Arden B. Walters

NOx decomposition and reduction by methane over La2O3

Abstract Nitric oxide reduction by methane was conducted in a quartz microreactor from 773 to 973 K over La2O3, a good methane oxidative coupling catalyst. La2O3 is a better catalyst than MgO and Li/MgO for this reaction because: (1) it has a much higher specific activity, (2) the presence of oxygen enhances the rate of reduction, and (3) it gives a selectivity to nitrogen that is close to 100%. Both Li/MgO and La2O3 catalyst systems have rates that increase continuously with temperature and exhibit no bend over. The apparent activation energy for nitric oxide reduction by methane over La2O3 was 24.4 kcal/mol (26.0 kcal/mol with oxygen present), and the reaction orders in CH4, NO and O2 were 0.26, 0.98 and 0.50, respectively. Unlike Li/MgO, La2O3 is also active for direct nitric oxide decomposition, but the activity is noticeably lower than that for nitric oxide reduction by methane. Both the direct decomposition of nitrous oxide and nitrogen dioxide and their reduction by methane were also studied on La2O3. The rate of direct nitrous oxide decomposition was quite high and essentially unaltered by the presence of methane; in contrast, the rate of nitrogen dioxide decomposition was very low. However, the rate of nitrogen dioxide reduction to nitrogen was greatly enhanced by methane, and the rate became comparable to that for nitric oxide reduction by methane in the presence of 1.0% O2. Methane oxidative coupling over this La2O3 catalyst was verified using oxygen and nitrous oxide as oxidants, whereas little or no coupling occurred when nitric oxide or nitrogen dioxide was used. At this time, the high NOx reductive activity of La2O3 and other coupling catalysts is attributed to the formation of surface methyl radicals.

NO reduction by CH4 in the presence of O2 over La2O3 supported on Al2O3

Dispersing La2O3 on δ- or γ-Al2O3 significantly enhances the rate of NO reduction by CH4 in 1% O2, compared to unsupported La2O3. Typically, no bend-over in activity occurs between 500° and 700°C, and the rate at 700°C is 60% higher than that with a Co/ZSM-5 catalyst. The final activity was dependent upon the La2O3 precursor used, the pretreatment, and the La2O3 loading. The most active family of catalysts consisted of La2O3 on γ-Al2O3 prepared with lanthanum acetate and calcined at 750°C for 10 h. A maximum in rate (mol/s/g) and specific activity (mol/s/m2) occurred between the addition of one and two theoretical monolayers of La2O3 on the γ-Al2O3 surface. The best catalyst, 40% La2O3/γ-Al2O3, had a turnover frequency at 700°C of 0.05 s−1, based on NO chemisorption at 25°C, which was 15 times higher than that for Co/ZSM-5. These La2O3/Al2O3 catalysts exhibited stable activity under high conversion conditions as well as high CH4 selectivity (CH4 + NO vs. CH4 + O2). The addition of Sr to a 20% La2O3/γ-Al2O3 sample increased activity, and a maximum rate enhancement of 45% was obtained at a SrO loading of 5%. In contrast, addition of SO=4 to the latter Sr-promoted La2O3/Al2O3 catalyst decreased activity although sulfate increased the activity of Sr-promoted La2O3. Dispersing La2O3 on SiO2 produced catalysts with extremely low specific activities, and rates were even lower than with pure La2O3. This is presumably due to water sensitivity and silicate formation. The La2O3/Al2O3 catalysts are anticipated to show sufficient hydrothermal stability to allow their use in certain high-temperature applications.

