A Nitrate Route for the low temperature Fast SCR reaction over a V2O5-WO3/TiO2 commercial catalyst

Article abstract of DOI:10.1039/b411613e

A novel mechanism is proposed for the Fast SCR reaction of NH₃, NO and NO₂ at low temperature involving the formation of ammonium nitrate as intermediate and its subsequent reaction with NO as the rate determining step.

Full text of DOI:10.1039/b411613e

View Online / Journal Homepage / Table of Contents for this issue
A ‘‘Nitrate Route’’ for the low temperature ‘‘Fast SCR’’ reaction over a
V O –WO /TiO commercial catalyst
2
5
3
2
a
a
a
b
Cristian Ciardelli, Isabella Nova, Enrico Tronconi,* Daniel Chatterjee and Brigitte Bandl-Konrad
b
a
Dip. di Chim., Mat. e Ing. Chimica ‘‘G. Natta’’, Politecnico di Milano, P.zza L. Da Vinci 32, I-20133
Milano, Italy. E-mail: enrico.tronconi@polimi.it; Fax: 139 02 2399 3318; Tel: 139 02 2399 3264
DaimlerChrysler AG Abteilung RBP/C, HPC: 096-E220, D-70546 Stuttgart, Germany.
E-mail: daniel.chatterjee@daimlerchrysler.com; Fax: 149 711 17 53915; Tel: 149 711 17 53915
b
Received (in Cambridge, UK) 29th July 2004, Accepted 6th September 2004
First published as an Advance Article on the web 8th October 2004
A novel mechanism is proposed for the Fast SCR reaction of
NH , NO and NO at low temperature involving the
addition to the Fast SCR, reaction (2). However, we notice that the
combination of reactions (3) and (4) already fully accounts for the
3
2
formation of ammonium nitrate as intermediate and its
subsequent reaction with NO as the rate determining step.
selective reduction of NO 1 NO by NH : reaction (3) selectively
2
3
reduces half of NO
2
directly to N , the remaining half being
2
converted to nitrate species; such nitrate species can react with NO
to form more NO according to the redox reaction (4); the extra
A key feature in the development of SCR catalysts for automotive
applications is the enhancement of the DeNOx activity at low
temperature. The standard SCR process is based on the following
2
formed NO
(3), thus completing the selective reduction of NO
2
can then react further with more NH
3
according to
. Notice in
x
to N
2
reaction between NH and NO:
3
fact that the sum of reactions (3) and (4) formally results in the
stoichiometry of reaction (2). Based on this picture, given an
equimolar mixture of NH and NO , with NO/NO ~ 1/1, we
4
NH
3
1 4 NO 1 O
2
A 4 N
2
2
1 6 H O
(1)
3
x
2
1
expect the following situations depending on the relative rates of
reactions (3) and (4): a) if (3) is faster than (4), we shall observe the
formation/build-up of ammonium nitrate, with only a limited
In the early ’80s Kato and co-workers found that the reaction
involving an equimolar NO and NO feed, with stoichiometry
2
occurrence of the ‘‘Fast SCR’’, i.e. only a partial reduction of NO
2
NH
3
1 NO
2
1 NO A 2 N
2
2
1 3 H O
(2)
x
6
: this is the situation actually observed by Koebel et al. at low
to N
2
temperatures (150 uC); b) if (4) is faster than (3), no formation of
NH NO will be detected, ammonium nitrate acting as a reactive
is considerably faster than reaction (1). The importance of the so
called Fast SCR reaction (2) has recently much increased, since the
possibility to install an oxidation precatalyst upstream of the SCR
catalyst, thus incrementing the NO /NO feed ratio, provides a
2 x
chance to boost the low temperature deNOx activity of SCR
devices for automotive exhaust after-treatment.
4
3
intermediate, and the occurrence of the ‘‘Fast SCR’’ reaction (2)
will be apparently observed, resulting in reality from the
simultaneous occurrence of (3) and (4): this is in line with the
6
data reported by Koebel et al. at higher temperatures.
We present herein transient low-T reaction experiments
In the last few years Koebel, Madia and co-workers have
extensively investigated the Fast SCR reaction over V O /WO /
TiO2 catalysts, both in powder and in monolith form.
performed with NH /NO/NO mixtures over
3
a powdered
2
2
5
3
2
–6
commercial V/W/TiO2 SCR catalysts (160 mg) in a quartz
microreactor while analysing the outlet gas concentration profiles
by a mass spectrometer, taking into account the response factors
and the overlapping fragmentation patterns of all the relevant
Concerning the catalytic mechanism, they reported reactivity and
spectroscopic experiments showing that NO2 enhances the
reoxidation of vanadia, regarded as the rate determining step of
7
5
species. The experiments herein reported were designed to address
the SCR reaction at low temperatures. Accordingly, they proposed
that the Fast SCR should be considered as a standard SCR
the role of reactions (3) and (4) in the catalytic mechanism of the
Fast SCR reaction. 1% water was added to the feed stream,
whereas no oxygen was included in the feed to simplify the
mechanistic interpretation: in fact preliminary experiments proved
that the O effect on the results herein reported is negligible.
2
Fig. 1 shows a composite transient response run at 170 uC: in a
2
reaction in which NO plays the role of a more efficient oxidizer for
the vanadium sites than oxygen. However, no direct kinetic
evidence was offered to prove that the enhanced rate of V
41
reoxidation can explain the order-of-magnitude enhancement of
the DeNOx rate observed in the presence of equimolar mixtures of
stream of He and 1% H O 1000 ppm of NO
2
the reactor (phase A). Upon step addition of 1000 ppm NH
2
were first admitted to
(phase
NO/NO
complexity of the chemistry resulting from addition of NO
SCR reacting system. In fact, on studying the reactivity of the NH
2
. Besides, this picture does not account for the greater
2
to the
3
3
/
6
NO/NO
reported the occurrence of two main reactions not observed with
NH /NO, namely the formation of NH NO from NH and NO2,
2
system over V-based catalysts at 150 uC, Koebel et al.
3
4
3
3
4
NH
and the redox decomposition of NH
NH NO u 2 NH 1 2 HNO
3
1 4 NO
2
A 2 NH
4
NO
3
1 2 N
2
1 2 H
2
O
(3)
4
NO
3
by NO,
2
(4.a)
(4.b)
4
3
3
3
NO 1 2 HNO
Reaction (4) is the sum of reactions (4.a) and (4.b)]
NH NO 1 NO A 3 NO 1 2 NH 1 H
3
2 2
A 3 NO 1 H O
[
2
4
3
2
3
2
O
(4)
Fig. 1 Ammonium nitrate formation and subsequent decomposition by
21
According to Koebel et al., reactions (3) and (4) occurred in
2
NO. T ~ 170 uC; GHSV ~ 90000 h ; H O ~ 1% in He.
2
7 1 8
C h e m . C o m m u n . , 2 0 0 4 , 2 7 1 8 – 2 7 1 9
T h i s j o u r n a l i s ß T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

