Co-Pyrolysis of Low-Density Polyethylene and Motor Oil—Investigation of the Chemical Interactions between the Components

Round 1

Reviewer 1 Report

The paper entitled “Co-Pyrolysis of low-density polyethylene and motor oil: investigation of the chemical interactions between the components” presents potentially interesting results but there are several aspects that require some clarification.

The firs one is what the improvements in relation to the previously published work that is described in reference 11, by the same authors, as all the data presented in section 3.1 is from that paper. Although the results have already been published, even this reproduction, which the authors justify as for clarity, is not clear. The authors state that the “results show a clear interaction between the components during their pyrolysis, leading to a stabilization of the mixtures”, indicated by “that the two maxima of the mass loss per unit of time are shifted to the right” but this is not clear from the figures and the authors should support this information by given the onset and maximum rate temperatures for the different samples. It seems clear that the mass loss associated with the oil moves to somewhat higher temperatures, but this is not at all clear for LDPE.

A second aspect, which cannot be clarified by reference 11 either, is the chemical nature of the motor oil that was used. In this paper the authors aim at identifying the species that are released in the thermal processing of the oil/LDPE mixtures and knowing the chemical nature of the materials involved is very important.

This is particularly important in an attempt to understand the observation that “during the pyrolysis of the motor oil, the evolved species are mostly detected at higher temperatures (around 400°C) than the temperatures at which the mass loss is observed with TGA (around 300°C)”. It is very relevant to know if the oil components decompose or just evaporates.

As the authors used MS to identify the components observed in gas-phase, just indicating the m/z ratio is rather limited and some sort of actual identification based on the mass spectrum should be given. The authors say that “the most present m/z ratios are 40, 55 and 70” and that these are “indicators of linear aliphatic hydrocarbons possessing 3, 4 and 5 carbon atoms”. The mass spectra that are obtained should be clarified as well as the reason why the authors suggest that the aliphatic hydrocarbons should be just linear when departing from a highly branched polyethylene.

This is also relevant to interpret section 3.4, where the presence of alcohols in the motor oil is assumed. Since detailed analysis of the products occurring after pyrolysis, the composition of the feedstock oils should be given and not just indicating that “that lubrication oils may contain both acids and alcohols” based on literature; were alcohols detected in the oils or in the pyrolysis of the oils? Note that the second equation in line 248 is not right as the unpaired electron seems to be lost from the left side to the right side of the equation.

In conclusion, the results are potentially interesting, but the discussion seems to not to be sufficiently supported by the data presented and the discussion should also be somewhat more supported, not only in data (some of which is missing, like the composition of the motor oils) but also in the literature – the authors mention reference 9 in the introduction but in the discussion there is no mention of literature data. There are more literature references related to co-processing of polyethylene and motor oil and other hydrocarbons.

Author Response

Dear Reviewer, 

Many thanks for your valuable comments, that we have tried to address in this revised version of the manuscript. We think that they helped to make the manuscript way better. We apologize for the delay but, due to the sanitary crisis, we are all working from home as the university is mainly closed (with limited access to the lab).

Here are our answers to your comments / questions :

The first one is what the improvements in relation to the previously published work that is described in reference 11, by the same authors, as all the data presented in section 3.1 is from that paper. Although the results have already been published, even this reproduction, which the authors justify as for clarity, is not clear. The authors state that the “results show a clear interaction between the components during their pyrolysis, leading to a stabilization of the mixtures”, indicated by “that the two maxima of the mass loss per unit of time are shifted to the right” but this is not clear from the figures and the authors should support this information by given the onset and maximum rate temperatures for the different samples. It seems clear that the mass loss associated with the oil moves to somewhat higher temperatures, but this is not at all clear for LDPE.

The results presented in figure 1 and 2 (the previous figure 1 has been split in two) have indeed already been published. However, for the sake of clarity, we reproduce them here, especially since they were not at all analyzed from the point of view of the interaction between the components in our previous work, which concerned the development of a co-pyrolysis device for used plastics and oils, in order to produce an alternative fuel.

