Plastics in Australian Archives: An Industry Survey Regarding Prevalence, Condition, and Preservation Strategies

ABSTRACT In recent decades, there has been an increased focus on the identification, storage, and treatment of polymer-based plastic materials in cultural heritage collections. While the need for the preservation of plastics in audio-visual collections is well established, there is evidence that unstable plastics are also associated with paper-based collections. This paper investigates whether libraries and archives need to direct preservation resources towards risks posed by plastics in paper-based collections. As there is no available documentation on the types and condition of associated plastics, Australian archives were surveyed using an online questionnaire. Respondents reported associated plastics in over 90% of archives, and ∼50% observed these to be in poor condition. It is concluded that preservation resources that are specific to vulnerable plastic materials in Australian paper-based collections need to be developed and communicated.


Plastics preservation in archives
Increasingly, the challenge of preserving plastics is becoming more pressing to the galleries, libraries, archives, and museums (GLAM) sector. Plastics, referring to semi-or fully synthetic mouldable polymeric materials, can display rapid and irreversible changes, requiring appropriate care to mitigate the loss of cultural value. Certain plastics are consistently found to be highly unstable and are referred to as 'malignant plastics': cellulose nitrate (CN), cellulose acetate (CA), plasticised poly(vinyl chloride) (PVC), polyurethane (PUR), and natural rubber (Williams 2002;Shashoua 2008). Research addressing plastics identification, storage, and treatment techniques include: the Preservation Of Plastic ARTefacts in museum collections (POPART) project in 2008-2012 (Lavédrine, Fournier, and Martin 2012); the biennial Future Talks conferences held since 2009 (e.g. Bechthold 2011); and ongoing projects Preservation of Plastics (Getty Conservation Institute 2017), COMPLEX (University College London 2021), NEMOSINE (NEMOSINE 2021), and PolyMuse (PolyMuse 2018). Such projects demonstrate the growing interest in addressing the many challenges encountered in plastics preservation.
Efforts since the 1990s have largely focused on modern materials found in museums and galleries. However, archives also have a strong record of preserving plastic heritage collections through industry-wide movements. Since the 1980s, archives have advocated for the mass and rapid transition of photographic film media to more stable forms, first in response to the short lifespan of CN, and then of CA (Houston 1994;Lavédrine, Gandolfo, and Monod 2003). Magnetic tapes with degrading CA bases became an additional concern by the early 2000s, presenting a similar need for data transfer (Hess 2008). Specialised storage techniques such as cold storage, acid scavenging, humidity control, and/or ventilation are used to prolong the lifespans of at-risk audio-visual materials, extending the window of opportunity for data transfer before inevitable degradation (Lavédrine, Gandolfo, and Monod 2003;Hess 2008). Archives have an impressive history of collective responses to plastics degradation, utilising storage and media transfer techniques to extend the usable lifespan of plastic-based collections.

