The following is a guest post by Preservation Research Fellows, A. Elizabeth Torres and Megan E. Zins. Elizabeth obtained her bachelor’s degree in Art: Paintings and Art History at The University of Texas at El Paso. She is an upcoming first year in New York University’s Master’s in Art Conservation program where she will focus on paintings conservation. Megan graduated with her bachelor’s degree in Chemistry with focus in Art History from The College of Wooster in Wooster, Ohio. She has recently joined the Library of Congress as a Preservation Technician in PRTD.
We joined the Assessing the Physical Condition of the National Collection (ANC) project in the fall of 2022, following a previous researcher, and have since gathered a substantial amount of data and have learned just as much. The ANC project, as its name suggests, seeks to survey the condition of books held in research libraries across the country to assess potential differences and/or similarities in the condition of these books. Specifically, the project is examining 500 of the same titles and editions from each of the five partner institutions—resulting in 2,500 books to assess! Due to the magnitude of this project, we concurrently fulfill a niche role independently and work collaboratively with colleagues in the Preservation Research and Testing Division (PRTD) here at the Library of Congress. Specifically, we carry out the chemical tests on a small sample from each book. However, our work on the project is not only limited to research inside the lab.
From the beginning of the project, it was clear that the same title was not always identical from every institution, as first expected. In some instances, the books did not even resemble one another by eye. On a more technical level, based on their Fourier-Transform Infrared (FTIR) spectroscopy data, these books contained paper types that were not consistent across all copies in a set. To find potential explanations regarding these differences, we examined the data that had been collected prior to us joining the project, including visual assessments, photo documentation, and optical and chemical properties of the books. Naturally, we each took a different approach when investigating this data, which reflected our unique backgrounds in the field of cultural heritage. We focused on historical information, both specific to individual books and across the time period studied (1840-1940); as well as the spectral data collected from the books, namely the fiber optic reflectance spectroscopy (FORS) and FTIR data.
Through examining the photo documentation, we began to notice unique differences in books within sets of the same title. In some instances, one or more books in a set would have a different publication date, place of publication, or even a different publisher altogether. Some books contained distinctive markings such as stamps, pictorial elements featured on the title page, and/or different printing fonts. In some instances, we found historical information that was particularly relevant. In one such set of four books, a single copy contained contrasting publishing information in addition to a unique note indicating that the book was in fact a reprinted copy. By conducting further research on the publisher listed, we learned that the company reprinted more affordable copies of popular novels and plays in the mid-19th century. This information provided a possible explanation as to why the reprinted copy seemingly contained a different paper type than the other three books in the set. This explanation became even more plausible, as the paper types were distinguishable by sight and touch. The reprinted copy’s paper appeared darker in color and felt coarser than the paper in the other copies.
Consulting the spectral data for this set further reinforced the clear distinction between the copies. By examining the FORS and FTIR data, we were able to identify some unique signals in the spectra that helped to distinguish the paper type of the reprinted copy from those of the other books. We then applied this kind of investigation of the spectral data to the other book sets that we studied. While some paper types were easier to distinguish through this optical data, others proved challenging, and in extreme cases, nearly impossible. The spectral data, in conjunction with historical information researched, allowed us to gather plausible reasons explaining the differences between some of the books in the sets we examined.
The distinctions that we observe when looking at the books themselves parallel the results of the miniaturized lab tests that we run on 5/8” paper samples, as results often vary between books that are meant to be the same copy. Specifically, we conduct tensile, pH, and size exclusion chromatography (SEC) experiments to gather more molecular-level data to help us potentially understand the overall condition of the paper.
Conducting these tests efficiently requires some planning in advance to keep the project moving. For example, tensile testing calls for the samples to be equilibrated for 24 hours to specific environmental conditions, known as TAPPI conditions, before the tensile strength can be tested. Because this equilibration period is a requirement for all samples, the effect of other environmental factors on the outcome of the tests is mitigated. Testing pH also requires planning in advance to account for the preparation of each sample. This preparation includes cutting very small strips from sample material and emulsifying them in a water-based solution. For every individual book sample, we produce and test two replicates of these mixtures, which increases the confidence in the measurement.
Running SEC experiments requires diligent attention to timing, as its preparation is more involved than the other tests. The purpose of these experiments is to measure the length of the cellulose chains of which paper is made. Therefore, dissolving the samples in a way that preserves these chains requires a slow approach that takes roughly two days to complete. After preparation, samples are run through our SEC instrument where each sample takes an hour to be measured. Even though these three lab tests can be run simultaneously, with over 2,500 samples to measure, timing really is everything!
As of the publication of this blog, we are still in the lab checking our watches and running chemical tests on the remaining samples of the ANC project. Our colleagues in PRTD have been working diligently on data analysis to begin making sense of all the data taken into account together, including the chemical data that we collect, along with visual assessments, photo documentation, and optical data. Look for more updates to in future blogs as the project wraps up!
Subscribe to the blog— it’s free! — learn how the largest library in the world preserves the coolest stuff in the world.