We agree that the results presented in section 3.1 would have deserved to be better commented in the first version of the paper, in particular from the point of view of the onset temperatures and positions of the peaks, to better support our message. Therefore, we have reworked the last paragraph of section 3.1, according to your comments, and hope it is clearer now.

A second aspect, which cannot be clarified by reference 11 either, is the chemical nature of the motor oil that was used. In this paper the authors aim at identifying the species that are released in the thermal processing of the oil/LDPE mixtures and knowing the chemical nature of the materials involved is very important. This is particularly important in an attempt to understand the observation that “during the pyrolysis of the motor oil, the evolved species are mostly detected at higher temperatures (around 400°C) than the temperatures at which the mass loss is observed with TGA (around 300°C)”. It is very relevant to know if the oil components decompose or just evaporates.

According to this comment, we have added a new section in the paper dedicated to the composition of the oils (section 2.2.), and we have tried to use at best the results presented in this new section in the rest of the paper, to support our message (especially in section 3.4 and 3.1). We have run several tests on the oils : HP mass spectrometer, FTIR, CHN and X-ray fluorescence (access to the lab was limited due to the university being close for the moment). Unfortunately, they do not give a precise overview of the chemical compounds present in the oils (except for the elemental analysis), due to their complexity. But, at least, they provide a useful information to the reader.  

Regarding the sentence: “during the pyrolysis of the motor oil, the evolved species are mostly detected at higher temperatures (around 400°C) than the temperatures at which the mass loss is observed with TGA (around 300°C)”, it was badly formulated and we think that this temperature difference is due to an artefact. We have rewritten the paragraph. 

As the authors used MS to identify the components observed in gas-phase, just indicating the m/z ratio is rather limited and some sort of actual identification based on the mass spectrum should be given. The authors say that “the most present m/z ratios are 40, 55 and 70” and that these are “indicators of linear aliphatic hydrocarbons possessing 3, 4 and 5 carbon atoms”. The mass spectra that are obtained should be clarified as well as the reason why the authors suggest that the aliphatic hydrocarbons should be just linear when departing from a highly branched polyethylene.

You pinpoint exactly one of the main issues the combination of gas analysis using mass spectrometry and TGA has: one tries to identify using time based spectra which mass fractions show the highest intensity. As the electron impact is rather high in these analyzers, a mass fraction for example of 40 is preceded by all sorts of other fragments having one or more hydrogen or carbon atoms less as these are fractionated. Although its shortcomings, due to its affordability, TGA-MS is a widely spread instrument to detect decomposition chemistry, or at least to identify the produced gaseous particles. It was entirely in our purpose to discuss these results to avoid any misunderstanding towards others in the field that caution should be taken while interpreting these results although commonly used and that other instrumentation is much more appropriate. As such, this analysis is far from redundant, but it is clear that the message was a bit lost in the last paragraph. Therefore, the phrasing at the end of the paragraph 3.2 has been changed to reinforce this message. Moreover, we have dropped the "linear" in the sentence you mention in your remark as it is true that the results do not allow affirming that the detected hydrocarbons are linear.  

This is also relevant to interpret section 3.4, where the presence of alcohols in the motor oil is assumed. Since detailed analysis of the products occurring after pyrolysis, the composition of the feedstock oils should be given and not just indicating that “that lubrication oils may contain both acids and alcohols” based on literature; were alcohols detected in the oils or in the pyrolysis of the oils?

Unfortunately, none of the techniques used in section 2.2 (or during the pyrolysis of the oils) did allow identifying alcohol or acid groups in the oils. Such detection unfortunately seems difficult due to the complexity of these oils, as well as the fact that these compounds are undoubtedly present in small quantities. We have reformulated section 3.4 (first paragraph) so as to be cautious and not positively advance a mechanism for the formation of aldehydes. We leave several ideas open to explain the formation of aldehydes and we support our proposition with two references from the literature which have observed similar behaviors.

Note that the second equation in line 248 is not right as the unpaired electron seems to be lost from the left side to the right side of the equation.

Indeed, it has been corrected. Many thanks for spotting this.

