Sustainable Resource Management: The End of Nickel Mining?

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

It's an interesting subject. Given the context and the strategic guidelines for nickel use, it is important to estimate the share of each nickel production method: mining and recycling.
However, it would be important to add and clarify several points in this article.
Firstly, the quantity of nickel ore available should be given, at least an estimate. This would enable the reader to position the different demand scenarios for this mineral for the different applications listed.
Secondly, the assumptions made and the factors not taken into account in the analysis should be explained. As the authors make a comparison with another study, these elements provide the reader with the means to analyse this comparison properly.
I would also like to see some changes made on specific points. Firstly, in introduction 1.1 on line 58. The time period is current, we're no longer in perspective, it would be good to check the figure with today's data. Secondly, in section 3.3, reference is made to the year 2040 for estimating GWPs. Why that year? An explanation of this choice would be helpful. In Figure 5, some parameters are shown in the red rectangle. We need to check that there is no duplication. If any parameters are duplicated, the reasons should be explained. In Figure 8, I think the actual curve of total nickel production to date should be added. This would give a positioning of the scenarios studied.
Finally, these elements would add to the possibilities for enriching the ‘Discussions’ section, particularly for the section dedicated to the comparison with the CSIRO study. The same would apply to the ‘Conclusions’ section, where further development would finalise this forward-looking work.

Author Response

Comment 1: Firstly, the quantity of nickel ore available should be given, at least an estimate. This would enable the reader to position the different demand scenarios for this mineral for the different applications listed.

Response 1: Agree. Currently, the estimated amount of nickel resources world-wide is 350 Mt onshore and 300 Mt offshore (British Stainless Steel Association, 2024; Nickel Institute, 2024). These values have been added to section 1.1, line 64-66.

 

Comment 2: Secondly, the assumptions made and the factors not taken into account in the analysis should be explained. As the authors make a comparison with another study, these elements provide the reader with the means to analyse this comparison properly.

Response 2: Agree. I have listed and explained all assumptions in section 3.4 (lines 329-342). I have listed them again here:

• 2020 is the current year (since there is much reference around 2020 as starting point, this was a simple way to incorporate certain assumptions).
• Pre-2021 split ratios of nickel use-cases are assumed to be the same than in 2021.
• Nickel demand (between 2050 and 2100) is driven by world population.
• Nickel within all products can be recycled again and again (infinite recycling).
• Once nickel is used as a certain product (e.g., EV battery), it will always be used as this product, and never for another product (e.g., stainless steel); i.e., “Once a battery, always a battery”. This is due to the process of extracting nickel from complex materials (like an EV battery) to revert it to a pure, raw form is energy-intensive and costly. The assumption was also made for model simplification purposes.
• Possible nickel exhaustion will only be an issue after 2100, i.e., resource exhaustion is not considered in this paper.
• All metal required to produce nickel products in the first year must come from mining only, and none is available for recycling before 1994.

Regarding the comparison with the other study (which has been added to section 5.4):

• CSIRO assumed a second life for EV batteries for some of the scenarios (pot. extending 5-10 y)
  • However, in this study, no second life has been assumed.
  • This was done for modelling simplification purposes.
• CSIRO assumes 2020 as the starting year (this also means that no recycling was assumed to happen before 2020).
  • However, in this study, the starting year of 1994 has been assumed (also assuming no recycling was done before 1994).
  • This was done to get a more accurate representation of the historic product flows of nickel – the more actual data used, the more accurate the model will be.
• The CSIRO study was not clear on which software it has used. In this study, Stella Architect was used. the different modelling techniques and software may also have lead to discrepancies between the two studies.

 

Comment 3a: I would also like to see some changes made on specific points. Firstly, in introduction 1.1 on line 58. The time period is current, we're no longer in perspective, it would be good to check the figure with today's data.

Response 3a: Agree. Line 58 [“One forecast suggests that there will be a 39% growth in nickel use for electric vehicle (EV) batteries between 2017 and 2025”] has been replaced by “Battery nickel use is still increasing for all EV types and is evident from the recent 8% increase over a 12-month period [ref adamas].”.  (line 57-59)

 

Comment 3b: Secondly, in section 3.3, reference is made to the year 2040 for estimating GWPs. Why that year? An explanation of this choice would be helpful.

