Applied Materials at Deutsche Bank Securities Inc. 2008 Alternative Energy Conference - Final

UNIDENTIFIED PARTICIPANT: We have the pleasure today of having Mark Pinto, who is the Chief Technology Officer for Applied Materials, who also has responsibility for the Energy and Environmental Services Group, and he is here to talk about that today. So without further ado, I'm just going to have Mark take the stage here, and I will go away.

MARK PINTO, CHIEF TECHNOLOGY OFFICER, APPLIED MATERIALS, INC.: Thanks very much. It is a pleasure to be here. I'm going to talk about our solar business at Applied Materials and an update. And really the theme here is going to be delivering lowest cost PV energy. So I will talk a little bit about the production of panels, but also a little bit more about balance of system and installation costs to the true cost of delivering PV electricity.

The obligatory Safe Harbor statement. And the forward-looking statements are based on estimates and projections and assumptions as of today, and we undertake no obligation to update these.

As an agenda, first I'm going to talk about an update on the SunFab line, our thin film production line based on 5.7 square meter panels. Our first customer panels have come out, to give you an update on our customer profile, an update on the technology, and an update on the delivered per watt costs. I am going to give our view of the PV electricity cost model, and why the thin film technology really is having such strong adoption. A short update on our crystalline silicon business, and then a summary.

So first we are in, as we discussed over the past few months, in a process of bringing up our first SunFab factories around the world. As of Monday of this week we have three customers who have produced their first panels. Two of them showed them at Intersolar a couple of weeks ago. So we are right on schedule.

In February we had a first announcement of a customer's customer who signed an agreement -- supply agreement with one of our customers, Signet Solar, and specifically mentioned the advantages of 5.7 square meter panels in the installation and total cost.

As well, at around Intersolar we had another customer's customer, this time with Moser Baer, announced a significant supply contract as well. So these are now happening at a regular rate.

On our announced contracts, we don't announce all the contracts. These are really up to our customers to announce. This is the current list of announced contracts. The sum is now over $3 billion in terms of contract backlog, and over 1.5 gigawatts of production volume. The most recent ones, BES Solar, one that we did a disclosure on earlier in the year; a gigawatt project in China; Sunfilm, which is significant because it is a first repeat order for tandem junction; Masdar, which I will talk about briefly in a second; and most recently, Moncada. Moncada is interesting because it is in Italy, another key region for PV, but also Moncada it is a major construction company in wind, and intends to build hybrid systems of wind and PV, as well as PV.

A month ago, approximately, we announced and our customer Masdar announced a project in Abu Dhabi. Actually the first line -- there is three lines, SunFab lines, one in Germany and two in Abu Dhabi in the phase one that will start in the 135 megawatts. These lines can be upgraded to 210 megawatts. And then it is part of longer-term project they have to go to gigawatt scale.

Significant in a number of respects, one, the size, two, where these panels are going to go in the Middle East and around this vision of a zero carbon, zero waste integrated city. And then also I think should not be overlooked is this is really the first high-tech nanomanufacturing in the Middle East. I think that is a fairly significant event. It has been talked about for integrated circuits for some time and happening in PV.

On the technology update, the first panel that was measured by a customer, Signet, the first 5.7 square meter panel -- module, you see a picture on the left there, May 23, 7.75% initial efficiency. This is initial module efficiency. Depending on how the degradation factors get measured and optimized, we believe 7.75 efficiency range is between 5.8 and 6.5. We have committed to customers 6% plus stabilized efficiency. So this as well on track. I am very pleased to see that that is the kind of result we get right out of the box.

On the tandem junction we are not yet announcing first modules from a customer, however, we have made significant progress on the R&D, in particular on the 5.7 square meter size. We have achieved now over 10% average cell efficiency -- initial cell efficiency on the full-size panels.

One of the key technology elements to managed here is the microcrystalline uniformity as part of the tandem junction. And we have now achieved results of less than 5% relative variation across the whole 5.7 square meter substrate. You can see in the upper right the measurement across the diagonal of the crystalline fraction, so you can see that it is very, very uniform. Just a reminder that we have mentioned in the past that in a lab we have achieved tandem junction module efficiency -- stable efficiencies as high as 9.3%.

