Zr-TUD-1: A lewis acidic, three-dimensional, mesoporous, zirconium-containing catalyst

Article abstract of DOI:10.1002/chem.200700725

A three-dimensional, mesoporous, silicate containing zirconium, Zr-TUD-1, was synthesized by a direct hydrothermal treatment method with triethanolamine as a complexing and templating reagent to ensure that zirconium was incorporated as isolated atoms. The mesoporosity of Zr-TUD-1 was confirmed by X-ray diffraction (XRD), N₂ sorption and high-resolution transmission electron micrograph (HR-TEM) studies. The nature and strength of the Lewis acid sites present in Zr-TUD-1 were evaluated by FTIR studies of pyridine adsorption and temperature-programmed desorption of ammonia. FTIR, X-ray photoelectron spectroscopic (XPS) and UV/Vis spectroscopic studies showed that, at Si/Zr ratios of 25 and higher, all the zirconium was tetrahedrally incorporated into the mesoporous framework, while at low Si/Zr ratios, a small part of the zirconium was present as ZrO₂ nanoparticles. Zr-TUD-1 is a Lewis acidic, stable and recyclable catalyst for the Meerwein-Ponndorf-Verley (MPV) reaction and for the Prins reaction.

Full text of DOI:10.1002/chem.200700725

FULL PAPER
DOI: 10.1002/chem.200700725
Zr-TUD-1: A Lewis Acidic, Three-Dimensional, Mesoporous, Zirconium-
Containing Catalyst
Anand Ramanathan,[a, b, c] M. Carmen Castro Villalobos,[a, d] Cees Kwakernaak,[e]
Selvedin Telalovic,[a] and Ulf Hanefeld*[a]
Dedicated to J. C. “Koos” Jansen, the inventor of TUD-1, on the occasion of his retirement
Abstract: A three-dimensional, meso-
porous, silicate containing zirconium,
Zr-TUD-1, was synthesized by a direct
hydrothermal treatment method with
triethanolamine as a complexing and
templating reagent to ensure that zirco-
nium was incorporated as isolated
atoms. The mesoporosity of Zr-TUD-1
was confirmed by X-ray diffraction
(XRD), N2 sorption and high-resolu-
tion transmission electron micrograph
(HR-TEM) studies. The nature and
strength of the Lewis acid sites present
in Zr-TUD-1 were evaluated by FTIR
studies of pyridine adsorption and tem-
perature-programmed desorption of
ammonia. FTIR, X-ray photoelectron
spectroscopic (XPS) and UV/Vis spec-
troscopic studies showed that, at Si/Zr
ratios of 25 and higher, all the zirconi-
um was tetrahedrally incorporated into
the mesoporous framework, while at
low Si/Zr ratios, a small part of the zir-
conium was present as ZrO2 nanoparti-
cles. Zr-TUD-1 is a Lewis acidic, stable
and recyclable catalyst for the Meer-
wein–Ponndorf–Verley (MPV) reaction
and for the Prins reaction.
Keywords: heterogeneous catalysis ·
isopulegol
·
Lewis
acids
·
mesoporous materials · zirconium
Introduction
in the presence of carbon–carbon double bonds or in oxida-
tions of alcohols as exemplified in the Meerwein–Ponndorf–
Verley reduction/Oppenauer oxidation (MPVO reaction). In
this reversible reaction that was first described as a reduc-
tion (Meerwein—Ponndorf–Verley), the Lewis acid
Lewis acids are versatile catalysts. They can promote the se-
lective transfer of hydride ions in the reduction of ketones
aluminium(III) isopropoxide is commonly used in stochio-
G
metric quantities and recycling is difficult and often impossi-
ble.[1] Therefore, heterogeneous catalysts have been devel-
oped as an alternative.[2–5] The mechanism of both the ho-
mogenous and the heterogeneous MPVOreaction has been
shown to proceed through a carbon-to-carbon hydride trans-
fer from an alcohol to a ketone or aldehyde (Scheme 1).[6–8]
The heterogeneous catalysts can also be used to racemise
benzylic alcohols; in this case, however, the reaction occurs
through an addition–elimination process.[6,9]
[a] Dr. A. Ramanathan, Dr. M. C. Castro Villalobos, S. Telalovic,
Dr. U. Hanefeld
Gebouw voor Scheikunde, Technische Universiteit Delft
Julianalaan 136, 2628 BL Delft (The Netherlands)
Fax : (+31)15-278-1415
E-mail: U.Hanefeld@tudelft.nl
[b] Dr. A. Ramanathan
Advanced Chemical Technology Division
Korea Research Institute of Chemical Technology (KRICT)
P.O. Box 107, Yusung, Daejeon 305–600 (Republic of Korea)
[c] Dr. A. Ramanathan
Moreover, Lewis acids catalyse such important reactions
as the carbon–carbon bond forming Prins reaction. Prins de-
scribed this reaction for the first time in 1917[10,11] and point-
Current address: Department of Chemistry
National Institute of Technology
Tiruchirappalli - 620015 (India)
[d] Dr. M. C. Castro Villalobos
Departamento de Tecnología Química y Ambiental
Universidad Rey Juan Carlos
Tulipµn, s/n, 28933 Móstoles, Madrid (Spain)
[e] C. Kwakernaak
Department of Materials Science and Engineering
Technische Universiteit Delft, Mekelweg 2
2628 CD Delft (The Netherlands)
Scheme 1. The Lewis acid catalysed Meerwein–Ponndorf–Verley reduc-
tion/Oppenauer (MVPO) oxidation occurs through carbon-to-carbon hy-
dride transfer.
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
961

