New insights into the origin of NO2 in the mechanism of the selective catalytic reduction of NO by propene over alumina

Article abstract of DOI:10.1039/a808465c

The formation of NO₂ as a reaction intermediate in the selective catalytic reduction of NO by C₃H₆ in the presence of a large excess of O₂ over alumina is shown not to occur through the direct oxidation of NO by O₂.

Full text of DOI:10.1039/a808465c

New insights into the origin of NO2 in the mechanism of the selective catalytic
reduction of NO by propene over alumina
Frederic C. Meunier,* John P. Breen and Julian R. H. Ross
Centre for Environmental Research, University of Limerick, Limerick, Ireland. E-mail: frederic.meunier@ul.ie
Received (in Cambridge, UK) 2nd November 1998, Accepted 24th December 1998
The formation of NO2 as a reaction intermediate in the
selective catalytic reduction of NO by C3H6 in the presence
of a large excess of O2 over alumina is shown not to occur
through the direct oxidation of NO by O2.
rate). Fig. 3(a) shows the varying molar ratios of NO2/NO
obtained from the results of the experiments shown in Fig. 2.
The NO2/NO ratio at the thermodynamic equilibrium of
reaction (1) at 813 K calculated using the initial concentration
values of NO and O2 is shown in Fig. 3(b). The conversion of
propene [Fig. 2(a)] increased with increasing W/F values,
reaching 100% conversion at W/F = 180 mg s cm23. The yield
of N2 [Fig. 2(b)] followed a similar trend to that of propene
conversion, increasing steadily with increasing W/F values and
then levelling off when all the propene was converted, i.e. at
Selective catalytic reduction (SCR) of NOx by hydrocarbons has
received much attention as a possible means to control the
emissions from diesel and lean-burn engines. The work by
Burch et al. has unveiled many of the features of the NO
reduction over Pt-based materials.1,2 Yet, the mechanisms of
SCR reactions over alumina, one of the most active metal
oxides for the SCR of NO by hydrocarbons,3 and alumina
promoted by metal oxides such as CoOx remain unclear.4 It has
been suggested that when using propene as a reductant, the
initial step of the SCR reaction is the oxidation of NO by O2 to
form NO2 which in turn reacts with the reducing agent to form
N2.5–7 The results of the experiments reported in this paper
bring a new insight into the formation of NO2 over alumina
during the propene-SCR of NO, resulting in a necessary re-
evaluation of the current theories on the reaction mechanism.
The g-alumina used in these experiments was supplied by
Alcan (AA400, 150 m2 g21) and was calcined in air at 923 K for
6 h prior to use. The reactants (NO and C3H6) and products of
reaction (CO2, CO, NO2, N2O, NH3) were analysed by FT-IR
spectroscopy (7 m path length Foxboro® gas-cell fitted in a
Nicolet® 550 FT-IR). No hydrogen cyanide nor any other
products were detected in the experiments reported here. The N2
concentration was calculated during steady state conditions on
the basis of a 100% nitrogen balance on the other N-containing
products detected. A quartz microreactor (4 mm internal
diameter) was used for the catalytic tests, the catalytic bed being
held in place by quartz wool plugs. No reaction was observed
when the reactants were passed through the reactor in the
absence of a catalyst bed. The results reported here were
obtained at steady state conditions and were found to be
reproducible.
Fig. 1 (a) Experimental NO2 yield (5) obtained during the oxidation of NO
over g-Al2O3 and (b) NO2 yield limit (—) associated with the thermo-
1
dynamic equilibrium of the reaction NO + O2 Ô NO2 as function of
2
temperature. (c) Experimental NO yield (2) obtained during the decom-
1
position of NO2 to NO + O2 over g-Al2O3 and (d) NO yield limit (---)
2
1
associated with the thermodynamic equilibrium of the reaction NO + 2 O2
Ô NO2 as function of temperature. Feed: 500 ppm NO or NO2 + 2.5% O2
in Ar, W/F = 60 mg s cm23
.
Fig. 1(a) shows the NO2 yield obtained during the oxidation
of NO (500 ppm) by O2 (2.5%) over alumina and Fig. 1(b)
shows the calculated NO2 yield obtained at the thermodynamic
equilibrium of the reactions described by eqn. (1) as a function
of temperature:
1
NO + 2 O2 Ô NO2
(1)
As it has often been reported in the literature, the alumina
catalyst exhibited very poor activity for the NO oxidation
reaction with yields of NO2 attaining equilibrium conversions
only at temperatures above 900 K. Fig 1(c) shows the NO yield
1
obtained during the decomposition of NO2 to NO + O2 (the
2
reverse reaction of that discussed above) and Fig. 1(d) shows the
calculated NO yield obtained at the thermodynamic equilibrium
of the same reaction as a function of temperature. The activity
of the alumina for the backward reaction of eqn. (1) was
somewhat greater than that of the forward reaction but yet again
the equilibrium limit was only reached at a temperature of about
900 K.
Fig. 2 shows the propene conversion [Fig. 2(a)] and yields of
the different N-containing products of the reaction between NO
(500 ppm), propene (500 ppm), and O2 (2.5%) over alumina at
813 K and at varying W/Fs (ratio of catalyst weight to feed flow
Fig. 2 Selective catalytic reduction of NO using propene over g-Al2O3 at
813 K: (a) propene conversion (-), (b) N2 yield (8), (c) NO2 yield (5), (d)
NH3 yield (D) and (e) N2O yield (3) as a function of the W/F. Feed: 500
ppm NO + 500 ppm C3H6 + 2.5% O2 in Ar, 200 mg of alumina.
Chem. Commun., 1999, 259–260
259

