Molecular Compounds and Their Spectra. IV. The Pyridine-Iodine

C. Reid, and R. S. Mulliken. J. Am. ... Sebastian M. Walter , Florian Kniep , Laxmidhar Rout , Franz P. Schmidtchen , Eberhardt Herdtweck , and Stefan...
0 downloads 0 Views 679KB Size
JOURNAL OF T H E AMERICAN CHEMICAL S O C I E T Y (Registered in U. S. Patent Office)

VOLUME76

(Copyright, 1954, by the American Chemical Society)

AUGUST 9, 1954

[CONTRIBUTION FROM

THE

NUMBER 15

LABORATORY OF MOLECULAR STRUCTURE AND SPECTRA, DEPARTMENT OF PHYSICS, THE UNIVERSITY OF CHICAGO]

Molecular Compounds and Their Spectra. IV. BY C. RE ID^^

AND

The Pyridine-Iodine System1

R. S. M U L L I K E N ~ ~

RECEIVED JANUARY 27, 1954 The visible and ultraviolet absorption spectra of dilute solutions of iodine plus pyridine in heptane have been studied, and the existence of an equilibrium with a 1:1 molecular complex Py.12 (“outer complex”) was demonstrated [K = 290 a t 16.7’, where K means (Py.Iz)/(Py)(Iz)]. The corresponding changes in heat content, entropy, and free energy (at 17’) in formation of the complex were determined t o be -7.8 kcal./mole, 15.5 cal./deg. mole, and -3.3 kcal./mole, respectively. The location and intensities of the 1% band (A,,, 4220 .&., emaX 1320) and of the charge-transfer band (A,,, 2350, emax 50,000) of Pya12 were determined. The X 4220 band shifts gradually, and increases in intensity, on adding more pyridine to the afbrementioned heptane solutions, until for pure pyridine solutions it has reached about X 3890, with emSx 2120, provided the solutions are not too dilute in iodine. These changes can most probably be attributed to a somewhat increased polarity and stability of the Py.12 “outer complex” in the polar solvent pyridine than in the non-polar solvent heptane. There is no evidence of the presence of the “inner complex” (PyI)+I- in more than small concentrations, but conductivity studies by Kortiim and Wilski indicate that appreciable small concentrations of its ions ( P y I ) + a n d I - are present in pure pyridine solutions of iodine. Additional studies in oery dilute solutions of iodine in pyridine show further interesting spectroscopic changes, which are discussed, but we feel that further experimental study will be needed using extreme precautions toward exclusion of side-reactions, moisture or impurities.

-

Introduction and Survey Recent studies have confirmed older ideas that in its violet solutions, iodine exists essentially free, but that in its brown solutions i t forms 1:1 molecular complexes with the solvent.3 The strong visible absorption of Iz vapor with maximum at 5,200 is essentially unchanged in ~ ~ v i o l e solvents, t~t but in solutions where i t forms complexes this peak is shifted toward shorter lengths; this accounts for the altered color. I n addition, a new very intense peak characteristic of the complex, first noted by Benesi and Hildebrand for aromatic at shorter wave lengths, usuallY solvents, in the ultraviolet. The interpretation of this new peak as a charge-transfer spectrum has Proved important for a clearer understanding of the ekCtrOniC structure of these complexe~.~ (1) This work was assisted by the Office of Ordnance Research under Project TB2-0001(505) of Contract DA-11-022-ORD-1002 with T h e University of Chicago. (2) (a) On leave of absence from T h e University of British Columbia, 1952-1953. Present address: Department of Chemistry, T h e University of British Columbia, Vancouver, Canada. (b) On leave of absence from T h e University of Chicago, 1952-1953; Fulbright Research Scholar a t Oxford University, 1952-1953. (3) See R. S. Mulliken, (a) THISJOURNAL,73, 600 (1950): 74, 811 (1952); (b) J. Phys. Chcm., 56, 801 (1952), for quantum-theoretical interpretation of molecular complexes and their spectra, and a comprehensive review, These are I, I1 and I11 of the present series.

