Lanthanide(III) nitrobenzenesulfonates as new nitration catalysts: The role of the metal and of the counterion in the catalytic efficiency

Article abstract of DOI:10.1002/ejoc.200400475

Lanthanide(III) complexes of p-nitrobenzenesulfonic acid, Ln(p-NBSA) ₃, m-nitrobenzenesulfonic acid, Ln(m-NBSA)₃, and 2,4-nitrobenzenesulfonic acid, Ln(2,4-NBSA)₃, were prepared, characterized and examined as catalyst for the nitration of benzene, toluene, xylenes, naphthalene, bromobenzene and chlorobenzene. The initial screening of the catalysts showed that lanthanum(III) complexes were more effective than the corresponding ytterbium(III) complexes, and that catalysts containing the bulky 2,4-NBSA ligand were less effective than the catalyst containing p-NBSA (nosylate) or m-NBSA ligands. Examination of a series of Ln(p-NBSA)₃ and Ln(m-NBSA)₃ catalysts revealed that there is a clear correlation between the ionic radii of the lanthanide(III) ions and the yields of nitration, with the lighter lanthanides being more effective. The X-ray single crystal structure of Yb(m-NBSA)₃·6H₂O shows that two m-NBSA ligands are directly bound to the metal centre while the third ligand is not located in the first coordination sphere, but it is hydrogen bonded to one of the water molecules which is coordinated to ytterbium(III). NMR studies suggest that this structure is preserved under the conditions used in the nitration reaction. The structure of Yb(m-NBSA)₃ is markedly different from the structure of the well-known ytterbium(III) triflate catalyst. The coordination of the nitrobenzenesulfonate counterion to the lanthanide(III) ion suggests that steric effects might play an important role in determining the efficiency of these novel nitration catalysts. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400475

FULL PAPER
Lanthanide(III) Nitrobenzenesulfonates as New Nitration Catalysts: The Role
of the Metal and of the Counterion in the Catalytic Efficiency
Tatjana N. Parac-Vogt,[a] Sophia Pachini,[b] Peter Nockemann,[a] Kristof Van Hecke,[a]
Luc Van Meervelt,[a] and Koen Binnemans*[a]
Keywords: Homogeneous catalysis / Lanthanides / Lewis acids / Electrophilic substitution / Nitration / Rare earths
Lanthanide(III) complexes of p-nitrobenzenesulfonic acid,
Ln(p-NBSA)3, m-nitrobenzenesulfonic acid, Ln(m-NBSA)3, bound to the metal centre while the third ligand is not loc-
and 2,4-nitrobenzenesulfonic acid, Ln(2,4-NBSA)3, were pre- ated in the first coordination sphere, but it is hydrogen bon-
NBSA)3·6H2O shows that two m-NBSA ligands are directly
pared, characterized and examined as catalyst for the nitra- ded to one of the water molecules which is coordinated to
tion of benzene, toluene, xylenes, naphthalene, bromoben- ytterbium(III). NMR studies suggest that this structure is pre-
zene and chlorobenzene. The initial screening of the cata-
served under the conditions used in the nitration reaction.
lysts showed that lanthanum(III) complexes were more effect- The structure of Yb(m-NBSA)3 is markedly different from the
ive than the corresponding ytterbium(III) complexes, and that
catalysts containing the bulky 2,4-NBSA ligand were less ef-
fective than the catalyst containing p-NBSA (nosylate) or m-
NBSA ligands. Examination of a series of Ln(p-NBSA)3 and
structure of the well-known ytterbium(III) triflate catalyst.
The coordination of the nitrobenzenesulfonate counterion to
the lanthanide(III) ion suggests that steric effects might play
an important role in determining the efficiency of these novel
Ln(m-NBSA)3 catalysts revealed that there is a clear correla- nitration catalysts.
tion between the ionic radii of the lanthanide(III) ions and the
yields of nitration, with the lighter lanthanides being more
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
effective. The X-ray single crystal structure of Yb(m-
Germany, 2004)
function as Lewis acids in aqueous media.[3] They are able
to catalyse numerous reactions in aqueous solutions, such
as DielsϪAlder reactions,[4] aldol reactions,[5] Michael ad-
dition [6] and allylation of carbonyl and imine compounds.[7]
The use of lanthanide reagents for the aromatic nitration
has been already demonstrated more than twenty years ago
by Ouertani et al., but this procedure required stoichio-
metric quantities of lanthanide() nitrates for the selective
nitration of phenols.[8] On the other hand, Waller et al. have
shown that only catalytic amounts of scandium() or lan-
thanide() triflates were needed for the successful nitration
of a range of simple aromatic compounds.[9,10] Unlike the
commonly used industrial procedure, that requires use of
potent mixtures of concentrated nitric and sulfuric acids,
this procedure uses catalytic quantities of lanthanide() tri-
flates as replacement for sulfuric acid and only one equiva-
lent of nitric acid. In this way, excessive acid waste streams
that are environmentally unfriendly and expensive to treat
are avoided, and the only by-product of the nitration reac-
tion is water. However, use of the lanthanide triflate cata-
lysts on a very large scale is unlikely due to their relatively
high cost and due to the fact that one needs to handle the
strongly corrosive triflic acid for their preparation.