ed out that the cyclisation of citronellal to isopulegol
(Scheme 2) described by him earlier that year[12] would also
follow the same mechanism. This cyclisation is an essential
step of the Takasago synthesis of menthol.[13] Industrially,
this Prins reaction is catalysed by stochiometric amounts of
Consequently, the preparation of Zr-containing mesopo-
rous materials is of great interest. However, the grafting of
zirconium(IV) propoxide onto MCM-41, MCM-48 and
SBA-15 is labour intensive.[33,38] Furthermore, several of the
zirconium-containing mesoporous materials are one-dimen-
sional, hence diffusion problems might once again arise.[37,41]
A less labour intensive preparation of zirconium-containing
mesoporous silicate again yields a material with a one-di-
mensional pore structure that limits diffusion.[42] Hence, the
mesoporous materials prepared so far are either not readily
accessible and/or one-dimensional. Both disadvantages need
to be overcome in order to provide a straightforward cata-
lytic system.
The recently reported three-dimensional sponge-like
meso
porous material TUD-1[43] has many advantages over
G
the conventional mesoporous materials HMS, MCM-41 and
MCM-48, such as a cost-effective one-pot synthesis proce-
dure, a tunable pore size and a better accessibility for sub-
strates and products. These properties of TUD-1 are ideal
for the incorporation of zirconium into the three-dimension-
al silica network either as isolated species or as zirconia
clusters supported on the pore walls of TUD-1, depending
on the Zr-loading. Earlier studies with other metals includ-
ing zirconium indicated that, at loadings of up to 4%, the
metal atoms are isolated in the framework of TUD-1.[44–51]
In the synthesis of Ti-TUD-1, triethanolamine was demon-
strated to be an ideal complexing agent for titanium.[44]
Indeed, this complex and also the zirconium complex were
isolated and fully characterised by others.[52] Triethanol-
Scheme 2. A) In the acid-catalysed Prins reaction, a new carbon–carbon
bond is formed between an olefin and an aldehyde, a propane-1,3-diol is
produced. B) The cyclisation of citronellal (1) to isopulegol (2) follows
the same mechanism.
solid ZnBr2 in an organic solvent, originally benzene.[13,14]
The ZnBr2 can be replaced by very selective homogeneous
catalysts, such as tris-(2,6-diarylphenoxy)aluminium,[15,16]
[1]
that resemble MPVOcatalysts. The removal of homogene-
ous catalysts is, in general, cumbersome and introduces
lengthy workup procedures.[17,18] Recently, the application of
catalytic amounts of zeolites and mesoporous materials as
catalysts for this reaction has attracted considerable atten-
tion.[19–27]
amine thus ensured the incorporation of titanium as isolated
Since the discovery of MCM-41 mesoporous materials,[28]
there has been tremendous development in the synthesis
and modification of the silica framework. By virtue of their
relatively large pores, mesoporous materials overcome the
size and diffusion limitations that the microporous zeolites
have. Acidity can be imparted to zeolites to obtain specific
catalytic properties for a targeted reaction. For example,
aluminium was incorporated to impart acidity[29–32], but
other acidic metals such as zirconium also have this
effect.[33–38] Zirconium is a particularly interesting metal,
since tetrahedrally coordinated zirconium is only Lewis
acidic and not Brønsted acidic.[8] In zirconia, zirconium has
proven not only to be acidic,[39] but it has also been demon-
strated to be very active in dehydration, hydrogenation, oxi-
dation and hydride exchange reactions.[5,40] The very low sur-
face area of zirconia was enlarged by supporting zirconium
on silica and by the synthesis of both microporous and mes-
oporous materials in which the framework is substituted
with zirconium. The presence of tetrahedrally incorporated
Zr4+ in the framework makes them Lewis acidic. Indeed
these materials can catalyse both the MPVOreaction and
the Prins reaction.[25,33] As has recently been shown for Zr-
atoms into the silica framework of Ti-TUD-1. It proved to
act similarly in the synthesis mixtures of many other M-
TUD-1s (M=metal) and specifically of Zr-TUD-1.[51] Thus,
during the synthesis, no zirconia clusters can form and the
metal is incorporated into the framework of the TUD-1 as
an isolated atom. In this way, it is ensured that the metal is
both isolated and fully accessible for the substrate. Inde-
pendently, this triethanolamine approach was also devel-
oped for the preparation of Zr-MCM-41 and labelled as the
“atrane route”.[41,53,54] El Haskouri et al. demonstrated that
triethanolamine complexes both silicon and zirconium
atoms, ensuring the preparation of isolated metal sites in
siliceous materials. Indeed, they could unambiguously show
that both silicon and zirconium atrane complexes were pres-
ent in an isolated form in the reaction mixture. Even when
the two atranes were mixed, no silicon–silicon, zirconium–
zirconium or bimetallic dimers were observed by fast atom
bombardment mass spectrometry (FAB-MS), strongly indi-
cating that the mesoporous materials are formed from mon-
omeric atranes.[41,54] However, the materials prepared by this
“atrane route” have the above-mentioned disadvantage of
diffusion limitations, which are not present in the three-di-
mensional TUD-1. The sponge like character of TUD-1 was
demonstrated, among others, by three-dimensional electron
tomography studies.[55] Furthermore, M-TUD-1 has proven
to be a particularly stable mesoporous material that can be
recycled.[44–51]
beta, it is the LUMOd orbital of zirconium that accepts
z2
electron density from a Lewis base,[8] the zirconium remain-
ing tetrahedrally coordinated in the framework. In both the
Prins and the MPVOreaction, the Lewis base is identical:
one of the lone electron pairs of the carbonyl group.
962
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972

Zirconium-Containing Catalysts
FULL PAPER
In this paper, we report the direct, hydrothermal, cost-ef-
fective synthesis of zirconium-containing three-dimensional
mesoporous silica Zr-TUD-1 with different Si/Zr ratios. A
detailed characterisation of these materials obtained by dif-
ferent techniques such as X-ray diffraction (XRD), N2 sorp-
tion, chemical analysis, high-resolution transmission electron
micrographs (HR-TEM) and UV/Vis, FTIR and X-ray pho-
toelectron spectroscopy (XPS) are discussed. These materi-
als are evaluated for their Lewis acidic catalytic properties
in the MPV reduction and the Prins reaction.
Results and Discussion
Zr-TUD-1 as a Lewis acidic, mesoporous material: A broad
intense peak at low angle (0.4–2 q) was observed in the X-
ray powder diffraction pattern for all calcined Zr-TUD-1
samples (Figure 1) demonstrating the mesostructured char-
Figure 2. High-resolution transmission electron micrographs (HR-TEM)
of calcined Zr-TUD-1 samples Zr-TUD-1(25) (top: bar represents
20 nm) and b) Zr-TUD-1(10) (bottom: bar represents 3 nm).
Figure 1. Powder X-ray diffraction patterns of a) Zr-TUD-1
TUD-1(50), c) Zr-TUD-1(25) and d) Zr-TUD-1(10).
A
mesoporosity. As expected, no crystalline zirconia particles
were observed up to a loading of 4 mol% of zirconium in
the TUD-1 matrix (Si/Zr=25). However, at higher loadings,
in Zr-TUD-1(10), a few crystalline nanoparticles were visi-
ble (Figure 2, bottom). The absence of zirconia crystals in
the XRD-spectrum of the same sample indicates that the
concentration of these nanoparticles must be low. The ele-
mental analysis (ICP-OES) and porosity measurements ob-
tained from N2 sorption studies at 77 K are listed in Table 1.
Elemental analysis indicated that the Si/Zr ratios in the cal-
cined samples were the same as those in the initial synthesis
acter of these materials. No evidence for crystalline ZrO2
was observed in the X-ray diffractograms, indicating that zir-
conium is incorporated into the framework. In line with the
effects of metal loading in Co-TUD-1[46] and Al-TUD-1,[47]
the peak intensity decreases slightly with an increase in the
Zr loading, indicative of the influence of Zr loading on the
integrity of the mesoporous structure. The HR-TEM of the
sample Zr-TUD-1(25) (Figure 2, top) further confirms the
Table 1. Physicochemical characterisation of Zr-TUD-1 with different Zr loadings.
Si/Zr ratio
product[a]
SBET
[m2 gꢀ1
Total pore volume
[cm3 gꢀ1
Pore diameter
[nm]
Micropore area
Micropore volume
[cm3 gꢀ1
gel
G
]
U
]
A
E
]
Zr-TUD-1(10)
Zr-TUD-1(25)
Zr-TUD-1(50)
10
25
50
10
25
51
676
764
771
753
0.4
1.23
1.2
<2
8.8
8.9
8
n.a.
<1
32
n.a.
<0.001
0.009
Zr-TUD-1
(100)
100
102
1.05
<1
<0.001
[a] From ICP-OES analysis.
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
963

