2232
E. J. MGEHAN, I. M. KOLTIIOFF, C. AUERBACII AND 11. MINATO
(CONTRIBUTION FROM
THE
Vol. 88
SCHOOL O F CHEMISTRY OF THE UNIVERSITY OF MINNESOTA, MINNEAPOLIS, blINNESOTA, .4ND BROOKHAVEN NATIONAL LABORATORIES, UPTON,NEWYORK]
OF
Reaction of Cumene Hydroperoxide with Hydrazine Induced by the Reaction Between Iron (111)EDTA and Hydrazinel r 2 BY E. J. MEEHAN,I. M. KOLTHOFF, C. AUERBACH AND H. MINATO RECEIVED NOVEMBER 11, 1960 The direct reaction between cumene hydroperoxide (CHP) and hydrazine is negligibly slow a t 0' and pH 9. Upon the addition of the ethylenediaminetetraacetate (EDTA) complex of iron(II1) a pronounced reaction occurs a t 0' in which the CHP is transformed quantitatively into acetophenone. From an analysis of the kinetic results it appears that the first step is the relatively slow reaction between iron(II1) EDTA and hydrazine, forming iron(I1) EDTA and hydrazyl radicals. Both iron(I1)EDTA and hydrazyl are oxidized by hydroperoxide, forming acetophenone (AP) and methyl radicals. The NzH, +CHa NzH,. and NzHa. ROOH -+- N2Hz. acceleration is probably due to the chain reaction CHs. AP CHa.. The effect of the EDTA complex of iron(II1) is markedly suppressed by sodium acrylate (M), which can be (n l ) M +CHIM,., the latter itself being terminated by reaccounted for by the competitive reaction CHI. M action with CHI..
+ + + -
+
Introduction In emulsion polymerization a t low temperature the initiating free radicals generally are produced by the interaction of a hydroperoxide and iron(II), the latter being formed continually by reaction of iron(II1) with a reducing agent. As an example, hydrazine is a suitable reductant for the ethylenediaminetetraacetate complex of iron(III), denoted as ir0n(111)EDTA.~ I n order to interpret the involved kinetics of emulsion polymerization, the kinetics of radical formation in homogeneous medium must be known. The mechanisms oi the reactions between iron(II1)EDTA and hydrazine4 and between iron(I1) EDTA and hydroperoxide6 have been established. The present paper deals with the system iron (111)EDTA-hydrazine-cumene hydroperoxide (CHP) in aqueous solution, in the absence and presence of sodium acrylate. Experimental Most of the experimental procedures and the chemicals have been described.' CHP (Hercules Powder Co.) was purified by the method of Hock and Lang.B The purity was found to be 90.570 by the arsenious' and iodometric methodss Acetophenone and acrylic acid were Eastman products. Each was purified by fractional distillation before use. The rate of disappearance of CHP from solutions containing iron(II1)EDTA and hydrazine a t pH 9 was measured polarographically with the dropping mercury electrode at -0.9 v. us. the s.c.e., using a Leeds and Korthrup Electrochemograph Type E. This potential is in the diffusion current region of both iron(II1)EDTA and CHP. After correction for the residual current the diffusion current was found proportional to the concentration in solutions of either CHP or iron(II1)EDTX over a range between 3 X 10-4 and 2 X lo-* M. In a mixture 0.05 ilf in borax and containing both electroactive species the observed diffusion current a t -0.9 v. was smaller than the sum of the individual diffusion currents of each species. This is caused by reaction of iron( 1I)EDTA formed upon reduction of iron(1II)EDTA with some CHP a t or near the electrode with formation of the RO. free radical, which rapidly disproportionates into acetophenone and the CHI. free radical. ( 1 ) This investigation was carried out under a grant from the National Science Foundation. (2) Initial experiments are described in the M.S. Thesis of H. Minnto, University of Minnesota, July, 1958. (3) I. M . Kolthoff and E. J. Meehan, J Polymer Sci., 9, 343 (1952). (4) H. Minato, E. J. Meehan, I, M. Kolthoff and C . Auerbach, J . A m . Chcm. Soc., 81, 6168 (1959). (5) W. L. Reynolds and I. M. Kolthoff, J. P h y s . Chem., 60, 996 (1956). (6) H. Hock and
S.Lang, Bcr., 77B,157 (1944). (7) D. C.Walker and H.S. C o n w a y , Anal. Cham., 25, 923 (1953). ( 8 ) V. R. Kokatnur and ILI. Jelling, J . Am. Chon. Soc., 63, 1432
(1941).
