Decomposition of DDT from Thermal Aerosol Dispersers JEROME GOLDENSON, H. E. SORTON, C. F. NOVAK,
-AND
SAJlUEL SASS
Technical C o m m a n d , A r m y Chemical Center, 2Wd.
T h e object of this work was to determine the extent of DDT decomposition on dispersion in oil solution by thermal aerosol insecticide generators. Results showed that the decomposition, in general, w-aslow and increased gradually with temperature to a maximum of about 17Yo at 700' F. At the normal temperature of operation of the generator (600" F.) it varied from 3 to 89'~. Results also showed that decomposition was mainly due to the tem-
perature of dispersal, the effects of air being negligible. At the low-er temperatures apparently only one chlorine atom was split off each decomposed DDT molecule, and at the higher temperatures from two to three chlorine atoms split off each decomposed DDT molecule. The reliabilit, of the chemical analyses for DDT on the oil portions of the samples collected at various teniperaturrs was checked by entomological anal?sis.
THE
The behavior of DDT on heating was investigated by Bcholefield and others ( 1 5 ) ,who determined the rate of decomposition in an atniosphere of nitrogen by measurement of the volume of hydrogen chloride evolved a t temperatures up to 260 O C. (500" F.)" They reported that the evolution of hydrogen chloride is only slight a t the nieltiiig point of DDT but is rapid in the neighborhood of 200" C. This previous work could be used only qualitative!y to predict the behavior of DDT in a niised oil solution of xyleiie, fuel oil Xo. 1, and fog oil No. 2, in the presence of small amourits of iroir from containers and generator parts wlien flsah-heated for intervals of less than a second x i t h steam and cooled in air. The small amounts of iron and the elevated temperaturw would be expected to cause marked decomposition, while the oils during ti short interval of flash heating wouid be expected to inhibit decompozition. To evaluate the deconiposition quautitatively, it mas considered necessary to make a c1ctcrni;nntion as close to actual conditions as practicable and to collect samples for analysis of DDT contmt truly repre.)- the hol aerosol as it is released frcini the Iiozzle (if tlie chpclrser into thi, atmosphere. As before. the r i i i i p u t of the diepmer was collectrd by condensation, and the oil a r i d water port ions were analyzed. Table I V gives analytical rrsult.: on t h e oil portions of the smiples; they indicate that t h e rele I J ~ ' the hot aerosol into the :itniosphere under actual operatiiig conditions does I toward decomposition of DDT. hut appears t o 11:ir h=n~ficialeffect at the higher trniperatures. This titwt,fic.i:i I vffect may posihly he due :o more rapid cooling action by i h r :iii,, In ronnection with tlitc run in 1)-liicli air was introducid, :I c . l i c ~ kwas made oi' t h r reriivi'ic by condriisation compared I O the input. Careful nitwuwnient shoivcd that 47 gallons of' oil CI i1u1ion arid water were tli;.perwcl, and slightly over 46 palloil.: (it' coridcnsatr !\-ere collcctcd Tc-hich corresponds to Cj&% IYIYJWY\-. The ratio of 35 gallon. o l oil solution t o 12 g a l l o ~ ~ofsW:LIer : I l l ( \ the results obtained o n lhr ;sine a a n i p h by the :tnnlysir of t h i b w t w portion for hyilroperi vhloridr iiitlicati~d.3s i i i thi, wronil
MANOMETER
-
-ORIFICE -AIR COMPRESSOR -MIXING VALVE
I
--
---
-?+
'
111
2175
INDUSTRIAL A N D ENGINEERING CHEMISTRY
October 1949
-COLD WATER
3
OVERFLOW
-r-
\PUMP , LWATER TANK '\ '-DOT SOLUTION TANK \ THERMAL AEROSOL DISPERSER-
CONDENSER
\
Figure 2. Schematic Illustration of Equipment for Determining I)I)T Decomposition When Dispersed by Thermal .ierosol Generators
test, that a t the lov-er temperatuws appt~reiitly(iiiiy oile t < J iivu chlorine atoms were split off each deronipowl DDT niolccule, and at the higher temperatures from trvo to three chlorine atoms rTere split o f f each tlrcompowl T i m i n i i l r c i i l e . T;rhle V ~Iiorvsthr.ze results. -1 check analj-sis by :til c.iitomolugie:il nielliud wit h rci:tcllr.,, carriiAd out on the oil portion of the s:impIe dippei,wd ai. 700' F. w a s introduced, indicattd t h a t th(i reliability of 1he c1ieniic:tl analytical method was not affwteti t)y the introtlurtion of air. This cherk annlysis led to the coiirlusion thxt thtx entornological effectiveness of the sainplr is :ipprosim:5tely t11:ir of a .5yo solution of DDT, and t h e syniptoiw ~ i ihc ' ro:achr,q riscd in the analyses ryere typical of roaches pri-oiicd with I ) I )T emuliion.
