Recovery of Iron from Copper Slag Using Coal-Based Direct Reduction: Reduction Characteristics and Kinetics

Round 1

Reviewer 1 Report

The article deals with an interesting topic of direct reduction of Fe phases in Cu slags.

Unfortunately, in its present form, it contains a large number of significant errors and understatements that make it impossible to accept. Basic information about the applied research methods (XRD, SEM-EDS, chemical analyzes) is completely lacking. In many cases, authors operate with scientific facts without supporting them with appropriate references.

Figures and tables require improvement, especially in terms of quality or elements they contain (lack of explanations or incorrect explanations in the captions, undescribed EDS peaks, etc.).

The entire publication requires linguistic verification by a native speaker. Together with the lack of an orderly style and some factual errors, these make it very difficult to familiarize yourself with the part of the results that matter.

Summing up, the article might be resubmitted only after its comprehensive analysis and correction.

Precise comments are within the attached file. 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have undertaken a study into the use of local copper smelter slag (containing some 48% iron) to produce metallic iron one assumes for steelmaking. No indication has been included in the study as to the possible economics of the use of this route for the production of iron as commercially traded iron ore contains 62-64% iron (from Brazil and Australia) and iron ores containing less than about 58% iron have difficulty finding a market. This information should form part of the Introduction which could be shortened.

The actual experimental study appears sound although there are some additional data that should be included, and these will be listed below. Once the technical aspects of the paper have been sorted out, there will need to be substantial editing required.

Some specific points that the authors should consider are:

1. Line 33 and 35, "tons" are mentioned. There are three "tons", the "long ton" containing 2240 lb, the "US or short ton" containing 2000 lb and the metric "tonne" containing 1000 kg. Which is meant?
2. Line 41/42. References 12 and 13 both refer to the use of slag in pavements.
3. The X-ray conditions under which Fig.1 was produced must be given. This reviewer suspects that the radiation was Cu K alpha with a Ni filter. XRD gives data on the crystal structure of the target material but very poor information on glassy material. Was the copper slag used in this study slow cooled or quenched? 
4. The thermodynamic data used to generate Fig. 6 must be referenced. It would appear that the assumption has been made that the reaction and product hat capacities remains constant and this is incorrect. The ordinate axis label  should be corrected.
5. Fig 4 shows a pelletization and a drying stage. What was the diameter of the pellets used in the experiments and under what conditions was the drying conducted? For example, were conditions sufficient to remove the water of constitution in the bentonite and dry the coal without the loss of coal gas and hydrocarbons? Were thw moisture contents of the pellets checked after drying?
6. The melting point of beta-fayalite is about 1220 C (Kubaschewski O et al. "Materials Thermochemistry" Pergamon Press 1993) so any reaction between the solid ferrous silicate and the fixed carbon in the coal would be extremely slow (requiring solid/solid reaction). At temperatures below the melting point therefore any reduction would need to be due to reaction with the volatile coal hydrocarbons or carbon monoxide. The presence of any  water would contribute. Fig 11 may need to be reinterpreted.
7. The statements made on lines 166 and 167 are contradictory and must be corrected.
8. What are the ordinate axes of the two diagrams shown as Fig 10?
9. What are the ordinate axe label on Fig 12? Is it ln k or ln K?
10. Finally, some 22 of the 34 references given are to Chinese authors. Are all necessary and is the referenced material generally available to the wider audience?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The article is devoted to the problem of obtaining iron from production wastes. This problem is urgent because of the increase in the amount of production wastes and their involvement in processing.

There are several questions and clarifications on the article:

1. Table 1 shows the chemical composition of the main components. The amount is 79.52%. What else is included in the composition? Since 20 percent is quite a lot and these components can have a significant impact on further processing.
2. The title of the article contains magnetic separation, in Fig. 4 also shows the operations of crushing, grinding and magnetic enrichment. But data on these operations are not given in the article. This information should be added.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors used reduction roasting to recover Iron from Fyalite slag. Slag was a product of copper extraction.  The authors obtained thermodynamic and kinetic parameters by experiments and formalized three kinetic models based on theoretical knowledge. 

1. The title is wrong. The author didn’t mention magnetic separation in the manuscript and therefore magnetic separation should be removed from the title. If magnetic separation is a key difference in iron recovery than an entire section in methods and an entire section in results should be added.
2. Figures must be better organized and much better described. A clear distinction between experimental points and calculated results should be made.
3. The statistical model should be provided in order to assess the reaction mechanism(s).
4. English requires substantial improvement, both grammar and style.

Line 46: Iron isn’t valuable metal.

Line 51: Sentence is unreadable, please change or delete.

Line64: Statement isn’t factual. Remove entire sentence

Line 67: What is ‘DRI’?

Figure 2: What is the peak between Ca and Fe at ~0.6Kev?

Table 1: Please, cheerfully check EDS results. EDS technique can’t give bonding characteristics of the oxygen compounds. Report elemental composition rather than oxides. Also, TFe should be defined in the table.

Line 84: Paragraph from line 84 to line 98 describes results and therefore doesn’t belong to the materials and methods section.

Table 2: sum is 100.11wt%

Figure 5: This is not a chart; it looks like a simplified cross-section of the furnace.

Line 174: Please specify clearly what is ‘the Fe grade’ and what you consider as ‘Fe recovery’.

