Available online 17 September 2015
Stick it on! Temporary Transfer Papers as Retouching Media for Ceramics Conservation
This paper describes the use of temporary transfer papers as an alternative method for the retouching of plaster fills applied to ceramic objects. It also presents the results of accelerated light ageing analyses. A transfer printed tureen was the basis for this investigation. The original pattern was identified on the object and sourced online. The found image was adapted on Adobe Photoshop to match the object's colour and shape. Preliminary tests investigated transfers’ workability properties and results were visually assessed. Given the good results, first steps were taken to assess their behaviour over time. Four types of water-based, cold-application transfers were tested for light fastness. A light dosimeter (Blue Wool Scale) placed alongside the samples suggests an equivalent exposure period of 50 to 100 years. Samples of these transfers (without printing ink) were aged under fluorescent light and colour change measured with a spectrophotometer. The four samples show average ΔE values under 1 (0.6-0.8) and over 1 (1-1.6) implying a very good light fastness quality if displayed with proper mounting.
The use of printed transfers to decorate ceramic objects became a reality since mid-18th century and rapidly spread across the globe. Oriental decorated porcelain, traditionally hand-painted and expensive to acquire, were reproduced on transfer and applied to several types of ceramic objects, carrying a little of that exuberance to most households . Although the process of making one transfer was also very laborious, the ability to be mechanically reproduced made transfer printed china inexpensive to those who could not afford the porcelain coming from the East.
Given their proliferation, conservators find themselves in constant presence of these objects at the workbench. If economically they might not have great value, conservators have been asked to conserve these pieces for their clients’ sentimental and historic interests. These objects have regularly been used domestically and show marks of time; often they were repaired to return to being used again and again.
In order to illustrate the temporary transfer papers procedure in this article, a Wood & Sons, transfer printed dinner service tureen will be used (Figure 1). It had a couple of sections detached, minimal staining and a large loss to the foot-rim. The process to reproduce this missing section is the basis for this investigation.
A commonly accepted method to reproduce the printed decoration is to apply a paint layer of background colour over a desired fill (plaster, resin, etc.) and hand-retouch the “dotted” transfer motifs . However meditative this might be, it is a time consuming process; often, private collectors do not want to pursue this costly approach and the fill is left “blank”.
Inspired by the tureen’s original manufacturing technique, it was conjectured whether a similar process could be used to achieve the same decorative finish, more quickly and efficiently than the hand-painted alternative.
The original transfer application requires special tools and equipment a ceramic conservator does not necessarily have, which are commonly seen in pottery and printing workshops. Therefore, this paper looks to cold-application transfers as an alternative. In this pursuit, we came across temporary transfer papers, often used for temporary tattoos, which can be applied to several surfaces (plaster, glass, human skin, wood, wax, etc.). These are also known as decals or water-slide transfers.
The use of transferable media has already been investigated for conservation purposes before. Roja de La Roja  developed a system of superimposing layers of coloured dotted sheets to recreate missing areas on painted sculptures and canvas. However, we were more interested in using a given pattern/image identified on the remaining object, digitally transform it and reproduce it integrally onto the reconstructed foot-rim section. Vitale and Messier  had successfully used digital imaging software as well for a similar purpose when reproducing and reprinting degraded historic wallpaper. Nheller and Youket  also investigated the use of Adobe Photoshop to isolate missing areas on paper and digitally recreate them. This project tries to combine the mentioned studies for the purpose of ceramics conservation on a 3D surface.
Very exciting results were achieved when the transfers were applied to the tureen’s previously made plaster fill (Figure 2). In this way, further analysis was carried on to understand these materials’ properties and assess their behaviour over time.
Therefore, the following sections describe the transfers, their use for the tureen case study and their workability properties. Furthermore, it also explores these materials’ light fastness qualities.
2. Temporary transfer papers
By restricting this research to water-based transfers applied in cold, we came across two main categories of water-slide, cold-application temporary transfers, which differ on end result and method of application. Cold-application transfers are easy to use and apply, working as simple stickers. Different suppliers were sourced online; each would have similar set of transfers only differing on the backing paper being opaque or transparent. Instructions outlined by the manufacturers were not changed and an exemplificative application process is described below for reference. Although no printing ink was tested at this point, both are printable on any desktop printer (using inkjet cartridges or a laser toner).
