706
ANALYTICAL CHEMISTRY Analyses of Known Amount of Aldrin in Compound 497
Table 111.
Aldrin Added to Compound 497, o/c
Aldrin Found. ?&
3.8 16.0 25.3
4.5 16.6 25.8
Table IY. Aldrin Added. P . P . M . 0 0 0 0 1 1 5 5 0
Recovery of Aldrin Added to Cow's Urine Aldrin Found, P.P.M. Uncorrected Corrected for blank for blank 0 20" ... 0 20" 0 62
3 6 0 0 0 0
0 60 1 10 1 00 5 45 5 36
...
0 40 0 41 0 90 0 93 4.91 4 84
Recovery Corrected for Blank, %
methyl ester] and Octacide 264 [the A'-octyl imide of bicyclo(2.2.1 )-5-heptene-2,3-dicarboxylic acid ] do produce a red color with an :tl)sorption maximum in the same region as that obtained in the analysis of aldriri. However, because Dimalone is an insect repellent :tnd Oct:ic.ide 2ti-l is a pyrethrum synergist, neither of these products is likc,Iy to be encountered in commercial mixtures of aldrin. The response to the colorimetric test for aldrin of some chemicals commonly used for insect control is listed in Table V.
Table
V. Response of Some Substances Used for Insect Control to Color Test for Aldrin
..
io
82 90 93 98 97
Interfering substances Noninterfering Substances Repellent Pyrethrum synergist Chlordan 0 0 DDT hIethoxychlor I1 Hexachlorocyclohexane ( B H C )
8
Calciilated as aldrin
//
two 5O-id. batches of hexane. Occasional emulsions were broken by centrifuging. The hexane extracts were dried with anhydrous sodium sulfate, filtered, evaporatively concentrated, and analyzed for aldrin as described under Procedure B. The results of these analyses are shown in Table I\'. Similar esperiments with human urine gave slightly better recoveries. For analysis of aldrin commercial dusts or wettable powders, a preliminary extraction with hexane is necessary. h Soshlet apparatus containing a sample large enough to be representative is satisfactory. A minimum of 30 minutes' extraction time has been used, and 1 hour is recommended. INTERFERENCES
The commonly used organic insect tosicants do not interfere in the analysis of aldrin by this new procedure. Hexane solutions of chlordan, DDT, methoxychlor, hexachlorocyclohexane (BHC), and toxaphene treated according to the procedure for determining aldrin gave a pale yellow color similar to that of the blank. Dimalone [bicyclo-(2.2.1)-5-heptene-2,3-dicarbosylic acid di-
6 Dimalone
I
XR
0 Octacide 264
ACKNOWLEDGMENT
The authors wish to acknowledge the assistance of Yuji A. Tajima of this laboratory in conducting the spectroscopic determinations and to Phebe Hines for aid in carrying out the analysis. LITERATURE CITED (1) Alder, K., and Stein, G., Ann., 485, 211 (1931). (2) C h e n . Eng. News, 26, 3854 (1948); (3) Lidov, R. E., Bluestone, H., Soloway, S.B., and Kearns. C. W . . Adaances in Chemistry Series, 1, 175 (1950). (4) Lindsay, R. O., and Allen, C. F. H., "Organic Syntheses." Vol. 22, p. 96, Xew York, John Wiley & Sons, 1942. (5) Schoutissen, H. A. J., J . Am. Chem. SOC.,55, 4535 (1933). RECEIVED April 12, 1949. Presented before the Division of Agricultural and Food Chemistry, Symposium on Economic Poisons, a t the 115th Meeting of the AMERICAN CHEMICAL SOCIETY, San Francisco, Calif.
Polarographic Determination of 0,O-Diethyl 0-p-Nitrophenyl Thiophosphate (Parathion) c. v. BOWEN A ~ FRED D I . EDWARDS, J R . Bureau of Entomology and Plant Quarantine, C'. S . Department of Igriccclture, Beltscille, Wd. Parathion may be determined quantitatively by means of the polarograph. The electrolysis is carried out in an acetone-water solution with 0.05 N potassium chloride as electrolyte, and 0.0170 gelatin as suppressor at 25' * 0.5" C. An accuracy of *170is obtained. Several commercial products were analyzed.
