INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER Daniel VEGA Departamento de Conservapto ¢ Restaure Faculdade de Ciencias e Tecnologia, Universidade NOVA de Lisboa, Campua de Caparica, Portugal Isabel POMBO CARDOSO Departamento de Conservagi ¢ Resanra & Requimte-CQFB Faculdade de Ciénciase Tecnologia, Universidade NOVA de Lisboa, Campus de Caparica, Portugal Leslie CARLYLE Departamento de Conersagio ¢ Restanro & Requinte-COEB Faculdade de Ciencias ¢ Tenologia, Universidade NOVA de Lisoa, Campus de Caparica, Portugal Abstract “This paper focuses on finding an inert formulation for pigmented wax-resin infils to be used for oil paintings on coppet Tivo pigmented wax-resin formulations currently in use are characterised: one supplied by one of the authors (C-PWR) and the other, 2 commercial product, Gamblin Pigmented Wax-Resin sticks (G-PWR), While both have excellent properties, they contain natural beeswax which has been reported to result in corrosion where in contact with copper. Acid number measurements confirm that both are acidic. Although ageing tests are still needed, the characterization and properties of two new formulations, both based on Techniwax 9426 microcrystalline wax with cicher Regalrez 1094 or 1126, are promising. Both new ‘waxtesin mixtures, with an acid number of O, are likey to bb inert in relation ro the copper and are anticipated to be chemically stable due to their composition. Keywords: oil painting on copper, infills, wax-resin, micro- crystalline wan, synthetic resins, beeswax Introduction Ichas been reported that paint losses on oil paint- ings on copper result from a chemical reaction be- ‘ween the copper support and the oil paint composite which includes both ground and paint layers (Horo- vitz 1986, 1996, 1999; Pavlopoulou 2004; Pavlopou- lou & Watkinson 2006). This was reinforced by ob- servation of over 80 oil paintings on copper where losses tended to be complete, leaving che metal an- derneath exposed (Vega 2016) (Fig. 1a & b). While the primary reason for re-integrating paint losses by inflling and inpainting is co restore unity to the im- age, there is also an argument for this step, as a means to reduce or halt further corrosion of the exposed metal (Phipps & Rice 1979; Zakipour er al. 199 Greywacz 2006), A review of publications on infill macerials in use or under consideration (see below) led to the conclusion that pigmented wax-resin could offer significant ad- vantages over previous options. However, formulations 67]DANIEL VEGA ~ ISABEL POMBO CARDOSO ~ LESLIE CARLYLE in use for oil paintings on other supports normally contain beeswax, which has been shown to corrode copper (sce below). ‘Therefore, one of the aims of the Masters thesis work described here, which was carried cout by Daniel Vega at the Universidade Nova de Lis- boa, was to develop a non-acidic pigmented wax-resin ture (Vega 2016). Properties of an ideal infill material In her PRD thesis on infil materials for the conser vation of paintings (2006), Laura Fuster Léper. iden fied the ideal properties of a successful infill (76-77): = compatible with the original materials (chemi- cally, physically and mechanically); capable of receiving an impression of texture (if necessary); = stable to relative humidity and temperature fluctuations; = removable mechanically (or at least removability is restricted to mild aliphatic solvents); and — long term physical and chemical stability In addition, infil for oil paintings on copper should also be capable of being applied in very thin layers to match the thickness of the paint composite (thin paint! ground layers are reported by Horovite 1999; Rizzo 2008; Broers 2002 and Vega 2015 see Fig. 1 a&b). As will be seen, a non-acidic pigmented wax-resin would fulfill these ideal properties Overview of previous infill materials The literacure falls into two categories: those cover- ing treatments (Table 1); and empirical trials with dif. ferent infill materials (Table 2). Table 1 demonstrates that most of the infills used. ‘were aqueous based. To act as a water barrier prior to the application of che infil, isolating layers were used in four ofthe six treatments, Tables 1 & 2 indicate thatthe acrylic co-polymer Paraloid B-72 was used most fie- quently for this, In empirical trials, (Table 2) Horoviea looked at three non-aqueous options, however she re ported problems with adhesion and handling during application. Table 2 shows that some non-conservation products were reported to perform well but the lack of information on their ingredients and the impossibility of assuring that the same formulations will be used by’ the manufacture beyond the product tested at the time’ of publication, means that these commercial materials cannot be relied on, either in terms of their longevity o:| in terms of repeatability. The Corrosive action of beeswax on copper ‘Two authors, Leegenhoek 1986, and Garrel 1992, refer to beeswax as an infill ingredient (Table 1). How- evet, Horovitz (1996) reported tha in her trials beeswax. promoted corrosion on copper. Beeswax’ corrosive ac ion on copper and copper alloys is reported elsewhere: in the conservation literature (Webb 2000; Scott 2002), “Table Published treatments of losses for oil paintings on copper Ifill Type Authors Isolation Layer Infill Materials aq | aq Leegenhoek, 1986 | Varnish Hide glue + Caos x Beeswax + CaCo3 x Garrel, 1992, Beeswax x ‘ScoreMoncrief, 1993 Fine Surface Polyfila x Berges, 1995 Microcrystalline wax + UVS | Mowilich 20 + solvents + pigments x Rerouch vaeaish Maule, 2002, B-72:0-48 + aetone Gelatine pusy x Broers, 2002 B.72 « xylene + BTA Lascaux Hydto-Grund 750 + pigments x ‘aq = non aqueous; aq = aqueous. Sex Appendix 1 for the identity of materials wheze known p=INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER. Table 2. ‘Summary of empirical ial by Horovir, Broets and Rizzo ‘Authors | Isolation | Infill Materials Infill Type ‘Author's Observations Layer nae Horovie,, [B72 B72 chalk a Unsatisfactory co apply, and may be disrupted 1996 by B72 isolating layer Plextol B500 + chalk | _X ‘Adhesion is difficult Beva 371 + chalk fe ‘Spontaneously peeled off upon drying Fine Surface Polyfilla X_ | Usefal where coarse vextue is required Rowney Acrylic X | Suitable for shallow Tosses bur rubbery and dificult to carve Liguitex Acrylic X | Very spreadable, suitable for large flac areas of Gesso shallow loss Brocrs, [B-72+BTA | Lascaux X | Exsy wo apply and carve 2002 Hydro-Grand 750 Rizzo, [B72 Modostue X__ | Good performance 2008 Gel Relief X | Easy wo apply: suitable when textured surfaces (Lefranc & Bour- axe req ge0is) Become cransparent afier drying ‘Tempera paint X__ | Fills up the loss with several applications ‘nag? nonaqueous; aq. aqueous. See Appendix | forthe idenciey materials where known Scott (2002) explains that carboxylic groups from the farcy acids present in beeswax interact with metal ions from the copper resulting in the formation of copper soaps (metal carboxylates). These organometallic com- pounds ate reported to be chemically unstable, ro