SPECTROPHOTOMETRIC STUDY OF
Oct. 20, 1964
T h e simple Huckel molecular orbital calculations obtained for the aliphatic Schiff bases indicate t h a t the system is polar with an appreciable negative charge on oxygen and with a partially delocalized positive charge. T h e same situation probahly occurs in the aromatic Schiff bases IT- and 1' with the possibility of delocalizing the charge into the aromatic system In the enol-imine form. the charge distri-
[ COsTRIBUTION F R O M
THE
DEPARTMENT O F
PHYSICAL
THE
SYSTEM Iz
+ Br-
4287
bution calculated for the aliphatic system suggests the system is less polar than the keto-amine form, primarily owing to the removal of the highly dipolar carbonyl group. T h e carbonyl oxygen would be a logical site for association with a hydrogen bonding solvent such as chloroform. Acknowledgment.-We wish to thank the Xilton Fund of Harvard University for generous financial support.
CHEMISTRY, HEBREWUNIVERSITY, JERUSALEM, ISRAEL]
A Spectrophotometric Study of the System 1,
+ Br-
BY EHUDEYALA N D AVNER TREININ RECEIVED MAY4, 1964
+
The ultraviolet spectrum of the system 12 Br- was thoroughly investigated. The shift of the main absorption band to shorter wave lengths on raising the concentration of Br- is ascribed to the equilibrium: 12BrBr IBr2I-. The spectroscopic and thermodynamic properties of the two complexes involved were determined. The origin of the electronic transitions is discussed.
+
+
~
Solutions containing Ip and Br- exhibit an intense absorption band, the location of which depends markedly on the concentration of Br-." Thus on raisin< the concentration of Br- from 2 X 10F3 to 3 -11, the peak of the band is shifted from 270 to 256 mp The spectrum a t low Br- concentration was assigned to 12Br-, this assignment being in agreement with the proposed origin of the band,' but the nature of the spectral shift on raising the bromide concentration remained unclear. The spectrum of IBr2- in aqueous solution displays an intense band a t about 254 mp. Therefore, Daniele2 proposed t h a t the spectral shift is due to IBr2-, which he considered to be photochemically produced in concentrated bromide solutions. However, we found the same spectral shifts irrespective of room and spectrophotometric illumination, so this cannot be the correct answer. But Rr- solutions as IBr2- is normally produced in Is expressed by the equilibrium
+
I?
+ 2Br-
+ I-
'IBr2-
(1)
or 12
+ Br-
12Br-
(2)
From the available thermodynamic data,4 we obtain a t 25'
which shows t h a t when (Br-)/(I-) is -lo6, practically no complex but IBr2- is present in solution. I n this wor!: we present the results of a detailed spectrophotometric study of the system I p Br-, from which we derive the spectra and some thermodynamic properties of IzBr- and IBr2-.
+
Experimental The absorption spectra were measured for ( a ) iodine in solutions of NaBr a t various concentrations; ( b ) the same, but (1) D . Meyerstein and A. Treinin, T v a n s . F a r a d a y S O L . 69, , 1114 (1963). ( 2 ) G. I?aniele, Cazs. c h i m . i i a l . . 90, 1082 (1960). (3) A. E . Gillam, T Y Q ~ FSa.r a d a y Soc.. 29, 1132 (1933). (4) W . M . Latimer, "Oxidation Potentials," Prentice-Hall, I n c . , Engle-
wood Cli;?~, N . J . , 1932.
