Investigation of cis- and trans-4-Fluoroprolines as Enantioselective Catalysts in a Variety of Organic Transformations

Article abstract of DOI:10.1071/CH14129

Stereoselective fluorination is known to rigidify the ring structure of L-proline, as a result of a combination of electrostatic and hyperconjugative effects associated with the C-F bond. This is a potential strategy for enhancing the enantioselectivity of proline-catalysed reactions. In this study, cis- and trans-4-fluoroprolines were investigated as catalysts in five different organic transformations, including examples of both enamine and iminium ion catalysis. Some significant differences in enantioselectivity were observed between the cis- and trans-isomers of the fluorinated catalysts, confirming that the ring pucker is important. However, no substantial improvements were observed relative to the parent catalyst, L-proline.

Full text of DOI:10.1071/CH14129

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2015, 68, 44–49
Full Paper
http://dx.doi.org/10.1071/CH14129
Investigation of cis- and trans-4-Fluoroprolines as
Enantioselective Catalysts in a Variety of Organic
Transformations
A
A
A
A
Daryl Q. J. Yap, Raju Cheerlavancha, Renecia Lowe, Siyao Wang,
A B
,
and Luke Hunter
A
School of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia.
Corresponding author. Email: l.hunter@unsw.edu.au
B
Stereoselective fluorination is known to rigidify the ring structure of L-proline, as a result of a combination of electrostatic
and hyperconjugative effects associated with the C–F bond. This is a potential strategy for enhancing the enantioselectivity
of proline-catalysed reactions. In this study, cis- and trans-4-fluoroprolines were investigated as catalysts in five different
organic transformations, including examples of both enamine and iminium ion catalysis. Some significant differences in
enantioselectivity were observed between the cis- and trans-isomers of the fluorinated catalysts, confirming that the ring
pucker is important. However, no substantial improvements were observed relative to the parent catalyst, L-proline.
Manuscript received: 10 March 2014.
Manuscript accepted: 12 April 2014.
Published online: 21 May 2014.
Introduction
enantioselective processes catalysed by L-proline have also been
developed.[1] However, the results are not always so impressive:
there are multiple reports of proline-catalysed reactions that
proceed with only moderate enantioselectivity.[2,5–8] A possible
reason for the imperfect selectivity observed in some cases is the
flexibility of proline’s five-membered ring (Fig. 1). Different
puckers of the proline ring would lead to different possible
transition state geometries, which may erode the stereoselectiv-
ity of organocatalytic reactions.
Thus, a potential strategy to enhance the enantioselectivity of
proline-catalysed reactions is to rigidify the five-membered ring
system. One way to achieve this is to exploit stereoselective
fluorination.[9,10] Fluorine preferentially aligns gauche to vici-
nal electronegative[11] or positively-charged[12] substituents,
because of a combination of hyperconjugative and electrostatic
interactions. This phenomenon has been observed in both cis-4-
fluoroproline and trans-4-fluoroproline (2 and 3, Fig. 1),[13] as
well as in their enamine and iminium derivatives.[14–16] This
leads to different preferred ring puckers for the cis- and trans-
isomers (Fig. 1), potentially offering a strategy for improving
the enantioselectivity of proline-catalysed reactions.
In this 21st century, it has become imperative for synthetic
chemists to develop processes that minimise harm to the envi-
ronment, either by controlling hydrocarbon consumption or
minimising overall waste. One way to approach this goal is by
developing more efficient catalysts. Within this context, much
attention has been directed towards organocatalysis, i.e. the
acceleration of reactions by small molecules that are comprised
exclusively of non-metal elements.[1] A particular focus has
been on chiral secondary amines: such molecules are versatile
entities because they can readily be converted into chiral
nucleophiles (enamines)[2] or chiral electrophiles (iminium
ions)[3] that can participate in a variety of enantioselective
processes.[1]
The archetypal organocatalyst is L-proline (1, Fig. 1), which
has the advantages of being cheap and readily available in
optically pure form. In a celebrated example of enamine
catalysis, Hajos and Parrish[4] reported in 1974 an intramole-
cular aldol reaction in which the product was delivered in an
impressive 97 : 3 enantiomeric ratio (er), catalysed by just
3 mol-% of L-proline (1). In the intervening years, many other
The concept of using fluorine to improve an organocatalyst’s
enantioselectivity has received considerable attention in recent
years.[17] Fig. 2 illustrates some examples of secondary amine
and N-heterocyclic carbene organocatalysts that owe their
enantioselectivity to conformational pre-organisation by fluo-
rine.[18–22] Surprisingly however, the literature on catalysis
using 4-fluoroprolines 2 and 3 is very scarce. In 2008, while
developing a total synthesis of the natural product hirsutene, List
and co-workers[23] investigated a series of L-proline derivatives
as catalysts in an enantioselective transannular aldol reaction.
They found that trans-4-fluoroproline (3) was the optimal
catalyst, delivering the desired aldol product in higher optical
ϭ
ϭ
ϭ
ϭ
1
2
3
ϭ
ϭ
Fig. 1. The flexible ring structure of L-proline (1) can be rigidified by
stereoselective fluorination (2, 3).
