A Catalyst-Directed Remote Stereogenic Center Switch During the Site-Selective Aldol Desymmetrization of Cyclohexanone-Based Diketones

Article abstract of DOI:10.1002/adsc.201600833

Site-selectivity, differentiating members of the same functional group type on one substrate, is an emerging tactic for shortened advanced building block and biomolecule synthesis. Despite its potential, site-selectivity remains less studied and especially so for ketone-based substrates. During this work ketone site-selectivity has been coupled with the chiral amine-catalyzed aldol desymmetrization of 4-keto-substituted cyclohexanones, allowing three stereogenic centers to form in the aldol product while leaving the acyclic ketone unreacted. Unique here, compared to all previous 4-substituted cyclohexanone desymmetrizations, is the first access to synthetically useful quantities of an epimeric (remote stereogenic center) aldol product. To demonstrate the value of these aldol products, we show their elaboration into eight keto-acetonide and one keto-lactone products. All compounds were isolated as single diastereomers and in high ee (≥96%). These efforts represent the first full characterization of aldol products with type III, Figure 2, relative stereochemistry, regardless of the enantiomer formed. (Figure presented.).

Full text of DOI:10.1002/adsc.201600833

COMMUNICATIONS
DOI: 10.1002/adsc.201600833
A Catalyst-Directed Remote Stereogenic Center Switch During
the Site-Selective Aldol Desymmetrization of Cyclohexanone-
Based Diketones
Thomas C. Nugent,a,* Peter Spiteller,b Ishtiaq Hussain,a
Hussein Ali El Damrany Hussein,a and Foad Tehrani Najafiana
a
Department of Life Sciences and Chemistry, Jacobs University Bremen, Campus Ring 1, 28759 Bremen, Germany
E-mail: t.nugent@jacobs-university.de
Institut fꢀr Organische und Analytische Chemie, Universitꢁt Bremen, Leobener Straße NW2C, 28359 Bremen, Germany
b
Received: July 28, 2016; Revised: September 12, 2016; Published online: && &&, 0000
Supporting information for this article can be found under: http://dx.doi.org/10.1002/adsc.201600833.
Abstract: Site-selectivity, differentiating members
of the same functional group type on one substrate,
is an emerging tactic for shortened advanced build-
ing block and biomolecule synthesis. Despite its po-
tential, site-selectivity remains less studied and es-
pecially so for ketone-based substrates. During this
work ketone site-selectivity has been coupled with
the chiral amine-catalyzed aldol desymmetrization
of 4-keto-substituted cyclohexanones, allowing
three stereogenic centers to form in the aldol prod-
uct while leaving the acyclic ketone unreacted.
Unique here, compared to all previous 4-substituted
cyclohexanone desymmetrizations, is the first access
to synthetically useful quantities of an epimeric
(remote stereogenic center) aldol product. To dem-
onstrate the value of these aldol products, we show
their elaboration into eight keto-acetonide and one
keto-lactone products. All compounds were isolated
as single diastereomers and in high ee (ꢀ96%).
These efforts represent the first full characterization
of aldol products with type III, Figure 2, relative
stereochemistry, regardless of the enantiomer
formed.