atack ‘metallic copper, and to promote further corrosion (Pat- erakis 1996, 2003; Scott 2002). Since four of the five fatey acids in beeswax, palmitic, linoleic, stearic, and oleic are the same as the major fatty acids in drying oils (van den Berg 2002), it follows that beeswax would havea similar effect on copper as drying oil. Drying oil is reported to form a green corrosion layer atthe inter face between the ol layers and the copper (Pavlopoulow ‘& Watkinson 2006; Oliveira 2015). Pigmented wax-resin infills Mixtures of wax and resin (natural or synthetic) in different proportions with the addition of pigments and fillers have resulted in infill materials thac are highly ver satile, Pigmented wax-resins (PWRs) were traditionally produced by conservators themselves. In recent years, pigmented and unpigmented wax-resin sticks have been ‘manufactured for conservation use by Robert Gamblin {gamblincolors.com/conservation-colors). Gamblin’ for- ‘mulation, which will be described below, is based on re- search by Christine Melntyre for her Masters thesis at Buffalo Stare College (McIntyre 201 1). For infills on copper, PWR has several advantages: is non-aqueous and is likely to remain relatively imps meable to moisture. PWR can be easily introduced the form of a fluid or a paste. [ti capable of forming a very thin layer that does nor shrink or crack (in the con- text of the inflexible copper support, mechanical crack- ing is not anticipated). For texturing, the fill can be shaped while warm, carved when cold, or the surface can be modified using a solvent with 2 low aromatic content (e.g. white spirit) Ic is casily removed mechan- ically or by using 2 solvent. Empirical crials indicate very effective adhesion to the copper metal. However, because the current formulations for wax-resin infils commonly contain beeswax, they have not been considered appropriate for paintings on cop- per. Furthermore, blooming, the development of thin whitish crystalline layer, has been reported on the sur- face of some of these PWRs, Bloom is associated with 6 |DANIEL VEGA ~ ISABEL POMBO CARDOSO ~ LESLIE CARLYLE beeswax, and has been reported on encaustic paintings, sculptures, and wax seals (Harley 1993, Novotna & Demovskova 2002, Bartl er al. 2015). Several studies have been carried out to establish the mechanism of blooming (Bartl etal. 2012, 2013, 2015). The most re- cent indicate that blooming is related to the migration and recrystallization of alkenes naturally occurring in beeswax (Bartl et a. 2015)' Studies on whether bloom will migrate chrough layers of pigmented resin used in inpainting on top of pigmented wax-resin fills has have not been found. Because of the significant advantages presented by PWR infil, it was decided to explore the possibility of finding a formulation which would eliminate the need for beeswax. To act as references for the characteristics sought in the new formula, wo successful wax formulations were chosen: a formula used by Carlyle in the early 1980s (C-PWR) and Gamblin Pigmented ‘Wax-Resin sticks (G-PWR). Both reference formula tions contain beeswax. ‘Wax-resin infill formulations Various authors (Murta 2007; Kemp 20095 Fuster Lépex 2012) report that throughout the 20th ceneury, formulations based on mixtures of natural waxes and. resins with pigments and fillers have been used exten- sively as an infill material for easel paintings and for other cultural objects. Fuster Lopez notes the substi- tution of natural resin with a synthetic which oc- curred most likely after the 1950s, when cyclohexa- none resin (originally Ketone Resin N, later Laropal K 80) was introduced (Fuster Léper 2012). PWR in- fills using beeswax with Ketone Resin N were in use at the Canadian Conservation Institute in the early 1980s? and a PWR formulation using beeswax, a mi- crocrystalline wax, “Be Square™ 195° and Larapol K 80 was used at Buffalo Stare College (Melntyre 2011). For her master thesis, McIntyre explored new PWR formulations aiming to find a substitute for Laropal K 80 afier its manufacture by Badische Anilin- und So- da-Fabrik AG (BASF) was discontinued in 2008 (McIntyre 2011). As noted above, her research was lacer used by Robert Gamblin in trials which resulted in his PWR sticks. ‘The PWR samples from Carlyle (C-PWR) consist of four parts beeswax to one part Ketone Resin N. Dry yent and chalk (calcium carbonate) were added to y the mele until the right consisteney was achieved (the consistency is very similar to G-PWR below). Carlyle provided three PWRs which had been prepared in the carly 1980s: a titanium white, a burnt umber and an, tunpigmented sample. Gamblin’s Pigmented Wax Resin Sticks (G-PWR) were purchased in 2009 and 2015. They consist of 3 parts beeswax, one part Be Square" 195 and one part Laropal A 81 (a urea aldehyde resin). Dry pigment is added in a proportion of one part pigment to one part WR By weight For the present study, sticks labelled Titanium white, Burne Umber and Neutral Base (an unpigmented stick) were chosen for direct comparison. with the C-PWR samples. ‘The individual ingredients Weeses Beeswax shows differences in composition depend- ing on the species and their food supply (Tulloch 1980, Aichhole & Lorber 1999, 20002). Tullock reported thac beeswax generally consists of: 14 per eent hyelroc bons (alkanes and alkenes), 35 per cent monoesters, 14 per cent Diesters, 3.3 per cent Triesters, 3.6 per cent Hydroxy monoesters, 7.7 per cent Hydroxy polyesters, 12 per cent free fatty acids, 0.8 per cent acid monoesters and 8.6 per cent of unidentified compounds (Tulloch 1980). These concentrations vary slightly depending upon the analytical procedures applied (Tulloch 1980; ‘Aichholz & Lorbeer 1999, 2000a&b; Jiménex. et al, 2004), Mills & White report a melting point for bees- wax of 63-64°C and softening point of 49 - 53.3 °C. (Mills 8 White 1999), Tulloch (1980) reports an acid value of between 16 and 24, however much higher acid values have been reported, from 16 to 125 (Warth 1956). Beeswax ean be purchased either refined or un~ refined. Acid treatments used during refining are te- ported to have little effect on the original acid value (Warth 1956), OF the synthetic waxes which have been used in some PWR formulations along with beeswax, and are candidates to replace beeswax in the new PWR formu- lation, there are two groups: macro-crystalline (eg. pat affin) and microcrystalline waxes (ManSoori et al 2003). The paraffins tend to be brittle and are therefore not suitable (Crowley & Laefer 2008, Laefer et al 2015); leaving the microcrystalline waxes. Pillon re~ ports on their composition: 30-60 carbon atoms and anINVES GATION AND TESTING TO DEVELOP AN INF average molecular weight of 500-800 g/mol, with 0-15 we per cent of n-paraffins; 15-30 wt per cent of iso-par- affins (branched); and 65-75 we per cent of naphthenes. “The melting point of microcrystalline waxes range from 60 t0 90 °C (Pillon 2008). Resins Resins usually have a medium molecular weight, an amorphous and often complex structure; and like waxes, they do not exhibit 2 