Potential plastics problem in paper collections
The vulnerability of analogue audio-visual formats is a clear motivation for prioritising the preservation of these materials. In contrast, paper collections in archives may appear relatively stable, as the lifespan of cellulose-based paper is generally longer than that of CN and CA (Lavédrine, Gandolfo, and Monod 2003, 16-21;Ahmad et al. 2020;Duran-Casablancas et al. 2021). However, paper-based materials can be associated with plastic materials at any point from manufacture to after acquisition, creating collection settings where paper and plastic are in close contact. Examples of known plastics in paper-based formats are scrapbooks, photographic albums, and laminated documents. For these materials, the recommended preservation approach is to remove or store separately the binding or lamination plastic if it may cause harm to the enclosed information, unless the intact form has historical significance or is too unstable for separation (Teper 2008;Woodward 2017). In the latter case, one recommended approach is to copy or digitise the information to minimise handling of the original.
Growing evidence suggests paper-based collections may contain additional plastic materials of concern, that are not currently addressed in collections care. The most obvious problem occurs when unstable plastics are not recognised in paper-based collections and are stored together for extended periods. In recent literature, there are at least three cases documenting the discovery of severe degradation in historical book covers or sleeves, affecting the associated paperbased materials (Campbell, Gerson, and Mosier 2013;Gatenby and Ritchie 2014;Whitehead 2017). In these cases, the plastics were only subsequently identified to be known unstable plastics: CN, CA, and PVC. Although these plastics have known vulnerabilities, there does not appear to be a practice of identifying malignant plastics in paper-based collections. It is unclear how much unstable plastic in the form of historical book covers and sleeves are contained within collections.
The second problem is more complex, as it stems from plastics which are currently considered stable for long-term use. Cross-infection occurs when the degradation of one material accelerates the degradation of another material and is the reason malignant plastics are considered a high risk. An example of this is when CN off-gases nitric acid as a degradation product, catalysing corrosion in neighbouring metals (Shashoua 2008, 180). For cellulose materials, crossinfection is found with the malignant plastics CN and CA (Curran et al. 2014). In the same study, cross-infection is additionally found with certain samples of poly (ethylene terephthalate) (PET), polyethylene (PE), polypropylene (PP), and polystyrene (PS). However, a different study on paper storage materials determined that the cross-infection effect with PET pockets is acceptably low, and that high-density PE crates beneficially produce a preserving effect (Lankester and Thickett 2021). As there is yet to be a consensus on the extent of cross-infection between paper and stabler plastics, further investigation is required. Cross-infection is also possible in the opposite direction, in which paper accelerates the degradation of plastic. Relatively stable plastics found in audio-visual materials such as polycarbonate (PC) and poly(methyl methacrylate) (PMMA) are vulnerable to vapourised acids and solvents, emitted by degrading paper and paper treated with biocides historically used in archives and libraries (Fenn 2003). Thus, plastics are both at risk and a potential risk when stored with paper-based materials.
In short, plastics which are incompatible with paperbased collections are potentially in need of preservation attention. As paper materials are common holdings of GLAM institutions, particularly libraries and archives, it is possible that institutions may need to direct attention to plastics in both audio-visual and paper-based collections.

Australian context
In Australia, it is unclear if plastics in paper-based settings warrant a wider preservation response. Documentation on the types of plastics collected and used in Australian paper-based archives and libraries is unavailable, making it difficult to assess the need for additional preservation actions. International literature addressing plastics in archive and library settings beyond audio-visual collections has been published by experts working in Canada, the United Kingdom, and the United States (Calmes 1993;Keneghan 2002;Fenn 2003) (see Table 1). Although there is a consensus that plastics require appropriate preservation, there is no data quantifying the proportion of paper-based collections affected. Determining the pervasiveness of plastics in paper-based collections assists in the formation of Australian collection policies and may provide an additional perspective to ongoing international discourse.
As part of the Australia-based PolyMuse project, it was of interest to determine the need for changes to plastics preservation in Australian paper-based collections. As an initial step, a survey was designed to capture knowledge present in the industry but not documented in the literature. Recent international surveys addressing other material types have been used to show current conservation methods (Teper and Straw 2011), to compare actual practice to theory (Alexopoulou and Zervos 2016), and to identify areas in need of more research (Soleymani, Ireland, and Table 1. Common plastic polymers found in archives (Calmes 1993).
McNevin 2015). In Australia, national industry surveys have been used to identify collection storage environments (Pagliarino 2018), and attitudes towards timebased media conservation (Sherring 2020). There have been no surveys on the topic of plastics in archives, libraries, or paper-based collections either internationally or within Australia. This paper describes the design, results, and implications of a survey with the aim of assessing the degradation and preservation of plastics in paper-based archives in Australia.

Participants
Archives were contacted using the 'Directory of Archives in Australia' (Rowell 2018), henceforth referred to as the 'Directory'. At the time of this survey in 2019, the Directory included 475 entries of Australian archives, listing institution names, contact details, and a unique identifier. To obtain a representative sample of the population and enable statistical testing, contacts were selected via simple random sampling. Numbers were generated by using RANDO-M.ORG (1998) and matched with identifiers in the Directory. Eligible entries with complete contact details were added to a contact list of 100 entries for the first round of recruitment, and three sets of 40 entries for subsequent rounds, in case additional recruitment was required to reach a higher number of responses.