with addition of a relatively small concentration of NaI; ( c ) IBr in solutions of 0.02 M HC104 and various concentrations of NaBr; ( d ) IBr in 10 M HC10, and a small concentration of KaCI. The spectrophotometric nicasurements were carried out with a Hilger Uvispek spectrophotometer in a thermostated cell compartment ( f 0 . 5 ' ) ; 1-cm. silica cells were used. Materials.-KaBr, NaI, and HC104 of A.R. grade were used without further purification. Resublimed 12 was further sublimed. T o prepare IBr we added excess of Brz (A.R.) to IZ and kept the mixture a t 50" for several hours. T h e IBr was sublimed from this mixture and its purity checked by determining its melting point (41'). Solutions of freshly sublimed IBr were used, but nevertheless they always appeared to contain excess In (see Results). Water was redistilled from alkaline permanganate and from dilute phosphoric acid in an all-glass still. Solutions.-The solutions of sets a and b were prepared from a saturated stock solution of Ip and solutions of KaBr and K a I , At the end of each measurement, the total concentration of the iodine was determined by adding few crystals of K I to the solution and measuring the absorbance of 1 3 - a t 352 mr.' We confirmed t h a t this absorbance was independent of further addition of K1. The solutions of set c were prepared by dissolving about 50 mg. of 1Br in 1 1M XaBr solutions and then diluting by a factor of 1/500. This procedure was used in order to rapidly dissolve the IBr. The dilute solution of IBr was used as a stock solution for preparing the mixtures, which contained HCIOl to prevent hydrolysis. The exact concentration of IBr was determined in the same way as done for 12, but now the excess of K I was added to a blank solution containing the same concentrations of IBr and S a B r but with no HC104 present. This was done to prevent autoxidation of I-. The solutions of set d were prepared by adding few crystals of NaCl to solutions of IBr in 10 1.I HClOa.
Results On raising the Br - concentration a t constant IZ concentration M ) , the peak of the absorption band is gradually shifted from 270 to 254 mp. The apparent extinction coefficient, e m a x = D m a x / ( I ? ) o is the total concentration of iodine] increases thereby till i t reaches a constant value of 5.4 X lo4 (Fig. 1). The following preliminary experiments seem to prove t h a t the 270-mp band is due to I2Hr-, which is converted to IBr2- on raising the Br- concentration. a . Job's method of continuous variations; tested X solutions of I? and NaBr indicate the foron mation of a 1 : 1 complex (Fig. 2). ( 5 ) P. Job, A n n . c h i m . ( P a r i s ) , 9 , 113, 135 ( 1 9 2 8 ) . I t should be realized t h a t t h e use of equal concentrations of reactants is not essentiai provided only a littie of both reactants enters into t h e complex.
EHUDEYALAND AVNER TREININ
4288
5 .O
i
Vol. 86
'2
i 0.25-
w
V
0.20
-
0.1 5
-
m CL
-
10-
0 Ln
m
Q
2 00
\
I
250
300
350
Fig. 1.-The absorption spectra of some polyhalides (the spectrum of IBrC1- on a n arbitrary scalei.
b. I n presence of a large concentration of Br-, Ia and IBr display nearly the same spectra (Fig. 1). The small difference between the two spectra may be due to some 12Br- present in the 12-Br- system. c. The iodide ion, even a t very low concentrations, has a pronounced antagonistic effect to that of Br- in shifting the spectrum back to longer wave lengths. For example, a solution of 3.3 X 10-j -11 If and 0.06 M NaBr displays a band with Am, a t 263 mp, which is shifted to 267 m p by adding lo-' M NaI (the absorption of 13- was subtracted as explained later). This result is readily understood by considering eq. 4. In order to suppress the formation of IBr2- in determining the spectroscopic and thermodynamic properties of I?Br-, we thus turned to study the system 1 2 NaBr NaI The iodide ion also acts to suppress the hydrolysis of Iz; addition of HCIOj had hardly any effect on the spectrum of this system. The System Ia NaBr Na1.-Two sets of experiments were carried out with this system: a . All the iodine was practically kept complexed by employing a relative large concentration of NaI. The ratio (Br-)/(I-) was such as to nearly eliminate the formation of IBr2-. The iodine was thus present as both 13-and 12Br-. their concentrations obeying the relation
+
+
+
+
(5) where K12Br-and K T t -are the stability constants of the corresponding complexes and F is the activity coefficient ratio. The absorbance of the system, a t wave lengths where 12Br- hardly absorbs, was compared to that of another solution containing the same concentration of Iz and XaI but with no NaBr present. Denoting the absorbance of the system b y D and that of the 13--solution by D AD i t follows t h a t
+
Table I contains the experimental results a t two wave lengths and the calculated value of FKI,-/ KI?B,, The approximate constancy of this ratio (with apparent accidental errors), despite the large
0.10-
1
I
I
a2 VOLUME
I
0.6
01 FRACTION
J
1
0.0
OF HALOGEN
+
Fig. 2.- Job's method applied to the systems: Is BrM ) and IBr BrM).
