The use of Lewis acids in the synthesis of 5-arylhydantoins

Article abstract of DOI:10.1016/j.jcat.2004.05.021

Different Lewis acids are able to promote the Friedel-Crafts reaction between 5-bromohydantoin and aromatic compounds. In the case of phenol, mixtures of ortho and para isomers are always obtained, with Mg(ClO₄) ₂ leading to the best selectivity. However, the best overall yield of 5-(hydroxyphenyl)hydantoin is obtained with YbCl₃. This method can be extended to other aromatic systems such as anisole and thiophene. These reactions give similar yields but proceed with total selectivity to 5-(4-methoxyphenyl)hydantoin and 5-(2-thiophenyl)hydantoin, respectively. The cationic exchange of Mg^II and Yb^III on anionic solid supports allows the preparation of very efficient heterogeneous catalysts for this reaction (productivity up to 600 mol of hydantoin per mole of Mg). These catalysts have practical advantages in that they can be recycled and reused.

Full text of DOI:10.1016/j.jcat.2004.05.021

Journal of Catalysis 226 (2004) 192–196
www.elsevier.com/locate/jcat
The use of Lewis acids in the synthesis of 5-arylhydantoins
a
a
a
a
a,∗
Carlos Cativiela , José M. Fraile , José I. García , Gustavo Lafuente , José A. Mayoral ,
a
b
Rachid Tahir , Antonio Pallarés
a
Departamento de Química Orgánica, Instituto de Ciencia de Materiales de Aragón, Universidad de Zaragoza-CSIC, E-50009 Zaragoza, Spain
b
DSM Deretil, E-04618 Villaricos, Almería, Spain
Received 29 March 2004; revised 12 May 2004; accepted 21 May 2004
Abstract
Different Lewis acids are able to promote the Friedel–Crafts reaction between 5-bromohydantoin and aromatic compounds. In the case of
phenol, mixtures of ortho and para isomers are always obtained, with Mg(ClO ) leading to the best selectivity. However, the best overall
4
2
yield of 5-(hydroxyphenyl)hydantoin is obtained with YbCl . This method can be extended to other aromatic systems such as anisole and thio-
3
phene. These reactions give similar yields but proceed with total selectivity to 5-(4-methoxyphenyl)hydantoin and 5-(2-thiophenyl)hydantoin,
II
III
respectively. The cationic exchange of Mg and Yb on anionic solid supports allows the preparation of very efficient heterogeneous cata-
lysts for this reaction (productivity up to 600 mol of hydantoin per mole of Mg). These catalysts have practical advantages in that they can
be recycled and reused.

2004 Elsevier Inc. All rights reserved.
Keywords: Lewis acids; Arylhydantoins
1
. Introduction
reaction between glyoxylic acid and urea—leading to allan-
toin [8,9]—followed by the reaction with phenol [10]. Both
steps require the use of a mineral acid, which can be re-
placed by solid acids [11]. However, neither the final yield
nor the selectivity to 5-(4-hydroxyphenyl)hydantoin are bet-
ter than those obtained in the direct synthesis. Another two-
step route involves the synthesis of p-hydroxymandelic acid
from glyoxylic acid and phenol and subsequent reaction with
urea [12,13]. From our own experience [14], although this
method may be an interesting alternative to the direct syn-
thesis, the final yield of 5-(4-hydroxyphenyl)hydantoin is
slightly lower and, once again, a large excess of an acid is
required.
In view of these problems, we considered the possibility
of using Lewis acids, rather than Brønsted acids, in catalytic
amounts. In this approach it is necessary to use different
starting materials and 5-halohydantoins were chosen as suit-
able substrates. We describe here the results of the reactions
between 5-bromohydantoin (2) and phenol (3) (Scheme 1)
with different homogeneous and heterogeneous Lewis acids.
The extension of this methodology to include other aromatic
compounds is also described.
