A New Indicator for pH 11 to 12 ROD O’CONNOR, WILLIAM ROSENBROOK, Jr., and GARY ANDERSON Department of Chemistry, Montana State College, Bozeman, Mont.
b Observation of the color change produced when an acetone solution of the potassium salt of the p-nitrophenylhydrazone of benzaldehyde was poured into water has led to elucidation of the chemical change involved. Benzaldehyde p-nitrophenylhydrazone exists in the yellow hydrazone form a t p H 11.3 or below, while at higher p H the proton is removed from the nitrogen, producing a red anion. The red color predominates in solutions of p H 1 1.7. Thus, a solution of benzaldehyde p-nitrophenylhydrazone or its potassium salt provides an indicator for the pH range 1 1 to 12.
>
T
of the pnitrophenylhydrazone of benzaldehyde, prepared by the method of Cuisa and Rastelli (S), was investigated as a possible water titrant since acetone solutions of the salt changed from red to yellow on addition of water. However, the reaction of the salt with water was found to be pH dependent, and not irreversible. The salt was then studied as a possible pH indicator, as the intense yellow color a t p H 11 or below contrasted sharply with the red color a t pH 12 and above. Concentrations of the salt necessary for sharp color change were low enough so that the pH of the solutions tested was not altered appreciably by addition of the indicator. HE POTASSIUM SALT
PROCEDURE
A solution of 6 grams per liter of the potassium salt of the p-nitrophenyl-
hydrazone of benzaldehyde in anhydrous methanol was used for the initial studies. One drop of this solution was required for optimum color production in 100 ml. of aqueous buffer solutions, and it produced less than 0.02 of a pH unit change in the reading. The Beckman Model H-2 pH meter was standardized using a pHydrion buffer of p H 8.0 (Micro Essential Laboratories, Brooklyn, N. Y.). Solutions of pH 9, 10, 11, 12, and 13 were prepared with aqueous KOH, and 100-ml. portions of each were tested with 1 drop of the indicator solution. The indicator produced a definite yellow color ,,A,( = 415 mp) a t pH 11 and below, and a definite red color (Amx = 515 mp) a t pH 12 and above. Potassium hydroxide solutions were prepared for the range 11.0 to 12.5, varying by 0.1 pH unit. The yellow color predominated to pH 11.3 and the red color above p H 11.7. The colors remained constant for about 30 minutes, but the red colors of solutions below pH 12.3 gradually faded with time as the hydrazone precipitated. As was expected, a solution of the p-nitrophenylhydrazone of benzaldehyde of equivalent concentration may be substituted for that of the potassium salt with identical results. A methanol solution, 15 grams per liter, of the N-methyl-pnitrophenylhydrazone of benzaldehyde, repared by the method of Cuisa and stelli (S), produced no red color until pH 14.
!iL
DISCUSSION
Apparently the colors obtained in basic solution of the p-nitrophenylhydrazone of benzaldehyde result, pri-
manly, from removal of the hydrogen from nitrogen by base (1, 4). Thus, the hydrazone form (I) is yellow in solution and the anion (11) is red. This production of the red color does not appear to involve reaction of the nitro-group itself with base ( 2 ) . This is shown by the absence of color formation by the N-methyl-p-nitrophenylhydrazone (111) until a thousandfold increase in pH is obtained. The pK. for the conversion of I to 11 has been calculated from the pH data to be a p proximately 11.5. H
CHa
LITERATURE CITED
(1) Bohlmann, F., Chem. Ber. 84, 490 (1951). (2) Bost, R. W., Nicholson, F., IND. ENQ.CHEM.,ANAL.ED. 7, 190 (1935). (3) Cuisa, R., Rastelli, G., Garz. Chim. Ital. 52, 11, 125 (1922). (4) Ragno, M., Ibid., 75, 193 (1945).
RECEIVED for review April 10, 1961. Accepted May 10, 1961.
