Document M4O6G64z6g7kQZ006xZpvQQva

704 INDUSTRIAL AND ENGINEERING CHEMISTRY Vol. 17, No. 11 dmp!c No, Tbl . . ,, ,Mw Spectrograph An.lyrii, Mole % .- . -. . ' H*S0* Anely.t., ,, . Mole 9 TToottaall1 Total CiH CiS.. CHi ' T&oCdHii n*rtHu Olefin ' Olefin X .. 14.0 0.5 0.3 . 85.2 a IS.9 0.3 0.3 65.5 3 13.3 0.5 0.1 86.0 . 4 7.8 o.e 0.0 91.5 0.0 0.0 O.l 0.0 - 14.3 14.2 1148..24 14.3 18.8 7.9 ' 8.1 sorption into the mercuric sulfate reagent. Results by the two methods differed by an-average of 0.7%, whioh is the order.of uncertainty of the former method. Analyses of four of these samples, .which happened to be free of ethylene bat contained higher olefins, ore inoieated in Table I. - ACKNOWLEDGMENT . . The authors are indebted to' W. A. Stover of this laboratory for making these 29 analyses available to them. D, Effect o? UnoROGsw and Carbon Monoxidb. Samples of pure hydrogen and carbonmonoxide and duplicate samples of a mixture of 51% carbon monoxide and 49% ethylene were ana lyzed with the mercurio sulfate solution in a pipet packed with vertical tubes, leaving the sample in the pipet 30 seconds be tween passes: " hiiti O 1 . 2 3-. 4 S 6 7 10 . Pure Hi Ml 100.0 100.0 loa.o 100.0100.0 100.0 100.0 . 100.0 100.0 Pure OO . Ml. 100:0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 . 100.0 Mnltart (46% Cilli) ML ML 100.0 62.0 71.4 63.2 66.2 52.3 51.1 60.fi 50.fi . -, 100.0 84.8 73.8 CB.6 58.8 53.8 51.1 50,9 50.6 E. Commercial Samples. ' Twenty-nine samples of experi mental'gases which varied in total olefin content from 2 to 30% were analyzed both by the mass speotrograph (8) and by ab* .' . literature cited -. (1) Adama, R., Homan, F. 1., and Sperry, W. N., J. Am'. Chon. Sob., 44,1781 (1922). . (2) Cuneo, J. F,, and Switzer, R. L., Inh. Enq. Chbu., Anal. Ed., IS, 508 (1943). (3) Ebori, J. J., JUd,, 14, 853 (1942). (4) Gooderham, W. J., J. Sod. Chm. Ind., ST, 390T (1938). (5) Eoiteema, Q,, Z. pfuM, Chtm., 17,604 (1895). - (6) Engel, G., and Hiboii, J,, Cktmio a Industrie, Special No. 286 (Feb, 1928). ". (7) ' hfstuazak, M. P., lin>. Eno. Czni., Aval. Ed., 10,354 (1938). (8) Washburn, H. W., Wiley, H. F, and Rook, S. Ibid;, 15, 641 - (1943). - (9) Winbladh, R., Ing. Vetenakaps. Akad, Handle No. 138, (1936). 'PxMiNTHD before th DMilon ol-Analytical Chemistry at the Spring Meet ingcl (he Philadelphia Section, Auzaioew Oawicln Sotaarr, June 19,1M5, and th.Diviolon of Petroleum Chebdetry at the Meeting-tn-Print of the Amsbioam Chuwioal. Sowrrr, September, IMS. Colorimetric Determination of DDT Cdor Test for Related Compounds . . MILTON S. SCHECHTER, S. B. SOLOWAY, ROBERT A. HAYES, AND H. L. HAUER . Bureau of Entomology and Plant Quarantine, Agriculture! Research Admlnlttratlon, U. $, Department of Agriculture, Belttvillc, Md. - A colorimetric method hei been developed for the eifiraetion of mg., arid they lack specificity. Furthermore, there is no method tmell amount* of DDT down to about 10 microgram*. ' The method baaed on chlorine determinations by which the amounts of jr,p'- Involve* Initniive nitration end the production of colpt* by .the DDT and o,p'-DDT present in mixtures eon bB estimated. nitrated product* In benzene plu* meihanollc sodium, methylate.. This colorreaction can alto be uted as a tort for degradation products The'terms used in tins paper to designate DDT and related compounds are as follows: The generic term `(DDT1originally of DDT and iome compound* related to it. .. abbreviated from didhlarodiphenyltriohleroethane, refers to the technical product, which ordinarily contains 70 to 77% of p,p- DDT [l-trichloro-2,2^bis(p-chloropheayl)efchane] and 15 to 26% THE extraordinary'development of the insecticide commonly of o,p-PDT (l-trichloro-2-o-chlorophenyl-2-i>-cUorophenyl)- known os DDT (1, 7) has made /the need for a sensitive- ethane]. One of the minor constituents is l,l-diohloro-2,2-bie(p- method of detection and determination rather urgent A ehlorophenyl)ethane which has been designated as p.p'-DDD (22). Gunther (11) has pointed out his error concerning the method which could detect small amounts of DDT would find term "p.p'-DDD" made m a previous article (10). This com application in such fields of study as spray-residue determina pound has been named "l)l-dichloro-2,2-bie(j>-cilorophenyl)- tions, water analyses, and pharmacological investigations. .. ethane" in the .present paper in conformity with the latest Much of the analytical work on DDT has depended on chlorine determinations. Either the "labile'' chlorine split out on de- Chemical Abstracts nomenclature. The chemical composition of technical DDT is described by Gunther (10) and by Haller, Bart lett. Drake, Newman, and others (IS). Debydroehlorinated hydiochlorination by alcoholic alkali can be determined, as p,p'-DHT fl,l-dichloro-2,2-bis(p-ohlorophenyi)ethylene] is a recommended by Neal et al. (SI) and by Gunther '(6), or else the decomposition product' and bisyi-chlorophenyl)acetic add. (8, total chlorine can be determined by some method such ae the Igj^vhich has Deen called p,p'-DDA, la a metabolite of p,p'- Parr bomb, Carius, or Umhoefer (SB), or