Modular optimization of enantiopure epoxide-derived P,S-ligands for rhodium-catalyzed hydrogenation of dehydroamino acids

Article abstract of DOI:10.1016/j.tet.2011.04.050

The optimization of P,S-ligands derived from enantiopure (2S,3S)-phenylglycidol for asymmetric rhodium-catalyzed hydrogenation of dehydroamino esters is described. The exceptionally high modular character of the (2S,3S)-phenylglycidol platform is demonstr

Full text of DOI:10.1016/j.tet.2011.04.050

Tetrahedron 67 (2011) 4161e4168
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Modular optimization of enantiopure epoxide-derived P,S-ligands for
rhodium-catalyzed hydrogenation of dehydroamino acids
,
,
b
a,b,
Xisco Caldentey a b, Xacobe C. Cambeiro a , Miquel A. Pericas
*
ꢀ
a Institute of Chemical Research of Catalonia (ICIQ), Av. Països Catalans 16, E-43007 Tarragona, Spain
Departament de Química Organica, Universitat de Barcelona, E-08028 Barcelona, Spain
b
ꢀ
a r t i c l e i n f o
a b s t r a c t
Article history:
The optimization of P,S-ligands derived from enantiopure (2S,3S)-phenylglycidol for asymmetric
rhodium-catalyzed hydrogenation of dehydroamino esters is described. The exceptionally high modular
character of the (2S,3S)-phenylglycidol platform is demonstrated by systematic modification of the ether
and thioether moieties, the skeletal aryl substituent and the stereo and regiochemistry of the ligands. An
experimentally useful method for the monitoring of hydrogenation reactions is also described and used
for obtaining relevant data of the catalytic systems studied.
Received 21 March 2011
Received in revised form 13 April 2011
Accepted 15 April 2011
Available online 21 April 2011
Keywords:
Asymmetric hydrogenation
P,S-ligands
Ó 2011 Elsevier Ltd. All rights reserved.
Rhodium
Dehydroamino acids
Modular ligands
1. Introduction
in many cases2 be brought to levels far superior to those intrinsically
arising from the enantioselectivity of the employed reaction by
Asymmetric catalysis is one of the most active research fields in
chemistry, having experienced an enormous growth in the last
decades. The huge and permanently growing number of chemical
processes suitable for asymmetric catalysis, as well as the large
variety of substrates to which they can be applied represent a per-
manent need for the discovery of new catalysts.
simple recrystallization,3 while the consequences of poor catalytic
activity (in terms of reactor/plant occupation) or inconvenient
reaction conditions (in terms of energy consumption) have
unavoidable, potentially very negative impacts on production costs.
From this perspective, comparison between very similar ligands,
which do not afford significant differences in terms of enantiose-
lectivity should take into account their catalytic activity as a key
factor, which strongly influences the economic viability of a dete-
rmined process, by saving time (and thus energy) and allowing
lower catalyst loadings to be used, which represents a very
important contribution to the global costs.
Recently, we have reported on the preparation and application in
Pd-catalyzed allylic substitution reactions4 of a family of modular
P,S-ligands5,6 derived from enantiopure (2S,3S)-phenylglycidol 1
(Fig. 1). This readily available, purely synthetic chiral edduct has
demonstrated in the last years to be an exceptionally versatile plat-
form for the preparation of diverse types of ligands.7 Encouraged by
these results, we envisioned to further study the behaviour of these
ligands by optimization of their structural parameters for their ap-
plication in Rh-catalyzed hydrogenation8 paying particular attention
tothe accuratecomparison of theircatalytic activities, representedby
the turnover frequency measured at half conversion (TOF1/2). Herein
we report the results from this systematic study, together with an
experimentally useful methodology for monitoring the hydrogena-
tion reactions based on the measurement of the gas consumed.
Preparation of combinatorial libraries of modular ligands for
asymmetric catalysis has been for years one of the key strategies for
fast progress towards high catalytic activity and enantioselectivity
for a broad variety of processes and substrates.1 In this context,
however, attention has been usually focused on enantioselectivity
only, prioritizing it over any other variable. Despite the undeniable
importance of methodologies for obtaining enantiopure compounds
for fine chemicals and pharmaceutical industries, catalytic activity is
a variable frequently overlooked by asymmetric catalysis academic
researchers. When a catalytic enantioselective process is analyzed
from the perspective of its potential application as a production tool
a different parameter; i.e., catalytic activity and its translation into
productivity, needs to be simultaneously considered. In this respect,
it is pertinent to recall here that enantiopurity of a synthetic material
(measured as either enantiomeric excess or enantiomeric ratio) can
* Corresponding author. Tel./fax: þ34 977 920 243; e-mail address: mapericas@
ꢀ
iciq.es (M.A. Pericas).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.04.050

4162
X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
almost all the cases (see below) the reactions proceeded to com-
pletion in relatively short reaction times (most frequently ranging
5e100 min), simply measuring conversion at a fixed time did not
constitute a practical method. Thus, we decided to use the hydro-
gen gas uptake in order to closely monitor the reactions. For this
purpose, computer-controlled reactors equipped with a pressure
probe where used (see Experimental section), this allowing accu-
rate measurements of the gas consumed.
Considering this accurate pressure measurement together with
the total amount of gas consumed (Fig. 2, left), corrected gas uptake
data in which non chemical gas expense (i.e., flushing reactors,
pressurizing, etc.) was excluded could be calculated (Fig. 2, right).
This data constitute a rough representation of the time interval in
which the reaction took place, allowing determination of the re-
action starting and ending points.
For transforming this data into a conversion against time plot,
a series of assumptions were made: (i) The dissolution of hydro-
gen gas in the reaction medium is faster than the reaction (so the
gas consumed in the solution is immediately replaced and this is
thus reflected in a small decrease in the gas phase pressure, which
is measured); (ii) the catalyst loading is small enough for
neglecting the gas consumed in reducing the catalyst; (iii) the
pressure variations throughout the hole process are small enough
to neglect the variation in the amount of hydrogen dissolved in the
reaction medium. These considerations allow us to directly relate
the gas uptake to conversion. Having in mind the 1:1 hydrogen to
substrate stoichiometry, conversion can be expressed as described
in Eq. 1.
Fig. 1. Modular construction of diverse ligands from a simple epoxide precursor.
2. Results and discussion
2.1. General strategy for the preparation of the ligands
The modular P,S-ligands (4) employed in this study were pre-
pared in a straightforward manner by ring opening with thiols of
different epoxyethers (2) derived from (2S,3S)-phenylglycidol, fol-
lowed by phosphinylation of the resulting alcohols 3 (Scheme 1), as
has been already described.4 This approach demonstrated to be
very reliable, allowing preparation of a wide variety of compounds
with good yields with little dependence on the particular structure
of the different building blocks.
nt ꢀ n
Conversionð%Þ ¼
0$100
(1)
nf ꢀ n0
where nt is the gas uptake (corrected) at a given time and n0 and nf
are the gas uptake data at the reaction starting and ending points,
respectively.
In this way, a conversion to time plot could be generated (Fig. 3).
From this graph, the time for half conversion could be determined
by linear interpolation (for accuracy, interpolation was done with
all the data between 40 and 60% conversion). Finally, the TOF1/2 was
determined by Eq. 2.
Scheme 1. Preparation of P,S-ligands from chiral enantiopure epoxides.
