Catalysis Z’oduy, 4 (1989) 173-185
`
`173
`
`REDUCTION OF NO, OVER ACID-LEACHED MORDENITE CATALYSTS
`
`Lars A.H. Andersson, Jan G.M. Brandin and C.U. Ingemar Odenbrand
`
`Department of Chemical Technology, Chemical Center, Lund University,
`
`Institute of Science and Technology, P.O. Box 124, S-221 00 Lund, (Sweden)
`
`ABSTRACT
`
`Selective catalytic reductions of NO, NO2 and mixtures of NO and NO2 over
`mordeninte catalysts were studied. The activity of mordenite catalysts with
`different Si/Al ratios, obtained through acid leaching, decreased with the Al
`content of the mordenite. The change in activity with temperature and acid
`leaching together with the changes in contents of Fe and Al indicate that
`Lewis acids are active sites. These Lewis acids could be either Fe ions or
`Lewis acids formed on dehydroxylation of Broensted acid sites. Activities of
`NO+reduction on leached mordenites were correlated to the amount of adsorbed
`NO measured by
`The activity in the reduction of NO revealed a maximal
`conversion at a NO /NO ratio of 0.5, indicating that t8e oxidation of NO or
`the decomposition gf Nlf, are the rate limiting step in the overall reduction.
`
`INTRODUCTION
`
`Mordenite, a zeolite with a high Si/Al ratio, is well suited for SCR reduc-
`
`tion of NOx (1,2). It has a structure of parallel main channels, from which
`
`side pockets lead to oval small channels parallel to the main ones. The
`
`crystal structure resists hydro-thermal breakdown as well as acid environ-
`
`ment. These are important properties of mordenite when used as a catalyst in
`
`flue gas treatment.
`
`Four coordinated aluminium to oxygen in the zeolite lattice creates a
`
`negative charge in the structure. The charge is balanced by counter ions,
`
`located in channels and pockets of the structure.
`
`Active centers of the catalyst are believed to be acidic centers
`
`(Broensted- and/or Lewis acid sites) on the internal surface. The chemical
`
`character of the counter ions affects catalytic acitivty. H-substituted
`
`mordenite exchanged with transition metal ions such as Cu or Fe has shown
`
`good catalytic properties in SCR of NOx with ammonia (2).
`
`Investigation of the influence of N02/NOx ratio on SCR with ammonia over
`
`V205/Ti02-Si02 has shown that the activity of NO, reduction is favoured by
`
`coexisting NO and NO2 in the reaction system. Our resuls show, that the ac-
`
`tivity is strongly enhanced by a ratio of 0.5 especially in the temperature
`
`region of 470-570 K. This information has been used in a patent claim (3).
`
`Shigeaki et al. (4) and Tuenter et al (5) have reported this effect in
`
`Exhibit 2025.001
`
`Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands
`SELECTIVE CATALYTIC
`IR.
`

`
`systems with N02/NOx ratios equal or greater than 0.5. According to Tuenter
`
`et al. NOx reduction N02/NOx ratio of 0.5 the activity was much higher
`
`compared to NO reduction over an industrial V205-W03-Ti02 catalyst.
`
`Compared to a V205/Si02-Ti02 catalyst, mordentie shows one order of magni-
`
`tude lower rate of reaction for NO reduction with ammonia (2,6). The differ-
`
`ence in catalytic activity increases as the temperature is decreased from 600
`
`to 470 K. In oxidation of NO H-mordenite has a higher catalytic acitivity
`
`than the compared catalyst (7,8).
`
`It is possible, that the relatively poor catalytic effect of H-mordenite in
`
`NO reduction with ammonia is caused by competitive adsorption between water
`
`and one of the nitrogen oxides. Mizumoto et al. have shown these effects in
`
`the reduction of NO with ammonia over Cu(II)NaY (9).
`
`The purpose of this study is to examine'the influence of the aluminium
`
`content in H-mordenite, changed by acid leaching, on the catalytic reduction
`
`of NO, NO2 and NOx with NH3.
`
`EXPERIMENTAL
`
`Catalyst
`
`The catalyst, a commercial H-mordenite "Zeolon 900 H" from Norton Ltd
`
`Zeolon 900 H, was leached by hydrochloric acid using two different methods.
`
`the first method different extents of leaching were achieved by varying
`
`the acid concentration indexed SLx (x=molar cone).