by a modification of the ' Winter method proposed, by Hall et at (IS). A search was made to find a suitable color reaction for DDT . The labile-chlorine method determines only.1 chlorine atom per wbioh could be made the basis of a colorimetric analytical method. molecule of DDT, whereas the total-ohlorine methods determine 5 Some of the exploratoiy work done fa this direction is outlined chlorine atoms per molecule. If DDT completely decomposes to below: - ' debydrochlorinated DDT, the former method would yield no Testa for the trichloromethyl group using pyridine and alkali, chlorine while the latter gro'up would determine 4 chlorine atoms resorcinol and alkali, or fi-naphthol'and alkali, as described by per molecule. If a total-chlorine, method is used as the sole Snell and Snell (S3), were all'negative! Boiling ethanolio silver . method of determination, no measure of decomposition .of the nitrate gave no precipitate of silver chloride. Nitration of DDT,- DDT can be obtained. Both labile and total, organic chlorine reductlon, and diozotlzation, followed by coupling with a suitable must be determined fa order to prove the presence of DDT or. compound, give a color (orange with y-naphthol). Although to detect its decomposition. - All these chlorine determinations this lino of attack could probably be developed into a method for pm into difficulty when .the amount of DDT Is less than about 1 the analysis ni tytvp -* not pursued further because it was PLAINTIFF'S EXHIBIT HARTOLDMON0030325 November, 1945 ANALYTICAL EDITION 705 believed to be subject to interference from many aromatic com pounds which could sIbo be nitrated, reduced, diazotized, and coupled. Intensive nitration followed by reaction of an acetone solution of the nitrated derivative with alkali gives a red color. This test would also be subjeot'to interference from many aromatic compounds (3). Intensely nitrated DDT gives a positive violet-red test when heated with the chemical-warfare - reagent DB:3 (23), which might be useful for field teste. Bes- oently two new colorimetrio methods for DDT have been de scribed (2, SB). _' . ' Hie method described in this paper depends on intensive nitra- - tion to polynitro derivatives and the production of intense colors upon addition of methanolic sodium methylate to a benzene solution of the nitration products (2). p,p'-DDT and p,p'DDD gjve blue colors, and o,p'-DDT gives a violet-red color. Degradation products of DDT, such as dehydrochloilhated p,p'- DDT and p,p'-DDA (3, 27), yield red colors. With the use of ` these reactions, si colorimetrio analytical method was developed. Although all the factors and possible variations at each stepof the analysis were not completely investigated, if is believed that the following description will satisfy the immediate' need for a sensi tive method Of analysis for DDT by chemists, entomologists, and pharmacologists' concerned with its wartime applications and public-health aspects. - ' ' APPARATUS Glass Beads, 2 or 3 mm. . Test Tubes, 22 X 175. mm., with rims, to be used for nitra tions. 'Sbpabatoky Funnels, 125-ml. capacity. The glass stoppers should be ground to fit very well. By attaching them- with Nichrome wire, they can be suspended loosely in the necks of the . funnels while solutions are being drained. Stopcooks should be greased occasionally with a good grade of. stopcock lubricant; vaseline is too thin. After each new greasing the excess mease should be removed by pouring ether or ohloroform into the funnel and rotating the stopcock as the solvent drains. After each, analysis the separatory funnels should be rinsed several times, with warm water before being used again. ' ' Glass Gooch-Cbuciblb Holders. Body about 25 mm. in. diameter and about 75 mm. long, stem about 30 mm. long. I. ' * . SOLVENTS AND REAGENTS Nitbatinq Acid. A mixture of c.P. fuming nitric acid (sp. gr, 1.48-1.60) and c.P, concentrated sulfuric acid (sp. gr. 1.84), 1 to 1' by volume. ' Sodium Hydroxide Solution, 2%. t Sodium Chloride Solution. Distilled water saturated with c.p. sodium chloride. Technical salt is unsatisfactory because of dirt and colored impurities extractable by ether. , Cotton. Extracted with acetone in a Soxhlet extractor, dried for several hours.at 105 to 110 C., and stored in a tightly stoppered bottle. . . Ether. TJ.S.P. grade distilled before use.. Ether that has- been standing long enough, to accumulate peroxides and alde hydes, or hasBeen recovered after ubb in this method is unsatis factory and should be purified before it is used again. - Benzene, o.p., dry. It is conveniently dned by distilling through a straight condenser until no more water distills-over with the benzene, and then replacing the condenser with a dry one and continuing the distillation. Benzene that has been used in tbiB method to dissolve the nitrated residues or to make dilutions thereof may be accumulated and recovered for reuse by distilla tion. . . Sodium Methylate Solution, 10.0 =*= 0.1% (concentrations are expressed as weight per unit volume throughout this paper) of sodium methylate in dry c.p. methanol (10.0 grams per -100 ml. of solution). An excellent method (18) of drying the methanol is to reflux with' magnesium turnings (5 to 10 grams per liter of methanol) and a small amount of iodine until the magnesium has completely dissolved and then to distill with the exclusion of moisture. The solution is prepared by diesolving the requisite amount of perfectly-clean Bodium or a good grade of powdered sodium methylate (available commercially) in the dried methanol with cooling, using a stirrer and a reflux condenser protected by a soda-lime tube. An aliquot of a dear portion of this solution should b.e diluted with water and titrated with standard hydro chloric acid, phenolphthalein being used as the indicator. The concentration of the solution should be' adjusted to 10.0 * 0.1% by tire addition of sodium or.sodium methylate or by dilution wiih dry methanol. .> . The sodium methylate solution that is added to the benzene to develop the color should be eolorless and optically clear. If the sediment does-not settle completely on standing, the solution should-be filtered or centrifuged. Occasionally.a turbidity or precipitate of crystalline material (probably sodium carbonate) will form when the centrifuged sodium methylate reagent , is added to the benzene solutions. This difficulty oan be obviated largely by cooling the standardized solution in a refrigerator for a day or two, -centrifuging while cold, and decanting into another container.- . .. Acetone, technical Redistilled before using. < ) PROCEDURE Prefaration of Sample fob" Analygis. Unless the total sample has very little DDT- (less than 100 micrograms),, it is advantageous to use a portion of the sample which contains a reasonably large amount of DDT (0.5 mg. to. several milligrams). It will then be4 possible to take an aliquot at the end of the prooedure for the -development of the color. Extract of strip the DDT from the sample with a suitable solvent and evaporate. Using acetone, transfer the residue or an aliquot thereof to a test tube for the nitration. -In some cases, the aliquot may be taken, direotiy from the extract before its evaporation. Care must be taken not to lose any of the sample meohanieally. during4 the evaporation of solvents prior to tbe. nitration. The best procedure for evaporating organic solvents is to add a glass bead, immerse the test tube about one third of its length in a steam bath, and shake gently until the glass bead bounces and ebullition starts. When the solvent has been com pletely boiled out, remove tbe last traces by inserting a glass tube attached to a source of vacuum one third of the way into the test tube for at least half a minute, while it is still being heated. Bn-' less'the solvent is completely removed,-it may reaot violently with the nitrating mixture in the next step of the procedure. If . benzene or an aromatic solvent has been uee'd, add 5 ml of ethanol .and evaporate to dryness in the same manner inorder to remove the aromatic solvent by azeotropic distillation. Nitration of Sample. Cool the test tube in a beaker of cold - water and with a pipet add 2.0 or 5.0 ml-, of the nitrating acid. Immerse the test tube one third to oue half its length in a steam bath and heat for 1 hour. Since nitrations of even small quanti ties of materials may sometimes be violent, safety .precautions should be observed. If there is much extraneous material, it is advisable-to place.the test tube in ice-cold water, add cooled nitrating acid, ana warm4 the tube cautiously to prevent a sudden or violent nitration. When the initial reaction has subsided, the tube may be heated at 100 with safety. After the 1-hour nitra-- tion, cool the-test tube in a beaker of.cold water, add 25 ml. of.ice- cold distilled water, and mix by gentle swirling. This stops the nitration, and the test tube may be left overnight if desired. Extraction of Nitrated Product. Rinse the contents of the test tube quantitatively through a Bmail funnel into a 125-ml, separatory funnel with about 25 mh of water from a wash bottle and 60. ml. of ether. A email,-irregularly Bhaped piece of glass placed .in the funnel used for the transfer will prevent the glass. bead from falling'into the separatory funneL Shake vigorously for at least 1 minute. After the layers, have separated clearly,- draw off and discard the lower layer. Wash the ether with 10-ml. portions of 2% aqueous sodium hydroxide until the washings are alkaline; one washing may be sufficient. Then wash tbe ether with two 10-ml. portions of salt solution. The final Balt wash should be drawn off as completely as possible. Pack a 0.75-inch plug of cotton tightly in a glass Gooch-crucable holder, moisten it with ether, and allow the ether solution from the separatory fun nel to filter slowly into a 125-mL Erlenmeyer flask. Rinse the i separatory funnel with 60 ml. of ether in four or five portions, passing tins ether through the cotton in the Gooch funnel. . If salt crystallizes in the neck of the separatory funnel, press the stopper of the funnel in place firmly with a rotating- motion to prevent, leakage' of ether. Add