Questionnaire design
Questions were written to cover four topics relevant to the aim of the survey (see Supplementary Material for full questionnaire): (1) Participant information: respondent details such as role, institution size, collection format types, and climate.
(2) Collection plastic types: types of plastic-containing objects and observed degradation. (3) Current strategies: preservation techniques currently used by the institution, such as cold storage and ventilation. (4) Useful resources: perceived usefulness of potential plastics preservation resources, such as digital databases and workshops.
A recent Australian survey (Pagliarino 2018) was consulted to aid the formatting of questions in section (1) Participant information. Additional care was taken to write all questions in a style suitable for an audience possessing a basic familiarity with collection terminology but no specialist training. For example, when describing degradation, conservation jargon was avoided in favour of everyday terms like 'colour change'.

Classifying plastic materials in archives
For section (2) Collection plastic types, it was important to group plastic objects into categories easily understood by the target audience. Most approaches to plastics conservation categorise plastics by polymer type (Quye and Williamson 1999;Shashoua 2008;Waentig 2008), providing an indication of expected degradation pathways. However, the identification of plastic polymers is a complex process, often aided by specialised training and/or analytical instruments which are not widely available to all collection settings. Collections are more often organised by physical format or historical context to aid discovery and access. For the purposes of this survey, it was more sensible to ask participants to identify plastics by physical format rather than polymer type.
Building on previous works which classified plastics in archives according to physical format (Calmes 1993;Keneghan 2002), collection plastics were divided into nine format subcategories (see Table 2), with the following alterations: . The category 'organisational aids' used by Keneghan (2002) was renamed associated materials, to include binding, protective, and organisational materials commonly found to be associated with paper-based materials that are not the primary support for information (e.g. synthetic book cloths and bindings, lamination sheets, storage boxes). . The subcategories visual and electronic information carriers were added under information carriers to accommodate more recent technologies such as synthetic papers and electronic solid-state storage.

Distribution
Participants were contacted in two rounds of recruitment using an open-source survey distribution platform called LimeSurvey. Each participant was asked to complete one questionnaire per institution. In the first round, the first 100 entries on the randomised contact list were contacted. After fewer than 20 responses were received after two weeks, a second round of recruitment of 40 entries were contacted. Additionally, four contacts recommended by questionnaire respondents were contacted. After eight weeks, 46 responses were obtained across the recruitment rounds, representing 10% of the Directory and giving a response rate of 32%. This response rate was considered acceptable, as it is close to the average response rate of 35.7% for organisational research (Baruch and Holtom 2008). Thus, recruitment was ended and the remaining contacts on the contact list were not contacted.

Data analysis
Responses were downloaded from LimeSurvey 3.0 in CSV format. Descriptive summaries and statistical tests were performed with IBM SPSS Statistics 26, a social science statistics programme. A Fisher's exact test was chosen for statistical testing, when determining whether there was an association between two categorical variables (McDonald 2014, 77-85). Graph visualisations were created with the open-source data visualisation library Seaborn 0.11.1.

Results
Data collection was completed over eight weeks from February to April 2019. Out of the 144 participants contacted, 43 full responses and 3 partial responses were obtained (results reported in parentheses are the percentage followed by the number of responses).

Institution type
Respondents reported the approximate number of staff members in their institution, defined as individuals who were financially compensated for ongoing full-or part-time work, selected from a list of predefined ranges. The largest group of respondents came from institutions with 100 or more staff members (39%, 18), followed by 1-4 staff members (30%, 14), and 5-9 staff members (11%, 5). Three institutions (7%) were fully volunteer run (no staff members). Overall, most respondents represent either large institutions with over 100 staff members, or small institutions with 1-4 staff members (see SM Figure 2). Respondents reported if their institution was structured as an independent archive, or as part of a larger organisation like a school or library. The majority (84%, 38) reported their institutions to be part of a larger organisation, rather than having archiving as the primary function of their institution (16%, 7).
Respondents selected the climate of their institution location according to a Köppen-Geiger climate classification map of Australia (Peel, Finlayson, and McMahon 2007). Almost all institutions represented (96%, 44) were located in temperate climate zones of Australia, with the remaining 2 respondents (4%) coming from dry climates. No responses were recorded from tropical climates.

