Silicone-supported Cinchona alkaloid derivatives as insoluble organocatalysts in the enantioselective dimerization of ketenes

Article abstract of DOI:10.1002/ejoc.201101320

A straightforward procedure is presented for the covalentimmobilization of ester and silyl ether derivatives of theCinchona alkaloid 10,11-dihydroquinidine within insoluble cross-linked silicone elastomeric films. These materials were effective heterogeneous organocatalysts in the asymmetric dimerization of ketenes, which provided chiral Weinreb β-ketoamides in 28-83 % yield and 79-99 % ee in the course of several recycles. A productivity/enantioselectivity protocol is also proposed to better assess the relative merits of soluble and supported asymmetric catalytic systems towards process intensification. A series of hydroquinidine derivatives embedded into elastomeric silicone films was prepared and used as insoluble organocatalysts in the asymmetric dimerization of ketenes. A protocol was introduced to compare the productivity/ enantioselectivity of the supported catalysts with those of their analogous soluble counterparts. Copyright

Full text of DOI:10.1002/ejoc.201101320

FULL PAPER
DOI: 10.1002/ejoc.201101320
Silicone-Supported Cinchona Alkaloid Derivatives as Insoluble Organocatalysts
in the Enantioselective Dimerization of Ketenes
[a]
[a]
[a]
[a]
Damiano Cancogni, Alessandro Mandoli,* Ravindra P. Jumde, and Dario Pini
Keywords: Asymmetric catalysis / Organocatalysis / Alkaloids / Ketene dimerization / Silicones / Supported catalysts
A straightforward procedure is presented for the covalent
immobilization of ester and silyl ether derivatives of the
Cinchona alkaloid 10,11-dihydroquinidine within insoluble
cross-linked silicone elastomeric films. These materials were
effective heterogeneous organocatalysts in the asymmetric
dimerization of ketenes, which provided chiral Weinreb β-
ketoamides in 28–83% yield and 79–99% ee in the course of
several recycles. A productivity/enantioselectivity protocol is
also proposed to better assess the relative merits of soluble
and supported asymmetric catalytic systems towards process
intensification.
Introduction
fined as the mol of product obtained per mol of catalyst)
[
3]
and the decrease of ee. Provided that some reduction of
the stereochemical purity of the product can be tolerated at
the synthesis stage, the use of diminished amounts of the
soluble system may be considered an equally valid option
for process intensification, as opposed to the development
of supported analogues. Under these circumstances, it is
clear that to improve its credibility proper comparison pro-
tocols have to be implemented and thoroughly used in this
kind of study.
From the early 70s, the increasing use of homogeneous
enantioselective catalysis has been accompanied by the de-
velopment of covalently immobilized analogues.[1] Stem-
ming from the exploration of alternative anchoring strate-
gies, chiral derivatives, and support materials, a large
number of effective insoluble polymer-bound (IPB) cata-
lysts and ligands are known to convincingly demonstrate
the viability of the approach at the proof-of-concept level
and clarify some important design elements required for the
attainment of high catalytic activity and enantioselectivity.
Nonetheless, the hesitation to adopt this strategy for
practical-scale applications is still strong.[ This reluctance
is partially dictated by cultural motivations, which is sus-
The general considerations stated above hold true when
the ligands and organocatalysts derived from the Cinchona
[4]
alkaloids 1–4 (Figure 1) are considered. By ingenious ex-
1b]
ploitation of the polyfunctional nature of the alkaloid core
and alternative immobilization techniques, a large array of
tained by the lack of evidence for decisive practical advan-
[1a–1h,3,5]
IPB alkaloid architectures have been reported,
tages for supported catalysts.[
1h,2]
A common concern for
some of which are summarized in Figure 2. However, with
a few exceptions where the native alkaloids (or simple deriv-
most of the disclosed IPB systems regards their prepara-
tion, which often appears too complex or poorly efficient
for realistic scale up. Even when this is not the case, the
limited efforts put into studying catalyst recycling can be
another problem that hampers a clear-cut answer to the
question of whether the repeated use of IPB systems can
actually compensate for their generally onerous obtain-
ment.
atives thereof) were effectively coupled to commercially
[5d–5f]
available functional resins (e.g. 5 and 6),
the prepara-
tion of most of the materials appears impractical. Typically,
this is a consequence of lengthy reaction schemes and re-
peated purifications in the synthesis of anchorable alkaloid
To further complicate this judgment, within some limits,
the repeated use of an IPB catalyst may have an overall
outcome analogous to the reduction of the loading of a
homogeneous one, i.e. the increase of productivity (P, de-
[
a] Dipartimento di Chimica e Chimica Industriale, Università di
Pisa, and CNR – ICCOM (Section of Pisa),
Via Risorgimento 35, 56126 Pisa, Italy
Fax: +39-050-221-9260
E-mail: alexm@dcci.unipi.it
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201101320.
Figure 1. Structures, acronyms, and numbering scheme for 1–4.
1336
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1336–1345

Silicone-Supported Organocatalysts
Figure 2. Typical anchoring positions for Cinchona alkaloid derivatives and selected IPB architectures (AlkO = QD, QN, DHQD, or
DHQN).
derivatives (e.g. 10)[3] or the low immobilization yield and in the heterogeneous asymmetric dimerization of ketenes.
limited alkaloid loading attained in the immobilization step In the course of this study, the problem of the comparison
[
5a,5c]
(
e.g. 8 and 9).
As pointed out by Kacprzak and of IPB systems with their soluble counterparts was also ad-
[
6]
Lindner, the development of IPB alkaloid derivatives for dressed, which led to the results presented in the following
practical uses requires a stronger emphasis on the efficiency sections.
of their preparation.
With this aim, we have recently developed a new protocol
Results and Discussion
for access to dimeric alkaloid derivatives linked to gel-type
or macroporous polystyrene supports.[
5g]
The strategy
Preparation of Anchorable DHQD Derivatives
proved highly effective in many respects, which included a
scalable preparation of the anchorable derivatives and their
Cinchona alkaloids have been coupled to linear poly-
efficient coupling to preformed resins in a click-chemistry (methylhydrosiloxane) (PMHS) by the groups of Siegel and
[
10]
step. Moreover, the resulting materials 11 were remarkable Bergbreiter.
organocatalysts in the asymmetric dimerization of ketenes
However, the main goal of these investi-
gations was the preparation of soluble materials, some of
and afforded chiral Weinreb β-ketoamides in 90–97% ee which were obtained by direct hydrosilylation of the 10,11-
[
7]
over 20 reaction cycles.[
5g]
double bond of the alkaloid. By contrast, the cross-linked
Nonetheless, the relatively complex structure of 11 does structure of the IPB materials explored here suggested the
not appear to be required for the aforementioned reaction introduction of a spacer group between the chiral moiety
because simpler monomeric 9-O alkaloid derivatives, such and the terminal alkene anchoring site. Similar to Siegel
[
10a]
as the propionate esters (Prop) and the trimethylsilyl (TMS) and coworkers,
or tert-butyldimethylsilyl (TBDMS) ethers are equally effec- suggested that it should be embedded into the 9-O substitu-
considerations of synthetic efficiency
tive organocatalysts.[
7a,7b,7f]
In addition, whilst polystyrene ent. This led to the design of the three DHQD derivatives
resins provide the IPB systems in the form of reasonably 16, 17, and 18, which can be considered as the anchorable
robust beads, in the case of immobilized bisoxazoline (Box) variants of the soluble Prop, TMS, and TBDMS organocat-
ligands we and the group of Nagashima[9] have recently alysts.
