Axcelis Technologies Inc

NASDAQ:ACLS

Earnings Call Transcript 2024-Q3

[Operator]

Good day, ladies and gentlemen, and welcome to the Axcelis Technologies call to discuss the company's results for the third quarter of 2024. My name is Karin and I will be your coordinator for today.

I would now like to turn the presentation over to your host for today's call, David Ryzhik, Senior Vice President of Investor Relations and Corporate Strategy. Please go ahead.

[David Ryzhik, executive]

Thank you, operator. This is David Ryzhik, Senior Vice President of Investor Relations and Corporate Strategy. And with me today is Russell Low, President and CEO; and Jamie Coogan, Executive Vice President and CFO.

If you have not seen a copy of our press release issued yesterday, it is available on our website. In addition, we have prepared slides accompanying today's call, and you can find those on our website as well.

Please note that included in our slide presentation is a table summarizing a correction to our historical backlog numbers, which Jamie will discuss in his prepared remarks. Playback service will also be available on our website as described in our press release.

Please note that comments made today about our expectations for future revenues, profits and other results are forward-looking statements under the SEC's safe harbor provision. These forward-looking statements are based on management's current expectations and are subject to the risks inherent in our business. These risks are described in detail in our Form 10-K annual report and other SEC filings, which we urge you to review.

Our actual results may differ materially from our current expectations. We do not assume any obligation to update these forward-looking statements.

Now I'll turn the call over to President and CEO, Russell Low.

[Russell Low, executive]

Good morning, and thank you for joining us for our third quarter 2024 earnings call. Beginning on Slide 3, we executed well in the third quarter, delivering revenue of $257 million and earnings per diluted share of $1.49. Overall, revenue was largely in line with our expectations as a strong sequential increase in revenue from our image sensor market offset a sequential decline in our power and general mature markets.

That said, bookings in the quarter were softer than we expected as we see some customers digest the investments made into global mature node capacity over the past few years. I'll provide more color on this later in my prepared remarks.

Turning to Slide 4, we show the breakdown of shipped systems revenue by segment, which is almost entirely comprised of mature nodes in the quarter, with power being the largest component.

Now let me review shipped system revenue by end-market, and we'll begin with the mature notes on Slide 5. Revenue from our power markets is approximately 57% of total, down sequentially from 63% in Q2 2024. Shipments to silicon carbide applications moderated in the third quarter. However, on a year-to-date basis for 2024, silicon carbide has been strong, growing year-over-year, reflecting continued build-out of capacity.

We expect fourth quarter revenue for silicon carbide to remain relatively consistent on a sequential basis. The long-term opportunity in silicon carbide remains an important growth driver for Axcelis. [ Yell ] estimates that silicon carbide market will grow from $2.7 billion in 2023 to $9.9 billion in 2029, or a 24% CAGR. And ion implantation is one of the most critical manufacturing steps to make a silicon carbide device.

We are well positioned as the market leader in implant for silicon carbide given the breadth of our portfolio and focused investments we've made in this market for several years. We are deeply embedded in our customers' technology road maps and widely engaged with customers to help them transition from 150 millimeters to 200 millimeters wafer capacity.

In addition, we are seeing interest in our solutions to enable customers transition to trench MOSFETs. The transition to trench architecture is a tailwind for Axcelis given the need for deeper implants require our high-energy tools where we are the technology and market leader.

From an end-market perspective, we are keeping a close eye on the transition from 400 volts to 800 volts electric vehicles over time, which is enabled by silicon carbide given better efficiency compared to traditional silicon. As many of you know, 800-volt EVs can deliver faster charging times and better battery efficiency. And as such, this can be an important catalyst for silicon carbide penetration into EVs, which ultimately translates to more ion implantation required.

In silicon IGBT, revenue in the third quarter was up sequentially, but remained generally muted as we expected, given the slow rate of recovery in the auto industry. And we anticipate demand to remain muted in the near term based on recent order trends. However, I'm pleased to say that we've received our first PO from a customer for our optimized Purion VXE implanter used for the critical silicon IGBT powered by [indiscernible] proton implant application.

Recall from our Investor Day in July, we talked about the benefits of this implant in reducing switching time in high-voltage operation of the silicon IGBT power devices for our customers. The adoption of this optimized implant based on our production-proven Purion VXE platform is a shiny example of the innovation engine at Axcelis. We identified a key application and customer need, designed a specific technology to address it, worked closely with a customer in qualifying our tool, and now intend to commercialize this to drive incremental revenue.

