In recent research at the Museo de Prehistoria of Valencia (Prehistory Museum of Valencia) and the Institut Valencià de Conservació i Restauració (Valencian Institute for Conservation and Restoration), we have developed different methods of reversible filling based on the use of synthetic films such as polyethylene terephthalate and polypropylene, materials commonly used in the field of document conservation. These methods have now been applied to a variety of archaeological glass collections. The results of this research project are set out in detail in this article.
Archaeological glass is an extremely delicate material that requires particular ability and care when treated by the conservator-restorer. Glass objects from archaeological origin are fragile, often very fragmented, and have very thin walls. Besides, they have undergone singular alteration processes when preserved in unsuitable environments (figure 1).
In this article we put forward new proposals that concern one of the most controversial processes carried out by conservator-restorers: the treatment of the missing areas. We have developed an alternative that, while being coherent with reversibility and minimal intervention criteria, a necessary prerequisite to any restoration intervention, does not prevent a reconstruction process that facilitates form legibility of the object.
Putting the concepts of reversibility and minimal intervention into practice: a question of criteria
Traditionally, the gap-filling has been understood as a process carried out in order to return form unity to a piece. The ICOM 2008 resolution defines it as a regular treatment in a restoration process that includes “all actions directly applied to a single and stable item aimed at facilitating its appreciation, understanding and use. These actions are only carried out when the item has lost part of its significance or function through past alteration or deterioration and are based on respect for the original material”1. But the dangers involved in that sort of direct action do not only arise from lack of manual ability and experience of conservator-restorers themselves, they can also be caused by the historical moment when actions take place: applied criteria, protocols, and materials used have varied with time.
If there is one thing we can currently learn from our recent past, that is the frequent mistakes made when, without awareness of negative effects, excessive intervention on cultural heritage objects is applied without absolute respect for the minimal intervention criteria. The damage done by professional conservator-restorer, when they justified excessive intervention to achieve a supposed improvement in the understanding of a piece, is precisely the reason why we should make the criteria of minimal intervention a priority, and consider it not only viable but also the alternative that is most coherent with strict respect for the original material conserved.
The Ministry of Culture, through the Instituto del Patrimonio Cultural de España (Cultural Heritage Institute of Spain) published ten criteria for restoration2. In relation to minimal intervention, the document says that “the principle of minimal intervention is crucial. Any manipulation of a piece involves risks, therefore we should limit ourselves to that which is strictly necessary and accept natural decay caused by time. Over interventionist treatments that can damage an object integrity should be rejected”. These recommendations also refer to the gap-filling process that according to the document should only take place “when it is necessary for the stability of the piece or for some of the materials that form part of it”. The controversial but indispensable reversibility criterion is also mentioned. Any report or publication regarding restoration must include it, even though its meaning can often create some misunderstandings . Products used for the fill-in process must be reversible but reversibility should not be a traumatic moment for the piece nor for the conservator-restorers themselves [2, pp. 60-61].
Nowadays, we have sufficient resources to make reversibility and minimal intervention criteria fit perfectly into the fill-in process. Acting with this in mind does not mean no intervention or that it is not possible to find alternatives that combine both respect for and legibility of the piece . We have tried to make both concepts compatible in our research.
Applied alternatives to the casting of missing areas in archaeological glass
Reconstruction of missing areas is often justified as a consolidation process of the piece. Its purpose is to improve the reading of the forms and the understanding of the piece as a historical document, where a gap is considered an interruption in the continuity of the form. Intervention might be necessary or advisable in some cases, especially when the stability of the piece is at stake. But we know that this is not always the case and that conservator-restorers are often subject to impositions or wrong criteria that find justification in considering that an incomplete piece cannot be understood or lacks aesthetic quality.
It is important to define certain areas before a process of conservation-restoration is carried out. We must know what the final destination of the piece is: storage, research, temporal or permanent exhibition. Once this has been established, a decision could be made regarding whether treatment should be preventive or if there is a need for a remedial approach. Other fundamental factors are the preservation state of the material and some of its characteristics, like glass thickness, and the size, shape and localization of gaps. In any case, there comes a moment when the professional will have to face the problem of a possible reconstruction of missing areas. What alternatives are there for that challenge?
