Good morning, and welcome to Entegris's 2022 Virtual Investor and Analyst Meeting. I'm Bill Seymour, VP of Investor Relations. Ana, please, advance. I'll pause here on the safe harbor. Just as some quick housekeeping items before we get started. For best performance, we recommend using probably a moot point at the moment, but we recommend using the latest version of Google Chrome, Firefox, Microsoft Edge. This event is being streamed, and we recommend that you listen via your computer speakers. There's no dial-in for the webcast. It's only available from the link that we provided. The webcast is being recorded, and you can access the recording via our IR website later this afternoon. Please note you can adjust the individual boxes within the webcast panel you are viewing today by clicking on the options in the upper right-hand side of the box.
The PDF of the presentation, you probably saw that, is also available for download. The Q&A will be done via the chat feature, so if you can use the Q&A box on the left side of the screen to send your questions, or you can email them to me directly at bill.seymour@entegris.com. Please put your name and your firm name in the question so I can acknowledge who's asking. We would expect today's session would last about an hour and a half, including Q&A. In today's session, to be clear, we will not be providing any update on the current quarter or the full year. The speakers today are Bertrand Loy. You can advance the slide, please, Ana. President, CEO, and Greg Graves, our CFO. Before we begin, I would like to remind listeners that our comments today will include some forward-looking statements.
These statements involve a number of risks and uncertainty, and actual results could differ materially from those projected in the forward-looking statements. Additional information regarding these risks and uncertainties is contained in our most recent annual report and subsequent quarterly reports that we have filed with the SEC. Please refer to the information on the disclaimer slide in the presentation. On this call, we will also refer to non-GAAP financial measures as defined by the SEC in Regulation G. You can find a reconciliation table in today's news release, as well as on our IR page of our website at Entegris.com. With that, I'll hand the call over to Bertrand.
Thank you, Bill. Good morning, everyone. Ana, if you could advance to the next slide where you can see the major themes we'll be covering today. Putting aside the short-term uncertainty, our conviction in the strong secular growth prospects of the semiconductor industry remains unchanged. Material science, material security are at the core of our value proposition, and they are increasingly important to our customers' technology roadmap. This will translate into growing Entegris content per wafer and help sustain strong top-line growth outperformance for the years to come. Naturally, another major area of focus for us for the next few years will be to unlock the full potential of our combination with CMC Materials, a transaction we closed in early July 2022. Today, we will reveal details around the timing of realization of our cost synergies.
We will describe the growth potential of the new combined platform and frame how we plan to deleverage our balance sheet and expand our EPS over time. Next slide, Ana, please. Maybe we can actually dive right in with the next slide. With the CMC acquisition, our platform, our Entegris platform, will now be comprised of four divisions of relatively similar sizes, and I will provide more detail around each of them in a moment. For now, let's focus on some of the other key facts on this slide. 80% of what we do is unit driven. These unit-driven products can be chemistries and materials that are consumed daily in semiconductor manufacturing fabs, or they can also be advanced filters and other consumable products that are single use and are replaced very frequently. The remaining 20% of our revenue are CapEx driven products.
They can be flow controllers, dispense systems, and other fluid handling components that we sell to equipment makers. Or they can also be wafer carriers, gas purification systems, or advanced fluid handling solutions that we sell when new fabs are built. You can see in the middle of the slide that our customer base is also very unique. Our solutions are widely used across the broad industry ecosystem, and it is particularly true for our advanced filters and advanced materials handling solutions, which are enabling unmatched levels of material purity from the point of bulk chemical manufacturing all the way to the final point of dispense onto the wafer in the fab environment.
As you can see further to the right, when it comes to our fab customers, we have greater exposure to logic foundry, representing about 70% of our fab revenues, while memory customers would represent the rest, approximately 30%. The final point I want to call out on this slide is that our mission as a company has not changed. Our purpose remains the same. As we will discuss, our combination with CMC Materials has added some very precious new capabilities, which will allow us to bring increasingly unique new solutions to the market faster. Anna, if you could advance to the next slide, please. Well, it's a slide that I felt compelled to insert given the prevailing nervousness about a potential industry slowdown.
It's a slide to reinforce our belief that Entegris is arguably one of the most resilient platforms in the semiconductor industry ecosystem. I like to think about resilience in a few different ways. As previously mentioned, 80% of what we do is unit-driven. That ties to wafer starts, so fab output, as opposed to the more volatile industry CapEx. Also, remember that in the past few years, we've been steadily increasing our content per wafer at the leading edge. As more wafer processing continues to migrate toward the leading edge, this will add another layer of buffering. Also remember that the solutions we develop are tailored to very unique customer process recipes. In other words, our solutions are sticky with high switching costs and very long tails.
Finally, because our solutions are broadly used across the industry ecosystem, our customer base is more diverse than traditional suppliers to the semiconductor industry. Now, I want to be very clear, I am not suggesting that we are immune to a downturn, because we're not. What I'm trying to press upon you is that our business is more stable relative to most other platforms in the industry. Let's go to the next slide, Anna, if you could. Our management team is second to none. Many years of industry experience managing very effectively across a broad range of industry conditions. It's also a team with very strong roots in the various legacy platforms comprising Entegris today. Dann Woodland, leading the new APS division, obviously joined us from legacy CMC Materials. You can see Mike Besnard or Jim O'Neill, formerly part of ATMI.
Stuart Tison or myself, formerly part of Mykrolis, and Bill Shaner and Greg Graves, who were part of the original Entegris platform. That goes without mentioning Sue, Neil, and Olivier, who bring the unique experience and rigor of large industrial platforms. This team has been working very effectively together for many years, and together we've been part of many complex integration processes. This great diversity of backgrounds, diversity of perspectives, is what shapes our culture. A culture that is intensely focused on the customer, intensely focused on collaborative innovation, and a team that is very committed to excellence in everything we do. I generally believe that this culture has been the source of our competitive advantage and will be the foundation to our future success.
