Properties and reactivity of conformationally constrained bicyclic diarylprolinol silyl ethers as organocatalysts

Article abstract of DOI:10.1002/ejoc.201402732

Bicyclic silyl ether derivatives 1, which are derived from simple chemical manipulations of trans-4-L-hydroxyproline, were recently proposed, by us, as conformationally constrained analogues of J?rgensen-Hayashi's catalysts 2. Despite the structural simil

Full text of DOI:10.1002/ejoc.201402732

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
FULL PAPER
DOI: 10.1002/ejoc.201402732
Properties and Reactivity of Conformationally Constrained Bicyclic
Diarylprolinol Silyl Ethers as Organocatalysts
Marco Lombardo,*[a] Lucia Cerisoli,[a] Elisabetta Manoni,[a] Elisa Montroni,[a]
Arianna Quintavalla,[a] and Claudio Trombini[a]
Dedicated to C.I.N.M.P.I.S. on the occasion of its 20th anniversary
Keywords: Synthetic methods / Organocatalysis / Low-loading catalysis / Nitrogen heterocycles
Bicyclic silyl ether derivatives 1, which are derived from sim-
ple chemical manipulations of trans-4- -hydroxyproline,
formance in iminium chemistry with respect to 2, but much
lower reactivity in enamine chemistry. The peculiar struc-
L
were recently proposed, by us, as conformationally con- tural features of 1 confer a pattern of reactivity between
strained analogues of Jørgensen–Hayashi’s catalysts 2. De- Jørgensen–Hayashi’s catalysts and MacMillan’s imidazolid-
spite the structural similarities, 1 displays remarkable per-
inones.
are easily prepared in a few synthetic steps from commer-
cially available N-Cbz-trans-4-l-hydroxyproline.[10] In par-
ticular, catalysts 1a and 1b (Figure 1) displayed the best
overall performances when tested in a few benchmark or-
ganocatalyzed asymmetric transformations relative to some
Jørgensen–Hayashi’s diarylprolinol silyl ethers (2a–2c, Fig-
ure 1), which are among the best organocatalysts proposed
for iminium/enamine activation mode.[11]
Introduction
Silylated diarylprolinols (Jørgensen–Hayashi’s cata-
lysts),[1] together with proline[2] and MacMillan’s imid-
azolidinones,[3] are the workhorses of enantioselective cova-
lent organocatalysis. They are used in asymmetric versions
of most classical organic reactions, such as aldol,[4] Man-
nich,[5] Michael[6] and many others.[7] In addition, several
combinations of some of these reactions in domino pro-
cesses have been reported for the enantioselective one-pot
construction of several stereogenic centers starting from
prochiral reagents.[8] However, a few drawbacks affect or-
ganocatalytic reactions, such as long reaction times, high
catalyst loading and the need to carefully optimize reaction
conditions for each specific application.
In the search for organocatalysts that perform better, we
recently proposed the introduction of ionic groups cova-
lently installed on the pyrrolidine framework as a strategy
to implement calalytic activity. Remarkable improvements
were observed in the asymmetric aldol addition by using
an ion-tagged proline[9a–9c] and in the Michael addition of
aldehydes to nitroalkenes by employing an ion-tagged
modified Jørgensen–Hayashi’s catalyst.[9d]
Figure 1. Structures of secondary amine organocatalysts 1 and 2.
By using 1 we successfully carried out the cyclopropan-
ation reaction of α,β-unsaturated aldehydes with α-bromo-
malonates,[12] the conjugate Michael addition of nitrometh-
ane to (E)-cinnamaldehyde,[13] and the Diels–Alder reaction
of α,β,γ,δ-unsaturated aldehydes.[14] The activity and selec-
tivity of 1 in the aforementioned reactions were always com-
parable to, if not superior than, 2.[10]
The presence of a 2,4-dioxa-3-sila-7-azabicyclo[4.2.1]-
nonane scaffold confers to 1 a superior stability towards
hydrolytic conditions. Catalysts 1 are indeed bench-stable
solids, making their storage and handling extremely practi-
cal and convenient. The higher stability towards desil-
ylation, an undesired reaction that compromises catalyst
performance,[15] makes their use particularly attractive
when acids are needed as co-catalysts and/or when pro-
By following a different approach to modify the diaryl-
prolinol framework, we recently reported a new family of
conformationally constrained bicyclic catalysts (1), which
[a] Dipartimento di Chimica “G. Ciamician”, Università degli
Studi di Bologna,
Via Selmi 2, 40126 Bologna, Italy
E-mail: marco.lombardo@unibo.it
www: http://www.chemistry.unibo.it/en
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201402732.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
M. Lombardo et al.
FULL PAPER
longed reaction times are necessary to achieve good conver- Table 2. Organocatalytic cyclopropanation reaction of (E)-4-nitro-
cinnamaldehyde 3a with dimethyl α-bromomalonate 4 with catalyst
sions. Herein we wish to report a more detailed study on
the organocatalytic activity and on the structural peculiari-
ties of catalyst 1a.
1a in different catalytic amounts.
Entry[a]
1a [mol-%]
t [h]
Conv.[b]
ee [%][c]
1
2
3
5
1
0.1
12
21
24
99
90
20
94
95
94
Results and Discussion
[a] Reactions run with 3a (0.12 mmol), 4 (1.1 equiv.), reported
amounts of 1a and 2,6-lutidine (1.2 equiv.) in dichloromethane
(0.5 mL) at room temp. [b] Conversion calculated with respect to
the starting aldehyde was checked by 1H NMR spectroscopy
(400 MHz) by taking samples (20 μL) at different time intervals
and diluting them in CDCl3 (0.5 mL) in a standard 5 mm NMR
tube. [c] Determined by chiral HPLC on crude products.
