Polystyrene-supported amino alcohol ligands for the heterogeneous asymmetric addition of phenyl zinc reagents to aldehydes

Article abstract of DOI:10.1016/j.tet.2005.07.112

A family of polystyrene-supported amino alcohols, characterized by a high catalytic activity in alkyl transfer from zinc to formyl groups has been successfully tested in the enantioselective addition of phenyl zinc reagents to aldehydes to afford chiral diarylmethanols. Enantioselectivities higher than 90% (mean ee 90.5%; eight examples) are recorded with aromatic aldehydes in what represents the first successful use of heterogeneous, polymeric reagents for enantiocontrol in the phenylation of aldehydes.

Full text of DOI:10.1016/j.tet.2005.07.112

Tetrahedron 61 (2005) 12111–12120
Polystyrene-supported amino alcohol ligands for the
heterogeneous asymmetric addition of phenyl zinc
reagents to aldehydes
b
David Castellnou, Montserrat Fontes, Ciril Jimeno, Daniel Font, Lluıs Sola,
b
a
a
a
´
`
b
Xavier Verdaguer and Miquel A. Pericas
a,b,
*
`
a
¨
Institute of Chemical Research of Catalonia (ICIQ), Av.Paısos Catalans 16, 43007 Tarragona, Spain
b
´ `
Departament de Quımica Organica, Universitat de Barcelona, 08028 Barcelona, Spain
Received 30 June 2005; revised 15 July 2005; accepted 15 July 2005
Abstract—A family of polystyrene-supported amino alcohols, characterized by a high catalytic activity in alkyl transfer from zinc to formyl
groups has been successfully tested in the enantioselective addition of phenyl zinc reagents to aldehydes to afford chiral diarylmethanols.
Enantioselectivities higher than 90% (mean ee 90.5%; eight examples) are recorded with aromatic aldehydes in what represents the first
successful use of heterogeneous, polymeric reagents for enantiocontrol in the phenylation of aldehydes.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
as the most convenient way to gain access to these structural
motifs.4
Enantiomerically pure diarylmethanols are important building
blocks for biologically active compounds.1 For example,
(R)-neobenodine, (R)-orphenadrine, or (S)-carbinoxamine
(Fig. 1) have been used for a long time as muscle relaxants
or antihistaminics.2 However, their preparation by asym-
metric synthesis is challenging, because the asymmetric
reduction of the appropriate diaryl ketone is usually
hampered by low ee’s due to the steric and electronic
similitudes between both aryl groups.3 Then, the enantio-
selective addition of an aryl fragment to an aldehyde (where
all groups are easily distinguishable by the catalyst) appears
Diphenylzinc could be in principle a suitable reagent for this
task, but it has the problem that the uncatalyzed background
addition to aldehydes is a significant competing reaction
with deletereous effects on the enantiomeric purity of the
resulting diarylmethanols. In any case, since the pioneering
work by Fu in 1997,5 several efficient homogeneous
catalytic systems for the enantioselective addition of
diphenylzinc to aldehydes have been successfully
developed relying in two basic strategies: the use of diluted
reaction conditions to increase the rate difference between
the catalyzed and the uncatalyzed processes,6 or the in situ
formation of a less reactive, mixed EtPhZn species
extensively studied by Bolm.7
Over the last years, our research group has been involved in
a project devoted to the synthesis of highly modular,
synthetic yet enantiopure b-amino alcohol ligands using the
Sharpless epoxidation of allyl alcohols or the Jacobsen
epoxidation of arylethylenes as the ultimate source of
chirality.8 The modular nature of these species has allowed
the simultaneous optimization of their catalytic activity and
enantioselectivity in different processes such as alkyl
transfer to carbonyls8,9 and imines,10 oxazaborolidine-
mediated reduction of ketones,11 transfer hydrogenation of
ketones,12 and allylic alkylation.13 Among these ligands,
readily available 2-piperidino-1,1,2-triphenylethanol (1)8b
depicts an excellent enantioselectivity/activity profile for
Figure 1. Some pharmaceutically active diarylmethanols.
Keywords: Heterogenous reaction; Ligand immobilization; Enantioselec-
tive synthesis; Aldehyde arylation.
*
Corresponding author. Tel.: C34 977 920 211; fax: C34 977 920 222;
e-mail: mapericas@iciq.es
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.07.112

12112
D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
alkyl transfer to aldehydes8b,14 and has also been found to be
a most efficient mediator for the enantioselective phenyl
transfer to these substrates,15 enantioselectivities of 91–99%
being achieved with only 1.5 mol% of 1 in reactions that
typically require less than 1 h for completion.
approach.6b Later on, Bolm immobilized a ferrocenyloxazoline
ligand onto polymeric supports and studied the
enantioselective phenyl transfer to p-chlorobenzaldehyde.
When the ligand was bound to an insoluble trityl chloride
resin, the polymer proved to be unsuitable for the catalysis
of the asymmetric process, and only racemic product was
obtained. On the other hand, when the ligand was bound to a
soluble MeO–PEG resin, the resulting catalyst led to the
addition product with high enantioselectivity.17 It is clear
from these results that low activity, insoluble resins fail to
induce the enantioselective reaction at a sufficient rate to
compete with the rather fast,15 uncatalyzed background
reaction, and that very active ligands should be designed to
achieve high enantiocontrol levels in this class of
heterogeneous reaction.