Collection type
Next, respondents indicated which collection object types most accurately represented their institution's collection compositions: (a) paper, (b) analogue audio-visual materials such as photographic film, (c) digital audio-visual materials such as optical discs, (d) a mixture of the three types, or (e) another type of composition. Almost half of the institutions (48%, 22) reported housing primarily paper-based records, and an equal proportion (48%, 22) housed a mixture of paper-based, analogue audio-visual, and digital audio-visual records. No institutions reported Table 2. A proposed classification of plastic physical formats for libraries and archives, drawing on Calmes (1993) and Keneghan (2002 primarily digital or analogue audio-visual records. Two respondents (4%) reported other collection compositions, which were described as 'paper, analogue, digital and three-dimensional objects' and 'paper and organics'. Paper-based collections were represented in all responses, with about half of the archives additionally collecting analogue and digital audio-visual material.
To test a hypothesis that there is a difference between the formats collected by mixed collections versus primarily paper-collecting institutions, a Fisher's exact test of independence was performed for each physical format (see SM Table 1). Two responses reporting 'other' collection compositions were excluded from the analysis due to the very small sample size. Based on responses, there is no significant difference between mixed and paper collections in the collection of any plastic format type (p < 0.05, Fisher's exact test). Therefore, overall collection composition is not associated with whether an institution collects any particular plastic physical format.

Degradation observed
Condition by physical format Next, respondents reported the observed preservation condition of plastic objects held in their collections. Using conditional logic, only the object types reported in the previous question were shown.
Respondents were asked to indicate whether each format had ever been observed to be in poor condition, which was defined in the questionnaire as 'requiring conservation attention'. As both the need for conservation and poor condition is subjectively identified, the inherently subjective nature of this question was an accepted limitation. Of the nine physical formats, the ones most commonly observed to be in poor condition were photographic bases (56%, 24 of 43) and document organisers (56%, 24 of 43), followed by book bindings (51%, 23 of 45), document protectors (48%, 21 of 44), and magnetic tapes (42%, 18 of 43). Least observed in poor condition and reported by fewer than 25% of respondents were mechanical sound carriers, optical discs, and digital storage devices (see Figure 2).

Degradation changes observed
Respondents were asked to select the types of degradation change observed in the formats in poor condition. Types of degradation reported varied by physical format (see Figure 3). For the five types of materials most reported to be in poor condition in the previous question, the following degradation changes were observed (presented in decreasing number of responses): . Photographic carriers (24 reporting): Odour change (49%, 21), colour (35%, 15), shape (30%, 13), hardness (30%, 12), structure (26%, 11), and texture . Plastic physical formats observed to be in poor condition. Associated materials were reported to be in poor condition more frequently than all information carrier types except photographic. (23%, 10). These changes are consistent with the symptoms of 'vinegar syndrome'. . Document organisers (24 reporting): Colour (67%, 16), structure (54%, 13), shape (42%, 10), and hardness (38%, 9). Compared to photographic bases, only a small number (8%, 2) reported odour changes in organisers, but colour and texture changes were comparable. . Book bindings (23 reporting): Hardness (57%, 13), colour (52%, 12), structure (52%, 12), and shape (30%, 7). Like organisers, the frequency in odour change was low (4%, 1). . Document protectors (21 reporting In the free response section, five respondents additionally noted data corruption for magnetic, optical, and digital storage formats. In two responses, the stickiness of plastic book bindings and document organisers were described to have resulted in adhesion with paper-based records. It is noted that there were five or fewer responses for mechanical, optical, and electronic carriers.

Malignant polymers
Respondents were asked if their collections contained malignant polymer types, due to the known instability of these plastics. Since polymer types are not easily identified, respondents were asked to respond to the best of their knowledge. Options for indicating being 'unsure or unknown' and having 'none' of the above polymers were included to distinguish between null responses.