Racemic 5-arylhydantoins are important intermediates in
the enzymatic production of (R)-2-arylglycines[1,2]. One of
the most important examples is 5-(4-hydroxyphenyl)hydan-
toin (4p), which is used in the preparation of semisynthetic
penicillins and cephalosporins [3,4].
In practice, 5-(4-hydroxyphenyl)hydantoin (4p) is ob-
tained by the amidoalkylation of phenol with urea and gly-
oxylic acid [5,6]. This method suffers from two main draw-
backs. First, a large excess of concentrated mineral acid
is needed in order to attain high conversions. Secondly,
a mixture of 5-(4-hydroxyphenyl)hydantoin (4p) and 5-(2-
hydroxyphenyl)hydantoin (4o) is always obtained. We ex-
plored the use of heterogeneous acids in this reaction [7] and
found that the solids did not improve on the conversion and
selectivity obtained in solution, although recycling and reuse
of the catalysts were possible. It is also possible to carry out
the reaction in two steps. One synthetic route involves the
*
Corresponding author. Fax: +34 976762077
E-mail address: mayoral@unizar.es (J.A. Mayoral).
0021-9517/$ – see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2004.05.021

C. Cativiela et al. / Journal of Catalysis 226 (2004) 192–196
193
and the suspension was stirred at room temperature for 24 h.
The solid was filtered, washed with methanol, and dried un-
der vacuum overnight prior to use. Owing to the low cation
content (see Table 3) the textural properties of the solids did
2
−1
not change to a great extent (surface areas 20–50 m g
lower).
2
.3. Synthesis of 5-arylhydantoins
Bromine (0.25 mL, 5.0 mmol) was slowly added (10 min)
by syringe to a stirred solution of hydantoin (1, 500 mg,
◦
5
.0 mmol) in anhydrous dioxane (2 mL) at 100 C under an
inert atmosphere. On completion of the addition the solution
was stirred for 40 min at the same temperature. The reaction
mixture was cooled and the resulting solution was added to a
mixture of the catalyst and aromatic compound (10.0 mmol).
Scheme 1.
2
. Experimental
◦
The reaction was stirred at 40 C for the appropriate time
2
.1. Synthesis and entrapment of 5-bromohydantoin (2)
(see Tables 1–3 for amounts of catalyst and reaction times).
The catalyst was removed by filtration and the solvent was
evaporated under reduced pressure. The products were crys-
tallized from dichloromethane.
Bromine (5.64 mL, 0.11 mol) was slowly added by sy-
ringe to a stirred solution of hydantoin (1, 10 g, 0.1 mol)
in glacial acetic acid (or dioxane, 40 mL) at 100 C under
◦
an inert atmosphere. On completion of the addition, the sys-
tem was opened to allow the release of HBr and the solution
was stirred for 45 min at the same temperature. The solu-
Table 1
Results obtained in the reaction between phenol and 5-bromohydantoin us-
ing homogeneous Lewis acids
a
◦
b
c
d
tion was cooled to 30 C and triethyl phosphite (23.95 mL,
Entry
Catalyst (%)
t (h)
% yield 4
4p/4o
0
.14 mol) was slowly added at such a rate that the tempera-
1
2
3
4
5
6
7
8
9
–
24
24
24
24
4
24
21
4
31
38
45
66
82
63
86
84
93
86/14
77/23
82/18
85/15
85/15
73/27
76/24
78/22
76/24
◦
Cu(OTf) (1%)
ZnBr (1%)
Mg(ClO ) (1%)
Mg(ClO ) (2%)
Yb(OTf) (1%)
YbCl (1%)
3
YbCl (2%)
ture of the solution remained below 40–45 C. The solution
was stirred for 90 min and the solvent was removed under
reduced pressure. Phosphonate 5 was crystallized from di-
ethyl ether. Yield: 83%. H NMR (d6-dmso, δ ppm, J Hz):
1
4
2
2
4
2
4
2
1
3
0.91 (s, 1H), 8.41 (s, 1H), 4.75 (d, 1H, J = 14.7), 4.08 (m,
H), 1.23 (t, 6H, J = 7.0).