From a general perspective, homogeneous catalysis has
notably contributed to the clean and efficient production of
chiral compounds as single enantiomers. Homogeneous
catalytic processes, however, are usually performed in a
batch manner, and work-up stages required for product
isolation and catalyst recovery are detrimental to their
overall sustainable characteristics. To solve this problem,
covalent anchoring of properly functionalized ligands to
polymeric supports has been widely applied.16 While this
method can ultimately allow performing catalytic enantio-
selective reactions in a continuous mode, it is usually
accompanied by a decrease in catalytic activity and
enantioselectivity with respect to structurally referable,
homogeneous ligands. Within this approach, attention has
been paid to the enantioselective arylation of aldehydes, but
only partial success has been achieved. In 1999, Pu and
co-workers used a soluble, rigid BINOL polymer to perform
the catalytic asymmetric addition of ZnPh2 to aldehydes
under much diluted conditions, achieving ee’s up to 92%
(Fig. 2). However, a high catalyst loading (40 mol%) was
needed to drive the reaction to completion, and some of the
main advantages associated to the use of polymer-supported
ligands (easy recovery and reuse) were absent from this
Using the highly active ligand (R)-2-piperidino-1,1,
2-triphenylethanol (1),8b as the basis of our design,18 we
have developed polystyrene supported amino alcohols 219
and 3,20 and have introduced for their designation the term
tail-tied ligands. Gratifyingly enough, both the catalytic
activity and the enantioselectivity exhibited by 2 and 3 are
among the highest ever recorded for supported ligands, and
this fact converts them into qualified candidates for the
achievement of enantiocontrol in the phenylation of
aldehydes.
We wish to report here the preparation of a new tail-tied
ligand (5), a regioisomer of 2 and 3, conceptually derived
from the highly active and enantioselective amino alcohol
Figure 2. Soluble and insoluble polymer-supported ligands employed in the enantioselective phenylation of aldehydes.

D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
12113
4,8c and its evaluation in the enantioselective ethylation of
aldehydes. In addition, the evaluation of 2, 3, and 5 in the
enantioselective phenylation reaction leading to the identi-
fication of the first insoluble, polymeric ligands successful
in the considered reaction is also reported.
When enantiomerically pure (S)-6 was treated with
diisopropoxytitanium diazide in benzene at reflux, a totally
regioselective ring-opening took place leading to azido-
alcohol (S)-7, arising from attack to the more heavily
substituted carbon of the epoxide. The crude azidoalcohol
was directly submitted to hydrogenolysis in methanol
(1 atm H2; Pd/C) to afford amino alcohol 8 in 85% overall
yield.
It is interesting to note that the amino group of 8 offers a
good possibility for structural diversity, since many
different groups could be installed on it by alkylation,
cyclialkylation, and reductive amination processes.8c In the
present case, the planned pirrolidine ring was constructed by
cyclialkylation with 1,4-diiodobutane in ethanol at reflux in
the presence of potassium carbonate. To achieve a good
yield in this reaction, it was necessary to maintain in the
reaction medium an excess of 1,4-diiodobutane during the
whole reaction time. In this way, a 65% yield of 9 was
obtained after 72 h, with periodical addition of alkylating
agent (up to 7 equiv). Finally, the cyano group in 9 was
reduced to hydroxymethyl through a two-stage process:
First, the aldehyde 10 was obtained in 92% yield by
treatment of 9 with DIBALH at K78 8C in hexane/ether
solution; then, the primary alcohol 11 was formed (65%
yield) by reduction of 10 with sodium borohydride in
ethanol at room temperature. For the anchoring of 11 to
polymeric supports, a chlorotritylated polystyrene resin
(Barlos’ resin)23 was selected as the most convenient
alternative24 (Scheme 2). Starting from a Barlos’ resin
with fo: 1.60, and using the standard anchoring conditions
(Scheme 2), a functionalized resin 5, with f: 0.90 (calculated
by nitrogen elemental analysis with the formula;
f: 0.714[%N]) was obtained. Since fmax for this particular
resin is 1.0619 the yield of the anchoring process turns out to
be 85%.
2. Results and discussion
2.1. Synthesis and anchoring to Barlos’ resins of (S)-1-
(4-(hydroxymethyl)phenyl)-2,2-diphenyl-2-(pyrrolidin-1-yl)-
ethanol
As we have already mentioned, a primary goal in the present
research was the synthesis of resins 5 which, on one side,
possess the characteristic of the ligand being anchored to the
polymeric support through a position remote from the active
center and, on the other side, exhibit a functional group
arrangement opposite to the one present in 2 and 3. We have
previously shown8c–d that the regiochemistry of the ring-
opening of triarylethylene oxides with nitrogen nucleophiles
can be exclusively directed to the more substituted carbon
by the use of diisopropoxytitanium diazide,21 although
some additional functional group manipulation can be
required. According to this strategy (Scheme 1), the
known epoxide 6, that is readily available in enantiomerically
pure form (O99.9% ee) by Jacobsen epoxidation22
and recrystallization from hexane, would be the starting
material for the synthesis. It is interesting to note that 6 is
also the starting material for the preparation of the supported
ligand 2.
Scheme 1. Enantioselective synthesis of amino alcohol 11, the precursor of resin 5.

12114
D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
Scheme 2. Anchoring of amino alcohol 11 to a Barlos’ resin.
Figure 3. Monitoring of the anchoring of amino alcohol 11 to a Barlos’ resin by gel phase 13C NMR. The NMR spectrum of the model compound 12 is included
for comparison.
The progress of the anchoring process can be easily
monitored by gel-phase 13C NMR.25 For comparison
purposes, the trityl-protected amino alcohol 12 was easily
prepared by treatment of 11 with N-tritylpyridinium
tetrafluoroborate in acetonitrile. The diagnostic region of
the 13C NMR spectra of both compounds (12 in solution; 5
in gel) is represented in Figure 3.
As it can be seen, resin 5 provides high conversion numbers
for the ethylation of aromatic aldehydes (13a–c) and for
aliphatic and a,b-unsaturated aldehydes not fully substi-
tuted at the a position (13e–f and 13d, respectively). With
Table 1. Enantioselective ethylation of aldehydes 13a–h mediated by resin
5 (8 mol%)
2.2. Evaluation of resin 5 as a ligand for enantioselective
ethylation and phenylation of aldehydes
Starting aldehyde
Conv (%)
ee (%)
Resin 5 was initially tested in the enantioselective
addition of diethylzinc to aldehydes (13). A represen-
tative set of aldehydes 13a–h, mostly containing difficult
substrates (aliphatic, a,b-unsaturated), was selected, and
the ethylation reaction leading to 1-propanols 14a–h was
performed in toluene at 0 8C, in the presence of a
8 mol% of 5. The results of these additions have been
summarized in Table 1.
