pK1 values of 8-hydroxyquinolines. The pK1 value is given by pKL = 5.25 - 5.90 X U where Z U represents the summation of u substituent constants obtained by the methNDd of Dewar and Grisdale (15). For quinoline-type compounds, Perrin has included a u constant of 0.06 for the benzenoid ring system. We have used this method to estimate pK1values of the compounds studied here; results are shown in Table VI. Calculated values are consistently lower than the experimental results. Careful examination of Perrin's own data also shows similar discrepancies for some 8-quinolinols We, therefore, conclude that the contributions due to the extra ring fused to the pyridine nucleus and to the OH group have been overestimated. The method does, however, predict closely related pKl's for the halosubstituted derivatives. Assigning a value of 0.0 for the u constant of the benzene ring brings the pKr values closer to the experimental resulls (Table VI).
__
Badger and Moritz (16) have shown that there is a linear relationship between the -OH stretching frequency of substituted 8-hydroxyquinolines and the Hammett u constants. Since Perrin's work indicates a linear relationship between u and pK1, we would expect that pK1 should show a linear relationship to Y O E . Such a relationship has been observed (17) for 8-hydroxyquinoline and some of its aza analogs. No correlation is, however, observed for the substituted 8-hydroxyquinolines, as is evident from the data of Table VII. In summary, thermodynamic pK values have been obtained for 5-bromo-, 5-chloro-, and 5-iodo-8-hydroxyquinolinein aqueous solution at 25 O C. Results have been compared with previous work and with suggested methods of estimation of pK values from substituent constants. Molecular orbital calculations have been carried out in order to compare electron distribution and n-electron energy differences with the observed values. RECEIVED for review July 8, 1966. Accepted Dec. 13, 1966.
(15) M. J. S . Dewar and P. J. Grisdale, J. Am. Chem. SOC.,84,3548
(1962).
(16) G.M.Badger and A. G. Moritz, J. Chem. SOC.,1958, 3437.
(17) F. J. C. Rossotti and H. S.Rossotti, J. Chem. SOC.,1958,1304.
Spectrophotometric Determination of Silver with Crystal Violet James J. Markham' Department of Inorganic and Structural Chemistry, The University, k e d s 2 , England
THELITERATURE conce wing the use of triphenylmethane dyes has been reported on by Hedrick and Berger ( I ) ; these investigators have described conditions for extracting perchlorate, periodate, and alkylbenzenesulfonate. Of the substances investigated they found crystal violet the best reagent and benzene the best solvent for this purpose. Both sulfite and cyanide were found to interfere with the reaction by decolorizing or reacting with the dye. In spite of the reaction between crystal violet and cyanide it, is possible to make a determination of silver in the concentration range to 10-4M by using crystal violet to extract the silver cyanide complex into benzene provided there is strict attention to procedure. EiXPERIMENTAL
Silver nitrate was stored as a 10-3M stock solution. Solutions of 10-4M and 10-5M were prepared by dilution shortly before use. Procedure. The solution which is 10-6 to 10-4M in silver is made 10-3M in sodium cyanide and 0.1M in sodium hydroxide. Exactly 10.0 ml of this solution are added to a glass stoppered tube together with 10.0 ml of benzene. Next 1.00 ml of approximately 10-3M crystal violet is added. The tube is stoppered immediately and shaken vigorously for exactly 15 seconds. The stopper is removed and the tube centrifuged for 60 seconds. The supernatant benzene layer is removed at once with a pipet and stored in a clean dry glass stoppered vessel. Between 30 and 40 minutes after removing the benzene layer, its absorbance at 600 mp is measured, RESULTS AND DISCUSSION
Apparatus. A Unicam SP500 spectrophotometer and
either 2.00-cm or 0.50-cm light path cells were used for all absorbance measurements. Borosilicate glass tubes approximately 2.5 mm X 100 mm with rounded bottoms and glass stoppered tops were used for extraction. Reagents. Crystal Violet, C.I. 42555, was obtained from British Drug Houses Ltd. All other chemicals were reagent grade. An approximately 10-3M solution of crystal violet was prepared by dissolving 82.2 mg in water and diluting to 200 ml.
The absorption spectrum of the crystal violet silver cyanide complex shows a maxima at 600 mp. A calibration curve should be prepared. A complication is that the reagent is being altered by reacting with the cyanide which is necessary for the formation of the silver cyanide anion. Nevertheless by strict attention to procedure it is possible to get reproducible results.