United States patents1 were very useful to better understand these papers’ elements and properties. However, further analyses need to be carried out to corroborate these evidences.
Type 1 – Temporary tattoo papers
These come in prepared sheets, ready to print. Consists of a non-woven synthetic film, over and pressure sensitive acrylic adhesive, temporarily adhered to a backing paper sheet. The motif is printed onto the non-woven film on any laser or inkjet printer (according to what is specified by the supplier). To apply, one cuts tight to the motif, soaks it in water and peels the transfer (printed motif/film/adhesive).
Type 2 - Water-slide transfer papers
These come with a set of backing paper (white or transparent) on to which the motif is printed, and an adhesive layer between two polyethylene adhesive carriers. After the motifs are printed onto the given paper, one of the carriers is removed and the remaining is turned over and applied to the printed surface. The other polyethylene carrier is then peeled, revealing the adhesive side that is then pressed against the desired surface. By dampening the backing paper, the realising agent is softened, the paper removed and the printed motif revealed.
Left to right:
Figure 1. Wood & Sons tureen before intervention. Note the foot-rim missing section.
Figure 2. Foot-rim plaster fill (left) and the same area after transfer application (right).
Figure 3. Digital manipulation of a same pattern dish from where the pattern was cropped and adapted.
3. Case study
Extensive bibliography can be found about materials and techniques for the making of fills for ceramics conservation, according to body type, purpose of intervention and object’s condition . However, a brief description of how the fill was made prior transfers’ application is considered important as to better understand the transfers’ scope as a retouching media for ceramics conservation.
The pattern was sourced online after being identified on the object. Some factories might have published their pattern catalogues; otherwise a photograph from a similar object can be used. Given the tureen's curved profile, it was thought it to be more efficient to find the same pattern on a flatter object (Figure 3), which would later be digitally manipulated, rather than photographing the tureen. Adobe Photoshop tools were used to adjust colour and saturation and re-shape the pattern to fit the fill's shape.
With regards to the foot-rim loss, a plaster removable fill was made to support the transfers’ application2 (Figure 2a). To prevent the plaster from becoming attached to the ceramic body, a layer of Parafilm (Figure 4) was placed over the break-edge and stapled with a stencil brush to improve the fill’s key definition. Liquid Crystacal ‘R’ plaster was introduced into a previously made dental wax mould in situ (Figure 5).
When set, the plaster was detached and coated with a background acrylic paint layer matching the object’s colour. Every application of paint and transfer was performed away from the object. After applying the transfer, a coating of acrylic varnish was needed to compensate the overall sheen.
Individual transfer workability tests were undertaken to better understand which of the two types would better suit this particular fill. The crucial characteristics were: maximum size of transfer application, adaptability to a curved profiled surface, and a sympathetic end result.
Left to right:
Figure 4. Parafilm applied to the break-edge.
Figure 5. Dental wax mould in situ with air vents for plaster insertion.
Figure 6. Transfers applied to plaster surrogates over a layer of acrylic background paint.
3.1. Workability tests
The transfer papers were tested onto plaster surrogates (Figure 6)(0.5 x 2 x 6cm) made out of Crystacal 'R', the same filling material used on the case study.
Normally, in creating a design motif, a retouching layer would be applied prior to creating the design: this layer can be coloured or clear but the medium would ultimately be the same. However, it was considered that these transfer papers may not require this stage or might even react to this layer after application and cause adhesion stresses. Therefore, both plaster and plaster with Golden Polymer UVLS Varnish were tested.
3.2. Workability results
Results were aesthetically promising. The samples were assessed visually right after application and periodically until a maximum of 90 days. The following points summarise the results of each paper type:
Type 1 - Temporary tattoo papers
• Good adhesion after 90 days at room temperature;
• Slight height inherent to the application process – the plaster fill needs to be recessed to compensate for this thickness;
• The film can have a yellowish colouration depending on supplier;
• Maximum size recommended is ~2cm2 due to printed film wrinkling;
• The adhesive swells/softens in acetone – the suppliers suggest the use of abundant soapy water to “wash off”, which is not adequate for conservation purposes;
• Can be coated with an airbrushed acrylic layer without directly affecting the inks and synthetic film below.