Introduction
In the past decade, a great number of organic synthetic
reactions involving lanthanide reagents have been ex-
plored.[1] Due to their unique electronic properties, the lan-
thanides offer new structural and reactivity patterns not
present in main group and d-transition metal chemistry.
Many important reactions such as carbonϪcarbon bond-
forming reactions, asymmetric catalysis, polymer synthesis,
etc., could be promoted by using various lanthanide cata-
lysts. Some of those procedures, which have been summar-
ized in several excellent reviews, play an important role in
modern organic synthesis.[2] Among the many useful lan-
thanide reagents, lanthanide() triflates have been recog-
nized as unique Lewis acids and they are currently of great
research interest. Unlike classic Lewis acids such as AlCl3,
BF3, TiCl4, SnCl4, etc., which decompose readily in the
presence of water, lanthanide triflates are stable and can
[a]
Katholieke Universiteit Leuven, Department of Chemistry,
Celestijnenlaan 200F, 3001 Leuven, Belgium
Fax: (internat.) ϩ32-16-327992
E-mail: Koen.Binnemans@chem.kuleuven.ac.be
[b]
Laboratory of Inorganic and General Chemistry, Faculty of
Nitration of aromatic compounds by the use of lantha-
nide reagents could also be achieved in fluorous solvents
Chemistry, Aristotle University of Thessaloniki,
Thessaloniki 54124, Greece
4560
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200400475
Eur. J. Org. Chem. 2004, 4560Ϫ4566

Lanthanide(III) Nitrobenzenesulfonates as New Nitration Catalysts
using lanthanide salts of perfluorooctanesulfonic acid as
FULL PAPER
Single crystal and thermal analytical studies of related
catalysts.[11] Although very small catalyst loading were suf- lanthanide complexes with p-toluenesulfonic acid, the so-
ficient for achieving moderate to good yields (40Ϫ60%) of called lanthanide tosylates, showed that this type of com-
nitrotoluene, this method is also unlikely to be utilized on plexes initially crystallize with nine water molecules in the
a large scale due to the high costs of fluorinated solvents first coordination sphere, and that seven water molecules
and their large global warming potential that could pose are lost upon heating at ca. 55 °C.[16] As the coordination
serious environmental problems.[12]
number of the lanthanides is always high (commonly 8 or
We recently reported that lanthanide() salts of p-tolu- 9), it has been proposed that upon removal of hydrated
enesulfonic acid efficiently catalysed the conversion of tolu- water molecules from the crystal, the sulfonate anions move
ene to nitrotoluene, but they were not very effective in the and act as multidentate ligands forming more metal-oxygen
nitration of aromatic compounds with electron-with- bonds in place of the removed water oxygen atoms. The
drawing groups.[13] In an attempt to increase the activity elemental analysis results of lanthanide() complexes with
of the catalyst we also explored lanthanide() salts of p- nitrosulfonic acids (Exp. Sect.) indicated that upon drying
nitrobenzenesulfonic acid, which are also called lantha- the crystalline solid at 40 °C in vacuo, some water of crys-
nide() nosylates.[14] Both p-toluenesulfonic and p-nitro- tallization is lost, but the complexes still contain two to six
benzenesulfonic acids have some major advantages com- molecules of water. Consequently, infrared spectroscopic
pared to triflic acid, because they are non-corrosive and data of the complexes showed strong and broad absorption
much cheaper to obtain. Although the acidity of p-nitro- in the region 3200Ϫ3600 cmϪ1, corresponding to the ϪOH
benzenesulfonic acid is at least five orders of magnitude stretching frequency of water molecules.
weaker than that of triflic acid,[15] its lanthanide()
Since it has been known that some sulfonic acids are cap-
salts seem to be sufficiently strong Lewis acids to catalyse able of promoting nitration reactions[17] it is important to
efficiently the nitration of electron rich and mildly electron demonstrate the absence of trace amounts of free nitroben-
poor arenes. However, contrary to nitrations catalysed by zenesulfonic acid in the lanthanide() catalyst. The infrared
lanthanide() triflates, in which the catalytic activity of spectra of lanthanide() complexes showed that the broad
triflate salts increases with decreasing ionic radius of the absorption between 2400 and 3000 cmϪ1, typical for the
lanthanide() ion,[9Ϫ10] the catalytic activity of lantha- OH stretching mode of protonated nitrobenzenesulfonic
nide() nosylate salts decreases with the decrease of the acid,[18] was absent in all cases. Furthermore, the aqueous
ionic radius of the lanthanide() ion. This intriguing differ- solution of lanthanide() complexes always had a close-
ence prompted us to examine the role that both metal to-neutral pH, which further confirms the absence of free
center and the counterion play in the reaction mechanism. nitrobenzenesulfonic acid in the isolated catalysts.
We also characterized these novel lanthanide() catalysts,
Screening of the Lanthanide(III) Catalysts
which are virtually unexplored compounds. To the best of
our knowledge, no literature data about chemical and struc-
tural properties of these reagents are available up to now.
