Controlling Mold on Library Materials with Chlorine Dioxide: An Eight-Year Case Study
By Pat L. Weaver-Meyers, Wilbur A. Stolt, and Barbara Kowaleski
The University of Oklahoma Libraries uses chlorine dioxide in solution as a wet wipe on moldy books, as a fumigant and in atmosphere-activated gas packs to control mold. Descriptions of the three delivery forms and results are provided. Chlorine dioxide toxicity is compared to some other commonly used preservation treatments.
Librarians and archivists have long recognized the many and varied enemies to paper. Insects, fungi, the environment, and people can be culprits in the destruction of books, archives and manuscripts. All present significant and unique challenges to preservationists. This article focuses on the problems associated with mold growth at the University of Oklahoma Libraries and the results of using chlorine dioxide in aqueous and gaseous form.
Articles discussing mold stress the importance of controlling the environment to succeed in the battle against mold. Modern facilities with sealed windows and computer-controlled heating, cooling and ventilation systems have created much better environments for the preservation of library materials. Yet, problems do occur when environmental controls fail.1
Review of the Web site http://palimpsest.stanford.edu/bytopic/mold/ maintained at Stanford by CoOL, Conservation Online, includes a thorough review of current recommendations on mold control. Emphasis is placed on vacuuming with HEPA filters. In Sandra Nyberg’s Invasion of the Giant Spore, available at this site, fumigation is not recommended because of human health concerns. However, Nyberg’s analysis of fumigation agents does not include any mention of chlorine dioxide.
Chlorine dioxide has a long history of use as a disinfectant and is accepted by the U.S. Environmental Protection Agency (EPA) Drinking Water Division in water treatment applications. Compared to phenols, bleach, glutaraldehyde, quaternary ammonium compounds and other disinfectants, chlorine dioxide is the most effective biocide on hard surfaces.2 In addition, it has been found to be 1,075 times more potent than ethylene oxide as a gas sterilant.3 However, it has not been discussed in the library literature as an alternative to ethylene oxide or other gaseous fumigants. Also, it has not been compared to phenol compounds (components of Lysol) as an effective aqueous wipe for books. This oversight may be due to the relatively recent research findings about chlorine dioxide gas and due to a common assumption that chlorine dioxide and bleach are the same. According to Walter Hardy, chlorine dioxide is often confused with chlorine or hypochlorite (bleach). However, chlorine dioxide uses chemical processes that are quite distinct against microbial organisms. Because of hyperactivity with sulfur compounds, chlorine dioxide or ClO2, has a very selective attack against the full range of microbes, from viruses to fungi. This selectivity of attack involves the oxidation of disulfide bonds occurring when sulfur-containing amino acids are used to crosslink two or more polypeptide strands within a structural or enzymatic protein. Enzymes must have a specific three-dimensional structure to catalyze their biochemical reactions. The destruction of the disulfide bonds causes the enzymes to “denature” or lose their shape and thus become dysfunctional. This prevents metabolism and breaks the structural integrity of the microbe.4
Although further testing of chlorine dioxide in gaseous form is needed, according to its current assessment by the U.S. Occupational Safety and Health Agency (OSHA) and other agencies, to date no mutagenic effects have been substantiated.5 Standards for both aqueous and gas forms can be found at http://www.epa.gov/iris/toxreviews/0496.pdf . These standards recommend worker exposure to the gas of .1ppm over eight hours. Short-term exposures can be as high as .3ppm. Aqueous solutions of chlorine dioxide have been more thoroughly tested because of its use in water treatment. It is frequently preferred in water treatment because it does not result in the formation of trihalomethanes, like chlorine does. In addition, recent uses of chlorine dioxide include the treatment of foods. Many studies exist and are documented by the EPA in the previously mentioned Web site about oral and skin exposure to chlorine dioxide. To date, OSHA recommended exposures to chlorine dioxide have been found to be relatively safe.
The University of Oklahoma Libraries’ main library building, the Bizzell Library, is a structure composed of three separate additions built in 1932, 1958, and 1982. The 1932 addition contains steel stack construction with suspended glass flooring and is known as “the decks” (see Photograph 1).
