Ryan Corbett

Thanks, Jim. Moving to slide 6, you will see the impact of the extended downtime on our production in the quarter. We produced 9,084 metric tons of REO and concentrate, about 16% less than a year ago and 19% lower than last quarter, which was the second-highest quarter of production in our history. This unplanned downtime was on top of our usual planned downtime of about one week for plant maintenance.
So as Jim mentioned, we had significant damage to the rigs in one of our thickeners soon after we completed our planned maintenance shutdown. Unfortunately, without these rigs operating, the whole upstream operation had to shut down. The lower production volumes, some impact of shipment timing, as well as more concentrate been refined to NdPr all combined to reduce REO sales volumes to 5,839 metric tons.
To give you a sense of the drivers of the sequential sales volume decline of roughly 3,500metric tons, approximately 60% of the decline was due to the lower upstream production. About 30% was due to higher consumption of concentrate into the midstream circuits as NdPr production ramped up. While the remainder was due to shipment timing.
The temporarily lower upstream sales volume, along with one-time repair costs, combined to drive rapid descaling in the upstream business and, therefore, much higher than normal production costs in the quarter. As Jim stated, we returned to strong efficient production in June and July.
In fact, we expect Q3, if current production trends remain intact, may be one of our best upstream quarters ever. Finishing off the discussion on the upstream, realized pricing in the quarter was approximately $4,183 per metric ton, in line with the guidance we provided back in early May.
Moving to our midstream KPIs, NdPr production more than doubled to 272 metric tons in the quarter, while sales of NdPr were in line with the prior quarter's production at 136 tons. NdPr realized pricing in the quarter was roughly $48 per kilogram. All of these metrics were in line with our guidance provided in May.
Looking forward to Q3, as of today, we would expect concentrate production volumes slightly above our typical levels in a non-shutdown quarter. Sales volumes will benefit from the higher production, but will be partially offset by the higher pull-through of concentrate into the midstream circuits as we ramp NdPr production volume, which, as Jim mentioned, we expect at least a 50% sequential increase in Q3, while sales volumes are expected to be approximately in line with Q2's production of 272 metric tons.
Recall, as we discussed in May and as you saw this quarter, NdPr sales volumes will lag production volumes by about one quarter on average, partially due to the metallization process and the initial build-out of our separated product sales channels. As for pricing, we would expect Q3 pricing to be down low-single-digit percentages sequentially for both REO and concentrate and NdPr oxide sales.
Moving to slide 7. On the far-left side, you can see the impact of the lower concentrate sales volumes to lower NdPr pricing and a longer sales cycle for NdPr sales on our revenue trends. It is important to remember that as we transition from upstream concentrate sales to NdPr oxide and metal sales, the concentrate sales will initially decline more quickly than our NdPr sales will ramp up.
We are in the heart of this transition period, a critical juncture we have long warned investors about, as we evolve into fully integrated miner and refiner. Unfortunately, the financial impacts of this transition were compounded by additional unplanned downtime and repair costs this quarter, alongside what would have already been a significant investment period.
Moving to the middle and right of the slide, we posted an adjusted EBITDA loss of $27.1 million and adjusted diluted loss per share of $0.17 in the quarter due to several factors. First, as we have discussed over the last few quarters, our midstream operation is currently subscale as we ramp production of NdPr and other separated products.
This quarter, we absorbed the impact of higher-than-target variable costs as we optimize production as well difficult fixed cost absorption at these lower production levels. This temporary period of subscale production resulted in us taking an inventory reserve of $11.8 million in the quarter. This relates to the currently elevated cost of production versus the continued pressure on market prices.
The second major impact in the quarter was the unplanned downtime, which showed up in the loss production and revenue and also combined to create both higher per-unit production costs and one-time repair expenses hitting the P&L. The concentrate unit cost of production was impacted by inefficiencies in the start-stop, lower fixed cost absorption, and higher repair and maintenance expenses from our planned maintenance shutdown.
This impacts both the concentrate cost of goods sold directly, but also found its way into are separated product production costs in the quarter, given the interconnectedness of our operation. Further, this quarter's earnings were impacted by several million dollars of costs associated with the repair of the thickener rigs, including for equipment, labor, and contractor costs. This impact, of course, we expect to be one-time in nature.
Lastly, recall that even at these low prices at normal production levels, our concentrate business generates $2,000 to $3,000 of gross margin per ton of REO in concentrate sold. As we ramp the refining operation, we are temporarily trading some level of profitable concentrate sales for temporarily unprofitable NdPr sales in the subscale state.
We are also investing working capital dollars into the transition as our sales and tolling channels for separating product is built. But as we have highlighted for the last few quarters, this is why we are being methodical in our midstream production ramp and we remain confident that over the long term, this is an investment we will be happy to have made.
Recently, we have made very good progress in optimizing the incremental variable costs we are seeing and, importantly, see a clear path to further optimization of variable costs and rightsizing of fixed cost absorption as we push forward in our ramp over the coming quarters. Importantly, we expect that the combination will reduce our per-unit production costs materially.
I want to reemphasize what Jim pointed out. Over the last 8 to 10 weeks, we have returned the upstream to its high-production, low-cost profile you are all used to seeing. We have also made significant progress in midstream optimization, both in terms of cost and production levels.
While our midstream cost structure won't be at target for several more quarters as we continue to ramp and react to bottlenecks, we see very clear signs of early success that will bear fruit over time. I would also point out that despite a challenging quarter financially, our balance sheet remains in great shape.
Again we received a $50 million customer prepayment for magnet precursor materials in April and expect another $100 million of additional prepayments over the next year. In addition, we'll earn and, for the majority, expect to receive cash for the various IRA tax credits approaching another $90 million through 2025.
This provides a further buttress to an already strong balance sheet as we complete Fort Worth and invest in the separated product transition. Following which, we expect operating cash flows to meaningfully improve.
With that, let me turn it to Michael to give you some of the technical details and updates on the operations. Michael?

