Near-infrared spectroscopic study of the interactions between water

COMMUNICATIONS. TO THE EDITOR ... Dimer formation in our system was observed and compared in ... tion are maximized in the cc14 system, viz., low dicl...
0 downloads 0 Views 493KB Size
COMMUNICATIONS TO THE EDITOR

449

volving sulfonated phthalocyanines in water,6 a similar mechanism was assumed to be operative in thesc systems. Monomer-Dimer Equilibrium. From each set of data, the pure monomer spectrum, pure dimer spectrum and the equilibrium constant can be calculated using a previously reported computer procedure.’ For each total concentration ct, the monomer conccntration cm, dimer concentration c d , and equilibrium constant K e , was found. The best fit was obtaincd at equilibrium constants of Keq = (1.58 0.09) X lo41. mol-‘ for the dye-benzene system and K,, = (2.97 0.02) X lo6 1. mol-’ for the dye-CC14 system. The best fit monomer and dimer spectra are shown in r’‘1 gure 2. The results of the analyses are tabulated in Table I. The equilibrium constants for the phthalocyanine in CC14 and benzene are in qualitative agreement with equilibrium constants reported for sulfonated phthalocyanines in ~ a t e r and ~ . ~the chlorophylls in Cc14.‘ The equilibrium constants obtained for these molecules were also realized by using the observed changes in the long wavelength electronic absorption bands as a function of concentration in the 10-6-10-4 M range. For example, the monomer-dimer equilibrium of the tetrasodium salt of 4,4”4”,4”’-tetrasulfophthalocyaninecobalt(I1) has been investigated by equilibrium spectrophotometric measurements5 and by stopped-flow relaxation techniquesg in aqueous solution. The equilibrium constant was found by both techniques to be (2.05 f 0.05) X lo5 1. mol-’. Sauer and Lindsaya have reported the dimerization of three chlorophylls in CC14 and found the association constants to be ca. 1 X lo‘ 1. mol-’. Dimer formation in our system was observed and compared in solvents other than CC1, and benzene. This was accomplished by measuring the ratio of the solution optical density a t 14,700 cm-’ relative to the 16,000 cm-1 density a t nearly equivalent total dye concentrations (ca. 2 X M ) . The order of decreasing phthalocyanine-dye dimerization in several solvent

*

*

WAVELENGTH Irng I

Figure 3. Comparison of melt cast solid-state absorption spectrum of dye with the benzene solution dimer spectrum.

systems is CC14 > benzene > tjoluenc > chloroform > dioxane > D H F > THF, etc. It is probably pertinent to note that the aggregation tendency of the dye is diminished in the solvents yielding the greatest dye solubility and, in general, the largest diclectric constant. To form the dimer, the dyc-dye interaction must be strong enough to overcome any other forces which would favor solvation of the monomer. Thus, the lower the diclectric constant of the solvent, the loss the screening of the dye-dye interaction by the solvent. Relative to benzene, the dye-solvent interaction is probably weaker in carbon tetrachloride. Thorefore, the conditions for maximum dye association or intcraction are maximized in the cc14 system, viz., low diclcctric constant and low solvation of the dye. By inspection of I’igure 3, it can be concluded that the solid state is also composed of Phthalocyanine molecules acting via similarly structured dimeric pairs. Thus, molecular interactions operative in the solid state can be simulated in solution if the solvent properties and structure-solubility characteristies of the dye are matched properly.

Acknowledgments. Stimulating discussions with Drs. D. F. Blossey and 1’1. S. Walker are acknowledged with pleasure. (9) 2. A. Schelly, R. D. Farina, and E. M. Eyring, J. Phys. Chem., 74,617 (1970).