The 1958 addition contains several offices and closed collection spaces. The materials affected by mold outbreaks in recent years have been materials contained in the 1932 decks and in a closed collection space in the 1958 addition. The materials contained in these spaces include older library materials still classified in Dewey and government documents materials. A vintage air handling systems specific to that addition services the decks. A separate air handling system services the 1958 addition, including a closed collection area. Humidity in the closed collection area has fluctuated more than relative humidity in the rest of the 1958 addition. Although air conditioning specialists have investigated these fluctuations, no explanation has been found for the frequent changing conditions in the closed collection area. The air in this area, because of the closed and infrequently used space, is often stagnant. Shelving, broken furniture, and unprocessed gift collections are stored with older materials in this area.
Changing humidity conditions in the decks have resulted from several mechanical failures in the air handling system and were also the result of roof damage that caused extensive leaks in 1991. Several mold outbreaks have occurred in the decks and the closed collection area. Although rapid repair of air handling equipment and stabilization of humidity levels was always the first choice, mold control could not be consistently achieved for a variety of reasons. Consultation with microbiologists led the Libraries to select chlorine dioxide as an antifungal agent with high biocidal activity and relatively low toxicity. Chlorine dioxide used in conjunction with humidity stabilization proved effective.
The University of Oklahoma Libraries began using chlorine dioxide as a solution to wipe moldy books in 1991. This choice was made because of concern about the long-term effects of other fungicides on stack management workers. When mold broke out in the decks after extensive roof damage in 1991, student workers were used to clean affected volumes. Although the process was initially begun using a thymol solution, concerns about the long-term carcinogenic effects of the treatment led to consultations with microbiologists. Chlorine dioxide was recommended as an agent that would be relatively safe for workers treating materials, with no residual complications for library users. It was also determined to be more effective in mold control than phenols.
The relative safety of chlorine dioxide made it easier to clean and dry books without moving them from the stack area to a workroom. Since the outbreak spread throughout several floors, among numerous volumes, teams could work more effectively and rapidly in the stacks. Concern about exposure to library users was deemed unnecessary and no areas of the library had to be closed. Wipe-down procedures were similar to other libraries’ treatments and consisted of wiping down the book covers and edges with the solution and allowing items to be fanned and air dried. If the mold growth infected the insides and spines, pipe cleaners were soaked in the solution and used to treat the interiors of the book spines. Shelves were also wiped with the solution. Workers wore plastic gloves, but no further safety precautions were recommended.
Chlorine dioxide is not bleach, but is an oxidizing and bleaching agent that has been used in the paper industry for many years. It is also used in both water and food treatment. In its use as a water-based solution to wipe moldy books, the concentrations of <.02% have little effect on hardback books. After repeated wiping, only slight bleaching in the fold on book spines has been noted. Items wiped once showed no visible change and most items wiped numerous times showed no visible change. Further analysis of the invisible effect of the solution on the long-term deterioration of the paper or bindings was not done. Simple vacuuming of paper items, as suggested by others, is probably preferable to wetting. However, numerous volumes of congressional hearings were treated with no observable damage.
Wipe down of materials in the decks continued for several months. However, mold did reappear. Repeated checks of the air handling systems revealed some problems that were resolved, but erratic performance of the system still left the decks vulnerable. Further consultation with microbiologists resulted in a decision to fog the deck area. Fogging was accomplished with a hand-held fogger. The operator walked the aisles spraying the stacks and floors with a <.02% solution of activated, stabilized chlorine dioxide. The intake and outtake vents of the air handling system were fogged as well, in the hope of reducing spore contamination in the ducts. Fogging took place while the library was closed, but closure of the stacks during regular hours was deemed unnecessary due to the lack of residual exposure problems and lack of residual odor when dried.
Since the fogging, approximately five years have passed and no further mold outbreaks have occurred in the decks. Although air regulation may be more stable, outages of the system continue to occur. A recent campus-wide air conditioning failure did not result in a mold outbreak in the decks. However, that same campus-wide problem caused a very substantial problem in the 1958 addition.
The closed collection area in the 1958 addition, as described earlier, also serves as a general storage area (see Photograph 2).
