Sheffield Resources Limited (SFX) Earnings Call Transcript & Summary

[Audio Gap] A good quarter this one, first quarter of cash flow positive. The format will be as usual. Please submit your questions via the Q&A throughout the presentation. Bruce will have around a 15-, 20-minute presentation, and we'll dive straight into that Q&A. So Bruce, why don't you take it away? And then we'll get engaged with the audience after.

Great. Thanks a lot, Peter, and thanks, everyone, for joining. Yes. As Peter said, certainly, the March quarter was a significant quarter for us. First, a significant turnaround in terms of zircon sales quarter-on-quarter or even, let's be honest, half on -- quarter-on-half. We -- having really struggled with sales in the second half of calendar year '24, very, very strong recovery in the March quarter as a result largely of having additional customers, which has made a big difference to our customer base, therefore, a broader base of people to sell to. And we saw that with the very strong sales in the quarter. As a result of those sales, first quarter of operating cash flow positive from the project, which was very pleasing. And then the other pieces which we'll focus a lot on today is we now have the forward plan. I think we've spoken on a few of these calls about the challenges we had and the need to come up with a plan for how we were going to ultimately overcome the oversize challenge, et cetera. And we obviously have come up with that now. And pleasingly, it doesn't require a lot of capital. So we'll talk through that, spend a bit of time on that to make sure it's understood what we're going to do and in what time frame and why we're confident that, that ultimately will deliver a robust production outlook for us in the future. So with that as sort of introductory remarks, sort of step through the usual presentation. Just at the start, I mean, as always, as I say, I presume most people on the call know who Sheffield Resources are. But we are mineral sands focused. Our primary asset is the Thunderbird project in -- which is part of the 50-50 Kimberly Mineral Sands joint venture, currently producing circa 900,000 tonnes a year of concentrates from 10 million tonnes of ore. We'll talk a lot about it today, but we do have a plan to grow that up to 16 million tonnes a year of mining, producing over 1 million tonnes of products per annum. And then that -- there's still more potential in the reserve. So ultimately, that will partly but not fully utilize the reserve at Thunderbird. And then we do have our other interests. We continue to work on the option in Brazil, progressing both the resource statement and the PFS and our ongoing interest in Capital Metals, who own the project in Sri Lanka, we are currently drilling. Just a couple of pictures for orientation as always. So left-hand side is a recent picture of mining. Actually, while I was on site in late March, near the end of one of our mining blocks, we can see the DMU, the pumping unit and the dozers. Those are working at the base of the ore there. So you get a pretty good feel of the total deposit from top to bottom, starting with sort of [ Pendang ] at the top and the [ Malaga ] sandstone and then into the ore layers, T1, T2 and then down to the base of the deposit. On the right-hand side, process plant, actually not really a picture of the process plant, but from the process plant looking over the ilmenite stockpile. So we had -- still had a pretty chunky ilmenite stockpile at the end of March, and that's the bit of it that was on site. In terms of the revised business plan, the way we sort of think about this is we talked last year a lot about -- or in previous years about having really 3 big challenges. One was the oversize resulting in lower Rougher Head Feed, our costs being higher than expected at FID. And while I don't have it on there, we had the challenge of selling zircon. Now clearly, with our results in the -- or the sales of zircon in the March quarter, we're a lot more confident about our ability to sell zircon on an ongoing basis. We have a broader customer base. So here, we're focusing on the production fixes. Now offsetting those negatives were some positives. We've had good process plant recoveries, and that continues. What's new, I guess, new information that we've gleaned over trials over recent months is we've been trialing running the mine at a higher push rate, effectively demonstrating whether our dozer fleet and DMU can actually sustain a higher mining rate. And using sort of 24-hour-and-longer tests have demonstrated that ultimately the dozer fleet and DMU can support a mining rate of up to 16 million tonnes per annum, and that would be after allowing for availability and so on. And then what we've also established is that the oversize material is essentially we're preferentially losing non-heavy mineral and certainly non-valuable heavy mineral so that there's not a lot of value in trying to further break down oversize. It's all about removing it as early as we can in the process. So based on that, the plan that's been developed is that we will be ultimately aiming to increase the mining rate to 16 million tonnes per annum, and we expect to achieve that on a quarterly basis, so full quarter run rate targeting the first quarter of FY '27. So effectively July to September of 2026. And there are some key steps we need to deliver that. And that's obviously a ramp-up over effectively 5 quarters from now, including the current quarter. And there's a number of elements to that plan that we need to execute. The first one was the increase in waste mining capacity. While we can sustain the sort of 10 million to 12 million tonnes that we've been saying 2.5 million to 3 million tonnes a quarter, with the existing waste mining, we would need to waste mine faster to sustain a higher mining rate. And so earlier in the year, we started drill and blast. We continue to see very good improvements in diggability with the drill and blast material. And as we transition to a new contractor, as part of that, we have a larger excavator being mobilized to site along with the larger truck fleet. So we're moving to a sort of 260-tonne excavator and 140-tonne trucks. And the expectation is that when that