View Online
2 3 2
B), NO and NH were seen to react with formation of N . At the
end of phase B about 490 ppm of NO and 520 ppm of NH were
2
3
converted while 250 ppm of nitrogen were produced, in line with
the stoichiometry of NH NO formation, reaction (3). The build-
4
3
up of NH
catalyst sample discharged immediately after phase B.
After the build-up of NH NO on the catalyst obtained by co-
4 3
NO upon the catalyst was confirmed by IR analysis of a
4
3
feeding NO and NH at 170 uC for about 1 hour (phase B in
2
3
Fig. 1), the NO
2
feed was shut down at t ~ 5300 s: correspondingly
the ammonia level increased to 1000 ppm and nitrogen was no
longer detected (phase C): in these conditions no reaction took
place. At t ~ 9800 s 1000 ppm of NO were admitted to the reactor:
conversion of both NO and NH was readily observed along with
3
2
the corresponding formation of N (phase D), both phenomena
however progressively decaying with time. This is explained
observing that according to reaction (4) NO can react with the
previously deposited ammonium nitrate producing NO and NH .
Fig. 2 Rate comparison between Fast SCR (2) and reaction of NH NO
4
3
2
3
21
2
with NO (4): T ~ 170 uC; GHSV ~ 90000 h ; H O ~ 1% in He.
We propose however that NO
according to (3) to give N and ammonium nitrate. Considering the
excess of NH present during phase (D), and assuming reaction (3)
to be faster than (4), the following scheme applies:
2 3
and NH further react in turn
2
the Fast SCR reaction (2), which converted 220 ppm of NH
10 ppm of NO and 110 ppm of NO generating 220 ppm of N
and the reaction of NH NO formation (3), which consumed the
remaining 230 ppm of ammonia and 230 ppm of NO producing
3
,
3
1
2
2
,
4
3
2
2
NH NO
4
3
1 NO A 3 NO
2
1 2 NH
3
2
1 H O
(4)
110 ppm of nitrogen. This means that during phase B ammonium
nitrate was being accumulated onto the catalyst; at the same time
the Fast SCR was progressing to an extent directly reflected by the
NO conversion level.
3
NH 1 3 NO A 3/2 NH NO 1 3/2 N 1 3/2 H O (3)
3
2
4
3
2
2
[
Reaction (5) is the sum of reactions (3) and (4)]
/2 NH NO 1 NO 1 NH A 3/2 N 1 5/2 H
Indeed the evolution of NH , NO and N in phase D of Fig. 1
agrees exactly with the stoichiometry of (5): NO and ammonia
signals are nearly overlapped giving evidence of an equimolar
consumption, and eventually rise from about 800 ppm to 1000 ppm,
i.e. their feed concentration, upon complete depletion of the
deposited ammonium nitrate. NO
negligible throughout phase D. The N
specular to NH , and then decreases as NH NO is progressively
converted. Notice that the nitrogen peak is of about 300 ppm, being
associated with a consumption of 200 ppm of NH and NO, in
accord with reaction (5). We emphasize that the overall experiment
of Fig. 1 resulted in the conversion of NH 1 NO 1 NO to N
and this clearly occurred via formation/decomposition of ammo-
nium nitrate.
In a dual experiment (not shown), performed at 140 uC, NO
alone was admitted to the reactor after NH
consequence, production of NO
detected. Without excess ammonia in fact the expected reaction
scheme, (4) 1 (3), yields:
During the second part of the run, phase C in Fig. 2, the NO
feed was shut down thus stopping nitrate formation. Nevertheless
NH and NO consumption as well as nitrogen production were still
2
1
4
3
3
2
2
O
(5)
3
noticed, in analogy with what observed in Fig. 1, phase D: this
indicates the occurrence of reaction (4) between the stored nitrate
and NO. The most important result shown in Fig. 2 is that the NO