Previous studies involving various reactions catalysed by
lanthanide() triflates have shown that ytterbium() tri-
flate is the most effective catalyst among all the lantha-
Results and Discussion
Table 1. Nitration of toluene in the presence of ytterbium() and
lanthanum() catalysts containing different counterions
Synthesis and Characterization of the Catalysts
Lanthanide() complexes of three different nitroben-
zenesulfonic acids: p-Nitrobenzenesulfonic (p-HNBSA), m-
nitrobenzenesulfonic (m-HNBSA), and 2,4-nitroben-
zenesulfonic (2,4-HNBSA) acids, were synthesized (Fig-
ure 1).
Conversion (%)[a]
Counterion
p-NBSA
m-NBSA
2,4-NBSA
YbIII
61%
58%
41%
LaIII
Ͼ 98%
Ͼ 98%
83%
Figure 1. Sulfonates used as counterions for lanthanide() cata-
lysts: p-NBSA ϭ para-nitrobenzenesulfonate; m-NBSA ϭ meta-
nitrobenzenesulfonate; 2,4-NBSA ϭ 2,4-dinitrobenzenesulfonate
[a] 1H NMR analysis, the product distribution of different nitrotolu-
ene isomers was in each reaction the same.
Eur. J. Org. Chem. 2004, 4560Ϫ4566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4561

K. Binnemans et al.
FULL PAPER
nides(III) triflates.[3] A screening of the entire series of lan-
The effect of catalyst concentration on the reaction yields
thanide() triflates for their catalytic activity in nitration was examined for the nitration of naphthalene. Within 30
also revealed a clear inverse correlation between the extent minutes, one equivalent of nitric acid and only 5 mol % of
of nitration reaction and the ionic radius of the trivalent the La(p-NBSA)3 catalyst were sufficient for the quantitat-
lanthanide ion.[10] The reaction we chose for the initial ive conversion of naphthalene to 1-mononitronaphthalene
screening of the catalytic activity of Ln(p-NBSA)3, Ln(m- and 2-mononitronaphthalene, while 2 mol % of the cata-
NBSA)3, and Ln(2,4-NBSA)3 catalysts was the nitration of lysts yielded 87% of the same products. (Table 3).
toluene. We first examined catalysts containing either ytter-
bium() or lanthanum() ions, as these two ions represent
two extreme examples of the lanthanide() series in terms
of ionic radius. To our surprise, when the nitration reaction
Table 3. Effect of the La(p-NBSA)3 catalyst loading on the extent
of naphthalene nitration
was quenched after four hours, the lanthanum() catalysts
gave in all the cases much higher yields than the corre-
sponding ytterbium() catalysts (Table 1).
Further experiments also showed that the lanthanum()
catalysts were consistently more efficient in the nitration re-
action of other aromatic compounds than the ytterbium()
catalysts, so we focused our attention on the former system.
Catalyst load(%)[a] [b]
Reaction yield
10%
5%
2%
1%
100
100
87
72
[a]
3 mmol of the substrate and 1 equiv. of 67% HNO3 in refluxing
[b]
1,2-dichloroethane for 30 min.
GC analysis; the product distri-
bution of 1-NO2 and 2-NO2 isomers was in each reaction 92:8.
Nitration of Arenes with La(p-NBSA)3, La(m-NBSA)3 and
La(2,4-NBSA)3 Catalysts
All nitration reactions were performed in 1,2-dichloro-
ethane at reflux temperature in the presence of a stoichio-
metric amount of 67% nitric acid and 10 mol % of the cata-
lysts. In the control experiments with no catalyst, only slow
reaction occurred and only a small amount of nitrated
products were observed. We initially anticipated that due to
the stronger electron-withdrawing ability of the 2,4-NBSA
group, La(2,4-NBSA)3 would have a stronger Lewis acidity
than La(p-NBSA)3 and La(m-NBSA)3. Therefore, La(2,4-
NBSA)3 was expected to be a more efficient catalyst than
La(p-NBSA)3 and La(m-NBSA)3. However, as Table 2
shows, in all reactions La(p-NBSA)3 and La(m-NBSA)3
were superior nitration catalysts compared to the La(2,4-
NBSA)3 compounds. In all the cases no dinitrated products
were observed, and in accordance with this, none of the
catalysts was effective for the nitration of nitrobenzene.
Effect of the Lanthanide(III) Ion on the Nitration Reaction
As already mentioned, the initial screening of the cata-
lytic activity showed that lanthanum() catalysts were su-
perior to ytterbium() catalysts. In order to elucidate the
role of the metal center in the nitration reaction, a series
of Ln(p-NBSA)3 and Ln(m-NBSA)3 catalysts with different
lanthanide() ions was examined as potential catalysts. Al-
though all complexes were found to exhibit catalytic
properties, marked differences were apparent. There is a
clear correlation between the ionic radii of the lantha-
nide() ions and the yields of the nitration, with the lighter
lanthanides being the more effective ones. As a result,
among Ln(m-NBSA)3 catalysts, lanthanum() (atomic
number 57) gave 80% conversion of p-xylene to nitro-p-xy-
lene over 16 h, whereas the ytterbium() (atomic number
70) catalyst yielded only 20% of the nitrated product over
the same period (Figure 2). A similar trend was observed
for the nitration of naphthalene catalysed by Ln(p-NBSA)3
(Table 4), although the relative differences in catalytic ac-
tivities between lighter and heavier lanthanides were
smaller (Table 3).