new equipment contractor is fully mobilized in May, by the end of May, we should be able to then increase our rate of waste mining significantly and build in-pit inventory to enable us, thus, to then increase the rate of mining with the DMU. Within the DMU itself, we've demonstrated that the DMU and the dozers can push the equivalent of 16 million tonnes and that the DMU can actually handle it. That's partly as a result of some modifications that have continued to be made to the DMU in terms of materials of construction of certain key bits and changes to make it more reliable and to handle the higher rate and the amount of oversize. And given the -- that we're not losing valuable heavy mineral in the oversize, there's an expectation that over time, what we'll do is likely reduce the screen size in the DMUs to remove more oversize in pit. And we are looking at whether there's an optimization of oversize handling. At the moment, we use excavators to move the oversize from the back of the pit to an interim stockpile, and then they get moved by excavator and truck to the waste or to build in-pit walls. And the opportunity may be to install a conveyor on the back of the DMU so that we don't rehandle the oversize waste. So that's an optimization point. Not a lot of capital involved in either of those, and it's just about getting the best possible performance out of the DMU. At that 16 million tonne rate, based on the trials we've done and our understanding of the oversize and all of those things, that would enable us to sustain Rougher Head Feed of the -- at the original design rate of 1,085 tonnes an hour. In reality, we're likely to vary the mining rate to keep the process plant full. Our expectation is that with the capacity to mine at 16 million tonnes an hour, we will fill the process plant, which means that what we will be doing is varying the mining rate depending on grade and so on, grade and oversize proportions, to make sure that we're delivering 1,085 tonnes an hour consistently to the wet concentrator. With 1,085 tonne feed and our current process plant performance, which is essentially above design on recoveries, we'd expect to produce between 900,000 and 950,000 tonnes a year of ilmenite concentrate and 220,000 to 240,000 tonnes a year of zircon concentrate. Both of those numbers are about 10% above the original design. So ultimately, while the expansion to the mine gets us to the designed Rougher Head Feed, we expect to end up with more than original designed products from that. We have sufficient haulage and port capacity to move that volume. In particular, with the port, the Kimberly Marine Supply Base, second wharf facility at Broome is currently under construction. And that creates additional capacity. Although the current port could handle it, having the second wharf would just give us more flexibility to ship out through Broome. Overall, given increased mining rate, increased production, we're expecting unit costs to come down. There's a certain portion of our cost that is fixed. And while aggregate cost will go up a bit with the extra volume, we're certainly expecting to see unit costs come down. And so we expect to see some cost efficiencies as a result of that expanded production. Pleasingly, we don't anticipate material CapEx or working capital being required to implement the new plan. The modifications we're talking about to the DMU are fairly minor sort of modifications. They sort of fall more into the category of sustaining capital rather than significant capital projects. So overall, that plan, given that it will get us additional products above original Stage 1 design, that defers the need to consider a Stage 2 expansion. And again, we -- while there is clearly the potential to expand production beyond even the 16 million tonnes within the reserve, we don't now anticipate the implementation of Stage 2 to be really this decade. It's likely deferred significantly beyond where was anticipated at FID. And that really reflects the fact that as a result of this expansion to 16 million tonnes per annum, we effectively have a low -- very low capital expansion that will give us additional production and so reduces the need for a Stage 2 expansion in the short term. So with that, that's a lot on the forward plan. What I'll do now is just talk a little bit about the quarter we just had in terms of production and then what does that look like when we map the forward plan as well so you can see a little bit of the 2 combined and then cover off a little bit on the cash generation. So in terms of the production in the quarter, this is our mined production. We fairly -- have been fairly consistently mining around that 2.5 million tonnes per quarter. As I've mentioned earlier, we have been working with Piacentini to make changes -- progressive changes to the DMU and operation and maintenance. And ultimately, we are comfortable that over time, we could -- well, even what we're doing now, we can increase availability. And then as part of our implementation of the higher push rate, we'll be able to push that to 3 million and beyond to 4 million tonnes per quarter ultimately. The WCP, so Rougher Head Feed, as we've said for a number of quarters, we continue to see 75% of our expected Rougher Head Feed largely as a result of the higher oversize. Grade typically has been above expected. It was a little bit lower in the March quarter. It reflected some edge effects. The particular block we were at was quite close to the edge of the deposit. We're still generally seeing grade above block model overall. In the process plant, so zircon, obviously, over time, we had forecast design capacity to slowly increase -- sorry, design recovery to slowly increase. We've sustained our high -- recoveries higher than design. So it's converging in, but the process plant continues to perform at or above its design, both for recovery and product quality. Ilmenite, similar story, higher recovery than design, generally higher ilmenite than expected. It's -- we don't do anything in particular in the way we operate the plant. It's just how the material is performing. And as a result of that, we have been getting higher than the base design -- or the base agreed price for this product. Although it's