U. Hanefeld et al.
gel. This excellent correlation of the Si/Zr ratio in the syn-
thesis gel with that in the product demonstrates the high
predictability of the synthesis method. This was also ob-
served earlier for other M-TUD-1 materials.[46–48]
From the XPS measurements, the observed binding ener-
gies of Zr 3d5/2 (183.2ꢁ0.1 eV) presented in Table 2 differ
significantly from the value reported for the reference
sample ZrO2 (182.4 eV), but are close to that of the refer-
ence sample ZrSiO4 (183.3 eV).[56] The binding energy of O
1s1/2 observed at 532.4ꢁ0.2 eV is also significantly higher
Three samples (Zr-TUD-1(100), Zr-TUD-1(50) and Zr-
A
TUD-1(25)) show a type IV isotherm (Figure 3), indicated
Table 2. The binding energy of the O1s 1/2, Zr 3d5/2 photoelectrons for Zr-
TUD-1 samples.
Spectral line
Zr-TUD-1(10)
Zr-TUD-1(25)
Zr-TUD-1(100)
A
O1s
Zr 3d5/2
532.2
183.1
532.4
183.2
532.5
183.2
1/2
than both ZrO2 (530.2 eV) and ZrSiO4 (531.4 eV),[36,56] but
approaches reported values of SiO2 with values of about
532.9 eV as presented in the National Institute of Standards
and Technology (NIST) X-ray Photoelectron Spectroscopy
Database.[57] This is in line with the fact that all samples con-
tain significantly more Si-O-Si bridges than Si-O-Zr bridges.
Moreover, it indicates that the amount of ZrO2 in Zr-TUD-
1(10), although detectable by HR-TEM, is low. The bind-
ing-energy values for Zr 3d5/2 and O1s are similar to those
1/2
observed for Zr in the MFI structure,[58] zeolite beta[59] and
mesoporous silicas.[56,60,61] These two observations are strong
evidence for the presence of zirconium in the framework of
the TUD-1 matrix, even in the case of Zr-TUD-1(10).
The IR spectra in the skeletal region of a KBr pressed
disc of Si-TUD-1 and Zr-TUD-1 (Figure 4) showed typical
Figure 3. N2 adsorption and desorption isotherms at 77 K and the corre-
sponding pore size distribution of Zr-TUD-1 with different Zr loadings.
by the large uptake of nitrogen at relative pressures between
0.5 and 0.9 p/p0 (due to capillary condensation in the meso-
pores). These samples also show a plateau at relative pres-
sures above 0.9 p/p0, indicating the absence of large meso-
pores, macropores or surface roughness in the measurable
range (20 nm to approximately 500 nm). Additionally, these
three samples have a larger pore size when compared to the
all silica TUD-1 (ca. 4 nm). Presumably, this is due to the
presence of the large zirconium atoms in the framework.
The presence of some percolation or networking effects can
also be deduced from a non-parallel adsorption and desorp-
tion isotherm.
Zr-TUD-1(10), on the other hand, showed an uptake of
nitrogen at a relative pressure of up to approximately 0.5 p/
p0 and no adsorption is observed above this point. This im-
plies the absence of large mesopores or macropores or sur-
face roughness in the measurable range. Since pore filling
occurs below the critical pressure of 0.43 p/p0, no hysteresis
was observed. This sample displayed very little mesoporosity
and a maximum pore-size distribution below 2 nm. As a
result of this, the pore volume is much lower than in the
other three Zr-TUD-1 samples; however, the surface area of
Zr-TUD-1(10) is comparable with the other three Zr-TUD-
1 samples. The presumption that the large Zr-atom would
cause an increase in pore size is not observed at this high
concentration of zirconium. Instead, a differently structured
material was formed with relatively small pores. This differ-
ence might be due to intensified interaction of the individual
Zr atoms in the structure or due to the nanoparticles identi-
fied by HR-TEM.
Figure 4. FTIR skeletal spectra a) Si-TUD-1, b) Zr-TUD-1
TUD-1(50), d) Zr-TUD-1(25) and e) Zr-TUD-1(10).
A
bands at 1093 cmꢀ1 and a shoulder at 1220 cmꢀ1 due to
asymmetric stretching vibrations of Si-O-Si bridges and at
798 cmꢀ1 due to symmetric stretching vibration of Si-O-Si.[35]
The peak at 972 cmꢀ1 (Si-TUD-1) is assigned to stretching
vibrations of terminal silanol (Si-OH) groups present at
defect sites.[62] In the Zr-TUD-1 samples, this peak is, to a
964
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972

Zirconium-Containing Catalysts
FULL PAPER
certain extent, due to Si-O-Zr stretching vibrations.[37,41,56]
The presence of both of these vibrations leads to a relatively
broad and less-resolved peak at 975 cmꢀ1 in the case of the
Zr-TUD-1 samples.
A temperature-programmed desorption (TPD) profile of
ammonia for various Zr-TUD-1 samples showed a broad
peak over the temperature range 120–5008C (Figure 5). De-
Figure 6. FTIR spectra in the OH group region of Si-TUD-1, Zr-TUD-1
(100), Zr-TUD-1(50), Zr-TUD-1(25) and Zr-TUD-1(10).
(3780 cmꢀ1) as described by JimØnez-López et al. and
Rakshe et al.[56,58] was detected for Zr-TUD-1 with Si/Zr=
25 or higher. This is good evidence that all the zirconium is
tetrahedrally incorporated into the framework. Zr-TUD-
1(10) does have a week peak in this area; most likely the
Figure 5. Temperature-programmed desorption (TPD) profile of ammo-
nia for various Zr-TUD-1 samples.
ꢀ
convolution of this broad peak clearly shows three peaks
corresponding to weak, medium and strong acid strength
(Table 3). The total acidity of the samples follows the load-
Zr OH groups are due to the small amount of zirconia
present in this material. All Zr-TUD-1 samples show a weak
ꢀ1
ꢀ
and broad shoulder at around 3650 cm , which is due to Si
OH groups at defect sites.[33] This peak is strongest in the
Zr-TUD-1(10) sample, indicating that this material is less
structured.
Table 3. Acidity derived from temperature-programmed desorption
(TPD) of ammonia for Zr-TUD-1 samples with different Si/Zr ratios.
FTIR studies of adsorbed pyridine were carried out to
access the type and strength of the acid sites. FTIR spectra
of Zr-TUD-1 samples (Figure 7) after pyridine desorption at
Zr-TUD-
1 (Si/Zr)
T [8C] (centre of
mmol NH3/g
Total
acidity
deconvoluted peak)
weak medium strong weak[a] medium[a] strong[a]
10
25
50
205
198
183
192
258
258
224
233
334
343
287
296
0.22
0.15
0.03
0.03
0.45
0.47
0.10
0.06
0.29
0.22
0.30
0.17
0.95
0.84
0.43
0.25
100
[a] Calculated from area of deconvoluted peak.
ing of zirconium: 0.954, 0.836, 0.433 and 0.247 mmolgꢀ1 of
ammonia for Zr-TUD-1(10), Zr-TUD-1(25), Zr-TUD-1(50)
and Zr-TUD-1(100), respectively. The small difference be-
A
tween Zr-TUD-1(10) and Zr-TUD-1(25) is most likely due
to the reduced surface area and the nanocrystals in Zr-
TUD-1(10) relative to the completely framework-incorpo-
rated zirconium in mesoporous Zr-TUD-1(25).
FTIR spectra of both Si-TUD-1 and Zr-TUD-1 samples
in the hydroxyl region (Figure 6) showed a band centred at
3745 cmꢀ1, which is attributed to terminal silanol groups.
The intensity of this band decreases with the zirconium
loading, indicating zirconium incorporation.[33] In addition,
Figure 7. FTIR spectra of Si- and Zr-TUD-1 samples after pyridine de-
sorption at 2008C: a) Si-TUD-1, b) Zr-TUD-1(100), c) Zr-TUD-1(50),
A
ꢀ
no band corresponding to terminal Zr OH groups
d) Zr-TUD-1(25) and e) Zr-TUD-1(10).
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
965