Among previously tested reactions, the cyclopropanation
of (E)-4-nitrocinnamaldehyde (3a) with dimethyl α-bromo-
malonate (4) afforded the best results with catalysts 1a or
1b, and gave better yields and slightly enhanced enantio-
selectivities relative to Jørgensen–Hayashi’s catalyst 2a. We
decided to examine this reaction in more detail not only by
using 1a in lower catalytic amounts, but also by employing
(E)-cinnamaldehyde 3b and electron-rich (E)-4-methoxy-
cinnamaldehyde 3c (Table 1).
The reaction conversion was virtually complete after 12 h
by employing 1a at 5 mol-% (Table 2, Entry 1), whereas
90% conversion was obtained after 21 h when lowering the
catalyst loading to 1 mol-% (Table 2, Entry 2). A very inter-
esting result was obtained when the reaction was tested by
using 1a at 0.1 mol-%, a catalytic loading seldom used in
asymmetric organocatalyzed transformations,[16] providing
an appreciable 20% conversion after 24 h and retaining the
same high level of stereocontrol (Table 2, Entry 3).
Table 1. Organocatalytic cyclopropanation reaction of α,β-unsatu-
rated aldehydes 3a–3c with dimethyl α-bromomalonate 4 with cata-
lyst 1a at 10 mol-%.
One of the most frequent criticisms regarding organocat-
alytic protocols is that a relatively large amount of catalyst
is typically required to obtain the desired product at a use-
ful reaction rate. To better elucidate the potential of catalyst
1a, we decided to explore its activity relative to 2a in the
cyclopropanation reaction of 3a at 0.1 mol-% catalytic
loading, and followed the conversion at different time inter-
vals. The results obtained are reported in Table 3 and plot-
ted in Figure 2.
[b]
Entry[a]
3
t [h]
Conv. [%]
Y [%][c] ee [%][d]
1
2
3
4
3a
0.5
1
2
65
77
88
99
92
94
4
5
6
7
3b
3c
1
2
4
68
80
90
84
80
96
95
Table 3. Organocatalytic cyclopropanation reaction of (E)-4-nitro-
cinnamaldehyde 3a with dimethyl α-bromomalonate with catalyst
1a or catalyst 2a at 0.1 mol-%.
8
9
2
4
71
86
Entry[a]
t [d]
Conv. [%] (1a)[b]
Conv. [%] (2a)[b]
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
10
11
21
45
56
62
66
71
73
76
80
81
11
19
24
30
34
39
42
44
47
47
[a] Reactions run with 3 (0.12 mmol), 4 (1.1 equiv.), 1a (10 mol-%)
and 2,6-lutidine (1.2 equiv.) in dichloromethane (0.5 mL) at room
temp. [b] Conversion calculated with respect to the starting alde-
1
hyde was checked by H NMR spectroscopy (400 MHz) by taking
samples (20 μL) at different time intervals and diluting them in
CDCl3 (0.5 mL) in a standard 5 mm NMR tube. [c] Isolated yields
after purification by flash chromatography on silica. [d] Deter-
mined by chiral HPLC on purified products.
In all cases examined, under the same experimental con-
ditions and catalyst loading, 1a gave comparable yields with
respect to catalyst 2a, but in a shorter reaction time.
Wang and co-workers reported that by using 2a, 5a was
obtained in 92% yield (94% ee) after 5.5 h (relative to
Table 1, Entries 1–4,), 5b in 88% yield (96% ee) after 21 h
(Table 1, Entries 5–7), and 5c in 85% yield (94% ee) after
24 h (Table 1, Entries 8–9).[12a] In terms of enantiocontrol,
[a] Reactions run with 3a (0.1 mmol), 4 (1.1 equiv.), catalyst
(0.1 mol-%) and 2,6-lutidine (1.2 equiv.) in dichloromethane
(0.4 mL) at room temp. [b] Conversion calculated with respect to
the starting aldehyde was checked by 1H NMR spectroscopy
(400 MHz) by taking samples (20 μL) at different time intervals
and diluting them in CDCl3 (0.5 mL) in a standard 5 mm NMR
tube.
Catalyst 1a was significantly more active than 2a, and
1a and 2a provided the same results. The higher activity of reached 56% conversion after 3 d and a notable 76% con-
catalyst 1a prompted us to explore the reaction with more version after 8 d. A value of 94% ee was obtained by ana-
reactive aldehyde 3a at lower catalytic loadings. The results lyzing the crude reaction mixture after 11 d, without any
obtained are reported in Table 2.
significant difference from the results obtained employing
2
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
Conformationally Constrained Bicyclic Diarylprolinol Silyl Ethers
Encouraged by these results we lowered the catalyst loading
to 5 mol-%, and again we observed complete consumption
of limiting oxindole 6 after 24 h. Under these conditions,
1
besides the signals of 7, we identified in the H NMR spec-
trum of samples (taken from the reaction mixture and di-
luted in CDCl3 without any work-up) some new signals,
probably a result of aldol intermediate 8 (Figure 3). The
relative trans configuration between the C2–C3 substituents
1
was tentatively assigned by analysis of the H NMR spec-
troscopic coupling constant (J H2-H3 = 13.6 Hz), but we
did not define the relative C3–C4 stereochemistry.
Figure 2. Activities of 1a and 2a in the cyclopropanation reaction
of (E)-4-nitrocinnamaldehyde 3a with dimethyl α-bromomalonate 4
in dichloromethane at room temp., with 0.1 mol-% catalyst loading.
1a at higher catalytic loadings. The same level of enanti-
ocontrol was displayed by 2a, but Jørgensen–Hayashi’s cat-
alyst afforded lower conversions in the same reaction time.
This difference was particularly noticeable in the earlier
stages of the cyclopropanation reaction, in which conver-
sions with 1a were always more than twofold those ob-
tained with 2a.
