Good morning everyone, thank you for attending this meeting of the Presentation of the 2022 Results of Eramet. In introduction, I will set the stage with some key messages. Our CFO, Nicolas Carré, will present the financial performance and the operational performance. I will come back to guide you on the strategic roadmap for the future and some words of conclusions and outlook for 2023. We are very happy, I mean, to present you, and very happy and very proud to present you these results today. As 2022 has been a landmark year for Eramet, with historical records in many areas. It's a record year in terms of financial performance, with a very strong deleveraging.
We have also delivered very strong CSR performance, recognized by many rating agencies and outperforming our roadmap targets in 2022 in many dimensions. We have finalized our strategic repositioning on our core mining and metal businesses. We are now relying on a very solid base of high quality and high cash generating assets, which enable us to accelerate our growth projects in metals for the energy transition. Let me highlight some of those great achievements, and let's start with the financial performance. We set one record after another. The first one is on our EBITDA.
The Adjusted EBITDA reached EUR 1.9 billion if we include the Weda Bay contribution that we have started to include in our EBITDA and report an Adjusted EBITDA. Weda Bay is becoming a very material contributor to the cash flow of Eramet. I mean, fair view of what is the cash the company can generate. The contribution of the Weda Bay has been EUR 344 million to the EBITDA in 2022. This EBITDA of EUR 1.9 billion has increased by 58% versus last year.
This huge increase in EBITDA is a combination of a very favorable price environment during the first half of the year, and also good operational performance and growth in mining volumes, triggering EUR 180 million of intrinsic performance. Despite sustained CapEx, we generated EUR 824 million of cash flow on our new scope, and this has enabled us to significantly reduce our debt, which reached EUR 344 million at the end of the year, the lowest level in the last 10 years. Our leverage ended up at 0.2, which is a much more comfortable level. Our first satisfaction is to be able to start 2023 with a much robust balance sheet and solid financial situation.
2022 has been a quite contrasted year in terms of economic environment with two periods. A strong first half with a very favorable price environment, followed by a severe slowdown in the second half, in a context of inflation and significantly rising energy prices. So over the year, the impact of the external factors have been positive by EUR 530 million, but with strong pluses, as you can see on this slide. A strong plus coming from the pricing, especially during the first half, almost EUR 1 billion coming from the commodity prices. Half of it coming from the manganese alloys prices, especially in the first part of the year.
We have also benefited from a favorable exchange rate with a positive impact of EUR 330 million. We had also a lot of strong minuses, and especially in the second part of the year, with a very high increase of our input cost, and especially the energy cost and the reductant cost for EUR 450 million. We had also, because of this huge surge in energy price, especially in Europe, we had to cut production in especially in our manganese alloys business and also the titanium business in Norway, that has had an impact of EUR 100 million. We had, as you all know, a high level of inflation, which had an impact of EUR 50 million overall.
All together, our EBITDA in S2, in the second semester, has been the half of the one of the first semester because of this squeeze effect and despite the significant intrinsic improvement that we have made during the year. As I said in introduction, in 2022, we have also successfully implemented our ambitious CSR roadmap and with significant progresses in all our KPIs beyond our initial targets, reaching 115% performance. Safety first is constantly improving with more than 20% again in 2022 versus 2021. We have divided our number of accidents by more than six in five years. Coming from far away, we are now proud to be among the very best of our sector in terms of safety.
We have also reduced our carbon intensity by 40% since 2018, significantly exceeding our initial 2023 target of 26%. Regarding biodiversity, we have constantly rehabilitated more land than we have cleared for in our mining operations over the last five years. We have started even to return some rehabilitated land to local communities in Senegal. We have returned 85 hectares this year to the Senegalese communities, fully rehabilitated after mining. Everywhere, we continue to invest massively to contribute to the well-being and economic development of the communities that are around us. This CSR progress was again recognized by extra financial rating agencies, positioning Eramet as a leader in its sector.
We are still ranked gold by EcoVadis among the top 3% in our sector. The CDP has upgraded our climate change rating from B to A -, positioning Eramet among the best in the industry. The group has continued to implement the very demanding IRMA standards in its activities. We have already conducted two self-assessment in 2022. Our first audit certification will take place in 2023. All our sites should be certified by IRMA by 2027. Building on these good results and on our repositioning on solid cash-generating assets, we are now accelerating on our very promising growth projects in energy transition. Here we can see the main ones.
In lithium, the production at our Centenario site in Argentina is scheduled to start late Q1 2024. We aim for an annual production of the 24,000 tons, which is the equivalent of 600,000 EV per year. Because the market is booming, the demand for lithium is higher and higher, and our deposit is very large and very rich, we have launched already a study to quickly triple the annual production capacity in Centenario. We aim to reach 75,000 tons at the end, so adding another 50,000 tons to the one already in construction.
Here we are in the study phase, and we aim to be able to make the final investment decision by the end of the year. In lithium as well, we are also cooperating with Électricité de Strasbourg on a production project in Alsace, with a target of around 10,000 tons. It's based on the geothermal lithium brines. In nickel, cobalt for batteries, so nickel and cobalt salts for batteries, we continue, I mean, our studies, the feasibility studies with BASF, on our Sonic Bay project, which is in Halmahera , Indonesia, close to the Weda Bay mine. Here we are aiming for a decision on this's investment by the end of the year.
Finally, also in the battery sector, we are partnering with Suez in the, what we call the ReLieVe project, which is a recycling project for batteries. Here also, we are progressing on the pre-industrial phase and feasibility studies. All these projects, short and mid-term, position us as a key player in the battery value chain for electrical vehicles. While we are accelerating on our growth projects, we will maintain our capital allocation discipline that we explained to you already last year, which is first to maintain the low leverage and a robust balance sheet. It's of course important in our industry. To focus on value-creative growth CapEx that are the building blocks of our new strategy while rewarding our shareholders.
In 2022, most of our free cash flow has been allocated to deleveraging, which was our first priority. Gross CapEx had been limited to EUR 200 million. This part will increase, as you will see, in 2023. Showing, I mean, the level of confidence we have in the future, we propose an increase of 40% of the dividend to EUR 3.5 per share. Now I will hand over to Nicolas, who will present to you the financial performance in more details and also the operational performance.
