UV-Vis microspectroscopy: probing the initial stages of supported metal oxide catalyst preparation.

Abstract

A UV-vis microspectroscopy methodology for monitoring the speciation and macrodistribution of catalyst-precursor species inside catalyst-support bodies at the initial stages of catalyst preparation has been developed. The setup is based upon optical-fiber technology and allows spatially resolved analysis of bisected catalyst bodies. The potential of this tool is demonstrated by two pore-volume impregnation studies involving Ni2+ d-d transition bands and Cr6+ charge-transfer bands.

Full text

Page 1

UV-Vis Microspectroscopy: Probing the Initial Stages of Supported Metal
Oxide Catalyst Preparation
Leon G. A. van de Water, Jaap A. Bergwerff, T. Alexander Nijhuis, Krijn P. de Jong, and
Bert M. Weckhuysen*
Department of Inorganic Chemistry and Catalysis, Debye Institute, Utrecht UniVersity, P.O. Box 80083,
3508 TB Utrecht, The Netherlands
Received September 13, 2004; E-mail: B.M.Weckhuysen@chem.uu.nl
The need for catalysts with well-defined activity, selectivity, and
stability is well recognized.1Rational design of improved catalysts
requires a detailed understanding of the structure-function cor-
relations of catalytic materials. Major advances in catalysis research
are only to be expected when the detailed nature (i.e., on the
molecular level) of the active species is known and the phenomena
occurring upon synthesis fully understood. As such, the develop-
ment of new synthetic strategies depends strongly on the emergence
of advanced characterization techniques for these materials and
preparation processes. The importance of the HRTEM technique
for the understanding and subsequent fast development of the zeolite
synthesis field illustrates this point.2
A large number of catalytic materials comprise metal (oxide)
species supported onto oxidic supports, where the local molecular
structure of the active component(s) and the macrodistribution over
the support are important factors in terms of catalyst performance.
Many of the phenomena occurring during catalyst preparation still
defy thorough understanding.3This lack of fundamental understand-
ing is the reason that improvements in the field of catalyst
preparation are most often not based on mastering the chemical
and physical processes occurring at multiple length scales, but,
instead, are the result of long trial-and-error processes.
In this communication, we present a UV-vis microspectroscopy
tool that enables monitoring, with a resolution of 100 µm or less,
of the molecular structure of catalytically active components, along
with their macrodistribution over support bodies. This study thus
aims at investigating catalyst preparation at multiple length scales
using UV-vis spectroscopy and is as such, to the best of our
knowledge, the first of its kind. We will show that valuable
additional information can be obtained, compared to conventional
spectroscopic studies involving supported catalysts in powder form.4
The UV-vis setup comprises a probe containing seven optical
fibers, one connected to a light source and six to a spectropho-
tometer (Figure 1). Spatially resolved measurements are performed
by remote-controlled automated stepwise motion of the measuring
cell, containing a bisected catalyst body. The setup is designed for
probing the metal-ion speciation of catalyst precursors along the
cross-section of bisected millimeter-sized catalyst bodies, allowing
the study of both the wet (i.e., impregnation) and the dry stages
(drying, calcination) of the preparation process. To illustrate the
broad range of catalytic systems for which this methodology may
be applied to, two examples are presented here, including Ni (d-d
transition bands) on γ-Al2O3and Cr (charge-transfer bands) on
γ-Al2O3.
Ni2+d-d transition bands have been monitored qualitatively after
pore-volume impregnation of γ-Al2O3pellets (L ) 3 mm) with
an acidified solution containing 0.5 M Ni2+and 1.5 M ethylene-
diamine (en) (pH ) 4.7). The UV-vis spectra, recorded 5 min
after impregnation, are depicted in Figure 2 (top 5 traces). The
spectrum of the impregnation solution (trace I) is included for
reference purposes. The spectrum recorded near the edge of the
cross-section (1500 µm) clearly resembles that of the solution. In
contrast, the spectra recorded further toward the center of the pellet
show a clear blue shift of the absorption bands; the ν2band is shifted
from 15 400 cm-1near the edge (1500 µm) to 17 400 cm-1in the
center of the pellet (0 µm), and the ν3band is shifted from 25 900
to 28 950 cm-1. The shift of these bands is caused by changes in
Figure 1. UV-vis DRS measuring cell. (a) UV-vis probe (fixed position);
(b) closed glass cell containing a small amount of water to ensure a relative
humidity of 100%; (c) sample holder; (d) UV-vis probe (enlarged)
containing seven optical fibers (Ø ) 100 µm), connected either to the light
source (e) or to the spectrophotometer (f).
Figure 2. γ-Al2O3pellets impregnated with 0.5 M (Ni(NO3)2+ 1.5 M en
+ HNO3(pH 4.7), 5 min after impregnation. The distance from the center
of the pellet is indicated. (I) Impregnation solution; (II) impregnated
powdered γ-Al2O3sample.
Published on Web 03/17/2005
5024 9 J. AM. CHEM. SOC. 2005, 127, 5024-5025
10.1021/ja044460u CCC: $30.25 © 2005 American Chemical Society