conversion level remained practically unaltered{ in going from
phase B to phase C: this proves that the Fast SCR (during phase B)
3
2
4 3
and the reaction between NO and NH NO (during phase C)
2
concentration is practically
progressed at the same rate, which rules out the hypothesis that
reaction (2) is consecutive to the nitrate decomposition, reaction
(4). On the contrary, this result is strongly suggestive of a scheme
where the Fast SCR, reaction (2), results from the sum of reactions
(3) and (4), reaction (4) acting as the rate determining step at the
investigated low temperatures. According to this picture, thus,
ammonium nitrate is not a terminal species, but behaves as an
intermediate of the Fast SCR reaction.
2
signal exhibits a maximum
3
4
3
3
3
2
2
,
8,9
Sachtler and co-workers recently proposed a nitrite inter-
mediate for the Fast SCR reaction but ruled out a route via nitrate
4 3
because of the well known stability of NH NO . However this
4
NO
3
build-up. As a
was observed and no NH was
conclusion does not consider the possibility of a reaction between
NH NO and NO, which in this work was demonstrated to be
2
3
4
3
2
active over a V/W/TiO catalyst already at low temperature.
NH NO
4
3
1 NO A NO
2
1 N
2
2
1 2 H O
(6)
Notes and references
Both the schemes above mentioned suggest that at T¡ 170 uC
reactions (3) and (4) occur sequentially, with reaction (3)
significantly faster than (4).
{
The small increase in the NO concentration level could be explained by
the influence of NH concentration on the equilibrium (4.a): during phase
C the increment of the ammonia concentration shifts the equilibrium to the
left causing a decrease of the available HNO , inhibiting reaction (4.b).
3
While the data in Fig. 1 agree with the assumption of nitrate
intermediates for the Fast SCR reaction (2), one could still
speculate that the behavior observed in phase D of Fig. 1 is
explained by the reaction of NO with deposited NH NO , reaction
3
1
A. Kato, S. Matsuda, T. Kamo, F. Nakajima, H. Kuroda and T. Narita,
J. Phys. Chem., 1981, 85, 4099.
4
3
(4), followed by an ‘‘independent’’ Fast SCR, reaction (2). For this
to apply, however, the Fast SCR, reaction (2), should be much
faster than reaction (4), otherwise a net production of NH and
3
2
3
M. Koebel, M. Elsener and G. Madia, Ind. Eng. Chem. Res., 2001, 40, 52.
G. Madia, M. Koebel, M. Elsener and A. Wokaun, Ind. Eng. Chem. Res.,
2002, 41, 3512.
NO
2
should be detected.
4 G. Madia, M. Koebel, M. Elsener and A. Wokaun, Ind. Eng. Chem. Res.,
2002, 41, 4008.
In order to compare the rates of reactions (2), (3) and (4), more
transient runs were performed in the simultaneous presence of
NH , NO and NO . Fig. 2 presents a typical experiment where
5
M. Koebel, G. Madia, F. Raimondi and A. Wokaun, J. Catal., 2002, 209,
59.
1
3
2
6
7
M. Koebel, G. Madia and M. Elsener, Catal. Today, 2002, 73, 239.
C. Ciardelli, I. Nova, E. Tronconi, B. Konrad, D. Chatterjee, K. Ecke and
M. Weibel, Chem. Eng. Sci., 2004, DOI: 10.1016/j.ces.2004.07.016.
Q. Sun, Z. Gao, B. Wen and W. M. H. Sachtler, Catal. Lett., 2002, 78, 1.
1000 ppm of NH
3
were fed to the reactor at 170 uC (phase A), then
500 ppm of NO and 500 ppm of NO were added to the feed
2
stream (phase B). At the end of phase B 450 ppm of NH
3
, 110 ppm
8
of NO, 340 ppm of NO were consumed producing 330 ppm of N .
2
These levels are in agreement with the simultaneous occurrence of
9 Y. H. Yeom, B. Wem, W. M. H. Sachtler and E. Weitz, J. Phys. Chem. B,
2004, 108, 5386.
2
C h e m . C o m m u n . , 2 0 0 4 , 2 7 1 8 – 2 7 1 9
2 7 1 9