`
`the seconed 2M HCl was
`
`used to leach the catalyst for 1,2 or 3 periods of two hours each (indexed
`
`SLEx x=periods). These treatments gave seven catalysts with different Si/Al
`
`ratios.
`
`TABLE 1
`
`Catalyst characterization
`
`Catalyst
`
`Si
`
`Al
`
`Fe
`
`d
`
`SBET
`
`Vpor
`
`(m2 g-l) (cm3 g-I)
`(Weight % hydrous)
`(nm)
`____________________~~~~~~~~~~~____________~~~______~~~__~~~~~
`
`Z900H
`
`SLl
`SL5
`
`SLconc
`
`SLEl
`
`SLEP
`
`SLE3
`
`31.0
`
`32.2
`31.3
`
`32.0
`
`32.2
`
`32.4
`
`34.0
`
`5.58
`
`5.06
`4.64
`
`4.64
`
`4.56
`
`3.97
`
`3.31
`
`0.50
`
`0.28
`0.25
`
`0.23
`
`0.23
`
`0.16
`
`0.09
`
`0.19489
`
`0.19477
`0.19466
`
`0.19471
`
`0.19469
`
`0.19447
`
`0.19415
`
`472.1
`
`493.0
`____-
`
`0.207
`
`0.211
`_____
`
`_____
`
`_____
`
`505.7
`
`528.7
`
`0.223
`
`0.239
`
`Exhibit 2025.002
`
`In
`In
`_____ _____
`

`
`175
`
`The catalysts were characterized by XRF, AAS, X-ray diffraction analysis,
`
`BET N2 adsorption and NH3 adsorption. A thorough description of catalyst
`
`preparation and characterization methods can be found in (8). A summary of
`
`catalysts and characteristics is presented in Table 1.
`
`Method and reaction conditions
`
`The catalytic properties of the mordenite samples were tested in a sta-
`
`tionary flow reactor. The procedure and the equipment have been described in
`
`a previous article (6). To avoid formation of NH4N03 and reaction of NH3 in
`
`the NOx-converter of the analysing equipment the sample flows were scrubbed
`
`to remove NH3 (7). The reactions
`
`which were investigated, were reduction
`
`with ammonia of NO, NO2 and mixtures of NO and NO2 at different NO/NOx
`
`ratios. The conversions were measured as a function of temperature or NO/NOx
`
`ratio.
`
`The flue gas was simulated by mixing the components, measured by flowmeters
`
`from gas cylinders with each component in N2 at known concentration. NO and
`
`total NOx were analysed by a Beckman 955 chemiluminisence instrument before
`
`and after passage through the reactor. The catalyst samples (0.1-0.5 g with a
`
`particle size 0.71-0.85 mm) were subject to a gas load of 50-60 1 (NTP) h-l
`
`at a total pressure of 200 kPa unless otherwise stated. The reaction condi-
`
`tions were varied according to Table 2.
`
`The effects of external mass transfer and channelling were shown not to
`
`influence the reaction rates of reduction of NO over a V205/Si02-Ti02 cata-
`
`lyst (6). These experiments were performed at higher rates of reduction and
`
`smaller amounts of catalyst were used than in the case of reduction over
`
`mordenites. Consequently the influence of these effects is negligible in the
`
`reduction of NOx over mordenites.
`
`TABLE 2
`
`Reaction conditions
`
`NO, Red
`NO2 Red
`NO Red
`______________________~~~~~~~~__~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`
`NO (wm)
`NO2 (rvm)
`NH3 (wm)
`O2
`(%)
`TEMP (K)
`N02/NOx
`(g)
`Catalyst
`S.V.
`(cm3 g -1 h-1)
`
`600
`110
`__
`__
`800
`150
`2
`2
`550-620 420-670
`0
`0
`0.1
`0.5
`500,000
`100,000
`
`-_
`100
`150
`2
`390-670
`1
`0.1
`500,000
`
`330
`45
`270
`50
`800
`150
`2
`2
`550-650 410-670
`o-1
`0.45
`0.1
`0.5
`500,000 100,000
`
`Exhibit 2025.003
`
`,
`

`
`176
`
`A net production of NO, was observed during preheating of N02+NH3+02 in N2
`
`at temperatures exceeding 590 K. Tests with an empty reactor resulted in
`
`production of total NO, amounting to 10% at 670 K. At the same time a change
`
`in N02/NO, ratio from 1 to 0.8 was observed at the reactor inlet. It is be-
`
`lieved, that partial reduction of NO2 by ammonia produces NO.
