Dec. 5, 1954
NOTES
602 1
of mannitol gave a boron content equivalent to 31.0 cc. of BsH9. The calculated Hz/BsHe ratio was 12.1. I n another experiment 6.56 cc. of BsH9 (by titration) yielded 78.4 cc. of Hz to give a ratio of 11.9.
region studied, had little effect on the spectrum of tetramethylammonium tetraiodochloride. It did lower the iodine bromide peak shown by tetramethylammonium iodobrornochloride a t 486 mp from a molar absorbancy index (molar extinction U. S. NAVALORDNANCE TESTSTATION coefficient) of 294 to 258. Absorption in the region PASADENA, CALIFORNIA 260-400 mp was increased a t the same time. Such changes might indicate some suppression of the Dissociation of Quaternary Ammonium Poly- dissociation and formation of the iodobromochloride halides in Trifluoroacetic Acid' ion especially since a peak of molar absorbency index around 3.5 x lo4 near 240 m p and one of BY ROBERT E. BUCKLESAND JACK F. MILLS molar absorbency index about 400 near 355 mp RECEIVED JULY 19, 1954 have been observed5for this ion in ethylene chloride During the investigation of quaternary ammon- and in acetonitrile. ium dichloriodides2 and of the quaternary ammonAttempts to suppress the dissociation in a 7.08 ium tetrachloroiodide~~ i t was found that the visible X M solution of tetramethylammonium and ultraviolet absorption spectra of the salts in triiodide with tetramethylammonium iodide (3.4 trifluoroacetic acid were identical with those of the X M ) led to the disappearance of the violet halogens expected from the complete dissociation color and the precipitation of the triiodide salt. of the polyhalide ion. I n Table I are summarized When 2.5 ml. of water was present with the trisimilar results for other quaternary ammonium fluoroacetic acid in 50 ml. of solution a red-violet halides. I n each case the hydrogen halide derived solution was obtained. The absorption spectrum from the most electronegative halogen atom present showed an iodine peak of molar absorbency index in the polyhalide ion must have been formed also. 706 a t 500 mp and very high absorption a t wave Little ionization of the hydrogen halide would be lengths below 400 mp which would be expected of expected in trifluoroacetic acid, which is such a triiodide ion. strong proton donor that the triiluoroacetate ion In the course of the investigation i t became necespresent would show little tendency to act as a sary to re-examine the absorption spectra of iodine, proton acceptor, and the acid itself would show bromine and iodine chloride in trifluoroacetic acid. even less. Consequently it is not unreasonable It was observed that the values for the molar that the concentration of the halide ion in each case absorbency indices reportedAa for these halogens would be so low that the molecular halogen present were consistently too low. Evidently a combinain the dilute solution used would not interact to tion of difficulties encountered in making up the give any detectable amount of polyhalide ion. stock solutions and standardizing them had not The dissociation cannot be explained by competi- been completely surmounted in the earlier investigative complexing of the molecular halogen with the tion. The higher values obtained in the present solvent since such acidic solvents as trifluoroacetic investigation and given in Table I actually are acid show4 little tendency to enter into such com- more consistent with the conclusions reached in plex formation. the earlier work than are the values which were reported a t that time. That is, the solutions of TABLE I the halogens in trifluoroacetic acid behave spectroMOLAR ABSORBENCY INDICES OF ABSORPTIONSPECTRA photometrically like those in non-complexing PEAKS OF HALOGENS AND QUATERNARY AMMONIUM POLYsolvents. This general relationship is shown in HALIDES IN TRIFLUOROACETIC ACID Fig. 1 for iodine bromide, which was not included Concn.. Compound in the original study. M x 108 A, mp am
1.44 512 815 I2 Meah-Ia 1.55 513 806 BLI~NI~ 1.40 513 805 MedNIs 0.64 512 1624 Me4NI4CI" 0.72 512 1682 13.9 411 160 Brz MedNBrS 14.2 410 159 BuaNBrl 12.8 410 157 IBr 11.1 187 308 MerNIBrz 7.7 487 314 Me4NIBrCI 7.4 486 294 a Iodometric analysis showed this compound to be impure. The percentage of reducible halogen was 3% high.
The addition of excess tetramethylammonium chloride, which showed little absorption in the (1) Work carried out under Contract No. AT(ll-1)-72, Project
No. 7 with the U.S. Atomic Energy Commission. (2) R. E. Buckles and J. F. Mills, THIS JOURNAL, 76,4845 (1954). (3) R.E. Buckles and J. F. Mills, ibid., 76, 3716 (1954). (4) (a) R. E. Buckles and J. F. Mills, ibid., 76, 552 (1953); ( b ) J. G. Bower and R. L. Scott, ibid., 76, 3683 (1953); (c) L. I. Katzin atid J . J. Katz, ibid , 7 6 , 6057 (1953).
350 400 450 500 550 600 Wave length, mp. Fig. 1.-Absorption spectra of iodine bromide (1) 5.3 X M in anhydrous acetic acid, (2) 6.6 X 10-3 M in carbon tetrachloride and (3) 1.11 X lo-* &f in trifluoroacetic acid. Experimental Part Materials.-The solvents and the halogens were purified as described4a for the earlier investigation. The polyio250
300
(5) A. I. Popov and R. F. Swenson, private communication
NOTES
602%
did& and the t r i b r o m i d e ~ ~were ~ ~ ' available from earlier investigations. Pure iodine bromide as well as samples of tetramethylammonium dibromoiodide (m.p. 192'), tetramethylammonium iodobromochloride (m.p. 204205') and tetramethylammonium tetraiodochloride (m.p. 108-109") were kindlysupplied by Dr. A . I. Popov and Mr. R. F. Swenson of this Laboratory. Stock Solutions.-Homogeneous solutions of polyhalides or halogens in trifluoroacetic acid could be prepared only when the solute and solvent were shaken together long and vigorously. Often periods of shaking several hours in length were necessary. Particular care was taken to use tightly stoppered containers when handling the solutions in order to avoid halogen losses caused by volatility. A 5- or 10-ml. sample was mixed with a t least 10 times its volume of water and the resulting aqueous solution was immediately subjected to an iodometric analysis for reducible halogen. Absorption Spectra Measurements.-Measurements were carried out a t 25' in calibrated silica cells of path length 1.OO i.0.01 cm. with a Cary Model 11 recording spectrophotometer. The uniform nomenclature and symbology suggested I,y the National Bureau of Standards* have been used throughout this note. (6) (a) R. E. Buckles, J. P. Yuk and A. I. Popov, THISJOURNAI., 74, 4379 (1952): (b) R. E. Buckles and J. P. Yuk. i b i d . , IS, 5048 (1953). ( 7 ) R. E. Buckles, A . I. Popov, W. F. Zelezny and R. J. Smith, ibid , 73, 4525 (1951). ( 8 ) National Bureau of Standards, Letter-Circular LC-857 (1947).
DEPARTMENT OF CHEMISTRY STATEUNIVERSITY OF IOWA IOWACITY,IOWA
Assuming the following equilibria ZnO -I- OHZnO 4-20H-
for which K1 = [ilfZn(OH)aKt =
[hfZntO€?)d--
+ H10 +&(OH),-
+ H20 +Zn(OH),--
x yZn(OH)a-l/[llfOEf- x YOH- x a H z O ] x YZ~(OH)~--][(A~~OOIT)* x (YOR-)' x
13Y T H E D F O R D P. DIRKSE, CLARENCE POSTMUS, JR., A S D
ROBERT 'I'ANDENBOSCH RECEIVEDJUNE 9, 1954
(4) nHzO1
(5 )
then
s = iIfZn(OH)t
+ [KI x
ATOH- x yOE-/yZn(OH)a-] f [KzX (MoH-)' X (YOE-)2/YZn(OH)r-] ((9 The activity of water is omitted here since in dilute alkaline solutions this value approaches one. It may further be assumed that in a given solution the activity coefficients of all singly charged ions are the same. Equation 6 then becomes s = M Z n ( O H 9 K1 x h f O € ? - Kz x (ACOEf- ) z x
+
+
(YOH-) 2/~znco~)r--( 7 )
From the Debye-Hiickel theory log Y ~ C I - I ) = -&4/(1 and log ~ * t ( i - z ) = - 2 A f i / ( 1
+ BaidG) + Baidid
where the numbers in parentheses refer to the type of ionic species present. From this it follows that log Y Z t C l - 2 ) = 2 log Y + C l - l )
A Study of Alkaline Solutions of Zinc Oxide
(2) (3)
( 8)
if ai(l - 1) = ai(1 - 2) which is a reasonable assumption for the systems studied here. From equation 8 it also follows that YrtCl--91
=
((3)
[Y*(l-l)l*
The following series of relationships can then he made (YOH-)' = (Y+)2(yoH-)2 = &]~4. = [Y+C1-111' =
In a recent paper' it was shown that in concen- YZn(OH)4-( Y f ) 2(YZn(OB)4)[ Y i C l - 2)1 a [y=!=Cl-1)16 1 trated alkaline solutions dissolved zinc is present (10) primarily as Zn(OH)r--. Since, however, claims [Y+cl- 0 1 2 have been made for the identification of a hydrogen Substituting this relationship in equation 7 zincate ion in such solution^^-^ an attempt was made O x -1lhHK2 x (1~~OH-)2/[Yi(l-11)12 to verify this in dilute alkaline solutions where s = ~ ~ ~ Zf o KI (11) such ions are more likely to occur. This attempt Although the Debye-Hiickel equation holds only involves the determination of the solubility of in dilute solutions still it is likely that a ratio of aczinc oxide in alkaline solutions. Since it was previously shown' that in strongly tivity coefficients determined from this equation alkaline solutions Zn(OH)4-- was the highest will hold up to somewhat higher ionic strengths. hydroxyl containing species of dissolved zinc it can Experimental be assumed that all the dissolved zinc in dilute The best All solubilities were determined a t 25 0.1'. alkaline solutions will be in the form of ZnO or grade of powdered zinc oxide, potassium hydroxide and soZn(OH)z, and Zn(OH)4--. Recently a dium hydroxide was used. The alkaline solutions were preby dilution of saturated solutions of potassium hyclaim has been made for the existence of Z I I ( O H ) ~ - ~ pared droxide and sodium hydroxide. Precautions were constantly ions in alkaline solution^.^ However, if the taken to exclude carbon dioxide. The solutions were stored authors would have used the mean activity instead in polyethylene bottles and after zinc oxide was added th(. oi the stoichiometric concentration of the hydroxyl mixtures were kept in the constant temperature bath for .L month. During this time they were shaken frrquently. ions in the Nernst equation, their results would Finally they were carefully filtered using fritted glass filtering ha\-e shown that Zn(OH)r-- is the highest hydroxyl crucibles. The analysis for zinc was done amprrometrically.6 The mean activity coefficients of the hydroxyl ioii containing zinc ion present. If S represents the total solubility of zinc in such were determined from the data given in the book of Harnetl and solutions, then Results .s-. - l J Z n ( O H ) z f l l I Z n ( 0 H ) a lll(ZnOH)4-(1) Figure 1 gives a summary of the data obtained. I t will be noted that the results for sodium hydrox(1) T . P. Dirkse, J Elecfrochc,n. SOC.,101, 328 (1954). ( 2 ) J. H. Kldebrand and W. G. Bowers, THIS J O U R N A L , 38, 785 ide solutions are more erratic than those for po(IYIO). tassium hydroxide. Extrapolation of these curves (X) G.U'. H e i x dnd I.;. A . Schumaker, 7'rans. Elc
