TO D E T E R M I N E

A SAMPLING

METHOD

INSECTICIDE

RESIDUES

ON S U R F A C E S

A N D ITS A P P L I C A T I O N TO FOOD-HANDLING R. B. L E I D Y ,

C. G. WRIGHT,

ESTABLISHMENTS andH.

E. D U P R E E ,

Jr.

Pesticide Residue Research Laboratory and Department of Entomology North Carolina State University, Raleigh, NC 27650, U.S.A.

(Received December 18, 1985) Abstract. K n o w n a m o u n t s o f acephate, chlorpyrifos, a n d diazinon were applied to Formica, unfinished plywood, stainless steel, and vinyl tile. Cotton-bail and dental wick materials were dipped in 2-propanol a n d " s w i p e d " over the treated surface area two time. More acephate was found on the second swipe compared to the first from vinyl tile, similar a m o u n t s on both swipes from plywood, and less on the second swipe from formica and stainless steel. The ratio of chlorpyrifos on Swipe 1 compared to Swipe 2 found with cotton-ball on both formica and stainless steel surfaces was equivalent (6 : 1), but a considerable difference was seen when two dental wick swipes were used. Residues of diazinon removed from formica and stainless steel were equivalent, regardless of the swiping material used. Residues of chlorpyrifos were detected by taking swipes of surfaces in two restaurants and a supermarket up to 6 m o after a prescribed application by a commercial pest control firm. The data show that measurable a m o u n t s of chlorpyrifos can be detected on surfaces not treated with the insecticide for at least 6 mo.

1. Introduction A number of studies have shown that insecticides can be detected in the ambient air o f buildings, houses, and even vehicles used to transport pesticides (Wright and Leidy, 1978, 1980; Wright et al., 1982). Even though a number of insecticides are considered nonpersistent they can remain inside structures for 6 weeks or more (Wright et al., 1983). Thus, it seems reasonable to assume that aerosols containing pesticides will settle onto surfaces within the structure. Jackson and Wright (1975) showed that prescribed treatments of rooms with chlorpyrifos and diazinon resulted in detectable residues being found at various locations within the treated room. Presently there are no tolerances in the United States for insecticides in either food or food-contact surfaces in buildings which manufacture, prepare, a n d / o r serve food products. As a result, the state o f Michigan has interpreted the absence of established tolerances to mean that the insecticide level will be zero (Oudbier, 1982). Regulatory programs have been established in some states to determine if insecticide residues were present; and if found, a pest control firm could be fined a n d / o r have its license revoked or suspended. This could happen even if the insecticides were applied by methods prescribed on the label. The state of North Carolina initiated a program to determine if insecticide residues were present in food-handling establishments, but the program was stopped because there were neither guidelines established for sampling nor were there set tolerances to determine if a food handler were in violation. The studies reported here were Environmental Monitoring and Assessment 9 (1987) 47-55. 9 1987 by D. Reidel Publishing Company.

48

R. B. LEIDY E T AL.

conducted (1) to determine how to obtain representative samples from several surface types (2) to determine the efficiency of two swiping materials to absorb chlorpyrifos and diazinon and (3) to use the sampling method to gather samples in commercial food-handling and -preparation establishments.

