By Andreas Riedl
All materials and products used outdoors are affected by weather: clothing, electrical cables, car paint, or a Photovoltaic (PV) module. Solar radiation (ultraviolet, visible, and infrared), moisture (rain, dew, and air humidity) and heat, the so-called primary weather factors act together with oxygen and cause the material to age. Of course, there are many secondary factors like soot from wildfires, biological agents, or salt in the coastal air that contribute to material degradation, in addition to the primary factors (see Figure 1). Weather ageing or ‘weathering’ is irreversible and usually changes the material’s or product’s properties in an undesired way. Weather-related ageing starts on the microscopic level, and is later detected as changes in appearance (e.g., visible corrosion, delamination, yellowing) or function (electrical performance, strength properties). Finally, the material or product fails, as it is no longer able to meet its intended purpose, however defined.
Product failure due to outdoor exposure is only a matter of time. It can’t be avoided but it can be delayed. The first step is to find out how exactly the material or product of interest reacts to the relevant weather factors. There is only one way to obtain this information: ‘Weathering testing’. Today, a variety of technologies and methods is available to carry out laboratory and outdoor weathering testing, as well as to measure property changes, interpret and correlate the weathering data, model the ageing behavior, and even predict lifetime.
Looking back, weathering technology actually started with fading of actor’s costumes caused by aggressive stage lighting about 95 years ago. After 20 years of laboratory light fastness testing of textiles with artificial light sources the first weathering instrument, the Atlas Weather-Ometer, was introduced in 1934. Since then, numerous testing standards, programs and strategies have been developed for weathering testing of paints, plastics, building materials, textiles, pharmaceutical and cosmetic products, automotive, military, aircraft, household, electric materials, components and others. The Atlas internal database lists more than 1,200 of such consensus-based standards and industry test methods.
Weathering comprises a special group of environmental durability tests. There are others, tailored to the environment of interest, like vibration tests for components and materials used inside machines or vehicles, or test protocols tailored to the conditions inside a car. Among the many different kinds of environments, weathering testing refers to the environment “Outdoors on earth’s surface”, where the conditions are determined by local weather and climate conditions (i.e., long-term average of weather patterns). Laboratory weathering tests aim at inducing similar degradations to those occurring in real-life end-use conditions in an accelerated but realistic, correlating and precise way. There are three main characteristics that are crucial for laboratory weathering tests:
Weathering Takes Time as the Maximum Stress Levels Have to Be Realistic.
In general, test conditions should be worst-case but realistic: Maximum observed real-life values must not be exceeded in order to avoid unrealistic degradation. If the maximum temperature measured on a PV module’s surface in operation is 70°C, this temperature should be applied as maximum for the laboratory test. If the testing temperature is unrealistically high, either unrealistic changes will be induced that never will happen in reality, or the product will be over-designed. Due to the general principle of not exceeding real-world worst-case conditions, the achievable acceleration is limited.1)
Primary Weather Factors Have to Act Simultaneously and in a Cyclic Way
As chemical reactions are usually temperature-dependent, the mechanism of the reaction sequences of photo degradation is different at different temperature levels. Additionally, the presence, absence, and level of moisture influence the ageing processes. Finally, cycling of temperature, humidity, and light--as in the real world--adds stresses to specimens that are crucial for good correlation. Numerous studies have shown that one can only expect realistic results with high probability if radiation, moisture, and heat (the primary weather factors) are applied simultaneously and in a cyclic format. Consequently, most advanced weathering standards include wet/dry, light on/light off as well as temperature cycling.
The Laboratory Light Has to Resemble the Sun to Avoid Unrealistic Degradation
Most polymers, including encapsulants, electrical cables, junction-box housings, films, and laminates are very sensitive to short-wavelength radiation. In addition, many photodegradation processes can be induced by visible light. Therefore, the UV and visible spectrum of the laboratory light source used has to be as similar to the sun as possible. Finally, in reality high temperatures on a product’s surface are caused by solar Infrared (IR) radiation. In contrast to heating a specimen by surrounding air in an oven, IR heating induces a temperature gradient through the material which may lead to additional mechanical stresses and different diffusion processes.