In general mature, revenue moderated in the third quarter, also consistent with expectations. We continue to monitor key end-markets, namely auto, industrial and consumer, which are drivers of our general mature segment, which have yet to show signs of recovery in the near term. That said, upon a macroeconomic rebound in the end-markets we serve, we would expect to benefit from the breadth of our customer base.

Turning to image sensors. Revenue was strong, driven by demand out of China, particularly for smartphone applications. Shipments to the image sensor market can be lumpy, and we expect revenue in the fourth quarter to normalize back to prior trends.

Turning to Slide 6. In advanced logic, we did not have any revenue in the quarter, but we continue to make progress with our evaluation systems, including our Dragon [indiscernible] at an advanced research institution in Europe, and are having meaningful conversations with our customers in the advanced logic space. As a reminder, this is a multiyear development effort, and we are encouraged by the market acceptance and customers' interest to date.

Moving to memory. We are seeing some early signs of activity from memory customers as they look to start adding some DRAM capacity. While we only sold 1 system in the third quarter, we expect additional revenue in the fourth quarter and are monitoring the scope and pace of market recovery as we look to 2025.

Generally, we see DRAM investments outpacing that of NAND primarily due to strong adoption of high-balance memory for AI applications, which is absorbing some DRAM capacity. As we noted last quarter, memory customers typically place purchase orders shortly before shipment. As a result, we have prebuilt some inventory for implanters to memory market and stand ready to respond as demand grows.

Turning to Slide 7. To summarize, our team executed well in the third quarter as we focus on what we can control. As referenced earlier, we've seen continued softness in customers' bookings below our prior expectations and moderation in growth expectations for our key markets in 2025. This is primarily tied to the digestion of capacity in our power and general mature markets, and particularly in China.

As a result, our preliminary expectation is for revenue in the first half of '25 to be lower in the second half of '24. Near-term dynamics aside, we remain very excited about the following long-term growth opportunities that lie ahead for Axcelis.

First, continued growth in adoption of silicon carbide as a key enabler of electrification, power efficiency and decarbonization. Electric vehicles are the poster child of what silicon carbide can do. We expect greater penetration into the EV market, but we also see silicon carbide proliferating to a wide array of applications such as industrial and renewable energy, to name just 2.

With the cost of silicon carbide wafers declining and device-makers generating better economy of scale from growing easy adoption of silicon carbide, we believe the lower cost profile overall silicon carbide devices will lead to more widespread adoption across a variety of applications. And this in turn drives the need for more ion implantation, which is foundational to silicon carbide.

Second, a cyclical recovery at memory and general mature markets, whilst spending on consumer electronics, auto and industrial rebounds. Third, share gains in the advanced logic market where we're exploring new applications for ion implantation in the middle of line and backend of line. And we are actively engaging with customers and driving interest with our evaluation units. Finally, penetrate the Japan market where we aim to extend the success we've made in power to other markets within Japan.

Summing it all up, we are well positioned to drive attractive long-term growth and profitability through the cycles. Finally, I want to thank our employees, customers, shareholders and partners for their continued support and trust of Axcelis.

With that, let me turn the call over to Jamie for a closer look at our results and outlook. Jamie?

[James Coogan, executive]

Thank you, Russell, and good morning, everyone. I'll start with some additional detail on our third quarter results before turning to our outlook for Q4.

Starting on Slide 8. Third quarter revenue was $256.6 million, with systems revenues at $201.1 million and CS&I at $55.5 million. This was largely in line with our outlook of $255 million.

As a reminder, CS&I is driven by our installed base and represents consumables, spares, services and upgrades. As we grow our installed base, we anticipate CS&I to deliver a steady and growing base of revenue and profitability in the coming years.

From a geographic perspective, China remained our strongest region at 71% of total shipped system sales, with the uptick quarter-over-quarter primarily due to higher sales to the image sensor market for smartphone production. We expect our revenue from China to decline sequentially in the fourth quarter and in the first half of 2025 compared to the second half of 2024 as customers digest the build-out of mature node capacity that has been built up over the past few years. Over the long term, we anticipate China to remain an important market for us, albeit at a lower growth rate than non-China revenue.

