The room temperature, stoichiometric conversion of N2O to adsorbed NO by Fe-MCM-41 and Fe-ZSM-5

Article abstract of DOI:10.1006/jcat.2000.3005

N₂O is a greenhouse gas, which is liberated into the atmosphere in various ways, i.e., de-NO_x treatments, combustion of fossil fuels and biomass, and adipic acid production. Thus, the conversion of N₂O to adsorbed NO by Fe-MCM-41 and Fe-ZSM-5 at room temperature and > 10^-5 mbar was observed. The adsorbed product was identified by its characteristic but different N=O stretching frequencies on these two materials, and also by chemiluminescent analysis after desorption at > 420 K. The other product of N₂O decomposition was nitrogen. This decomposition was thermodynamically favorable, due not least to the heat of absorption of the adsorbed NO formed. Catalytic N₂O decomposition, forming nitrogen and oxygen, did not occur until > 770 K, probably by a different mechanism. Three samples were compared, i.e., Fe-MCM-41-SYN, Fe-MCM-41-PS, and Fe-ZSM-5, providing relative intensities that were normalized both for wafer thickness and iron content. The results were similar for Fe-ZSM-5 and for Fe-MCM-41 where the iron was introduced after framework synthesis. A much lower figure was noted, however, for Fe-MCM-41-SYN, suggesting that some of the iron in this sample was buried in the structure during synthesis. This correlated with the less well ordered nature of this sample. The similar results for Fe-MCM-41-PS and Fe-ZSM-5 suggested that the reaction might occur at isolated Fe(II) ions and iron-oxo nanoclusters.

Full text of DOI:10.1006/jcat.2000.3005

Journal of Catalysis 196, 126–133 (2000)
doi:10.1006/jcat.2000.3005, available online at http://www.idealibrary.com on
The Room Temperature, Stoichiometric Conversion of N2O to Adsorbed
NO by Fe-MCM-41 and Fe-ZSM-5
,2
Gerd Grubert, ,1 Michael J. Hudson,† Richard W. Joyner, ,2 and Michael Stockenhuber
Catalysis Research Laboratory, Department of Chemistry and Physics, Nottingham Trent University, Clifton Lane, Nottingham NG11 8NS,
United Kingdom; and †Department of Chemistry, University of Reading, Box 224, Whiteknights, Reading RG6 6AD, United Kingdom
Received April 25, 2000; revised July 10, 2000; accepted July 17, 2000
by hydrocarbons, particularly when platinum is an impor-
tant catalytic component, has generated renewed interest
in nitrous oxide decomposition (7).
Nitrous oxide (N2O) interacts at room temperature and low pres-
sure (>10 5 mbar) with Fe(II) species in both Fe-MCM-41 and Fe-
ZSM-5, to form adsorbed nitric oxide. The adsorbed product is iden-
With others(8), we are also interested in the use ofnitrous
oxide asan oxidant with transition-metal-containingporous
catalysts such as Fe-ZSM-5 and Fe-MCM-41 (9, 10). We
here report the observation, surprising but unambiguous,
that these materials convert nitrous oxide into adsorbed
nitric oxide at room temperature.
==
tified by its characteristic but different N O stretching frequencies
on these two materials, and also by chemiluminescent analysis af-
ter desorption at T > 420 K. It is proposed that the other product
of N2O decomposition is nitrogen. This decomposition is thermo-
dynamically favourable, due not least to the heat of adsorption of
the adsorbed NO formed. Catalytic N2O decomposition, forming
nitrogen and oxygen, does not occur until T > 770 K, probably by a
c
different mechanism.
2000 Academic Press
2. EXPERIMENTAL
Key Words: nitrous oxide (N2O); conversion to nitric oxide (NO);
MCM-41; Fe-ZSM-5; iron.
