Effect of High-Energy Milling on Ceria-Zirconia’s Redox Properties

Round 1

Reviewer 1 Report

The manuscript is devoted to the production of materials based on cerium and zirconium oxides using a high-energy milling.

The manuscript contains the following comments:

1. The introduction does not indicate the problem being solved. It is simply listed that cerium and zirconium oxides can be prepared by various methods. But this is a known fact, since these oxides and materials based on them have been successfully used for several decades.

When describing the milling method, the authors talk about its promise, but say nothing about its limitations, often associated with the production of nanoparticles.

The introduction needs to be improved to identify the problem being solved.

2. About the description of the porous structure of materials.

For a complete and correct description of the porous structure, one should begin with an analysis of the adsorption isotherms. What pores and shapes are observed in this material?

To characterize the porous structure, generally accepted designations should be used: specific surface area = S_a, pore volume = V_pore. Also makes sense what pore volume (micro, meso- or summary) are we talking about?

In Fig. 4, the pore size is indicated in A, in Fig. 15 - already in nanometers. In Fig. 4 is a slit-like shape (the width of the slit pore), in Fig. 15 is already the diameter of a cylindrical pore. Why? It makes sense for the authors to understand the results obtained.

3. To describe the phase composition of such oxides, it would be interesting to see the Raman spectra.

4. In the manuscript, the authors call the resulting material a nano-sized powder with a size of 0.3 - 1.1 microns. Is it really nano-sized?

General impression: The article is poorly structured. The objects of research are not pure cerium and zirconium oxides; they also contain lanthanum and praseodymium oxides, as you will learn about only on page 14.

First, we are talking about the properties of the CeZr40 composition, then CeZr20. Different methods were used for characterization (or not all results are given). Therefore, it is quite difficult to compare the results obtained.

Recommendations: the results obtained must be structured, analyzed and clearly described. The results need to be compared with data available in the literature.

Minor editing of English language required

Author Response

We appreciate your reviewing the manuscript draft and providing us with valuable comments. We have carefully considered the comments and tried our best to address every one of them.

1. The manuscript structure has been changed.
2. Changes to “introduction” have been made.
3. SSA/PV (“production” slang) is changed to generally accepted Sa and Vpore. Vpore = total pore volume (added to “methods”). N2 isotherms of CZ20 and CZ40 belong to class IV. Materials are mesoporous. Figure 4 has been edited.
4. PIDC doesn’t have Raman. The standard characterization is XRD. Customer requirement is “single phase” after ageing at 1100oC according to XRD.
5. By some classifications particles below 1 micron are nano, by other those below 0.1 micron. In automotive catalysis we operate with particles sizes typically from 5 micron to 20 micron. GPF/DPF requires particles less than 1 micron. So, they differ from standard powders and regarded as “nano”.“
6. Material” section is at the end of the manuscript according to the Journal structure. We duplicated CZ compositions in text. 
7. All comparative testing has been done using CZ20 and CZ40 composition (as well as several others to check if “fast oxygen mobility” effect is generic). We have HRTEM characterization data only for CZ40 composition. So, we included HRTEM data as illustration. Enhancement of oxygen mobility after steam-jet milling was observed also for other CZ compositions with CeO2 content from 15% to 50%. Effect doesn’t depend on composition.

Reviewer 2 Report

The manuscript reported the effect of different mode of milling on ceria-zirconia redox properties. The results are interesting, I think it can be published after minor revision based on the following commments.

 1.      The authors claimed that thermal stability of steam jet milled nanomaterials remains sufficiently high SSA-1000/6 >35 m2/g and PV-1000/6 > 0.15 cc/g in page 5 line 83 and 84. It can be seen evidently from Table 1 that the thermal stabilities of the milled samples exhibit worse thermal stabilities than that of the unmilled powder. Why can you say that? Furthermore, the SSA-1000/6 is 34 m2/g but not >35 m2/g.