Thank you, Christel. Good morning, everyone. I have quite a lot of numbers for you and pretty exciting numbers, so that's good news. I will describe in more details what Christel was summarizing in her introduction. Let's start with a summarized dashboard of our key financials. Christel has said it's written on the top of the slide, 2022 was a record year for Eramet, and it was a record year for different aspects. It was a record EBITDA. By definition, it was a record Adjusted EBITDA, which is the new indicator we are disclosing. It is important because you will see how important Weda Bay has become for our financial performance and for our cash generation.
It's been a record for net income, for the share of the group, it reached EUR 740 million in 2022. It's been a record of free cash flow generation. This also led to the very significant deleveraging we managed to achieve in 2022. It was following already a successful 2021 year on this regard. This is a continuation of a great trajectory. We will come back to. It enabled us, as Christel has been mentioning, to decrease our debt to EUR 344 million, knowing that it's leading to a leverage of 0.2. Leverage being the net debt on EBITDA, our capability to reimburse the debt.
We're gearing the debt to the equity at 15%. If you remember, not that far ago, it was above 100%, so it's clearly a great achievement. How did we handle this performance? I think this slide, and sorry, it's pretty heavy, but I think it's important to spend some time on that. Clearly, we have benefited in 2022 from very positive external factors. We've said it was positive by EUR 533 million. For the people who have in mind what we presented in July for the half year results, the same box was actually EUR 729 million at the end of H1. What does this means?
This is actually re-reflecting what we have said, what Christel has said about the fact that the economic environment has actually shifted, and that's something we actually said in July with decreasing selling prices and at the same time rising input costs. That's why we have seen this change in trend between the two halves, and by the way, leading to a total different situation in terms of calendarization or sequencing seasonality of our performance in 2022 versus what it is usually. Usually, we are much stronger in H2 than in H1. Here, in terms of pure financial performance, we generated EUR 1 billion of EBITDA in H1 and EUR 500 million in H2.
Again, overall in 2022, very significant and positive, let's say, market environment, it would have been useless without our capability to actually seize it. That's the big and important message to provide in this slide. Okay, we have had this half a billion positive impact of external factors, but at the same time, we have actually generated close to EUR 200 million of intrinsic performance on the Adjusted EBITDA. The two main factors are actually coming from the continued increase of our manganese ore production and sales out of Comilog.
The second piece being, that's also why it is very important to have this picture in the Adjusted EBITDA, was the outstanding performance of the Weda Bay mine, which as Christel has already mentioned, has doubled its level of sales between 2022 and 2021. That's the high-level summary. Not everything is green, as you can see. On the intrinsic performance piece, we need to highlight the fact that SLN was negative by EUR 34 million versus 2021, which was already a challenging year. Let's face it, we'll come back to that. 2022 has been even more challenging, especially in terms of weather conditions and in terms of electricity, leading to a very complicated financial situation for this entity.
Back to the evolution of our net income. I won't spend too much time on that, but it's always important to see that we have been able to more than double our net income in 2022 versus 2021. Despite an overall EUR 350 million unusual charge, impairment, and also provision, coming from SLN, given the situation I was describing before.
We took an impairment for EUR 221 million in 2022, of which 124 is actually impacting our group share of the net income. And we also took EUR 126 million provision on the Erasteel value to put it at the value corresponding to the offer we have received and which will enable us to indeed divest Erasteel in the course of 2023. Also, that's part of our strategy, we have been able to balance our investments, our CapEx in 2022, so it has increased significantly versus 2021.
It was intended and, clearly, the intention was to both invest in what we call sustainable CapEx or maintenance CapEx. Just as a matter of reference, the EUR 236 million you see here is more or less corresponding to the depreciation level we have, on a yearly basis. It's, what I will say is a good management of our assets, and it's also highlighting what we said before, that 2021, and it was the case in 2020 also, was a pretty limited level of investment given, again, the amount of CapEx we have to, or the amount of, investment we need, to sustain, our activities.
On top of that, we have invested, overall, I will come back to the portion of Centenario, EUR 352 million of growth investments, of which EUR 126 million is for our mine, manganese mine in Gabon. It is to sustain a growth which has been pretty impressive in the last four years and we'll see, at a later stage, what I mean by impressive growth in the last four years. Also, EUR 42 million coming out of our railways operations in Gabon, for SETRAG, to be able to transport all the growth of our ore, but also the growth of the other activities in Gabon.
That's a big portion of the EUR 200 million you see in the orange box. The EUR 152 million corresponding to Centenario, so to the lithium project, is actually the acceleration of the investments we have restarted at the end of 2021. It is in line with what we are expecting, so the project is progressing at the expected pace. As you may remember, and this is something which is important to highlight again, all of it in 2022 has been funded by our partner, Tsingshan, in this regard. No impact overall to our net debt. This being said, this is part of the overall CapEx as we consolidate fully this project into our books.
Talking about cash, one item also is important to highlight. If you look at these numbers, the first thing you may actually think is that it was a less good management of our working capital than in the previous year because, yes, we have increased the level of working capital by more than EUR 100 million in 2022 versus 2021. Yes, if we take just the calculation of number of days of turnover, it has increased. We may say that it was not the best management.
The other way to look at it is, first one, if you look at the evolution within the year, it has actually decreased by EUR 200 million between end of June, when we said it was artificially high given the context of prices, et cetera, and the end of 2022, where it has actually decreased because this level of prices was lower. The other way also to look at it, so as I was saying, overall, it's an increase by EUR 131 million of value versus the end of 2021. Very interesting. Out of this EUR 131 million, 80 is coming from the inventory of manganese alloys products in our plants at the end of 2022.
Very interesting to note that, actually in volumes, this inventory was going down by 20%. This is, this was, close to 120,000 tons at the end of 2021. It was down to lower than 100,000 tons, actually 95,000 tons at the end of the year. What does this mean? This is actually the reflection of the impact of the additional input cost we have been describing, so higher reductants, higher manganese ore cost because it was more in terms of value consumed in H2 2022 than in H2 2021, and higher energy costs. All of this was leading to a very significant increase of our inventory.
If we restate for that, you will see that actually we have been able to keep a very well-managed working capital, as it has been implemented in 2020 and 2021, considering the fact that generally speaking, we are in a growing environment, we generally speaking generate more working capital. That's really something I wanted to precise because this is also another topic on which we have performed a very good management in 2022. Between the great EBITDA performance, Adjusted EBITDA performance, the very well-controlled CapEx, which was in line with our guidance, with also this solid working capital management I've just been describing.