Si-MCM-41 was prepared in a heterogeneous system fol-
lowing the procedure of Gru¨n et al. (11). Cetyltriammoni-
umbromide (CTAB, Aldrich, 99% , 2.4 g) was dissolved in
water (120 ml) at 299 K, followed by the addition of ammo-
nia solution (8 ml, 25% ) and tetraethylorthosilicate (TEOS,
Aldrich, 98% , 10 ml). After the solution was stirred for
1 h, the precipitated sample was dried overnight at 353 K
and calcined at 823 K for 12 h. To introduce iron and/or
aluminum atoms during the synthesis, Fe(NO3)3 9H2O
(0.95 g) and/or Al(NO3)3 9H2O (1.12 g) were added to the
CTAB–water solution. To achieve a better distribution of
the metal atoms in the framework of the MCM-41, TEOS
was added before the ammonia, as it is known to form a
complex with the metal ion. The samples prepared by this
method are designated Fe-MCM-41-SYN and Al-MCM-
41-SYN.
1. INTRODUCTION
Nitrogen has a very rich oxide chemistry, forming ox-
ides in which it exhibits all of the formal oxidation states
from one to five. Although all are thermodynamically un-
stable with respect to decomposition into their constituent
elements, they usually display high kinetic stability and
are pernicious atmospheric contaminants. Most attention
has focused on nitric oxide and its atmospheric oxidation
products NO2 and N2O4 (NOx). However, nitrous oxide
(N2O) is a greenhouse gas which is liberated into the atmo-
sphere in a number of ways, including de-NOx treatments,
combustion of fossil fuels and biomass, and adipic acid
production.
Alternatively, iron was introduced into MCM-41 after
framework synthesis, usingiron acetylacetonate [Fe(acac)3]
in toluene, a method developed by Kenvin and white for the
preparation of iron oxide species on silica (12). Si-MCM-
41 (5 g) was dispersed in toluene (Aldrich 99.8% , 200 ml),
Fe(acac)3 (Aldrich 99% , 1.5 g) was added, and the mixture
was stirred at room temperature for 24 h. After filtration
and washing the resulting material was calcined at 823 K
for 12 h, and this material is designated Fe-MCM-41-PS.
Iron was introduced into H-ZSM-5 using a triple-
exchange process described previously (13). For each
exchange, H-ZSM-5 (3 g) was slurried for 24 h at room
There is an extensive literature on the catalytic decompo-
sition ofnitrousoxide into itselements, and thisreaction can
be carried out with varying degrees of efficiency by many
simple oxides (1, 2) and also by transition metal exchanged
zeolites (3–6). The recognition that N2O can be a signifi-
cant by-product of the selective catalytic reduction of NOx
1 Present address: Institut fu¨r Angewandte Chemie Berlin-Adlershof
e.V., Richard-Willsta¨tter-Str. 12, 12484 Berlin, Germany.
2 To whom correspondence should be addressed. R.W.J.: Fax: 44-115-
948-6838. E-mail: richard.joyner@ntu.ac.uk. M.S.: Fax: (44)-115-948-6694.
E-mail: michael.stockenhuber@ntu.ac.uk.
126
0021-9517/00 $35.00
c
Copyright
2000 by Academic Press
All rights of reproduction in any form reserved.

CONVERSION OF N2O TO ADSORBED NO BY Fe-MCM-41
127
TABLE 1
temperature in an iron nitrate solution (500 ml, 0.1 M) and
subsequently dried at 373 K.
Iron Content, Surface Area, and Mean Pore Size
for the Materials Studied
The structures of the MCM-41-derived materials were
characterised by X-ray diffraction using a Spectrolab CP5
Series 3000.120 powder diffractometer with nickel-filtered
Cu K radiation, physical adsorption of nitrogen (Mi-
cromeritics, ASAP 2000), and transmission electron mi-
croscopy (TEM) using a Philips CM20 instrument operat-
ing at 200 keV. The materials were also examined by in situ
infrared spectroscopy. An ATI Research Series FTIR spec-
trometer measuring in transmission mode was equipped
with a reaction chamber which has also been described be-
fore (13). Samples were initially activated at 773 K in a
1
Surface area/m2g
BJH pore size/A
˚
Sample
Fe/wt%
Si-MCM-41
—
4
3.5
0.45
1050
900
28
26
27
—
Fe-MCM-41-SYN
Fe-MCM-41-PS
Fe-ZSM-5
850
a
—
a Nitrogen physisorption does not provide meaningful surface areas for
microporous materials.
which will affect both the total pore volume and the surface
area (17).