3
16
When the bromination was carried out in acetonitrile
a
under reflux, the only isolated product was the dimer 6.
Reaction conditions: in situ preparation of 5-bromohydantoin from hy-
◦
1
dantoin (5 mmol) and bromine (0.3 mL) in dioxane (2 mL) at 100 C;
reaction with phenol (10 mmol) and catalyst at 40 C.
H NMR (d6-dmso, δ ppm, J Hz): 11.18 (s, 1H), 11.01 (s,
◦
1
H), 8.19 (s, 1H), 5.78 (s, 1H), 3.95 (d, 1H, J = 17), 3.75
b
Amount of catalyst with respect to starting hydantoin.
Isolated yield. Calculated from the starting hydantoin.
Determined by H NMR spectroscopy.
13
(
1
d, 1H, J = 17). C NMR (d6-dmso, δ ppm): 170.8, 170.7,
c
56.7, 156.5, 63.0, 47.0.
d
1
2
.2. Preparation of the catalysts
Table 2
Results obtained in the reaction between 5-bromohydantoin and different
aromatics using homogeneous Lewis acids
2
−1
a
β-Zeolite (CP814E, surface area 680 m g ) in its am-
monium form was purchased from Zeolyst.
The fluorosulfonic-typesupport (surface area 550 m g
b
c
Entry
Aromatic
Anisole
Metal salt (%)
ZnBr (1%)
t (h)
% yield
2
−1
,
1
2
3
4
5
6
7
8
24
24
21
16
24
21
20
16
27
45
67
60
33
60
71
75
2
3
−1
pore volume 0.75 cm g , pore diameter 55 Å), represented
as SiO2–CF2SO3H and prepared from (HO)3Si(CH2)3–
CF2)2O(CF2)2SO3K [15], was a generous gift from DuPont
Research and Development. Prior to exchange, SiO2–CF2–
SO3H was transformed into its sodium form by passing a
solution of 2 M NaCl through a column of the solid un-
til neutral pH was obtained. The solid was then washed
ZnBr (5%)
2
Mg(ClO ) (5%)
4
2
(
YbCl3 (5%)
ZnBr (1%)
Thiophene
2
ZnBr (5%)
2
Mg(ClO ) (5%)
4
2
YbCl (5%)
3
a
Reaction conditions: in situ preparation of 5-bromohydantoin from hy-
◦
with deionized water and dried under vacuum at 150 C
◦
dantoin (5 mmol) and bromine (0.3 mL) in dioxane (2 mL) at 100 C;
reaction with phenol (10 mmol) and catalyst at 40 C.
b
◦
overnight.
A solution of the metal salt [YbCl3 or Mg(ClO4)2,
mmol] in methanol (10 mL) was added to the support (1 g)
Amount of catalyst with respect to starting hydantoin.
Isolated yield. Calculated from the starting hydantoin.
c
1

194
C. Cativiela et al. / Journal of Catalysis 226 (2004) 192–196
Table 3
Results obtained in the reaction between aromatic compounds and 5-bromohydantoin using heterogeneous Lewis acids
a
c
d
Entry
Aromatic
Catalyst
Loading
Run
t
% yield
4p/4o
−
1 b
(h)
(productivity)
(mmol g
)
1
2
3
4
5
6
7
8
9
0
1
2
3
Phenol
SiO –CF SO Mg
0.074
1
2
3
1
2
1
2
1
2
1
2
1
2
8
15
19
16
22
20
20
16
21
15
24
15
21
82 (554)
75 (507)
76 (514)
86 (328)
76 (290)
72 (655)
70 (636)
88 (389)
70 (310)
38 (345)
28 (255)
34 (309)
26 (236)
78/22
76/24
77/23
79/21
81/19
78/22
75/25
79/21
81/19
–
2
2
3
SiO –CF SO Yb
0.131
0.055
0.113
0.055
0.055
2
2
3
β-Mg
β-Yb
β-Mg
β-Mg
1
1
1
1
Anisole
–
–
–
Thiophene
a
◦
Reaction conditions: in situ preparation of 5-bromohydantoin from hydantoin (5 mmol) and bromine (0.3 mL) in dioxane (2 mL) at 100 C; reaction with
◦
phenol (10 mmol) and catalyst (100 mg) at 40 C.