From these principles it becomes clear that, for instance, the damp-heat test as specified in IEC 61646 and IEC 612152),3) is NOT a weathering test. Simulated solar radiation is missing and the combination of temperature and humidity (85% relative air humidity at 85°C) can be found nowhere on earth, except, perhaps, in an Icelandic geyser. The damp heat test just tells the experimenter how the module reacts to damp heat. Ten times the damp heat test only gives information on how a module appears after multiple iterations of damp-heat testing. Although often misunderstood, the damp-heat test is only intended “to determine the ability of the module to withstand the effects of long-term penetration of humidity” 2). Similar to the humidity/freeze or the thermal cyclic tests, it does not provide any information about ageing or lifetime of PV modules under realistic environmental conditions.
The obvious goals of weathering testing are to rank materials and products (relative testing), and to determine the ageing performance compared to a known standard and find about lifetime of a product (absolute testing). The underlying objectives in most cases are to save time, e.g. for market introduction, and/or reduce the risk of premature failure occurring within the warranty period.
How Long Will a PV Module Last?
The only way to answer this question is to do weathering testing. As almost all manufacturing industries have been doing weathering testing for decades, there are tens of thousands of weathering instruments in operation in laboratories around the globe, and one would assume that there are weathering standards existing for PV modules, as well as studies, inter-laboratory tests, and publications on the aging behavior of PV modules tested according to these standards. The truth is: There is not a single weathering testing standard for full PV modules in existence, neither in ISO or IEC nor in ASTM, DIN, BSI, CEN, CENELEC, etc. A surprise? Why is the PV module industry not doing proper weathering testing? Is it because nobody is interested in how long PV modules actually last? Is nobody asking for data about the lifetime of PV modules?
From many meetings with industry representatives, and from what we have learned in PV conferences and trade shows, three possible reasons for the existence of this situation will be discussed:
Reason 1: The PV Industry Is Relatively Young
Although PV modules have been in existence since the 60’s and 70’s, they were initially only intended to power satellites in earth orbit. Actually, the real push for terrestrial PV did not start until the beginning of the 90’s with Germany’s 1,000-roofs program, and a couple of years later with a similar program in Japan4). Other countries followed. China is now one of the biggest PV module producers, and the Obama administration recently launched a giant program to support renewable energy in the U.S.A.
During its path upwards, the young and quickly expanding PV industry had many other topics of higher interest than the long-term durability of their products: managing the raw material (silicon) supply, developing new technologies like thin-film, competing with other energy sources, dealing with feed-in tariffs, increasing the efficiency of solar cells, building factories, introducing modern manufacturing methods. Lifetime of modules was literally years away and just did not make it into most agendas.
Only recently, as investments and risks grew, financers have started to ask more aggressively for bankability criteria, including lifetime, in order to protect their investments.
Reason 2: The PV Industry Has not Been Used to Weathering
Many engineers, developers, and managers in the PV industry have their background in the electronic, semi-conductor, spacecraft or other industries that historically developed very few products intended for use outdoors. Most of these products, such as batteries, circuit-boards, or displays were never intended for outdoor use, and if so, were usually protected from weather by housings.
There are not as many experts in the PV industry who know about the scientific and technical concepts and the application of weathering testing, compared to, for instance, the paint industry. In addition, some weathering keywords are used in a different way causing misunderstandings: e.g., ‘degradation’ (polymer degradation by impact of weather vs. degradation of a solar cell’s efficiency over time). Very often, the difference between a climatic chamber (delivering and controlling temperature and air humidity) and a weathering chamber (delivering and controlling temperature, air humidity, AND radiation) is not understood.
The PV industry has to manage two conflicting challenges: Fast rates of market growth and technology innovation on the one hand, and on the other hand, the fact that the industry is offering products with expected lifetimes of 20, 25, or 30 years.
Reason 3: There Are Technical Challenges to Weathering Testing of PV Modules
There are several challenges that make weathering testing of PV modules a not-so-simple task. According to Akira Terao of SunPower5) these are:
1) 25-year warranty
2) Ill-defined field conditions
3) Harsh and varied outdoor conditions
4) Materials used near their limits
5) Limited acceleration factor → long tests
6) Large samples, small sample size
7) Subtle polymer chemistry
8) Cumulative effects, positive feedback loops
This means: Time exhaustive laboratory tests with advanced weathering technology (see Figure 3) and limited statistical options. It becomes obvious that both manufacturers as well as standards developing organizations have been hesitant to implement weathering tests for PV modules in the past. Instead, the focus has been on looking at the different single stress factors causing module failures and specifying the IEC qualification test program, each test looking at one or two of these single factors only.