Table 2. Nitration of aromatics in the presence of different lantha-
num() catalysts
Conversion (%)[a][b]
Arene
La(p-NBSA)3 La(m-NBSA)3
La(2,4-NBSA)3
This trend is opposite to the trend observed in lantha-
nide() triflate catalysts in which there is an inverse corre-
lation between the catalytic activity and the ionic radii of
the lanthanide() ion.[10] It has been reasoned that the elec-
trostatic interaction between the lanthanide ion and nitric
acid, which is crucial for the formation of the nitronium
ion (NO2ϩ) as the de facto nitrating agent, increases as the
polarising power (i.e. charge-to size-ratio) of lanthanide ion
increases. However, it is obvious that in the case of lantha-
nide() nitrobenzosulfonate catalysts the electrostatic inter-
actions do not play the decisive role in the extent of ni-
tration, since the complex with the least polarizing lantha-
nide (i.e. the lanthanum ion) is the most effective. The iso-
mer distributions for the nitration of arenes are nevertheless
consistent with the distribution expected for the direct elec-
toluene[c]
Ͼ 98
Ͼ 98
Ͼ 98
77
Ͼ 98
80
Ͼ 98
Ͼ 98
58
74
32
18
19
naphthalene[d] Ͼ 98
benzene
82
Ͼ 98
78
m-xylene[e]
p-xylene
bromobenzene[f] 74
chlorobenzene[g] 76
69
70
[a]
3 mmol of the substrate with 10 mol % of the catalyst and 1
[b]
equiv. of 67% HNO3 in refluxing 1,2-dichloroethane for 16 h.
GC/1H NMR analysis; the product distribution of nitro isomers
was independent on the catalyst. [c] Isomer distributions (%): ortho:
[d]
52; meta: 6; para: 42.
Isomer distributions (%): 1-NO2: 91; 2-
[e]
[f]
NO2: 9. Isomer distributions (%): 4-NO2[:g]87; 2-NO2: 13. Iso-
mer distributions (%): ortho: 42; para: 58.
(%): ortho: 44; para: 56.
Isomer distributions
4562
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 4560Ϫ4566

Lanthanide(III) Nitrobenzenesulfonates as New Nitration Catalysts
FULL PAPER
lysts. In lanthanide() triflates, the triflate counterions are
outer sphere and are merely spectator ions. X-ray crystal
structures of hydrated lanthanides(III) triflates have been
reported for the entire lanthanide series, and they all show
that nine water molecules occupy the first coordination
sphere and that triflate counterions are in the outer
sphere.[20] The solution studies on lanthanides(III) triflates
suggest that this structure is preserved in solution and that
the triflates are essentially spectator ions.[21] We were able
to determine the X-ray single crystal structure of Yb(m-
NBSA)3·6H2O, and this structure shows that the coordi-
nation environment around the ytterbium() is occupied by
two oxygen atoms from the sulfonate group of m-NBSA
and six oxygen atoms from water molecules (Figure 3,
Table 5, Table 6). The coordination polyhedron around yt-
terbium() can be described as a slightly distorted quad-
ratic antiprism. The third m-NBSA anion is not located in
the first coordination sphere, but forms hydrogen bonds
with two of the coordinated water molecules. The dihedral
angle between the nitro group and the plane of the phenyl
ring of the non-coordinating 3-nitrobenzenesulfonate mol-
ecule is with 20.1(10)° noticeable greater than the angle for
the two coordinating molecules with 2.5(11)° and 6.1(9)°,
respectively.
This difference in structure compared to ytterbium()
triflate may explain the significant differences in the ef-
ficiency of these two catalysts. In order for the interaction
between lanthanide() ion and the nitric acid to occur, dis-
placement of one of the ligands in the first coordination
sphere is necessary. While in the case of Yb(OTf)3 there is
no steric hindrance for this interaction, it is conceivable that
in the case of Yb(m-NBSA)3, the bulky m-NBSA ligands
present in the first coordination sphere may hinder this in-
teraction. This may also explain why Ln(2,4-NBSA)3 cata-
lysts with bulky counterions are inferior compared to Ln(p-
NBSA)3 and Ln(m-NBSA)3.
Figure 2. Effect of the lanthanide() ion in Ln(m-NBSA)3 on the
conversion of p-xylene to nitro-p-xylene. The ionic radii of the lan-
thanide() ions are shown for the sake of comparison
Table 4. Effect of various Ln(p-NBSA)3 for the nitration of naph-
thalene
LnIII(%)[a] [b]
Conversion
La
100
Nd
Sm
Eu
Gd
Dy
Tm
Yb
97.1
88.6
93.9
83.4
82.4
82.1
72.5
[a]
3 mmol of the substrate with 10 mol % of the catalyst and 1
equiv. of 67% HNO3 in refluxing 1,2-dichloroethane for 30 min. [b]
GC analysis; the product distribution of 1-NO2 and 2-NO2 isomers
was in each reaction 92:8.
trophilic attack by the nitronium ion. The mechanism by
which the nitronium ion is formed presumably involves the
binding of nitric acid to the lanthanide() ion, resulting in
proton liberation due to the strong polarization by the me-
tal. The liberated proton reacts further with another mol-
ecule of nitric acid yielding a nitronium ion in the classical
manner.[19] It is at this point important to note that the
identity of the counterion is crucial for the catalytic activity.