`
`5N02 + 2NH3 +7NO
`
`+ 3H20
`
`This reaction has previously been reported for reduction of NO2 in the
`
`same experimental setup (7). The change in N02/NOx ratio at the reactor inlet
`
`influences the overall rate only at temperatures higher than 620 K.
`
`The NO2 reduction experiments exhibited a hysteresis effect when they were
`
`performed at increasing or decreasing temperatures. Lower conversions were
`
`obtained, when data points were taken at increasing temperatures compared to
`
`points measured at decreasing temperatures. The hysteresis appeared below
`
`550 K. The reason for this behaviour is believed to be the formation of
`
`unstable ammonium nitrate.
`
`Mass transfer limitaions
`
`order to evaluate the extent of intraparticle mass transfer, NO-reduc-
`
`tion was performed over ZgOOH at constant W/FNo, but at two concentrations of
`
`NO (570 ppm and
`
`ppm). NO conversions at 650 K were 95 and 62% at the
`
`first and the second concentration levels respectively.
`
`the reaction order
`
`is lower than 1, which is the order of mass transfer rate, a decreased
`
`reactant concentration will give an increased mass transfer influence on the
`
`overall reaction rate (robs). An Arrhenius plot of the reaction rate at the
`
`two concentration levels with the rate constant (kobs) derived from the
`
`expression for a first order reaction in an integral reactor, gives an
`
`overall activation energy (Eobs).
`
`At mass transfer limited reaction conditions the effectiveness factor (n)
`
`approaches the value of the inverse Thieles module (for n<O.l). Under these
`
`circumstances the intrinsic activation energy (Eintr) can be derived, if n in
`
`the over all reaction rate expression is substituted by the inverse Thieles
`
`module (11).
`
`The influence of the activation energies of diffusion (Ediff), and of
`
`reaction (Eintr) on Eobs is then given by:
`
`E
`intr
`
`= 2*E
`obswEdiff
`
`Exhibit 2025.004
`
`In
`110
`If
`

`
`At the lower concentration level (110 ppm NOx and 150 ppm NH3) n is assumed
`
`to be less than
`
`CH4 in mordenite (12), 7 kJ mole-l, is used
`
`0.1. If Ediff of
`
`as an approximation for Ediff for NOx, the values for NO, NO2 and NOx reduc-
`
`tions presented in Table 3 are obtained.
`
`TABLE 3
`
`Reduction
`
`Rate
`
`E
`
`E.
`
`E
`
`Temperature
`
`(mole g-lh-') (kJ :88e-l)
`
`(kJ :i:e'-l)
`
`NO
`
`NO2
`NOx(*)
`
`3.6*10-7
`
`4.4*10-7
`
`31+2
`
`41+7
`
`23+4
`
`55+4
`75+14
`
`39+8
`
`5823
`____
`
`2721
`
`500-600
`
`530-620
`
`500-600
`
`*N02/NOx=0.45
`
`No result of NO2 reduction over 0.5 g Z900H is presented since 600 ppm
`
`concentration of NO2 caused formation of NH4N03 in the system. In reduction
`
`of NO the calculated Eintr is close to the value of Eobs from reaction over
`
`0.5 g Z900H. This indicates that the influence of mass transfer is negligible
`
`in reductions of 600 ppm NO over 0.5 g Z900H and that the assumption of <O.l
`
`is valid. As the rates of reduction for NO2 and NO, are faster than that of
`
`NO, n should be smaller. Therefore the calculated Eintr for reduction of NO2
`
`and NOx should be proper estimations of the true activation energies as well.
`
`RESULTS
`
`The reduction of NO
`
`Results presented here were obtained at experimental conditions as stated
`
`above as level 1 (0.5 g catalyst, 600 vppm NO, 750 vppm NH3 and 2% O2 in N2).
`
`Figure 1 shows the influence of temperature on the conversion of NO. The
`
`acitivity increases with temperature but decreases with increasing si/Al
`
`ratio.
`
`Exhibit 2025.005
`
`. .
`(kJ %eml)
`(K)
`1 .o*10-7
`

`
`178
`
`Figure 1 NO-conversion versus temperature: Gas load 100,000 Ncm" g-l h-l
`
`Pressure 200 kPa over 0.5 g catalyst; NO 600 ppm, NH3 800 ppm O2 2 % N2.