7
vacuum (10 mbar), in air, or in hydrogen. To allow
quantitative comparisons, absorptions were normalised us-
ing overtone lattice vibrations with frequencies between
2200 cm 1 and 1700 cm 1 (14). The intensities of the over-
tone bands in these MCM-41 samples correspond well with
the literature (15), and are also used to obtain an estimation
of the relative wafer thickness of ZSM-5 samples. The valid-
ity of this approach was established in separate experiments
comparing lattice vibration intensity and wafer mass.
Nitrous oxide decomposition experiments were per-
formed in a fixed-bed microreactor that has been described
previously (16). Product analysis was performed by us-
ing a chemiluminescent NOx analyser (Signal Instruments
Model 4000) and a gas chromatograph fitted with a ther-
mal conductivity detector (Pye Unicam PU4550). Catalysts
were activated in flowing helium at 773 K for 1 h. The re-
actant mixture used to study the decomposition reaction
comprised 1000 ppm of nitrous oxide in helium. Powdered
catalysts were sieved to a mesh size of 0.6–1 mm, and the
bed volume was 1 ml.
The infrared spectra of all of the mesoporous MCM-41
materials studied are characterised by a broad feature at
3600 cm 1, which is due to clustered silanol groups or nests,
and a much sharper band at 3742 cm 1, which originates
from stretching vibrations of terminal Si–OH groups (18).
1
In Al-MCM-41 the intensity of the band at 3600 cm is
significantly decreased compared to that of purely siliceous
1
samples, whereas the band at 3742 cm is increased. For
iron-containing materials, spectra are largely independent
of whether activation takes place in a vacuum, in air, or in
hydrogen, and representative results for the two Fe-MCM-
41 samples and Fe-ZSM-5 are shown in Fig 2. The intensity
of the broad feature at ca. 3600 cm 1 is reduced when iron
is introduced after synthesis compared to the parent Al-
MCM-41 material, which is consistent with earlier observa-
tions (14). Introduction of iron during synthesis results in a
decreased intensity for this band. This is surprising, because
introduction of iron is expected to increase the concentra-
tion of defect sites, which should result in a higher concen-
tration of hydrogen-bonding silanol groups. By contrast,
the band at 3742 cm 1 is reduced for the material modified
with iron acetylacetonate, compared to the parent MCM-
3. RESULTS
All samples show four characteristic X-ray diffraction 41. In purely siliceous samples modified with Fe(NO3)3 in
features, at 2 = 2.0 , 3.7 , 4.2 , and 5.9 , consistent with a methanol solution, this band increases in intensity (14).
Fe-ZSM-5 has a sharp band at ca. 3610 cm 1, due to
˚
pore diameter of ca. 30 A (11). However, where iron has
been introduced during the synthesis step, the intensities bridged Brønsted acid sites, which is absent in all of the
of the two peaks at the highest diffraction angles are re- MCM-41 materials. Also, the intensity in the 3700 cm 1 re-
duced, reflecting a decrease in the ordering of the hexag- gion is much lower in Fe-ZSM-5 than in any of the MCM-41
onal pore structure. This is consistent with changes in the materials.
electron micrographs shown in Fig. 1, where the areas of
The results on the interaction with nitrous oxide are now
well-ordered hexagonal pores are smaller than in the par- presented. Figure 3 shows the FTIR spectrum of nitrous
ent MCM-41 material. It is also important to note, as Fig. 1 oxide at 1 mbar pressure, measured in the absence of any
demonstrates for Fe-MCM-41-SYN, that none of the mate- solid sample. There are several components in the range
1
rials show evidence of particulate iron oxides. The BET sur- 2190–2260 cm which fuse into the expected doublet as
face areas and the BJH pore sizes obtained from nitrogen the pressure is increased to 10 mbar (19). In all of the
adsorption are listed in Table 1, which also gives informa- remaining spectra, the contribution of gas-phase N2O has
tion on the iron content of the materials studied. Introduc- been removed by subtraction. No nitrous oxide adsorption
tion of iron, either during or after synthesis, decreases the was detected on MCM-41 or Al-MCM-41, so any bands at-
surface area by up to 20% . The loss of surface area is be- tributable to N2O adsorption result from interaction with
lieved to result from the introduction of iron into the pores, iron-containing sites.