a glass bead to the Erlenmeyer flask, warm the flask on a steam'bath with a gentle swirling motion until the bead starts bouncing, and recover or evaporate the ether com pletely. While the flask is still being heated, insert a -glaas tube connected to & source of vacuum two thirds of the way into the flask for atleast half a minute;, thehremove'the flask and stopper it. The analysis may be interrupted at this point if desired. The whole extraction procedure must be done carefully to avoid any loss, such as ether sprayed from the separatory funnel when the stopcock is opened to release pressure or when tire glass stopper is removed. This type of loss can he minimized by allow- . ing time for tbe ether to chain away from the stopcock or the stopper before performing these operations. 4 - HARTOLDMON0030326 706- INDUSTRIAL AMD ENGINEERING CHEMISTB7 ' Vol. 17, No. II Development op Color. At this stage there is a choke of procedures, depending on the amount of DDT expected, the amount of solution necessary for use in making the photometrio measurements, and whether it is desired to have some solution left to repeat the photometrio measurements. ' . Procedure /. Add accurately measured amount of benzene-- for example, 5.00 ml.--to the residue in the Erletuneyer Bask . and swirl gently until it is dissolved. Use a volume of benzene at least equal to one third the volume necessary for use iia the ab sorption cell or tube of the photometer. With a pipet add 2 volumes.(10.00 mL for 5.00 mL of the benzene solution) of the sodium methylate reagent to 1 volume of. benzene solution. Swirl gently until-the solution is homogeneous, pour into the absorption cell or tube of the photometer, and prepare to make the most important measurements 15 minutes after the sodium methylate reagent has been mixed with the benzene. This pro cedure should do used only when it is known that the amount of , DDT is very low and in. the range where the color developed wBJ ' be suitable for direct measurement in the photometer. If there is a possibility that the color developed will be too dark for direct measurement, it is preferable to use procedure 2 rather than add more benzene and sodium methylate to the colored solution to dilute it.. ' ' Procedure 8. Add a measured amount of benzene--for ex- ample,'25.00 ml--to the Erlenmeyer flask and swirl gently until - the residue is dissolved. To an aliquot--for example, 5.00 ml-- addtwicerts volume ofsodium methylatereagent, mixthoroughly ' by gentle swirling, and poor into the absorption cell or tube. In some oases it is possible to mbs the solutions directly in the ab sorption cell or tube. If the color ia too deep, a photometric measurement may be made to' obtain a rough estimate. Dilute - part' or all of the remaining benzene solution to a more suitable volume before removing a new aliquot for development of the color.. If tbe color is too light for good- photometric measure ment, rinse the pipet used for the first transfer with benzene into the Erlemneyer flask, evaporate all the solvent on the'steam bath.' swirling the flask gently to start the bead bouncing, and, when all tbe benzene is evaporated, remove the last tracesby inserting a ' glass tube attached to a source of vacuum. This residue in the Erlenmeyer flask should now be treated as in procedure 1. PBOXOunraio Measubements. Bpectrophotometrio orphoto* metric measurements should be made at themost important wave lengths or with the most important filters ss close as possible to 15 minutes after the .sodium methylate solution has been mixed with the benzene. Measurements at other wave lengths or with other filters can be made just before or after the moBt significant readings have been taken. -Absorption cells or tubesshould be stoppered tightly. ' Absorp tion cells usually have glass covers or Btoppers, but if test tubes are used, as in-many routine photometrio measurements,-rubber stoppers washed free of sulfur are preferable to cork stoppers, eontaot with which will turn the solution yellow.- Since the solutions on which optical measurements are made are strongly alkaline, absorption cells constructed with alkali-resistant cement Bhould be used. The solutions should be left in the cells no longer than is necessary to make photometrio measurements, after which the cells should be cleaned immediately. Although it might be ' expected that the alkaline solutions would attack and etch glass Cells, no such difficulty has been experienced during several months of use. - -- under the particular conditions ofpreparation of tbe sample and nitration employed by the analyst.. It 1b advisable to get rid of as much extraneous material as possible before analyzing sam ples. In this connection solvents' which extract less extraneous material than others may be used. ' ' NiraxTioK or SaUtix. A number of nitrating mixtures other than the recommended 1 to 1 fuming nitric add-concentrated sulfurio aoid were, tried, suoh as 1 to 1 red fuming nitrio add concentrated sulfuric acid, 1 to I fuming nitric asid-25% fuming sulfuric add, 1 to 1 pad fuming nitrio add-25% fuming