[
8]
found that the switch to cross-linked silicone supports pres-
The synthesis of the silyl ethers 17 and 18 required the
ents advantages in terms of handling and catalytic perform- initial preparation of the corresponding chlorosilanes 13
ance.
and 15, achieved as depicted in Scheme 1 by the reaction of
These considerations prompted us to investigate the co- dichlorodimethylsilane with the Grignard reagents obtained
[
11]
valent immobilization of monomeric 9-O derivatives of from the halides 12 or 14.
0,11-dihydroquinidine (DHQD-H, 3) within insoluble chemicals, a practical advantage of this route was that pure
elastomeric silicone films and their use as organocatalysts enough 13, 14, and 15 could be obtained in acceptable (un-
In addition to using cheap
1
Eur. J. Org. Chem. 2012, 1336–1345
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1337

D. Cancogni, A. Mandoli, R. P. Jumde, D. Pini
FULL PAPER
optimized) yields by a simple and scalable procedure, which Preparation of DHQD Derivatives in Cross-Linked Silicone
involved high-vacuum distillation as the purification tech- Elastomeric Films
nique.
Previous work on the immobilization of chelating Box
ligands within elastomeric silicone films demonstrated the
advantage of using highly active nanostructured solvated Pt
[
8]
catalysts in the hydrosilylation step. However, the reduced
coordination ability of 16–18 and literature precedents[
suggested that commercial Pt catalysts could also fit the
needs of this work.
10a]
To test this hypothesis, 17 was reacted in C D6 with
6
PMHS (x = y ca. 6, Scheme 3) in the presence of a small
amount of the commercial Karstedt catalyst (Si–H:17:Pt =
1
13
1
000:200:1). After 5 h at 50 °C, H and C NMR spec-
troscopy indicated complete conversion, which was evi-
denced by the disappearance of the olefin and allyl signals
of the side chain of the alkaloid. In addition, the spectra of
the final solution showed neither degradation of the alka-
loid core nor significant isomerization of the terminal
double bond of 17 to the hydrosilylation-refractory internal
Scheme 1. Synthesis of 13 and 15. (i) Mg, Et
hydrofuran (THF), 50% from 12; (iii) acetone, Et
HCl, ZnCl , 50% from 12; (iv) Mg, THF, 70 °C, then Me
cat. CuCN, THF, 43% from 14.
2
O; (ii) Me
O, then 37%
2
SiCl ,
2 2
SiCl , tetra-
2
2
2
Next, the synthesis of the anchorable alkaloid derivatives
was carried out (Scheme 2). For the preparation of the 9-O
undecenoic ester 16, a modification of the reported pro-
cedure for the analogous Prop derivative was used,[ which
consisted of the reaction of 3 with commercially available
[
13]
positions.
The covalent linkage of the chiral unit to
PMHS was also confirmed by the growth of two broad res-
7a]
1
1
1
0-undecenoyl chloride without additional base. Similarly,
7 and 18 were prepared by the reaction of 3 with 13 and
5 under the conditions reported for the synthesis of the
[7a,12]
corresponding TMS and TBDMS ethers.
After chro-
matographic purification, fractions of 16, 17, and 18 were
obtained in fair to good yields as clear oils that gave a single
spot by TLC. The identities of the three alkaloid derivatives
1
were confirmed by electrospray mass spectrometry and H
1
3
and C NMR spectroscopy. In the case of 16, the spectro-
[
10a]
scopic constants were identical to those published,
which included the occurrence of single sets of sharp reso-
1
13
nances in the H and C NMR spectra. On the contrary,
the spectra of 18 were more complicated and showed two
distinct sets of sharp resonances (with a ratio of ca. 76:24)
1
13
in the H and C NMR spectra. Given the sample homo-
geneity, these observations are best explained as a conse-
9
quence of the restricted rotation around the C –O and O–
Si bonds of hindered 18. Interestingly, similar conclusions
have been reported for other Cinchona alkaloid TBDMS
silyl ethers, which showed a ca. 75:25 ratio of rotamers by
1
[12]
H NMR spectroscopy.
Scheme 3. Preparation of P1–P5.
Scheme 2. Synthesis of 16, 17, and 18. Reagents and conditions: (i) 1.3 equiv. 10-undecenoyl chloride, CH
2
Cl
N, 30 mol-% 4-(dimethylamino)pyridine (DMAP), N,N-dimethylformamide (DMF), r.t., 48 h, 62%; (iii)
N, 10 mol-% DMAP, DMF, r.t., 72 h, 53%.
2
, r.t., 48 h, 75%; (ii)
1
.3 equiv. 13, 2 equiv. Et
.3 equiv. 15, 4.5 equiv. Et
3
1
3
1338
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1336–1345

Silicone-Supported Organocatalysts
onances at 23.4 and 17.9 ppm in the 1 C NMR spectrum, Table 1. Feed composition and characterization data for P1–P5.
3
which, by comparison with literature values,[ identified
14]
Alkaloid
derivative
PMHS
[x/y]
Composi-
tion
Elasto-
meric film [mmolg ]
Film loading
the CH CH Si fragment of the polymer-bound alkaloid.
[a]
–1 [b]
2
2
Although these initial experiments proved useful to vali-
date the chemistry involved in the anchoring procedure, ge-
1
6
6
6:6
6:6
0:26
6:6
3:1:1 (25)
5:1:2 (30)
3:1:1 (35)
3:1:1 (25)
3:1:1 (25)
P1
P2
P3
P4
P5
0.28Ϯ0.007
0.19Ϯ0.004
0.57Ϯ0.004
0.24Ϯ0.007
0.26Ϯ0.04
1
lation and some gas evolution was noticed in the C D sam- 16
6
6
1
7
8
ple shortly after the NMR spectroscopic measurements.
The poor solubility of the gelled material hampered a de-
tailed investigation of the phenomenon, which was likely a
result of side reactions of the excess Si–H units of PMHS
1
6:6
[
a] Molar ratio Si–H/alkaloid derivative/PDMS-DV lw (wt.-% of
PDMS-DV hw in the feed in parentheses). [b] Alkaloid content
from nitrogen elemental analysis (average of two determinations).
[
13,15]
after the complete conversion of the vinyl groups of 17.
Because the observed phase separation made the prepa-
ration of uniform elastomeric films by the previously devel-
oped two-step approach unfeasible,[ an investigation of a
single-step procedure was performed. Toluene solutions of
mass recovery of purified insoluble materials, which was
much lower for P3 than for the other cross-linked films (22
vs. 74–90% of the feed weight, respectively). Together with
the trends in the degree of functionalization, these results
affected the absolute immobilization yield of 16, 17, or 18,
which was 28% in the case of P3 and 45–56% for P1, P2,
P4, and P5.
8]
16, 17, or 18, PMHS, low and high molecular-weight divi-
nyl-terminated silicones (PDMS-DV lw and hw, respec-
[
8]
tively), and the Karstedt catalyst were gently heated over-
night under air in a PTFE-lined vessel (Scheme 3).