precise melting point (Mildenberg er al. 1997). Synthetic resins cover a wide range of materials obtained by the polymerization of hydrocarbons. René de la Rie has extensively studied synthetic resins suitable for conservation use primarily in relation to painting varnishes (de la Rie & Mc- Glinchey 1990, de la Rie et al. 2002, Maines & de la Ric 2005), As noted above, one of the first synthetic resins to replace dammar resin in PWRs was cyclohexanone, but itis no longer made. For this work, along with the Lara- pol A 81 used in McIntyre’ formula, three other sy thetic resins were tested for the new wax-resin formula- tion: Laropal A 110, and Regalrez 1094 and 1126 (these resins, introduced by de la Rie were chosen based on stability testing as well as their easy availability). Laropal A, developed by BASE, is obtained by che condensation reaction between urea and aliphatic aldchydes (Von der Goltz et al. 2012). Information combined from de la Rie et al. 2002, Maines & de la Rie 2005, and online BASF Safety Daca sheets January 2016, provide the fol- lowing: the molecular weight (MW) of Laropal A 81 ranges from 3640 to 4300; the glass transition tempera- ture (Tg) is 57°C; and the melting range (m.c) is be- tween 80-95 °C. Laropal A 101 has a much lower MW: 2979, and a higher Tg: 73 °C. The melting range is also higher: between 95-110 €C. Regalrez, developed by Hercules Inc. (and currently produced by Eastman Chemical Co), is alow molecular ‘weight resin resulting from the polymerization of hy- drogenated styrene monomers. The chief difference be- rwween Regalrez 1094 and Regalrez 1126 is their molec- ular weight. This modifies their properties: Regalrez 1094 has an average MW of 900 gimol, a'Tg of 41 °C and melting point of 95 °C; whereas Regalrez 1126 has an average MW of 1250 glmol, a Tg of 67 °C and a melting point of 124 °C. Regalrez 1094 has a lower vis- cosity than Regalrez 1126 (de la Rie & McGlinchey 1990; Regalrez” 1094 & 1126 datasheets). FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER Experimental: a comparison with reference PWRS and the new formulations Both PWR reference samples demonstrated excel- lent performance with very similar working properties in terms of application and adhesion to copper sub- strates, therefore the aim was to match their key proper- ties in the new formulation. ‘Unpigmented samples were stuclied frst, so thar re- sults would not be masked by the presence of pigments or filler. The same tests were then used to study the pig- mented versions, and all tests were then applied to the new formulations for comparison. Physical/chemical estigations included: chemical characterization by ro Fourier Transform Infrared Spectroscopy (FTIR) and FTIR Attenuated Total Reflectance (FTIR-ATR), melting range, glass transition tempera- ture (Tg), and acid number. Other important parame- ters for determining the suitability of the infills were: workability, removability, and hardness. Chemical characterization }-FTIR and FTIR-ATR analysis were carried out on the PWR reference samples, the new pure materials, and new formulations, co establish their composition and evaluate their durability based on the compounds present. Spectra acquisition for both methods are de- scribed in Appendix 2, for che spectra see Figs 3.1, 3.2, 3.3 in Appendix 3, Melting range and Tg A Reichert Thermovar melting point microscope (Appendix 4) was used to determine the melting range ofthe materials. These measurements establish the tem- perature at which the PWR is workable. Softening point cannot be close to room temperature, since that ‘would incur the risk of future deformation/distortion in ambient conditions, and the melting point cannot be to0 high since the PWR must be safely applied without the potential to cause hear damage to the paint. The use of this instrument involved observation during con- tolled heating: the temperature was registered when the material flows for the first time (softening point) until the moment it becomes completely liquid (mele- ing poind) Similarly, che glass transition temperature (Tg) is important: a'Tg close to room temperature indicates a myDANIEL VEG! material which is soft and flexible but susceptible co dire absorption, while a high Tg is associated with brittle materials (Horie, 2010). Tg and melting range can be established using ASTM 4419 with Differen- sial Scanning Calorimetry (DSC). DSC also provides data related to the degradation temperature of the PWRs. At the time of writing results were not yet available but DSC measurements are being planned for the near furure. Acid number ASTM D1386 describes the procedure for deter- mining the acid number of waxes and was used for test- ing the PWR reference samples and wax-resins being, considered for the new PWR formulation. ‘The acid ‘number is determined by the number of milligrams of | potassium hydroxide (KOH) necessary to neutralize cone gram of sample dissolved in solvent. In the study being reported here, this correlates to the presence of carboxylic groups derived from Fatty acids in beeswax. Deuails of the procedure from ATSM D1386 are repro- duced in Appendix 5. Workability ‘Two fragments from old oil paintings on copper thought co date from the 18 century’ were used for testing the workability of the PWRs, Some of the paint composite losses on the fragments were isolated firs using Paraloid B-72 diluted in toluene (15 per cent wiv), PWRs were heated and applied to losses with and without the isolation layer using a ‘wax carving pencil’ (Appendix 4) Remavability ‘Mechanical removal ofthe different PWR infils was, tested using a wooden stick. Two mild aliphatic solvents, {Shellsol D and white spirit) were also tested for remov- al using solvent. Hardness ‘The hardness of the PWR is an importan parameter for judging its suitability. Extremely hard. substances ‘would not be appropriate for filling losses surrounded by oil paine composites with significantly different physical properties. The ASTM International standard for measuring the hardness of waxes is D1321 which je ISABEL POMMO CARDOSO - LESLIE CARLYLE requires a penetrometer. Unfortunately, it was not pos- sible to gain access to this instrument in Portugal. “Therefore, empirical testing using a needle to penetrate the surfaces of the PWRs was used. Hardness testing was carried out by one individual to achieve a level of uniformity in the results ‘The selection of waxes, resins and pigments for the new formulation Five microcrystalline waxes were chosen based on their melting ranges which were similar to the ingredi- cents in the PWR reference samples. The British compa- ny Kerax (kerax.co.uk), kindly provided samples for this study with the melting ranges required (Table 3). Melntyre found Laropal A 81 10 be the best option for replacing Ketone Resin N in the PWR formula she was using, as it provided effective adhesive qualities and. flexibility (McIntyre 2011). For this research Laropal A. 101 was also investigated to explore the effect of chang- ing MW and melting range. ‘Melintyre had eliminated Regaltez 1094 because she found that it made a slightly brittle WR which did noe mix well with pigments, In the present study Regalrez 1094 was included because different microcrystalline ‘waxes were being tested which could improve its perfor- mance, Because Regalrer 1094 is reported to ext long-term stability (de la Rie & McGlinchey 1990, Smith & Johnson 2008) and is currently widely avail able, a re-evaluation was felt to be important. Regalrez 1126 was included to establish whether ic would pro- vide any advantages over Regalrez. 