Now let me turn to our bookings and backlog in the quarter. As we noted in our earnings press release, during our preparation of third quarter financial statements, our internal financial team identified an error in the past calculation of our quarterly backlog dating from 2019 through the second quarter of 2024. This relates to the manner in which we treated prepayments on some of our systems orders that led to a double-counting of these amounts, which has now been corrected.

Specifically for the second quarter of 2024, we previously reported backlog of $994 million, which, following this review, is corrected to $879 million. It's important to note, with the exception of the backlog figures, this change did not result in any correction to any other previously reported financial information. We have provided the corrected backlog figures for 2019 through the second quarter of 2024 on Slide 13 of our earnings presentation.

As a reminder, our backlog may not include bookings received from memory customers given the short lead times following receipt of order and also do not include expected revenues associated with our CS&I business.

Turning to the third quarter. Our bookings were $84 million, and we ended the quarter with a backlog of $758 million.

Turning to Slide 10 for additional detail on the third quarter. Gross margin was 42.9%, which came in slightly below our target of 43.5%, primarily a result of systems mix as well as slightly lower CS&I revenue, which carries gross margin above our consolidated average.

Operating expenses totaled $63.1 million or 24.6% of revenue. During the period, we recorded a bad debt expense associated with the receivable from a small gallium nitride device manufacturer in Europe that filed for bankruptcy. This caused our total operating expenses to exceed our target of $60 million in the quarter. The net charge associated with this totaled approximately $3.4 million and resulted in an $0.08 impact to diluted earnings per share. Excluding this charge, operating expenses were $59.6 million or 23.2% of sales, which was in line with our expected performance for the period.

As a result, operating profit was $46.9 million, reflecting an 18.3% operating margin. We estimate the negative impact of the onetime charge to be approximately 130 basis points to operating margin.

We generated approximately $8.5 million in other income, primarily a result of interest income and the benefit from foreign currency. Our tax rate in Q3 was 12%, slightly more favorable than our outlook.

Our weighted average diluted share count in the quarter was 32.7 million shares, and this reflects continued execution on our share repurchase program where we exit the third quarter with $145 million remaining in share repurchase authorization.

This all translates into a diluted earnings per share of $1.49, which exceeded our outlook of $1.43. Higher-than-expected EPS stemmed from the slightly higher revenue, FX gain and a favorable tax rate, partially offset by the onetime charge I discussed earlier, as well as the slightly lower gross margins.

Moving to our cash flow and balance sheet. We generated $42 million of free cash flow in the quarter. In fact, through the first 9 months of 2024, despite revenues declining 7% on a year-over-year basis, our free cash flow grew by 49%, reflecting our focus on working capital management. As a result, we ended the third quarter with $579 million in cash, cash equivalents and short-term investments on hand.

Moving to our fourth quarter outlook on Slide 11. We expect revenue in the fourth quarter of approximately $245 million. We expect fourth quarter gross margins to be approximately 42.5%, with operating expenses estimated at approximately $60 million. We expect our tax rate to be approximately 15%, leading to an estimated diluted earnings per share of approximately $1.25.

In summary, we are pleased with the performance delivered by our team thus far in 2024. Our cash generation remains strong. We are engaging with customers across a number of key growth opportunities. And we are investing in our product road maps while maintaining discipline in our overall cost structure. All of this, when coupled with our strong balance sheet, put us in position to capture the growth opportunities that lie ahead and drive long-term value creation for shareholders.

With that, operator, may we have the first question, please?

[Operator]

[Operator Instructions] Our first question comes from Craig Ellis of B. Riley Securities.

[Craig Ellis, analyst]

I wanted to start with inquiry on order activity, and hoping you can put more context around some of the slides in the deck. So the question is, can you just frame up what you've seen with orders on an end-use basis and maybe a geographic basis, starting with the momentum that exists in the business exiting 2Q, how things played out in 3Q, and what you've seen fourth quarter to date?

[James Coogan, executive]

Yes. So Craig, thanks for the question. As we look at it, and as we said in our commentary in the prepared remarks, the softness that we're seeing in the general mature and power space is driving the lower order activity. And as a result, that's really informing our view -- our preliminary view on the first half of 2025, with that being lower than the second half of 2024.

As we entered the quarter, we did have an expectation for bookings to be higher, but we continue to see that order activity, although there's a lot of discussions, we see the placement of those purchase orders continue to push out.