Now, after this long preamble, I would like to move to the next section and start addressing the impact of our combination with CMC Materials. And let me start with the updated sales growth algorithm for the new combined company. Top line growth is always on our mind as a management team, and this has been a major area of focus during our integration planning phase, as you can imagine. Let's go to the next slide, and maybe the punchline first here. We expect to deliver on a larger platform with similar growth of organic top line growth outperformance versus what we presented last year. To be clear, here you're looking at a pro forma organic-only growth rate. As mentioned before, despite the short-term uncertainty, we expect the semiconductor industry will continue to enjoy a secular growth rate of twice GDP growth.
We expect to outperform the industry by 3%-6% organically. Why is that? Well, first and simply put, we believe that materials science and materials purity will be the major drivers of future chip performance. As such, we expect greater Entegris content per wafer over time, which will drive our growth in excess of the industry. Second, we expect to gain market share as we continue to compete effectively and as we continue to contribute our unique portfolio of capabilities, and we combine all of that into highly differentiated mission-critical solutions for our customers. Greg will show some illustrative models and provide some additional details in his own section. For now. Let's go to the next page and discuss some of these drivers at high level.
As you can see on this slide, it took the semiconductor industry nearly 40 years to reach the $500 billion mark. A level of revenue that the semiconductor industry reached last year. There is growing consensus that it will take less than 10 years to double that number, as we expect global semiconductor revenue to reach $1 trillion by 2030, give or take a year. The digitalization of our lives is expected to continue to accelerate. Will reshape in very fundamental ways, everything around us, from our transportation and healthcare systems to the way we work and interact with one another or interact with machines. All of which will drive exponential demand for faster, more energy efficient and more reliable semiconductors.
To get to this $1 trillion level, we expect many more fabs to be built, and we expect a steady increase in fab production. Now, again, I want to be clear that I'm not dismissing a potential industry contraction short term, and we will manage through this effectively as we have in previous industry down cycles. As we do this, as we manage through this potential soft patch, it would be essential that we remember that demand will snap back to the trend line, and it will or may snap back to the trend line faster and harder than before. My message here is simply that the semiconductor industry is a great place to be despite the short term uncertainty. This is particularly true if you are Entegris, as you will see on the next slide. If you could advance, Anna, please.
This slide is built from the customer's perspective, looking at the Entegris portfolio. Think about it. Our customers have a major overarching goal, and that is to improve chip performance. Performance attributes for semiconductor devices can be faster computing, greater energy efficiency, lower cost per bit, lower cost per gate, or better device reliability over time, et cetera. Our customers have two major tools in their toolbox to achieve this. More complex chip architectures and miniaturization of the critical dimensions on the wafer. The technology roadmaps of the industry leaders are calling for new materials with better structural and electrical properties, as well as more precise polishing and etching chemistries. Of course, they need ever greater process purity to achieve optimal yield and optimal cost.
At Entegris, we operate at the intersection of material science and materials purity, and this is precisely what our customers need right now. The punchline of this slide is that the compounding process complexity of our customers' technology roadmaps will be the source of exciting growth opportunities across the Entegris platform. I would like to talk about our rate of outperformance in the context of our expectations for each of our four divisions. Let's start with the two divisions that are not directly impacted by the CMC integration. Anna, let's go to the next slide, please. We're starting with our microcontamination control division, which develops advanced filters and purifiers. These solutions have become essential for our customers in their quest for yield optimization.
As feature sizes continue to shrink and as new materials are adopted, the permissible sizes and classes of contaminants, as well as their concentration levels, are becoming very, very stringent. As a result, the push to achieve ever greater process purity translates into greater usage of advanced filters, greater frequency of replacement for those filters, and also includes the introduction of more points of filtration further upstream in the fab ecosystems. All of which is expected to drive an outperformance level of five to seven points for our Microcontamination Division. Given the growing importance of these solutions, we expect to sustain operating margins in the mid-30s%. Next slide, please. Our Advanced Materials Handling Division has printed some very strong results in the past three years, and we expect this strong performance to continue.
In this division, we develop solutions to enable safety and to protect the integrity, the purity and stability of chemistries, wafers and masks during transportation to the fabs and during processing in the fab environment. While these solutions are different, the drivers are very similar to what I just covered in the, you know, the previous slide for the MC division. Defect control, defect prevention has become increasingly complex and important for our customers. Our customers, as a result of that, are driving new purity standards within their ecosystems and within their fab environments. That ultimately drives demand for a broad range of AMH products, including our fluid handling and wafer carrier products, as well as our EUV pods and high purity chemical containers.
Please also note on this slide that this division is the one that is most exposed to the industry CapEx. About 60% of the revenue is CapEx driven. Going forward, we expect top line growth of two to four points in excess of the industry and operating margins in the low to mid-20s. Now let's turn to the two divisions impacted by the combination with CMC Materials. Next page, please. This is about our new SCEM division, which represents approximately 23% of our revenue. Here we added the QED, ITS, and PIM business from legacy CMC Materials to SCM, and we moved out all preexisting CMP-related products previously part of SCM.
For example, our Entegris post CMP cleaning chemistries or the recently acquired Sinmat and Viasep slurries have been transferred to our fourth division, APS, and I will review APS next. Let's stay with SCEM for now. With this new scope, our SCEM division has a 90/10 mix of unit CapEx for its revenue. The primary mission of the SCEM division is to develop newer and better performing materials and chemistries that will improve the performance of next generation semiconductors. The device architectures are becoming incrementally complex, so we need and we expect the adoption of these enabling new materials to accelerate. As a result of that, we expect the SCEM division to grow three to six points above the industry.
