 So hello, everyone. I'm very happy to be here today to present the research I completed as my thesis project at Queen's University on the aging properties and potentials of materials used for scratch repair on the polymethyl methacrylate support of face-mounted photographs. Now, Clara already did a wonderful overview of face-mounted photographs, so I'll just do a very simple recap. In the most basic sense, face-mounting is a finishing process for dye coupler or inkjet-printed photographs. Patented in 1972 by Heinz-Suvila Brilhardt in Switzerland in a process called diasec, it eliminates the need for heavy and hazardous glass frames by irreversibly adhering the image directly to the backside of a polymethyl methacrylate, or PMMA, acrylic sheet. This licensed method uses a moisture curing silicone sealant as an adhesive to set the photograph in the anoxic environment between the photographic emulsion and the acrylic. There are, however, some other unlicensed face-mounting methods that use different types of adhesives or transparent double-sided adhesive films of varying thicknesses to secure the photograph to the PMMA. In general, face-mounting allows the photograph to be adequately protected from harmful deteriorating agents, such as ultraviolet radiation, airborne pollutants, fingerprints, and mold, while the rigidity of the PMMA can allow the print to be displayed without a distinct frame, thus opening the image to the viewer with no subconscious barrier. Commonly used by many contemporary artists to showcase their work, face-mounting allows the colors from the photograph to seemingly permeate fully throughout the acrylic, giving the completed work an absorbing and wet look that the viewer can fully enjoy. In terms of general research on face-mounted photographs, many authors have already given great focus to the subject. In 2004, a global face-mounting initiative comprised of many member institutions and partners was set up to establish guidelines for the preservation and conservation of face-mounted photographs. Some research has been dedicated to the image layer of the print, categorizing and analyzing different types of damage that can occur, ranging from slight to severe, from surface contamination to damage from transport, as we all saw in Alex's presentation. The effect of these problems has been carefully and meticulously described in many publications. Also documented have been experiments designed to monitor the effect of different types of storage in gallery settings on the aging characteristics of these prints. As they are a combination of different materials requiring specific storage conditions, these experiments have been vital to the safekeeping of these delicate and sometimes quite large and obviously expensive works. Some studies have even tried to measure the extent that the adhesive contained in these prints will off-gas and, in turn, how it can affect any degradation and color changes that occur. Otherwise, certain authors have focused their attention on the rigid surface finish. As PMMA is so easily scratched, abraded and susceptible to damage from certain solvents, it has become a constant problem for institutions wanting to display, maintain and care for these prints. Researchers, like Clara and her team, have put many materials through a wide range of tests to examine the effect that different combinations of cleaning products and cleaning methods will have on the acrylic. Unfortunately, their tests were too successful, demonstrating that no matter the product, there will always likely be some small level of scratching produced on the surface. Since the finishing process irreversibly adheres the photograph to the PMMA, the acrylic sheet becomes inseparable in the viewing and interpretation of the object. While the acrylic can sufficiently protect the photograph from physical harm, the susceptibility of the PMMA to abrasions will cause just as many distractions to the preservation and presentation of these prints. When the acrylic is severely scuffed or cracked, the print in its entirety would need to be fully replaced, ultimately becoming quite an expensive cycle, especially since polishing away a scratch on the surface isn't always a viable option. Building off the work that has previously been done on scratches produced by cleaning brings us into the reason for this research, as my original focus lent towards reducing the visibility of scratches on the surface. A scratch on a smooth acrylic is composed of a depressed center trough and rough and uneven ridges created by the displaced acrylic material. When the raised ridges on either side of the trough reflect light haphazardly, the scratch is visually disturbing. By modifying or reducing the scratch ridges, as has been stated in other research, and filling the trough with a material that has similar properties as PMMA, the visibility of these scratches should be reduced. The problem here lies in finding a film material that will visually act as PMMA does with time with no adverse deterioration and reflect light in a similar way. Fortunately for me and my research, this intention is not a new idea, as quite a few researchers have already begun investigating the idea of using adhesives as film materials. Each adhesive that has been tested was chosen specifically based on a number of factors. These are, and a number of these materials are on the screen for you to see. In order to replicate a PMMA surface, the ideal material should have a similar refractive index, surface gloss and color tone, or a lack of color, and ultimately be invisible over time. Furthermore, to be an ideal filler, the material should be relatively easy to manipulate and be able to flow readily into the very narrow and minuscule surface area of the scratch. While considering all these factors, certain materials continue to produce a favorable range of positive