2.      I suggest the authors add the XRD pattern of the unmilled samples into the Figure 5. And an explanation why the peak positions are different should be given.

3.      In the part of 3.2. Characterization methods, ICP was used to measure the elemental composition of mixed oxide samples, however, I cannot find the data in the main text.

Author Response

We appreciate your review and valuable comments on the manuscript draft.

1. Completely agree that milling has a negative effect on CZ materials’ thermal stability. PIDC is a manufacturing company which uses milling techniques for making nanomaterials (zirconia, yttria stabilized zirconia, ceria gadolinia, etc.) for ceramic-types applications where final product should be practically non-porous. For gasoline and diesel particulate filters customer requirement is materials’ particle size – d50 should be less than 1 micron. Higher surface area and porosity of milled material is a definite benefit for performance. We found that PIDC milled materials retain surface area after ageing at 1000oC for 6 hours at around 30-45 m2/g, depending on milling depth. Such SSA values are comparable with SSA of competitors materials without milling. That’s why we made a statement about “sufficiently” high thermal stability of milled CZ materials.

Statement is removed.

2. For pure CeO2 2Theta position of the 1^st peak is close to 28 degrees and for pure ZrO2 (cubic or monoclinic phase) is close to 30 degrees. CZ20 and CZ40 are solid solutions and the position of their first peak is between 28 and 30 degrees depending on CeO2 content (as well as on the amount and type of rare earth dopants).

Added XRDs of milled and un-milled material.

3. Removed.

Reviewer 3 Report

In this manuscript the authors investigated the effect of different types of milling reducing ceria-zirconia (CZ) powders with two different C/Z ratio to nanoparticles on the morphological, physical and chemical properties of the materials with a special focus on Oxygen Storage Capacity (OSC).

This would be a rather interesting paper but, however, it is very confusing.

First of all, they made a poor analysis of the literature which is very wide on this topic. Moreover, references must be definitely updated.

The aim of the work at the end of the introduction is missing.

Moving to the experimental section:

All acronyms and symbols (d50 included) must be defined. Please, replace the Greek letter m instead of u reporting the particle dimension (micron) in the text.

They made (or report) some experiments on CZ20 and some others on CZ40 never explaining why they selected a given composition and which is the reason of this choice. Since they did not report any comment about the different composition, it would be better to analyse only a single composition (CZ20 or CZ40).

They did the same about the temperature of calcination. They initially investigated materials calcined at 700, 1000 and 1100°C but they selected only those calcined at 1000°C for further experiments without explaining their choice.

I suggest to remove parameters that are not discussed.

Table 1 “Aged specific surface area”. SSA cannot be aged and not even fresh. The materials are aged or fresh.

Line 84 – What does “SSA-1000/6>35 m2/g and PV-1000/6>15 cc/g” mean? What does the number 6 represent?

Figure 5. How do the authors explain the shift of all peaks of the black XRD pattern with respect to the red pattern?

CZ crystalline phase must be clearly defined with chemical formula and PDF number of the corresponding file.

TPR-H2. Although the order of reducibility evaluated by TPR-H2 should be theoretically proportional to OSC, TPD-O2 would provide a more direct measure of this property. Moreover, this would provide much more useful information for catalytic soot oxidation which is reaction where soot actually reacts with oxygen produced by the catalytic materials. The same instrument they used for TPR-H2 can be easily used also for TPD-O2.

Line 157-158. 595°C is the temperature of the uncatalyzed oxidation of carbon black. This must be reported as well as the corresponding reference (the authors often do not cite references when they mention results from literature).

Figure 9. The authors did not consider that, in addition to the higher OSC, the SJM CZ consists of much smaller particles thus strongly increasing the contact area with Printex-U. This can significantly contribute decrease the temperature of oxidation.

Figure 13. Unmilled CZ20 (blue profile) missing in the legend.   

Table 5. What is Tmax reported in the Table? TPR? Soot oxidation?