2. Materials and M e t h o d s

2.1. SAMPLINGTECHNIQUES Four materials commonly found in food-handling establishments, were selected for the initial study: formica, fir plywood (unfinished), stainless steel, and vinyl tile. Each piece of material was 5.0 by 35.6 cm, which was determined from the ability to apply a single uniform spray of insecticide. A stainless steel template with a 2.5 by 31 cm (1.0 inch by 12.0 inch) slot cut from the center was placed over the surface to be treated. It was felt that a swipe with this dimension would obtain a representative amount of insecticide found on a 100.0 cm2 surface. In addition the 1.0 by 12.0-inch slot could be fabricated readily, since only a standard 12-inch ruler is required to measure the area. A known concentration of acephate (15.0 I~g m L - l in n-hexane) was placed uniformly within a scribed 2.5 by 30.5 cm area in the center of the slot, and the solvent was allowed to evaporate. Two absorbent materials were used to wipe the surface: cotton balls (weight, ca. 0.4 g) and dental wick (2.5 by 1.0 cm diam., weighing ca. 0.5 g). Both were purchased from commercial vendors and used without further cleaning, because it was found that extracting the materials as described below gave no extraneous peaks when analyzed. Each swiping material was dipped in 10 mL of 2-propanol, squeezed against the side of the glass vial with forceps, and wiped up and back along the longitudinal axis of the treated area. To determine if the insecticides were transferred to the sampling material, a second "swipe" was wiped up and back over the treated area. Each swipe was placed in a separate vial (precleaned with soap and water, acetone rinsed, distilled water rinsed and dried at 200 ~ containing 10 mL of 2-propanol, sealed, placed in an insulated box containing dry carbon dioxide and the vials were taken to the laboratory for analysis. In the second sudy, formica and stainless steel plates were treated with chlorpyrifos and diazinon as described above, The following concentrations of each insecticide were used: chlorpyrifos: 20 lxg m L - l , diazinon: 10 gg mL-1, 15 ~tg mL-1. Each treatment was replicated five times. Both the cotton ball and dental wick swipes were used. The third study involved sampling a large supermarket and two restaurant kitchens (one in a hotel) for residual chlorpyrifos. A history of insecticide application was obtained for each facility from the pest control firm routinely servicing the three accounts, and it was agreed that no chlorpyrifos application would be made during the sampling period. Based upon the areas treated by the service technicians, six sampling sites (eg walk-in cooler doors, preparation tables, exhaust hoods) were

A SAMPLING METHOD TO DETERMINE INSECTICIDE RESIDUES ON SURFACES

49

selected in each location. In the supermarket, one sampling site was a large exhaust fan over a stove in the delicatessen section, because residues could be adsorbed to the grease and oils in this area. Bendiocarb and diazinon were substituted for chlorpyrifos for the remainder of the sampling period. Originally, samples were to be taken at 2-mo intervals for 1 y, but the cooperating applicator firm was underbid by a competitor and samples were collected for only 6 mo. The three accounts had been treated with chlorpyrifos by the new applicator prior to the time for the 8-mo samples. Temperature and relative humidity were recorded in each building prior to sampling. Single swipe samples, using the cotton balls, were taken, as described above, on several surface types in six locations in each of the three buildings (Table I). In addition, a cotton ball was exposed to the air at each sampling site and placed immediately in 10.0 mL of 2-propanol to use as a control to ensure that no interferences were absorbed from the air. Cotton balls were selected because of the data generated from the second study. Subsequent samples were taken at 2-mo intervals from the same sites with a minimum of 10 cm spacing between areas sampled previously, to ensure that no overlap occurred. 2.2. ANALYTICAL TECHNIQUES Samples were returned to the laboratory and placed immediately into a Soxhlet thimble (35 by 80 mm O.D.) contained in a 500 mL soxhlet apparatus. The 10 mL

TABLE I Sites in commercial food handling establishments sampled for residues of chlorpyrifos a Location

Surface

Sample Description No.

Restaurant kitchen

Stainless steel Stainless steel Stainless steel Formica Ceramic tile Structural clay tile Stainless steel Stainless steel Stainless steel Formica Ceramic tile Structural clay tile Stainless steel Stainless steel Stainless steel Formica Ceramic tile Ceramic tile

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Hotel Kitchen

Supermarket

Door surface inside walk-in cooler Tabletop Overhead exhaust hood frame Tabletop of cabinet Archway dividing kitchen Floor baseboard Door surface inside walk-in cooler Tabletop Overhead exhaust hood Tabletop Wall 2.5 cm from ceiling Floor baseboard Door surface inside walk-in cooler Tabletop in produce preparation area Overhead exhaust hood in delicatessen section Tabletop-delicatessen section Restroom wall avobe sink Restroom wall below sink

a Single swipes taken at 2-mo intervals. (Successive samples taken from surface adjacent to previous, at 10 cm interval to prevent overlap.)