I should also mention silicon has a very strong roadmap. There are research results in the literature that go up to over 13% in stable module efficiencies in an R&D environment. Our target is to get in at 8% stable tandem junction efficiency with these initial fabs at a good cost point and then drive that upward in terms of efficiency and lowering cost.

Also, we have achieved the required film morphology on zinc oxide-based TCO in our R&D tool. We have an R&D tool in Alzenau in Germany, and we also have one at a customer site. It is our intention to offer TCO capability either in-line or to glass suppliers who can use the same TCO capability and supply a customer. We think the business model will develop on its own. Customers may decide to do the glass coating or they may choose to partner with a glass vendor. Especially relevant when you get to gigawatt size, where you can put a float line right onto the gigawatt fab and do the coating either on the float line side or on the fab side. So we have very, very high confidence on our initial sign-offs and our technology roadmap, the tandem junction technology.

I want it talk about really the drivers of the business. What are the fundamental pillars of making this technology work, this business model work? It is really about two things to start. It is not about efficiency or size. Those are things that help drive these. It is really about lowering the production cost and lowering the balance of systems cost. You add the two together, that is the total delivered cost.

Our objective, we have said before, to deliver less than $1 per watt in 2010 in production cost. That is a function of materials utilization, the process, the equipment cost, the size of the module, like in display as we have demonstrated with big HDTVs drives down the cost, the efficiency, and the factory scale.

On the other side, our objective is to also get the balance of systems cost below $1 per watt by 2010. That is a function primarily of installation scale, what kind of techniques you can use, -- I will show you shortly -- the location, the module size, which as we have discussed in the past has -- can drive down the cost significantly, and also efficiency where our target is greater than 10% -- greater than or equal to 10% in 2010.

But to really drive the industry you have to have those two pieces and a supply chain and materials scalability, and so that means abundant materials, low cost, safe and reliable, with the ability to supply gigawatts per year and gigawatt per project capability in 2010. So we have a good head start on this because we can build off of our display supply chain. It is especially relevant for big projects in Asia, because that is where most of the display equipment goes.

So again, with these three pieces, production cost, balance of systems cost, and the scalability to deliver, our objective is and our belief is we're going to lower -- that we're going to deliver the lowest overall cost per watt on the thin film for large-scale installations.

Where are these drivers on the production cost? Some of it comes from the watts per module, the efficiency. That comes from further optimization of the tandem junction and delivering this textured advanced TCO. Materials cost reduction, utilization of gas, thinner device layers. Actually the TCO and thinner layers go together. As you texture TCO, you can refract the light. The light can then go in a parallel direction to the surface of the glass, and so you can use thinner layers. That is the kind of physics and design we have to implement.

Scaling, we get cost effects from size of the facility, better equipment utilization, volume purchasing and operations, the labor and other facilities, and then higher throughput, yield and uptime of the line. With all those things we are on a roadmap to go from $1.48 per tandem junction single line to $0.85 tandem junction gigawatt scale by 2010.

Just to mention the single junction lines that we're talking about are around $1.20 a watt. They have lower upfront costs and higher balance of systems costs today. Most of what we see going forward will be the tandem junction, as the tandem junction cost comes down.

A very key advantage that we have discussed in the past year on the size is also not just the production cost advantage that you get, but also the installation cost. I'm going to show you an example. This case compares two implementations, assuming a fixed and constant common efficiency of 9%. That if you use conventional size modules versus SunFab modules, you use eight times less modules, more than -- less than half the brackets, eight times less cabling connections, more efficient support rails. And you can see the cost savings that you get. Again, this is for same efficiency.

If you compare then the total net cost that comes out of all of this, when you factor in land, everything else, you see a net of greater than 17% reduction by going to these big modules. And that is equivalent to 2 to 3 percentage points in efficiency. If you have a module -- a small module or a large module that is 10% efficient, it is the equivalent of 12 to 13% smaller module because of the balance of systems cost that you get from going to the larger module.

Now I'm going to show you -- we get questions about the effectiveness -- is there really something behind this balance of systems cost? What I'm going to show you now is how this can be implemented. This is the video we did with an installer. (video being played).

Again, when you are working on a farm, a relatively large size installation, this is isn't for your rooftop that you're going to go out and try to carry these things up the ladder. When you have got a big enough installation, you can use this kind of equipment. It is not custom-made. It is widely available. And you can see it takes about a minute and a half and four or less people -- we believe you can get down due to two to that job. Which again gives us -- the modeling shows that we can get to these costs of less than $1 a watt installed.

