Special Post – Microfading Workshop!

On June 26th 2014, Project Conservator Julia Sybalsky and Graduate Intern Gisella Campanelli attended the Microfading and Light Management Study Day, a half-day workshop and demonstration organized by Paul Whitmore at Yale University’s Institute for Preservation of Cultural Heritage (IPCH), and sponsored by IPCH and the Andrew W. Mellon Foundation.

After catching an early morning train from New York to West Haven, Connecticut, we arrived at Yale’s West Campus Conference Center for the Microfading and Light Management Study Day, hosted by the Institute for Preservation of Cultural Heritage. We enjoyed two presentations. The first talk, presented by Paul Whitmore, summarized the basics of microfading: what a microfade tester does, how it is constructed, and how and why one might use one. In the second, Bruce Ford, presented case studies from the National Museum of Australia in which microfading informed the revision of overly restrictive lighting guidelines. The result of his study was a cost-effective policy that protects light sensitive materials and materials of special significance while increasing access to collections in general.

These presentations were followed by an informal tabletop demonstration of the microfade tester. Paul gave us a comprehensive tour of the instrument and explained how to interpret the spectral data, whose stability is the measure of the relative permanence of the color. Tests on several of the inks from a set of “permanent” marking pens made the point that the label is not always an accurate description of stability.

A selection of topics covered in the course of the day’s presentations and demonstration are summarized below.

Background

Light has the potential to cause irreversible damage to objects. It can be especially deleterious to colorants, resulting in visible fading. Conservators understand that light-sensitive objects have a finite display life. Nowadays, collecting institutions often adopt guidelines to indicate the lighting conditions and durations of exhibitions of classes of objects (e.g. oil paintings, watercolors, metal sculpture, etc). These policies assume that objects within a class share the same, known stability. Some of those assumptions have been brought into question.

FadeExample_woodblock

Two copies of a Japanese woodblock print by Utamaro. When they were created, they were genetic identical twins, made from the same materials applied in the same way, probably by the same people. Then the two copies were separated at birth, with one image experiencing significant fading (by being exposed to light). Such experiences demonstrate the profound change in the object from light exposure, but also show that some colors, like the green on the kimono, are not very light-sensitive. Knowing the stability of the colors on an individual object, which may have no identical twins from which to gain information, is the challenge that microfading addresses.

In his presentation, Bruce discussed the way in which museums have traditionally managed this ‘display-destroy’ dilemma (table 1).

Table 1. Lighting requirements for display across three institutions

Institution Light level (lux) Display/Total (years)
Tate 80 4/8
V&A 50 2/10
CIE157 50 1/10

He argued that these guidelines are essentially based on guesswork and assumptions, and challenged this dogmatic approach by raising questions such as: How efficient are these standards? Are they cost-effective? Do they limit public access to objects of significance? Are they counter-productive? Do they exist in order to protect artifacts or to protect conservators?

Bruce proposed that the Micro-Fading Tester (MFT) can be used as a reliable tool to predict a colorant’s rate of fade over time. This information, along with an appreciation for the object’s significance, public demand, and aesthetic value, will better inform decisions about that object’s exhibition.

Sensitivity_significance

A structured significance evaluation can help prioritize conservation efforts and improve communication between conservators and other museum professionals. In this example, significance is used in a risk assessment aimed at assessing an item’s likely future demand for exhibition as well as its importance to the museum. Ref: Russell, R. and K. Winkworth. 2009. Significance 2.0 – a guide to assessing the significance of collection. Collections Council of Australia Ltd.

Microfading

Some colorants are more sensitive to light than others. But how do you know the sensitivity of a pigment or dye used on a specific artifact? The conventional ways to assess fading risks to objects come with a price. One way is to exhibit  the object and watch to see if its colors fade, either to the naked eye or by tracking the changes with a color measuring device. This approach obviously risks incurring some degree of fading in order to provide the information about the sensitivity to further light exposure. A second approach is to identify the materials present on the object, and to replicate those materials in all their detail in a sample that can then be exposed to light and evaluated. The challenge here is to identify the composition of the colored materials in every important detail, which can be very difficult for natural colorants which may have altered over time. Replicating such a substance is also not usually practical. This approach then becomes difficult and time-consuming, and it is impractical for application to a large number of objects. The last approach is to do spot testing, to shine light on a small area of color and measure whether the color is altered by that light exposure. That is the approach taken in the microfading tester, and the device has the added essential feature that it can be done so sensitively that the tested colors are not changed visibly: it is essentially nondestructive, so the presentation surfaces of objects can be tested safely.