(Figure 1).[3a,b,d] Among diketone substrates, 1,2- and
1,3-diketones possess electronically interconnected
carbonyl moieties making them susceptible to intra-
Keywords: aldols; desymmetrization; diketones;
epimer switch; organocatalysis; site-selectivity
Ketone transformations are common, reliable, and
represented by a large variety of reactions.[1] Never-
theless, the use of prochiral diketones for enantiose-
lective synthesis is limited due to the difficulty of dif-
ferentiating: (i) non-equivalent ketone carbonyl moi-
eties, a topic of site-selectivity,[2,3a,c,e,4] or (ii) equivalent
Figure 1. Prior examples of highly enantioselective diketone
ketone carbonyl units, a topic of desymmetrization differentiation.[3]
Adv. Synth. Catal. 0000, 000, 0 – 0
1
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
molecular hydrogen bonding (enol-ketone), metal hydroxyproline (3) catalysis (Figure 2, left panel),[3e]
chelation control opportunities, and so forth. By con- was unique because it was the first to show that dike-
trast, 1,4- and higher diketones would generally lack tones, based on 4-substituted cyclohexanones, can be
these attributes, represent good prototypes for exam- site-selectively desymmetrized, but the product ste-
ining ketone carbonyl differentiation, and are the reochemistry followed the same well-defined relative
focus of this manuscript. From these diketones, we and absolute stereochemistry as noted for all previous
are aware of only five studies[3,4] that demonstrate aldol desymmetrizations of 4-substituted cyclohexa-
highly enantioselective reactions while concomitantly nones.[9] Common to all prior studies, two dominant
illustrating site-selectivity or desymmetrization stereochemical outcomes were always noted as the
(Figure 1).[5,6]
relative stereochemistries of type I (major, often
The examples in Figure 1 show selective methyl greater than 80% yield) and II (minor) aldol prod-
ketone reactivity in the presence of internal acyclic ucts,[13] see Figure 2 (middle panel). A small subset of
ketones, and the enamine-based catalysis achieve- the fifty eight publications describe and quantify
ments therein may reflect Barbasꢃ 2001 observation other stereochemical outcomes as minor products, see
that 2-butanone has good reactivity under (S)-proline aldol products III and IV of Figure 2.[10b,12b,14] From
catalysis while an internal ketone, e.g., 3-pentanone, those studies, two reveal >5% yield for aldol prod-
displayed a complete lack of reactivity.[7] In an effort ucts of type III,[14b,c] and the study by Plusquellec is
to expand beyond methyl ketone site-selectivity, we the only one to indicate the formation of a type IV
recently demonstrated that cyclohexanone carbonyls aldol product (8% yield).[14c] Specifically, in 2011 Cꢄr-
can be reacted over methyl ketone carbonyl moieties dova noted one example in which a 3:1 ratio of aldol
in 4-keto-substituted cyclohexanones (Figure 1, I (68% yield) to aldol III (23%) products were found
Nugent, R=Me, n=2).[3e] In doing so, we demonstrat- when examining the desymmetrization of 4-methylcy-
ed the first examples in which a non-hindered methyl clohexanone (10 equiv.) with 4-nitrobenzaldehyde
ketone remains unreacted. Here we detail an exten- under 10 mol% TBSO-threonine catalysis (Figure 2,
sion of that study in which an alternative amine cata- catalyst 4).[14b] One year later, Plusquellec noted that
lyst, picolylamine (PicAm) 1 (Figure 2, left panel),[8] is 4-methylcyclohexanone (5 equiv.) could be desymme-
used to form the first realistic quantities of a non-ac- trized with 3-nitrobenzaldehyde under 2 mol% (R)-3-
cessible epimeric aldol product (Figure 2, middle pyrrolidinol (not shown) catalysis in a 1.0M aqueous
panel, type III).
solution of a sugar derivative. During the Plusquellec
All enantioselective transformations of mono-4- study, the greatest aldol type III yield was 22%, while
substituted cyclohexanones must concomitantly entail the type IV aldol product was noted, for the first
a desymmetrization (Figure 2, middle panel), and in time, in 8% yield. Unfortunately, both of those prod-
2007 Gong disclosed a highly selective aldol variant.[9] ucts were racemic.[15] Perhaps unsurprisingly, the low
It used an efficient prolinamide catalyst 2 (Figure 2), yields of these compounds precluded their isolation in
and fifty-eight organocatalyst-based publications have pure form by Cꢄrdova or Plusquellec. As such, no
followed in which the products, typically from 4- aldol products of type III or IV, or analogs thereof,
methyl,[10] 4-tert-butyl,[10b,d,11] or 4-phenylcyclohexano- have ever been fully characterized.
ne[9,10b,11b,12] starting materials, have been benchmarked
In summary, regardless of the R substituent on
against each other under the use of alternative cata- a mono-4-substituted cyclohexanone (Figure 2, middle
lysts and reaction conditions. As such, our recent in- panel), their desymmetrization only infrequently pro-
vestigation of 4-keto-substituted cyclohexanone sub- vides aldol products III or IV, and then only in minor
strates 5–8 (Figure 2, right panel), under TBDPSO-4- quantities. Furthermore, only when 4-methylcyclohex-
Figure 2. Left to right panels: catalysts (1–4), generic relative stereochemical outcomes for 4-substituted cyclohexanone de-
symmetrization (types I–IV), and diketones examined during this study (5–8).