So how do we translate that then into the business opportunity? Why is that significant? You've got to then convert this to delivered PV electricity cost. We have developed models around this. The models include a number of effects. It includes a factor for the amount of installation or radiation that you get, including the effects of things like diffuse effects. Thin film modules -- thin film silicon modules absorb -- actually absorb more light per day because they can make use of more of the diffused light at the beginning and the end of the day.

It includes effects of module lifetime, includes effects of degradation, includes the cost of capital and the term -- finance term. Overall system efficiency, typically something like 80% for various losses in the system. It includes the cost of the inverters, as well as the lifetime replacement of the inverters. Land, the scale, operating and transmission cost. And you can also include various incentives, although I'm not going to include them in the analysis I'm going to show you.

The specific analysis here is going to be for an average location in California, around 1,800 peak sun hours, and two scenarios of installations -- one crystalline silicon for midsize installations, and then a second based on industry analysts cost estimates. And then a second based on our cost estimates for the large thin film panels, 5.7 square meter panels with the installation techniques you have just seen.

So for California, I'm showing two bounds of the electricity rates starting in 2007. The peak rate, which is very relevant to solar because the peak rate, which is in effect right now, is when there is maximum sunlight, and that is when solar is most effective, unlike wind, where wind and the peak rate do not correspond typically. And then a base rate, which is the new generation rate in California based on gas turbine generation, to compare against the minimum base load or base energy rate.

This is going to be done in price, so we have included margin for all the people in the food chain here. So this is not cost. This is going to be an analysis based on an estimate of prices.

So the first datapoint will show is where we are with crystalline silicon. So this includes no incentives, no ITC, no feed-in tariff. It includes a cost of capital -- real cost of capital of 4%, something in real terms with a homeowner, where he is generating it on his house and not making a profit to sell it to someone else. What you see there is a crossover of the peak rate next year, and then going down into the middle of this band of utility rates over the next few years.

On the other side, the thin film using the same assumption, a 4% cost of capital, no incentives, assuming it is a big installations so you can use these mechanized techniques and large panels, you can see that there is a substantial improvement in the cost -- generated cost of electricity, in fact, well below the average utility rate by 2010 in California, going to basically the same as the base rate in 2014.

Now realistically, however, unless you are a business that you're going to build this solar installation yourself, typically you're going to do a PPA or do some other kind of power supplier, 4% is not the right cost of capital then for a project where you're going to get debt financing and equity. So if we go somewhere in between, say an 8% cost of capital, you can see it shifts the curve approximately by two years. So that is then the payback. But you're still completely in the zone between grid peak rate, even today in 2008, between grid peak rate and new generation rate. Again, going towards the new generation rate within five to six years.

So this is really what is driving the interest in this technology. Again, this is no feed-in tariffs, no investment tax credit, just raw numbers of where we are versus grid electricity with this kind of installation cost and production cost of the modules. So as a result, we believe that SunFab will deliver the lowest overall electricity cost for these large-scale installations, lower than any other thin film technology.

A brief word on our crystalline silicon business. Crystalline silicon, if you look, the segment that it serves the best is existing residential rooftops -- thin film, large-scale installations. There is a gray area in between where, depending on various prices and efficiencies, they may compete. But they're very strong in their own segments, residents for crystalline and large-scale farms for thin film.

In crystalline our focus is slightly different. We are focused on differentiated tools for value-added steps in cell production and wafer production. The key for crystalline silicon is to develop the capability to handle -- produce and handle thin wafers, high efficiency at large scale and low production cost. You want all of those things together. And that is where the roadmap needs to go to make crystalline silicon continue down that cost path I have just showed you.

We currently addressed 42% of the ingot through module CapEx in crystalline silicon. Somewhere around $0.50, $0.60 a watt in terms of spend. We have achieved substantial progress on increasing the production capacity. You may recall, we have done two acquisitions in the past 12 months in this space. The one floor that you see on the left is PWS, the wafering business we acquired in Switzerland, the former HCT Company. And on the right is Baccini Cell Systems, BCS. They were acquired at the beginning of February -- and closed at the beginning of February. Market leading products, good technology, and a key here, good margins.