The MFT delivers a focused 0.3mm beam of high-intensity light from a xenon lamp onto a tiny spot on the surface of the object and progressively measures any color changes that take place. In doing so, it allows a quick determination of whether an object contains light sensitive materials without need for their prior identification. The machine has two fiber optics angled at 45° degrees to the test surface. One fiber optic supplies the incoming beam from the xenon lamp. The incoming light is filtered to remove both infrared (IR) and ultraviolet (UV) radiation, which cause damage that is usually minimized in museum and gallery lighting environments. The object absorbs some of this incoming light, while the rest is reflected into a spectrophotometer via the second fiber optic. The spectrophotometer separates the wavelengths in this reflected light and produces a spectrum. As the spectrum is continuously updated during the test period, color measurements are recorded and used to calculate. ∆E, a measure of color change since the initial spectrum was collected. Any non-zero ∆E values indicate that the colorant being assayed is being altered by the light delivered. If the color does exhibit a color change during the test, the rate of that change is compared to the rate of fading of  Blue Wool standards, cloths whose fading rates are known and have been adopted as the international yardstick for color lightfastness. There are eight cloths in that scale, with Blue Wool #1 being least stable and Blue Wool #8 the most stable. “Fugitive” colors are considered to be in the Blue Wool #1-#3 range.

MicrofaderMounted

A microfading tester in use, measure the light stability of the colors in a lithograph. ©Paul Whitmore.

This incoming beam delivers light up to fifty times more intense than sunlight, exposing the tiny test area to a light dose equivalent to what the average museum object might receive over 5-10 years. By comparing the rate of increasing ∆E values in the test colorant to rates for Blue Wool standards exposed in the same way, the conservator is able to make informed predictions about the future behavior of the test material over time: will it retain its color for a long time like a Blue Wool 8, or is it more fugitive than a Blue Wool 1? To fully understand an artifact’s light sensitivity, each color on the object should be tested, and the exhibition policy for the object based on the stability of the most sensitive color.

Textile with emitted light from microfader visible on the surface.

Photomicrograph showing size of area tested with microfader light beam, here being directed onto a coarse-woven textile.

Microfade testing is considered an essentially nondestructive technique, as it does not leave a visible mark on the object. Not only is the test area minute, but more importantly, color change is monitored in real time by the operator and the experiment can be stopped if it looks as if visible change might occur. Meaningful results can be obtained without exceeding a ∆E value of 5, while we are only able to visually detect color differences in these tiny spots when the changes are larger than about 15. Damage can be avoided by observing ∆E values in real time throughout the duration of the test. The test usually runs for five minutes, but should be stopped earlier if ∆E approaches 5. There can be a temperature change of 5-10°C within the test location. This is usually inconsequential for most objects, but may require consideration for waxes, plastics, or other materials with low melting points. Paul explained that such test sites will often re-solidify on cooling without any visible damage, but the results of the test would not be reliable measures of color stability.

Benefits of Micro-fading

The advantage of using the MFT over traditional practices is that it eliminates the need for guesswork because it is measuring the sensitivity to exhibition lighting of colorants that are present on the actual object, rather than a simulation. Objects that can be micro-fade tested include paintings, prints (even through glass), manuscripts, photographs, 3D objects, textiles, and furs/taxidermy – basically any surface that one can make a color measurement on.

Fader_manuscript

Microfade testing at the National Library of Australia. ©Bruce Ford.

Fader_taxidermy

Microfade testing of taxidermy at the Horniman Museum and Garden, London, UK. ©Bruce Ford.

Fader_thru_glass

Microfade testing of Japanese prints, through glazing. Otago University, New Zealand. ©Bruce Ford.

Bruce explained that prior to revision of their lighting guidelines, the National Museum of Australia (NMA) typically displayed materials thought to be light sensitive for 2 years per decade. After a subsequent ‘resting’ period, objects would be returned to display, and the cycle repeated. Through his studies using micro-fading, Bruce was able to determine that numerous objects presumed to be light sensitive were in fact more stable than the 2 years exposed in 10 restriction implied (and could therefore be displayed for longer periods), while for a smaller number of items the same restriction was too generous. By measuring fading rates, very fugitive colorants were able to be identified and better protected without the need to restrict access across the board. By being more selective about restrictions using a combination of microfading and a structured significance assessment, the museum has been able to save thousands of dollars in the cost of exhibit changeovers.