Evaporation of the solvent resulted in the casting of ro-
bust cross-linked films, which were swollen with an excess
of 1-hexadecene in toluene and heated again to cap any
residual Si–H group. Continuous extraction with THF and
CH Cl and drying under vacuum afforded P1–P5
Homogeneous Asymmetric Ketene Dimerization
To test the suitability of the new chiral derivatives of this
2
2
(Table 1), which were characterized by IR spectroscopy work in the asymmetric dimerization of ketenes, homogen-
(Supporting Information) and microanalytical determi- eous catalysis experiments were initially carried out. With
nation of the nitrogen content. IR spectroscopy ruled out this aim, 16, 17, and 18 were employed under the conditions
the presence of significant amounts of unreacted Si–H and described by Calter and coworkers for the in situ generation
C=C units, which was demonstrated by a lack of appreci- of ketenes from the acid chlorides 19a–c and one-pot open-
–
1
able absorptions at 2280–2400 and ca. 1600 cm , respec- ing of the intermediate β-lactone dimers 20a–c to the corre-
[
16]
tively.
of the ester derivative in the elastomeric network was also mation).
confirmed by the presence of a weak carbonyl stretch at In the dimerization of methylketene from 19a (Table 2,
In the case of P1, P2, and P3, the incorporation sponding Weinreb amides 21a–c (Supporting Infor-
[
7f]
–
1
around 1748 cm . IR spectroscopy proved less informative Entries 1, 4, and 5), all three modified catalysts, 16, 17, and
for P4 and P5 due to the lack of strong bands of the alka- 18, appeared to be capable of providing 21a with 94–98%
loid and side chain fragments in the spectral regions not ee, albeit in slightly reduced yields (56–64%) with respect
obscured by the polysiloxane backbone. Nonetheless, nitro- to those published (65–79%). Interestingly, in the case of
gen elemental analysis demonstrated the substantial incor- 17 the enantioselectivity level matched that reported for the
poration of the chiral derivative for all of the prepared ma- corresponding TMS ether of quinidine (TMS-1), which sug-
[
7f]
terials with alkaloid contents ranging from 0.19– gests that neither the introduction of the longer side chain
.57 mmolg– (Table 1). The comparison of the experimen- in the silyl group nor the saturation of the alkaloid 10,11-
1
0
tal loadings with the values calculated from the feed com- double bond has a major impact on the stereoselectivity.
positions revealed that the degree of functionalization of A similar comparison for the other two organocatalysts
P1–P5 was more sensitive to the Si–H content of the PMHS is less straightforward because published data refer to the
precursor and Si–H/alkaloid ratio than to the nature of the use of Prop or TBDMS derivatives of the quinine pseudo-
chiral derivative. Indeed, the use of different soluble precur- enantiomer 2, which afford 69 and 94% ee, respectively, in
[
7f]
sors (16, 17, or 18) but the same Si–H/alkaloid molar ratio the dimerization of the ketene from 19a.
3:1) and PMHS copolymer (x = y ≈ 6) in the preparation Nonetheless, 16 and 18 provided 94 and 98% ee, respec-
of P1, P4, and P5 resulted in similar anchoring levels (60, tively, in the same reaction (Table 2, Entries 1 and 5). In
5, and 59% of the theoretical values, respectively). On the addition to the strong influence of the alkaloid peripheral
contrary, the increase of the Si–H/alkaloid ratio to 5:1 or positions, which includes the diastereomeric relationship
(
5
[
17]
the switch to pure methylhydrosiloxane homopolymer (x = between the chiral cores,
, y ≈ 26) caused a significant increase of the incorporation these findings was the possibility to attain synthetically use-
of the same chiral derivative 16 in P2 and P3 (76 and 129% ful ee values in the quinidine series by the use of the simple
of the theoretical values, respectively). ester derivative 16. This conclusion was corroborated by
the practical implication of
0
The attainment of a higher than expected loading for P3 two further runs with 16 in the dimerization of homologous
was due to the significantly reduced incorporation of polys- ketenes from 19b and 19c, which led to the corresponding
iloxane components in the film. This was confirmed by the Weinreb amides with 96–97% ee (Table 2, Entries 2 and 3).
Eur. J. Org. Chem. 2012, 1336–1345
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1339

D. Cancogni, A. Mandoli, R. P. Jumde, D. Pini
Table 2. Homogeneous and heterogeneous asymmetric catalytic dimerization of ketenes from acid chlorides.
FULL PAPER
Entry
Catalyst
Acid chloride
t
1
[h]
t
2
[h]
Product
Yield [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
9
16
16
19a
19b
19c
19a
19a
19a
19b
19c
19a
19a
19a
19b
19c
19a
19b
19c
6
6
24
6
6
6
6
24
6
6
6
6
24
6
6
2
2
24
2
2
2
2
24
2
2
2
2
24
2
2
21a
21b
21c
21a
21a
21a
21b
21c
21a
21a
21a
21b
21c
21a
21b
21c
57
83
52
56
64
58
88
79
60
64
53
69
67
57
62
50
94
96
97
97
98
94
96
95
95
93
94
99
99
98
95
96
16
17
18
P1
P1
P1
P2
P3
P4
P4
P4
P5
P5
P5
10
11
12
13
14
15
16
24
24
[a] After isolation by flash chromatography. [b] By HPLC with chiral stationary phases; the prevailing enantiomer had S configuration.
For a better comparison of the productivity of homogen- Heterogeneous Asymmetric Ketene Dimerization
eous and heterogeneous catalytic systems (vide infra), a
series of experiments was also carried out with variable
amounts of the soluble TMS ether 22 (Figure 3). The re-
sults of these tests revealed that the reaction of methylke-
tene could tolerate a twofold reduction of the catalyst load-
The heterogeneous catalysis experiments were carried out
with 5 mol-% of P1–P5. The conditions were the same as
described above for the homogeneous runs with the only
variation that, due to the lack of an active role of the alka-
loid units to open the chiral β-lactone intermediate,[ the
successive reaction steps were performed in two distinct ves-
7b]
ing below the literature conditions (5 mol-%)[7f] without any
major erosion of the ee of the product. However, a further
decrease of the amount of alkaloid resulted in a rapid de-
gradation of performance, with an abrupt decrease in iso-
lated yields and ee values when the loading was reduced
below 1 mol-%.
[
5g]
sels. As detailed elsewhere,
after stirring for the time t1,
the solution that contained the chiral dimer 20a–c was sepa-
rated from the organocatalyst film by siphoning, treated
with N,O-dimethylhydroxylamine and 2-pyridone, and
stirred for the time t to give the final product 21a–c. Stan-
2
dard filtration experiments were carried out concurrently
(Supporting Information), which ruled out a significant
homogeneous contribution to catalysis from species leached
into the solution.
Screening of P1–P5 in the reaction of propionyl chloride
(19a) showed that all of the silicone-supported derivatives
effectively promoted the asymmetric dimerization of meth-
ylketene (Table 2, Entries 6, 9–11, and 14). Despite the in-
soluble nature of the catalytic system, the isolated yields of
the chiral dipropionate product 21a proved satisfactory and
are comparable to those afforded by the soluble compounds
1
6–18 after identical t and t reaction times. Excellent re-
1 2
sults were also achieved in terms of enantioselectivity, with
ee values as high as 93–98%. As expected from the prelimi-
nary homogeneous runs, the stereochemical efficiency of
the supported organocatalysts was somewhat dependant on
the structure of the immobilized alkaloid derivative and
reached optimal results with TBDMS-like P5 (Table 2, En-
try 14). Different film compositions and catalyst loadings
appeared to have a minor impact on the catalytic properties,
Figure 3. Asymmetric dimerization of methylketene in the presence
of 22 (5 mol-% data are for TMS-2, from ref.[7f]; for the other con-
ditions see Table 2).