Materials testing
Participants reported whether they conducted any type of material compatibility or safety testing on new materials used for the storage and display of collections. A large majority of respondents (80%, 32) never conduct testing on new materials. 15% (6) reported sometimes conducting testing, and 5% (2) always conduct testing. Six participants did not respond to this question. Thus, testing of new materials is likely not commonly employed by archives.
Preservation strategies for non-photographic plastics Participants identified whether they used any of a list of eight preservation strategies to preserve plastics in poor condition. Strategies from photographic and plastics conservation were drawn from recommendations by Lavédrine, Gandolfo, and Monod (2003) and Shashoua (2008) respectively. A distinction between photographic film and other plastic materials was made as existing strategies have placed an emphasis on photographic collections.
Options provided for non-photographic film were: cold or cool storage, desiccants, ventilation, separation of plastics by polymer type, monitoring techniques, replacement of aged storage materials with new ones, removal of the plastic component, digitisation, and none. By a margin of about 30% (see Figure 5), the most commonly reported preservation strategies for non-photographic film collections were: the replacement of aged plastic storage materials with new ones (78%, 36), digitisation (70%, 32), and removing aged plastic components (63%, 29). Remaining techniques, used by 33% or less of institutions, included cold or cool storage, reported by only 15 respondents (33%). Five respondents (11%) reported that no collection care strategies were used at all for plastics in poor condition.
When a test was performed to determine if there was a difference in preservation techniques employed by paper versus mixed collections (see SM Table 2), no significant difference was found (p < 0.05, Fisher's exact test). Current results indicate that the employment of preservation techniques for non-photographic film plastics is not associated with whether a collection is primarily paper-based or mixed.
Preservation strategies for photographic film Using conditional logic, the next three questions were only shown to respondents who indicated collecting photographic carriers earlier in the survey.
Respondents were asked to indicate which collection care strategies were employed for film. Strategies provided were the same as for non-photographic objects in the previous question, except that choices for removal of aged plastics components and replacement of aged plastic storage materials with new ones, were changed to the use of plastic storage materials for film and the duplication of film materials. Strategies chosen (see Figure 6) were digitisation 34 (79%), cold or cool storage 14 (33%), and no strategies 4 (9%).
When a Fisher's exact test was performed to compare the techniques employed by paper and mixed collections, a significant difference was found   for only one technique (see Table 3). Mixed collections were more likely to employ duplication methods than paper collections (p = 0.046, Fisher's exact test). No significant difference was found for the remaining techniques (p < 0.05, Fisher's exact test).
Institutions that reported collecting photographic film were asked to report the temperature, humidity, and respective fluctuations of their film storage environments. Institutions were able to report up to two storage environments. Institutions reported that films were stored at an average temperature of 19.3 ± 8.6°C (mean value ± standard deviation) (37 responses from 29 respondents, 14 missing), and at a relative humidity of 45.8 ± 9.8% (37 responses from 29 respondents, 14 missing). On average, the reported temperature fluctuation was 3.9 ± 2.6°C (30 responses from 24 respondents, 19 missing) and the reported humidity fluctuation was 8.2 ± 6.0% (30 responses from 24 respondents, 19 missing). As seen from a distribution plot (see Figure 7), only two institutions stored film at 0-4°C (cold storage), and only four between 4-12°C (cool storage), as defined by the Image Permanence Institute (IPI) (Adelstein 2009). This number is much smaller than the number of respondents who reported implementing cold or cool storage for film in a previous question.
Respondents that reported their institution not using cold or cool storage for film were asked to select from a list of reasons preventing implementation. In order of frequency, reasons selected were high financial cost (66%, 19), inappropriate infrastructure or space (41%, 12), a small or insignificant collection size (41%, 12), lacking manpower for preparation (31%, 9), no knowledge or familiarity with the technique (21%, 6) and no time (21%, 6). No respondents selected the reasons 'access', 'achieved with other means', or 'not required'. In a free-response box, three respondents stated that the cold storage of the film collection was not considered an institutional priority.

Potential resources
Participants scored three resources for their potential usefulness at their institution: a digital reference database of plastic materials, a workshop for plastics identification, and standards for plastics preservation. More than half of respondents gave all three resources a score of 4 or 5, where 5 indicated the highest usefulness (see Figure 8, combined responses with next question). On average, the usefulness of resources scored in decreasing order were standards for plastics preservation (3.95), workshop for plastics identification (3.85), and a digital reference database of plastic materials (3.82).
Next, participants scored five areas of plastics knowledge on how much they were needed at their institution: identification techniques, condition analysis techniques, degradation patterns, treatment techniques, and storage techniques. Again, more than half of respondents scored all five areas a 4 or 5 (see Figure 8). The mean average of the top three techniques in decreasing order are storage techniques (3.85), degradation patterns (3.82), and condition analysis techniques (3.78).

