Plasma-Assisted Gasification of Cellulose via Dielectric Barrier Discharge

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Line 225: the mist like volatile products that are trapped in the plasma region, contributes to the mass yield 
with increasing DBD power as the volatile products are decomposed.
The reason for the volatile products to be trapped at the plasma region was not elaborated clearly. 
With no DBD activated, there are no electrons in the plasma region.
How will the electrons attach to the mist particles and get captured in the electric field, when there is no electric field from the DBD? Nearly 50% or more of mass is trapped within this mist particles.

Line 295 and Figure 7: about 10 to 50% of mass (depending on the plasma power) are not accounted for, due to condensation at the quarts tube walls. Was there a difference in mass of the quarts before and after the experiments? This could be a heavy components of tar, maybe polycyclic aromatic hydrocarbons or higher ringed-aromatics that can only be devolatilized at temperatures above 600degC. This temperature which is higher than the DBD plasma (lower than 240degC). In terms of identifying the solvents, maybe try Sulfolane, which is used commercially to extract aromatics.

Overall, the article is scientifically sound, but requires further experiments and elaboration to address gaps in the data. The application of the research outcome should be made clear as well. The problem statement seemed to be the long gasification time of 92h at 50degC, and the IR+DBD gasification will reduce the time to 3 min. The flow rate, if not mistaken, was only 0.5 LPM, not high enough to prove the concept at pilot or commercial scale. 

 

Author Response

Thank you very much for your kind and thoughtful comments. We have revised the manuscript according to your comments and provided our response below.

1. Line 225: the mist like volatile products that are trapped in the plasma region, contributes to the mass yield with increasing DBD power as the volatile products are decomposed. The reason for the volatile products to be trapped at the plasma region was not elaborated clearly. With no DBD activated, there are no electrons in the plasma region. How will the electrons attach to the mist particles and get captured in the electric field, when there is no electric field from the DBD? Nearly 50% or more of mass is trapped within this mist particles.

As you pointed out, even without DBD plasma, about half of the mist-like products condensed inside the quartz tube, simply because levoglucosan has a high boiling point (385 °C). This is not due to electrostatic trapping by DBD plasma. To clarify this point, the boiling point information was added to the relevant section (line 225). To emphasize that electrostatic trapping does not occur without DBD plasma, we added an explanation in lines 250–252 of the manuscript. In addition, we replaced the term "electrostatic separation" with the more commonly used expression "electrostatic precipitation (ESP)" (line 247) and updated the corresponding reference [27] accordingly.

 

2. Line 295 and Figure 7: about 10 to 50% of mass (depending on the plasma power) are not accounted for, due to condensation at the quarts tube walls. Was there a difference in mass of the quarts before and after the experiments? This could be a heavy components of tar, maybe polycyclic aromatic hydrocarbons or higher ringed-aromatics that can only be devolatilized at temperatures above 600degC. This temperature which is higher than the DBD plasma (lower than 240degC). In terms of identifying the solvents, maybe try Sulfolane, which is used commercially to extract aromatics.

After the experiment, the quartz tube was washed with methanol and dried, and its weight was precisely measured using an analytical balance. The difference from the pre-experiment weight was defined as the methanol-insoluble fraction, which is shown in Figure 7. As shown in Figure 6, this methanol-insoluble fraction appeared brownish in color, but as indicated in Figure 7, its mass was not very large. Even after accounting for the methanol-insoluble fraction, the methanol-soluble fraction (washed from both the gas bag and quartz tube), and the gas-phase products, there remained mass gaps of approximately 10–50%. Since the experimental system was closed, the total mass should theoretically sum to 100%, but this was not the case.

This mass gap is one of the challenges in cellulose pyrolysis and we have not yet identified the products responsible for the observed mass gap, as described in lines 301–302. This issue of mass gaps has received little attention in previous studies, despite its potential significance. The missing mass may plausibly be due to intermediate compounds that exist in the gas phase of the gas bag, which are not well extracted by solvents and yet are not small enough to be detected by micro-GC. Although this remains a hypothesis, we added this point to the manuscript in lines 304–307.

In this study, we refer only to the mass of the methanol-insoluble fraction (tar and coke), and no chemical analysis of it was conducted. Your suggestion to use sulfolane for extraction is a good idea, and we would like to consider applying it in future work to analyze this methanol-insoluble fraction.

