2374
H. G. HEALAKD
JOHN
MAY
[COSTRIBUTION FROM THE L N V E R S I T Y O F
Vol. 80
BRITISHC O L U M B I A ]
A Spectrophotometric Study of the Stability of Lead(1V) in Hydrochloric Acid Solutions' BY H. G. HEALAND
JOHN
MAY
RECEIVED ACGUST26, 1967 The absorption spectra of lead(I\:) solutions are described. Equilibrium constants for The lead( IV)-leatl(IIj-chlt~ri~~c system in hydrochloric acid solutions of 0.10 to 1.00 S concentration, a t 19.5 and 2 5 O , are reported, as wcll 3 s specific r a t e for the decomposition of lead(1V) in 1.00 -V hydrochloric acid at several temperatures.
Introduction Wescott2 studied the equilibrium between lead(IV), lead(I1) and chlorine in hydrochloric acid solutions saturated with lead(I1) chloride, by means of chemical analysis. He concluded that lead(1V) takes the form of the ion PbCIS- in such solutions. His work was done before absorption spectroscopy had come into general use for studying the constitution of solutions, before i t had become usual in equilibrium studies to take account of, or eliminate, the effect of varying ionic strength, and a t only one temperature (25"). We have studied this equilibrium by making use of the previously undescribed ultraviolet absorption of lead(IV), and have determined equilibrium constants a t 19.5 and 25.0". \Ve also give some data on rates of decomposition of lead(1V) a t several temperatures near room temperature, which have not been reported previously. A11 our measurements were performed with solutions in 1.00 il: hydrochloric acid, 1.00 iV perchloric acid, or mixtures of these at a total concentration of 1.00 iV. The results show that lead(1V) exists in a t least two forms in solution, but support IVescott's conclusion about the predominant form in hydrochloric acid. Our investigation was of an exploratory character only, and has revealed a need for further work which we are not in a position to undertake a t present. Cesium hexachloropiumbate(IY), CsZPbCls, was employed throughout as a source of lead(1V). I t was preferred to the better known ammonium salt because of its superior keeping qualities (it is stable up to 250" and unaffected by moist air), and because in solutiom of the ammonium salt there is a risk of a reaction between ammonium ion and the chlorine formed by decomposition of the hexachloroplumbate(IV), which would have made i t difficult to obtain reliable equilibrium data. The measurements were all made in the concentration range to df, so that the small solubility of the cesium salt was no disadvantage. Experimental Cesium hexachloroplumbate(1V) was made by passing chlorine into a suspension of 0.5 g . powdered AI< lead(I1) chloride in a solution of 5 g. of AR cesium chloride in 6 nil. of water, at 100'. n'hen all the lead(I1) chloride had dissolved, the yellow microcrystalline precipitate of cesium hexachloroplumbate(1V) which had formed was filtered, washed well with 0.2 A' hydrochloric acid saturated n-ith chlorine, drained and dried in a vacuum desiccator. The oxidizing power of the product, when dropped into standard iron(I1j solution, and its loss in weight on heating to 275', were within 29% of the theoretical for CsaPbCls. The quantities, 3-30 nig., of the salt required for standard solutions (1) Part of this inlxstigation r a s carried o u t b y John hlay a s one of the requircments for the Honours R A . degree of t h e University of British Columbia. ( 2 ) E. \T \\-escott, 'I'liis J O U R N . ~ L , 4 2 , 1:183 (1920).
(usually in volumes of 25 or 50 nil.) were weighed out directly by means of a semi-microbalance and dropped into acid of the final concentration required. Redistilled water, redistilled reagent grade hydrochloric acid, and reagent grade 6OChperchloric acid were employed in the solutions. In all equilibrium and kinetic measurements, steps were taken to ensure that all the chlorine formed in the decomposition of lead(I\') remained in the solution. Solutions were prepared and stored in calibrated flasks which were filled completely and glass-stoppered, mixing being accomplished by shaking glass beads about in the flask. Optical cells were also filled so as to leave little or no air space, and promptly stoppered. Spectra were plotted by means of a Cary model 14 recording spectrophotometer, using 1 cm. or 10 cm. quartz optical cells. The sample compartment \vas thermostated t o i 0.2". Equilibrium data were obtained a t 19.5 and at 25", by allowing standard cesium hexachloroplumbate(1V) solutions to stand in completely filled stoppered flasks for several hours at the appropriate temperature in darkness, then quickly measuring the residual lertd(IV) concentration by means of the absorbance at 307 mM. The results obtained were independent of the period of standing, between 2.26 and 16 hours. Kinetic runs were carried out with solutions in the optical cells. I n order to minimize temperature changes arising from the handling of cells and pouring of solutions a t the beginning of the runs, the following procedure was adopted. Twenty-eight nil. of the acid to be used in the run w:i\ placed in a 10-cni. cell of 33.5-nil. capacity, anti allowed to come to the temperature planned for the run. Mean\\ hile a standard cesium hexachloropluxnbate(Il~)solution [ J i about 6 X lo-: molar, in the same acid, was brought t o the same temperature. Five mi. of this solution then ma5 pipetted quickly into the cell, and mixed by rocking the small air bubble back and forth. The dilution of the 31ready equilibrated solution shifted the equilibrium, causing more lead(1V) to decompose. 'The change in lead(I\-j concentration Tyas followed by measuring t h e absorbancc a t 307 inp. I t made no difference to the decomposition rate, or to the final equilibrium concentration of lead(IV), irhetber the cell n n s continuous1~-illuminated with 307 inp light throughout the run, or only briefly a t intcrvals of several minutes.
