J. E. LUVALLE AND A. WBISSBBWBR
1576
[ COMXUNICATION NO. 1138FROM THE KODAK
VOl. 69
LABORATOREUS]
Odddon Processes. XlX.' Quinom Catdyds in ths Hydroquinones
of
BY J. E. LUVALLEAND A. WEISSBERGER The autoxidation of durohydroquinone is catalyzed by duroquinone, which is formed in the reaction or added to the Similar autocatalysis and catalysis take place with pseudocumohydroquinone, but here the reaction rate does not always increase in proportion to the amount of quinone added.a If pseudo-cumoquinone is added or formed in higher concentrations, the catalytic effect of additional quinone appears to decrease until it becomes negligible, i e . , it reaches saturation. The expression "quinone catalysis" is used as a brief term for the designation of the catalysis of the oxidation of a compound, R, by the bivalently oxidized product, T, op by a 'dmilar compound. It is a special case of the equilibrium R + T z 2 S
(1)
sium cyanide and potassium thiocyanate, respectively, in order to inhibit heavy metal catalysis.a A very useful and clean method for the reduction of quinones is the treatment with hyclrPbine. 2,S-Xyloitydmquinone.LEsstatap IC.&+ Co. p-xyloquinone (18.6 g.; 0.1 mole) wp9 dsdved m 100 ml. of hot 95% alcohol, and the solution c d e d to 50". Successive small portions of a mixture of 85% hydrazine hydrate solution with an equal volume of water wet(? then added until the solution was almost colorks~. Nine to ten ml. of the hydrazine solution was required. The temperature was kept a t 60" during the addition. The alcohol solution was then heated to bailing, 100 ml. 2f boiling water was added, and the mixture cooled to 0 The hydroquinone was filtered off, washed with water anddriedintheair; yieldwas 11.4g. (81%). Recrystallized from 40% alcohol, it melted at 213-216". Pseudo-camohydroquinons.@-Pseudo-cumohydmquinonewas prepared from pseudo-cumenol-6 by the method of Smith'gnd co-workers. The over-all yield of quinone was 76%, based on the pseudo-cumenol-6 used. The pseudo-cumoquinone was reduced to the hydroquinone by the hydraaine hydrate method; yield was 76%. After sublimation, folkmed by two r e c r y s t a b tions from toluene, the pseudo-cumahydroquhone melted at 169-170".
.
Quinone catalysis and its saturation at rather low quinone concentraoiOns appear to be of fundamental importance in the autoxidation of enediols of hydmquinones and of related immp~unds.~J~~ Calculations The quinone acts as a catalyst through the In the preceding papers of this series, it was formation of highly reactive semiquinones'.* The frequently found that the total a m m t of oxygen satulrution of the quinone catalysis is explained absorbed was greater or smaller tban that calcuw i t h the attainment of a steady state by the semi- lated for qUinone.1.' The attsinment of Gturation by the R 0 2 +T Hz02 (2) quinone catalysis should depend upon the ratio of (T)/(R)l rather than upon the concentration of where R is the substrate, e.g., hydroquinone, and T. Dependency on the latter had been surmised T is the oxidation product, e.g., quinone. The by James, &dl and Weissberger.8 They also deviations axe caused by side reactions and subfound some indications, but no definite proof, that sequent reactions, respectively. Fop instance, the xylohydmquinones might exhibit an auto- reactions of R and of the semiquinone with hydrocatalytic period. Both observations needed gen peroxide diminish the total oxygen uptake, checking, and the pFsent paper gives new data whereas reactions of the quinone, or of a reaction on pseudo-cumohydroquinone, 2,5dmethyl- product of the quinone, with oxygen increase the hydroquinone, toluhydroquinone, and hydro- oxygen uptake. The choice of the proper volume quinone, and a discussion which is based on the of absorbed oxygen at t m is therefore often &cult and may even be arbitrary. Such a choice is preceding paper. avoided by Guggenheim's methodE€or the analyExperimental sis of rate measurements. In this method, two Materials and Techniques sets of data are recorded, a set, X i , a t time, ti, and Reagents, unless noted otherwise, and techniques for another X'; at time, t'i, where t'i = ti T. The the measurement of oxygen absorption and pH were accuracy of the method increases with 7 . Values identical with those described in preceding papers of this Of log (X'i Xi) are then plotted against t i . series.