Document kDkdrObJwGBbxMyZ988vV2Z8E

o 4>- 2-4-72. . jVc^/'5^ 4- // /?7 Barrier Coatings to Prevent Polychlorinated Biphenyl (PCB) Contamination of Silage. L. B. Willett Department of Dairy Solenee Ohio Agricultural Researoh and Development Center Polyohlorlnated biphenyls or PCB'a, also often oalled by their trade names Aroolors (Monsanto), have been reoently In the news as "new" chemical contaminants found In livestock feed and subsequently in meat, eggs, poultry, fish and milk. The discovery of these compounds has caused considerable concern among those associated with animal agriculture and considerable financial loss to those who have been directly affected by this contaminant. PCB's are a family of stable chlorinated hydrocarbons which upon analyses may appear quite similar to DDT. They, as all synthetic chlorinated hydrocarbons, are Illegal oontamlnants of human food. PCB's are quite ohemloally Inert and are not broken down by acids, alkalies, or oorroslve ohemloals. They have a high boiling point and do not readily burn - only vaporise. Thus PCB's have the necessary physloal and chemloal characteristics for persistence and accumulation up the food chain. Polyohlorlnated biphenyls were first described In the scientific literature In l88l, and oonsnerolal production was successfully achieved In 1930. Between 1930 and 1969 the C3DA listed 55 different Industrial or agricultural applications of PCB compounds, many of which represented potential sources of environmental, food and feedstuff oontamlnatlon. Seen of the produots that have oontalned PCB's Include: eleotrloal Insulating fluids, tires, MQNS 070666 o 2- prlnters Ink*, flame retardant paints and fabric*, oarbonl*** If-duplicating p*pr, aone lnaaotloldes and ooncrete coatings. Interestingly enough, PCEPa were only discovered as an envlrotmiontal oontanlnant In 1966. A large number of slloa In Ohio and neighboring states have been coated to prevent leakage and erosion of the walla with a material that contained PCB's. The PCB found In greatest Quantities In these ooafelng materials Is Aroolor 1254* Oils PCB material migrates from the silo coating Into the silage, then la consumed by and accumulated in the livestock. Therefore, the PCB residues can ultimately be deposited In human food products produced by llvestook fed the contaminated silage. If the oon- oentratlon of this residue exoeeds The Pood and Drug Administration "administrative guideline" the product oan no longer be marketed. The mechanism by whloh PCB's from Aroolor-oontalnlng ooatlngs oontamlnate the allage, and in turn the llvestook,was originally thought to be only mechanical by flaking off of contaminated chips. It is not uncommon for these ohlps to contain $,000 to 10,000 parts per million (ppm) of Aroclor. However, recent work at the University of Kentucky has shown that the Aroclor in the ooatlng materials Is soluble In silage juloes. The replacement of contaminated silos with new ones would represent a considerable financial loss to the dairyman or oattle- man. Therefore, studies are being conducted at the Ohio Agricul tural Research and Development Center to determine how effective ooatlng materials may be to prevent further PCB contamination of silage. The results of the first trial are presented In this rspor HONS 070667 o 3 ) Two Unas of Investigation were used In this study. A silo which had been previously treated with an Aroclor 1254 containing ooatlng was used as the major structure. The silo was divided with sheets of Masonite Into test sections, the walls were wire brushed to remove loose material, and the surfaces were coated with materials (table 1) considered Impervious to the PCB In the silo walls. Untreated and wire brushed test sections were left as controls. The silo was filled to the top of the experimental seotlons with chopped corn (32% DM). Four months later the spoilage was removed and silage samples were collected at three levels In the silo and at five dlstanoes from the silo wall. The samples were analyzed by gas-ehromatography (with electron capture detector) for Aroclor 1254 In order to determine the effectiveness of the coating materials and to determine the vertical and horizontal distribution of any PCB contamination. In other studies oonorete silo blocks were painted with a ooatlng that contained Aroclor 1254* Then the blooks were reooated with the materials designed to prevent PCB migration. These blooks. were sailed In 55 gallon barrels of chopped corn for four months. The silage was then sampled at three depths and at three distances from the blooks and analyzed for Aroclor 1254. The resulta of the teats oonduo ted In the silo are shown In figure 1. In these tests all of the coating materials reduoed the amount of Aroclor 1254 that migrated Into the silage when ooraparec to the untreated oontrol. However, only coatings I and VIII