Volume 21, Number 1
May 1997

The pH of New Library Books, 1976-1993:
A Compilation of Survey Results

Librarians and preservation people have found that the most practical way to find what proportion of all books are printed on alkaline paper is to do a sample pH survey of incoming books at a large library. (It is not possible to monitor incoming books for permanence, however, because there are no simple tests for the other three specifications-tear resistance, lignin content, and alkaline reserve.)

At least nine such surveys have been reported or mentioned in the preservation literature in the last 18 years. They range in sophistication from amateurish to professional, and in number of volumes surveyed from 38 to thousands. This is a summary of these surveys. The focus is on books printed in the U.S., though all the surveys included books from other countries. Except where noted, pH was determined by means of spot tests with chlorophenol red, a pH indicator.

1981. At the Allerton Park Institute in Illinois, Gerald Lundeen reported what he had learned from chemist George B. Kelly about pH testing at the Library of Congress: "The Library of Congress has been monitoring the pH of books coming into its collection and finds that about 25 percent of the American books and about 50 percent of the European books are made of alkaline paper. Five years ago less than 1 percent of the books tested were made with alkaline paper." No details were ever published about this testing program, though George Kelly did say in a telephone conversation that they were using a flat-head electrode to find the pH of the intact pages.

European paper mills made more alkaline paper than mills in the U.S. at that time because good quality chalk (CaCO₃) was widely available there, and could be used as an extender for scarce fiber.

An alkaline size (AKD) had been put on the market in the U.S. about l954 for the first time, but few mills used it. Another (ASA) was marketed in the early 1960s, and like the first, was little used. EPA regulation of effluent from paper mills in the 1960s had some effect, however, because conversion to alkaline papermaking enabled the mills to recycle their white water (from the paper machine), and discharge less of it to the environment. EPA can probably take credit for the estimated increase in alkaline book paper from 1% to 25% in the 1970s.

1982. The Council on Library Resources' Committee on Production Guidelines for Book Longevity, perhaps drawing on the same source of information used by Gerald Lundeen, estimated that 25% of paper manufactured in America for book production was acid-free, and listed 11 manufacturers of acid-free paper.

1983. As a course project when I was studying preservation administration at Columbia University, I surveyed incoming library books, and published preliminary results in the February 1984 issue of the Abbey Newsletter on p. 2. About 62% of the U.S. publications in the 236 books surveyed had a pH of over 6.7, according to the chlorophenol red spot test. (The sample was in two parts: a trial run of 133 volumes and a main sample of 103 volumes. Sixty-five percent of the U.S. books in the trial run, and 58% of those in the main sample, were acid-free.) The figure for all books tested, regardless of country of origin, was lower: 41%.

Books were also tested for presence of an alkaline reserve by taking a small sample, dropping HCl onto it, and watching for effervescence. Most of the alkaline paper samples tested did effervesce, but some did not, perhaps because the carbonate particles had been sized along with the fibers in the machine. This would have made the particles waterproof, of course, and kept them from interacting with the dilute acid. Still, 78% of the acid-free books from all countries in the trial run, and 64% of the acid-free books in the main sample, were shown to have an alkaline reserve.

Over 50 publishers were found to be using alkaline paper in books published between 1977 and 1983. Most of these publishers were American.

Spring/summer 1988. Three students (Curtin, Harger and Yasue) did a follow-up survey on the pH of incoming books at Columbia University Library. The statistical work was competently done with the help of faculty from the University's Mathematics Department, and over 800 monographs were tested for presence of an alkaline reserve as well as for pH. Of all U.S. imprints, 66% tested alkaline, and 29% of these had either the infinity sign or an alkaline paper notice on the verso of the title page. Of the U.S. imprints published in 1987 or 1988, 78% were alkaline.

Foreign books were analyzed too. Japanese imprints were 75% alkaline; so were 48% of Asian books, 44% of West European and 29% of Middle Eastern imprints. No alkaline imprints were found in books from Africa, South and Central America, Eastern Europe, the Soviet Union, or Australia.

