2424
W. F. BARTHEL, B. H. ALEXANDER,P.
IIOP03H2is present in high concentration, hydrolysis by way of an intermediate analogous to -4 would lead to the alcohol, ROH, and monomeric metaphosphoric acid, HP03. But presumably such an intermediate will not lose ROH, with the bonding electron pair accompanying oxygen, as readily as will the negatively charged intermediate, A. In strong acid, one or more additional mechanisms, which require C-0 cleavage, dominate the hydrolysis. Thus the distinctive pH-rate profile for the phosphate hydrolysis may be explained provided that (at moderate acidities) proton transfer to the alcoholic oxygen atom is an obligatory part of the process. The mechanism for the hydrolysis of simple phosphate esters also can be formulated without the additional water molecule postulated in A so that the hydrolysis goes by way of a four-membered, rather than a six-membered, ring. However, the intermediate A bears somewhat the same rela-
+
hydrated to HpO-P03- simultaneously with the decomposition of A. These formulations can also explain the pH-rate profile. -4 mechanism somewhat analogous to that shown in equations 5 and 6 can explain the hydrolysis promoted by lanthanum hydroxide gel. Since the rate is considerably increased (see Table IV) by a substituent in the &position in the ester, the oxygen or nitrogen atom of this substituent is prob(21) F. H. Westheimer and W. A. Jones, THISJ O U R N A L . 63, 3283 (1941).
[CONTRIBUTIOK FROM
THE
VOl. 77
ably coordinated with lanthanum ion; since the rnethoxy group causes an increase in rate, the ionization of the substituent is certainly not a necessary part of the process. (Choline phosphate is a special case; perhaps here the positive charge of the nitrogen atom has a favorable electrostatic effect upon the rate of hydrolysis.) A plausible mechanism for the hydrolysis of methoxyethyl phosphate is shown in equation 7 .
I
+
tionship to the zwitterion, R-0-Pod, as the cyclic, H hydrogen-bonded intermediate in the decarboxylation of beta ketoacidsZ1bears to its corresponding zwitterion ; the six-membered ring appears a priori more probable than the other possibilities. Alternatively, the hypothetical ion, POs-, might be
GIANG, ~ N DS . .l. HALL
‘P‘
I
(B)
pas-
+ 1-1~0+Er2P04-
(ti,
Alternatively, the postulated intermediate, B,may hydrolyze directly to HZP04- or may rearrange directly to a lanthanum phosphate complex. I n eq. 7 the positive lanthanum ion replaces, as to essential function, the proton of the monoanion, ROP03H-, in the hydrolysis shown in eq. 5 . Bamann pointed out that the effect of lanthanum ion in promoting the hydrolysis of a-glyceryl phosphate can be observed only in alkaline solution where the gel, La(OH)3, is present as a separate phase. The exact statement of mechanism for a heterogeneous reaction is extraordinarily difficult a t the present time; a more accurate formulation must, in all probability, await a determination of the kinetics of metal-ion promoted hydrolysis of a phosphate ester in homogeneous solution. C A M H R I D G E , hIASS.
ENTOMOLOGY RESEARCH BRANCH, ACRICUL I C R A L RESEARCH SERVICE, U. S. DEPARTMENT OF AGRICULTURE]
Insecticidal Phosphates Obtained by a New Rearrangement Reaction1” BY by. F. BARTHEL, B. H. ALEXANDER, P. A. GIANGAND S. A. HALL RECEIVED DECEVBER 10, 19541b Dehydrochlorination of 0,O-dialkyl 2,2,2-trichloro-l-h~-dro~~ eth) lphosphonatcs causes rearrangement t o dialkyl 2,2dichlorovinyl phosphates. The products of dehydrochlorination were compared with compounds of known structure obtained by another synthesis and found to be identical. Othcr isomeric structures were considered and ruled out by chemical a n d physical data. S e w compounds are reported, several of which possess high insecticidal activity. 0,O-Diethyl 2,2dichloro-1-hydroxyethylphosphonate was converted to the systemic insecticide diethyl 2-chlorovinylphosphate.
