HOMER ADKINS,R. M. ELOFSON, A. G. Rossow
3622
[COXTRIBUTION FROM THE
LABORATORY O F ORGANIC
AND
C. 6. ROBINSON
voi. 41
CHEMISTRY O F THE UNIVERSITY O F WISCONSIN]
The Oxidation Potentials of Aldehydes and Ketones BY HOMERADKINS,RICHARD M. EL OF SON,^^ A, G. ROSSOW’~ AND COWINC. ROBINSON The relative strengths of ninety carbonyl compounds as oxidizing agents have been ascertained by determining the concentrations a t equilibrium in systems of the type
latter against isopropyl 9,10-anthraquinone-2carboxylate. Limitations and Reliability of Data There are several factors which determine and A1 (0C4Hg-t)~ limit the reliability of the procedures on which the R’zCHOH R2CHOH Rz’CO. RpCO data in Table ‘I are based. The method for the determination of the concentration of the reacThe experimental observations on which the data tants a t equilibrium depends upon a t least one of in the table are based were made by F. W. COX,^ the two carbonyl compounds in a reaction mixture Robert H. B a k e ~ - , ~ ?G.~ ROSSOW,~ A. C. C. Robinson6 showing a depolarization potential on a polaroand Richard hl. E l o f ~ o n . The ~ last named has cal- graph. Thus direct comparisons must include culated all the values given herewith, and has an aldehyde, or a quinone, or a ketone with unconcluded7 that except for benzaldehyde, the saturation in the 2,3-position. The concentration values previously published are high by 33 m.v. of such carbonyl compounds can be determined so The oxidation potentials were calculated on accurately with a polarograph4 that if the anthe basis of the oxidation potentials of certain alytical determination were the only source of quinones. Three quinones of known oxidation error the calculated oxidation potential would in potentials8 were included in the series, i. e., general be reliable within 1 or 2 mv. g,lO-anthraquinone, 154 mv., l-chloroanthraThe greater the difference in oxidation potential quinone, 174 mv. and isopropyl 9,10-anthra- between the two carbonyl compounds in the requinone-2-carboxylate, 222 mv. One ketone, action mixture, the less accurate is the calculated 2,7-dichlorofluorenone, was equilibrated against oxidation potential. Small errors in the analytical each of the three quinones. The oxidation po- determination will cause relatively large errors tential, Eo, of 2,7-dichlorofluorenone was cal- in the calculated oxidation potential when the culated from the oxidation potential of a quinone difference between Eo and E,’ is more than about E: using the expression Eo = E: f RTINF 50 mv. This follows from the fact that if two In K where K is the equilibrium constant for carbonyl compounds A and B differ in potential the reaction given above. For 25’ the expression by 100 mv. then a t equilibrium there will be becomes Eo = E : 0.0296 log K . only 2% of the ketone A having the higher poThe oxidation potential of the ketone was found tential and 98y0 of B. If the difference in poto be 157, 156 or 160 mv. depending upon the tential is 80 mv. there will be 4.2y0 of A, for quinone used in comparison. Fluorenone and 60 mv. 8.870 -4, for 50 mv. 12.570 A, for 30 mv. benzophenone have also been equilibrated against 23.5y0 A, for 20 mv. 31.5% A and for 10 mv. each other and against 9,lO-anthraquinone. 40.5y0 A. In order to minimize errors arising These ketones were in turn equilibrated against from this source, fluorenone, 119 mv., 2,7-dichlorocyclohexanone and the latter against 2,7-dichloro- fluorenone, 157 mv., and isopropyl benzoylfluorenone. The values of Eo for 2,7-dichloro- formate, 282 mv., were preferred for the measurefluorenone, calculated from these series of com- ment of carbonyl compounds having potentials parisons, were 154 and 159 mv. The average in the three ranges. Cyclohexanone, 162 mv., of the five values for 2,7-dichlorofluorenone is has been used in equilibrations where the polaro137 mv. This figure has been used as the basis graphic curve for 2,7-dichlorofluorenone was obfor calculating the oxidation potential of most of scured by that of the other carbonyl compound the carbonyl compounds listed in the Table I, in the reaction mixture. Benzaldehyde, 192 since 2,7-dichlorofluorenone was equilibrated di- mv., and trimethylacetaldehyde, 21 1 mv., were rectly or indirectly against all the other ketones also useful in determining the oxidation potential and aldehydes except chloral and the alpha-keto of other aldehydes in the range of 186 to 270 esters. These keto esters and chloral were equi- mv. librated against isopropyl benzoylformate and the The most serious limitation upon the accurate 11j IYisconsin .-\lunini I