The structure of the ligands was modified (see below) as to provide
optimal catalytic activity and enantioselectivity in the hydrogenation
of methyl Z-a-acetamidocinnamate (Z-MAC). This reaction was used
as a benchmark for the structural optimization of P,S-ligands 4, per-
formed systematically according to their catalytic performance
(expressed in terms of turnover frequency measured at half conver-
sion (TOF1/2) and enantiomeric excess of the product obtained).
50
l$t1=2
TOF1=2
¼
(2)
where l is the catalyst loading expressed in percentage respect to
the substrate and t1/2 is the time for half conversion previously
determined.
2.2. Monitoring of the hydrogenation reactions by the gas
uptake data. Determination of TOF1/2
In addition to serving for the determination of reaction times
and TOFs, the conversion versus time plots obtained allowed ob-
servation of the kinetics of the process. In all the cases herein
With the purpose of accurately comparing the different ligands
tested, a measure of the catalytic activity was necessary. Since in
Fig. 2. Raw (left) and corrected (right) gas uptake versus time plots for a model hydrogenation reaction.

X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
4163
In all the cases, the reaction proceeded cleanly to complete
conversion in short reaction times, giving place to the pure hy-
drogenated product in quantitative yield. Ligands 4ca and 4da,
bearing more bulky benzhydryl and trityl substituents, showed
higher catalytic activities than the less hindered 4aa and 4ba. On
the other hand, little influence on the ee was observed, ligand 4aa
affording the highest.
2.4. Optimization of the reaction conditions
With these results in hand, we set on to optimize the reaction
conditions before going on with the study of the different modules
in the ligands. With this aim, some experiments were run in order
to identify the best metal precursor. A series of cationic rhodium
complexes of the general formula [Rh(diene)2]X were studied,
varying independently the diene ligands and the counterion (X).
The results of these tests are summarized in Table 2.
Fig. 3. Conversion versus time plot for a model hydrogenation reaction and linear
regression calculated with the central points (red).
Table 2
Influence of the precursor rhodium species on the hydrogenation of Z-MAC with
studied, except when alcohol type solvents were used (see Section
3.3), the curves obtained were close to perfectly linear. This implies
the reaction rate is independent of the substrate concentration. On
the other hand, the rates were linearly dependent on the hydrogen
pressure. This rate equation (order 0 on the substrate, order 1 on
hydrogen) is consistent with the oxidative addition being the rate
limiting step of the reaction.
ligand 4aaa
Entry
Diene
X
t (min)
TOF1/2 (hꢀ1
)
ee (%)
1
2
3
4
5
nbd
cod
cod
cod
cod
BF4
ꢀ
69
100
56
19
20
109
68
137
318
291
79
74
79
79
82
OTfꢀ
BF4
PF6
SbF6
ꢀ
ꢀ
ꢀ
a
All the reactions performed under the conditions described in Table 1, with the
appropriate metallic complex and ligand 4aa.
2.3. Modification of the alkoxy group
Not surprisingly, the diene ligand in the precursor complex
seemed to have little influence on activity and enantioselectivity of
the resulting catalysts (entries 1 and 3), which is consistent with
these ligands not being present in the catalytic species. No in-
duction period was observed in any case, indicating that both di-
enes are readily hydrogenated and dissociated.
As an entry to our investigation in rhodium-catalyzed hydro-
genation, we tested a series of P,S-ligands of the general structure 4
bearing alkoxy substituents (OR1) with increasing size (Scheme 2).
For the catalytic tests, Z-MAC was chosen as a reference substrate,8
and the reactions were carried out under standard conditions in
THF, with the catalysts generated in situ from cationic bisnorbor-
nadiene rhodium (I) tetrafluoroborate and under 20 atm of hy-
drogen. The results obtained are summarized in Table 1.
On the other hand, remarkable variations were registered when
the counterion was changed (entries 2e5). The best enantiose-
lectivity was registered using the SbF6
only a moderate decrease in activity respect to that recorded with
PF6 (entry 4).
Thus, using SbF6
ꢀ anion (82% ee, entry 5), with
ꢀ
ꢀ
as the counterion, the influence of variables as
solvent, hydrogen pressure and temperature were studied, with the
results shown in Table 3.
Table 3
Optimization of the reaction conditions for the rhodium-catalyzed hydrogenation of
Z-MAC with ligand 4aaa
Scheme 2. Variation of the ether R1 substituent in epoxide-derived P,S-ligands 4.
Entry
Solvent
P (bar)
T (ꢁC)
t (min)
TOF1/2 (hꢀ1
)
ee (%)
1
2
3
4
5
6
7
8
Toluene
CH2Cl2
THF
MeCN
IPA
MeOH
THF
THF
THF
THF
THF
THF
20
20
20
20
20
20
5
10
50
10
10
10
30
30
30
30
30
30
30
30
30
50
75
100
116
16
59
367
291
w3c
85d
53
76
82
36
63
20
79
82
77
63
58
55
Table 1
20
eb
116
110
75
Rhodium-catalyzed asymmetric hydrogenation of Z-MAC with P,S-ligands 4. Effect
of the ether R1 substituenta
124d
79
38
8
15
6.7
3.5
156
773
563
1706
4680
9
10
11
12
Ligand
t b (min)
TOF1/2c (hꢀ1
)
eed (%)
4aa
4ba
4ca
4da
69
63
36
36
109
120
198
208
79
75
77
77
a
All the reactions performed under the conditions described in Table 1, with
[Rh(cod)2]SbF6 and ligand 4aa.
b
The reaction proceeded to 23% conversion in 8 h, and reduction of the solvent to
ethylamine was observed by 1H NMR.
a
c
All the reactions run with 50 mg of substrate and 0.1 M concentration.
Time for reaction completion, monitored by the gas uptake data.
TOF calculated at 50% conversion.
TOF at 23% conversion.
The reaction rate was not constant in these cases, but it fastly decayed with time,
b
c
d
in contrast with that observed for aprotic solvents. This suggests alcoholysis of the
phosphinite taking place in the reaction medium.
d
Determined by HPLC with a chiral stationary phase.

4164
X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
The nature of the solvent turned out to be an important factor,
Finally, ligands resulting from the incorporation of alkyl thiols
were tested (4ah, 4ai and 4aj). All of them showed a catalytic ac-
tivity in the same order of that observed with 4ac. In terms of
enantioselectivity, the ligand bearing an iPr group (4ah) turned out
to be the best one, affording the product in 82% ee. Interestingly,
4ah showed also a rather high (397) TOF1/2 value.
the results obtained for the reaction in THF being optimal. In ace-
tonitrile, conversion was very low even after 8 h, due to hydroge-
nation of the solvent as a competitive process. Alcohols, such as
iPrOH or MeOH resulted also detrimental for the reaction. On the
other hand, no improvement was obtained by changing pressure or
temperature.
2.6. Modification of the skeletal carbon substituent
2.5. Variation of the thioether moiety
Having in mind that the sulfur atom becomes a stereogenic
centre upon coordination, whose configuration is mainly controlled
by steric repulsion between the thioether (R2) and the skeletal aryl
(Ar) substituents,5b,9 we thought that a bulkier aryl substituent was
likely to increase the enantioselectivity by more strongly fixing the
configuration of this newly formed chiral centre. With this in mind,
ligand 6 was prepared from epoxide 59 by the same sequence de-
scribed before for ligands of type 4 (Scheme 4).