Page 2

the coordination sphere of Ni2+due to changes in the local pH
upon diffusion of the acidic solution into basic γ-Al2O3. The en
ligands are protonated in the impregnation solution and are not
coordinated to Ni2+. Toward the center of the γ-Al2O3pellet, the
pH increases, deprotonation of protonated en occurs, and water
ligands are exchanged for en. By comparing the position of the
UV absorption band in the center of the pellet (17 400 cm-1) with
that of spectra measured in solution at varying pH, the local pH in
the center of the pellet was estimated to be ∼6.2. Hence, the UV-
vis spectra reveal detailed information not only on the distribution
and local coordination environment of Ni2+but also indirectly on
the local pH value. Trace II in Figure 2 shows the UV-vis spectrum
of a sample of crushed γ-Al2O3, 5 min after impregnation with the
same solution. This spectrum can be regarded as the average
spectrum of the spectra recorded across the cross-section and is
very much solution-like. In other words, measurements on impreg-
nated γ-Al2O3powders do not reveal the detailed information on
local differences inside catalyst support bodies during impregnation
that can be obtained by analyzing bisected γ-Al2O3catalyst bodies
with the developed UV-vis optical fiber setup.
The transport of chromium oxide through γ-Al2O3bodies after
impregnation with a CrO3solution is a slow process. The UV-vis
(DRS) spectra recorded 6 h after impregnation of γ-Al2O3pellets
with a 0.01 M CrO3solution (the pH adjusted to 6.1) show charge-
transfer bands of chromate (CrO42-) species, found at around 36 350
and 26 650 cm-1.5In Figure 3a, the 26 650 cm-1band is depicted
at different positions across the surface of a bisected support body.
The band positions do not shift throughout the pellet, indicating
that no species other than CrO42-, such as Cr2O72-(which has an
absorption band at 22 200 cm-1), are present. To be able to quantify
the CrO42-distribution throughout the pellet, the charge-transfer
band at 26 650 cm-1was used to create a calibration curve. The
CrO42-concentration profiles at different points in time inside
γ-Al2O3bodies have been quantified using this calibration curve
and are depicted in Figure 3b. After 2 h, a sharp drop in Cr
concentration over the first 500 µm from the edge was observed.
After 72 h, the CrO42-distribution was found to have reached an
equilibrium value close to the calculated 0.011 mmol Cr/g γ-Al2O3
(1.1 mL of the 0.01 M CrO3solution per gram of γ-Al2O3).
The spectroscopic information obtained with the optical fiber
UV-vis setup presented here highlights the potential of this tool
in probing the local metal-ion speciation throughout catalyst-
support bodies in the early stages of catalyst preparation. The
changes occurring to catalyst-precursor species during preparation
are often a function of subtle changes in the ratio of different
components, pH,6and reactivity of the support.7Until now, these
phenomena inside catalyst bodies have not been described with
spatial resolution, required in the quest for catalysts with narrowly
defined properties. It is envisaged that UV-vis microspectroscopy
may contribute toward a better understanding of the preparation
process of supported catalysts, involving supports such as alumina
and metal-oxide species with characteristic d-d or charge-transfer
transitions. Moreover, a more detailed understanding of the prepara-
tion process of more complex catalysts, such as bimetallic (Co)-
MoS2-γ-Al2O3catalysts, may be within reach. In certain cases,