Implications of respondent demographics
All respondents confirmed that their institutions have paper-based collections. Approximately half of institutions housed primarily paper-based collections, while the other half collected a mixture of paper, analogue, and digital audio-visual collections. Thus, it is likely that current results are representative of paperbased collections in Australian archives.  As expected for a survey targeting archive institutions, most reported respondent roles were archivists (70%), followed by librarians, conservators, and other roles. As no preferred respondent was specified during the distribution of the survey, these results are not an indication of employment rates. However, as only three responses from conservators were received, this result indicates that a conservation strategy targeted to the archive industry must be useful to professionals from a wide range of specialties. This is consistent with the reported historical trend of preservation tasks in archives and libraries being performed by in-house cultural heritage professionals such as archivists and librarians, with only a minority able to employ professional conservators (Cunha and Cunha 1983, 97-110;Harvey and Mahard 2014, 1-9).
Diversity of responding institutions was observed in the range of staff numbers reported, indicating that both large, well-staffed institutions through to small, volunteer-run institutions are represented. Most archives were part of a larger organisation (84%) rather than functioning as independent archiving organisations, indicating that a range of institutional agendas must be accommodated. An industry-wide strategy must assume a wide range of collection settings.
Most responding archives were located in temperate climates (96%), a reflection of a higher density of institutions in the temperate metropolitan zones of Sydney, Melbourne, Brisbane, and Perth listed in the Directory. However, tropical and dry regions cover the majority of Australia (Peel, Finlayson, andMcMahon 2007, 1642). There is a need to obtain more data from institutions outside of metropolitan zones and temperate regions to determine if there are differences in indoor climate control practices (e.g. artificial HVAC systems or passive building envelopes) and the subsequent effect on plastics in collections.

Implications of format and plastic types
Responses indicated the potential pervasiveness of plastic materials across all types of Australian archives. Photographic, magnetic, and optical carriers were reported by over 90% of institutions. Similarly, over 90% of institutions collected the associated material Figure 8. Combined results from usefulness of potential resources for plastics management and need for areas of plastics conservation knowledge (n = 38), in decreasing order of average score for each resource or knowledge area. 1 = not at all useful/ needed, 5 = very useful/needed. types of book bindings, document protectors, and organisers (see Figure 1). Crucially, no statistical difference was found in the types of plastics reported by archives holding mainly paper-based records or mixed collections.
Results indicate that plastic degradation is potentially pervasive in paper-based collections. As expected from existing knowledge of at-risk media types, photographic film and magnetic tapes were reported by the highest proportion of respondents when identifying information carriers requiring conservation attention, at 56% and 43% respectively. Perhaps less expected was that similar numbers of respondents reported observing degradation in associated materials (see Figure 2). In other words, more collections reported poor condition in associated materials than for all information carrier types, with the exception of photographic materials. Note that the survey results do not indicate the proportion of the collections in poor condition, but rather the proportion of respondents collecting the respective materials who have ever observed the relevant materials in poor condition. As noted in the POPART surveys of three museums in the UK and Europe, the proportion of objects in poor condition varies depending on collection type and historical period (Lavédrine, Fournier, and Martin 2012, 136-138). Based on the frequency of reported plastic associated materials in poor condition, further examination of the collection contexts, manufacture, and age of plastic associated materials may reveal currently unmitigated factors contributing to degradation, including cross-infection.
Types of reported degradation varied between physical formats (see Table 4), falling into three categories when compared with existing literature: (1) In some formats, the reported changes directly matched degradation signs described in existing literature. Reported changes for photographic carriers and magnetic carriers were consistent with CA film displaying 'vinegar syndrome' and PUR tapes displaying 'soft binder syndrome' respectively (Hess 2008;Adelstein 2009). For document protectors, the reported changes reflect literature on historically introduced materials made with known malignant materials such as in CA lamination (Woodward 2017) and CN sleeves (Whitehead 2017).
(2) Some formats had degradation changes described in the literature that were not reported by participants. Visual carriers showing hardness changes and optical carriers showing shape changes were not reported, although the remaining changes for these formats were noted. As these changes represent more advanced degradation stages, namely the embrittlement of resin-coated papers and warping of optical discs (Wagner 1999;Iraci 2005), it follows that these symptoms are less commonly observed. (3) Responses for some formats appeared to describe degradation changes not documented in the literature. However, as the number of responses of this nature are low, these are unlikely to be areas of high concern: (a) In electronic carriers, structure and hardness changes were reported by respondents, while no information on physical changes could be found in the literature. This is likely explained by the higher risk of data corruption and technological obsolescence compared to physical degradation risks in flash formats (Leggett 2014). A high risk to data integrity in electronic carriers is supported by several respondents reporting data loss in the free response section. Careful handling may be a sufficient response to prevent the cracking of digital carriers reported by a minority of respondents. (b) Optical carriers were reported to have texture changes in addition to structure changes. Texture changes are not documented in the literature. It is unclear if this is an undocumented change or a misreporting of scratches or delamination. (c) Most of the noted changes in bindings and organisers match with the degradation of plasticised PVC in these formats (Teper 2008;Tétreault 2017). However, odour changes were also reported in these formats by a small number of respondents, in contrast to known degradation signs. It is unclear if the respondents were reporting the noticeable odour of PVC, or degradation changes in PVC or other plastics. Further investigation of these associated plastics may reveal new information. Overall, there is a close match between the industry knowledge and existing literature on plastics degradation, with minor differences reported by a small number of respondents regarding bindings, organisers, optical carriers, and digital carriers.
Respondents recognised the presence of malignant polymers in collections, with most institutions reporting collecting CA, PVC, PUR, and rubber. As these polymers are present in many physical formats, this result suggests that general knowledge of plastic types is common in archives. Only eight respondents were unable to report polymer types in collections due to being unknown or unsure. Results on malignant plastics correlate with the high percentage of collected physical formats reported in the previous questions.

