688 Analyst, October, 1968, Vol. 93, $$. 688-690 Consistency of RSt Values of Six Organophosphorus Pesticides Resolved by Thin-layer Chromatography in the Presence of Plant Extracts without Elaborate Clean-up C. E. MENDOZA, P. J. WALES AND D. F. BRAY (Research Laboratories, Food and Drug Directorate, Ottawa 3, Ontario, Canada) This paper reports statistical analyses of the ratio of the distance travelled by each of five organophosphorus pesticides to that travelled by parathion in the presence of plant extracts, without elaborate clean-up. Mixtures of azinphos-methyl, carbophenothion, diazinon, ethion, malathion and mevinphos in the presence of apple, beet, carrot, lettuce or pea extracts were evaluated with reference to the parathion internal standard. GALANOS and Kapoulasl enumerated the sources of variation in the determination of RF (the distance travelled by a compound divided by the distance travelled by the solvent front) values of chemical compounds.Dhont and de R ~ o y , ~ and Severin: found that the Rst values of some 3,5-dinitrobenzoates on plates coated with silica gel were consistently and conveniently obtained- Distance travelled by the studied substance X Rst = Distance travelled by the reference standard st Our observation that R, values were greatly affected by the changes in room humidity concurred with that of Reiche14; the R, values obtained were low in winter and high in summer even at constant temperature of the chromatographic tanks. In addition, the same compound spotted on a thin-layer chromatographic plate at different positions along the origin line gave slightly different spot locations, although the solvent front moved evenly.6 Because of such limitations, R,t values were calculated to substantiate further the identity of the pesticides in plant extracts without comparing each spot with the pesticide standards. As the compounds used belonged to different organic classes, R,t was used instead of RF0.l Parathion was chosen as the standard reference because its migration rate was intermediate between the slow and fast-moving compounds; its R, value is 0.41.In addition, it is stable and sensitive to the enzyme-detection technique.5 Rst values of azinphos-methyl, carbo- phenothion, diazinon, ethion, malathion and mevinphos were calculated in the absence or presence of extractives from beets, apples, carrots, lettuce and peas.The R,t values were evaluated to determine the degree of variation in the migration pattern of the six pesticides on plates with a 450-p thick silica gel layer. EXPERIMENTAL PREPARATION OF THE SAMPLE- Each 50-g plant sample was extracted and partitioned according to McLeod, Mendoza, Wales and McKinley.6 The combined partition fraction was fortified with azinphos-methyl at 1-00 p.p.m., carbophenothion at 0.50 p.p.m., diazinon at 0.10 p.p.m., ethion at 0.10 p.p.m., malathion at 8.00 p.p.m., mevinphos at 0.25 p.p.m. and parathion at 1-00 p.p.m. The fortified extract was made up to 200 ml with hexane; a portion was concentrated 20 times under a gentle flow of nitrogen. THIN-LAYER CHROMATOGRAPHIC PROCEDURE- Aliquots equivalent to 100, 150 and 200 mg of plant samples were spotted in duplicate on each thin-layer chromatographic plate, coated with a 450-1" thick layer of MN-Kieselgel G-HR (silica gel).The pesticides on the thin-layer chromatographic plates were resolved at 25" C in a glass tank with 15 ml of acetone diluted to 100 ml with hexane; the solvent front 0 SAC and the authors.in the where MENDOZA, WALES AND BRAY 689 The thin-layer chromatographic - enzyme was allowed to reach a distance of 15 cm. inhibition technique5 was used to detect the pesticides on the plates. STATISTICS- The following model was used for the statistical analysis of the Rst values of the pesticide presence of plant co-extractives- Yijkl = p + pi + Dj(i) + Lk(ij) + El(ijk), Yijkl is the observed Rst value for each pesticide; p is the over-all mean; Pi is the fixed effect of the i t h plant sample; Dj(i) is the random effect of the j t h day in the ith plant sample; Lk(ij) is the random effect of the kth plate within the jth day in the ith plant sample; El(ijk) is the random variation between spots of the same pesticide on the same plate.and Because only one R,t value was observed on each plate for each standard pesticide (without plant extracts), a slightly different model was used for the analysis of these data- where the effects p and Dj(i, are as defined before; Yijk = p + Gi + Dj(i) + Rk(ij), Gi is the effect of the ith grouping (arbitrarily grouped according to plant sample); Rh(ij) is the residual source of variation. and RESULTS AND DISCUSSION The chromatograms showed that the distance travelled by the pesticides was not affected Table I shows the mean RSt values of the standard pesticides and of the same pesticides In all instances, the variance among the six values for by plant extracts or by the size of the aliquots spotted.in the presence of plant extracts. each pesticide, within plates, is small. TABLE I MEAN Rparathfon VALUES OF SIX ORGANOPHOSPHORUS PESTICIDES* Pesticide A I I Azinphos- Carbo- Mevinphos methyl Malathion Parathion Diazinon Ethion phenothion Standard? 