In a previous paper2we described a series of 0 , O dialkyl - 2,2,2 - trichloro - 1- hydroxyethylphosphonates (I),which are readily prepared by condensing chloral with a dialkyl hydrogen phosphite. The lower members of this series possess insecticidal activity, and the methyl ester (I, R = CH3), as “Bayer L 13/69,’’ has been widely tested. Recently (1) (a) This work was conducted in part under funds allotted by the Department of the Army to the Department of Agriculture. (b) Original manuscript received November 11, 1954. (2) W. F. Rnrthrl, P. A. Giang and S. A . Hall, TRTS J O I J R H A I . , 76, 41Rli (186.1).
Mattson, Spillane and Pearce3a have reported on their original discovery that “L ‘*/69’) undergoes dehydrochlorination in the presence of alkali, to yield a volatile, more highly insecticidal ester, to which they assigned the enol structure, 11. They s~bsequently~b assigned the keto structure, 111, to the dehydrochlorination product on the basis of infrared absorption data and formation of an (3) (a) A. M. Mattson, J. T. Spillane and G. W.Pearce. Abstract of paper for presentation a t the 126th Meeting of the American Chemical Society a t New York, N. Y.,September 12-17, 1951; (b) ibid., pnpw as artiiallv pr?srntwl
May 5 , 1955
2425
lNSECTICIDhL PHOSPHA‘I‘ES
phosphoric acid. He also found that the esters took up 1 mole of chlorine to yield the expected tetrachloro derivatives VI. We repeated the preparaI I1 tion of V by the Perkow route from trialkyl phosO H 0 phites and chloral, and found the physical proper/I I I1 ties of the liquids essentially the same as those of (RO)zP-C-CC12 (RO)zP-O-CH=CCI. the corresponding esters obtained by dehydrochlo‘0’ 1v V rination of I. We then made tetrachloro derivatives 0 0 VI (R = CH3, C2Hs) of the compounds made by 11 11 each route and found them identical. The infrared (RO)*P-OCHClCC13 ( C ~ H ~ O ) ~ P - O C H = C H C spectra ~ of the dimethyl esters prepared by the PerVI VI1 kow method, and the dehydrochlorination method osazone with 2,4-dinitrophenylhydrazine. The were the same. Although we could find no other osazone was described as identical with that of reports in the literature of a rearrangement of a glyoxal. The same osazone is obtained from di- phosphonate to a phosphate, it is evident that this chloroacetaldehyde. Presumably dichloroacetalde- rearrangement takes place hyde was split off from the dehydrochlorination 0 0 product on treatment with 2,4-dinitrophenylhydraI( 11 - KC1 zine. (RO)2PCHOHCCla (RO)zP-OCH=CCl~ We have prepared the dehydrochlorination prodI V ucts by treatment of I with aqueous sodium hydroxWe also prepared diethyl 2-chlorovinyl phosphate ide, or in the case of higher members of the series (VII), (although in poor yield) by dehydrochlorin(where R = CSH7 or above), with alcoholic sodium hydroxide, removing 1 mole of hydrogen chloride. ation and rearrangement of O,O-diethyl2,2-dichlo0 The dehydrochlorination products of I, which disIt tilled under vacuum, are mobile esters heavier than ro-1-hydroxyethylphosphonate, (CsH60)2PCHOHwater. These did not react with 2,4-dinitrophenyl- CHC12. We then prepared VI1 in fairly good yield hydrazine or semicarbazide hydrochloride. How- by the Perkow route from triethyl phosphite and diever, the dehydrochlorination product of “L 13/ chloroacetaldehyde. The esters obtained by both 59” (I, R = CHa) was cleaved with p-nitrophenyl- routes were identical. The vinyl rearrangement hydrazine to yield the corresponding glyoxal osa- thus can be extended to p-dichloro-a-hydroxyphoszone. We then synthesized I11 (R = CHa, CzHs), phonates. It is of interest to note that VI1 has according to the general procedure described by been reported6 to be a systemic insecticide, alKabachnik and RossiYskaya4as though no preparative method is given. 