+
+
+
Acetophenone yields a wave with a diffusion current at -1.7 v. which is proportional to the acetophenone concentration. This wave was not found in solutions of pure CHP but was observed in mixtures of iron(II1)EDTA and CHP. The current giving rise to the wave a t 1.7 v. is kinetic in nature. The height of this wave gives a measure of the rate of reaction between iron(I1)EDTA and CHP and is determined more accurately than the "negative kinetic current" a t -0.9 v. which is equal to the difference between the current calculated, assuming no reaction occurs, and the observed current. When the effect of drop time upon the diffusion currents was taken into account, the sum of the diffusion currents a t -0.9 and -1.7 v. was found equal to the calculated sum of the diffusion currents a t -0.9 v. In mixtures which were 0.001 ill in iron(II1)M in CHP the observed curEDTA and 0.3 to 1.1 X rent a t -0.9 v. corresponded to an apparent decrease of the CHP concentration of 10% both a t 0' and 25'. This decrease was found proportional to the CHP concentration under the above conditions. Thus, the current a t -0.9 v. serves as a reliable indicator of the concentration of CHI' in solutions containing iron(I1I)EDTA. In blank experiments it was shown that hydrazine is polarographically inactive in the voltage range used.
-
Results Reaction Between CHP and Hydrazine.-The direct uncatalyzed reaction is negligibly slow. For example in a solution 0.001 M in CHP, 0.05 M in hydrazine, 0.006 M in EDTA and 0.05 M in borax (pH 9), only 5% of the CHP was found to react in 30 minutes a t 25'. The reaction was much faster in the absence of EDTA, the half-time being a few minutes. This large increase in rate could be attributed to a trace of copper which catalyzes the reaction and which is made harmless by EDTX. This effect is so great that the catalysis codd be made use of analytically for the determination of traces of copper. The reaction a t 25' was foulid practically instantaneous in the presence of 5 X 10-5 A4 copper(II), but with an excess of EDTA this concentration of copper had no effect on the rate. Oxygen also exerts a pronounced accelerating effect on the reaction of CHP in the presencc of EDTA. The reaction between C H P and hydrazine has not been studied further, but i t has been established that under all the conditions used in the present work (particularly presence of EDTA, absence of oxygen) the reaction is slower by several orders of magnitude than the iron(II1)EDTX-hydrazine reaction.4 Disappearance of CHP in Presence of Iron(II1)EDTA and Hydrazine.-The reaction rate was measured a t 0" and 25" in solutions 0.05 M in
REACTION OF CUMENE HYDROPEROXIDE WITH HYDRAZINE
May 20, 1961
borax, 0.5-1 X Miniron(II1)EDTA and 0.010.05 M in hydrazine. The solutions contained about 10-8 M free EDTA and sufficient sodium sulfate to make the ionic strength unity. Separate experiments showed that variation of the ionic strength from 0.2 to 1 did not affect the rate appreciably. Results of typical experiments a t 5 X M iron(II1)EDTA and three concentrations of hydrazine a t 0' are shown in Fig. 1. Brief induction periods, possibly due t o traces of oxygen, were observed both a t 0 ' and 25'. Initial reaction rates (vo) are given in Table I. These were obtained graphically allowing for the induction periods.
I
20
TABLE I DISAPPEARANCE O F CHP I N PRESENCE OF IRON(III)EDTA AND HYDRAZINE, p H = 9.0
(1) 0' Molar concn. X 10' Molar concn. CHP FeyNtHl NaA"
X
ka k11FeY-l X l o b e [ N d h l X 10s
I 40
I
60
I 80
2233
I 100
Y 120
MINUTES.
Fig.1.-Reaction of CHP a t Oo, pH 9.0; Fe(II1)EDTA = 5.17 X lo-' M; hydrazineconcentration: curve 1,O.Ol M; curve 2, 0.02 M; curve 3, 0.05 M.
