Ammonium hydroxide attacks panels

“In one place, I saw a roof gutter that had corroded completely after ten years,” reports Willi Vaaßen, Head of the Regenerative Energies division at TÜV Rheinland when interviewed about the aggressiveness of ammonia. The pungent gas develops in livestock barns when animal excrements rot. The concentration is especially high in pigsties and chicken coops. There the ammonia gas burns the eyes, causes fits of coughing and the smell stings in the nose. To prevent health damage, the barn air is extracted and transported off via the roof – where solar panels are installed. Then a creeping chemical reaction sets in on the panels, caused by the ammonia: racks and frames corrode, the glued connections on the sockets become discolored, glass turns dull. The barn air makes the panels age faster and may lead to damage that results in loss of performance and, in the worst case, arcing.
As a result, many farmers are uneasy about their planned investments in PV on their barn roofs. That is why the German Agricultural Society (DLG) developed a method at the end of 2009 to test the resistance of solar panels to ammonia. Module manufacturer Schott Solar assisted the society, which also tests food products and agricultural engineering. In May 2010, TÜV Rheinland then introduced its own ammonia test method.
Integrated roof systems are especially vulnerable to ammonia, because the foil on the back of the panels is fully exposed to the barn. Ammonia fumes may penetrate the foil and damage the strip conductors there. But even if farmers install a rooftop system, they still cannot prevent the aggressive ammonia attack: the air still penetrates through the shingles or is transported outside via an exhaust shaft.
This leads to corrosion on parts of the aluminum frames as well as discoloration and damage to the silicon adhesions with particular frequency, as Jörg Althaus, Head of the Photovoltaic Module Qualification division at TÜV Rheinland, observed in tests. “The damage is partially cosmetic. An imminent risk is not to be expected, but weakening of materials always bears a risk,” he reported in a workshop titled “Solar panels and their resistance to ammonia”.

Liquid vs. gaseous

Since the end of April, TÜV Rheinland has operated an accessible testing chamber. Up to twelve solar modules can be tested simultaneously in the chamber. The small test samples have to suffer the same fate as large modules. The results can be transferred: “These solar modules are only about half the size of the originals, but otherwise everything is the same from the junction box to the frame,” explained Willi Vaaßen.
At the beginning of the test, the floor of the test chamber is covered with water. While ammonia gas is fed into the chamber, a heater in the floor heats up the water until it evaporates, which results in almost 100 percent air humidity. Now the gas mixes with the steam and produces corrosive ammonium hydroxide, a liquid ammonia-water mixture.
“It’s as if the modules were standing in a washhouse,” is how Vaaßen describes the test procedure. After eight hours in the ammonia vapor at an ambient temperature of up to 60 degrees Celsius, the chamber is cooled down again. “The condensation that now develops and covers the modules is the essential difference to other tests,” explains Vaaßen. Under real conditions, the gaseous ammonia also compounds to ammonium hydroxide, for example when condensation develops on the inner bottom side of the panels, either through rain, fog or thawing. Ammonium hydroxide is much more aggressive than the pure gas other institutions use for testing.
The newly drafted standard of the TÜV test is based on the so-called Kesternich test (DIN 50081), which is used to simulate the effects of acid rain on metals and coatings. The new ammonia test procedure resembles it down to the last detail, the only difference being that sulphur dioxide is used with the Kesternich test. TÜV Rheinland has refined the test even more and raised the testing temperature from originally 40 to 60 degrees Celsius to approximate the usual operating temperature of photovoltaics modules.
Each testing institute has different ideas about how strong the ammonia concentration has to be in the tests in order to simulate the actual conditions and time periods as close to the real conditions as possible. The concentration is stated in ppm, meaning a ratio of one millionth per volume unit. In pigsties, the ammonia concentration is approximately 40 to 50 ppm.