To expand the study on the activity of 1a in organocata-
lytic transformations, we decided to test it in a different
enantioselective iminium/enamine cascade reaction. The re-
action chosen was between 3-substituted oxindoles and α,β-
unsaturated aldehydes catalyzed by 2a in the presence of an
equimolar amount of benzoic acid as co-catalyst, which was
recently reported by Barbas III and co-workers.[17] In par-
ticular we chose to examine the reaction of N-Boc-3-(2-
oxopropyl)-oxindole 6[18] with 3b (Scheme 1), a reaction
that affords complex spiro-cyclopenteneoxindole 7 in 95%
ee and 9:1 dr after 24 h at 37 °C with catalyst 2a at 10 mol-
%.
Figure 3. Structures of intermediate 8 and product 7 and the 1H
NMR spectrum (400 MHz) inset relative to the reaction between 6
and 3b catalyzed by 2a at 5 mol-% in toluene at room temp. in the
presence of benzoic acid (5 mol-%), after 24 h.
Although practically the same rate of conversion relative
to 6 was achieved after 24 h, catalyst 1a and 2a behave dif-
ferently with respect to the relative ratios between 7 and 8
at different time intervals. By using 2a the 7/8 ratio was
67:33 after 24 h, 80:20 after 48 h, 85:15 after 4 d and finally
95:5 after 7 d. However, by using catalyst 1a the 7/8 ratio
was 20:80 after 24 h, 25:75 after 48 h, 37:63 after 4 d and
only 45:55 after 7 d (Figure 4). It should be noted we were
not able to isolate intermediate 8 because, after reaction
work-up and/or purification by flash chromatography on
silica, only dehydrated product 7 was quantitatively reco-
vered from the reaction mixture.
Scheme 1. Organocatalytic iminium/enamine cascade reaction of
N-Boc-3-(2-oxopropyl)oxindole 6 with (E)-cinnamaldehyde 3b.
Figure 4. 1H NMR spectroscopy (400 MHz) insets relative to the
reaction between 6 and 3b catalyzed by 1a (upper traces) and 2a
(bottom traces) at 5 mol-% in toluene at room temp. in the presence
of benzoic acid (5 mol-%), after 1 and 7 d.
We initially performed the reaction under similar condi-
tions reported by Barbas III, but at a temperature of 20 °C
instead of 37 °C. In this case we observed that both catalyst
When we lowered the catalyst loading to 2.5 mol-% a
1a and 2a at 10 mol-% quantitatively afforded desired prod- much slower reaction resulted, with conversions based on
uct 7 after 24 h, with 94% ee. A better diastereomeric ratio starting oxindole 6 of 50% (7/8 = 15:85) for catalyst 1a and
1
(dr Ͼ 95:5) was observed by H NMR spectroscopic analy- of 44% (7/8 = 57:43) for catalyst 2a after 24 h. Unfortu-
sis of the crude reaction mixtures obtained at room temp. nately, when we checked the conversions after 48 h, the re-
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
M. Lombardo et al.
FULL PAPER
actions seemed to have stopped, showing again a value of
This different behavior may be explained by considering
50% (7/8 = 20:80) for catalyst 1a and of 46% (7/8 = 59:41) that the much higher steric encumbrance of amine 1a de-
for catalyst 2a. Further lowering of the catalytic loading creases its basicity and makes the elimination reaction less
gave much slower reactions and with very low conversions efficient, even when catalyzed by weak acids. (Figure 5). In-
even after prolonged reaction times. When the reaction was deed, the apparent conjugate acids pKa calculated by using
tested with 0.5 mol-% catalytic loading no product forma- ACD/Percepta software resulted 7.9Ϯ0.5 for 1a and
1
tion was detected by H NMR spectroscopy for both 1a 8.5Ϯ0.4 for 2a.[20]
and 2a, even after 7 d.
We next expanded the study on the catalytic activity of
Catalysts 1a and 2a displayed almost the same activity in 1a to the enantioselective Michael addition of aliphatic al-
this iminium/enamine cascade reaction when benzoic acid dehydes to nitroalkenes, a classic workbench reaction com-
was used as an equimolar co-catalyst, but they showed monly employed to test the performance of new organocat-
markedly different behavior in the dehydration reaction alysts in enamine chemistry.[21]
from 8 to 7. Because the dehydration reaction should be
We recently employed, for the same reaction, an ion-
accelerated by acid-catalysis and because the relative tagged modified Jørgensen–Hayashi’s catalyst[9d] and, fol-
amount of benzoic acid was always the same in the two lowing the protocols reported, we tested 1a in dichloro-
sets of reactions, this different behavior may be ascribed to methane or in water as the solvent with and without ben-
different basicity of the catalysts employed. Although no zoic acid as co-catalyst. Invariably the product formed with
general rule can be simply derived to determine the efficacy excellent values of enantio- and diastereoselectivity, but sur-
of an acid additive on the activity of a determined catalyst prisingly in all tested conditions only very low yields of the
and/or reaction, Seebach and Hayashi recently studied in desired product were obtained in short reaction times.
detail the effect of acid co-catalysts in the Michael addition
To better understand the reasons for this poor reactivity,
of aliphatic aldehydes to nitroalkenes catalyzed by 2a.[19] we studied the enamine formation with both catalyst 1a and
They noted that the best results are obtained when acids in 2a by H NMR spectroscopy.[22] We choose phenyl-acetal-
1
the 4–6 pKa range are used. The acid additive plays many dehyde (12) as the electrophilic reaction partner because the
fundamental roles to foster the organocatalytic cycle by in- corresponding enamine with 2a has been extensively studied
creasing reaction rates, but the use of acids that are too by the groups of Seebach[23] and Mayr.[24] So by simply mix-
strong (pKa Ͻ 3) inhibit the reaction, probably as a result ing equimolar amounts of 12 and 1a, or 2a (0.030 mmol)
of a lower concentration of the free amine catalyst left in in CDCl3 (0.6 mL) in a standard 5-mm NMR tube we were
solution.
able to confirm that corresponding enamines 13 and 14
To verify the role of the acid in the dehydration reaction, were quantitatively formed almost instantaneously (Fig-
we analyzed the reactions catalyzed by 1a and 2a at 5 mol- ure 6).