 

3. Overall, the article is scientifically sound, but requires further experiments and elaboration to address gaps in the data. The application of the research outcome should be made clear as well. The problem statement seemed to be the long gasification time of 92h at 50degC, and the IR+DBD gasification will reduce the time to 3 min. The flow rate, if not mistaken, was only 0.5 LPM, not high enough to prove the concept at pilot or commercial scale.

As mentioned above, elucidating the mass gaps is not straightforward; we regard it as a broader and fundamental challenge related to cellulose pyrolysis. Further investigation will be necessary to address this issue.

Regarding the application of our research outcome, as you pointed out, a key point lies in the enhancement of the gasification rate by applying plasma treatment to volatile products from cellulose pyrolysis. The sample amount of 0.1 g and the gas flow rate of 0.5 LPM were chosen for fundamental studies at the laboratory scale. We are also interested in future scale-up studies toward practical implementation. To emphasize these points, we added explanatory sentences in lines 342–349 and 367–370.

 

As an additional revision not related to your comments, the arrow indicating the IR heating period in Figure A2(b) was incorrect and has been replaced with the correct version.

Reviewer 2 Report

Comments and Suggestions for Authors

In this manuscript, the authors present a present a interesting study on cellusose gasification assisted by DBD plasma discharge. I find this work timely, the investigation of plasma assisted gasification of waste is still scarce and much needed.

The discussion of results is clear and insightful. I recommend that the paper can be considered for publication in hydrogen  after undergoing the revisions suggested.

1) A Physicochemical characterization of the samples in terms elemental composition, calorific value, moisture content would be useful to evaluate the gasification process. A thermogravimetric analysis (TGA) combined with differential scanning calorimetry (DSC),would also be useful  to  provided insights into the thermal stability and decomposition of the samples.

2) The authors refer to the presence of moisture content in the chromatograms, which is presumably coming from the air contamination in the micro-GC system. It could be also due to moisture content in the samples. What is moisture content in the samples after the drying process? How long was the drying process?

Author Response

Thank you very much for your kind and thoughtful comments. We have revised the manuscript according to your comments and provided our response below.

1. A Physicochemical characterization of the samples in terms elemental composition, calorific value, moisture content would be useful to evaluate the gasification process. A thermogravimetric analysis (TGA) combined with differential scanning calorimetry (DSC), would also be useful to provided insights into the thermal stability and decomposition of the samples.

The Whatman cellulose used in this study contains >98% α-cellulose and <0.007% ash, making it approximately pure cellulose. We added this information to subsection 2.1 (line 95). The elemental composition of pure cellulose is approximately (C₆H₁₀O₅)ₙ and its lower calorific value is around 17 MJ/kg; since these properties are well known, we considered it unnecessary to include them in the manuscript. As for the moisture content, it is negligible after drying, as described in our response to your next comment.

As you pointed out, combined thermal analysis is a powerful tool for elucidating pyrolysis mechanisms. We are therefore actively studying the pyrolysis of cellulose and wood using TG-DSC and TG-MS, and some results will be submitted soon. However, these analyses mainly target primary pyrolysis and are beyond the scope of the present study, which focuses on plasma-assisted secondary reactions occurring in the gas phase.

 

2. The authors refer to the presence of moisture content in the chromatograms, which is presumably coming from the air contamination in the micro-GC system. It could be also due to moisture content in the samples. What is moisture content in the samples after the drying process? How long was the drying process?

The cellulose sample was dried to constant weight in an oven at 105 °C. The drying duration was overnight, which was sufficient for the sample to reach a constant mass. ISO 3130 defines this drying process as the standard method for preparing an absolutely dried sample. We added this explanation to subsection 2.1 (lines 98–100) to emphasize that the cellulose used in this study was well dried. However, even this absolutely dried sample may retain only residual bound water (typically less than 1%). In practice, complete removal of all water from cellulose is nearly impossible.

In addition, as you pointed out, the water detected in the micro-GC originates not only from air contamination but also from residual water in the cellulose sample and from water generated during pyrolysis. We added this information to the manuscript (line 177).

 

As an additional revision not related to your comments, the arrow indicating the IR heating period in Figure A2(b) was incorrect and has been replaced with the correct version.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Responses to 1. about trapped mist volatile products: authors have addressed the comments adequately with elaboration and clarification of their statements.

Responses to 2. about 10 to 50% of mass not accounted for: based on own experience, gas bags are probably not the best way to transfer the gas samples to the GC. Losses through the bags might occur. Even with custom made 'gas bottles'. Direct piping is the best way the reviewer understands if lab environment/arrangement does not permit this to happen.

Responses to 3 about scale-up: addressed in the manuscript.

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

I recommend to accept the article in its current form.