It's Not Emerald (Simply) Being Green

Most folks in book circles have heard about the use of the toxic emerald green (copper acetoarsenite) pigment on 19th-century clothbound books. If you haven't, you can read up on the current state of bibliotoxicology research on the University of Delaware's Poison Book Project website.

The topic can be a sticky one for a number of reasons. There are the well-established health risks associated with handling these bindings unknowingly or improperly, and increasing visibility to the topic is a great way to facilitate proper identification, training, handling, and storage protocols to protect both library patrons and staff. The late surge of public interest in the history of bookbinding processes—not simply in the age or content of books, but in their materiality—is a special delight for me as a conservator.

On the other hand, many reports on this topic tend to focus on the sensational, and seem more interested in sparking a panic than on providing useful context and practical handling advice. I've seen folks go so far as to say that all green books from the 19th century contain arsenic, which is both blatantly false and frankly damaging to actual research. Overblown statements like this, even when well-intended, dilute the warnings for a specific group of items presenting an acute health risk; "all green books" undoes researchers efforts to draw attention to a very specific shade of green used during a particular time period and even with preference in certain geographies, according to current research. This makes it more difficult for everyday library users to recognize the risk and less likely to respond appropriately when handling an actual poison book—a cry wolf situation which leads to burnout and apathy.

This is all to say that while the risks are real and there's certainly nothing wrong with being over-prepared (handing with nitrile gloves, washing hands and contact surfaces before and after handling any antiquarian books, etc), there is nuance to the poison book phenomenon, and recognizing it makes people safer and more informed than writing off an entire century's worth of library stacks. I found this to be especially true after conducting a small analytical survey of 19th-century green paper-covered books from the Schweben Buntpapier collection, using X-ray fluorescence spectrometry to identify the elemental composition of several bright green bindings.

While research in the field of bibliotoxicology is ongoing and new information continues to be published, there are a few key points when it comes to identifying these books in the wild:

• cloth-covered books are generally a vibrant, bright blue-green color (rather than the much more common dark forest green) have a great deal of gold tooling, in line with generic publisher's cloth bookbinding trends

• paper-covered books are often a pale minty green color and can also include printed or tooled gold designs

• peak usage for the pigment on bookbindings is between the 1840s and the 1860s

• copper acetoarsenite degrades with light exposure to a dull brown color, often recognizable on the spines of emerald green cloth books

There are some factors that break these rules:

• copper-arsenic pigments were actually produced quite early in Germany (1775) and used on textblock edges during the 18th century

• some books with earlier publication dates may have been rebound during the 19th century with emerald green-colored components

• non-arsenical copper- and chromium-based green pigments mixed with white pigments (especially lead) can create hues resembling emerald green

Here are the books in my collection which were tested... can you tell based on the color and publication info which are arsenical?

Nineteenth-century green books tested for arsenic via X-ray fluorescence.

From left-right, top row: 24.093 - Germany 1800, 23.034 - Germany 1816, 25.006 - Germany 1822; middle row: 25.080 - Germany 1835, 25.012 - United States 1846, 24.036 - Germany 1855; bottom row: 22.036 - Germany 1855, 25.048 - United States 1856, 24.046 - United States 1870)

---

And here are the results!

Looking at all of these books side-by-side in concert with their elemental identification, new insights take shape. Firstly, as the books are arranged chronologically from 1800 to 1870, we have confirmed emerald green across that range, in objects dating 1822, 1855, and 1870. The Poison Book Project lists 1840s-1860s as the strongest date range for the use of emerald green in bookcloth, but its presence may lean earlier for textblock edge decoration and paper bindings.

All of the books which did not contain arsenic contained either copper, lead, chromium, or a combination of these. The two earliest examples (24.093 and 23.034) seem to have been colored with copper only, probably using a malachite-based pigment. Comparing these two books with the others tested revealed a blueish brightness to the green hue which was not apparent to me when the books were purchased. This, in addition to their dating from the first decades of the 19th century, may make them more recognizable in the wild as non-arsenical green paper bindings.

All three of the positively emerald green-colored bindings involve surface-coloring, i.e. the color was applied after the paper sheet was formed. This is true also of emerald green bookcloth—the pigment is mixed with a paste binder and driven into the weave of the cloth; the fibers are NOT dyed, despite the prevalence of term in media coverage of the phenomenon. This distinction is important because surface-bound pigments are more readily displaced by physical wear and handling than solubilized dyes—the nature of emerald green as a surface coloring informs its potential for harm through the displacement and subsequent inhalation or ingestion of pigment particles. While it is possible to color papers in the vat with pigments, I've not yet seen an example of emerald green being used in this way.

Ultimately, these variables and the results of the survey suggest that color comparison alone is an unreliable method of identifying emerald green on bookbindings. Analytical testing is the preferred method of establishing the presence of heavy metals—in the case of most home libraries, it's safest to bag and use nitrile gloves for all suspicious materials. The Poison Book Project scientists strongly discourage the use of at-home arsenic test kits meant for potable water supplies as they create a serious disposal and groundwater contamination problem. In the Schweben Buntpapier collection, tested and confirmed arsenical books are shelved in sealed polyethylene bags and only come out for analysis.