C O M M U N I C A T I O N S TO THE E D I T O R

Near-Infrared Spectroscopic Study of the Interactions between Water and Acetone Publication wsts borne completely

by The

Journal of

Physical Chemiatry

Sir: McCabe, Subramanian, and Fisher have recently published “A Near-Infrared Spectroscopic Investigation

of the Effect of Temperature on the Structure of Water.” In order to support their interpretation, they have cxamined the near-infrared spcctra of water-acetone mixtures. We had studied absorption spcctra, betxvecn 1000 and 11000 cm-I, of H20, DsO, and HOD a8 frcr (1) W. C. McCabe, S. Subramanian, and H.1.‘. Fisher, J. PhUs. Chem., 74,4360 (1970). The Journal of Physical Chemistry, Vol. 76, N o . 8, 1972

450

COMMUNICATIONS TO THE EDITOR I

Id500

A

I

18500

I

I

15000



IlOOO

Xtir

idooo

9’500

7078 5246

p,

10326

r

i’F”

i \ l

s 16650

5

\ 1050s

1

i,

’-‘e,

,.‘

p’

10503

-d.-

-

0 I ,

i 1. b

0 5283

* d \

.a

0

1 3500

,

A

I

3700

I

6400 6400

I

I

6800 800 P

L

I

I

7200

I

J (cm-’1

10000 I

10500 I

D

Figure 1. Spectra of H20 molecules diluted in mixtures acetone-carbon tetrachloride. In Figures B, C, and D, the abscissas are linear with the wavelengths; however the maxima of the bands are expressed with cm-l. The ordinates on the left of the spectra correspond to 0.1 or 0.01 unity of absorbance for the curvesor -.-, respectively. The dashed spectra are doubtful on account of the absorption of the solvent. Acetone molar fraction in the mixture acetone-carbon tetrachloride: a, 0; b, 0.065; c, 0.13; d, 0.26; e, 0.57; f, 1; in Figure D (f), acetone-& has been used. Figure A: HzO concentration ( M ) : a, 0.008; b, 0.01; c, 0.02; d, 0.04; e, f, 0.2; length: of the cell (cm): a, 3; b, 0.5; c, d, 0.2; e, 0.0383; f, 0.0217. Figure B: HtO Concentration ( M ) : a, 0.002; b, 0.02; c, 0.05; d, 0.04; e, 0.03; f, 0.2; length of the cell (cm): a, 100; bl, 10; bz, 5; c, 3; d, 5; e, 1; f, 0.1. Figure C: HzO concentration ( M ) : a, 0.007; b, 0.01; c, 0.05; d, 0.12; e, f, 0.2; length of the cell (cm): a, 100; b, c, d, e, 10; f, 1. Figure D: HzO concentration ( M ) : c, 0.05; d, 0.12; e, 0.5; f, 1; length of the cell (cm): c, d, e, f, 10.

molecules and in interaction with bases in dilute solut i o n ~ . ~Our ,~~ overall experimental evidence seems a priori in contradiction with their interpretation of the 7075-cm-l absorption band of water in acetone. The aim of this paper is a demonstration of this statement from the spectra of ternary mixtures water-acetonecarbon tetrachloride. The spectra have been recorded using, under 4000 cm-I, a Beckman ir 12 spectrometer and, above 4000 cm-l, a Cary Model 14 equipped either with a 0-2 or a 0-0.2 absorbance potentiometer. The spectral slitwidth was less than one tenth of the band half-width. The stabilization temperature of the solutions in the beam was about 30”. The precision of the wavenumbers measured at the maxima of the absorption bands is approximately 1 cm-l for narrow bands and 5 cm-1 for The Joumal of Physical Chemistry, Vol. 76,N o . 3, 1978

broad bands. The shape and frequency of some of the bands vary with water concentration. Solutions have been prepared with spectroquality acetone and carbon tetrachloride from Prolabo. These solvents have been dried over 4 A Molecular Sieves and the mixtures made in a drybox in order to control the water content of the solutions. We have also used acetone-da (99.7%) from C. E. A. France, but we were unable to study ternary solutions (H20-acetone-dscarbon tetrachloride) because of a fast exchange, probably catalyzed by trace amounts of HC1 in carbon tetrachloride. (2) A. Burneau, “ThBse de troisiQmecycle,” University of Paris VI, Paris (1970) ; A. Burneau and J. Corset, Chem. Phys. Lett., 9,99 (1971) (3) (a) A. Burneau and J. Corset, J . Chim. Phys. Physicochim. Bbl., 69, 142 (1972); (b) ibid., 69, 153 (1972); (c) ibid., 69, 171 (1972). I

