Formaldehyde Residues Following Fumigation of Leafcutting Bee Cells and Neutralization with Ammonia Gas By David Ostermann Introduction Formaldehyde fumigation (using paraformaldehyde product) for many producers in Manitoba is their first treatment choice for chalkbrood control, over bleach or heat treatments. Formaldehyde fumigation of live cells is conducted in the spring in conjunction with the incubation of the cells for development of adults and subsequent alfalfa pollination that season. While it is very effective at controlling leafcutting bee pathogens, there are health concerns with formaldehyde gas as it is toxic and overexposure may be fatal (EC/HC 1999). It may be harmful by inhalation, ingestion, or skin absorption. For these reasons it is important to avoid dangerous exposure to formaldehyde and to follow the product Handling Precautions as indicated on the label. The current formaldehyde-based chalkbrood control practice is to expose the leafcutting bee cells and equipment to high concentrations of formaldehyde gas for a 24-hour period, then ventilate the fumigation area for 48-72 hours, and longer if necessary. Neutralisation with ammonium bicarbonate quickly reduces formaldehyde concentrations in the air without adversely affecting chalkbrood control and without hurting development of the leafcutting bees (Lafreniere 2001). It is generally accepted that formaldehyde gas reacts with ammonia gas to produce hexamine (synonym hexamethylenetetramine) (and potentially other amine gases). While hexamine has been identified as a harmless substance (Kawamata and Kodera 2004; Ostermann 2005), other sources indicate that hexamine is an irritant to eyes, nose, and skin, is hazardous (Dugan and Serago 2005; NJDHSS 1999), and may decompose to formaldehyde in the presence of perspiration (ie sweat; slightly acidic 4-6.5) (Dugan and Serago 2005), and therefore should be handled with caution. In the solid form, hexamine is a white, crystalline powder with a mild ammonia odor.

This study assesses the levels of formaldehyde gas in the air that may be left behind following the fumigation of live leafcutting bee cells, neutralization with ammonia gas, and venting, at a local producer operation, and discusses the safety risk associated with residual substances. Methods The fumigation trials were conducted in conjunction with cell fumigation and incubation practices at a local producer operation. The study involved two rooms equipped with ceiling fans for good air circulation, exhaust fans, door vents, and temperature and humidity control systems. The rooms were pre-conditioned to increase RH and bring temperature to 23°C. Prior to fumigation, the rooms were entirely sealed using plastic sheeting and duct tape. Both rooms contained trays with live loose leafcutting bee cells and were fumigated with formaldehyde and ammonia (eg. paraformaldehyde and 1 ammonium bicarbonate). The experiment was conducted over a 20-day period (up to incubation day 15) in June, 2005. The procedure for the formaldehyde fumigation portion of the experiment followed the label recommendation for paraformaldehyde fumigation (Goerzen 2002) of alfalfa leafcutting bee nest material and leafcutting bee cells. Paraformaldehyde prills, at a rate of 1.10 lbs per 1000 ft 3 , were heated to about 200°C+ in electric frying pans for 4 hours then left for 20 more hours. Next, ammonium bicarbonate crystals, at a rate of 1.43 lbs per 1000 ft 3 (30% more than paraformaldehyde product (see Addendum) were heated to about 200°C in electric frying pans for 4 hours then left for 2 more hours before venting. The paraformaldehyde product was purchased from Cleartech in Winnipeg. The ammonium bicarbonate was purchased from Univar Canada Ltd. in Winnipeg. Addendum: Luftman (2005) reports that the determined theoretical weight ratio of ammonium bicarbonate (AB) to paraformaldehyde (PF; 90%), for neutralization, is between 1.55 and 1.58 (i.e. 60% more AB than PF product), and therefore, this greater amount of AB may provide additional PF neutralization. Ammonium carbonate (AC) may also be used and the weight ratio is between 1.06 and 1.10 (i.e. 10% more AC than PF product). Neutralization efficacy results may vary with operation. This method is not registered on the paraformaldehyde label. Refer to product label Handling Precaution; personal protection is important, including full-face respirator, gloves, and impervious clothing. Adequate aggressive ventilation is important to help reduce levels of any hazardous residual gasses. Approximately 6 million live cells were put in each room. The dimensions of the rooms were approx. 24’x13’x10’. Following the fumigation and neutralization processes, the rooms were ventilated with a 12” exhaust wall fan (Princess Auto; 800 max. CFM rating) and door vents for a period of 48 hours. After the rooms were ventilated, the temperature in the room was increased to 30°C and water put on the floor of the rooms to increase the relative humidity for incubation of the bees. A timer connected to the fan was used to provide additional venting (10 min. periods, 14 times a day) to incubation day 7 when dichlorvos strips for parasites were placed in the room. Biosign biological indicators (Getinge Canada Ltd.), for monitoring steam and ethylene oxide sterilization cycles, were used to determine if the fumigation process was successfully executed. Two Biosign indicators were placed in each room during formaldehyde fumigation and processed in a laboratory to determine viability. In the incubator rooms, an area (approx. 