sulfurio acid, and 1 to 1 concentrated nitrio add-concentrated sulfuric acid.. The last mixture gave odors that were too light, and none of them seemed to have any particular advantage over the recom mended mixture. . . . ,' Although the nitration seems to be completed in less than 1 hour, it was considered that in many applications of the method a 1-hour period of heating with, the nitrating mixture would give more oomplete destruction of extraneous material. . The nitrating mixture destroys to a large extent many plant extracts, oils, etc., by oxidation and conversion to alkali-soluble products, which are removed when theether solution ia washed with aqueous sodium hydroxide. However, some interfering substances are not de stroyed completely, ' The amount ofnitrating acid used is not critical Where small- samples are used and the amount of extraneous material is not large, 2 ml. can be employed; where larger amounts of extraneous material are regularly encountered, 5 ml are preferable. Since the quantity of acid used will make a slight difference in the calibration curves on-known amounts, these ourves Bhould be pre pared on the basis of whatever amount of add is to be used for analysis ofsamples. - - The nitration of an organic compound rarely' gives a 100% yield of a single produot. -Usually a number of isomeric nitrated products are formed, and products of lower and higher nitration are sometimes present. Ordinarily a certain amount of material In any application of the method it is important to run a blank analysis on a sample of the Bame type of material being analyzed ' which has not been treated with-DDT. The results, in terms of DDT or extinction values (never in terms of per cent transmiB* aion), should be applied as corrections to the values obtained at- eaoh wave length, or filter used in the analysis of the DDT- treated samples. If appropriate blanks are not run, the results of the analysis may ho high. Blank analyses' should be made by diluting the blank runs in the seme manner as the DDT-treated . - samples, or else the corrections should be calculated to the same weight ofuntreated material us used in tbe analysis of the treated - material. ' DISCUSSION OF THE METHOD . 1 Pbepjlbation or Sample. DDT decomposes with evolution of hydrogen chloride when heated at a high temperature, and it may decompose at 100 or lower in the presence of traces.of cer tain catalysts, such as ferric chloride or iron (0), To minimize the possibility of decomposition, solvents may be removed from samples at room temperature by means of a draft of air. Testa should be made to demonstrate that there is no decomposition . Figure 1. Filter Photometer Color Curves A. Pare t*traitltrM>JJ -DOT , B. 0.1 ms. of pp -ODT carried Uroagh readied C. 0.1 ntf. of tadnleal DDT-carrltd throaih authod 0. Pur* tctrin[tro-o,p'-DDT' E. 0.1 ras. at ojr'.ODT canted thmajh malted HARTOLDMON0030327 November, 1945 ANALYTICAL EDITION ' 707 near 11% of sodium methylate, as-measured with the No. 58 filter and hear 13% as measured with the No. 51 filter. The absorption curve of a benzene solution of tetranitro-o,p'-DDT plus methanolic sodium methylate (5.0 grams of sodium' per 100 ml. of solution) exhibits two absorption peaks (22), one at 590 and the other at 511 millimicrons. It is interesting to note from Fig ure 2 that each peak is affected-in a different manner by different concentrations of sodium methylate. Unless there is a. special interest in determining o,p '-DDT, or the relative amounts of p,p- and o,p'-DDT in mixtures, there is no advantage in using higher than the recommended 10.0% sodium methylate reagent, since the sensitivity with regard to p.jri-DDT is thereby.de creased. . ' The color may be due to the following type of reaction product, one of tbe structures of the resonance hybrid of the complex from tetranitro-p,p'-DDT being shown: Figure S. - Effect of Concentration of Sodium ' .Methylate on Intensity.of Color - is oxidized to degradation products, or even completely oxidized When p.p'-DDT and o,p`-DDT are carried throughthe method, the main products are the tetranitro compounds described by Schechter and Haller (22), as borne out by a comparison of photo metric measurements given in Figure 1, . Extraction op Nitbated Product, Benzene would be ad vantageous for the extraction, in that the color couldbe developed in the extract directly. However, for routine use ether is prefer-. able because it gives more rapidly a clear separation of the - layers. If the shaking.is vigorous and the layers are permitted to separate clearly, a single extraction with ether is as good as a double extraction, within the precision of the. method. To pre vent possible decomposition of the chromogenio compounds, con tact with the aqueous alkali should be no longer than necessary for thorough extraction and separation of the layers, Shaking the ether with saturated saltLSolution washes out tt alkali and also partially dries the ether. Filtration through oven-dried cot ton prevents any salt droplets from coming through and further dries the ether. . ' Development