In summary: For several reasons, real weathering testing has not been implemented in the PV industry yet, and internationally agreed weathering testing standards are not yet in existence. As the PV market will continue to grow, as competition increases, and as more and more utility-size projects require large capital investments, knowledge of service life and environmental durability of PV modules becomes increasingly more important. The only way to get this information is by means of weathering testing.
What Are the Requirements for a PV Module Weathering Testing Program? Why Does the IEC Program Not Meet Them?
Of course, the design verification test programs of the IEC qualification tests for crystalline (IEC 61215) and thin-film (IEC 61646) photovoltaic modules are useful to avoid design flaws and limit the ‘infant mortality’ of modules. The actual life expectancy of modules so qualified will depend on their design, their environment and the conditions under which they are operated.
From what has been stated above, an accelerated but realistic laboratory testing program to simulate environmental aging of PV modules has to meet the following five key conditions:
1) The laboratory test equipment used has to be big enough to hold a full module.
2) All relevant stresses (heat, moisture, radiation, salt, freeze-thaw, corrosion, etc.) have to be applied on the same module under test, as they are in the real world.
Again, as mentioned above, the IEC standards3),4) apply different stresses to different modules under test, so that no module experiences the complete set of environmental stresses.
3) The primary weather factors heat, moisture, and radiation have to be applied simultaneously and in a cyclic manner.
The IEC standards do not include such a test. The included outdoor exposure test, of course, delivers the primary weather factors, but is much too short and, by nature, is not able to achieve accelerated ageing.
4) The test program has to be long enough to achieve sufficient accumulation of stresses and damage.
Nobody can predict the acceleration factor of a laboratory test prior to actually doing the test and comparing to outdoor results. Nevertheless, the literature reports acceleration factors of the laboratory test compared to outdoor exposure in a wide range between 4 and 276) or even between 2 and 507), depending on the laboratory test, the actual outdoor location, and the material of interest. This means that it is very unlikely that a test program which is much shorter than a year will age the PV modules sufficiently to provide sufficiently useful information.
5) In addition to the laboratory accelerated test, identical modules should be exposed outdoors under benchmark conditions for reference.
Accelerated laboratory test data only provide part of the story. It is crucial to compare and correlate these results with aging data that were produced in well-defined outdoor conditions. The two benchmark outdoor locations that have been historically established as references by many industries, are Phoenix, Arizona for hot-arid and Miami, Florida for warm-humid climate.
Recently, Atlas Material Testing Technology has developed a test program for PV modules which meets the above key requirements: Atlas 25PLUS TM.
Actually, the IEC design qualification tests and environmental durability or weathering tests are not competing concepts but may be used in combination. The ideal situation would be carrying out the IEC qualification test program before AND after an environmental durability test. This strategy would provide the information that not only a new module can withstand a hailstorm but also one that is many years old. Similar test concepts are used in other fields, for instance, stone chipping resistance testing of car paints before AND after weathering.
Andreas Riedl is Global Manager of Consulting Solutions and Global Manager of Solar Energy Competence Center at Atlas MTT GmbH (www.atlas-mts.com).
1) ISO/DIS 4892-1:2008-12 Plastics ? Methods of exposure to laboratory light sources ? Part 1 General guidance, paragraph 4.2.4
2) IEC?61215:2005 Crystalline silicon terrestrial photovoltaic (PV) modules?- Design qualification and type approval
3) IEC 61646:2008 Thin-film terrestrial photovoltaic (PV) modules?- Design qualification and type approval
4) “PV solar power: The last years in retrospect and perspectives for the next 25 years”, Dr. Winfried Hofmann, president EPIA, PV Symposium Bad Staffelstein, OTTI, March 2, 2010
5) “Modules: Remaining Reliability Challenges”, Akira Terao, SunPower, Workshop II on Accelerated Aging Testing and Reliability in Photovoltaics, April 1 and 2, 2008, Denver, CO, NREL & Sandia
6) Fundamentals of Weathering, Atlas Seminar, Dr. Jorg Boxhammer, 2008, Comparison of some calculated and measured acceleration factors
7) How long do I have to run my automotive test?, Warren Ketola, 6th International Automotive Weathering Symposium, October 2008, Detroit
For more information, please send your e-mails to firstname.lastname@example.org.
ⓒ2011 www.interpv.net All rights reserved.