Carbon Aerogel-Supported Iron for Gasification Gas Cleaning: Tars Decomposition

Round 1

Reviewer 1 Report

In the manuscript, the authors prepared catalyst of Fe on carbon aerogel for toluene and naphthalene decomposition. The possible mechanisms for tar degradation were proposed. Toluene and benzene are reversible during toluene decomposition, and the activation energy for toluene decomposition is very high. The activation energy for naphthalene was estimated considering (and not considering) the deactivation effect. In both cases, the iron oxidation occurs and loss the catalytic activity. Here are some comments and questions.

1. Writing:
   1. It is better to have proper definition on abbreviation or terms, even though the terms may be well-known, such as Cmears in Mears criterion, space velocity, LHHW, MARI, MFC, etc.
   2. In Line 195, Figure 4, 4a, and 4b are not representing the reduction of conversion.
2. Why pellets have better Fe dispersion in Figure 2?
3. Is there any figure for "propylene is the main species detected" in Line 183? Without figures or data reported, it’s difficult to show that the statement is correct.
4. Is there any evidence to show that “toluene and benzene are reversible” in Line 176? The authors tried to use the spike of benzene in toluene decomposition as the evidence. However, the decrease in toluene with the spike in benzene in the first 2 hours implies the effectiveness in iron catalyst only survives for 2 hours. The zero benzene flow and the steady flow of toluene mean no decomposition, not the reversibility in toluene and benzene. In addition, considering the non-linear Arrhenius plot, very large activation energy, and iron oxides, the simple answer that the catalyst is not functional after 2 hours seems legit.
5. For naphthalene decomposition, the authors claimed that a sharp decrease at 665°C implies “a higher level of Fe oxidation” in Line 298. However, the TOF of 665°C in Figure 7 is similar to TOFs of other temperatures. If the statement is right, 665°C has iron oxidation, and has a similar TOF, it means Fe oxidation also occurs at other temperatures.
6. Thus, in Figure 8, it is very improper to fit Arrhenius plot by only 2 points, by saying that high temperature 665°C suffers oxidation so it can be excluded, as the “deactivation effect estimation” in Line 314.
7. In Line 353, the authors wrote “On the other hand, naphthalene decomposition over Fe/CAG-ps produces benzene as the main product, evidenced in a rise of benzene concentration fed together with naphthalene.” However, there is no figure, like Figure3, showing that benzene concentration rise with naphthalene.
8. In addition, for toluene decomposition Figure 3 of flow vs. time was plotted. What’s the reason for naphthalene decomposition in Figure 7, TOF was plotted against with time?

With the comments above, I recommend rejection to Catalysts. Thank you.

Comments for author File: Comments.pdf

Author Response

 

Dear Reviewer

On behalf of the authors, I would like to thank you for giving us the opportunity to send a revised version of our manuscript entitled “Carbon Aerogel-Supported Iron for Gasification Gas Cleaning: Tars decomposition”. According to your directions herein we provide a detailed reply to all the referee`s questions and suggestions. Finally, we want to thank you for your valuable comments, which have contributed to improve the quality of our manuscript.

 

With kind regards,

Corresponding Author

Prof. Dr. L.E., Arteaga-Pérez,

Laboratory of Thermal and Catalytic Processes (LPTC-UBB).

Chile

E-mail: [email protected] 

 

REVIEWER-1

In the manuscript, the authors prepared catalyst of Fe on carbon aerogel for toluene and naphthalene decomposition. The possible mechanisms for tar degradation were proposed. Toluene and benzene are reversible during toluene decomposition, and the activation energy for toluene decomposition is very high. The activation energy for naphthalene was estimated considering (and not considering) the deactivation effect. In both cases, the iron oxidation occurs and loss the catalytic activity. Here are some comments and questions.

1. Writing:

1. It is better to have proper definition on abbreviation or terms, even though the terms may be well-known, such as Cmears in Mears criterion, space velocity, LHHW, MARI, MFC, etc.

According to your recommendation we have included all the abbreviation´s definitions the first time each is used.