0- 32 0.52 0.76 1.00 1.11 1.22 1.47 With plant extractives3 0.30 0.51 0.75 1.00 1-11 1.22 1.47 * Rparathlon = the distance travelled by a compound divided by the distance travelled by parathion; MN-Kieselgel G-HR, 450-p thick layer, 15 ml of acetone diluted to 100 ml with hexane solvent system, 25" C room temperature.t Each value is a mean of 20 determinations. Each value is a mean of 100 determinations, except that for diazinon, which is a mean of 99 values. Table I1 shows that plant samples were not a significant source of variation but the days and the plates gave variation larger than that between individual pesticide spots on the same plate. The significant variation for days and plates may be attributed to changes on room humidity on different days, when the plates were made and used in chromatography. Table I11 indicates that the groupings (arbitrarily grouped according to plant sample) did not give a significant source of variation for any pesticides, with the exception of malathion.The significant mean square for groupings for malathion was most likely caused by the low mean square for days within groupings. For ethion, the variation from day to day was only significant at the 5 per cent. level, and may be attributed to changes in room humidity on different days.690 MENDOZA, WALES AND BRAY TABLE I1 ANALYSES OF VARIANCE OF THE &, VALUES OF THE PESTICIDE STANDARDS IN THE PRESENCE OF PLANT EXTRACTIVES Mean squares for each pesticide Azinphos- Source d.f. Mevinphos methyl Malathion Diazinon Ethion Plant samples . . . . 4 0.012972 0-007275 0.001450 0-003225 0.001942 Dayslplant samples . . 5 0.010328* 0-016460t 0.004380* 0.004040t 0-003564* Plates/days/plant sam- ples . . .. . . 10 0.0022377 0.002270t 0.000920t 0.000370 0.000556 Between spots/days/ plant samples .. 100; 0.000367 0.000663 0-000190 0.000227 0.000380 (0.001352) (0.002112) (0.000599) (0*000557) (0.000660) d.f. = Degrees of freedom. * Significant at Po.oa. t Significant a t Po.ol. $ Degrees of freedom for diazinon = 99. Figures in parentheses are sums of variance components. 7 Cabo- phenothion 0.014200 0*023680* 0.004570t 0.000675 (0.0029 16) TABLE I11 ANALYSES OF VARIANCE OF THE R,t VALUES OF THE PESTICIDE STANDARDS Mean squares for each pesticide I- A \ Azinphos- Carbo- Groupings . . . . 4 0.002712 0*001338 0.001220* 0.001580 0*000500 0.00415 Dayslgroupings . . 5 0.000895 0.001530 0.000180 0-000820 0.001060t 0.00454 Residual/days/groupings 10 0.000725 0.000910 0.000600 0.000250 0.000310 0.00144 Source d.f. Mevinphos methyl Malathion Diazinon Ethion phenothin (0*000805) (0.001220) (0.000600) (0.000535) (0-000685) (0.0029 d.f.= Degrees of freedom. * Significant a t Po.o6 when tested against the days, not significant when tested against the residual. t Significant a t Po.o5. Figures in parentheses are sums of variance components. The variance components, calculated from the R,t values of standards, shown in Tables I1 and I11 (figures in parentheses), were similar to those of the same pesticides in the presence of plant extractives. This similarity permitted the use of individual error terms from the first analysis to test the significance between individual standard values and means of the six values (per individual pesticide in the presence of plant extractives) per plate.Only five differences out of 120 observations were found greater than the 5 per cent. least significant difference values. This indicates that no significant difference exists between the R,t values of a pesticide standard and of the same pesticide in the presence of plant extractives, both determined on the same plate on the same day. Analyses showed that R,t values for the six organophosphorus pesticides were consistent, regardless of the presence of plant extractives and of the location of the spot origin. The effect of the origin location was negligible when the test compounds and the parathion reference standard were contained in the same sample spot. When close control of humidity is a prob- lem in the determination of R, values, R,t values relative to parathion on the same path of migration can, therefore, be used to substantiate further the identity of the six pesticides studied. We thank Drs. K. A. McCully, W. P. McKinley and E. Somers, and Miss W. van Walbeek for their constructive criticism of the manuscript. REFERENCES 1. 2. 3. 4. 5. 6. Galanos, D. S., and Kapoulas, V. M., J . Chromat., 1964, 13, 128. Dhont, J. H., and de Rooy, C., Analyst, 1961, 86, 527. Severin, M., J . Chromat., 1967, 26, 101. Reichel, W. L., Ibid., 1967, 26, 304. Mendoza, C., Wales, P., McLeod, H. A., and McKinley, W. P., Analyst, 1968, 93, 173. McLeod, H. A., Mendoza, C . , Wales, P., and McKinley, W. P., J . Ass. Ofl. Analyt. Chem., 1967, Received February 22nd, 1968 50, 1216.