[eqn. (1)]. A mechanism involving the direct reaction of NO, O2
and NO2 could not exceed thermodynamic limits. Therefore, the
current model of the alkene-SCR reaction over alumina-based
catalysts needs to be re-evaluated.5–7
At least two alternative solutions could be proposed to
explain the formation of the high concentrations of NO2
observed. First, a mechanism involving consecutive steps could
occur in which a nitro or a nitrite compound, e.g. nitromethane
CH3NO2, is formed along with CO (or CO2) from the reaction
of propene, O2 and NO [eqn. (2-1)]. Organic nitro or nitrite
species have often been quoted as possible intermediates of
SCR reactions.8 The subsequent oxidation of this organo-NOx
species would yield NO2 [eqn. (2-2)]
3
C3H6 + 2 O2 + 2 NO ? 2 CH3NO2 + CO
(2-1)
(2-2)
7
3
CH3NO2 + 4 O2 ? NO2 + CO2 + 2 H2O
Overall, eqn. (2-1) and (2-2) combine to give:
Fig. 3 Selective catalytic reduction of NO using propene over g-Al2O3 at
813 K as a function of the W/F: NO2 to NO ratio (a) experimental (2) and
C3H6 + 5 O2 + 2 NO ? 2 NO2 + 2 CO2 + CO + 3 H2O (2-3)
1
(b) at the thermodynamic equilibrium of the reaction NO + O2 Ô NO2
2
This mechanism is supported by the fact that thermodynamic
calculations show that the reactions given by eqn. (2-1) and
(2-2) are strongly exergonic over the temperature range
investigated here (e.g. DG°813K = 2298 kJ mol21 and 2734 kJ
mol21, respectively). Regarding a second possible mechanism,
the oxygen needed to form the NO2 could be supplied by
another molecule of NO, i.e. in a disproportionation reaction,
and not by the O2. An example of mechanism of SCR of NO
involving a disproportionation of NO to N2 and NO2 was
proposed by Smits and Iwasawa .9 These authors suggested that
one of the roles of the hydrocarbon, NO and O2 was to form a
radical molecule R* (e.g. 1-nitro-sec-propyl) which in turn
could propagate the disproportionation reaction:
(---). Feed: 500 ppm NO + 500 ppm C3H6 + 2.5% O2 in Ar, 200 mg of
alumina.
W/Fs greater than 180 mg s cm23. The striking feature of Fig.
2 is the sharp rise in the production of NO2 [Fig. 2(c)] as soon
as the conversion of propene reaches 100%. At propene
conversions lower than 95%, the presence of NO2 was not
detected (i.e. < 2.5 ppm). Upon completion of the conversion of
propene, the NO2 yield reached a value of approximately 25%
(corresponding to 127 ppm of NO2). With further increase in the
values of W/F, the NO2 yield gradually decreased. The gradual
decrease in the concentration of NO2 was accompanied by a
similar increase in the concentration of NO (not shown in Fig.
2) while the concentrations of all the other products of reaction
were either steady (N2) or decreased (N2O and NH3). Inter-
estingly, Fig. 3 shows that in the region of 100% propene
conversion (W/F ! 180 mg s cm23), the experimental values of
the molar ratios of NO2/NO were greater than 1. Hence, these
values were more than 10 times greater than the value of the
thermodynamic equilibrium ratio of NO2/NO (i.e. 0.09) calcu-
lated at the same temperature. The same results were obtained
whether the W/F was varied with increasing or decreasing
values.
Further experiments carried out using the same reactant
concentrations, at constant W/F (60 mg s cm23) and varying
temperature also showed that, at the temperature of 100%