o-Methoxybenzaldehyde (13a)
m-Tolualdehyde (13b)
p-Fluorobenzaldehyde (13c)
Cinnamaldehyde (13d)
Heptanal (13e)
3-Phenylpropanal (13f)
2-Ethylbutanal (13g)
a-Methylcinnamaldehyde (13h)
95
90
94
89
93
94
50
55
86
90
84
71
90
89
91
89

D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
12115
respect to enantioselectivity, good results are obtained for
aromatic aldehydes. However, the most noteworthy
enantioselectivities are those obtained with aliphatic and
a,b-unsaturated aldehydes, where many homogeneous and
heterogeneous ligands fail. Encouraged by these results, we
decided to test the use of resin 5 for the enantioselective
phenylation reaction leading to diarylmethanols 15. The
phenyl transferring system developed by Bolm and co-
workers,7 that involves the use of a Ph2Zn/Et2Zn mixture
and has provided excellent results in the enantioselective
phenylation mediated by the monomeric ligand 1 was also
used in this case.15 The reactions were initially tested on a
limited set of aromatic aldehydes (13i–13l) by using a 10%
molar amount of catalyst. Since our primary interest was on
enantioselectivity, no attention was paid to optimization of
reaction time. The results of this study have been
summarized in Table 2.
determined first, working on p-tolualdehyde (13k) and
performing the reactions in toluene (for optimal resin
swelling) at room temperature (Scheme 3).
Scheme 3. Optimization of ligand amount in the phenylation of
p-tolualdehyde mediated by resin 2.
It was already clear from these experiments that 2 was a
much better ligand than 5 for the asymmetric phenyl transfer
reaction. With respect to the optimal amount of ligand, it
was decided to perform the reaction with a 10% molar
amount of 2 in order to secure the highest possible
enantioselectivity in the shortest reaction time. It is
important to recall here that this level of ligand loading is
the usual one in phenyl transfer reactions with homo-
geneous, monomeric ligands.
Table 2. Enantioselective phenylation of aldehydes 13i–l mediated by resin
5 (10 mol%)
Starting aldehyde
Conv (%)
ee (%)
Next, the phenyl transfer reaction was performed on a
representative family of aldehydes under the optimized
conditions. To test the preparative merits of the procedure,
the diarylmethanol products 15 were isolated and quantified
after each reaction. Results arising from this study have
been summarized in Table 3, where the enantioselectivities
recorded with the homogeneous ligand 1 under identical
experimental conditions have also been included for
comparison.
o-Fluorobenzaldehyde (13i)
o-Tolualdehyde (13j)
p-Tolualdehyde (13k)
90
99
99
99
43
38
48
42
m-Methoxybenzaldehyde (13l)
A first aspect of these results to be highlighted is that the
heterogeneous ligand 5 is able to induce enantioselectivity
in the phenylation reaction, albeit to a moderate level. It thus
appears that the strategy of structural modification of
ligands known to be very active and enantioselective to
allow anchoring to a polymeric matrix without perturbing
the catalytic center (tail-tied ligands) can provide a solution
for the problem of the enantioselective phenylation of
aldehydes with heterogeneous ligands. In any case, when
the behavior of the polymer-supported ligand 5 is compared
with that of its homogeneous counterpart 4,8c it becomes
evident that in this case the anchoring process provokes
some decrease in the enantioselectivity characteristics of the
homogeneous ligand. As we have previously shown,19–20
this is not the case for polymer-supported ligands 2 and 3,
conceptually derived from amino alcohol 1. In view of the
results obtained in the preliminary evaluation of resin 5, its
use as a catalytic ligand for the catalytic enantioselective
phenylation of aldehydes was abandoned. Alternatively, the
evaluation of resins 2 and 3 with the same purpose was
undertaken.
Table 3. Enantioselective phenylation of aldehydes mediated by resin 2
(10 mol%)
Starting aldehyde
Yield
(%)
ee
(%)
ee with 1
(%)
a-Methylcinnamaldehyde (13h)
o-Fluorobenzaldehyde (13i)
o-Tolualdehyde (13j)
p-Tolualdehyde (13k)
m-Methoxybenzaldehyde (13l)
Pivalaldehyde (13m)
75
99
98
96
74
78
86
81
87
85
91
87
90
80
91
90
94a
98
99
98
—
92
97
96
Biphenyl-4-carbaldehyde (13n)
2-Naphthaldehyde (13o)
a Reaction in toluene at 0 8C.
As it can be readily seen, high yields of diarylmethanols 15
are obtained in the phenyl transfer reaction mediated by
resin 2. Even more importantly, a uniformly high
enantioselectivity is recorded in the reactions, the mean ee
of the resulting products 15 being 87.6%.
2.3. Evaluation of resin 2 as a ligand for enantioselective
phenylation of aldehydes
2.4. Evaluation of resin 3 as a ligand for enantioselective
phenylation of aldehydes
According to precedents in the enantioselective ethylation
of aldehydes with this family of polymer-supported ligands,
a resin with a rather high cross-linking level (2% DVB) and
a functionalization level (f) of 0.35 mmol ligand/g was used
in this study. The optimal molar amount of resin was
While it is clear that resin 2 depicts a very interesting profile
as a ligand for the catalytic enantioselective phenyl transfer
to aldehydes, it is also true that its synthesis (as in the case of

12116
D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
Scheme 4. Two-step assembly of ligand 3 from its precursors.22
resin 5) from commercial precursors is rather lengthy. This
observation, that could be of practical interest if the
application of these resins at a larger scale was considered,
boosted the development of resin 3, that can be straight-
forwardly assembled from its fragments: enantiomerically
pure triphenylethylene oxide, piperazine, and a Merrifield
resin, as shown in Scheme 4.20
corresponding diarylcarbynol 15k was obtained with
enantiomeric purities of 92, 94, and 91%, respectively. It
is thus confirmed that, at least for 13k, 10 8C represents the
optimal temperature for reaction. To simultaneously gain
information on the kinetics of the process at different
temperatures, the forementioned experiments were per-
formed with continuous monitoring of the reaction progress
by in situ FTIR spectroscopy. This was done with an
immersible DiComp ATR diamond probe, and the
disappearance of the band corresponding to the carbonyl
group of 13k was analyzed. We have represented in Figure 4
the evolution of this band in the experiments at 0, 10, and
23 8C in the presence of 5 mol% of 3.