`
`Figure 2 Polynomial fits of Arrhenius plots of the observed rate constant
`
`for the reduction of NO: Gas load 100,000 Ncm3 g-'h-l; Pressure 200 kPa over
`
`0.5 g catalyst; NO 600 ppm, NH, 800 ppm, 0, 2 % in N,; Temperature 495-600 K.
`L
`L
`
`Activation energies 5823
`
`moye-'.
`
`0 05 I AkORSAN:: 2 3 ” RELATIVE NO?ONVE&lON &ID. (%y
`
`Figure 3
`
`Figure 4
`
`Figure 3 Reduction rate constant ratios versus IR absorbance ratios at 2165
`11
`cm
`
`for NO with values of catalyst SLcons as standard: Gas load 100,000
`
`Ncm3 g-'h-l; Pressure 200 kPa over 0.5 catalyst; NO 600 ppm, NH3 800 ppm,
`
`O2 2 % in N2, Temperature 640 K.
`
`-1 -1;
`h
`Figure 4 NO reduction versus NO oxidation: Gas load 100,000 Ncm' g
`
`Pressure 200 kPa over 0.5 g catalyst; NO 600 ppm, NH3 800 ppm, O2 2 % in N2.
`
`Exhibit 2025.006
`
`kJ
`

`
`Arrhenius plots of experiments are presented in Fig. 2. In this case kobs
`
`is derived from a 0.6 order rate equation, since diffusion is found to be
`
`negligible. Deviations of the fits from experimental data were less than 5 %.
`
`The plots are parallel within experimental errors and the intercepts
`
`decrease with increasing leaching except for the most leached sample SLE3.
`
`The Arrhenius plots indicate a quantitative change in the number of active
`
`sites, and that qualitative changes are small for the different catalysts.
`In infrared studies of adsorbed NO on leached mordenites published in the
`work on NO oxidation (8) the IR peak at 2165 cm -' was assigned to formed NO+
`species on the catalyst surface. In Figure 3 the rate constants of reduction
`
`(first order reaction rate) are plotted against the absorbance at 2165 cm-'
`
`for mordenites with different Si/Al ratios. Absorbances and rate constants
`
`are shown as ratios relative to the values of SLconc. The NO reductions were
`perfomed at 650 K, and the IR measurements were done at room temperature. The
`
`correlation coefficient between absorbance and rate in Figure 3 is 0.95.
`
`Figure 4 is a plot of the reduction of NO versus the oxidation of NO for
`
`different mordenites at three temperatures. The conversion for reduction is
`
`lower than that of oxidation for all catalysts but Z900H at 550 and 595 K. At
`
`640 K the correlation between the reduction and the oxidation is good, as the
`
`value of conversion for oxidation approaches half the value of that for
`
`reduction.
`
`The reduction of NO2
`
`Due to the mentioned formation of ammonium nitrate, the evaluation of NO2
`
`reduction will be limited to data from experimental conditions of 0.1 g
`
`catalyst 100 ppm N02, 150 ppm NH3 and 2% 02. The data points were taken at
`
`descending temperatures.
`
`The influence of temperature on NO2 conversions is shown in Figure 5. The
`
`degree of leaching has no significant effect on conversion except for SLE3 in
`
`the temperature range of 520 to 620 K. At temperatures exceeding 620 K homo-
`
`genous or wall catalysed partial reduction of NO2 to NO (7) makes an evalua-
`
`tion of experimental data uncertain. At low temperatures (450-490 K) there is
`
`a weak trend of activity increasing with degree of leaching. Figure 6 shows
`Arrhenius plots, assuming an overall first order reaction (If the overall
`reaction is heavily mass transfer limited, the first order mass transfer will
`
`be dominating). Deviation of the fits from the experimental data is less than
`6 %. In the interval 530-620 K the average Eobs is 412 7 kJ mole -1
`
`Exhibit 2025.007
`
`179
`.
`

`
`180
`
`-1 -1
`h
`Figure 5 NO2 conversion versus temperature: Gas load 500,000 Ncm3 g
`
`Figure 5
`
`Figure 6
`
`Pressure 200 kPa over 0.1 g catalyst; NO2 110 ppm, NH3 150 ppm, 02 2 % in N2.