Response 3b: The International Energy Agency (IEA) has estimated the nickel demand by scenario for the year 2040, specifically. This year was therefore selected as a key point in the study, since there is forecast data available for this year. An explanation of this was also added to the text in lines 184-189.

 

Comment 3c: In Figure 5, some parameters are shown in the red rectangle. We need to check that there is no duplication. If any parameters are duplicated, the reasons should be explained.

Response 3c: Figure 5 highlights the recycling model. The inputs into this model are the “total metal”, “recycle rate”, “lifetime (year mean)” and the “lifetime (year standard deviation), whereas the output is the “recycled metal”. There is no duplication of these parameters used in the model. The definitions of these terms have been added below to clarify the confusion:

• “Total metal” represents the amount of nickel currently present in a certain product which can be recycled once it reaches end-of-life.
• “Recycle rate” is the rate at which a certain product gets recycled. It determines the percentage of nickel that will be recycled after the product’s end of life. The recycling rates are assumed to increase with time.
• Lifetime (year mean)” is the mean number of years after which a product will be recycled (this value was assumed to stay constant).
• “Lifetime (year standard deviation)” was assumed to account for model variability. For stainless steel, batteries and other products, standard deviations of 3 years, 2 years and 2 years were assumed, respectively.
• “Recycled metal” represents the final amount that have been recovered from the recycling process for a specific year.

 

Comment 3d: In Figure 8, I think the actual curve of total nickel production to date should be added. This would give a positioning of the scenarios studied.

Response 3d: Thanks for the suggestion. Figure 8 shows the amount of Ni supply through mining as a percentage of total supply. The latest (2021, USGS data) data point was added to the graph (see figure below) to see how we are tracking in terms of production via mining vs recycling. This data point suggests that recycling of nickel is not occurring as much as the model assumed. Based on the scenarios, recycling would have been at around 20% (80% mining), but is currently at only 5% (95% mining) of the total nickel supply.

A short description has been added as a footnote between lines 388-389.

Comment 4: Finally, these elements would add to the possibilities for enriching the ‘Discussions’ section, particularly for the section dedicated to the comparison with the CSIRO study. The same would apply to the ‘Conclusions’ section, where further development would finalise this forward-looking work.

Response 4: Agree. I have added some text to the discussion (section 5.4) and conclusion (section 6) sections.

 {please see pdf document attached for additional figures}

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1.     Line 57, the author mentioned EV batteries so the author should have written (EV) after electric vehicle on line 48 rather than on line 58.

2.     The topic 2.2 “Medium change case scenario (“SDS”)”, the author should start a new page.

3.     For the case 2.3 “Rapid change case scenario (“1.5 C”)”, this was add on the discussion about the reason on this temperature but for the case of prediction, it was provided to 2050 while the objective was predict to 2100.

4.     Could you please explain more about the equations 3 and 4?

5.     Line 265, “standard deviations of 3 years, 2 years and 2 years were assumed.”  It appears that the author has written the same thing twice.

 

6.     The table2, superscript after (2100) and (2050), what does it mean? 

Author Response

Comment 1: Line 57, the author mentioned EV batteries so the author should have written (EV) after electric vehicle on line 48 rather than on line 58.

Response 1: Agree. This has been updated and highlighted.

 

Comment 2: The topic 2.2 “Medium change case scenario (“SDS”)”, the author should start a new page.

Response 2: Agree. This has been updated and highlighted, but other text changes might shift positions.

 

Comment 3: For the case 2.3 “Rapid change case scenario (“1.5 C”)”, this was added on the discussion about the reason on this temperature but for the case of prediction, it was provided to 2050 while the objective was predict to 2100.

Response 3: The BHP Climate Change Report has estimated the nickel demand for this scenario (achieving net zero, or the ‘1.5°C scenario’) for the year 2050. This year was therefore used as a key point in the time period, since there is data available for this year. Beyond 2050, there were limited demand forecast information available, so from 2050 to 2100, the demand was assumed to follow a similar trend to world population growth. This text is added to text manuscript in lines 186-189.

 

Comment 4: Could you please explain more about the equations 3 and 4?