Materials and Methods:

calcination of materials never mentioned

Impregnation with PGM not described

No amount of both CZ and carbon black (and their ratio) reported in the description of oxidation tests

Author Response

We appreciate your reviewing the manuscript draft and providing us with valuable comments. We have carefully considered the comments and tried our best to address every one of them.

1. The manuscript structure has been changed.
2. CZ materials are known and widely used for decades – commodity products. Agree, that there is a lot of literature (and patents) on CZ materials. All fundamental research has been done in 90-th and early 2000. Our objective is very modest - to provide interesting data on fast oxygen mobility in CZ to the academy and industry community. As a manufacturing company we are not doing fundamental research but will be glad to work with the academy. Literature has been updated.
3. The aim of the work has been stated at the end of introduction.
4. Added particle size analysis in “characterization” section with explanation of d50 meaning.
5. CZ20 and CZ40 are PIDC commercial products.
6. Calcination temperature. All CZ materials have been calcined at 700oC first (“fresh” materials). Additional calcination at 1000oC is wash coaters “standard” for CZ porosity characterization and evaluation of thermal stability. Ageing at 1100oC is typically used for CZ phase stability evaluation. Some companies may use lower or higher temperatures (from 950 to 1150oC).
7. Agree that SSA can’t be “aged” or “fresh”. Corrections made.
8. SSA-1000/6, etc. Changed to Sa after 1000oC for 6 hours.
9. For pure CeO2 (cubic) 2Theta position of the 1^st peak is close to 28 degrees and for pure ZrO2 (cubic or monoclinic phase) is close to 30 degrees. CZ20 and CZ40 are solid solutions and the position of their first peak is between 28 and 30 degrees depending on CeO2 content (as well as on the amount and type of rare earth dopants). These compositions are not in PDF database.

Added Figures with XRDs of milled and un-milled material for CZ20 and CZ40. The main idea of these graphs is to show that CZ materials withstand milling without phase disproportionation.

10. TPR-H2. The use of TPR-H2 is typical for characterization of CZ materials for CZ manufacturers and wash coaters.
11. The soot oxidation graph has been removed. Agree, CZ particle size should influence change of soot oxidation temperature.
12. Added description of CZ calcination and ageing conditions.
13. Added description of CZ impregnation with PGM.

Round 2

Reviewer 1 Report

The authors of the manuscript corrected many comments, but some questions remained:

1. In Fig. 3 the designation of the ordinate axis did not appear - what size distribution is shown in this figure?

2. The description of the source materials raises even more questions. Why were the starting materials subjected to heat treatment up to 700C before milling? How and with what can the obtained data be compared?

Minor editing of English language required

Author Response

Thank you very much for taking the time to review this manuscript again. Please find the our responses on your comments.

Typo in Figure 3 has been corrected.

All washcoaters require not freshly precipitated CZ materials but calcined. They have specific requirements to surface area (~70-100 m2/g), porosity (depending on application) and residual water content (LOI<1-3% at 1000oC). Freshly precipitated ceria-zirconia based hydrous oxides could have Sa 200-300 m2/g, pore volume >1 cm3/g and LOI>40%. So, all commercial grade products are calcined by the manufacturer to meet washcoaters requirements.

“Fresh” or calcined (at 700oC) materials have been used in all milling experiments. Jar milling and Eiger milling give CZ aqueous slurries. After drying slurry in oven at 110-130oC CZ powders have LOI>10%, which doesn’t allow compare them with the reference. Pretty much the same situation is in the case of steam jet milling. The final product is “dry”, but LOI is higher than in the reference CZ. In order to get comparable results all milled CZ materials have been calcined similar to reference (starting materials) at 700oC. Such treatment gives the idea how milling affects Sa and pore volume. Additional ageing of milled materials at 1000oC and 1100oC show how milling impacts thermal stability (sinterability) and phase stability.

It’s worth noting that repeated calcination of “fresh” un-milled CZ at 700oC has an extremely negligent effect on surface area and pore volume. Significant sintering of CZ materials used in TWC application starts at temperatures higher than 1000oC.