The key has been to increase their production capacity. With precision wafering we have increased their capacity 90% since acquisition. And with Baccini, since the acquisition we have increased the monthly shipments over 100%. And we will be recognizing the first revenue for Baccini in the current fiscal quarter.

I would just note on the right picture you see of Baccini, that is one of their four production floors. When I was there in January there are only two lines of equipment running along vertical axes. You can see that floor is crammed with equipment today and meeting the demand -- increasing demand of customers.

We mentioned on our last conference call that our current crystalline silicon product shipment runrate is around $200 million per quarter. I would emphasize that this is an existing business. This is not one where we're waiting for fabs to sign-off. This has the cadences much more like our other businesses, and with products that have been out in the field, so you should see some of those results flowing through in the coming quarters.

So with that, let me just conclude. On the thin film side, again, we are well on track with our projects. Three customers with initial panels, meeting the specs, even on the first panels, good contract momentum, now over 1.5 gigawatts of announced capacity, over $3 billion of backlog. We believe we're well on our way to be the number one thin film technology in 2010, with over 1 gigawatt in production.

And we have two customers now that have reordered lines, and as I said, three customers who have panels coming out. And on the crystalline side, a growing business. The crystalline market is growing at over 30% per year, and we're running at a $200 million runrate on shipments. So I think [you will] see strong performance over the next quarters from that business.

With that, thank you very much and happy to take questions.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: So two concentrating technologies, concentrating photovoltaics and concentrating thermal. Typically this -- there is some overlap potentially. Personally, I think this, even in the areas of overlap, this is going to be stronger. But that aside, the concentrating technologies really only work well in direct sunlight. Typically on the large side they have to be very, very large projects. It doesn't have the ability to be distributed throughout the grid. It doesn't have the ability to be distributed into areas that have less direct sunlight. So I think this is a much more versatile technology.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: Two questions. One question is how the customers will evolve. Will they evolve towards maybe some of the traditional semiconductor manufacturers as this thing gets into bigger scale? And the second question about flexible.

So the first question, I think, yes, the customer landscape will change, but I'm not sure I can predict how. At this point I would not pick the semiconductor makers as the leaders in this necessarily. It may be possible that some of them do. The market is quite different. The technology is quite different. It is about -- you don't get any -- much added value at all about the design; it is all about production cost.

Also, on the thin film side you can make the argument that it is better -- although large is good for factories, distributed is better, because it is more like the glass industry where you're going to have -- this is going to turn into construction material.

I also think that there is value in people participating in different parts of the value chain, and maybe even on a small scale. You look at company like Moncada that is going to combine it with wind, they have some inherent advantage. Some people have regional advantages. There definitely an advantage to going big though. So in some markets for sure, I think the winner will be somebody who does big, but it may not be somebody necessarily that comes from a semiconductor background.

Crystalline could be a little bit different. But crystalline already, I think there are some significant players there who have gained scale. I look at Q-Cell as an example -- Suntech. So again it is -- without a breakthrough, I'm not sure that it is that easy for anyone else -- or I guess maybe acquisition -- to step into this.

On the Flex side, I would -- we do have equipment that -- we have a Web business. It is part of the EES Group. The current applications are mostly around consumer and electronic packaging, everything literally from potato chip bags to touch screens. There is an opportunity, and we are certainly looking at flexible solar. I would say it is a market though that is still immature. There is one significant supplier of flexible solar modules today.

I am not clear how big, especially if the crystalline silicon prices come down with more poly, how much market is uniquely served right now by that technology. So it is something we're looking at. Clearly if you could raise the efficiency and lower the cost I think from where it is today, it would open up more of a market. But as a -- where it is at, it is may be a niche. How big that niche is I think time will tell.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: The question is, are any of the customers in the U.S. currently in our list. The answer is no, but we're ready to do that. We are certainly looking for opportunities to engage. There are places in the U.S. I think, again, a lot of the comments at the lunch break about the political situation, also the smart metering and smart grid will have a big impact.

You look at a place like Hawaii, tremendous opportunity there with where they need to go. Obviously California, another environment where it makes a lot of sense. So we think it is going to happen. Very hard to predict right now. I would say that is true across the whole solar spectrum.

From the point of view of certification, we have done some certification of small modules in the U.S. We have not done the large ones yet, but they could be submitted at any point.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: The answer is yes. I think as you go further out in time the tandem becomes more cost effective. Today it is almost balanced. Tandem cost more to produce, but is lower in balance of system cost. For single junction it costs less to produce, but you have to use more panels. So it kind of balances out to first order.