On the margin front, you should know that many higher margin products have been transferred out of SCM to APS, and lower margin products have been added from legacy CMC Materials. The net impact is an operating margin target in the low 20s%. This new mix is really the reason for this decline versus the previous era. We expect the operating margin will continue to improve as new products with better margins are introduced to the market over time. Next slide, please. Our final slide will be our Advanced Planarization Solutions division, which will focus on the CMP module, broadly defined to include post CMP cleaning and other electronics materials. The mission of the APS team will be to position Entegris as the undisputed leader in consumables used in and around the CMP module.
The historical growth rate has been a bit slower than the rest of the Entegris platform, but we believe we can reignite growth and sustain two to four points in excess of the industry as we unlock the full potential of the APS platform. Polishing is one of the most challenging processes in the fab, with a lot of complex interactions between the wafer, the slurries, the pads and pad conditioners. Entegris is now the only supplier of the entire solution set. This will allow us to better understand and then optimize these interactions. The result will be better polishing rates, lower cost of ownership for the customer, which we expect will translate into market share expansion and ultimately growth in excess of the industry. We expect these, you know, incrementally differentiated solutions to deliver very attractive margins in the mid- to high-20s%.
This is probably a good pivot to move to the next slide and start discussing a little bit more the CMC integration. Before I dive into any details, let me just say that we started the integration planning the minute we announced the transaction. We've been following the same methodology that we've been using for every single acquisition we've made in the past decade. A methodology that has served us very well and that has allowed us to unlock great value from the ATMI acquisition and many more since then. Let's go to the next slide, where I want to talk a little bit about the core principles guiding our integration work.
They are listed here on the top row, and everything flows from that. Through this integration, we want to maintain a strong focus on the customer and not let the integration become a distraction for the organization. As a company, as a management team, we also believe in common systems, which, once implemented, are the foundation for our operational and financial excellence. This is also why we believe in speed. We obviously want to quickly realize the cost synergies we committed to, but we value speed, most importantly, because as I was mentioning, we want to minimize the distraction caused by the integration, and we want to allow the organization to quickly focus on the positives of the combination. In other words, turn the page on the headcount reductions and refocus the collective energy onto the customer and the revenue synergy initiatives.
I can tell you that the response by the APS team to these integration guiding principles has been overwhelmingly positive, and this bodes very well for our common future and future success. As an example, I just would like maybe to flag that the organizational structure of the new Entegris platform was largely finalized during the integration planning pre-close. We communicated to each Entegris employee within two weeks of closing if and how the integration would impact them individually, which is very, very fast by any standards. As you can see, we've been making very fast progress on many critical fronts. I will probably start diving a little bit in each and every one of these, and maybe we can start on the next page with the cost synergies.
You should know that we expect to migrate to a common ERP platform by early summer 2023. This will allow us to achieve our cost synergy targets by the second half of 2023. We expect to spend approximately $75 million to realize these synergies. The graph illustrates at a high level the timing of realization. All numbers are run rates, so a run rate of $40 million achieved by year-end 2022. A run rate of $60 million by the second quarter of 2023, and a synergy run rate in excess of $75 million by the end of 2023. We expect approximately 250 positions to be eliminated, and I would like to stress that 30% of the impacted personnel will be positions at Legacy Entegris.
In other words, as we always do, we're using this combination as an opportunity to improve our talent pool wherever possible. Next slide, please. Sorry for this call. Before I start here, let me be clear that these revenue synergies are included in our growth algorithm. I think on page 10 for Entegris and page 16 for APS, those formulas take into account these synergies. Please, no double counting. That said, I want to call out these opportunities separately. First, because the revenue synergies are significant, and also because these synergies are the source of great excitement within the team, and these revenue synergies are why this combination is so compelling for both our customers and investors.
As you can see, we expect cross-selling opportunities to be realized in the next one to three years. The work has already started soon after closing, and we have started to record some early wins. Within a two to five year timeframe, co-optimized new products will be developed and introduced to the market, and in parallel of that, a broader end-to-end solution selling strategy will be introduced, which will benefit all four divisions. All of which combined is expected to generate incremental annual sales of $75 million by 2025 and will exceed $200 million by 2028. I will not break these numbers down for you, but allow me maybe to expand a little bit more on the concept and the objective of this end-to-end solution strategy. On the next slide, please.
The way we built this slide is, you know, reminding you that our customers really need best-in-class technology for sure, but they also need fast time to solution. After all, they want to be first to market with new device performance, and that means, for example, be first to market with incremental layer counts in 3D NAND or first to market with gate-all-around at two nanometer or below. Our end-to-end solution strategy is our answer to this customer's need. The new films the industry is hoping to adopt will be thinner, and the materials will be harder to polish. Our SCM division is developing these new films, and APS will have now an early access to film composition. This will allow our APS division to quickly engineer new abrasive solutions and optimize the interaction of the various components in the CMP module.
In turn, the new abrasive solutions will likely be harder to wash away after polishing. Our development teams working on post-CMP cleaning chemistries, which are now part of APS, will have unique insight in the slurry blend and will be able to quickly optimize our post-CMP formulated cleans. I think you get the point of this continuum of capabilities, which will of course also include best-known methods of filtration and industry-leading in-line metrology around the CMP module. Our goal here is to offer to our customer best-in-class technology with optimum time to yield. You should know that our offering, our capabilities are unique in the industry, and we believe they will be the source of new differentiating and faster time to solution, which will drive market share gains in the served applications. Next slide, please.
Another major area of focus has been the assessment of the CMC Materials portfolio. During the past 10 months, we have objectively assessed the strategic fit and the attractiveness of all parts of the CMC Materials portfolio. Simply put, our goal was to define if Entegris was the best owner for the various parts of the business. You can see on this slide the types of criteria we used for this exercise. What I can share with you today is that all portfolio decisions have been finalized. What is also true is that we are still defining how and when to communicate and implement these decisions.