T
HE only reported method for the estimation of 0,O-diethyl
0-p-nitrophenyl thiophosphate (parathion) is a colorimetric procedure ( 1 ) based upon the reduction of the nitro group to an amino group with subsequent diazotization and coupling with N (1-naphthyl)-ethylenediamine to produce a color that may be measured. This procedure has had application in the determination of spray and dust residues where 90% recovery is satisfactory, but is not suitable for use in the assay of technical materials. Consequently, reliable and sensitive methods of analyqis are greatly needed for this new and highly toxic material in insecticidal formulations. Because the reduction occurs so readily with zinc in the above-mentioned procedure and nitrobenzene (6)was the first organic compound tc be investigated with the polaro-
graph, it uas considered probable that parathion would be easily reduced a t the dropping mercury electrode and thus be determined by this means. APPARATUS
A Sargent Model X X I polarograph was used in this investigation. The reduction was carried out in an H-type electrolysis cell with it saturated calomel reference cell in one arm ( 6 ) . .4 thermostatically controlled water bath maintained the cell a t 25' * 0.5' C. During the recordifig of the polarogram the air stirrer was stopped in order to eliminate vibration and the heating system was disconnected a t the bench outlet to remove the possibility of stray current effect ( 3 ) . It was observed that other operating
V O L U M E 2 2 , NO. 5, M A Y 1 9 5 0
707 Tqble I.
Analysis of C o m m e r c i a l S a m p l e s of Technical Parathion Sample No. Parathion Found, % Sample No. Parathion Found, % 1
8.4.7 83.9 85.3 Av. 8 4 . 6 92.3 91.5 92.0 Av. 9 1 . 9 98.1 97.5 97.7 Av. 9 7 . 8
2
3
4
93.1 92.0 93.2 Av. 9 2 . 8
5
94.4 93.5 95.3 Av. 9 4 . 4
~~
-. I
I
2
3
I
I
I
4 5 6 PARATHION-mgflOO
I
7 ml.
I
I
8
9
IO
Figure 1. S t a n d a r d Curves for Parathion D e t e r m i n a t i o n Sensitivity, microampere per millimeter, A , 0.020;
B, 0.030r
Material Commercial dusta Sample 1 (25%)
Parathion Present, %
,..
Sample 2 (25%)
PREPARATION OF STANDARD CURVES
The 0,O-diethyl 0-p-nitrophenyl thiophosphate w e d in the preparation of the standard curves was obtained by isolation from a high-purity technical parathion according to the method devised by Edwards and Hall ( d ) . I t was a crystalline material that melted sharply at 6 " C. The physical constants were in agreement with those published by Fletcher et al. (4). A sample of 0.4863 gram of this purified 0,O-diethyl O-p-nitrophenyl thiophosphate was dissolved in acetone to make 1 liter of standard solution. A 20-ml. aliquot, containing 9.73 mg., W&B placed in a 100-ml. volumetric flask and 30 ml. of acetone were added. Then 0.35 gram of potassium chloride and 0.6 gram of acetic acid were dissolved in about 25 ml. of water and added to the acetone solution; 0.01 gram of gelatin was dissolved in a few milliliters of water by warming, cooled, and added to the above, and the solution was brought to the mark with water. (This solution is 0.05 N with respect to potassium chloride, and 0.1 N with respect to acetic acid, and contains 0.01% of gelatin and 50% of acetone in water.) The acetic acid was added to prevent any hydrolysis of the ester during the electrolysis. The sample side of the H cell was emptied and rinsed by means of suction without being removed from the thermostat bath. The used mercury was retained in the suction flask. The cell was rinsed well with acetone and then with a portion of the sample solution before being filled with the sample solution. Prior to the electrolysis oil-pumped nitrogen was bubbled through the sample solution for 10 minutes to remove dissolved oxygen. The nitrogen was passed through a 1 to 1 acetone-water solution before i t reached the sample solution. F6r electrolysis the dropping mercury electrode was placed firrdy in the cell and the polarograph set to record the wave a t 0 to -1.5 volts a t a sensitivity of 0.020 microampere with maximum damping. Waves were recorded in duplicate for 0.020-, 0.030-, and 0.040-microampere sensitivity to allow for considerable leeway in the sire of the sample. The sensitivities of the polarograph refer to the microamperes corresponding to 1-mm. deflection of the recorder. Polarographic waves were obtained in the same manner for 7.29-, 4.86, and 2.43-mg. samples of parathion. Figure 1shows the average wave
~
Parathion Found Soxhlet Flask extraction, % extraction, %
.. 24.4 24.1 24.4 Av. 2 4 . 3
Sample 3 (1%)
...
