Am.
Chem. Soc. 2010, 132, 9188. One entry regarding amino-catalyzed
aziridination of b,b-disubstituted enals was reported in: d) H. Jiang,
B. Gschwend, Ł. Albrecht, K. A. Jørgensen, Org. Lett. 2010, 12,
5052.
[9] J. Clayden, N. Greeves, S. Warren, P. Wothers in Organic Chemistry,
Oxford University Press, USA, 2001, pp. 462–474.
Chem. Eur. J. 2011, 17, 3842 – 3846
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3845

K.-A. Jørgensen et al.
[10] For the first report on amino-catalyzed enal epoxidation reactions,
see: a) M. Marigo, J. Franzꢆn, T. B. Poulsen, W. Zhuang, K. A. Jør-
gensen, J. Am. Chem. Soc. 2005, 127, 6964; for other reports, see:
b) H. Sundꢆn, I. Ibrahem, A. Cꢅrdova, Tetrahedron Lett. 2006, 47,
99; c) S. Lee, D. W. C. MacMillan, Tetrahedron 2006, 62, 11413; for
an improved protocol see Ref. [8c].
Received: December 1, 2010
Revised: January 28, 2011
Published online: March 4, 2011
3846
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 3842 – 3846