properties and results, as well as demonstrating challenges that still need to be faced with application, creating further areas to explore. The experiments conducted by previous conservators and scientists therefore allowed me to narrow down my potential fillers to three materials that would undergo even more vigorous testing in my experiments. My film materials, an acrylic co-polymer, a two-part epoxy and a UV curing adhesive were chosen and tested based on their ability to replicate an unscratched PMMA surface after an aging process that simulated museum storage and gallery exhibit conditions. Building on previous research, I examined the characteristics of these materials when used in conjunction with surface modification techniques that try and reduce the visibility of scratches on PMMA. An initial pilot study analyzed the covering ability and working ease of each of the three materials. Half of a deep scratch area on PMMA was blocked with cellophane tape, that's the left section of the images, while the other half was covered with the material. Paralloid B72, a common material in conservation labs, is a co-polymer of ethyl methacrylate and methylacrylate and has a refractive index of 1.49, the closest of the three to PMMA, which is 1.4899. Since many organic solvents can cause crazing in PMMA, the B72 in my experiment was a 20% mixture in one to four hexane and toluamine. This combination of solvents creates a very thin solution and allows the B72 to flow into the surface scratches. And while these solvents are safe for PMMA, they have dissolved the adhesive in the tape, which is why it has bled out onto the area, as you can see. And yet the covering ability of the material is still quite strong. It has proved to be a worthwhile scratch reducer except for application problems due to quick solvent evaporation. The second, hexadol, is often a first choice in terms of glass repair and conservation. Here, the low viscosity of this compound again makes it ideal for flowing into thin scratches. With a refractive index of 1.52, hexadol was able to closely match and cover minor scratches in the PMMA. However, remaining bubbles from mixing the epoxy appear quite noticeable in the deep scratch. Finally, with the lowest viscosity and longest working time, dimax, the UV curing urethane acrylate-based decorative adhesive coating was expected to provide the greatest covering ability of the three materials. Though it was easy to work with since it wouldn't harden until cured and easily flowed into the thinnest scratches, the curing process itself proved to be more difficult than expected since I was unable to acquire UV lamps with enough power. Therefore, the curing time went from the expected 20 seconds to over an hour of UV exposure, resulting in dust attracting to the surface. Even with the incredibly long exposure time, the dimax with refractive index of 1.5 was still able to cover all minor scratches and a good portion of the deep scratch. Even with, oh, I already read that. Overall, the three materials continued to present good initial covering ability and visual scratch reduction prior to introducing any surface modification techniques. My experimental samples were braided using a custom-made mechanical rub test machine that was designed for a previous student research project. A small weight with sandpaper at the PMMA interface was passed over the surface of the samples to produce relatively uniform scratches, ranging between 50 and 100 microns in width. The rub test machine allowed me to create gentle scratches and gentle and relatively shallow scratches that emulated damage from repeated cleaning rather than the deep gouges that could be caused from transportation or handling disasters. As you can see from the slide, the scratching produced rough ridges on the scratch troughs, around the scratch troughs that reflect light and become visually distracting from the smooth surface. In order to alter the surface characteristics of the scratch ridges and reduce their overall visibility, I chose to introduce two surface modification techniques on my samples. A flattening strategy involved the localized application of heat with a microscopy roller to soften and reduce the scratch ridges. By heating the displaced PMMA, the ridges were smooths and any sharp edges were reduced. The second technique, a scraping method, quickly and easily created a smooth surface over the existing scratches. Using a razor blade that was pulled mechanically over the surface, the ridges were quickly and easily removed entirely to create a surface theoretically free of any sharp, light-refracting material. Both of these methods altered the way that light was reflected and demonstrated how the reflection of light from the scratch ridges plays into the visual disturbance of scratches on the surface of face-mounted photographs. After initial sample preparation, each of the fill materials was applied to a sample in each surface modification category, creating nine areas of comparison. These scratch unmodified samples were braided using the rub test machine and the fill material was applied. These scratch-flattened samples were scratched and underwent flattening prior to adhesive application and the scratch-scraped samples were similarly abraded, scraped, and then coated. After adhesive application, the samples were subjected to thermal aging for an equivalent age of 100 years in museum storage and then to light aging to mimic 100 years in gallery exhibition. To calculate accurate degradation factors, the thermal aging times were based on the thermal degradation activation energy of each material. Analysis of the samples was completed at each stage of the experiment with a spectrophotometer to detect any transition in B star values, that is, any change from blue to yellow, and a gloss meter that measured the change in surface roughness after the