50

R.B.

LEIDY ET AL.

of 2-propanol were added and the sample container was rinsed with two 10 mL portions of ethyl acetate. The rinse and an additional 200 mL of ethyl acetate were added to the Soxhlet apparatus and extracted 4 h at eight (8) turnovers per hour. After cooling, the solvent was evaporated to 2 to 3 mL at 40 ~ under reduced pressure and transferred quantitatively to 12 mL tubes for analysis by gas-liquid chromatography. The chromatograph was a Tracor Model 222 equipped with a flame-photometric detector operated in the phosphorous mode. Gasses to the detector were hydrogen and air at 50 and 80 mL min -1, respectively. The column was U-shaped glass (183 by 0.2 cm (I.D.)) packed with 4% SE-30 + 6% Q F - 1 on Gas Chrom Q (100/120). Nitrogen was the carrier gas at a flow rate of 60 mL m i n Temperatures were as follows: inlet, 200 ~ detector, 200 ~ oven: 195 ~ for acephate: 185 ~ for chlorpyrifos and diazinon. To confirm that the peaks seen were the compounds, a second column (U-shaped glass, 122 by 0.2 cm (I.D.) packed with SP-2250 on Supelcoport (80/100) was used. Oven temperatures were as follows: Acephate, 180 ~ chlorpyrifos and diazinon (170 ~ Data were quantitated by the peak height method using standards of known concentration. The efficiency of the analytical method was determined by adding known amounts of three insecticides, individually, to the swiping material and analyzing the fortified samples by the same method. Using the same pipette, each insecticide was added to a 12-mL tube and diluted with solvent (spike) to determine recovery values. Two fortified swipes of each type, two spikes, plus an untreated swipe of each material were analyzed with each set of samples during the preliminary study. Two fortified cotton-ball swipes, two spikes, and an untreated cotton ball were analyzed with each set of samples obtained from the commercial establishments.

3. Results and Discussion

A number of materials and solvents could be used to conduct swipe tests on surfaces. It was felt that two criteria had to be met: that the material and solvent be readily available and that they be nondestructive to the surface being swiped. Cottoncontaining materials and 2-propanol were chosen for these reasons. Too, neither the cotton ball nor the dental wick required any cleanup prior to use; they could be manipulated during sampling and in the laboratory with minimum effort; they were transported easily, and had a large surface area-to-size ratio. Two-propanol was chosen because of its degree of polarity relative to other solvents and its relative nondestructiveness to surfaces when compared to solvents such as ethyl acetate and methylene chloride. When known amounts of acephate, chlorpyrifos, or diazinon were added to untreated cotton balls and dental wicks and analyzed, the recovery values were approximately equivalent among insecticides (Table II). Average recoveries from the three insecticides on cotton bails and detal wicks were 98.7 and 95.3%, respectively. Because the cotton materials were free of interfering compounds, no cleanup was

A SAMPLING METHOD TO DETERMINE INSECTICIDE RESIDUES ON SURFACES

51

required after the extraction and concentration steps, which enhanced recovery averages. When 15 Ixg of acephate were applied to four different surfaces, the second swipe contained a considerable amount of acephate (Table III). For vinyl tile, more acephate was found on the second swipe than on the first. The amounts found on the unfinished plywood were equivalent. Apparently the solvent from the first swipe causes the acephate, which is entrapped in the sample matrix, to solubilize and move toward the surface where it was removed by the second swipe. A comparison of the ratio of the amount of acephate on swipe 1 to swipe 2 between the two materials showed that the cotton ball swipe trapped more than the dental wick, probably because of the larger surface area. The dental wick material, being rolled into a cylinder, was not unrolled; thus, less surface area was exposed. Based on the data for acephate, two materials, formica and stainless steel, were used in the second experiment; and known amounts of chlorpyrifos and diazinon

