Ionization of Strong Electrolytes. VII. Proton Magnetic Resonance and

Chem. , 1959, 63 (1), pp 101–103. DOI: 10.1021/j150571a026. Publication Date: January 1959. ACS Legacy Archive. Note: In lieu of an abstract, this i...
2 downloads 0 Views 326KB Size
PROTON MAGNETIC RESONANCE OF IODIC ACID

Jan., 19Fj9

101

TABLE 111 A COMPARISON OF KINETICDATAFOR THE DECARBOXYLATION OF MALONIC ACID AND TR~CHLOROACETIC ACIDIN VARIOUS AMINES. CHANQE OF AS* AND AH* WITH CHANQE IN SOLVENT Change of solven-

r

From

(1) (2) (3) (4)

Aniline Aniline Aniline Quinoline

To

o-Toluidine 0-C hloroaniline Quinoline 8-Methylquinoline

--A(AH+) Trichloroacetic acid

-700 - 130 -520 - 1670

composition of the trichloroacetate ion than for that of malonic acid. A comparison of the data (references 7, 8 and Table 11) reveals that this is actually true in each case. For the decomposition of trichloroacetic acid in aniline AH* is 2400 cal. lower than for that of malonic acid; in o-toluidine

(ca1.)-

---A(AS+)

Malonic acid

Trichloroacetic acid

-2300

-4.25 -4.78 f4.98 -10.84

- 1200 -320 - 160

----& -.e)( Malonic acid

-2.59 -2.47 +2.09 -8.10

1900 cal. lower; in o-chloroaniline 2200 csl. lower; in quinoline 2760 cal. lower; in 8-methylquinolinc 2100 cal. lower. Acknowledgment.-The support of this research by the National Science Foundation, Washington, D. C., is gratefully acknowledged.

IONIZATION OF STRONG ELECTROLYTES. VII. PROTON MAGNETIC RESONANCE AND RAMAN SPECTRUM OF IODIC ACID BY G. C. HOOD,A. C. JONES AND C. A. REILLY Shell Development Company, Emeryville, California Received August i l , 1965

The proton magnetic resonance shifts and intensities of Raman lines in solutions of iodic acid have been measured and used t o calculate degrees of dissociation. The thermodynamic dissociation constant is found to be 0.18 at 30" in good agreement with the results of classical measurements.

The most satisfactory methods for the determination of the degree of dissociation of strong electrolytes are based on measurements of nuclear magnetic resonance (n.m.r.) shifts or measurements of Raman i n t e n ~ i t i e s ~ l -These ~ methods are free of the problem of ionic interaction which is inherent in classical methods. Comparison of the results obtained by the n.m.r. and by the Raman technique showed excellent The dissociation of iodic acid has now been investigated by both the n.m.r. and the Raman methods. Iodic acid is of special interest since its thermodynamic dissociation constant of about 0.18 is approximately the upper limit of the region where classical methods can also be expected t o yield meaningful dissociation constants of univalent electrolytes.' Thus a comparison of all methods is possible. Experimental The samples were prepared from reagent grade iodic acid which was not further purified. Nuclear magnetic resonance shifts were measured at 40 megacycles per second with a Varian Model V-4300 spectrometer equipped with a sample spinner and a field stabilizer. The magnetic field was swept through resonance by in'ecting a small voltage into the sensing circuit of the field staiilizer The shift of the protons in each sample relative to that or water at 30' was determined as described previously.' One important modification, however, was the use of two ref0. Redlich, Chem. R e m . 39, 333 (1948). H. S. Gutowsky and A. Saika, J . Chem. Phyr., 21, 1688 (1953); Hood, 0. Redlich and C. A. Reilly, ibid., BB, 2067 (1954); BS, 2229 (1955). (3) 0. Redlich and G. C. Hood, Disc. Faraday SOC., 24, 87 (1957). (4) A. Krawetz, Doctoral Thesis, University of Chicago, 1955. (5) G. C. Hood and C. A. Reilly. J . Chem. Phys.. BT, 1126 (1957). ( G ) 0. Redlich, Monatsh, 86, 329 (1955).

erence compounds. The shifts in the more concentrated solutions (down to 1.850 molar) were measured directly with respect to that of water. The shifts in the more dilute solutions were so small that they were difficult to measure directly. For these solutions the shifts were measured relative to that of the ring protons in toluene. The shift of these protons from that of water at, 30' was determined in separate experiments to be 65.5 & 0.1 C.P.S. (to lower applied field). The iodic acid shifts relative to water then were calculated. This procedure allowed the determination of shifts with a precision of 5 X 10-9 of the resonance frequency. The magnetic susceptibility per gram x of each solution was determined by means of a Gouy balance. The density d of each solution also was measured and the bulk magnetic susceptibility then calculated from the relation K = xd (1) The Raman spectra of the solutions were obtained with a standard Cary Model 81 recording spectrometer. Degrees of Dissociation.-The assumptions involved in obtaining degrees of dissociation from magnetic resonance measurements have been described previously.a The degrees of dissociation were calculated from the shifts according to