NMA_savings

Illustration of potential cost savings using exhibit guidelines revised based on microfade testing. Green = extended display, yellow = unchanged display guidelines, red = displayed less than 2 years per decade.

The MFT can also be used in an anoxic environment to determine whether display in such conditions will have any appreciable benefits for the object in terms of color preservation. Some colorants may fade more rapidly in the absence of oxygen, for example Prussian blue and iron gall inks. Interestingly both of these examples undergo a reaction with oxygen which opposes the light-driven color loss but which cannot be measured using microfading or any other accelerated light exposure method. Again, this demonstrates how under-informed assumptions about reducing light damage can be unjustifiable and costly.

Conclusion

Data from microfading tests can supplement other information about the artifact to better inform decisions impacting an artifact’s display life. Using an evidence-based approach to manage display policies can prevent wasteful use of precious funds. Such resources can be redirected towards more needy causes.

How to Access a MFT

MFTs are available for purchase as kits at around $20K. Alternatively, one may contact a nearby colleague or institution that owns one to inquire about arranging for access. Currently, there are 24 MFTs in use worldwide. More information is available through Bruce Ford’s website, www.microfading.com.

Advertisements

What’s the Plan?

The research team had a good sense of the relative lightfastness of the Orasol dyes as a result of the renovation research, but how long could we expect the dyes to last inside an actual diorama environment? How could we better understand the ease of removal of the dyes from the animal hairs – is there a way to manipulate the dyes’ ability to penetrate or “fix” to the substrate? Could the dyes potentially cause the hairs to degrade faster inside the harsh diorama environment, or do they block the light and slow down damage – or both?

These questions acted as the foundation to the current collaboration and the focal point of the project research. Since beginning the project last fall, the majority of the work to date has focused on questions concerning lightfastness. The research team has decided to narrow future investigation to the Orasol line of dyes, rather than bringing new materials into the research plan. While the research team also looked into water soluble dyes, fiber reactive dyes, other metal-complex dyes, and some acid dyes, we were the most interested in applications that minimized the use of water.  Many of these alternative colorants are water-soluble and not appropriate for use on water-sensitive materials, require rinsing to remove excess dye and dye-bath residues, or their appearance during application does not represent their final color. The Orasol dyes remained the most promising among the materials available, as they lack the major drawbacks of some of the alternative colorants. The experimental work began in late summer 2014, starting with a large amount of sample preparation.

Lightfastness and Measuring Color Change

The first phase of the research plan tackles the question of how resistant the dyes are to fading – determining the lightfastness according to established standards (ASTM D4303) when applied without a binder or substrate that might interact with the dye. The testing done during the renovation project used wool flannel, an unstable substrate that also can discolor when aged, which could complicate being able to detect color change of the dye. For our current round of tests, we are replacing the wool flannel substrates with quartz plates, an inert substrate not expected to interact with the dye. The plates will be prepared according to methods used for testing in the paint industry – we’ll get into more detail about sample prep in future posts. Lightfastness test results will allow colors to be ranked and ASTM lightfastness ratings to be assigned to each dye color.

Intern Ersang Ma preparing dye for application to quartz plates

Intern Ersang Ma preparing dye for application to quartz plates.

Additionally, we will be developing an aging protocol that mimics ASTMD 4303, but more closely reflects the UV-filtered light conditions, temperature, and RH that are recorded inside of the AMNH habitat dioramas. This will give us a ranking of the dyes when exposed to an environment that is more similar to what the taxidermy is experiencing inside the diorama.

150  quartz plates with dyes dissolved in various solvents and airbrushed onto quartz plates for the first round of lightfastness testing.

150 quartz plates with dyes dissolved in various solvents and airbrushed onto quartz plates for the first round of lightfastness testing.

On going color measurements throughout both tests will let us better characterize the rate of color change – then we can give projections about how the dyes will hold up over time in our diorama lighting environment. Our initial light dosage calculations indicate a rough equivalence between 1 day of exposure in the environmental aging chamber at these test standards, and 1 year inside the most brightly lit dioramas.

We will also be looking into what impact, if any, solvent choice has on lightfastness and the rate of color change. Early testing during the renovation project suggested that not only can the color appearance of these dyes differ when they are applied in different solvents (as illustrated in the images below), but that the solvent choice may also affect lightfastness.