We think that calcination profiles chosen give the best comparative results.

Reviewer 3 Report

1. The manuscript structure has been changed.

The authors deeply modified the manuscript but one of the main issues still exist: they did not select only one compositions (CZ20 or CZ40). This is fundamental because the composition affects catalyst properties but it is not one of the parameters studied in this work.

2. CZ materials are known and widely used for decades – commodity products. Agree, that there is a lot of literature (and patents) on CZ materials. All fundamental research has been done in 90-th and early 2000. Our objective is very modest - to provide interesting data on fast oxygen mobility in CZ to the academy and industry community. As a manufacturing company we are not doing fundamental research but will be glad to work with the academy. Literature has been updated.

The authors added some more recent references.

The academic community is very glad to work with manufacturing companies as well. This is fundamental for applied research, however, both communities should try to cross data and results on a common basis and if people from industry choose to publish in journals mostly read by the academic community they should seek to adapt to the “academic rules”.

3. The aim of the work has been stated at the end of introduction.

The authors could stress the ease of the method to increase oxygen mobility which is the strength of the work.

4. Added particle size analysis in “characterization” section with explanation of d50 meaning.

5. CZ20 and CZ40 are PIDC commercial products.

This was clear also in the original version of the manuscript. Nevertheless, the Ce content is a parameter the authors did not investigate which, however, could affect the catalytic and redox properties.

The authors themselves assert that “OSC depends mainly on composition. OSC increases linearly with ceria increase in CZ up to ~30-35%CeO2 and then remains practically 120 unchanged at 1.0-1.1 mM H2/g for materials with higher CeO2 content”

As a consequence, they should limit the study to one of the two commercial products because CZ20 contains 20%CeO2 and CZ40 40%CeO2.

Moreover, in the new Fig.5 a much higher crystallinity of CZ20 is evident looking at the intensity scale.

So, what is the reason for including CZ40 just creating confusion?

6. Calcination temperature. All CZ materials have been calcined at 700oC first (“fresh” materials). Additional calcination at 1000oC is wash coaters “standard” for CZ porosity characterization and evaluation of thermal stability. Ageing at 1100oC is typically used for CZ phase stability evaluation. Some companies may use lower or higher temperatures (from 950 to 1150oC).

All experiments have been done on materials aged at 1000°C. Materials aged at 1100°C appear only in Fig. 4. Please, explain in the text that jar milling and in addition to more severe aging conditions (1100°C) do not modify the crystalline structure.

7. Agree that SSA can’t be “aged” or “fresh”. Corrections made.

8. SSA-1000/6, etc. Changed to Sa after 1000oC for 6 hours.

9.     For pure CeO2 (cubic) 2Theta position of the 1^st peak is close to 28 degrees and for pure ZrO2 (cubic or monoclinic phase) is close to 30 degrees. CZ20 and CZ40 are solid solutions and the position of their first peak is between 28 and 30 degrees depending on CeO2 content (as well as on the amount and type of rare earth dopants). These compositions are not in PDF database. Added Figures with XRDs of milled and un-milled material for CZ20 and CZ40. The main idea of these graphs is to show that CZ materials withstand milling without phase disproportionation.

In the new Figs.4 and 5 the authors aligned the two patterns. Which was the error in the original version?

They should report the PDF of pure CeO2 and ZrO2 and explain the shift they observe with respect to pure compounds. Once again, if CZ40 is one of the materials investigated they should report if this shift is different depending on the composition.

10. TPR-H2. The use of TPR-H2 is typical for characterization of CZ materials for CZ manufacturers and wash coaters. 

If the authors do not want to carry out additional TPD-O2 experiments, they must at least explain that TPR-H2 can provide just a trend of oxygen mobility but it does not provide a direct measure of OSC and that temperatures in TPR analysis  can be very different from those required to oxidize soot.

11. The soot oxidation graph has been removed. Agree, CZ particle size should influence change of soot oxidation temperature.