With these results in hand, we turned our attention back to the
optimization of the different structural variables on the phosphi-
nyloxy thioether P,S-ligands. With this purpose, a series of ligands
was prepared by using different thiols as nucleophiles in the ring
opening step (Scheme 3).
Scheme 4. Modification of the skeletal aryl (Ar) substituent.
Test of this ligand in hydrogenation (Scheme 5), unfortunately,
showed no improvement respect to the phenyl-substituted ana-
logue 4ac. The product of hydrogenation was obtained with slightly
decreased ee (81%) in a significantly longer reaction time.
Scheme 3. Modification of the thioether R2 group in epoxide-derived P,S-ligands 4.
All these ligands were tested in the reference reaction using the
conditions described in Table 3, entry 3 (i.e., [Rh(cod)2]SbF6 as the
metal source, in THF, at 30 ꢁC and under 20 bar hydrogen pressure).
The results obtained are shown in Table 4.
Scheme 5. Rhodium-catalyzed hydrogenation of Z-MAC with ligand 6.
Table 4
Influence of the modification of the R2 thioether substituent on the catalytic per-
formance of epoxide-derived P,S-ligands 4 on the rhodium-catalyzed hydrogenation
of Z-MACa
2.7. Variation of regio- and relative stereochemistry
Ligand
t (min)
TOF1/2 (hꢀ1
)
ee (%)
Finally, as another approach for introducing diversity in our li-
gands, we decided to modify the relative configuration of the ligands,
as well as their regiochemistry. Thus, the syn diastereoisomer of 4ac
was prepared, as well as both diastereoisomers of the C2-opening
regioisomeric product 10.
For the preparation of 10, a sulfur 1,2emigration was used,
through formation and opening of an episulfonium ion under
Mitsunobu conditions (Scheme 6).7c,10
On the other hand, syn-10 and syn-4ac were prepared from the
cis epoxide 1112 (Scheme 7). Although the opening of the trans
epoxide 2a proceeded in a completely regioselective manner, as
explained before, the situation was completely different with the
cis stereoisomer. Thus, opening of the cis epoxide 11 afforded
a mixture of both regioisomers syn-9 and syn-3ac. After separa-
tion, phosphinylation gave place to the desired ligands syn-10 and
syn-4ac.
4aa
4ab
4ac
4ad
4ae
4af
4ag
4ah
4ai
20
40
17
13
17
5
40
17
18
19
291
182
284
620
349
2249
148
397
432
331
82
76
84
59
73
64
80
82
72
74
4aj
a
All the reactions performed under the conditions described in Table 3, entry 3,
with the appropriate ligand in each case.
Ligands bearing aryl substituents with different substitution
patterns were tested and compared to the original 4aa. A most
favourable compromise of different steric parameters was attained
with ligand 4ac, bearing a 3,5-dimethylphenyl substituent on sul-
fur, which afforded complete conversion to the hydrogenated
product in short time and with 84% ee.
All three ligands gave place to rhodium complexes that were
active as catalysts for the model reaction (Table 5).
Regarding the operation of electronic effects, it is worth to note
the extremely fast reaction recorded with ligand 4af, bearing an
electron-rich 4-methoxyphenyl substituent. Unfortunately, this
dramatic increase in catalytic activity was associated with a signif-
icant decrease in enantioselectivity.
Although all the ligands were active in the reaction, particularly
syn-4ac, which afforded a w6-fold increase in reaction rate com-
pared to 4ac, in all the cases the enantioselectivities were also
significantly reduced. It is worth to note that with ligand syn-10,
with S configuration in the chiral centre in position
a to sulfur, the

X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
4165
the electronic properties of the sulfur substituent, as shown by the
very high catalytic activities registered with ligands syn-4ac (pre-
senting a 10-fold increase respect to its anti analogue 4ac) and 4af,
bearing an electron-donating substituent on sulfur.
Scheme 6. Preparation of regioisomeric ligand 10 through formation of an epis-
ulfonium intermediate.
Fig. 4. Data dispersion of % ee versus TOF1/2 for the rhodium-catalyzed hydrogenation
of Z-MAC with all the ligands studied under the optimized conditions.
3. Conclusion
Scheme 7. Preparation of epoxide-derived P,S-ligands with syn relative configuration.
In summary, a complete, systematic study on the influence of
the different structural parameters of epoxide-derived P,S-ligands
family 4 has been done, demonstrating the extremely modular
nature of these ligands. Catalytic activity and kinetics data obtained
by monitoring the reactions gas uptake, together with the enan-
tioselectivity data, have allowed determination of the key struc-
tural variables (steric and electronic properties of the sulfur
substituent and relative stereochemistry), which will lead to the
development of new and more efficient ligand families.
Table 5
Effect of the regiochemistry and relative configuration on the catalytic performance
of P,S-ligands for rhodium-catalyzed hydrogenation of Z-MACa
Ligand
t (min)
TOF1/2 (hꢀ1
)
ee (%)
4ac
10
syn-4ac
syn-10
17
11
7
284
1139
1765
629
84
38
61
17
45b
a
All the reactions performed under the conditions described in Table 3, entry 3,
with the appropriate ligand in each case.
b
The product was obtained with opposite absolute configuration (S).
4. Experimental section
4.1. Materials
product was obtained with the opposite absolute configuration.
Thus, it is this centre the one controlling the stereochemical out-
come of the reaction.
Unless otherwise stated, all commercial reagents were used as
received and all reactions were carried out directly under open air.
Reactions under inert atmosphere were performed using standard
Schlenk techniques with oxygen and water free solvents and re-
agents, obtained as follows: DMF, THF, DCM, toluene and acetoni-
trile were obtained from a solvent purification system, with
<5 ppm of water (Karl-Fischer analysis); methanol, 2-propanol,
hexane and ethyl acetate were dried by prolonged contact with
2.8. Analysis of the (2S,3S)-phenylglycidol-derived P,S-ligands
structural space
All the data collected in the former sections (3.1e3.6) afford
a comprehensive sight on the structural space available for P,S-
ligands derived from (2S,3S)-phenylglycidol. A representation of
the TOF1/2 and ee obtained in the hydrogenation reaction with each
individual ligand (Fig. 4) allows us to determine the main structural
variables leading to high enantioselectivity or catalytic activity.
Thus, the higher enantioselectivities where those registered for
some of the 4ax-type ligands, derived from direct ring opening of
epoxide 2a with thiols, followed by phosphinylation, bearing a xylyl
(4ac), phenyl (4aa) or 2-propyl (4ah) substituents on the sulfur
atom. On the other hand, the results make obvious the dramatic
effect on catalytic activity of two main variables, i.e., the relative
stereochemistry of the two stereogenic centres in the ligand and
ꢁ
activated 4 A molecular sieves and degassed by bubbling argon;
ꢁ
deuterated chloroform was dried by prolonged contact with 4 A
molecular sieves, distilled and degassed by vacuumeargon cycles
at low temperature, then stored under inert atmosphere protected
from light; triethylamine was dried by prolonged contact with
powdered calcium hydride, distilled, degassed by bubbling argon
and stored over pieces of calcium hydride under inert atmosphere;
chlorodiphenylphosphine was degassed prior to use by high vac-
uumeargon cycles to remove hydrochloric acid; hydroxythioethers
were dried prior to phosphinylation by azeotropic distillation of
toluene.