This was leading to this record free cash flow in 2022, amounting to EUR 824 million if we consider the continuing activity, so the new activities and the future activities of the group. If we, of course, take into account on top of that, the CapEx and OpEx of the Centenario project, which have been funded by our partner, it's above EUR 1 billion that we have been able to generate in the year 2022, which is a cash conversion, and it's always a very good factor indicator to follow, a cash conversion out of the Adjusted EBITDA generated in 2022 of 53%. That's also a very solid and strong performance to highlight.
It is important, of course, that's something we've said, it helps, it's critical. It was what we were willing to do to be able to have a much more robust balance sheet and to deleverage our balance sheet altogether. I won't describe all the numbers on this heavy slide, this is helping to understand how we moved to from this net debt close to EUR 1 billion at the end of 2021 to EUR 344 million euros, sorry, at the end of 2022.
Knowing that, if you add up the EUR 30 million of external debt, which is still belonging to Aubert & Duval, the reported net debt, not considering the IFRS 5 norm at the end of the year was EUR 374 million. It's also helping to have a very positive liquidity situation, and that's critical for two things, to be able to address the future maturities and also to be able to finance the very significant growth projects we have in our roadmap. We started in the last year, as I was describing, and will continue in the coming years.
The first thing, in terms of liquidity altogether, we have a situation of EUR 1.6 billion of pure cash, even close to EUR 1.7 billion of pure cash at the end of the year. On top of that, we have also more than EUR 900 million of undrawn RCF facility, which is also available cash we could have in case of necessity. All of it is leading to EUR 2.6 billion liquidity at the end of the year.
The other good news, because we try to have more good news, is that we have also been able to refinance our term loan, which was expiring in 2024, until 2027, and also increasing the amount to EUR 480 million versus before EUR 350 million, out of which EUR 270 million has been drawn at the beginning of this year. If we take into account this additional liquidity, we have a EUR 2.8 billion liquidity pro forma as we speak. We don't have significant maturity, and that's the other piece of good news, coming in 2023. We have some in 2024. We have, of course, 2025.
Clearly, that's why we have now, and it is the right time to work on making sure that we can refinance these coming maturities, and that's something we'll keep working on in the coming weeks. Today, I can say that we have a high level of liquidity. We don't have significant maturity coming in 2023. The last piece about our financing situation, which is also good to know and to highlight in such moving environment, is the fact that 2/3 of our debt is actually at fixed rate. It's helping to also reduce the impact of the very significant rising interest rates we have seen in the course of last year.
This leads also to confirm what has been said in the course of last year, which is the capital allocation policy. What have we said? We have said first priority was to deleverage. Again, it's important to remind the key numbers. We have decreased our debt by more than EUR 600 million in 2022. In two years, it's close to EUR 1 billion of deleveraging, of reduced net debt, as also Christel has been mentioning at the beginning of the presentation. Second piece of our objectives and of our policy is to invest in gross CapEx. That's something we have started in 2021, and it was especially focusing on the manganese ore.
Again, I would describe the effect in our, in our overall production of manganese ore. It's continued in 2022. The EUR 200 million you see here is just the piece outside of lithium, because lithium is fully financed by Tsingshan. It will further continue in 2023 with all the projects that Christel will describe later on in the presentation. That's the second piece, and that's why it was so important to be in the situation of robustness, of financial robustness, we have at the end of 2022. A third priority we defined, and that's also leading to what we have decided in terms of dividend, is to reward our shareholders.
That's something we are doing by proposing an increase of the dividend per share by 40% versus the end of 2021. It is totally in line with what we said and that's a confirmation also of the success of the strategy we have defined. More in details in terms of operational performance now. I've mentioned a few numbers, but it will provide here a bit more details about what it means in terms of pure operational performance. We have had an outstanding year in terms of mining operations. Manganese ore was up by 7% versus 2021.
Let's keep in mind that we have faced actually almost a week of missed production at the end of 2021 because of a landslide happening in Gabon, which is now fully solved. That's something which is also enabling to understand the overall performance. That's one piece. Second piece, I've said it, but also very important to highlight, the performance of Weda Bay in terms of mining was more than twice higher and bigger than in 2021. In three years' time, we started this production at the end of 2019. In three years' time, we have been able to become the biggest nickel mine in the world, which is absolutely amazing. It's above 20 million tons in 2022.
It will be even much more in 2023, as you may have seen in our release and as Christel will describe later on in terms of guidance. SLN, I've said it before, was challenging, clearly, because of adverse weather conditions. La Niña was continuing in 2022. It was the second year. It was even worse even in terms of quantity of rain. It was more than twice, close to three times the amount of rain we faced in 2022 and in 2021 versus what is the usual situation of rain in New Caledonia.
Despite this situation, we have been able to handle and to manage a stable amount of ore production at 5.4 million tons, and also a slightly increasing level of export at 3 million tons. It's slightly because it was 2.95 million in 2021, and it increased to 3.05 million in 2022. In terms of ferronickel production, it was also slightly higher at 41,000 tons, comparing to 39,000 tons in 2021. Knowing that on top of the weather situation, we faced complicated situation of electricity availability during last year.
All of it clearly not a good year for SLN, but we need to acknowledge the fact that there were external factors which were very unusual, and we managed to keep the production either stable or in slight increase. Mineral sands, same. Numbers could seem low because in both situation, we have a reduction versus 2021. Two things to be precise and which are absolutely important to understand this performance. For GCO, it was in a year of lower grade versus what we enjoyed in 2021. We have actually improved our overall efficiency, thanks to an additional capacity of dry mining, which actually supports the level of production of the dredge. Nothing to be worried about concerning this performance.
We know also that, by the way, 2023 will be a lower grade situation than in 2022. The good news is that 2024 and 2025, we know, we will address much richer areas than what we have had in 2022 and 2023. Concerning ETI, you may have in mind the name TTI, we changed it to ETI, E being Eramet, in last year. For ETI, it was a reduction also, but without the evolution of the electricity cost we faced at the end of the year, it would have been up actually this number. The performance of the plant was great until the end of August, when the cost of energy was surging.