I think as all of you will understand, there is some tension between, on the one hand, the desire to simplify the portfolio, reduce the debt, and improve our financial profile, and on the other hand, the desire to maximize proceeds for these divestitures. We have a solid balance sheet, and while we recognize that timing the market is possible, we are ready to be patient and smart about our next steps. You should know that no benefit from potential divestitures is included in the financial analysis included in this deck, simply because we did not want to speculate one way or another. Let me now turn the stage to Greg Graves for the financial section. Greg?
Thank you, Bertrand. Today, I'm gonna talk a little bit about the capital structure and then about our profitability model moving forward. Next slide, please. Starting with the capital structure. This is our capital structure post-transaction. It doesn't show up in the second quarter 10-Q because we didn't close until after the quarter ended. We have about $6.1 billion in debt. Left-hand side's permanent debt. Right-hand side, we'll be paying down the bridge loan first, which currently has a rate of a little over 7%. You look at the magnitude of the debt, I don't wanna underplay it. $6.1 billion is a lot of debt. What gives me some comfort though is we've done everything possible to de-risk that debt.
First of all, we have no maintenance covenants on the debt. So there's no. If EBITDA doesn't reach a certain level or paydowns don't meet a certain pace, we don't have an issue. The second piece is we have very minimal required paydowns. In the next 18 months, the only thing we have to pay down is that bridge loan and $25 million of the term loan. Finally, I just wanna comment on the interest rate structure of the debt. Initially, the term loan is typically a floating rate instrument. At the beginning of Q3, we hedged that interest rate, and so only about $500,000 of that term loan is floating rate. $2 billion of it is fixed at essentially a SOFR rate of a little less than 3%.
The rate on that will be about 600 basis points. The paydown. That hedged portion steps down consistent with our anticipated paydown of the debt. What I want you to take away from this slide, no question of a large quantum of debt. When you think about the structure of that debt, we think we've done everything possible to de-risk that. Next slide, please, which I believe Bertrand covered in the period of time where I was having technical difficulties. The one thing I do wanna highlight on this slide, though, is number two, the paydown in debt and capital levels. I mean, harbor no illusions, that is priority number one. I mean, we're obviously gonna make investments in our business.
At the end of the day, we know that deleveraging is critical for a number of reasons, not the least of which is that with the interest rates where they are is the significant EPS benefit. Each $100 million of debt pay down of the term loan saves us about or it was about 3.5 cents of EPS. That shorter-term loan, the bridge loan that's at 275, and its rate's over 7% today, the pay down on that will be even more accretive to EPS. With that, I'm gonna assume Bertrand's covered the rest of what's on this slide, and we'll move to the next slide. I do wanna talk about our cash flow and the environment there and kinda where we are.
You look at our cash flow for the last 12 months, it has been suboptimal. I wanna point out that the reasons for that were not a surprise to us. They were decisions that we had made. There's also reasons that those decisions that we've made historically, things are gonna improve as we go forward. First of all, on the left-hand side of the page, our CapEx, which is typically run about 7%, historically run about 7% of revenue, in the current year, on a pro forma basis, it's in the mid-teens. I mean, we'll spend about $500 million in capital as Entegris standalone, and then CMC would be above and beyond that.
As we move forward, as a percentage of revenue, and I would just say also on an absolute basis, I mean, we expect 2023 CapEx to be down from the levels that we had in 2022, and we will normalize at a level of 9%-10% of sales on the CapEx side. Point here is, you know, 2022 represents peak CapEx. We'll manage it. We're not gonna go back to that 7% level where we ran historically, but we will migrate down to 9%-10% of revenue. On the inventory side, same thing. This is on a pro forma basis with CMC. Our turns in 2022 are right around three times. That's down from a little bit over 3.5 times.
Our expectation is that over time, we will get back to that 3.5 level. Again, this decline in turns was not an accident, I guess I would say. I mean, we intentionally invested in inventory to be able to protect our supply chain and deliver for our customers. When I talk about protecting our supply chain, if you look at the increase in inventory, the majority of that has been in raw materials or inputs to be able to support our output. The takeaway from this slide is, not terribly pleased with the cash flow in 2022, but feel very confident that our cash flow will improve as we move out into 2023 and beyond. Next slide, please.
Just looking at the debt pay down algorithm, we are, as I said earlier, committed to rapid pay down and deleveraging the balance sheet. What's gonna drive that pay down, obviously, is the cash that we generate through improving EBITDA over time. I'm not suggesting that it will necessarily improve next year, but I'm saying we will improve EBITDA over time. I talked about the lower CapEx and the improvements in net working capital. We've got opportunity today. Our liquidity position is very strong. We've got about, on any given day, between $550 million and $600 million in cash. A little more than $300 million of that is in the U.S., so we've got an opportunity to repatriate additional cash back to the U.S, and then obviously potential divestitures.
The slide that I'm showing here, in terms of our repayment curve, does not take into account any potential divestitures. You know, based on our strategic plan model, I mean, this is what things look like. If we do achieve divestitures, it will be an improvement on this. The other thing I want you to take away is we do have a track record. This is not a sort of playbook that's new to us. Those of you who were around back in the days of ATMI, which we did in 2014, our initial leverage levels went to a similar point. You know, two years post-ATMI, our gross leverage was down to about three times.
Three years after the acquisition, it was about two times, which is relatively similar to what you see here. The assumption here in 2020 is that it by 2025, we'll have paid down about $1.4 billion in debt. On an organic basis without, you know, without taking into account, divestitures. Next slide, please. Let's shift from the balance sheet and leverage and de-leveraging to our financial targets as we move forward. Next slide, please. We talked about Entegris being a value compounder. Our growth algorithm, Bertrand laid it out. He didn't put numbers behind it, but I'm gonna put illustrative numbers behind it. Sales growth over this three-year algorithm compounded approximately 9%. The assumptions behind that is that with the the industry would grow about 4.5%.