4166
X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
4.2. Instrumentation
CDCl3) 6.82 (s, 2H), 3.88 (dꢂd, 1H, 2J¼11.4 Hz, 3J¼2.9 Hz), 3.81 (d,
1H, 3J¼2.3 Hz), 3.56 (dꢂd, 1H, 2J¼11.4 Hz, 3J¼5.3 Hz), 3.46 (s, 3H),
3.13e3.16 (m, 1H), 2.35 (s, 6H), 2.26 (s, 3H). 13C NMR (100 MHz,
CDCl3) 137.4 (C), 137.0 (C), 130.3 (C), 128.6 (CH), 72.6 (CH2), 59.3
(CH3), 58.3 (CH), 54.4 (CH), 20.8 (CH3), 19.6 (CH3). HRMS (ESþ): m/z
found: 229.1205 (MþNa), calculated for C13H18O2Naþ: 229.1204. IR
2977, 2887, 1729, 1615, 1453, 1374, 1315, 1199, 954, 850, 816, 703.
All flash chromatography was carried out using 60 mesh silica
gel and dry-packed columns. Catalytic tests in hydrogenation were
performed in a computer-controlled AMTEC SPR-16 parallel reactor,
equipped with a pressure probe that allows registering the gas
uptake data from the reaction. NMR spectra were registered in
a Bruker Advance 400 Ultrashield spectrometer in CDCl3 at room
temperature, operating at 400.13 MHz for 1H (100.63 MHz for 13C
and 162.18 MHz for 31P). Chemical shifts are reported in parts per
million referred toTMS (1H and 13C) or 85% phosphoric acid (31P). IR
spectra were recorded on a Bruker Tensor 27 FT-IR spectrometer.
Elemental analyses were performed on a LECO CHNS 932 micro-
analyzer at the Universidad Complutense de Madrid, Spain. Melt-
4.4.2. (1R,2S)-1-(3,5-Dimethylphenylthio)-1-mesityl-3-methox-
ypropan-2-ol (6b). To a solution of the epoxyether 5b (30 mg,
0.15 mmol) and sodium hydroxide (30 mg, 0.75 mmol) in dioxa-
neewater (10:1 v/v) was added the corresponding thiol (113 ml,
0.75 mmol). The mixture was warmed up to 80 ꢁC and stirred at this
temperature for 120 min while monitoring the progress of the re-
action by TLC. After disappearance of the epoxide, the mixture was
allowed to cool down to room temperature; 10 mL of water was
added and the mixture was extracted with CH2Cl2 (3ꢂ15 mL). The
combined organic extracts were dried with Na2SO4, the solvent was
removed under vacuum and the crude was purified by flash chro-
matography on SiO2 eluting with 9:1 hexaneeethyl acetate to ob-
tain the title product as a white solid (42 mg, 81% yield). 1H NMR
(400 MHz, CDCl3) 6.98 (s, 2H), 6.87 (s, 1H), 6.84 (s, 1H), 6.78 (s, 1H),
4.77 (d, 1H, 3J¼10,3 Hz), 4.33e4.38 (m, 1H), 3.82 (dꢂd, 1H,
2J¼9.7 Hz, 3J¼2.6 Hz), 3.63 (dꢂd, 2J¼9.7 Hz, 3J¼5 Hz), 3.32 (s, 3H),
€
ing points were determined using a Buchi melting point apparatus.
FAB mass spectra were obtained on a Fisons V6-Quattro in-
strument, ESI mass spectra were obtained on a Waters LCT Premier
instrument and CI and EI spectra were obtained on a Waters GCT
spectrometer. Optical rotations were measured on a Jasco P-1030
polarimeter. High performance liquid chromatography (HPLC) was
performed on Agilent Technologies chromatographs (Serie1200),
using Chiralpak OJ-H column and guard column.
4.3. Experimental procedures for the preparation of P,S-
ligands
2.52 (s, 3H), 2.24 (s, 9H), 2.16 (s, 3H), 2.03 (d, 3J¼4.43 Hz, OH). 13
C
NMR (100 MHz, CDCl3) 138.3 (C), 137.4 (C), 136.9 (C), 136.7 (C), 135.7
(C), 134.5 (C), 133.0 (C), 131.2 (CH), 129.6 (CH), 129.1 (CH), 128.9
(CH), 74.0 (CH2), 72.6 (CH), 58.9 (CH3), 50.7 (CH), 21.3 (CH3), 21.3
(CH3), 21.1 (CH3), 20.8 (CH3). HRMS (ESþ): m/z found: 367.1710
4.3.1. General procedure for the ring opening of epoxides by thio-
ls4. To a solution of the epoxide (1 mmol) and sodium hydroxide
(2 mmol) in dioxaneewater (10:1 v/v) the corresponding thiol
(2 mmol, 2 equiv) was added. The mixture was heated at the in-
dicated temperature, and reaction progress was monitored by TLC
until disappearance of the starting epoxide was observed (ca.
20e90 min). The mixture was then allowed to reach room tem-
perature, when 10 mL of water were added, and the mixture was
extracted with CH2Cl2 (3ꢂ15 mL). The combined organic extracts
were dried over Na2SO4, the solvent was removed under reduced
pressure and the crude product was purified by flash chromatog-
raphy on silica gel, eluting with hexaneeethyl acetate mixtures to
give the desired product in pure form.
(MþNa), calculated for C21H28O2NaSþ: 367.1708. [
a
]
D26
ꢀ266.73 (c
0.2, CHCl3). IR 3444, 2916, 2859, 1599, 1579, 1450, 1376, 1191, 1121,
1097, 1062, 1032, 848, 689.
4.4.3. ((1R,2S)-1-(3,5-Dimethylphenylthio)-1-mesityl-3-methox-
ypropan-2-yloxy)diphenylphosphine (6). Under inert atmosphere,
b
-hydroxysulfide 6b (20 mg, 60
were dissolved in toluene (0.28 M) at room temperature. Then,
triethylamine (10 l, 70 mol) and chlorodiphenylphosphine (11 l,
60 mol) were sequentially added via syringe. The reaction was
mmol) and DMAP (0.7 mg, 6 mmol)
m
m
m
m
stirred for 20 min before the solvent was removed under vacuum,
4.3.2. General procedure for the phosphinylation of
b
-hydroxy thio-
and the reaction was diluted with 95:5 hexaneeethyl acetate
(0.5 mL, dried over 4 A molecular sieves and degassed with argon).
ethers4. To
a
Schlenk flask containing -hydroxy thioether
b
ꢁ
(0.55 mmol) and DMAP (0.055 mmol, 0.1 equiv) in toluene (0.28 M)
at room temperature, NEt3 (0.66 mmol, 1.2 equiv) and chlor-
odiphenylphosphine (0.55 mmol, 1.01 equiv) were added via sy-
ringe and the reaction was stirred for 20 min. The solvent was
removed under reduced pressure and the reaction crude was di-
luted with 95:5 hexaneeethyl acetate mixture (0.5 mL). The
resulting slurry was loaded onto a plug of silica and purified by
flash chromatography (1ꢂ3 cm, 95% hexane, 5% ethyl acetate).