U. Hanefeld et al.
2008C exhibits bands at ꢂ1610, 1547, 1447 and 1490 cmꢀ1
corresponding to acid sites.[9,58,59] Pyridine coordinated to
Lewis acid sites gives rise to two signals (ꢂ1610 and
1447 cmꢀ1). A combination signal is associated with both
Brønsted and Lewis acid sites (pyridinium ion and coordi-
nated pyridine, respectively) at 1490 cmꢀ1 and a signal indi-
cative of only Brønsted acid sites (pyridinium ion at
1547 cmꢀ1). The bands at ꢂ1610 and 1447 cmꢀ1 are present
in higher intensity than in the reference material (Si-TUD-
1) and increase with zirconium loading, proving the Lewis
acidity of the materials. Moreover, the band at 1547 cmꢀ1 is
absent at low zirconium loading Si/Zr=100) and very weak
even at Si/Zr=25. Only in Zr-TUD-1(10) does this peak
become more prominent. This might be due to Brønsted
ed to 260 and 340 nm. The band near 210 nm was assigned
to the presence of isolated zirconium in a tetrahedral envi-
ronment.[35,41] Again, similar to the XRD spectra, the zirco-
nia nanoparticles visible in Zr-TUD-1(10) by HR-TEM are
not evident in the UV/Vis spectrum. This indicates that the
concentration of zirconia nanoparticles must be very low.
Zr-TUD-1 as a Lewis acidic catalyst: Having established the
tetrahedral nature of the isolated zirconium atoms in Zr-
TUD-1, its catalytic activity was investigated in two acid-cat-
alysed model reactions, the MPV reduction of 4-tert-butylcy-
clohexanone (3) (Scheme 3) and the carbon–carbon bond
forming Prins reaction (Scheme 2).
ꢀ
acidic sites of the zirconia particles in this sample or the Si
OH groups at defect sites. In other words, the increase in
zirconium content up to Si/Zr=25 induces Lewis acidity in
the samples. Only traces of Brønsted acid sites were ob-
served. This means that, as planned, zirconium induces the
strong acid sites observed by NH3–TPD and that these are
due to Lewis acid sites. This is in line with the high Lewis
acidity of Al-free Zr-beta and other zirconium-containing
microporous silicates.[9,58,59]
Scheme 3. The Zr-TUD-1 catalysed Meerwein–Ponndorf–Verley (MPV)
reduction of 4-tert-butylcyclohexanone (3) proceeds diastereoselectively.
Diffuse reflectance UV/Vis spectra of the Zr-TUD-1 sam-
ples showed, in each case, a weak and broad peak between
240 and 280 nm with a tail at around 350 nm (Figure 8). For
When the heterogeneous MPV reduction of 4-tert-butylcy-
clohexanone (3) is performed with zeolites such as H-
beta,[2,63] Sn-beta[3,8] or Zr-beta,[4,5,8,59] a strong selectivity for
the cis alcohol (cis-4) is observed. This high selectivity for
the thermodynamically less favoured product is clearly relat-
ed to the spatial confinement within the pores of the zeo-
lite.[8] In mesoporous materials such as Zr-MCM-41, Zr-
MCM-48 and Zr-SBA-15,[33] no spatial confinement is pres-
ent and the thermodynamically determined dominance of
the trans product (trans-4) is observed. Initial studies with
Zr-TUD-1 and also Al-TUD-1 confirmed this behav-
iour.[51,64] However, Zr-TUD-1(25) was not a very active cat-
alyst under the originally described reaction conditions
(2 mmol substrate in 4 mL isopropanol). Indeed, its activity
was much lower than for the earlier reported Zr-MCM-41,
Zr-MCM-48 and Zr-SBA-15,[33] although comparable reac-
tion conditions were used. In an earlier study, the impor-
tance of the solvent in the homogeneous MPV reduction
had been demonstrated.[65] Therefore, the Zr-TUD-1(25)
catalysed reduction of 3 was studied in coordinating and
non-coordinating solvents (Figure 9). Alcohols coordinate
well to the zirconium and thus block the active site, conse-
quently tert-butyl alcohol inhibited the reaction (Table 4,
entry 5). When the less coordinating ether 1,4-dioxane was
Figure 8. Diffuse reflectance UV/Vis spectra of various Zr-TUD-1 sam-
ples (left) compared with commercial ZrO2 (right).
these samples, the peaks are assigned to O2ꢀ!Zr4+ charge-
transfer interactions with Zr in low coordination either iso-
lated or present in small ZrxOy clusters in the silica network
of TUD-1.[60] This is further supported by the fact that these
spectra of Zr-TUD-1(25), Zr-TUD-1(50) and Zr-TUD-1
Table 4. Effect of the solvent on the MPV reduction of 3.
Solvent
3/isopro-
Conver-
Select-
ivity [%]
trans-4/
cis-4
panol[a]
sion [%][b]
1
2
3
4
5
isopropanol
toluene
toluene
1,4-dioxane
tert-butyl alcohol
1:26
1:1
1:3
1:1
1:1
83.0
79.5
97.5
37.3
4.54
>99
>99
>99
>99
>99
87:13
76:24
83:17
82:18
90:10
A
ZrO2. Due to the low zirconium concentration, the expected
signal at approximately 210 nm is not visible. The sample of
Zr-TUD-1(10) is different, since the spectrum displays a
sharp band at 216 nm and the broad bands are slightly shift-
[a] Molar ratio. [b] After 3 days reaction at 808C.
966
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972

Zirconium-Containing Catalysts
FULL PAPER
Figure 10. a) Conversion of 3 with Zr-TUD-1(10), Zr-TUD-1(25) and
Zr-TUD-1
A
1
ACHTREUNG
ene (3.55 mL) with three equivalents of isopropanol at 808C.
The acid-catalysed Prins reaction of citronellal (1) leads
to cyclisation, yielding isopulegol (2) and its stereoisomers
5–7. Due to its all-equatorial orientation of the substituents,
2 is the thermodynamically favoured product (Scheme 4).
Figure 9. The solvent has a significant influence on the catalystsꢁ activity
in the MPV reduction of 3 (3.85 mL solvent, 50 mg Zr-TUD-1(25),
808C).
employed, the reaction pro-
ceeded faster than in tert-butyl
alcohol (entry 4 versus 5). Tolu-
ene, however, did not inhibit
the reaction. When three equiv-
alents of isopropanol in toluene
were used, the reactions pro-
Scheme 4. Through the Prins reaction, citronellal (1) is cyclised to isopulegol (2) and its diastereoisomers 5, 6
and 7.
ceeded smoothly (entries 2 and
3 versus 5). Isopropanol as a
neat solvent inhibited the reac-
tion due to coordination, too (entry 1) and, therefore, tolu-
ene was applied as the solvent. There seems to be little in-
fluence of the solvent on the trans:cis ratio of 4, which is
similar to results reported earlier for Zr-MCM-41, Zr-MCM-
48 and Zr-SBA-15.
When Zr-TUD-1 with different Si/Zr ratios was employed
for the MPV reduction, Zr-TUD-1(25) displayed the highest
activity per milligram of catalyst (Figure 10A). Evidently,
the larger pore volume and diameter (Table 1) gives it an
advantage over Zr-TUD-1(10), which had a higher loading
of Lewis acid sites. However, only the turn over frequency
(TOF) reveals the true activity of the catalysts. The highest
Similarly, based on the preferred all-equatorial transition
state (Scheme 2B), it is also the kinetically favoured prod-
uct.[20] In the first cyclisation studies performed with zeo-
lites,[19,20] diastereoselectivities were low. Therefore, it was
first proposed that the cyclisation takes place on the surface
of the catalysts;[19] later calculations demonstrated that it
can take place inside the pores of the zeolites.[20] Surprising-
ly, diastereoselectivities did not differ among the zeolites H-
beta, H-MORD, H-ZSM-5 and H-Y[19,20,24,39] and mesopo-
rous materials, although spatial confinement by the micro-
pores of the zeolites might be expected to improve the ste-
reoselectivity. With the introduction of tin and, in particular,
zirconium into zeolite beta, the diastereoselectivity could be
improved significantly, with Zr-beta being the catalyst of
choice.[21,25,66] This superior selectivity was ascribed to the
larger metal ion.[25] However, this improved selectivity is
coupled to a reduced reactivity, possibly due to diffusion
limitations.
activity per catalytic site was observed for Zr-TUD-1(100),
A
clearly demonstrating that, as postulated above, no diffusion
limitations occur in TUD-1 (Figure 10B). Again no influ-
ence on the trans:cis ratio of 4 (ꢂ83:17 for all the catalysts)
was observed. Even though Zr-TUD-1 clearly displayed a
much higher activity under these reaction conditions and ac-
cepts sterically very demanding substrates,[51] the overall ac-
tivity was lower than that of comparable mesoporous zirco-
nium-containing silicates.[33]
Industrial grade rac-1 (containing 6% rac-2) was submit-
ted to a Zr-TUD-1(25)-catalysed Prins reaction at different
temperatures in different solvents. As in the above-de-
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
967