COMMUNICATIONS TO THE EDITOR I n the coursc of a systematic study of overtone and combination vibrational transitions of water interacting with bases of different strcngths, we have examined, among others, the case of acetone. The spectra of Figure 1 can be interpreted without difficulty with thc help of a detailed analysis of similar spectra related to interactions HzO-hexamethylphosphorotriamide (HMPT).3a It can be seen that the spectra of the water molecule at low concentration in mixtures of acetone and carbon tetrachloride are clearly dependent on the molar fraction of the mixture. The evolution of the spcctra is explained by equilibria between different species: the water molecule quasi free in pure carbon tetrachloride (curvcs a) yields mostly a 1-1 complex when acetone is added up to a molar fraction of 0.25 (curves d) . When acetone concentration is still higher a 1-2 complcx appears and becomes predominant at a molar fraction of 0.57 (curves e). This 1-2 complex is practically the only species present when water is diluted in pure acetone (curves f ) . From the study of the stretching fundamental vibrations, different aut h o r ~had ~ already reached such conclusions for water interacting with different types of organic bases. The infrared spectra of the water molecule in carbon tetrachloride solution (curves a) have already been disthe water molecule is considercd as “quasi free” and is only submitted to weak nonspecific interactions. It is still able to rotate as indicated by wings on the sides of the bands, the one at higher frequency being more intense.3a When acetone is added (curves b and c) the absorpLions due to free molecules disappear and are replaced by new ones due to the 1-1 complex H-O-H. . B (B stands for an acetone molecule). The relatively narrow bands are mainly related to the free OH bond:5 ~ ( 0 0 1 )at 3689 cm-l, ~(011)at 5288 cm-l, 4021) and ~(002) at about 6850 and 7150 cm-l, ~ ( 0 0 3 )near 10500 cm-l. The wider bands are mainly due to the hydrogen bonded OH: ~(100)at 3530 cm-I and ~(110) at, 5139 cm-l. The decoupling between the two vibrators, however, is less pronounced than in the 1-1 complex with a strong base like HMPT.aa,b -From an acetone molar fraction of 0.25 (curves d to f ) some new rather wide absorptions are due to the formation of a 1-2 complex B. . .H-0-H. . . B at the expense of the 1-1 Complex, as can be Seen from the intensity decrease of the narrow bands. Moreover the shoulder towards 5250 cm-I shows that a slight amount of 1-2 complexes might be present from an acetone molar fraction of 0.13 (curve e, Figure lB).38. At an acetone molar fraction of 0.57 (curves e), the bands at 3685, 5286 and 7139 cin-l on the one hand, and at 3617,5247 and about 7060 cm-l on the other hand, show clearly the coexistence of 1-1 and 1-2 complexes. In pure acetone (curves €), the second group of absorptions is the only one left, due to the almost exclusive presence of 1-2 complexes. The question is whether the whole

451 spectrum (f) must be related only to interactions between water and acetone. For instance, Karyakin, et al., have assumed that, for relatively high water concentrations, the band at 7078 cm-l is due to HzO-acetone interactions and those near 6852 and 6509 cm-l to H20-H20 complexes, called “water in quasi emulsion,” involving an indeterminate number of molecules.6 The shape of the spectrum is as a matter of fact dependent on the whole water concentration. Nevertheless at the low concentrations used in Figure 1 (0.2 M ) it is likely that the transitions of spectrum (f) correspond to 1-2 complexes between water and acetone, with the exclusion of more complex aggregates. The bands at 6852 and 7078 cm-I are assigned to ~(021)and ~(101) in agreement with Greinacher, et Q Z . , ~ but the whole region 6400-7500 cm-l (Figure 1 C) needs a more detailed s t ~ d y . ~In~the , ~other ~ regions the bands at 3540, 3618, and 5246 cm-I are due to ~ ( 1 0 0 ) )~ ( 0 0 1 )and ~ ~(011)for HzO in the 1-2 complex; the shoulder near 5150 cm-I could correspond to ~(110). Figure 1 also shows that the more the sum of the quantic vibrational numbers of the final state increases the easier it is to detect the free vibrators, as we have already observed.aa It can indeed be noticed that the ratio of the extinction coefficients of the bands corresponding to the free and associated vibrators varies with the considered spectral regiona8,9 For instance, for an acetone molar fraction 0.57 (curves e), the narrow band at 3685 cm-’ is much less intense that the neighboring group of absorptions, whereas the bands at 5247 and 5286 cm-1 have a similar absorbance; finally, in the region of the first and second overtones of the stretching vibrations (Figure 1 C and D), the characteristic bands of the free vibrator clearly predominate. This is why the sharp change of slope near 10500 cm-I, displayed by the absorption of water in acetone-de