0 The infrared spectra’ of the dimethyl esters of I, I1 I11 and V were compared. A peak a t 3.21 p of I is (R0)zPOR ClCCHClz + I11 RCl characteristic of an associated hydroxyl group; a t The a-ketophosphonic esters, I11 (R = CHI, one-tenth dilution this peak is diminished, while a CpHb), thus obtained, readily reduced Fehling rea- sharp peak a t 2.93 p, characteristic of a free hygent and reacted vigorously with p-nitrophenylhy- droxyl, is increased. I n confirmation of this we drazine to give the expected osazones. They also acetylated I (R = C1H6)proving that it is indeed a exhibited markedly different physical properties phosphonate. The spectrum of I11 shows a band a t (including infrared spectra) from the isomeric com- 5.88 p, attributable to the carbonyl group. This pounds derived from I by dehydrochlorination. band is absent in V. The spectra of VA and VB, The keto structure I11 is, therefore, excluded for prepared, respectively, by dehydrochlorination and the dehydrochlorination compounds. The latter by Perkow’s5 method, are identical. I n Table I gave no active hydrogen by the Zerewitinoff we have listed the properties of the dialkyl chloromethod, nor could they be acetylated as could the vinyl phosphates, V and VII, with the yields obparent compounds I. Infrared spectra also failed tained by the dehydrochlorination procedure and to show the presence of a hydroxyl group in the de- by the Perkow method. hydrochlorinated derivatives of I. Thus, structure I1 is excluded. The epoxide structure, IV, was elimExperimental8 inated because of addition of 1 mole of chlorine Dialkyl 2,2-Dichlorovinyl Phosphates (V). Method A . with no formation of hydrogen chloride. The possi- By Dehydrochlorination and Rearrangement of the Correbility of a rearrangement reaction on treatment of I sponding Dialkyl 2,2,2-Trichloro-l-hydroxyphosphonates.with alkali to yield dichlorovinyl phosphates V was The phosphonate I (0.1 mole) was partially dissolved in then considered. Perkow,6 in connection with the (6) R . A. Corey, S. C. Dorman, W. E. Hall, L. C. Glover and R . R. preparation of some new anti-cholinesterases, re- Whetstone, Science, 118, 28 (1953). These infrared spectra have been deposited as Document numcently described a rearrangement reaction between ber(7)4458 with the AD1 Auxiliary Publications Project, Photoduplicatrialkyl phosphites and chloral with elimination of 1 tion Service, Library of Congress, Washington 25, D. ‘2. A copy may mole of alkyl chloride, to yield what proved to be V. be secured by citing the Document number and by remitting $1.25 for I n support of the phosphate structure, Perkow photoprints, or 51.25 for 35 mm. microfilm in advance by check money order payable to: Chief, Photoduplication Service, Library found that acid hydrolysis of the esters yielded or of Congress. 0
0
ij
I1
(RO)*PCHOHCC13 (R0)2PC(OH)=CCIz
00
11 I1
(R0)zPCCHClz I11
-
+
+
(4) M. I. Kabachnik and P. A. Rossiiskaya, Isvcsfiya Akad. Nauk 0f.kh.n. (No. 4), 364 (1945). (5) W. Perkow, Bcr., 87, 755 (1954); W. Perkow, K . Utterich and Pr. Meyer, N n f r ~ r i u i . s r ~ ~ z s r h a f39, l c n ,353 (1982).
Dr. Jonas Carol of the Food and Drug Administration kindly determined the infrared spectra. (8) Physical properties, yield and analyses not chnwn are listed in ‘Tahlr 1.
2426
Vol. 77 TABLE I 0
I1
DIALKYL CHLUROVISYI. PHOSPHATES ( RO)2P-0--CH=CC12 Yield, h
R
CHa CzlIs C3H7
hfethod"
A B A I3 A
B $-CII-III A
B C4H9
A B
B.p.C
%
oc
57 86' 51
120 120d 131-132 132-133" 107 114 99-105 108-111' 125-128 124-131
9i"
36 70' 26 7 3 53
79
Ilm.
14 1I
14
14 0.3 1.1 0.7 0.9 0 .X
0.2
?I
1,4524 1 453:3 1 ,447SI 1.1473 1.44