U. Hanefeld et al.
scribed MPV reduction of 3, coordinating solvents inhibited
the reaction. In 1,4-dioxane, the reaction proceeded slug-
gishly but conversions >99% with selectivities >98% for
the cyclisation products 2, 5, 6 and 7 were achieved at 808C
within two hours (Figure 11A). In toluene, the reaction pro-
Figure 12. a) Cyclisation of industrial grade rac-1 (2 mmol) catalysed by
50 mg Zr-TUD-1(10), Zr-TUD-1(25) or Zr-TUD-1
ACHTREUNG
TUD-1(10), Zr-TUD-1(25) and Zr-TUD-1
ACHTREUNG
dustrial grade rac-1 (TOF=mol 1 reactedmol Zrꢀ1 minꢀ1
(4 mL) at 808C.
) in toluene
Figure 11. Cyclisation of industrial grade rac-1 (2 mmol) catalysed by of
Zr-TUD-1 (25) (50 mg) in a) 1,4-dioxane (4 mL) or b) toluene (4 mL) at
different temperatures.
Table 5. Cyclisation of industrial grade rac-1 (2 mmol) catalysed by
50 mg Zr-TUD-1(10), Zr-TUD-1(25) or Zr-TUD-1(100) in toluene
A
(4 mL) at 808C.
Catalyst
t [min]
Conver-
sion [%]
Select-
ivity [%]
Molar ratio of
isomers 2/5/6/7
ceeded significantly faster, and even at 408C conversions
>95% were realised within three hours with selectivities
>98% for the cyclisation products 2, 5, 6 and 7 (Figure 11
Zr-TUD-1 (10)
Zr-TUD-1 (25)
Zr-TUD-1 (100)
Zr-TUD-1 (10)
Zr-TUD-1 (25)
Zr-TUD-1 (100)
30
30
30
60
60
60
96.6
99.5
84.7
99.4
99.5
95.5
99.6
99.5
91.2
99.5
99.5
99.6
64.9:26.7:5.8:2.6
63.7:28.2:5.2:2.8
63.1:30.0:4.9:2.3
69.4:22.9:5.4:2.3
64.1:27.3:5.8:2.8
61.7:29.2:6.0:3.0
B). Diastereoselectivities of 2/(5+6+7) were always around
G
65:35. When compared to Zr-beta, a significant improve-
ment in rate is observed, as might be expected in a virtually
diffusion-limitation free material as TUD-1. Due to the lack
of spatial confinement, the diastereoselectivity in Zr-TUD-1
is, however, lower than in Zr-beta. It might also be noted
that, in contrast to our results, a coordinating solvent,
namely tert-butyl alcohol, was the solvent of choice for Zr-
beta.[25,66]
When Zr-TUD-1 with different Si/Zr ratios were com-
pared for the cyclisation of rac-1 in toluene, Zr-TUD-1(25)
had the highest activity per milligram of catalyst (Figure 12
A). Similar to the MPV reduction, the smaller pore diame-
ter and pore volume (Table 1) of Zr-TUD-1(10) reduced
the activity of the Lewis acid sites significantly and the per-
rac-2 was established in a recycling experiment. Cycle times
of 30 min were chosen in order to ensure that a possible loss
of activity could be detected.[67] After five cycles, only a
small loss of activity of the Zr-TUD-1(25) was detectable
(Figure 13). Given the crude nature of the starting material,
the catalyst was calcined to remove any inhibitors contained
in the feed. Full activity of the Zr-TUD-1(25) was regained.
In line with this, no leaching of zirconium could be detected
(ICP-OES, detection limit 0.002 ppm). To confirm that the
reaction was truly heterogeneous, hot filtration studies were
performed. Upon removal of the catalyst at 808C after
15 min of reaction (94% conversion), no further reaction
was observed in the filtrate, confirming the heterogeneous
nature of the catalysis.
The solvent studies for the MPV reduction had shown
that alcohols could act as an inhibitor of Zr-TUD-1. As 2
and its diastereoisomers are alcohols, the cyclisation of 1
might be product inhibited. Since the crude rac-1 contains
rac-2, this might have been the reason for the observed loss
of activity in the recycling experiments (Figure 13). There-
fore, the Prins reaction of optically pure (+)-citronellal was
compared with that of optically pure (+)-citronellal with
10% optically pure (ꢀ)-isopulegol (Figure 14, Table 6). The
formance of Zr-TUD-1(100) was even poorer. By determin-
U
ing the TOF, the real activity per active site was revealed
(Figure 12B). Zr-TUD-1 (100) showed the highest TOF fol-
lowed by Zr-TUD-1(25). Clearly, this activity per active site
of Zr-TUD-1 (100) in this very fast reaction was only possi-
ble due to the three-dimensional mesoporous character of
the material that prevents diffusion limitations. The selectiv-
ity for the cyclisation of all three catalysts was (with >99%)
excellent (Table 5). The diastereoselectivity towards 2 was
always around 65:35; the reduced pore diameter of Zr-
TUD-1 (10) did not have an influence on this ratio.
The stability of Zr-TUD-1 as a catalyst for the cyclisation
of industrial grade rac-1 containing approximately 6% of
968
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972

Zirconium-Containing Catalysts
FULL PAPER
Conclusion
The straightforward synthesis of a zirconium incorporated
three-dimensional mesoporous silicate, Zr-TUD-1, with dif-
ferent Si/Zr ratios was achieved by means of a direct hydro-
thermal method with triethanolamine as the template. This
is significantly more efficient than the grafting approach de-
scribed for Zr-MCM-41, Zr-MCM-48 and Zr-SBA-15,[33] and
of similar ease to the preparation of one-dimensional Zr-
MPS carriers that were used as support for Fischer–Tropsch
catalysts.[42] Zr-TUD-1 was shown to possess a high surface
area, and no detectable zirconia phase was observed both in
XRD and TEM for Si/Zr ratios of 25 and higher. Further in-
crease in Zr content led to the formation of some ZrO2
nanoparticles and some loss of mesoporosity evidenced by
TEM and N2 sorption studies. The framework incorporation
of Zr at all concentrations was evident from FTIR, XPS and
UV/Vis studies. Thus, a highly predictable and short synthe-
sis of a novel mesoporous silicate with Zr incorporated into
its framework was achieved. The properties of these materi-
als such as high surface area, pore sizes similar to Si-TUD-1
and high Lewis acidity mean that Zr-TUD-1 can be utilized
as a purely Lewis acidic catalyst for a variety of reactions.
Zr-TUD-1 was tested in the MPV reduction of 4-tert-bu-
tylcyclohexanone (3) and in the carbon–carbon bond form-
ing Prins reaction. In the MPV reduction and in the Prins
reaction, a clear solvent dependence of the catalytic activity
was observed, with non-coordinating toluene being the sol-
vent of choice. While the selectivity for the MPV reduction
of 3 was excellent, the trans diastereoselectivity in this re-
duction was, as expected for a mesoporous material, low. In
the Prins cyclisation of citronellal (1) to isopulegol (2), Zr-
TUD-1 revealed its full catalytic potential. Both conversion
of 1 and selectivity for the Prins cyclisation were >99%.
Due to the three-dimensional sponge-like pore structure of
Zr-TUD-1, no diffusion limitations were observed. Accord-
ingly, Zr-TUD-1 has a very high activity, which is particular-
ly evident when comparing the TOFs of the different Zr-
Figure 13. Recycling experiments with Zr-TUD-1(25) (50 mg) in the cyc-
lisation of industrial grade rac-1 (2 mmol) in toluene (4 mL) at 808C,
30 min per cycle.
Figure 14. Cyclisation of a) (+)-1 (2 mmol) and b) (+)-1 (2 mmol) con-
taining 0.2 mmol (ꢀ)-2 catalysed by Zr-TUD-1(25) (50 mg) in toluene
(4 mL) at 808C.
TUD-1s; Zr-TUD-1(100) had the most active catalytic sites
A
(Figure 12B). While the diastereoselectivity of Zr-TUD-1
still needs to be improved, recycling experiments with indus-
trial grade citronellal (1) demonstrated the great stability of
Zr-TUD-1; any reversible loss of activity was due product
inhibition. This stability was underlined by the fact that no
leaching of zirconium could be detected. As a result, the
new, readily prepared, three-dimensional, Lewis acidic Zr-
TUD-1 is a promising catalyst for the Prins reaction.
Table 6. Initial rates for the cyclisation of (+)-1 (2 mmol) and (+)-1
(2 mmol) containing (ꢀ)-2 (0.2 mmol) when catalysed by Zr-TUD-1(25)
(50 mg) in toluene (4 mL) at 808C.
Initial rate [mmolmLꢀ1 minꢀ1
]
(+)-citronellal
(+)-citronellal + (ꢀ)-isopulegol
0.075
0.061
initial rates in the Zr-TUD-1 (25) catalysed reaction were
different, the reaction containing product being slower.
Compound 2 inhibited its formation from 1. This product in-
hibition contributed to the reversible deactivation of Zr-
TUD-1 as was observed in the recycling experiments.
Experimental Section
Synthesis of Zr-TUD-1: Zr-TUD-1 was synthesized with triethanolamine
(TEA) as chelating agent in a one-pot procedure based on the sol–gel
technique. Zr-TUD-1 with different Si/Zr atomic ratios was synthesized
by adding appropriate amounts of tetraethyl orthosilicate (TEOS, 98%,
Aldrich) to zirconium(IV) propoxide (Aldrich, 70 wt.% in 1-propanol)
and of 2-propanol (30 mL). After stirring for a few minutes, appropriate
amounts of a mixture of TEA (97%, ACROS) and water was added fol-
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
969