12. Added description of CZ calcination and ageing conditions.
13. Added description of CZ impregnation with PGM.

Points 1. 4. 5. 7. 9. 11. 12. 13. OK

Author Response

Thank you very much for taking the time to review this manuscript again and making new valuable comments. Please find our responses below.

1. The manuscript structure has been changed.

The authors deeply modified the manuscript but one of the main issues still exist: they did not select only one compositions (CZ20 or CZ40). This is fundamental because the composition affects catalyst properties but it is not one of the parameters studied in this work.

The investigation of the catalytic performance of CZ materials as function of their chemical composition was not our objective.  This was done long ago, and I am sure that it’s difficult to add something new to already known. The main objective was to show that “fast oxygen mobility” effect (not reported before, new) depends only on milling conditions and doesn’t depend on CZ type. That is why results for two types of CZ materials have been presented. CZ20-type (or zirconium rich CZ) is used for Rh-catalysts and CZ40 for Pt/Pd-catalysts. CZ20 and CZ40 are designed for different applications and represent two main CZ groups in TWC (based on fundamental research carried out on CZ catalytic performance). We show that mechanochemical activation of both CZ compositions can result in the appearance of “fast oxygen mobility”.

1. CZ materials are known and widely used for decades – commodity products. Agree, that there is a lot of literature (and patents) on CZ materials. All fundamental research has been done in 90-th and early 2000. Our objective is very modest - to provide interesting data on fast oxygen mobility in CZ to the academy and industry community. As a manufacturing company we are not doing fundamental research but will be glad to work with the academy. Literature has been updated.

The authors added some more recent references.

The academic community is very glad to work with manufacturing companies as well. This is fundamental for applied research, however, both communities should try to cross data and results on a common basis and if people from industry choose to publish in journals mostly read by the academic community they should seek to adapt to the “academic rules”.

We hope that your valuable comments will help in “adapting academic rules”.

1. The aim of the work has been stated at the end of introduction.

The authors could stress the ease of the method to increase oxygen mobility which is the strength of the work.

As was stated, not all types of milling but only those under severe conditions could generate the effect. So, this is not easily achievable. Maybe this is the reason why this phenomenon has not been reported before by other researchers.

1. Added particle size analysis in “characterization” section with explanation of d50 meaning.

1. CZ20 and CZ40 are PIDC commercial products.

This was clear also in the original version of the manuscript. Nevertheless, the Ce content is a parameter the authors did not investigate which, however, could affect the catalytic and redox properties.

Ce content in CZ has been investigated, but with the different goal – to show that fast oxygen mobility is a generic effect and depends only on milling conditions. We added OSC values for un-milled and milled CZ20 and CZ40 (lines 219-220, 226-227, 231-233). Indeed, increase in CeO2 content in CZ from 20% to 40% results in OSC increase from 0.7 mM/g to ~1.0-1.05 mM/g. Theoretical OSC for CZ40 is ~1.30 mM/g. For CZ materials with CeO2 content higher than 30-35% not all CeO2 is redox active.  

The authors themselves assert that “OSC depends mainly on composition. OSC increases linearly with ceria increase in CZ up to ~30-35%CeO2 and then remains practically 120 unchanged at 1.0-1.1 mM H2/g for materials with higher CeO2 content”

As a consequence, they should limit the study to one of the two commercial products because CZ20 contains 20%CeO2 and CZ40 40%CeO2.

Can not agree with this suggestion (see what has been stated above).

Moreover, in the new Fig.5 a much higher crystallinity of CZ20 is evident looking at the intensity scale.

XRD patterns serve for illustration that even severe milling doesn’t affect crystal phase. XRD peak intensities are very similar in un-milled and milled samples of the same composition. It’s known fact that in Zr-rich CZ materials crystallites grow faster than in Ce-rich under similar ageing conditions. As a result, peak intensities for CZ20 are higher in comparison with CZ40.

So, what is the reason for including CZ40 just creating confusion?