+
variation in ionic strength, is easily explained by eq. 5 . The activity coefficient products in the numerator and the denominator shpuld be nearly the same and should vary in the same way with the ionic strength. Thus we can put F = 1. Using the value KI,- = 824,6we obtain K19Br-= 10.5 f 0.4 J-' a t 20'. TABLEI FOR EVALUATING FKIa-/KlrBr-a RESULTSNECESSARY ( S a B r ) , -M
1 .0 1 5 2 0 2 0 2 0 2 0 2 5 3.0 3.5 According
(NaI),M
ADiD at 352 mp
0.06 0.06 0.06 0.09 0.12 0.15 0.06 0 06 0.06 to eq. 5 and
0.228 0.234 0.324 0.343 0.450 0 440 0.280 0 277 0 208 0.216 0.163 0.154 0.520 0 520 0.672 0.680 0.755 0 746 6 ; ( I 2 ) , = 3 X 10-6 M , 20".
A D / D at 370 mg
FKI,-/KI,B~(average)
72 75 75 80 79 84 80 74 78
b. In this set the concentration of I - was small (4 X 10+ M), the concentration of Iz was about 1 0 F Mj and that of Br- was varied between 0.04 and 0.14 AI. Under these conditions the iodine was only partly complexed and (IBrz-)/(IaBr-) did not exceed The contribution of 13- to the absorption was evaluated as follows: the absorbance was measured at 30%and 344 mp, where 13- has the same extinction coefficient (Fig. 1). Denoting the contribution of 12Br- and 13- to the absorbance a t 344 mp by D I ? B , and Dra--,respectively, we obtain D W = D I > B ~-t DI,- and D302 = P D I 1 ~ r - DI,-. The constant P was determined from the spectrum of 12Br- a t an unknown concentration. For this purpose a solution was prepared which contained about lW3 M 1 2 , 1 0 F 3 -14 Br-, and M I-, and its absorption was measured against a solution of the same concentration of Ia and I-. Under these conditions only a little of the
+
(6) G. Daniele, Gnzz. r h i m . t i d , 90, 1068 (1960)
SPECTROPHOTOMETRIC STUDY OF
Oct. 20, 1964
THE
SYSTEM I2 4- Br-
I
4289 I
I
I
L n ( l 0 D)
0.8 -
Ln( K/100)
,8
5.0
-
L
r 0
0
8-1.2
L.0
-
-
-0.9 3.3
32
-
2.0
I
I
I
Fig. 4.-The I2
35
3.4 1 /T
x
10' deg:l
plot of the van't Hoff equation for the systems: (curve I ) and IBr Br- (curve 11).
+ Br-
+
30
I
6 9 12 A.ICI~M . 3.--Scott's method applied to the system I? Br-, in the presence of a little I - to suppress the formation of I B r 2 - a t 20". 3
+
Izwas complexed, so that the sample and reference solutions contained nearly identical concentrations of I3-. The value of /3 thus determined is i . In this way D1,and hence the contribution of IS- to the absorption at any other wave length could be determined. Moreover, the total iodine concentration, apart from that of Id-, could also be determined. Denoting the latter by B , Scott's equation' takes the form
AB
A
D
E
-=
-+
-1 KE
(7)
4.0
2.0
(BrFig. 5.-&ott's
60 x
8.0
1 0 3 ~
method applied to the system IBr various temperatures.