4ft 2 ) was marked out on the floor with duct tape and wiped clean before fumigation and neutralization, to see if solid residue could be observed following neutralization. To monitor levels of formaldehyde in the air over the course of the experiment, ¼ inch plastic tubing was run from the outside to the centre of each room, 5 ft off the ground. The amount of formaldehyde in the air was measured using three methods including Dräger pump and tubes (Dräger Safety AG & Co. KGaA; formaldehyde 0.2 and 40 ppm), 2 SKC (XAD-2 (2-hydromethylpiperidine); category #226-118) tubes (SKC Inc.) and pump and the PortaSens II system (AFC International Inc.; sensor part #00-1040). The SKC tubes and pump were purchased and serviced by Maxxam Analytics Inc. Laboratory analysis was done according to NIOSH method 2541. Measurements were taken during fumigation, neutralization, and following ventilation, up to incubation day 15, on which day temperature was reduced to halt bee development. Air from the rooms was drawn through the SKC tubes for 30 mins., then sent to the Maxxam Analytics lab for analysis, with results available within a week or so. At the time of this experiment, no system was found to be immediately available to measure hexamine. Relative humidity and temperature were recorded with HOBO H8 loggers (Onset Corporation Inc.). Given the limitations of the PortaSens II and the Dräger tube monitoring techniques, only the SKC tube results are discussed below as these readings are considered more reliable under the conditions of this experiment. Results and Discussion When comparing the different formaldehyde gas measurement systems used in this experiment, there are some points that should be discussed. The PortaSens II system has some advantages in that it is electronic, does not use disposable tubes making it possible to take many samples without additional cost, and gives a reading in approx. 5 mins; however, the formaldehyde-specific sensor is costly, and the PortaSens II system, as indicated in the accompanying literature, may be affected by other gases in the room, such as ammonia, and therefore is less valuable in this situation. The Dräger tube system is less expensive than the SKC tube system and gives a reading in about a min.; however the measurements on the Dräger tube can be difficult to read and the system does not necessarily distinguish between formaldehyde including a closely related chemical, hexamine. It is not clear whether the Dräger tube reading would represent the sum of formaldehyde and hexamine levels in the air in this experiment. The SKC tube method is believed to be very accurate at measuring air-borne chemicals; however it is the most expensive system of the three used in this experiment, and it takes much longer to get a reading. Fig. 1 shows the differences in three formaldehyde gas measurement systems. Due to problems with collecting data from one room, where formaldehyde couldn’t be detected and the cause of the problem remained unknown after examination, only data from the second room is discussed below. During fumigation of the live leafcutting bee cells in this experiment (before incubation), levels in the air reached 150 parts per million (ppm) (n=1), then dropped quickly to 6.5 ppm (n=1) after neutralization, then got down around 3.0 ppm (n=1) after 48 hr of full ventilation of the room (Fig. 1). During incubation (after neutralization and 48 hr ventilation), levels of formaldehyde in the air ranged from 1.7 to 4.1 ppm (n=5) to 17 days after fumigation (incubation day 13) with the lower levels being recorded later in the experiment. On incubation day 15, to hold development of the bees, the temperature in the room was reduced to 15 o C, and the level of formaldehyde also dropped to 0.2 ppm (n=1). Relative humidity in the room ranged from 50-70% until about incubation day 7 and was affected by timed venting over this period. From incubation day 7 to 15, relative humidity ranged from 60-70%. The temperature in the room remained near 30 o C from the start of incubation up to day 15. 3 The Biosign biological indicators indicated that the formaldehyde fumigation was effective in killing the indicator bacteria. In the marked area on the floor, and around the room, no obvious white residue powder was observed, where hexamine residue may have been expected. As a general guideline, 2 ppm is the 15 min. short term exposure limit (STEL), and 0.75 ppm is the 8-hr day permissible exposure limit (PEL) for formaldehyde gas for people (OSHA 2002). These results suggest that when treating alfalfa leafcutting bee cells, the level of formaldehyde gas that may be present following neutralization and venting may still pose a health risk to producers who spend time inside the room, checking their cells, applying treatment for parasites, or putting lids on their trays, for example. The residual formaldehyde levels observed in this experiment differ from the results of Lafreniere (2002). When working with eclosed leafcutting bee cells, in a different incubator, and measuring residual formaldehyde with Dräger tubes, Lafreniere observed no measurable levels in the air when temperature was increased following neutralization. Although the presence of formaldehyde could be sense in the air (ie. slight odour and irritation of the eyes), the levels could not be measured using Dräger tubes sensitive to 2 ppm. The differences may be due to the use of live bee cells versus eclosed cells, greater volume of organic material with the live cells, the addition of water on the floor and differences in weather conditions at the time of the experiments. It is not clear what levels of hexamine if any remained in the air in this