Journal compilation Ó CSIRO 2015
www.publish.csiro.au/journals/ajc

Cis- and trans-4-Fluoroprolines as Enantioselective Catalysts
45
purity (er . 99 : 1) than was obtained with either cis-4-
fluoroproline (2) or L-proline (1) itself. A subsequent theoretical
study by D´ıaz and Goodman[14] confirmed that the Cg-exo
transition state was favoured, and this contributed to the high
enantioselectivity. Surprisingly, this work remains the only
reported example of a reaction catalysed by 4-fluoroprolines.
Accordingly, the aim of this study is to investigate cis- and
trans-4-fluoroprolines (2 and 3, Fig. 1) as catalysts in a variety of
organic transformations, to determine whether rigidifying the
proline ring structure can be a general strategy for enhancing
enantioselectivity. This study will focus on examples from the
literature in which L-proline (1) has previously been utilised
successfully but with only moderate enantioselectivity: an
Results
Synthesis of 4-Fluoroproline Catalysts
The first task was to procure the requisite catalysts,
4-fluoroprolines 2 and 3. Both of these compounds were synthe-
sised from trans-4-hydroxyproline according to the methods of
Chorghade and co-workers[24] with slight modifications.[25] With
the catalysts 1–3 in hand, attention was next turned towards their
application in several enantioselective reactions.
Aldol Reaction
The first organocatalytic process to be investigated was an
intermolecularaldolreactionbetween 9 and10 (Table1, entry 1).
This reaction, which is an example of enamine catalysis, was
previously reported by List and co-workers to deliver the
product 11 in er 88 : 12.[2] In our hands, the product 11 was found
to be very susceptible to dehydration, so the reaction was closely
monitored by TLC and worked up immediately upon comple-
tion. In this manner, the purified aldol product 11 was obtained
in 38 % yield, which was somewhat lower than List’s reported[2]
yield of 68 %. The reaction was then repeated using cis-4-
fluoroproline (2) and trans-4-fluoroproline (3) as the catalysts
(Table 1, entries 2 and 3), with the only difference in each case
being the addition of one equivalent of triethylamine to neu-
tralise the hydrochloride salts of 2 and 3. The latter reactions
resulted in yields of 56 and 47 %, respectively.
intermolecular aldol reaction,[2] a Robinson annulation,[5]
a
Mannich reaction,[6] a Michael addition reaction,[7] and a
multicomponent synthesis of a pyran.[8]
4
5
6
High-performance liquid chromatography (HPLC) with a
chiral stationary phase was employed to determine the optical
purity of the aldol product 11. After extensive method develop-
ment a partial separation of the enantiomers of 11 was able to be
achieved:[25] this revealed that the catalyst L-proline (1) had
delivered the aldol product in er 77 : 23 (cf. the literature result[2]
of er 88 : 12). HPLC analysis of the aldol products 11 obtained
with cis-4-fluoroproline (2) and trans-4-fluoroproline (3)
revealed optical purities of er 73 : 27 and ee 79 : 21, respectively
(Table 1, entries 2 and 3). Thus, a slight improvement in
enantioselectivity was observed with trans-4-fluoroproline (3),
while a decrease in enantioselectivity was observed with the
catalyst cis-4-fluoroproline (2).
7
8
Fig. 2. Examples of fluorinated organocatalysts.[18–22]
Table 1. Organocatalysed aldol reaction
-
ϩ
9
10
11
Robinson Annulation
Entry
Catalyst
Isolated yield of 11
Optical purity of 11
The second organocatalytic reaction to be investigated was a
variant of the Robinson annulation (Table 2). This reaction,
which is an example of enamine catalysis, was previously shown
by Bui and co-workers to give the product 15 in er 85 : 15.[5]
1
2
3
1
2
3
38 %
56 %
47 %
er 77 : 23
er 73 : 27
er 79 : 21
Table 2. Organocatalysed Robinson annulation
-
ϩ
12
13
Catalyst
14
Isolated yield of
15
Entry
Conversion of
Optical purity of
14-15
15
15
1
1
41 %
25 %
8 %
nd
5 %
1 %
er 85 : 15
er 79 : 21
er 86 : 14
er 85 : 15
er 86 : 14
2
3
4
5
2
3
1 %
1A
3A
26 %
9 %
nd
AReaction performed with purified 14 and 20 mol-% of catalyst.

