Insulation from empty bottles

An early prototype of the insulation bar. Image: Hochuli advanced

But although it has been established for more than four decades, the insulation web has potential for improvement. An Empa team led by Michel Barbezat and Giovanni Terrasi from the Mechanical Systems Engineering lab is working on a new type of product – together with experts from the metal construction company Hochuli in Wigoltingen, which has founded its own spin-off named "hochuli advanced" to commercialize the new developments.

The highlight of the "Alpet" insulation web: Inside the glass-fiber-reinforced plastic is a foam strip made of polyethylene terephthalate – PET, in other words, from recycled bottles.

Numerous air pores within this layer insulate efficiently: The thermal conductivity of the prototypes averages around 0.1 W/mK, depending on the web width – far less than a standard insulation web made of another polymer, polyamide (around 0.25 W/mK), and also significantly lower than high-end products available today.

Under the microscope: The largest pores (black) have a diameter of about half a millimeter. Image: Hochuli advanced

"As many as possible" is the answer, in order to achieve a high insulating effect. Only: "Too many" would have jeopardized the stability of the prepared PET strand, because its encasement with the black plastic takes place at temperatures of up to 300 degrees and under high pressure.

The researchers also adapted the thickness of this "sheath" to the necessities: as thin as possible for a good insulating effect – but thick enough so that the web would later also withstand mechanical loads, including the post-treatment of the finished overall profile, such as powder coating or anodizing.

From numerous samples, the developers distilled seven variants for testing – altogether an insulation web about 1,000 meters in length – and from these the final prototype as the basis for the finished products. An important step in the production process was also achieved along the way: The web welds virtually by itself due to the heat – without the need for another "seam" to be glued.

"This is already a big advantage," says metal construction engineer Frank Hochuli from Empa's industrial partner. "There are no local weak spots where material could come off. And the fewer steps, the cheaper the product."

A lengthy process that paid off in the end, as Empa's Michel Barbezat confirms. "Technically, we certainly have a very good chance," he says. The measured values and the product, which can be easily recycled despite the use of two different materials, are already convincing. And the experts are also confident about the necessary long-term stability over many years, which his team tried to estimate with experiments using the innovative Stepped Isostress Method (see box).

Simulation of the load on the new web, but still using data with a different material for the cladding. Image: Empa

How does a plastic component deform over a period of 30 or more years – under continuous load and large temperature differences? Common creep tests in the laboratory over 1,000 hours under combined thermal-mechanical loads do not allow any serious statements to be made.

In order to be able to make long-term predictions, however, new methods have been developed and refined in recent years, including the Stepped Isostress Method (SSM), which Empa experts applied to the new insulating web.

In simple terms, the deformations from numerous individual tests with increasing loads are added up to form a master curve that provides information about longer periods of time.

This procedure can also be carried out with a single measurement performed with gradually increasing loads – which makes it easier to investigate critical temperature ranges at the same time.