+

s = Av/40 g Q = + 2 . 6 0 ( ~ 0.700) X lo6

+

+

(2) (3)

~ / p 0191 (1 - a)sa (4) where Av is the measured resonance shifts in c.P.s., g is the bulk magnetic shielding correction, k and -0.700 X 108 are the bulk magnetic susceptibilities of the sample and of water, respectively, s is the total chemical shift, s1 is the shift of HoO+, sz is the shift of undissociated HIOa, a is the degree of dissociation and p is the stoichiometric mole fraction of hydrogen in HsO+ on a total hydrogen basis. The variation of s with (shown in Fig. 1 for lower concentrations) is similar to tgat of nitric acid.2 The straight line portion of the curve (c > 1.85, l p > 0.052) is attributed to the equilibria between undissociated monomer and associated species. The curved portion of the curve (c < 1.85, p < 0.052) is attributed to dissociation. The situation can be seen more readily in Fig. 2 where s / p is shown as a

102

G. C. HOOD, A. C. JONES -4ND C. A. REILLY

Vol. 63

TABLE I DISSOCIATION OF IODIC ACID Susceptibility corm.,

f',

moles/l.

P

R.

0.034 .076 ,104 .160 .362 .662 1.020 1.318 1.850 3.292 5.800 7.53 7.53

0.00094 .00198

-0.003 - .006 - .008

.00258 .00427

-

.OlOO .OH3 .0287 .0380 .0520 .0980 .200 .311 .324

-

-

0.3GO

Resonance shift, 8

Degree of dissociation, a (n.m.r.)

0.88 .80 .73 .68 .3Y .22 * 12 .08 0

0.012 .024 .030 .048 ,090

.01 .03 .05 .09 .12 .20 .27 .40 .59 .60

I

14

.20 ,25 .31 .57 1.14 1.67 1.70

...

...

...

...

Sodium iodate s o h

a/c

-1.710 -1.540 -1.460 -1.400 -1.353 -1.307 -1.470

...

... ... ...

... ...

...

log

K6 -0.788 - .841 - .892 .905 -1.138 -1.279 -1.415

-

...

...

... ...

Raman intensity ratio,

R

Degree of dissociation, a (Raman)

2.30 2.18 2.00

(0.88) .86 .81

1.50 1.33 1.23 1.22 1.19

.56 .34 .14

...

...

...

...

...

..

3.69

...

....

...

.10 0

...

... ...

... ...

function of p . This interpretation is substantiated by Ra0 man spectral data which are discussed later. Th? value 13.70 was obtained for s1 by utilizing the expression s1 = lims/p P-0 (5) This value for the shift of H30+is to be compared with the values 13.1, 11.8 and 9.2 determined previously26for HzS04, HNOa and HClO, solutions, respectively. The value 6.00 for sz was taken directly from s at c = 1.85 where the solution ap ears to consist predominantly of undissociated monomeric%IOa (Table I). A small error in this value is not too serious since the degree of dissociation calculated is not particularly sensitive to sz. The experimental data and the degrees of dissociation calculated from equation 4 are given in Table I. The Raman spectra of concentrated aqueous solutions of iodic acid have bands with Raman shifts of 330, 630 and 780, with a shoulder indicating the presence of a band near 825 cm.-I (Fig. 3). As the concentration of iodic acid is reduced, the 780 cm.-l band shifts to 800 cm.-l and as the concentration is further reduced the intensity at 800 cm.-l increases relative to the shoulder at 825 cm.-'. The intensity of the 630 cm.-l band also decreases more rapidly than linearly with concentration below 1 molar. I n the most dilute solutions examined, the spectrum is predominantly a single band at 800 cm.-l. This band is also observed in solutions of NaIOs and may be ascribed to both undissociated acid and iodate ion. The rather weak bbnd at 330 cm.-l persists in Apectra of all acid concentrations as well as in sodium iodate solutions. Typical spectra are given in Fig.

P.

Fig. 1.-Observed

log

chemical shift of iodic acid.

15 -SI

11

4 Pl '. I

3.

7 0

I

I

I

I

I

I

I

0 ,

P.

Fig. 2.-Evaluation

of

SI and s2.

WAVE NUMBER. cm,-'.

Fig. 3.-Raman

spectra of iodic acid and sodium iodate.

The inter retation of the spectral changes with dilution is that the s#hft of the band at 780 cm.-l corresponds to the equilibrium shift of associated iodic acid to undissociated monomer. The subsequent increase in intensity of the 800 cm.-l band relative to that at 825 cm.-l corresponds to the dissociation of iodic acid to iodate ion. These observations differ from the older data of Rad in that at no point in the concentration interval from 7.5 to 0.03 molar is the intensity of the 825 cm.-l band greater than the intensity of the 800 cm.-1 band as Rao reported for 6 to 3 molar solutions. Calculation of degrees of dissociation from Raman spectra requires quantitative intensity measurements which, unfortunately, are affected by a variety of instrumental factors and sample properties.8J I n the present case, the principal sources of error in the relative intensity measurements are the effects of variations in the optical absorption and refractive index of the various solutions. One method of dealing with this situation is to consider that the spectrum of each solution is recorded on a different unknown intensity scale; therefore, although it is not possible to cofnpare the observed intensities from different spectra, the ratios of intensities at given spectral positions may be compared. (7) N. R . Rao, Indian J . Phvs., 16, 71 (1942). (8) H. J. Bernstein and G. Allen, J . Opr. SOC.Am.,46, 237 (1955). (9) A. C. Jones, Doctoral Thesis, University of Chicago, 1955.

