Sb-Containing Metal Oxide Catalysts for the Selective Catalytic Reduction of NO_x with NH₃

Round 1

Reviewer 1 Report

The authors present binary or ternary mixed oxides of Sb, Ce and Zr with equimolar amounts of each element as potential catalysts for the selective catalytic reduction of NOx by NH3.
In the introductory section, sufficient background on other mixed oxide systems for NOx reduction is given, although language and style of this first part are very bad, some parts even appear to be automatically translated and this section urgently requires improvement (i.e. it almost made me stop reading the rest of the paper).

Line 105: The instrument used was probably a *Bruker* D8 diffractometer

Figure 1: In section 2.2, the feed composition is not uniquely defined. The steam concentration is given subject to the condition "when used", line 93. Therefore, the authors should specify if steam was present in the experiment resulting in Figure 1. Also the aging conditions should be defined. Line 151 mentions 700°C, 4h, but the gas atmosphere is not defined. Especially the presence of steam during aging is crucial.

In section 2.2 no information is given about the temperature program of the test. The diagrams imply a stationary test at discrete temperature levels. What is the equilibration time at each temperature (the performance of SCR catalysts is often dependent on NH3 saturation and may require some equilibration time to become stationary)?
Do the authors have data on the time on stream stability of their catalyst (e.g. results of repeated temperature ramps)? This is especially important for the tests with SO2. Usually a certain SO2 exposure is required until SO2 becomes harmful. What was the SO2 exposure in the tests with SO2? Was it comparable for all catalysts?

Related to performance data: The authors only report NO or NOx conversion, but no information on the selectivity of the reaction is provided. How much N2O is formed by the SbCeZrOx system? Is N2O formation comparable to state of the art V/TiO2 type catalysts (which are in commercial use) or is N2O make (a potent green house gas) higher?

Maybe it would be worthwhile to also include at least BET and XRD results for aged samples. What is the aging mechanism? Just sintering and loss of surface area or is phase segregation observed as well?

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Editor/Authors,

The submitted article “Sb-containing metal oxide catalysts for the selective catalytic reduction of NO_x with NH₃” is very well written, and the results are sufficiently discussed. The discussion is sufficient to support the conclusions. The methods and results are both interesting and novel in my opinion. However, the author can correct a few typos and comments which are highlighted in the PDF file.

The paper can be accepted for publication in the International Journal of Catalysts subject to minor revisions.

Please check the comments in the attached PDF file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

In the present manuscript, Sb-containing catalysts were studied in the selective reduction of NOx with NH3, where the addition of Zr has been shown to increase the catalytic activity and the thermal stability and resistance of the catalyst against H2O and SO2. Various characterization methods have been applied to understand the different reactivity of the samples. Thus, an increase in acidity and reducibility has been observed with the addition of Zr to the catalyst. The authors conclude, based on in situ DRIFT studies, that the NO oxidation and NH3 adsorption doesn’t compete in the SbCeZr sample, promoting in that way NOx reaction. In addition, different reaction mechanism (Eley-Rideal and Langmuir-Hinshelwood) have been postulated as well as the participation of bidentate nitrates as active intermediate species.

 The study has been well done, the samples well characterized, but the in situ DRIFT studies need to be recheck. Spectra resolution is poor and some of the conclusion need to be verified and/or discussed in more detail.  Authors should be aware that the NOx reduction is a wide studied reaction, with a lot of discussion about active intermediate species and sites, reaction mechanism, role of NH3 adsorbed on Lewis and Brönsted sites, role of NO2, nitrites and nitrates. Under that perspective, some conclusion have been set in the present manuscript, which need to be revised and/or discussed according to the literature. For instance:

• The role of NH3-L versus NH4+- B species. The authors base their argument that NH3-L species may be involved in the reaction mechanism, according to the faster disappearance of the NH3-L IR band versus the NH4+-B IR band. However, this is hard to visualize in Figure 9. It would be useful to show the time dependant evolution of the IR Band intensity (better IR area) in one graph. (time in the x axe, and IR area in the y axe).

• In the Co-adsorption experiments of NO+O2, the authors assign the IR bands to the formation of bridge nitrate, bidentate nitrate, and monodentate nitrate. But what about NO2, nitrites, N2O4, NO+,…?. NO2 has been speculated as active intermediate specie, and should be detected in the NO+O2 IR experiment and discussed.

• The IR band in fig 10 is quite broad. Several bands have been identified located at 1559, 1539 and 1519 cm-1. How accurate is it?

• The authors postulate an Eley-Rideal mechanism where adsorbed NH3 on the catalyst surface react with NO/NO2 in the gas phase based on figure 9. In my opinion, this conclusion has to be re-considered, and additional experiments are needed. 200ºC is a too high temperature where the stabilization of reaction intermediate or adsorbed species is strongly limited and the reaction kinetic is too fast to visualize intermediates. In my opinion, the experiment should be done by lowering the temperature to 100 ºC for example, and then slowly increase to 200ºC. In that case, intermediate species, as for instance NH4NO2 may be identified. If that is the case, a Langmuir-Hinshelwood mechanism may be considered.