1340
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1336–1345

Silicone-Supported Organocatalysts
as proved by the rather uniform performances of the mate-
rials that contain the same undecenoyl ester derivative
within the variable polymer architecture of P1–P3 (Table 2,
Entries 6, 9, and 10).
The promising results of the asymmetric dimerization of
methylketene were confirmed by using P1, P4, and P5 in
the reaction of ketene homologues from 19b and 19c
(Table 2, Entries 7, 8, 12, 13, 15, and 16). In these cases,
acceptable yields were obtained under standard conditions
together with ee values that matched, or slightly surpassed,
those afforded by the soluble alkaloid derivatives in this
[
7f]
work (Table 2, Entries 2 and 3) or from the literature.
Overall, the covalent immobilization into elastomeric sili-
cone films proved capable of preserving the high enantio-
selectivity and satisfactory activity of 16–18 to a large ex-
tent. Therefore, the cross-linked polysiloxane network of
P1–P5 appears well suited for the heterogeneous catalysis
of the reaction under examination, with no appreciable im-
pact on the asymmetric induction ability of the supported
chiral units and the possibility for the species in solution to
access them. Concerning this latter point, even if the rate-
limiting step in the homogeneous reaction (ketene forma-
tion) is known to be independent of the alkaloid derivative
[
7f]
concentration, the attainment of dimerization products
with substantial yields and ee values requires the involve-
ment of the chiral organocatalyst. For this reason, the oc-
currence of critical diffusion problems in P1–P5 appears
unlikely.
Figure 4. (a) ee of the product 21a in the recycle runs of P1–P5
for the conditions, see Table 2; 21a was obtained in 28–65% iso-
(
lated yield). (b) Enantioselectivity vs. productivity for soluble (22)
and IPB (P1–P5) catalysts.
Catalyst Recycling and Productivity Evaluation
As already found with Box ligands,[8] the film shape of not preclude the attainment of 90–91% ee in the following
[
18]
P1–P5 allowed their prompt separation from the reaction three cycles and 79% ee in the final one.
mixture by siphoning. This feature greatly facilitated the Although the repeated use of P4 appears capable of im-
study of recycling, which was evaluated by selecting the re- proving the organocatalyst productivity over the analogous
action of 19a as the benchmark. homogeneous systems, the assessment of this aspect cannot
The experiments (Figure 4, a) demonstrated that each of be separated from the concurrent evaluation of the enantio-
the IPB systems could promote at least five consecutive di- selectivity attained under alternative conditions. For this
merization runs, which still led to 21a with substantial ee reason, the available data from the benchmark homogen-
values. Nonetheless, some significant differences were noted eous and heterogeneous methylketene dimerization runs
[
5g]
in the propensity of the diverse materials to retain their ini- were converted into P@ee value pairs
and, for the sake
tial catalytic efficiency on recycling. In particular, although of comparison, plotted in an enantioselectivity vs. produc-
the ester-type systems P1–P3 underwent a fast drop of per- tivity (ee-P) graph (Figure 4, b).
formances in successive runs (28–33% yield, 79–83% ee in
As far as the homogeneous reactions with 22 are con-
the fifth cycle; yields not shown in Figure 4), the silyl ether cerned, the P@ee points were directly calculated from the
materials P4 and P5 continued to provide nearly unchanged catalyst amount employed in each run and the recorded
results in this short reaction series (45–55% yield, 94–95% yield and ee of 21a (Figure 3). The outcome of this analysis,
ee in the fifth cycle).
shown as the broken curve in Figure 4 (b), revealed that the
Because the preparation of P4 required a more conve- use of 2.5 mol-% of 22 could afford a result (P ≈ 11@97%
nient soluble precursor than that of P5, it was selected for ee) that compares favorably with that attained under stan-
further testing. Eight more runs were performed with P4 to dard literature conditions (e.g. P ≈ 8@97% ee from the pub-
[
7f]
give the additional data included in Figure 4 (a). The results lished data with 5 mol-% of TMS-2). However, due to
of these experiments showed that the IPB organocatalyst the anticipated relationship between the amount of alkaloid
could be employed in thirteen successive cycles, and still and catalytic performance (Figure 3) this favorable trend
provide 21a in 45% isolated yield (data not shown) even in was not valid on moving to lower catalyst loadings. In this
the last run. Concerning the enantioselectivity, a moderate respect, in addition to the reduction in enantioselectivity,
negative trend was observed after the sixth cycle, which did the relatively large P achieved by the use of 0.1 mol-% of
Eur. J. Org. Chem. 2012, 1336–1345
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1341

D. Cancogni, A. Mandoli, R. P. Jumde, D. Pini
FULL PAPER
22 (P ≈ 35@55% ee) actually corresponded to the isolation cess on reduction of the catalyst loading. Because neither
of impractically small amounts of 21a (7% yield). Accord- of the two results could be easily predicted a priori, the
ingly, in the reaction, a moderate reduction of the catalyst significance of the P@ee analysis described above is evi-
loading appears to be a viable option to increase the pro- dent. In this respect, this kind of analysis is not expected to
ductivity of the alkaloid derivative, which cannot be pur- be a universal criterion to decide in favor of any of the
sued much below the 1 mol-% threshold (P ≈ 17@93% ee, two options in a real application as many other factors (e.g.
35% isolated yield).
preparation simplicity and cost, turnover frequency, actual
The establishment of the ee–P relationship for the soluble product yield, etc.) could decisively come into play. None-
catalyst set the stage for a better assessment of P1–P5. For theless, in cases like this where the chiral catalyst is by far
this purpose, ee and yield data of the heterogeneous runs the most expensive chemical in the transformation, the
were converted into P @ee pairs, defined as the producti- evaluation of the P@ee/P @ee relationships for the homo-
n
n
vity (P ) and weight-averaged enantiomeric excess ( e¯ e¯ ) for geneous and heterogeneous cases may be of prime impor-
n
a virtual gross-sample of 21a obtained by pooling together tance for a fair assessment of the relative merits of the two
[
19]
those isolated up to the nth run (Supporting Information). alternatives towards process intensification.
Examination of the results shown as solid curves in Fig- As a final comment, the polystyrene-supported dimeric
ure 4 (b) revealed that two limiting scenarios have to be systems 11 appear to provide superior results to P1–P5. In
considered, which depend on whether the priority is given fact, 11 showed a significantly better recycling profile than
to the maximization of the product ee or to the catalyst P1–P5 and allowed the attainment of significantly higher
productivity. In order to maximize the ee, P5 was the only productivities, even in the high ee range (e.g. P ≈ 57@97%
8
[
5g]
IPB system that could rival the homogeneous catalysts ac- ee and P20 ≈ 135@95% ee).
For reasons that are un-
ross a number of reaction cycles. Although a complete ee– known at present but possibly related to the different chem-
P analysis for the TBDMS soluble system was not per- ical stability of the linkage that connects the 9-O substituent
formed in this work, the P ≈ 26@97% ee reached with P5 with the chiral core, the polystyrene-supported dimeric sys-
5
(
Figure 4, b) appears substantially superior to P ≈ 6@97% tems 11 remain the reference materials for the hetero-
ee and P ≈ 7@94% ee provided under standard conditions geneous catalysis of the asymmetric dimerization of ketenes.
by 18 (Table 2, Entry 5) and TBDMS-1, respectively.