Michael Rosenthal

Thanks, Ryan. On slide 8, we have a drone shot from above the pit. You could see the upper rim of the pit in the foreground, with the separations pad on the right side. For those of you who haven't seen it before, the very long building on the right is our light rare earths separation facility where we separate NdPr from lanthanum and any remaining [series].
Moving to our operations. Second-quarter performance was severely impacted by rig damage in one of our thickeners that Jim and Ryan described. This issue impacted production for approximately three weeks on top of the planned one week semi-annual maintenance outage during the quarter.
Our teams, supported by certain contractors and service providers, worked tirelessly and around the clock to restore full operation. We resumed partial operations with some concentrate production only 10 days after the incident. Though for the duration of the repairs and improvements, concentrate quantity, grade, and stability were meaningfully impacted.
The reduction in upstream production, reduced concentrate sales, and curtailed feedstock for midstream operations for several weeks ran mid-May into early June. This was easily the biggest disruption our operation has sustained in the past seven years.
While we are not happy about the incident, the creative and coordinated efforts of our team through this challenge were remarkable and inspiring. And we finished the quarter having produced more than 9,000 tons of REO and concentrate.
Once normal operations were restored, we accelerated the progress we have been making on all fronts. And it's quite satisfying to see the momentum we are building in many areas.
Concentrate production per operating hour improved, and June represented one of our best months ever with higher production at higher grade. In addition, we are beginning to realize some of the efficiency gain in our midstream business that I discussed on previous calls.
Notwithstanding the extra downtime that impacted feedstock for the midstream operations into June, we were able to approximately double NdPr oxide production quarter over quarter. NdPr outside quality has been very good, as is the quality of NdPr metal that is now being consistently delivered to our magnet maker customers.
Mechanical reliability of our new assets remains one of the biggest challenges in the midstream operation. But we are seeing noticeable improvement, largely corresponding to process maturity and that certain supply chain impediments are resolved.
As I look ahead to the balance of the year, we will remain focused on uptime and executing certain projects in our upstream business and balancing uptime, process optimization, and production costs in our midstream circuits. Strong concentrate production has continued into Q3.
We are optimistic that non-capital-related optimizations we have implemented this year have moved out grade recovery curve favorably. These optimizations may free up real estate in our concentrator facility for future improvements and/or growth.
Separately, we have to upstream 60K-related capital projects that will begin commissioning this quarter. So we wouldn't count on any benefits until next year.
The first is a large-scale pilot of alternative flotation equipment. The second is the more sizable investment in our grinding circuit that we hope will ensure a tighter and more favorable grind distribution that should improve recovery and throughput potential.
In the midstream circuits, with improved mechanicals stability, we are starting to achieve higher availability. As our teams gain experience and confidence, our execution is improving and we are able to work through issues that previously slowed production.
However, as we have done so, we have identified a handful of process bottlenecks, and have straightforward plans to address many of these through the balance of the year. Our processes to recycle our wastewater is a good example.
These circuits remove impurities similar process wastewater, allowing us to recycle the water and to minimize or eliminate the need for freshwater in our midstream processes. They are therefore critical circuits for water balance, environmental sustainability, and operating cost.
We're not running to their potential. They have proved a modest impediment to higher throughput and uptime. In recent weeks, we have greatly improved their performance in all operating environments, allowing for greater uptime in the rest of the plant in subsequent periods.
We are also laser focused on improving the NdPr recoveries in the leach circuit, where, despite success in cerium rejection, we see a huge opportunity for improved NdPr recovery and even better cerium rejection. Incremental progress will be a major contributor to better leverage of our reagent consumption site-wide.
The impact of this is material, as reagents represents our largest production cost category. Another opportunity was realized in late Q2 and early Q3.
We completed the replacement of both turbines in our power plant that had reached the end of operating life. This was completed with less interruption to operations than feared and allows us greater power production capability at lower cost.
We now expect to be able to sustain our current operations on one turbine for a larger portion of time, even as we ramp the chemical plant further, resulting in lower heat rate and better than previously assumed cost performance. We are very pleased with the outcome and congratulate our utilities, E&I, process controls, and commissioning teams on the excellent job.
Summing up the above and assuming the continuation of current trends, we are expecting a strong quarter of concentrate production and at least 50% quarter over quarter NdPr oxide production growth in Q3 versus Q2. July results put us on this path.
With that, I will turn it back to Jim.