Results The Absorption Spectrum of Lead(IV).-Sincc lead(I\-) in hydrochloric acid decomposes to ;t considerable extent even during the time required to make up a solution arid scan the spectrum, :L pure lead(IY) spectrum cannot be directly observed, and has to be deduced from the experimental spectra of mixtures of lud(1V) with lead(1I) and chlorine. The spectra shown iil Fig. 1 were arriT-ed t i t by the fol!owing procedure. T o begin with, some method of finding the instantaneous lead(1V) concentration a t each wave length of the spectruni was required. Figures 1 and 2 show that there is a peak in the lead(1V) spectrum (at 303 nip in 0.1 A T HC1-0.9 LV HC104, and a t 307 mp in 1.0 N HCl) in a region where neither the P b + + ion nor any of its chloro-complexes absorbs appreciably. The absorption of chlorine is also relatively weak at this wave Ieiigtli. The
May20, 1958
STABILITY OF LEAD(IV)IN HYDROCHLORIC ACIDSOLUTIONS
I
lh 20
I?\
2375
scanning. The rate of decline of absorbance at 307 mg then was determined, and thence the instantaneous lead(1V) concentration a t each of a series of wave lengths in the spectrum of the mixture. The initial lead(1V) concentration being known, also the stoichiometric relationship Pb (I V) -+ Pb(I1) Cln, the instantaneous concentrations of lead(I1) and chlorine a t each wave length were calculated next. The absorptivity of lead(I1) in acid of the same composition was separately measured (Fig. 2), and that of chlorine obtained from the data of Zimmerman and S t r ~ n g . The ~ absorbance a t each wave length due to lead(1V) was then calculated by subtracting the lead(I1) and chlorine absorbances from the total absorbance a t that wave length. This, together with the previously calculated lead(1V) concentrations, gave the absorptivities plotted in Fig. 1. Spectra were measured a t 19.5" rather than 25") because the lead(1V) decomposes inconveniently quickly a t 25'. Zimmerman and Strong's data on the chlorine spectrum a t 25" were assumed to be accurate enough a t 19.5'. There are no peaks in the visible spectrum, though concentrated lead(1V) solutions look pale yellow because of absorption in the tail of one of the ultraviolet bands : the ultraviolet spectrum shows two prominent peaks. The stronger, a t 208 mp) coincides almost exactly with similar peaks in the spectra of lead(I1) in hydrochloric or perchloric acids. The other peak comes a t just over 300 mp. I t appears to shift toward shorter wave lengths as the hydrochloric acid concentration is reduced, though the shift (307-303 mp) between 1.00 and 0.10 N HC1 is not pronounced. In 0.35 iV hydrochloric 2cid the peak was found a t about 306 mp; the spectrum was not measured completely at this concentration. I n 1.00 *V perchloric acid with no added chloride, however, (Fig. 3), the peak appears
+
0 200
300
400
A,mP
Fig. 1.-Calculated absorption spectra of lead( IV), at 19.5': (1) in 1.00K HC1; (2) in 0.lOX HC1-0.90 N HClOd; e = molar absorptivity, plotted on a linear scale.
1
0 200
300
400
A, m,u. Fig. 2.-Measured absorption spectra of lead(I1) at 19.5': (1) in 1.00 N H C l ; ( 2 ) in 0.10 NHCl-0.90 S H C 1 0 , ; E = molar absorptivity, plotted on a linear scale.