b
Determined by plasma emission spectroscopy.
c
Isolated yield. Calculated from the starting hydantoin. Productivity is the number of molecules of product produced per catalytic site.
d
1
Determined by H NMR spectroscopy.
2
.4. NMR spectra (d6-dmso, δ ppm, J Hz)
-(4-Hydroxyphenyl)hydantoin (4p): 1H NMR: 10.69
5
(
6
1
s, 1H), 9.51 (s, OH), 8.28 (s, 1H), 7.10 (d, 2H, J = 8.5),
1
3
.76 (d, 2H, J = 8.5), 5.00 (s, 1H). C NMR: 174.7, 157.4,
28.0, 126.3, 115.4, 60.9.
5
-(2-Hydroxyphenyl)hydantoin (4o): 1H NMR: 10.59
(
2
1
s, 1H), 9.75 (s, OH), 7.98 (s, 1H), 7.12 (m, 2H), 6.78 (m,
H), 5.11 (s, 1H). 13C NMR: 175.1, 157.8, 155.9, 130.2,
29.6, 122.4, 118.9, 115.6, 58.5.
5
-(4-Methoxyphenyl)hydantoin (7): 1H NMR: 10.73
Scheme 2.
(
s, 1H), 8.33 (s, 1H), 7.22 (d, 2H, J = 8.5), 6.95 (d, 2H,
13
phosphonate (5) [19]. Dioxane and acetic acid were suit-
able solvents for this reaction and both led to high yields
of phosphonate (> 80%). A high temperature is necessary
for bromination and solvents with lower boiling points, such
as THF, did not give any bromination at all. In some cases
ZnBr2 has been used as promoter in this bromination [18],
but differences were not observed when dioxane or acetic
acid were used. In acetonitrile the use of ZnBr2 allows the
isolation of a dimer (6), which indicates the role of the ad-
jacent N–H bond. In order to assess the importance of this
N–H bond, the bromination of N,N-dibenzylhydantoin was
tested under the same conditions. The recovery of unaltered
starting material clearly shows the importance of this func-
tionality.
J = 8.5), 5.08 (s, 1H), 3.76 (s, 3H). C NMR: 174.5, 159.2,
1
57.4, 127.9, 126.8, 114.0, 60.7, 55.1.
1
5
-(2-Thiophenyl)hydantoin (8): H NMR: 10.87 (s, 1H),
8
7
1
.58 (s, 1H), 7.53 (d, 1H, J = 5.1), 7.11 (d, 1H, J = 3.3),
13
.03 (dd, 1H, J = 5.1, 3.3), 5.47 (s, 1H). C NMR: 173.1,
57.0, 138.8, 127.1, 126.1, 125.9, 57.4.
3
. Results and discussion
3
.1. Synthesis of 5-bromohydantoin (2)
5
-Chloro- [16] and 5-bromohydantoin [17] had been de-
scribed in the literature as intermediates in the synthesis of
allantoin and 5-butoxyhydantoin, respectively. These com-
pounds were prepared from hydantoin by reaction with
SOCl2 or Br2. In both cases the 5-halohydantoins were de-
scribed as unstable and they were preferably used with-
out purification, a process also described in a more recent
patent [18]. Given the problems encountered in the synthe-
sis of 5-chlorohydantoin, 5-bromohydantoin was chosen as
the starting material and it was prepared by reaction with
bromine in different solvents (Scheme 2). Triethyl phosphite
was added in order to entrap the 5-bromohydantoin as the
On the basis of the information outlined above, diox-
ane was chosen as the best solvent for bromination and the
Friedel–Crafts reaction with phenol was carried out without
isolation of 5-bromohydantoin.
