Ben Welter - Monday, March 25, 2019
A PCM-based thermal energy storage system installed in an 11-story building at an Australian university used just 15 percent of its heat storage capacity to shift peak cooling load, according to researchers who monitored system performance for 25 months.
A team led by Morshed Alam of Swinburne University of Technology in Hawthorn found that the PCM reduced chiller cooling load by 12 to 37 percent in winter but remained inactive in summer, partly because the ambient temperature was not cold enough to charge the PCM tank. The tank was designed to reduce the daytime cooling load on the chiller by 33 percent.
The results of the study are reported in "Energy saving performance assessment and lessons learned from the operation of an active phase change materials system in a multi-storey building in Melbourne," published in Applied Energy earlier this year.
"The factors that contributed to the underperformance of active PCM system," the researchers concluded, "include mismatch between designed and actual operation of the PCM system, inefficient operation logic of the system, poor material quality, and limited knowledge of maintenance staffs during the operation stage."
The TES system, installed in the 11th floor of Swinburne's Advanced Manufacturing and Design Centre, completed in 2015, was designed to minimize the daytime cooling load on the chiller and increase the building energy efficiency. The system includes a 5x4x2-meter tank filled with 5,120 FlatIce PCM panels made by PCM Products Ltd. Each HDPE panel is 500x250x45 mm and is filled with a salt hydrate PCM with the melting temperature of 13–15 °C. Water is used as the heat transfer fluid.
The researchers reported two problems with the PCM: a "very high degree of supercooling" that slowed the solidification process and a measured latent heat capacity (53 joules per gram) that was much lower than the manufacturer's specification (160 J/g).
Alam, a senior research fellow at Swinburne's Centre for Sustainable Infrastructure, Department of Civil and Construction, answered a few questions about the research in an email interview.
Q: This the first comprehensive report I've seen that analyzes the actual performance of a PCM/TES system in a commercial building. Do you know of others? For example, has the TES system at Melbourne CH2 undergone this kind of analysis?
A: "I am aware of the PCM system installed in CH2 building in Melbourne. But no such study was carried out to understand the performance of the system. The system went out of order within 5-6 years."
Q: Any theory on why the PCM's thermal storage capacity of 53 J/g, as measured using differential scanning calorimetry, did not match the manufacturer's specified capacity of 160 J/g?
A: "The DSC test is widely used by the researchers around the world to test PCM capacity. But it may not be an appropriate approach to test PCM thermal energy storage capacity. That is why we are now in the process of conducting an experimental study where 6 degree Celsius chilled water will be supplied in a small PCM tank. We will measure the flow rate, melting and solidification temperature and time required to completely solidify and melt the PCM. Based on the result we will optimize the system."
Q: Have the researchers' recommendations been put into practice?
A: "We are now working with our facilities department to put the recommendations in practice. We need to finish the optimization and cost-benefit analysis study. Then we can define the ideal operating condition of PCM in this building. We will be able integrate those maybe before next summer in Australia."
I shared results of the paper with Zafer Ure of PCM Products Ltd. and asked him to comment on the authors' observations on supercooling and specified-vs-measured latent heat capacity.
"We have been supplying the same PCM for the last 20 years and many of the installations are 10~15 MWh levels," Ure wrote. "You cannot use the DSC machine to work out the latent heat for sulphate-based hydrated salts, it only picks up one of the chemicals in the mixture and gives you the data for that chemical. This is a well-known handicap for the DSC application for sulphate-based PCM testing. Moreover, DSC sample is less than gram quantities and this is a mixture of multiple salts, nucleating agent, stabilizers and thickening agent so if you do not pick up the correct mixture, which is very difficult unless you pick it up from the reactor vessel while you are making the PCM (i.e. while the agitators and mixers working at the correct temperature (some salts crystallize at room temperatures), you may not be able to pick up true sample.
"They took the sample out of one of the containers and no idea where and how they took the sample. If you do not have the true sample, especially lack of nucleating agent in the very small gram quantity sample, you can get it very wrong data as this is the case for their DSC. We would not ship any PCM unless QA release the goods and our records show the product supplied was within the standard capacity level. The actual latent heat can only be established using air test and actual freeze and melt profiles.
"If they managed to charge and discharge the tank fully they would have measured the tank performance which would have shown the true TES capacity. Sounds like their chilled water design and control did not allow that and therefore there is no way of evaluating the TES tank performance."