a fixed price contract, the price is per percent titanium dioxide, and the price is actually higher if the product is high quality. So where we've been -- if we're selling at a plus 40% shipping, a plus 40% ilmenite concentrate to Yansteel, we actually get more than $123 a tonne, and we've seen that in the recent quarter. So when we combine sort of what's actually been happening in the history with what we're forecasting going forward, essentially, this is the tonnes mined. We've been mining it at 2.5 million tonnes. We expect in the current quarter with improved availability to be pushing up a bit beyond that. And then ultimately, as I said, in the first quarter of FY '27, aiming to be able to sustain 4 million tonnes per quarter of mined material. So a significant increase over where we've been. And as I said, with the current oversize, 75% oversize, we expect to get -- sorry, 75% of forecast Rougher Head Feed, that will translate into achieving our design Rougher Head Feed of 1,085 tonnes an hour. So effectively, expanding to 4 million tonnes per quarter fills our process plant. On zircon concentrate production, so again, production and sales. Production has been growing steadily. I guess the variability has been on sales, and we're all aware of that. And as you see, the Q3 sales of over 80,000 tonnes was not just a record but effectively represented pretty much 6 months of sales at our current production rate. A big catch-up on the -- particularly the sales from the first quarter of FY '25, and that really reflected the fact that we have more customers. We're forecasting in the Q4 to ship 40,000 to 50,000 tonnes, essentially ship what we produce potentially a bit of the remaining excess inventory, but there's not a lot of excess inventory left. We're down -- we had just over 20,000 tonnes of inventory at the end of March, and we would expect normal inventory levels to be about 10,000 tonnes, sometimes a bit higher if there's a shipment early in the following -- due and a bit less of a shipment has just been made. And then with the Q1 FY '27, as I said, we're expecting 55,000 to 60,000 tonnes a quarter of zircon concentrate, which is the 220,000 to 240,000 tonnes per annum, and we're expecting to sell what we produce. Obviously, the obvious observation we might have had in the past as well, can you be sure you can sell the extra product. Now while markets are always uncertain, et cetera, the fact that we shipped over 80,000 tonnes in the quarter does give us confidence that ultimately selling 55,000, 60,000 tonnes a quarter is not -- it's something we've done effectively. So we can be reasonably confident that as we grow that production, that there should be a market there for that product. We come to ilmenite. Sort of less similar story, a bit more variability in production. This just reflects the fact that we -- the ilmenite production largely is what it is. We don't do -- it comes out the side of the process plant. We produce what we produce. Generally shipping what we've been producing in the March quarter. We deliberately deferred one ilmenite shipment basically to prioritize zircon sales, and there was a short outage at Broome late in the month. So we expect to catch that up in the current quarter. We are looking to ship significant volume in the current quarter, in fact, what would be equivalent to or above our record shipments for ilmenite and to be producing sort of in line with what we've been doing, if not a little bit above. And then by FY '27, Q1 FY '27, looking at that, 900,000 to 950,000 tonnes a year equivalent of zircon -- of ilmenite concentrate, all of which will go to Yansteel under the existing take-or-pay agreement. In terms of cash flow for Q3, very different cash flow chart to the one we've been showing previously. Fairly obviously, revenue being higher than OpEx on a cash flow basis means we generated operating cash flow of $32 million. It's worth noting that the revenue and the offtake prepay are understated in this chart because we did settle part of the prepay with physical product. So there's no cash effect. So in fact, the operating cash flow was actually higher than the -- than is shown in this chart. The repay -- we repaid one of the 2 facilities from Yansteel. We still have $8 million -- sorry, yes, we still have an $8 million facility in place, and we did settle part of -- as I just said, we did settle some of the prepay with physical product. We continue to have the ability or support of Yansteel. And to the extent they're required, we would expect to be able to -- or KMS would expect to be able to utilize a prepayment facility from Yansteel to support short-term working capital needs for the business. I mentioned before inventory, we do still have, I'd say, a little bit above normal inventory for both products: zircon, about 10,000 tonnes; ilmenite, normal ilmenite shipments. It's probably about right. We normally expect to have about 1 month of stock. 75 is a little bit of -- 1-month shipment of stock, 75 is a bit above that. But timing of shipment, it will always be a bit above or a bit below depending on the timing of the shipment. Most of the CapEx is going into the tailings storage facility and the in-pit wall construction. So effectively, it's an in-pit. It's taking waste and depositing it to create the walls, which will eventually form our in-pit tailings cells. And probably last point to make on this, we did pay 2 interest payments in this quarter because we had deferred an interest payment, first of all, from September to December, and then December was into the current quarter. We paid -- sorry, into the March quarter. We paid both of those in this quarter. So there was no interest outstanding at the end of the quarter. Cash ended up more or less where it started basically because we've used a significant amount of the operating cash flow developed to repay the offtake and clear -- ensure that we were up to date with all of our interest payments. But generally, I mean, a very, very healthy quarter but recognizing the fact that we sold significantly more zircon than we would normally sell given that we were on -- it was a catch-up quarter. And I think that's probably a good place to leave it. Peter, happy to take questions.