Present address, Chemistry Department, Villanova University, Villanova, Pa. 19085
Because of the great stability of the hydroxyargentocyanide
ANALYSIS FOR BOUND SILVER
~
(1) C . E. Hedrick and B. A. Berger, ANAL.CHEM., 38,791 (1966). VOL 39, NO. 2, FEBRUARY 1967
241
ion (2), it seems reasonable to suspect that an alkaline cyanide solution might be able to remove silver from less stable complexes of silver. If such is the case, it should be of advantage in the analysis of trace quantities or organic bound silver in that preliminary wet ashing or dry ashing could be avoided. To test this hypothesis a mild silver protein, “Argyrol,” was analyzed gravimetrically and found to contain 21.7 % silver. An aqueous solution of this material was prepared to give
(2) I. M. Kolthoff and J. T. Stock, J. Am. Chem. Soc., 78, 2081 (1956).
samples containing 10.9 pg of silver in 10.00 ml of solution which was also 10-3M in sodium cyanide and 0.1M in sodium hydroxide. Analysis was then carried out according to the standardization procedure given above. Triplicate results gave 9.94, 10.3, and 9.78 pg of silver, respectively, for an average value of 10.0 pg corresponding to a precision of 2 % and an error of 8%.
RECEIVED for review August 29, 1966. Accepted November 14, 1966. Financial assistance was provided by the United States Department of Health, Education, and Welfare Fellowship 1 F3-WP26, 091-01,
Polarography of Uranium by the Catalytic Nitrate Fathi Habashi and GeorgeAnn Thurston Department of Metallurgy, Montana College of Mineral Science and Technology, Butte, Mont.
THE CATALYTIC ACTION of uranyl ion on the reduction of nitrate ion at the dropping mercury electrode was first reported by Kolthoff, Harris, and Matsuyama in 1944 (I). Since this wave is about 100 times higher than the normal polarographic wave, Crompton, Tichenor, and Young in 1949 (2) explored the possibility of utilizing it in determining traces of uranium. They concluded that the wave is highly susceptible to interference by minute amounts of anions and cations. It was only in 1955 that Hecht et al. (3) made use of this wave in determining traces of uranium in surface waters after a careful separation of uranium from other ions by strongly basic anion exchange resins. Since little work was done in this area, and we were interested in determining traces of uranium in sulfuric acid solution, we studied the catalytic wave in this medium.
i
4 c
EXPERIMENTAL
Stock solutions of uranyl acetate, sodium nitrate, sulfuric acid, and nitric acid were prepared. Different solutions of known concentrations of uranyl ion, nitrate ion, and sulfuric acid were prepared by dilution, and then polarographed. Solutions of nitric acid containing variable amounts of uranyl ion were also prepared and polarographed. A SargentHeyrovsky Polarograph Model XI1 was used, dropping times were 1 drop/3 seconds, and the temperature was controlled in a water bath at 25’ C. RESULTS
In sulfuric acid medium the following points were observed: At low NO3- concentration (0.0001-0.001M) the limiting current when plotted against the uranyl ion concentration shows an adsorption curve (Figures l a and lb). At [NOa-] = 0.01M, the limiting current is a nonlinear function of uranyl
(1) I. M. Kolthoff, W. E. Harris, and G. Matsuyama,J. Am. Chem. SOC.,66 1782 (1944). ( 2 ) C . E. Crompton, R. L. Tichenor, and H. A. Young, U.S. Ar. Energy Comm. Rept. AECD 2704 (1945); declassified 1949. (3) F. Hecht, J. Korkisch, R. Patzak, and A. Thiard, Mikrochimica Acta, 1956, 1283.
242
ANALYTICAL CHEMISTRY
U r o n i urn C o n c e n l r a t i o n ,
M /1
Figure 1. Effect of uranium concentration on the limiting current a) [NO3-] = 0.0001M, [H~SOII= 0.1M b) [NOS-] = 0.001M, [HzSOa] = 0.lM C) [NOa-] = 0.01M, [HzS04] = 0.1M
ion concentration (Figure IC). However, under these conditions, it is possible to determine uranium in this medium with reproducibility 1 5 %. The half-wave potential varies with the NO3- concentration as shown in Table I. At [N03-] = 0.1 M or higher, the half-wave potential becomes more negative and apparently coincides with the decomposition of the acid. For this reason the wave is completely distorted and is of no analytical value.