1094 (Table 4). ‘The ratio of wax 10 resin and the procedure for combining. materials ‘Melntyre's successful formula, adopted by Gamblin, called for four parts wax to one part resin by weight. “Therefore, initial trials for the new formulation were based on this ratio (which turned out to give good con~ sistency with the new formulations). Guidelines from ‘Melntyre’s thesis for producing the mixtures were fol- owed (McIntyre 2011) with further advice provided by’ Robert Gamblin (via email correspondence). ‘Materials were weighed on a digital precision balance (Appendix 4). ‘The wax was melted using a bain-matie, and the resin, which had been reduced to powder, was slowly added while sticring with a magnetic stirrer.Table 3, Microcrystalline waxes selected Code [Description [Melting Range (O) KEWI_|_Techniwax 9211 84.88 KTW2_| Techniwax 9265 6 KTWS_| Techniwax 9300 76 KTWA | “Techniwax 9356, 7278 KEWS | Techniwax 9426 72.78 “Table 4. Synhetic resins selected Code |Description —_| Melting Range (°©) i Laropal A 81 8095 12 | Laropal A101 95-110 Ri | Regaler 1094 95 R2 | Regalrer 1126 124 As noted above, based on the colours available from the C-PWR samples, titanium white and burnt tumber (Table 5) were chosen for all testing to allow a comparison between the reference samples and the new formulations. While the Meintyre-Gamblin reci- pe called for 1:1 pigment to wax-tesin by weight, for the new formulations, a similar consistency 0 the PWR references was achieved by adding titanium white and burnt umber in a ratio of one part pigment {0 two parts wax-resin by weight. Results “The first significant finding was that it was not pos- sible to obtain a mixture of the microcrystalline waxes chosen with either Laropal A 81 or A 101. After six hours of mixing and several repetitions, a viable wax-res- in mixture could nor be formed. However, when bees- wax was melted with microcrystalline wax frst then the «TION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER Laropal (cither A 81 or A 101) was added, @ mixture was easily achieved after 25 minutes. This demonstrated thar beeswax is a necessary ingredient when Larapol A. 81 oF 101 resin is used. In contrast, achieving a mixture of each of the two Regalrez resins with each of the five microcrystalline waxes was very straightforward. All 10 different combi- nations formed viable mixtures which took 17 to 40 minutes to prepare (Table 6). In addition, the 10 new formulations for WRs mixed very well with pigments and formed a uniformly coloured PWR with the pig- ‘ments remaining in suspension both in the liquid and. solid phase. ‘Table 6 also reports the acid number for the PWR reference samples and the new formulations, As expect- ed, because of the presence of beeswax, the reference samples proved to be acidic. Not surprisingly, che acid number is higher for che WR formula wih the highest proportion of beeswax (C-PWR). The acid number for all new formulations was zero. Properties of the new formulations Melting ranges Melting range measurements of the 10 new formu- lations all showed a lower meleing point than the pure products alone. This i in accordance with the literature which reports a common phenomenon known as ‘Mele- ing point depression’: ehe melting point of @ mixture of ‘wo different materials is never the average of the melt- ing points of the two initial products, but is always low- er than the melting point of the material present in the greatest amount (Isayev 2010). However, it was also shown in this work that the resin with the highest melting point consistently affects the overall melting point of the mixture by raising it one to three degrees Celsius more than when resin with the lower melting point was used (compare MPs for formu- las with RI & R2 Table 6). As a result six of the new ‘Table 5. Pigments selected (from Kremer Pigmente) Code | Kremer pigment number | __ Deseripti Colour Index ‘Chemical Description PL 46200 ‘Ticanium Whice Pwo7789 [TIO = Natural brown earth, con- p2 40610 Burne Umber, Cyprus PBr8772___ | Nine manganese oxiler 3 |DANIEL VEGA ~ ISABEL POMBO CARDOSO ~ LESLIE CARLYLE Table 6 ‘Unpigmented WR reference samples compared with differene combinations of microcrystalline waxes and Regalrez resins Workin | Misc | Timero [Range ofWonabifiy Toy | aay of prepare silt ‘Hardness Codes | components | Caines) aes eae ‘Workabiliry | Number CrwR A 68 yes 16,056 2 G-PWR 2 © ve 11,299 2 KTWLRI v 25 4 70 yes 0 Hae KTWIR v 40 % 73 yes a 4 KIW2R v 7 25 30 20 a 1 KTW2.RO v 7 28 BA 20 a 1 KTW3:R1 a 21 24 50, no o 1 KTW3:R2 Vv 22 4 52. no, o 1 KIWERI ¥ 18 3 7 20 A 1 KTW4.Re v 18 40 36 20 Q 1 KIWS:RI ¥ 22 a 64 yes @ 2 KOTW5.Ra v 25 4 @ ve a 2 Hardness: 1. Sofi; 2. Medium soft; 3. Medium hard; 4, Hard ‘Table 7. Temperature range for reference PWRs formulations were discarded as their melting range was and sclected WR formulations deemed to be too low (in fact, four of these formula- tions were very soft and cacky ae room temperature). Te Range for Workability Four formulations, KTW1-R1, KTW1-R2, KTW5-R1 Pigment [Softening | Melting | and KTW5-R2 were found to have melting ranges sim- Point (CC) | PoineO) | ilar to the WR reference samples. These consisced of PL a @ microcrystalline waxes Techniwax 9211 (KTWI) ot P2 2 o Techniwax 9426 (KT WS) with the resins Regalrez. 1094 Pr @ 70 (RI) and Regalrer 1126 (R2). 7] 2 7 Ie is imporeane to note chat the introduction of a pigment was found co increase the melting point of the KTWERI PI 42 7 mixcures by one or wo degrees Celsius (Table 7). P2 2 70 KTWLR2 u 7 ze Workability and hardness (Characteristics which were assessed empirically: work KTWS:RL PL = = ability and hardness, also revealed interesting, results and. Po au 7 contributed to the choice of the final two formulations. KIWSIR2 PL 2 o Regarding worlabiliry, both the PWR reference 2 a 70 samples and two of the remaining new formulations were found to have che same application and working parameters: the temperature required for softening was 529C. Application was straightforward and efficient us- ing the wax carving pencil; the PWRs were easily lev- clled wich gentle pressure while the infils were still ‘warm; and ic was possible to work che infills at room 1INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER, temperature by slight warming and to provide rexture using dental rools ora thin wooden stick. ‘The removal of excess material in the loss or around its perimeter could be carried out mechanically or with solvents. [¢ was found that most of the excess PWR