% in the absence of benzoic acid. Starting oxindole 6 was
completely consumed after 24 h and in both cases aldol
product 8 was the major component of the reaction mix-
tures. Again different 7/8 ratios were found: 30:70 (versus
67:33 with benzoic acid) for 2a and 7:93 (versus 20:80 with
benzoic acid) by using 1a (Figure 5).
1
Figure 6. H NMR spectra (400 MHz) insets relative to enamines
13 and 14 in CDCl3.
The (E)-configuration of the double bond for both en-
amines was confirmed by inspection of the H1Ј-H2Ј cou-
pling constants (J = 14.0 Hz for 13, and 13.9 Hz for 14).
Figure 5. 1H NMR spectra (400 MHz) insets relative to the reac-
The preferred s-trans conformation of sterically constrained
tion between 6 and 3b catalyzed by 1a (upper trace) and 2a (bottom
sc-endo enamine 13 was supported by a strong anisotropic
shielding of the H1Ј proton, upfield-shifted to 5.80 ppm, in
trace) at 5 mol-% in toluene at room temp. without benzoic acid,
after 24 h.
4
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
Conformationally Constrained Bicyclic Diarylprolinol Silyl Ethers
accordance with the studies made by Zeitler and Gschwind increasing sp3 character was determined for less reactive en-
on the preferential conformations of 2a-derived enamines amine 17 (Δ = 0.293 Å).[24]
1
in solution.[22] The H NMR spectrum of enamine 14 was
To verify the pyramidalization degree of the nitrogen
in complete agreement with the one reported in the litera- atom in enamine 13, we calculated the M06–2x/6-311G(d,p)
ture for the isolated compound.[23]
optimized geometries of enamines 13, 14 and 17. From the
Once the quantitative formation of both the enamines in Cartesian coordinates of nitrogen atom and of the three
solution was confirmed, we added a stoichiometric amount attached carbons, the pyramidalization parameter (Δ) was
1
of β-(E)-nitrostyrene 10 to each one and recorded the H easily calculated (Figure 7).[26,27]
NMR spectrum again after 24 h. Although enamine 14 and
10 were completely consumed, enamine 13 was left com-
pletely untouched.
Mayr recently determined the relative reactivity towards
carbocations of several secondary amines derived enamines
of phenylacetaldehyde 12, such as simple pyrrolidine en-
amine 15 and MacMillan 1st and 2nd generation imidazolid-
inones derivatives 16 and 17. The relative reactivity order
(15 Ͼ 14 ϾϾ 16 Ͼ 17) roughly correlated to the 13C NMR
Figure 7. Vector notation and formula used to calculate the altitude
Δ of the trigonal pyramid defined by the nitrogen atom (N) and by
the three attached carbons (Ca–Cc).
spectroscopic chemical shifts of the C2Ј carbon. The more
C2Ј is downfield-shifted, the lower its electron density and,
as a consequence, its nucleophilicity. On this basis, 16 and
17 are up to 100 times less nucleophilic than 14.
In catalyst 1a derived enamine 13, the C2Ј carbon reso-
nates at δ = 100.0 ppm, exactly in between enamines 14 and
17 (Δ = –2.9 ppm; Table 4).
Again an almost planar sp2 nitrogen was calculated for
enamine 14 (Δ = 0.022 Å), in very good accordance with
the value derived by Seebach from X-ray crystal structure
data.[23] However, increased sp3 character was determined
for enamine 17 (Δ = 0.221 Å), in very good accordance with
the values determined from X-ray crystal structure data of
MacMillan’s catalyst-derived enamines.[24] The calculated Δ
value for enamine 13 derived from 1a (Δ = 0.159 Å) was
intermediate between the values of 14 and 17, in agreement
with the trend of the 13C NMR spectroscopic chemical
shifts.
Table 4. 13C NMR spectroscopic chemical shifts (CDCl3) of en-
amines 13–17.
The M06-2x/6-311G(d,p) optimized geometries of the
most stable conformers of enamines 13, 14 and 17 are dis-
played in Figure 8.
Entry
Enamine
δ(C2Ј) [ppm]
1
2
3
4
5
15
14
13
16
17
97.4[a]
97.2[b]
100.0
101.9[a]
102.9[a]
[a] Data taken from ref.[24] [b] Data taken from ref.[23] and con-
firmed by 13C NMR spectroscopy (100 MHz).
The relative reactivity of enamines can be also related to
the spn character of the nitrogen atom, because the higher
the sp2 character, the more efficient the p-π orbital interac-
tion results thus raising the electron density on C2Ј. To
quantitatively define this interaction, Dunitz introduced the
pyramidalization parameter Δ, which is defined as the dis-
tance between the nitrogen atom and the plane formed by
the three attached carbons.[25] An almost planar sp2 nitro-
Figure 8. M06-2x/6-311G(d,p) optimized geometries of the most
stable conformers of s-trans-(E)-configured enamines 13, 14 and
17.
To confirm this order of reactivity, we studied the ad-
gen was found in enamine 14 (Δ = 0.037 Å),[23] whereas dition reaction of propanal (9) to nitrostyrene (10) in
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
M. Lombardo et al.
FULL PAPER
and 5a). General Procedure: To
a
solution of 1a (3.9 mg,
dichloromethane as the solvent with catalysts 1a, 2a and
MacMillan’s imidazolidinone 18 at 10 mol-%, both with
and without an equimolar amount of benzoic acid as the
co-catalyst (Scheme 2).