Brilliant. Thanks, Bruce. Just to start off, in terms of the zircon sales for this quarter, how much of that was due to new customers? And how much of it -- well, obviously, we had Chinese New Year this quarter as well. Just trying to get a feel for sort of why we saw so much demand in Q1.

Look, it was largely due to new customers. I mean we're still shipping some to existing customers. But effectively, yes, Chinese New Year was in there, which meant there were some timing effects. Actually, Chinese New Year was in late January. We shipped a lot of product in February and March post Chinese New Year, which is quite common. The -- we've certainly seen -- I'm trying to think in that quarter, but I think most of the 86,000 tonnes or 81 dry tonnes went to new customers rather than existing customers. So we effectively use that new demand to get -- to introduce product and get people to take -- who hadn't taken the product before to take material volumes.

Okay. And following up, we've got an audience question here, which follows on quite nicely from that one. The realized zircon price increased by $1 per tonne despite Iluka quoting a 7% decrease in premium zircon. What was the driver for Thunderbird's resilient zircon pricing, noting that contained ZrO2 and produced zircon concentrate was slightly stronger than previous periods?

Yes. Look, so there's a little bit of a quality effect taking place. We did, I think, ship a little bit more. But in reality, I know there's -- a lot of the market commentary was we've seen prices going down Q-on-Q. What we saw was that -- we've said it previously. We've seen our price in the first quarter. The March quarter would be similar to what it was in the December, and it was, and we're seeing that now. We haven't seen a lot of pushback from customers at current pricing levels. I think part of it is that ultimately, concentrates -- I mean we're shipping into China. Concentrates is what most of the -- I mean China is becoming a concentrate market. So I think there is a different pressure on zircon pricing than there is on concentrate pricing necessarily. I also think as well, we saw a bit of a catch-up effect. I think we saw other people's prices were still coming down perhaps to meet the market a little bit. So certainly, we're sort of seeing a stable pricing environment at the moment for the concentrate products.