128
GRUBERT ET AL.
FIG. 1. Transmission electron micrograph of Fe-MCM-41-SYN, in which 4% weight of iron is introduced during framework synthesis.
Figure 4 shows spectra obtained during nitrous oxide ad-
The observation of NO adsorption on these materials
sorption on a range of materials at 303 K, after activation in after exposure to nitrous oxide (N2O) is entirely unex-
a vacuum at 773 K. At the higher pressures studied, weak pected. An experiment has therefore been carried out in
bands attributable to nitrous oxide adsorption are observed order to eliminate the possibility that NO adsorption re-
in the range 2230–2300 cm 1. More surprising and inter- sults from its presence as an impurity in the N2O sup-
esting is the presence of much stronger bands observed at ply, even though the manufacturer’s specification is less
1
1
ca. 1830 cm on Fe-MCM-41, and ca. 1876 cm on Fe- than 1 ppm of NO impurity. Figure 5 compares the re-
6
ZSM-5. These positions are identical to those observed sults of exposing Fe-MCM-41-SYN to 10 mbar of NO
previously when nitric oxide (NO) interacts respectively and to 10 5 mbar N2O (equivalent to 10 11 mbar NO impu-
1
with Fe-MCM-41 and Fe-ZSM-5 (13, 14). A further broad rity), and similar intensities are observed at ca. 1825 cm
but weak feature is observed in the range 1620–1630 cm
.
1
,
Since the N2O interacts at such a low pressure to yield
which has previously been identified as being due to ad- spectra characteristic of adsorbed NO, any possible role
sorbed nitrogen dioxide or nitrate ions (20–22) formed by for NO as an impurity in our N2O supply can be ruled
disproportionation reactions.
out. The spectra clearly arise from the room temperature

CONVERSION OF N2O TO ADSORBED NO BY Fe-MCM-41
129
is detected when oxidised samples of Fe-MCM-41 materials
are exposed to N2O, and this is not surprising, given that
Fe(III) species have previously been demonstrated not to
adsorb NO to any significant extent (13, 14).
4. DISCUSSION
In what follows, we consider first the properties of the
hydroxyl groups and the location and structure of the iron
species associated with the MCM-41 materials, and then
discuss the unusual conversion of nitrous oxide to adsorbed
nitric oxide that we have observed.
In agreement with earlier observations (14), the intensity
of the hydroxyl stretching vibrations is significantly greater
in the mesoporous MCM-41 material than in the microp-
orous ZSM-5. However, even on MCM-41 materials con-
taining aluminum, no bands assignable to bridging acidic
hydroxyl groups were observed. Interestingly, the broad
1
band at 3600 cm which is attributed to clustered SiOH-
like groups in hydroxyl nests is weakened in our Fe-MCM-
41 materials compared to the case in the parent MCM-41,
irrespective of whether iron is introduced during or after
FIG. 2. FTIR spectra of samples activated in vacuo at 773 K:
(A) Fe-MCM-41-PS; (B) Fe-MCM-41-SYN; and (C) Fe-ZSM-5.
conversion of N2O to NO in the presence of iron-containing
materials.
Once formed, nitric oxide may be desorbed by heating. In
a fixed-bed microreactor, Fe-MCM-41-SYN was exposed to
1% N2O in helium at room temperature for 75 min. The ni-
trous oxide flow was then switched off, and the sample tem-
perature was increased in flowing helium. A small burst of
NO was detected by the NOx analyser between about 420
and 470 K. This is the same temperature range where we
have previously reported NO desorption from these mate-
rials (13, 14). Other experiments in the microreactor show
that the N2O decomposition reaction does not become cata-
lytic until a temperature of ca. 770 K. The products of the
catalytic decomposition are nitrogen and oxygen, and no
nitric oxide is observed.