U. Hanefeld et al.
lowed by addition of tetraethyl ammonium hydroxide (TEAOH,
35 wt%, Aldrich) as a mineraliser under vigorous stirring. This yielded a
was recorded with a pass energy of 89.45 eV and a step size of 0.5 eV.
The spectra were evaluated with Multipak 6.1 A software (Physical Elec-
tronics).
gel with a molar ratio of SiO2/xZrO2/
G
U
ACHTUNG-
TREUNG
Catalytic experiments: All chemicals were purchased from Aldrich, Jans-
sen or Acros. For the catalysis experiments, the anhydrous solvents and
solids were used as received, all other liquids were dried and distilled
prior to use. Zr-TUD-1 catalysts were activated in the presence of air at
up to 6008C at a temperature ramp of 1 Kminꢀ1 and subsequent heating
at 6008C for 10 h. The experiments were performed in dried glassware
under a nitrogen atmosphere.
room temperature for 12–24 h, dried at 988C for 12–24 h, followed by hy-
drothermal treatment in a Teflon lined autoclave at 1808C for 4–24 h and
finally calcined in the presence of air up to 6008C at a temperature ramp
of 1 Kminꢀ1 and subsequent heating at 6008C for 10 h. The Zr-TUD-1
obtained could be used for catalysis immediately or was, upon standing
for a prolonged period of time, reactivated by repeating the calcination
procedure. Samples with a Si to Zr ratio of 100, 50, 25 and 10 (denoted
For the Meerwein–Ponndorf–Verley (MVP) reductions of 4-tert-butylcy-
clohexanone (3; 2 mmol), solvent and isopropanol (4 mL; e.g., 4 mL iso-
propanol or 3.85 mL solvent and 0.15 mL isopropanol=2 mmol or
3.55 mL toluene and 0.45 mL isopropanol=6 mmol) and 1,3,5-triisopro-
pylbenzene (internal standard, 0.1 mL) were loaded into a Schlenk flask
containing activated Zr-TUD-1 catalyst (50 mg). The reaction mixture
was immediately immersed into an oil bath at 808C and the reaction was
followed by taking samples of 20 mL. After the solvent screening, all
MPV reductions of 3 were performed with toluene (3.55 mL) and isopro-
panol (0.45 mL), that is, with a ratio of 3 to isopropanol of 1:3, at 808C.
as Zr-TUD-1(100), Zr-TUD-1(50), Zr-TUD-1(25) and Zr-TUD-1(10),
respectively) were prepared.
G
Material characterisation: Chemical analyses of Si and Zr were per-
formed by dissolving the samples in an aqueous solution containing 1%
HF and 1.3% H2SO4 and measuring them by inductively coupled plasma
optical emission spectroscopy (ICP-OES) on a Perkin Elmer Optima
3000DV instrument. Powder XRD patterns were obtained on a Philips
PW 1840 diffractometer equipped with a graphite monochromator using
CuKa radiation. The textural parameters were evaluated from volumetric
nitrogen physisorption at 77 K. Prior to the physisorption experiment,
the samples were dried in vacuo at 3008C, and the nitrogen adsorption
and desorption isotherm was measured on a Quantachrome Autosorb-6B
at 77 K. Transmission electron microscopy (TEM) was performed by
using a Philips CM30T electron microscope with a LaB6 filament as the
source of electrons operated at 300 kV. UV/Vis spectra were collected at
After one catalytic experiment, the catalyst was filtered off and the sol-
vent evaporated. The trans:cis ratios of the 4-tert-butylcyclohexanols (4)
obtained from this Meerwein—Ponndorf–Verley reduction of 3 was then
1
1
established by H NMR spectroscopy. The H NMR spectrum of the mix-
ture was recorded on a Varian-INOVA 300 MHz spectrometer in CDCl3
with TMS as an internal standard and was in agreement with the litera-
ture spectra of trans-4 and cis-4.[68] The ratio between cis-4-tert-butylcy-
clohexanol (cis-4) and trans-4-tert-butylcyclohexanol (trans-4) was deter-
mined from the well-separated signals of the H in the 1-position. For cis-
4 this H is equatorial and has a shift of d=4.032 ppm, for trans-4 this H
is axial with a shift of d=3.510 ppm. The trans:cis ratio for 4 was 83:17.
Based on this analysis, the retention times for the diastereoisomers of 4
in the gas chromatography (GC) could be unambiguously assigned. Reac-
tions were followed by GC with a Shimadzu GC-17 A gas chromato-
graph, equipped with a 25 m”0.32 mm chiral column Chrompack Chira-
sil-Dex CB, split injector (1/97) at 2208C, a Flame Ionisation Detector at
2208C and He as carrier gas. The correct retention times observed
(1208C isotherm) are: 1,3,5-triisopropylbenzene (internal standard):
3.9 min; 4-tert-butylcyclohexanone (3): 4.7 min; cis-4-tert-butylcyclohexa-
nol (cis-4): 5.6 min; trans-4-tert-butylcyclohexanol (trans-4): 5.9 min. The
retention times previously reported in reference[51] accidentally give the
reverse order for the diastereoisomers cis-4 and trans-4. Under these con-
ditions and with the correct assignment, the trans:cis ratio for 4 deter-
mined by 1H NMR spectroscopy was confirmed. When the MPV reduc-
tion of 3 was performed with microporous H-beta as catalyst, the major
product was cis-4[2] confirming the assignment of the retention times.
room temperature on
BaSO4 as reference.
a Shimadzu UV-2450 spectrophotometer with
Temperature-programmed desorption of ammonia measurements were
carried out on a Micromeritics TPR/TPD 2900 equipped with a Thermal
Conductivity Detector (TCD). The sample (30 mg) was pretreated at
5508C in a flow of He (30 mLminꢀ1) for 1 h. Afterwards, pure NH3
(40 mLminꢀ1) was adsorbed at 1208C for 15 min. Subsequently, a flow of
He (30 mLminꢀ1) was passed through the reactor for 30 minutes to
remove any weakly adsorbed NH3 from the sample. This procedure was
repeated three times. Desorption of NH3 was monitored in the range of
120 to 5508C at a ramp of 10 Kminꢀ1
.
FTIR spectra of self-supported wafers and KBr-diluted wafers of Zr-
TUD-1 samples were recorded on a Nicolet AVATAR 360 FTIR instru-
ment. Skeletal FTIR spectra were carried out with KBr-diluted wafers in
the region 1500–650 cmꢀ1. Acid-strength distribution was evaluated by IR
spectra of adsorbed pyridine. It was carried out on self-supported wafers
of Zr-TUD-1 samples (15–25 mgcmꢀ2
) after evacuation (5008C, 2 h,
10 Pa) in a custom made vacuum cell with CaF2 windows. Then the cell
was cooled down to room temperature, and the IR spectra were collect-
ed. As water and any adsorbed materials were removed by this treat-
ment, hydroxyl spectra were obtained from this sample. Then, the tem-
perature of the cell was raised to 1008C, and the sample was then
brought into contact with pyridine vapour (20 Torr) for 30 min. The tem-
perature of the cell was raised to 1508C, with vacuum applied for 30 min
to remove physisorbed and loosely bound pyridine. Subsequently, the
temperature of the cell was raised to 2008C under vacuum for another
30 min. This procedure ensures the removal of all physisorbed pyridine
(with only chemisorbed pyridine available for IR analysis). The cell was
then cooled to room temperature, and the differential IR spectra were
collected (Figure 7a). All of the spectra were recorded at room tempera-
ture with a resolution of 2 cmꢀ1 averaged over 500 scans.
The Prins reaction experiments were carried out at 80, 60, 40 and 258C
and using 50 mg of the catalyst. The reaction mixture consisted of
2 mmol of (ꢁ)-citronellal (rac-1) or enantiopure (+)-citronellal (+)-1,
toluene or 1,4-dioxane (4 mL) as solvent and 1,3,5-triisopropylbenzene
(internal standard, 0.1 mL). After the solvent and temperature screening,
the ideal reaction conditions, 808C and toluene, were used. The reaction
was followed by taking samples of 20 mL for GC analysis.
After one catalytic experiment, the catalyst was filtered off and the sol-
1
vent evaporated. The H NMR spectrum of the mixture was recorded on
a Varian-INOVA 300 MHz spectrometer in [D8]toluene with TMS as an
internal standard. The spectrum of the main component (approx. 65% as
1
The XPS measurements were performed with a PHI 5400 ESCA provid-
ed with a dual Mg/Al anode X-ray source, a hemispherical capacitor ana-
lyser and a 5 keV ion gun. The input lens optical axis to the analyser was
at a take off angle of 158 with respect to the sample surface normal. The
input lens aperture used was 3.5”1.0 mm. All spectra were recorded with
unmonochromated magnesium radiation. The X-ray source was operated
at an acceleration voltage of 15 kV and a power of 400 W. A survey spec-
trum was recorded between 0 and 1000 eV binding energy with a pass
energy of 71.95 eV and a step size of 0.25 eV. The spectra of the separate
photoelectron and Si-Auger electron lines were recorded with a pass
energy of 35.75 eV and a step size of 0.2 eV. The Zr-Auger electron line
determined by H NMR spectroscopy, by using the signal of the H in the
1-position of 2 at 3.270 ppm) was identical to that of a commercial
sample of 2, while the other components were isomers 5, 6 and 7. Due to
overlying signals, their separate concentrations could not be accurately
determined. All other reactions were followed by GC. The samples ob-
tained were analyzed with a VARIAN 1177 gas chromatograph equipped
with an injector at 2508C, a 30 m”0.25 mm VARIAN FactorFour column
and using a FID detector at 2508C. The retention times observed (15 min
1008C isotherm, then with a ramp of 100 Kminꢀ1 until 1758C, then iso-
therm at 1758C for 2 min) are: 1.4 min toluene, 4.8 min citronellal,
5.6 min neo-isopulegol, 6.1 min isopulegol, 6.4 min iso-isopulegol, 6.9 min
970
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972