[7f]
In order to make a comparison with the catalyst produc-
tivity under the alternative conditions, i.e. when some re-
duction of the enantioselectivity was deemed acceptable, the Conclusions
average enantiomeric purity achieved at the end of the cy-
cles with P4 (90% ee) was arbitrarily selected as the refer-
A convenient procedure has been described for the coval-
ence level. With this choice, the interpolation of the data ent immobilization of different ester or silyl ether mono-
for the soluble organocatalyst (Figure 4, b) led to an estima- meric derivatives of 10,11-dihydroquinidine into cross-
tion of P ≈ 18@90% ee, which corresponds to the use of linked insoluble films. In spite of the unconventional sup-
ca. 0.75 mol-% of 22 (Figure 3); by contrast, the cumulative port structure, the IPB systems obtained by this approach
productivity of the heterogeneous system could be directly were highly enantioselective and reasonably recyclable or-
obtained as the rightmost point of the corresponding curve ganocatalysts in the heterogeneous asymmetric dimeriza-
for P4 (Figure 4, b) and was P13 ≈ 67@90% ee.
tion of ketenes. This allowed the preparation of chiral Wein-
Comparison of these results demonstrated that recycling reb β-ketoamides with 79–99% ee in up to 13 reaction cy-
afforded nearly three times more product than 22 at re- cles.
4
duced loadings, the amount of chiral organocatalyst and
The problem of a sound comparison between IPB sys-
the enantiomeric purity of the final amide 21a being the tems and their homogeneous counterparts has been ad-
same. Interestingly, this perhaps unimpressive increase in P dressed by the introduction of a productivity/enantio-
was not the only advantage of the IPB approach because selectivity protocol that, in essence, requires the extended
the use of the supported system also allowed the isolation reuse of the recoverable catalyst on the one hand, and the
of 21a in a nearly pure form, without the need for chroma- evaluation of the corresponding homogeneous system at
tographic purification, and with a substantially higher yield progressively lower loadings on the other. Because this kind
(51% overall) than with the soluble organocatalyst at low of analysis allows a better critical evaluation of the results
loadings (ca. 27% yield, Figure 3). than the customary description of (a few) recycle runs
In this scenario, the immobilization of the chiral organ- alone, we feel that the proposed procedure could become a
ocatalyst was more effective than the reduction of the load- standard tool for further studies in the field of recoverable
ing of the corresponding soluble derivative, both in terms enantioselective catalysts.
of higher productivity of the expensive alkaloid catalyst and
Given the straightforward preparation of the alkaloid
better use of the other chemicals involved in the catalyzed monomeric precursors 16–18 and the suitability of hydro-
reaction. This conclusion is related to the features of the silylation chemistry for the covalent anchoring of organic
reactions under examination, which include a reasonably derivatives and catalysts within capillary channels,[ the
good recycling profile for the IPB systems (especially P4 extension of the approach described in this work to micro-
and P5) and the specific response of the homogeneous pro- fluidic devices is currently underway.
20]
1342
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1336–1345

Silicone-Supported Organocatalysts
colorless oil (3.45 g, 61% over two steps); b.p. 106–115 °C/
Experimental Section
1
0
1
2
.15 Torr. H NMR (300 MHz, CDCl
.52 (s, 6 H), 1.66–1.85 (m, 2 H), 1.88–2.16 (m, 2 H), 4.76–5.16 (m,
3
): δ = 1.07–1.51 (m, 14 H),
General: See the Supporting Information for the materials, general
methods, instrumentation, and analytical procedures.
13
H), 5.81 (ddt, J
a
= 16.9, J
): δ = 25.26, 29.08, 29.27, 29.52, 29.68,
29.87, 32.54, 33.96, 46.25, 71.30, 114.25, 139.27 ppm. C14
b
= 10.1, J
c
= 6.7 Hz, 1 H) ppm.
C
Chlorodimethyl(undec-10-enyl)silane (13): A 100 mL three-necked
flask fitted with a reflux condenser, dropping funnel, nitrogen inlet,
and magnetic stirring bar was charged with degreased magnesium
turnings (0.716 g, 29.4 mmol), dry diethyl ether (10 mL), and a
crystal of iodine. The flask was wrapped with a tissue cloth and a
solution of 11-bromoundec-1-ene (6.25 g, 26.8 mmol) in dry diethyl
ether (15 mL) was added dropwise over 2 h to the rapidly stirring
suspension so as to maintain a slight but regular reflux in the con-
denser. The dropping funnel was rinsed with dry diethyl ether
NMR (50 MHz, CDCl
3
H27Cl
(230.82): calcd. C 72.85, H 11.79, Cl 15.36; found C 73.01, H 11.82,
Cl 15.28.
Chlorodimethyl(2-methylpent-4-en-2-yl)silane (15): A 100 mL three-
necked flask fitted with a reflux condenser, dropping funnel, nitro-
gen inlet, and magnetic stirring bar was charged with magnesium
turnings (0.330 g, 13.7 mmol, degreased by washing with dry di-
ethyl ether and dry THF (5 mL). After the addition of 1,2-dibro-
moethane (0.2 mL) the flask was heated at 90 °C, and a solution
of 14 (2.61 g, 11.3 mmol) in dry THF (5 mL) was added dropwise
over 2 h. The dropping funnel was rinsed with dry THF (3 mL),
and the resulting dark solution was stirred at 70 °C for 4 h, where-
upon Ͼ85% conversion of the halogen substrate was observed by
GC. After the small amount of residual magnesium settled, the
clear supernatant and two THF rinses (2 mL) were cannulated un-
der nitrogen into a second 100 mL three-necked flask that con-
(3 mL) and the resulting mixture was stirred overnight at r.t., to
give a mid-brown solution of the Grignard reagent, which con-
tained small amounts of unreacted magnesium and showed Ͼ98%
GC conversion of the starting bromide. After the solid settled, the
clear supernatant and two THF rinses (4 mL) were cannulated un-
der nitrogen into a second 100 mL three-necked flask that con-
tained a rapidly stirring solution of dichlorodimethylsilane (6.9 mL,
57 mmol, 2.1 equiv.) in dry THF (10 mL). The initially clear solu-
tion was stirred overnight at r.t. under nitrogen to give a thick
white suspension, which was diluted with dry n-hexane (20 mL)
and filtered under nitrogen through a mid-porosity glass frit. The
residue on the frit was washed with dry n-hexane (7ϫ 5 mL) and
the combined filtrates were concentrated with a rotary evaporator.
The crude compound was distilled under reduced pressure with a
tained
1
2
a
stirring solution of dichlorodimethylsilane (1.61 g,
2.4 mmol, 1.1 equiv.) and copper cyanide (15 mg, 0.24 mmol,
mol-%) in dry THF (5 mL). The resulting brownish clear solution
was stirred under nitrogen for 24 h at 70 °C. After cooling to r.t.,
the solution was diluted with dry n-hexane (30 mL) and filtered
under nitrogen through a medium porosity glass frit. The residue
on the frit was washed with dry n-hexane (2ϫ 5 mL), and the com-
bined filtrates were concentrated with a rotary evaporator. The
crude compound was distilled under high vacuum with a Claisen
apparatus connected to an oil diffusion pump. After discarding an
abundant forerun, a fraction was collected (b.p. 40–73 °C/3–
Claisen apparatus to give 13 as a clear, colorless oil (3.30 g, 50%);
1
b.p. 73–74 °C/0.2 Torr. H NMR (300 MHz, CDCl
3
): δ = 0.40 (s, 6
H), 0.69–0.94 (m, 2 H), 1.13–1.57 (m, 14 H), 1.90–2.17 (m, 2 H),
.84–5.10 (m, 2 H), 5.81 (ddt, J = 16.9, J = 10.2, J = 6.7 Hz, 1
H) ppm. C NMR (75 MHz, CDCl ): δ = 1.83, 19.17, 23.15, 29.12,
9.30, 29.40, 29.64, 29.67, 33.14, 33.99, 114.25, 139.29 ppm.