3
5
.2. Use of homogeneous Lewis acids in the synthesis of
-arylhydantoins
An excess (2 eq) of phenol and a Lewis acid catalyst
were added to a solution of 5-bromohydantoin (Scheme 1).

C. Cativiela et al. / Journal of Catalysis 226 (2004) 192–196
195
1
The results of these reactions are gathered in Table 1. It has
been reported [18] that this reaction takes place without a
catalyst, but in our experience the yield is low (31%, en-
try 1). The use of ZnBr2 improves the result (45%, entry 3)
but the yield is still lower than that reported in the litera-
ture [18]. In view of this result, several different Lewis acids
were tested. Cu(OTf)2 (entry 2) was even less efficient than
ZnBr2. In contrast, Mg(ClO4)2 led to 66% yield after 24 h
error of H NMR spectroscopy). When the para isomer
alone is taken into account, the yields obtained from phe-
nol and anisole are very similar. This finding opens the way
to the synthesis of pure 5-(4-hydroxyphenyl)hydantoin from
O-protected phenols.
Thiophene was the other aromatic system tested (Table 2,
entries 5–8) and the yields were slightly higher, up to 75%
3
with YbCl . This increase in yield is consistent with the
(
entry 4). This result was improved by using a larger amount
of catalyst (2%), which gave rise to 82% yield after only 4 h
entry 5).
Another interesting family of Lewis acids are the lan-
higher reactivity of thiophene and even ZnBr is an effi-
cient catalyst in this case. The reaction is also completely
2
(
selective and in this case gives the 2-substituted thiophene.
These results demonstrate the potential application of this
method, given that the enzymatic kinetic resolution of this
hydantoin would lead to enantiomerically pure α-amino-2-
thiopheneacetic acid.
thanides [20,21], which allow a higher coordination number
and, in many cases, are compatible with water. Yb(OTf)3
(
entry 6) showed an activity similar to that of Mg(ClO4)2
whereas YbCl3 was clearly better (entry 7). In fact, the use
of 2% of the latter catalyst gave 93% yield after 16 h (en-
try 9).
3
5
.3. Use of heterogeneous Lewis acids in the synthesis of
-(4-hydroxyphenyl)hydantoin (4p)
One of the most important aims from the point of view
of the industrial synthesis of 5-(4-hydroxyphenyl)hydantoin
is the para/ortho selectivity [7]. The Lewis acids promote
the reaction, with para/ortho ratios in the range 73/27 to
One way to improve this methodology concerns the abil-
2
+
ity to recycle the catalyst. With this aim in mind, Mg
and Yb3 ions were exchanged onto anionic solid sup-
ports to obtain heterogeneous Lewis acids. Two different
supports were investigated. First, a fluorosulfonic support
with a rather high exchange capacity, prepared from silica
and (HO)3Si(CH2)3(CF2)2O(CF2)2SO3K [15]. This support
has proven useful in the exchange of chiral bis(oxazoline)-
copper complexes [23]. Secondly, a microporous β-zeolite
was employed in an attempt to improve the para/ortho ratio
through shape selectivity [24]. The results obtained in the
synthesis of 5-arylhydantoins with these heterogeneous cat-
alysts are gathered in Table 3.
+
8
5/15 and Mg(ClO4)2 is always more efficient than YbCl3
with regard to selectivity. In this case the Lewis acids do not
present the ortho preference shown in the synthesis of hy-
droxymandelic acid [22]. On the contrary these results are
in agreement with the 82/18 ratio obtained in the synthe-
sis from other cyclic intermediates, such as allantoin [11],
promoted by Brønsted acids, and they are generally worse
than the 86/14 ratio obtained in the direct synthesis [7].
