The preservation of sound archives: A computer science based approach to quality control

Most tape is made by coating the surface of a plastic film base or backing with a paint or “ink” containing magnetic particles homogenized in a binder that adheres to the film and dries by evaporation. Figure7.2 shows a general cross section of a tape. The coating is usually so thin that it is only a small fraction of the overall tape thickness [80, p. 97]. Although the base is magnetically inert, it must meet stringent requirements of thickness, strength, and stability, in order to give a reasonable playing time in a compact reel. A tape about 50 µm thick fills a 18 cm diameter reel with 365 meters of tape, which equals to a half hour at 7.5 inch/s. In addition to the base and binder, most tapes produced since the early 1980s show a back coating that acts as an aid to tape packing on the hub and reel [45]. The degradation of all types of audio carriers is mainly due to their intrinsic chemical instability [66], aggravated by inadequate handling, storage conditions and poor manufacture. The effects of the process of degradation are multiple and carrier-dependent. One of the problems most often observed in magnetic tapes, especially Open-reel tapes, is known as Sticky Shed Syndrome (SSS, see Appendix B for a definition): this is confirmed by the survey in Figure 7.1 (second value from left, “sticky shed”), and also by the author’s experience with the archives of the Fondazione Arena di Verona (see Section 11.1 of Chapter 11) and of the Scuola Normale Superiore di Pisa (see Section 11.2 of Chapter 11).
Organic materials are commonly prone to spontaneous chemical decay, and metallic particles in the presence of oxidizing compounds are subject to oxidation [79].
The importance of remedial actions finalized at making audio media playable is revealed within the field of preservation by the number of scientific and non-scientific sources discussing them. But at the same time it shows how the approaches are fragmented and how seldom the information builds upon scientific knowledge: “baking methods have been refined in diverse studios and labs by many different practitioners, they vary significantly in terms of equipment, temperature, and duration.” [81].

After treating a set of tapes, or “baking” them as is often heard, the author has observed that some were not responding to the treatment at all, while others required longer treatments to achieve an improvement. Therefore the author decided to have a deeper insight on thermal treatment, considering that the object of the treatment were unique copies of unpublished performances by artists the calibre of Luciano Pavarotti, for which the risk of collateral damage must be avoided completely. Despite the popularity of thermal treatment, the material gathered by the author formed a contradictory set of information, from which basically emerged that no-one clearly knows what happens in the incubator. It all started in 1993, with the US Patent assigned to the Ampex Systems Corporation [82] which presented “a method of heat treating magnetic recording media comprising maintaining the magnetic recording media at a sufficiently elevated temperature for a sufficient time to overcome the adverse consequences of undesirable shed, stickiness or squeal”. It recommended “a temperature of at least 50 ◦C for at least 8 hours”, however “lower temperatures and/or shorter times can restore media sufficiently to enable playback”. The factor that determines the temperature and the duration of the treatment is the state of deterioration of the audio document, but the patent does not provide specific indications to plan adequate treatments. It is generally agreed that the presumed responsible for the sticky syndrome is hydrolysis (hydro “water” and lysis “separation”), a reaction between the polymers of which the tape is made and water from atmospheric moisture: “when a sufficient number of [polymeric chains] have been hydrolyzed and broken, the binder becomes undesirably weakened due to degeneration of molecular weight” (see also [83]).

The patent curiously refers to its own results as “unexpected” and “surprising” , besides admitting that the theory behind it is not ascertained: “it is believed that the heat restoration process according to this invention, to some extent, may re-verse certain hydrolysis” (italics of the author). According to this “belief”, more recent sources claim that thermal treatment is aimed at “removing the moisture that has accumulated in the binder” – although apparently this is only “thought” to happen. Other sources from the archival community admit that the data they possess about thermal treatment “is merely anecdotal and will require further study.” [78, p. 24]. Other open questions about the treatment are: the recommendable humidity level in the incubator during the treatment, which is not indicated in any of the sources; the duration of the benefits (“incubating the tape returns the tape to a playable condition for weeks or months after treatment” [84]; “this remedy is temporary; the tape will revert over time.” [85]); and the risks involved with the treatment. “Although some report having 20-or-more successful ’bakes’, there is no published or documented information on how many times a tape can be baked, cycling back and forth between the sticky-firm-sticky succession before it fails completely or before the signal is distorted or altered beyond use”. [45] In fact, there is “little knowledge about how exposure to increased heat may impact the tape artifact itself”. [81]

In the author’s experience the treatment has generally produced satisfying results, nevertheless within the scope of a scientific methodology a deeper understanding of the reasons why and how the treatment works is required, and the author believes that a deeper understanding will save a lot of documents from the consequences of unaware treatments – in terms of temperature, duration, and equipment (instead of a precision incubator, relatively expensive, “the most common equipment is the American Harvest Snackmaster Pro FD50 Food Dehydrator” [81]; even the home made solution of the hair-dryer-in-a-cardboard-box is said to “work well” [86]. The author’s aim is to prevent improvised treatments on precious audio documents, and to achieve a scientific protocol for tape restoration. In order to do this, a set of chemical analyses have been planned in collaboration with the Department of Industrial Engineering (Chemical sector) of the University of Padova. The first step has been the characterization of the tapes, because only a precise knowledge of their properties would provide the understanding of their behavior at varying environmental conditions. Characterizing the tapes has been necessary as most brands and models were unknown or uncertain: very few tapes show the brand/model on the back surface (among the tapes analyzed in this work, only two have been identified by their names, see Figures 8.8(a) and 9.1). This information is usually derived from the reel or the box, although “a tape’s box may not be original and may not accurately indicate the type of tape” [79, p. 8].
Eight tape segments have been selected and analyzed with a Fourier Transform InfraRed spectroscopic analysis (FTIR) and a ThermoGravimetric Analysis (TGA). The next Section presents the analytic techniques, while the experimental results are described in Chapter 8 and discussed in Chapter 10.