(4) P. Saumagne and M. L. Josien, Bull. SOC.Chim. Fr., 813 (1958); G. V. Yuknevich, A. V. Karyakin, and A. V. Petrov, Zh. Prikl. Spektrosk., 3, 142 (1965); S. C. Mohr, W. D. Wilk, and G. M. Barrow, J. Amer. Chem. Soc., 87, 3048 (1965); D. N. Glew and N. 5. Rath, Can. J. Chem., 49, 837 (1971). (5) Cold transitions are written v(m, 82, 4, quantic vibrational numbers between brackets being those of the final state. The indexes 1 and 3 are used for stretching vibrations, 2 for bending.as (6) A. V. Karyakin, A. V. Petrov, and Yu. Gerlit, Vodorodnaya Svyaz, 181 (1964); Chem. Abstracts, 62,4783e (1965); A. V. Karyakin and A. v. Petrov, Sovrem. Metody Anal., 185 (1965) ; Chem. Abstracts, 649 7367d (1966). (7) E. Greinacher, W. Lilttke, and R. Mecke, Ber. Bunsenges. Phys. Chem., 59, 23 (1955). (7a) NOTEADDED IN PROOF. Since we have submitted this communication, we have shown that the absorption a t about 6500 em-’ is not to be assigned only to HzO molecules: it corresponds to a combination transition simultaneously involving the vibrations of hydrogen-bonded water and acetone molecules [A. Burneau and J. Corset, J. Chem. Phys., 56, 662 (1972)l. (8) Analogous observations about the intensity of overtones have been done by different authors from the study of diatomic g r o u p ~ . ~ ~ g (9) W. Ltkttke and R. Mecke, Z. Elektrochem., 53, 241 (1949); R. Mecke, Disc. Faraday Soc., 9, 161 (1950); M . Couzi, “Thhse de troisihme cycle,” University of Bordeaux, Bordeaux (1966) ; C. Bourderon and C. Sandorfy, 64th Can. Chem. Conf., Halifax (1971).

The Journal of Physical Chemistry, Vol. 76, No. 3,1979

452

COMMUNICATIONS TO THE EDITOR

(curve f , Figure 1D) could correspond to traces of 1-1 complexes. The spectra of ternary solutions water-acetone-carbon tetrachloride show that, for acetone molar fractions lower than 0.25, water and acetone interact essentially by following a 1-1 stoichiometry. On the contrary, in pure acetone, water at low concentration is almost totally involved in 1-2 complexes and the band around 7078 cm-’ must be related to such complexes, contrary to the hypothesis of McCabe, et al. These authors have indeed made an erroneous extension to water-acetone binary solutions of results obtained, by diff erent techniques, with ternary mixtures water-acetone-1,2dichloroethane where the acetone formal concentration was lower than 0.6 M (molar fraction c 0.05).’0