few minutes of each experiment in the presence of acrylate the reaction rate was relatively larger than the subsequent rate, undoubtedly due to a .. trace of a reducing impurity in the acrylate. After .. this brief initial period the reaction showed the .01 0 . 1 same kinetic behavior as in the absence of acrylate, .02 0.1 with the values of ko given in Table I. (2) 25' I n separate experiments i t was shown that addi5.3 .. 1.2 7.1 5.170.01 .. tion of 0.1 M sodium acrylate had no effect on the 9.4 .. 2.4 7.1 5.17 .02 .. rate of reaction of iron(1II)EDTA and hydrazine .. 6.0 7.1 5.17 .05 . . 12 a t Oo, pH 9. 5.4 .. 1.1 6.0 4.S .01 .. Stoichiometry of CHP-Iron(I1)EDTA Reaction.6.0 4.83 .02 .. 7.6 .. 2.2 When the corkentrations of iron(I1)EDTA and 13 .. 5.6 6.0 4.S .05 C H P (absence of hydrazine) in the reaction mixture Sodium acrylate. * Initial rate of reaction, mole I.-' are of the same order of magnitude, the molar min.-' Limiting zero order reaction rate, mole 1.-1 rnin.-' reaction ratio is about unity. No induced deObtained by extrapolation to zero time. composition occurs and the reaction products are Under the experimental conditions a t 2 5 O , the rate AP and ethane. Under the conditions described was practically zero order with respect to CI-IP in this paper, iron(I1)EDTA is formed slowly and until 60-70y0 had reacted. From then on there its concentration is small compared to the initial was a gradual decrease in rate. I n the experi- concentration of CHP. The following experiment ments a t 0' the reaction was between zero order was carried out to see whether any induced decornand first order with respect to CHP. For such position of C H P occurs under such conditions. T o M in CHP, kinetics, - dc/dt = KO kfc, and a plot of log (c 20 rnl. of an air-free solution 7 X ka/kf) vs. t should yield a straight line of slope kf. 0.05 M in borax was added 5.0 ml. of air-free 5.3 X M iron(I1)EDTA in 0.05 M borax. The addiBy trial it was found that such linear plots could tion was carried out continuously by means of a be obtained with values of k0lrt.t of the order of The corresponding values of the zero order constant syringe over a period of five minutes. The concentration of unreacted C H P was determined ko are given in Table I. In a number of instances the amount of aceto- polarographically. Duplicate experiments yielded phenone (AP) was measured polarographically a t a reaction ratio, moles iron(I1) consumed/moles the end of the reaction. Within the experimental C H P consumed, of 1.09 f 0.04, which is the same error (50/,) the C H P was converted quantitatively as found when equivalent concentrations of C H P to AP. Application of the chromotropic acid test and iron(I1)EDTA are mixed rapidly. for f ~ r m a l d e h y d emodified ,~ to include the determiDiscussion nation of methanol, showed that no detectable amount of either substance was present a t the end The first step in the reaction between iron(II1)of the reaction. EDTA (Fey-) and hydrazine is the reversible Sodium acrylate has a pronounced retarding reaction While in its absence the reaction a t effect a t 0'. FeyN~HI Fey-&HI. H + (1, -1) Oo was 90% complete in 100 minutes, in the presence of 0.1 A4 acrylate only 45% reaction oc- I n a medium 9.3% in ethanol the bimolecular rate curred in 126 minutes (5 X lo-* M iron(II1)- constant of the forward reaction 1 is 0.090 and 2.3 EDTA, 0.01 M hydrazine). Reacted C H P was l.-mole-' min.-l a t 0 ' and 25', respectively, the converted quantitatively to AP. During the first hydrazine concentration being that of the free base.4 Ethanol reduces the rate somewhat, and (9) C . E. Bricker and H. R . Johnson, I n d . Eng. Chcm., Anal. E d . , the value of the rate constant in the absence of 17, 400 (1946). 7.2 7.2 7.2 8.0 8.8 8.8
+
5.38 5.38 5.38 10.0 5.00 5.00
0.01 .02 .05 .01
..
1.2 2.3 3.6 2.7 0.4d 0.7d
..
0.19 .14 .l3 .29 .ll .28
0.05 .ll .27 .10 .05 .10
+
+
+
+
2234.
E. J. ~ I E E H A I. N 11. , KOLTHOFF, C. AUERBACH AND H. NINATO
ethanol is estimated to be about 0.1 and 2.5 1.mole-l min.-l a t 0 and P5', respectively. Thus in X in ironi1II)EDTA and lop2 a solution 5 X AM in hydrazine the rate of formation of iron(I1)EDTA and hydrazyl radicals in reaction 1 is 0 5 x mole I.-' min.-' a t 0'. Values of kl[FeY-][N2H4]for other experimental conditions are given in Table I. Iron(1I)EDTX is produced also in the subsequent fast oxidations of N?H3,, N2H2. and N2H., SO the maximum rate of formation of iron(1I)EDTA is four times the rate of reaction 1. The rate constant for reaction 2 between iron(I1)EDTA and C H P (ROOH) a t 0 ' is 1.44 X lo4 1.