height of this peak (duly corrected for absorption by chlorine) was therefore used as a measure of the lead(1V) concentration. The absolute value of e, the molar absorptivity of lead(IV), a t 307 mp, was found in the first instance by plotting the decay of the peak against time, and extrapolating back to find the height a t the time when the standard solution was mixed. There are several uncertainties in this extrapolation, and fortunately an independent check on the value of E so obtained was available. The equilibrium ratio [Pb(II)1. [C12]/[Pb(IV)] = K' (see next paragraph), calculated from the experimental data, can only be independent of the Pb(II)/Pb(IV) ratio if a correct value of E is employed in the calculations. It was found in fact, that the value of E a t 307 mp obtained as the mean of two back-extrapolations considered reliable, 0.97 X 10') did give K' values independent, within experimental error, of the Pb(II)/Pb(IV) ratio, for solutions of HCI concentration between 0.10 and 1.00 N , and a t temperatures of 19.5 or 25". It was therefore adopted. We consider that the value E = 0.9'7 X l o 4 is probably reliable within 3 or 4y0 for the solutions with hydrochloric acid concentration 0.35 N and higher, but may be somewhat more in error than this for 0.1 N hydrochloric acid, since the spectrum a t this concentration is appreciably different (Fig. l ) , and our K' values for this concentration are not so consistent as a t higher concentrations (Table I). The spectrum of a partially decomposed solution of lead(1V) was plotted a t a known rate of
10
1
Fig. 3.-(1) Absorption spectrum posed solution of CsZPbCle in 1.00 iV M. (2) The same solution nine M PbClz in 1.00 N HC10,. X solutions in 1 cm. cells.)
A, m,u. of a partially decomHC104, initially 0.9 X minutes later. (3) 0.9 (-ill spectra at 19.5",
to have shifted right down to 250 mp approximately. The Lead(1V)-Lead(I1)-Chlorine Equilibrium.One cannot write a single equation for this equilibrium, since lead(I1) exists in hydrochloric acid solutions as a mixture of complexes containing (3) G. Zimmerman and F. C . Strong, (1957).
T H I 3 JOURNAL,
7 9 , 20133
TABLE I EQUILIBRIUM COSSTANTS FOR THE LEAD(IV)-LEAD(II)CHLORINE EQUILIBRIUM T = temp., "C.; N I = moles HC1/1., N Z = moles HC10J1. C1 = equilibrium moles Pb(IV)/l. X 104; C2 = equilibrium ~ moles Pb(I1) or C1*/1. X lo4; K' = C I ~ / C(equilibrium constant). K'
x
G
T
Xi
NI
CI
1 9 . 5 10 . 3
1.00
Zero
1.06 0,566 0,192
1.57
1.05 0.54 0.25
1.79
1.26 0.741 0.211
5,24 4.11 2.21
1.29 2 19 2.28 2.31
1.43 0.997 .436
10.9 9.05 6.02
8.33 8.21 8 31
0.70
0.35
0.30
0.65
0.10 0.90
2j.O f 0.2
1 . 0 0 Zero
,230 ,170 ,087
108
{ ;:;: { :::; 1.28
19.7 17.1 12.9
169 171 201
,305
4 25 3.29 2.42
1.88 1.86 1.91 3.27 12.1
,960 ,580
0.70
0.30
.GO4
4.44
,35
.65
,688
9.10
.20
.SO
,723
17.3
.10
.90
,169
19.5
41.6 226
Solutions equilibrated 4 hours. brated 2.26 hours.
+ Cln + nC1-
*
k 2.26 f 0.08
8.31 0 . 1
170 & 10
This expression should hold as long as the chlorine formed in the decomposition is entirely retained in the solution, which was the case in these experiments. Actually, i t fitted the experimental data very well, good straight line plots being obtained for log
3.3 f 0.1 12.1 i0 . 4 41.6 i 1 . 5 225 i 15
Pb(1Y)
PbCls-
=
= specific rate of forward reaction C = instantaneous lead(1V) concn. CO = initial lead(1V) concn. C m = final equilibrium lead(1V) concn. t = time
1.88 i 0 . 0 6
For the very dilute lead solutivns employed in this work, in which the relative concentrations of the different complexes of a given oxidation state are independent of the total concentration of that state, i t follows that [total Pb(II)] [total Clz] [total Pb(IV)]
as a unimolecular reaction opposed by a bimolecular one leads to the expression
1 . 2 3 f 0.06
including one atom each of Pb(I1) and Pb(IV) and one molecule of chlorine (free or combined as Cl3-). Two such equations might, for example, be PbCl+ + Clz + 2C1- J_- PbC15and PbC12 $- C1z $- C1-
ePb(I1) + Cla
Mean x 103
from one to four chloride ions with the uncomplexed ion,4 chlorine exists partly as Clz and partly as C13-,3while lead(1V) (as shown above) must be present as at least two species. Only the chlorinetrichloride ion equilibrium, of these three equilibria of complex formation, has been fully investigated. However, for any stated group of ionic species there will be an equilibrium equation of the form Pb(I1)
Pb(1Y)
K'
Solutions equili-
a
The Rate of Decomposition of Lead(IV).Standard kinetic treatment of the reversible reaction
K'
IC' should not depend on any of the concentrations on the left-hand side, but will be a function of the chloride ion concentration. This relationship fits the experimental data well. From the practical standpoint, the most useful mode of presentation of the data seems to be as a series of k" values for various chloride ion concentrations. These are given in Table I. (4) H. Fromherz and K.-H.-Lih, 2 . physik. Chenr., 8153, 321 (1031); A. I. Bipgs, 1-1. S.Parton and I