2. In Line 195, Figure 4, 4a, and 4b are not representing the reduction of conversion.

The authors agree with the referee. Accordingly, we have moved the text in line 195 to line 250 (after Fig. 4) and modified it as:

“On the other hand, steps - in the proposed mechanism fails to explain the reduction of TOF represented in Figure 4, thus it can be assumed that C atoms and traces of aromatic rings remain on the catalyst surface, which contribute to the C deposition via polymerization reactions. This last point was witnessed by Korus et. al. who detected the formation of polyaromatics from toluene under similar temperatures as those used here [30]”.

1. Why pellets have better Fe dispersion in Figure 2?

Both catalysts were prepared using incipient wetness impregnation. Our hypothesis is that this difference in the dispersion between pellets and fines is related to their different heat response during the reduction process. When the pellet is reduced, the heat transfer to the interior is lower than for unshaped catalyst (fines), leading to a milder reduction reaction and hence to the formation of smaller particles.

According to your comment we have included the following statement in the revised manuscript (L162 to L168):

“The slight difference between the dispersion measured for Fe/CAG and Fe/CAG-ps is attributed to the thermal response of the pellets during the reduction. In this case the heat transfer limitation to the pellet interior leads to a milder reduction, thus to a lower cluster size. However, regardless the shape of the catalysts both dispersions are lower than 10%, which allow us to rule out the effect of surface coordination on the catalytic activity for tar decomposition. In other words, any difference observed in the catalytic activity can be attributed only to the nature of the active phase”

Is there any figure for "propylene is the main species detected" in Line 183? Without figures or data reported, it’s difficult to show that the statement is correct.

We make the identification of this specie by direct sampling from the reaction line to a GC/MS. However, we do not prepare a figure with this data as it was only a chromatogram along with the resulting ionization pattern to confirm the identity of the detected peak (m/z = 41).

1. Is there any evidence to show that “toluene and benzene are reversible” in Line 176? The authors tried to use the spike of benzene in toluene decomposition as the evidence. However, the decrease in toluene with the spike in benzene in the first 2 hours implies the effectiveness in iron catalyst only survives for 2 hours. The zero benzene flow and the steady flow of toluene mean no decomposition, not the reversibility in toluene and benzene. In addition, considering the non-linear Arrhenius plot, very large activation energy, and iron oxides, the simple answer that the catalyst is not functional after 2 hours seems legit.

We do not agree with referee´s point. If so, then how to explain the result shown in Figure 3b, where only Benzene is fed and both Benzene and Toluene are observed and consumed simultaneously? If catalyst is deactivated by Toluene, it cannot show this performance. Moreover, Figure 3a (only Toluene is fed) shows that B is formed up to ~1h (usually catalyst surfaces need time to reach the steady state condition of coverage), then B is totally consumed, and Toluene concentration reach a steady state because only a fraction of Toluene react via Benzene formation and consumption, which suggest that the amount of catalyst (active surface) is not enough to convert all the Toluene fed, instead of the catalyst deactivation suggested by the referee. Furthermore, the molar flow of Toluene in steady state condition in the exhaust gases is lower than the molar flow in the feed, which indicate that Toluene decomposition is occurring on the catalyst surface stably during the last 10-12 hours of experiment.

1. For naphthalene decomposition, the authors claimed that a sharp decrease at 665°C implies “a higher level of Fe oxidation” in Line 298. However, the TOF of 665°C in Figure 7 is similar to TOFs of other temperatures. If the statement is right, 665°C has iron oxidation, and has a similar TOF, it means Fe oxidation also occurs at other temperatures.

Not necessarily Fe oxidation occurs at the other (lower) temperatures. Please, note that all TOF values shown in Figure 7 were calculated by assuming that all exposed Fe atoms in the catalyst are active for naphthalene decomposition. However, if a fraction of Fe is oxidated at 665ºC, then the amount of active exposed Fe is being super estimated leading to a lower-than-expected TOF value; if the real active exposed Fe would be able to be estimated, the TOF value would be higher at 665ºC than at 620ºC according to Arrhenius equation. On the other hand, the comparison with CO addition at 565ºC is not fair in this case and has another explanation.