PROTON MAGNETIC RESONANCE OF IODIC ACID

Jan., 1959

103

-0.71

0

I 0.2

I 0.4

C,

Fig. 4.-Degree

I

I

0.8

1.0

I

0.6

I 1.2

I

,

4

0

0.2

0.4

Fig, 5.-Dissociatfon

of dissociation of iodic acid.

+ +

YACAJA~~

YICIJISOO

Y A C A J ~ Z ~

YICIJISZ~

0.6 C,

The degrees of dissociation of iodic acid can be computed from the ratio R of Raman intensities a t two spectral positions (800 and 825 cm.-l) at which the relative contributions from HI03 and from IO3- differ appreciably. If one assumes that, after the spectrum has been corrected for the presencC of water and material which may contribute a flat background component, the remaining observed intensity is due only to the contributions from the species HI03 and Io3-, then the intensity ratio for a given solution is

R=

5

1.4

MOLES/^.

(6)

The symbols CA and CI refer to the concentrations of the undissociated acid HI03 and IO3-, respectively. The specific Raman scattering coefficient J represents the intensity of the Raman emission from a solution of unit concentration of the molecular species and a t the spectral position designated by the subscripts. The term y represents the factor to reduce the different intensity scales of the spectra from which the respective J’s were derived to a common intensity scale, namely, that of the spectrum of the solution under consideration. Since CA/CI = (1 a)/a,expression (7) is derived easily

-

The values of RI and RA were obtained from Raman spectra of solutions of NaI03and of the undissociated monomer acid, respectively. It has been assumed here, as in the n.m.r. analysis, that HI03 exists as essentially undissociated monomer in the 1.85 molar solution. The ratio of y’s in equation 7 can only be obtained by the assumption of the value of 01 obtained from the n.m.r. method for one solution. The 0.034 molar solption with an a of 0.88 was used. Having thus obtained the necessary parameters, the degrees of dissociation for the remaining solutions were calculated from the Raman data. As shown in Table I and Fig. 4, the Raman data are not inconsistent with the n.m.r. data but the difficulties inherent in the Raman method in this case preclude a more quantitative comparison. Dissociation Constant.-The thermodynamic dissociation constant K was determined from the n.m.r. data by extrapolation of log Kfi to zero concentration (Fig. 5) Kfi = a/c(l - a ) (8) The symbol p represents the activity coefficient of undissociated iodic acid, a is the activity of iodic and c is the stoichiometric concentration in moles/l. The activities of the solutions were computed from activity coefficients deter-

1

1.0

0.8

,

MOLES/I.

constant of iodic acid

TABLE I1 DISSOCIATION CONSTANT OF IODIC ACID Temp.,

Method

OC.

Freezing point Conductivity Conductivity, freezing point Conductivity Conductivity Solubility Indicatcr N.m.r.

0 18 25

25 25 25 25 30

I

K

Ref.

0.262 .19 .18

11 11 12

.17 .1686 .163 .167 .18

13 14 15 16 This inveatigation

mined irom freezing points.Io The value of 0.18 determined for K may be compared with the values obtained by classical methods (Table 11).

Conclusions The dissociation constant 0.18 determined from n.m.r. data, and supported by Raman data, is in good agreement with previous classical determinations. As su gested earlier,’ a dissociation constant of this agnitude is probably close to the limit of reliability for classical methods with uniunivalent electrolytes. Previous data have shown that dissociation constants near unity (trifluoroacetic and heptafluorobutyric acids) are not obtainable by classical m e t h ~ d s . ~ J ~ ~ ’ ~

Bn

(10) E. Abel, 0.Redlich and P.Hersch, Z. physik. Chem., A170, 112 (1931). (11) E. Abel, 0. Redlich and P. Hersch, ibid., A170, 112 (1934). (12) V. Rothmund and K. Drucker, ibid., 46, 827 (1903). (13) L. Onsager, Physik Z., 28, 277 (1927). (14) R . M. Fuoss and C. A. Kraus, J . A m . Chem. Soc., 6.5, 47G (1933). (15) S. Naidich and J. E. Ricci. ibid., 61, 3268 11939). (16) R. von Halban and J. Brill, H e h . Chim.Acto, 27, 1719 (1944): C. A . , 60, 791 (1946). (17) G. C. Hood and C. A. Reilly, J . Chem. Phus., 28, 329 (1958). (18) T. F. Young and A. C. Jones, Ann. Rev. P h y s . Chem., 8, 278 (1 952).