46
D. Q. J. Yap et al.
Initial experiments in our hands revealed that the purity of the
precursor 12 was critical, and this reagent was therefore distilled
immediately before use. Even so, the reaction remained very
low-yielding in our hands, with the product 15 obtained in only
5, 1, and 1 % yield in the reactions catalysed by L-proline (1), cis-
4-fluoroproline (2), and trans-4-fluoroproline (3), respectively
(cf. the literature yield[5] of 57 %, catalysed by 1). The extremely
low yields of 15 were partly because of the need for multiple
rounds of column chromatography, since the triketone 14 and
product 15 eluted very close to one another. However the extents
of conversion, which were estimated by 1H NMR spectroscopy,
were also low: 41, 25, and 8 % for the reactions catalysed by 1–3,
respectively. This suggests that the fluorinated catalysts exhibit
decreased reactivity (see below). In an attempt to improve the
yields of 15, the organocatalytic reactions were repeated with
purified triketone 14 and with a higher catalyst loading (Table 2,
entries 4 and 5). The isolated yields of 15 were somewhat
improved under these conditions, although still lower than the
literature figure.[5]
The optical purity of the Robinson products 15 were
determined using HPLC with a chiral stationary phase. The
result of er 85 : 15 obtained with L-proline (1) agrees with the
literature report[5] (Table 2). For the fluorinated catalysts, cis-4-
fluoroproline (2) decreased the optical purity quite significantly
to er 79 : 21, while trans-4-fluoroproline (3) delivered the
product with a similar or very slightly higher level of enantios-
electivity (er 86 : 14) than the parent catalyst 1. This result seems
to mirror the observations from the previous organocatalytic
reaction (Table 1), where catalyst 2 exhibited lower enantios-
electivity while catalyst 3 led to a possible marginal improve-
ment. The reaction with a higher loading of catalyst 3 (Table 2,
entry 5) did not give any further improvement.
This is almost identical to the result obtained under the same
conditions in this project, but it is obvious that the fluorinated
catalysts (2 and 3) resulted in considerably lower optical purity.
Michael Addition
The next organocatalytic reaction to be investigated in this
project was a variant of the Michael addition (Table 4). This
reaction, which is an example of iminium ion catalysis, was
previously developed by Hanessian and co-workers and gave the
product 21 in er 81 : 19.[7] When this reaction was repeated in
our laboratory the Michael product 21 was generated in 25 %
yield, which was similar to the literature yield[7] of 30 %. The
yields produced by the fluorinated catalysts (2 and 3) were
somewhat lower, with a decrease to 13 % in the reaction
catalysed by cis-4-fluoroproline (2) and 14 % in the reaction
catalysed by trans-4-fluoroproline (3).
HPLC with a chiral stationary phase was found to be unsuit-
able for determining the optical purity of the Michael product 21,
since the enantiomers could not be resolved. Therefore, an
alternative method was investigated in which ketone 21 was
converted into a pair of diastereoisomeric acetals through reac-
tion with a chiral diol (see Experimental section). The relative
amounts of the diastereoisomeric acetals were quantified by
13C NMR spectroscopy, and this revealed that the Michael
product 21 was obtained in er 82: 18 using L-proline (1) as the
catalyst (Table 4, entry 1). This is very similar to the value of
er 81: 19 reported in the literature.[7] The reaction catalysed by
cis-4-fluoroproline (2) delivered the Michael product 21 in a
decreased optical purity of er 79: 21, while the reaction catalysed
by trans-4-fluoroproline (3) delivered the Michael product 21 in
an almost identical optical purity of er 81: 19. In an effort to
further investigate whether catalyst 3 could represent any
improvement over catalyst 1 in the Michael addition, the reaction
Mannich Reaction
The next organocatalytic reaction to be investigated was a var-
iant of the Mannich reaction (Table 3). This reaction, which is an
example of iminium ion catalysis, was previously developed
by List and co-workers and delivered the Mannich product 18 in
er 87 : 13.[6] In our hands, the yield of the product 18 obtained
in the reaction catalysed by L-proline (1) was 28 %, significantly
lower than the literature yield[6] of 74 %. When substituted with
the fluorinated catalysts, cis-4-fluoroproline (2) yielded a slight
decrease to 23 % while trans-4-fluoroproline (3) yielded a dra-
matic decrease to 6 % (Table 3).
The enantiomers of 18 were readily separated by HPLC with
a chiral stationary phase, and this enabled the optical purity of
the products to be measured at er 86 : 14, 65 : 35, and 70 : 30 for
the reactions catalysed by 1–3 respectively. The literature[6]
optical purity of 18, obtained using L-proline (1), was er 87 : 13.
Table 4. Organocatalysed Michael addition
-
ϩ
19
20
Catalyst
21
Entry
Isolated yield of 21
Optical purity of 21
1
2
3
4
1
25 %
13 %
14 %
29 %
er 82 : 18
er 79 : 21
er 81 : 19
er 77 : 23
2
3
3A
AReaction performed with 20 mol-% of catalyst 3.
Table 3. Organocatalysed Mannich reaction
-
ϩ
ϩ
10
16
17
18
Optical purity of 18
Entry
Catalyst
Isolated yield of 18
1
2
3
1
2
3
28 %
23 %
6 %
er 86 : 14
er 65 : 35
er 70 : 30

Cis- and trans-4-Fluoroprolines as Enantioselective Catalysts
47
Table 5. Multicomponent synthesis of a pyran
-
ϩ
ϩ
22
23
24
25
Entry
Catalyst
Isolated yield of 25
Optical purity of 25
1
2
3
1
2
3
98 %
85 %
97 %
er 51 : 49
er 50 : 50
er 51 : 49
was repeated with 20 mol-% of catalyst 3 (Table 4, entry 4).