TABLE II Recoveries of three insecticides from untreated cotton balls and dental wick swipes a Insecticide

Material

A m o u n t Added

No. Samples

Range

(~tg)

Acephate

Cotton Ball

Dental Wick

Chlorpyrifos

Cotton

Dental Wick

Diazinon

Cotton Ball

Dental Wick

5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20 5 10 15 20

Average Recovery

(07,) 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

98-115 95-113 96-99 92-97 95-99 82-95 95-106 87-94 96-104 88-95 93-105 96-101 90-99 91-102 87-97 94-104 94-103 92-100 86-98 96-107 88-99 96-107 90-99 87-98

106 106 98 94 98 86 100 92 101 93 98 97 96 97 93 99 99 97 94 102 94 99 96 93

a Insecticide applied individually to a single cotton ball or dental wick, solvent allowed to evaporate, and material extracted by analytical procedure.

52

R.B. LEIDY ET AL.

were added to these surfaces and swiped with both materials (Table IV). The ratio of chlorpyrifos found on formica was equivalent to that for stainless steel swiped with the cotton balls, whereas there was a considerable difference between the dental-wick swipes. The amount of chlorpyrifos removed from the stainless steel by the second dental wick was greater than that for the second cotton ball, which caused the ratio difference. This could be due to pipetting errors, swiping variations, the amount of alcohol left on the surface from the first swipe, or the amount of alcohol contained in the second swipe. The amounts of diazinon removed from the two surfaces were approximately equivalent regardless of the swiping material used. If the ratios of insecticide found on swipe 1 to that found on swipe 2 are compared, diazinon residues were more equivalent than those of chlorpyrifos. The data show that most of the chlorpyrifos is removed by the first swipe. This could be due to a lack of penetration into the sample matrix because of the molecule of chlorpyrifos interacting primarily at the surface of the material. Diazinon, being more volatile, could penetrate and move into the matrix; and when the alcohol from the first swipe came into contact, residues of diazinon migrated back to the surface and were trapped by the second swipe. Based on these data, a third study was done to determine if residues of chlorpyrifos

T A B L E III Residues o f acephate swiped from different surfaces a Surface type

Swab No.

Material Cotton Ball

Dental Wick

A m o u n t Found

Formica

Plywood (untreated)

Stainless Steel

Vinyl Tyle

1 2 Ratio, 1 2 Ratio, 1 2 Ratio, 1 2 Ratio,

Swab 1 : 2

Swab 1 : 2

Swab 1 : 2

Swab 1 : 2

(~g)

(~g)

8.9 + 0.2 b 5.3--+0.1 1.7 4.8 -+ 0.4 4.7_+0.3 1.0 5.9 -+ 0.2 2.9-+0.1 2.0 6.6 -+ 0.4 7.6-+0x5 0.9

6.3 --+0.1 4.4-+0.3 1.4 4.6 -+ 0.2 5.5_+0.5 0.8 7.7 + 0.1 6.5-+0.1 1.2 4.7 -+ 0.6 7.6_+0.4 0.6

a Fifteen gg analytical grade acephate spread over a 2.5 by 30.5 surface; solvent allowed to dry and surface swiped with 2 cotton balls or 2 dental wicks, each in 2 directions along the longitudinal axis. There were 5 replications. b -+ S.E.