0.012 mmol, 10%) in CH2Cl2 (0.5 mL) was added methyl bromo-
malonate (0.13 mmol, 0.019 mL, 1.1 equiv.), 2,6-lutidine
(0.14 mmol, 0.021 mL, 1.2 equiv.) and finally 4-nitro cinnamal-
dehyde (0.12 mmol, 21 mg). The reaction was stirred at room tem-
perature for the specified time. Product 5a was purified by flash
chromatography on silica gel (cyclohexane/ethyl acetate, 8:2). 1H
NMR (400 MHz, CDCl3): δ = 9.59 (d, J = 3.8 Hz, 1 H), 8.19 (d,
J = 8.8 Hz, 2 H), 7.44 (d, J = 8.2 Hz, 2 H), 3.88–3.83 (m, 4 H),
3.54 (s, 3 H), 3.47 (dd, J = 7.6, 3.8 Hz, 1 H) ppm. Conversions
1
were determined with H NMR spectroscopy by using the doublet
at δ = 9.59 ppm of the product and the doublet at δ = 9.79 ppm of
the starting aldehyde. The racemic product was synthesized under
the same conditions with racemic Jørgensen–Hayashi’s catalysts 2a.
The enantiomeric excess was determined by chiral HPLC after
derivatization of the product with Ph3P=CHCO2Et. Separation
conditions in chiral HPLC: AD 90:10 n-Hex/IPA for 15 min then
80:20 in 10 min, 0.5 mL/min, 40 °C, λ = 230 nm; tr = 32.6 min
(major), tr = 34.7 min (minor).
Scheme 2. Organocatalytic Michael addition of propanal 9 to β-
(E)-nitrostyrene 10 with catalysts 1a, 2a and 18.
Organocatalytic Cyclopropanation Reaction of (E)-Cinnamaldehyde
3b with Dimethyl α-Bromomalonate 4: (Table 1, Entry and 5b). Fol-
lowing the procedure reported for 5a, 5b was isolated by flash
chromatography on silica gel (cyclohexane/ethyl acetate, 8:2). 1H
NMR (400 MHz, CDCl3): δ = 9.53 (d, J = 4.4 Hz, 1 H), 7.88 (t, J
= 7.8 Hz, 1 H), 7.35–7.24 (m, 4 H), 3.87–3.84 (m, 4 H), 3.50 (s, 3
H), 3.42 (dd, J = 7.5, 4.4 Hz, 1 H) ppm. Conversions were deter-
mined with 1H NMR spectroscopy by using the doublet at δ =
9.53 ppm of the product and the doublet at δ = 9.74 ppm of the
starting aldehyde. The racemic product was synthesized under the
same conditions with racemic Jørgensen–Hayashi’s catalysts 2a.
The enantiomeric excess was determined by chiral HPLC after re-
duction of the aldehyde to alcohol with NaBH4 in MeOH. Separa-
tion conditions in chiral HPLC: IC 70:30 n-Hex/IPA, 0.5 mL/min,
40 °C, λ = 230 nm; tr = 14.4 min (minor), tr = 20.2 min (major).
Jørgensen–Hayashi’s catalyst 2a was by far the most
active one and afforded complete conversion to desired
product 11 in less than 1 h both with or without benzoic
acid. However, under both conditions tested, MacMillan’s
imidazolidinone catalysts 18 did not afford any traces of 11,
even after 48 h. As expected, an intermediate reactivity was
displayed by catalyst 1a, which afforded 48% yields of 11
after 48 h, with and without benzoic acid and with the same
values of enantioselectivity obtained by using 2a (ee 93%
without benzoic acid and ee 95% with benzoic acid as the
co-catalyst).
Conclusions
Organocatalytic Cyclopropanation Reaction of (E)-4-Methoxy-
cinnamaldehyde 3c with Dimethyl α-Bromomalonate 4: (Table 1, En-
try 9 and 5c). Following the procedure reported for 5a, 5c was iso-
lated by flash chromatography on silica gel (cyclohexane/ethyl acet-
Catalyst 1a was rationally designed with the aim to ob-
tain a more stable analogue under hydrolytic conditions of
Jørgensen–Hayashi’s catalyst 2a, with which it shares a
strict structural resemblance. From studies that explored its
1
ate, 8:2). H NMR (400 MHz, CDCl3): δ = 9.48 (d, J = 4.5 Hz, 1
reactivity in classic iminium/enamine activation modes, we H), 7.16 (d, J = 8.4 Hz, 2 H), 6.84 (d, J = 8.3 Hz, 2 H), 3.85 (s, 3
H), 3.79–3.81 (m, 4 H), 3.51 (s, 3 H), 3.35 (dd, J = 7.5, 4.8 Hz, 1
H) ppm. Conversions were determined with 1H NMR spectroscopy
by using the doublet at δ = 9.48 ppm of the product and the doublet
at δ = 9.67 ppm of the starting aldehyde. The racemic product was
synthesized under the same conditions with racemic Jørgensen–
Hayashi’s catalysts 2a. The enantiomeric excess was determined by
chiral HPLC after reduction of the aldehyde to alcohol with
NaBH4 in MeOH. Separation conditions in chiral HPLC: AD
90:10 n-Hex/IPA for 10 min, then 80:20 in 15 min, 0.5 mL/min,
40 °C, λ = 230 nm; tr = 23.3 min (major), tr = 26.0 min (minor).
noticed remarkably different behavior of the two catalysts,
mainly owing to the higher steric encumbrance of 1a, which
results from its conformational rigidity.
A careful study of some recently reported organocata-
lyzed domino reactions, together with a more detailed
structural characterization, allowed us to determine that 1a
is highly efficient in iminium chemistry, but a less powerful
enamine catalyst.