Unanimity of criteria is hard to achieve, above all regarding the process that conditions the piece appearance when it is eventually presented. When we make a diagnosis for an object and establish the percentage that has been lost, we have to act responsibly and decide among different proposals that could be considered. The first thing to be ascertained is whether a casting of the gaps is really necessary. No intervention could, in fact, be a good option, especially in those cases where the piece can be easily read (figure 2).
From left to right:
Figure 1. Group of archaeological glass objects. Cycladic Museum (Athens, Greece).
Figure 2. A piece without casting of missing areas. Casa Romei (Ferrara, Italy).
Figure 3. New support to reconstruct the missing base of the glass. Casa Romei (Ferrara, Italy).
Figure 4. The original artifact rests on an internal support system. Corinto (Greece).
Figure 5. Support system made of moulding resin which bears original fragments. Museo de Valladolid (Spain).
We may also decide that only a partial intervention is necessary, with occasional fillings that strengthen strategic areas to give stability to the piece. Or we may opt for a no integration proposal and use other means of supporting the piece instead (figures 3-5). There are different types of supports used for glass made of synthetic resins  or with blown glass [5; 6, p. 160]. In some cases, instead of completing the object form, support is minimized and its presence is reduced to some elements that not only hold the piece but in some way help to imagine the area of the object that has been lost (figure 6-7).
This approach requires a radical change of attitude not only in the case of conservator-restorers, who are the first to be convinced of the many advantages of that decision, but also on the part of archeologists, museum directors and the general public who must learn to really value this alternative and be aware that nowadays it is possible to create 3D digital reconstructions that can be used as a complement for a better understanding of pieces, avoiding thus the need of acting directly on them.
From left to right:
Figure 6. External support to hold a glass artifact. Hadrian's Library (Athens, Greece).
Figure 7. An internal support with a recreation of the base made in the Institut Valencià de Conservació i Restauració (Valencia, Spain). Photography Pascual Mercé.
Figure 8. Casting of the missing glass made of plaster. Archaeological Museum of Haniá (Crete, Greece).
But among traditional options there is also the total intervention, the complete reconstruction of gaps in the object for conservation, aesthetic or exhibition related reasons. Many different products have been used in the fill-in process. For instance, glass from other objects reused for this purpose, materials traditionally used for ceramics, such as plaster (figure 8), acrylic resins (Technovit 4000, Plastogen G), polyester (GTS from Vosschemie, C-32 from Canuts), polyurethane (Crystal Clear 200) or epoxy resins (Ablebond 342-1, Fyne-bond, Araldite 2020, Hxtal NYL-1, Epotek 301) [6, pp. 153-159; 7, pp. 286-304; 8, pp. 76-95; 9-11] (figures 9-10). In recent years epoxy resins have been, without a doubt, the products most frequently used and research has focused on analyzing their long term aging [12, 13]. This process involves making models (generally silicones, modeling clays, dental waxes or clays) and further work on the resin finishing in contact with the piece (figures 11-12). Proposal for reconstruction of large gaps with resin by means of molds made from the piece have been occasionally put forward. A replica of the lost area is obtained, worked on and then adhered to the piece . S. Koob experimented with detachable fills as well: “[…] the making an intermediate fill or casting with plaster of Paris. The plaster will be removed from the object and then molded in silicone rubber, from which an epoxy fill or replacement fragment will then be made. This can be joined to the original object with B-72” [8, pp. 95-104].
From left to right:
Figure 9. Fill in resin in archaeological glass. Museu de Conimbriga (Portugal).
Figure 10. Yellowing process of filling resin. Museo Arqueológico de Santa Pola (Alicante, Spain).
Figure 11. Polishing coloured epoxy resin made at the Institut Valencià de Conservació i Restauració (Valencia, Spain).
Figure 12. A casting with coloured epoxy resin made at the Institut Valencià de Conservació i Restauració (Valencia, Spain).