`
`Fig. 6 Polynomial fits of Arrhenius plots of the observed rate constant for
`the reduction of NO?: Gas load 500,000 Ncm3 g -' h -I; Pressure 200 kPa over 0.1
`
`g catalyst; NO2 IlO-ppm, NH3 150 ppm, O2 2 % in N2.
`
`Figure 7
`
`15 1.6 I/-P 100: 19
`Figure 8
`
`Figure 7 NO, conversion versus temperature at N02/N0, ratio 0.45: Gas load
`100,000 Ncm3 g-lb-l; Pressure 200 kPa over 0.5 catalyst; NO, 600 ppm, NH3
`
`800 ppm, O2 2 % in N2.
`
`Figure 8 Arrhenius plots of the observed rate constant for the reduction of
`NOx at N02/NOx ratio of 0.45: Gas load 100,000 Ncm3 g -' h -l; Pressure 200 kPa
`over 0.5 g catalyst; NO, 600 ppm, NH3 800 ppm, O2 2 % in N2; Temperature
`
`505-595 K.
`
`Exhibit 2025.008
`
`‘O” TEMP&ATURE~) ‘O”
`;
`

`
`The reduction of NOx
`
`Data points were taken from NOx reductions as a function of temperature at
`
`a constant N02/NOx ratio of 0.45 and over 0.5 g of Z900H, SLEZ, and SLE3.
`
`Fig. 7 shows the conversion of NO, as a function of temperature and Fig. 8 is
`
`an Arrhenius plot of the same data.
`
`The activity is decreasing, as the Si/Al ratio increases. (Fig.7 and 8).
`
`The.change in slopes of In(k) versus l/T in Fig.8 is probably caused by an
`
`increasing effectiveness factor, as the total amount of active sites de-
`
`creases with increasing leaching. The activation energies were:
`
`E
`obsZ900H
`
`EobsSLE2
`E
`obsSLE3
`
`= 26.7 kJ/mole
`
`= 30.9 kJ/mole
`
`= 32.2 kJ/mole
`
`The NO, reductions at varying N02/NOx ratio were investigated at 585 K.
`
`This was done with 110 ppm on all catalyst samples in order to prevent
`
`ammonium nitrate formation at high N02/NOx ratios.(Fig. 9). The conversions
`
`of NOx versus the N02NOx ration in Fig.9 were valcano shaped. The rate for
`
`the reduction of NOx at N02/NOx=0.45 was higher than both the reductions of
`
`NO and of N02. The ratios between apparent rates for SLEl were:
`
`rNO:rNO :rNO =1:3.6:4.1.
`
`Figure 9 NOx conversion versus NO2 fraction; Gas load 500,000 Ncm3 g'lh-l;
`
`Pressure 200 kPa over 0.1 g catalyst; NO, 110 ppm, NH3 150 ppm, O2 2 % in N2;
`Temperature 585 K.
`
`DISCUSSION
`
`The reduction of NO
`
`According to the NOx reduction mechanism proposed by Brandin et al (10) the
`
`active sites are Lewis acids. Lewis acids in mordenite are either dehydroxy-
`
`lated Broensted acid sites and/or metallic counter ions having a charge
`
`Exhibit 2025.009
`
`181
`

`
`exceeding +l. If one of these sites alone is catalyticly dominating, the
`
`observed activity would depend strongly on either Fe or Al concentrations in
`
`the catalysts.
`
`A correlation between the change in conversion of NO with Fe or Al content
`
`in the catalysts was made at 570 and 650 K (data from Table 1 and Fig.1) The
`
`correlation coefficients were 0.88 for Fe and 0.86 for Al at 570 K; and at
`
`650 K they were 0.18 and 0.96. The correlation is fair for both Fe and Al at
`
`570 K. At 650 K the fit to the Al content is good, but the correlation to the
`
`Fe content is lost. The change in correlation with change in temperature are
`
`in opposite directions for Fe and Al contents. An explanation for this change
`
`could be, that Fe counter ions are Lewis acids independent of temperature,
`
`and that they constitute the dominating part of Lewis acids at low tempera-
`
`tures. The remaining part of Lewis acids is formed from dehydroxylated
`
`Broensted acid sites in the vincinity of aluminium atoms in the mordentie
`
`structure. The dehydroxylation is increased with temperature (10). Thus the
`
`change in correlation with temperature is explained.