• According to figure 11 the authors conclude that bridge nitrate species participate in the reaction mechanism while bidentate and monodentate nitrate species doesn’t do. This is hard to visualize it from fig 11. Moreover, as indicated above, the role of nitrite species in the reaction mechanism need also be discussed

• In figure 11, the IR band associated to NH3-L species overlap with the initial IR band associated to bridge nitrate specie. Thus, the formation of NH3-L species is difficult to assess in the experiment of figure 11. This rise some doubts about the concept of competitive adsorption of NH3 and NOx.

In addition to the above exposed fundamental questions, details in the IR section are missed. For example the temperature at which each experiment have been performed (For instance the NO+O2 co-adsorption (section 3.3.3. and Figure 10), the adsorption of NO+O2 followed by introduction of NH3, (section 3.3.4 and Figure 11), the adsorption of NH3, (section 3.3.1 and Figure 8). Also, sample pre-activation has been modified in the different techniques, i.e. pretreatment at 500 ºC under flow of Ar in NH3-TPD and NO-TPD, pretreatment at 300 ºC in the H2-TPR and DRIFT studies, while no activation have been performed in the catalytic studies. According to the TPR profile, activation at 500ºC in an inert atmosphere could reduce the surface in some extension.  It is important not to increase too much the preactivation temperature and use always the same conditions. On the other hand, please revise the discussion, for instance Figure 12 in line 434 doesn’t exist. Figure 10 in line 429, should be Figure 9, as well as Figure 12 in line 434, should be figure11.

Also, some inconsistence appears comparing the TPD-NH3 with the IR-NH3. Looking to the scale bar of figure 8 it looks that the amount of Brönsted and Lewis sites in the SbCe sample (figure 8b) is approximately  10 times lower than in the SbCeZr sample (Figure 8a). However in the TPD-NH3 data of figure 7, the difference is only a factor 2. The authors should think about this observation. I think the problem is in the IR spectra because it is known that in DRIFT the IR intensity is strongly dependant on sample particle and grain size, homogeneity of the sample,….. This is very important when comparing and discussing IR band intensity from different samples, as for instance the SbCe and SbCeZr sample.

Finally, the Y axe in the IR spectra should be Kubelka Munck, not absorbance, or have the spectra been acquired in absorbance?

In conclusion, in my opinion, the manuscript need to be revised in detail and some part redone before publication.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The discussion of the IR data are still ambiguous. The following conclusions are not supported from experimental data and need to be checked again before publication of the manuscript.

1. Bridge nitrate as active specie. Fig.11 in the manuscript is hard to visualize. The authors supplied a new figure (Fig.R7), but still the decrease in the intensity of the IR bands due to bridge nitrate species (1605 and 1245 cm-1) is hard to visualize. Moreover, the IR band at 1245 cm-1 increase in intensity.
2. No competitive adsorption between NH3 and nitrates in the SbCeZr samples (explaining the enhanced catalytic activity). As stated before, there is no appreciable change in the IR band intensity due to bridge, bidentate or monodentate nitrate species after NH3 co-adsorption (Data in fig.11 and Fig S7). A new band appear at 1440 cm-1 corresponding to NH4+-B, however NH3-L (characterized by an IR band at 1198 cm-1 according to figure 8) is not formed. The authors assign the slight increase of the 1253 cm-1 IR band due to NH3-L species. That could be, but in that case the IR band at 1198 cm-1 should be clearly detected, which is not the case. Under this premise, I would say that NH3 adsorption on Lewis sites compete with nitrate species. NH4+-B are formed (which interestingly are absent in the SbCe sample). Accordingly, the authors could still postulate “a dual site mechanism” in the SbCeZr sample, but assuming  NH4+-(B) and NOx-(L) as active sites. Please, consider this option.
3. NH3 on Lewis sites as active site. The faster disappearance of the NH3-L IR band in figure 9 and in Figure R4 could also be explained by NH3-L as reservoir of NH4+-B sites, behaving the last one as real active site. In deed the slope of the NH4+-B line in FigR4, has an inflexion point at around 10 min. Please consider this option
4. E-R mechanism. I still consider 200ºC too high for mechanistic studies. Furthermore the new figure (Fig.R6) supplied by the authors makes not sense. NH3 is clearly detected in figure 9 of the manuscript, which has been acquired at 200ºC. Therefore, it has to be detected at 100ºC in Fig.R6. In addition, bridge nitrate species has two bands at 1605 and 1247 cm-1, and both bands should be detected in the spectra of figure R6 after 60 min of reaction. Please revise those data.

Finally,:

The reaction temperature should be included in the figure captions of Fig.8-11. 

Decovolution of Fig.10, need to be revised. Some components are missing.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 3

Reviewer 3 Report

The results are more convincing now. The authors have answered and discussed with detail all the question risen by the reviewer. I consider the manuscript has improved considerably, specially in the discussion of the reaction mechanism.

I will thank the authors for his efforts and I consider the manuscript suitable for publication in the present form.