In any case, these results represent a 70% yield of 5-(4-
hydroxyphenyl)hydantoin with only 2% catalyst, a level
close to the 77% obtained with 2.2 eq HCl (i.e., 220%) in
solution.
The promising results described above encouraged us to
extend this methodology to the reaction of 5-bromohydan-
toin with other aromatic compounds (Scheme 3). Anisole
was the first choice given its similarity to phenol, albeit
with a lower reactivity. The two best homogeneous cat-
alysts, Mg(ClO4)2 and YbCl3, together with ZnBr2 were
tested in this reaction (Table 2, entries 1–4). As expected,
the yield was lower than in the case of phenol, up to 67%
with Mg(ClO4)2, but with the important advantage that to-
tal para selectivity was achieved (within the experimental
The four heterogeneous Lewis acids were tested in the
synthesis of 5-(4-hydroxyphenyl)hydantoin.A small amount
of catalyst was used (100 mg) and this level represents
very few Lewis acid sites (0.11 to 0.26% catalyst). Sur-
prisingly, the reactions were very efficient, with yields of
5
-(hydroxyphenyl)hydantoin in the range of 72 to 88% (en-
tries 1, 4, 6, and 8). These yields are indicative of very
high turnover numbers, leading to a site productivity (mole-
cules formed per site) ranging from 328 in the case of
SiO2–CF2SO3Yb to 655 with β-Mg. The differences in the
productivities of the two cations can be explained in terms
of the different loading of each catalyst. This good behav-
ior contrasts with the results obtained in the homogeneous
phase, where 1–2% catalyst is necessary in order to obtain
such reaction yields and, consequently, the productivities
(
38–86 molecules per site) are much lower. Although several
factors can be proposed to explain this behavior, the effect of
site isolation may be important in this case as it would allow
the efficient use of all the metal atoms as Lewis acid sites.
Moreover, recycling of the heterogeneous catalysts is
possible and only a slight reduction in the catalytic activ-
ity was observed with the recycled systems. Indeed, a third
reaction was tried with SiO2–CF2SO3Mg and this led to a
total productivity after 3 runs of 1575 molecules per site,
Scheme 3.

196
C. Cativiela et al. / Journal of Catalysis 226 (2004) 192–196
an excellent result for a Friedel–Crafts reaction with rather
complicated molecules.
The weight productivity in such processes is also an im-
portant factor and 16.9 g of 5-(4-hydroxyphenyl)hydantoin
indebted to the M.C.Y.T. for a postdoctoral grant. We are in-
debted to Dr. Mark A. Harmer, from DuPont Research and
development, for the generous gift of the fluorosulfonic sup-
port.
(
22.4 g of the isomers mixture) was produced per gram of
SiO2–CF2SO3Mg after three reactions (under nonoptimized
conditions). This productivity may be increased after further
recycling experiments and opens the way to possible indus-
trial applications.
References
[1] S. Takahashi, T. Ohashi, Y. Kii, H. Kumagai, H. Yamada, J. Ferment.
Technol. 57 (1979) 328.
In the case of zeolite catalysts, the activities and produc-
tivities are very similar to those obtained with the fluoro-
sulfonic support. Unfortunately, a significant improvement
in the para/ortho selectivity was not observed. Given its
high productivity, zeolite β-Mg was chosen as the catalyst
for the reaction between 5-bromohydantoin and the other
two aromatic molecules (entries 10–13). It can be seen that
the lower reactivities of anisole and thiophene reduce the
isolated yields to 38 and 34%, respectively. It should be
noted that these figures represent productivities of more than
[
2] A.S. Bommarius, K. Drauz, U. Groeger, C. Wandrey, in: A.N. Collins,
G.N. Sheldrake, J. Crosby (Eds.), Chirality in Industry, Wiley, Chich-
ester, 1992, p. 371.
[3] L.B. Crast Jr., US patent 3 489 750 (1970), to Bristol-Myers.
[4] J.H.C. Nayler, A.A.W. Long, H. Smith, T. Taylor, N. Ward, J. Chem.