Given this situation, given also the fact that the selling prices were going down on the market, we took the decision actually to adjust the production to be able to keep a positive margin situation. That's why we have this minus 10% versus the previous year. That's something which actually summarizes the evolution of the overall environment last year versus the previous year. It's really a story of two tales and a clear evolution, very strong one at the beginning of the year. In almost all products, a sharp, a very sharp decrease in the second piece of the year. That's helping also to understand why we have had this evolution of internal, of external factors, between the first half and the second one.
It's been the case for manganese ore. Yes, overall it's been up by 13%, but as you can see, it was up significantly in the first half, very significantly down in the second half. Nickel ore was a bit less moving in terms of prices. This being said, it's also a negative evolution in the second half versus the first half. Also very interesting to note that even if the selling price is lower for what we produce and sell in Indonesia for Weda Bay, keeping in mind that there is a ban of export in Indonesia. Everything which is produced out of Indonesia is actually sold in Indonesia.
Knowing that the level of production of nickel is increasing dramatically there. That's to secure everyone about the fact that there is absolutely no question about the possibility for us to sell these volumes. Even if the level is lower, the increase of the value has been much higher in 2022 than for the ore we are actually selling out of SLN. It's been up by 35%. Manganese alloys. This is likely the most dramatic evolution. It started in the second half of 2021. It continued in the first half of 2022, even if we could say at a slightly lower pace.
This being said, much, much stronger than the first half of the year before. As you can see, it started to decrease significantly in the second half. Overall, we have had in the year, the same as for manganese ore, a positive evolution for all products, commodities and refined. It's been much lower in the second half. This was actually the main driver of the negative evolution of the external factors in the second half of 2022 versus the second half of 2021. Concerning ferronickel and NPI. One thing which is also important to remind, we've said it during the H1 results, but that's something we need to highlight again.
There is a significant disconnect since what happened in March of last year on the LME market between the LME index and the selling price of the ferronickel. It's for the entire market. It's not specific for us. It is clearly the evolution for the entire market. You can see that the lines green and orange were very close up to the beginning of last year. The deviation has been dramatic starting at the end of Q1. You can see that due to this situation that LME could not be a reference anymore. Our selling price and the selling price of ferronickel on the market has been much closer to the index of NPI, the SMM.
This is what you can see, the evolution between the black and the orange line. At the same time, same, it's to summarize what we have said already a couple of times, whilst we were seeing a rollercoaster in terms of selling prices, we have seen that the input costs were rising very significantly all through 2022, especially on reductants. The good piece of news is that we have started to see a slight reduction at the end of the year. This being said, it remains still at a much higher level than the overall level of 2021 and even historical levels.
That's something which is impacting heavily the production cost, especially in our manganese alloys plants and also in the Doniambo plant out of New Caledonia. At the same time, fuel was also increasing sharply all through 2022 because it was up by 40% overall. That's something also which is slightly reducing at the end of the year. It's a bit softening also at the beginning of this year. We have still a significant burden which is to be supported. It's especially the case for Doniambo because this plant is getting power out of power plants using fuel. That's why it generates also a significant burden for this activity.
The thing I would like to highlight out of it, you don't see here the evolution of electricity spot price because we have been able to actually adjust the level of production, especially for manganese alloys and ETI, as I was mentioning before, to the level of electricity we are actually hedging and for which we have a pretty low level of cost. We have been able to minimize the impact of cost. It has been a little bit to the expense of level of production, but it was actually helping to keep producing something creating value for the company. It's important to remind that, overall, we have 80% of our power purchases which are protected. It's been the case in 2022.
It will still be the case in 2023. Freight cost, I won't spend too much time on that, because at the end of the day, it has been actually through the year, and if we compare to 2021, the more stable input cost we have seen. It was increasing in the second half of 2021, keeping also pretty similar level in the first half of last year, but it's been reducing in the second half. Overall, actually the burden is more or less stable in terms of index.
We have been able to reduce actually our cost thanks to the implementation of barging and transshipment equipment in Gabon, enabling us actually to load vessels four times the size of what we were loading before. It was helping to reduce significantly the cost per unit in our activity in Gabon. At the end of the day, I will describe quickly what it meant for our different operations. Manganese, I will start with this still very important portion of our business.
Has been by far an incredible year because thanks to the performance out of Gabon, the very good management of the cost in manganese alloys and the strong pricing environment I've been describing, we've been able to generate for this BU a EUR 1.4 billion EBITDA. 1.5x what we generated in 2021, which was already a very positive and significant year. In terms of free cash flow also, it's been an amazing year because EUR 835 million cash was generated by this business unit. It's 1.7x higher than what we achieved in 2021.
Also, that's something you will see for each business unit, we have been able to perform very positive and strong CSR activities, developing nursery activities for vegetation in Gabon. It's been done thanks to a very innovative system of drones. That's really something we are very proud of, and that's something we'll continue to develop in the coming years. The 17 hectares is just the beginning, and that's something we really developed going forward. Also, something which is important to mention when we talk about manganese alloys, outside of the great management we have done in terms of electricity consumption. This is green electricity for a big portion of that.
All the plants in Norway, our plant in France, our plant in Gabon, are actually using electricity, which is renewable, coming from renewable energy. This is what is enabling us to actually have a CO2 intensity, which is 60% lower than the rest of the industry. That's really something also we are proud of. We have one plant currently which is not using renewable energy, which is in the U.S., but for which we are targeting and we are developing projects to be able to use in the future, renewable energy as well. In terms of overall market environment, won't spend too much time there because I was already describing what was happening through the year.
One important thing I would like to highlight and something we said at the end of June or following the, our publication at the end of H1 is concerning the spread of our ore, high quality ore of manganese versus the one of lower quality. Usually, historically, it was a spread of $0.50 per DMTU. It has been increasing up to $2 per DMTU. It's been slightly reducing or it's been reducing at the end of last year because of higher inventory, especially in China. It was more or less back to the same level of what we could have done historically. We are not surprised by that. At the beginning of this year, it's actually up again. It's above $1.
It's even closer to $1.5. This is back to what we said in July. The fact that we are confident that the market has shifted and we have a much better premium, a much higher premium out of our manganese ore versus what we had historically. Moanda, I was promising, teasing twice already the fact that I will come back to the historical evolution. It's always good to come back to that. Look at this evolution between 2018 and 2022. In these four years, we have been able to increase the production by 80%, eight- zero, from 4.3 to 7.5. This is not over because we are targeting to increase to a higher amount.