Zirconium-Containing Catalysts
FULL PAPER
neoiso-isopulegol and 13.9 min 1,3,5-triisopropylbenzene. The products
were identified on the basis of their retention times by comparison with
commercial 1 and 2. The other three isomers were assigned according to
reference [20].
[27] P. Mertens, F. Verpoort, A.-N. Parvulescu, D. De Vos, J. Catal. 2006,
243, 7–13.
[28] C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli, J. S. Beck,
Nature 1992, 359, 710–712.
[29] D. Trong On, D. Desplantier-Giscard, C. Danumah, S. Kaliaguine,
Appl. Catal. A 2001, 222, 299–357.
[30] R. Ryoo, S. Jun, J. M. Kim, M. J. Kim, Chem. Commun. 1997, 2225–
2226.
[31] R. Mokaya, W. Jones, Phys. Chem. Chem. Phys. 1999, 1, 207–213.
[32] B. Bonelli, B. Onida, J. D. Chen, A. Galarneau, F. Di Renzo, F.
Fajula, E. Garrone, Microporous Mesoporous Mater. 2004, 67, 95–
106.
The recycling experiments of Zr-TUD-1(25) (50 mg) for the conversion
of (ꢁ)-citronellal rac-1 (2 mmol) in toluene (4 mL) as solvent and 1,3,5-
triisopropylbenzene (internal standard, 0.1 mL) were performed at 808C
for 30 min, then the catalyst was separated from the reaction mixture,
dried and submitted to a new reaction. After the fifth cycle the Zr-TUD-
1(25) was calcined at 6008C before recycling. Results are given in
Figure 13.
[33] Y. Z. Zhu, S. Jaenicke, G.-K. Chuah, J. Catal. 2003, 218, 396–404.
[34] S. Gontier, A. Tuel, Appl. Catal. A 1996, 143, 125–135.
[35] M. S. Morey, G. D. Stucky, S. Schwarz, M. Frçba, J. Phys. Chem. A
1999, 103, 2037–2041.
Acknowledgements
[36] D. J. Jones, J. Jimenez-Jimenez, A. Jimenez-Lopez, P. Maireles-
´
Torres, P. Olivera-Pastor, E. Rodryguez-Castellon, J. Roziere, Chem.
The authors would like to thank Dr. Ugo Lafont of DCT/NCHREM,
Technische Universiteit Delft (The Netherlands), for performing the elec-
tron microscopy investigations. Financial support from NWO(Mozaïek
fellowship) for S.T. is thankfully acknowledged. Stimulating discussions
with Dr. J. A. Peters (TU Delft) are gratefully acknowledged. All this
work would not have been possible without Dr. Koos Jansen, the inven-
tor of TUD-1. We thank him for many fruitful discussions and advice.
Commun. 1997, 431–432.
[37] X. X. Wang, F. Lefebvre, J. Patarin, J. M. Basset, Microporous Meso-
porous Mater. 2001, 42, 269–276.
[38] A. Tuel, Microporous Mesoporous Mater. 1999, 27, 151–169.
[39] G. K. Chuah, S. H. Liu, S. Jaenicke, L. J. Harrison, J. Catal. 2001,
200, 352–359.
[40] S. H. Liu, S. Jaenicke, G. K. Chuah, J. Catal. 2002, 206, 321–330.
[41] J. El Haskouri, S. Cabrera, C. Guillem, J. Latorre, A. Beltran, D.
Beltran, M. Dolores Marcos, P. Amoros, Chem. Mater. 2002, 14,
5015–5022.
[42] M. Wie, K. Okabe, H. Arakawa, Y. Teraoka, Catal. Commun. 2004,
5, 597–603.
[43] J. C. Jansen, Z. Shan, L. Marchese, W. Zhou, N. van der Puil, T.
Maschmeyer, Chem. Commun. 2001, 713–714.
[44] Z. Shan, E. Gianotti, J. C. Jansen, J. A. Peters, L. Marchese, T.
Maschmeyer, Chem. Eur. J. 2001, 7, 1437–1443.
[45] R. Anand, M. S. Hamdy, U. Hanefeld, T. Maschmeyer, Catal. Lett.
2004, 95, 113–117.
[46] M. S. Hamdy, A. Ramanathan, T. Maschmeyer, U. Hanefeld, J. C.
Jansen, Chem. Eur. J. 2006, 12, 1782–1789.
[47] R. Anand, R. Maheswari, U. Hanefeld, J. Catal. 2006, 242, 82–91.
[48] M. S. Hamdy, G. Mul, W. Wei, R. Anand, U. Hanefeld, J. C. Jansen,
J. A. Moulijn, Catal. Today 2005, 110, 264–271.
[49] M. S. Hamdy, G. Mul, J. C. Jansen, A. Ebaid, Z. Shan, A. R. Over-
weg, T. Maschmeyer, Catal. Today 2005, 100, 255–260.
[50] R. Anand, M. S. Hamdy, P. Gkourgkoulas, T. Maschmeyer, J. C.
Jansen, U. Hanefeld, Catal. Today 2006, 117, 279–283.
[51] A. Ramanathan, D. Klomp, J. A. Peters, U. Hanefeld, J. Mol. Catal.
A 2006, 260, 62–69.
[52] M. K. Sharma, A. Singh, R. C. Mehrotra, Ind. J. Chem. Sect. A 2001,
40, 568–572.
[53] S. Cabrera, J. El Haskouri, C. Guillem, J. Latorre, A. Beltran-Porter,
D. Beltran-Porter, M. Dolores Marcos, P, Amoros, Solid State Sci.
2000, 2, 405–420.
[54] D. Ortiz de Zµrate, A. Gómez-Moratalla, C. Guillem, A. Beltrµn, J.
Latorre, D. Beltrµn, P. Amorós, Eur. J. Inorg. Chem. 2006, 2572–
2581.
[55] K. P. de Jong, A. J. Koster, A. H. Janssen, U. Ziese, Stud. Surf. Sci.
Catal. 2005, 157, 225–242.
[1] ”Transfer Hydrogenation Including the Meerwein–Ponndorf–Verley
Reduction”: D. Klomp, J. A. Peters, U. Hanefeld, in Handbook of
Homogeneous Hydrogenation, Vol.
3 (Eds.: J. G. de Vries, C. J.
Elsevier) Wiley-VCH, Weinheim, 2007, Chapter 20, pp. 585–630.
[2] J. C. van der Waal, E. J. Creyghton, P. J. Kunkeler, K. Tan, H. van
Bekkum, Top. Catal. 1997, 4, 261–268.
[3] A. Corma, M. E. Domine, S. Valencia, J. Catal. 2003, 215, 294–304.
[4] Y. Z. Zhu, G. K. Chuah, S. Jaenicke, J. Catal. 2006, 241, 25–33.
[5] G. K. Chuah, S. Jaenicke, Y. Z. Zhu, S. H. Liu, Curr. Org. Chem.
2006, 10, 1639–1654.
[6] D. Klomp, T. Maschmeyer, U. Hanefeld, J. A. Peters, Chem. Eur. J.
2004, 10, 2088–2093.
[7] R. Cohen, C. R. Graves, S.-B. T. Nguyen, J. M. L. Martin, M. A.
Ratner, J. Am. Chem. Soc. 2004, 126, 14796–14803.
[8] M. Boronat, A. Corma, M. Renz, J. Phys. Chem. B 2006, 110,
21168–21174.
[9] Y. Z. Zhu, K. L. Fow, G. K. Chuah, S. Jaenicke, Chem. Eur. J. 2007,
13, 541–547.
[10] H. J. Prins, Chem. Weekbl. 1917, 14, 932–939.
[11] H. J. Prins, Chem. Weekbl. 1919, 16, 1072–1074.
[12] H. J. Prins, Chem. Weekbl. 1917, 14, 627–630.
[13] R. Noyori, Adv. Synth. Catal. 2003, 345, 15–32.
[14] Y. Nakatani, K. Kawashima, Synthesis 1978, 147–148.
[15] T. Iwata, Y. Okeda, Y. Hori, US 6,774,269 B2, 2004.
[16] M. Friedrich, K. Ebel, N. Gçtz, DE 10 2004 063 003 A1, 2006.
[17] L. Veum, U. Hanefeld, Chem. Commun. 2006, 825–831.
[18] R. A. Sheldon, I. Arends, U. Hanefeld, Green Chemistry and Catal-
ysis, Wiley-VCH, Weinheim, 2007.
[19] M. Fuentes, J. Magraner, C. de las Pozas, R. Roque-Malherbe, J.
Perez Pariente, A. Corma, Appl. Catal. 1989, 47, 367–374.
[20] P. Mäki-Arvela, N. Kumar, V. Nieminen, R. Sjçholm, T. Salmi, D. Y.
Murzin, J. Catal. 2004, 225, 155–169.
[21] A. Corma, M. Renz, Chem. Commun. 2004, 550–551.
[22] F. Iosif, S. Coman, V. Pârvulescu, P. Grange, S. Delsarte, D. De Vos,
P. Jacobs, Chem. Commun. 2004, 1292–1293.
[23] D.-L. Shieh, C.-C. Tsai, C.-W. Chen, A.-N. Ko, J. Chin. Chem. Soc.
2003, 50, 853–860.
[24] D.-L. Shieh, C.-C. Tsai, A.-N. Ko, React. Kinet. Catal. Lett. 2003, 79,
381–389.
[25] Z. Yongzhong, N. Yuntong, S. Jaenicke, G.-K. Chuah, J. Catal. 2005,
229, 404–413.
[26] Y. Nie, G.-K. Chuah, S. Jaenicke, Chem. Commun. 2006, 790–792.
[56] A. Infantes-Molina, J. MØrida-Robles, P. Maireles-Torres, E. Finoc-
chio, G. Busca, E. Rodríguez-Castellón, J. L. G. Fierro, A. JimØnez-
López, Microporous Mesoporous Mater. 2004, 75, 23–32.
[57] C. D. Wagner, A. V. Naumkin, A. Kraut-Vass, J. W. Allison, C. J.
Powell, J. R. Rumble, Jr., NIST X-ray Photoelectron Spectroscopy
Database, Version 3.4 (Web Version) http://srdata.nist.gov/xps/,
2003.
[58] B. Rakshe, V. Ramaswamy, S. G. Gegde, R. Vetrivel, A.V. Ramasw-
amy, Catal. Lett. 1997, 45, 41–50.
[59] Y. Zhu, G. Chuah, S. Jaenicke, J. Catal. 2004, 227, 1–10.
Chem. Eur. J. 2008, 14, 961 – 972
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
971