We have previously shown that Fe(II) present in both
ZSM-5 and MCM-41 can easily be converted into Fe(III)
by exposure to oxygen. On the other hand, activation of
oxidised samples in a vacuum leads to reduction of Fe(III)
FIG. 3. Gas-phase spectrum of nitrous oxide, measured at 1 mbar
speciesto Fe(II) (13, 14). However, no adsorbed nitricoxide pressure.

130
GRUBERT ET AL.
FIG. 4. (a) Normalised, difference FTIR spectra observed for Fe-MCM-41-SYN on exposure to N2O at room temperature and different
5
4
3
2
1
pressures: (A) 10 mbar; (B) 10 mbar; (C) 10 mbar; (D) 10 mbar; (E) 10 mbar; (F) 1 mbar. (b) As for (a), but for Fe-MCM-41-PS:
5
4
3
1
(A) 10 mbar; (B) 10 mbar; (C) 10 mbar; (D) 10 mbar; (E) 1 mbar. (c) As for (a), but for Fe-ZSM-5. The noisier appearance of these
spectra is due to the substantially lower iron content of this sample compared to the iron-MCM-41 materials—see Table 1.
synthesis of the framework. This is an indication that iron framework, and the slight intensity decrease in the intensity
1
species are preferentially bound to these silanol nests, as of the band at 3740 cm suggests a partial interaction of
we have reported for iron introduced after synthesis into iron species with more isolated–OH groups.
purely siliceous MCM-41. Thus, the incorporation of iron
Previous detailed studies of ZSM-5 (13) and purely
during the MCM-41 synthesis appears to occur at strained siliceousMCM-41(14) showthat the iron cationsmayadopt
positions in the walls, where several hydroxyl groups may a number of different structural environments. In ZSM-5,
occur in very close proximity.
Significant differences in the hydroxyl spectra are ob- species, and the NO absorption band occurs at 1880 cm
iron-oxo nanoclusters of typical size Fe4O4 are the majority
1
.
served, dependent on whether iron is introduced from the A small number of isolated cations also occur, for which
acetylacetonate dissolved in toluene or from a methano- the NO absorption band is at 1840 cm 1. In Si/MCM-
lic solution of iron nitrate. In methanolic solutions, 41(MeOH), iron is present mainly in the form of iso-
1
the intensity of the band at 3732 cm was increased lated cations, on which the NO absorption band occurs at
by treatment with methanol, whereas it was decreased by 1823 cm 1. Only a very small number of nanoclusters exist
treatment with iron acetylacetonate in toluene. This is con- in this material (14). Since in both cases, Fe-ZSM-5 (large
sistent with our findings (to be reported in detail elsewhere) concentration of clusters) and Fe-MCM-41 (mostly iso-
that methanol reacts with the MCM-41 framework to form lated), significant conversion of N2O to NO occurs, we be-
bound methoxy species and more isolated hydroxyl groups lieve that both species, isolated cations, and clusters are act-
(14). The toluene solvent does not react with the MCM-41 ing as sites for the reaction. UV–visible spectroscopy and

CONVERSION OF N2O TO ADSORBED NO BY Fe-MCM-41
131
chiometric but it is highly efficient, and we have presented
experiments that allow us to rule out impurity effects or
other experimental artefacts.
In the MCM materials, differences in the absorption ca-
pacities for both NO and N2O are observed, dependent
on whether iron is introduced during (SYN) or after (PS)
the framework synthesis step. For Fe-MCM-41-SYN, bands
due to N2O are found only at higher pressures, compared to
those for Fe-MCM-41-PS. At lower pressures, the intensity
of the NO band is higher in Fe-MCM-41-SYN, but lower
at higher pressures. We believe that N2O is first adsorbed
at an iron cation and then converted to NO, which remains
adsorbed on the same site. Conversion of N2O to NO seems
to be faster than that for Fe-MCM-41-PS, because at lower
pressure the intensity of NO bands is higher and the inten-
sity of the N2O bands is lower for Fe-MCM-41-SYN than
for the postsynthesis sample. Thus, more N2O has been con-
verted to NO. We do not understand the reasons for these
small differences in the reactivity of the two samples. They
FIG. 4—Continued.
EXAFS measurements to be reported elsewhere suggest
that in all of the MCM-41 materials studied here iron is
present mainly in the form of isolated cations. The extent
of iron aggregation, however, is likely to increase with the
transition metal loading.