%&.@ The temperature of the kinetic experiments was 19.97 0.02". Buffers were present in a concentra- The resulting curve is a straight line for firsttion of 0.20 M , and the solutions were 0.004 M in potas- order reactions. The rate constant, k, is proportional to the slope, s, of this b e , b = -2.50s. (1) Part XVIII, LuValle and Weissberger,THIS JOURNAL, 6S, 1667 It may be recalled that in terms of the tra(1947). ditional way of calculating rate constants of (2) James and Weissberger, ibid., 60, 98 (1938). (3) Jpmu,Sndl and Wdasberger, ibid., 60, 2084 (1938). first-order reactions, k = 2.30s: where s' is the (4) (a) James and Weissatper, ibid., 6l, 442 (1939); (b) Weinsslope of the plot of log (a x ) against t, a being berg- and Tbmaas, ibid., (w, 1661 (1942); (c) Weissbtl.ger, Thomas,
+
+
+
-
f
-
and LuVrIk, &d., 66, 1489 (1943); (d) Weimberger, LuValle and Tkomss, a., 68, 1984 (1943). (6) Kornfdd and Weissberger, ibjd., 61, 360 (1939). (6) Wdedbagr+, Mainz and Strasser, &r., 6S, 1942 (1929).
Mr.E. C. Armstrong. (7) Smith and -workers, J . Org. C h . , 4, 318 (1939). (8) Guggcnhcim. Pit& Mag.,0, 638 (1926). (6a) Fpr these preparations, we are indebted to
July, 1947
QUINONE
CATALYSIS m AUTOXIDATION OF HYDROQUINONES
1577
0.6
0.4
0.2 0.0
4
.
0.2
I
3
0.6 0.4
0.2 0.0
-0.2
2
4
6
8 1 0 1 2
f , minutes,
2
4
6
8 1 0 1 2
t , minutes.
8.33 * 0.02; +curno. Fig. l.-Guggenheim curves for pseudo-cumohydroquinone (3.33 X lo-* M ) in oxygen; quinone added: (a) 0.00; (b) 0.416 X lo-* M; (c) 1.67 X 10-8 M; (d) 3.33 X lo-* M. The point at which R = T is indicated by +, except in cases where Ro = To.
value of x / a for the time at which k , Le., the slope of the Guggenheim curve, becomes constant (Fig. 1). The d u e s of k are given in column 9. The Guggenheim curves. for several initial values of T are given in Figs. 1 and 2. The experiments of Fig. 1were made under oxygen, those of Fig. 2 were carried out under air. The data of Table VI of James, Snell and Weisbergera are given in Fig. 3. The data show that under oxygen the saturation of the quinone catalysis occurs in every case when R = T, irrespective of PH and of the initial values (Z/U)R-T = (a - T0)/20 ’ (3) of (R) and (T). Within the experimental error, where TOis the initial amount of T present in the the rate constant a t saturation, i e . , the maximum reaction mixture. This equation is valid as long and constant value of k, is independent of the as the semiquinone formation constant is small. initial concentrations of the quinone and the hydroquinone, The constant rate is first-order Results with respect to the initial concentration of pseudoPseudo-cumohydroquinone.-The measure- cumohydroquinone, and very closely first-order ments were made in 20% ethanol, for reasons of with respect to the concentration of oxygen solubility. The rate data are summarized in throughout the reaction. Under air (Fig. 3), the Table I. value of k becomes constant at 15, 15 and 19% Colurnms 1 and 2 give the initial and the final completion of the reaction a t pH 7.40, 7.64 and PH,respectively. The oxygen presswe of the 7-84, respectively, Le., the steady state is estabexprimmts is Wed in c o h 3. The initial lished earlier in air than in oxygen. concentration of p s m d w d r o q u i n o n e , the 2,5-Xylohydrq&one.- James, Snell and amount of pexdo-amoquinonttinone added, and the Weissberger2 noted a slight induction period in ratio of both f%Brqmm& at the beginnkg of the several experiments with xytohydroquinones withreaction am given in rolumns 4, 5 and 6, respec- out establishing it beyond doubt. The new rate tively. cohuntl 7 lists the value of %/a for the data for one of the xylohydroquinones, 2,5time at which R = T, and colum 8 gives the dimethylhydroquinone, were obtained in 33y0 the total substrate and x the reacted portion. Guggenheim’s m e is analogous to the customary log (a - x ) us. time plot, and deviations from firstorder reactions affect the shape of both curves in the same way. The f r & h of the reaction cowtfilded is calculated as %/a,where u is the initial number of moles of R and x is the number of moles of oxygen absorbed. The fraction of the reaction completed for the time ut whkh R = T
J. E.LUVALLEAND A. WEISSBERGER
1578
$-~UMOHYDROQUINONE IN 20% 4 2 3 R motesjliter Pos, 9H
1
?E!