Soc. C 10 (1971) 1920.
[
[
[
[
[
5] T. Ohashi, S. Takahashi, T. Nagamachi, K. Yoneda, H. Yamada, Agric.
Biol. Chem. 45 (1981) 831.
6] K. Yoneda, T. Ohashi, T. Nagamachi, H. Fukumitsu, S. Takahashi, US
patent 4 230 869 (1980), to Kanegafuchi.
7] C. Cativiela, J.M. Fraile, J.I. García, B. Lázaro, J.A. Mayoral, A. Pal-
larés, Appl. Catal. A 224 (2002) 153.
3
00 molecules per site, again impressive for Friedel–Crafts
8] W. Mesh, O.-A. Grosskinsky, N. Lösch, German patent 1939924
alkylation reactions. The solids can be reused with only a
slight reduction in activity, leading to overall productivities
of 550–600 molecules per site.
(1969), to BASF AG.
9] K. Schermanz, K. Fitzinger, Eur. patent 0 460 412 (1991), to Chemie
Linz.
[
10] Y. Christidis, French patent 79 14562 (1979), to S.F. Hoechst.
3
.4. Conclusions
-Bromohydantoin has proven to be a valuable interme-
[11] C. Cativiela, J.M. Fraile, J.I. García, B. Lázaro, J.A. Mayoral, A. Pal-
larés, Green Chem. 5 (2003) 275.
[12] C. Eguchi, H. Kakizaki, H. Iwagami, British patent 2 012 756 (1979),
5
to Ajinomoto.
diate in the synthesis of 5-arylhydantoins. The reaction with
aromatic systems is easily promoted by catalytic amounts
of both homogeneous and heterogeneous Lewis acids. The
heterogeneous catalysts not only give rise to very high pro-
ductivities, but they can also be efficiently recycled with only
a slight decrease in activity. Although the reaction with phe-
nol leads to para + ortho mixtures, other aromatics allow
the synthesis of only one isomer with total selectivity. This
method opens the way to the synthesis of enantiomerically
pure arylglycines by enzymatic kinetic resolution of the cor-
responding 5-arylhydantoins.
[
[
13] Y. Christidis, Eur. patent 8 547 (1979), to S.F. Hoechst.
14] C. Cativiela, J.M. Fraile, J.I. García, B. Lázaro, J.A. Mayoral, A. Pal-
larés, manuscript in preparation.
[15] M.A. Harmer, Q. Sun, M.J. Michalczyk, Z. Yang, Chem. Commun.
(1997) 1803.
[16] J. Abblard, A. Meynaud, Bull. Soc. Chim. Fr. (1971) 942.
[17] D. Ben-Ishai, G. Ben-Et, A. Warshawsky, Heterocycles (1970) 1289.
[18] H. Yasuda, Eur. patent 647 630 (1995), to Showa Denko K.K.
[19] N.A. Meanwell, H.R. Roth, E.C.R. Smith, D.L. Wedding, J.J.K.
Wright, J. Org. Chem. 56 (1991) 6897.
[20] S. Kobayashi (Ed.), Lanthanides: Chemistry and Use in Organic Syn-
thesis, Springer, Berlin, 1999.
[
[
21] G.A. Molander, Chem. Rev. 92 (1992) 29.
22] A.J. Hoefnagel, J.A. Peters, H. van Bekkum, Rec. Trav. Chim. Pays-
Bas 115 (1996) 353.
Acknowledgments
[23] J.M. Fraile, J.I. García, C.I. Herrerías, J.A. Mayoral, M.A. Harmer,
J. Catal. 221 (2004) 532.
This work was made possible by the financial support of
the Spanish M.C.Y.T. (project PPQ2000-0322-P4). R.T. is
[24] F. Ramôa Ribeiro, F. Alvarez, C. Henriques, F. Lemos, J.M. Lopes,
M.F. Ribeiro, J. Mol. Catal. A 96 (1995) 245.