That's different than what we would've shown you last year in terms of the industry growth and takes into account the broad industry expectation that there'll be some near-term softness. MSI, we expect to grow about 6% over the next three years. We expect capital to be flattish. When I'm putting these expectations out, I'm really talking about what industry analysts are expecting. I mean, we don't have any special window into the industry that's different than the forecasters or different than other industry participants. Then our growth above the industry, approximately 3%-6% on that algorithm. You blend the 6% MSI over the next three years with a flat CapEx environment. Let's be clear that we don't expect those to be straight lines. Those are compounded numbers.
You get a 4.5 number. You take the midpoint of our above market at 4.5, and it gives us a 9% number. 9% growth with our algorithm of 40% flow-through to the EBITDA line will result in 12% EBITDA, which is a 32% EBITDA margin. Note that's up from about the 30% pro forma that we talked about this year. We expect to achieve that through you know, modest improvements in gross margin, obviously leverage on the OpEx line. We don't expect to get leverage on the SG&A or R&D line.
In fact, the R&D line is negative to the leverage 'cause as Bertrand said, our goal is ultimately to get to 8%-9% from the 7%, that we've, you know, in the area that we've been running at. Then obviously $75 million of cost synergies. The thing that's really changed in the model though is now when we shift from the EBITDA growth to the EPS number, so we say in excess of $6 by 2025. What's different here is for the first time we have or, I guess we had it back at the time of ATMI, but there's significant leverage in two points between the EBITDA line and the EPS line. The first is obviously de-leveraging the balance sheet and the impact on earnings per share.
We talked about that for every $100 million of term loan we pay down, that's $0.035 of EPS. The other thing is our pro forma tax rate today is approximately 20%. We expect that to move down to 17% by 2025. That'll be a pretty linear, like 19% next year, 18% the following year, and 17% by the time we get to 2025. That's a function of moving the CMC businesses onto our business model and really brings us in line with a rate that's not too far off of our historical norms. Slightly lower than our historical norms. Next slide, please. Just the annual target model. There's nothing new or magic here.
It follows what we've talked about for a very long period of time. It's a really important concept for us because when we plan our business, this target model and achieving this 40% flow-through to the EBITDA line is kinda really the holy grail for us. It has been for a long period of time. It's served us very well historically. If you look over the last five years, we've been right around that 40% number. Maybe just a hair south of it over the last five years. In general, we have achieved that. It's not always linear. I mean, there have been years where the flow-through has been 30%, and there have been years where the flow-through is 45%.
Over time, we would expect to achieve that 40%. We're confident in that because, A, we're gonna achieve some synergies which will help. In terms of maintaining SG&A, as we've talked about in the past or we have a number of initiatives that will, including our shared services operations, that will allow us to continue to be efficient on that SG&A line. Take this target model, flip it back to the prior page, 9% CAGR coming off a $4 billion pro forma base gives you a revenue number of about $5.2 billion.
You know, pick a column here, and that takes you to that greater than between the $5 billion and the $5.5 billion, that takes you to the greater than $6 billion number that I. Or excuse me, $6 EPS number that I talked about. Key assumptions in here is the interest expense in this model. Remember, this is not a forecast, it's a static model. The assumption here is that interest costs have gone from, you know, roughly where they are today on a pro forma basis in that 3.30%-3.40% range to 2.75%, and then it reflects the lower tax rate that I spoke of previously. Next slide, please.
I just wanna unpack and compare sort of what we talked about in the 2021 Analyst Day, where the people are gonna say, "Gosh, we showed a 15% growth number there, we're showing a 9% growth number here. What's different?" Well, really what's different, if you take that 2021 number and roll it out to 2025, our growth rate, that $5.2 billion comes to a 22% growth versus the 15% we showed you last year. Obviously, the biggest difference is that M&A bar, which we said would be two to three points of growth last year in our Analyst Day last year, accounts for about 13 points of growth over the over this four year period. We've moderated the growth from what I'll call our emerging markets.
That's a function of where we were sitting last year. We had much higher hopes for the life sciences business. We were in the thick of COVID, and that business was doing very well. It continues to do well. We continue to be very positive on it, but we've dialed the expectation back to some degree there. The Entegris outperformance, the light blue bar, which was 4% last year, is 4.5% this year, and then obviously tempered the industry growth from that 6%-ish number last year to the 4.5% number I talked about earlier today. We just wanted to unpack that growth for you. With that, we'll move on to the next slide. For me, before I turn it back over to Bertrand, I just wanna wrap up with a couple of points.
First, Bertrand talked a bit about this, but the resilience of the business model. We're about 80% tied to unit-driven products, which we're exposed to the secular growth, obviously, in chip demand, but we think that's a very good place to be. Just to give you an illustration, if you go back and look at 2019, industry units down about 7%, we were down about 3% organically. We've demonstrated that resilience before. We're only about approximately 10% exposed to WFE. The number to be exact is 12% if you look at the donut chart that Bertrand showed earlier. That's our exposure to front-end equipment. We do have other capital exposure. We tend to think the capital side in terms of new fab construction and capital more broadly will be more resilient than WFE overall.
We are talking about increasing our outperformance. We've increased it at a midpoint from 4%-4.5%. Bertrand talked about the revenue synergies, the increasing importance of what we do, and so that outperformance is a number that we feel good about. We talk about the 450 at the midpoint. I just wanna highlight again that historically, if you look at the average outperformance in 2019, 2020 and 2021, it's 440 basis points. That is based on publicly available numbers. It's not, you know, our own internal numbers. It's based on unit growth per lengths and CapEx growth per Gartner.