But there are other less technical aspects that, unfortunately, are sometimes forgotten such as the high financial cost of many of those products, their short life, toxicity and questionable reversibility. In general, these products are not reversible; therefore we believe that the use of a primer between the original piece and the new material, in order to make removal easier, should be not just necessary but compulsory.
Besides, we should not forget the technical complexity of the fill-in process; mistakes can easily occur. The process requires extreme precision, not just for the preparation stages (making the mold) but also when pouring the resin and in the polishing that later takes place. These are all risky actions when they are performed on an archaeological object of extreme fragility.
Conscious of the problems involved in the application of these reconstruction methods, which use traditional materials, we are researching in our laboratories, a proposal that might solve the question of reconstruction and conform to reversibility and minimal technical difficulty requisites.
A reversible fill-in method: detachable films
The market offers a great deal of synthetic compounds made for industries whose activities differ a lot from those normally carried out in the area of cultural heritage. Conservators have gradually been getting materials that had originally been created for other purposes. Therefore, each new product incorporated to our resources requires a number of studies that make sure that is not damaging for the materials it might be in contact with, and to judge the suitability of new ideas.
The use of detachable films in this proposal is not new (figure 13), but it has not been sufficiently studied to be considered a generalized practice. In some areas of conservation, acrylic resin sheets have been made using products such as Paraloid B-72, Technovit 4004A [15, 16] or slow hardening epoxies (AY 103, Araldite 2020, Hxtal NYL-1, Epotek 301) [17-19]. These can be modeled while they are becoming hard, in order to give them the shape of the gap and then stick them to the piece as if they were fragments [7, pp. 304-306; 8, pp. 104-106].
There are publications that have mentioned the much less researched alternative of making detachable films with acrylic sheet precast (Perspex, Plexiglass). Some authors define these materials as less manageable than others and not very suitable for aesthetic reasons [6, p. 161; 7, p. 304; 8, p. 104; 20]. Although it is true that the use of detachable films has its limitations, we have investigated it as a proposal in relation to reversibility and minimal intervention requisites. We have used particular materials and methods and we outline the results we have obtained below.
There are many comparative studies for the gap-filling resins used for interventions on glass, for their virtues and qualities. But we cannot find contrasted analyses for different type of sheets, results, possibilities, or the limitations in their use, even in the cases when they are presented as an alternative. Therefore, we have based our selection of the laminated products that we have used on analyses focused on other applications and materials within the field of conservation.
Research on the causes for glass deterioration has found evidence of an acceleration process related to environment acidity [21, pp 79-80; 22]. Other objects of a different nature such as paper and metal, share that sensibility but with some differences concerning direct effects. We have found detachable films of various types that have been used, and analyzed, for application in those specializations for more than fifty years. In our proposal, we are putting forward incorporating some of the results obtained in analyses originally focused on, among other things, applications of lamination treatment for paper or for storage systems for metals, to the area of glass materials.
Some compounds, for instance, those derived from polyvinyl acetate (PVA), polyvinyl chloride (PVC) and cellulose acetate (CA), turned out to be suitable from the aesthetic and morphologic point of view, and for their malleability and the fact that they are easy to handle. But they have all been rejected because of the damage they can cause to glass surfaces. Plasticizing elements used in the manufacturing process are the main agents for acid vapor emissions that make them brittle and tacky [23, p. 15].
Eventually we considered two compounds as the most tested and verified as harmless for our work: polyethylene terephthalate (PET) and polypropylene (PP). PET has been used since the mid-20th century and has been the object of several studies related to the multiple applications it has had, due to its optimal results. In the conservation world, the use of PET is fundamentally associated with archival work and graphic document treatment. Acceptance of this product in these fields shows it is the most suitable amongst those we know. This is, to a great extent, due to the absence of plasticizing elements in its manufacturing process which avoids later emissions because “the semicrystalline nature is the basis for the excellent resistance to chemicals” [24, p. 68].
We have found that PET made in an uncoated, biaxially oriented, and polished form presents the best and more secure results within the different varieties. This is the case of Melinex, the product we use. Melinex shape and surface does not alter at least until 120 ºC and does not change its tackiness for forty days at 37’7º C [25, Tables 1-2].