Analysis of Products from the Electrolytic Oxidation of Acetate Ion in Acetonitrile SIR: As reported by Geske (8), the oxidation of acetate ion at platinum electrodes in acetonitrile solution occurs a t electrode potentials low enough so that no significant oxidation of the solvent interferes. This fact was discovered independently in these laboratories in 1960 and is being used to advantage in B chronopotentiometric study of the kinetics of the electron1282
ANALYTICAL CHEMISTRY
transfer process in the electro-oddation of carboxylate salts. To establish the principal reaction taking place and to make certain that acetonitrile serves as an inert solvent, an anode product analysis was undertaken for the electrolysis of a solution of tetrabutylmnmonium acetate in acetonitrile with tetraethylammonium perchlorate as the supporting electro-
lyte. Because of the more general interest of these analytical results, they are presented in this preliminary report. A discussion of the electrochemical kinetics of the oxidation of carboxylate salts will be presented later. EXPERIMENTAL
Reagents. The tetrabutylammonium acetate used was in the form of
double crystals containing 50 mole % ' acetic acid and was prepared by neutralizing aqueous tetrabutylammonium hydroxide with a large excess of acetic acid, evaporating to dryness a t 60" C.,and recrystallizing from benzene to obtain large white needles. These double crystals, unlike most tetraaikylammonium carboxylate salts, are not hygroscopic and are convenient to handle. Their melting range after several recrystallizations was 113-17' C., in agreement with the melting point of 116" reported for tetrabutylammonium acetate prepared by a similar procedure (6). If any loss of acetic acid took place below 113" C., i t was not accompanied by a visible change. The composition of the crystals was determined by acidimetric titration of the acetate with perchloric acid in anhydrous acetic acid, by titration of the acetic acid with sodium hydroxide in aqueous solution, and by controlledpotential coulometric oxidation of the acetate in an acetonitrile solution. The controlled - potential coulometric analysis, in agreement with the report of Geske (8), indicated that a oneelectron oxidation of the acetate takes place. The acetic acid present did not interfere, since acetic acid is not oxidized below the anodic background potential in acetonitrile solutions of tetraethylammonium perchlorate. The anode potential was held at +1.4 volts us. Hg/HgZS04/K$04 (sat. as.) for the controlled-potential coulometric analysis. Practical grade acetonitrile was purified by four or more distillations from phosphorus pentoxide, followed by one from potassium carbonate (16). It was stored in a siphon bottle fitted with a greaseless Teflon stopcock in the delivery tube and protected from the atmosphere Kith a drying tube. Tetraethylammonium perchlorate was prepared by the method of Kolthoff and Coetzee (IS) and was recrystallized from water five times. Apparatus and Procedure. The products of the electrolytic oxidation of acetate in acetonitrile were identified by sweeping them from the anode compartment of the electrolysis cell with a stream of helium and analyzing the effluent gas by adsorption chromatography on silica gel. A current of 40.0 ma. was provided by a constantcurrent source of Lingane's design (14) to a platinum gauze anode immersed in 20 ml. of acetonitrile to which 0.5 gram of tetrabutglammonium acetateacetic acid double crystals and 0.5 gram of tetraethylammonium perchlorate had been added. The electrolysis was carried out a t room temperature. The electrolysis cell is shown in Figure 1. The anode was separated from the cathode by a salt bridge consisting of a 4- to 6-em. length of 6-mm. (id.) Tygon tubing that had been rendered conductive by soaking it for several days in a solution of sodium perchlorate in acetonitrile and then passing current through it to draw more electrolyte into the polymer. The bridge resistance was of the order of 5 kilohms. During the electrolysis helium was
G D E
/
W Figure
1.
Electrolysis cell
A. Anode compartment 6. Platinum gauze anode C. Sat. aq. K&04 salt bridge to reference electrode D . Helium inlet E. Gas outlet F. Cork sealed with epoxy resin G. Agar-aq. K2S04 gel H. Cothode 1. Treated Tygon tubing 1. Solid glass plug K. Cathode compartment
bubbled into the anode compartment at a constant rate (*2Q/o) which was monitored by the pressure drop across a fritted-glass plug in the gas stream. Upon leaving the anode compartment, the gas stream was passed through a sulfuric acid bath to remove acetonitrile vapor and then to a sample-collection vessel. After the start of the electrolysis, 5 to 10 minutes were allowed for attainment of a steady state; gas samples were then collected at intervals and analyzed. Throughout the electrolysis the anode potential stayed between 1.3 and 1.5 volts us. Hg/ Hg,SOd/K&O, (sat. aq.), increasing with the depletion of acetate ion. In this potential range no appreciable oxidation of the solvent takes place: The background current in the absence of tetrabutylammonium acetate was 0.2 ma. a t an electrode potential of 1.5 volts and did not rise to 0.4 ma., or 1% pf the current used in the electrol IS, until a potential of 1.7 volts was reacgd. RESULTS AND DISCUSSION
The products observed and their yields were: carbon dioxide, 91 f 6%; 3%; and methane, 3.7 f ethane, 77 0.4%. The yields, based on theoretical
*
yields of 1 mole per faraday for carbon dioxide and methane and 0.5 mole per faraday for ethane, were calculated from the helium flow rate, the per cent composition of the effluent gas, and the current. The limits of error are 95% fiducial limits, assuming normal distribution of the sample population about the true value. The random error is attributed partly to the gas analysis and partly to fluctuations in cell operation, due to slight variations in temperature and helium flow rate during the electrolysis, which changed the steady-state partition of COZ, C2H6, and C& between the solution and the gas. Fourteen samples in all, from five different runs, were analyzed. The principal reaction that occurs is a simple Kolbe coupling. The formation of ethane as the main product has been observed in the electrolytic oxidation of acetate under suitable conditions in such vaned solvents as water (If?), ethylene glycol (9), acetic acid (11, 18), methanol (4, 17, 18), ethanol (18), and even fused salts (19). Considerable evidence supports the hypothesis of a free-radical mechanism for this reaction (19).