However this did not improve the enantioselectivity.
trans-4-fluoroprolines (catalysts 2 and 3), particularly in the
aldol reaction (Table 1) and the Robinson annulation (Table 2).
These results mirror List’s previous observation[23] that trans-4-
fluoroproline (3) gave substantially higher enantioselectivity in
an intramolecular aldol reaction than cis-4-fluoroproline (2).
Together, these findings confirm that the proline ring shape is an
important determinant of selectivity.
However, no substantial improvements over the parent cata-
lyst, L-proline (1), were observed in any of the reactions
investigated in this study. Cis-4-fluoroproline (2) tended to
decrease the enantioselectivity relative to L-proline (1) in all
cases. The results with trans-4-fluoroproline (3) were more
variable; in some cases the enantioselectivity was reduced
relative to the parent catalyst 1 (e.g. Table 3), while in other
cases the enantioselectivity was approximately equal to, or
slightly higher than, that of 1 (e.g. Tables 1, 2 and 4). This
suggests that rigidifying the proline ring structure can be
beneficial in some cases, but that rigidity is not the only factor
that is important for enantioselectivity.
In addition, it became clear during this study that fluorina-
tion seemed to reduce the catalysts’ reactivity in most cases
(e.g. Tables 2, 3 and 4). This is possibly attributable to the
electron-withdrawing nature of the fluorine substituent, which
presumably lowers the nucleophilicity of the amino group of
the catalyst.[16] Unfortunately, the design of this study did not
allow us to disentangle the different effects of fluorination
(i.e. rigidification and electron-withdrawing nature).
Finally, difficulties were encountered in the multicomponent
synthesis of pyran 25 (Table 5). Mechanistically, this process is
more complex than the other reactions investigated in this work;
our inability to reproduce the literature enantioselectivity of this
process highlights that subtle variations in experimental condi-
tions may dramatically affect reaction outcomes in certain
circumstances, adding an additional challenge to the study of
organocatalysis.
Multicomponent Synthesis of a Pyran
The final organocatalytic reaction examined in this work was the
multicomponent synthesis of a chiral pyran (Table 5). This
reaction, which is an example of enamine catalysis, was previ-
ously developed by Elnagdi and co-workers and reportedly gave
pyran 25 in er 85 : 15.[8] In our hands, this process delivered the
pyran 25 in 98, 85, and 97 % yields for the reactions catalysed by
1–3, respectively (Table 5). These yields were slightly higher
than the literature report[8] obtained with catalyst 1 (83 %).
Although the obtained yields of pyran 25 were gratifying
(Table 5), the determination of their optical purity was other-
wise. In the literature report,[8] pyran 25 was obtained in
er 85 : 15 using L-proline (1) as the catalyst, but in this work
under identical conditions pyran 25 was obtained in only
er 51 : 49 as determined by HPLC with a chiral stationary phase.
To rule out the possibility of any inaccuracy associated with our
HPLC method, we also employed the literature method[8] for
determining theoptical purity of 25. Thus, the1H NMR spectrum
of 25 was recorded in the presence of the chiral shift reagent
europium
tris[3-(heptafluoropropylhydroxymethylene)-(þ)-
camphorate]. Under these conditions, the pyran ring hydrogen
of 25 (4.7 ppm) resolved into two components, integration of
which confirmed that pyran 25 was essentially a racemate. The
batches of 25 obtained with cis-4-fluoroproline (2) and trans-4-
fluoroproline (3) were also found to be racemates (Table 5,
entries 2 and 3). One possible explanation is that epimerisation
occurred during these reactions, but we were not able to
conclusively determine whether this was the case.
Discussion
The hypothesis tested in this study was that rigidifying the five-
membered ring structure of L-proline (1) could be a general
strategy for achieving higher enantioselectivity in organocata-
lytic reactions. Five different organic transformations were
Conclusion
investigated: an aldol reaction,[2] a Robinson annulation,[5]
a
Cis-4-fluoroproline (2) and trans-4-fluoroproline (3) have been
investigated as enantioselective organocatalysts in a variety of
organic transformations, and their performance has been com-
pared with L-proline (1). Significant differences in enantios-
electivityare observedinsomecases between2 and3, confirming
that the ring pucker can be an important determinant of selec-
tivity. However, in general the fluorinated catalysts 2 and 3 did
not exhibit any substantial improvements relative to the parent
catalyst, 1. It was also observed that fluorination tended to reduce
the catalysts’ reactivity, presumably because of the fluorine’s
electron-withdrawing nature. These results suggest that future
applications of 4-fluoroprolines in organocatalysis may continue
to be limited to specific, isolated examples.[23]
Mannich reaction,[6] a Michael addition reaction,[7] and a multi-
component synthesis of a pyran.[8] These reactions included
examples of both enamine and iminium catalysis, and they were
chosen for investigation because they have previously been
reported to be catalysed by L-proline (1) but with only moderate
enantioselectivity. The strategy employed here for rigidifying
the ring structure of L-proline was to incorporate a fluorine atom
stereospecifically at the 4-position, since this has previously
been shown to enforce either a Cg-exo or a Cg-endo ring pucker
depending on the fluorine stereochemistry (Fig. 1).[13–16]
The first noteworthy outcome of this study is that some
significant differences in er were observed between cis- and