[Craig Ellis, analyst]

Got it. And then just following up on the point you made, Jamie, about the first half of '25. Can you talk about, within the view the company now has, what are some of the positives half-on-half and one half '25? What are some of the headwinds? And how does that stack up both on a geographic basis and on an end-use basis?

[Operator]

One moment.

[Russell Low, executive]

In gen power -- okay, sorry. So what I was saying was that we do see the first half of '25 being lower than the second half of '24. We do see like a digestion period for general mature and power, particularly in China. So I'd say the bright spots would be memory. So while we only revenued 1 machine for memory in Q3, we are seeing more activity in Q4. And we expect some of that to continue into the first half.

[James Coogan, executive]

Yes. And thinking through some of the headwinds and challenges, right, the softness in that digestion, the market for a long time now, we've been sort of outrunning some of the cyclical downturns that other peers in our space have experienced. And as we think through 2025, barring a meaningful recovery, 2025 could be down year-over-year relative to 2024. .

But this is a time for us to continue to invest in our road maps and our technologies and work with our customers, Craig. So we see this as, and the strength of our balance sheet and our cash generation, put us in a position to be able to strengthen the company during the period to exit -- accelerate the exit out of the cycle as it occurs.

[Russell Low, executive]

Yes, I think that's a really good point. This is a great opportunity to get bandwidth from our customers and to work really closely with them on new products, so that when the recovery comes, we're ready to go with new products and we can gain market that way.

But just to be clear, the visibility into '25 is not great. Naturally, come our Q4 earnings call, we'll give you a lot more granularity. But this is kind of what we're seeing right now.

[Operator]

Our next question comes from Jack Egan of Charter Equity Research.

[Jack Egan, analyst]

So a few days ago, there was a report that some other capital equipment companies are directing their suppliers to kind of eliminate the use of Chinese components. I'm not really sure to what degree Axcelis uses Chinese suppliers for your components, but have you taken any similar actions? Or do you plan on taking any similar actions?

[James Coogan, executive]

Yes. I mean I think, Jack, not aware of these other reports from the other providers by any means. But what I can say is that our supply chain, on a fairly consistent regular basis, is reviewing our supply base. We are looking to find to improve the localization where we can to limit cross-border shipping as much as we possibly can, to reduce tariffs and duties and freight costs associated with the supply chain base.

We have a very strong supply base here in the United States and in other parts of the world outside of China. We've done a very meaningful job of second-sourcing into low-cost regions and low-cost countries where appropriate. So all I can really say on this is we actively manage the supply chain on a, basically, daily basis here relatively consistently. And we've developed some fairly robust plans, which are outlined in our margin expectations to continue to grow the margins of the business through better and more efficient utilization of the supply chain.

[Jack Egan, analyst]

Got it. Okay. And then I guess this was kind of more last quarter, but you mentioned that the rebound in general mature kind of needed to see a better consumer spending to return to strength. And I wasn't sure if there's really a way to quantify this, but to what degree is general mature leverage to the 3 end-markets you mentioned? So automotive, industrial and consumer. I mean, is consumer always the biggest driver of that segment? Or is it just kind of particularly important right now because of some of the weakness in those other end-markets?

[Russell Low, executive]

Good question. Thank you. So I don't know what the ratios are between consumer, industrial and automotive regarding mature. But what we do know is that each one of those drives demand very much. I mean when you think about a car, it's now basically a computer on wheels with all of the infotainment and all of the automation and the electric ones have complete powertrain. Industrial has been weak. And obviously, you know consumer has been weak as well I mean you can see that through phones and laptops.

So we do think that once those markets start to recover, and 2023 was a great year for those, once those markets start to recover, we can see a rebound in that business. But I can't tell you which end-market has the biggest opportunity.

[Operator]

Your next question comes from Ross Cole of Needham & Co. LLC.

[Ross Cole, analyst]

I wanted to dive a little bit deeper into your GaN opportunity given that you mentioned the European customer had a little bit of a negative outcome there and led to the increased OpEx. How do you see this changing going forward?

[Russell Low, executive]

Okay. So thanks for the question, Ross. So GaN, so if I look at wide-band [ gap ] semiconductors, obviously, that's kind of like the 2 exciting areas are silicon carbide and GaN. They have very different spaces. GaN is typically higher frequency, lower power than, say, silicon carbide. So we're expecting both to have their applications.