That's the complexity of the guidance we have provided for 2023. One could say that it's stable versus 2022. Yes, we have faced this landslide issue, which was actually impacting almost the entire month of January. It's back to normal. We are confident that we are totally back to a normal situation since the end of January. Due to that, we have adjusted our target for this year, for this coming year, knowing that if we wouldn't have had to adjust it, you can make quickly the math, we would have again increased our production by around 7% versus what we achieved already in 2022.
If we go back now in terms of transportation performance in last year versus the year before, it's been even more impressive because it was up by 10%. Knowing that here also the performance has been impacting down by one week at the end of last year because of the landslide. The last piece to highlight here is we have a cash cost which is indeed pretty stable between the two years. We have had some inflation clearly impacting the cost. And that's important, why do we invest and why do we keep investing is to also be able to be more efficient and more productive in the way we have generated the growth.
A portion of the growth we have generated in the past few years was thanks to dry mining. We will move to the more usual way we are producing our ore, especially manganese ore, using washing plants. These washing plants will come into service in 2023. This will actually help to have a lower cash cost than what we have had in the past two years. Here you can see also what is the impact of the reduction of freight costs thanks to the transshipment I was mentioning before. It was helping to reduce the cost by 10%. Manganese alloys, as I said earlier, we have actually adjusted the level of production to adjust to the high electricity cost.
It was much more significant by definition in the second half than in the first half. Overall, yes, we have had a lower production by 9% versus 2021. It was down by 22% in the second half versus the same period, versus the first half of 2022. It was no issue in terms of production. It was no overall concern. It was just to be the more agile to the market evolution, both in terms of potential demand and also to avoid producing at a too high electricity cost. That's what was leading to this situation.
It's very interesting actually to see the graph on the right-hand side because it's actually highlighting how impactful are the other items than manganese ore now in the manganese alloys' cost of production. As you can see, the orange curve, which is actually the margin per DMTU we have versus ore comparing to our selling price, is actually higher than what we had almost all through 2021. This being said, due to the fact that reductants especially have increased significantly, the overall level of margin is lower, and that's really something I would like to highlight on this graph.
This is really the point that, yes, the margin on the ore itself is pretty good, but we are in a situation in the market and environment today where the reductants have a much higher impact and negatively impacting our margin, hence the need to be very agile in the way we handle our production to be able to keep generating value for this business. Nickel BU saw an outstanding performance once again of Weda Bay, and that's really something I would like to highlight. When we said why do we want to move to an Adjusted EBITDA? Here, look. We have generated out of the EUR 1.9 billion of Adjusted EBITDA in 2022, EUR 344 million coming from Weda Bay.
This was leading to more than EUR 200 million of cash, EUR 237 million, as you can see. This is why we believe that this is the right economical indicator to look at, because this is how important the performance of this mine is leading. The same comment as I was making before for manganese on the fact, outside of the great financial performance we generated in 2022, we have also keep contributing strongly to our CSR goals and strategy. In Indonesia by contribution to local communities by doing different things in terms of waste collecting, equipment purchasing, in terms of donation of equipment to universities and also financing of scholarships.
Our goal, as we do in every country, is to be able to support the local communities, that's something we have done. The same, by the way, for the IWIP park, we have supported strongly the local environment, the local communities. In New Caledonia, we have had a ratio of rehabilitation which is comparing the areas of rehabilitated space versus the deforested ones of 1.5. It mean that we are rehabilitating quicker than what we are potentially deforesting. Nickel, here really, I was providing before the more important message, saying that yes, LME, which is something everybody was looking at before, was strongly moving in the year of 2020.
Two very significant positive evolution, but now totally disconnected from the market. That's why it is important to look at more the evolution of the ferronickel prices, and especially of the NPI index, which has been much more stable last year than the one of the LME. Weda Bay, this summarizes what I was describing before. This 21 million tons is actually more than double what has been sold in 2021. It is actually continuing. That's a good news because we are announcing and we are confirming that our target for 2023 is above 30 million tons. Above 30 million tons. I really want to repeat that. It's a mine which started its operation in Q4 2019.
We managed to achieve this mine that we are operating in less than four years. That's how impressive the performance is. The other thing which is also important to keep in mind is most of the value coming out of Weda Bay is actually the mine, the ore. It's not the plant. The ratio is largely above 80% of the value is coming from the mine. This is how important also is this evolution, this expected evolution. One item to precise though is it's part of this increase is coming from what we call conservation ore or low grade, which is expected to be a significant volume in 2023.
It's still very accretive, generating significant value, but lower margins than what is the high grade that was produced so far. SLN, I was commenting the biggest numbers or evolutions before. One item I would like to highlight there is concerning the cash cost. It's being again significantly up in 2022, reaching $8.2 per pound. That's actually the impact of the energy and other inflation items. If you add up this one, $1.3, and also the inflation for $0.7, you have $2, which is coming from there.
Yes, there was a positive currency effect, but that's to highlight the effects and the impacts of the higher input cost we have faced on this business. Last but not least, mineral sands. I was describing before what was challenging the volumes for both GCO and ETI last year. It was still also a record year in terms of financial performance for this activity, for this business unit. Look, it was up to EUR 184 million EBITDA, which is +34% year-over-year. It was also a very solid pricing environment. I will show that later. Free cash flow was lower. The main reason being that here, the working capital was extraordinarily managed at the end of 2021.
We are not able to repeat the same situation in 2022. On top of that, we have had to actually move part of the distribution of the TiO2 products from the distributor which was holding the inventory to our own distribution, so which was creating additional working capital. I spent some time on that because you may wonder why we don't see actually the effect in free cash flow versus the effect of EBITDA. This being said, you can also see that 2021 was an outstanding conversion cash from EBITDA, and it remains above 50% in 2022. It's still a very solid contribution to our cash which comes from this business unit.
Same, we have also very strong CSR activities and performance there between the production of waste recovered at GCO by producing a product Ilmenite 56, which is actually coming from the two product we were selling before. Now we have this product, 56% that we can actually recoup and sell. The fact that we have returned 85 hectares to the Senegalese state after rehabilitation, asking the communities what kind of rehabilitation they wanted to have. This is just a start. It will be close to multiply by 10 in the coming years. This is also a very strong performance and activity I wanted to highlight. That's what I was promising in the previous slide.