A 1% solution of the Orasol Red BL dye in MEK (methyl ethyl ketone). The dye is unevenly distributed and with distinct droplets. On the macro level, the swatch is a dark mauve.

A 1% solution of the Orasol Red BL dye in MEK (methyl ethyl ketone) applied to wool textile. The dye is unevenly distributed and with distinct droplets. On the macro level, the swatch is a dark mauve.

A 1% solution of the Orasol Red BL dye in methanol. The same dye delivered in methanol results in a more level film and a brighter color; the darkening you see is a product of higher hairs catching the airbrush more and getting more deposition.

A 1% solution of the Orasol Red BL dye in methanol. The same dye delivered in methanol results in a more level film and a brighter color; the darkening you see is a product of higher hairs catching the airbrush more and getting more deposition.

A 1% solution of the Orasol Red BL dye in PGME (Propylene glycol monomethyl ether). The dye seems to be pulled through the interstices of the weave.  In fact, if you flip the fabric over, the color looks the same on the front and the back.

A 1% solution of the Orasol Red BL dye in PGME (Propylene glycol monomethyl ether). The dye seems to be pulled through the interstices of the weave. In fact, if you flip the fabric over, the color looks the same on the front and the back.

Variations in how the dye is deposited on the substrate might influence the lightfastness of the dye – our initial accelerated aging testing will include dyes in five common solvents.

In addition to tests of the dye on quartz plates, we also need to test the dyes on fur in order to better understand the rate at which we might expect recolored taxidermy to fade or discolor. Accelerated aging tests will be conducted on samples of mounted fur, and the final selection of colors, solvents, and testing conditions will be based on the results from testing the quartz plate samples. From a long list of fur initially considered for testing, examples were narrowed down to only bison, fox, and deer hairs to represent wiry, slick, and hollow-haired fur-bearing mammals. We then had to develop a mounting system that could secure the fur samples against disturbance from handling and air turbulence. It was a challenge to trap both ends of every fiber so that they can be dyed in place while still being able to reliably take consistent color measurements of the same surfaces over the course of testing – hairs mustn’t dislodge because once they are dyed, a lost hair would expose the undyed hairs below and skew the color change measurements. We’ll reveal our mockups and final mount method in our post about sample preparation!

Equipment Purchases

Paul Whitmore watches Alan J. Boerke, a representative from the Q-Lab Corporation, during a training session at AMNH in February 2014.

Paul Whitmore watches Alan J. Boerke, a representative from the Q-Lab Corporation, during a training session at AMNH in February 2014.

For the lightfastness testing during the renovation project, we were able to partner with the Art Conservation Department at Buffalo State College to utilize their artificial aging chamber and spectrophotometer. In support of our current in-depth investigation, the grant funds allowed the AMNH to purchase several major pieces of equipment, including a Q-Sun Xenon Arc Accelerated Aging Test Chamber Xe-3 and an X-Rite Ci62 handheld spectrophotometer. Installation of the aging chamber posed some of greatest challenges in planning, and required extensive assistance from the museum’s plumbers, electricians, and carpenters from our facilities department. We’ll get into some the details of those challenges in a future post.

Orasol dyes and related colors

Over the last several months of planning, we’ve also discovered that it is difficult to obtain small quantities of the BASF Orasol dyes. Few retailers market them, and those that do carry stock that may be many years old. Other retailers use the Orasol name to sell related dyes manufactured by other companies. The true chemical and behavioral equivalence of these dyes is unknown – two dyes with the same name but different manufacturers may have differences in color, solubility, or lightfastness, among other traits. BASF has added and discontinued colors in their Orasol line over time – currently the line includes 26 colors. The previous testing as part of the renovation project included about half of the colors available – we are working to procure samples of all 26 colors. The upcoming lightfastness testing will include both new and old BASF Orasol dyes, as well as related non-BASF dyes – a total of 46 colors in 5 solvents.

And that’s just the lightfastness testing using the aging chamber! We’re expecting these initial cycles of testing to begin next month and continue through the fall and winter, and we’ll be making adjustments to the total number and types of samples to be tested along the way. There are some other avenues of testing we will be conducting, including investigating the rate of color change under real diorama lamps (as opposed to the xenon lamps used in the testing chamber), as well as exploring the possibility of being able to manipulate dye penetration and investigating the long-term effect of dyes on the hair itself.