It is a closed stack area with a small collection of older materials or materials deemed highly collectible. This area has had repeated mold infestations that have been treated quickly and effectively with a wiping solution. This area has never been fogged, because the collection is so small that control is normally easily maintained. However, in the summer of 1997 during a campus-wide air conditioning failure, control of mold in this area became impossible.
Moving materials out of the closed stack was not deemed advisable, because of concern about contaminating other areas with a high mold spore count. Since air conditioning was not working throughout the building, limiting exposure in the major portion of the collection was most important. After several weeks of daily treatment of volumes in this area, and repair of the air handling system, mold continued unabated. Consultants suggested that the mold spore count was so high in the collection that ordinary control measures (humidity control and temperature control) would not work. The plethora of stored furniture, boxes, shelves, and other items made it impossible to clean the area and provided too many opportunities for the mold population to reinfect the treated book shelves a few feet away.
Although fogging would probably have worked, a newly available alternative was provided by Engelhard Corporation as a test.6 This product, a self-activating packet (Aseptrol™) that reacts with humidity in the air, was used. The packet is a loose powder sachet, which will begin releasing chlorine dioxide in 90 minutes after exposure to ambient relative humidity. This sachet will release 1/2-2 vppm of chlorine dioxide per gram of material in a one gallon volume with 10cc/minute of continuous air changeover. The sachet will continue to release at a constant rate for 15 days before exhausting at 85% relative humidity and over 30 days at 31% humidity.7
Approximately three packets were hung with wire ties between shelves (see Photograph 3).
Regular inspection of the collections revealed a slight chlorine-like odor, but no other effects were noted. However, mold continued to grow until three additional packets were hung. Once a total of six packets were hung in an area with approximately 12,000 sq. feet, total control of mold growth was achieved. No further mold infestations have occurred in the following six months. Exposure to workers was limited because the collection is closed. Exposure rates are controlled by the humidity. Relative humidity in the area did not reach 85%, which means that the release rates were slower than the maximum rate available from the packets. No odor could be detected elsewhere in the libraries (odor threshold for chlorine dioxide is .1ppm). As a result, no limitations were placed on public use of the rest of the collections even through the closed collection shared an air handling system that serviced the entire addition.
The gas packets proved to be highly effective and required the least staff hours to deploy. It took about five minutes to open the packages and hang the packets. The method was preferable to fogging, because it dispersed the chlorine dioxide without wetting the floors or materials. This kept humidity fluctuations down. Further, the gas packets continued to disperse chlorine dioxide in response to the relative humidity and were able to penetrate all the nooks and crannies of the storage area. Because the packets disperse chlorine dioxide faster as relative humidity increases and decrease dispersion rates as the humidity lowers, it is ideal for situations in which humidity control is known to fluctuate. A further advantage is that control does not require constant monitoring and intervention.
Although environmental control is the best way to control mold infestations, air-handling systems do fail. Vacuuming, the current recommended technique, reduces mold spore levels and removes mycelium residue from books. However, it does not kill mold spores. It is also rather labor intensive. More importantly, exposure to high levels of mold represents a human health hazard.8 Our experience with chlorine dioxide’s inhibition of future mold growth after fogging (five years) and after use of gas packets (six months, to date) suggests a strong sporicidal activity.
These results, coupled with recent findings that confirm chlorine dioxide’s relative safety, suggest that preservationists should reexamine fumigation. New substances and more test results make it appropriate to look at fumigation again and consider adjusting the current recommended procedures with regard to human health hazards. In particular, the health hazards of uncontrolled mold may outweigh the use of “safe” fungicides.
Further research on chlorine dioxide’s effect on the long-term deterioration of paper needs to be undertaken to assure treatment is as safe as visual observation of our treated collections has suggested. However, in disaster recovery situations that require immediate, safe control of mold growth, chlorine dioxide gas packs or fogging can help contain an outbreak before air-handling systems can be repaired. The current research reviewed in this article suggests that the risks of long-term human health effects, either mutagenic or carcinogenic, for the use of chlorine dioxide are low. The use of this substance, particularly with the gas packet technology, is attractive. Convenience, low labor costs and its potential to avert disastrous mold outbreaks make this strategy worth serious consideration.