This is showing the evolution of the market for especially zircon, which is the highest, sorry, value creative item for our GCO business. As you can see, this one was actually the single one which was showing almost all through 2022 a positive evolution. It's been slightly reducing in the back end of the year, but still remaining at a very high level and much higher levels than what we had the year before. That also was helping to of course, to this very strong financial contribution I was describing before. It was only possible thanks to the fact that we were very efficient and were able to seize actually this pricing evolution.
Here is just a description of the production and I was already providing the details before, so I won't provide much more details, leaving some time for the rest of the presentation. I will hand it over now to Christel talking about our outstanding strategic transformation.
Thank you, Nicolas. Now I will zoom a bit on the strategic roadmap going forward. We communicated to you our new strategic roadmap in 2022. As you know, it is built around two pillars. The first pillar is to grow in metal supporting global economic development. We are talking about manganese ore and alloys for the carbon steel, construction, infrastructure and automotive. Nickel for stainless steel and mineral sand, as you know, so titanium and zircon, mainly for the construction industry. These are resilience markets where we have demonstrated leadership and continuous value accretive organic growth in the past and will continue, I mean, to grow organically and generate cash in those businesses.
The second pillar of our strategy is to sustainably develop the critical metals for the energy transition. Here we are talking about lithium and nickel cobalt for the batteries and also the battery recycling. These are fast-growing markets with huge potential, where we have great resources, good knowledge and we have the right CSR positioning to become a leader in that area. This CSR positioning is really an advantage and more and more especially when we are talking about the EV industry. On this second pillar, we are positioned on the very upstream of the value chain on mining and first extraction of metal, for example, in Indonesia and up to the refining of the metal for lithium in Argentina.
On the very end of the value chain, the recycling part, which is absolutely key, because, as you know, because of the energy transition, the world will need a lot of metals. We'll need to responsibly extract those metals. We need also to make sure that once extracted, we keep them and we recycle them and we transform them again into metals that can be used in the batteries. It's what we are aiming to in our strategy. Let's start with the lithium. With the faster than expected development of the EV industry, we are facing a continued strong growth momentum for lithium demand. Each time the demand forecasts are revised, they are revised up.
Despite the fact that there are more and more new project coming on stream, we see the deficit of lithium and we see it for quite a long time and which has brought the price to a very high level above $70,000 per ton over the last year. We see also the long-term consensus price being increased now up to $18,000 per ton of lithium carbonate, which is significantly higher than what we had in our, in our forecast only a couple of years ago. In this area, we have this first-class deposits in Argentina. A very big deposit, 10 million tons of resources with a life of mine above 40 years.
We have a process that has been tested on site in an industrial pilot now in real condition for the last three years. Quite robust, a new DLE process. We have started the construction of the first plant of 24,000 tons of lithium carbonate. We plan to commission this first plant end of Q1 2024 and to be at full capacity at the latest mid-2025. This project, as you know, it's things that we have figures that we have shared with you already, is very well positioned in the 1st quartile of the cash cost curve. With very low cash cost. It shows very attractive metrics.
With the new long-term price, we are targeting long-term EBITDA of around $300 million, with obviously a very high IRR. The project CapEx, since the restart of the project, is estimated at $550 million. We will be able to provide with this first plant about 500,000 EV batteries. As the demand is there and growing, the deposit is very large and rich, as I've said. The value creation is huge. We have accelerated and already started the studies to triple the capacity of the plant. Adding an additional 50,000 tons per year of capacity to the initial 24,000.
We are targeting at the end 75,000 tons per year of lithium carbonate. The idea is to be able to make a decision by the end of the year on this second phase. We target production if in 2027 to be at full capacity. The second big project, as you know, is in nickel cobalt for the batteries. This project is aimed to produce nickel cobalt MHP for batteries, taking advantage of the big Weda Bay deposit, which has a lot of limonite resources. We are in partnership with BASF. This plant should have a capacity of around 60,000 tons per year of nickel and 66,000 tons of cobalt.
We are in the detailed feasibility study. We aim to take a final investment decision in the second half of this year, by the end of the year, we still expect to start production as planned in 2026. This project will meet the highest CSR standards, it's very important because we will build this plant according to the highest IRMA standards, which will differentiate quite significantly this plant from its competition, mainly Chinese project in Indonesia. Especially, we are planning dry stacking of the waste of the plant instead of deep-sea tailing disposal or big dams, which is the plan of a lot of our competitors.
We are also involved, as I said, in the recycling business, partnering with Suez to become a key player of the battery recycling value chain in Europe, starting in France. It's a project that is still in feasibility study. The partnership with Suez, in fact, covers the upstream part, the production of black mass. Here we are in a pre-industrial phase. The second part of the refining of the black mass will be a pure Eramet project. Here we are building the industrial demo plant at the Eramet research facility. We plan to be in production for the full setting by 2027 in France.
If we make the decision, it will be probably located in Dunkerque to be close to the battery hub that is being built there. Finally, in lithium, as you may have seen, we have signed an MOU with Électricité de Strasbourg. We are studying the development of the production of lithium in Alsace from geothermal brines. Eramet will deploy its DLE, so direct lithium extraction process that has been proven to work on the geothermal brines. Électricité de Strasbourg, of course, will provide its knowledge of the geology in Alsace and its knowledge of production of geothermal energy. It's an interesting project, even if it's smaller than what we have in Argentina.
We aim to produce 10,000 tons of lithium carbonate per year, which is about 15%-20% of the French needs. It will be a very, very green lithium because it comes from geothermal, so it will be even positive in term of energy, and it's green energy. In term of course, of freight, it will be produced in France, so lithium will not cross the Atlantic to come to the French battery producer. It will be really positioned as the greener lithium, probably produced in France. As a conclusion and some outlook for 2023.
As you have seen, we are starting 2023 with a much more robust company than we used to be before. We need that because 2023 will be more challenging than 2022 in term of macro economy and with lower commodity prices and still high energy and reductant costs. The environment will be less favorable than 2022. Eramet as you have seen is well positioned, is more robust. We still well positioned to continue to deliver new mining production records in Weda Bay first, and as Nicolas said, we plan above 30 million tons of nickel ore, so +50% versus 2022. being able to market 15 million tons of low-grade ore.
In manganese ore in Gabon, we will still target more than 7.5 million, despite the loss of one month of production due to the landslide on the railway. We will continue to build the future and, sorry, I should have moved to this slide. Build the future and fuel the future growth of the company with an increase in CapEx, EUR 600 million of CapEx plan for 2023. EUR 300 million being dedicated to growth.