Although structurally similar to 2a, its reactivity seems
to be more similar to the first generation MacMillan’s imid-
azolidinone catalyst, able to work efficiently in iminium
chemistry, but not completely unreactive when used in en-
amine activation modes. Investigations to explore the reac-
tivity of 1a in transformations in which MacMillan’s cata-
lysts offer the best performances are underway and will be
reported in due course.
Low-Loading Organocatalytic Cyclopropanation Reaction of (E)-4-
Nitrocinnamaldehyde 3a with Dimethyl α-Bromomalonate 4:
(Table 3). To avoid excessive weighting errors, stocks solutions of
reagents were freshly prepared before each run and mixed by using
volumetric syringes. Solutions (0.01 m) of each catalyst were pre-
pared by dissolving 1a or 2a (3.3 mg) with CH2Cl2 in a 1 mL volu-
metric flask (solutions A1 and A2). A solution of 2,6-lutidine (1 m)
was prepared by dissolving the reagent (0.35 mL) with CH2Cl2 in
a volumetric flask (3 mL; solution B). A solution of 2-bromo-di-
methylmalonate 4 (1 m) was prepared by dissolving the reagent
(0.44 mL) with CH2Cl2 in a volumetric flask (3 mL; solution C). To
3a (0.1 mmol, 17.7 mg) in CH2Cl2 (0.16 mL) were added solution C
Experimental Section
Organocatalytic Cyclopropanation Reaction of (E)-4-Nitrocinnam-
aldehyde 3a with Dimethyl α-Bromomalonate 4: (Table 1, Entry 4
6
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
Conformationally Constrained Bicyclic Diarylprolinol Silyl Ethers
(0.11 mL), solution B (0.12 mL) and solution A1 or A2 (0.01 mL)
and the reactions were stirred at room temperature and checked
for conversion at the reported time intervals. Conversions were de-
termined with H NMR spectroscopy by using the doublet at δ =
9.59 ppm of the product and the doublet at δ = 9.79 ppm of the
[1]
a) H. Jiang, L. Albrecht, G. Dickmeiss, K. L. Jensen, K. A.
Jørgensen, in: Comprehensive Enantioselective Organocatalysis,
vol. 1 (Ed.: P. I. Dalko), Wiley-VCH, Weinheim, Germany,
2013, p. 33–50; b) H. Gotoh, Y. Hayashi, in: Sustainable Catal-
ysis (Eds.: P. J. Dunn, K. K. Hii, M. J. Krische, M. T. Wil-
liams), John Wiley & Sons, New Jersey, 2013, p. 287–316.
H. Kotsuki, N. Sasakura, in: Comprehensive Enantioselective
1
starting aldehyde.
Organocatalytic Iminium/enamine Cascade Reaction of N-Boc-3-(2-
oxopropyl)oxindole 6 with (E)-Cinnamaldehyde 3b. General Pro-
cedure: The catalyst (1a or 2a; 3.3 mg, 0.01 mmol, 10%) and ben-
zoic acid (0.6 mg, 0.01 mmol, 10%) were added to a solution of 3-
substituted oxindole 6 (29 mg, 0.1 mmol) and (E)-cinnamaldehyde
3b (0.0126 mL, 0.1 mmol, 1 equiv.) in toluene (0.4 mL, 0.25 m). The
reaction was allowed to stir at room temperature and the conver-
[2]
[3]
[4]
Organocatalysis, vol.
1 (Ed.: P. I. Dalko), Wiley-VCH,
Weinheim, Germany, 2013, p. 3–31.
R. Mahrwald, in: Comprehensive Enantioselective Organocata-
lysis, vol. 1 (Ed.: P. I. Dalko), Wiley-VCH, Weinheim, Ger-
many, 2013, p. 69–95.
a) D. Gryko, D. Walaszek, in: Stereoselective Organocatalysis
(Ed.: R. Rios Torres), John Wiley & Sons, New Jersey, 2013,
p. 81–127; b) G. Guillena, in: Modern Methods in Stereoselec-
tive Aldol Reactions (Ed.: R. Mahrwald), Wiley-VCH,
Weinheim, Germany, 2013, p. 155–268.
a) X.-H. Cai, G. Huo, X. Bing, Eur. J. Org. Chem. 2012, 3,
258–266; b) M. Benohoud, Y. Hayashi, in: Science of Synthesis
Asymmetric Organocatalysis, vol. 1 (Eds.: B. List, K. Ma-
ruoka), Thieme, Stuttgart, 2013, p. 73–134.
Y. Zhang, W. Wang, in: Stereoselective Organocatalysis (Ed.:
R. Rios Torres), John Wiley & Sons, New Jersey, 2013, p. 147–
204.
U. Scheffler, R. Mahrwald, Chem. Eur. J. 2013, 19, 14346–
14396.
a) C. M. R. Volla, I. Atodiresei, M. Rueping, Chem. Rev. 2014,
114, 2390–2431; b) H. Pellissier, in: Asymmetric Domino Reac-
tions, RSC Publishing, 2013, 354, p. 251–467.
a) M. Lombardo, S. Easwar, F. Pasi, C. Trombini, Adv. Synth.
Catal. 2009, 351, 276–282; b) E. Montroni, S. P. Sanap, M.
Lombardo, A. Quintavalla, C. Trombini, D. D. Dhavale, Adv.
Synth. Catal. 2011, 353, 3234–3240; c) E. Montroni, M. Lom-
bardo, A. Quintavalla, C. Trombini, M. Gruttadauria, F. Gia-
calone, ChemCatChem 2012, 4, 1000–1006; d) M. Lombardo,
M. Chiarucci, A. Quintavalla, C. Trombini, Adv. Synth. Catal.
2009, 351, 2801–2806.