Gallium nitride is significantly smaller right now as an opportunity compared to silicon carbide. One thing I would say is that with silicon carbide, it's really intense on our ion implantation steps and doping. While we do dope gallium nitride devices, it's not nearly the same intensity. So for us, silicon carbide and silicon IGBTs are very interesting because they drive a lot of implant steps. Gallium nitride, it does drive implant steps; it is currently what I would consider a niche application.

[James Coogan, executive]

Yes. And this was a very -- as you'd imagine, Ross, this is a very customer-specific issue, and I don't think it foretells anything broadly about GaN or power device segment broadly.

[Russell Low, executive]

Yes. It was a small customer in Europe. It wasn't one of the big guys that have talked about gallium nitride.

[Operator]

Our next question comes from Tom Diffely of D.A. Davidson & Co.

[Thomas Diffely, analyst]

I appreciate the chance to ask a question here. Just wanted to focus on the transition of silicon carbide from 150 to 200 for some of your customers. I'm curious, what is the opportunity for you for a fab that's transitioning versus new greenfield fab?

[Russell Low, executive]

Tom, thanks for the question. So it's actually interesting. 150 to 200 transition seems to actually be picking up speed. We're working with many, many customers on 200, and a lot of them have actually taken delivery of 200 millimeter capabilities. So this is the thesis we have and this is kind of what we're seeing, is that most people are going to bring up their 150, optimize their processes, optimize their yields. And then as the availability of 200-millimeter wafers becomes available and the quality continues to improve and the price continues to drop, they will bring up their 200-millimeter capacity on brand-new machines. They're not going to transfer straight away. So they'll have the 150-millimeter running and then they'll bring up the 200-millimeter.

Once the 200 millimeter is up and stable and more cost-effective, I think you would expect to see the 150 potentially get retrofit, assuming the rest of the equipment of the fab is able to be retrofit. Because remember, a lot of these fabs that people started with were old fabs that they're reutilizing. So there will be some fabs that will upgrade and some that will not be able to.

[Thomas Diffely, analyst]

Just to clarify though, your tools are 150, 200 capable with a bit of a kit, is that right?

[Russell Low, executive]

Yes, they are. So all of their tools are [ fill ] upgradable. So if you think about that as an upgrade opportunity, then we have a large installed base and that would be a great opportunity for our aftermarket business in the future. But we still are expecting to sell a lot of new tools into greenfield as well.

[Thomas Diffely, analyst]

Okay. Great. No, that helps to clarify. And then just a follow-up question on the margin side. When you look at the margins down about 150 basis points year-over-year, is that simply overhead absorption? Is it product mix? Could you provide a little more color there?

[James Coogan, executive]

The answer to those is yes, yes and yes. So we had -- it's a little bit of overhead absorption, it's a little bit of systems mix, and slightly lower CS&I volume period-over-period. So it's systems, CS&I and a little bit of overhead absorption just given the lower volumes in the period, Tom. .

[Thomas Diffely, analyst]

Okay. But hasn't been any kind of unusual pricing pressure?

[James Coogan, executive]

Nothing that we've seen materially. I think we always get challenged on price from customers to do better and find opportunities. But this really does come down to systems mix, CS&I mix of between systems and CS&I, and then ultimately, the overhead absorption just given the lower volume.

[Operator]

Our next question comes from Christian Schwab of Craig-Hallum Capital Group.

[Christian Schwab, analyst]

Great. I guess as far as memory is concerned, you guys seem to have increased enthusiasm about the recovery of DRAM spending, while other memory guys have been more excited about an increase in NAND spending on the movement to 300 layer this year. So I'm just wondering if you can give us further clarity on your enthusiasm for DRAM.

[Russell Low, executive]

Sure. Christian, it's Russell. So I think that the big thing is that we sell implant as placed on wafers out, whereas when you go to like 200, 300 more layers, that actually increases the kind of density, if you like, of many tools. So [ Deponit ] wins big when they add more layers. For us, it doesn't make a significant difference. So we're all about more capacity coming online.

So the reason we're excited about DRAM is that HBM has taken up a lot of DRAM capacity that has then pushed up the utilization. So what we're seeing from some of our customers is now an investment in DRAM. The existing fabs are kind of being slightly optimized. You're getting a few tools to optimize the new -- the existing fabs. But really, the big opportunity for us is going to come when they actually have the new fabs come online.