27ClSi (246.90): calcd. C 63.24, H 11.02, Cl 14.36, Si 11.38;
found C 63.20, H 11.00, Cl 14.65.
4
a
b
c
13
3
2
–5
1
6
ϫ10 mbar) that, by H NMR spectroscopy and GC–MS, con-
13
C H
tained ca. 68 wt.-% of 15 together with 12-methyltridec-1-ene (≈ 20
wt.-%) and minor amounts (Ͻ5 wt.-%) of 12-methyltridec-1,12-
1
1
2-Chloro-12-methyltridec-1-ene (14): A solution of (undec-11-en-
-yl)magnesium bromide was prepared as detailed above from mag-
diene and 14. The mixture (1.39 g ≈ 30% of 15) was used directly
1
in the next step. H NMR (200 MHz, CDCl
3
): δ = 0.37 (s, 6 H),
nesium turnings (0.912 g), 11-bromoundec-1-ene (8.00 g,
4.3 mmol), and dry diethyl ether (10 + 15 mL). After cooling with
an ice bath, dry acetone (3.8 mL, 52 mmol, 1.5 equiv.) in diethyl
ether (5 mL) was added dropwise over 1 h. The dropping funnel
was washed with diethyl ether (5 mL) and, after stirring at r.t. for
0.96 (s, 6 H), 1.04–1.50 (m, 16 H), 1.90–2.27 (m, 2 H), 4.84–5.19
(m, 2 H), 5.82 (ddt, J = 16.9, J = 10.1, J = 6.6 Hz, 1 H) ppm.
C NMR (50 MHz, CDCl ): δ = –0.73, 22.28, 23.79, 29.01, 29.21,
29.57, 29.69, 29.77, 30.72, 33.88, 38.27, 114.14, 139.20 ppm.
3
a
b
c
1
3
3
HRMS (EI ): m/z calcd. for C16H
33ClSi [M]+ 288.2040; found
+
15 min, the contents of the flask was hydrolyzed with saturated
288.2047.
ammonium chloride (25 mL). The organic components were ex-
tracted into diethyl ether (3ϫ 10 mL), and the combined organic
phases were dried with sodium sulfate. Removal of the volatiles
with a rotary evaporator and at 0.05 Torr afforded crude 2-meth-
yltridec-12-en-2-ol (7.10 g) as a clear pale-yellow oil that was di-
rectly used in the next step.
9
-O-[Dimethyl(undec-10-en-1-yl)silyl]-10,11-dihydroquinidine (17):
A 10 mL Schlenk tube was charged under nitrogen with hydroquin-
idine (0.250 g, 0.766 mmol), dry DMF (1.0 mL), triethylamine
(
3
0.21 mL, 1.5 mmol), 4-dimethylaminopyridine (30 mg, 0.25 mmol,
0 mol-%), and 13 (0.28 g, 1.0 mmol, 1.3 equiv.). The mixture was
magnetically stirred at r.t. for 48 h and then diluted with toluene
(5 mL). The organic phases were washed with water (2ϫ 5 mL)
and dried with anhydrous sodium sulfate. The volatiles were re-
moved with a rotary evaporator, and the residue was purified by
Crude 2-methyltridec-12-en-2-ol (5.00 g) was placed into a 100 mL
two-necked flask fitted with a dropping funnel and a magnetic stir-
ring bar. After cooling with an ice bath, a solution of zinc chloride
(
4
3
6.42 g, 47 mmol) in 37% HCl (6.5 mL) was added dropwise over flash chromatography (SiO
2
, AcOEt/MeOH = 95:5) to give 17
(0.255 g, 62%) as a pale yellow viscous oil. R = 0.51 (SiO , AcOEt/
= +1.1 (c = 0.51 g/100 mL, CH Cl ). MS (ES ):
m/z = 537.8 [M + H] . H NMR (200 MHz, CDCl ): δ = –0.03 (s,
0 min with rapid stirring. The dropping funnel was rinsed with
7% HCl (3 mL), and the resulting orange mixture was warmed
f
2
2
1
+
MeOH, 95:5). [α]
D
2
2
+
1
to r.t. with vigorous stirring until the consumption of the alcohol
substrate was observed by GC analysis (1 h). The contents of the
flask was diluted with water (15 mL) and extracted into light petro-
leum ether (3ϫ 15 mL). The organic phases were dried with cal-
cium chloride, and the volatiles removed with a rotary evaporator
to give an oil. The crude product was purified by distillation at
reduced pressure with a Claisen apparatus with a short Vigreux
column. After discarding the forerun, 14 was obtained as a clear
3
3 H), 0.08 (s, 3 H), 0.53–0.55 (m, 2 H), 0.90 (t, J = 7.1 Hz, 3 H),
1.10–1.75 (m, 14 H), 1.98–2.08 (m, 2 H), 2.68–2.98 (m, 4 H), 3.08–
3.22 (m, 1 H), 3.94 (s, 3 H), 4.86–5.02 (m, 2 H), 5.68–5.89 (m, 2
H), 7.22 (br. s, 1 H), 7.34 (dd, J
a b
= 9.2, J = 2.5 Hz, 1 H), 7.51 (br.
s, 1 H), 7.99 (d, J = 9.0 Hz, 1 H), 8.71 (d, J = 4.4 Hz, 1 H) ppm.
1
3
C NMR (50 MHz, CDCl
3
): δ = –1.45, 12.08, 16.93, 19.69, 23.26,
25.32, 26.47, 27.16, 29.53, 29.57, 33.50, 33.87, 37.46, 50.55, 51.12,
Eur. J. Org. Chem. 2012, 1336–1345
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1343

D. Cancogni, A. Mandoli, R. P. Jumde, D. Pini
FULL PAPER
5
1
5.96, 60.66, 73.06, 100.64, 114.15, 118.70, 121.79, 126.24, 131.87, Heterogeneous Catalytic Asymmetric Ketene Dimerization. General
39.26, 144.50, 147.42, 147.65,158.11 ppm. C33 Si (536.87): Procedure: A 50 mL Schlenk tube with a magnetic stirrer and stop-
cock side arm was charged under nitrogen with P1–P5 (88–263 mg,
.050 mmol, 5 mol-%). After sealing the tube with a septum, dry
-O-[Dimethyl(2-methyltridec-12-en-2-yl)silyl]-10,11-dihydroquinid-
CH Cl (10 mL) was syringed into the tube and the polymer film
52 2 2
H N O
calcd. C 73.83, H 9.76, N 5.22; found C 73.40, H 9.62, N 5.17.