It has been shown that simple lanthanide salts such as lan-
thanide() chlorides or nitrates are poor nitration cata-
lysts.[10] The counterion needs to be the conjugate base of
a sufficiently strong acid so that nitric acid is preferentially
protonated, producing a nitronium ion.
Unfortunately, we were not able to isolate crystals of lan-
thanum() catalysts of a quality good enough for the sin-
gle-crystal X-ray analysis, so it is not possible to tell if in
these complexes counterions are directly bound to the lan-
thanide() centre or if they are merely spectator ions. Even
if the counterions are located in the first coordination
sphere, the steric hindrance around larger lanthanum()ion
should be lesser than the smaller ytterbium() ion. How-
ever, it is possible that lanthanum() complexes of nitro-
benzenesulfonic acids follow the same structural pattern as
analogous lanthanum() tosylate complexes, in which there
are nine water molecules occupying the first coordination
sphere while the counterions are not interacting with the
metal centre.[22] This type of structure would be consistent
with the high activity of the lanthanum() catalyst, since
Role of the Counterions in the Nitration Reaction
How can we explain that the extent of nitration is the the absence of bulky counterions in the first coordination
smallest in the case of strongly polarizing lanthanide ions? sphere poses no steric hindrance for the interaction between
Since lanthanide() nitrobenzenesulfonates and lantha- the lanthanum() and nitric acid. In contrary, the ytter-
nide() triflate catalysts differ only in the nature of the bium() tosylate complex has a structure very similar to
counterion, it is likely that the anions play the decisive role the structure of Yb(m-NBSA)3 in which two ligands are
in catalytic reaction. Therefore, it is helpful to consider the directly bound to ytterbium().[16] The analogy between
interactions of these anions with the lanthanide() cata- the structures of the lanthanide() nitrobenzenesulfonate
Eur. J. Org. Chem. 2004, 4560Ϫ4566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4563

K. Binnemans et al.
FULL PAPER
Figure 3. Molecular structure of Yb(m-NBSA)3·6H2O
and lanthanide() tosylate complexes is not surprising, if
one takes into account the similarities of the nitroben-
zenesulfonate and tosylate structures and their similar pKa
values.
Table 5. Crystallographic data for the compound Yb(m-
NBSA)3·6H2O
Formula
C18H12N3O22S3Yb
891.56
0.2 ϫ 0.2 ϫ 0.1
colorless
monoclinic
C2/c
56.55(3)
7.624(4)
14.158(8)
93.28(3)
6094(6)
8
Formula mass
Dimensions (mm)
Color
Solution Studies of Ln(p-NBSA)3 and Ln(m-NBSA)3
Catalysts
Crystal system
Space group
1H NMR spectra of Ln(p-NBSA)3 and Ln(m-NBSA)3
complexes were measured in order to gain more infor-
mation about the structure of these complexes in solution.
In aqueous solutions, proton resonances of the ligand
showed the same chemical shifts as in the fully dissociated
nitrobenzenesulfonic acids, implying that there was no
binding of the nitrobenzenesulfonate ligand to the lantha-
nide() ion, so that complexes existed as purely lantha-
nide() aqua ions. The concentration of water in the NMR
measurements was about five orders of magnitude greater
than the concentration of the catalyst. Since the lantha-
nide() ions have a high affinity toward water molecules, it
is not surprising that under the condition in which there is
a huge excess of water they tend to exist as aqua complexes.
However, the water solution clearly does not represent
the experimental conditions applied in the nitration
reaction. The nitration reactions were performed in 1,2-
dichloroethane, and the only source of water comes from
the 67% solution of the nitric acid. A simple calculation
indicates that under reaction conditions the catalyst and the
water were present in approximately equimolar amounts.
Attempts to record proton NMR spectra under conditions
which exactly mimic the nitration reaction conditions were
unsuccessful, due to the low solubility of lanthanide()
catalysts in chlorinated solvents. As an alternative, acetone
was used as solvent and the proton NMR spectra of Eu(m-
NBSA)3 complexes were recorded in different [D6]acetone/
D2O ratio. As Figure 4a shows, in the presence of a small
amount of D2O (less than 5 volume%) two sets of peaks
with the intensity ratio of 2:1 were observed. The larger set
˚
a (A)
˚
b (A)
˚
c (A)
β (°)
3
˚
V (A )
Z
Dcalcd. (g·cmϪ3
θmax. (°)
)
1.944
70.72
8.499
3480
µCu-Kα (mmϪ1
F(000)
)
Measured reflections
Unique reflections
Observed reflections [Fo Ͼ 4σ(Fo)]
Parameters refined
R1
ωR2
R1 (all data)
23853
5725
4578
424
0.0815
0.2240
0.0972
0.2388
1.068
ωR2 (all data)
GooF
Table 6. YbϪO distances for Yb(m-NBSA)3·6H2O
˚
Atoms
Distance (A)
Yb1ϪO3
Yb1ϪO16
Yb1ϪO21
Yb1ϪO8
Yb1ϪO19
Yb1ϪO17
Yb1ϪO18
Yb1ϪO20
2.276(5)
2.311(5)
2.319(5)
2.330(4)
2.346(5)
2.374(5)
2.394(5)
2.416(5)
4564
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 4560Ϫ4566

Lanthanide(III) Nitrobenzenesulfonates as New Nitration Catalysts
FULL PAPER
of peaks which appeared in the region between δ ϭ 0 and larger lanthanum() ion. Similarly, the steric reasons may
2 ppm could be assigned to m-NBSA ligands which are explain why lanthanide() catalysts with the bulkier 2,4-
bound to europium(). Due to this interaction m-NBSA nitrobenzenesulfonate ligand are less efficient than the cor-
protons experience a large paramagnetic shift. The smaller responding meta- and para-nitrobenzenesulfonate lantha-
set of peaks appeared in the region between 7.5 and 9 ppm nide() catalysts.