re duced the contamination level to below that of the wire brushed oontrol surfaoe. Statistical analyses to determine If the observed HONS C7C668 oo 4- dlffsrenoes are significant hava not yet bean conducted. The raaaona why coatings I and VIII provided the bast barrier to the FCB are not yat thoroughly understood. It la suspactad that the olvsnta used In the other eoatlnga tested (II, V, IX) dissolved more Aroolor from the silo wall than did the lsoprophyl alcohol and water used as solvents In coatings I and VIII respectively. When Aroolor becomes dissolved by the solvent of a new ooatlng It can than become a component of the new surface. In all test and control saotlons the majority of the con tamination was detected in the silage within 30 cm (1 ft.) of the silo wall. Likewise, the oonoentratlon of Aroolor was highest at the base of the silo. For an unknown reason all samples collected one meter (about 3 ft.) from the base had low levels of Aroolor. The levels of Aroolor 1254 In the silages from the blook experiments were not as high as those observed In silage from the silo (figure 2). The major reason for this was that the Aroolor 1254 eontalnlng ooatlng usod on the blooks oontalned less Aroolor than the coating that was used on the silo. All but three of the coating materials reduced the quantity of Aroolor 1254 oontamlnatlon of the silage to below that of the control blook. Aroolor oontamlnatlon observed adjacent to ooatlng I was higher than the control level. This cannot be readily explained slnoe this material reduced the oontamlnatlon In the silo. The solvents In coatings III and VTI probably dissolved the Aroolor In the base ooatlng and oaused the Increased FCB residue In the silage. hONS 070669 uo 5- The results of this blook study lndloate that the thiokness of the oover coating may be an Important consideration when . electing a coating to oover PCB contamination. This can. be readily seen by comparing ooatlng system III vs. 17 and II vs. IX. In both oases the latter coating system Included both a sealer and a top ooat, and the contamination was reduoed. The differences In effectiveness between ooatlngs VI and VIII, even though they are greatly different In composition, msy well be the thick ness of the ooatlng. Only a thin barrier was provided by a single ooatlng of 71, whereaa a single eoat of VIII was quite thick. These studies have not yet provided enough Information to reoommend a particular ooatlng or type of ooatlng to oover the PCB oontamlnatlon In silos. Riese trials are continuing In order to determine how long the ooatlng will remain bonded to tho silo wall and also to determine If the Aroelor will migrate through a ooatlng when used continually. The studies have provided Information about some of the neoessary properties of a coating If It Is to be used as an effeotlve barrier to PCB reslduos. These studies have Indicated that It Is not desirable to use a ooatlng whloh oontalns a solvent In whloh Aroslor 1254 Is very soluble. It also appears that the ooatlng thiokness Influences the effective ness of the ooverlng. Ihe author would like to thank the following organizations for their support of this research. Aotlonl Plastic Ino., Charlotte, Michigan Ohio Milk Producers Federation, Columbus, Ohio Ohio Silo Company, Wooster, Ohio Silver Bros. Inc., Wooster, Ohio Standard Dry Wall Products Inc., Miami, Florida United Cooperatives Inc., Alllanoe, Ohio HONS 070670 TABLE 1. The Identlfloatlons, principle components, Brlnolple solvents end test locations of the barrier coatings tested. Coating Humeral I IX III IV V VI . VII VIII IX XI Coating Type Furfuryl Vinyl Principle Component or Resin Purfuryl alcohol resin Vinyl resin Epoxy sealer Epoxy resin Epoxy sealer + Epoxy resin Epoxy topooat Epoxy sealer Epoxy resin Epoxy topooat Water epoxy Epoxy resin Thermoplastic Alkyl aeromatlc hydrocarbon hydrocarbon Hydraulic cement Cement Vinyl sealer Vinyl resin + Vinyl topcoat Vinyl resin Epoxy sealer + Water epoxy Epoxy resin Epoxy resin Principle Solvent Isoprophyl alcohol Methyl lsobutyl ketone Xylene Xylene Cellosolve Water Kerosene Water Methyl lsobutyl ketone Methyl lsobutyl ketone Xylene Water Where Tested silo -SToeE XX XX X X XX X X XX X XX MOIHS 07067i HONS 0 7 0 6 7 2 Figure 1. The parts per million (ppm) of Aroolor 1251* (dry matter basis) In corn silage from experimentally 00ated, wire brushed and untreated aeotlons of PCB contaminated silo. See table 1 for a desorlptlon of the coating materials used. M " meter 0.30 H =1 ft. CM = centimeter 2.54 CM = 1 In. Figure 2. The parts per million (ppm) of Aro- elor 1254 (dry matter baels) In oorn silage from 55 gallon barrels that contained oonerete silo blocks. The blocks were treated with an Aroclor 1254 containing coating then reooated with the experimental coatings. See table 1 for a des cription of the coating materials used.