Based on all this price expectation, this macroeconomic uncertainties, but also all the good things that we are providing in our operation, we target an Adjusted EBITDA around EUR 1.2 billion in 2023. As a conclusion, thanks to this solid fundamentals that we have built over the last five years of transformation of this group, and thanks to the finalization of our repositioning towards the high cash generating mining and metal businesses. The group is focusing on the development of its project to produce the metals required for the energy transition and to meet the need of the fast growing market in the coming years.
Thank you very much, for your attention during this presentation, and I think we have still some time for the questions.
Hello. Nicolas Montel, Portzamparc. First of all, I have a question on the manganese side. In terms of miner manganese ore demand in China, what do you expect in 2003? In terms of price, how do you see the market? Also the same thing for alloy. We see a high decrease in production from steelmaker in Europe. How do you see the demand? You have explained that you have decreased your volume because of spot electricity price, but what we see right now is that they are decreasing. What is your plan for 2023 on volume for alloy?
My last question would be, you talk some issue in Weda Bay in Q4 in the production. It was decreased a little bit. Can you explain what's happened there? Thank you.
Just on the manganese market, manganese ore market, so it's, as you know, manganese is used mainly in carbon steel. We don't see the carbon steel market growing in 2023. We see it more slightly decreasing, let's say be flat in China and slightly decreasing in the rest of the world due to all what you said about the energy cost and slowdown of the economy in the world. So, an overall, let's say slightly decreasing market with some dynamism in China. And we have seen a quite good start of the year in China. Maybe too good, I would say.
There was really a rally at the beginning of the year in China, building on the sentiment that because they'd freed up the COVID restriction, they everything will restart very fast. Now they are realizing that it's not restarting as fast as they thought. There is a little bit of slowdown again, but we expect China to be relatively dynamic. It's the same for the manganese alloys production. In term of pricing, you know that we are following the pricing of the consensus, which is planning a $5.2 per DMTU for the manganese ore high grade in 2023, which is about -15% versus the average of 2022.
It's in line with what we have experienced the second half of the year, 2022. Presently, the price is higher than that, because also we have stopped the production in Gabon for one month, and we are today the biggest mine in the world. When the biggest mine is stopping, of course it has an impact on the prices. We don't know exactly what will be the outlook for the year. As an overall, I mean, feeling today is that the year is starting better than what we anticipated only a few months ago. It's we are not in recession as some planned some months ago.
The market is obviously not at all at the level of 2022, it's not that bad. In terms of stainless steel, anticipating on another question, it's we are expecting the growth and momentum in China especially. Regarding your question on manganese alloys production, we have to be very agile and it's what we have done in 2022 very successfully, depending on the market and you see that even if the steel industry is still quite low in terms of capacity today in Europe, they are slowly restarting some capacity and we're expecting Q2 to see some new capacity restarting. We will adapt to the market.
We are market leader in Europe, especially for refined alloys. We are market leader in the U.S., so we, it's our duty to make sure that we adapt to the market demand and also to the energy price. You're right to say that today, for refined alloys, even the spot market, the spot price of energy, is okay to produce at a positive margin, refined manganese alloys, which was not the case only a few weeks ago. We are adapting, and we are permanently optimizing our production to make sure that we continue to make margin on this business. On the Weda Bay, do you want to take it?
Yeah, sure. It was a slight slowdown in Q4. It's now back to normal. I really want to emphasize what I was mentioning before concerning the Weda Bay performance is that already it was the case in 2021 that the biggest value is actually coming from the mine and not from the plant. It's even more the case in 2022. It will be even more the case, of course, in 2023. As you could have seen, we target to increase the level of production of the mine by 50% versus what it was in 2022. The overall level of NPI production out of the plant is, generally speaking, remaining stable.
Yes, there was a slight issue at the end of last year, but again, no worry about the ability to reach the target in 2023.
I can tell you that the beginning of the year is quite good in Weda Bay. Other questions? There is one in the room here.
Yes, Maxime Kogge from Oddo. I have a first question on your cash cost, and perhaps then I will have another question on your growth projects. You generally don't give any guidance on cash costs for the coming year. Still, given the tailwinds we are seeing currently, like lower freight, lower reductants, also it's a bit, it's still a bit volatile, and higher production, would you see the cash cost position of SLN, Weda Bay, improve significantly next year? Perhaps the same question for Comilog, given that here too, you have the benefit of lower freight and lower, and washing machines that are supposed to improve the productivity?
Take it.
Maxime, thanks for the question. I will answer the last one, which is the easiest one. Clearly, yes, we are indeed expecting a lower cash cost, and we don't provide any guidance indeed, but we, you can expect and we are expecting a lower cash cost for manganese ore. For the FOB, indeed, thanks to the implementation of the washing plants, which is making us more productive, more efficient. As you spotted it, indeed, for the CIF cost, it will further improve thanks to the stability overall of the freight market and the fact that the transshipment I was describing before is indeed having a full year impact in 2023, which was only partial last year. That's for the Comilog piece.
Concerning the rest for SLN, yes, we can expect an improvement of the cash cost because it's a bit easing on the input cost. It's a bit easing. It's not dramatically easing. It's not back to what we could have experienced in the previous years, but it's expected to be better than 2022. The second factor, that's also the value of cash costs, which have a nominator and a denominator. The denominator, which is the volume, as you have seen, is expected to increase pretty significantly in 2023. We have said that we are targeting more than 45,000 tons. It was 41,000 in 2022. By definition, so it should improve.
Last but not least, as we have said it, given the financial situation of SLN, there has been an emergency plan to focus on the fixed costs, which also is expected to show reduction versus the overall cost of SLN, versus 2022.
For Weda Bay, clearly, we don't expect significant deviation of the cash cost versus last year. Just really the thing I would like to highlight is what I was mentioning quickly earlier, part of the increase of the volumes of ore will actually come from a lower grade. Again, still significantly value created, but creative, but not exactly at the same level as the one we have had on the higher grade ore that we were producing since the beginning. Generally speaking, overall, not an expected significant deviation in terms of cash cost for Weda Bay.
Thank you. The second question is on your growth projects. There's a lot of eagerness from the market on this project. Maybe perhaps can you tell us what are the remaining steps you need to achieve before taking a decision both on Sonic Bay and on the Alsace project? I understand that on the Sonic Bay project, everything is or almost everything is done on a technical level. Perhaps this is more a question of finance scheme now that you would like to complete. On Alsace, I mean, you made a significant headway recently announcing a memorandum of understanding, but still, I mean, the final decision will come...