Adv. Synth. Catal. 0000, 000, 0 – 0
2
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
anone was used, were yields of up to 23% for aldol tonide building blocks as single diastereomers and in
product III observed. There are no current examples high ee (Table 1).[20]
with larger substituents, located at the 4-position, that
To gain access to the keto-acetonide products,[21] we
provide >5% yield of type III relative stereochemis- took advantage of the well-known fact that b-hydroxy
try products. In particular, we stress here that gaining ketones are selectively reduced over simple ketones
access to these remote stereogeneric center epimeric when employing NaB(OAc)3H.[22] The stereochemical
products (III or IV) would be laboriously inefficient outcome of these type of reductions has been dis-
via (i) any other synthetic approach; (ii) post modifi- cussed elsewhere,[22] but it should be noted that car-
cation of aldol products I or II; or (iii) the application bonyl hydride delivery occurred, predominately, from
of an enantiomeric catalyst coupled with post modifi- the opposite face of the cyclohexanone ring for 9
cation.[16] With this communication, we change that versus 10. The resultant keto-1,3-diols (not shown),
dynamic by showing that useful, albeit modest, yields were chromatographically unobtainable as pure dia-
of pure type III relative stereochemistry compounds stereomers using EtOAc/petrol ether eluent systems,
can now be accessed.
nonetheless, collection of the major diastereomer in
With that perspective, we begin our discussion by mediocre diastereomeric and chemical purity, 85–90%
noting that picolyl amine (PicAm) catalyst 1 provides after chromatography, did allow their conversion to
nearly equal quantities of two major aldol products, 9 keto-acetonides when treated with 2,2-dimethoxypro-
and 10, in combined yields of approximately 90% pane (20–30 equiv.) under the mildly acidic conditions
(1H NMR analysis) during the enantioselective de- of catalytic pyridinium p-toluenesulfonate (5.0 to
symmetrization of diketones 5–8 (Scheme 1).[17] Aldol 7.5 mol%) in CH2Cl2.[23] The keto-acetonides attracted
products 10a–h contain the relative stereochemistry our attention because these structures always permit-
as found in all previous studies, that is, type I, and ted chromatographic isolation of a single diastereo-
have been previously synthesized.[3e] They are conse- mer that could be fully characterized.
quently not a focus of this manuscript and are not fur-
As shown in Table 1, an array of 4-ketosubstituted
ther discussed. On the other hand, structures 9a–h cyclohexanone-based diketones (2.0 equiv.) reacted
possess the difficult to access type III aldol stereo- with a handful of aromatic aldehydes (1.0 equiv.) in
chemistry and were readily isolated and characterized, the presence of 10 mol% of either (R)- or (S)-PicAm
albeit as their corresponding keto-acetonide (11a–h) 1. Note that most of the keto-acetonide products
and keto-lactone (12) products and are shortly dis- were synthesized with (R)-PicAm 1, which produces
cussed (Table 1). The PicAm 1 catalyst consequently the same relative stereochemistry as aldol type III,
permits an epimer switch, albeit non-selective, at the albeit for the opposite enantiomeric form. The overall
remote stereogenic center of the aldol product yields, for the three step diketone to keto-acetonide
(Scheme 1).
transformations (Scheme 1), ranged from 25–34%
Cyclohexanone-based aldol products, like aldol 9 (Table 1). At the low end, a 25% overall yield repre-
and 10, often undergo non-selective epimerization at sents, on average, a 63% yield for each of the three
their a-carbon (Scheme 1, carbon 2) upon exposure reaction steps: aldol, reduction, and keto-acetonide
to silica gel.[18,19] Adding to this challenge, a majority formation. On the high end, a 34% overall yield rep-
of our anti and syn aldol products had very similar Rf resents a 70% yield for each reaction step. Placed in
values (TLC). As such, we found it practical to lock the context of no prior access, in pure form, to these
in the aldol stereochemical information by simply highly enantioenriched compounds with four stereo-
using the worked-up, crude, aldol products for our genic centers, the current yields may be considered, if
next reactions. In this manner, we showed that the not yet practical, then perhaps as enabling the first
aldol products can be elaborated into useful keto-ace- speculative applications for natural product or medici-
Scheme 1. Aldol products 9 and 10 are formed in near equal quantities as the major reaction products under PicAm 1 cataly-
sis. Keto-acetonides 11 were isolated and characterized, see Table 1. Note: 2,4-DNBSA=2,4-dinitrobenzenesulfonic acid.