1
sions were determined by H NMR spectroscopy. When (E)-β-ni-
trostyrene was completely consumed, the reaction mixture was
quenched at 0 °C with HCl (1 m, 2 mL) and the aqueous phase was
extracted with ethyl acetate. The combined organic phases were
dried (Na2SO4) and the solvents evaporated under vacuum. The
product was purified by flash chromatography on silica by eluting
with cyclohexane/ethyl acetate mixtures. 1H NMR (400 MHz,
CDCl3): δ = 10.07 (s, 1 H), 7.73 (d, J = 8.3 Hz, 1 H), 7.12–7.09 (m,
4 H), 6.78–6.76 (m, 2 H), 6.72–6.69 (m, 1 H), 6.20 (d, J = 7.6 Hz,
1 H), 4.60 (s, 1 H), 3.04–2.99 (m, 2 H), 2.42 (s, 3 H), 1.66 (s, 9
H) ppm. The racemic product was synthesized under the same con-
ditions with racemic Jørgensen–Hayashi’s catalysts 2a. The enan-
tiomeric excess was determined by chiral HPLC: IC 80:20 n-Hex/
IPA, 1.0 mL/min, 40 °C, λ = 214 nm; tr = 19.8 min (minor), tr =
39.8 min (major).
[5]
[6]
[7]
[8]
[9]
Organocatalytic Michael Addition of Propanal 9 to β-(E)-Nitrostyr-
ene 10. General Procedure: Propanal (0.0144 mL, 0.2 mmol,
2 equiv.) was added to a solution of catalyst (1a, 2a or 18,
0.01 mmol, 10%), β-(E)-nitrostyrene 10 (14.9 mg, 0.1 mmol) and
eventually benzoic acid (0.6 mg, 0.01 mmol, 10%) in CH2Cl2
(0.4 mL, 0.25 m). The reaction was allowed to stir at room tempera-
ture and the conversions were determined by 1H NMR spec-
troscopy. When the starting materials were completely consumed,
the reaction mixture was quenched with HCl (1 m, 2 mL) and the
aqueous phase was extracted with ethyl acetate. The combined or-
ganic phases were dried (Na2SO4) and the solvents evaporated un-
der vacuum. The product was purified by flash chromatography on
[10]
[11]
M. Lombardo, E. Montroni, A. Quintavalla, C. Trombini, Adv.
Synth. Catal. 2012, 354, 3428–3434.
a) K. L. Jensen, G. Dickmeiss, H. Jiang, L. Albrecht, K. A.
Jørgensen, Acc. Chem. Res. 2012, 45, 248–264; b) M. Nielsen,
D. Worgull, T. Zweifel, B. Gschwend, S. Bertelsen, K. A.
Jørgensen, Chem. Commun. 2011, 47, 632–649.
1
[12]
[13]
a) H. Xie, L. Zu, H. Li, J. Wang, W. Wang, J. Am. Chem. Soc.
2007, 129, 10886–10894; b) R. Rios, H. Sundén, J. Vesely, G.-
L. Zhao, P. Dziedzic, A. Córdova, Adv. Synth. Catal. 2007,
349, 1028–1032; c) I. Ibrahem, G. L. Zhao, R. Rios, J. Vesely,
H. Sundén, P. Dziedzic, A. Córdova, Chem. Eur. J. 2008, 14,
7867–7879.
a) Y. Hayashi, T. Itoh, H. Ishikawa, Angew. Chem. Int. Ed.
2011, 50, 3920–3924; Angew. Chem. 2011, 123, 4006–4010; b)
S. K. Ghosh, Z. Zheng, B. Ni, Adv. Synth. Catal. 2010, 352,
2378–2382; c) I. Mager, K. Zeitler, Org. Lett. 2010, 12, 1480–
1483; d) Y. Wang, P. Li, X. Liang, T. Y. Zhang, J. Ye, Chem.
Commun. 2008, 1232–1234; e) L. Zu, H. Xie, H. Li, J. Wang,
W. Wang, Adv. Synth. Catal. 2007, 349, 2660–2664; f) C. Pal-
omo, A. Landa, A. Mielgo, M. Oiarbide, A. Puente, S. Vera,
Angew. Chem. Int. Ed. 2007, 46, 8431–8435; Angew. Chem.
2007, 119, 8583–8587.
Z.-J. Jia, H. Jiang, J.-L. Li, B. Gschwend, Q.-Z. Li, X. Yin, J.
Grouleff, Y.-C. Chen, K. A. Jørgensen, J. Am. Chem. Soc.
2011, 133, 5053–5061.
M. H. Haindl, M. B. Schmid, K. Zeitler, R. M. Gschwind,
RSC Adv. 2012, 2, 5941–5943.
F. Giacalone, M. Gruttadauria, P. Agrigento, R. Noto, Chem.
Soc. Rev. 2012, 41, 2406–2447.
K. Albertshofer, K. E. Anderson, C. F. Barbas III, Org. Lett.
2012, 14, 5968–5971.
silica by eluting with cyclohexane/ethyl acetate mixtures. H NMR
(400 MHz, CDCl3): δ = 9.72 (d, J = 1.7 Hz, 1 H), 7.37–7.27 (m, 3
H), 7.18–7.13 (m, 2 H), 4.80 (dd, J = 5.5/12.7 Hz, 1 H), 4.68 (dd,
J = 9.3/12.7 Hz, 1 H), 3.81 (dt, J = 5.5/9.1 Hz, 1 H), 2.90–2.69 (m,
1 H), 1.00 (d, J = 7.3 Hz, 3 H) ppm. The racemic product was
synthesized under the same conditions with racemic Jørgensen–
Hayashi’s catalysts 2a. The enantiomeric excess was determined by
chiral HPLC: IC 90:10 n-Hex/IPA, 1.0 mL/min, 40 °C, λ = 214 nm;
tr = 15.1 min (anti, minor), tr = 22.0 min (syn, minor), tr = 25.6 min
(syn, major), tr = 27.4 min (anti, major).