There is evidence4-’ that iodine forms especially stable complexes with pyridine and related cornpounds. Waentig4 reported golden crystals, which he attributed to Py.12, crystallizing from a saturated solution of iodine in pyridine. From heats of solution Hartley and Skinner3 estimated the heat of formation of Py.12 in solution to be about 7.95 kcal./mole, much larger than for other types of iodine complexes. Similarly, the enhancement of the dipole moment in the formation of Py.12 is exceptionally large.B Further, the change in the infrared spectrum of py when it goes into py,12 is much greater5 than the corresponding effect in the case of complex-forming solvents of other types. Audrieth and Birr8 reported that so~utionsof iodine in pyridine show high electrical conductivities, which increase with tirne to asymptotic (4) (a) P. Waentig, Z . physik. Chem., 6 8 , 513 (1909); Chatelet, A n n . chim., [ l l ] 2, 1 2 (1934): H. Carlsohn, Z . angew. Chem., 45, 580 (1932); 46, 747 (1933). (5) K. Hartley and H. A. Skinner, Trans. Faraday SOC.,2 6 , 621 (1950). (6) Y. K. Syrkin and K. M. Anisimowa, Doklady Akad. Nnuk. SSSR, 69, 1457 (1948); G. Kortiim, J . chim. phys., 49, C127 (1952); G. Kortiim and H. Walz, Z.Eleklrochcm., 67, 73 (1953). (7) D. L. Glusker, H. W. Thompson and R. S. Mulliken, J . Chem. P h y s . , 21, 1407 (1953), and references given there; also further unpublished results of Mr. Glusker. Also W, Luck, Z. Elekfrochcrn., 69. 870 (1952), especially table IV. (8) L. F. Audrieth and E. J. Birr, THISJOURNAL, 66,GO8 (1933).

3869

C. REID A N D K . S.~IUI.I,IKRN

3870

values. According to them the molar conductivity based on 1 2 is so high in dilute solutions that i t cannot be explained by simple dissociation into I + (or P y I + ) and I . ’I‘hey suggested instead thc formation of a ternary electrolyte Py.12

PY++

+ 21-

However, recent work of Kortiim and Wilski,9 using very great precautions to keep moisture excluded, indicates that iodine in freshly prepared solutions in pure pyridine at concentrations in the neighborhood of 10F molar gives only a small conductivity, though larger than for most iodine complexes.’O They find, however, that this increases with time, and attributes the effect to a slow iodination in the ring; the effect is strongly catalyzed by platinum sponge. Kleinberg, Yandern‘erf and associates have made a spectrophotometric investigation of solutions of iodine in pyridine (also in quinoline). They too conclude that a very slow iodination in the ring occurs; this should liberate I- ions, which may form Is- ions with 12. Mulliken1?in 1932 suggested that when 1: is dissolved in pyridine the following should be considered as the primary reactions

+

~y I: “outer complex” fast ~ y . 1 2I_ ( P ~ I ) + I“inner coinpleu”

PY

(I)

12

(PyI)+I-

PyI+

(2) (3 1

+ I-

Vol. 76

by a spectrophotometric investigation, first, of equilibrium (1) at varying low concentrations of Py and I? in a non-polar solvent medium; second, of the combined cyuilibria ( l ) , (2) and ( 3 ) in a polar medium (perhaps pyridine itself, or preferably a different polar solvent). These two phases of the present work are reported in sections I and I1 below. I n section I equilibrium (1) was successfully studied in heptane solution. The visible iodine band of the outer complex Pya12 was located a t the exceptionally strongly shifted position of X 4,220 (for free iodine it is at X 5,200), and the expected charge-transfer band a t X 2,350. The equilibrium constant for (l), and the heat of formation of Py.12, were determined. This work confirms other indicat i o n that ~ ~ Py.12 ~ ~ is an exceptionally tightly bound outer complex. Taking into consideration the observed dipole moments4 of P y (2.28 D )and of Py.12 (4.5 D),and assuming a geometrical ~ t r u c t u r e ’ ~ somewhat as shown in Fig. 1, one can estimate that the outer complex Py.12 may easily have as much as perhaps 257, dative character. That is, in the type of formulation given by Alulliken3 +(PY.IZ) = a+o(Py,Iz) no-bond