Adv. Synth. Catal. 0000, 000, 0 – 0
3
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
Table 1. Type III keto-acetonide and lactone products from (R)- or (S)-PicAm catalysis.[a]
En- Keto-aceto-
try nides
(11)
Aldol product data[a]
Keto-acetonide product
data
Reaction Conversion
Time [h]
(1H NMR)[b]
Aldol 9/10
anti/syn (type I &
III to II & IV)[c]
Overall yield of 11
ee
(type III to
(from aldehyde)[e]
I)[c,d]
1
28
96
95
95
72
1.00:1.20
1.00:1.17
1.00:1.23
1.17:1.00
>24:1
>24:1
>24:1
8:1
28
27
32
25
97
2
40
30
48
>99
3
97
4[f]
97
5
25
40
69
94
95
94
1.00:1.30
1.00:1.32
1.13:1.00
>24:1
>24:1
>24:1
30
32
34
>99
6
96
7[g]
96
nal chemistry goals. Of equal or higher significance, strate 4-methylcyclohexanone.[24] Under unchanged
a convincing proof of concept for epimeric product reaction conditions a 1.2:1.0 ratio of 9i (shown) to 10i
1
formation has been established and opens a path, via (not shown) was noted by crude H NMR, and 9i was
catalyst refinement, for more selective epimer isolated after silica gel chromatography in 46% yield
switches.
(Scheme 2). This yield doubles the best previously re-
To establish that PicAm 1 can catalyze useful yields ported for 9i,[14b] and we have fully characterized this
of type III aldol products, beyond those studied here: compound for the first time (see Section 5, Support-
5–8, we additionally investigated the benchmark sub- ing Information). This result firmly establishes that
Adv. Synth. Catal. 0000, 000, 0 – 0
4
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
Table 1. (Continued)
En- Keto-aceto-
try nides
(11)
Aldol product data[a]
Keto-acetonide product
data
Reaction Conversion
Time [h]
(1H NMR)[b]
Aldol 9/10
anti/syn (type I &
III to II & IV)[c]
Overall yield of 11
ee
(type III to
(from aldehyde)[e]
I)[c,d]
8[f]
33
94
96
1.22:1.00
1.00:1.20
9:1
30
31
98
98
9[h]
28
>24:1
[a]
The aldol reactions were performed with (R)-PicAm 1, entries 4 and 8 used (S)-PicAm 1. For reaction details, see
Scheme 1 and the Supporting Information.
[b]
Reaction aliquot (1H NMR), integration of aldehydic (limiting reagent) resonance versus the combined integration for
the benzylic resonances of all anti- and syn-aldol products.
[c]
[d]
[e]
1
See Figure 2 for relative stereochemistry. Crude H NMR data: ratio of the two major 2,3’-anti products 9 and 10.
Section 4 of the Supporting Information is dedicated to verifying the 9/10 ratios (1H NMR expansions).
The yield is the overall yield from three reaction steps: aldol, reduction, and keto-acetonide formation. Mmol of pure
keto-acetonide product 11 versus mmol of the aldol limiting reagent (aldehyde)ꢀ100%.
(S)-PicAm 1 catalyzed the aldol reaction for this keto-acetonide. Note that the (S)-PicAm 1 catalyst provides the same
enantiomer of 10, a type I aldol product, as when the (2S,4R)-TBDPSO-4-hydroxyproline catalyst (Figure 2, catalyst 3) is
used.[3e]
[f]
[g]
[h]
Ar equals p-trifluoromethylphenyl.
Lactone formation occurred after treatment of aldol product 9a with mCPBA (5.0 equiv.) in CH2Cl2, see Section 3 of the
Supporting Information.
the noted epimer switch is a general phenomenon for 11 failed to form crystalline material. Aldol product
4-substituted cyclohexanones. 9i is a solid but in our hands X-ray suitable crystals
Finally, the syntheses of the diketone starting mate- could not be formed. A stepwise approach was then
rials (5–8) are straight forward and proceed in good taken to first confirm the relative stereochemistry by
overall yields. We note that diketones 5 and 6 are syn- extensive NMR experiments (see the Supporting In-
thesized after two reaction steps from commercially formation, Sections 6 and 7) and then establishing the
available phenols.[25]
absolute stereochemistry based on circular dichroism
As discussed earlier, aldol products of type III have studies (see the Supporting Information, Section 8).