The resulting slurry was loaded onto a plug of silica and purified by
flash chromatography (1ꢂ3 cm, 95% hexane, 5% ethyl acetate) to
yield product 6 as a clear oil (23 mg, 75% yield). 1H NMR (400 MHz,
CDCl3) 7.35e7.40 (m, 2H), 7.23e7.28 (m, 3H), 7.10e7.14 (m, 1H),
6.97e7.01 (m, 4H), 6.82 (s, 1H), 6.62e6.70 (m, 4H), 4.96 (d, 1H,
3J¼10,8 Hz), 4.64e4.7 (m, 1H), 3.83 (dꢂd, 1H, 2J¼10 Hz, 3J¼1.03 Hz),
3.69 (dꢂd, 2J¼10 Hz, 3J¼4.4 Hz), 3.01 (s, 3H), 2.44 (s, 3H), 2.23 (s,
6H), 2.20 (s, 3H), 2.14 (s, 3H). 31P NMR (162 MHz, CDCl3) 117.9. 13
C
NMR (100 MHz, CDCl3) 143.6 (C, d, J¼17.9 Hz), 142.2 (C, d, J¼14 Hz),
138.3 (C), 136.9 (C), 136.6 (C), 136.0 (C), 136.0 (C), 134.2 (C), 130.9
(CH), 130.0 (CH, d, J¼21.9 Hz), 130.0 (CH, d, J¼21.9 Hz), 129.5 (CH),
128.9 (CH), 128.8 (CH), 128.7 (CH), 128.3 (CH), 127.9 (CH, d,
J¼7.1 Hz), 127.5 (CH, d, J¼6.6 Hz), 82.1 (CH, d, J¼19.1 Hz), 73.6 (CH2,
d, J¼2.9 Hz), 58.2 (CH3), 50.1 (CH, d, J¼6 Hz), 21.5 (CH3), 21.4 (CH3, d,
J¼2.5 Hz), 21.1 (CH3), 20.8 (CH3). HRMS (ESþ): m/z found: 529.2313
4.4. Preparation and characterization of new compounds
4.4.1. (2S,3S)-2-Mesityl-3-(methoxymethyl)oxirane (5b). A solution
of the epoxide 59 (300 mg, 1.56 mmol) in DMF (3 mL) was added via
cannula to a suspension of sodium hydride (44 mg, 1.81 mmol) in
DMF (2 mL) at ꢀ20 ꢁC under inert atmosphere. The mixture was
(M), calculated for C33H38O2PSþ: 529.2330. [
CHCl3). IR 2970, 2920, 1738, 1435, 1375, 1229, 1217, 1127, 1094, 1053,
739, 697.
a]
ꢀ123.7 (c 0.33,
D26
stirred for 20 min, and methyl iodide (126
ml, 2.03 mmol) was sy-
ringed into the mixture. After being stirred for 4 h at ꢀ20 ꢁC, the
mixture was allowed to reach room temperature and stirred for
a further hour. MeOH (10 mL) and brine (10 mL) were added. The
aqueous solution was extracted with CH2Cl2 (3ꢂ20 mL). The com-
bined organic extracts were dried and concentrated in vacuo. The
residual oil was purified by flash chromatography using hex-
aneeEt2O (9:1 to 7:3) as eluent to give 300 mg (93%) of the epox-
4.4.4. (1S,2R)-2-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propan-1-ol (9). A solution of 3ac (150 mg, 0.5 mmol), PPh3
(657 mg, 2.48 mmol) and 4-nitrobenzoic acid (93 mg, 0.55 mmol) in
toluene (4.5 mL) and THF (4.5 mL) under inert atmosphere was
D27
]
yether (5b) as an oil. [
a
ꢀ13.4 (c 0.44, CHCl3). 1H NMR (400 MHz,
cooled down to ꢀ20 ꢁC. Then, DEAD (403
ml, 2.48 mmol) was

X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
4167
syringed into the solution. The mixture was stirred at ꢀ20 ꢁC for
3 h, allowed to reach room temperature and stirred for further 14 h
at room temperature. The solvents were removed in vacuo, and the
residual oil was chromatographed using hexaneeEtOAc (95:5) as
eluent to give 157 mg (80% yield) of a mixture of regioisomers 7 and
8 with a 1:7.5 ratio.
1.37 mmol) and NaOH (55 mg, 1.37 mmol) at 55 ꢁC for 2 h. After
purification by flash chromatography (9:1 hexaneeethyl acetate),
both regioisomers could be separately isolated (syn-3ac, 80 mg, 57%
yield, syn-9, 53 mg, 38% yield).
4.4.8. (1R,2R)-1-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propan-2-ol (syn-3ac). 1H NMR (400 MHz, CDCl3) 7.19e7.29 (m,
5H), 6.89 (s, 2H), 6.82 (s, 1H), 4.27 (d, 1H, 3J¼8.4 Hz), 4.04e4.09
(m, 1H), 3.4 (dꢂd, 1H, 2J¼9.8 Hz, 3J¼3.1 Hz), 3.27 (s, 3H), 3.17
(dꢂd, 1H, 2J¼9.8 Hz, 3J¼6 Hz), 3.00 (br s, OH), 2.20 (s, 6H). 13C
NMR (100 MHz, CDCl3) 139.7 (C), 138.3 (C), 133.2 (C), 130.3 (CH),
129.3 (CH), 128.4 (CH), 128.2 (CH), 127.4 (CH), 73.6 (CH2), 72.9
(CH), 59.1 (CH3), 58.1 (CH), 21.1 (CH3). HRMS (ESþ): m/z found:
325.1241 (MþNa), calculated for C18H22O2NaSþ: 325.1238. IR
3438, 2914, 1599, 1579, 1491, 1451, 1190, 1120, 1074, 962, 847, 748,
699, 686.
To this mixture (141 mg, 0.31 mmol) in CH2Cl2 (2 mL) at
ꢀ20 ꢁC under inert atmosphere, DIBAL-H 1 M in hexanes (2.5 mL,
2.5 mmol) was added dropwise. After stirring at ꢀ20 ꢁC for 19 h,
brine (20 mL) and CH2Cl2 (30 mL) were added carefully, and the
mixture was stirred vigorously. The organic phase was separated
and the aqueous layer extracted with CH2Cl2 (2ꢂ30 mL). The
combined organic extracts were dried and concentrated in vacuo.
The residual oil was chromatographed using hexaneeEtOAc
mixture (99:1 to 95:5) as eluent to give 55 mg (58% yield) of 9. 1H
NMR (400 MHz, CDCl3) 7.32e7.35 (m, 4H), 7.26e7.29 (m, 1H), 6.99
(s, 2H), 6.87 (s, 1H), 4.93e4.96 (m, 1H), 3.73 (d, 1H, 3J¼5.2 Hz),
3.61e3.66 (m, 1H), 3.47e3.54 (m, 1H), 3.35 (s, 3H), 2.27 (s, 6H). 13C
NMR (100 MHz, CDCl3) 141.4 (C), 138.7 (C), 133.6 (C), 130.0 (CH),
129.3 (CH), 128.2 (CH), 127.6 (CH), 126.4 (CH), 75.0 (CH), 73.0
(CH2), 59.1 (CH3), 55.5 (CH), 21.1 (CH3). HRMS (ESþ): m/z found:
325.1239 (MþNa), calculated for C18H22O2NaSþ: 325.1238. IR
3438, 2914, 1599, 1579, 1491, 1451, 1190, 1120, 1074, 962, 847, 748,
699, 686.