Okay. Good. Obviously, let's talk a little bit more about the business improvement plan. Ramping up to 16 million tonnes per annum, have you done any testing on the plant to see how it handles that sort of throughput?

Yes. So that's part of what we've done as the tests have been run. So you can imagine, if we've been pushing the equivalent of 16 million tonnes per annum into the DMU, we've been getting the undersize generated. And so we actually have operated the -- so in those test periods, we've had close to and around the design feed, and that's allowed us to see what will the wet concentrator do if you put 1,085 tonnes an hour in it. And out of that, what I'd say is we've demonstrated most bits were fine. As always, when you run flat out like that, you identify the odd bottleneck that effectively the way it was -- the way things are set up. And so we've used those trials to identify those and make minor tweaks in the plan to handle it. So we're very confident the wet concentrator can handle that production. And we've also taken the opportunity to run the CUP at the higher rate to reflect the fact that cannot actually consume all of the concentrate if we -- of the HMC we produce. And equally, we've seen that as well. So we've been able to run trials to demonstrate that the system end-to-end, at least over a short period of time, can sustain the 16 million tonnes per hour. And there's no particular reason why we wouldn't expect it to sustain it over longer once we get to that consistent production rate. So essentially, the way we think about it today is if you think about that ramp-up, why you don't go straight to 16 million tonnes is ultimately we need to relax the waste constraint and then get ahead. We need to create some input inventory, sustain -- get that sustained and running so we know we can clear the waste fast enough and then ultimately build up the mining rate steadily to the 16 million tonnes so that as you're stepping up, if you're going to run at 12 million tonnes per annum equivalent for a full quarter, do that, make sure that everything is working and just keep moving it up. So it's a sustained sort of careful responsible ramp-up rather than just try to run -- running too fast, too quickly.

Yes. Good. There's a few more questions here. There have been a few questions that I think would probably not be allowed on the ASX listing rules. So just to bear in mind that, obviously, we can't say things that haven't been press released or pretty material. But following on from that, there is another question there that is really about the confidence level. So essentially, they're saying, obviously, the 16 million tonnes per annum target using the same dozer fleet and DMU, what gives you the confidence that you can get a 60% increase in productivity?

Effectively, we've demonstrated it. When I was on site, we were pushing at that rate that the plant handled it. There was no sign that the DMU was struggling to handle that sort of volume. So I think we've done the trials to show that. We're continuing to do them. We're doing periodic trials at a higher rate. The more you demonstrate it, the more you learn. But certainly, there's no -- all the trials indicate that that's a sustainable rate, that the system can handle it, that there's nothing -- we're not going to break anything by doing it. Yes, it just -- we've demonstrated that it's possible.

Okay. Good. Something else I wanted to touch on as well was, obviously, the plant has been running above design capacity at the moment. Do you think -- how confident are you that, that can continue doing so when it is at max capacity or...

Above design recovery rather than above design capacity. Yes. I think, look, again, what we've seen when we've had higher feed, obviously, you have to -- people appreciate that when you're running, your set points and so on will change. But what we've seen is that when we've run -- when we've had those opportunities to run at those higher rates that the recoveries -- the plant has still performed well. And what we've seen in terms of bottlenecks has been more about where you see the volume. Do you have enough? Are you able to -- does the material clear quickly enough from somewhere? And then you just have to look at like orientation of an outlet pipe, those sorts of things. So we haven't really seen any indication that the overrecovery is because the plant is running under throughput that when we've had high throughput through the plant, we've still got good recoveries.

Okay. One here from Chris Baker. How did the wet season impact production?