James Litinsky

Thank you, Michael. Let's turn to slide 9 with the recent shot of the magnetics facility at dusk. To wrap up, this was undoubtedly our worst financial performance quarter as a public company. I hope I have always been clear that these kinds of quarters can happen once in a while.
But as I said earlier, beyond the surface results, we made a substantial amount of progress. Our team is spectacular and relentless. Notwithstanding MP's unwavering nature, pricing conditions remain outside our control.
We do know that no producers are thriving at these prices, not even in China. These market conditions have now destroyed most of the [hope for] projects from just a couple of years back. Despite the efforts and investments of many governments, Chinese control over the vast majority of the supply chain remains.
Rare earths, therefore, continue to be at the center of a highly charged geopolitical environment. In June, China tightened regulations over its domestic industry. Meanwhile, in the West, leaders pursue new protective measures have their own.
And with an election season in full swing, the criticality of our supply chain is one of the very few things upon which politicians in Washington agree. How this shakes out over a quarter or two is always uncertain. But what remains certain is the strategic, irreplaceable value of the platform we are building in MP.
In recent weeks, our team has turned a corner at Mountain Pass both on cost structure and ramp, and we are now positioned to scale volumes and complete our evolution as a fully integrated producer of refined rare earths. The timing aligns well with the customer development activities.
As I mentioned at the outset, we recently added another household name automaker to our growing client list. Between this new agreement and new and expanded agreements tied to our Sumitomo distributorship, a significant portion of our anticipated NdPr production at full capacity is now committed.
With our products now qualified by some of the most scrupulous end-use manufacturers and magnet makers outside of China, MP's position as the American champion is now cemented. Our downstream magnetics business is approaching a similar inflection point and, later this year, will return large scale rare earth metal production to the United States for the first time in many decades.
In July, Michael and I celebrated the seventh anniversary of the acquisition of the Mountain Pass mine and the founding of MP Materials. We are incredibly proud of the team we've built, the progress we have made, and even more excited about the future. In a geopolitically charged, increasingly electrified world with billions of AI-enabled robots whether on wheels or legs, MP's products, technology, and platform matters. I believe our long-term success is necessary and inevitable.
With that, we are happy to take your questions. Operator?

Question and Answer Session

Operator

Thank you. (Operator Instructions) David Deckelbaum, TD Cowen.

David Deckelbaum

Thanks, Jim, Ryan, and Michael. Thanks for taking my questions and thanks for all those details. I'm glad to see you guys put this quarter behind you.
I'm curious, Michael, if you can talk a little bit about just this first upstream 60K milestone with the pilot testing around a flotation. That test, I think, is anticipated to begin this quarter in the third quarter. How do you think about that in relation to the impact on your current operations?

Michael Rosenthal

Thanks for the question. In terms of this initial test, we don't expect it to have any material impact on the operation. We'd be running a midstream through this equipment to see the performance, but we do expect it to ultimately lead to additive results even as a pilot, and then we can look to scale that if the results are successful.
We've already tried similar equipment, slightly smaller scale, and have been very pleased with the results. The grinding circuit investment that I mentioned is slightly bigger (inaudible) the potential for more temporary interference with the current operation. But like I said, I'm very excited about that one, particularly, next year.