PP responded to tasting in a less conclusive way than PET. Therefore, it is considered an acceptable compound to be used in conservation but that should be tested further in order to confirm the results. It is a chemically inert material and it is not toxic, but it is vulnerable to sun light effects.
PP is considered acceptable, provided that is manufactured, as in the case of PET, without plasticizing additives. PP is easy to manipulate and to work with and can have a hazy and matt finish, characteristics that in some cases turn out to be an advantage for gap-filling in archaeological glass that has partly lost transparency.
We have carried out tests with these two materials, both at the Institut Valencià de Conservació i Restauració and the Museo de Prehistoria of Valencia. The tests concern the application of the materials as sheets in the fill-in process of glass from archaeological origin. We have opted for PET (Melinex) in the case of an islamic lamp (figure 14). We have placed the sheet to support a group of fragments that were poorly supported by the base. PP (Plakene) was used for the loss compensation of a medieval chalice that had partly lost its transparency and that had several small gaps (figure 15).
From left to right:
Figure 13. A detachable film solution. British Museum (London, UK).
Figure 14. PET detachable film as a support in an islamic lamp. Institut Valencià de Conservació i Restauració (Valencia, Spain). Photography Pascual Mercé.
Figure 15. Chalice with PP detachable film. SIP Archive of the Museo de Prehistoria of Valencia (Spain).
Figure 16. Preparations to make detachable films. SIP Archive of the Museo de Prehistoria of Valencia (Spain).
The sheet is made without having any contact whatsoever with the piece. This avoids excessive manipulation. Besides, it means that we do not have to be over concerned with the possibility of mistakes and with the possibility of having to repeat the process. In the first place, we should choose the material best suited to the piece (figure 16), and then decide about the sheet thickness (figure 17). There are different options for each of the products: 75 to 175 microns for PET and 300 to 1,200 microns for PP. We have also made colour tests for these materials. It might be interesting to colour them slightly in order to harmonize the materials with the treated piece. The application of a mix of pigments and Paraloid B-72 in ethyl acetate has been successful and we can give the material an aesthetic finish closer to the original if we wish to do so. This might depend on characteristics of the piece and the differentiation criterion we choose to apply. But being a totally reversible system, we can eliminate the application with no complication if we decide to do so. In fact, the main advantages of these systems are the possibility of changing and eliminating sheets and the reduction to a minimum of the risks involved in changes.
From left to right:
Figure 17. Selection of the PP detachable film and its colour in connection with the original. SIP Archive of the Museo de Prehistoria of Valencia (Spain).
Figures 18. Adhesion of one PP detachable film once coloured and cut. SIP Archive of the Museo de Prehistoria of Valencia (Spain).
Figure 19. The scalpel indicates one of the missing areas made with PP detachable film. SIP Archive of the Museo de Prehistoria of Valencia (Spain).
The first step in the process of making a sheet for gap-filling is to obtain the gap profile. We place a thin acetate sheet on the original piece and faithfully mark on it the gap contour with a permanent marker. This acetate sheet will be used as a pattern to cut the PET or PP sheet later as accurately as possible; but mistakes can always be rectified. In order to adapt the material to the curve shape of a gap, we can heat the sheet by means of a hot air blower at low temperature. This will work provided that the curvature is not too pronounced.
Once we have checked that the fitting of the sheet is optimal, we only have to fix it in the gap, adhering the fragment to the glass walls with resin (figures 18-19). Following our reversible material criterion we chose to use as adhesives Paraloid B-72 or Mowital B60HH at 20 %; we have obtained excellent results with both of them.
Nowadays we have sufficient resources to make reversibility and minimal intervention criteria fit perfectly into the fill-in process. This can be done using materials that are harmless and stable in the long term. Acting with this in mind does not mean no intervention or that it is not possible to find alternatives that combine respect for and legibility of the piece. In our research into the use of PET and PP sheets for the integration of gaps in archaeological glass, we have tried to make the concepts mentioned above compatible, and we believe to have obtained satisfactory results.