1. Thus, in Figure 8, it is very improper to fit Arrhenius plot by only 2 points, by saying that high temperature 665°C suffers oxidation so it can be excluded, as the “deactivation effect estimation” in Line 314.

We agree with the reviewer, and according to your comment, in the revised manuscript we have avoided to report any values for the activation energies.

1. In Line 353, the authors wrote “On the other hand, naphthalene decomposition over Fe/CAG-ps produces benzene as the main product, evidenced in a rise of benzene concentration fed together with naphthalene.” However, there is no figure, like Figure 3, showing that benzene concentration rises with naphthalene.

According to your suggestion we have included a table in the Annexes, where the variation of the benzene flows is reported for all the explored conditions (Table A4). This new table was properly cited in the text.

1. In addition, for toluene decomposition Figure 3 of flow vs. time was plotted. What’s the reason for naphthalene decomposition in Figure 7, TOF was plotted against with time?

The initial idea was to show the reduction in the TOF with the operation time and temperatures. However, attending the referee´s suggestion we have updated the Figure 7 and now the naphthalene flow is plotted again reaction time.

 

 

Reviewer 2 Report

The manuscript devoted to the decomposition of different tar components is very interesting and presents very clearly the resultants obtained by the authors.

I recommend to publish this manuscript but I want some more complementary information:

1. Is the authors performed some TGA analysis to the spent catalysts?
2. Is the authors performed the catalytic tests for more day?
3. A regeneration step is necessary in catalysis to see if the catalyst recover his activity.

Author Response

Dear Reviewer

On behalf of the authors, I would like to thank you for giving us the opportunity to send a revised version of our manuscript entitled “Carbon Aerogel-Supported Iron for Gasification Gas Cleaning: Tars decomposition”. According to your directions herein we provide a detailed reply to all the referee`s questions and suggestions. Finally, we want to thank you for your valuable comments, which have contributed to improve the quality of our manuscript.

 

With kind regards,

Corresponding Author

Prof. Dr. L.E., Arteaga-Pérez,

Laboratory of Thermal and Catalytic Processes (LPTC-UBB).

Chile

E-mail: [email protected] 

REVIEWER-2

The manuscript devoted to the decomposition of different tar components is very interesting and presents very clearly the results obtained by the authors.

I recommend to publish this manuscript but I want some more complementary information.

Authors thanks your positive recommendation.

1. Do the authors performed some TGA analysis to the spent catalysts?

Authors performed TGA of the catalysts. These results were reported in a previously published paper [1], thus we did not include it here. In addition, we performed some TGA analysis to identify carbon deposition, but the technique was not suitable for such a low carbon quantities, thus we studied this phenomena by SEM-EDX. 

[1] Catalysts 2018, 8(9), 347; https://doi.org/10.3390/catal8090347.

 

2. Do the authors performed the catalytic tests during more days?

No, we performed the catalytic test only for 18 hours as it is presented in the manuscript. An interesting tetst could be the continuous operation until catalyst fail.

3. A regeneration step is necessary in catalysis to see if the catalyst recover his activity?

According to our observations, (18 hours of operation) the catalyst does not require for regeneration. However, for a longer operational time, we suggest the recovery of the surface by using a mild oxidizing gas to eliminate carbon deposition.

 

Reviewer 3 Report

The manuscript deals with the preparation, characterization and testing of novel iron-based catalysts for tar removal. The manuscript suffers from lacks avoiding its publication in the current form and it must be improved according to the comments below. In particular, several experimental tests are required in order to provide reliable kinetic results.