Implications of current strategies
In combination with findings from the previous sections, there appears to be three areas in need of greater attention regarding strategies for the preservation of plastics in archives: (1) Resources for storage (2) Expertise and training (3) Preservation standards Increasing resources for storage Considering the potential pervasiveness of plastics degradation in archives, the first area in need of attention is to direct more resources to physical storage. Respondents reported a high need for storage techniques, rating it the highest of five provided knowledge areas (3.85 out of 5). Associated plastics, commonly used in the storage of other collection materials, were observed to be in poor condition by many respondents. Additionally, in comparison to techniques recommended by the literature (see Table 5), many archives use fewer techniques applied in storage settings, such as monitoring, ventilation, and the use of desiccants. Only 33% of institutions reported using cold or cool storage for film and nonfilm plastics in poor condition (15 and 14 respondents respectively), and only six of the storage temperatures reported for photographic film qualify as cold or cool storage environments as defined by the IPI (Adelstein 2009). It is unclear if the disparity in reported cold or cool storage usage versus storage temperatures is due to a misunderstanding of the term, or due to respondents utilising cold storage at other institutions. Regardless, there is a general trend towards implementing data preservation rather than physical preservation in the surveyed archives. Directing greater attention to storage techniques would address multiple concerns raised in this survey.
A lower emphasis on physical preservation in current practice is likely a reflection of the impracticality of implementing appropriate storage environments for all collections. Respondents in this survey reported that high financial cost was the most common barrier to the implementation of cold or cool storage of film. The problem of limited resources for preservation work has long been acknowledged, particularly as the archive sector in Australia has suffered chronic funding cuts (Arrow 2021). Since the already limited resources are prioritised for the digitisation of at-risk audio-visual materials (National Film and Sound Archive of Australia c. 2018), the impact of resource limitation is a crucial factor in the discussion of any recommendations in archives. This financial resource gap can be addressed through advocacy to government, upper management levels, and public stakeholders regarding the importance of storage and physical preservation in archives. Additionally, there is the need for close collaboration between conservators, archivists, and administration to gain institutional and industry support.