could be gently removed with the point of a wooden stick without causing damage to the paine surface. This mechanical removal could then be followed by cleaning svith white spisis or wich Shellsol D where necessary In addition, it was found that gloss, in the case of the new formulacions, like the PWR reference samples, could be adjusted easily, and if required, a high gloss can be achieved by buffing the surface with cotton wool or soft fabric. “The new infills, like che PWR reference samples, could be easily coated with a varnish based on Regalrez 1094 diluted in white Spiri (20g resin ro 80 mls whice spirit). Nevertheless, since the infill material is sensitive to most solvents and extensive brush application could disrupt or remove the surface texturing, it may be prux dent co apply the varnish by spraying. Hardness cesting using a scalpel and 2 needle con- firmed that the two new formulas chosen for theit prop- erties in workability were also the most similar to the reference samples in terms of hardness, ‘The two forrnu- Jas chosen both consisted of four parts by weight of “Techniwax 9426 microcrystalline wax (code KTW5), id one part of Regalrex: either Regalrez. 1094 (RI) or Regalrez 1126 (R2). The dry pigment ratio for the new formulations which best matched che PWR references is one part pigmenc to fwo parts WR. Stability ofthe new formulations Reports on testing in che literature indicate that mi- crocrystalline waxes (Rowe 2006; Gad 2008), and the synthetic resins Regalzez 1094 and 1126 can be consid ered stable (Maines & de Ja Rie 2005). Preliminary pPTIR and FTIR-ATR analyses of the new formulations indicate tha che materials are likely ro remain stable since spectra from the pure materials as well as the final mix- tures show only the presence of saturated hydrocarbon bonds (Appendix 3). These are known to be the leat re- active bonds found in organic compounds (Martel 2004), In contrast, the spectra from the reference samples C-PWR and G-PWR show the presence of functional groups which are chemically more reactive, Of the func tional groups present, carbonyl groups are bonded to esters, ketones and aldehydes. The esters are from the beeswax present in both formulations. Ketones, present in the C-PWR can be attributed to the polyeyclohexa- none resin used in its formulation, and the aldehydes, present in che G-PWR sample, are associated with La- ropal A 81. Also associated with the Larapol A 81 and present in the G-PWR sample, are unsaturated bonds and hydroxyl groups (Appendix 3). Future research Based on the #FTIR and FTIR-ATR results, itis likely thac the new formulations will prove to be more chemically inert than the PWR reference samples. However, ie will be important to undertake phorochem- ical ageing tests of both the reference samples and the new formulacions with the individual materials alone and in mixture (Ferseira er al 2010). While the acid number of the reference samples suggest that chey will promote corrasion on copper due to the beeswax they contain, this has yer to be demonstrated. Ie is also important to note ¢hat a recent study un- deraken of Laropal A 81 and Regalrez 1094, which made use of different and more sensitive analytical in- strumentation, provides new information about previ- ously undetected features: in Laropal A 81 the presence of polar groups, and in Regalrez 1094, of double bonds, {Bonaduce eral, 2013). In the later case further research js requited in zelation co the use of Regalrez resins in _ixcures with microcrystalline wax. Blooming is evident on both the C-PWR and the G-PWR reference samples, however to date, it has not been observed on the surfaces of the new PWR formula- tions, nor was bloom observed by Mcintyre on naturally aged samples with microcrystalline wax alone (McIntyre 2011). Testing of the two new PWR formations for pos- sible bloom development in ambient temperatures with cycles of low and high relative humicicy is planned. Conclusions The characterization of the PWR reference samples; the Master thesis by Christine Mclncyres and the im- poreant suggestions made by Robert Gamblin regarding, the preparation of che pigmenced wax-resin mixture, al formed the basis for these preliminary trials and testing co explore and develop @ new war-resin formulation suitable for infill on paintings with a copper substrate, 195renee pete DANIEL VEGA ~ ISABEL POMBO CARDOSO = LESLIE CARLYLE “The reference samples from Carlyle and from Gam- bilin’s Pigmented Wax-Resin sticks proved 10 be very important for guiding the choice of raw macerials to be used in the new formulation, and as anticipated, high- lighted the potential problem of beeswax as an ingredi- enc for infills on oil paintings with a copper substrate. ‘The characterization of the the new formulations made with Techniwax 9426 mi- crocrystalline wax with Regalrez 1094 or 1126, sug- gest that these new wax-resin mixtures with their acid umber of ze10 ae likely to be chemically stable in relation to 2 copper support. Since Regaltez. 1094 has already been widely adopted in che conservation field as a varnish for oil paintings, and is therefore easily available, it is the most likely candidate of the cwo Re- galrer resins studied to be used in the new PWR for- mula, However, as noted above, further testing is re- quired, especially on copper. ividual macerials in Ik is very encouraging that the working qualities of the new formulations were found co be very similar to the PWR reference samples. The newly formulated PWR fils ate easy to apply and achieve an excellent re- sult regarding the infilling of very shallow losses. Fur- thermore, their application temperatures and subse- quent softening points were in the range suitable for use on oil paintings. References Aichhol, R. and Lorbeer, B. 1999. “Investigation of comb wax of honeybees with high-temperature gas chromatog- raphy and high-temperature gas chromatography-chemi- cal ionisation mass spectrometry. I: High-temperacure gas, chromatography’, Joural of Chromatography A8SS (2): 601-615, Aichhok, R. and Lorber, E. 2000a, ‘Investigation of comls wax of honeybees with high-temperature gas chromacog- raphy and high-temperature gas chromatography-chemi- cal ionisation mass spectromenry. II: High-temperacure gas chromatography -chemicl ionisation mass spectrom xy, Journal of Chromatography ABS3 (1-2): Aichbole R, Lorbeer E, Pechhacker H, Httinger E. 2000b, “The chemical compounds of beeswax fiom Apis species. Papec presented at the Proceedings of 4 Asian Apicultur- al Association International Conference, Kathmandu, 152.154, Bard, B. Trejbale, J. Durovié, M. Vasickovéd, S. Valterové, I. 2012. “Analysis of efflorescence on surface of beeswax seals’, Journal af Cultural Heritage, 133): 275-284, ag Bard), B. Hevline, J. Teebald, J. Durovié, M. 2013. “Eflores- cence on the surface of beeswax seals: A calorimettic study’, Thermachimica Acta 566: 292-297, Barth, B; Kobera, L, Drabkovi, K. Durovié and Brusc, J. 2015. “Wax bloom on beeswax cultural heritage objects: Exploring the causes of the phenomenon’, Magnetic Res nance in Chemisty 53: 509-513. Berger, G. 1995, ‘Inpainting Media and Varnishes Which Do Not Discolos. Pare I: Preparation for Inpainting’, Te pic. sare restorer 8:5: Bonaduce, l. Colombini, M.P Degano, I. Di Girolamo, F La ‘Nasa, J. Modugno, F and Orsini, 8. 