So I think you've probably been reading in the news that some of these fabs have been pulled in and they're going to be used for DRAM. So we're excited about the DRAM utilization going up and driving additional demand.

NAND is still quiet for us, again, and is wonderful for the [ Deponit ] guys. For us, it's the same implant intensity, whichever node it is. Does that help, Christian?

[Christian Schwab, analyst]

That does. So when would you -- you're building capacity, so would we assume that increased DDR5 wafer starts in essence for DRAM. I think that's what you're talking about, you would anticipate that maybe in the summer going into, hopefully, a better smartphone and PC sales cycle in the second half of '25? Am I putting the pieces of the puzzle together correctly there? .

[Russell Low, executive]

So yes, just kind of take a step back. So as you know, we are strong in Korea. So yes, DDR5 and advanced nodes are where we're going to be focused. The older nodes like DDR4 and DDR3 are probably going to be manufactured in China. I think we've probably said before that we're really not participating in that. So we're very focused on the advanced stuff.

The transition from all the different nodes hasn't really changed the implant intensity. So again, it's the number of wafers out that's important. And we can expect that there will be new capacity coming on in 2025, as you mentioned. I think some of our big customers have mentioned the pull-in of fabs or the reuse of fabs. So we are expecting to see DRAM orders.

Just to kind of give you a data point. So we talked about there's only 1 machine in Q3. We are expecting more activity in Q4. But that activity is what I would consider optimizing existing capacity just to get the most out of the 4 walls, Really the big CapEx is going to come from the greenfields, which has been publicly noted.

[James Coogan, executive]

Yes. And Christian, I'll just add that kind of the scope and scale of that recovery is difficult to predict exactly when that's going to occur. But we'll have more -- we should -- as we get greater visibility into 2025, we'll have more information on our views on memory as part of our fourth quarter earnings call.

[Operator]

Our next question comes from Jed Dorschmeier of William Blair.

[Mark Shooter, analyst]

You have Mark Shooter here on for Jed this morning. Just touching on what you're seeing in power in China. There are -- they seem to be like the last bastion of EV growth. So could you provide some color on the capacity and the digestion in that region that you're seeing? What needs to occur? .

[Russell Low, executive]

Mark, it's Russell. So yes, I mean -- so I would say go separate the EV mobile market, which they are very big on, versus the actual domestic supply of silicon carbide chips. So we've mentioned before there's like 20 different customers. The Tier 1s have a little bit more capacity, but a lot of them are Tier 2 companies that don't have a lot of capacity.

One of the things we're seeing is that, as companies have brought in their mini lines, and they're ramping those mini lines, they've taken a while to get the [indiscernible] down across the whole fab, improve their productivity, their yields and the reliability.

So I think it may not be so much a pause because they've got too much capacity, because I'm not sure that market is completely supply-demand driven. What their aspirations are to supply domestically silicon carbide to their own silicon carbide EVs, so think more of it as a taking a time to mature the processes and ring out the processes.

[Mark Shooter, analyst]

Okay. That makes sense. I appreciate the color, Russell. Can you touch on, as a follow-up, like how much of your backlog is China sick? And have you seen any changes to the patterns and pushouts or cancellations there due to the things you're talking about?

[James Coogan, executive]

Yes. So we don't give specific breakdown on the backlog by segment or region, Mark. But I think what we've noted in the commentary and what we said in our prepared remarks is that the softness in general mature, power, and particularly in China, is, in the third quarter here, is what was driving the lower bookings rate. So yes, that is going to feed into our expectations for the first half, that preliminary view we have on the first half of 2025 relative to the second half of 2024.

And when we think about that, it wouldn't be unreasonable to assume that Q1 could be lower than Q4, obviously. I think that's an obvious statement. But Q1 could be lower than Q4 as a result of that.

Broadly speaking, 2025, we need that recovery in the general mature and power space to kind of see growth. Again, unless there's some meaningful recovery there, we could be down year-over-year. But we'll have more full year commentary as we prepare our results, get greater visibility into 2025, and have our earnings call in the fourth quarter.

[Operator]

Our next question comes from David Duley of Steelhead Securities.

[David Duley, analyst]

I was wondering -- I saw that the China revenue, well, I think, was 71% of revenue in the third quarter. Could you give us an idea of what you think the percentage will be in Q4 and in the first half of next year? I'm assuming that the Chinese revenue is what is going to be down in the first half of next year, but maybe I'm wrong.