48
D. Q. J. Yap et al.
Experimental
(S)-8a-Methyl-3,4,8,8a-tetrahydronaphthalene-1,6(2H,7H)-
dione (15)
General
Catalyst (5 mol-%) was added to a solution of ketone 14
(0.950 g, 4.84 mmol) in anhydrous DMSO (5.0 mL) and stirred
for 120 h. The mixture was then dissolved in ethyl acetate
(200 mL), washed with water (10 ꢀ 40 mL), dried (MgSO4) and
concentrated under vacuum at 658C. Purification by flash
chromatography (3 : 2 n-hexane/ethyl acetate followed by 5 : 1
n-hexane/ethyl acetate) provided ketone 15.
Data for ketone 15, obtained using catalyst 1: Yellow oil
(76.9 mg, 5 %); [a]D þ68.0 (c 0.25, toluene), er 85 : 15 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH 96 : 4, flow rate
All reactions were performed in oven-dried glassware under a
nitrogen atmosphere with magnetic stirring unless otherwise
stated. Anhydrous solvents were freshly prepared as follows:
˚
solvents were dried over activated finely ground 4 A molecular
sieves for 1 h and then filtered. All other reagents were pur-
chased in the highest available quality and used as supplied.
‘Concentration under vacuum’ refers to the removal of volatile
solvents under reduced pressure by means of a rotary evaporator
and water bath (408C, unless otherwise stated) and subsequent
drying under high vacuum (,0.1 mmHg) at room temperature.
Solution phase reactions were monitored by TLC using Merck
aluminium-backed silica gel 60 F254 (0.2 mm) TLC plates. TLC
spots were visualised under short-wave UV light (254 nm) and
stained with Goofy’s stain, ninhydrin, or potassium permanga-
nate dips. Flash column chromatography was performed using
Davisil 40–63 mesh silica gel. The eluent is stated as volume-to-
volume ratios. Melting points were determined using a Mel-
Temp. II device. Optical rotations were measured using a Perkin
Elmer model 341 polarimeter (l ¼ 589 nm; l ¼ 1 dm, and c
expressed in grams per 100 mL). NMR spectra were obtained at
298 K using a Bruker Avance III 300, 400, 500, or 600 MHz
instrument. High resolution mass spectra were recorded at the
Bioanalytical Mass Spectrometry Facility (UNSW) using
an Orbitrap LTQ XL ion trap MS in positive ion mode
using an electrospray ionisation (ESI) source. HPLC was
performed using a Daicel Chiralcel OD-H analytical column
(250 mm ꢀ 4.6 mm ID). HPLC samples were run at isocratic
elution, and retention times were confirmed by comparison with
racemic samples.
0.8 mL minꢁ1
, l 280 nm, tR(major) 31.2 min, tR(minor)
34.2 min) (lit.[28] [a]D þ68 (c 1.5, toluene), er 85 : 15). dH
(400 MHz, CDCl3) 5.79 (d, J 1.5, 1H, CH), 2.72–2.63 (m, 2H,
CH2), 2.47–2.37 (m, 4H, 2 ꢀ CH2), 2.12–2.05 (m, 3H, CH2),
1
1.71–1.59 (m, 1H, CH2), 1.40 (s, 3H, CH3). H NMR data in
accordance with literature values.[29]
Data for ketone 15, obtained using catalyst 2: Yellow oil
(1.8 mg, 1 %); [a]D þ40.0 (c 0.25, toluene), er 79 : 21 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH, 96 : 4, flow rate
1
0.8 mL minꢁ1, l 280 nm). H NMR identical to that described
above.
Data for ketone 15, obtained using catalyst 3: Yellow oil
(4.0 mg, 1 %); [a]D þ4.0 (c 0.25, toluene), er 86 : 14 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH, 96 : 4, flow rate
1
0.8 mL minꢁ1, l 280 nm). H NMR identical to that described
above.
(R)-4-((4-Methoxyphenyl)amino)octan-2-one (18)
A suspension of catalyst (35 mol-%), p-anisidine (135 mg,
1.10 mmol), freshly distilled valeraldehyde (0.11 mL,
1.0 mmol), and acetone (10 mL) was stirred at room temperature
for 48 h. The reaction was filtered and concentrated under
vacuum without heating. Purification by flash chromatography
(5 : 1 n-hexane/ethyl acetate) provided ketone 18.
(R)-4-Hydroxy-4-(4-nitrophenyl)butan-2-one (11)
Acetone (1.0 mL, 14 mmol) was added to a solution of catalyst
(20 mol-%) and p-nitrobenzaldehyde (73.0 mg, 0.483 mmol) in
DMSO (4.0 mL). The mixture was stirred at room temperature
overnight. Saturated NH4Cl (10 mL) was then added, and the
mixture was extracted with ethyl acetate (3 ꢀ 15 mL). After
washing with water (5 ꢀ 10 mL) and brine (10 mL), the organic
solution was dried (MgSO4) and concentrated under vacuum
without heating. Purification by flash chromatography (3 : 1
n-hexane/ethyl acetate) provided the title compound.