However, in spite of the success in adapting flat sheets to small curvature of gaps, we are aware that it is there where the limitation of our proposals lies. PET and PP sheets modify their plasticity with the application of heat and can be slightly modelled to adjust their flat shape to a tridimensional object. But we cannot say that this is possible when the volumes to be reintegrated have more complicated shapes, or are rather large, angular or have pronounced curves. In those cases where sheets cannot follow the voluptuousness of the piece shape, we will have to put forward other alternatives for partial reintegration as a means of support or an adaptation of the resin sheet approach suitable for those shapes. Our system also finds limitations in objects or areas where glass is rather thick. We might not be able to get that thickness with this type of material unless we join several sheets.
However, we are committed to interventions that follow the line of research described above, whose main objective is to develop gap-filling systems which are easily reversible and that do not represent a risk for the conservation of the pieces.
We would particularly like to thank Carmen Pérez and Helena Bonet, directors of the Institut Valencià de Conservació i Restauració and the Museo de Prehistoria of Valencia, respectively, for their support for this research project and their understanding when it comes to value the criterion of respect for the original piece in the intervention on heritage.
1 Terminology to characterize the conservation of tangible cultural heritage, Resolution adopted by the ICOM-CC membership at the 15th Triennial Conference, New Delhi, 22-26 September 2008, available at URL (accessed 20th April 2011)
2 Free translation from Spanish from Decálogo de la Restauración - Criterios de Intervención en Bienes Muebles, available at URL [pdf] (accessed on 20 April 2011)
 B. Appelbaum, “Criteria for treatment: reversibility”, Journal of the American Institute for Conservation 26(2), 1987, pp. 65-73, available at URL
 J. Barrio, “Evaluación crítica de los principios en arqueometría, conservación y restauración de los vidrios arqueológicos”, Patina 12, 2003, pp. 53-64
 M. Favre-Félix, “Ambiguïtés, erreurs et conséquences: «Rendre l’œuvre lisible»”, Ceroart 3, 2009, pp. 2-16, available at URL
 M. Quiñones López, and J. García Sandoval, “Restauración de vidrio arqueológico. Montaje de vidrio arqueológico sobre resina en las lámparas de la sinagoga de Lorca para su exposición”, XX Jornadas de Patrimonio Cultural de la Región de Murcia, 2009, pp. 267-275, available at URL [pdf]
 M. E. Ortiz Palomar, “Tratamiento para la conservación, restauración y exposición de vidrios antiguos: la reintegración de vidrio con vidrio”, Boletín del Museo Zaragoza 13, 1994, pp. 303-312
 M. Bailly, “Le verre”, in La conservation en archéologie. Méthodes et pratique de la conservation-restauration des vestiges archéologiques, M. C. Beducou (coord.), Paris, 1990, pp. 120-162
 S. Davison, Conservation and Restoration of Glass, Butterworth-Heinemann, Oxford, 2003, pp. 284-307
 S. P. Koob, Conservation and Care of Glass Objects, Archetype Publications, London, 2006, pp. 75-110
 G. Lemajič, “Advantages of using a transparent PVC mould used in the process of replacing missing pieces on hollow glass objects”, Diana 10, Department for Preventive Conservation, National Museum Belgrade, 2004-2005, pp. 154-159
 L. Fernández, L. Schönherr, M. Pugès, “Productes i tècniques per la reconstrucció de vidre arqueològic”, Quaderns tècnics de l’MHCB: Conservació i Restauració 2, 2007, pp. 63-79
 B. Martínez Pla, “Restauración de alzata de vidrio y cobre dorado perteneciente a un juego litúrgico del Real Colegio Seminario del Corpus Christi del Patriarca (Valencia)”, Preprints of the 17th International Meeting on Heritage Conservation, Fundación La Llum de les Imatges, Conselleria de Cultura i Esport, 2008, pp. 501-504
 N. Tennent, “Clear and pigmented epoxy resins for stained glass conservation: light ageing studies”, Studies in Conservation 24(1), 1979, pp. 153-164
 J. L. Down, “The Yellowing of Epoxy Resin Adhesives: Report on High-Intensity Light Aging”, Studies in Conservation 31(4), 1986, pp. 159-170
