Acid-Base Indicators in Methanol. II. The Dissociation of

Publication Date: January 1939. ACS Legacy Archive. Cite this:J. Am. Chem. Soc. 1939, 61, 1, 16-19. Note: In lieu of an abstract, this is the article'...
0 downloads 0 Views 317KB Size
16

I. M. KOLTHOFF AND L. S. Guss

VOl. 61

N-ALKYLSACCHARINS Melting point of derivs. Iodide Bromide Chloride

132 94

132 94

74

74

134

134

Analyses

r

Formula

Found

%N

7.23

CsHyOsSN CsHoOaSN

% C and H Found Calcd.

Calcd.

7.11

C, 50.93 H, 4.64 C, 52.89 H, 5.45 C, 53.21 H, 5.17

51.18 4.26 53.33 4.89 53.33 4.89

C, 54.80 H, 5.72

55.23 5.44

C, 37.78 H, 3.08 110.111 CirHliOaSN C, 61.57 H, 4.11 175.5 C, 52.90 Ci4HiaOsSNz H, 3.52 ' Compounds not previously described. The melting points were taken with a calibrated thermometer.

37.25 2.76 61.54 4.06 52.83 3.17

74

COHi10aSN

CioHiiOiSN

38-39.5 75 81

38 75 81

38

58

58 98 99

58 98

5.81 5.84

CiiHisOsSN CiiHisOsSN CiiHisOsSN CizHisOaSN CioHioOaSN CoHsOsSNBr

rides, nor with ethylene dichloride.

5.85 5.85

S 12.50 6.34

12.60 6.28

give N-alkyl saccharins which are useful for the purposes of identification.

Summary Sodium saccharin reacts with alkyl halides to

DETROIT, MICHIGAN

RECEIVED JUNE 20, 1938

[CONTRIBUTION FROM THE SCHOOL OF CHEMISTRY OF THE INSTITUTE OF TECHNOLOGY OF THE UNIVERSITY OF MINNESOTA]

Acid-Base Indicators in Methanol. 11. The Dissociation of Triphenylcarbinols in Methanol BY I. M. KOLTHOFF AND I,. S. Cuss'

A number of methoxytriphenylcarbinols were prepared by Lund.2 He investigated their properties, while Kolthoff studied and advocated the use of the penta-, hexa- and heptamethoxy derivatives as acid-base indicators. The acid forms of these indicators are red, the alkaline forms colorless. In aqueous medium the indicator equilibrium is represented by the following equation

+

ROH H colorless

+

z R+ red

+ HzO

(1)

In methanol as a solvent, there is the establishment of a n equilibrium between the two basic forms of the indicator ROH CHaOH I _ ROCHs HzO (2) with the equilibrium constant

+

+

k =

UROH

~ R O C H :a&o

I n this equation, a stands for activity. The activity of methanol in the presence of small amounts of water, as well as in anhydrous solutions, can be considered constant. From the above, it follows that the indicator equilibrium of the methoxytriphenylcarbinols in methanol is determined by equations (1) and (4) ROCHs

+HC

__

Rf

+ CHsOH

(4)

When the indicator constant is determined in the usual way,4 we measure KI =

a H + (UROH

+

(I.ROCH8)

aR+

(5)

which will greatly depend on the water content of the methanol. According to equation (1) UH+ UROH

(3)

(1) From a thesis submitted by L. S. Guss to the Graduate School of the University of Minnesota in partial fulfilment of the requirements for the degree of Doctor of Philosophy, May, 1938. (2) H. Lund, THISJOURNAL, 49, 1346 (1927). (3) I. M. Kolthoff, i b i d . , 4B, 1218 (1927).

KRoH

=

-

and, according to (4) ._

(4) I . M. Kolthoff and L. S. Guss, ibid., 60, 2516 (1938).

(7)

Jan., 1939

From ( 5 ) , (6) and (7), i t follows that KI = KROH~ H : O

+ KROCHS

(8)

I n the present study, we have determined the values of KROH, KRocH, and k in methanol. As will be evident from the experimentally determined value of k , the indicator equilibrium in anhydrous methanol may be represented by equation (7), if the total concentration of the indicator is small. The addition of water will tend to convert ROCH3 into ROH, thus it will shift the color of the indicator in a buffer solution to the alkaline side. If r denotes the ratio of alkaline to acid form of the indicator in pure methanol, we find (9)

where concentrations of the forms of the indicator are used instead of activities. I n a previous study,4i t was demonstrated that, in buffered solutions, the indicator equilibria in methanol are unaffected by the addition of traces of water. In a given buffer then, KROCH, and aCHaOHz+ are unchanged. Hence, in equation (9), r will remain constant even on addition of water. Starting with a given concentration of indicator, we determine experimentally the ratio [ROCHsl 4- [ROHI = r f [R+I

