A B.B. King recording brings PSI scientists and tape preservation experts together to rethink magnetic tape degradation

The Montreux Jazz Digital Project was launched in 2007 to preserve and digitise the Montreux Jazz Festival’s audiovisual heritage and is today carried by EPFL, the Claude Nobs Foundation and the Montreux Jazz Festival. The 1967–2013 Claude Nobs Collection, inscribed on UNESCO’s Memory of the World Register in 2013, comprises around 14,000 recordings. During its preservation, a particular challenge emerged with a group of Pyral CJ87 magnetic tapes originating from a private collection. Among them was a unique 48-minute recording of B.B. King’s 1980 concert. For deteriorated Pyral tapes of this type, a specialised cleaning, playback and digitisation procedure was developed at the United Music Foundation in Geneva, allowing over a hundred Pyral recordings to be recovered and digitised. The practical solution existed – but exactly why such an elaborate treatment was necessary remained an open scientific question.

The B.B. King tape presented a striking contradiction. It showed none of the characteristic macroscopic symptoms of sticky-shed syndrome: it wasn't sticky, its magnetic coating was not delaminating, and ‘baking’ – the standard heat treatment for sticky-shed tapes – did not restore its playability. Yet chemical analysis at the Paul Scherrer Institute revealed the same signature of polyurethane binder hydrolysis associated with sticky-shed syndrome. How could a tape show the chemistry of sticky-shed syndrome without behaving like a sticky-shed tape? Building on the unusual behaviour observed during its preservation, researchers at PSI investigated what was happening inside deteriorating magnetic tapes. 

Different symptoms, same underlying chemistry

The investigation was led by Jack Harrison, a postdoctoral fellow at PSI, who conducted the research together with PSI colleagues, the EPFL Cultural Heritage and Conservation Platform and internationally recognised experts in magnetic tape preservation. Using a combination of infrared spectroscopy, optical microscopy and scanning electron microscopy, the PSI team uncovered a surprising result: tapes exhibiting apparently different forms of degradation all showed the chemical fingerprints of polyurethane binder hydrolysis. At the same time, microscopy revealed physical differences that spectroscopy alone could not capture. The underlying chemistry may be shared, but differences in surface morphology, layer interactions and secondary degradation processes help determine how the damage appears in practice, and why the same treatment does not work for every tape.

The Pyral tape from Montreux, originally recorded by sound engineer Philippe Zumbrunn,  provided the key to resolving the puzzle. Although baking reduced the chemical signatures associated with binder hydrolysis, the recording still could not be played. The researchers found that the polyurethane binder had hydrolysed not only in the magnetic coating, but also in the tape's carbon back coat. Their measurements confirmed an interpretation that had emerged from the earlier tape treatment at the United Music Foundation: material from the degraded back coat had transferred onto the adjacent magnetic recording surface, producing the black residue observed during playback.

From symptoms to cure

For decades, archivists have classified degraded magnetic tapes according to their behaviour during playback, and conservators often rely on experience to decide which approach is most appropriate. These results show that tapes cannot be reliably diagnosed from their playback behaviour alone. Identifying these microscopic differences makes it possible to design remediation strategies based on the actual cause of the degradation, rather than on its observable symptoms.

The researchers also found that degradation is far from uniform. Even within a single tape, the degree of chemical degradation can vary considerably across its surface, meaning that measurements at a single location may not accurately represent the condition of the recording. Reliable diagnosis therefore requires measurements at multiple locations across the tape, rather than relying on a single-point measurement.

To establish a reliable benchmark, the team compared the degraded archival tapes with a modern Recording The Masters (RTM) SM911 tape, representing the current state of the art in analogue tape manufacturing. The modern tape provided a reference against which the chemical and microscopic changes in historical tapes could be objectively assessed.

The study brought together expertise spanning physics and materials science, conservation science and practical tape restoration. The collaboration included Alain Dufaux (Executive Director, EPFL Cultural Heritage and Conservation Platform) and internationally recognised tape preservation specialists David Hadzis (United Music Foundation, Switzerland), Richard L. Hess (Aurora, Ontario, Canada), Kelly Pribble (Kvibe Enterprises, USA), and Frédéric Ménétrier (Audio Gecko, France), combining state-of-the-art materials characterisation with decades of experience restoring historically important recordings. Hadzis had previously developed the preservation procedure used for the deteriorated Pyral CJ87 tapes from the Montreux archive. The work was carried out within the BRIDGE Discovery project Play It Again, led by PSI scientist Sebastian Gliga and jointly funded by the Swiss National Science Foundation and Innosuisse.

Beyond explaining the behaviour of a single B.B. King tape, the work provides a scientific framework for diagnosing magnetic tape degradation. By moving beyond symptom-based classification towards understanding the underlying chemical and microscopic mechanisms, the approach allows conservators to select targeted remediation strategies before attempting playback, reducing the risk of damaging both irreplaceable recordings and increasingly rare playback machines. For archives around the world engaged in a race against time to preserve twentieth-century audiovisual heritage, this represents an important step towards more reliable and evidence-based preservation.