David Deckelbaum

Appreciate the color. Ryan, I'm curious just as we navigate this year of weak prices. I think you guys highlighted still a robust cash position even with the downtime this quarter. I'm just talking about, I think, the $150 million or so of prepayment and I guess some tax receipts that you anticipate. Can you just rehash the timeline and the conditions that you have to meet in order to receive those payments?

Ryan Corbett

Yeah. Sure, David. Happy to. As you saw this quarter, we received the first $50 million of that $150 million set of prepayments for the downstream business. The way to think about the upcoming $100 million is that is unlocked by further operational progress on the downstream business as we move towards commercial production of metal in Fort Worth and continue to build up our metalmaking capabilities to support magnet making there.
And so we expect to receive the totality of that over the next 12 months. In terms of the incremental dollars that we flagged that are part of the various tax credits that we intend to receive, you saw, actually in Q4 of last year, a receivable of nearly $20 million, which relates to a 45X tax credit which we expect to receive in the not-too-distant future here or upon filing of our 2023 tax return. So that's a pretty near-term receipt.
The other elements of 45X relate to the ongoing 45X credits that we'll earn via production over the course of this year and next. In addition to that, as we announced, I guess it was last quarter, is the $58.5 million of 48C tax credit for our investment in Fort Worth.
We expect to likely receive about $30 million of that in cash next year. And the remainder of that is also available for us to monetize in a variety of different fashions earlier in 2025 time frame. Or we can allow that to play out over a longer period of time on. But suffice to say, we've got a lot of opportunity to continue to support really healthy liquidity and the balance sheet as we invest here.

David Deckelbaum

Thanks, guys.

Operator

Laurence Alexander, Jefferies.

Hi. Good evening. This is Kevin (inaudible) on for Laurence. Thank you for taking my questions. I'm not sure if you can answer the specifics around this. I guess I just wanted to get a sense of maybe the size of the commitment that you have with the automaker that you announced today or maybe the length of time of the agreement? And really, the same for the Department of Defense supply contracts. I mean, just any specifics that you can offer would be helpful. Thank you.

Ryan Corbett

Yeah, sure. It's Ryan; I'll take that. In terms of the OEM commitment, that is a multi-year commitment with deliveries starting this year and ramping through '25 in the following year. We expect over the medium term that this commitment will represent a double-digit percentage of our total targeted output. So it's a pretty significant commitment for us, which we're very excited about.
And I think stepping back and taking both this commitment and the DoD purchase agreement -- I mean, contracts, which I'll give you some details about in a moment, I think it really is emblematic of some of the comments that Jim made about the importance of what we're doing.
I think we're winding to many, many years ago when we started on this mission. The thought that a global automotive OEM would be reaching all the way upstream to directly make this type of commitment was unheard of.
And so I think that that speaks to the progress that we've made both as MP and the increasing importance of our industry and what we produce. The DoD agreement is a pretty modest agreement. So that was announced by them as well.
It's about $11 million commitment from them for the National Defense stockpile. But again, emblematic of continued support from Department of Defense, which were very happy to see.

Certainly. Thank you. I appreciate that. And I guess as my last question, I guess in terms of the Chinese economy and the weakness there, possible stimulus or puts and takes. Obviously, there's a flow through into pricing. I guess, just curious to get what you guys' take on the economy. Your outlook in the Chinese economy would be helpful. Thank you.

James Litinsky

Sure, yeah. The Chinese economy is still pretty challenging. I think we've been seeing writ large around earnings season this year across corporate America. We see that to property. Everything related to it is very weak. It's really just an overall negative environment with respect to industrial production and consumer spending.
So we don't really have any different views than what you've been hearing around the board with respect to what's going on in China. I would say, though, as it relates to the underlying NdPr demand. Anecdotally, we do see Chinese industry growing abroad, particularly in the global South, in South America, sub-Saharan Africa.
So when you think about the pace of what's happening in the Chinese economy, I would say that one mitigant is that Chinese industry is attempting to make up for that with growth around the world, particularly in our space. And so I do think that -- and again, as you know, I always try to caveat these answers with respect to commodity prices.
We don't know. But I do think that in the medium to long term, the things that we're seeing continue to show that the medium and long term demand picture is very bright from China all the way around the world.

Okay. Thank you very much.

James Litinsky

Sure.

Operator

Matt Summerville, D.A. Davidson.