and was assigned to one m-NBSA ligand which is not di-
rectly bound to europium(). Upon addition of D2O (ca.
Experimental Section
40 volume%), the larger set of peaks disappeared and all
m-NBSA resonances appeared in the region between 7.5
and 9 ppm (Figure 4, b).
Materials and Methods: Reagents were obtained from Aldrich
Chemical Co. Inc., Acros Organics or Rhodia Electronics and Ca-
talysis, and used without further purification. The p-nitroben-
zenesulfonic acid is commercially available, but it can be also ob-
tained very cheaply, in a one-step procedure starting from sulfanilic
acid.[23] 1H NMR spectra were run with a Bruker Avance 300, op-
erating at 300 MHz. GC analyses were performed at ThermoFinni-
gan Trace GC. Elemental analysis was performed on a CE Instru-
ments EA-1110 elemental analyser.
General Procedure for the Synthesis of Ln(p-NBSA)3, Ln(m-NBSA)3
and Ln(2,4-NBSA)3
The compounds were synthesized by the reaction of 1.1 equivalent
of the corresponding lanthanide() oxide, and six equivalents of
the sulfonic acid in aqueous solution. After stirring the solution in
boiling water for 30 minutes, the excess of oxide was removed by
filtration. The filtered solution was either left to crystallize or was
evaporated to dryness and the resulting solid was dried in a vacuum
oven at 50 °C overnight.
Results
for
Ln(p-NBSA)3·xH2O:
La(p-NBSA)3·2H2O
Figure 4. Proton NMR spectra of Eu(m-NBSA)3 in [D6]acetone
containing (a) 5 vol% D2O, and (b) 40 vol% D2O. The large peaks
at ca. 2.2, 3.0 and 4.6 ppm belong to the solvents
(C18H16LaN3O17S3): calcd. C 27.66, H 2.06, N 5.37; found C
27.37, H 1.97, N 5.15. Nd(p-NBSA)3·3H2O (C18H18N3NdO18S3):
calcd. C 26.86, H 2.26, N 5.22; found C 26.42, H 2.27, N 4.98.
Sm(p-NBSA)3·2H2O (C18H16N3O17S3Sm): calcd. C 27.26, H 2.03,
N 5.29; found C 27.16, H 1.97, N 5.29. Eu (p-NBSA)3·5H2O
(C18H22EuN3O20S3): calcd. C 25.47, H 2.61, N 4.95; found C 25.16,
H2.64, N 4.65. Gd(p-NBSA)3·5H2O (C18H22GdN3O20S3): calcd. C
25.32, H 2.59, N 4.92; found C 24.84, H 2.53, N 4.63. Dy(p-
These experiments suggest that under condition in which
the amount of water is limited, the solid-state structure is
preserved in the solution, and that two m-NBSA ligands
bind to the lanthanide() ion. Only upon adding large
amount of water, these ligands become spectator ions.
NBSA)3·3H2O (C18H18DyN3O18S3): calcd.
C 26.27, H 2.21,
N 5.09; found C 26.34, H 2.31, N 4.91. Er(p-NBSA)3·2H2O
(C18H16ErN3O17S3): calcd. C 26.70, H 2.00, N 5.20; found C 27.20,
H 2.01, N 5.22. Tm(p-NBSA)3·5H2O (C18H22N3O20S3Tm): calcd.
C 24.979, H 2.56, N 4.85; found C 24.75, H 2.56, N 4.68. Yb(p-
NBSA)3·5H2O, (C18H22N3O20S3Yb): calcd. C 24.86, H 2.55, N
4.83; found C 24.10, H 2.67, N 4.45.
Conclusions
In summary, we have demonstrated that efficient lantha-
nide() nitration catalysts can be prepared from para- and
meta-nitrobenzenesulfonic acids. These acids have some
major advantages compared to triflic acid as they are non-
corrosive and cheaper to obtain. The catalytic efficiency of
their lanthanum() complexes for the atom economic ni-
tration of arenes is comparable to the catalytic efficiency of
Elemental Analysis Results for Ln(m-NBSA)3·xH2O: La(m-
NBSA)3·4H2O (C18H20N3O19S3La): calcd. C 26.45, H 2.46, N 5.14;
found
(C18H16N3O17PrS3): calcd. C 27.60, H 2.06, N 5.37; found C 27.68,
H 2.33, N 5.45. Nd(m-NBSA)3·4H2O (C18H20N3NdO19S3): calcd.