Type 2 - Water-slide transfer papers
• Good adhesion but remains tacky after 90 days at room temperature – tackiness only occurs where no ink is printed, i.e. where the adhesive remains active;
• Can reduce tackiness with French chalk or acrylic varnish;
• Need to reverse image before printing (no size restriction);
• The adhesive swells/softens in acetone (as Type 1);
• Can be coated with an airbrushed acrylic layer without directly affecting the inks and adhesive below.
There was no difference in adhesion between plaster surrogates with or without an acrylic paint layer. As far as colour is concerned, because there is no white ink in a common desktop printer, all the “white” areas we see on the computer screen will take the colour of the fill underneath, thus the airbrushed acrylic layer influences the visual end result. Type 2 was elected the most suitable for this case due to the large size and curvature of the fill.
The completed plaster fill was then bonded to its correct place with Paraloid B72 (40% w/w in Acetone). No attempt was made to hide the new section; it remains visibly and physically different from the original though integrated (Figure 2b).
4. Accelerated light ageing
When designing the accelerated ageing analyses, it was decided to focus on what was left attached to the fill’s surface after the transfer had been applied and all backing papers disposed of, i.e. the synthetic film/adhesive, and investigate printing inks and coatings at another stage. Moreover, the choice was made to expose the transfers to light, the environment’s main variable to which such objects would be exposed to whilst on display. There was no intention to control humidity or temperature levels at this point.
4.1. Sample preparation
Four different transfers were tested:
Type 1-A: Temporary tattoo film by Ink to Print (http://www.inktoprint.co.uk)
Type 1-B: Water-slide decal paper by Crafty Computer Papers (http://www.craftycomputerpaper.co.uk)
Type 2-A: Temporary tattoo kit by Papilio (http://www.papilio.com)
Type 2-B: Tattoo papers by Crafty Computer Papers
An average of four pieces of the material (2x5 cm) were cut and applied according to instructions given (discarding the printing steps) to a piece of PMMA (poly(-methyl-methacrylate)). Colour measurements were made over a white opaque surface.
4.2. Accelerated light chamber
An accelerated light chamber set up was built according to British Standard BS1006:1990 guidelines with slight alterations to the exposure rack (Figure 7). Fluorescent full-spectrum lighting was preferred in order to keep temperature levels down so that thermal ageing effects on the samples below would be minimised. Samples were displayed flat on a tray, underneath light bulbs (Philips 500w 235-245ml) at a distance of 12cm and were left under constant exposure for 115 days, totalising 2760 hours.
4.3. Blue Wool Scale
A light dosimeter (Blue Wool Scale) was put alongside the transfer samples in the light chamber rack as indicated in British Standard BS1006:1990 . The Blue Wool Scale is made of 8 pieces of wool fabric tinted with blue dyes of different known light fastness rates. The bottom half was covered with a black board to hide strips from light in order to serve as control. By assessing which of these strips had the least change, an equivalent amount of light exposure can be estimated. The light dosimeter ranges from 1 (Very poor light fastness) to 8 (Excellent light fastness).
4.4. Colour Measurement
A Minolta CM2300d spectrophotometer with standard illumination D65 simulating daylight with colour temperature 6504K and an observer angle of 10˚ was used to measure the colour change. The colour scale CIE1976 L*a*b* was used based on the Opposite-Colour theory. Three values are given by the spectrophotometer: the L* axis stands for luminosity (dark=0 to light=100); the a* axis plots colour from red to green; the b* axis from yellow to blue. The three values together give a quantitative description of colour in a 3-dimensional colour space. A single value of Delta E (ΔE), which represents a total colour difference based on the change of L*, a* and b* values (before and after light exposure), was calculated according to the following equation, where Δ represents the change of each given variable: ΔE = √[(ΔL*)2 + (Δa*)2 + (Δb*)2].
ΔE values higher than 1 are generally accepted as the minimal noticeable colour change by the untrained eye. If lesser than 1, change is normally not visible, between 1 and 2 is only noticeable to the trained eye and over 2 begins to be generally noticeable . By using a spectrophotometer we can assure that colour change is systematically and accurately registered.