`
`The effect of exchanged counter ions at low temperatures is pronounced in
`2+
`NO reduction over Cu
`
`exchanged 2900H. A 30 % Cu exchanged Z900H increased
`
`the activity one order of magnitude at 470 K compared to Z900H, while the
`
`activities were the same at 650 K (10). This is the same effect as observed
`
`for Fe counter ions mentioned above.
`
`The correlation between the absorbance of NO and the rate of reduction for
`
`NO in Fig. 3 supports the idea, that NO is adsorbed on Lewis acid sites.
`
`The method used for evaluation of mass transfer influence in this work
`
`results in values of the activation energies, but it excludes an evaluation
`
`of the pre-exponential factors since low effectiveness factors induce large
`
`errors in the calculation of the prexponential factor. The activation
`
`energies obtained will contain all temperature dependent properties of the
`
`rate limiting reaction step, such as diffusion, adsorption and reaction.
`
`key step of the reduction of NO is assumed to be the oxidation of
`
`NO, and that reactants adsorb competitively on the catalyst surface, as
`
`proposed by Mizomoto et al. (9), the rate expression would be of Langmuir
`
`Hinshelwood type.
`
`The rate expression would consist of:
`
`1. A rate constant
`
`2. An adsorption expression with the competitively adsorbed species in
`
`the numerator
`
`3. An active site constant.
`
`Exhibit 2025.010
`
`182
`If the
`

`
`183
`
`The adsorption constants of ammonia and water will disappear from this
`
`expression in a pure oxidation (without ammonia and water).
`
`The activation energy for the oxidation, as reported previously (8), was
`
`46 kJ mole-' at 470-530 K. Ered at the same conditions, but at 540-620 K, was
`
`58 kJ mole-l. The adsorption coefficients for ammonia and water are not
`
`present in the evaluation of the rate constant for the NO oxidation. The
`
`temperature regions also differ. As the amount of active sites (Lewis acids)
`
`increases with temperature, the catalytic activity will increase with tem-
`
`perature. This will yield a higher activation energy in the reduction of NO
`
`than in the oxidation of NO.
`
`The activities of both the oxidation and the reduction of NO in Fig. 4 are
`
`changed regularly with the degree of leaching. At 550 and 595 K the rate of
`
`reduction is slower than rate of oxidation for all catalysts but Z900H. The
`
`slow reduction rate is probably due to competitive adsorption of H20 and/or
`
`NH3. The influence of competitive adsorption decreases with increased tem-
`
`perature, which yields a better correlation between the reduction and the
`
`oxidation conversions at high temperature Fig. 4.
`
`The reduction of NO2
`
`An explanation for the increased activity with leaching at low temperatures
`
`could be, that NH3 is adsorbed on Broensted acid sites and combine with NO2
`
`forming a surface bound NH4N03
`
`which decomposes to N20 (Fig. 5). The propor-
`
`tion of Broensted acids is increased, along with the acid leaching of the
`
`catalyst because Fe ions are substituted with protons.
`
`the NO2 recuduction the decomposition of NO2 should be the rate limiting
`
`step according to the proposed mechansism (10). The activation energy of
`
`decomposition is 88 kJ mole-'
`
`at 0.5 g Z900H and 600 ppm NO2 (8). The
`
`observed EN02red is 41: 7
`
`mole -' at 0.1 g Z900H and 100 ppm N02. Assuming
`
`that n<O.l would give an Eintr = 75(f14) kJ mole -I. This is close to the
`
`experimental value of NO2 decomposition.
`
`The reduction of NOx
`
`The mass transfer limitation is coupled to the reaction rate according to
`
`Chatelier's principle. That is as intrinsic reaction rate increases, the
`
`system will respond with a decrease in mass transfer efficiency. This leads
`
`to a suppressed volcano shape as well as suppressed differences in activity
`
`between the leached catalysts (Fig.9).
`
`The fact that NOx reduction at N02/NOx ratios orther than zero or one was
`
`faster than NO or NO2 reduction indicates a reaction mechanism involvning
`
`Exhibit 2025.011
`
`,
`In
`kJ
`

`
`both species at the same time. The effect is more pronounced on the NO rich
`
`side in NO, conversion where the slope of the volcano curve is steeper.
