Online information source for semiconductor professionals

When the term "traceable calibration standards" gets mentioned, it's not likely to trigger a giddy rush of excitement among most professionals in the semiconductor, flat-panel display, and related micro/nano industries. But without such measurement-accuracy-enhancing tools, process and metrology engineers would be comparing apples to oranges far too frequently, with no way to be absolutely certain that one piece of equipment's critical dimension readout matches another's, or that the film thickness and resistivity numbers they see on an instrument are consistent with those of every other similar system.

The photovoltaics industry has its own calibration standards needs, especially when it comes to measuring those all-important solar-cell and module conversion efficiencies. By properly and consistently monitoring and calibrating the solar simulators used to determine those efficiencies, the quality and accuracy of the measurements can be drastically improved, which has enhanced revenue and performance-reliability ramifications all the way up the PV system validation value chain.

VLSI Standards, a company that's synonymous in the chip-, panel-, and hard disc-making sectors with the tools and services required for the heady world of NIST-traceable measurements, announced last week that it's entered the PV manufacturing fray, opening the doors of an accredited solar calibration lab at its San Jose HQ as well as offering a couple of versions of an NREL-traceable solar reference cell product.

VLSI Standards' Tortonese seeks solar-cell measurement accuracy.

I've known VLSI's Marc Helvey for years, as much for his dogged determination in marketing and selling his company's products and services around the globe as for his lifelong loyalty to the San Francisco Giants and his glee in giving me grief about my baseball team and archrivals of the Giants, the Los Angeles Dodgers. After I saw the press release announcement, I emailed Marc and his boss, Marco Tortonese, president of the company, with a list of follow-up questions about VLSI's new solar side.

Here are most of my Q's and Marco's A's:

How long have you been working on the solar products and service, and what are the reasons for launching now?

About two years ago, we realized that we were selling our existing semiconductor products into the photovoltaics industry. I am referring to our traditional step-height standards, which are used to calibrate the profilometers that control the texturization of solar cells; film thickness standards, used to calibrate the ellipsometers that control antireflective coatings; and resistivity standards, used to calibrate the machines that measure doping.

We made a conscious decision to focus our efforts on solar about a year ago when we made a capital investment with the addition of a photovoltaic calibration lab. Our first product introduction was a certification service for Solar Reference Cells, which was released in conjunction with our own reference cell in order to provide a turnkey solution for manufacturers who do not want to build their own. A reference cell is used to calibrate the solar simulators that measure the efficiency of solar cells or panels at the end of a production line.

This product makes a lot of sense for us. VLSI Standards' mission is to make a positive impact in the world by improving the quality of industrial production through accuracy in measurement. We found that we can have a big impact on the solar industry by helping our customers measure the efficiency of solar cells with accuracy. Efficiency is the single most important measurement in the photovoltaic energy market, as it ties directly to the economic value of the product throughout its supply chain, from cell to panel to final installation. Therefore, it is critical to measure cell efficiency accurately and early on in the manufacturing process, in the solar-cell manufacturing plant, before errors propagate all the way through a completed installation.

When did you actually begin offering the products and the service, and what has been the response so far? Which do you expect to be more popular, the service or the cells?

We officially entered the market on July 15, the first day of Semicon West. The market has responded enthusiastically. The industry had been relying on national laboratories or research institutes to provide these calibration services, and was really waiting for a commercially viable and accredited laboratory to step up to the plate and provide this calibration service on a reliable commercial basis.

We are engaged with some of the major manufacturers of solar simulators. They are interested both in our Solar Reference Cell and in our calibration services. We expect that some of our customers will build their own reference cells and will [want] to have us certify them. Others will prefer a turnkey solution, where we provide both the cell and the certification. We have chosen to offer the reference cell and the certification service as two separate products, providing customers the maximum flexibility in managing their supply chain.

What are some of the main differences you've learned about, between the calibration and certification needs in the solar and semiconductor industries as well as the infrastructure in general?