Matt Summerville

Yeah, thanks. A couple of questions. Just back to -- Jim, I'd like to just pick your brain a little bit. What was the criteria you used to make the strategic decision to align yourself with this particular automotive OEM for this agreement?
And to Ryan's comment, double digits can mean a lot. Is there a way that you can maybe narrow that down a bit? And more so, also curious as to whether or not this is causing -- panic's maybe too strong a word, but causing other OEMs to come knocking on the door, so to speak, or lining up at the door to [fund you]?

Ryan Corbett

Sure, Matt. I'll steal that one from Jim quickly. You're right, double digit is pretty broad. I can narrow that down to, say, low double digits. And obviously, we want to respect our customers' wishes and not get into specific details on any individual contract.
But I think that as we look around the world, we certainly are seeing a lot of interest in our commodity, as I mentioned. And I think that what we are most focused on is maximizing the realized price that we receive for a product that we believe is incredibly important. And obviously, as we look at contract structure or pricing and all those things that go together, we are going to prioritize customers that are thoughtful over the long term and are looking at this as partners.
And so that's certainly what we did with General Motors with the downstream business. That's absolutely what we're doing with other OEM partners on the midstream business. And so it's something that we're certainly very pleased about.
In terms of other automakers panicking, I wouldn't put it quite at panic at this point. But certainly, there's a whole host of things that we've seen over the course of the last several months that really, I think, reenergized despite all of the headlines and the Wall Street (inaudible) on the electric vehicle trend. We've seen a lot of activity.
Certainly, the tariff announcement by the current administration, that maybe caused a little bit of -- maybe you can call it panic and in terms of automakers really being interested in what we're doing from the magnetic side on. But overall, we're really pleased to continue to build up our book of customers with blue-chip automakers.

Matt Summerville

Thank you. And then maybe for Michael, is there any way to talk about, based on the progress you're seeing or you saw in Q2 , where you're tracking thus far in Q3 in your goal relative to that 50% production output increase of NdPr oxide? Is there any way to talk about where you hope to exit the year on a run rate basis in terms of NdPr production?

Michael Rosenthal

It's still early, so it's hard to give a clear answer in that. I'd rather not. But I would say that whereas in the past, we saw greater bottlenecks to production, we're now really working on optimizing and balancing the throughput -- sorry, availability of equipment and getting that up and then pushing the throughput further. Whereas in the past, we may have seen more bottlenecks to availability, those are being released. And we expect to see significant improvement.
The data that July puts us on that path to the 50% growth is probably where I'll leave it, but I do believe there's significant opportunity for continued growth in the fourth quarter. Jim might have (inaudible) that. .

James Litinsky

Yeah. The one thing I'd add is I think the reason you're probably sensing a little bit of a shift in tone from us on this is, I think, where we've progressed on understanding and getting our arms around the various moving pieces on increment variable cost. We're at the point, from a cost structure perspective, in the midstream business where getting to our targeted cost structure at this point is more a denominator issue.
So this is more about pushing forward on our volume ramp up will effectively have the cost structure take care of itself. And so, obviously, Michael talked through some of the steps along the way. But we're very, very pleased to see that progress. And so I think that enables us to be even more tactical in how we approach the ramp here over the course of the year.
And so as Michael said, it's early to give specific numbers on volume because there are so many different ways to get to a particular volume number. We want to get there in the most efficient way possible, which is consistent with what we told you over the last couple of quarters.

Ryan Corbett

And then I'll just add since -- to do the trifecta here, so you hear from all three of us on this one. I think -- obviously, this tough quarter numbers aside, to see the scale of this ramp happening in the -- ordinarily, if you just look at our industry or similar industries, when you're ramping up a multi-billion-dollar refinery, these things take time. And ordinarily, if you were looking at this with respect to another business that might not have an existing operation within it, you'd obviously see numbers in ramp and you wouldn't think much of it.
And so the fact that the scale of this ramp is happening as impressively as it is, with Michael on the team and Mountain Pass, I think, is -- I'm very pleased of what's happening out there. And I really do I think, again, tough quarters number aside -- and obviously, we have to keep executing. But this is a really, really impressive ramp that is happening. And obviously, hopefully, we expect that to remain the case throughout the rest of the year.

Matt Summerville

Great. Thank you, guys.

Operator

George Gianarikas, Canaccord Genuity.