NO decomposition and reduction by CH4 over Sr/La2O3

Both NO decomposition and NO reduction by CH4 over 4%Sr/La2O3 in the absence and presence of O2 were examined between 773 and 973 K, and N2O decomposition was also studied. The presence of CH4 greatly increased the conversion of NO to N2 and this activity was further enhanced by co-fed O2. For example, at 773 K and 15 Torr NO the specific activities of NO decomposition, reduction by CH4 in the absence of O2, and reduction with 1% O2 in the feed were 8.3·10−4, 4.6·10−3, and 1.3·10−2 μmol N2/s m2, respectively. This oxygen-enhanced activity for NO reduction is attributed to the formation of methyl (and/or methylene) species on the oxide surface. NO decomposition on this catalyst occurred with an activation energy of 28 kcal/mol and the reaction order at 923 K with respect to NO was 1.1. The rate of N2 formation by decomposition was inhibited by O2 in the feed even though the reaction order in NO remained the same. The rate of NO reduction by CH4 continuously increased with temperature to 973 K with no bend-over in either the absence or the presence of O2 with equal activation energies of 26 kcal/mol. The addition of O2 increased the reaction order in CH4 at 923 K from 0.19 to 0.87, while it decreased the reaction order in NO from 0.73 to 0.55. The reaction order in O2 was 0.26 up to 0.5% O2 during which time the CH4 concentration was not decreased significantly. N2O decomposition occurs rapidly on this catalyst with a specific activity of 1.6·10−4 μmol N2/s m2 at 623 K and 1220 ppm N2O and an activation energy of 24 kcal/mol. The addition of CH4 inhibits this decomposition reaction. Finally, the use of either CO or H2 as the reductant (no O2) produced specific activities at 773 K that were almost 5 times greater than that with CH4 and gave activation energies of 21–26 kcal/mol, thus demonstrating the potential of using CO/H2 to reduce NO to N2 over these REO catalysts.

NO reduction by CH4 over rare earth oxides

Abstract NO reduction by methane was studied between 773 and 973 K. All the REO catalysts tested were active for this reaction in both the absence and presence of O2. Activities increased continuously with reaction temperature and no deactivation or bend-over was observed at high temperatures except for Sm2O3, over which complete combustion of CH4 occurred in the presence of O2. The specific activities for NO reduction to N2by CH4 were higher than those for NO decomposition, showing that CH4 enhances NO conversion. CH4 reduction of NO gave selectivities to N2 that were near 100% for all the catalysts except Sr/La2O3, Sm2O3 and Sr/Sm2O3, over which 5–20% N2O was formed. Except for CeO2, the presence of O2 promoted the rate of NO conversion to N2. Overall, Sr/La2O3 had the highest specific activity for NO reduction by CH4 in either the absence or presence of O2, with respective values of 4.6 × 10−3 and 13 × 10−3 μmole N2/s/m2 at 773 K. Turnover frequencies (TOFs) under these two sets of conditions, based on NO adsorption at 300 K, were 0.78 × 10−3s−1 and 2.3 × 10−3s−1, respectively. Activation energies fell between 22–32 kcal/mole for all the REOs. The best REO catalysts correlated with those best for the oxidative coupling of methane. On either a specific activity or a TOF basis, the best REO catalysts were comparable to Co-ZSM-5.

Distinguishing surface and bulk electron charge carriers for ZnO powders

Abstract A two-charge-carrier model that assumes coexisting high-mobility, low-concentration bulk, semiconduction electron charge carriers and low-mobility, variable-concentration surface-trapped-electron charge carriers is used to explain measured electrical and chemisorption properties of ZnO powders. This model resolves two serious quantitative issues not explained by the single-charge-carrier-type model used for our previously reported studies. Not explainable using a single-charge-carrier model are (1) wide variations in measured electron mobility values due to variations in surface treatments and (2) calculated electron number densities too low to match measured electron-transfer chemisorption results. Our two-charge-carrier model for ZnO powders assigns high-mobility bulk electrons to n-type ZnO semiconduction and low-mobility surface electrons to ( V o ) 2− and ( V o + ) − surface oxide ion electron trapping vacancies. This model results in a high variation in surface electron number density due to surface treatments, while the mobilities for both the bulk and surface charge carriers remain constant. The model also calculates much higher surface electron number densities that better match charge-transfer chemisorption results. The two-charge-carrier model is expected to have significant importance in explaining chemisorption and catalysis on ZnO and other similar powder oxides. In particular, the two-charge-carrier model can yield two to three orders of magnitude higher calculated concentrations of surface electrons than for the single-charge-carrier model for any powder with coexisting high-mobility, semiconduction, bulk charge carriers and variable concentrations of low-mobility, surface charge carriers.