Data for ketone 18, obtained using catalyst 1: Yellow oil
(71.5 mg, 28 %); [a]D þ4.8 (c 3.2, CHCl3), er 86 : 14 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH, 95 : 5, flow rate
1.0 mL minꢁ1
, l 315 nm, tR(major) 16.4 min, tR(minor)
10.0 min). dH (400 MHz, CDCl3) 6.76 (m, 2H, p-methoxyphe-
nyl), 6.57 (m, 2H, p-methoxyphenyl), 3.75–3.69 (m, 4H, OCH3
and NCH), 3.28 (br s, 1H, NH), 2.65 (dd, J 5.5, 16.3, 1H, C(O)
CHH), 2.56 (dd, J 6.4, 16.3, 1H, C(O)CHH), 2.12 (s, 3H,
CH3CO), 1.56–1.51 (m, 2H, C(N)CH2CH2), 1.44–1.26 (m,
Data for ketone 11, obtained using catalyst 1: Yellow solid
(38.4 mg, 38 %); mp 55–588C (lit.[26] 57–608C); [a]D þ28.6
(c 0.35, CHCl3), er 77 : 23 by HPLC with a chiral stationary
1
4H, CH2–CH2–CH3), 0.88 (t, J 6.9, 3H, CH2CH3). H NMR
data in accordance with literature values.[6]
phase (n-hexane/i-PrOH, 95 : 5, flow rate 1.0 mL minꢁ1
,
l 280 nm, tR(major) 37.4 min, tR(minor) 39.4 min) (lit.[27] [a]D
þ46.2 (c 1.0, CHCl3), er 88 : 12). dH (400 MHz, CDCl3) 8.16
(d, J 8.6, 2H, p-nitrophenyl), 7.52 (d, J 8.6, 2H, p-nitrophenyl),
5.24 (m, 1H, CHOH), 3.71 (br s, 1H, OH), 2.85–2.83 (m, 2H,
CHCH2), 2.20 (s, 3H, CH3). 1H NMR data in accordance with
literature values.[27]
Data for ketone 18, obtained using catalyst 2: Yellow oil
(58.6 mg, 23 %); [a]D þ1.3 (c 2.4, CHCl3), er 65 : 35 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH, 95 : 5, flow rate
1
1.0 mL minꢁ1, l 315 nm). H NMR spectrum identical to that
described above.
Data for ketone 18, obtained using catalyst 3: Yellow oil
(15.8 mg, 6 %); [a]D þ1.9 (c 0.52, CHCl3), er 70 : 30 by HPLC
with a chiral stationary phase (n-hexane/i-PrOH, 95 : 5, flow rate
Data for ketone 11, obtained using catalyst 2: Yellow solid
(58.5 mg, 56 %); mp 56–598C; [a]D þ37.1 (c 0.35, CHCl3),
er 73 : 27 by HPLC with a chiral stationary phase (n-hexane/
1
1.0 mL minꢁ1, l 315 nm). H NMR spectrum identical to that
1
i-PrOH, 95 : 5, flow rate 1.0 mL minꢁ1, l 280 nm). H NMR
described above.
identical to that described above.
(R)-3-(Nitromethyl)cyclopentanone (21)
Data for ketone 11, obtained using catalyst 3: Yellow solid
(48.0 mg, 47 %); mp 58–608C; [a]D þ45.7 (c 0.35, CHCl3),
er 79 : 21 by HPLC with a chiral stationary phase (n-hexane/
i-PrOH 95 : 5, flow rate 1.0 mL minꢁ1, l 280 nm). H NMR
A mixture of 2-cyclopenten-1-one (44ml, 0.53mmol), nitro-
methane (60 ml, 1.1mmol), 2,5-dimethylpiperazine (60mg,
0.53 mmol), and catalyst (4 mol-%) was stirred in chloroform
previouslypassedthrougha bedofbasicalumina (4.0 mL)for62 h
1
identical to that described above.

Cis- and trans-4-Fluoroprolines as Enantioselective Catalysts
49
at room temperature. The mixture was then diluted with dichlor-
omethane (16mL) and washed with 3 % aqueous HCl (10 mL).
The organic extract was dried (MgSO4) and concentrated under
vacuum without heating. Purification by flash chromatography
(1:1 n-hexane/ethyl acetate) provided ketone 21.