...
Wettable powders Sample 1 (25%)
...
..
Sample 2 ( 2 5 % )
...
23.3 23.3 23.6 Av. 2 3 . 4
10.0
...
Sample 2
12.2
12.2 12 n 12.1 Av. 1 2 . 1
Sample 3
29.1
29.2 29.3 29.3 Av. 2 9 . 3
C, 0.040
electrical appliances, such as hot plates on the same bench, had a stray current effect on the polarograph.
~
T a b l e 11. Analysis of P a r a t h i o n Formulations
Synthetic dusts Samplezl
24.0 24.0 24 1' Av. 2 4 . 0 24.0 24 I 24 0 Av. 24 0 0.97 0.92 0.94 A v . 0.94 24.0 23.7 23.9 Av. 2 3 . 9 23.9 23.7 23.8 Av. 23.8 9.7 9.7 9.8
Av.
9.7
height for from 2 to 10 determinations a t each coneentration plotted against the concentration to give the standard curves. After some of the standardization polarogram had been obtained, it was decided that considerable time could be saved by using an aqueous stock solution of twice the normality of potassium chloride and acetic acid desired in the sample to be analyzed instead of weighing these materials for each determination, T h e gelatin was weighed fresh each day. ANALYSIS OF COMMERCIAL PRODUCTS
Technical parathion samples were analyzed in the same manner as the standard samples (Table I). For dust formulations a t least 1 gram was taken for a sample and made to volume with acetone to obtain an extract co. taining approximately 1 mg. per ml. of parathion based on the manufacturer's claims. The sample WBB shaken intermittently for 1 hour, and allowed to stand for 10 minutes, and a portion was centrifuged in a glass-stoppered tube until clear. An aliquot of this clear solution to contain approximately 10 mg. was taken and the procedure described wa8 followed. The recovery of parathion from dusts by this procedure was checked by extracting two of these commercia1,dusts in a Soxhlet apparatus. The results were found to be within the limite of accuracy of the method, aa shown in Table 11. Dusts of known parathion content prepared in this laboratory were
,
ANALYTICAL CHEMISTRY
708 analyzed after Soxhlet extraction, and the recovery as shown in Table I1 was found to give results also within the limits of accuracy of the method. DISCUSSION
Normal curves for the polarograms were obtained with the technical materials as well as with the purified sample. The decomposition potential of -0.30 volt and a half-wave potential of -0.39 volt were obtained against the saturated calomel electrode. pNitrophenol, which is a major contaminant of the technical parathion, does not reduce at the dropping mercury electrode until after the parathion has been completely reduced, and consequently does not interfere with the curve obtained in the analysis. T h e decomposition and half-wave potentials for pnitrophenol under the conditions for the determination of parathion were found to be -0.45 and -0.68 volt, respectively. Diethyl p-nitrophenyl phosphate, the oxygen analog of parathion, was investigated to ascertain whether it would interfere, if present, in the dhtermination of parathion. It was found, however, under the conditions used in this method to have a decomposition potential of -0.37 volt and a half-wave potential of -0.47 volt. A mixture consisting of one third parathion and two thirds osygen analog, instead of giving the anticipated broken wave beginning a t the decomposition voltage for parathion, gave a
normal curve with a decomposition potential of -0.34 volt. This indicates an interference if small amounts of the oxygen analog should be present, a situation not likely to occur with present methods of synthesis ( 4 ) . The polarographic method of analysis of parathion as described here has an accuracy of * 1%, and 2 mg. of 0,O-diethyl O-pnitrophenyl thiophosphate per 100 ml. of solution are apparently a minimum concentration for the sensitivities investigated. However, the polarograph used is equipped with resistors, so that a sensitivity of 0.003 microampere per millimeter may be used. At this sensitivity it would be possible t o obtain a sufficient wave height to determine parathion at a concentration of less than 1 p.p.m. LITERATURE CITED
(1) Averell, P. R., and Norris, M. V., ANAL.CHEM., 20, 753-6 (1948). (2) Edwards, F. I., and Hall, S. A , Ibid., 21, 1567 (1949).
