Abstract

Under prolonged storage, motion picture film may adhere within the reel, forming a solid mass; this is referred to as blocking. In response, the National Film and Sound Archive of Australia (NFSA) has developed conservation treatments to successfully deal with each of the mechanisms of blocking, and conducted several projects to treat large numbers of affected films.

Keywords

film, blocking, gelatin, conservation, treatment.

Background

For storage, motion picture film is wound onto reels or simple cores; also known as bobbins. The tension under which the film is wound has an impact on the chemical stability of cellulose ester-based films during long-term storage (Bigourdan, 1997). The more tightly wound the film, the lower the probable life expectancy.

However, as well as decomposition, tightly wound films may suffer from a strong adhesion between the layers of film. This is referred to as blocking. Blocking may be sufficiently strong enough to cause serious physical damage to the emulsion and base layer of the film. Additionally, blocking may occur in cellulose ester- and polyester-based films.

Before either a blocked cellulose ester- or polyester-based film may be safely unwound prior to projection, or duplication, the condition must be dealt with. There are several different mechanisms of blocking depending on not only the wind tension but also the chemical condition of the film. Broadly, we have found that there are three main types of blocking:

•  Gelatin cross-linking between layers

•  Exuded additive chemicals forming a cementing layer

•  Decomposition by-products forming a cementing layer

External blocking or cementing agents, such as animal excrement, and structures, such as insect carton, are not considered in this paper, although blocking due to these reasons may also be occasionally encountered.

The Role of Gelatin in Blocking

Since gelatin, both as the film emulsion and a major component of the backing layer, was the point of the adhesion between film layers, the majority of the research focused on the behaviour of gelatin.

Gelatin is a manufactured product formed by the combination of amino acids into a complex compound in long chains. Gelatin has many unique properties including significant response to changes in pH. Gelatin is a macromolecule that can be viewed, simplistically, as a combination of carboxylic and amino groups. Changes in pH alter the protonation of the groups. At the isoelectric point, ~pH 4.8, all the carboxyl groups are protonated, whereas the amino groups are not, giving the system a balanced charge (Kowalski, 1972). As the pH drops below the isoelectric point, the amino groups in the gelatin structure become protonated—that is, have a net positive (+) surface charge. Therefore, at the isoelectric point, there is a balance of small charges (+ and −) along the length of the chain. In this condition, the chains are tightly coiled. As the pH increases or decreases from this state, the balance of charge alters. Lowering the pH increases the prevalence of positive charges along the chain causing the gelatin chain to uncoil as the charges repel each other. Inversely, as the pH is increased there is an increase in the prevalence of negatively charged sites, and the chain once again progressively begins to uncurl.

Interlayer Gelatin Cross-Linking

This is the most commonly encountered form of blocking and may be the result of:

•  a disaster involving water

•  improper drying after processing

•  tight wind and storage under high, relative humidity

•  natural shrinkage and subsequent increase in tension within the film reel

•  exposure to extreme heat—for example, fire

The blocking mechanism is thought to be the cross-linking of the gelatin as the result of the gelatin chains being brought into close proximity and placed under pressure so that the charges on each surface may interact from one layer to the other.

The severity of the adhesion may be such that the bond between the layers of gelatin becomes stronger than the bond between the emulsion and the film base. Any attempt to unwind a film in this condition will tear the emulsion from the base, or even tear the base.

There is an additional caveat for polyester-based prints. While the longitudinal strength of polyester is high, the internal bonds holding the longitudinal polyester chains together is comparatively weak. If even a lightly blocked film is unwound, the polyester base will fracture and form voids inside the base polymer that permanently blemish the film.

The considered approach to this problem was to minimise the interlayer adhesion by raising the pH to promote a net negative charge on the adhering surfaces, causing the gelatin surfaces to slightly repel one another. Alkali processed gelatin, typical in photographic emulsions, shows an inflection point and significant reduction in the slope of the swelling curve around pH 8 (Sheppard, 1942). Kodak gave advice on two benign photographic solutions that worked in this region: Hypo Clearing Agent (HC-1) and Rewash (PB-6).

To identify any problems due to extended soaking in these solutions, test frames, colloquially known as “China Girls,” were sought from as many different film emulsion types as possible. Additional test frames were printed on contemporary stocks. Each frame was measured using a Macbeth densitometer using the appropriate status filter. The test frames would provide information on density and colour balance changes due to the treatments. A pin register device was used to ensure that the same location was measured on the test frame for consistency.

After soaking in the solutions for time periods ranging from a few minutes to days at a time, the test frames were placed in an accelerated aging chamber for 49 days and the density measured.

The average density loss in each dye layer, in the negative test material, was in the order of 10%. It was noted that print stock typically had a lower dye loss with an average of 5%.