George Gianarikas

Hi, everyone. Good afternoon and thank you for taking my questions. I guess I just want to start with -- I was wondering if the song choice was intentional during the waiting period. I assume -- I'm old enough to know what it was. And hopefully, things will only get better. So my first question --

James Litinsky

I'll take that real quick. Just so you know, that's -- I appreciate you recognizing the song. That is a song about resilience and powering forward. And so we're very methodical about our song selection, as you know, every quarter.
So I think one of these days, we'll have to put together a Spotify list of all of our pre-quarter songs so you have them in one place. But that was a deliberate selection.

George Gianarikas

Nice, nice. So maybe to focus on the OEM decision for oxide. I'm curious as to what -- if you are the ones who drove the decision around oxide sales versus magnets sales or is that something that the OEM chose?

Ryan Corbett

Sure, George. It's Ryan. That's something that is mutual, I would say. The reality, though, is that with our current design capacity in Fort Worth, we're fully committed there at this point. And so to bring another significant automotive OEM on would entail incremental capital investment there, which is something that we do look very hard at.
But obviously, our focus at this point is delivering for GM. I think the thing to think about is the size of our upstream business, at least for the near and medium term, is many, many multiples of the size of the downstream business. And so what we're focused on is ensuring, as I said, aligning ourselves with partners that may be long-term partners across all pieces of the business over the course of time.
But certainly, we need to prioritize -- given we've done a great job on the downstream in terms of sales, we need to prioritize continuing to one push forward in line with the right partners on the midstream.

George Gianarikas

And do you know, perchance, who will be making the magnets for that OEM partner?

Ryan Corbett

I wouldn't want to comment specifically. I think what I'd say about that in general, though, is we have seen some continued development in the market for ex-China magnet production in broader Southeast Asia. And so that opens up opportunities for us, both to sell directly to those existing and future magnet makers as well as, of course, these types of agreements where we're providing the material directly to the end customer.

George Gianarikas

Thank you. Maybe as the last question, someone asked a previous question the state of Chinese demand. But I'm curious if you could share your thoughts around Chinese supply. Clearly, some of the major refiners are losing money at current levels.
Just curious as to, Jim, what your thoughts on the sustainability of that and whether or not we could see a little bit of a supply -- at least less growth or a supply reduction going forward. What are your thoughts on what exactly is happening there? Thank you.

James Litinsky

Sure, sure. Well, with my usual caveat that it is always very difficult to read the tea leaves in China, what I would say is that it's very clear that there's losses and, I think, there's a struggle in the supply chain there. Because nobody is doing well in this environment with prices where they are. That is very clear.
I do think that -- there were some recent headlines about China cracking down on illegal production. And I think that there's intent of the government to continue with, frankly, what's been going on for the last few years to crack down on illegal behavior inside and outside the country.
And I do think that, when you think about having a fully burdened cost of production, when you get a lot of that out, that is constructive for pricing and competence and, frankly, better for the environment. And so I think in the backdrop of what is clearly a challenge to macro environment, there are a lot of good trends that suggest that when we see the upcycle, that it should be good.
And then lastly and just going back to what I said earlier, we do continue to see substantial international investment from major Chinese downstream producers. I think we don't see those -- in the US, we see a lot of headlines, obviously, about our OEMs and our market. But let's not forget, obviously, China is the largest auto market in the world. But then there's the rest of the world, and you're seeing real penetration there from the Chinese OEMs and around factories and localized production.
And so I do think when we think about the upstream supply, the Chinese industry has to continue to supply its producers going around the world. But at some point, you do cross that threshold where they're not going to want to have big upstream losses to be supplying local competitors sitting next to them in factories that they have.
So I do think again, with, always, the caveat that nobody knows about prices, I do think that the pendulum -- which, obviously, was very excited a couple of years ago one way has swung too far the other way. And it should recover at some point.

George Gianarikas

Thank you.

Operator

Lawrence Winder, Bank of America Merrill Lynch.

Lawson Winder

Yeah. Thanks, operator. It's Lawson Winder at Bank of America. Thanks very much and thanks for the presentation today. I wanted to ask just to get your thoughts on the magnetic business. And with commercial production being targeted for the end of this year, how do you think about becoming EBITDA positive?
Is that something that will happen this year? Do you think about that happening in '25? And then, what's the ramp on the EBITDA contribution from the magnetic business?