0
9
2
2
ine (18): A 10 mL Schlenk tube was charged under nitrogen with
hydroquinidine (0.430 g, 1.32 mmol), dry DMF (1.0 mL), triethyl-
amine (0.83 mL, 6.0 mmol), 4-dimethylaminopyridine (16 mg,
was allowed to swell with stirring for 5 min. N,N-Diisopropylethyl-
amine (170 μL, 1.0 mmol) and 19a–c (1.0 mmol) were sequentially
injected through the septum and the mixture was stirred at 500 rpm
0
1
7
.13 mmol, 10 mol-%), and 15 (0.66 g, 68% purity, ca. 1.5 mmol, at r.t. After the time t (Table 2), the clear supernatant was can-
1
.1 equiv.). The resulting mixture was magnetically stirred at r.t. for
nulated under nitrogen into a dry Schlenk tube and the polymer
2 h and then diluted with toluene (5 mL). The organic phases were
film washed with CH Cl (5 mL). The combined organic phases
2
2
washed with water (2ϫ 5 mL) and dried with sodium sulfate. The
were treated with HN(OMe)Me (37 μL, 0.50 mmol) and 2-pyridone
volatiles were removed with a rotary evaporator, and the residue
(4.7 mg, 0.05 mmol), and the resulting solution was stirred at room
was purified by flash chromatography (SiO
2
, AcOEt/MeOH = 95:5)
temperature for the time t (Table 2). For the ee determination, a
2
to give 18 (0.406 g, 53%) as a pale yellow viscous oil. R
f
= 0.59 sample of the reaction mixture (0.20 mL) was passed through small
21
(
SiO
2
, AcOEt/MeOH, 95:5). [α]
D
= +1.0 (c = 0.53 g/100 mL,
pad of silica gel with n-hexane/AcOEt = 2:1 (3ϫ 1 mL), evaporated
with a nitrogen flow, and dissolved in 2-propanol for HPLC analy-
+
+
1
CHCl
3
). MS (ES ): m/z
600 MHz,CDCl
s, 3 H), 0.90 (t, J = 7.4 Hz, 3 H), 0.92 (s, 3 H), 0.94 (s, 3 H), 0.99–
=
579.9 [M
+
H] .
H
NMR
sis.[
5g,7f]
The remainder of the solution was washed with concen-
(
(
3
): major rotamer, ca. 76% δ = –0.33 (s, 3 H), 0.15
[7f]
trated pH 7 buffer solution, dried (Na SO ), and evaporated to
2
4
1
1
2
.58 (m, 21 H), 1.64 (br. s, 1 H), 1.72 (m, 1 H), 1.92–2.10 (m, 3 H),
.65–2.75 (m, 1 H), 2.76–2.94 (m, 3 H), 2.96–3.03 (m, 1 H), 3.94
give 21a–c as turbid oils that were nearly pure by H NMR spec-
troscopy (Figure S3). The polymeric film recovered after can-
nulation was washed with dry CH Cl (2ϫ1 mL) and used directly
(
J
7
8
s, 3 H), 4.89–5.02 (m, 2 H), 5.60 (d, J = 3.1 Hz, 1 H), 5.81 (ddt,
= 16.9, J = 10.2, J = 6.7 Hz, 1 H), 7.19 (d, J = 2.4 Hz, 1 H),
.36 (dd, J = 9.2, J = 2.6 Hz, 1 H), 7.55 (d, J = 4.5 Hz, 1 H),
.02 (d, J = 9.2 Hz, 1 H), 8.73 (d, J = 4.5 Hz, 1 H) ppm. Minor
2
2
a
b
c
in further catalysis cycles.
a
b
Supporting Information (see footnote on the first page of this arti-
cle): Procedure for the homogeneous catalysis runs, spectra of new
compounds, appearance and IR spectra of the polymer films, and
elaboration of the P/ee data.
rotamer, ca. 24% δ = –0.42 (s, 3 H), 0.11 (s, 3 H), 0.80 (s, 3 H),
0
1
1
.82 (s, 3 H) 2.45–2.53 (m, 2 H), 2.55–2.64 (m, 1 H), 3.35–3.49 (m,
H), 3.92 (s, 3 H), 4.80 (d, J = 9.2 Hz, 1 H), 7.10 (d, J = 4.3 Hz,
H), 7.32 (dd, J = 9.2, J = 2.7 Hz, 1 H), 7.86 (d, J = 2.7 Hz, 1
a b
1
3
H), 7.98 (d, J = 9.2 Hz, 1 H), 8.64 (d, J = 4.2 Hz, 1 H) ppm.
NMR (50 MHz, CDCl ): major + minor rotamers δ = –4.29, –4.05,
3.89, –3.75, 12.18, 20.65, 21.23, 21.40, 22.86, 22.91, 23.81, 23.98,
C
Acknowledgments
3
–
2
2
5
1
1
1
This work supported by the University of Pisa, Ministero dell’Uni-
versità e della Ricerca (MIUR) (project: Sintesi e Stereocontrollo
di Molecole Organiche per lo Sviluppo di Metodologie Innovative
di Interesse Applicativo – PRIN 2007) and Consiglio Nazionale
delle Ricerche – Istituto di Chimica dei Composti Organometallici
5.28, 25.88, 26.24, 26.31, 26.62, 27.29, 27.56, 29.04, 29.26, 29.62,
9.79, 29.85, 30.89, 33.92, 37.71, 38.76, 38.86, 49.62, 49.92, 50.55,
1.49, 55.45, 55.80, 61.05, 61.15, 73.15, 79.64, 100.51, 105.00,
14.19, 119.03, 121.26, 121.41, 121.67, 126.40, 127.13, 131.53,
31.95, 139.30, 144.40, 145.53, 147.46, 147.59, 148.42, 156.57,
(ICCOM-CNR).
58 2 2 2 2
57.96 ppm. C36H N O Si·0.02CH Cl (580.63): calcd. C 74.51,
H 10.07, Cl 0.24, N 4.82, O 5.51, Si 4.84; found C 74.59, H 10.15,
N 4.81.
[
1] a) Chiral Catalyst Immobilization and Recycling (Eds.: D. E.
De Vos, I. F. J. Vankelecom, P. A. Jacobs), Wiley-VCH,
Weinheim, Germany, 2000; b) Q.-H. Fan, Y.-M. Li, A. S. C.
Chan, Chem. Rev. 2002, 102, 3385–3466; c) M. Benaglia, A.
Puglisi, F. Cozzi, Chem. Rev. 2003, 103, 3401–3429; d) F. Cozzi,
Adv. Synth. Catal. 2006, 348, 1367–1390; e) M. Benaglia, New
J. Chem. 2006, 30, 1525–1533; f) A. F. Trindade, P. M. P. Gois,
C. A. M. Afonso, Chem. Rev. 2009, 109, 418–514; g) J. Lu, P. H.
Toy, Chem. Rev. 2009, 109, 815–838; h) T. E. Kristensen, T.