Molecular nitrous oxide is held only weakly both by the
Fe-MCM-41 materials and by Fe-ZSM-5, and significant ab-
sorption bands are observed in the infrared spectra only at
pressures of ca. 10 1 mbar or higher. Frequencies observed
are typically higher than those of gas-phase N2O, but the na-
ture of bonding to the iron species is not clear. The results
presented, however, show that nitrous oxide interacts at
room temperature, both with Fe-MCM-41 and Fe-ZSM-5,
to form adsorbed nitric oxide. Conclusive evidence for the
==
formation of adsorbed NO is that the main N O stretching
frequency is known to occur at quite different positions in
the infrared spectrum when it is adsorbed on Fe-MCM-41
1
(1823 cm 1 (18)), or nanoclusters in Fe-ZSM-5 (1880 cm
(13)). These are also the band positions that we observe
when N2O is allowed to react with these materials, as shown
in Fig. 4. The reaction to form adsorbed NO occurs even at
FIG. 5. FTIR difference spectra of the NO adsorption region for Fe-
MCM-41-SYN after adsorbing: (A) N2O at 10 5 mbar pressure; (B) NO
the lowest N2O pressure studied, 10 5 mbar. It is only stoi- at 5 10 6 mbar pressure.

132
GRUBERT ET AL.
TABLE 2
may reflect subtle changes in the iron coordination envi-
ronment indicated by small changes in EXAFS parameters
and IR stretching frequencies.
At the highest pressures studied, we believe that the ma-
jorityofaccessible iron cationsare covered with NO or N2O.
Although the iron content in Fe-MCM-41-PS is lower than
that when iron is introduced during framework synthesis,
a larger concentration of NO is observed, as the iron sites
are thought to be more accessible.
Relative Intensities of the Infrared Adsorption Bands for
Adsorbed NO Observed on the Different Materials Studied,
Normalised for Different Iron Contents and Wafer Thicknesses
Sample
Fe-MCM-41-SYN
0.024
Fe-MCM-41-PS
0.045
Fe-ZSM-5
0.050
Relative intensity
Reaction between adsorbed nitrous oxide and the iron-
containing materials to form adsorbed nitric oxide has not
been reported before. As noted above, the reaction takes
place when the materials are in their reduced form, and
iron is present as Fe(II). We therefore propose that the
reaction is
A measure of the extent to which the N2O to NO conver-
sion occurs can be gained by examining the intensity of the
absorption bands due to NO. Table 2 compares the three
samples studied, providing relative intensities that are nor-
malised both for wafer thickness and iron content. The re-
sults are similar for Fe-ZSM-5 and for Fe-MCM-41 where
the iron has been introduced after framework synthesis.
A much lower figure is noted, however, for Fe-MCM-41-
SYN, suggesting that some of the iron in this sample has
been buried in the structure during synthesis. This corre-
lates with the less well ordered nature of this sample. The
similar results for Fe-MCM-41-PS and Fe-ZSM-5 suggest
that the reaction may occur at both isolated Fe(II) ions and
iron-oxo nanoclusters.
N2O → NO(ads) + 0.5N2.
The fate of the nitrogen atoms is not known. They may
remain as an adsorbed species; however, they most prob-
ably combine and desorb as N2. This reaction sequence
is thermodynamically favourable, since the free energy
change for
2N2O → 2NO + N2 (all species in the gas phase)
is 1G2o98
=
35.7 kJ mol 1. The adsorption of NO on Fe(II)
ACKNOWLEDGMENTS
species is exoenergetic, further increasing the thermody-
namic driving force.
We are very grateful to Philippe Trens (Reading University) for help
with the MCM-41 synthesis and to Ian Burton (previously at Nottingham
Trent University, now at the Queen’s University of Belfast) for carrying
out the nitrous oxide decomposition experiments. We also acknowledge
the financial support of the Engineering and Physical Sciences Research
Council.