+ Br-
at
where A is the Br- concentration, K is the stability constant of IzBr-, D is the contribution of I J 3 - to the that of D . Figure 4 shows the v a n ' t Hoff plot a t 270 total absorbance, and E is its extinction coefficient at mp. From the slope of the line we obtain AH = - 1.37 the corresponding wave length ( e of 1 2 is relatively very kcal./mole. small a t the wave length region studied). Figure 3 The System IBr NaBr HC104.-This system shows a plot of the results a t 2 i U mp (Amax) according was investigated in order to determine the spectrum to Scott's equation. By the method of least squares, and thermodynamic properties of IBr2-. HC10, the following parameters were derived a t 20': K I % B ~ - (0.02 M ) was introduced t o suppress the hydrolysis of = 10.5 M-', emax = 4.4 X lo4 M - l cm . . Thevalues IBr. Job's method5 applied to this system (Fig. 2) of K I * B ~derived by the two methods are thus in indicates the formation of a 1 : 1 complex. excellent agreement though the range of ionic strength Scott's method was applied to this system a t, , ,A is very wide. This indicates that the salt effect on ( 2 5 3 . 5 mp). I n the range of concentrations studied K 1 2 ~ is r smaller than what was previously assumed.8 (including t h a t of HClOd), the salt effect on The spectrum of IzBr-, based on the derived value of is very small,6 and so no effort was taken to keep the emax, is recorded in Fig. 1. To determine AHr of the ionic strength constant. complex, a simple method was employed The same The apparent extinction coefficient F,,, is defined as type of measurement which was used for determining the ratio between the observed absorbance and the the spectrum of IzBr- was conducted at several total halogen concentration, as determined by its temperatures. When the concentrations of IZ and final conversion to Is- (see Experimental). IBr Br- are both small (about .If),we can put decomposes to Iz BIZ; furthermore, it appears that the IBr used was mixed with excess of Iz. At low Brconcentration only the IBr is cornplexed, while a t high concentrations the Iz is also converted to IRr2-, and indeed, a relatively high concentration of Br- (>O.l where (I2)0 and (Br-)o are the total concentrations of AT) was required for gmax to reach its saturation value iodine and bromide, Assuming that the extinction though the stability constant of IBr2- is quite large. coefficient does not vary with temperature, the variaThis explains the behavior of the system as displayed tion of K with temperature should be proportional to in Fig. 5 : the extinction coefficient derived from the (7) R.I.. Scott, Rec. trau, c h i i n . , 76, 787 (1956). slope of the line (-3.5 X lo4) is much smaller than that (8) A. E. H. Appelman, J. P h y s . C h e m . , 66, 324 (1961)
+
+
+
EHUDEYALAND AVNERTREININ
4290
Vol. 86
TABLE I1 SPECTROSCOPIC A N D THERMODYNAMIC DATAFOR SOMEPOLYHALIDES IN SOLUTIONS~ hum,,,
Complex
emax, 10'
kcal.
M-1 m
K, M - '
- 1
A H i , kcal.
4 . 4,b3.54,"4.06d 10.5,' 14.gC - l . 3 7 , b -1.48" I2Br106,' 108,' 102d 5 . 5 , * 5 04,' 5 . q d 480,b 444' -10.8' IBr2-113,bs"112d IBrC1122.5,b 121d 3 .8 i e 43.5Q 1x1-h 115 1. 0 5 +1.0 Present work (20'). See ref. 2, 25'. For some early data see "Stability Constants," Part 2, Chemical Society, London, 1958, f See ref. 8 (21'). d A . I. Popov and R. F. Swensen, J . .4m. Chem. Soc., 77, 3724 (1955) (in acetonitrile). e See ref. 3. 0 J. H. Faull, See ref. 1. J . A m Chem. Soc., 56, 522 (1934) (25').