conversion of propene, there was a sharp increase in the yield of
NO2. At 843 K, propene conversion was not complete and NO2
was not detected; however, at 873 K propene was fully
converted and NO2 yields of 30% were obtained (this value
again being well in excess of levels obtained from equilibrium
calculations). The sharp increase in the NO2/NO ratio at the
temperature of 100% conversion of propene was the same
regardless of whether the reaction temperatures were increased
or decreased.
4 NO (+R*) Ô N2 + 2 NO2 (+R*)
(3)
The thermodynamics of the disproportionation reaction de-
scribed by eqn. (3) essentially favour the right hand-side of the
equation at temperatures below 600 °C (and are independent of
the concentration of O2) and therefore this mechanism could be
responsible for the high yields of NO2 observed. Hence, one of
the possible role of the alumina (a known mild oxidiser) during
alkene-SCR reactions would be to allow the formation of
organic-nitro or nitrite species [eqn. (2)] and/or the formation of
a hydrocarbon-based propagator [eqn. (3)]. Experiments using
N18O and 16O2 are planned and should give some evidence of
the origin of the oxygen in NO2 (i.e. by observing either
N18O18O or N18O16O) to determine which of the two models
proposed is the most likely.
The conclusion of this work is that the formation of NO2
during the propene-SCR of NO over alumina is not obtained by
direct oxidation of NO by O2 but that a more complex
mechanism has to be considered.
Part of this work was funded by the European Community,
through the Environment and Climate Programme, Contracts
E5V5-CT94 and ENV4-CT97-0658.
It is evident from the results of experiments described in this
paper that NO2 could not be observed when propene was
present in the product streams. This can be explained by the
work of a number of different authors,4,6,7 who have shown that
alumina is very active for the reduction of NO2 by propene to
form N2. However, upon complete combustion of the propene,
the NO2 could not be reduced and was therefore detected in the
reactor effluent. Fig. 2 and 3 show that the maximum NO2/NO
ratio occurred at a W/F value of 200 mg s cm23, and at higher
W/Fs the ratio decreased towards the equilibrium ratio. This was
probably due to the slow decomposition of the NO2 to NO and
O2 (in the absence of propene) which Fig. 1(c) shows can occur
over the alumina catalyst. Surprisingly, the catalytic data
reported on alumina in this paper and the corresponding
thermodynamic calculations show that most of the NO2
observed upon completion of propene conversion during the
SCR of NO did not arise from the direct oxidation of NO by O2
Notes and references
1 R. Burch and T. C. Watling, Catal. Lett., 1997, 43,19.
2 R. Burch, P. Fornasiero and T. C. Watling, J. Catal., 1998, 176, 204.
3 H. Hamada, Catal. Today, 1994, 22, 21.
4 J. Yan, M. C. Kung, W. M. H. Sachtler and H. H. Kung, J. Catal., 1997,
172, 178.
5 H. Hamada, Y Kintaichi, M. Sasaki and T. Ito, Appl. Catal., 1991, 70,
L15.
6 H. Hamada, Y. Kintaichi, M. Inaba, M. Tabata, T. Yoshinari and H.
Tsuchida, Catal. Today, 1996, 29, 53.
7 N. Okazaki, S. Osada and A. Tada, Appl. Surf. Sci., 1997, 121/122,
396.
8 C. Yokoyama and M. Misono, J. Catal., 1994, 150, 9.
9 R. H. H. Smits and Y. Iwasawa, Appl. Catal. B, 1995, 6, L201.
Communication 8/08465C
260
Chem. Commun., 1999, 259–260