I think the other thing that is positive about where we sit is we use the term evenly exposed across end markets, and we're a microcosm of the industry overall. I mean, 'cause we're selling across the supply chain, across the industry, and so our exposure, people ask us, "How are you exposed to automotive? How are you exposed to handsets?" Our exposure is consistent with the industry more broadly. The exposure both to the mainstream as well as the leading edge. The model, like I said, has historically been more resilient. Again, on a quarter-to-quarter basis, it's hard to measure the resilience, but in the aggregate, it's that we've demonstrated that time and again. We think about where we are from a model perspective.
We have a model that has a fair amount of flexibility in it. You look at our cost of goods sold, about 50% of that is raw materials. Another, 10% plus is related to, variable labor. The cost structure on the cost of goods sold line is about 60% variable. We've got significant variable costs within the OpEx side of the business. Finally, you know, we've got experience sort of on the deleveraging side. I talked about seeing the movie before when we did ATMI a few years ago, and same thing on the synergy side. I mean, as Bertrand said, we're running the same playbook that we've run historically, and it's been a successful playbook.
We've got a great team driving and running the integration, so quite confident on the achievement of the synergy. The bottom line is, I mean, we're proactively managing the business. You know, we're balancing managing for the near term and investing in the future. With that, I'm gonna pass it back over to Bertrand, who will wrap up, has a couple of finishing slides.
Thank you, Greg. If we could maybe move to. Yes, perfect. Corporate social responsibility. This is a program that has been a great
Focus, and a great source of pride for all of us at Entegris. The results that you can see here and the goals are relating to legacy Entegris only. We will obviously adjust all of them to include the impact of the CMC acquisition, and you should expect the new version of this framework to be published by summer 2023. The overall framework is not gonna change. Our focus on innovation, safety, personal development and inclusion, as well as sustainability, are very much interwoven in our value proposition, our business strategy, and frankly, who we are as an organization. We've made great progress on many fronts since we started the program three years ago.
The updated framework that you will see soon will likely include new goals and very likely also include, in some cases, more ambitious goals for 2030. It will be on the same four pillars of innovation, safety, PD&I, and sustainability. Let's go to the next and final slide as we wrap things up here. I wanna leave you with some final thoughts. Entegris is a resilient unit-driven platform. Our value proposition has never been more critical to the industry technology leaders, and this will allow us to sustain exciting levels of organic growth. I think we talked about that also extensively.
In the next few years, our singular focus will be to realize the full potential of our combination with CMC Materials, and by that I mean reignite growth in the legacy CMC platform, paying down debt, and accelerate EPS growth and value creation for Entegris investors. More generally, let me tell you that as a management team at Entegris, our overarching goal is to create the electronics materials platform of reference. Thank you. This concludes our presentation, and I will turn back the call to you, Bill. I believe you will be moderating the Q&A session, right?
All right, we'll start the Q&A here. First question from Sidney Ho from Deutsche Bank, and again, you keep sending them through the chat feature or email me. Sidney says, you know, the healthy share in CMP slurries, number two in pads, how do you think about the potential share gain in pads for the future?
It certainly will be an area of focus for us. We believe that as we understand better the interactions between slurry pads and pad conditioners, we will be in a position to optimize each of those individual components to great effect for the customer. By that, I expect better finish, better polishing rates, and lower cost of ownership. All of which, I believe, will translate into share gains on the slurry front, on the pads, and the pad conditioner fronts.
Okay, I'll do another one from Sidney. There's been a lot of development in the industry in advanced packaging. Haven't heard Entegris talk a lot about advanced packaging. What's Entegris's position and prospects, advanced packaging?
Yes, it is a fair observation. I think today, if you look at the growth algorithm that we presented to you, there is very little coming from advanced packaging. I think that this will be an area where I could envision those opportunities to continue to expand for us. Advanced packaging is becoming more automated, will require greater precision and greater levels of purity, all of which will be ripe opportunities for Entegris. I think there is one area where we have already been very active, and that's really around you know, wafer carriers for the back end.
I think there's a little bit of that maybe included into the AMH growth algorithm, but that's probably the only area right now that we have included into those numbers that we presented to you. I think that more will be available to us over time.
Okay, next question is from Toshiya from Goldman. To what extent, maybe just a update on how we're seeing the CHIPS Act, and then to what extent is anything that would come of that incorporated in our long-term financial model?
The CHIPS Act obviously is a very important piece of legislation here in the U.S. But there are very similar types of programs that have been enacted in Japan and Korea and Europe. I think it's good at many levels. First, it will drive greater levels of investments and new fab construction projects globally. That will benefit us, even though, again, as we said, you know, the CapEx part of our business is lower today. It's 20%, but it remains 20%. As you know, it's a very important driver for our AMH division in particular. New fab construction projects and that's positive.
The other side of it, obviously, is that as a critical participant to the semiconductor industry, Entegris is eligible to some of those funds, and we are actively assessing what and how this should impact our long-term manufacturing investments. I think that as of right now, we have not really, you know, assumed that we would get subsidies, but it's fair to say that we will be actively engaging with the U.S. administration if we choose to make additional investments in the U.S.
David Silver asked a question. You know, at what point does Entegris become involved when a customer builds out a new facility? Is it at the start-up or the ramp-up, or how does that essentially flow?
Well, all of the above. I mean, we are obviously, you know, we have different solutions that are introduced into those new fab construction projects at different times. The first product lines typically would be our fluid handling systems, which are used in the chemical loops in the sub-fabs. About the same time, we would expect to start also shipping some of the gas purification systems that will equip those advanced fabs. We'd have to wait a little bit longer. At the same time as some of the first tools are getting into the fabs, we would expect the first orders for our wafer carriers, our FOUP products. Of course, we see the full benefit of a new fab when they start running wafers and start using our full suite of chemistries, materials, and advanced filters.