In addition to the ease of its preparation, resin 3 was shown
to be a most efficient ligand for the enantioselective ethyl
transfer to aldehydes, with catalytic activity and enantio-
selectivity that did not show any decrease with respect to the
referable, homogeneous ligand 16.20,26
Two aspects of this graph deserve a comment: On one hand,
the important acceleration experienced by the reaction when
the temperature increases from 0 to 10 8C, that has
necessarily to obey to a combination of physical (mass
transport) and chemical (kinetic) factors. On the other hand,
the high catalytic activity exhibited by 3 at 10 8C or above,
that leads to complete conversion in only 50 min. Keeping
in mind the possibility of a future use of 3 in a continuous
flow system, we also wanted to test if reaction time could be
further reduced if catalyst loading was increased. To this
end, the reaction at 23 8C was repeated with a 10 mol%
catalyst loading. The progress of this reaction has been
represented in Figure 5 along with that of the experiment
with 5% catalyst loading at the same temperature.
According to these precedents, resin 3 was an ideal
candidate for a successful ligand in enantioselective phenyl
transfer to aldehydes. As in the case of 2, a preliminary
screening confirmed these expectations (Table 4).
Table 4. Preliminary screening of resin 3 (10 mol%) in the enantioselective
phenylation of aldehydes
It is interesting to observe that the time required for
complete conversion is essentially divided by a factor of 2
when catalyst loading is increased from 5 to 10%. This is
clearly indicative that even much shorter contact times
could be sufficient for complete conversion at higher
catalyst loadings, and tells in favor of 3 as a suitable
candidate ligand for enantioselective phenylation in
continuous flow systems. As an additional point, it is to be
mentioned that the ee of the resulting arylcarbynol 15k
increases in only 1% (from 91 to 92%) while increasing
from 5 to 10% catalyst loading. As a result of these
observations, a set of optimized practical conditions for the
use of 3 in enantioselective phenylation reactions was
developed (5 mol% 3, toluene, 10 8C, 2 h) and tested on a
diverse set of aldehydes. The results of this study have been
summarized in Table 5.
Starting aldehyde
ee (%)
a-Methylcinnamaldehyde (13h)
o-Fluorobenzaldehyde (13i)
o-Tolualdehyde (13j)
m-Methoxybenzaldehyde (13l)
Pivalaldehyde (13m)
2-Naphthaldehyde (13o)
90
84
91
89
89
90
Next, some key parameters related to the use of 3 in the
reaction were optimized. On the first place, since it is known
that phenyl transfer from zinc to carbonyl groups usually
presents an isoinversion temperature,27 and the temperature
for optimal enantioselectivity had been previously
established as 10 8C working with ligand 1 in the addition
to p-tolualdehyde (13k),15 the optimization was repeated for
ligand 3 working on the same substrate. By using a 5%
molar amount of 3 in reactions at 0, 10, and 23 8C, the
As inspection of Table 5 reveals, excellent results are
obtained for p-substituted substrates under this set of
experimental conditions. On the other hand, since the
studied reaction is in general highly responsive to small
variations in experimental conditions (temperature, solvent,
catalyst amount), the possibility that higher enantiomeric

D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
Figure 4. Progress of the phenylation of 13k mediated by 3 (5 mol%) at different temperatures.
Figure 5. Progress of the phenylation of 13k mediated by 3 at 23 8C with different catalyst loadings.
12117
excesses can be achieved for some of the substrates under
different experimental conditions can not be excluded. In
this respect, it is illustrative to compare the results obtained
for 13h, 13m, and 13o using either 10 mol% (Table 4) or
5 mol% (Table 5) of resin 3.
3. Summary and outlook
In summary, the first polymeric heterogeneous ligands
(2 and 3) for the highly enantioselective phenyl transfer to
aldehydes have been developed. The high catalytic activity
depicted by these ligands, probably arising from a design
where the handle used for the anchoring of the monomers to
the polymer backbone introduces a minimal perturbation on
the catalytic center, appears to be key to this behavior.
With respect to enantioselectivity, 2 and 3 appear to be
complementary in many aspects, and experimental
conditions have been found for efficiently controlling the
enantioselectivity of the phenylation of aromatic aldehydes
(90.5% mean ee, eight examples) by a proper ligand choice.
Although fewer examples have been studied, only marginally
inferior results are recorded with a-substituted aliphatic
Table 5. Enantioselective phenylation of a selected set of aldehydes
mediated by resin 3 (5 mol%) under optimized reaction conditions
Starting aldehyde
Yield (%)
ee (%)
2-Ethylbutanal (13g)
83
82
100
97
75
79
94
68
a-Methylcinnamaldehyde (13h)
p-Tolualdehyde (13k)
Pivalaldehyde (13m)
Biphenyl-4-carbaldehyde (13n)
2-Naphthaldehyde (13o)
p-Methoxybenzaldehyde (13p)
p-Chlorobenzaldehyde (13q)
69
100
75
92
85
O99
82
80

12118
D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
and a,b-unsaturated aldehydes. With respect to catalytic
activity, reaction times of only 25 min have been
determined by on-line FTIR analysis through the use of
10 mol% amounts of ligand 3. As an application of this
property, the development of flow systems for the
continuous enantioselective phenylation of aromatic
aldehydes is being actively pursued in our laboratories
and will be reported in due course.