been previously reported, but we know of no evi- The short tether at the 4-position of keto-acetonide
dence that establishes their relative or absolute ste- 11e (Table 1, entry 5), in combination with the confor-
reochemistry. We now rigorously address this point. mational rigidity imposed by the interlocked cyclo-
We were thwarted from a stereochemical assign- hexane and acetonide rings of these compounds,
ment via X-ray analysis after several keto-acetonides made it a good candidate for determination of its rel-
ative stereochemistry via NMR measurements. To do
so, first the proton and the carbon chemical shifts
were assigned to the corresponding atom numbers of
11e with the aid of the correlations in the COSY and
the HSQC and by taking the measured shift values
into account (see the Supporting Information, Sec-
tions 6 and 7). The two-dimensional structure of 11e
was then confirmed by analysis of the HMBC correla-
tions. The relative stereochemistry of 11e was elabo-
rated by analysis of the JH,H coupling constants in the
Scheme 2. Benchmark substrate (4-methylcyclohexanone)
examination.
1H NMR and the COSY spectra and analysis of the
Adv. Synth. Catal. 0000, 000, 0 – 0
5
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
NOE correlations obtained from the NOESY experi- thereof was unequivocally established as having the
ment (see the Supporting Information, Section 5). relative and absolute stereochemistry of ent-11e,
The protons at carbons C-2 to C-7 of the cyclohexane albeit without the 4-substitutent and its associated
ring show coupling constants typical of a chair confor- remote stereogenic center.[19] During this current
mation (Figure 3), since the protons are either axially work, 11e was formed with (R)-PicAm, thus the C-
1 and C-2 stereogenic centers of the initial aldol prod-
ucts 9e and 10e should be anti and have the same ster-
eogenicity as depicted for 11e in Figure 3.
With the type III aldol product structure estab-
lished, the next question was why PicAm-1 catalyzes
a significant increase in their formation. Scheme 3 de-
picts the expected enamine and aldol transition states
that would provide the type I and III aldol products.
Because aldol products 9 and 10 are formed in essen-
tially equal quantities (Table 1), it seems logical that
transition states C and D must be very similar in
energy. Why this is true for PicAm-1, while proline-
based catalysts overwhelmingly favor the correspond-
Figure 3. Selected NOE correlations and numbering of 11e.
ing transition states that would mimic D over C, is
or equatorially oriented on account of their coupling clearly the focus of a future computational study.
constants. Proton H-1 couples to the axial proton H-2 In conclusion, we have expanded the functional
3
with a coupling constant of JH,H =6.2 Hz, indicating group diversity of these aldol products by demonstrat-
that these protons are gauche oriented to each other. ing that 4-substituted, keto-carbonyl containing, cyclo-
A characteristic NOE from H-1 to the methyl protons
at C-12 and the absence of an NOE between H-1 and
H-3 show that H-1 is not oriented in the same direc-
tion as H-3. Since there are key correlations between
the protons H-15/19 of the aromatic ring and H-3eq,
H-2ax and H-7eq the six-membered ring containing the
acetal moiety is most likely present in a boat like con-
formation and the aromatic ring is oriented as shown
in Figure 3. This assignment is further supported by
a key NOE correlation between the proton H-3eq and
those of the methyl group at C-13. Finally, the relative
stereochemistry at C-6 is confirmed by the coupling
pattern of H-6 which shows it to be axial.
The absolute stereochemistry was established by
two different approaches. First, we generated the the-
oretical circular dichroism spectrum for the enantio-
mer represented by keto-acetonide 11e. This was
compared to the experimentally obtained circular di-
chroism spectrum, in n-pentane, and the two were
found to be in general agreement regarding their ab-
sorption shifts, intensity, and positive or negative attri-
butes about the x-axis, which displayed characteristic
Cotton effects. This result supports the indicated ab-
solute stereochemistry as shown in Figure 3 and
Table 1 (entry 5). Section 8 of the Supporting Infor-
mation is dedicated to a description of the CD study
and the conclusions drawn here. In that section we
also show the theoretical CD spectrums of related
diastereomers so a comprehensive overview can be
made.
The second piece of supportive evidence for this
absolute stereochemistry is that (S)-PicAm 1 has been
previously used to catalyze the reaction of cyclohexa-
none and 4-nitrobenzaldehyde. The aldol product products from 4-substituted cyclohexanones.