High vs. low concentration

In order to be able to show the strain on the modules in time lapse, DLG tests with an ammonia concentration of 750 ppm. TÜV Rheinland on the other hand uses a concentration that is eight times higher: 6,667 ppm. “We use such a high concentration to speed up the test procedure. This shortens the ageing process,” explains Vaaßen. This is also why TÜV Rheinland only tests for 480 hours instead of the 1,500 hours of DLG. “On the other hand, the ammonia concentration becomes highly diluted by the introduced water in our test, and this is where we have to step up, because after complete condensation, the ammonia concentration in the chamber is practically zero,” says Vaaßen.
DLG however considers its concentration of 750 ppm as much closer to reality. “TÜV Rheinland bases its test on a material testing standard, not on the experiences gained in an agricultural environment,” states Winfried Gramatte, Project Manager at the DLG test centre. The association therefore sees no need to change its test procedure. “We have calculated our values exactly. As it is possible to describe damage to the module caused by chemical attacks, temperature and humidity mathematically, we are able to perform a simulation that maps 20 years of operation of a solar installation on a barn in time lapse,” says Gramatte. This way, the test values obtained closely reflect reality.

More ammonia tests

Besides DLG and TÜV Rheinland, there also are other institutes that offer an ammonia test, for example the Swiss testing institute Société Générale de Surveillance (SGS). The society takes a sportive approach towards the squabbles about the test values and offers an ammonia test in which the concentration, the temperature and the air humidity can be individually adjusted. In this test, the ammonia resistance of the modules can be tested according to DLG, TÜV, or individual criteria. “In addition, we can introduce four more pollution gasses like chlorine, nitrogen oxides, hydrogen sulphide and sulphur dioxide,” says Jörn Brembach, Business Manager Photovoltaics at SGS. To put it exactly, the dispute about the right concentration is based on a dispute about the acceleration factors. No one can say exactly how long a test has to be performed at what concentration in order to simulate a 20-year lifespan. TÜV justifies its high values with missing experience. The high concentration is an initial approximation value, which may still change with increasing experience, says Althaus.

Creating international standards

In order to create a uniform standard for ammonia tests, TÜV Rheinland wants to establish its test as the future international standard IEC 62716 by the name “Ammonia corrosion testing of photovoltaic (PV) modules”. The draft was already submitted to the International Electrotechnical Commission (IEC) where it is presently being discussed. However, it will still take two to three years until it will be included in the IEC guidelines in whatever form. So it will still take some time until it becomes internationally valid.
DLG can’t keep up in this respect. However, Gerhard Kleiss, Head of Corporate Quality at SolarWorld, does not see any need to do so. The DLG test is also accepted in the Benelux countries, and even in the USA it is quickly explained what DLG is and what it does, he says. Konrad Fredrich, Product Manager at Solon, has a different opinion. For him, the international standard gives different boards the possibility to become involved and to keep the test procedure adaptable. “At DLG, the ammonia test is specified and cannot be changed. Several times we had suggested to consider other requirements in the test, but were always turned down,” says Fredrich. He would have welcomed the participation of DLG in the draft of the IEC standard. DLG argues that it cooperates for example with the Association for Electrical, Electronic & Information Technologies (VDE) in the photovoltaics sector. This association is involved in the IEC and also represents the interests of DLG.

The experts in the companies also still disagree about the informative value of the individual tests. Fredrich of Solon sees the TÜV test at a clear advantage: “Because the DLG test does not consider any temperature change, it can be assumed that only the TÜV test is presently able to separate the wheat from the chaff.” System manufacturer Conergy on the other hand wants to do completely without the TÜV test, because its own modules had already proven their resistance to ammonia in the DLG test. Schott Solar, co-developer of the DLG test, is of the same opinion: “The ammonia concentration in the DLG test already maps a worst case scenario, which does not happen as such in reality. Therefore, the TÜV test has no added value for us,” says Thomas Block, Product Manager at Schott Solar. SolarWorld on the other hand takes a dual approach. Although it already has a DLG seal, its modules are currently tested by TÜV so as not to unsettle buyers. Now it remains to be seen how the demand will develop, once the TÜV Rheinland test has become an IEC standard.