Further experimentation identified that sodium polymetaphosphate, a component of HC-1, was an effective swelling agent and caused the least dye loss when used by itself—that is, without the sulphite preservatives also found in HC-1. In solution, at a concentration of approximately 1.5%–2% w/v, the pH fell close to the 8–8.5 range. However, without the preservative, this solution required frequent changing, optimally on a daily basis. This protocol had the added advantage of preventing excessive bacterial action occurring on the films being treated.

It was noted that some films turned the solution slightly magenta. There were several possible reasons:

•  Aged gelatin containing the upper magenta dye layer was dissolving

•  Magenta leuco dye was being converted to a visible magenta form

•  Residual cyanine sensitising dye

UV-visible spectrophotometry was used to determine the most likely underlying cause of the magenta colouration. Samples of image forming magenta dye exhibited a single peak absorbance at 530 nm, whereas, samples of the magenta water showed peaks at 499 and 547 nm, closely aligning to the absorbance spectrum of the sensitising dye. Kodak confirmed that while these dyes were mostly removed during processing, it was feasible that some may remain under some circumstances.

A further consideration was stabilising the film after treatment. Rewashing was required to remove the residual chemical from the film prior to returning to long-term storage.

Two problems became apparent:

•  In the original processes, the dyes were stabilised by formaldehyde in the final bath

•  The emulsion showed evidence of reticulation

There was no solution to the first problem, as formaldehyde is a now listed as a hazardous chemical, so the films were left unstabilised. The second problem was controlled with very careful attention to solution temperature.

Exuded Additive Chemicals Forming a Cementing Layer

Additives are incorporated in the film base polymer for several reasons, such as to assist in the removal of other manufacturing additives and solvents, as a flame retardant, and as a plasticiser. Also, the additives are not chemically bonded within the base polymer.

Fourier transform infrared spectrophotometry (FT-IR) analysis of the crystal residues found on severely decomposed films identified triphenyl phosphate (TPP) as the dominant species. Kodak researcher, Dr. Tulsi Ram (1990) notes, “degradation of CTA polymer under ambient conditions is accompanied by its ability to pick up more water in the deacetylated regions which facilitates exudation of the solid TPP to the surface.”

The positive charges on the gelatin surfaces provide bonding sites for the exuded TPP. The exuded TPP, firstly, bonds to the gelatin layers by hydrogen bonding. As the quantity of TPP increases, crystals form in situ via covalent bonds or van der Waal forces (Atkins, 1986), and thus create a linkage between the plasticizer-coated surfaces, adhering the two layers firmly.

The increase in free acid, in the decomposing film, lowers the film pH below the isoelectric point. A pH-induced change was considered too risky to attempt due to the instability of the gelatin. The most feasible approach was to either remove the TPP cementing layer or break the crystalline structure.

TPP is highly soluble in lower alcohols such as ethanol and has a comparatively low melting point.

The first approach sought was to dissolve the TPP cementing layer without extracting the TPP remaining in the film base. A solution of 5%–10% ethanol in isopropyl alcohol (IPA) was dripped onto the point of adhesion as the film was unwound. TPP is only sparingly soluble in IPA. This is a slow process, as it takes a while for sufficient TPP to dissolve, and the residual acetic acid undergoes an ester reaction with the alcohol forming an acetate ester and water. The damp film requires drying before it can be safely rewound. This is a successful but slow treatment.

TPP has a comparatively low melting point (~50°C); however, there are risks to the film at this temperature. This temperature will possibly cause the deteriorated gelatin emulsion to pass through its T_g. Additionally, the rise in temperature may move the emulsion beyond the safe region for photographic materials and suffer plastic, or nonreversible, deformation (McCormick-Goodhart, 1996). This may lead to delamination of the emulsion during handling. Aware of the risks, severely decomposed and blocked films were placed in an incubator at slightly above the literature melting point and left for 100 minutes, sufficient to raise the internal temperature above 50°C. As expected, the TPP melted, and the films were very easy to unwind; unfortunately, the liquid TPP made handling the films awkward.

However, both methods occasionally struck persistent adhesions. Several mechanisms were hypothesised for this problem. The most feasible mechanism is that these adhesions are regions of locally hardened gelatin that are strongly cross-linked and have hardened (Somasundaran, 2006, Bigni, 2004) prior to the onset of decomposition, and are not as affected by the pH change as other parts of the emulsion. This may be supported as the persistent adhesions appear to fall in regions of suspected greater compression caused by vertical storage orientation (fig. 1).

Decomposition By-Products Forming a Cementing Layer

One method commonly used to synchronise sound to image in motion picture film, apart from optical tracks, was to apply a thin magnetic track, commonly called a mag stripe, to the film.