We have not been able to make the N2O decomposition
reaction catalytic at low temperatures. Nitric oxide does not
desorb at room temperature and therefore self-poisons the
potential catalyst against further activity. At higher temper-
atures, once NO does desorb from the catalyst, we believe
that the interaction of nitrous oxide with the iron cations,
already weak at room temperature, has become too weak
for it to adsorb and react.
REFERENCES
As with many oxide materials containing transition metal
cations, Fe-MCM-41 and Fe-ZSM-5 show activity for the
catalytic decomposition of nitrous oxide into its elements,
but only at temperatures in excess of ca. 700 K (1, 3, 4).
The course of that interaction is thus quite different from
that reported here, and the mechanism is also likely to be
different, almost certainly involving the fission of the N–O
bond, rather than the split into N and NO fragments that
we observe here. Our results suggest that a cyclic tem-
perature swing process for the conversion of N2O to NO
can be envisaged, whereby adsorbed NO is formed by in-
teraction of N2O with Fe-MCM-41 at room temperature,
and the product desorbed by heating to a temperature of
ca. 420–470 K. This could be practically advantageous, as
the selective reduction of NO to nitrogen can be carried out
at lower temperatures than are normally needed for N2O
decomposition.
1. Kapteijn, F., Rodriguez-Mirasol, J., and Moulijn, J. A., Appl. Catal. B:
Environ. 9, 25 (1996).
2. Xie, S., and Lunsford, J. H., Appl. Catal. A: General 188, 137 (1996).
3. Akbar, S., and Joyner, R. W., J. Chem. Soc., Faraday Trans. 1 77, 803
(1981).
4. Rauscher, M., Kesore, K., Mo¨nning, R., Schweiger, W., Tissler, A.,
and Turek, T., Appl. Catal. A: General 184, 249 (1999).
5. Mauvezin, M., Delahay, G., Coq, B., and Kieger, S., Appl. Catal. B:
Environ. 23, L79 (1999).
6. Valyon, J., Millman, W. S., and Hall, W. K., Catal. Lett. 24, 215 (1994).
7. See, e.g.: Tanaka, T., Okuhara, T., and Misono, M., Appl. Catal. B:
Environ. 4, L1 (1994).
8. Sugiyama, S., Abe, K., Hayashi, H., Matsumura, Y., and Moffat, J. B.,
J. Mol. Catal. A: Chem. 144, 347 (1999).
9. Panov, G., Kharitonov, A. S., and Sobolev, V. I., Appl. Catal. A: General
98, 1 (1993).
10. Pophal, C., Yogo, T., Yamada, K., and Segawa, K., Appl. Catal. B:
Environ. 16, 177 (1998).
11. Gru¨n, M., Unger, K. K., Matsumoto, A., and Tsutsumi, K., Microp-
orous Mesoporous Mater. 27, 207 (1999).

CONVERSION OF N2O TO ADSORBED NO BY Fe-MCM-41
133
12. Kenvin, J. C., and White, M. G., Langmuir 7, 1198 (1991).
13. Stockenhuber, M., and Joyner, R. W., J. Phys. Chem. B 103, 5963
(1999).
14. Stockenhuber, M., Hudson, M. J., and Joyner, R. W., J. Phys. Chem. B
104, 3370 (2000).
15. Jentys, A., Kleesdorfer, K., and Vinek, H., Microporous Mesoporous
Mater. 27, 321 (1999).
16. Connerton, J., Stockenhuber, M., and Joyner, R. W., J. Chem. Soc.,
Chem. Commun. 185 (1997).
18. Jentys, A., Pham, N. H., and Vinek, H., J. Chem. Soc., Faraday Trans.
92, 3287 (1996).
19. See the Galactic Industries database of infrared spectra at www.
galactic.com.
20. Hoost, T. E., Laframboise, K. A., and Otto, K., Appl. Catal. B: Environ.
7, 79 (1995).
21. Kucherov, A. V., Slinkin, A. A., Kondreatev, A. A., Rubinsthein,
T. N., and Minachev, K. M., Zeolites 5, 320 (1985).
22. Aylor, A. W., Larsen, S. C., Reimer, J. A., and Bell, A. T., J. Catal. 157,
592 (1995).
17. Mokaya, R., and Jones, W., J. Mater. Chem. 9, 555 (1999).