+
CHsCOO- = CHaCOO. eCHICOO. CHs. COa 2CH3. C~HB =I
+
An attempt was made to detect the presence of reaction intermediates by means of rapid chronopotentiometry with current reversal ('7). No reversecurrent wave was observed under conditions where any reducible species having a lifetime of 10 milliseconds or greater would have been detected. This was true even when the reverse-current was much smaller than the forward (anodic) current. This indicates either that the radicals produced a t the electrode have lifetimes shorter than 10 milliseconds or that they are not reduced before the cathodic background. Qualitatively similar results were obtained by Geske (8) using ordinary (slow) currentreversal chronopotentiometry . Methane is a commonly observed side product in the electrolytic oxidation of acetate. Its formation in aqueous solution has been shown by deuterium tracer studies to involve attack by methyl radicals on acetic acid or acetate ion; ethanol and methanol are preferentially attacked if added to the electrolysis solution (1). Therefore, in the presence of a large preponderance of acetonitrile one might reasonably expect the acetonitrile to serve as the major source of hydrogen atoms. The following reaction is thus regarded as the probable origin of the observed methane: CHI.
+ CH3CN = CHI + .CH2CN
The free radical formed in this reaction could couple with other radicals present VOL. 33, NO. 9, AUGUST 1961
1283
to form succinonitrile or propionitrile. No attempt waa made to detect these. The methane yield was not affected by adding enough glacial acetic acid to the electrolysis solution to raise the acetic acid concentration to 0.5M. The initial concentration resulting from the acetic acid present in the tetrabutylammonium acetate crystals was 0.07M. I t is not surprising that the measured yield of carbon dioxide is not 100% and that the sum of the yields of methane and ethane is not equal to the carbon dioxide yield, in view of the free-radical reaction mechanism. Simple free radicals, because of their instability, are notoriously nonselective in their attack on neighboring molecules. The literature shows that a number of side reactions could consume the radicals without resulting in the formation of the major products. Such side reactions could lead to the formation of esters (11, 12, 16, 17), diacyl peroxides and peroxy acids (8, 3, 6, IO), olefins (4, II), and alkyl perchlorates (IO).
SIR: The direct exposure technique of Wilzbach (8) for labeling organic compounds with tritium has received considerable attention recently because it is a simple and economical method of obtaining radioactive tracers for application in petroleum chemistry, biochemistry, and other similar fields. Means for accelerating the exchange of tritium for hydrogen have been reported more recently by Lemmon and coworkers (2), Westermark, Lindroth, and Enander (6),Dorfman and Wilzbach (I), and Mottlau (3). These advances have made the general application of tritium-labeled compounds as tracers even more attractive. Olefins, as a class of hydrocarbons, are of considerable interest in petroleum research. However, these compounds have proved difficult to label with tritium by the direct exposure technique! with or without acceleration. This is particularly true of terminally bonded mono-olefins because of saturation of the double bond with tritium, which produces a labeled paraffin rather than substitution of tritium for hydrogen (4,5). The Bureau of Mines has used two techniques with considerable success in preparing tritium-labeled terminally bonded mono-olefins. These techniques 1284
ANALYTICAL CHEMISTRY
ACKNOWLEDGMENT
The chromatographic analyses were performed with the kind assistance of R. G. Rinker and Y. L. Wang. LITERATURE CITED
.