initial
final
mm.
8.30 8.23 8.15 8.15 7.95 8.00 8.37 8.36 8.32 8.31 8.32 8.31 8.31 8.30 8 32 8.33 8.32 8.26 8.27 8.25 8.22 7.68 7.70 7.72 7.70 7.70'
8.33 8.27 8.20 8.18 8.00 7.93 8.37 8.36 8.32 8.30 8.32 8.30 8.32 8.82 8.32 8.37 8.32 8.28 8.31 8.29 8.23 7.76 7.72 7.73 7.72 7.73
733.7 726.6 741.9 742.5 740.7 737.1 737.4 736.8 734.0 725.8 735.7 728.1 727.3 154.0 154.0 736.5 726.6 742.2 743.0 743.3 736.5 732.5 734.3 732.5 729.9 748.0
x 1oz 3.33 3.33 3.33 3.33 3.33 3..33 3.33 3.33 3.33 3.33 3.33 1.67 1.67 3.33 3.33 3.33 3.33 1.67 1.67 1.67 1.67 3.33 3.33 3.33 3.33 3.33
VOl. 69
TABLE I ET~~NO POTASSIUM L, PYROPHOSPHATE B ~ E R 6
x
t
6
T,
108
.. 3.33
..
3.33
.. 3.33
..
1.67
..
R T Cahd.
k = const. Obsd.
0.0 1.0 0.0 1.0 0.0 1.0 0.0 .5
0.50 .00 .50 .00 .50 .00 .50 .25
0.5
.50
.5
.o
..
3.33
..
3.33 1.67 0.416 1.67 0.833 3.42 3.33 3.33 6.67 1.67 0.832
0.50 ,125
.o
.50 2.05 2.00 1.0 2.0 0.5 .25
..
.o
9
.5 . .25
.00 .00 .50
.o .o
.50 .00 .25 .44 .00 .25
.17 .00 .25 .41
.5
.o
.o
.2
.o .o .o
.o
.25 .375 .50
.o
8
.o
.00 .00 .00
0.206 .175
.3 .4 .5
.5
.M)
k
min. -1
1109 .122 .0598 .0562 .212 .200 .194 .215 .233 ,156 ,216 ,0518 .0506 .218 ,217 .208 ,195 ,244 ,256 ,0346 .0322 ,0352 ,0324 .0302
.5
.o
.00
9
(da)
h1tlaIly
1.0 1.5 0.0 2.0 0.0 1.00
3.33 5.00
8
(%/a)
T!R
moles/liter
TABLE I1 2,5-xYLOEYDROQUINONE IN 33% ETHANOL 1
3
2
Initial
Final
Buffer
9.33 9.30 9.30 9.31 9.32 9.33 9.34 8.89 8.88 8.92 7.95 8.39 8.46 8.68 8.75
9.39 9.39 9.37 9.36 9.32 9.34 9.34 8.88 8.90 8.94 7.98 8.39 8.50 8.60 8.77
Na-p-phenol sulfonate Na-p-phenol sulfonate Na-+phenol sulfonate Na-$-phenol sulfonate Na-p-phenol sulfonate Na-+phenol sulfonate Na-p-phenol sulfonate Na-p-phenol sulfonate Na-p-phenol sulfonate Na-p-phenol sulfonate KeHPOi &PPI
&P,G Na-p-phenol sulfonate Na-p-phenol sulfonate
4
5
6
R.