4.1.2. (S)-4-(2-Amino-1-hydroxy-2,2-diphenylethyl)-
benzonitrile (8). The crude azide 7 (6.21 g; 18.4 mmol)
was dissolved in MeOH (100 mL) and added via canula to a
suspension of 10% Pd/C (0.76 g) in MeOH (100 mL) under
hydrogen (1 atm). After 15 h stirring at room temperature,
the reaction mixture was filtered through a pad of Celite to
remove the catalyst, and MeOH was evaporated under
vacuum. The residue was purified by column chromato-
graphy on Et3N pre-treated SiO2 (2.5% v/v) eluting with
hexane/ethyl acetate mixtures of increasing polarity to
afford (S)-8 (4.40 g) in 85% yield [from (S)-6] as a white
solid. Mp: 166 8C. [a]2D3 K249.5 (c 1.02, CHCl3). 1H NMR
(300 MHz, CDCl3) d: 2.2–2.8 (br s, 2H), 5.6 (s, 1H), 6.8–7.7
(m, 14H) ppm. 13C NMR (75.4 MHz, CDCl3) d: 65.9 (C),
76.9 (CH), 111.0 (C), 118.8 (C), 126.6 (CH), 127.0 (CH),
127.2 (CH), 127.5 (CH), 128.0 (CH), 128.5 (CH), 130.9
(CH), 144.4 (C), 145.3 (C), 145.4 (C) ppm. IR (film, NaCl)
ymax: 3478, 3350, 3290, 3090, 3060, 2228 cmK1. MS (CI,
NH3) m/e: 315 ([MC1]C, 100%), 316 ([MC2]C, 23%).
Elemental analysis: calcd for C21H18N2O: C, 80.23; H, 5.77;
N, 8.91. Found: C, 80.21; H, 5.75; N, 8.92.
4. Experimental
4.1. General
Optical rotations were measured at 23 8C (concentration in
g/100 mL). Melting points were determined in open
capillary tubes and are uncorrected. IR spectra were
recorded as film between NaCl plates or by KBr pellet
techniques.1H and 13C NMR spectra in solution were
recorded in CDCl3. Carbon multiplicities have been
assigned by distortionless enhancement by polarization
transfer (DEPT) experiments. The NMR gel samples were
prepared as follows: the appropriate mass of resin was
placed in a 5 mm NMR tube, and the mass volume of
solvent was added. When the solvent had been absorbed,
small additional fractions of solvent were added to obtain a
homogeneous gel. The so-prepared samples were allowed to
stand for 8–12 h before recording the spectra. 13C NMR gel
phase NMR spectra were recorded at 75.4 MHz in CDCl3.
Elemental analyses were carried out by the ‘Servei
4.1.3. (S)-4-(1-Hydroxy-2,2-diphenyl-2-(pyrrolidin-1-yl)-
ethyl)benzonitrile (9). Amino alcohol (S)-8 (3.49 g,
11.1 mmol) was dissolved in absolute ethanol (21 mL).
Anhydrous potassium carbonate (3.10 g; 22.4 mmol) and
1,4-diiodobutane (2.92 mL, 22.2 mmol) were added to the
solution, and the resulting mixture was heated under reflux.
Over 3 days, additional 1,4-diiodobutane (7.6 mL;
57.4 mmol) was added in portions to the refluxing reaction
mixture. Afterwards, the reaction mixture was cooled down
and filtered, and ethanol was removed at reduced pressure.
The residue was purified by column chomatography on
Et3N pre-treated SiO2 (2.5% v/v) eluting with hexane/ethyl
acetate mixtures of increasing polarity to afford (S)-9
(2.60 g) in 64% yield as a white solid. Mp: 71 8C. [a]D23
`
d’Analisis Elementals del C.S.I.C. de Barcelona’. Tungsten(IV)
oxide was used in the resin analyses to ensure total
combustion of the samples. DMF, piperidine, and CH2Cl2
were distilled from CaH2 and stored under N2. Hexane,
THF, and Toluene were distilled from Na and stored under
N2. Barlos resins were obtained from commercial sources.
Online FTIR analysis were performed with a React IR-4000
instrument fitted with an immersible diamond (DiComp)
ATR probe from Mettler Toledo.
1
C39.6 (c 1.00, CHCl3). H NMR (300 MHz, CDCl3) d:
1.4–1.8 (br s, 4H), 2.2–2.6 (br s, 4H), 5.9 (s, 1H), 6.9–7.5
(m, 14H) ppm. 13C NMR (75.4 MHz, CDCl3) d: 22.1 (CH2),
45.9 (CH2), 71.8 (CH), 110.3 (C), 119.0 (C), 126.3 (CH),
126.9 (CH), 127.0 (CH), 127.5 (CH), 128.4 (CH), 130.5
4.1.1. (S)-4-(2-Azido-1-hydroxy-2,2-diphenylethyl)-
benzonitrile (7). Enantiomerically pure (O99.9% ee)
(S)-6 (4.8 g; 16.1 mmol), prepared according to a reported
procedure, were dissolved in 40 mL anhydrous benzene,
and added to a freshly prepared suspension of diisopro-
poxytitanium diazide (5 g; 19.9 mmol) in anhydrous
benzene (40 mL) under reflux. After 320 min, the mixture
was cooled down, benzene was removed under vacuum, and
the residue was dissolved in diethyl ether (50 mL). Aqueous
5% H2SO4 (50 mL) and the mixture was vigorously stirred
for 60 min. Phases were then separated, and the aqueous one
extracted with diethyl ether (4!50 mL). The combined
organic extracts were dried (Na2SO4), filtered and evaporated
to afford 6.21 g of crude (S)-7, that was submitted to
azide reduction without further purification. [a]2D3 K35.8
(c 1.01, CHCl3). 1H NMR (200 MHz, CDCl3) d: 2.78 (br s,
1H), 5.73 (s, 1H), 7.01–7.44 (m, 14H) ppm. 13C NMR
(50.3 MHz, CDCl3) d: 75.8 (C), 78.4 (CH), 111.4 (C), 118.7
(C), 127.3 (CH), 127.8 (CH), 128.1 (CH), 128.4 (CH), 128.5
(CH), 128.7 (CH), 130.9 (CH), 139.3 (C), 140.0 (C), 144.3
(C) ppm. IR (film, NaCl) ymax: 3457, 2228, 2109 cmK1. MS
(CI, NH3) m/e: 359 ([MC19]C, 26%), 358 ([MC18]C,
100%).
(CH), 130.7 (CH), 146.2 (C) ppm. IR (film, NaCl) ymax
:
3400, 2228 cmK1. MS (CI, NH3) m/e: 368 (MC, 100%), 369
([MC1]C, 28%). Elemental analysis: calcd for C25H24N2O:
C, 81.49; H, 6.57; N, 7.60. Found: C, 81.50; H, 6.57; N,
7.62.