[James Coogan, executive]

Yes. So I think, David, it's a great question. The elevated levels here in the third quarter were largely attributable to the image sensor business within the region as we supported 1 of our customers there who's building out some incremental capacity for smartphone.

I think if you were to kind of normalize that, we had said coming into the year that we'd be somewhere in that sort of 40% to 60% range by quarter, so going into Q4, we would expect that to come down, the China exposure, to come down back into those relative levels. Obviously, mix by region is going to be a primary driver of that once we get there.

Going into 2025, you're right, and as we said in the prepared remarks, I would say we are seeing softness in general mature and power broadly. And then particularly, we're seeing some softness in China as we look to the first half. So we would expect that, as a percentage of sales, right, we have to see where the ultimate mix is to determine what that would be as a percentage of total sales. But we are seeing, I think, preliminarily, we're seeing the actual dollar values come down.

[David Duley, analyst]

Okay. And then as far as silicon carbide in China, in the first half of next year, would you expect that to be down versus the second half of 2024? Or is silicon carbide still strong in China in the first half of next year?

[Russell Low, executive]

So I think we're expecting silicon carbide in China to be down in the first half of next year compared to the second half of this year. And I think we are attributing that to a digestion phase. So we've still got a large backlog. We have seen a number of customers kind of push out that backlog, push out the delivery dates, which probably is weighing a little bit on our book-to-bill, right? Why buy -- why place another PO when you still got outstanding POs that you're waiting for.

But I do think it's a digestion period. And I think it's a period of time where you're going to see the maturation of the processes occur. So the yield improves, the reliability improves. And then once that's got more maturity, then I think you're going to see some of these companies expanding.

Because remember, a lot of these are like pilot lines, mini lights. A lot of the Tier 2s aren't actually in high volume. So they want to make sure they got their act together before they drive meaningful capacity.

[James Coogan, executive]

Yes. So the customers are focused on reliability at this point so that they can more meaningfully provide into the domestic EV production market and drive the cost down to be able to support that self-efficiency goal. And we're partnering with our customers in that region to help them do that.

And so during this time of digestion, we actually are going to continue to make investments in that portion of the business, work with their customers and their road maps. And again, like we said a little bit earlier, continue to position us to accelerate on the sort of return to growth inside of the markets.

[Russell Low, executive]

I think that's a great point, Jamie. So we've kind of said that the silicon carbide customers in China are underweight relative to the opportunities for EVs. And naturally, their goal is to become a world provider of silicon carbon. So they've got a long way to go, very much in the early innings. So we still see this as a cycle on what is otherwise a strong secular growth opportunity for us.

[David Duley, analyst]

Okay. A couple of follow-ons. Just maybe you could help us with the utilization rates of your equipment. You could hear on the U.S. and European SiC conference calls of your customers, business is not really growing at this point or they're in digestion mode or trying to figure out lowering their CapEx and whatnot.

So I'm just kind of curious what the utilization rates are in the U.S. and Europe? And then are they much different in China than they are in those geographic regions?

[Russell Low, executive]

Are you talking -- any given market type? You're talking about memory or certain -- which specifically are you talking about, Dave?

[David Duley, analyst]

Just in general, the utilization rates of your tools in the marketplace, if you'd like to care. I'm more interested in silicon carbide than anything else, but however you'd like to answer the question would be great.

[Russell Low, executive]

I just want to understand.

[James Coogan, executive]

Yes. I think broadly, what we're seeing is that the continued softness, and obviously, with memory being where it is right now, we are probably seeing a little bit of a spike in some utilization as tools are being used to help drive HBM. But that is limiting the sales of new tools, and so they start adding capacity, as Russell noted.

I think as we look at general mature and expectations for power, you're going to see utilizations, given our CS&I numbers are down period-over-period, that will probably give you some indication of the utilization rates, is that's going to be an indicator of overall utilization, is the spares, consumables opportunities we have inside of that mix.

So again, although we don't comment on specific customers, regions or markets, I think generally we're comfortable saying that utilizations are down.

[Operator]

Thank you. This concludes the question-and-answer session. I would now like to turn it back to David Ryzhik for closing remarks.

[David Ryzhik, executive]

Thank you. I want to thank everybody for joining us on this call. Operator, you can close the call.

[Operator]

Thank you so much for your participation in today's conference. This does conclude the program. You may now disconnect.