3212
Yol. 77
h-OTES
TABLE I A B S O R P T I O S O F I O D I S E I N SOME ‘“ON-ACTIVE”
SOLVENTS
Solvent
Xax., mfi
Ref.
Chloroform Carbon tetrachloride
520 520 517 518 520 51s
2 2 3
Carbon disulfide
4
2 5
I n the course of several investigations on the behavior of iodine in various solvents, the author noted two instances wherein iodine dissolved in solvents t o give colors that were completely unexpected. This note is the result of these observations. (A) Solvents Containing Nitrogen Atoms, Such as Pyridine, Quinoline, or Amines, Dissolve Iodine to form brown Solutions-These solvents are termed “active” solvents. Previous studies have revealed that in these cases the iodine reacts with the solvent.6 When iodine was dissolved in completely fluorinated amines such as (C4F9)3N, (CaFb)zN(C3F7)and ( C Z F ~ ) ~the N ,resulting solution was violet, showing an absorption maximum a t 520 mp. This band is consistent with other investigations for iodine in violet solutions (see Table I), Besides the maximum a t 520 mp, there is also observed a slight maximum a t 350 inp. The latter peak, however, is of much lower optical density and probably indicates some sort of interaction between the solvent and the iodine. The surprising fact these fluorinated amines yield violet solutions becomes intelligible when one stops to consider the nature of these compounds. In these amines both steric and electronic effects must be considered. The three rather large alkyl groups surrounding the nitrogen atom and the presence of so numerous a group of highly electronegative fluorine atoms tremendously decrease the availability of the free pair of electrons present on the nitrogen atom. Hence, the nitrogen is no longer able to bond with the iodine as, for example, in the case of pyridine where the addition compound (C5H6N)z.Izhas been i ~ o l a t e d . ~ (B) Iodine Dissolves in Phosphorus(II1) Chloride to Give violet Solutions.-Phosphorus(II1) chloride is a highly reactive but unsaturated molecule. It readily adds oxygen from the air in the cold to forin POCIS. If refluxed with sulfur in the presence of aluminum chloride, P e l 3 readily adds sulfur to form PSC13.8 Chlorine adds easily to PC13 to give PC15. Bromine also adds to PC13 to form the low melting solid PC13Br2.9 However, it has not been possible to add iodine to PC13 to form the hypothetical PC1312. The failure to isolate an adduct of iodine with PC13 may be due to the insta(‘2) H. Rigollot, Cornpi. r e n d . , 112, 38 (1891). (3) H. A. Benesi and J. H. Hildebrand, THISJOTJRKAL, 7 1 , 2703
(1949). (4) W. R . Brode, ibid., 48, 1877 (1920). ( 3 ) J. Gioh, 2. a l f o r g . a t k e r n . C h e m . , 162, 287 (1927). (6) J . Kleinberg, E. Colton, J. Sattizahn and C. A . VanderWerf, THISJ O C R N A L , 7 5 , 412 (1953). (7) M.Chatelet, Compl. r e n d . , 196, 142 (1933). I;. R n o t z , dsterv. Ciiem. z.,SO, 128 ( I W Y ) , F. E p h r a i m , “Inotganic Chemistry,” Gurney and J a c k w n , London, 1949, F i f t h ed., p . 705.
bility of the yet unknown PC1312. Another reasonable explanation would be that the iodine molecule is simply too large to approach close enough to the phosphorus atom to form an addition compound and thus increase the covalency of phosphorus to five. That iodine does not react with PC1, is evidenced not only by the violet color of the resulting solution and the absorption band a t 520-500 mp but also by the persistence of the violet color even after refluxing iodine and Pc13 for two hours. Experimentall’Jall Materials.-The completely fluorinated aniines were obtained through the courtesy of the Miiinesota Mining arid Mfg. Co. and are colorless, odorless liquids. They were used directly as received. Iodine was J. T. Baker C.P. gradc nhich was stored in a desiccator over phosphoric anhydride and used without further purification. Phosphorus(II1) chloride was Merck reagent material, boiling range 7575”, which was taken from a freshly opened bottle. Apparatus and Procedure.-The spectrophotometric measurements were made on the Beckman quartz spectrophotometer, model DU. A pair of 1 cm. matched silica cells were used, one of which contained the solvent as the blank. The iodine solution was prepared prior to measurement and stored in the dark except for the period of use. Slight heating was necessary to effect solution of iodine in the amines. All solutions were ca. l W 3 JI. (10) T h e assistance of Mr. E. G. Vassian in iilitnining t h e measurements in PClr is hereby ackoffledged. (11) T h e measurements of iudine in t h e fluorinated amines were performed b y t h e author as p a r t of his L1.S.thesis, University of Kansas, 1952, under t h e direction of Dr. Jacob Kleinberg.