Again, we really benefit from new fab constructions over the lifetime of a fab, from the very early beginning of the construction project all the way to obviously, you know, when the wafers are being produced. Those fabs, as you know, depending on the type of fab, can be used and can be in activity for, you know, five to six years for a memory fab at a particular node, all the way to 10-15 years for some of the logic fabs.
Yeah. A couple from Mike Harrison, Seaport. You know, we touched on it a little bit, but have you seen any cross-selling wins so far after the acquisition of CMC, and what kind of platforms? Anything to any color there?
Yes, we have. That has given us obviously and the team a lot of confidence that many of the hypotheses that we developed as part of this integration plans were real and of great interest for the customer. I can only share maybe one anecdote, but there would be many more. For the sake of time, I would just say that one particular customer, longstanding slurry customer of legacy CMC Materials, was facing some severe defectivity challenges. Shortly after close, we commissioned the slurry and filtration experts to spend several days on the ground, and we were able to redesign the process and introduce a number of different filtration points and different types of filters, which improved significantly the yields for that particular customer. Again, here it's the very beginning, right?
We're taking an existing slurry and existing filter, but we are, you know, sharing our application knowledge with the customer and helping them improve yields. I would expect the customer benefits to become even greater and even more compelling as we start to co-develop and co-optimize slurries and filters, you know. Again, as I was mentioning, extending that to the fab conditioners downstream to the post-CMP cleaning solutions as well. You know, a lot of the hypothesis behind that end-to-end solution strategy are not things that we are hoping for. They are hypothesis that we've been already validating in the field with our customers.
I think we're getting another one from Mike, but I think Mike Harrison. I think we're getting a few similar ones, but what are you hearing from your customers? What are the kind of demand trends you're seeing in the market right now?
Look, we don't really have any particular unique perspective on this. We are very actively comparing notes with all of our customers across the ecosystem. We are all recognizing some softening for sure, which likely will extend into 2023. The depth of the softening, the duration of the softening remain entirely unclear at this point, and I don't think I want to be speculating. Instead, I would like to go back to factual information. The first fact is that as a management team, and you've heard it from me, you've heard it from Greg, as a management team, we know how to manage through uncertainty. We've done that very effectively many times before. The second fact is that our platform and Entegris is highly resilient.
As I said, it doesn't mean that we are immune to a downturn, but it means that relative to most other participants, our platform will prove more stable and more resilient. Again, as a team, we are ready to address whatever industry conditions we may be facing going into 2023.
A question from Aleksey Yefremov from KeyBanc. Any signs of slowing node transitions over the next two years in response to the macro environment?
I mean, we've seen some signs of delays, but they are really mostly, you know, customer specific as opposed to part of a broader trend. No, we have not seen any delays as of right now, at least, that would be directly related to deteriorating, you know, macro environments. If we do see something like that, we'll share that with you in due time. As of right now, no, we have not seen anything like that.
Another one from Aleksey. Do you expect your outperformance in a theoretical to be symmetrical in an up and a down market?
Well, you know, it's a tough one. I'm not really 100% sure. I can lay out the conditions that may actually sway the outperformance closer to the lower end of the range and the conditions that could actually sway that to the higher end of the range. That's why we have a range, right? It's because, you know, we don't control the timing of adoption of the new nodes. We obviously don't control the success of those new nodes, which to a large extent will be a function of the, you know, the macroeconomic environment in which those new products will be launched. Again, that's why we have a range. If you look back in time, you would see that our outperformance over the last four years has been 440 basis points.
At times it was higher than that, at times it was lower than that. As Greg said, that's not something that we measure on a quarterly basis because there's just too much noise in the information. If you look at it on an annual basis over time, you would see that, you know, we've been performing within that range. It's really mostly a function of the timing of node transitions.
Okay. We'll go to Kieran from Mizuho. Can you discuss capital deployment priorities? I know we talked about this in the presentation, but if there's an appetite for M&A, what areas would you look at?
I think we've been very, very clear, both Greg and I, that right now the priority is to deleverage the balance sheet, so pay down the debt. As you could tell from the projections presented by Greg, it's gonna take us a couple of years to get there. You know, don't expect us to make any large scale acquisition or mid-size acquisitions. There may be some technology tuck-ins, but that's really not a priority for us at this point. The priority is to unlock the full potential of CMC Materials. It's exciting. It's a very exciting time.
It's a very busy time for all of us in terms of system migrations, in terms of putting all of the right commercial strategies in place to unlock the top line growth, and then more importantly, really to start marshaling resources to develop the next generations of deposition materials with the matching polishing solutions and the optimized for CMP cleaning solutions on the back end of all of that. Again, very exciting times. We have plenty to keep us busy, and M&A will not be a priority for the next two years at least.
Next question from Amanda Scarnati from Citi. What's the biggest swing factor in your driving your growth above market? More confidence in SAM expansion, market share, anything from a product point of view that you would point to?
I think it depends, you know, each division would have probably or would trigger a different answer. I would say that for APS, it would mostly be market share gains. There would be a little bit of SAM expansion in the wide bandgap substrates, so the you know SiC, gallium nitride substrates polishing, but really mostly market share gain. I would say that for the other divisions, it's mostly SAM expansion. A little bit of market share gain over time.
A few questions from Paretosh. I'll try to read this verbatim. "I was hoping to understand what sort of incentives your customers have to look for cheaper solutions from your competitors. And then what percentage of total manufacturing costs for semiconductors typically represented by Entegris?" It's kind of a combo question.
If you think about the cost of our solutions on a per wafer basis, it still doesn't really add up to very large dollar amounts. We see steady trends, so what we do is very important, but what we sell are very small quantities of very critical materials, or again, very critical enabling chemistries and filters. I think the incentive for our customers to go for lower cost solutions, I mean, it's really at the end of the day, it's up to us to make our offerings so compelling and to provide our customers such a, you know, unique process advantage when they use our slurries, our chemistries, our filters that they just actually wanna buy from us.