It is very common to find that the mag stripe has blocked the adjacent layer of film and unwinding the film will cause the stripe to become severely damaged (fig. 2).

There has been much research into the causes and treatments of blocking of magnetic media, such as audio and video tape. Where blocking occurs in magnetic media, the cause is often hydrolysis of the binder. However, the treatment for this problem, low relative humidity rejuvenation, is not successful with mag stripe on film.

The mag stripe was commonly made using cellulose nitrate as the binder (Kolb, 1961), unlike the PE/PU or PVC binders in magnetic tape. Analysis of the stripe by FT-IR and simple solubility tests confirmed that cellulose nitrate was the binder.

Examination of blocked areas under a microscope indicated that the adhesion was the result of a distinct layer that effectively cemented the two layers of film together (fig. 3). This layer was not present in nonblocked sections of stripe.

Samples of the cementing layer were analysed by FT-IR, melting point, and solubility. Analysis indicated that the cementing layer was primarily cellulose nitrate, silicon oils, and gelatin, but TPP was not involved in the adhesion. It was hypothesised that the cementing layer was formed by partial hydrolysis of the cellulose nitrate binder, and the acidic by-product was interacting with the adjacent gelatin layer.

Water was found to be successful in softening the cementing layer sufficiently to unwind the film, but the treatment is slow and laborious, with controlling the application and managing drying time. A solution of water, Photo-Flo, and isopropyl alcohol was formulated. This simple solution showed superior penetration of water into the adhesion and some improvement in the drying time; however, some persistent adhesions that could cause damage during handling remained.

FT-IR analysis had not given relative proportions of the two main constituent components of the adhesive layer, cellulose nitrate and gelatin. To specifically target these components two broad classes of enzymes were investigated:

1. amylases targeting the β1,4 glycosidic linkages found in cellulose, and

2. proteases targeting the peptide linkages within proteins.

It was hoped that targeting persistent adhesions with enzymes would enable unwinding without damage.

Since both components are found in the original film formulation, there is a risk involved in using enzymes to target the adhesive layer; however, it was felt that this avenue of research might discover a viable treatment on the proviso that a tightly controlled application method was developed.

Unfortunately, neither family of enzymes gave encouraging results within a reasonable time frame, and given the difficulty in controlling the application, and risks of enzymes nonspecifically attacking other components of the film, this appeared to be a dead end. Of the two enzymes, proteases showed the most promise, although this was also felt to be the riskiest, with severe damage to the emulsion possible.

Conclusions

The balance of the charges on the gelatin chain is believed to be the key to understanding blocking.

While gelatin is at the core of each blocking mechanism, the treatment for each targets a different aspect of gelatin behaviour. Persistent adhesions may occur with all mechanisms of blocking, and these are most probably related to additional cross-linking and subsequent hardening as gelatin ages.

MICK NEWNHAM is currently the Manager of Conservation and Research, at the National Film and Sound Archive of Australia (NFSA) based in Canberra, Australia. For the past twenty years, Mick has been providing consultancies and training in audiovisual collection management and preservation on behalf of the NFSA and organisations such as UNESCO, SEAPAVAA, ASEAN and ICCROM for collections across the world. Mick is a lecturer with the Charles Sturt University’s online audiovisual preservation course and a Visiting Fellow at the University of Melbourne, Grimwade Centre for Cultural Materials Conservation.

References

Atkins, P. W. (1986). Physical Chemistry, 3rd Ed. Oxford University Press. pp. 10.

Bigi, A., Panzavolta, S., and Rubini, K. (2004). Relationship between triple-helix content and mechanical properties of gelatin films. Biomaterials. Nov. 25(25). pp. 5675.

Bigourdan, J. L., and Reilly, J. M. (1997). Environment and enclosures in film preservation. Final report to the Office of Preservation. National Endowment for the Humanities, Grant #PS 20802–94. pp. 15.

Kolb, F. J., Lovick, R. C., Peer, J. R., and Weigel, E. M. (1961). Precision striping of 8mm magnetic film. Journal of the SMPTE. Vol. 70. pp. 611–617.

Kowaliski, P. (1972). Applied Photographic Theory. J. Wiley and Sons.

McCormick-Goodhart, M. (1996). The allowable temperature and relative humidity range for the safe use and storage of photographic materials. Journal of American Archivists. Vol. 17(1).

Ram, A. T. (1990). Archival preservation of photographic films—A perspective. Polymer Degradation and Stability. Vol. 29. pp. 24.

Sheppard, S. E., Houck, R. C., and Dittmar, C. J. (1942). The sorption of soluble dyes by gelatin. J. Phys Chem. 46. pp. 158–176.

Somasundaran, P. (2006). Encyclopedia of surface and colloid science, 2nd Ed. Vol. 4. CRC Press.