Scheme 3. Possible transition states for type I and III aldol
Adv. Synth. Catal. 0000, 000, 0 – 0
6
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
(1H NMR), was recorded. The aldol product was used in the
next reaction step without further purification.
Note: Exposure to silica gel chromatography often results
in reduced diastereoselectivity and aldol product yield and
is discouraged.
hexanones (5–8) can be used as starting materials.
Furthermore, we have shown that a non-selective
epimer switch occurs and provides the first reasonable
access to either enantiomeric form of a new relative
stereochemistry for these aldol products. Their con-
version to the corresponding acetonide (four stereo-
genic centers) and lactone (three stereogenic centers)
products represents the first realistic starting point for
their planned use in target-based synthesis. Further-
more, no other approaches, e.g., transition metal-
mediated or enzyme-based, can currently surpass the
organocatalyzed reaction results regardless of the de-
sired stereochemical outcome (type I-IV).
Finally from a catalysis perspective, we have taken
an obscure chemical observation, low yields of type
III aldol product formation,[14b,c] to a level where it
can now be imagined that a selective epimer switch,
at the remote stereogenic center of these aldol prod-
ucts, is within reach via rational catalyst design. This
is significant because there are no known step effi-
cient replacements for the formation of these epimer-
ic products.
Acknowledgements
This work was generously supported by the Deutsche For-
schungsgemeinschaft (German Science Foundation) under
award no. NU 235/6-1, and Jacobs University Bremen. We
are very thankful for low and high resolution MS measure-
ments provided by Professor Dr. Nikolai Kuhnert. IH thanks
the Higher Education Commission and Hazara University
Mansehra, KPK Pakistan for award grant number: F.No.HU/
PQDP/2012/194.
References
[1] a) T. C. Nugent, M. El-Shazly, Adv. Synth. Catal. 2010,
352, 753–819; b) M. R. Monaco, G. Pupo, B. List, Syn-
lett 2016, 27, 1027–1040; c) F. Huang, Z. Liu, Z. Yu,
Angew. Chem. 2016, 128, 872–885; Angew. Chem. Int.
Ed. 2016, 55, 862–875.
Experimental Section
[2] For a broader discussion of site-selectivity, see: P. M.
Tadross, E. N. Jacobsen, Nat. Chem. 2012, 4, 963–965.
[3] a) J. Zhou, B. List, J. Am. Chem. Soc. 2007, 129, 7498–
7499; b) J. Zhou, V. Wakchaure, P. Kraft, B. List,
Angew. Chem. 2008, 120, 7677–7886; Angew. Chem.
Int. Ed. 2008, 47, 7656–7658; c) R. C. Simon, B. Gri-
schek, F. Zepeck, A. Steinreiber, F. Belaj, W. Kroutil,
Angew. Chem. 2012, 124, 6817–6820; Angew. Chem.
Int. Ed. 2012, 51, 6713–6716; d) M. Skoupi, C. Vaxe-
laire, C. Strohmann, M. Christmann, C. M. Niemeyer,
Chem. Eur. J. 2015, 21, 8701–8705; e) T. C. Nugent,
F. T. Najafian, H. A. E. D. Hussein, I. Hussain, Chem.
Eur. J. 2016, 22, 14342–14348.
[4] A single example of double aldol site-selectivity has
been demonstrated, see Table 2, entry 8, compound 3h
(10 mol% catalyst loading, 67% yield, 9:1 dr, 81% ee)
within: Y. Shimoda, T. Kubo, M. Sugiura, S. Kotani, M.
Nakajima, Angew. Chem. 2013, 125, 3545–3548; Angew.
Chem. Int. Ed. 2013, 52, 3461–3464.
Experimental Details
Altogether112 pages of detailed Supporting Information are
associated with this manuscript.
General Procedure for Aldol Products (9 and 10)
To a dry 2.0-mL screw-cap reaction vial were added dike-
tone (1.0 mmol, 2.0 equiv.), aldehyde (0.5 mmol, 1.0 equiv.)
and (S)- or (R)-PicAm 1 as a 2,4-dinitrobenzenesulfonic
acid 1:1 salt (MW=550.58 gmolÀ1, 0.05 mmol, 10.0 mol%,
27.5 mg). After stirring for 5 min distilled water (0.50 mL)
was added. This reaction mixture was then stirred and
heated at 458C until a starting material conversion of
ꢀ95% was noted unless otherwise stated. Reaction progress
was monitored by aliquots (1H NMR). Reaction conversion
was determined by integrating the aldehydic resonance (sin-
glet, ꢁ10.0 ppm) versus the combined integration of the
benzylic proton resonance (doublets, both found between
4.50–5.50 ppm) of the syn- and anti-aldol products. Reaction
times ranged from 25–69 h, see the individual descriptions
for the specific reaction time.