Over time the tandem junction will be more cost effective. You'll get more watts out of the factory for the same production costs or less. And then on the other end, you will continue to lower the balanced system cost as you increase the efficiency. So there is more ways to increase the efficiency of tandem junction. Single junction is very limited in that respect. So bottom line is, yes, we believe in 2009 there should be a crossover. And that is what we are seeing as most of the repeat orders or additional orders going from 2009 forward will be tandem junction.

In terms of the capital cost, no change there. The only difference since we began this product is that, obviously as you to a bigger factory, one of the ways you're saving is like an IC factory in better matching of equipment throughput to the scale of the fab. So you better optimize the equipment against that. So when you look at the CapEx ratio for a large fab, it is more favorable in terms of dollars per CapEx. I can basically use the number we did in the disclosure, that we had around about a gigawatt scale fab that we disclosed was a $1.9 billion U.S. contract. And it goes up from there as you go to smaller and smaller lines. It will go over $2 and up to the mid $2 range.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: No, not at all. So thin film modules in general, they have been out there. There is for instance a site in Sacramento that has been operating for 25 years. Thin film photovoltaics were invented in 1976 by Dave Carlson at RCA Labs and Solaicx after that -- I mean the former BP Solar.

So the technology, the actual packaging technology that we are using is based on very stringent automotive standards. And we do test in excess of what the certification agencies do. We have our own lab, which we're getting certified in China, that does all sorts of tests, from wind shear factor to humidity, damp humidity to, you name it. So, no, I don't see that at all. It is very, very, very stable, reliable technology.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

MARK PINTO: I could follow up and give you the exact numbers, but we exceed all of the IEC standards in terms of, as I said, snow loading. And obviously one of the issues I have heard is, I am not sure that we can withstand, or any class panel can withstand a high school baseball player's line drive. So that has been one of the markets that has been good for flexible solar because it tends to be more robust.

But in terms of typical weather or extreme weather conditions, all of those things we have huge sand loading and different kind of tests for that. So no issues. It is due to the rail bonding. I think that is -- it is a very elegant, cost effective solution. It doesn't require a frame around the outside, which can collect water or other kinds of materials that could accelerate the degradation. So we are very -- very strongly believe this is a reliable solution.

UNIDENTIFIED AUDIENCE MEMBER: (Inaudible question - microphone inaccessible).

[Thomson Financial reserves the right to make changes to documents, content, or other information on this web site without obligation to notify any person of such changes.

In the conference calls upon which Event Transcripts are based, companies may make projections or other forward-looking statements regarding a variety of items. Such forward-looking statements are based upon current expectations and involve risks and uncertainties. Actual results may differ materially from those stated in any forward-looking statement based on a number of important factors and risks, which are more specifically identified in the companies' most recent SEC filings. Although the companies may indicate and believe that the assumptions underlying the forward-looking statements are reasonable, any of the assumptions could prove inaccurate or incorrect and, therefore, there can be no assurance that the results contemplated in the forward-looking statements will be realized.

THE INFORMATION CONTAINED IN EVENT TRANSCRIPTS IS A TEXTUAL REPRESENTATION OF THE APPLICABLE COMPANY'S CONFERENCE CALL AND WHILE EFFORTS ARE MADE TO PROVIDE AN ACCURATE TRANSCRIPTION, THERE MAY BE MATERIAL ERRORS, OMISSIONS, OR INACCURACIES IN THE REPORTING OF THE SUBSTANCE OF THE CONFERENCE CALLS. IN NO WAY DOES THOMSON FINANCIAL OR THE APPLICABLE COMPANY OR THE APPLICABLE COMPANY ASSUME ANY RESPONSIBILITY FOR ANY INVESTMENT OR OTHER DECISIONS MADE BASED UPON THE INFORMATION PROVIDED ON THIS WEB SITE OR IN ANY EVENT TRANSCRIPT. USERS ARE ADVISED TO REVIEW THE APPLICABLE COMPANY'S CONFERENCE CALL ITSELF AND THE APPLICABLE COMPANY'S SEC FILINGS BEFORE MAKING ANY INVESTMENT OR OTHER DECISIONS.]