`
`The activation energy for the oxidation of NO on Z900H is 46
`
`the temperature 470-530 K (8). ENOred is 55kJ mole-I
`-1
`mole
`
`and ENO+N02red is 27
`at 500-600 K. The order of activation energies are
`
`ENOxred
`
`<E
`
`NOox < ENOred
`
`compared to the order of reaction rates
`
`rNOox <
`
`'NOred < 'NOred'
`
`The different orders could be explained by the fact that NO reduction involve
`
`new species, NH3 and water, which can be competitively adsorbed on active
`
`sites. The presence of H20 have been reported to adsorb competitively with
`
`NO2 (13) on zeolites resulting in NO oxidation beyond thermodynamic equili-
`
`brium.
`
`CONCLUSIONS
`
`Conclusions on NO,reductions over acid leached mordenite
`
`1. The catalytic activity of mordenite decreases with increasing Si/Al
`
`atomic ratio in the reductions of NO and of NOx at N02/NOx = 0.45
`
`2. The decreased activity results in no change of activation energy, but a
`
`decrease in the preexponential factor. This indicates the change in
`
`activity is due to a change in the total amount of active sites along with
`
`the aluminium content.
`
`3. The activity can only be correlated to the aluminium content above 590 K.
`
`Higher temperatures give better correlations for Al and worse for Fe.
`
`This implies, that both the reduction of NO and NOx (N02/NOx = 0.45) are
`
`catalysed over Lewis acids. We propose that there are two types of Lewis
`
`acids, metallic counter ions with a charge exceeding +l, and dehydroxy-
`
`lated Broensted acids, that are active in catalytic reduction of NO and
`
`NO,.
`
`4. The rate for the reduction of NO, at N02/NOx = 0.45 is higher than both
`the reductions of NO and of N02. This indicates a reaction mechanism for
`the reduction that involves both NO and NO2 in equal amounts.
`
`Exhibit 2025.012
`
`184
`kJ mole-’ at
`kJ
`.
`

`
`ACKNOWLEDGEMENTS
`
`This Study was supported by The National Energy Administration of Sweden.
`
`We were kindly helped in catalyst preparation and characterization by Mr.
`
`M. Persson and Mrs. 8. Svensson. The IR-measurements were performed by Dr. B.
`
`Rebenstorf, Inorganic Chemistry 1, Lund University.
`
`REFERENCES
`
`1
`
`2
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`10
`
`11
`
`12
`13
`
`J.R. Kiovsky, P.B. Koradia and C.T. Lim, Ind. Eng. Prod. Res. Dev.,
`19 (1980) 218.
`P.MJ Hirsch, Environmental Progress, 1 (1982) 24.
`C.U.I. Odenbrand, J.G.M. Brandin, L.A.H. Andersson and S.T. Lundin,
`Swedish Patent Appl. nr 8701159-9, March 1987.
`Fuel Sot. of Japan (Nenryo Kyoaishi),
`K. Shigeaki, Y. Toru and S. Eiji,
`56 (1977) 834.
`G. Teunter, W.F. van Leeuwen and L.J.M. Snepvangers, Ind. Eng. Chem.
`Prod. Res., 25 (1986) 363.
`C.U.I. Odenbrand, S.T. Lundin and L.A.H. Andersson, Applied Catalysis, 18
`(1985) 335.
`C.U.I. Odenbrand, L.A.H. Andersson, J.G.M. Brandin and S.T. Lundin,
`Applied Catalysis, 27 (1986) 363.
`C.U.I. Odenbrand, L.A.H. Andersson and J.G.M. Brandin, Catal. Today,
`4 (1989) 155.
`M. Mizomoto, N. Yamazoe and T. Seiyama, J. Catal., 59 (1979) 319.
`J.G.M. Brandin, C.U.I. Odenbrand and L.A.H. Andersson, Catal. Today,
`4 (1989) 187.
`C.N. Satterfield, Mass Transfer in Heterogeneous Catalysis, M.I.T.Press,
`London, 1970, p. 129.
`C.N. Satterfield and A.J. Fabretti Jr., A.I.Ch.E.J., 13 (1967) 731.
`H. Arai, H. Tominaga and J. Tsuchiya, in G.C. Bond, P.B. Wells and F.C.
`Tompkins (Editors), Proc. Sixth Int. Congr. Catal. 1976, Chem. Sot.,
`Letchworth, 1977, p. 997.
`
`Exhibit 2025.013
`
`185