A SAMPLING METHOD TO DETERMINE INSECTICIDE RESIDUES ON SURFACES

53

could be detected in commercial food-handling establishments treated by a commercial pest control firm (Table V). Initial samples were taken 18, 14, and 9 days after the last application to the restaurant kitchen, hotel kitchen, and supermarket, respectively. Temperature and humidity measurements were taken prior to sampling and averaged 24 ~ and 66~ (restaurant-kitchen); 25 ~ and 55070 (hotel-kitchen); 22 ~ and 67070 (supermarket), respectively. Single swipes were taken because the control study showed that most of the chlorpyrifos would be removed by the first swipe. The largest amount of chlorpyrifos found was along the floor baseboard (structural clay tile) in the restaurant kitchen. At 6 mo 0.019 ~tg cm - 2 was detected at this location and measurable amounts were found in four of the six sites. Very low amounts were detected in the hotel kitchen and supermarket for 2 mo, and residues were below the detectable limits at 4 and 6 mo. The restaurant kitchen had a large infestation of German cockraches Blattella germanica (L), and had been treated more

T A B L E IV Residues of chlorpyrifos and diazinon swiped from two surfaces a Insecticide

Surface type

A m o u n t added (~g)

Swab No.

Material Cotton Ball

Dental Wick

A m o u n t found (.g) (.g)

Chlorpyrifos

Formica

20

Stainless steel

20

Formica

10

Ratio, Swab 1 : 2

Ratio, Swab l :2 Diazinon

2 Ratio, Swab 1 : 2 Stainless steel

10 2 Ratio, Swab 1 : 2

Formica

15 2 Ratio, Swab 1 : 2

Stainless steel

15 2 Ratio, Swab 1 : 2

18.5• 3.0• 6.2 10.8• 1.8• 6.0 5.2• 4.3• 1.2 5.9• 3.9• 1.5 9.8• 3.5• 2.8 11.6• 5.7• 2.0

o

17.0• 3.1• 5.5 9.8• 5.2• 1.9 7.2• 5.7• 1.3 6.1• 3.8• 1.6 9.2• 4.2• 2.2 8.0• 4.3• 1.9

Analytical grade chlorpyrifos or diazinon spread over a 2.5 by 30.5 cm surface; solvent allowed to dry and surface swiped with two cotton balls or two dental wicks, each in two directions along the longitudinal exis. There were five replications. b -+S.E. a

54

R.B. LEIDY ET AL.

frequently than the other establishments, which could explain the presence of measurable amount of chlorpyrifos at 6 mo. These data show that measurable amounts of chlorpyrifos can be detected on surfaces which were not treated with the insecticide for at least 6 mo and support data in another study (Wright et al., 1983) in which measurable amounts of chlorpyrifos and diazinon were detected on surfaces 41 days after a prescribed crack and crevice application. There is a need to determine an action level considered safe by those Federal and State agencies responsible for establishing acceptable levels. Realistic insecticide levels must be set, because prescribed applications following label instructions will result in residues being deposited on nontarget surfaces which, if sampled by an inspector, can result in fines or license suspension to the pest control firm.

TABLE V Residues of chlorpyrifos from various surfaces in commercial food handling establishments ~ Location

Type Surface b

Sample Time after application No.C (Months) 0 2 Concentration (Ixg cm -2)

Restaurant Kitchen

Hotel Kitchen

Supermarket

Stainless Steel Stainless Steel Stainless Steel Formica Ceramic Tile Structural Clay Tile Stainless Steel Stainless Steel Stainless Steel Formica Ceramic Tile Structural Clay Tile Stainless Steel Stainless Steel Stainless Steel Formica Ceramic Tile Ceramic Tile

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

0.001 0.003 0.008 0.001 0.001 0.009 0.002 0.001 0.001 0.001 0.001 0.001 0.001 0.001 < 0.001 0.001 0.001 < 0.001

4

6

(Ixg cm -2)

(~tg cm -2)

(~g cm -2)

0.003 0.002 0.005 0.006 0.002 0.003 0.002 0.002 0.001 0.003 0.003 0.002 0.002 0.001 0.001 0.001 0.001 0.001

0.001 0.001 0.001 < 0.001 d 0.002

A sampling method to determine insecticide residues on surfaces and its application to food-handling establishments.

Known amounts of acephate, chlorpyrifos, and diazinon were applied to Formica, unfinished plywood, stainless steel, and vinyl tile. Cotton-ball and de...
410KB Sizes 0 Downloads 0 Views