(13) ’ J. ~,Kolthoff. I. hl.. Coetzee. J. F.. Am. Chem. *Soc.79; 870 (1957). (14) Lingane, J. J., ANAL. CHEM. 26, 1021 (1954). (15) Miiller, E., “Methoden der Or-
ganischen
Chemie (Houben-Weyl),”
Auflage 4, Band I, Teil2,828 (1959). (16) Petersen, J., 2. physik. Chem. 33, 99 (1900). ( l i j Salauze, J., Compl. rend. 180, 662 (1925). (18) Shukla, S. N., Walker, 0. J., Trans. Faraday SOC.28, 457 (1932). (19) Weedon, B. C. L., Quart. Revs. 6, 380 (1952). C. D. RUSSELL \ - - - - , -
(1) Clusius, K., Schanzer, W., 2. physik. Chem. 192A, 273 (1943). (2) Denina, E., Ferrero, G., de Paolini, F. S., Gazz. chim. ital. 68, 443 (1938). (3) Fichter, F., Buess, H., Helv. Chim. Acta 18, 445 (1935). (4) Fichter, F., Meyer, R. E., Ibid., 16, 1408 (1933). (5) Fichter, F., Zumbrunn, R., Ibid., 10, 869 (1927). (6) Fuoss, R. M., Cox, N. L., Kraus, C. A., Trans. Faraday Soc. 31,749 (1935). (7) Furlani, C., Morpurgo, G., J. Electroanal. Chem. 1,351 (1960). (8) Geske, D. H., Ibid., 1, 502 (1960). (9) Glasstone, S., Hickling, A., J. Chem. SOC.1936,820. (10) Hallie, G., Rec. trau. chim. 57, 152 (1938). (11) Hopfgartner, K., Maatsh. 32, 523 (1911). (12) Kolbe, H., Ann. Chem. Liebigs 69, 257 (1849).
are described below. The first is more efficient but less vemtile, in that it requires a specific starting material which may not be readily available. This method involves the saturation of one olefin bond in a terminally bonded diolefin with tritium. The second method combines the Wilzbach exchange labeling with a simple organic synthesis. APPARATUS
The only special apparatus required, aside from that proposed by Wilzbach or others for the direct exposure labeling procedure, is purification equipment that can separate a terminally bonded mono-olefin from its corresponding diolefin or paraffin. In the bureau laboratories, gas-liquid chromatography was used for this separation. It employed a 12-meter by ‘/l-inch column packed with Chromosorb P impregnated with propylene carbamate in a ratio of 5 to 1 by weight. A series such as n-hexane, 1-hexene, and 1,Shexadiene is readily separated with this apparatus. EXPERIMENTAL PROCEDURES A N D RESULTS
Approximately 0.34 gram of 1,5hexadiene was exposed to a tritium atmosphere of 3.7 curies for 1 hour with a potential of 15 kv. a t 3 ma. An accelerated labeling procedure similar to
FRED C. ANSON
Gates and C r e e Laboratories of Chemistry California Institute of Technology Pasadena, Calif. RECEIVEDfor review April 19, 1961. Accepted June 5 1961. Contribution 2699, Gates and &ellin Laboratories of Chemistry. Work supported b the U. S. Army Research Office under &rant No. DA-ORD-31-12461-G91 and by the National Science Foundation in the form of a fellowship held by CDR.
that described by Lemmon (2) was used. The sample incorporated 52 mc. of the tritium, giving a specific activity of 150 mc. per gram, as assayed by liquid scintillation counting (7). A portion of the exposed material then was processed by gas-liquid chromatography to separate the 1-hexene that had been produced. This 1-hexene had a specific activity of 1500 mc. per gram. None of the other components of the exposed sample was collected, but ionization chamber current measurements of the effluent from the GLC unit indicated that a small quantity of tritiated n-hexane and 1,5-hexadiene also could have been obtained. The yield of 1-hexene as indicated by integrating the area beneath the peaks from a gas-liquid thermal conductivity detector was 0.5%. A second sample of 1,Shexadiene (0.77 gram) was exposed a t room temperature to an atmosphere of tritium gas (4.1 curies) for 8 weeks, by the conventional Wilzbach technique (8). This time interval was selected only for convenience in scheduling laboratory work. A total of 1.29 curies of activity was incorporated into the sample, giving a specific activity of 1680 mc. per gram. This. was separated and purified as before. The material charged to the GLC unit represented 35 mc. of activity. The 1-hexene and 1,5-hexadiene peaks were trapped. The 1-hexene component had a total activity of 1.5 mc.
