1 INTRODUCTION

Damage to works of art and historical records on paper is a major focus of conservation efforts. The Library of Congress, for example, estimated in 1984 that 25% of its collection (some 3 million volumes), had become too brittle for circulation. Recognition of the degradation of paper is not new: in 1338 A.D. King Pedro IV of Spain issued a decree demanding the improvement of paper manufacturing (McCleary 1981). Paper degradation is largely due to the acid hydrolysis of the glycosidic linkages in cellulose (Smith 1975). Paper may be acidic because of the bleaching products, rosin sizing, and alum, and components of incompletely processed fibers such as lignin, hemicellulose, and resin extractives resulting from the manufacturing process. Paper may also become acidic by absorption of acidic atmospheric pollutants. Paper deacidification neutralizes the acidic content of paper and leaves an alkaline reserve for the uptake of acidic pollutants. To preserve works of art and historical records, paper deacidification has become a protection measure widely practiced in the conservation community (Mihram 1986; Bredereck et al. 1990; Lienardy and Van Damme 1990; and references therein).

The major acid pollutants—sulfur dioxide (SO2) and nitrogen dioxide (NO2)—are ubiquitous in urban air worldwide (Bennett et al. 1985; DeKoning et al. 1986), with typical ambient concentrations of 5–50 parts per billion (ppb) for SO2 and 50–200 ppb for NO2. Reaction products of sulfur dioxide and nitrogen dioxide in the atmosphere and on surfaces include sulfuric acid (H2SO4) and nitric acid (HNO3), which may contribute to the degradation of paper.

The degradation of paper by SO2 has been investigated by Jarrell et al. (1936). Edwards et al. (1968), Langwell (1955, 1959), Atherton et al. (1973). Hudson et al. (1964), and, more recently, by Daniel et al. (1990). These studies have involved SO2 concentrations of 1,000–10,000 ppb that is, 20–2,000 times higher than those found in ambient air. A potential problem with using such high SO2 concentrations, at which SO2 readily forms sulfuric acid aerosol in air at ambient humidity, is that the observed damage may be due to uptake of H2SO4 rather than SO2. Another problem with these high SO2 concentrations is that the alkaline reserve (e.g., carbonate) is depleted rapidly, and the observed damage is no longer relevant to deacidified paper. Thus it is important to study the SO2-paper system at much lower SO2 concentrations relevant to polluted air. In this way no sulfuric aerosol is produced, and therefore chemical and physical changes, if any, can be unambiguously attributed to uptake of SO2.

To our knowledge, no one has directly measured the expected products of SO2 adsorption—sulfite, bisulfite, and sulfate—on paper. Nor have the reaction products resulting from exposure of paper to nitrogen dioxide been investigated. Such product studies are important for paper conservation, since to understand the chemistry taking place when paper is exposed to air pollutants one must know what compounds accumulate in the paper as reaction products.

Accordingly, our study was aimed at investigating the nature and yields of the reaction products resulting from exposure of deacidified and untreated paper to ambient levels of sulfur dioxide and nitrogen dioxide. Two types of paper were tested white wove and newsprint, each with three types of treatment—untreated, deacidified using an aqucous method, and decidified using a nonaqucous method. Four exposures were camed out, one to purified air (control, duration 11 weeks), one to purified air containing SO2 (87 ppb, 29 weeks), one to purified air containing NO2 (92 ppb. 13 weeks), and one to purified air containing both SO2 and NO2 (108 and 98 ppb. respectively, for 13 weeks). Paper samples were withdrawn from the exposure chamber weekly or biweekly and were analyzed for sulfate, bisulfite, sulfite, nitrate, and nitrite. Humidity was controlled to the extent provided by a large bed of silica gel in the air purification system used and daytime relative humidity was typically 60 � 10%. Within that range, variations in RH have no effect on the chemical reactions between paper and SO2 and/or NO2. This range of RH also provides a realistic model for the many buildings and historic houses in which paper is stored and which are not equipped with heating, ventilation, and air conditioning or other means of achieving strict humidity control.