Ryan Corbett

Hi. Sure, this is Ryan. I'll take that. We do expect, once we are producing product in the downstream business -- and just to clarify in terms of the production targets, we aim to be in production with precursor products with metal at the end of this year. We're still targeting production of magnets in the end of 2025.
But in the downstream business, when we do start production -- and so I think the fair read through is for 2025. We absolutely do expect a positive EBITDA contribution from that business nearly immediately as we bring it on. Certainly, we've been investing in that business over the last several years and several quarters.
And you've seen, within some of our G&A and R&D type of line items on the P&L, increased investments there. But we do expect, as we start making commercial product there, to be both gross profit and nicely EBITDA positive in that business.
In terms of your question on ramp and timing, we haven't gotten into specific details there. As you'd imagine, with any type of launch of a new product and new process not too dissimilar to the midstream operation, we will take time to reach our full targeted throughput. And so we expect to embark on that very methodically as well.
And so that will be -- of course, reaching our targeted throughput of 1,000 tons at (inaudible)magnets on will be over a multi-year period.

James Litinsky

And I would just add -- let me just add, particularly, given our tough quarter that hopefully is now behind us after this call. But as a reminder, given that question that -- when we think about Fort Worth, a little over two years ago, that was a grassy field. And this is a highly complex product.
And so the team has been executing really well. And to think that now, we're in this place where, yeah, we're going to be EBITDA positive so quickly, building something from scratch going into some business that is brand new for us and, frankly, the West is really remarkable. So we're moving as quickly as we can on that front.

Lawson Winder

Yeah. Great point, Jim. Thanks for that. And then also thinking about that 2025 from a CapEx point of view, obviously, it's early and I know you guys probably aren't prepared to provide specific guidance. But just directionally thinking about what you provided for guidance for '24, can you perhaps speak to the rough magnitude and direction of the CapEx move, like for '24 to '25?

James Litinsky

Yeah. I think it's early for us to get into any 2025 discussion. Obviously, as you would expect, we are laser focused on return on invested capital and ensuring that we've got a robust and healthy balance sheet. But at this point, it's early to talk about 2025.
We're halfway through '24. And you've seen our results from a capital deployment perspective there. And what I would say is that we gave a range of [200 to 250] in CapEx for 2024. And just looking at the progress to date, we're likely coming in at the low end of that, certainly with our focus on trying to be as capital efficient as possible in the current pricing environment.

Lawson Winder

Okay. Thank you very much.

Operator

Bill Peterson, JPMorgan.

Bill Peterson

Yeah, good afternoon. Thanks for providing all the details. A few questions. So first, coming to the -- back to the magnet discussions. So discussed commissioning the prototype production. I guess, can you provide additional color on the progress thus far?
Have you been producing material that are meeting any initial customer specifications or expectations? And then maybe perhaps you could shed some light on the next steps for the next few quarters in terms of optimizing performance in manufacturing scale-up?

Ryan Corbett

Sure. I'll try to give you some color there. I think I would split the discussion on progress and process development into a conversation around metal versus a conversation around magnets. We discussed, I think it was last quarter or quarter before, a successful commercial scale pilot of our metal making technology, energy and equipment in a pilot facility.
That is what we're bringing to bear in the Dallas, Fort Worth, facility by the end of this year. And so we feel very good about that, given the fact that we proven out our approach to that.
As it relates to magnets specifically, I think the beauty of building this business from scratch is that we've been able to engage with our customer, I think, in a very thoughtful way to think about commercialization of this from the very get-go. And so from our perspective, thinking about the types of magnets that we'll produce, the mix of magnets, the number of SKUs, the performance characteristics, et cetera.
And we've been focused on ensuring that we are biting off what we expect to be able to chew here in the near term. And so we are absolutely not approaching this from the perspective of taking on 100 different SKUs. We're thinking about this absolutely from being able to commercialize rapidly.
What we've seen from the pilot production facility so far is -- and again, I don't really want to pat ourselves on the back too much on this. But we have gone from metal to a finished magnet at performance characteristics that we're pretty happy with already in this plant.
And so when we talk about being in production with magnets at the end of 2025, we get this question a lot. How are you going to be able to do that? Are you going to be bidding spec? And so if we're already making magnets at the pilot plant, clearly, we're working on dialing in all of those processes, our process development, and technology starting now through the end of next year to be sure that we're ready to produce at commercial scale for GM.
It's not lost on us that the automotive supply chain is not the easiest to qualify into. And so that also explains part of the significant hiring that we've had done in Fort Worth to ensure that we're ready for qualification, feedback, et cetera.