In conclusion, the results are potentially interesting, but the discussion seems to not to be sufficiently supported by the data presented and the discussion should also be somewhat more supported, not only in data (some of which is missing, like the composition of the motor oils) but also in the literature – the authors mention reference 9 in the introduction but in the discussion there is no mention of literature data. There are more literature references related to co-processing of polyethylene and motor oil and other hydrocarbons.

According to your comment and the elemental analysis of the oils, we have tried to improve section 3.4 (see our answer two comments above)

Best regards,

Benoit Haut, on behalf of the authors

Reviewer 2 Report

Dear Authors,

in my opinion this is a well designed and written manuscript containing interesting results.

Here below you will find my suggestion to further improve the quality of the manuscript.

Thermogravimetry coupled with Fourier Transformed InfraRed (TGA-FTIR) experiments were also realized, but they did not bring any valuable additional information on these interactions.

If TGA/FTIR analysis was performed some results should be reported. For instance chlorine derivatives should be clearly revealed in oils, if present. Otherwise, if any result is shown about TGA/FTIR, please delete this sentence.

Formation of aldehydes: the formation of aldehydes by pyrolysis of LDPE has been reported by Shah et al., Energy Conversion and Management 51 (2010) 2791–2801, DOI: 10.1016/j.enconman.2010.06.016. It could be useful for the Authors to read cite this manuscript.

Line 7. I suggest to mention also the thermal stabilization effect of inorganic materials on the lignocellulose fraction in construction and demolition waste with respect to furniture waste (Castaldo et al., Polymers 2019, 11, 1604; DOI:10.3390/polym11101604)

Line 8. I suggest to mention also the effect of compatibilizing agents that induce a thermal stabilization of plastic and cellulose in composites (Avella et al., J Therm Anal Calorim (2010) 102:975–982, DOI 10.1007/s10973-010-0836-3)

Line 92 and 93 Titles of Section 3 and subsection 3.1 are repeated two times. Please delete.

Line 122 -124: Figure 1 caption “Left hand side: 45 weight% mixture of LDPE and motor oil. Right hand side: 67 weight% mixture of LDPE in used oil. Both mass and mass loss per unit of time are presented as functions of the temperature”.

1. a) I suggest to use “Left column” and “Right column” instead of “left hand side” and right hand side”.
2. b) “45 weight% mixture of LDPE and motor oil” is not clear. Is it “mixture of LDPE (45 wt%) and motor oil (55 wt%)”? The same for LDPE/used oil.
3. c) Mass loss per unit of time is not useful as the heating rate is not reported. Please add the heating rate or please report the bottom graphs in Mass loss (wt% / °C). In any case, I suggest to replace % with wt% in the y axis.

The elemental analysis of the chars of oil and used oil at the end of TGA could be useful to have an indication on the different composition of the two materials in terms of inorganics (metal oxides). Indeed, the presence of inorganic components (metals, metal oxides) in the oils could also affect the pyrolysis of LDPE because metal oxides could act as catalysts (Kahn et al., Journal of Analytical and Applied Pyrolysis, 2017, 124, 195-203, DOI: 10.1016/j.jaap.2017.02.005).

Author Response

Dear Reviewer, 

Many thanks for your valuable comments, that we have tried to address in this revised version of the manuscript. We think that they helped to make the manuscript way better. We apologize for the delay but, due to the sanitary crisis, we are all working from home as the university is mainly closed (with limited access to the lab).

Here are our answers to your comments / questions :

"Thermogravimetry coupled with Fourier Transformed InfraRed (TGA-FTIR) experiments were also realized, but they did not bring any valuable additional information on these interactions." If TGA/FTIR analysis was performed some results should be reported. For instance chlorine derivatives should be clearly revealed in oils, if present. Otherwise, if any result is shown about TGA/FTIR, please delete this sentence.

You are right. This sentence has been deleted. 

Formation of aldehydes: the formation of aldehydes by pyrolysis of LDPE has been reported by Shah et al., Energy Conversion and Management 51 (2010) 2791–2801, DOI: 10.1016/j.enconman.2010.06.016. It could be useful for the Authors to read cite this manuscript.

Indeed, thanks for this valuable reference. We have added it as a support to the discussion in section 3.4.