4.4.9. (1S,2S)-2-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propan-1-ol (syn-9). 1H NMR (400 MHz, CDCl3) 7.27e7.43 (m, 5H),
6.92 (s, 2H), 6.84 (s, 1H), 4.98 (m, 1H), 3.51e3.56 (m, 1H), 3.4e3.47
(m, 3H), 3.35 (s, 3H), 2.24 (s, 6H), 1.56 (br s, 1H). 13C NMR (100 MHz,
CDCl3) 141.1 (C), 138.6 (C), 133.7 (C), 129.8 (CH), 129.2 (CH), 128.2
(CH), 127.7 (CH), 126.6 (CH), 74.0 (CH), 73.2 (CH2), 59.0 (CH3), 57.5
(CH), 21.1 (CH3). HRMS (ESþ): m/z found: 325.1241 (MþNa), cal-
culated for C18H22O2NaSþ: 325.1238. IR 3052, 3027, 2916, 2893,
1599, 1580,1492, 1480, 1466,1453,1434,1129, 1094,1074,1026, 967,
848, 741.
4.4.5. ((1S,2R)-2-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propoxy)diphenylphosphine (10). A procedure analogous to that
described for compound
0.165 mmol), DMAP (2 mg, 0.016 mmol), NEt3 (28
and chlorodiphenylphosphine (31 l, 0.165 mmol). After purifica-
6
was applied, using
9
(50 mg,
m
l, 0.2 mmol)
4.4.10. ((1R,2R)-1-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propan-2-yloxy)diphenylphosphine (syn-4ac). A procedure analo-
gous to that described for compound 6 was applied, using syn-3ac
m
tion, compound 10 was obtained as a clear oil (53 mg, 78% yield).
1H NMR (400 MHz, CDCl3) 7.56e60 (m, 2H), 7.3e7.36 (m, 7H),
7.17e7.26 (m, 6H), 6.8 (s, 2H), 6.78 (s, 1H), 5.18e5.21 (m, 1H),
3.61e3.67 (m, 2H), 3.44e3.49 (m, 1H), 3.23 (s, 3H), 2.18 (s, 6H). 13C
NMR (100 MHz, CDCl3) 142.2 (C, d, J¼15.9 Hz), 142.2 (C, d,
J¼18.9 Hz), 139.1 (C, d, J¼2.23 Hz), 138.2 (C), 134.6 (C), 131.0 (CH, d,
J¼22.7 Hz), 130.3 (CH, d, J¼21.5 Hz), 130.0 (CH), 129.4 (CH), 128.9
(CH), 128.8 (CH), 128.2 (CH, d, J¼7.1 Hz), 128.0 (CH, d, J¼6.6 Hz),
127.9 (CH), 127.9 (CH), 127.8 (CH), 81.4 (CH, d, J¼21.6 Hz), 72.0
(CH2), 58.6 (CH3), 56.0 (CH, d, J¼6.3 Hz), 21.1 (CH3). 31P NMR
(162 MHz, CDCl3) 114.9. HRMS (ESþ): m/z found: 509.1673 (MþNa),
calculated for C30H31O2NaPSþ: 509.1680. IR 3052, 3027, 2916,
2893, 1599, 1580, 1492, 1480, 1466, 1453, 1434, 1129, 1094, 1074,
1026, 967, 848, 741.
(48 mg, 0.16 mmol), DMAP (2 mg, 0.016 mmol), NEt3 (26
ml,
0.19 mmol) and chlorodiphenylphosphine (29 l, 0.16 mmol).
m
After purification, syn-4ac was obtained as a clear oil (55 mg, 71%
yield). 1H NMR (400 MHz, CDCl3) 7.46e7.55 (m, 4H), 7.24e7.36
(m, 8H), 7.16e7.18 (m, 3H), 6.82 (s, 2H), 6.74 (s, 1H), 4.55 (d, 1H,
3J¼5.6 Hz), 4.38e4.44 (m, 1H), 3.63 (dꢂd, 1H, 2J¼10 Hz,
3J¼4.4 Hz), 3.27 (dꢂd, 1H, 2J¼10 Hz, 3J¼5.6 Hz), 3.09 (s, 3H), 2.16
(s, 6H). 13C NMR (100 MHz, CDCl3) 143.1 (C, d, J¼18.3 Hz), 142.2
(C, d, J¼14.8 Hz), 140.0 (C), 138.1 (C), 134.9 (C), 130.8 (CH, d,
J¼22.5 Hz), 130.2 (CH, d, J¼22.3 Hz), 129.1 (CH), 129.0 (CH), 128.9
(CH), 128.7 (CH), 128.4 (CH), 128.1 (CH, d, J¼5.1 Hz), 128.1 (CH),
128.0 (CH, d, J¼5.8 Hz), 127.1 (CH), 82.9 (CH, d, J¼19.1 Hz), 73.4
(CH2, d, J¼3.5 Hz), 58.7 (CH3), 56.3 (CH, d, J¼5.9 Hz), 21.1 (CH3).
31P NMR (162 MHz, CDCl3) 117.9. HRMS (ESþ): m/z found:
487.1838 (MꢀH), calculated for C30H32O2PSþ: 487.1868. IR 3052,
3027, 2916, 2893, 1599, 1580, 1492, 1480, 1466, 1453, 1434, 1129,
1094, 1074, 1026, 967, 848, 741.
4.4.6. (2R,3S)-2-(Methoxymethyl)-3-phenyloxirane (11). A solution
of ((2R,3S)-3-phenyloxiran-2-yl)methanol11b (150 mg, 1 mmol) in
DMF (1.4 mL) was added via cannula to a suspension of sodium
hydride (44 mg, 1.1 mmol) in DMF (2 mL) at ꢀ20 ꢁC under inert
atmosphere. The mixture was stirred for 20 min before methyl
4.4.11. ((1S,2S)-2-(3,5-Dimethylphenylthio)-3-methoxy-1-phenyl-
propoxy)diphenylphosphine (syn-10). A procedure analogous to
that described for compound 6 was applied, using syn-9 (38 mg,
iodide (82 mL, 1.3 mmol) was syringed into the mixture. After being
stirred for 4 h at ꢀ20 ꢁC, the mixture was allowed to reach room
temperature and stirred for further 2 h. MeOH (10 mL) and brine
(10 mL) were then added, the aqueous solution was extracted with
CH2Cl2 (3ꢂ15 mL) and the combined organic extracts were dried
and concentrated in vacuo. The residual oil was chromatographed
using hexaneeEt2O (9:1 to 7:3) as eluent to give 140 mg (86% yield)
of epoxyether 11 as a clear oil. All the spectroscopic data matched
those previously described.11a
0.13 mmol), DMAP (1.6 mg, 0.013 mmol), NEt3 (21
ml, 0.15 mmol)
and chlorodiphenylphosphine (23 l, 0.13 mmol). After purification,
m
syn-10 was obtained as a clear oil (45 mg, 74% yield). 1H NMR
(400 MHz, CDCl3) 7.5e7.54 (m, 2H) 7.27e7.4 (m, 7H), 7.17e7.23 (m,
6H), 6.67 (s, 1H), 6.63 (s, 1H), 5.29 (dꢂd, 1H, J¼9.4, 2.9 Hz),
3.38e3.46 (m, 1H), 3.27e3.34 (m, 2H), 3.02 (s, 3H), 2.08 (s, 6H). 13C
NMR (100 MHz, CDCl3) 142.1 (C), 141.9 (C, d, J¼4.1 Hz), 140.4 (C, d,
J¼2.1 Hz), 138.3 (C), 135.3 (C), 131.2 (CH, d, J¼23.2 Hz), 130.3 (CH, d,
J¼22.0 Hz), 129.5 (CH), 129.0 (CH), 128.7 (CH), 128.4 (CH), 128.2 (CH,
d, J¼7.3 Hz), 128.1 (CH, d, J¼7.0 Hz), 127.9 (CH), 128.6 (CH), 127.3
(CH), 80.1 (CH, d, J¼21.2 Hz), 72.3 (CH2), 58.4 (CH3), 57.4 (CH, d,
J¼6 Hz), 21.1 (CH3). 31P NMR (162 MHz, CDCl3) 115.4. HRMS (ESþ):
m/z found: 509.1665 (MþNa), calculated for C30H31O2NaPSþ:
1H NMR (400 MHz, CDCl3) 7.28e7.37 (m, 5H), 4.14 (d, 1H,
3J¼4.3 Hz), 3.40e3.44 (m, 1H), 3.33 (dꢂd, 1H, 2J¼11.4 Hz 3J¼4.4 Hz),
3.28 (s, 3H), 3.22 (dꢂd, 1H, 2J¼11.4 Hz 3J¼6.5 Hz).