Yes. I mean we -- I think everyone knows we did have the one -- we formally advised the market that we had a gas-related shutdown when the road north of Port Hedland closed, and we didn't get LNG export imports for a few days. Overall, I think the wet season is certainly from a practical perspective, the guys on site, I mean, it's a challenge managing through the wet season. We had what would probably be considered a normal wet. There was quite a lot of rain. The pit gets wet. We -- there's a lot of trackability issues and things like that. But ultimately, they managed through that fine. We had the gas interruption. I think we had one other significant rain event where we actually suspended mining operations for a period of time just to protect the plant. We have a very pretty sophisticated wet weather planning process in terms of progressively suspending noncritical activities. And what we did find a number of occasions during the year -- during the wet season, we will suspend haulage operations. You can imagine an unsealed road, the big heavy trucks, 120-tonne net loads. When you're running them on the wet road is not good for the road. So what -- we more frequently suspended trucking operations during particularly heavy rain or immediately after and then started up again. But we've always allowed for that in our operating model. So it didn't ultimately cause any delay. So I think overall -- not because it's easy. There was a lot of good work by the operating team, but they managed through the wet season very well.

Okay. Good. Another question here. Bruce, what sort of percentage reduction in unit mining and processing costs are you expecting post the implementation of the improvement program in Q1 FY 2026?

FY '27 would be when we would expect to see that full cost. Look, we haven't put an estimate of that out. So I'm not going to give one here on the call. I mean people can -- you can imagine, common with a lot of mining operations, circa 50% fixed cost, 50% variable. So if you increase production by 60%, you can kind of get a feel for what sort of unit cost decreases you might see.

Okay. Good. Can you please explain the increase in C1 cash costs for the March quarter, noting how inventory management impacts it?

Yes. Look, so there's a lot of effects with cash in any given quarter. Some of it's timing of payments and so on. And there is inventory effects. You can imagine from a pure accounting perspective, when you sell a lot more zircon in a quarter, you're writing in a lot more higher-value product, which is being held as inventory. Overall, I think we still see sustained -- we still believe that, that $55 million to $60 million a quarter at the current production rate is where costs are. You can have little timing effects here and there. So we're comfortable that, that is the true cost base of the business.

Okay. One here on China and -- has the Chinese slowdown caused any problems at all?

Look, we haven't seen any real -- I mean I don't think there's a new slowdown in China. We're all aware that the Chinese economy had been slower over time. Certainly, we saw pretty decent demand, have seen pretty decent demand so far this year. There's -- I guess we all know there's a bit more uncertainty now than there was a month ago. But we haven't -- I guess at the moment, the people we sell to sell their product domestically. A large part of their customer base is domestic. So we're probably looking at second or third order effects of the various things that are going on in the global economy at the moment. So we certainly haven't seen any obvious impact in demand at this stage.

Okay. Another question here. So current HM grades of around 20%, obviously a lot higher than the resource grade of 11%. Is that something that you think you can sustain?

Yes. I mean we might remember the original logic for a Stage 2 expansion was that you would -- ultimately, grades decline over the life of mine. And so the original Stage 2 expansion was you double the mine capacity, but you only had about a 50% increase in products, and that was effectively offsetting that grade decline. So I think fundamentally, over the 30-year -- 30-odd year mine life, as grades decline, you would expect that we would increase the mining rate to sustain production. Now whether you increase it -- increase the mining rate to increase production even further is a different question. So it just reflects the fact that like most operations, you start -- you mine higher grades first. I mean we're not -- certainly not high grading the deposit. The high-grade area is quite large. But ultimately, you're starting in the highest-grade areas. And then over the life of the mine, you will see average mine grade come down.

Okay. And another question here. So really what they're asking is, would 2 DMUs be potentially more effective at running at, say, 50%, 70% than running one at, say, 90%? What was the rationale? Obviously, there's [indiscernible] there.

I think you always have this question about, but would you spend $20 million to add capacity you don't really need. I mean if your existing facility can handle that, there's also some complexities about a second DMU involves a second mining phase, a second set of pipe work. You need to waste mine in a second area and so on. So there's a working capital implication in that. So I think it's not to say that you would never go for a multiple location mining. But ultimately, we're pumping -- when we get to 16 million tonnes, what we're pumping to the process plant is what we would have always expected to do from one DMU. So we wouldn't -- it's not obvious that the answer would be to spend more capital to end up with the same production rate.

No, that makes sense. Is Yansteel happy to take an additional 40% ilmenite from 2027? Will the additional tonnage fall under the take-or-pay?