op Color. Benzene is a better solvent-than methanol for the nitrated residue and is miscible with 2 volumes of the sodium methylate-methanol reagent. . The ourves shown in-Figure 2 illustrate the effect of the concentration of sodium methylate on tho intensity of the.color ae measured with an Aminco type F filter photometer (see Results for a description of the filters). p,p'-DDT, o,p'-DDT, and' technical DDT were carried through the procedure, and the color was developed on aliquots of the final benzene solutions (0.10 mg. per 5.00 ml. of benzene) with the uso of different concentrations of methanolic sodium methylate. The results indioato that, for p,p'~DDT and technical DDT,, the most intense color was developed close to 10.0% of sodium methylate in methanol as measured Mth the No. 58 filter, and this concentration was adopted in the authors' work.- The maximum intensity for the o,p'-DDT was developed The reaction of polynitro compounds with sodium alcoholates has been investigated by Jackson and Earle (IS) and by Meisen- heimer (19). '. Photometric Measurements. Filter, photometers are sub ject to a number of difficulties and sometimes give deviations from Beer's law because of broadness of the bands passed by their fil ters, stray light effects, etc. Their limitations must therefore be kept in mind. The difficulties and sources of error in filter photometry are adequately discussed by-States and Anderson (24) and by Hamilton (14). Hogness et al. (IS) and Brode (4) give good discussions of absorption spectrophotometry. In gen eral, better results can be obtained with a spectrophotometer than with a filter photometer. ' . - ' The absorption peak when p,p'-DDT is carried through the method using 10% sodium methylate reagent is at 598 milli microns, and the two peaks given by o,p-DDT are at 590 and 506 millimicrons (unpublished data). To determine p,p'-DDT or technical, DDT, measurements should be made at the .wave length orfilter that gives the maximum absorption for p,p'-DDT (ca. 596 millimicrons) obtained on the instrument used by the analyst. Calibration curves should be made with knownamounts of the type of.material to be,determined, whether it is p,p'-DDT or some batch of technicalDDT. . .- .Sohechter and Haller (22) indicated that it would be feasible to determine the relative amounts of p,p'- and o,p'-DDT in mix tures of the two. Although technical DDT has as its major con stituents p,p'-DDT (about 70 to 77%) and o.p'-DDT (about 15 to 25%), it does contain small amounts of other compounds and unidentified material [Haller etall (IS)]. The composition may vary with the method of manufacture, and even from one hatch . to the.next. While calculations of the amounts of p,p - and o,p'. DDT, assuming that these are the only two compounds present, give results of the right order of magnitude, the effect of the . minor constituents, such as p,p'-DDD, should not be ignored. These calculations can be made if analytical calibration curves are prepared for known amounts of eaeh of the two isomers at two suitable wave lengths or filters (in tho range of 595 to 800 millimicrons and in the range.of 500 to 510 millimicrons). Such computations from photometric data on mixtures are adequately disoussed by Miller (20), Knudsen et al. (17), and others. It should be emphasized that these calculations will hold only over the regions of the calibration ourves which follow Beer's law, so that a spectrophotometer or a photometer having filters of narrow wave length range should be UBed. HARTOLDMON0030328 708 INDUSTRIAL AND ENGINEERING CHEMISTRY V61. 17, No.'ll , RESULTS . . > An Aminoo type F photometer was used with . filters 36, 68, 53, 61, 46, and 42, having wave lengths of rnwrimnm transmission at 650, 680, 630, 614, 460, and 424 millimicrons, respectively, and- test tubes 2 cm, in diameter. All photo ' metric readings were made on 6410 ml.- of the \ benzene solution plus 10.G0 ml. ofsodiummethylate reagent and were converted to extinction values, 2. O 6 sz p V ' t. MICROGRAMS OF P, P - DDT Figure 3. Analytical Calibration Curvet lor p.p'-DDT Aa,. VM' WHh Mo. 5S RlUr 6s int ' For more accurate determination of the percentage of p,p'- DDT in technical DDT, the crystallization procedure of Cristol etaL (6) probably is more suitable. . ' In applications of the oolorimetric method it is advisable to pre pare calibration curves and also to make readings at a number of wave lengths or filters. There Is usually less'interference from extraneous materials at the higher wave lengths, sincemanyinter ferences, such as those from plant extracts, exhibit an absorption, ourve which shows gradually increasing absorption with de creasing wave length.