4.1.4. (S)-4-(1-Hydroxy-2,2-diphenyl-2-(pyrrolidin-1-yl)-
ethyl)benzaldehyde (10). A solution of DIBALH
(2.7 mL, 2.7 mmol) in hexane was added dropwise to a
solution of (S)-9 (0.252 g, 0.70 mmol) in hexane (6.5 mL)
and diethyl ether (2 mL) at K78 8C. After 1 h, ethyl acetate
(1 mL) was slowly added, and the reaction mass was
allowed to heat to room temperature. After 20 min, a
saturated solution of NH4Cl (3 mL) was added, and the
resulting mixture was stirred at room temperature for 2 h.
The crude was then filtered through a pad of Celite, the two
phases were separated, the aqueous phase was extracted
with ethyl acetate (3!10 mL), and the combined organic
phases were dried (Na2SO4) and concentrated under
vacuum. The residue was purified by column

D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
12119
chomatography on Et3N pre-treated SiO2 (2.5% v/v) eluting
with hexane/ethyl acetate mixtures of increasing polarity to
afford (S)-10 (0.231 g) in 92% yield as a white solid. Mp:
pressure, and the residue was purified by column chomato-
graphy on Et3N pre-treated SiO2 (2.5% v/v) eluting with
hexane/ethyl acetate mixtures of increasing polarity to
afford (S)-12 (68 mg) in 83% yield as a white solid. Mp:
1
67 8C. [a]2D3 C51.1 (c 1.00, CHCl3). H NMR (300 MHz,
79 8C. [a]2D3 C34.0 (c 0.55, CHCl3). H NMR (300 MHz,
1
CDCl3) d: 1.5–1.8 (br s, 4H), 2.2–2.8 (br s, 4H), 5.9 (s, 1H),
6.8–7.6 (m, 14H), 9.8 (s, 1H) ppm. 13C NMR (75.4 MHz,
CDCl3) d: 22.1 (CH2), 45.9 (CH2), 72.1 (CH), 74.6 (C),
126.2 (CH), 126.8 (CH), 126.9 (CH), 127.4 (CH), 128.2
(CH), 128.5 (CH), 130.6 (CH), 130.8 (CH), 134.9 (C), 147.8
(C), 192.2 (CH) ppm. IR (film, NaCl) ymax: 3380, 2834,
1697 cmK1. MS (CI, NH3) m/e: 371 (MC, 100%), 372
([MC1]C, 27%).
CDCl3) d: 1.5–1.8 (br s, 4H), 2.3–2.6 (br s, 4H), 3.9 (s, 2H),
5.9 (s, 1H), 6.6–6.7 (m, 2H), 6.8–7.6 (m, 27H) ppm. 13C
NMR (75.4 MHz, CDCl3) d: 22.2 (CH2), 45.9 (CH2), 65.6
(CH2), 72.4 (CH), 74.6 (C), 86.8 (C), 125.4 (CH), 126.0
(CH), 126.6 (CH), 126.9 (CH), 127.1 (CH), 127.7(CH),
128.1 (CH), 128.7 (CH), 130.9 (CH), 137.5 (C), 138.9 (C),
144.2 (C) ppm. IR (film, NaCl) ymax: 3385 cmK1. MS (CI,
NH3) m/e: 243 (Ph3CC, 100%), 615 (MC, 9%), 616 ([MC1]C,
4%).
4.1.5. (S)-1-(4-(Hydroxymethyl)phenyl)-2,2-diphenyl-
2-(pyrrolidin-1-yl)ethanol (11). NaBH4 (71.4 mg,
1.88 mmol) was added to a solution of (S)-10 (176 mg;
0.5 mmol) in absolute ethanol (3.5 mL). After 0.5 h at room
temperature, saturated aqueous NH4Cl solution was added
dropwise, and the resulting aqueous phase was extracted
with CH2Cl2 (3!5 mL). The combined organic phases
were washed with water (3!5 mL), dried (MgSO4), filtered
and concentrated under vacuum. The residue was purified
by column chomatography on Et3N pre-treated SiO2 (2.5%
v/v) eluting with hexane/ethyl acetate mixtures of increasing
polarity to afford (S)-11 (113 mg) in 65% yield as a white
4.2. General procedure for the enantioselective addition
of ZnEt2 to aldehydes catalyzed by resin 5
Twenty three milligram of resin 5 (8 mol%, f: 0.90) were
suspended under a nitrogen atmosphere in 125 mL of
anhydrous toluene. After swelling for 24 h under slow
stirring, 0.5 mL (0.5 mmol) of diethylzinc 1 M in hexanes
were added. The mixture was cooled to 0 8C, and 125 mL of
a 2 M solution of four aldehydes (0.062 mmol of each one)
in hexanes were added dropwise. After 6 h at 0 8C, the
reaction was quenched with saturated aqueous NH4Cl,
extracted with dichloromethane (3!15 mL), and the
aqueous extracts dried over Na2SO4. The resulting solution
was analyzed by GC using a chiral b-DEX capillary column
and a FID detector. The analysis method was developed
using a racemic sample. For the particular analytical
conditions for each alcohol, see Ref. 8a.
1
solid. Mp: 79 8C. [a]D23 C18.0 (c 1.00, CHCl3). H NMR
(300 MHz, CDCl3) d: 1.5–1.8 (br s, 4H), 2.2–2.7 (br s, 4H),
4.5 (s, 2H), 5.9 (s, 1H), 6.6–6.8 (m, 2H), 6.8–7.5 (m,
12H) ppm. 13C NMR (75.4 MHz, CDCl3) d: 22.2 (CH2),
45.9 (CH2), 65.0 (CH2), 72.3 (CH), 74.5 (C), 125.3 (CH),
126.0 (CH), 126.6 (CH), 126.7 (CH), 127.1 (CH), 128.4
(CH), 130.9 (CH), 131.0 (CH), 137.3 (C), 139.4 (C), 139.5
(C) ppm. IR (film, NaCl) ymax: 3397 cmK1. MS (CI, NH3)
m/e: 373 (MC, 100%), 374 ([MC1]C, 29%).