experiment as no system was used to specifically measure this chemical; however, an estimation of Dräger readings minus SKC readings suggests the presence of hexamine particularly in the first part of the experiment. No occupational exposure limits have been established for hexamine; however, this does not mean that this substance is not harmful. Hexamine is an irritant to eyes, nose, and skin, is hazardous, and may decompose to formaldehyde in the presence of perspiration. In an experiment looking at the levels of formaldehyde in an empty incubator room conducted in March 2005, it was found that some formaldehyde may remain in the room following formaldehyde fumigation, neutralization, ventilation, and sweeping of the floor. These findings may have been the result of the leaking of some formaldehyde gas, as well as specific incubator characteristics. Even though neutralization with ammonia gas significantly reduces the concentration of formaldehyde in the air, the presence of any residual formaldehyde is likely influenced by a number of factors which may not be easily assessed. These include factors related to the incubator, materials and conditions in the incubator, and venting efficacy, for example. The results of the studies suggest that formaldehyde gas may become trapped or absorbed in the incubator (eg. in cells, trays, ducts, or walls, etc.), then be released into the air, as temperature increases (above approx. 15 o C). It has been reported that formaldehyde gas 4 may be absorbed or remain on the surface of materials at least for a short period of time (Luftman 2005; Braswell et al. 1970). It is not clear what effect the high humidity air from outside (ie. during some rainy weather at the time) may have affected the venting of (hydrophilic) formaldehyde from the room in the experiment with the cells. Because of the hydrophilic nature of formaldehyde, temporary water on the floor of the incubator to increase humidity is not ideal in this experiment; however, it is not an uncommon practice and therefore represents a potential situation. A humidifier with a compartment of water would likely pose less of a conflict. The level of human sensitivity to formaldehyde gas, or ability to detect the gas in the air usually by smell or irritation to eyes, is believed to be around 0.1-2.0 ppm, where eyes are usually most sensitive. Those who are less sensitive to detection of the gas may be at a greater risk of significant exposure. Furthermore, the odour of leafcutting cells can mask the odour of formaldehyde so that its presence is less obvious. Due to the complexity of the chemical reactions, the results, and the limited scope of this experiment, more research may be warranted in this area. The presence of water or moisture in the room, including moist venting air, may be difficult to control and understand, yet continues to be a conflicting factor when fumigating with and neutralizing formaldehyde gas. The results also emphasize the importance of using caution when fumigating and wearing a full-face respirator and protective clothing to minimize any eye, nose, or skin absorption exposure to residual substances. Venting while in the room also helps reduce exposure to low residual gasses. Also, follow the paraformaldehyde product Handling Precautions as indicated on the label. Thanks to Greg Shirtliff for his generous help and participation with this study. Funding for this project was provided through the Covering New Ground initiative and in-kind contributions from the Manitoba Forage Seed Association. Thanks also to Dr. Currie at the University of Manitoba for the use of his Dräger and PortaSens II measuring systems. For more information about this study including information about different products and treatments, please contact David at (204) 945-3861. 5 Fig. 1 Levels of formaldehyde as measured by three systems, as well as changes in room temperature and relative humidity, during incubation of cells up to day 15, following fumigation, neutralization and venting for 48 hrs. Additional venting occurred (14 times a day for 10 min. periods), using a timer on the exhaust wall fan, up to incubation day 7 (approx. June 21), when parasite strips were placed in the room. Literature Cited Braswell J.R., D.R. Spiner, and R.K. Hoffman. 1970. Adsorption of formaldehyde by various surfaces during gaseous decontamination. Applied Microbiology 20(5): 765-769. Dugan, M. and E. Serago. 2005. Hexamine – MSDS sheet. Website: Last accessed Feb. 21, 2006. Environment Canada & Health Canada (EC/HE). 1999. Priority Substances List Assessment Report: Formaldehyde. 102 pp. Goerzen, D.W. 1992. Paraformaldehyde Fumigation of Alfalfa Leafcutting Bee Nest Materials. Saskatchewan Agriculture Development Fund. Kawamata, S. and H. Kodera. 2004. Reduction of formaldehyde concentrations in the air and cadaveric tissues by ammonium carbonate. Anatomical Science International, 79:152-157. Lafreniere, R. 2001. Manitoba leafcutting bee extension report. Forage Seed News, winter, 32-35. Lafreniere, R. 2002. Safer techniques of decontaminating leafcutting bee equipment using formaldehyde based products. Covering New Ground report. Luftman, H.S. 2005. Neutralization of formaldehyde gas by ammonium bicarbonate and ammonium carbonate. Applied Biosafety 10(2):101-106. New Jersey Department of Health and Senior Services (NJDHSS). 1999. Hexamine – hazardous substance fact sheet. Website: Last accessed Feb. 21, 2006. OSHA, 2002. Formaldehyde fact sheet. Ostermann, D. 2005. Leafcutting bee research and management. Forage Seed News, spring/summer, 15&46. February 2006 (Updated January 2008) 7