S c a o o L O F CHEMISTXY GEORGIA IXSTITUTE O F ’rECHNOLOGY ilTLBSTA, GEORGIA
The Oxidation of Oxalic Acid and Oxalate by Bromine and the R61e of the Positive Bromine Ion BY Y , KNCLLER AYD B.PERLMUTTGR-HAYMANI RECEIVED SOVEMBER 23, 1954
Bromine in aqueous solution is known to oxidize oxalic acid and oxalates to carbon dioxide, itself being reduced to hydrobromic acid. X o substitution occurs. The kinetics of this reaction have been investigated by several Josefo~icz,~ and later Griffith, McKeown and Winn5 concluded the rate-determining step to take place between the acid oxalate ion arid hypobromous acid. Liebliafsky and h9akower6 found for slightly acid solutions of hypobromous acid a rate constant agreeing well with this mechanism. More recently it has been pointed out by Hinshelwood7 that the dependence of the rate of reaction on pH is equally compatible with the assumption that the rate-determining step takes place between the divalent oxalate ion and a bromine cation, and this reaction is shown to have a more plausible electronic mechanism than the reaction assumed previously. I n the present paper the reaction has been reinvestigated under carefully controlled conditions a t values of (1) ’ro whom inquiries should b r addresqed. (‘7) M Roloff, Z . p k y s i & . C h r i n . 1 3 , 340 (1894). (3) A Berthoud and H. Bellenut, J . c h i r n . p h y s . . 21, 308 (1921). (AJ E Josefowicz. Rocz c h ? x , 8 , 193 (1028). (31 1%.0. Griffith. A. AlcKeow-n and A. G. LV’inn, T r a m . P a r a d a s S O C . ,28, 107 (1932). (iii 1%. A. Liebhafsky and R l i a k u w e r , i b i d . , 2 9 , A97 (1933). ( 7 ) C X. €Iinqhelwor,d, J . Chcin. S o r . , 0 9 1 (19171.
June 20, 1955
NOTES
p H of 0.8 to 5, and the two mechanisms proposed in the literature are discussed.
and
Experimental In order to avoid losses of bromine due t o evaporation, the reaction mixtures were prepared in small glass-stoppered bottles (4-7 for each experiment). Only one sample was taken from each bottle a t a suitable time after pipetting the necessary quantity of hypobromite solution into a known volume of a solution containing all the other components of the reaction mixture. The progress of the reaction was followed by iodometric titration, using 0.01 N thiosulfate. The pH was kept constant by the use of buffer solutions (phosphate, chloroacetate or acetate), or, in the strongly acid range, by addition of perchloric acid. The pH was measured with a glass electrode and a Beckman Model G p H meter. The hypobromous acid was prepared by mixing yellow mercuric oxide arid bromine water, decanting and distilling in vacuo. The experiments were carried out a t 20 =!= 0.1' (unless otherwise stated), in ordinary glass vessels, and in the absence of daylight and of direct electric light. Unless otherwise stated, the ionic strength was 0.15-0.24; changes within these limits had been found not to influence the reaction rate. The initial bromine concentration was 4.8-6.3 mmoles 1.-*.
The experimental results can be summarized in the following statements: (a) The buffers used were found to exert no specific influence. (b) The reaction was confirmed to be of the first order with respect t o total oxalate. (c) The effect of varying the initial bromide concentration is shown in Table
Results The assumptions t h a t hypobromous acid or the bromine cation are the reactive oxidizing agents both lead to the conclusion that the reaction rate must be equal to dr/dt = k*[Br,](b - x)/[Br-]
(1)
where b is the initial concentration of the oxalate, x its decrease after a time t , and k* a rate constant which depends only on P H . Assuming the concentration of all other oxidizing species t o be negligible in comparison with those of molecular bromine and the tribromide ion, we can get the following approximation
c
3213
= (3K3 f 4a -!-2C)(2K3
+
2U
f C)/(Ka
+ +6 + U
C)
TABLE I INFLUENCE OF INITIALBROMIDECONCENTRATION OX THE RATECONSTANT fiH
b , mmole I.-1
1.97"
1.06 x 0.91 x 0.86 x 4.39 x 4.29 X 4.13 X 3.81 x 3.83 x 4.00 X 3.87 X 3.72 x
20.6 41.2 82.4 5.15 10.3 20.6 41.2 82.4 20.6 41.2 82.4 164.8
39.8
2.64-2.68
k*, s e c . 3 mole-' 1.1
c , mmole I.->
19.9
10-5 10-5" 10-5
10-5
10-5 10-5
IO+ 10-4 3.81 x 10-4 a In the experiments a t low PH, k* calculated for the first two or three measurements is somewhat higher than the value for the remainder of the reaction; only the latter is given in the table. * Ionic strength 0.28. 4.54-4. 56b
9.96
I.9 (d) The influence of p H is shown in Fig. 1, where log k* is plotted us. PH (circles) (initial bromide concentration: 41 mmole 1.-'). (e) In hypobromous acid solutions-in the presence of silver ion to prevent the formation of free bromine-the reaction is almost instantaneous a t 0".