obtained a t high concentration of Br- ( 5 . 5 X lo4) and it varies with temperature. However, the derived value of K I ~ r 2(determined from the ratio between the slope and intercept of the line) does not depend on the concentration of IBr. At 20' K I B ~ *= - 480 M-l, Figure 4 includes the van't Hoff plot for IBr2-, from which we derived A H 1 = - 10.8kcal. I
I
I
I
As a check for the validity of our results and interpretation, we constructed the theoretical absorption band for an arbitrary mixture of Iz and Br- a t 20'. The following equations were used
+ BrI2 + BrIBr + BrI2
I
z z
(Iz)o= ( 1 2 )
-
0.10 -
1
2Cl
I
252
I
I
I
260
268
27 6
I2BrIBr
+
+
The System IBr NaCl HC104.-This system was studied for obtaining the spectrum of IBrC1and its comparison with that of IBrZ-. The spectrum was measured against a reference solution containing the same concentrations of HClOl (10 M ) and IBr. It is recorded in Fig. 1 on an arbitrary scale.
Discussion Table I1 summarizes the present and some previous results. Our result for K I B ~ *is- in good agreement with some previous results, which were derived by a distribution-between-two-phases method. On the other hand, our result for 12Br- is appreciably lower than those derived before by a few methods, including spectrophotometric. This is probably because all early works ignored the presence of IBr2- in the 1 2 f Br - system.
=
10.5
K
=
2.2 X
(ref. 8 )
K = 480
IBr2-
+ (I2Br-I + (IBrz-) + ( I B r )
+
(since (I-) = (IBr2-) ( I B r ) ) . Assuming t h a t (IBr) = 0, the equations can be easily solved to give the concentrations of the light absorbing species, LBr- and IBr2-. The absorbance a t any wave length cIBr2-(IBr2-). is then simply D = q9Br-(IzBr-) Figure 6 presents the theoretical and experimental spectra for a mixture of 0.03 M NaBr, 3.03 X low5 M 1 2 , and 0.02 M HC10, (to prevent hydrolysis). The theoretical band is more intense by about 10% than the experimental one, b u t both display nearly the same Amax (263 mp). This seems to us a rather satisfactory agreement. It was already shown' t h a t 12Br- may be considered as a charge-tiansfer complex and its spectrum to originate from an internal electronic transition of the donor-acceptor type
+
hv
12-BrFig, 6,-The theoretical and experimental spectra of a solution containing 3 x 1 0 - 2 M NaBr, 3.03 x 10-6 M I?, and 2 x 1 0 - 2 M HC104 at 20".
+ I-
K
+12--Br
A
interpretation may be given to the structure and spectra of IBr2- and IBrCl-, where the acceptor is probably IBr, and indeed the difference between the transition energies of these two complexes (-9 kcal.) is the same as between that of IzCI- and IzBr(see Table 11). Considering the transition energies of the two pairs 12Br.--IBrz- and 12CI--IBrC1-, it appears t h a t the vertical electron affinity of Iz is ca. 7 kcal. larger than t h a t of IBr. [A somewhat larger difference (-10 kcal.) results from the spectra of the complexes of 1 2 and IBr with pyridine, 2-picolinej and 2 , 6 - l ~ t i d i n e . ~Thus, ] in both IBrz- and IBrCl-, the halide ions (Br- and C1-) are attached to the same atom, probably the iodine atom of IBr which acts as its positive pole. This was already suggested before from a free energy consideration.10 X-Ray study of solid IBrC1- salt has already shown t h a t the C1ion is attached to the iodine atom. (9) A. I. Popov and R . H. Rygg, J. A m . Chem. Soc., 79, 4622 (1957). (10) K . I,. Scott, i b i d . , 7 6 , 1550 (1953). We are indebted to t h e (11) K.C. I,. Mooney, Z. K r i s l . , 98, 322 (1937). referee for drawing our attention t o the X-ray work.