Supporting Information (see footnote on the first page of this arti-
cle): M06-2x/6-311G(d,p)-optimized coordinates of the most stable
conformers of s-trans-(E)-configured enamines 13, 14 and 17; cop-
ies of 1H and 13C NMR of enamines 13 and 14; chiral HPLC traces
of tested reactions.
[14]
Acknowledgments
[15]
[16]
[17]
Financial support by the University of Bologna, the Ministero
dell’Università e della Ricerca (MIUR) (Rome, PRIN 2009) and
the European Union, COST Action (CM0905, Organocatalysis) is
acknowledged.
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
M. Lombardo et al.
FULL PAPER
[18] W. G. Rajeswaran, L. A. Cohen, Tetrahedron 1998, 54, 11375–
11380.
Mayr, S. Lakhdar, B. Maji, A. R. Ofial, Beilstein J. Org. Chem.
2012, 8, 1458–1478.
[19] a) K. Patora-Komisarska, M. Benohoud, H. Ishikawaa, D.
Seebach, Y. Hayashi, Helv. Chim. Acta 2011, 94, 719–745; b)
D. Seebach, X. Sun, M.-O. Ebert, W. B. Schweizer, N. Purkaya-
stha, A. K. Beck, J. Duschmalé, H. Wennemers, T. Mukai-
yama, M. Benohoud, Y. Hayashi, M. Reiher, Helv. Chim. Acta
2013, 96, 799–852.
[20] The apparent pKa values were calculated under standard condi-
tions (25 °C and zero ionic strength) in aqueous solutions for
every ionizable group within organic structures by using the
ACD/pKa DB, version 12.00, software (Advanced Chemistry
Development Inc., Toronto, Canada, www.acdlabs.com).
[21] a) Y. Hayashi, H. Gotoh, T. Hayasi, M. Shoji, Angew. Chem.
Int. Ed. 2005, 44, 4212; Angew. Chem. 2005, 117, 4284; b) Y.
Hayashi, T. Okano, S. Aratake, D. Hazelard, Angew. Chem. Int.
Ed. 2007, 46, 4922; Angew. Chem. 2007, 119, 5010; c) S. Zhu,
S. Yu, D. Ma, S. Zhu, S. Yu, D. Ma, Angew. Chem. Int. Ed.
2008, 47, 545–548; Angew. Chem. 2008, 120, 555; d) J. Burés,
A. Armstrong, D. G. Blackmond, J. Am. Chem. Soc. 2011, 133,
8822–8825.
[25]
[26]
J. D. Dunitz, X-ray Analysis and the Structure of Organic Mol-
ecules; Cornell University Press, London, 1979.
a) T. P. Radhakrishnan, I. Agranat, Struct. Chem. 1991, 2, 107–
115; b) S. P. Gavrish, J. Comput. Chem. 2012, 33, 2173–2179.
[27] It should be noted that although Dunitz pyramidalization fac-
tor is one of the most used crystallographic index for describing
the distortion from planarity of a trivalent atom, its value de-
pends on the lengths of the three bonds incident at the central
atom and thus can give different values depending the nature
of the groups bounded to it. This is not completely correct
because, in principle, pyramidalization should be considered
simply as the tendency of an atom to deviate from planarity.
The nitrogen sp2 character of enamine 13 relative to enamine
14 and 17 was thus evaluated also by using π-orbital axis vector
(POAV) methodology, which considers only the angles formed
by δ-orbitals with the central atom π-orbital. a) R. C. Haddon,
J. Am. Chem. Soc. 1986, 108, 2837–2842; b) R. C. Haddon,
Acc. Chem. Res. 1988, 21, 243–249; c) R. C. Haddon, J. Am.
Chem. Soc. 1990, 112, 3385–3389. We determined a pyrami-
dalization angle θp = 0.93° for 14, which corresponds to an
index of hybridization n = 2.00 (spn), whereas θp = 5.67° with
n = 2.06 for 13 and θp = 8.25° with n = 2.13 for 17.
[22] M. B. Schmid, K. Zeitler, R. M. Gschwind, Chem. Sci. 2011,
2, 1793–1803.
[23] U. Groseˇlj, D. Seebach, D. M. Badine, W. B. Schweizer, A. K.
Beck, I. Krossing, P. Klose, Y. Hayashi, T. Uchimaru, Helv.
Chim. Acta 2009, 92, 1225–1259.
[24] a) S. Lakhdar, B. Maji, H. Mayr, Angew. Chem. Int. Ed. 2012,
51, 5739–5742; Angew. Chem. 2012, 124, 5837–5840; b) H.
Received: June 10, 2014
Published Online: ᭿
8
www.eurjoc.org
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 0000, 0–0

Job/Unit: O42732
/KAP1
Date: 19-08-14 11:09:02
Pages: 9
Conformationally Constrained Bicyclic Diarylprolinol Silyl Ethers
Organocatalysis
Bicyclic silyl ether derivative 1a, which is
easily prepared from trans-4-l-hydroxypro-
line, is much more stable toward protodesi-
lylation than Jørgensen–Hayashi’s catalyst
2a. Despite their structural analogy, 1a be-
haves as a very efficient catalyst in iminium
chemistry, but is much less active than 2a
in the enamine activation mode. Here, 1a
exhibits a higher or equivalent level of ster-
eocontrol.
M. Lombardo,* L. Cerisoli, E. Manoni,
E. Montroni, A. Quintavalla,
C. Trombini ....................................... 1–9
Properties and Reactivity of Confor-
mationally Constrained Bicyclic Diarylpro-
linol Silyl Ethers as Organocatalysts
Keywords: Synthetic methods / Organocat-
alysis / Low-loading catalysis / Nitrogen
heterocycles
Eur. J. Org. Chem. 0000, 0–0
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
9