Oil Uikpersion Temp., O F. 6.32 6,29 6.28 6.27 6.2.3 ti. 23 0.19 6.08 5.79 .i. 67 5.15
...
(J , 3
0.6 0.8 I.3 1.1 2.1 3.8 8.4 10 3
18.;
fi, 35 (i.31 ti.18 6.28 (i.22 (i . 23 ti.17 p 07 ,J.
!ii
3 , 73 .,i2 i
0.05 0 10
37; 400 425
450 $75
.100 550 600 2.50
,00
I:1 I: I 1:l 1:J .I . 1 -. 2:1 .,'.. . 1 I! t o 3 : l
-'to8:1
-.I .. L
0 15 11. 2.5 11, 3!1
0,70 ( 1 , 9.5 2.05 8.15
2176
INDUSTRIAL AND ENGINEERING CHEMISTRY EKTOMOLOGICAL B I E l H O D
milliliter of the oil sample was emulsified with 9 ml. of 5% gun1 acacia solution, and the emulsion was injected into ten roaches at a dosage of 5.8 cu. mm. per gram weight,. T h e resulting mortality in 40 hours was 80%. From standard curves previously determined, this mortality corresponds to a dosage of 30 nig. of D D T per kg. of roach weight, or 30 micrograms of D D T per gram of roach weight. This indicates that 5.8 cu. mm. of oil emulsion contains 30 micrograms of D D T or 5.17 micrograms per cu. mni. As the dilution in preparing the emulsion is 1 to 10, there xre then 5.17 X 10 (dilution factor) or 51.7 microgramsof D D T per cu. mm. of the oil sample, or 5.17% (a specific gravity of 1 was also wc:d in preparing the standard curves). The chemical analytical w l u c for this sample was 5.27y0 D D T (Table IV). 01iv
DISCUSSION O F RESULTS
.\!though these tests were conducted at temperatures up to 700' F.?the normal operating tcmperature of this generator is theriiiostatically controlled in the range around 600' F. The results of the three tests indicated that the percentage of D D T decomposed at t h r w tcmptmtures varied from about 3 to S%, whicli is considered not t u be excessive for this method of dispersal. It is quite low in comparison to decompositions encountered viith dispersal devices such as candles, in which the tlrcomposition 11:~skieen found t o be 25% and higher. Tahlcs I11 and 11- intiicatc. that) even at the highe5t temperaturesohpervc'd,rioriec)f thtLDDT was coniplctel!-pyrolyzrd; at the low-cr teniperaturrn apparently only i)ii(' chlorine atom n-:is ~plir off each decomposed D D T molrcule, arid a t the higher tcmpersturrs from tv-u t o t l i i w chlorine atonib n-rre split off each drc , ~ i n l ~ i ~DDT s ~ d riiolecule. klowevei., f!oni the inwc.tic.itia1 st mdpc,int, t lie eliminntioii of oiily one chlorine atom :is hy(1rochloric acid results in a conipountl nrhich is ineffcctivct :IF an insecticide. Some question might be raised as to the effects on vegetation of the hydrogen chloride released by the small amount of D D T decomposed when dispersed by the thermal aerosol method. -1cdculation, such as the following, shows that the amounts of hydrogen chloride rplea,vd arr so .ma!! ns to he considered negli-
Vol. 41, No. 10
gible for practical pur'poses: 46.7 grams of hydrogen chloride are released hy every pound of decomposed DDT. rissuming decomposition of S%, which is the maximum found at 600" F., there would be 3.7 grams or 0.008 pound of hydrogen chloride released for every pound of D D T dispersed a t 600 o F. If the do+ age is 0.25 pound of D D T per acre, the hydrogen chloride released v-odd be 0.9 gram or 0.002 pound prr acre, xliirli i < thought l o he an insignificant amount. ACKIVOWLEDGRI ENT
The check entomo1ogic;tl analyses were performed by Leigh E. Chadwick of the Entomology Section, Medical Division. Arm? Chemical Crnter. LITERATURE CITED
C'hem. J . , 3, 56 (.July 194s).
