Direct Decomposition of NO into N 2 and O 2 over C-Type Cubic Y 2 O 3-Tb 4 O 7-ZrO 2

Catalytic activities for direct NO decomposition were investigated over C-type cubic Y2O3-Tb4O7-ZrO2 prepared by a co-precipitation method. The NO decomposition activity was enhanced by partial substitution of the yttrium sites with terbium in a (Y0.97Zr0.03)2O3.03 catalyst, which shows high NO decomposition activity. Among the catalysts synthesized in this study, the (Y0.67Tb0.30Zr0.03)2O3.33 catalyst exhibited the highest NO decomposition activity; NO conversion to N2 was as high as 67% at 900 ̊C in the absence of O2 (NO/He atmosphere), and a relatively high conversion ratio was observed even in the presence of O2 or CO2, compared with those obtained over conventional direct NO decomposition catalysts. These results indicate that the C-type cubic Y2O3-Tb4O7-ZrO2 catalyst is a new potential candidate for direct NO decomposition.


Introduction
Nitrogen oxides (NO x ) are not only harmful to human beings, but are also responsible for photochemical smog and acid rain when present in relatively high levels in the atmosphere.NO x is mainly produced by the high-temperature combustion of fossil fuels such as petroleum in the engines of vehicles and ships, or coke in large-size boilers of factories.NO x species in exhaust gases emitted at high temperatures are composed principally of thermodynamically stable NO and a negligible amount of NO 2 .Therefore, studies on the catalytic decomposition of NO x should focus on NO.
Several NO reduction processes have been proposed for NO x removal.Among them, selective catalytic reduction (SCR) methods employing ammonia or urea have been extensively studied and applied in diesel engines and large-size boilers [1].The SCR methods have sufficient NO decomposition efficiency and a stable reaction process at high temperatures.However, separate specialized equipment is necessary to supply the reducing agents.Moreover, it is absolutely essential to ensure secure control systems due to the high toxicity and flammability of ammonia.
In contrast to the above processes employing ammonia or urea, direct catalytic decomposition of NO into N 2 and O 2 (2NO → N 2 + O 2 ) is the most ideal route for NO x removal, because no reducing agents and no special equipment are required.A number of materials, such as zeolites [2,3], perovskites [4][5][6], and other complex oxides [7][8][9][10][11][12][13][14][15][16][17][18], have been proposed as active catalysts for direct NO decomposition.However, the NO decomposition activity of these conventional catalysts is significantly decreased in the presence of O 2 and CO 2 , due to strong adsorption of these molecules on the surface of the catalysts.Oxygen molecules produced by NO decomposition as well as those present in the gas phase adsorb on the catalyst surface and interfere with the catalytic reaction.In addition, a number of active direct NO decomposition catalysts contain highly basic alkaline and alkaline earth ions in the lattice.Among them, catalysts containing Ba typically exhibit high NO decomposition activities, because NO is acidic and adsorption of NO on the surface of the catalyst is significantly enhanced with increasing basicity of the catalyst.However, the high basicity also facilitates adsorption of CO 2 .As a result, catalyst poisoning by CO 2 adsorption becomes a serious problem.
To realize high activities for direct NO decomposition, it is important to design a novel catalyst that is not influenced by the presence of CO 2 or O 2 .Accordingly, we focused on C-type cubic rare earth oxides.Rare earth oxides can form three types of crystal structures, A-(hexagonal), B-(monoclinic), and C-types (cubic), depending on the ionic size of the respective rare earth element [19].Among these, the C-type cubic structure has the largest interstitial open space.It has been generally accepted that large open spaces in a catalyst play an important role in direct NO decomposition, and for this reason the C-type cubic rare earth oxide is suitable as a direct NO removal catalyst.Furthermore, in our previous works, we found that several C-type cubic rare earth oxide catalysts can exhibit high activities for direct NO decomposition even in the presence of O 2 [20][21][22][23][24][25] and that exclusion of alkaline earth ions from the catalyst lattice is significantly effective to induce CO 2 tolerance [25,26].

Catalyst Preparation
The C-type cubic (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ (x = 0, 0.10, 0.20, 0.30, and 0.40) catalysts were synthesized by a co-precipitation method, where the zirconium content was fixed at 3 mol% for the reason mentioned above.A stoichiometric mixture of 1 mol•dm -3 Y(NO 3 ) 3 , 0.1 mol•dm -3 Tb(NO 3 ) 3 , and 0.1 mol•dm -3 ZrO(NO 3 ) 2 aqueous solutions was added to a 1.0 mol•dm -3 ammonium carbonate solution with stirring.The pH of the mixture was adjusted to 10 by dropwise addition of 9% ammonia solution.After stirring at room temperature for 6 h, the resulting precipitate was collected by filtration, washed several times with deionized water, and then dried at 80˚C for 6 h.The powder was then ground in an agate mortar and finally calcined at 900˚C in air for 6 h.