Vol. 83
iron(III)EDTA,*it may be expected to be oxidized more easily by the strong oxidizing agent hydroDeroxide ROOH i NzH3' +RO. S?Hz. + H20 (5) Reactions 3, 4 and 5 constitute a chain reaction between hydroperoxide and hydrazine. Undoubtedly NzH2. and N2H. radicals can also be oxidized by hydroperoxide. The above mechanism corresponds to two simultaneous modes of disappearance of hydroperoxide : (a) a direct reaction with iron(I1)EDTA and hydrazyl radicals produced in ( l ) , zero order with respect to CHP a t sufficiently high concentration Fey-ROOH +Fey- + RO. + OH- ( 2 ) of CHP, the maximum rate of the zero order remole-' min.-', independent of p H over a wide action being 4 X the rate of ( l ) ,and (b) a chain range.5 I t is known that k-, is smaller than k l , reaction 3, 4, 5. It is impractical to attempt to so in the presence of sufficient C H P all the iron(I1)- derive the kinetics in the general case, in which EDTA produced in (1) reacts with C H P and the allowance is made for the oxidation of the three back reaction (-1) does not occur. No induced hydrazine-derived radicals ( N2H3., N2H2. and decomposition of CHP occurs in mixtures of C H P N2H.j by two oxidants, and for the possibility of and iron(I1)EDTA (molar reaction ratio is unity), the back reaction -1 a t low concentrations of and the direct reaction between C H P and hy- hydroperoxide. Because of experimental difficulties drazine is negligibly slow. Therefore if C H P i t was not possible to use the polarographic method disappeared only by reaction with iron (11)EDTA, a t much higher concentrations of CHP than used the maximum rate of disappearance would be four in the present work. For this reason no attempt times the rate of (1). Also, the reaction would be has been made to analyze the kinetics in detail. zero order with respect to CHP, when sufficient Qualitatively it is plausible that the chain seCHP is present. This situation appears to be ap- quence should be first order with respect to hydroproached a t 25'. The decrease in rate after peroxide, so the combined zero-first order kinetics 60% of the C H P has reacted may be attributed to observed a t 0' may be understood. Allowing for the fact that the back reaction (- 1) can no longer the uncertainty in deriving ko from the rate data, be neglected. the agreement in order of magnitude between The most interesting results are obtained a t Oo, ko and kl[Fe'rT-][NzH4] a t 0' is entirely satisa t which temperature the values of DO are from 13 factory. The chain initiating reaction 4 is comto 29 times the rate of (l), and the reaction appears petitive with the dimerization of CHi, to ethane. to be between zero and first order with respect to From the experimental results it appears that a t CHP. Various mechanisms to account for the 25' the dimerization is favored over reaction -2. accelerated reaction have been considered. Both As mentioned in the Experimental part sodium the acceleration and the kinetics could be ac- acrylate does not affect the rate of reaction of ironcounted for by the reaction (1II)EDTA and hydrazine. This is conclusive proof that acrylate does not react with either RO. +-1P CH3, (3) iron(1I)EDTS or with N2H3.; if such reaction f ollowed by occurred, the back reaction -1 would be supCH3. + ROOH +RO. CHaOH pressed and the over-all reaction rate would be inHowever, no methanol could be detected in the creased. The suppressing effect of acrylate (AI) reaction mixture and, in addition, the slow addition on the induced reaction of hydroperoxide with of iron(I1)EDTA to C H P did not lead to any auto- hydrazine can be accounted for by catalytic decomposition. The following mechaCH3, + M + ( n - 1) M --+ CHJln. (6) nism appears plausible, Iron(1I)EDTA formed in which is competitive with (4). (1) reacts with C H P (2) and RO. thus formed The present research has been discontinued. decomposes (3). In the absence of hydrazine the The experimental results reported in this paper methyl radicals self-terminate to form ethane. are new and unexpected. However, more work In the presence of hydrazine reaction 4 is assumed would be desirable to substantiate the above interCH3. + KzHd ---+ CHI + S?Ha' (4) pretation. This is analogous to the abstraction of hydrogen Acknowledgment.-Grateful acknowledgment is by methyl radical from dimethylhydrazine, which made of the experimental assistance of Mary K. has been postulated to occur in the photolysis of Reck, who worked during the summer of 1960 on a the latter substance. lo Sitice hydrazyl radical National Science Foundation Undergraduate Reis easily oxidized by the weak oxidizing agent search Participation Program and who repeated the rate measurements described in ref. 2. ( l o ) W. L. Kay and H. A. Taylor, J . Chem. P h y s , 10, 497 (1942).
+
+
+
+