2013, ‘Mas specto- rmettic echniques for characterizing low-molecu- lar-weight resins used as paint varnishes’ Analytical and Bioanaiptical Chemisty 405: 1047-1065. Broers, N. 2002, La restauration et la conserotion de la pein- ture sur cute, Exempleconcret de restauration dane eune ‘prince sur cuivre. Final report, Instieut Supérieur des Beaux-Arts, Saint-Luc, Lidge, Belgium. Broers, N. 2003. Paintings on Copper interactions between copper supports and the materials sed in their preparation dard paint Layers. MA Dissertation, Department of Arts, University of Northumbria, Newcastle, UK. Crowley, A and Laefer, D.E 2008, ‘Understanding, Micro- crystalline Waxes for the Seismic Protection of Are Ob- jeces Journal of the American Instcute for Conseraaton 47(2): 119-138, Laefet, D.R, Kennedy, J-E., O'Flynn, K. and Stancon, KT. 2015 ‘Structure ~ Property Relationships in Microcrystal- line Waxes for Museum Seismic Protection Applications, Frogres in Rubber, Plates and Recycling Technology 31 (4): 281-295. De la Rie, ER. and C.W. McGlinchey, C.N@, 1990. ‘New ‘Synthetic Resins for Picture Varnishes’ in Cleaning, Re- touching and Coatings: technology and practice for easel ‘paintings and polychrome seulpinre. Consributions t0 the Brusels Congress 3-7 September 1990, J. Mills and B Soni (eds.) 168-173, London: International Instcute For Conservation of Historic and Artistic Works. De la Rie, R. Lomax, S.Q., Palmer, M. and Aimes, CA. 2002. ‘An investigation of che photochemical stability of films of the ureaaldhyde resin Laropal A 81 and Laropal A 101’. Paper present at Preprints of che 13% Triennial Meeting of the ICOM Committee for Conservation, Vol Il, International Council of Museums, Rio de Janei- 10, 881-887. Ferreira, Joana L., Melo, Maria J., Ramos, Ana M. 2010. PVAe paints in works of art a photochemical ap- proach — Pare 1’, Pelymer Degradation and Stability 95: 453-461INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER, Fuster Léper, L. 2006, Euaudo de la idoncidad de las masilas de relena en e ateaniento de laginas en pintura sobre len- 20, Evoluciin histricotéeniea y andlisis ico-mecinico, PhD dissertation, Department of Conservation and Res- toration, Universidad Politéenica de Valencia, Spain Fuster Léper,L. 2012. ‘Filling’, in Conservation of ese paint- ings, Joyce Hill Stoner and Rebecca Rushfild (eds) 586- 6685. Uniced Kingdom: Routledge Gad, S. (ei) 2008. Pharmaceutical Manufectiving Hand- book: Production and Procewes, New Jersey: John Wiley & Sons Ine Garrel, C1992. ‘Procesos de restauracién en la pintura sobre ‘obec. Paper present at IX Congreso de Conservacién y Restauracién de Bienes Culturales, Sevilla, 17, 18, 19 y 20 de sepriembre de 1992. 154-163 Sevila: Secretaria del IX Congreso de Conservacin y Restauracién de Bienes Caleurales. Greywacz, C. M. 2006, Monitoring for Gaseous Polltants in ‘Meum Enviromens, Los Angeles: The Geeey Conserva- tion Instieute, Harley, C. 1993. A note on the crystal growth on the surface of a wax acifact Studies in Conservation 38(1): 65-66. Horie, CV, 2010. Materiak for conserucion. Organic consoli- dlant, adbesives and coating, 2" ed. London: Routledge “Taylor & Francis Group. HHorovic, 1. 1986. "Paintings on copper supports: rechniques, decerioraion, an conservation, The Conseruator, 10: 44-8. Horovie, I. 1996. The Consolidation of Princings on Copper Suppor, in ICOM Commitee for Conseracton, 11th Ten nial Meting, Fdinbuagh, Scotland 1-6 September 1996: pre- prin. J. Bridgland (ed) 276-81. London: James & Jes. Horovie, I. 1999. "The materials and techniques of Furopean paintings on copper support in Copperas Canvas: Tio Centuries of Masterpiece Paintings on Copper. 1575-1775. Phoenix Art Museum. M. Komanecky (ed.) 63-92. New York: Oxford Universicy Press Hotovie, 1. 2012. ‘Copper as a support for easel paintings’, in Conservation of Easel Painting, | H. Stoner 8 R. Rush- feld (eds) 99-106, United Kingdom: Routledge. lsayer, A. (ed) 2010. Encyclopedia of Polymer Blends. Volume 1: Fundamentals, Boschste; Wiley-VCH Verlag GmbH & CoKGat. Jiméner, JJ, Bernal, JL, Aumente, S., del Nozal, MJ, Marti, MT. and Bernal, |. 2004. ‘Quality assurance of commercial beeswax. Part 1. Gas chromatography-clec ‘won impact ionisation mass specttomerry of hydrocar- bons and monoeste’, Journal af Chromatography ALO24: 147-154, Kemp, J. 2009. Fils for the Repair of Marble. A Brief Sur vey, Journal of Architectural Conservation 15(2): 59-78, Leegenhock, 1. 1986, Les eableans peints sur eure. Origine, constitution, conservation. 3rd year project. Insicut Fran cise de Restauration des oeuvres ar, Pais, France ManSoori, G.A. Barnes, H.L. and Webster, G.M. 2003. "Pe- troleum waxes in Fuels and Lubricants Handbooks Tech nology properties, performance and testing, George E Tor ten; Steven R Westbrook and Rajesh J Shalt (eds,) 525- 556. West Conshohocken, Pa: ASTM Incernational Maines, CAA. De a Rie, R. 2005. 'Size-excusion chromatog- raphy and differential scanning calorimetry of low molec- ular weight resins used as varnishes for paintings, Progress inn Organic Coatings 52: 39-45. Mantel, B. 2004, Chemical Risk Anadis. A practical Hand- book, London: Kogan Page Science. Maule, L. 1999. The Gadarene Swine by Paul Brit 3% year project, Hamilton Keer Insitute, Cambridge University Unied Kingdom, Melneyre, C. 2011. Development ofa pigmented wares fil formation for the conseroation of paintings. MA Disset- tion, Art Conservation Department, Buffalo Sate Col- lege, New York, United States of América Mildenberg, R. Zander, M. Collin, G. 1997. Hdrocarbo Resins. New York: VCH Publishers Ine Mills, J. and White, R. 1999. The Organic chemisory of m= scum objects, 2" ed, Oxford: Butterworth-Heinemann, Mucea, E.E. 2007, “The Use of Wax-Resin in Conservation “Treatments of Gilded Surfaces’ Conserution. The online magazine 11: 83-98. Novotna, P, Demnovskova, J. 2002. ‘Surface Crystllsation com Beeswax Seals, Resteuntear 23(4): 256-268. Oliveira, M.L. 2015. A technical investigation ofan oi paint- ing on capper support, including astudy or consolidants for treatment. MA dissertation, Department of Conservi- sion and Restoration, Faculdade de Cignctas e Tecnole- gia, Universidade Nova de Lisboa, Monte da Caparica, Portugal Pacerakis, AB. 1996. ‘Conservation: preservation veisus analysis, Studies in Conservation 4\supl): 143-148, Parerakis, A.B, 2003. “The Influence of Conservation Trea ments and Environmental Storage Factors on Corro- sion of Copper Alloys in the Ancient Athenian Agora’, Journal of the American Instsuse for Conservation 42: 313-339. Pavlopoulou, L-C. 2004. Oil Paint on Copper, « Study of Decay Mechanisms. MSc Dissercation, Cardiff Univer sity. UK, Paviopoulou, L-C. and Watkinson, D. 2006, “The degrada- tion of el painted eopper surfaces, Reviews in Conseroa- sion 7: 55-65. 