Left to right:
Figure 7. The light chamber set up. Photo by Ronnie Kam.
Figure 8. Blue Wool scale after 115 days of light exposure.
5. Results after Accelerated Light Ageing
5.1. Blue Wool Scale
Figure 8 shows the Blue Wool scale after ageing; there is a visible change in colour towards the left side but it becomes less noticeable as it reaches the other end. Table I summarizes the spectrophotometer results of the same scale after exposure to light. According to the specification of ΔE greater than 1, a Blue Wool scale 6 was achieved, where the least colour change was measured (ΔE=5.7). The ASTM (American Society for Testing Materials) conversion table for the given Blue Wool scale  indicates an equivalent period of 50 to 100 years with 100 megalux-hours of illumination before the change becomes noticeable to the naked eye. This conversion can only be seen as an estimate and considering an indirect source of light and exposure of 12 hours /day.
5.2. Temporary transfer papers
The transfers’ colour change results measured with the spectrophotometer are shown on Table II. These show that Type 1 transfers have the greatest change in colour (ΔE>1).
Although colour change is less noticeable on Type 2 samples, the adhesive remains active and tacky after ageing. Moreover, Type 1 and Type 2 show a ΔE value where change is considered “unnoticeable to the untrained eye” or “invisible” respectively.
Important steps were taken with this investigation towards the use of this kind of material for reinstating of large decorative areas of loss in the conservation of ceramic objects. Far from exhaustive, this investigation sheds a light on the benefits of their use, their applicability in conservation and discusses long-term properties.
The light ageing analyses suggest a more stable performance of Type 2 over a period of 50 to 100 years with proper lighting and mounting conditions that museums and art galleries possess, although further analysis need to be carried out concerning humidity and temperature levels as well. Within this time period the adhesive layer should not discolour to a degree that it becomes noticeable. Before adapting this technique, inks and coatings should also be investigated for long-term ageing.
The author wishes to thank his tutors David Dorning and Lorna Calcutt at West Dean College for their guidance and invaluable help whilst a student and as a graduate, particularly with regards to the use of the accelerated light equipment. Also special thanks to Professor Norman Billingham (University of Sussex), Whitney Baker (University of Kansas Libraries), Ronnie Kam and Abigail Bainbridge (West Dean College).
 K. Petrie, Ceramic Transfer Printing, A & C Black, London, 2011
 V. Oakley and K. Jain, Essentials in the care and conservation of historical ceramic objects, Archetype, London, 2002
 J. M. de la Roja de la Roja, Sistema de reintegración cromática asistido por medios transferibles obtenidos por procedimientos fotomecánicos: aplicación en la restauración de pintura de caballete, PHD thesis, Universidad Complutense de Madrid, 2004, http://eprints.ucm.es/1750/
 T. Vitale and P. Messier, Historic Wallpaper Digitally Remastered: Early Twentieth-Century Block-Printed English Wallpaper in the Yin Yu Tang House at the Peabody Essex Museum, The Book and Paper Group Annual, 23, pp. 109-113, 2004
 D. Nheller and M. Youket, A Tale of Two Facsimiles: Incorporating Digital Technology in Conservation Treatments, The Book and Paper Group Annual 21, 2002, pp. 91-94
 S. Buys and V. Oakley, The Conservation and Restoration of Ceramics, Butterworth-Heinemann, Oxford, 1993
 BS1006 Methods of test for colour fastness of textiles and leather, British Standards Institution, London, 1990
 Electronics for Imaging, Delta E, Delta H, Delta T: what does it mean?, http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.171.205 (accessed 1 June 2015)
 ASTM D5383 – 02 Standard Practice for Visual Determination of the Lightfastness of Art Materials by Art Technologists, ASTM International, West Conshohocken, PA, 2003
1 Selected patents: US5958560 (1999), US6264786 (2001), US6074721 (2000) and US6793999 (2004).
2 Materials and techniques hereby used are in line with standard best practices and Health & Safety measures discussed in extensive bibliography and own professional experience.
London, United Kingdom