Line 7. I suggest to mention also the thermal stabilization effect of inorganic materials on the lignocellulose fraction in construction and demolition waste with respect to furniture waste (Castaldo et al., Polymers 2019, 11, 1604; DOI:10.3390/polym11101604)

Line 8. I suggest to mention also the effect of compatibilizing agents that induce a thermal stabilization of plastic and cellulose in composites (Avella et al., J Therm Anal Calorim (2010) 102:975–982, DOI 10.1007/s10973-010-0836-3)

These two references have been added in the introduction. Thanks for providing them.

Line 92 and 93 Titles of Section 3 and subsection 3.1 are repeated two times. Please delete.

It has been done. Thanks for spotting this. 

Line 122 -124: Figure 1 caption “Left hand side: 45 weight% mixture of LDPE and motor oil. Right hand side: 67 weight% mixture of LDPE in used oil. Both mass and mass loss per unit of time are presented as functions of the temperature”.

1. a) I suggest to use “Left column” and “Right column” instead of “left hand side” and right hand side”
2. b) “45 weight% mixture of LDPE and motor oil” is not clear. Is it “mixture of LDPE (45 wt%) and motor oil (55 wt%)”? The same for LDPE/used oil.
3. c) Mass loss per unit of time is not useful as the heating rate is not reported. Please add the heating rate or please report the bottom graphs in Mass loss (wt% / °C). In any case, I suggest to replace % with wt% in the y axis.

According to your different comments, we have improved Figure 1 and its caption (and, actually, we have splitted Figure 1 in two).

The elemental analysis of the chars of oil and used oil at the end of TGA could be useful to have an indication on the different composition of the two materials in terms of inorganics (metal oxides). Indeed, the presence of inorganic components (metals, metal oxides) in the oils could also affect the pyrolysis of LDPE because metal oxides could act as catalysts (Kahn et al., Journal of Analytical and Applied Pyrolysis, 2017, 124, 195-203, DOI: 10.1016/j.jaap.2017.02.005).

We have added a new section in the manuscript about the oil compositions, including the elemental analysis (see section 2.2). It helped us to further discuss our results (see section 3.4), notably by pointing out the presence of oxidized metals in both oils, oxidized metals which have been shown to contribute to the formation of aldehydes during the catalysis of LDPE in a previous study (actually, it is the paper from Shah et al. that you mentioned above).

Best regards,

Benoit Haut, on behalf of the authors

Reviewer 3 Report

Dear  Authors

After studying this paper, I state the minor revision. Some suggestion are below:

• Line 92 and 93 is repeated
• Have you done experiments with different used oil? It is known that waste oil can have different properties.
• Quality of figures should be improved. The descriptions on the axes are of a different style.

Author Response

Dear Reviewer, 

Many thanks for your valuable comments, that we have tried to address in this revised version of the manuscript. We think that they helped to make the manuscript way better. We apologize for the delay but, due to the sanitary crisis, we are all working from home as the university is mainly closed (with limited access to the lab).

Here are our answers to your comments / questions :

Line 92 and 93 is repeated

Indeed, thanks for spotting this. It has been corrected.

Have you done experiments with different used oil? It is known that waste oil can have different properties.

Our used oil origins from an as wide as possible sourcing from very different usages in different types of motor blocks (as is stated in paragraph 2.1: "Used oil is a mixture of oils with a large variety of lifetime in different motor blocks. It has therefore undergone stress, friction, and exposition to high temperatures of unknown nature"). Therefore it is not simply a used oil of a certain kind, which would have indeed made the necessity of testing different used oils. We rather consider this used oil as the prototype of dirty oil the recyclers do obtain after passing different garages and waste fields. We have done tests with this used oil collected at different times and they still show the same behavior. Also, we show in this article that the same stabilization behavior is obtained with a new oil; so we think the point is very general.

Quality of figures should be improved. The descriptions on the axes are of a different style.

The quality of the figures has been improved in this new version of the manuscript (no longer in Word, but in LateX). Regarding the axes, we have tried to improve their homogeneity (which is not always easy as some of them origin directly from the analysis software on which we performed the analysis itself).

Best regards,

Benoit Haut, on behalf of the authors