- ' While a calibration ourve at a wave length of 340 millimicrons, or with a filter at 660 millimicrons, will have a lower sensitivity for determining p,p'-DDT, the results read from such a curve will have least interference from extraneous materials and practically none from possible decomposition or degradation products of p.p'-DDT, Fadwg or THB Colons. The use of impure solvents and re agents can give rise to serious difficulties, Technical benzene sometimes contains impurities which contaminate the distillate with hydrogen sulfide, and this causes rapid fading of- the de veloped color. Contamination of the benzene or of the sodium methylate reagent with sulfur, such eb that from rubber stoppers, will increase the rate of. fading. . Sulfur can be removed from . stoppers by boiling them in strong sodium hydroxide solution, washing, and drying, > The presence of water in the benzene used to dissolve the nitrated residue, or of water or sodium hydroxide in the Sodium methylate solution, will also cause rapid fading. With good re agents and solvents the authors have found the faffing to be 2 to. 3%, in terms of either p,p'-DDT or technical DDT, 1 hour after . the first reading. Six per cent during the first hour should be re garded ae the maximum permissible amount of fading. While' - methanolio sodium bydroxi.de will produce the blue color, the rate of fading la so fast that satisfactory photometric) measurements cannot he made. The oolors given by p,pVDDT, e,p'-DDT, and technical DDT develop fully in 2 to 3 minutes and then fade very slowly. The red colors given by some of the possible decomposition or degrada tion products of DDT require about IQ minutes to develop fully and are relatively stable. . otlos (itaSa)1 l>*>and 6 show photometric measurements made with the various filters in the photometer; they do not represent Bpeotropbotometrio curves. Figure 1 1 shows the readings obtained when 0.10 mg. of p,p'-DDT, o,p'-DDT, and technical DDT were carried through the method and the colors de . veloped according tp procedure 1. For compari- . son, pure, teIranitro-p,p'-DDT' and tetrarritrot o,p-DDT were dissolved In benzene, 5.00-rnL - aliquotscontaining0.15 mg. (equivalentto0.10mg. ' of p,p'-DDT and o,p'-DDT, respectively) were mixed with 10.00 ml. of sodium, methylate re . agent, and the reeultB plotted (dottedlines)'in the. ' . same figure. .- ' Figure3 illustratesthetype of analytical calibra tion ourve obtained! with; email amounts of p,p'- DDT (20, 60, and 100 micrograms) when 2:0 mL of nitrating acid were used and the folor was de veloped according to procedure l At each concentration the average deviation-with the No. 68 filter in a number of runs by three workers was about 2 mlorograms and the maximum devia tion about 4 micrograms. A. Beckman DU spectrophotometer gave straight-line calibration curves for the same amounts. The deviation from Beer's law when the No. 65 filter was used is'due to the absorption characteristics of this filter. This "stray light" 8 ' S3 51 46 FILTER NUMBERS Figure 4, Filter Photometer Color Curves A. 0.1 of, el d.SyAoehlerlnalid r>,p'-DDT , o. 0,1 mj. el 4,4rydVcmoiDbft*ej>h*ior*, C. 0.1 m$ oF p*-DDA {>. 0,1 mg. of 4c4 dlehlorebenzohydroi . 0,6 mp. of o-dkMereb*nx*r>* F> 0,068 ms, pr pu/e 4,4 dIchlofO-8#J 5,5 4t/4nttroo*nzophnene & <U mte.ot)fl4kht9T&>34"bltCp<h\Qi09kiMl)a&*nt, ot p<p *DUD He 0.1 ms, or otrfp<hiorepbnvl)mtlMB Ewh Pempewid #* mt(*o ihrcufh nethpd txcp| F HARTOLDMON0030329 t:* November, 1945 ANALYIICAl EDITION . r ' 709 effect 5s described by Stfttes and Anderson (4). When 10 mg. of DDT were used, and the color was developed after proper dilution into a readable photometric range according to procedure 2, ft calculated result of 9.9 mg. was obtained. Figure 4.shows the results obtained by applying the color test' to a number of compounds related to DDT, using procedure 2. p,p'-DDD gives a blue color practically identical to that of p,p'~ DDT but 6lightly less intense. -. Some of the possible breakdown produots of p,p'-DDT--such as dehydrochlorinated p,p'-DDT, 4,4'-dicbJorobenzophenone (JO), 4,4'-diehlorobenaohydrol, bis(p-chlorophenyl)methane (7), andp,p-DDA (B, 7]--give red colors with negligible absorption when the No. 65 filter 1b usBd. Consequently there could be little or no interference from these breakdown products on the analysis for p,pVDDT as read from its calibration curve at this filter.' In ^ an experiment with a mixture containing 0.050 mg. of p,j>'-DDT and 0.025 mg. of dehydroohlorinatod p,p'-DDT, the result when read from the analytical calibration curve at the No. 65 filter indicated the presence of 0.050 mg. of p,p'-DDT with no inter ference from the dehydrochlorinated p,p'-DDT. However, in the case of technical DDT, If there is considerable decomposition, it would be difficult to calculate or interpret the results because of the complexity of the system, at least four components (p,p'- and o.p'-DDT and their dehydrochlorinated derivatives) being present. .. If the results for DDT read from calibration curves at several filters agree after being corrected for the blank'analysis at each filter, it is evident that the DDT has not decomposed to any appreciable extent. The red colors are due to the formation of a considerable amount of 4,4'-dichloro-3,3'r5J5'-tetramtrobenzO' phenone (probably accompanied by isomers and other nitro derivatives) during the nitration of these degradation products of - DDT. The extinction values on the color given by synthetic 4,4'-dicbloro-3,3',5,5'-tetranitroben2ophenone inbenzene solution plus sodium methylate are also given (dotted line) in Figure 4 for comparison. ' '' 4-Chlaro-3,5-dJnitrobenzoic acid has been detected as another product of the nitration of dehydrochlorinated p,p'-DDT, 4,4'- dichlorobensophenone, and p,p'-DDA. In the analysis of these- compounds the 2% alkali washes of the ether solutions were acidified and extracted with ether, the ether was washed with salt solution and evaporated to dryness, and the residues were dis solved in benzene.