C 26.27, H 2.45, N 5.10; found C 26.20, H 2.43, N 4.68. Sm(m-
C
26.48,
H
2.34,
N
4.89. Pr(m-NBSA)3·2H2O
ytterbium() triflate. We have shown that contrary to trifl- NBSA)3·4H2O (C18H20N3O19S3Sm): calcd. C 26.08, H 2.43, N
5.06; found C 25.98, H 2.37, N 4.60. Eu (m-NBSA)3·4H2O
(C18H20EuN3O19S3): calcd. C 26.03, H 2.42, N 5.06; found C 25.95,
H 2.38, N 4.70. Gd(m-NBSA)3·4H2O (C18H20N3O19S3Gd): calcd.
C 25.86, H 2.41, N 5.02; found C 25.70, H 2.35, N 4.61. Dy(m-
NBSA)3·4H2O (C18H20DyN3O19S3): calcd. C 25.70, H 2.39, N 4.99;
ate salts, a simple charge-to-size ratio of a lanthanide()
ion cannot be used for the relative reactivity of the lantha-
nide nitrobenzenesulfonate salts as nitration catalysts. The
evidence from the solid state and solution studies suggest
that under the reaction conditions, the nitrobenzenesulfon-
ate ions are likely to be bound to the lanthanide() ion,
which could cause a steric hindrance for the interaction be-
tween a molecule of nitric acid and the lanthanide() ion.
Because of the steric effects, the catalysts with the small-
found
C
25.69,
H
2.36,
N
4.56. Er(p-NBSA)3·H2O
(C18H14N3O16S3Er): calcd. C 27.30, H 1.79, N 5.30; found C 27.20,
H 2.01, N 5.22. Tm(p-NBSA)3·5H2O,(C18H22N3O20S3Tm): calcd.
C 24.98, H 2.56, N 4.85; found C 24.75, H 2.56, N 4.68. Yb(p-
NBSA)3·4H2O (C18H20N3O19S3Yb): calcd. C 25.38, H 2.37, N
sized ytterbium() ion are inferior to the catalysts with the 4.93; found C 25.76, H 2.51, N 5.04.
Eur. J. Org. Chem. 2004, 4560Ϫ4566
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4565

K. Binnemans et al.
FULL PAPER
Elemental Analysis Results for Ln(2,4-NBSA)3·xH2O: La(2,4-
NBSA)3·6H2O (C18H21LaN6O27S3): calcd. C 21.87, H 2.14, N 8.50;
[3]
[4]
S. Kobayashi, M. Sugiura, H. Kitagawa, W.W. L. Lam, Chem.
Rev. 2002, 6, 2227Ϫ2302.
S. Kobayashi, I. Hachiya, M. Araki, H. Ishitani, Tetrahedron
Lett. 1993, 34, 3755Ϫ3758.
found
C
21.43,
H
2.24,
N
7.91. Yb(2,4-NBSA)3·6H2O
(C18H21N6O27S3Yb): calcd. C 21.14, H 2.07, N 8.21; found C 20.76,
H 2.17, N 7.69.
[5] [5a]
[5b]
S. Kobayashi, Chem. Lett. 1991, 2187Ϫ2190.
S. Arseni-
yadis, Q. Wang, D. Yashunsky, T. Kaoudi, A. R. Brondu, R.
[5c]
Potier, Tetrahedron Lett. 1995, 36, 1633Ϫ1636.
S. Kobaya-
Single Crystal X-ray Data for Yb(m-NBSA)3·6H2O: A single crystal
was obtained by slow evaporation of an aqueous solution of the
ytterbium() complex. A crystal of dimensions 0.2 ϫ 0.2 ϫ 0.1 mm
was mounted in a nylon loop for data collection at 110 K with a
SMART 6000 diffractometer equipped with CCD detector using
shi, I. Hachiya, J. Org. Chem. 1994, 59, 3590Ϫ3596. [5d] S. Ko-
bayashi, I. Hachiya, Tetrahedron Lett. 1992, 33, 1625Ϫ1628.
[5e] S. Kobayashi, T. Wakabayashi, S. Nagayama, H. Oyamada,
[5f]
Tetrahedron Lett. 1997, 38, 4559Ϫ4562.
S. Kobayashi, T.
Busujima, S. Nagayama, Chem. Commun. 1998, 19Ϫ20.
E. Keller, B. L. Feringa, Synlett 1997, 842Ϫ842.
˚
[6]
[7] [7a]
Cu-Kα radiation (λ ϭ 1.54178 A). The images were interpreted and
integrated with the program SAINT from Bruker. The structure
was solved by direct methods and refined by full-matrix least-
squares on F2 using the SHELXTL program package.[24] Non-hy-
drogen atoms were anisotropically refined and the hydrogen atoms
in the riding mode with isotropic temperature factors. The refine-
ment was complicated by the high crystal mosaicity. CCDC-246951
contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge via www.cam.ac.uk/
conts/retrieving.html (or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax:
ϩ 44-1223-336033; or deposit@ccdc.cam.uk).
I. Hachiya, S. Kobayashi, J. Org. Chem. 1993, 58,
[7b]
6958Ϫ6960.
38, 853Ϫ856.
T. Akiyama, J. Iwai, Tetrahedron Lett. 1997,
[8]
[9]
M. Ouertani, P. Girard, H. B. Kagan, Tetrahedron Lett. 1982,
23, 4315Ϫ4318.