• Authors refer to tar removal from a syngas produced by gasification. However, CO2 and H2O, unavoidable components of a gasification stream, are not taken into account. Their co-feeding can significantly modify the reaction mechanism and tar conversion into lighter products.
• About kinetic measurements and Arrhenius plots. The Authors generally provide linear interpolation between two points; due to the experimental errors, this procedure provides unreliable results. At least four points are required to obtain a reliable interpolation. Accordingly, statistical parameters (as, RMSE and confidence intervals) must be provided.
• Lines 234-239. It should be underlined that a negative activation energy can be associated to an exothermic equilibrium reaction as rate-determining step (a common example is the homogeneous NO oxidation to NO2, showing an apparent negative activation energy). This hypothesis is more probable considering that the negative activation energy is observed at high temperature, where reactant adsorption (an exothermic equilibrium reaction) could be limiting.
• Section 4.9 is devoted to the description of the experimental procedure. Lines 442-446 should be moved to another section (I suggest, in the introduction)
• There are several typos and errors.

Author Response

Dear Reviewer

On behalf of the authors, I would like to thank you for giving us the opportunity to send a revised version of our manuscript entitled “Carbon Aerogel-Supported Iron for Gasification Gas Cleaning: Tars decomposition”. According to your directions herein we provide a detailed reply to all the referee`s questions and suggestions. Finally, we want to thank you for your valuable comments, which have contributed to improve the quality of our manuscript.

 

With kind regards,

Corresponding Author

Prof. Dr. L.E., Arteaga-Pérez,

Laboratory of Thermal and Catalytic Processes (LPTC-UBB).

Chile

E-mail: [email protected] 

 

 

REVIEWER-3

The manuscript deals with the preparation, characterization, and testing of novel iron-based catalysts for tar removal. The manuscript suffers from lacks avoiding its publication in the current form and it must be improved according to the comments below. Several experimental tests are required to provide reliable kinetic results.

• Authors refer to tar removal from a syngas produced by gasification. However, CO₂ and H₂O, unavoidable components of a gasification stream, are not taken into account. Their co-feeding can significantly modify the reaction mechanism and tar conversion into lighter products.

Authors agree with the referee on this comment. However, here we studied the decomposition and surface interaction of individual tar´s model molecules and the role of some species in the gasification gas on this reaction mechanism. This for sure will provide valuable information on gas cleaning but requires further analysis to validate their efficiency under quasi-real operation.

• About kinetic measurements and Arrhenius plots. The Authors generally provide linear interpolation between two points; due to the experimental errors, this procedure provides unreliable results. At least four points are required to obtain a reliable interpolation. Accordingly, statistical parameters (as, RMSE and confidence intervals) must be provided.

We agree with the reviewer. According to your comment and considering that this calculation is not of paramount importance to validate the hypothesis we have discuss in this paper, we decided to eliminate it from the revised manuscript.

• Lines 234-239. It should be underlined that a negative activation energy can be associated to an exothermic equilibrium reaction as rate-determining step (a common example is the homogeneous NO oxidation to NO2, showing an apparent negative activation energy). This hypothesis is more probable considering that the negative activation energy is observed at high temperature, where reactant adsorption (an exothermic equilibrium reaction) could be limiting.

The referee´s comment have sense; although, for homogeneous reactions this kind of changes are different. We revised our discussion on this, and another explanation might be that at higher temperatures and lower toluene inlet concentrations, the adsorption step become slower and consequently, the RDS. In fact, a RDS resulting in a negative activation energy is commonly associated to exothermic equilibrium reactions -like the adsorption step-, thus the behavior showed in Figure 4 can be also explained by a change, of this nature in the reaction pathways.

Considering your question, we have included a statement on that in the revised manuscript.

• Section 4.9 is devoted to the description of the experimental procedure. Lines 442-446 should be moved to another section (I suggest, in the introduction).

We eliminated this section as it is redundant with respect to lines 70 to 75.

• There are several typos and errors.

We carefully revised the document for eliminating typos and errors.

 

 

Round 2

Reviewer 1 Report

Now we have seen the fact based on Figure 3 that there is benzene in the first few hours and at the same time toluene is gone. One answer is that toluene is decomposed into benzene by the catalyst, and further benzene is consumed and converted back to toluene. The other answer is that the toluene is decomposed into benzene by the catalyst, and the catalyst becomes inactive (because of deposition?).

It is just surprising that the catalyst reduced the activation energy of toluene→benzene, and vise versa at the same time.

Reviewer 3 Report

The revised manuscript is suitable for publication