Characterization
The catalysts were characterized by X-ray powder diffraction (XRD; Rigaku SmartLab) with CuKα radiation.XRD patterns were recorded in the 2α range from 10˚ to 70˚.The sample compositions were analyzed by X-ray fluorescence spectrometry (XRF; Rigaku, ZSX100e) and the specific surface area was measured by the Brunauer-Emmett-Teller (BET) method using nitrogen adsorption at −196˚C with a Micromeritics TriStar 3000 adsorption analyzer.

Catalyst Test
The NO decomposition reaction was carried out in a conventional fixed-bed flow reactor with a 10-mm-diameter quartz glass tube.A gas mixture of 1 vol% NO and He (balance) was fed at a rate of 10 cm 3 •min −1 over 0.5 g of catalyst.The W/F ratio, where W and F are the catalyst weight and gas flow rate, respectively, was adjusted to 3.0 g•s•cm −3 .The gas composition was analyzed using a gas chromatograph (Shimadzu GC-8A) with a thermal conductivity detector (TCD), a molecular sieve 5 A column for NO, N 2 , and O 2 , and a Polapak-Q column for N 2 O separation.The activity of each catalyst was evaluated in terms of NO conversion to N 2 .
The effect of the presence of O 2 or CO 2 was measured by mixing each gas species with the reactant gas.The concentrations of the additional gases and NO were controlled by changing the feed rate of He as the balance gas to maintain a total reactant flow rate of 10 cm 3 •min -1 .

Temperature Programmed Desorption
Temperature-programmed desorption (TPD) measurements of O 2 was carried out after adsorption of O 2 at 600˚C for 1 h.After heating the catalyst in a flow of He (30 cm 3 •min −1 ) at 600˚C for 30 min, the catalyst was exposed to O 2 (1 atm) at the same temperature for 1 h, and then cooled to 50˚C.After evacuation at 50˚C for 30 min, the catalyst was heated under a flow of He at a heating rate of 10˚C•min −1 and the desorbed gas was monitored using a gas chromatograph with a catalysis analyzer (BELCAT-B BEL JAPAN).In the case of CO 2 -TPD, the catalyst was heated in a flow of He (30 cm 3 •min −1 ) at 600˚C for 1 h, and subsequently in a flow of H 2 (30 cm 3 •min −1 ) at 600˚C for 30 min.The catalyst was cooled in a flow of He to 50˚C, and was then exposed to CO 2 (1 atm) at this temperature for 1 h.After evacuation at 50˚C for 30 min, the catalyst was heated under a flow of He at a heating rate of 10˚C•min −1 .

Characterization of the Catalysts
Figure 1 shows XRD patterns of the (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ catalysts (x = 0, 0.10, 0.20, 0.30, and 0.40).C-type cubic rare earth oxide with a single phase structure (PDF-ICDD 41-1105 for Y 2 O 3 ) was successfully obtained for all samples and no crystalline impurities were observed.The catalyst compositions, as determined using XRF, the lattice constants, and the BET surface area of the (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ catalysts are summarized in Table 1.The cubic lattice parameter of (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ gradually decreased with the x value, because ionic sizes of Y 3+ and Tb 4+ with a six-fold coordination are 0.1040 nm [27] and 0.0900 nm [27], respectively.When the smaller Tb 4+ occupies the lattice position of Y 3+ in (Y 0.97 Zr 0.03 ) 2 O 3.03 , the lattice parameter  decreases monotonically with increasing Tb 4+ content.
The results indicate that C-type cubic solid solutions were successfully formed for all samples.The BET specific surface area was slightly affected by the introduction of Tb 4+ into the (Y 0.97 Zr 0.03 ) 2 O 3.03 lattice.

NO Decomposition Activity
Figure 2 depicts the temperature dependencies of the N 2 yield for the (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ catalysts (x = 0, 0.10, 0.20, 0.30, and 0.40).Initial NO decomposition activity appeared at 550˚C and the N 2 yield increased monotonically with reaction temperature.The formation of N 2 O was not detected between 400 and 900˚C.Figure 3 shows the dependence of NO conversion into N 2 at 900˚C on the composition of the C-type cubic (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ catalysts.The catalytic activity increased with the x value, and the highest catalytic activity was obtained for the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 composition, where the N 2 yield obtained over this catalyst was 67%.