gives’ a shift of -90 cm-’ towards lower frequencies for each of the v1 and v3 vibrations from those for “free” water in CCI,. However, in the overtone region, the shift for the v(101) band for the 1-2 complex from that of “free” water in CClh is only 85 cm-l. This shift is about half the expected value. The anharmonicity factor would not, we believe, account for this drastic reduction. This, taken together with the fact that the intensity of bonded OH is higher than frec OH in the fundamental while the opposite is true in the overtone r e g i ~ nsuggests ,~ that the peak at 7078 cm-’ could be a composite of a sharp peak (at -7075 cm-I) due to a small amount of a 1-1 complex and a small broad peak at -6980 cm-’ due to the major 1-2 complex. We extended the arguments of a 1-1 complex in water-acetone 1,2-dichloroethane mixtures6 to water-inAcknowledgments. We wish to thank Professor M. acetone mixtures to show the spectral contribution at L. Josien for helpful discussions and Rilrs. R. nil. 7075 cm-1 is due to a 1-1 complex but did not deny the Moravie for her assistance with the experimental work. existence of a 1-2 complex for small amounts of water in large amounts of acetone. There was also another (10) T. F. Lin, 8. D. Christian, and H. E. Affsprung, J . Phys. Chem., 69, 2980 (1965); T,F. Lin, S. D. Christian and H. E. Affbasis for assigning the 7075 cm-l Gauwian component sprung, ibid., 71, 1133 (1967). in Figure 5 of our paper to a 1-1 complex. Worley and LABORATOIRE DE SPECTROCHIMIE ANDRBBURNEAU* Klotz,6 in their near-infrared studies of HOD in D20, MOLBCULAIRE JACQUES CORSET assigned the peak at 7063 cm-’ to the overtone of free UNIVERSITB PARIS VI OH. Our water-in-acetone spectra had a major peak PARISVe, FRANCE at 7075 cm-’. Recognizing a small difference due to isotopic substitution and dielectric constant of the RECEIVED SEPTEMBER 22, 1971 medium, we assigned the 7075-cm-’ band to the v( 101) of the 1-1 complex. Reply to “Near-Infrared Spectroscopic Study

of the Interactions between Water and Acetone,” by Burneau and Corset Publication costs assisted by the Veterans Administration

Sir: Burneau and Corset’s studies’ of CC14-acetonewater mixtures in the medium and near-infrared spectral regions demonstrate the sequential formation of 1-1 and 1-2 water-acetone complexes as the mole fraction of acetone is increased. While their assignments are correct in the fundamental region, they are subject to certain doubts in the overtone region. The assignment of the 7163-crn-‘ band to ~(101)is in agreement with other studies.2 They do not consider the passibility of a v(200) band due to the bonded OH in the 1-1 complex while assigning the 7150-cm-l band to ~(002)of the free OH in the 1-1 complex and the 7078 cm-l band to ~(101)of the 1-2 complex. The shoulder at 7050 cm-l observed at moderate acetone concentrations moves toward higher frequencies and appears as a peak at 7078 cm-’ in pure acetone as solvent. The peak at 7078 Cm-’ (Similar to Figure 5 in our paper3) is broad and asymmetric and can be resolved into two Gaussian peaks-one at 7075 cm-’ and another at -6980 cm-1. This is obvious from Figure 5 in our Papers3 In the fundamental region, the 1-2 complex The Journal of Physical Chemistry, Vol. 76, No. 3,107%

(1) A. Burneau and J. Corset, J. Phys. Chem., 76, 449 (1972). (2) D. I?. Stevenson, ibid., 69, 2145 (1965). (3) W. C. McCabe, 8. Subramanian, and H. F. Fisher, ibid., 74, 4360 (1970). 43, 134 (1967). (4) W. A. P. Luck, Discuss. Faraday SOC., (5) T. F. Lin, S. D. Christian, and H. E. Affsprung,J. Phvs. Chem., 69, 2980 (1965); T. F. Lin, S.D. Christian, and H. E. Affsprung, ibid., 71, 1133 (1967). (6) J. D. Worley and I. M. Klote, J . Chem. Phys., 45, 2868 (1966).

VETERANSADMINISTRATION HOSPITAL KANSAS CITY,MISSOURI 64128

S. SUBRAMANIAN

H. F. FISHER*

RECEIVED OCTOBER21, 1971

Bicipital Relaxation Phenomena Publication coats assisted by the National Institutes of Health

Sir: I n investigating the kinetics of metal-ligand reactions using the temperature-jump relaxation method, we have recently observed unusual behavior in the relaxation curves which does not appear to have been



(1) (a) D. B. Rorabacher, Inorg. Chem., 5, 1891 (1966); (b) W. J. MacKellar and D. B. Rorabacher, J. Amer. Chem. ~ o c . ,93, 4379 (1971); (c) D. B. Rorabacher and C. A. Melendez-Cepeda, ibid., 93, 6071 (1971); (d) D. B. Rorabacher and R. B. Crue, Abstracts INOR-43,162nd National Meetingof the American Chemical Society, Washington, D. C., Sept., 1971; (e) F. R. Shu and D. B. Rorabacher, Inorg. Chem., accepted for publication.