Note: Extending the reaction time often results in de-
creased diastereoselectivity through a-keto epimerization.
Reaction work-up: The reaction mixture was transferred
to a separatory funnel containing distilled water (25–35 mL)
by excessive extractive addition of CH2Cl2 (9ꢅ1.5 mL) to
the reaction vessel. After this initial extraction from water,
the water was further extracted with CH2Cl2 (6ꢅ20 mL).
Combined organic extract was dried (Na2SO4), filtered and
concentrated (rotary evaporator bath temperature should
not exceed 288C to minimize the risk of a-keto epimeriza-
tion). The crude aldol product was then exposed to high
vacuum drying and after 2–3 h the dr, anti/syn ratio
[5] Triketone site-selectivity coincidentally represents the
foundation of modern amine catalysis, namely the
Hajos–Parrish–Eder–Sauer–Wiechert
intramolecular
aldol cyclizations under proline catalysis, see: a) U.
Eder, G. Sauer, R. Wiechert, Angew. Chem. 1971, 83,
492; Angew. Chem. Int. Ed. Engl. 1971, 10, 496;
b) Z. G. Hajos, D. R. Parrish, J. Org. Chem. 1974, 39,
1615–1621.
[6] For literature overviews of triketones, see: a) S. Mu-
kherjee, J. W. Yang, S. Hoffmann, B. List, Chem. Rev.
2007, 107, 5471–5569; b) B. Bradshaw, J. Bonjoch, Syn-
lett 2012, 23, 337–356.
[7] K. Sakthivel, W. Notz, T. Bui, C. F. Barbas III, J. Am.
Chem. Soc. 2001, 123, 5260–5267.
[8] For the synthesis of the picolyl amine (PicAm) 1, see:
T. C. Nugent, A. Bibi, A. Sadiq, M. Shoaib, M. Umar,
Adv. Synth. Catal. 0000, 000, 0 – 0
7
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
asc.wiley-vch.de
F. T. Najafian, Org. Biomol. Chem. 2012, 10, 9287–
9294.
[15] See Table 3, entry 4 of ref.[14c]
[16] Related to this discussion, selective conversion of an
anti-aldol product of type I to the corresponding syn-
aldol product II, or vice versa has, to the best of our
knowledge, never been demonstrated. See ref.[13] for
further information.
[9] Gong initially reported, ref.[9a], a type III, aldol struc-
ture, but corrected it to a type I relative stereochemis-
try in a corrigendum (Additions and Corrections at
Chem. Commun.) which is ref.[9b] Later, Singh, ref.[9c]
,
in error, used Gongꢃs original type III relative stereo-
chemical assignment to assign his product: a) J. Jiang,
L. He, S. W. Luo, L.-F. Cun, L.-Z. Gong, Chem.
Commun. 2007, 736–738; b) J. Jiang, L. He, S. W. Luo,
L.-F. Cun, L.-Z. Gong, Chem. Commun. 2008, 6609–
6614; c) M. R. Vishnumaya, V. K. Singh, J. Org. Chem.
2009, 74, 4289–4297.
[17] The remaining product balance in these aldol reactions
is aldol products II, and IV, generally in a combined
yield of ꢂ5% with approximately the another ꢂ5% of
the limiting reagent remaining (the aldehyde).
[18] It is more often than not that cyclohexanone-based
aldol products undergo partial, but non-selective epi-
merization, upon exposure to silica gel. See page 4
within the Supporting Information of ref.[19]
[10] For representative examples, see: a) D. Gryko, W. J.
Saletra, Org. Biomol. Chem. 2007, 5, 2148–2153; b) X.
Companyꢄ, G. Valero, L. Crovetto, A. Moyano, R.
Rios, Chem. Eur. J. 2009, 15, 6564–6568; c) C. Wu, X.