4.4.7. Thiolate ring opening of cis epoxide 11. A procedure analo-
gous to that described for compound 6b was applied, using the cis
epoxide 11 (75 mg, 0.46 mmol), 3,5-dimethylbenzenethiol (192 ml,

4168
X. Caldentey et al. / Tetrahedron 67 (2011) 4161e4168
3. For solid materials arising from metal-catalyzed reactions, crystallization is
compulsory in order to avoid metal impurities, irrespectively of the enantio-
selectivities of the reactions leading to them.
509.1680. IR 3052, 3027, 2916, 2893, 1599, 1580, 1492, 1480, 1466,
1453, 1434, 1129, 1094, 1074, 1026, 967, 848, 741.
ꢀ
4. Caldentey, X.; Pericas, M. A. J. Org. Chem. 2010, 75, 2628e2644.
5. For pioneering work on modular phosphinite-thioether ligands, see (a) Evans,
D. A.; Campos, K. R.; Tedrow, J. S.; Michael, F. E.; Gagne, M. R. J. Am. Chem. Soc.
2000, 122, 7905e7920; (b) Evans, D. A.; Michael, F. E.; Tedrow, J. S.; Campos, K.
R. J. Am. Chem. Soc. 2003, 125, 3534e3543.
4.5. General procedure for the asymmetric hydrogenation of
Z-MAC
ꢂ
mmol) and
6. For other approaches to P,S-ligands, see: (a) Herrmann, J.; Pregosin, P. S.;
Salzmann, R.; Albinati, A. Organometallics 1995, 14, 3311e3318; (b) Hiroi, K.;
Suzuki, Y. Tetrahedron Lett. 1998, 39, 6499e6502; (c) Enders, D.; Peters, R.;
Runsink, J.; Bats, J. W. Org. Lett. 1999, 1, 1863e1866; (d) Hauptman, E.; Fagan, P.
J.; Marshall, W. Organometallics 1999, 18, 2061e2073; (e) Nakano, H.; Okuyama,
Under inert atmosphere, ligand 4aa (1.38 mg, 3.0
[Rh(cod)2]SbF6 (1.67 mg, 3.0 mol) were dissolved in dry, degassed
m
THF (1.2 mL). The mixture was stirred at room temperature for
30 min to obtain a yellow solution. Separately, also under inert
atmosphere, another solution was prepared containing methyl Z-
MAC (62 mg, 0.28 mmol) in dry, degassed THF (1.7 mL).
A stainless steel reactor was conditioned by warming it up to
110 ꢁC and pressurizing with N2 to 50 bar, followed by venting
(9 cycles) and cooling down again to room temperature. Then, the
catalyst solutionwas introduced (1 mL, 2.5 mmol,1 mol %) via syringe,
immediately followed by the substrate solution (1.5 mL, 0.25 mmol).
The system was flushed with H2 and pressurized to 20 bar.
ꢀ
ꢂ
Y.; Hongo, H. Tetrahedron Lett. 2000, 41, 4615e4618; (f) Pamies, O.; Dieguez, M.;
Net, G.; Ruiz, A.; Claver, C. Organometallics 2000, 19, 1488e1496; (g) Alexakis,
ꢀ
A.; Benhaim, C. Tetrahedron: Asymmetry 2001, 12, 1151e1157; (h) Pamies, O.; van
ꢂ
Strijdonck, G. P. F.; Dieguez, M.; Deerenberg, S.; Net, G.; Ruiz, A.; Claver, C.;
Kamer, P. C. J.; van Leeuwen, P. W. N. M. J. Org. Chem. 2001, 66, 8867e8871; (i)
ꢂ
Mancheno, O. G.; Priego, J.; Cabrera, S.; Arrayas, R. G.; Llamas, T.; Carretero, J. C.
~
ꢂ
J. Org. Chem. 2003, 68, 3679e3686; (j) Mancheno, O. G.; Arrayas, R. G.; Carre-
tero, J. C. J. Am. Chem. Soc. 2003, 126, 456e457; (k) Nakano, H.; Yokoyama, J.-i.;
Okuyama, Y.; Fujita, R.; Hongo, H. Tetrahedron: Asymmetry 2003, 14,
2361e2368; (l) Tu, T.; Zhou, Y.-G.; Hou, X.-L.; Dai, L.-X.; Dong, X.-C.; Yu, Y.-H.;
Sun, J. Organometallics 2003, 22, 1255e1265; (m) Molander, G. A.; Burke, J. P.;
The reaction was monitored by the gas uptake data. When no
more H2 was consumed, the reactor was vented and opened to air
and the reaction solution was concentrated under reduced pressure
to obtain a pale brown solid identified by 1H NMR as pure product.
This crude could be purified by a filtration through silica gel,
eluting with hexaneeethyl acetate mixtures. After removal of the
solvents, the product was obtained in quantitative yield as a col-
ourless crystalline solid. A sample (1 mg/mL in hexaneeIPA 93:7)
was analyzed by HPLC on a chiral stationary phase (82% ee).