nd Wilson, I. R . , J . Econ. Entomol., 40,3U!j iI ! I $7 ' , (4) (hlliiis, D. L.. and Glasgow. R . D., I b i d . , 39,241 (1946). (5) Fleck, E. E,, and Haller, H. L., IND.EXG.CHEN.,37, 403 (1H45). (6) Fleck, E. E., and Haller, H . L., J . .4m. Chon. S O C . ,66, 2095 f 1944). ii)Gcer. H., and Scol-ille, H., J r . . S a t l . Research Council (\-. ,,sect Comm. R p p t . 132 (1945). gow, Ti. D., and Collins, D. L., J . Econ. Entomol., 39, 227
(1946). Gunther, F. -4.. ISD. ENC.CEiEar., ~ A L ED., . 17, 149 (1945:. Hiiffman, C . If-., Saeser, C. R., and Hartnett, J. G., TUMR 1241, U. P. Dept. Commerce, Office of Tech. Services. K e p t . PB 19862 (19461. i l l ) L a t t a . 11.. . I . Ecopi. Eiifvrnol., 38,66s (1945). 112) LeClnir, J . E., ISD. E s c . CHEM.,hs.~.. ED.,18,i 6 3 (1946). (13) Xorton, H . E . , TDLfR 1304, U. P. Ilept. Commerce, Office o f Tech. Sen-ices, R e p t . PB 53246 (1947). i 14) Scheciiter, M .S., arid Hnller, FT. L., J . -4m.Chem. SOC.,66,2129 (1944). (15) Bcholefield, P. G., Bowden, S. T.. aiid Jones, W.J., J . SOC.Cheni. I n d . , 65, 384 (1946). (16) War Department, Washington, D. V., 'I31 3-381, Generator. Smoke, hlechanical, XI2 (1944). (9) (10)
~ < > : c L I \P I )
I l r < ~ r m h c r31, 1948
Fungicidal Activity of Bisphenols P 4 U L B. I\3ilRSH, IIIARY L. BI-'I'LER, ASD BERKICE S. CL4RK S . D e p a r t m e n t of .4gricirltz~re,Beltscille, M d .
1 n continudtion of experinleiits previous11 reported (IP), t h i r t j -nine additional bisphenols and closely related compounds have been tested for fungicidal actiFitj as mildew prebentives on cotton fabric, bringing the total number of compounds tested to sevent5-three. Somc of these compounds are much more effective per unit weight on fabric t h a n others as fungicides, and the trend toward high act i \ i t ? is correlated with certain features of chemical structure. The data do not appear to warrant the selection of any siiigle bisphenol from among the group tested as having unique high potency b u t suggest rather t h a t high potenry is related to a certain generic t1pe of structure within the group. For convenience in presenting the data, 2,2'-methj lenebis(4-chlorophenol) has been selected arbitrarily from among t h e more active compounds and used as a point of reference for comparison with other compounds i n both a c t i \ i t j and chemical structure. Bis-CHCHg-, phenolic bridges consisting of -CHl-, -CHC&--, -CH=CII-, and -Sh a t e been found comor -SOT patible with high activity, whereas -SO-
bridges are much less satisfactory. The presence ot' halogen atoms in all four positions ortho to t h e bisphenolich? droxjls is consistently accompanied by low activity. Bromine has been found less desirable from the standpoint of fungicidal potenc) than chlorine i n bisphenols of high total halogen content. Bisphenols with a chlorothymol t ? pe of structure, with unusually high molecular weights, a coniplete lack of halogen, or ether linkages blocking both phenolic hydroxyls have been found low in activitj -
PIWYIOUS report ( 1 2 ) described experiments which dealt with the fungicidal activity of the bisphenols in tests as mildew preventives on cotton fabric. The present paper gives results of a continuation ef this earlier work. The previous report pointed out certain relations between chemical structure and fungicidal activity within the bisphenols. iimong the thirty-four diff erent compounds tested, several highly active materials were' notcd in a certain generic structural group, typified for conveil-