7]DANIEL VEGA ~ ISABEL. POMBO CARDOSO ~ LESLIE CARLYLE Phipps, PB. Band Rice, DW. 1979. “The role of water in umospheric eorcosion’, Corrosion Chemistry, American Chemical Society Symp. 89: 235-261. Pilon L.Z. 2008. Laterfcial Properties of petolewm product. Now Yorke Taylor & Francis Group. Regaltez™ 1094 Hydrocarbon Resin, Accessed 23 Occober 2016. Available ac: hep://www.eastman.com/Pages/Pro- ductHome.aspx?product~71049324 Regalrez’ 1126 Hydrocarbon Resin, Accessed 23 October 2016, Available ac: hup://www.kremer-pigmente.com/ edia/pdfl67284e pdf Rizao, L, 2008. Come restanrare i dipint su vol, tela, carte, rane, vetro, Roma: Edup Rowe, RC. Sheskey, BJ, Owen, S.C. (eds) 2006, Hland- book of Pharmaceutical Exeipients, 6 edition. Washing. con DC: Pharmaceutical Press and American Pharma- ceutical Assocation, Scott, DA. 2002. Capper and Bronze in Art— Corvesion,colo- ‘avn, conseration. Los Angeles: The J. Paul Getty Test Scort-Monetief, A. 1993. ‘Copper as canvas. Conservation and reconstruction of wo landscapes on copper atribut- cd to F P Ferg at Southampton Art Gallery’, The picture restorer 3: 20-1 Smith, G.D. and Johnson, R. 2008. ‘Seip “Tes - Solubilisy ‘Data for the Removal (and Application) of Low Molecu- lar Weight Synthetic Resins Used as Inpaincing Media and Picture Varnishes, WAAC Newsleter 30(1): 11-18. Tadros, TE 2005. Applied Surfactants Principles and Applica- ons. Weinheim Wiley-VCH Verlag GmbH &¢Co. KGaA. “Tilloch, A.P. 1980. “Beeswax-composition and analysis. Bee World 61:47 62. 1 vvan den Berg, J:D. 2002, Analytical chemical seudies on radi. sional linseed oi pains, Molart Report 6, PAD dissereation, University of Amsterdam, Netherlands Vega, D. 2015, Study of the Ground Ler in Oil Paintings on ‘Copper. Unpublished student project supervised by De Leslie Caryle and co-supervised by Dr Isabel Fombo, De- partment of Conservation and Restoration, Faculdade de Ciéncias ¢ Tecnologias, Universidade Nova de Lisboa, Costa de Caparica, Porcugal Vega, D. 2016. Oil painsing on copper: characterisation of the copper support and the feasibility of wing pigmented waseree {in nfl for pins loss reinegratin, MA Dissettation, De- partamento of Conservation and Restoration, Faculdade dde Cigncias ¢ Tecnologias, Universidade Nova de Lisboa, Monce da Caparica, Porcugal. Von der Goltz, M., Proctor, R.G., Whitten, J., Mayes, L. Myers, G., Hocnigswald, A., Swicklik, M. 2012. “Var- nish as part of the conservation reatment of easel paine- ings in Conseroation of easel paintings, Joyce Hill stoner and Rebecca Rushfeld (eds) 635-658. United King- dom: Routledge. ‘Warth, AH. 1956. The chemisy and tclmalogy ed. New York: Reinhold Publishing Co. ‘Webb, M, 2000. Leegue: technology and conservation; a com: _prebensive guide to the tecaology and consereation of Asian ‘end European lacquer. Oxford: Burcerworth-Helnemann Whitcen, J. 1995. ‘Regalrez 1094: Properties and Uses, WAAC Newsleter 17(0): 11-12. Zakipous, S., Tidblad, J. and Leygiaf; C. 1995. ‘Aeroepheric corrosion effecs of SO, and 0, on laboratory exposed cop pee’ Jounal ofthe Electrochemical Socesy 142(3): 757-760. of wases, 2INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COUPER 1a: Oi painting on copper, Virgin Mary, 17 century private collection, 1b: detail of che jn losses showing the exposed copper subs trate, The paint composite (pai/ground layer) is in the range ofc.30 jm which i simile to shat Fx ina scudy of 12 179 8 18" century 5) oil paingings on copper (V Appendix 1: Ingredient identification from Tables 1 & 2 Commercial products The ingredients in the following commercial prod uucts were not identified by the authors at the time of publication: Fine Surface Polly Filla; UVS Retouch Varnish; Rowney Acrylic, Liquitex Acrylic Gesso, Modos- tue, Leffane & Bourgeois Gel Relief, Lascaux Hy- dro-Grund 750, Tempera paint Conservation products listed and idemtfied in current lit cevaturefwebstes = Mowilith 20: Poly vinyl acetate resin based on low molecular weight solid vinyl acetate homopolymer [1] — B-72: Paraloid B. based on a copolymer of ethyl methaerylate/methyl acrylate (EMA/MA) in a ratio of 70:30 [2] — B-48: Paraloid B-48, ‘Thermoplastic acrylic resin based on butyl acrylatefmethyl methacrylate (BA/ MMA) copolymer (22-4496BA) [2] = BTA: Benzotriazole; (C,H.N,). a nitrogen heterocy lie derivative (3] 72, ‘Thermoplastic acrylic resinDANIEL VEGA ~ISABEL POMBO CARDOSO - LESLIE CARLYLE = Plextol B500: Acrylic dispersion [4] based on the copolymer EA/MMA (66% EA) [2] = Beva 371: Adhesive mixture based on the copoly- mer ethylene/vinyl acetate. Original composition: {60% solvents: toluene and naphta with 40% solids: «wo ethylene vinyl acetate copolymers (Elvax 150 re- sin, and A-C Copolymer 400), cyclohexanone resin, (Kerone Resin N or Larapol K 80), a phthalate ester of hydroabietyl alcohol (Cellolyn 21) and paraffin wax. [5,6]. This product produced after 2010 con- tains a substitute synthetic resin for the cyclohexa- none resin and is referred to as Beva 371b [7] Notes 1. Kremer product information sheet for “Polyvinyl Acetaze 20" Ac- ‘esed 2] July 2016. Available at: chep:/Iorewkremerpigmene ‘om/media/pdfi67000e p> 2. Down, J. 2015, The evaluation of selected poly(vinyl acetate) and acrylic adhesives A final research upeat, Stain Conservation 6011) 33-54 43, Sease, C1978, ‘Benzowianle: a review for conservator, Studies fn Conservation 230): 76-85, 4. Hove, C. V, 2010, Materials fr Comereaton: Organ Conol- dents, Adhesives and Coatings, 2nd od. Oxford: Burterworth-Hei- 5. Plocger,R, De la Rie, E, McGlinchey, C., Palmer, M., Maines, CA, Chiantors, 0. 2014, "The long-term seabilry of a popular heatsal adhesive for the conservation of pained cukural objec Palymer degradation and stabiliy, Pohmer Degndarion and Sa bility 107): 307-313, 6, McGlinchey C. Ploeger, R. Colombo, A. Simonucti, R. Palmer, M, Chianore, O. Procter, R. Lavédrne, B, De la Ri, E. 2011 “Lining and Consolidations Adhesives: Some new developments and areas of facure researc. Paper present a the Proceeding of ‘Symposiun 2011 ~ Adhesives and Consolidans for Conserv tion: Reseach and Applications. October 17-21 October. Cane ca: Orava 7. Bera teormulazed formula, Accessed 23 October 2016, Available at) chtspnww.consersaionsupportaystems.comystem/ asset rmuh/New Beva Formula> 1 Appendix 2: Analytical instrumentation and conditions Fourier Transform Infrared Spectroscopy (i-FTIR) Infrared spectra were acquired using a Nicolet Nex us spectrophotometer coupled to a Continumm micro- scope (15x objective) with a MCT-A detector cooled by liquid nitrogen. The spectra were collected in trans- mission mode, beeween 4000-650 cm-I, resolution set- ting 4em-1 and 128 scans, using # Thermo diamond anvil compression cell. The spectra are shown here as acquired, without corrections or any further manipula tions, except for the removal of the CO absorption at ca. 2300-2400 cm-I. FTIR- Attenuated Total Reflectance (FTIR-ATR) ‘The FTIR analysis was performed using an Agilent 4300 Handheld FTIR Spectrometer with a diamond ATR head . Spectra were obtained covering the 4000- (650 cm-I range, with spectral resolution of 4 cm-1. The collected spectra were obtained with 64 scans before Fourier transform. Dara was acquired by MicroLab™ software and data was matched by Polymers and Poly- mer Additives P/N 30002 ATR-PTIR Spectra database, from S.T. Japan (Europe Gmbh).INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER, Appendix 3: FTIR analysis Absorbance Absorbance 40007 3800 3000” 38607000 1800 1000 aoe 7260000 ato ‘Wavenumber (cmt) Wavenumber (cm’) Big A-1.