Line 185: Sentence makes no sense, please fix typos.

Line 237: Terms ‘fitting’ and ‘fitted’ are confusing. Straight lines were calculated.

Figure 12: Figure requires more clarification. What is the physical meaning of the results? What are units at y axis if any? If straight lines were calculated than information regarding linear correlation (R2) would be interesting.

Figure 13: What are the various kinetic models? Do you refer to EQs 2-4?

Table 5: same as above, not clear.

Line 248: How did you make this conclusion? Which statistical model did you use to back up statement?

Lines 265 to 277 and Figure 14. Figure 14 makes no sense whatsoever. I am kindly suggesting authors remove Figure 14 and Table6. If you keep Fig.14 please provide an extensive and detailed explanation.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have significantly improved the manuscript, which is impressive considering the short period, but there are still some errors that need correction before it can be finally published.

1.) There is still not enough information about sampling. How many samples were collected? What was their weight etc?
2.) The most important comment: figures still are not of the best quality. I do not mean resolution here but how they were prepared. The authors should compare them with figures in some high-impact publications. For example Figure 2 should be separated into four parts (a, b, c, d) which should be explained in caption. On the SEM image, it is impossible to see the scale. The same applied to figure 2 (and figure 10) where we do not see the scale at all, and the first two images are not explained in the caption.
3.) line 115 “As shown in Fig.1, the crystal minerals have some diffraction peaks.” You shouldn’t say that - it is obvious.
4.) Line 279 Authors misunderstood my concerns about the “slag phase”. I meant that you shouldn’t use such a term - you need to be precise. Is it a pyroxene, Ca-feldspar, or amorphous phase?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have made a very great improvement both to the technical content and the readability of the paper. They are to be commended. However they should review the manuscript again taking account of the comments made.

1. This reviewer commented previously on the use of "tons". The authors confirmed that it was "metric tons' or "tonnes" that was meant yet in most places "tons" were still used. This reviewer would prefer to see "tonnes" used throughout.
2. Line 120. Table 1. If the slag contains 68.20% by weight FeO, this amounts to 53.0% Fe not 47.81% as shown. This needs correcting.
3. Comment is made in line 123 that copper can be recovered. An indication should be given as to how this can be managed or any mention of copper recovery removed.
4. The authors may wish to recalculate the slag basicity remembering that FeO is a basic oxide. This reviewer believes the slag to be nearly neutral or slightly basic.
5. Sulfur in the coal is present probably as FeS₂ which is converted to Fe₂O₃ in the ash. The authors may wish to revise Table 2 as currently the sum of all items shown   comes to 100.
6. The fixed carbon is shown as 52.59% while in Table 2 as 52.56%. A correction needs to be made.
7. Fine dried coal forms part of the green pellets (line 150). What is meant by the statement on line 155 that "green pellets were roastedby adding coal"?
8. It would make sense for the subscripts on the Standard Gibbs Energies of Reaction for Equations 11-18 to be the same as the equation numbers rather than 1-8.
9. Figure 6 shows the Standard Gibbs Energies of Reaction for the various reactions of interest to be a linear function of absolute temperature. The integration of the van't Hoff Equation shows that the only way this can occur is for ΔC_p to be zero. This generally is not true so at best the lines shown are approximations. This needs to be stated.
10. Have the authors considered coal volatiles acting as reductants? At what temperature did the coal start to lose volatiles?
11. Finally, a useful piece of information that needs to be given is a typical iron product assay so that potential use could be judged. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

The authors have taken account of the comments and questions raised by this reviewer and generally made appropriate corrections to this draft of their paper. They should be commended for this. However there are still some aspects of the paper that need attention and these are listed below:

1. In line 3 of the Introduction, "ton" is still used. This needs to be corrected.
2. The authors in their response to this reviewer's question on the slag basicity indicated that they considered that it was the basicity of the remaining slag at the end of the iron reduction process that was important and hence the need for the addition of lime into the green pellets. This needs to be clearly stated. Unless this is done, it would appear that it is the basicity of the feed copper slag that needs adjusting. When one calculates the basicity of the copper slag, the iron oxide must be taken into account as a basic oxide. 
3. The authors in their response to this reviewer's comment on the thermodynamic equations relating Standard Gibbs Free Energy changes to absolute temperature stated "When ΔG^o is greater than zero, the reaction can proceed spontaneously". This is incorrect! ΔG^o needs to be negative for the reaction equilibrium constant to be one or greater. However the main point that this reviewer wanted to make was that the only way that ΔG^o  can be a linear function of absolute temperature is if ΔC_p (products-reactants) is zero! This can be shown from the integration of the van't Hoff Equation. Such conditions do not apply to any of the reaction equations shown and should be clearly stated. A reference should be given as to the source of the thermodynamic data used.
4. The coal used in this study was a bituminous relatively high volatile coal. The authors state that at about 450^oC, the coal begins to decompose with the evolution of organic gases and liquids most if not all of which would be capable to oxide reduction. This needs to be discussed.
5. Table 4 shows the chemical composition of the DRI product, metallic iron together with any adhering slag phase. A very important piece of information that is not given is the actual composition of the iron product. For example what are the C and Si contents?
6. Finally, did the authors measure the melting point af the feed copper slag/ Liquid slag consists of Si-O anions together with metal cations . 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf