The Extraction of Ferric Chloride by Isopropyl Ether. II. A

The Extraction of Ferric Chloride by Isopropyl Ether. II. A Conductometric Study of the Ether Layer. Donald E. Campbell, Herbert M. Clark, Walter H. B...
0 downloads 0 Views 358KB Size
NOTES

506

Vol. 62

accord with the results of previous who have indicated that water is directly associated with the iron complex which is dissolved in the ether layer during extraction of iron from strong hydrochloric acid solutions of ferric chloride. It has been proposed2that all of the water is associated with the hydrogen of the complex acid, HFeC14, possibly as (H30+.4H20)FeC14-. Friedman and Taubes have shown by means of conductance measurements that an analogous salt, NH4GaC14, dissolved in anhydrous ether exhibits the characteristic behavior of strong electrolytes dissolved in solvents of low dielectric constant. It was therefore of interest to examine the conductance behavior of ether layers from the iron extraction system.

DH

1.2

0-Dilution8 of Ether Phare 0-Dllutlonr

a-Pd

It is apparent that the crossover in catalytic activity occurs a t the point where the rate-determining step changes from a n atomic desorption to a slow discharge. These experimental results show the relationship between the electronic configuration (effect on the heat of adsorption of hydrogen atoms) and the rate-controlling step of the hydrogen-producing mechanism. These results confirm previous results'O which indicated that on platinum and palladium electrodes both hydronium ions and water are simultaneously reduced to hydrogen atoms on the cathode surface. As the hydronium ion concentration in the double layer is decreased, the primary source of hydrogen ions changes from hydronium ions t o water. T H E EXTRACTION OF FERRIC CHLORIDE BY ISOPROPYL ETHER. PART 11. A CONDUCTOMETRIC STUDY OF THE ETHER LAYER' BY DONALD E. CAMPBELL, HERBERT M. CLARKAND WALTERH. BAUER Depavtment of Chemistry Rensselaer Polytechnic Institute, Troy, New York Received December 7,1967

It has been demonstrated,2 in the extraction of ferric chloride from aqueous hydrochloric acid solutions, that five molecules of water must accompany each molecule of HFeC14 extracted in order to dissolve this complex in isopropyl ether. This is in (1) Abstraotedf rorn the Ph.D. Dissertation rf Donald E. Campbell, Rensselaer Polyteohnic Institute, Troy, N. Y., June, 1952. (2) A. H. Laorene, D. E. Campbell, S. E. Wiberley and H. M. Clark, THIS JOURNAL,60, 901 (1956).

btainrd

~b%!1$~c'3 0.4M FeClg 0-Dllutlonr of E t h w Phone Obtolned at 7 M HCI and 0 . I M F*Cl3

i

Fig. 3.-Effect of pH on the catalytic activity of a- and 6-Pd-H cathodes. Inset shows an expanded scale for low values of the rate constant, k.

1 z!!w%h%$ a t 7M HCI and

8"

0.8

2 ha

2

0.4

0.0

-3

-2

-1 Log ( M F ~ ) . Fig. 1.-Equivalent conductance of dilutions of ethereal iron extraction phases prepared by extraction of 7 M hydrochloric acid solutions containing varying amounts of iron.

Experimental Procedure.-Ethereal iron extraction phases were prepared by equilibration of aqueous solutions, containing varying concentrations of ferric chloride and hydrochloric acid, with equal volumes of isopropyl ether at 20" for 48 hours with frequent shaking. The ether layer was separated from the equilibrium aqueous phase by means of a siphon, and analyzed for chloride and iron as previouslya described. Water in the ether layer was determined by a modified Karl Fischer titration, described by Laurene.7 Dilutions of the ethereal iron extraction phases were made with isopropyl ether, which had been equilibrated over an equal volume of hydrochloric acid. The hydrochloric acid concentration in each case was the same as that used to prepare the given ethereal iron extraction phase. Other diluents, such as dry isopropyl ether, or isopropyl ether saturated with water, caused appearance of multiple ether layers. The iron concentrations in the diluted solutions were determined from the dilution ratio and the iron concentration of the original (3) S. Kato and R. Ishii, Sei. Papers Inst. Phys. Chem. Researclr, Tokyo, 36, 82 (1939). (4) J. Axelrod and E. H. Swift, J . A m . Chem. SOC.,6 2 , 33 (1940). ( 5 ) R. J. Myers, D. E. Metzler and E. H. Swift, ibid., 7% 3767 (1950). ( 6 ) H. L. Friedman a n d H. Taube, ibid., 72, 3362 (1950). (7) A. H. Laurene, Anal. Chem., 24, 1490 (1952).

*

507

NOTES

April, 1958

1.o

0.8

3 0.6 s

0-Dilution8 of Ether a t 7 M HCI @-Dilutions of Ether a t SM HCI 0-Dilutions of Ether at 4 M HCI

Phase Obtained and 0.2M FeC13 Phasr Obtalned and 0 . 2 M FeCI3 Phase Obtained and 0 . 2 M FeC13

ba

0.4

0.2

0.0 -6

-1

-. Fig. 2.-Equivalent

conductance of dilutions of ethereal iron extraction phases prepared by extraction of 0.2 M ferric chloride solutions containing varying amounts of hydrochloric acid.

ethereal iron extraction phase. The water in the diluted solutions was determined either by direct analysis or from the water analyses of the equilibrated ether diluent and the ether extraction phase. The specific conductances of the resulting series of diluted solutions were measured in a stoppered conductance cell constructed out of a standard taper 24/40 female joint. The lower end of the joint was blown into a bulb which enveloped the electrodes, and mercury wells were sealed to opposite sides of the bulb. A groove was ground in the stopper and a hole drilled in the side of the female joint so that there would be no pressure exerted on the solution in the cell when the stopper was put in place. Conductances were measured a t 20.00 iz 0.01' with a General Radio 650-A impedance bridge equipped with 1000 cycle oscillatoramplifier. A Dumont 208-B cathode ray oscilloscope was employed as the null point detector.

equivalent of iron and from its dependence upon the ethereal iron concentration shown in Figs. 1 and 2 it is evident that the iron complex which is extracted into the ether layer in the ether extraction of iron is ionic, Moreover, it is seen that the ionic nature of the extracted iron complex does not depend upon the initial conditions of extraction. The dilution curves shown in Figs. 1 and 2, which were obtained under widely differing initial conditions, are identical in shape and nearly coincident over the same range of iron concentration. The conductance curves shown in Fig. 2, obtained for diluted extraction phases prepared at the lower initial hydrochloric acid concentrations, Results are slightly displaced in the direction of higher The dilution studies were carried out on two dif- equivalent conductance, This effect is ascribed to ferent groups of ether extraction phases. I n the the solution of water in the ether solvent. Such first group, three different ether phases were ob- water is to be distinguished from the water which tained by extraction of aqueous solutions which is known2 to be associated with the iron complex. were initially 0.1, 0.2 and 0.4 ill in iron, but which From work done on the ternary system, isopropyl were all 7 M in hydrochloric acid. The results of ether-HC1-HZO,* isopropyl ether equilibrated over the conductance measurements made on this first an equal volume of hydrochloric acid below 9.5 M group are shown in Fig. 1. In the second group, dissolves more water the lower the hydrochloric three ether phases were prepared by extraction of acid concentration. aqueous solutions which were initially 7, 6 and 4 M When the dilutions were carried to low iron conin hydrochloric acid but were all 0.2 &I in iron. centrations, minima in the conductance-dilution The conductance data for diluted solutions of this curves were found to occur, as shown in Fig. 2. group of ether extraction phases are plotted in Fig. 'Friedman and Taubes also observed conductance 2. When the solvent conductance exceeded 1% minima in their conductometric study of solutions of the specific conductance of the ethereal iron solu- of the analogous compound, NH4GaC14, in ethyl tions, the specific conductance was corrected for ether. The occurrence of conductance minima the conductance of the solvent. The latter was has been shown by Fuoss and Krausg to be characwas considered to be isopropyl ether equilibrated teristic for highly polar compounds dissolved in over an equal volume of hydrochloric acid at the media of low dielectric constants, and has been assame concentration used in the extraction. (8) D. E. Campbell, A. H. Laurene and H. M. Clark, J . Am. Chem. Discussion SOC.,74,6193 (1952). From the magnitude of the conductance per (9) C . A. Kraus and R. M. Fuoss, ibid., 66, 21, 2387 (1933).

508

NOTES

cribed to the formation of electrostatically associated neutral ion pairs. At high concentrations in isopropyl ether, the tetrachloroferrate ion2 may be largely associated in the form of triple ions such as + and [FeC14[ H ~ O + ~ H ZFeCl4-H30f.4Hz0] O &0+4Hz0 FeC14-] -, associated in neutral iron pairs as [H30+.4Hzo FeC14-1, and dissociated to a lesser extent as the ions H30+.4H20and FeCI4-. With dilution, the conductance would be expected to go through a minimum as the triple ions dissociated and the concentration of neutral ion pairs increased. The final increase in conductance would follow an increased dissociation of neutral ion pairs with high dilution. Thus, the appearance of such minima in the conductance-dilution curves of the ethereaI phases of the iron extraction system provides supporting experimental evidence for the (6 polymerization" theory which Myers, Metzler and Swift5 have proposed to account for the variation in the iron distribution coefficient with total iron concentration in ethereal extraction of iron from aqueous solution having constant initial hydrochloric acid concentrations. Acknowledgment.-The authors are indebted to Lewis G. Bassett for his many suggestions during the course of this work. This work was partially supported by the U. S. Atomic Energy Commission under Contract No. At(30-1)-562.

Vol. 62

Tetrazo1es.-All were prepared from amide and hydrazoic acid by a method similar to that described by Harvill, et aZ.* Analytical data are presented in Table I. All tetrasoles were sublimed twice before making any physical measurements. Amides.-These were prepared in the usual manner. I n Table I, the melting points of the amides are listed; all were recrystallized twice from ethanol. Physical Measurements and Calculations .-This experimental part is essentially that of Kaufman, et al. The apparatus was calibrated with benzene, the dielectric constant of which was taken to be 2.2750 a t 25°.4 The calculated data are presented in Table 11, where €1 and VI, the dielectric constant and the specific volume of the solvent, respectively, are the intercepts while CY and p are the slopes of the straight lines obtained by plotting solution dielectric constant and specific volume, respectively, against weight fraction of solute.6 P m is the solute polarization a t infinite dilution calculated from the Halverstadt and Kumler relationship6 while Rz is the molar refraction of the solute for the D-sodium line and p is the dipole moment calculated from the approximate Debye equation /A 0.01281 X lo-'* [ ( P a - Rz)T]*/z (1)

Discussion From the known moments of nitrobenzene (3.97 D),1-phenyl-5-methyltetraaole (5.64 D) and l-pnitrophenyl-5-methyltetrazole (3.20 D) it follows that the angle between the phenyl-NOz bond and the moment of 1-phenyl-5-methyltetrazole is approximately 34". Therefore, 34" is also the angle between the p h e n y l 4 bond and the moment of 1phenyl-5-methyltetrazole. In the following manner, Kofod, et al.,' obtained a value of 0.8 D for the Ph-N bond assuming a value of 1.3 D for the GEOMETRIC CONSIDERATIONS OF H-N bond. The experimental value for unsubTETRAZOLE DERIVATIVES FROM stituted pyrrole is 1.80 D while that of N-phenylpyrrole is 1.32 D. Therefore, since the two rings DIPOLE MOMENT DATA are co-linear, the Ph-N bond must be 0.48 D MARTINH. KAUFMAN AND ALANL. WOODMAN less than the H-N bond or 0.8 D. Similarly the Physical Chemi8ti-g Branch, U. S. Naval Ordnance Test Station, China CH3-N bond is 0.12 D more than the H-N bond Lake, Califmnia or 1.42 D. Received December IO, 1067 If the value of 0.8 D for the phenyl-N moment in Hill and Sutton' in their study of electric dipole 1-phenyl-5-methyltetrazole is utilized, a value of moments of some sydnones applied vector addi- 5.00 D for the moment of 5-methyltetrazole and tion to the solution of geometric problems. From 39" for the angle between the phenyl-N bond and known moments and moments of aromatic com- the moment of 5-methyltetrazole is obtained. pounds differing only by para substitution, they From a similar treatment of 1-methyl-5-p-nitrodeduced various angles between substituted phenyl phenyltetrazole (3.87 D) the following informabonds and the direction of the dipole moments tion was obtained. The angle between the phenylof the parent compound. A qualitative picture NOz bond and, therefore, the phenyl-C bond and the of the hybrid nature of tetrazoles was recently re- moment of 1-methyld-phenyltetrazole is approxiported by Kaufman, et aL2 The purpose of the mately 43". The angle between the phenyl-C bond present study was t o investigate the geometry of and the moment of 1-methyltetrazole is approxitetrazoles by means similar to those of Hill and Sut- mately 45" while the dipole moment of l-methylton. With this idea in mind, chloro-, bromo- and tetrazole is calculated to be 5.45 D. The experinitrophenyltetrazoles were prepared and their mental value of 5.38 D was reported for this comdipole moments measured in benzene solution at pound by Jensen and Friediger.* 25". From the above data, predictions about the moments of phenyltetrazoles differing only by para Experimental substituents may be made; e.g., from the known Preparation and Purification of Materials.-Mallinckrodt analytical reagent grade benzene was purified by crystallizing moments of chloro- (1.59) and bromobenzene twice, rejecting a small portion each time, and then fractionally distilling after standing several days over freshly cut and ribboned sodium. All was rejected but the center cut which was re-distilled in the same manner before use. The specific volume parameter for the solvent was obtained for each run from a least squares examination of the specific volume, weight fraction data and was found to be constant t o f0.0295. (1) R. A. W. Hill and L. E. Button, J . Chem. Soc., 746 (1949). (2) M.H. Kaufman, F. M. Ernsberger and W. S. McEwan, J . A m . Chem. SOC.,18, 4197 (1956).

(3) E. K. Harvill, R. M. Herbst, E. C . Schreiner and C . W. Roberts, J . O r g . Chem., 16, 668 (1950). (4) A. S. Brown, P. M. Levin and E. W. Abrahamson, J . Chem. Phys., 19, 1226 (1961). (5) G. Hedestrand, Z . physik. Chem., B2, 482 (1929). (6) I. F. Halverstadt and W. D. Kumler, J . A m . Chem. Soc., 64,2988 (1942). ( 7 ) H. Kofod, L. E. SutDon and J. Jackson, J . Chem. SOC., 1467 (1952). (8) K. A. Jensen and A. Friediger, Kgl. Danske Vddenalcab.Selskab. Math. /us. Medd., 20, 1 (1943).