The objective was to show that “fast oxygen mobility” effect depends only on milling conditions and doesn’t depend on CZ type. That is why results for two types of CZ materials have been presented.

1. Calcination temperature. All CZ materials have been calcined at 700oC first (“fresh” materials). Additional calcination at 1000oC is wash coaters “standard” for CZ porosity characterization and evaluation of thermal stability. Ageing at 1100oC is typically used for CZ phase stability evaluation. Some companies may use lower or higher temperatures (from 950 to 1150oC).

All experiments have been done on materials aged at 1000°C. Materials aged at 1100°C appear only in Fig. 4. Please, explain in the text that jar milling and in addition to more severe aging conditions (1100°C) do not modify the crystalline structure.

Ageing at 1100oC is necessary for CZ phase stability evaluation. Materials aged at 1100oC appear in Fig 4 and Fig 11.

The statement in the manuscript text is different – ”jar milling doesn’t change CZ20 crystal phase”. Crystal phase remains tetragonal as there is no peak splitting, no appearance of new peaks or peak shoulders formation after milling.

1. Agree that SSA can’t be “aged” or “fresh”. Corrections made.

1. SSA-1000/6, etc. Changed to Sa after 1000oC for 6 hours.

9. For pure CeO2 (cubic) 2Theta position of the 1^stpeak is close to 28 degrees and for pure ZrO2 (cubic or monoclinic phase) is close to 30 degrees. CZ20 and CZ40 are solid solutions and the position of their first peak is between 28 and 30 degrees depending on CeO2 content (as well as on the amount and type of rare earth dopants). These compositions are not in PDF database. Added Figures with XRDs of milled and un-milled material for CZ20 and CZ40. The main idea of these graphs is to show that CZ materials withstand milling without phase disproportionation.

In the new Figs.4 and 5 the authors aligned the two patterns. Which was the error in the original version?

There were no errors.

Figure 4 show XRD profiles for un-milled and jar milled CZ20. This graph was added to the revised manuscript.

Figure 5 – is TPR-H2 profiles for un-milled and jar milled CZ20. This graph was added to the revised manuscript.

Figure 11. The previous version had a graph with XRD profiles for steam jet milled CZ20 and CZ40. One of the reviewers suggested showing XRD profiles of un-milled and milled materials to get visual impression that milling doesn’t change phase. Four XRD profiles on one graph is too much, so, we made two graphs with XRDs for CZ20 and for CZ40 (milled and un-milled). There is no peak position shift for CZ20 and CZ40 after milling.

They should report the PDF of pure CeO2 and ZrO2 and explain the shift they observe with respect to pure compounds. Once again, if CZ40 is one of the materials investigated they should report if this shift is different depending on the composition.

Reporting PDF of pure ZrO2 and CeO2 is irrelevant to this research, as we don’t use or compare these oxides. Both studied CZ materials have tetragonal structure (already known fact) and their unit cell parameters are more relevant.

There is no XRD peaks shift after milling. Peak shift would indicate that chemical composition has been changed after milling. This does not happen.

I am really surprised that XRD profiles could make confusion. We made just a statement for Eiger milling that it doesn’t affect phase stability. We can do the same for steam jet milling (and remove Figure 11). But this statement should be in the manuscript as researchers working in this field will have questions about phase stability.

1. TPR-H2. The use of TPR-H2 is typical for characterization of CZ materials for CZ manufacturers and wash coaters. 

If the authors do not want to carry out additional TPD-O2 experiments, they must at least explain that TPR-H2 can provide just a trend of oxygen mobility but it does not provide a direct measure of OSC and that temperatures in TPR analysis  can be very different from those required to oxidize soot.

Lines 107-109. “TPR-H₂ allows to determine available oxygen storage capacity (OSC) and based on T_max position gives indication on CZ oxygen mobility”. We are aware that TPR-H2 can’t be used for direct measurement of oxygen mobility, and we didn’t make such a statement in the manuscript.