 E. Risser, “A new technique for the casting of missing areas in glass restoration”, Journal of Conservation & Museum Studies 3, 1997, DOI:10. 5334/jcms.3973
 R. F. Erret, “The repair and restoration of glass objects”, IIC Bulletin of the American Group 12, International Institute for Conservation, 1972, pp. 48-49
 P. Jackson, “Restoration of an Italic glass oinchoe with Technovit 4004A”, Conservator 7, 1983, pp. 44-47
 L. Hogan, “An improved method of making supportive resin fills for glass”, Conservation News 50, London, 1993, pp. 29-30
 D. Ling, “Conservación de vidrio hueco en el British Museum de Londres”, Jornadas Nacionales sobre Restauración y Conservación de Vidrios, Fundación Centro Nacional del Vidrio, 2000, pp. 135-143
 S. Davison, “Reversible fills for transparent and translucent materials”, Journal of the American Institute for Conservation 37 (1), 1998, pp. 35-47, available at URL
 I. Gedye, “Pottery and glass: the conservation of cultural property”, Museums and Monuments 11, UNESCO, Paris, 1968, pp. 109-113
 L. Osete, Estudios de procesos de corrosión de vidrio y vidriados arqueológicos y caracterización de sustancias filmógenas tradicionalmente utilizadas en su restauración, Facultad de Químicas, Universidad de Valencia, 2005
 J. M. Fernández Navarro, El vidrio, Consejo Superior de Investigaciones Científicas, Madrid, 2003
 B. Cope, “Transparent plastic film materials for document conservation”, Paper Conservation News 93, Institute of Paper Conservation, 2000, pp. 14-15
 L. Bottenbruch (ed.), Engineering Thermoplastics. Polycarbonates, polyacetals, polyesters, cellulose esters, Hanser Gardner Publications, Munich, Vienna, New York, 1996
 T. O. Taylor, “The use and identification of plastic packaging films for conservation”, The Book and Paper Group Annual 4, The American Institute for Conservation, 1985, available at URL
About the authors
Betlem Martínez graduated from the Department of Fine Arts, Universidad Politécnica de Valencia, specializing in conservation in 1997. She has expanded her knowledge through a number of courses since 1995; she has taken the Máster Oficial en Conservación y Restauración de Bienes Culturales - at the Universidad Politécnica de Valencia specializing in archaeological materials in 2010. She has been working in public and private projects related to her specialization since 1998 both as part of a private enterprise working in restoration of archaeological materials and, since 2006, at the Conservation and Restoration Service of Diputación de Castellón and the Institut Valencià de Conservació i Restauració de Bens Culturals.
Trinidad Pasíes (PhD) is a restorer at the Museo de Prehistoria of Valencia. She graduated in Fine Arts in 1992, specializing in restoration. Since then she has been expanding her knowledge of archaeological material treatment doing research and working at different international centres such as the Atelier de restauration de mosaiques (France); Opificio Delle Pietre Dure and ICCROM (Italy); Parque de Tikal (Guatemala); Ministry of Culture (Greece). She has directed a large number of archaeological conservation and restoration interventions at national level. She has been working as a teacher since 1996. In 2007 Dr. Pasíes completed the Máster Oficial en Conservación y Restauración de Bienes Culturales at the Universidad Politécnica de Valencia. She has participated as a researcher in Research, Development and Innovation projects and her work has appeared in several national and international publications.
Maria Amparo Peiró
M. Amparo Peiró graduated from the Department of Fine Arts, Universidad Politécnica de Valencia in 1999, specializing in restoration. Since then she has expanded her knowledge and experience in the field by means of grants and projects in Italy. Since 2002 her work is mostly focused on archaeology. She collaborates in projects with Museo de Prehistoria of Valencia. In 2010 she continued her professional training with the Máster Oficial en Conservación y Restauración de Bienes Culturales, with a final project on archaeological lead. She currently combines her work at Museo Arqueológico of Burriana and teaching activity with various restoration projects of archaeological material for private and public enterprises.