Mm.
within only five, four years. Maybe can you also here guide us through the various challenges you still face in that area?
Okay. On Alsace, we are really at the pre-feasibility study. We are quite early in the process. We have tested the process of Eramet on the geothermal brine, which is very different from the one we have, you have in South America because it's hot. We operate at 80 degrees, and it's quite aggressive in term of content. We had to prove that our process could work. This is what we have done. Now we have to do all the pre-feasibility study, including how many pits we need to connect, huh? Because one pit cannot provide 10,000 tons of lithium. We need to connect several pits.
We need to look at all the basins that we will have and how we'll reinject and where we will reinject the depleted brine. I mean, this geology and technical things and also all the permitting, yeah, because we are in Alsace. We are in a middle of a very populated area. It's not at all the same as on the Argentinean Salar. All the local permitting will take some time. That's why we are cautious today on the timetable. It could go faster, but we are cautious on the timetable between now and the FID. On Sonic Bay, it's not that much a financing.
As you have seen, we have a more robust balance sheet and the capabilities to be able to finance. We look at alternative financing as well, but it's not what is slowing down the project right now. We are working together here also with BASF on the detailed feasibility study on the project execution strategy and the contracting strategy with whom we will contract the different parts of the plant. Again, we have very high CSR standards, so we have to make sure that the contractors could fulfill these standards, which is not always obvious.
We have, I mean, contracting strategies that we have to finalize, and we have also all the permitting to get in Indonesia, which sometime is taking more time than expected. It's what we need to finalize to be able to make the final investment decision hopefully by the end of the year. Any other question? Oh, yeah. Yes.
Merci. Hello, this is Andrew from BOC. Thanks very much for delivering this interesting presentation today. I just got a quick question on SLN in New Caledonia, the Mandat Ad Hoc process. I'm just wondering if we can get any updates to an disclosable extent about this issue. When we search online, there are a lot of worrying voices on this issue in the press.
It's we are not anymore in Mandat Ad Hoc. It's a technical, but we are in Conciliation, which is a step ahead of the Mandat Ad Hoc. You know, it's about the same, but the timeframe is more constrained. That being said, I mean, it's in SLN, as I said, we are working right now together with the French government and the Caledonian authorities and Caledonian communities in order to try to solve and build, I mean, the future of SLN and find structural solutions to structural problems.
Because today what we have done with our emergency plan, short term and the very limited loan that the French state has given to SLN is, we have a patch that help us to continue to run without, I mean, having an issue of being fearing being bankrupt. It's not giving any long-term solution to SLN. As you know, there are issues regarding the energy price, electricity price in Caledonia, which is very high compared to the competition. We have also issues regarding the access we have to our to our mines, with a lot of societal issues.
At the end of the day, we need to not to have patches and emergency plans after plan. We need to find long-term solution. It's what we are doing in 2023 together with the French government and the Caledonians. There are other question?
No. No. Okay, we have some questions on the chat box.
Yes, we have a few question from the web. Going back to lithium in Argentina, is your cash cost for lithium in Argentina still at the same level, or could it be different when you will start production?
As you know, there is inflation. What we know is that, we are still by far in the first quartile and at the beginning of the first quartile, I would say, of the cash cost curve. We are reviewing the OpEx for the future, but it will remain very competitive and due to the fact that the long-term price for lithium is at $18,000. When we announced the project, we were talking about a cash cost of $3,600 per ton. Even if we have some inflation on that, we remain very, very low compared to the price.
What could you do to accelerate the start of production, considering that the current price is very attractive?
It's a very good question. What we are doing is that, we are accelerating and the hiring and training of the people ahead of the plan start to make sure that the ramp-up will be quicker than what we had planned, if possible. We hope to be able to have a quicker ramp-up is the first thing that we want to do. The second thing is what I've announced, is that, we were planning to have a phase II, of course, but not that fast.
With all the need of the market and the very good price, we think that it's, it is the right time to accelerate and to build the phase II, earlier than what we had anticipated at the beginning.
Maybe if I may to complete on that one is also you have seen that for this phase II, we are saying that it's double the size of phase I, so it's 50,000 tons. There was before, what we said at the end of Q3 was actually more or less the same size. That's also what we are doing. It's accelerating, but also making sure that we have a much higher quantity than what we are initially targeting.
This project looks very profitable. How would the Argentinian government react to you overearning?
We have an agreement on with the, of course, we have a contract with the Argentinian government on royalties, which is a percentage of. I mean, there are different levels of royalties because they are the level of the federal government, and there are some at the level of the local provinces. All together, they are fixed, and they are percentage of sales or results. At the end of the day, if we make more money, they will make more money. It's royalties in percentage.
Looking at Sonic Bay, how much CapEx would the JV with BASF cost?
We have not given this figure. This CapEx will be in line with what we see on the market. Probably a bit higher because again, of our higher CSR standards, which at the end will cost a bit more than what we can see on some Chinese projects. It's something that we are, I mean, we are figuring out and that we will disclose at the time of the FID. People have to be careful that very often when our competitors announce cost for their projects, they don't take into account the site plans.
For example, the power plant, the acid plants that are supposed to feed the asphalt plant itself. Very often they disclose only the cost of the asphalt plants. The project, of course, will include a power plant and an acid plant as the others. We have to compare what is comparable.
Finally, regarding lithium in France, could you give more details or metrics regarding this project, such as cash cost or estimated CapEx?
It's too early. What we have in mind is that the cash cost will be very low because we have the proceeds of the sale of the geothermal energy. Altogether, the cash cost should be quite competitive. What will be higher than what we have in Argentina is the CapEx per ton because in fact, we have to build a higher capacity. Because the brine has a lower grade. The brine in Alsace are around 180 mg per liter. And in fact, in we are around 500 mg in Argentina. It's very different concentration, so it requires more CapEx per ton.
It seems that we have answered all the questions. Thank you again for being there, for your attention, and I hope that you are with us recognizing that again, Eramet has moved to another level in terms of robustness, in terms of, and is ready now in order to enter into this new age of metals with a very interesting growth project. Thank you very much, and I will see some of you in the next two days for more detailed question. Have a good day.