Increasing expertise and training
Responding archives show a need for preservation training and guidelines from external sources. Respondents rated all areas of plastics knowledge listed in the survey as highly needed. Additionally, as materials safety and compatibility tests are reportedly never conducted by 80% of respondents, it is likely that existing resources (e.g. Pasiuk 2004; Tétreault 2017; National Archives of Australia c. 2018) are commonly used to inform decisions regarding safe storage materials. Both existing and potential instructional information on appropriate preservation techniques are useful for responding archives.
Greater attention to the distribution of plasticsfocused knowledge and resources is the second way of meeting current industry needs. As seen from the literature included in Tables 4 and 5, much of plastics preservation research assumes a background in conservation or conservation science. The few resources that consolidate plastics guidelines for an archive setting (e.g. Calmes 1993;Keneghan 2002) are unlikely to be available to the general public due to the publication format. Targeted information distribution for non-conservators with websites, fact sheets, videos, etc., would increase the accessibility of existing plastics storage recommendations. Additional training and education may be used to increase the number of archive workers equipped to care for plastics, such as through the provision of training workshops for relevant roles. Instructional materials suitable for archives would benefit the preservation of both audio-visual collections and paper-based collections with associated plastics.

Establishing standards
The third area in need of attention is the development of plastics preservation standards. Respondents rated the existence of plastic preservation standards as the resource that would be most useful to their work (3.95 out of 5). Current standards like the National Archives of Australia standard for the storage of non-digital archival records (National Archives of Australia 2018) and ISO 11799 (International Organization of Standardization 2015) address high-risk information carriers such as magnetic tapes and photographic film, but not other types of plastic materials. In current standards, advice on associated plastics is limited to avoiding the use of polymers with known instability. Standardisation of practices for degrading plastics would aid the industry in targeting attention and resources. This paper is unable to comprehensively cover recommendations for standards development, which would be better discussed in another article. However, two areas stand out as potential avenues for standardisation. Firstly, systematic identification and monitoring of plastics would benefit many collections, decreasing the potential for unexpected degradation and ensuing problems. The presence of malignant plastics is already recognised by respondents and should be coupled by regular observation of changes. Secondly, the use of plastic associated materials, even if composed of polymers that appear stable, should be re-evaluated. Considering the stilldeveloping research on cross-infection between paper and plastic, more investigation is necessary before determining if the benefits provided by plastic storage materials outweigh the risks. As the industry understanding of plastics evolves, so should its guidelines and standards for care and preservation.

Future studies
Plastics in paper-based collections have not previously been studied through systematic surveys. Future studies are recommended to increase understanding of this collection type. Firstly, more details regarding common plastic storage materials used in paper collections are needed. Ongoing research is identifying the age, manufacturers, and composition Table 5. Preservation strategies from literature presented by format (✓: indicated by literature, ◯: non-applicable for the format). of associated plastic materials through collection surveys. Secondly, the development of standards for the storage of plastics in paper-based settings is needed. Identification of plastic types and degradation patterns specific to associated materials must particularly be examined. Protocols for planned loss and guidance on expected lifetimes must be developed as part of preservation recommendations. Lastly, any research must be followed by the effective communication of results to the GLAM industry, the provision of plastics-specific training for archive professionals, and/or an increase in industry access to plastics conservation experts, to ensure the dissemination of information.

Conclusion
Plastics are a point of vulnerability in Australian paperbased archives. Australian archives were surveyed with an online questionnaire to determine the types of plastics in collections, conservation condition, management strategies set in place for plastics in poor condition, and desired resources for plastics management. Results indicate that plastics are present irrespective of whether they are part of archives that are primarily paper-collecting or containing a mixture of audio-visual and paper materials. Degradation changes observed in plastics were closely matched with known trends in the literature. A large proportion of archives reported observing the poor condition of associated plastics such as book bindings and document sleeves in collections, in addition to plastic information carriers such as film and tape. Based on survey results, there is a need for: (1) increased allocation of resources towards storage strategies for plastics in collections, (2) increased allocation of resources and training towards industry knowledge of plastics preservation, and (3) the development of preservation standards for plastic materials in archives. While archive professionals display an awareness of the types of plastics in their collections, they are not always able to appropriately preserve them. Greater financial, educational, research, and institutional support are key factors for improving the preservation of plastic collections.
The continued presence of degrading plastics in archive collections without mitigation places entire archive collections at risk. Without appropriate intervention, future generations may witness the loss of not only the plastic chapter of human history, but also of collections saved alongside these materials.

Ethics
Approval for conducting human research was granted by the University of Melbourne's Faculty of Arts Human Ethics Advisory Group on 31 January 2019, under project ID number 1852382.2. Written consent for participation in the study was obtained from all participants.