U. Hanefeld et al.
[60] E. Rodríguez-Castellón, A. JimØnez-López, P. Maireles-Torres, D. J.
Jones, J. Roziꢂre, M. Trombetta, G. Busca, M. Lenarda, L. Storaro,
J. Solid State Chem. 2003, 175, 159–169.
[64] In reference 51 the trans:cis ratio of 4 for the Zr-TUD-1 and Al-
TUD-1 catalysed MPV reductions is accidentally reported as cis:-
trans ratio.
[65] D. Klomp, K. Djanashvili, N. Cianfanelli Svennum, N. Chantapariya-
vat, C.-S. Wong, F. Vilela, T. Maschmeyer, J. A. Peters, U. Hanefeld,
Org. Biomol. Chem. 2005, 3, 483–489.
[66] Y. Nie, W. Niah, S. Jaenicke, G. K. Chuah, J. Catal. 2007, 248, 1–10.
[67] C. Simons, U. Hanefeld, I. W. C. E. Arends, R. A. Sheldon, T.
Maschmeyer, Chem. Eur. J. 2004, 10, 5829–5835.
[68] R. J. Abraham, J. J. Byrne, L. Griffiths, M. Perez, Magn. Reson.
Chem. 2006, 44, 491–509.
[61] D. Fuentes-Perujo, S. Santamaria-Gonzalez, J. Merida-Robles, E.
Rodríguez-Castellón, A. JimØnez-López, P. Maireles-Torres, R.
Moreno-Tost, R. Mariscal, J. Solid State Chem. 2006, 179, 2182–
2189.
[62] M. A. Camblor, A. Corma, J. Perez-Pariente, J. Chem. Soc. Chem.
Commun. 1993, 557–559.
[63] E. J. Creyghton, S. D. Ganeshie, R. S. Downing, H. van Bekkum, J.
Mol. Catal. A 1997, 115, 457–472.
Received: May 11, 2007
Published online: November 8, 2007
972
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 961 – 972