where the impregnated species do not change throughout the support
body and over time, even quantitative analysis can be performed.
When combined with spectroscopic studies on catalyst performance
under reaction conditions,8the fate of a supported metal oxide may
be studied throughout its whole life cycle. Finally, combining it
with other local probing techniques, such as Raman microscopy,9
may even further enhance the potential of this tool. UV-vis is
complementary to Raman spectroscopy in cases where multiple
species are present, such as Co and Mo, where Co2+can be
monitored with UV-vis and Mo6+with Raman spectroscopy.
The analysis of (biological) microscopic samples, using FT
infrared microspectroscopy with high spatial resolution (10-100
µm), is well documented.10In the future, UV-vis microspectros-
copy may have potential toward applications different from
heterogeneous catalysis, such as for biomedical and materials
science purposes.
Acknowledgment. B.M.W. acknowledges NWO-CW for a
VICI grant, whereas T.A.N. is grateful to a NWO-STW VIDI grant.
Supporting Information Available:
synthetic method details (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
Experimental details and
References
(1) (a) Fujdala, K. L.; Drake, I. J.; Bell, A. T.; Tilley, T. D. J. Am. Chem.
Soc. 2004, 126, 10864-10866. (b) Fujdala, K. L.; Tilley, T. D. J. Catal.
2003, 216, 265-275. (c) Bell, A. T. Science 2003, 299, 1688-1691. (d)
Schlo ¨gl, R.; Abd Hamid, S. B. Angew. Chem., Int. Ed. 2004, 43, 1628-
1637. (e) de Jong, K. P. Curr. Opin. Solid State Mater. Sci. 1999, 4, 55-
62.
(2) Thomas, J. M.; Terasaki, O.; Gai, P. L.; Zhou, W.; Gonzalez-Calbet, J.
Acc. Chem. Res. 2001, 34, 583-594.
(3) (a) Neimark, A. V.; Kheifez, L. I.; Fenelonov, V. B. Ind. Eng. Chem.
Prod. Res. DeV. 1981, 20, 439-450. (b) de Jong, K. P. CatTech 1998, 3,
87-95. (c) Lekhal, A.; Glasser, B. J.; Khinast, J. G. Chem. Eng. Sci.
2004, 59, 1063-1077.
(4) (a) Negrier, F.; Marceau, E.; Che, M. Chem. Commun. 2002, 1194-1195.
(b) d’E Äspinose de la Caillerie, J.-B.; Kermarec, M.; Clause, O. J. Am.
Chem. Soc. 1995, 117, 11471-11481. (c) Bru ¨hwiler, D.; Frei, H. J. Phys.
Chem. B 2003, 107, 8547-8556. (d) Carriat, J. Y.; Che, M.; Kermarec,
M.; Verdaguer, M.; Michalowicz, A. J. Am. Chem. Soc. 1998, 120, 2059-
2070.
(5) Weckhuysen, B. M.; Wachs, I. E.; Schoonheydt, R. A. Chem. ReV. 1996,
96, 3327-3349.
(6) Schreier, M.; Regalbuto, J. R. J. Catal. 2004, 225, 190-202.
(7) Carrier, X.; Lambert, J.-F.; Kuba, S.; Kno ¨zinger, H.; Che, M. J. Mol.
Struct. 2003, 656, 231-238.
(8) (a) Weckhuysen, B. M. Phys. Chem. Chem. Phys. 2003, 5, 4351-4360.
(b) Nijhuis, T. A.; Tinnemans, S. J.; Visser, T.; Weckhuysen, B. M. Phys.
Chem. Chem. Phys. 2003, 5, 4361-4365.
(9) Bergwerff, J. A.; Visser, T.; Leliveld, B. R. G.; Rossenaar, B. D.; de Jong,
K. P.; Weckhuysen, B. M. J. Am. Chem. Soc. 2004, 126, 14548-14556.
(10) (a) Wetzel, D. L.; LeVine, S. M. Science 1999, 285, 1224-1225. (b) Tobin,
M. J.; Chesters, M. A.; Chalmers, J. M.; Rutten, F. J. M.; Fisher, S. E.;
Symonds, I. M.; Hitchcock, A.; Allibone, R.; Dias-Gunasekara, S. Faraday
Discuss. 2004, 126, 27-39.
JA044460U
Figure 3. (a) UV-vis spectra recorded 6 h after impregnation of γ-Al2O3
pellets with a 0.01 M CrO3solution (pH 6.1) across the surface of a bisected
support body. (b) CrO42-concentration profiles inside γ-Al2O3bodies at
different points in time: (A) 2 h, (B) 6 h, and (C) 72 h after impregnation.
C O M M U N I C A T I O N S
J. AM. CHEM. SOC. 9 VOL. 127, NO. 14, 2005 5025