4.3. Typical procedure for the enantioselective phenyl
transfer to aldehydes catalyzed by resins 2, 3, and 5
4.1.6. Anchoring of amino alcohol 11 to a Barlos resin:
resin 5 (fmax: 1.06) from Barlos’ resin with an initial
substitution level of 1.60 mmol Cl/g. Diisopropyletyl-
amine (0.15 mL, 0.88 mmol) was added to a mixture of
aminodiol 11 (171 mg, 0.46 mmol) and the resin (232 mg,
0.38 mmol of active Cl) in CH2Cl2 (2.5 mL), under
nitrogen, at room temperature. After smoothly stirring for
24 h, the resulting mixture was filtered, washed with DMF
(2!10 mL), DMF/water 1:1 (4!10 mL), water (4!
10 mL), pH 9 Na2CO3/NaHCO3 buffer (4!10 mL), water
(8!10 mL), MeOH (4!10 mL), toluene (4!10 mL) and
CH2Cl2 (4!10 mL), and dried under vacuum to constant
weight to afford 0.301 g (100%) of resin 12 (f: 0.896). 13C
gel-phase NMR (75.4 MHz, CDCl3) d: 22.2 (CH2), 40.4
(CH), 45.9 (CH2), 65.7 (CH2), 72.3 (CH), 74.5 (C), 86.2 (C).
Anal. Calcd for fmax: N, 1.43. Found: N, 1.30. Anchoring
yield: 85%.
In first place, a mixture of 293 mg (1.33 mmol) of ZnPh2
and 333 mg (2.7 mmol) of pure ZnEt2 was dissolved in
25 mL of anhydrous toluene. Then, the corresponding
weight of resin, according to f and to the desired molar
amount, was suspended in 6.4 mL of the ZnPh2/ZnEt2
solution under argon, and allowed to swell for 1 h. After
cooling to 10 8C, 59 mL (0.50 mmol) of p-tolualdehyde were
added. After 2 h, the reaction was quenched with saturated
aqueous NH4Cl, filtered under vacuum to remove the
catalyst, extracted with CH2Cl2 (3!15 mL), dried
(Na2SO4), and solvents removed under vacuum. The
diarylcarbynol was obtained in quantitative yield. Enantio-
meric excess was determined by HPLC with a Chiralcel OD
chiral column. For the particular analytical conditions for
each alcohol, see Ref. 15.
4.1.7. (S)-2,2-Diphenyl-2-(pyrrolidin-1-yl)-1-(4-(trityloxy-
methyl)phenyl)ethanol (12). A solution of (S)-11 (50 mg,
0.13 mmol) and N-tritylpyridinium tetrafluoroborate
(66 mg, 0.16 mmol) in acetonitrile (0.6 mL) was kept
under nitrogen at room temperature for 24 h. Diethyl ether
(5 mL) was then added, and the resulting solid material was
separated by filtration. Solvents were removed at reduced
Acknowledgements
We thank DGI-MCYT (Grant BQU2002-02459), DURSI
´
(Grant 2001SGR50), Fundacion Ramon Areces and ICIQ
Foundation for financial support. D.C. and D.F. thank MEC
for fellowships, and C.J. and L.S. thank MEC for Torres
Quevedo contracts.

12120
D. Castellnou et al. / Tetrahedron 61 (2005) 12111–12120
`
11. Puigjaner, C.; Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.;
Riera, A. J. Org. Chem. 1999, 64, 7902–7911.
References and notes
´ `
12. (a) Pasto, M.; Riera, A.; Pericas, M. A. Eur. J. Org. Chem.
´ ´
2002, 2337–2341. (b) Ferrer, M.; Pasto, M.; Rodrıguez, B.;
1. (a) Bolshan, Y.; Chen, C.-Y.; Chilenski, J. R.; Gosselin, F.;
Mathre, D. J.; O’Shea, P. D.; Roy, A.; Tillyer, R. D. Org. Lett.
2004, 6, 111–114. (b) El Ahmad, Y.; Maillet, P.; Laurent, E.;
`
Riera, A.; Pericas, M. A. Tetrahedron: Asymmetry 2003, 14,
1747–1752.
´
Talab, A.; Teste, J. F.; Cedat, M. J.; Fiez-Vandal, P. Y.;
`
13. Pericas, M. A.; Puigjaner, C.; Riera, A.; Vidal-Ferran, A.;
´ ´
Gomez, M.; Jimenez, F.; Muller, G.; Rocamora, M. Chem.
Dokhan, R.; Ollivier, R. Eur. J. Med. Chem. 1997, 32,
205–218. (c) Botta, M.; Summa, V.; Corelli, F.; Di Pietro, G.;
Lombardi, P. Tetrahedron: Asymmetry 1996, 7, 1263–1266.
(d) Stanchev, S.; Rakovska, R.; Berova, N.; Snatze, G.
Tetrahedron: Asymmetry 1995, 6, 183–189. (e) Toda, F.;
Tanaka, K.; Koshiro, K. Tetrahedron: Asymmetry 1991, 2,
873–874. (f) Meguro, K.; Aizawa, M.; Sohda, T.; Kawamatsu,
Y.; Nagaoka, A. Chem. Pharm. Bull. 1985, 33, 3787–3797.
2. (a) Harms, A. F.; Nauta, W. T. J. Med. Pharm. Chem. 1960, 2,
57–77. (b) Sperber, N.; Papa, D.; Schwenk, E.; Sherlock, M.
J. Am. Chem. Soc. 1949, 71, 887–890.
Eur. J. 2002, 8, 2164–2178.
´ `
14. Garcıa-Delgado, N.; Fontes, M.; Pericas, M. A.; Riera, A.;
Verdaguer, X. Tetrahedron: Asymmetry 2004, 15, 2085–2090.
` `
15. Fontes, M.; Verdaguer, X.; Sola, L.; Pericas, M. A.; Riera, A.
J. Org. Chem. 2004, 69, 2532–2543.
¨
16. For recent reviews, see: (a) Brase, S.; Lauterwasser, F.;
Ziegert, R. E. Adv. Synth. Catal. 2003, 345, 862–929.