Development of the microwave Hall effect technique using an ESR spectrometer and a network analyser

An electron spin resonance (ESR) system was modified to conduct microwave Hall effect (MHE) measurements, but the ESR capability of the system was fully retained using interchangeable equipment. An X-band microwave frequency, a square-law detector, and phase-sensitive detection of the ESR signal were utilized during the modification. The ESR cavity was replaced with a bimodal MHE cavity and source modulation was used instead of field modulation. A cancelling channel was added to cancel out the non-ideal power at the secondary mode. A vector network analyser was used to tune the cavity and to measure the mobility of high mobility samples. A bimodal TE112 cylindrical cavity and a TE102 rectangular cavity were designed and constructed, and both gave good results. Silicon semiconductor samples, which had previously been well characterized, were used to calibrate the system and calculate the calibration constants.

NO reduction by CH4over La2O3: temperature‐programmed reaction and in situ DRIFTS studies

The chemistry between NOx species adsorbed on La2O3 and CH4 was probed by temperature‐programmed reaction (TPR) as well as in situ DRIFTS. During NO reduction by CH4 in the presence of O2, NO3- does not appear to activate CH4, thus either an adsorbed O species or an NO2- species is more likely to activate CH4. In the absence of O2, a different reaction pathway occurs and NO- or (N2O2)2- species adsorbed on oxygen vacancy sites seem to be active intermediates, and during NO reduction with CH4 unidentate NO3-, which desorbs at high temperature, behaves as a spectator species and is not directly involved in the catalytic sequence. Because reaction products such as CO2 or H2O as well as adsorbed oxygen cannot be effectively removed from the surface at lower temperatures, steady‐state catalytic reactions can only be achieved at temperatures above 800 K, even though formation of N2 and N2O from NO was observed at much lower temperature during the TPR experiments.

NO reduction over La2O3 using methanol

Nitric oxide (NO) reduction by methanol was studied over La2O3 in the presence and absence of oxygen. In the absence of O2, CH3OH reduced NO to both N2O and N2, with selectivity to dinitrogen formation decreasing from around 85% at 623 K to 50–70% at 723 K. With 1% O2 in the feed, rates were 4–8 times higher, but the selectivity to N2 dropped from 50% at 623 K to 10% at 723 K. The specific activities with La2O3 for this reaction were higher than those for other reductants; for example, at 773 K with hydrogen a specific activity of 35 μmol NO/s m2 was obtained whereas that for methanol was 600 μmol NO/s m2. The Arrhenius plots were linear under differential reaction conditions, and the apparent activation energy was consistently near 14 kcal/mol with CH3OH. Linear partial pressure dependencies based on a power rate law were obtained and showed a near‐zero order in CH3OH and a near‐first order in H2. In the absence of O2, a Langmuir–Hinshelwood type model assuming a surface reaction between adsorbed CH3OH and adsorbed NO as the slow step satisfactorily fitted the data, and the model invoking two types of sites provided the best fit and gave thermodynamically consistent rate constants. In the presence of O2 a homogeneous gas‐phase reaction between O2, NO, and CH3OH occurred to yield methyl nitrite. This reaction converted more than 30% of the methanol at 300 K and continued to occur up to temperatures where methanol was fully oxidized. Quantitative kinetic studies of the heterogeneous reaction with O2 present were significantly complicated by this homogeneous reaction.

Microwave absorption measurements of the electrical conductivity of small particles

A microwave absorption technique based on cavity perturbation theory is shown to be applicable for electrical conductivity measurements of both a small, single-crystal particle and finely divided powder samples whenσ values fall in either the low (σ<0.1 Ω−1 cm−1) or the intermediate (0.1 ≤σ≤ 100 Ω−1 cm−1) conductivity region. The results here pertain to semiconductors in the latter region. If the skin depth of the material becomes significantly smaller than the sample dimension parallel to theE-field, an appreciable error can be introduced into the calculated conductivity values; however, this discrepancy is eliminated by correcting for the field attenuation associated with the penetration depth of the microwaves. A modification of this approach utilizing the skin depth allows a first-order correction to be applied to powder samples which results in the accurate measurement of absoluteσ values, and results with doped Si powders are compared toσ values obtained from one small single particle using this microwave technique as well as reported DCσ values determined with single crystals. The use of this microwave absorption technique with small particles having high surface/volume ratios, such as catalyst supports and oxide catalysts, under controlled environments can provide fundamental information about adsorption and catalytic processes on such semiconductor surfaces. An application to a ZnO powder demonstrates this capability.