modification techniques. The success or failure of the fills was evaluated based on the ability of the fill material to replicate the original PMMA surface through color and gloss and visually reduce the surface scratches. Now, with all the testing and analysis that took place over several months and my experiments, I was left with a great deal of numbers, unit values, and graphs. Therefore, to make it easier on everyone, I'm only going to present some of my results here. In terms of the results for color change, this graph illustrates the average transition in B star for the change in the level of blue for each of the chosen adhesives acting as fill materials over the surface of the PMMA. The graph on the screen indicates data levels from the beginning of the experiment after thermal aging and finally after light aging. The blue line demonstrating changes in the unmodified PMMA itself shows a slight increase in the amount of blue. However, since the change of less than one is considered negligible and unnoticeable, it is unnoticeable by the naked eye. In regards to the adhesives themselves, the parallel B72 had the least change in blue throughout the experiment, while the dimax had the largest blue decrease, mainly apparent after light aging. Hexadol also became less blue, however, less so than the dimax. I should mention here that a decrease in blue doesn't necessarily mean a transition to yellow unless it's so severe that the value becomes negative, in which in those cases it hasn't. When comparing the final B star values at the end of the experiment, the dimax displayed the closest final color to that of the unscratched PMMA followed by the hexadol. The parallel B72 resulted in a final color most different from the PMMA, however, these final values are still within the error range and are not considered noticeable by the human eye. In terms of surface gloss, the resulting effects of the modification techniques are about as graphically apparent as the preliminary color changes. The chart on the screen demonstrates the extent to which scratching in red specifically reduces the surface gloss of PMMA, as shown in blue. The extensive gloss decrease with scratching is slightly improved with the flattening treatment, shown in green, and even more so by the scraping method in purple, which does actually decrease the surface sheen. Oh, sorry, it increases it, if only by a small amount. As theorized, this can be described by the effect each treatment had on the scratch ridges. By smoothing out the ridges, the surface became more uniform and the ridges diffracted light more evenly, thus reducing the visibility of scratches. While removing the ridges completely, created a flat and highly reflective PMMA surface from which there was no sharp flakes to create any visual disturbance. While these results can be more visually noticeable, the compiled data statistically demonstrated that any modification is fairly indistinguishable from an unmodified surface. When examined with a stereo microscope, the effectiveness of the surface modifications when paired with the fillers becomes more apparent. On the slide, I have inserted some images at high magnification to better illustrate the covering ability of the materials over surface scratches. The scratch unmodified and scratch-flattened DMACC samples were the most visually effective in reducing scratch ridges. Though the B72 scratch-flattened samples, as well as some of the Hexdahl samples were also effective scratch reducers on the more shallow abrasions, unfortunately, when considering each sample set, there didn't seem to be a method or material that continuously reduced the visibility of scratches in the same way. For the purposes of my experimentation, I'll conclude by saying that the samples in my research had no perceptible color change in any of the tested fill materials after accelerated aging. However, the Paralloid B72 produced a noticeably matte surface while the Hexdahl became even glossier than the unmodified PMMA. Due to the aforementioned curing problems with the DMACC, too much dust had accumulated on the sample surface to be considered effective in reducing visual disturbances on the PMMA. Redoing this experiment with correctly cured DMACC samples would be essential to discovering its potential as a substitute acrylic surface. In this experiment, the filled scratches were always somewhat visible, even though the refractive index was similar to that of PMMA. Possibly due to internal reflection within the scratch itself, as stated in most of the other research, handling an application of these materials was still a problem. While the low viscosity materials could flow easily into the scratches, regulating the desired fill amount to stop overflow onto the surface was difficult. The brush absorbed too much adhesive to create a light enough flow and the syringes were too large and the material too viscous to be able to apply correctly. In regards to the surface modification techniques, both flattening and scraping away the ridges resulted in an increase of uniform light reflection and a corresponding increase in surface gloss. While neither of these techniques exactly replicated a piece of unscratched PMMA, the scraping treatment by removing the ridges completely created a flat surface without large material disruptions. While the lack of great changes within the materials themselves presents these adhesives as good candidates for future fill material research, the experimental stage still contains too much risk to be able to apply successfully to works of art just yet. I would like to acknowledge all those involved in my research, process from Queens University and at the Royal Bridge Columbia Museum, as well as my colleagues at the Art Institute of Chicago and those responsible for planning this meeting who have helped make this presentation possible. Thank you.