From equation (3) we see that r' = r ( l

+ kaB,o)

(10)

Values of r' were determined experimentally a t various concentrations of water. When these are plotted against the water concentration, the intercept will give the value of r and the slope, krfHzO, in which f H 2 0 is the activity coefficient of water in methanol. I n order to obtain a value of KRoH in methanol which will correspond to its value in water, we have chosen pure water as the standard state of water. I n dilute solution, f&O is equal to 0.0761, according to U n r n a ~ k . ~The activity coefficients of the other constituents may be considered to be unaffected by small amounts of water. Hence, by the above procedure, we find the value of k, and of r. From the latter and equation ( S ) , KROCH, can be calculated and, hence, KROH. The indicator constants of penta- and hexamethoxytriphenylcarbinols were determined in pure methanol in the same manner as described in the previous paper,4 the extrapolation method 1.5)

17

DISSOCIATION OF TRIPHENYLCARBINOLS IN METHANOL

A. Unmack, 2. dhvsik. Chem.. 129. 349 (1937).

to zero ionic strength being used for finding the ratio of activities instead of concentrations. It may be added that the values of the various constants do not require the use of entirely waterfree methanol, as by extrapolation of the y' values obtained in the presence of larger excesses of water, the value of r a t a water content of zero is found. Experimental The methods of purification of reagents already have been described. * The indicators were samples furnished by H. Lundai3 while working a t the University of Michigan. Solutions of water in methanol were prepared by weighing the required amount of conductivity water in a calibrated volumetric flask of 50-ml. capacity and making up to the mark. Colorimetric comparisons were made with a Dubosq colorimeter. Pentamethoxy Red.-The indicator equilibrium in water-free methanol was determined in trichloroacetate buffers, the concentration of the indicators in the buffers being 2 X 10-6 M . The results are given in Table I, and graphically represented in Fig. 1.

TABLE I PENTAMETHOXY REDIN TRICHLOROACETATE BUFFER CAoetate

0.0080 .0080 .0080 .0080 ,0080 ,0080 .008 .008 .008 .008 ,008 .008 .008 .008 .008 .008 .008

Salt added

LiCl NaBr LiCl NaBr LiCl NaBr NaBr LiCl LiCl NaBr LiCl

Csdt

0.011 .018 .055 .088 .110 .177 .177 .221 .331 .354 .442

v'ji

0.090 ,090 .090 .090 .090 ,090 .14 .I6 .25 .31 .345 .43 .43 .48 .58 .60 .67

CB/CA

3.6 2.5 1.96 1.60 1.35 1.17 2.5 2.5 2.5 2.5 2.5 2.5 6.2 2.5 2.5 6.2 2.5

CIb/CIs

1.47 0.91 .74 .55 .505 .43 .665 ,665 .55 .37 .38 .41 .695 .25 .22 .43 .183

9K

0.39

.44 .43 .46 .43

.435 .57 .57 .66 .83 .82 .74 .95 1.00 1.06 1.16 1.13

The extrapolated value of pK' is 0.2. As PKA of trichloroacetic acid is 4.9,6 we find for PKROCHa a value of 5.1. The effect of water up to a concentration of 1.8 M was determined in a buffer with ionic strength of 0.0080 (CB/CA = 2.5). The results are given in a graph in Fig. 2, in which r' is plotted against the concentration of water. Hexamethoxy Red.-The indicator equilibrium in pure methanol was determined in salicylate buffers. The results are given in Table I1 and graphically in Fig. 3. From the extrapolated value of pK' of 0.65 and PKA of salicylic acid of 7.9; we find PKRoCHs = 7.25. The effect of salts on the indicator equilibrium was also (6) H. Goldschmidt a n d H. Aartlott, Z . physik. Chem., I l l , 317 (1925). (7) H. Goldschmidt and F. Aas. ibid.. 112. 429 11924)

1. M. KOLTHOFF AND L. S. Gus

1s

Vol. 61

0.6 1.2 1.8 Concentration of water. Fig. 2.-Effect of water on the indicator ratio of pentamethoxy red in trichloroacetate buffer. 0

0.2 0.4 0.6 Square root of ionic strength. pig. 1.-Pentamethoxy red in trichloroacetate buffer. 0

determined in m- and p-nitrobenzoate buffers and was found almost identical with that in salicylate buffers The effect of water up t o a concentration of 1.8 molar was determined in a salicylate buffer and also in a m-nitrobenzoate buffer, the two determinations giving an average value of k in satisfactory agreement. The data obtained in the salicylate buffers are plotted in Fig. 4.