(3) English, F. L., Ibid., 20, 889-91 (1948). (4) Fletcher, J. H., Hamilton, J. C., Hechenbleikner. I., Hoegberg, E. I., Sertl, 3. J., and Oassaday. J. T., J . A m . Chem. SOC.,70, 3943-4 (1948). (5) Kolthoff, I. M., and Lingane, J. J . , "Polarography," N e w York, Interscience Publishers, 1946. (6) Shikata, M., Trans. Faraday Soc., 21, 42-52 (1925). RECEIVEDOctober 1, 1949. Presented before the Division of Agricultural and Food Chemistry, Symposium on Economic Poisons, a t the 115th Meeting of the BUERICAN CHEMICAL SOCIETY. San Francisco, Calif.
Quantitat ive Determination of Certain Flavonol-3-glycosides By Paper Partition Chromatography THOMAS B. GAGE AND SIMON H. WENDER, University of Oklahoma, Norman, Okla. Binary mixtures of rutin, quercitin, isoquercitin, robmin, and xanthorhamnin and ternary mixtures of rutin, quercitin, and isoquercitin have been separated quantitatively b y paper partition chromatography. The mixtures, containing 10 to 40 micrograms of each pigment, were chromatographed on Whatman No. 1 filter paper with n-butanol-acetic acid-water (40-10-50 volume %). For absorption spectra determinations, blank filter paper strips were carried through the chromatographic procedure. The individual pigment zones (located under ultraviolet light) and corresponding areas of the blank strips were cut out and extracted by capillary leaching for 1 hour with 0.5% aqueous alumin u m chloride solution. The optical density of the flavonoid-aluminum chloride complexes a t their
S
EVERAL approaches to the quantitative determination of
flavonoid pigments have been offered in recent years. Porter, Brice, Copley, and Couch ( 4 ) have developed a spectrophotometric assay method for mixtures of rutin and quercetin based upon their absorption maxima in the ultraviolet region. This method is excellent for samples known to contain only rutin and quercetin, but it has not yet been adapted for mixtures where other flavonoid pigments are also present. A colorimetric method of analysis for total content of quercetinlike substances has been described by Wilson et al. (7). This method was later adapted for fluorometric analysis by Glazko and bo-workers (9). The latter two methods depend upon the color or fluorescence of certain flavonoid pigments in the presence of boric acid and are valuable
respective absorption maxima was determined with the Beckman Model DU spectrophotometer. Per cent recoveries, calculated on the ratio of optical density of each recovered pigment to the optical density of the aluminum chloride complex (containing the same initial amount of nonchromatographed flavonoid pigment), ranged between 92 and 95%. The flavonoid standards may each be carried through the chromatographic procedure on individual paper strips and the aluminum chloride complex formed by the leaching process. This procedure adjusts the standard for the losses occurring on the filter paper during chromatography. Recoveries of 98 to 101% have been calculated for mixtures after correcting in this manner for losses occurring during chromatography.
in the determination of total content of quercetinlike substances. The amounts of individual flavonoid compounds present in mixtures, however, generally cannot be determined by the latter two procedures. Porter, Dickel, and Couch (5)have used the absorption spectrum of the rutin-aluminum chloride complex to determine rutin in urine. The authors noted that the presence of other flavonoid pigments interfered with the determination of rutin because of their absorption maxima in the vicinity of the rutin-aluminum chloride complex. Wender and Gage (6) have recently adapted the method of paper partition chromatography (I), coupled with the use of chromogenic sprays, t o the qualitative separation and identification of mixtures of flavonoid pigments. This method has now