Effect of the Presence of O 2 or CO 2
The effect of the presence of O 2 or CO 2 on the N 2 yield  obtained over the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst at 900˚C was also examined, and the results are presented in Figure 4.In the presence of O 2 , the N 2 yield for (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 decreased from 67 to 53% until the content reached 1 vol%, but became almost constant in the range from 1 to 5 vol%.As a result, NO decomposition activity as high as 47% was maintained, even in the presence of 5 vol% O 2 , and the activity did not deteriorate during the 10-h catalytic test.This conversion ratio in the presence of 5 vol% O 2 (47%) was higher than that for C-type cubic (Yb 0.50 Tb 0.50 ) 2 O 3±δ (35%) [25].
In contrast, the effect of CO 2 on the NO decomposition activity was relatively larger than that for O 2 , but the decreasing tendency of the N 2 yield with the increase in the partial pressure of CO 2 was similar to that observed for O 2 .Since CO 2 is acidic, adsorption of CO 2 on the basic sites of the catalyst surface will inhibit NO adsorption on the open space sites of the catalyst.However, for the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst, a high N 2 yield of 36% was maintained even in the presence of 5 vol% CO 2 , which is higher than those for the (Yb 0.50 Tb 0.50 ) 2 O 3±δ (34%) [25], Ba 0.8 La 0.2 Mn 0.8 Mg 0.2 O 3 (20% at 850˚C) [12], and La 0.8 Sr 0.2 CoO 3 (10% at 800˚C) catalysts [15].Furthermore, the catalytic activity was recovered when the catalytic test was carried out again in the absence of CO 2 (1 vol% NO/He), which indicates that the decrease in NO decomposition activity is caused by the adsorption of CO 2 on the catalyst.In addition, the XRD patterns of the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst were the same before and after the reactions.

O 2 and CO 2 Desorption Profiles
As mentioned above, the present (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst showed relatively high catalytic activity even in the presence of O 2 or CO 2 .To facilitate direct NO decomposition, it is important that these coexisting gases desorb from the surface of the catalyst easily.Therefore, desorption behavior of O 2 and CO 2 adsorbed on (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 was characterized by TPD measurements.
Figure 5 shows O 2 and CO 2 desorption profiles (O 2 -TPD and CO 2 -TPD) for (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 .In references [5,6,12], catalysts that demonstrated lowtemperature O 2 desorption exhibited high activities for NO decomposition.In the present case, a single O 2 de-  sorption peak was observed at 470˚C for the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst, whereas that for (Yb 0.50 Tb 0.50 ) 2 O 3±δ was 510˚C.In general, catalysts with weak oxygen adsorption exhibit higher NO decomposition activities.Therefore, the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst showed high NO decomposition activity even in the presence of O 2 .
In the case of CO 2 , the CO 2 desorption peak was observed at 115˚C, which is much lower than those seen for Ba 0.8 La 0.2 Mn 0.8 Mg 0.2 O 3 (700˚C) [14] and La 0.8 Sr 0.2 CoO 3 (750˚C) [15].The adsorption strength of CO 2 correlates with the desorption temperature observed in TPD, and the lower the CO 2 desorption temperature is, the weaker the adsorption strength becomes.Although the weakly adsorbed CO 2 may block the active site for NO decomposition in the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst, the adsorption strength of CO 2 on this catalyst is quite weak.As a result, the NO decomposition activity was not significantly suppressed by the CO 2 coexistence.This consideration is supported by the result mentioned above that the N 2 yield obtained at 900˚C in the presence of 5 vol% CO 2 over the present (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst (36%) was almost equivalent to that over (Yb 0.50 Tb 0.50 ) 2 O 3±δ (34%), because the CO 2 desorption peak was observed at 115˚C and 125˚C for the former ant the latter, respectively.

Conclusion
C-type cubic (Y 0.97-x Tb x Zr 0.03 ) 2 O 3.03+δ (x = 0, 0.10, 0.20, 0.30, and 0.40) catalysts, in which the yttrium site was partially substituted with terbium, were found to exhibit high catalytic activity for direct NO decomposition.The highest catalytic activity was obtained for (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 .It is noteworthy that the catalytic activity was maintained at a high conversion ratio even in the presence of O 2 or CO 2 .Therefore, the (Y 0.67 Tb 0.30 Zr 0.03 ) 2 O 3.33 catalyst is expected to be a new potential candidate as a direct NO decomposition catalyst.