T.
mm.
moles/ liter x loa
moles/ liter
152.2 155.0 154.5 152.0 151.0 156.0 154.2 153.0 162.8 156.4 737.0 733.0 733.0 730.2 731.1
3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33 3.33
Po,,
ethanol solution. They are summarized in Table 11, which corresponds to Table I. In air, no autacataXySis was detected. The Guggenheim curves for three runs, in which the xyloquinone cmcentration was varied axe given in Fig. 4a. Fur the start and the larger part of the reactions, the data are fitted by a straight line. For the last 25% of the reactions, the rate drops below that expected for a fust-order proc-
x
108
7
T!R
min. -1
0.0
0.00 .00 . 00
0.265 .276 .274 ,286 ,274 ,285 .272 ,0425 .w5 .0425 ,0059 .042 .037
.o
0.416 416
.13 .13 .25 .25 .50 .00 .12 .25
.834 I834 1.67 0.416 .834
k
R = const.
... I
(da)
initially
...
.o
...
0.0
3.33
1.0
.o
1.0
.00
.o
.o .o .o .o .o 19 22 0.0
17 0.0
.lll
,115
ess. This drop is much more pronounced than for pseudo-cumohydrquinone. Experiments under mygm revealed autocatalytic periods extending through 15 to 20% of the reaction. When quinone is added to make Ro = To,the reactions are first-order from the start. Guggenheim curves are given for s e d values of p H in Figs. 4b, 4c and 4d. In the latter part of these reactions, a departure from first-order
July, 1947
1579
QUINONE CATALYSIS IN AUTOXXDATION OF HYDROQUINONES 20
0.4 0.2
0.0
r: 0.2
s I
5
0.6
3
0.4 O0.06 'O81
0.2
4
0.04 -
0.0 -0.2 20 30 40 Time, minutes. Fig. 2.4uggenheim curves for pseudo-cmohydroquinone (3.33 x 10-8 M) in air; j Z 8.33 * 0.02; cumoquinme added: (a) 0.00; (b) 3.33 X M. 0.2. -c mdicates polnt at whkh T/R 10
0.16
+
takes place, which increases with increase in pH (Figs.4c and 4d). The departure froan jirst-order in the latta portion of the ceaetkm is accompanied by an incneese in the W oxygen absorbed beyond the theoretical a.p@ke, It is probab€y =sed by the removal of T by rcmceiOn with oxygen,hydro-
0.12
1 I
oxygen
/---
0.08
0.04
1
/ I
4 I
I
,
4
3c
0.1 0.2 0.3 0.4 Fraction of reaction completed. Fig. 3.-k values against fraction of reaction completed for $-cumohydroquinone: (a) PH 7.40; (b) PH
7.84; (c) PH 7.84.
data of,James, Snell and they did not reveal any The maations follow a
and of t ~ l u ~ d r and o ~tolaquinone ~ o ~ when mixed become htskntaneodv amber: 2.5xyklhydroq@one and 2,&xyb6hone form am& color sonnewBat slower thafl the mono-
an
methyl compounds, and pseudo-cumohydroquinone and pseudo-cumoquinone form an amber color still rapidly but mu& slower than the dimethyl compounds. Durohydrquiuone and duroquinone show no color change upon mixing. When solutions of varying molar ratio of the respective quinones a d hydroquinones are mixed, and the total concentration is kept constant, the most intense color is obtained when the ratio is unity. If the volume of the solutions is increased, the intensity of the color decreases more rapidly than would follow from Beer's law, indicating that the color arises from a dimer rather than from a monomeric radical." According to Michaelis,lo durohydroquinone and duroquinone do not form a dimer but readily (91 Michaelis and Schubcrt, Chrm. Rev..S¶. 437 (1938). (10) Michaelis, Schubcrt, R a k , Kuck and Granick, T H I S JOURNAL.
w, 1678 (1938).