I think, you know, we are really spending a lot of time educating our customers to looking at really the full cost of ownership and looking at the yield benefits as opposed to the individual cost of a product. I think that as we build this very complete and very compelling application knowledge, and as we use that to develop highly differentiated solutions that are all becoming increasingly mission critical and are becoming increasingly industry standards, I would expect our customers to for sure continue to have, you know, local lower cost competitors that they will keep alive to keep us honest.
At the end of the day, we want to believe that our value proposition will be so compelling that they will wanna, you know, they will want to bet on Entegris, because they view Entegris as the surest path to, you know, best in class technology and faster time to yield. That's really at the heart of our value proposition. For any one of you understanding the semiconductor industry, you understand that this is really at the heart of what our customers are trying to accomplish.
A follow-up from, Paretosh Misra. "What are your thoughts on the next CapEx investments post the Taiwan facility?
Do you want me to take that, Bill, or do you want Greg to? I don't know. It's fine either way.
Yeah. Oh, you go ahead. Yep.
Well, again, I think I just said it's really continuing to invest in our growth. That means invest in capacity, invest in manufacturing capabilities as well, because our customers continue to expect greater process capabilities from Entegris. Of course, continue to invest into our tech centers. I mean, those tech centers have proven to be essential in our ability to engage effectively with customers in Korea, in Taiwan, and elsewhere in the world. As we are invited to collaborate on more opportunities with our customers, we wanna be sure that we have adequate capabilities in terms of local tech centers as well as local manufacturing capabilities.
Okay, this is, I'm not sure about the firm, but Mr. Sapra asks, "What is Entegris' exposure to silicon carbide market? What sort of solutions do you have, and what's your sort of outlook?
I think if the question is really around silicon carbide wafers for power electronics, I would say that, you know, we are engaged with those customers at different levels. I mean, obviously, when new fabs are being built, it provides a great opportunity for our gas purification systems, which are very important for those operations. And then of course, we are very actively involved in supplying some of the slurries for wafer polishing and some of the pads that are used there as well. This is a market that is still fairly small today, but it's growing very, very fast. That was what led us to the acquisition of Sinmat and the spin-off from Viasep. We have been able to make great progress, great market share gains on those applications.
We believe that the CMC Materials pads have also some very unique attributes that make them ideal for those types of applications. I think we have a hopefully you know a lot of great stories to share with you as the industry continues to grow and those applications become really mainstream.
I've got an unattributed question. We've touched on this a little bit. What are the key medium-term technology inflections in logic and memory, and why is Entegris so well-positioned?
Well, I think we discussed that at length for the legacy Entegris platform during the previous analyst day. You know, it's really about enabling materials that customers will be able to deposit in thinner films in 3D NAND architectures. By doing so, you know, help them find solutions around the highest aspect ratio challenges. It's about really developing new selective etching solutions that can really very effectively and uniformly etch those very thin and tall structures in 3D NAND. You know, if you think about gate-all-around in logic, obviously this is a materials intensive process. That will open up a lot of opportunities from a material standpoint from a selective etching chemistry standpoint.
Of course, all of those architectures are going one way, and that's really, you know, miniaturization, so it's smaller and smaller dimensions. Another challenge will open up the doors to, you know, to many opportunities for our microcontamination platform. When it comes to the APS division, I think that there are equally exciting opportunities, for, you know, gate-all-around, at two nanometer, if you consider that, and you add to that the power rail architecture that our customers are also contemplating, that should add about 20%-30% more CMP polishing steps, many of which will be tungsten. It's very good and very exciting for our APS division.
If you turn to the, you know, the 3D NAND architectures, you know, new strings of memory will have to be stitched up to get to 200+ layers, 300+ layers, 400+ layers. Every customers have a slightly different architecture, but we expect that every time you stitch up, you know, new strings of memory, we could see an increase of about 20%-ish. Again, it's different customer by customer, but an increase in polishing steps as well. Many reasons for us to also, you know, be very excited about the technology inflections for APS in both memory and logic.
A question on what is, in broad terms, Entegris' exposure leading versus mainstream?
You know, this is a very simple question and yet a tough one to answer. I would say that, you know, the ballpark number I would give you is probably between 40% to 50%, 55%. The way I get there is we have about 35%-40% of our revenue that are coming from products that have been introduced in the last 5 years. Clearly those products are used in leading edge. We also know that leading edge fabs continue to use some older generation of products as well. I would say that 55% is probably a good number, but it's an approximation.
The final question, Todd Edlund. This is a good one. How do you see and how do you manage your innovation PIM? Do new products typically have a higher margin? This will be the final question.
Typically, the new products have higher margins, and the higher margin is really a function of making sure we select hard-to-solve problems that our customers are facing. As we come up with the right solution in a timely fashion, typically our customers are willing to reward us with improved margins. That's, you know, that's typically how we manage the opportunity PIM, is we do not allow ourself to focus on easy-to-solve problems because it means that everybody can do it, and that means that there's very little value at the end for the customer and for ourselves. One of the major criteria for us is really to find the adequacy between a hard problem to solve and some very unique capabilities that we have at Entegris.
That's one of the reasons we are so excited about this combination with CMC Materials. We are really adding to the Entegris platform, the leader in CMP solutions. By broadening our capabilities, I think it will translate to many more opportunities to contribute unique value to the roadmap of our customers. As we have discussed, we expect that to translate into growth outperformance over time.
Thank you, Bertrand, and thank you everyone for joining today. I would be happy to answer any questions that you have following the event or ones that we didn't get to. Please reach out to me if you have any questions. Thank you and have a good day.