(b) Special Issue on Recoverable Catalysts and Reagents;
Gladisz, J. A., Ed. Chem. Rev. 2002, 102, 3215–3892. (c)
Chiral Catalyst Immobilization and Recycling; de Vos, D. E.,
Vankelekom, I. F. J., Jacobs, P. A., Eds.; Wiley-VCH:
Weinheim, 2000. (d) Polymeric Materials in Organic
Synthesis and Catalysis; Buchmeiser, M. R., Ed.;
Wiley-VCH: Weinheim, 2003.
3. Ohkuma, T.; Koizumi, M.; Ikehira, H.; Yokozawa, T.; Noyori,
R. Org. Lett. 2000, 2, 659–662. And references therein.
4. For a review, see: Bolm, C.; Hildebrand, J. P.; Mun˜iz, K.;
Hermanns, N. Angew. Chem., Int. Ed. 2001, 40, 3284–3308.
5. Dosa, P. I.; Ruble, J. C.; Fu, G. C. J. Org. Chem. 1997, 62,
444–445.
17. Bolm, C.; Hermanns, N.; Claben, A.; Mun˜iz, K. Bioorg. Med.
Chem. Lett. 2002, 12, 1795–1798.
18. For a precedent, see: Vidal-Ferran, A.; Bampos, N.; Moyano,
6. (a) Huang, W.-S.; Pu, L. J. Org. Chem. 1999, 64, 4222–4223.
(b) Huang, W.-S.; Hu, Q.-S.; Pu, L. J. Org. Chem. 1999, 64,
7940–7956.
`
A.; Pericas, M. A.; Riera, A.; Sanders, J. K. M. J. Org. Chem.
1998, 63, 6309–6318.
7. (a) Bolm, C.; Hermanns, N.; Hildebrand, J. P.; Mun˜iz, K.
Angew. Chem., Int. Ed. 2000, 39, 3465–3467. (b) Bolm, C.;
Kesselgruber, M.; Hermanns, N.; Hildebrand, J. P.; Raabe, G.
Angew. Chem., Int. Ed. 2001, 40, 1488–1490. (c) Bolm, C.;
Mun˜iz, K. Chem. Commun. 1999, 1295–1296. (d) Hermanns,
N.; Dahmen, S.; Bolm, C.; Brase, S. Angew. Chem., Int. Ed.
2002, 41, 3692–3694. (e) Bolm, C.; Rudolph, J. J. Am. Chem.
Soc. 2002, 124, 14850–14851. (f) Rudolph, J.; Hermanns, N.;
Bolm, C. J. Org. Chem. 2004, 69, 3997–4000.
` ´ `
19. Pericas, M. A.; Castellnou, D.; Rodrıguez, I.; Riera, A.; Sola,
L. Adv. Synth. Catal. 2003, 345, 1305–1313.
`
20. Castellnou, D.; Sola, L.; Jimeno, C.; Fraile, J. M.; Mayoral,
`
J. A.; Riera, A.; Pericas, M. A. J. Org. Chem. 2005, 70,
433–438.
21. Caron, M.; Carlier, P. R.; Sharpless, K. B. J. Org. Chem. 1988,
53, 5185–5187.
22. Brandes, B. D.; Jacobsen, E. N. J. Org. Chem. 1994, 59,
4378–4380.
23. Barlos, K.; Gatos, D.; Kallitsis, J.; Papaphotiu, G.; Sotiriu, P.;
`
8. (a) Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A.
`
J. Org. Chem. 1997, 62, 4970–4982. (b) Sola, L.; Reddy, K. S.;
`
Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A.;
´
Alvarez-Larena, A.; Piniella, J. F. J. Org. Chem. 1998, 63,
` `
7078–7082. (c) Reddy, K. S.; Sola, L.; Moyano, A.; Pericas,
¨
Wenging, Y.; Schafer, W. Tetrahedron Lett. 1989, 30,
3943–3946.
24. Fragmentation of the 1,2-amino alcohol moiety was observed
under the basic conditions required for anchoring of 11 to
Merrifield resins. For precedents, see Refs. 8d and 19.
25. (a) Giralt, E.; Rizo, J.; Pedroso, E. Tetrahedron 1984, 40,
4141–4152. (b) Look, G. C.; Holmes, C. P.; Chinn, J. P.;
Gallop, M. A. J. Org. Chem. 1994, 59, 7588–7590. (c)
Murphy, M. M.; Schullek, J. R.; Gordon, E. M.; Gallop, M. A.
J. Am. Chem. Soc. 1995, 117, 7029–7030. See also Refs.
18–20.
M. A.; Riera, A. J. Org. Chem. 1999, 64, 3969–3974. (d)
` `
Reddy, K. S.; Sola, L.; Moyano, A.; Pericas, M. A.; Riera, A.
´
Synthesis 2000, 165–176. (e) Jimeno, C.; Pasto, M.; Riera, A.;
`
Pericas, M. A. J. Org. Chem. 2003, 68, 3130–3138.
`
9. (a) Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A.
Tetrahedron Lett. 1997, 38, 8773–8776. (b) Jimeno, C.;
`
Moyano, A.; Pericas, M. A.; Riera, A. Synlett 2001,
`
1155–1157. (c) Fontes, M.; Verdaguer, X.; Sola, L.;
`
Vidal-Ferran, A.; Reddy, K. S.; Riera, A.; Pericas, M. A.
26. For highly active analogues of 3, immobilized on silica by sol–gel
synthesis and grafting, see: Fraile, J. M.; Mayoral, J. A.;
` `
Serrano, J.; Pericas, M. A.; Sola, L.; Castellnou, D. Org. Lett.
Org. Lett. 2002, 4, 2381–2383.
10. (a) Jimeno, C.; Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.;
Riera, A. Tetrahedron Lett. 1999, 40, 777–780. (b) Jimeno, C.;
`
2003, 5, 4333–4335.
27. Buschmann, H.; Scharf, H.-D.; Hoffmann, N.; Esser, P.
Angew. Chem., Int. Ed. Engl. 1991, 30, 477–515.
` `
Reddy, K. S.; Sola, L.; Moyano, A.; Pericas, M. A.; Riera, A.
Org. Lett. 2000, 2, 3157–3159.