Data for ketone 21, obtained using catalyst 1: Yellow oil
(12.4 mg, 17 %); [a]D þ56.4 (c 0.39, CHCl3), er 82 : 18 by chiral
derivatisation with (7R)-2,3-dimethyl-7-(nitromethyl)-1,4-
dioxaspiro[4.4]nonane (lit.[7] [a]D þ66.9 (c 0.60, CHCl3),
er 81 : 19). dH (400 MHz, CDCl3) 4.51–4.43 (m, 2H, CH2NO2),
3.00 (m, 1H, CH–CH2), 2.53 (dd, J 7.7, 18.3, 1H, CH2), 2.40
(m, 1H, CH2), 2.33–2.22 (m, 2H, CH2), 2.01 (dd, J 10.4, 18.4,
1H, CH2), 1.70 (m, 1H, CH2). 1H NMR data in accordance with
literature values.[30]
Data for ketone 21, obtained using catalyst 2: Yellow oil
(9.4 mg, 13 %); [a]D þ55.0 (c 0.40, CHCl3), er 79 : 21 by chiral
derivatisation with (7R)-2,3-dimethyl-7-(nitromethyl)-1,4-
dioxaspiro[4.4]nonane. 1H NMR spectrum identical to that
described above.
Data for ketone 21, obtained using catalyst 3: Yellow oil
(10.4 mg, 14 %); [a]D þ52.3 (c 0.44, CHCl3), er 81 : 19 by chiral
derivatisation with (7R)-2,3-dimethyl-7-(nitromethyl)-1,4-
dioxaspiro[4.4]nonane. 1H NMR spectrum identical to that
described above.
[4] Z. G. Hajos, D. R. Parrish, J. Org. Chem. 1974, 39, 1615. doi:10.1021/
JO00925A003
[5] T. Bui, C. F. Barbas, Tetrahedron Lett. 2000, 41, 6951. doi:10.1016/
S0040-4039(00)01180-1
[6] B. List, J. Am. Chem. Soc. 2000, 122, 9336. doi:10.1021/JA001923X
[7] S. Hanessian, V. Pham, Org. Lett. 2000, 2, 2975. doi:10.1021/
OL000170G
[8] N. M. Elnagdi, N. S. Al-Hokbany, Molecules 2012, 17, 4300.
doi:10.3390/MOLECULES17044300
[9] L. Hunter, Beilstein J. Org. Chem. 2010, 6, 1. doi:10.3762/BJOC.6.38
[10] X.-G. Hu, L. Hunter, Beilstein J. Org. Chem. 2013, 9, 2696.
doi:10.3762/BJOC.9.306
[11] (a) K. B. Wiberg, M. A. Murcko, K. E. Laidig, P. J. MacDougall,
J. Phys. Chem. 1990, 94, 6956. doi:10.1021/J100381A008
(b) D. O’Hagan, C. Bilton, J. A. K. Howard, L. Knight, D. J. Tozer,
J. Chem. Soc., Perkin Trans. 2 2000, 605. doi:10.1039/B000205O
(c) C. R. Briggs, D. O’Hagan, H. S. Rzepa, A. M. Z. Slawin, J. Fluor.
Chem. 2004, 125, 19. doi:10.1016/J.JFLUCHEM.2003.08.011
[12] (a) A. M. Sun, D. C. Lankin, K. Hardcastle, J. P. Snyder, Chem. – Eur.
J. 2005, 11, 1579. doi:10.1002/CHEM.200400835
(b) C. R. Briggs, M. J. Allen, D. O’Hagan, D. J. Tozer, A. M. Z. Slawin,
A. E. Goeta, J. A. K. Howard, Org. Biomol. Chem. 2004, 2, 732.
doi:10.1039/B312188G
(c) N. E. J. Gooseman, D. O’Hagan, M. J. G. Peach, A. M. Z. Slawin,
D. J. Tozer, R. J. Young, Angew. Chem. Int. Ed. 2007, 46, 5904.
doi:10.1002/ANIE.200700714
[13] (a) D. B. Hobart, J. S. Merola, Acta Crystallogr. E 2012, 68, o2490.
doi:10.1107/S1600536812031741
(b) S. K. Holmgren, L. E. Bretscher, K. M. Taylor, R. T. Raines, Chem.
Biol. 1999, 6, 63. doi:10.1016/S1074-5521(99)80003-9
[14] J. D´ıaz, J. M. Goodman, Tetrahedron 2010, 66, 8021. doi:10.1016/
J.TET.2010.08.003
[15] D. A. Bock, C. W. Lehmann, B. List, Proc. Natl. Acad. Sci. USA 2010,
107, 20636. doi:10.1073/PNAS.1006509107
[16] Y. P. Rey, L. E. Zimmer, C. Sparr, E.-M. Tanzer, W. B. Schweizer,
H. M. Senn, S. Lakhdar, R. Gilmour, Eur. J. Org. Chem. 2014, 1202.
doi:10.1002/EJOC.201301730
2-Amino-4,6-diphenyl-4H-pyran-3,5-dicarbonitrile (25)
A mixture of benzaldehyde (70mL, 0.69 mmol), malononitrile
(45.5 mg, 0.689mmol), and catalyst (10 mol-%) was stirred at
room temperature for 2 min. Nitrile 24 (100mg, 0.689 mmol) was
then added and the mixture was refluxed in ethanol (2.0mL) for
6 h. After concentrating under vacuum, purification by flash
chromatography (2: 1 n-hexane/ethyl acetate) provided pyran 25.