ꢂ
Carroll, P. J. J. Org. Chem. 2004, 69, 8062e8069; (n) Verdaguer, X.; Lledo, A.;
ꢂ
ꢀ
Lopez-Mosquera, C.; Maestro, M. A.; Pericas, M. A.; Riera, A. J. Org. Chem. 2004,
ꢂ
69, 8053e8061; (o) Guimet, E.; Dieguez, M.; Ruiz, A.; Claver, C. Tetrahedron:
Asymmetry 2005, 16, 959e963; (p) Khiar, N.; Suarez, B.; Valdivia, V.; Fernandez,
I. Synlett 2005, 2963e2967; (q) Cheung, H. Y.; Yu, W.-Y.; Lam, F. L.; Au-Yeung, T.
T. L.; Zhou, Z.; Chan, T. H.; Chan, A. S. C. Org. Lett. 2007, 9, 4295e4298; (r) Sola,
J.; Reves, M.; Riera, A.; Verdaguer, X. Angew. Chem., Int. Ed. 2007, 46,
ꢂ
ꢂ
ꢀ
ꢂ
5020e5023; (s) Lam, F. L.; Au-Yeung, T. T. L.; Kwong, F. Y.; Zhou, Z.; Wong, K. Y.;
Chan, A. S. C. Angew. Chem., Int. Ed. 2008, 47, 1280e1283; (t) Khiar, N.; Navas, R.;
ꢂ
ꢂ
ꢂ
Suarez, B.; Alvarez, E.; Fernandez, I. Org. Lett. 2008, 10, 3697e3700 and refer-
ences therein.
ꢀ
7. (a) Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A. J. Org. Chem. 1997, 62,
4970e4982; (b) Puigjaner, C.; Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera,
A. J. Org. Chem. 1999, 64, 7902e7911; (c) Jimeno, C.; Moyano, A.; Pericas, M. A.;
Riera, A. Synlett 2001, 1155e1157; (d) Pasto, M.; Riera, A.; Pericas, M. A. Eur. J.
1H NMR (CDCl3, 400 MHz):
d
1.99 (s, 3H), 3.09 (dd, J¼13.9, 5.7 Hz,
ꢀ
1H), 3.15 (dd, J¼13.9, 6.0 Hz, 1H), 3.73 (s, 2H), 4.89 (dt, J¼7.8, 5.7 Hz,
1H), 5.91 (br s, 1H), 7.09 (m, 2H), 7.28 (m, 3H). HPLC: Chiralpak OJ-H
column, hexaneeIPA 93:7, 1 mL/min, 216 nm; tR¼18.1 min (S),
28.8 min (R).
ꢀ
ꢂ
ꢀ
ꢀ
Org. Chem. 2002, 2337e2341; (e) Pericas, M. A.; Puigjaner, C.; Riera, A.; Vidal-
ꢂ
ꢂ
Ferran, A.; Gomez, M.; Jimenez, F.; Muller, G.; Rocamora, M. Chem.dEur. J. 2002,
ꢂ
ꢀ
8, 4164e4178; (f) Jimeno, C.; Pasto, M.; Riera, A.; Pericas, M. A. J. Org. Chem.
2003, 68, 3130e3138; (g) Jimeno, C.; Vidal-Ferran, A.; Pericas, M. A. Org. Lett.
2006, 8, 3895e3898; (h) Rodríguez, B.; Pasto, M.; Jimeno, C.; Pericas, M. A.
Tetrahedron: Asymmetry 2006, 17, 151e160; (i) Popa, D.; Puigjaner, C.; Gomez,
M.; Benet-Buchholz, J.; Vidal-Ferran, A.; Pericas, M. A. Adv. Synth. Catal. 2007,
349, 2265e2278; (j) Cattoen, X.; Pericas, M. A. J. Org. Chem. 2007, 72,
3253e3258; (k) Alza, E.; Bastero, A.; Jansat, S.; Pericas, M. A. Tetrahedron:
Asymmetry 2008, 19, 374e378; (l) Fernandez-Perez, H.; Pericas, M. A.; Vidal-
ꢀ
ꢂ
ꢀ
Acknowledgements
ꢂ
ꢀ
€
ꢀ
This work was funded by MICINN (Grant No. CTQ2008-00947/
BQU), DIUE (Grant No. 2009SGR623), Consolider Ingenio 2010
(Grant No. CSD2006-0003) and the ICIQ Foundation. X.C. and X.C.C.
thank MEC for FPU fellowships.
ꢀ
ꢂ
ꢂ
ꢀ
€
ꢀ
Ferran, A. Adv. Synth. Catal. 2008, 350, 1984e1990; (m) Cattoen, X.; Pericas, M.
A. Tetrahedron 2009, 65, 8199e8205; (n) Subirats, S.; Jimeno, C.; Pericas, M. A.
ꢀ
Tetrahedron: Asymmetry 2009, 20, 1413e1418; (o) Popa, D.; Marcos, R.; Sayalero,
ꢀ
S.; Vidal-Ferran, A.; Pericas, M. A. Adv. Synth. Catal. 2009, 351, 1539e1556.
Supplementary data
8. For reviews, see: (a) Brown, J. M. In Comprehensive Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 2000; (b) Nagel,
U.; Albrecht, J. Top. Catal. 1998, 5, 3; (c) Albrecht, J.; Nagel, U. Angew. Chem., Int.
Ed. 1996, 35, 407e709; (d) Takaya, H.; Ohta, Y.; Noyori, R. Catalytic Asymmetric
Synthesis, Chapter 1; VCH: Wenheim, 1993; (e) Nugent, W. A.; Rajanbabu, T. V.;
Burk, M. J. Science 1993, 259, 479e483.
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2011.04.050. These data in-
clude MOL file and InChiKey of the most important compounds
described in this article.
ꢂ
ꢂ
ꢂ
9. (a) Bayon, J. C.; Claver, C.; Masdeu-Bulto, A. M. Coord. Chem. Rev. 1999, 193e195,
ꢂ
ꢀ
73e145; (b) Masdeu-Bulto, A. M.; Dieguez, M.; Martin, E.; Gomez, M. Coord.
Chem. Rev. 2003, 242, 159e201; (c) Mellah, M.; Voituriez, A.; Schulz, E. Chem.
Rev. 2007, 107, 5133e5209; (d) Pellissier, H. Tetrahedron 2007, 63, 1297e1330.
References and notes
ꢀ
10. Medina, E.; Moyano, A.; Pericas, M. A.; Riera, A. Helv. Chim. Acta 2000, 83,
972e988.
1. (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley: New York, NY,
1994; (b) Seyden-Penne, J. Chiral Auxiliaries and Ligands in Asymmetric Syn-
thesis; John Wiley: New York, NY, 1995; (c) Comprehensive Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 2000; (d)
Asymmetric Synthesis, 2nd ed.; Ojima, I., Ed.; VCH: New York, NY, 2000; Special
Issue on Asymmetric Catalysis Acc. Chem. Res. 2000, 33; (e) Blaser, H. U. Chem.
Rev. 1992, 92, 935e952.
2. A paradigmatic example is the Monsanto process for l-DOPA, initially relying on
an asymmetric hydrogenation with 88% ee. Knowles, W. S. Angew. Chem., Int. Ed.
2002, 41, 1998e2007.
11. For a similar strategy applied to formation of aziridinium instead of epis-
ꢀ
ꢀ
ulfonium intermediates, see: (a) Sola, L.; Vidal-Ferran, A.; Moyano, A.; Pericas,
M. A.; Riera, A. Tetrahedron: Asymmetry 1997, 8, 3685e3689; (b) Sola, L.; Vidal-
Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A. Tetrahedron: Asymmetry 1997, 8,
3985.
ꢀ
ꢀ
12. The product has been described in its racemic form in (a) Harada, T.; Akiba, E.;
Oku, A. J. Am. Chem. Soc. 1983, 105, 2771e2776 the starting epoxide has been
described in; (b) Hoshino, Y.; Yamamoto, H. J. Am. Chem. Soc. 2000, 122,
10452e10453.