Fu, S. Li, Eur. J. Org. Chem. 2011, 1291–1299; d) S. Li,
C. Wu, X. Fu, Q. Miao, Ind. Eng. Chem. Res. 2011, 50,
13711–13716; e) S. Li, C. Wu, X. Long, X. Fu, G. Chen,
Z. Liu, Catal. Sci. Technol. 2012, 2, 1068–1071.
[11] For representative examples, see: a) P. Dinꢆr, M.
Amedjkouh, Org. Biomol. Chem. 2006, 4, 2091–2096;
b) B. Rodrꢇguez, A. Bruckmann, C. Bolm, Chem. Eur.
J. 2007, 13, 4710–4722; c) W. Clegg, R. W. Harrington,
M. North, F. Pizzato, P. Villuendas, Tetrahedron: Asym-
metry 2010, 21, 1262–1271; d) M. North, P. Villuendas,
Org. Lett. 2010, 12, 2378–2381.
[12] For representative examples, see: a) Q. Zhang, X. Cui,
L. Zhang, S. Luo, H. Wang, Y. Wu, Angew. Chem.
2015, 127, 5299–5302; Angew. Chem. Int. Ed. 2015, 54,
5210–5213; b) F. J. N. Moles, G. Guillena, C. Nꢈjera, E.
Gꢄmez-Bengoa, Synthesis 2015, 47, 549–561.
[13] See Table 2, entry 19 of the following citation for a rare
example of high type II, product selectivity. We are un-
aware of any other examples like this, see: Y. Qian, X.
Zheng, Y. Wang, Eur. J. Org. Chem. 2010, 3672–3677.
[14] a) M. Majewski, I. Niewczas, N. Palyam, Synlett 2006,
2387–2390; b) G. Ma, A. Bartoszewicz, I. Ibrahem, A.
Cꢄrdova, Adv. Synth. Catal. 2011, 353, 3114–3122; c) A.
Bellomo, R. Daniellou, D. Plusquellec, Green Chem.
2012, 14, 281–284; d) F. J. N. Moles, A. BaÇꢄn-Cabal-
lero, G. Guillena, C. Nꢈjera, Tetrahedron: Asymmetry
2014, 25, 1323–1330.
[19] T. C. Nugent, M. N. Umar, A. Bibi, Org. Biomol.
Chem. 2010, 8, 4085–4089.
[20] Only one lactone product is noted here (Table 1,
entry 9). In fact we synthesized four more lactone prod-
ucts, but their isolation in greater than 90% chemical
purity was not possible in our hands. We abandoned
the further pursuit of these lactones because of these
problems. We speculate that the ring opened forms
may allow this problem to be circumvented, see ref.[3e]
for an example during an Alzheimer drug candidate
synthesis.
[21] Note that ortho-substituted aromatic aldehydes, for ex-
ample, o-nitrobenzaldehyde, smoothly formed the cor-
responding aldol products, but those aldol products re-
sisted reduction under our standard conditions with
NaB(OAc)3H.
[22] a) D. A. Evans, K. T. Chapman, E. M. Carreira, J. Am.
Chem. Soc. 1988, 110, 3560–3578; b) S. H. J. Thompson,
M. F. Mahon, K. C. Molloy, M. S. Hadley, T. Gallagher,
J. Chem. Soc. Perkin. Trans. 1 1995, 379–383.
[23] K. Mori, S. Maemoto, Liebigs Ann. Chem. 1987, 863–
869.
[24] Since Gongꢃs 2007 disclosure of enantioselective orga-
nocatalytic aldol desymmetrization of mono-4-substi-
tuted cyclohexanone substrates, fifty-eight manuscripts
based on the same subject have been published. From
those, forty-eight describe the aldol desymmetrization
of 4-methylcyclohexanone, making it the benchmark
substrate.
[25] See the Supporting Information of ref.[3e]
Adv. Synth. Catal. 0000, 000, 0 – 0
8
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
A Catalyst-Directed Remote Stereogenic Center Switch
During the Site-Selective Aldol Desymmetrization of
Cyclohexanone-Based Diketones
9
Adv. Synth. Catal. 2016, 358, 1 – 9
Thomas C. Nugent,* Peter Spiteller, Ishtiaq Hussain,
Hussein Ali El Damrany Hussein, Foad Tehrani Najafian
Adv. Synth. Catal. 0000, 000, 0 – 0
9
ꢂ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÞÞ
These are not the final page numbers!