Data for pyran 25, obtained using catalyst 1: Yellow
solid (0.202 g, 98 %); mp 159–1628C (lit.[8] 162–1638C); [a]D
ꢁ4.0 (c 0.25, CHCl3), er 51 : 49 by HPLC with a chiral station-
[17] L. E. Zimmer, C. Sparr, R. Gilmour, Angew. Chem. Int. Ed. 2011, 50,
11860. doi:10.1002/ANIE.201102027
[18] (a) C. M. Marson, R. C. Melling, Chem. Commun. 1998, 1223.
doi:10.1039/A801718B
ary phase (n-hexane/i-PrOH, 90 : 10, flow rate 1.0 mL minꢁ1
,
l
280 nm, tR(major) 33.5 min, tR(minor) 23.3 min).
dH (400 MHz, DMSO) 7.81–7.79 (m, 2H, ArH), 7.62–7.53 (m,
3H, ArH), 7.46–7.42 (m, 2H, ArH), 7.37–7.35 (m, 3H, ArH),
7.31 (s, 2H, NH2), 4.84 (s, 1H, CH). 1H NMR data in accordance
with literature values.[8]
(b) C. M. Marson, R. C. Melling, J. Org. Chem. 2005, 70, 9771.
doi:10.1021/JO0513414
[19] C. Sparr, R. Gilmour, Angew. Chem. Int. Ed. 2010, 49, 6520.
doi:10.1002/ANIE.201003734
[20] (a) C. Sparr, W. B. Schweizer, H. M. Senn, R. Gilmour, Angew. Chem.
Int. Ed. 2009, 48, 3065. doi:10.1002/ANIE.200900405
(b) I. G. Molna´r, E.-M. Tanzer, C. Daniliuc, R. Gilmour, Chem. – Eur.
J. 2014, 20, 794. doi:10.1002/CHEM.201303586
[21] D. A. DiRocco, K. M. Oberg, D. M. Dalton, T. Rovis, J. Am. Chem.
Soc. 2009, 131, 10872. doi:10.1021/JA904375Q
[22] Y. P. Rey, R. Gilmour, Beilstein J. Org. Chem. 2013, 9, 2812.
doi:10.3762/BJOC.9.316
[23] C. L. Chandler, B. List, J. Am. Chem. Soc. 2008, 130, 6737.
doi:10.1021/JA8024164
Data for pyran 25, obtained using catalyst 2: Yellow solid
(0.175 g, 85 %); mp 160–1628C; [a]D þ4.0 (c 0.25, CHCl3),
er 50 : 50 by HPLC with a chiral stationary phase (n-hexane/
1
i-PrOH, 90 : 10, flow rate 1.0 mL minꢁ1, l 280 nm). H NMR
identical to that described above.
Data for pyran 25, obtained using catalyst 3: Yellow solid
(0.200 g, 97 %); mp 157–1598C; [a]D þ8.0 (c 0.25, CHCl3),
er 51 : 49 by HPLC with a chiral stationary phase (n-hexane/
1
i-PrOH, 90 : 10, flow rate 1.0 mL minꢁ1, l 280 nm). H NMR
[24] M. S. Chorghade, D. K. Mohapatra, G. Sahoo, M. K. Gurjar,
M. V. Mandlecha, N. Bhoite, S. Moghe, R. T. Raines, J. Fluor. Chem.
2008, 129, 781. doi:10.1016/J.JFLUCHEM.2008.06.024
[25] See Supplementary Material.
identical to that described above.
Supplementary Material
Synthetic procedures, NMR spectra, and HPLC traces are
available on the Journal’s website.
[26] Z.-X. Du, L.-Y. Zhang, X.-Y. Fan, F.-C. Wu, C.-S. Da, Tetrahedron
Lett. 2013, 54, 2828. doi:10.1016/J.TETLET.2013.03.087
[27] K. Sakthivel, W. Notz, T. Bui, C. F. Barbas, J. Am. Chem. Soc. 2001,
123, 5260. doi:10.1021/JA010037Z
References
[28] P. Buchschacher, A. Fu¨rst, J. Gutzwiller, Org. Synth. 1985, 63, 37.
doi:10.15227/ORGSYN.063.0037
[1] D. W. MacMillan, Nature 2008, 455, 304. doi:10.1038/
NATURE07367
[29] R. Pedrosa, J. M. Andre´s, R. Manzano, C. Pe´rez-Lo´pez, Tetrahedron
Lett. 2013, 54, 3101. doi:10.1016/J.TETLET.2013.03.123
[30] S. V. Pansare, R. Lingampally, Org. Biomol. Chem. 2009, 7, 319.
doi:10.1039/B812038B
[2] B. List, R. A. Lerner, C. F. Barbas, J. Am. Chem. Soc. 2000, 122, 2395.
doi:10.1021/JA994280Y
[3] K. A. Ahrendt, C. J. Borths, D. W. MacMillan, J. Am. Chem. Soc. 2000,
122, 4243. doi:10.1021/JA000092S