J. E.LUVALLEAND A. WEISSBERGER
1580
Vol. 69
+0.6 +0.4 $0.2
0.0
7 -0.2
h
+0.6
3 +0.4 +0.2
0.0
-0.2
1 .40 4 8 12 16 20 24 Time, minutes. Fig. 4 . 4 ~ n h e i m curves for 2,5-dimethylhydroquinone(3.33 X IO-a M): (a) in ak; $H 830 * 0.02: -, 295methylquinone = 0.00; -eO-,3,i$dimethylquinone = 0.416 X lo-* M ; -@-@-, 2.5-dimethylquinone = 0.832 X '01 M. In oxygen: (b) pH 7.95; (c) pH 8.39; (d) pH 8.64. 10
20
30
form the semiquinone. The available evidence indicates that dimer formation is hiddered by me$hylation of the aromatic nucleus and increases in che d e s , drouhydrquinone, pseudo-cumohydroquinone, xylohydroquinones, toluhydroquinone and hydroquitlone.
0.9 0.8
h
r; 0.7 -
ci
I 0.6
-
I
d 0.5 *
' M
0.4
0.3 0.0 0.1
followed by a gradual increase in density. A s p e d r o p h o t e inve&iga* &these systems therefote requires very rapid r n a k ~ ~or~a ~ ~ ~ t s kinetic W y whkb permtts exttqmbtion of the r c d t s to the h e of mixing.11 The gmdual &mge appears to be connected substance with the reactivity of quimnede. is unstable in Iydmqkontaihing solvents k the dark,l* abd even more 50 in the ligzp.t, The instability is partidady great in a.lk&ae s&?ion. Quinone re8et9 rapidly with hflrmn peroxide to f o r a hydroxyqtiin.oae.l* Substitution' w3th methyl in the ring has a great stabilizing effect on uinone. Dumis stable even in strong y alkdixe SO~Utions,2*sand dmohydroquindne was therefore chosen for the initial investiggtions of the autoxidation of hydroquinones.2 The cataIytic activity of paeudo-ctuncquhone is not seriously impaired by standing a t pH 7 to 8.5 for the length of time required far the autoxidation of the hydroquinone. Solutions of pseudo-cumohydroquinone which have undetgoae autoxidation below PH 8.5 remain light yellow for some time after
-
I
\
10 15 20 Time, minutes. Fig. Ei.-Guggenheim c u ~ for e hydroquinone at PH 8.41. 5
Some s p e ~ experiments ~ ~ near~ the ~ (11)cA h w maebb of t& trpd deacribad chrac+ [ J . Franklin m, 466 (lQ4O)l which ahodd k dhbk for than measureneutral point showed that the instantaneous Ins;., ments is under conetiuetion in time Laboratories. darkening of the solutions of hydroquinone(12) Braude, J . Cknr. Soc., 44M flQ45). quinone and toluhydroquinonetoluquinone is (18) Reintiers a d Dingenune Rrc. em. drdm., 68,209 (19%).
QUINONE CATALYSIS IN AUTOXIDATION OF HYDROQUINONES
July, 1947
completion of the experiment. Most of the hydrogen peroxide formed is present at the end of the autoxidatian, and the total oxygen uptake is nearly the theoreticaL8 Thus, pseudo-cumouinme is relatively stable. The stability of -A, maPlOmetsOlQudtlOPb and quinone decreases. With decreasing number of methyl groups, the solutions discolor increasingly during the autoxidation, less hydrogen peroxide is present at the end of the reactions, and the observed uptake of oxygen increases beyond the theoretical value. Discussion If steric hindrance retards or prevents dimer formation, it can be presumed that it also retards or prevents formation of the complex, T.S. The autoxidation af durohydroquinone2 proceeds by the mechanism of Class I-A-3 of Part XVIII.' This mechanism is the limiting case of Class I-E for a stable quinone with no tendency to form a dimer or a complex with the semiquinone. Pseudo-cumoquinone is fairly stable under the conditions of the experiment, but forms a quinhydrone with pseudo-cumohydroquinone. In the beginning of the autocatalytic period, the rate of autoxidation of pseudo-cumohydroquinone is independent of the oxygen concentration. The rate in the region of saturation of the quinone catalysis is somewhat less +in first-order with respect to oxygen. The end of the autocatalytic period is determined by the ratio of (T)/(R). The length of the autocatalytic period in air is about onethird as long as in oxygen. These properties place pseudo-cumohydroquinone in Class I-El with k7 (T) %(02). The Guggenheim curves of Fig. 1 are of the type of the log (a - x ) curves of Fig. 2 of the preceding paper.' Equations (18) and (%a) of the preceding paper explain why the steady-state rate and the length of the autocatalytic period are not strictly proportional to the oxygen pressure. Homologs of hydroquinone which are less protected by methyl groups than pseudo-cumohydroquinone have a greater tendency to form quinhydrone, and the respective quinones are less stable than pseudo-cumoquinone. An irreversible loss of T takes place during the autoxidation of xylohydroquinone, t&ahydroquinone and hydroquinone; and the reactions proceed according to Class I-E of the preceding paper' [K.r(T)2LKn(O2)]. 2,5-Dirmethylhydroquinone shows in oxygen an unmistakable autocatalytic period extending over approximately 20% of the total reaction. In air, no autocatalysis could be observed. This behavior can be expected, in view of the greater reactivity of the xyloquinone as compared with the trimethylquinone. It parallels the dependency of the autocatalytic period on the oxygen presstlre, which was discussed above. Xylohydroquinone is in its behavior intermediate between the trimethyl compound and toluhydroquinone [MT)2k7(0$1.