where a is the initial concentration of the oxidizing agent as obtained by titration, c the initial stoichiometric bromide concentration, and Ka the dissociation constant of the tribromide ion.8 Inserting expression (2) into eq. 1 and integrating, we get a ( b - x)
1
k* = ; [ A
where
--n In -___ b ( n - x)
+
+ + + + +
+
b - B In -b - %
1
A = (2a C) (2a c K) B = [2u(2a I b 3K3 4c) Jb(3b 4c 2K3) c(7K3 6 ~ ) 2Ka2I/(K3 a b c)
+
+ +
+
2
3
4
5
PH. Fig. 1.-Dependence
of rate constant on pH; log k * X X 10'O
+ + + + + 1 0 6 (circles), log k*F X 10'0 (crosses) and log k*uH (squares) us. PH. a + K312 - 4(n f 2 x ) l ' h + c + 3~ - + Ka Discussion
(8) From t h e exact expression
1/[&l
=
[(C
X
X)(C
2K3(a
- x)
Q.
An inspection of Table I shows the rate constant k* to be almost constant for con1/[Rrll = stant p H , over a fairly wide range of bromide con(a + (' 2x + K 3 ) ( C+ + a + K 3 ) centration. Thus, with respect to the dependence of K 3 ( a- x ) ( c + x a + K3) the rate on bromide concentration, we corroborate on (This expression is justified provided 4(a - x ) ( c f P x ) / ( K a 4- c + the the Jindings Of prmious au'hors74-6 and a + x ) < < l , a condition which is easily seen t o hold for all values of c . ) either hypobromous acid or the bromine cation m ay Combining this expression with .. one gets
+
[Br-]/[Brzl = 1
+ ( c + 3% - a)/[Brp]
we obtain ( 2 ) . (The exact form of equation 1 can also be integrated: t h e result, however, is too cumbersome t o be of a n y practical use,)
(9) T h e value of Ks used for t h e evaluation of k* is 5.8.10-2 (R. 0. Griffith, A. McKeown a n d A. G. Winn. Trans. Faraday Soc., 28, 101 (1932): G. Jones and S. Backstrom, THISJOURNAL, 56, 1517 (1934)).
3214
NOTES
Vul. 77
be the oxidizing agent (This conclusion is confirmed by the high rate of the reaction in hypobromous acid solution). I n the former case, the meaning of k* becomes
It is this fact which makes the two mechanisms kinetically indistinguishable; factors such as bromide concentration, pH or ionic strength will influence the reaction rate in exactly the same way on either assumption. l 3 We can, however, draw some k * = + n H Y R ,KhK3 [k(HBrO + HzCzOd ~ H ~ Y H C ~ O , ' conclusions from the value of the rate constants which the latter assumption would lead to. Now, k ( a B r o + HC204-1 KIOH-t k(aaro + c20,-) K ~ K ~ Y H c ~ o ~ - / ~ known. Various attempts the value of K B is~not "fc*04-1 ( 5 ) to estimate its upper limit,'4,15though based on where K h is the hydrolysis constant of bromine, a H doubtful assumptions, give results compatible with the hydrogen ion activity, K1 and K I are the first the value which Bell and Gellesi6 obtained from and second dissociation constants of oxalic acid, thermochemical data and a consideration of certain analogous reactions; they found [H2Br0+I [Br-]/ respectively, the y's activity coefficients, and considered correct to within a few F = F(na) = U I T ~ Y H C ~ O ~ - I ~ I O HK~K~Ync~o,-/"fc~oi- [Br2] = powers of ten. The value of [Br+] [Br-]/[Br2] Now, if all the three oxalate species were oxidized was found to be still much lower. Although the a t approximately equal rates, k*aH would be inde- assumption of HzBrOf existing in solution seems pendent of pH. I n Fig. 1, the logarithm of this to offer the most plausible explanation of certain quantity is plotted vs. p H (squares), and is seen to reactions, 17,18no direct experimental evidence for its decrease sharply a t $H > 3.2. On the other existence is available. Accepting the experimental only results of Korosy and SzCkelylg we can conclude hand if, as assumed by previous k ( H B r O + H C ~ O ~ - )has an appreciable value, k * F that K B