-FTIR-ATR speccrum ofthe wactesin KTW5:R2and ofthe puze materials: micros stalline Techniwa 19426 (KTWS) and Realtez 1126 (R2). Specta show the presence of hydrocarbon bonds only 8 8 i i i : g g = < Stet = | 4000 3800 3000 28002000 1500 1000 1800 1600 1400 1200 1000 800 Wavenumber (cm) Wavenumber (cm“*) Fig. 82. jc FTIR spectrum ofthe unpigmented wax-resin from Carle (C-PWR) with reference specs of besa and of cyclohexanone sein. The specea show the presence of carbonyl groups associated to ketones (in poly ‘yelohexanone) and estes (in beesw), 21 |= p DANIEL VEGA ~ ISABEL. POMBO CARDOSO — LESLIE CARLYLE Booswax aT Jt 003 sab So Beka 00 00 O80 1066 0 —io5 08 Tae8— 30 ‘Wavenumber (em™") Wavenumber (cm) Fig. 83, FTIR spectra from Gamblin Pigmented Wax Resin Seick, Neural Base G-PWR and reference spectra of beeswax, aspect ofthe microcrystalline wax Techniwax 9426 (KTWS}) for ‘comparison and a reference spectra of Latopal A 81. Spectra show che presence of exrbonyl groups sociated with aldhydes (in Laopal A 1) and esters (in beeswax) as well as unsaturated bonds (in Laropal A)INVESTIGATION AND TESTING TO DEVELOP AN INFILL FORMULA SUITABLE FOR OIL PAINTINGS ON COPPER, Appendix 4: Equipment and suppliers controlled speed, making the process of obtaining. an homogeneous mixcures more efficient. Wax carving pencil: This unit was originally designed to carve and mould a variety of waxes used in the Equipment — Digital precision balance: Precisa’, model 500M 20006 — Magnetic Scirrer wich Heating: Heidolph*, model MR Hei-Standard. This laboratory device employs @ rotating magnetic field to cause a stir bar immersed. in a liquid (contained in a glass vessel) co spin under dencal industry. The variety of carving tips and easy temperature control allow for precise waxing work “Temperature range is berween 50 to 200°C. For using with PWR, ic is available through Gamblin (www. gamblincolors.com/conservation-colors/) Laropal” ASI (L1) Laropal” A110 (12) Pigments: Ticanium White and Burnt Umber araloidTM B-72 MATERIAL (PROJECT CODE IN BRACKETS) bahdieetaiy ae “Techniwax samples 9211 (TW) mae 9265 (KTW2) ms 9300 (TW) worwikerax.co.uk Keecncdants 9356 (KTW4) . ss 926 (KTW5) ‘Gamblin Pigmented Wax Resin sticks (G-PWR) | Gamblin Conservation Colours vowgamblinclors.comfeonssrva- 20089 and 2015 tion-colors Regalo? 1094 (RD) Regalree" 1126 (R2) Kremer Pigmente GmbH & Co. ais KG igmentede Paincings laboratory supplies Potassium hydroride BioXtra, »85% KOOH basis (P5958 FLUKA) thloric acid ACS reagent, 379% (320331 FLUKA) Phenolphthalein [ACS reagent (105945) 96% ethanol Paris. Pa., ACS reagent, reag. Ph. Bur, 96% (o/s) (2294 RIEDEL-DE HAFN) Sigma-Aldrich vwww.sigmaaldrich.com FCTScientfic lb FCT-PhotoChemistry Lab Xylene Reagent grade (214736 ALDRICH) Paincings laboracory supplies Fine sable-haired brush ‘Coiman Watercolour Round III Series Winsor & Newton Paintings laboracory supplies 209 |DANIEL VEGA ~ ISABEL POMBO CARDOSO ~ LESLIE CARLYLE Appendix 5: Standard ‘Test Method for Acid Number (empirical) of synthetic and natural waxes “The standard D13865, published by the American Society of the Association for Testing and Materials (ASTM), was followed for this test. Information from. the standard has been extracted here: Scope: — This test method covers the determination of che acid ‘number of synthetic waxes and natural waxes. The number is obtained by direct titration of the material and indicates the amount of fiee acid resent. = This test method, using an ethanol-xylene mixture, is applicable to all natural waxes, including carnau- ba. ‘The test method is also applicable to oxidized microcrystalline waxes, oxidized Fischer-Tropsch, oxidized polyethylene, and montan estes. = The values stated in ST units are to be regarded as standard, No other units of measurement are includ- ced in this standard. Significance and use ‘This rest method is used to determine the property of acid functionality. Acid functionality determines the utility of the wax as well as being a significant Quality Control test. Reagents and materials — Purity of Reagents: Reagent-grade chemicals or equivalent, as specified in Practice E200, shall be use in all tests. = Ethanoic Potassium Hydroxide, Standard Solu- tion: Dissolve approximately 5.6 g of potassium hy- droxide in 5.6 g of distilled water. Dilute with USS- D3A denatured ethanol or 95% ethanol ro 1000 sm. Standarize with 0.1 N hydrochloric acid. — Phenolphthalein Indicator Solution (10 g/litre): Dissolve 1 g of phenolphthalein in 100 mL. of USS- DBA, denatured ethanol ot 95% ethanol. ~ Xylene Procedure 1, Transfer 1 co 2 g ofthe sample, weighed to the near «st 0,001 gro a 250-mL acid-value flask. Add 40 ml 1 of xylene. Heat on a hor plate or water bath co put the sample into solution. Occasional swirling may be necessary. 2. Add 3 to 5 drops of phenolphthalein indicator solue tion and titrate the hot solution to the frst persistent pink colour. The end point is taken when the pink colour remains for at least 10s. Swit! the flask vigor- ously during the tiation. If precipitation of waxes ‘occurs during titration, reheat the sample, The ti- tration should be carried out as quickly as possible Record the number of millilitres of standard alkali solution used. Warning - to avoid saponification, do not reheat che solution during chis operation. 3, Determine the blank titration value by repeating the procedure without the addition of sample material 4, Run a total of three titrations of sample being tested and one blank. Caleulation Calculate the acid number as follows: Acid number = (A ~ B) x Nx (56.1)! A= millilitres of alkali solution required for titration of the sample, B= nullilicees of alkali solution required for citration of the blank sample, N= normality of the alkali solution, and C= grams of sample used. Notes 1. Dr Barth at The Insitute of Chemical Technology (Prague, has sued bloom development on beeswax in natural ageing condi: tions by storing simples ar 4 fora minimum of 3 years (2015, 2013, 2008). 2, Personal communication Dr Carlyle, 2015. Barbara Klempan in- troduced a PWR formula to paintings conservators at CC afer hae ining in Copenhagen. [3 Robere Gamblin kindly shared this information withthe authors to asi in the curent study 4, These fngments, 224 x 38 mm and 430 x 50 mm respectively were donated frm a private collection for study (Vega 2016). 5. The edition used was approved and published in Occober 2015) Originally approved in 1995 as D1386-55T. The most recent previous approved standard was published in 2010 as D1386.