Response 4: All the nickel that is mined in a certain year, gets split into the three use-case (or categories) of products that nickel can be used for. These products are 1) stainless steel, 2) EV batteries, and 3) other. Based on literature, 69% of the total nickel mined is used for stainless steel production, 7% for EV batteries, and the rest for other products. It is predicted that the 69% of stainless steel will decline to 45% use of the nickel mined, and that the 7% EV batteries will increase 41% (both by the year 2040). ‘Other’ will, again, just be the difference between the two. In order to get a linear decline from 69% to 45% over 20 years (or linear growth from 7% to 41%), a ramp function is used, in other words, assuming constant linear change from start (2021) to finish (2040).

This means that, for stainless steel, constant decline from 69% to 45% over 19 years (from 2021 to 2040) would mean reducing the split ration with 1.26 constantly for 19 years, until the 45% split ratio is achieved. This is further explained in the equations below, for stainless steel and EV batteries.

{please see attached pdf}

Where Split ratio_i refers to the initial split ratio in 2021 (which was 69% for SS and 7% for EV batteries), Split ratio_f refers to the final split ratio in 2040 (which was 45% for SS and 41% for EV batteries). Year_i and Year_f is the initial and final years corresponding to the split ratios, i.e., 2021 and 2040.

These values (-1.26 and 1.79) correspond to the values used in equations 3 and 4, where the linear ramp equation is expressed {please see attached pdf for equations / figures}:

 Let’s assume the specific split ratio for SS for a certain year, ‘t’ (e.g., 2035) needed to be calculated, then:

% _{stainless, t} = 69 – 1.26·(t–2021) = 69 – 1.26·(2035–2021) = 69 – 1.26·(14) = 51.3%.

This is shown in the figure below, but is also clear from Figure 2 in the manuscript.

  {please see attached pdf}

I added a paragraph below to the manuscript (lines 250-255) to explain the abovementioned.

 

Comment 5: Line 265, “standard deviations of 3 years, 2 years and 2 years were assumed.” It appears that the author has written the same thing twice.

Response 5: This part is written correctly. The standard deviation assumed for stainless steel is 3 years, the standard deviation assumed for EV batteries is 2 years, and the standard deviation assumed for the other products is also 2 years.

 

Comment 6: The table2, superscript after (2100) and (2050), what does it mean?

Response 6: These superscripts refer to the two notes added at the end of the page. They indicate more details about the demand forecast for that specific case.

The first note “(2100)¹” indicates that all demand forecasts for 2100 were calculated using world population growth forecasts from 2040 to 2100.

The second note “(2050)²” indicates that the specific value (9500) adheres to the 370% cumulative demand forecast, as per the BHP Climate Change Report.

 

[Please see attached PDF for assitional figures / equations.]

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper examines the flow dynamics of nickel supply and demand using a modeling approach. The key findings underscore the critical role of nickel mining in addition to the recycling of nickel products. This provides valuable insights into the global nickel industry. The following comments require the authors' attention.

1.     Please provide further clarification on the reasons for selecting the STEPS, SDS, and 1.5°C scenario models, as introduced.

2.     Some graphs, such as Figures 10 and 11, have low resolution, and the text within them is not easily readable due to color and font size issues.

Author Response

Comment 1: Please provide further clarification on the reasons for selecting the STEPS, SDS, and 1.5°C scenario models, as introduced.

Response 1: The STEPS, SDS, and 1.5°C scenarios are based on similar scenarios used by the IEA (e.g., “The Role of Critical Minerals in Clean Energy Transitions") and also used by companies like BHP (“Our portfolio in a 1.5°C scenario”), etc. These scenarios are used to evaluate the effect of varying future metal demands, from minimal change predicted (STEPS) until significant change seen under the 1.5°C scenario. Similar scenarios were also used by the CSIRO study referred to within the paper (“Known unknowns: the devil in the details of energy metal demand”).

Because the future is unknown, using these three scenarios provides a good way to establish different potential future circumstances, and estimate what the possible outcomes will be under each of the different scenarios.

 

Comment 2: Some graphs, such as Figures 10 and 11, have low resolution, and the text within them is not easily readable due to colour and font size issues.

Response 2: I attached higher resolution figures for Figure 10. a), b), and c), as well as for Figure 11. a), b), and c).