&m
-
1581
Toluhydroquinone and hydroqpbaae do not pennit the direct observation of autocatglyais under an conditions so f [b(T)>RrrOr)l. This is d y expected in view of the b a d of &e autocatalytic period with tion of the ring and is to &-&-order dependency with respect to oxygen. This extrapolation is suppo~edby an estimate of the extent of the autocatalytic periods of the various hydroquinones on *e basis of the steadystate autoxidation rates. It has been shorn that the steady-state rate is proportional to the semiquinone concentration. Inasmuch as (S) is proportional to the product, (T) and (S) a t the end of the autocatalytic period is proportional to the length of the latter, R in the steady state is proportionat to &e length of the autocatalytic period, provided ( R ' ) is the same for all of the hydroquinones, ;.e., the p K ' s are identical. The pK's for the several hydroquinones have not been measured. The acidity of phenol decreases somewhat w i t h increasing methylation, but the change from phenol to pseudo-cumenol is less than one order of magnitude in $KO1* The pK's of hydroquinone and the various methylhydroquinones are probably also of the same order of magnitude. From the values of James, Snell and Weissberger* and the extent of the autocatalytic period for pseudo-cumohydroquinone (50% completion of the reaction), we calculate the autocatalytic periods in Table 111. (R-),1*21*1b
TABLE I11 EXWNTOF TEE AUTOCATALYTTC PERIOD UNDER OXYGEN Compound
Relative rates"
Hydroquinone 1.00 Toluhydroquinoae 3.9 2,3-Xylohydroqdnone 10.5 2,6-Xylohydroqninone 18.2 2,6-Xylohydroquinone 17.0 \k-Cumohydroquinone 31.0 a Data of James, Snell and Weissberger.'
Extent of autocatalytic period (calcd.)
%
1.6 6 16 29 27 50
The assumption that (R-)is the same for all of the hydroquinones w ill tend to make the values somewhat low. However, the estimated value for the length of the autocatalytic period of 2,5-dimethylhydroquinone is about 40% bigher than the experimental value. This indicates that factors are of importance, which are neglected in the assumption that (T) at the end of the autocatalytic period is proportional to k. The existence of such factors is obvious if it is realized that the assumed propoponality presupposes that the energies of activation and the entropies of dimer formation, (14) (a) Bader, Z. physik. Chcm., 8, 290 (1890); (b) Hantzsch and Farmu, B e . , 119,3088,8089,9101(1899); (c) Kuhn and Waeserman, E&. Chim. Aclu, 11, 38 (1928): (d) Swnrrmback and Egli, ibid., 17 1176 (1934); (e) Summabell, J . Chcm. Soc., 996 (1934).