Yes. So the -- because we're still in -- well, yes is the simple answer. It does to both. It does fall under the take-or-pay. Yansteel are -- we wouldn't be doing it if Yansteel weren't supportive of taking the extra product. It's really not -- if we -- if you break it down, we're really talking about 10% more than design, not 40% more than they expected. So ultimately, they were expecting to get a certain volume, and they're going to get a little bit more than that, not 40% more. So certainly, they're very supportive of the plan to increase production and take the ilmenite.

Okay. That makes sense. Another question here, just asking about the customer base. Is it all within China? Or have any of the new customers come from outside of China?

In terms of shipments, all to China so far. We have -- or KMS have both signed up or have MEPs in place for at least one non-China customer. They have not yet taken any product. I think there's an important distinction to draw that a lot of the market outside of China is ceramics. And the quality requirement for ceramics is -- well, first of all, ceramic market is pretty weak. And the quality requirement is quite exacting, particularly for material like ours that without an acid wash largely presents a standard grade. So China is a good market for it. There's a lot of the chemical and non-ceramic applications. A very deep market, so an ability to blend and handle that, and in some cases, people looking at doing acid wash to increase the quality of at least some of that ilmenite to true premium. So there is a strong natural fit for our product with the China market. So while we continue to work with potential customers outside, their general ability to take large volumes of our product is limited by the markets they ultimately sell into.

Yes, yes. That makes sense. We've only got a few questions left here. So if anyone does have any questions, please do submit now. Otherwise, we will be wrapping up. Bruce, would you mind just talking about the debt and sort of how you're looking at potentially paying that down over the next couple of years?

Yes. So look, the first -- the debt facility, the 2 facilities, NAIF and Orion. The Orion facility is repaid or repayments start first on that. It's the commercial debt. NAIF is longer dated. The first payment for the Orion debt is due in mid -- in June. Certainly, the -- we're working towards making that payment. That's the intention. I think what I would say for people is if you think about that forward plan, we've got a production profile now in terms of final products, which is different to what we had when we did the original debt sculpting. The payments -- the June payment is not a large payment. It's USD 11 million. But then the shape of the payments was that the largest payment was actually in December this year and then sort of reducing down. That repayment schedule was based on a certain production profile. And we would expect that, I mean, essentially with lenders, a discussion about not a refinancing or that sort of thing but actually how you align the payments to the production profile. So while we're certainly comfortable that the asset can generate the cash it needs to repay that debt, you have to look at the timing at which those payments are made. So they're relatively long-term facilities. And there's scope to change the payment profile, which we would expect is -- it's a fairly normal process to go through with lenders now that you're actually operating, what is the right payment profile for the debt given what the asset is actually generating in terms of cash. Also recognizing that the price environment we're in is lower than it was anticipated to be, and that's also a consideration in working out what the optimal timing of debt repayments is.

Okay. Cool. Any news on South Atlantic at all? Obviously, I know it's been a big quarter for the business improvement plan and sales and whatnot. But is that still on the radar?

We're progressing the resource -- predominantly the resource work following the drilling, both at Retiro and at Bujuru and the PFS. And certainly, we are quite close to having a resource for that. We've been progressing it as appropriate in the background. It's not been a big focus for us, but the work has continued on that in the background.

Okay. Nice. Okay. I think that's it for questions, Bruce. So I guess any final thoughts before we wrap things up?

No. Look, I think certainly appreciate the chance to provide an update. As I said at the start, very pleasing quarter in terms of the transformation of zircon sales made a big difference. Cash flow positive for the first time is a great place to be in and a medium-term plan to really get the operation back to where we wanted it to be, which is that 1,085 tonnes an hour of feed, and that will be a robust business. So it's good to have both the short-term performance but also that medium-term plan to address the production problems or challenges we faced in the first year. We've now got a plan to overcome those and fortunately a plan that doesn't require a lot of capital.

Perfect. Okay. Well, look, we have been recording. So if those who want to view this afterwards, it would be on the YouTube channel and over the social media channels that Sheffield have. And I'm sure if there's any other questions, there's info e-mail at the bottom there. It will be on the website. So I'm sure you can get in touch, and Bruce or Mark will respond. But thank you, everyone, for attending, and thank you, Bruce, for the presentation.

Right. Thanks very much. Thank you.