F. J. Waller, A. G. M. Barrett, D. C. Braddock, D. Ramprasad,
Chem. Commun. 1997, 613Ϫ614.
F. J. Waller, A. G. M. Barrett, D. C. Braddock, R. M. McKin-
nell, D. Ramprasad, J. Chem. Soc., Perkin Trans. 1 1999,
867Ϫ871.
[10]
[11]
[12]
M. Shi, S-C. Cui, Chem. Commun. 2002, 994Ϫ995.
S. K. Fischer, M. McFarland, ‘‘Alternatives to CFCs and Global
warming: A systems approach to evaluating net contributions’’,
Materials Research Society, 1992, vol. 17, p. 39.
T. N. Parac-Vogt, K. Deleersnyder, K. Binnemans, J. Alloy.
Comp. 2004, 374, 46Ϫ49.
General Procedure for the Lanthanide(III) Nitrobenzenesulfonate-
Catalysed Nitration of Arenes: To a solution of the respective lan-
thanide() catalyst (5 mol %, 0.3 mmol) in nitric acid (67%, d ϭ
1.41, 0.200 mL, 3.0 mmol) was added a solution of the aromatic
compound (3 mmol) in 4 mL of 1,2-dichloroethane. The mixture
was stirred and heated at reflux. After a given period of time the
solution was cooled and diluted with water. The yellow organic
phase was dried with MgSO4, the solvent was evaporated and the
[13]
[14]
[15]
[16]
[17]
[18]
T. N. Parac-Vogt, K. Binnemans, Tetrahedron Lett. 2004, 45,
3137Ϫ3139.
D. C. French, D. S. Crumrine, J. Org. Chem. 1990, 55,
5494Ϫ5496.
Y. Ohki, Y. Suzuki, T. Takeuchi, A. Ouchi, Bull. Chem. Soc.,
Jpn. 1988, 61, 393.
C. L. Coon, W. G. Blucher, M. E. Hill, J. Org. Chem. 1973,
1
residue was analysed by H NMR or gas chromatography.
38, 4243Ϫ4248.
E. Pretsch, T. Clerc, J. Seibl, W. Simon in Tabellen zur Struktu-
raufklärung organischer Verbindungen mit spektroskopischen
Methoden, Springer, Berlin, 1976.
Acknowledgments
T. N. P. V. and K. B. are Postdoctoral Fellows of the FWO-Flanders
(Belgium). Financial support has been provided by the K. U.
Leuven (VIS/01/006.01/2002/-06/2004 and GOA 03/03) and by the
FWO-Flanders (Krediet aan Navorsers to K. B.). The authors
thank Rhodia electronics and Catalysis for a gift of rare-earth
oxides.
[19] [19a]
G. A. Olah, R. Malhotra, S. C. Narang, Nitration: Meth-
[19b]
ods and Mechanisms; VCH Publishers, New York, 1989.
K. Schofield, Aromatic Nitration; Cambridge University Press,
Cambridge, 1980.
[20] [20a]
J. M. Harrowfield, D. L. Kepert, J. M. Patrick, A. H.
[20b]
White, Aust. J. Chem. 1983, 36, 483Ϫ492.
A. Chatterjee,
E. N. Maslen, K. J. Watson, Acta Crystallogr., Sect. B 1988,
[1]
44, 381Ϫ386.
‘‘Lanthanides: Chemistry and Use in Organic Synthesis’’ (S. Ko-
[21]
[22]
B. P. Hay, Inorg. Chem. 1991, 30, 2876Ϫ2884.
D. L. Faithfull, J. M. Harrowfield, M. I. Ogden, B. W. Skelton,
K. Third, A. H. White, Aust. J. Chem. 1992, 45, 583Ϫ594.
M. Vincent, J. Maillard, M. Benard, Bull. Soc. Chim. Fr.
1962, 158.
SAINT, Manual Version 5/6.0, Bruker Analytical X-ray Sys-
tems Inc.: Madison, Wisconsin, 1997. SHELXTL-NT, Manual
Version 5.1, Bruker Analytical X-ray Systems Inc.: Madison,
Wisconsin, 1997.
bayashi, Ed.), Springer, Berlin, 1999.
[2]
[2a] H. B. Kagan, J. L. Namy, Tetrahedron 1986, 42, 6573Ϫ6614.
[2b]
[2c]
G. A. Molander, Chem. Rev. 1992, 92, 29Ϫ68.
Molander, C. R. Harris, Chem. Rev. 1996, 96, 307Ϫ338.
G. A.
[23]
[24]
[2d]
G.
A. Molander, C. R. Harris, Tetrahedron 1998, 54, 3321Ϫ3354.
[2e]
[2f]
A. Krief, A. M. Laval, Chem. Rev. 1999, 99, 745Ϫ777.
H. J. Liu, K. S. Shia, X. Shang, B. Y. Zhu, Tetrahedron 1999,
[2g]
55, 3803Ϫ3830.
P. G. Steel, J. Chem. Soc., Perkin Trans. 1
2001, 2727Ϫ2751. [2h] J. Inanaga, H. Furuno, T. Hayano, Chem.
Rev. 2002, 102, 2211Ϫ2225.
Received July 5, 2004
4566
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 4560Ϫ4566