Bill Peterson

Yeah, thanks for that. Next question is on robotics. And I know -- I think your entry song last quarter was around robotics. But you have discussed on a few quarters of just being an area for growth. But I guess, have you begun direct discussions with any companies in the robotics value chain or magnet suppliers in that space?
I guess, thinking about potential supply agreement, given you spoke to the OEM one earlier and you also spoke earlier that some of your -- I guess, a large chunk of your volumes are going to eventually be spoken for. Just thoughts around what could happen in that space.

James Litinsky

Sure. And great question. I spend a lot of time thinking about this because as we look around at all the headlines and investment that are going into AI and now, in robotics, what they call physical AI, there's no question that there's a lot of excitement in this area.
And so we expect this to be a gamechanger demand stream. Obviously -- and I hope I've said this thoughtfully the last couple of quarters when we've discussed this, obviously, this a few years down the line as far as something that is of scale. I mean -- but I would say that this is all happening so fast.
The advancements that we're seeing out there -- and obviously, you can follow a lot of this stuff yourself. It's pretty remarkable. And without mentioning whether or not we've had discussions with a particular robotics company or producer or whatever, certainly, you can look at who the leaders are. And obviously, I would say one of the most notable leaders out there that is talking about this is certainly Tesla with Optimus.
And if -- we've seen discussion that Musk believes that that will lead to $30 trillion of Tesla market cap. And we're talking about 10 billion or 20 billion humanoid robots. And what I would say is if you believe that that is one-tenth directionally correct, one-twentieth directionally correct, it is a gamechanger for rare earth magnetics, multiples to what EVs were.
Because there's typically two to five times the amount of rare earth magnet content in a humanoid robot than there would be in an EV. And then we're talking about many multiples of potential production. Again, this is all long term, but this is happening so fast that it's really exciting. But again, I caveat it with this is something that is a few years out.

Bill Peterson

Understood. Thanks again.

Operator

Thank you. Benjamin Kallo, Baird.

Davis Sunderland

Hey, guys. Good afternoon. This is David Sunderland on for Ben. Thank you for sneaking me in here at the end. Just to rewind and maybe at all the way back to beginning. Jim, in your prepared remarks, you talked about NdPr production costs coming down materially over the next couple of quarters.
I'm just wondering if you or Michael could give some color as to what materially means, maybe if we should expect a linear decline or step changes, and anything you can say qualitatively about the levers for this cost improvement. Thanks, guys.

James Litinsky

I am going to hand that one to my CFO. I'll tap out of that.

Ryan Corbett

Yeah, no. I'm happy to take it. We made a couple of comments on this. And I think the important thing to think about in terms of our progression on the cost structure or is that we talked about, for the last couple of quarters, really getting our arms around the incremental variable costs and the items that were driving that, and ensuring that we were not ramping in the face of very suboptimal incremental variable costs.
One important thing that we've found over the last quarter or so is really getting our arms around that. And so we're not where we want to be. But I think about it as -- when we started to ramp the plan, it was trying to figure out where the pins are. Now, we know exactly where the pins are; we just need to knock them down.
And so we know what we need to execute on. And the team is working tirelessly to execute on them as it relates to the variable costs. And so what that really leaves us with is -- with the data that we've seen on a circuit-by-circuit basis over the last several months, what this becomes is a denominator thing.
This is really about ensuring that we are able to maintain the proper uptime. And then as we check that box, continue to push throughput and rinse and repeat and bring more denominator into that equation. Because that's going to be really what gets us to our targeted cost structure. And so a lot of it really depends on exactly what the volumes are and exactly how we get there.
And so it's tough to give you a ton of specifics other than in all the scenarios that we've modeled out, we absolutely, from what we see now, believe that as we drive volume towards our targeted throughput, we will get to our targeted cost structure. It's just a matter of us continuing to execute over the next several quarters.

Davis Sunderland

Thank, Ryan. Appreciate it, guys.

Operator

Thank you. And I would now like to hand the call back over to Mr. Litinsky for any final comments.

James Litinsky

Yes. Thank you, and thank you, everyone. And I guess I will officially say good riddance to this Q2. And I would say for those of us who spend a lot of time on these things, I would say there are quarters that are -- there are bad quarters that are bad, and then there are bad quarters have a lot of good in them.
And I would put this in the latter category. We really responded extraordinarily well to some unanticipated downtime, and things are really humming. And we feel very good about the progress that we made last quarter, that we're making this quarter, and that we're excited to talk to you next quarter . So thanks, everyone.

Operator

Thank you. This will conclude the MP Materials second-quarter 2024 earnings call and webcast. You may now disconnect your lines.