- On addition of sodium methylate reagent a vi olet-red color waB developed in each case. The some color was developed when p-chlorobenzoie arid was carried through the method with the omission of the alkali wash. A comparison of photometric readings (Figure 6) obtained on these solutions with readings (dotted line) made on a-sample of synthetio.4-ahloro- 3,5-dinitrohenzoic. arid in benzene plus sodium methylate reagent indicated that each of the solutions contained this compound. The fact that p,p'-DDA yields 4,4'-dlchloro-3,3'l5,5'-tetra- - nltrobenzophenone and 4-chloro-3,5-dimtrobenzoic acid on in tensive nitration is rather remarkable, since it involves decar boxylation of the DDA in addition to nitration and oxidation reactions, The dicUorotetr&nitrobenzophenone formed is neutral and remains in the ether, in the analytical method, while the chlorodinitrobenzoic acid goes into the alkali wash. y.p'-DDA has been shown in pharmacological studies (7) to be a metabolite of p,p'-DDT, and a method for its detection and estimation has considerable importance in these investigations. Use can' be made of its acidic properties to separate DDA from DDTand any neutral degradation produots priorto thB application of this oolori- metrio test. It doeB not seem, to be possible at present.to differ- ontlate some of the other breakdown products of DDT, such as dehydrochlorinated p,p'-DDT and 4(4'-dichlorobenzophenone, on the basis of this color test alone. In auch cases supple mental chlorine determinations on the samples would be ofvalue. Bls(p-chlorophenyl)sulfone, a minor constituent of technical DDT, aDd p-dJohlorobenssene were found to give no color when Figsie 5. Detection ol 4-Chloro-3,3-dlnibobtnzols Acid A. Ran ilkall wtit Is andyik el 0J mf. at pji'-DDA B.. 0.1 M. ef>M,4kMe);3,5'lbeinleseH . C. 0,05m. ................................ .` SoSlLii carried through the analytical procedure. o-Dichlorobenzene gave an orange color (Figure 4). Care should be taken in inter preting results of this method whon-aramatic halogen compounds that might interfere are known to be present; however, there are not likely to be any such materials in spray residues. This method is being applied to the analysis of spray residues, - water samples, etc. The chemistry involved in the nitration of DDT audita breakdown products and the reactions of these and related nitro derivatives together, with pertinent speetrophoto- metric data are also being investigated. . .' LITERATURE CITED (I) . Annand, J. Scon, Entamd., 37, 125 (1944). . (2) Bailee and Payne, Ind. Eng. Ceem., Ansi,. Ed., 17,438 (1245). (3) Boat and Nicholson, Hid., 7, 290 (1B35). . (4) Brode, "Chemioal Spectroscopy", 2nd ed^ New York, John. Wiley & Sons, 1943. ' (5) Cristol, Hayes, and Haller, Iot. Eng. Cana., Anax. Ed., 17. 470 <1946). . (6) Fleck and Haller, J. Am. Chon. See., 66,2095 (1944), (7) Froelioher, Soap A Sanit, Chm,, SO (7), 115 (1644), : (8) Grummitt, Buck, and Stearns, J. Am. diem. Soe., ST, 1ES (1945). . (9) Gunther, Inn. Eng. CjSBU., Anax. Ed., 17,142 (1945).. (10) Gunther. S. Chm. Education, 22,238 (1245).' ` (II) IbU 22,372 (1945). . (12) Hell, Scheohter, and Fleck, U. S. Bur. Entemol. Plant Quaran tine ET-211 (1244). (18) Holler, Bartlett, Drake. Newman, and others, J. Am. Chm. Soe.,6T,mi(mS). ' (14) Hamilton, Imd. Ebo. Chxji., Anax. Ei>., 16,133 (1944). (15) Hogneaa, Zaoheile, and Sidwell, J. Phi/t. Chan., 41, 379 (1937). (16) Jackson and Earle. Am. Chm. J., 29, .89 (1903). . (17) Knudaen, Meloche, and Juday, Ind. Eno. Ohmi.. Anal, Ed.. 12, 716 (1940). ' * (18) Lund and Bienrufn, Bet.. 64,210 (1631). . (IB)' Maiseaheimer, Ann., 323, 205 (1902). - (20) Miller, "Quantitative Biological Spectroscopy", Minneapolis, Burgess Publishing Co., 1039. (21) Neal, von Oettlngen, and others, U. S, Pub. Health Service, - Pub. Health Bepte., Supplement 177,2 (1844). (22) Schechter and Haller, J. Am. Chm. floe., 68, 2129 (1944). (23) Snell and Snell, "Colorimetric Methods of Analysis", Vol. 2, New Vork, D. Van Noetrond Co., 1937. (24) States end Anderson, J. Optical Soe. Am., 32, 3SS (1042).. (26) Stiff end CeetUlo, Science. lot, 440 (1945). (26) Umboefer, Inn. Eno, Chum., Anax. Ed., IS, 883 (1843). (27) White and Sweeney, U. S. Pub. Health Service, Pub. Health Repl*-, BO, 66 <1945). (28) Zals, J. Chem. Education, .21,489 (1944), ' Past of this work wsa done under t transfer of funds, recommended by the Committee an Medical Research, from the Odes at SeleatlBe Reecoreh and Development to the Bureau of Entomoloay and Plant Quarantine, HARTOLDMON0030330