The use case for calibration is remarkably similar between solar and semi, and probably any other manufacturing industry for that matter. If you measure a critical parameter in a production process, that metrology tool must be calibrated to give accurate readings, whether it is the critical dimension of a 25-nm transistor gate or the efficiency of a solar cell. Even more so if the measurement is directly related to a final product specification, and the purchase price of the product is tied to that specification, as is the case for efficiency measurements of solar cells and panels. From a business model perspective, we have observed that solar production facilities tend to purchase turnkey factories more than their semi counterparts. That makes it more important for us to partner with OEMs and system integrators.

What are the most important requirements for a certification product in solar? Can you explain how the certification/calibration process works?

Calibrating a solar reference cell is not an easy task. It requires competencies in a variety of metrological areas, from radiometry to spectral photometry, electrical measurements, thermometry, and dimensional measurements. All of those parameters have an effect on the final certified values. Each of the measurements we perform must be traceable to the International System of Units through an unbroken chain of measurement comparisons to fundamental standards, and measurement errors in each of these comparisons must be propagated rigorously. In particular, the fundamental radiometric standard for the photovoltaic industry is maintained by an international body that performs extensive direct black body radiation measurements from the sun in Davos, Switzerland. We were able to tie to those standards through the National Renewable Energy Laboratory in Colorado.

In practice, we have created a solar simulator that is traceable to those same measurements in Davos. For us to calibrate a reference cell we must first perform all the checks and balances on our solar simulator to ensure that it matches the standard test condition for spectral emission and intensity, then we measure the output of the cell in a traceable way under those controlled conditions.

Do you manufacture the reference cells or outsource that?

We have a dual-manufacturing strategy. We can manufacture our own, but we also outsource.

What special manufacturing requirements are involved?

The most typical special manufacturing requirement that we have encountered for reference cells is that the cell should be a fairly close match to the cells being measured in the manufacturing plant. For example, you cannot use a single-crystal reference cell to calibrate a solar simulator that is then used to measure efficiency of amorphous silicon cells. This is because single crystal and amorphous silicon have different spectral responses. Therefore, we have to simulate the material properties, for instance, using appropriate color filters to obtain a good spectral match to our customers' products.

Alternatively, we can use our customer's own special materials and create reference cells for them. This will allow us to provide calibrated reference cells for all kinds of conventional and new exotic materials, from crystalline to thin films, from organic to multijunction devices, provided we use the correct approach.

In a typical cell or module factory, how would the manufacturer use the reference cells? How many would be needed for a typical volume line?

A manufacturer would put the reference cell under a solar simulator and adjust the intensity of the light until the current output of the reference cell matches the value that we certify. At that point the manufacturer is assured that the solar simulator reproduces the irradiance of the standard test conditions, which are defined as 1000 watts per square meter, or one sun in common solar terminology. Typically this process is automated by the manufacturer of the solar simulator and frequently the reference cell is integrated with the simulator. In this case, there needs to be a reference cell for every solar simulator in the plant.

What kind of impact will the use of these cells have on conversion efficiencies, traceability, and other solar metrics?

The benefits to the industry and to the consumers from using accurate and consistent efficiency measurements should be pervasive. For example, within a factory, it is important to match the efficiency measurements of all the simulators used by the different production lines. A common ruler for efficiency will be important in addressing disputes between buyers and vendors of photovoltaic products. We have purchased solar cells that did not meet their efficiency specifications, while on the other hand I have friends who have been pleasantly surprised with their rooftop installations supplying more electricity than they had contracted for. Overall I believe everyone will benefit from the use of a common ruler, and that is what we are here to provide.

What role did NREL play in development of the products and service?

We have worked very closely with both NREL and NIST [National Institute of Standards and Technologies] in our product development. NREL has provided a traceability path to the international radiometric standards. NIST has provided us with an accreditation to the ISO/IEC 17025 standard, which is pretty much a requirement for any calibration laboratory to play the game. To our knowledge, we are the only U.S. commercial laboratory to have achieved this accreditation for photovoltaic devices.