Hansen, Eur. J. Org. Chem. 2010, 3179–3204; i) Heterogenized
Homogeneous Catalysts for Fine Chemicals Production, vol. 33
Preparation of P1–P5 by Anchoring 16–18 on to Cross-Linked Sili-
cone Films. General Procedure: A solution of 16, 17, or 18 (0.40–
0.75 mmol), PMHS, PDMS-DV lw, and PDMS-DV hw (for the
exact ratios, see Table 1) in toluene (14–20 mL) was poured into a
2
flat PTFE-lined vessel (90–150 cm base area). A solution of the
Karstedt catalyst in toluene (1.0 mm, molar ratio SiH/Pt = 1000)
was added dropwise and, after swirling to complete the mixing of
the solutions, the vessel was covered with a glass lid and heated
under air for 1 h at 50 °C. The temperature was then raised to 70 °C
to effect solvent evaporation and polymer cross-linking. After cur-
ing overnight, the resulting solid film was swollen with a solution
of 1-hexadecene (0.10 mL) in toluene (10 mL), and the volatiles
were evaporated by gently heating for 1 h at 70 °C. The polymer
film was moistened with little THF to facilitate the detachment
from the vessel and transferred to a metal-net thimble, which was
placed into a Kumagawa device. After continuous extraction over
(
Eds.: P. Barbaro, F. Liguori), Springer, Dordrecht, The Ne-
therlands, 2010; j) Catalytic Methods in Asymmetric Synthesis
Eds.: E. Gruttadauria, F. Giacalone), John Wiley & Sons,
011.
(
2
[
2] a) O. Gleeson, R. Tekoriute, Y. K. Gun’ko, S. J. Connon, Chem.
Eur. J. 2009, 15, 5669–5673; b) C. W. Jones, Top. Catal. 2010,
53, 942–952.
[3] A. Mandoli, D. Pini, M. Fiori, P. Salvadori, Eur. J. Org. Chem.
005, 1271–1282.
[
2
4] a) K. Kacprzak, J. Gawronski, Synthesis 2001, 961–998; b) C.
Palomo, M. Oiarbide, R. Lopez, Chem. Soc. Rev. 2009, 38,
2
2 2
d, first with dry THF and then with dry CH Cl , the almost
colorless and nearly transparent P1–P5 (Figure S1) was dried un-
der reduced pressure (0.05 Torr). Calliper measurement of the re-
sulting soft, elastic films indicated an approximate thickness of
632–653; c) T. Marcelli, H. Hiemstra, Synthesis 2010, 1229–
1279.
[
5] a) C. E. Song, J. W. Yang, H.-J. Ha, Tetrahedron: Asymmetry
0.2 mm. The material was characterized by IR spectroscopy (Fig-
1
997, 8, 841–844; b) C. Bolm, A. Maischak, Synlett 2001, 93–
ure S2) and elemental analysis. Elemental analysis found (N%): P1
95; c) F. Bigi, S. Carloni, R. Maggi, A. Mazzacani, G. Sartori,
0
.77, P2 0.53, P3 1.59, P4 0.66, P5 0.73.
344
G. Tanzi, J. Mol. Catal. A 2002, 182–183, 533–539; d) S.
1
www.eurjoc.org
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2012, 1336–1345

Silicone-Supported Organocatalysts
France, D. Bernstein, A. Weatherwax, T. Lectka, Org. Lett.
pedia of Chemical Technology, 5th edition, vol. 22, pp. 547–626,
John Wiley & Sons, Hoboken, New Jersey, 2006.
2
005, 7, 3009–3012; e) B. Thierry, J.-C. Plaquevent, D. Cahard,
Mol. Diversity 2005, 9, 277–290; f) R. Chinchilla, C. Najera, [14]
B. P. S. Chauhan, J. S. Rathore, J. Am. Chem. Soc. 2005, 127,
F. J. Ortega, S. Tari, Tetrahedron: Asymmetry 2009, 20, 2279–
2
13738–13738.
286; g) R. P. Jumde, A. Mandoli, F. De Lorenzi, D. Pini, P.
[
[
15] J. Y. Corey, Adv. Organomet. Chem. 2004, 51, 1–52.
Salvadori, Adv. Synth. Catal. 2010, 352, 1434–1440.
16] P. L. Launer, in: Silicone Compounds Register and Review (Ed.:
B. Arkles), Petrarch Systems, Inc., Bristol, UK, 1987, available
at http://www.gelest.com/Library/11Infra.pdf (June 1st 2011).
17] a) L. Mink, Z. Ma, R. A. Olsen, J. N. James, D. S. Sholl, L. J.
Mueller, F. Zaera, Top. Catal. 2008, 48, 120–127; b) J. Lai, Z.
Ma, L. Mink, L. J. Mueller, F. Zaera, J. Phys. Chem. B 2009,
113, 11696–11701.
[18] In response to a query from a referee, rapid magnetic stirring
caused some breakage of the polymer films into smaller par-
ticles. Even in the last cycle this problem did not prevent the
rapid separation of the reaction mixture by siphoning.
[
[
6] K. M. Kacprzak, N. M. Maier, W. Lindner, Tetrahedron Lett.
2006, 47, 8721–8726.
7] a) M. A. Calter, J. Org. Chem. 1996, 61, 8006–8007; b) M. A.
Calter, X. Guo, J. Org. Chem. 1998, 63, 5308–5309; c) M. A.
Calter, F. C. Bi, Org. Lett. 2000, 2, 1529–1531; d) M. A. Calter,
X. Guo, W. Liao, Org. Lett. 2001, 3, 1499–1501; e) M. A.
Calter, X. Guo, Tetrahedron 2002, 58, 7093–7100; f) M. A.
Calter, R. K. Orr, W. Song, Org. Lett. 2003, 5, 4745–4748.
8] A. Mandoli, M. Lessi, D. Pini, C. Evangelisti, P. Salvadori,
Adv. Synth. Catal. 2008, 350, 375–379.
[
[
[
[
9] Y. Motoyama, T. Nishikata, H. Nagashima, Chem. Asian J.
2
011, 6, 78–82.
10] a) M. S. DeClue, J. S. Siegel, Org. Biomol. Chem. 2004, 2, 2287–
298; b) M. A. Grunlan, K. R. Regan, D. E. Bergbreiter, Chem.
[19] For the sake of completeness, Calter and coworkers described
conditions for the homogeneous dimerization that provide up
[7d]
2
to P ≈ 92@99% ee (ref. ). However, these protocols require
the external generation of methylketene and the use of low tem-
peratures (–78 °C) and are, therefore, less convenient than the
alternative procedure followed in this work.
Commun. 2006, 1715–1717.
[
[
[
11] a) A. Shirahata, Tetrahedron Lett. 1989, 30, 6393–6394; b)
C. G. L. Khoo, J. B. Lando, H. Ishida, J. Polym. Sci. Polym.
Chem. Ed. 1990, 28, 213–232.
[20] a) O. Trapp, S. K. Weber, S. Bauch, W. Hofstadt, Angew. Chem.
2007, 119, 7447; Angew. Chem. Int. Ed. 2007, 46, 7307–7310;
b) D. Wistuba, V. Schurig, LC-GC 2009, 22, 60–69; c) S. K.
Weber, S. Bremer, O. Trapp, Chem. Eng. Sci. 2010, 65, 2410–
2416.
12] a) S. J. Rowan, J. K. M. Sanders, J. Org. Chem. 1998, 63, 1536–
1
546; b) S. J. Rowan, D. J. Reynolds, J. K. M. Sanders, J. Org.
Chem. 1999, 64, 5804–5814.
13] M. Butts, J. Cella, C. D. Wood, G. Gillette, R. Kerboua, J. Le-
man, L. Lewis, S. Rubinsztajn, F. Schattenmann, J. Stein, D.
Wicht, S. Rajaraman, J. Wengrovius, in: Kirk-Othmer Encyclo-
Received: September 9, 2011
Published Online: January 25, 2012
Eur. J. Org. Chem. 2012, 1336–1345
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1345