+ blC’1(PY+-I2-1 dative

(4)

with a? 0.75, b 2 = 0.25. I n eq. 4, because of the asymmetry (Fig. 1) and unusual strength of the complex, the dative function $1 may be already ap+I -

proximately of the structure C5H5N - I with the The “outer complex” Py.12 in (1)would be a molec- X + bonded to one I atom nearly in the P y plane ular complex of the usual type. The “inner com- (N-iodopyridinium ion) leaving the other I atom as plex” in (2) would be an essentially ionic structure an I- above the plane.I4 -4n outer complex with (N-iodopyridinium iodide). I t was suggested that, an exceptionally large amount of dative character in iodine solutions in pyridine, the pyridine has a may well account for the fact7 that complex fordouble role, acting as an electron donor toward 1 2 mation causes greater changes in the infrared specin reaction ( l ) , and as a polar medium in assisting trum in the case of P y than for any other knolm cases (except the related picolines). reactions ( 2 ) and ( 3 ) . \Vhen the work reported in section I1 was underThe present research was undertaken in the hope of studying these two roles of pyridine separatelv taken, i t was with the thought,12 suggested by the conductivity studies of Audrieth and Birr,s t h a t in T pure Py, acting as a polar medium, (a) equilibrium for reaction (2) lies almost completely to the right; but (b) the reaction proceeds only very slowly, over :i high potential barrier; and that as fast as (PyI) +I - is formed, reaction 3 proceeds largely to the right. However, the recent work of Kortiim and Wilski7 indicates that ions PyI+ and I- are formed a t once in I1 solutions in Py, in definite relatively small equilibrium concentrations, and that a later slow increase in ionic concentration is due to slow sidereactions. Taken in connection with our spectrophotometric results in sections I and IIA and the Fig. 1. discussion presented in IL4, the work of Kortiiin (9) G. Kortiim and H. Wilski, Z . p h y s i k . Chenz., 202, 35 ( l ! l 5 3 ) . and Wilski indicates that in the absence of side-reSee also Kortum, ref. 6. actions most of the iodine would remain as Py,12, ( I O ) They find an ionic dissociation constant (PyI L)(I-)/(Py.12) of but that a small portion of i t has a t once undergone about 4.6 X 10-8, which corresponds to about 2% ionization a t 10-6 molar iodine. This may be compared with 1.2 X 10-11 for (H201) +. reaction (2) followed by (3), or else perhaps the di( I -) I(HrO.12) as determined by R. P. Bell and E. Gelles [J. Chem. Soc., rect ionization 2734 (1951)l and smaller values (see ref. 9) for the benzene and difast

,,/\ \

oxane complexes. However. i t seems not impossible that some of the alcohols may have larger values [ c f . L. I. Katzin, J . Chem. Phys., 21, 400 (19S3) 1. (11) (a) R . Zingaro, C. A. VanderWerf and J. Kleinherg, THIS J O U R N A L , 73, 88 (1951); (b) J. Kleinberg, E. Colton, J. Snttiznhn :ind C. A. VnnderWerf, ibid., 75, 447 (1953). (12) Rrferencr X a , p. 818: rpf, 311,p p . X I ? , X I 0

(13) This is based on general consideration? previously advanced by hfulliken (ref. 3). (11) T h e Py would then be acting as an n donor in the terminology o f ref. 3b. However, t o a slight extent, it probably acts ?irnultan?ously :I$ a =-donor (like benzene in its iodine complex; cf. forltnote 43 ( 1 1 1 page 818 of ref. 3a. $1 in rq. 4 would then involve R mixtiire of mciilily ?? with R littlr T donor action b y the P v

SPECTRA OF

Aug. 5, 1954 Py.12

PyI+

+ 1-

THE

PYRIDINE-IODINE SYSTEM

3871

(5)

Further discussion will be given in section IIA. Experimental C.P. pyridine was refluxed with chromium trioxide for several hours t o remove traces of picolines, dried by NaOH, and distilled from magnesium perchlorate. C.P. iodine was sublimed and kept in a desiccator. Solvents were purified by the methods described by Potts.16 Absorption measurements were made in a Beckman spectrophotometer, using lo-, 1-, 0.0296- and 0.0109-cm. cell thicknesses. Apart from the use of cells with fairly well fitting lids, no precautions were taken to avoid moisture uptake during a run. No lids a t all were possible in experiments using spacers to decrease cell thickness.

I. The Py.12 Complex in a Non-polar Solvent The equilibrium (1) was studied in very dilute 12) in heptane (>99.9Oj, by solutions (