ILSEL. HOCHHAUSER AND HENRYTAUBE
1582
semiquinone formation and semiquinone autoxidation we identical for hydroquinone and for the v d m me"tfiy1 kydroquinones. In analogy with the h&gs far 2,5-&methylhydroquinone, the extrapdatzti d u e s for the leflgth of the autocatalytic periott for toluhydroquinone and for hydroquinone may therefore be too high. summary 1. The autoxidation of pseudo-curnohydroquinone, 2,5-dimethylhydroquinone, toluhydro-
Vol. 69
quinone and hydroquinone is investigated further. 2. The saturation of the quinone catalysis in the autoxidation of pseudo-cumahydroquinone is explained. 3. The suggestion is corroborated that the apparent indifference of the autoaidstion of hydroquinone and its lower homologs to the presence of the respective quinones is c a d by the fact that the quinone catalysis reaches saturation at rather low concentrations of the quinones. ROCHESTER 4, NEWYORKRECEIVED FEBRUARY 20, 1947
[ CONTIUBUTION FROM DEPARTMENT OF CHEMISTRY, CORNELL UNIVERSITY]
The Direct and Sensitized Photochemically Induced Reaction of Chlorine a d O d i c Acid. Comparison with the Chemically Induced Reaction BY ILSEL. HOCHHAUSER AND HENRY TAUBE Previous studies of the reaction
+
+
+
H ~ C C ~ OClr ~ +2H' 2C12C02 have shown that the spontaneous reaction' pro-
ceeds slowly in 2M hydrochloric acid and that ferrous ion3 acts as an inducing agent by the intiation of a chain reaction. While the extensive data on the ferrous ion induced reaction have been completely interpreted by a mechanism involving atomic chlorine and an oxalate free radical, it has appeared to us desirable to seek a more defmite conclusion as to the identity of the active intermediates. In the work reported here, this has been attempted by studying the kinetics of the reaction above induced by light, and comparing for the chemically and photo induced reactions such properties of the reaction paths as specific rate, energy of activation and the eflect of inhibitors. It has thus been established with reasonable certainty @at the same intermediates operate in both systems; from a consideration of the methods of production and of the properties of the intermediaries, it likewise seems certain that they are, indeed, mainly atomic chlorinea and the oxalate free r a d i ~ a l . ~ A principal goal in these investigations is to establish absolute values for the various specific reaction rate ratios of the intermediate substances (e. g., kn/kl/s, k&s'/2, see Discussion) not only because the values themselves possess interest, but also because they may be regarded as properties of the intermediates useful in identifying them in other systems. It wr"ll be evident that to calculate the absolute values of such ratios from the expehental data, the actual rate of intermediate production must be known. This rate may (1) Gri5th and McKeown, Trass. Fareday Soc., P I , 618 (b32). (2) Taube, THIS JOURNAL. 68, 611 (1946). (3) Cf.ref. 2, fmtnote 2. (4) The oulate free radical will be represented an &or- rathet than HCIOI. T h e assumption that H&or is a strong add seema reasonable from a consideration of ita electronic configurption.
be considered as the product of two factors: the rate at which the inducing agent, li@t quanta or chemical substance, is introduced, and the efficiency of the primary intermediate production process. The fvst factor can easily be measured, the second can in special cases be i n f d . If, for example, the effects produced by wideiy Mering means for intermediate produetion, such as light and a reducing agent, agree quantitatively (a. e., one quantum behg equivdent in its effect to two ferrous ions) it would seem fairly safe to conclude that the efficiency is unity for each type of process. While the possibility of accidental agreement on a value of the efficiency less than unity exists, it seems remote, and can be eliminated almost completely if a third method of intermediate production gives quantitative agreement. These arguments have been applied in reaching some tentative conclusions about the efficiency of the three primary processes which have been compared.
Experimental Apparatus.-As light source for most experiments, a C-H-1 Mazda a. c. mercury vapor lamp was used; for a few runs, a Hanovia 550W d. c. lamp was employed. The energy input was checked with a voltmeter and, in some series, the constancy of radiant output was checked by means of photronic cells. The variation in voltage never exceeded 3% during a continuous series of experiments. A lens and a round-bottom flask served to concehtrate the beam on the reaction cell. Directly in front of the cell were placed light Mters and a shutter. To obtain substantially monochromatic ligbt, Btera were used: 3650 d Corning Glass filter no. 584; 4047 A., Wratten filter no. 36; 4360 d., Corning Glass tiltem no. 585 and no. 038. The.reaction cell w w eylinMcd in shape and was painted black, except for the front and back windows, which were made of