Aug. 1988. The Provo (Utah) Public Library used a prototype Abbey pH Pen to test the pH of 38 new acquisitions, all hard cover. They found that 66% of books published in the U.S. were over pH 6.7.

1989. Randall Butler, a library science student at Brigham Young University, surveyed incoming 1987 publications, both hardcover and softcover. He had three purposes: to determine the percentage of alkaline commercial American imprints, to check which had permanence statements in the books, and to "determine the frequency and reliability of acid-free ISBN qualifiers" (by which he may have meant the statement on the verso of the title page that the paper was alkaline or permanent). Results were presented in his paper, entitled "Here Today ... Gone Tomorrow: A pH Investigation of 1987 Library Acquisitions." His results were also published in the library literature and are often cited, which is a pity because he seems to have been confused about the pH scale, and the definition of acid and alkaline.

He concluded that 67.5% of the books surveyed were alkaline, and this would have been in line with other surveys about this time, except that this survey also included softcover books. However, too many aspects of this survey do not add up, so one should not rely on it too heavily.

1991. Akio Yasue, coauthor of the second Columbia survey, returned to Japan after earning his preservation certificate at Columbia, and wrote a report of pH surveys at the National Diet Library for the July 1992 issue of CAN.

The National Diet Library had been doing surveys of monographs every year since 1986. This one included 600 volumes that came in during one week in August. A pH of 6.5 was taken to be alkaline. Two samples were measured: commercial and government publications submitted as required by law to the Diet Library, a national depository; and a select group of volumes of research value. Apparently all the books tested were published in Japan. Of the commercial and government publications tested, 71% were alkaline; of the research publications, about 85% were alkaline.

Not only the 1991 survey is described in this report, but the 1986 and 1990 surveys too. It is hard to follow the descriptions, because they are intertwined, and the surveys that provided the data for the tables and graph are not clearly identified. In the 1990 survey, they measured pH for each of 20 journals and monographs in five different ways: with two different flathead electrodes, two kinds of pH indicator strips, and the cold extraction method. The cold extraction test gave generally higher pH readings, one or two points higher than the other four methods. Such comparisons are rare in the technical literature, so this is good data..

1991. For five years, 1987 to 1991, the National Library of Medicine kept track of which journal titles in the Index Medicus were printed on alkaline paper . During that period, they went from 4% to 49%, a remarkable accomplishment by the NLM. They had gone to some trouble to convince publishers that the need for permanent paper in these journals was greater than the publishers' economic need for the usual thin groundwood paper on which journals are usually printed. Findings were reported, with a table and a graph, in the December 1991 Abbey Newsletter.

1991. Ohio State University surveyed incoming monographs published in 1989, 1990 and 1991, using the Abbey pH Pen. William Studer, the library director, reported the results in the context of ARL libraries' need for deacidification programs, in A Roundtable on Mass Deacidification, 1992. The data are reported in a single table, by country, hardcovers vs. paperbacks, year of publication, and percent alkaline. They found that 95% of the US hardcover imprints for 1991 were alkaline, while only 66% of the U.S. paperbacks that year were alkaline. It averaged out to 85%.

The change in percent alkaline for all 28 nations covered by the survey rose rapidly over those three years: 58% to 64% to 67%. For the U.S. alone, the rise was even steeper: 48% to 81% to 85%.

1993. A third pH survey was performed at Columbia University by Lee Dirks and Jacalyn Mignogna. Sample size chosen was 382: 219 hardcovers and 163 paperbacks, chosen from acquisitions going through library binding or shelf-processing procedures. Each volume had to have been published in 1990 or later. U.S. hardcover imprints for the three-year period were found to be 85% alkaline. (No figure was given for U.S. imprints for 1993 alone; data was not analyzed for any single year by country.) U.S. nonacademic press imprints were not far behind those from U.S. academic presses. Acid-free notices were carried by 68% of the academic press imprints.