TABLE I1 HEXAMETHOXY REDI N SALICYLATE BUFFER csalioylste

0.0060 .0060 .0060 .010 .006 .006 .a06 .006 .006 .006

.006 .006 .006

Salt added

LiCl NaBr LiCl NaBr LiCl NaBr LiCl NaBr LiCl

./T

csalt

0.011 .022 .055 .062 .110 .122 .221 ,182 .331

CB/CA

Clb/tla

The methoxytriphenylcarbinols act as basic indicators and their behavior should be similar to that of methyl yellow or neutral red. From the general equation PK

$'K'

- IOgfIa

lOgfIb

-h

f B

10gfA (11)

we see that the slope of the curve obtained on plotting @K against the square root of the ionic

fiK

0.078 0.415 1.06 -0.41 .078 .48 1.15 - .38 .078 ,575 1 . 3 - .35 .10 .95 1.82 - .28 .13 .415 0.87 - .32 .195 .415 .60 - .16 .25 .415 .515 - .09 .26 .415 .48 - .07 .34 .415 .41 .01 .36 .415 .41 .01 .48 .415 .31 .13 .435 ,415 .34 .08 .58 .415 ,265 .19

A summary of the values of the various constants calculated from the results is given in Table 111.

TABLE 111 CALCULATION OF CONSTANTS OF METHOXYTRIPHENYL-

Pentamethoxyred Hexamethoxy red

Intercept Slope

0.2 0.4 0.6 Square root of ionic strength. Fig. 3.-Hexamethoxy red in salicylate buffer. 0

CARBINOLS

Indicator

k

PKROCH~PKROH

1 . 0 0.48 6 . 4 5 . 1 1.17 1.10 12.5 7.25

4.3 6.15

Jan., 1939

MOLECULAR STRUCTURE

OF

HYDROGEN FLUORIDE

strength should approach +4, as the ionic strength approaches zero. That this is the case is seen from Figs. 1 and 3. From the values of k, i t is seen that in mixtures of methanol and water, containing a large amount of the latter component, the methoxytriphenylcarbinols are largely present in the ROH form. From the practical point of view, the values of the constant

19

3 .O

2.6

2.2 L

are more interesting than the k values. For the pentamethoxy indicator, k' is equal to 0.48, and for the hexamethoxy indicator, 0.94. Finally, i t may be mentioned that PKRoH increases more when going from water to methanol as a solvent than that of other cation acids. The PK of the latter increases by about one unit.' In water, the PK values of penta- and hexamethoxy red are 1.85 and 3.3,8respectively. In methanol, the pK values increase by 2.45 and 2.85, respectively.

1.8

1.4 1 .o0

1.2

0.6

1.8

Summary

Concentration of water. Fig. 4.-Effect of water on the indicator ratio of hexamethoxy red in salicylate buffer.

1. The effect of salts upon the color of pentaand hexamethoxytriphenylcarbinols in methanol corresponds to that of a system of uncharged basecation acid. 2. In pure methanol, the carbinols are present

mainly in the form gf the methyl ether. The equilibrium constant of the reaction between the methyl ether and water has been evaluated. 3. The values of P K R o c H , and P K R o H of the carbinols in methanol have been determined.

(8) I. M. Kolthoff, THISJOURNAL, 49, 1218 (1927).

MINNEAPOLIS, MINN.

RECEIVED OCTOBER 3, 1938

[CONTRIBUTION FROM THE LABORATORIES OF THE PENNSYLVANIA STATE COLLEGE AND PRINCETON UNIVERSITY]

The Molecular Structure of Hydrogen Fluoridel BY S. H. BAUER,J. Y . BEACH^

AND

J. H. SIMONS

Structural studies of hydrogen fluoride may be vapor density data (over the pressure range 56 to traced back for more than half a ~ e n t u r y . That ~ 760 mm. and temperature range -39 to 88') rests the compound is polymerized in the vapor phase on the assumption of the single equilibrium has been known since the early work of Mallet4 n H F = (HF), Simons and Hildebrand demonstrated that some with n = 6 and of the polymers are of an order higher than tetlog Ks f40'000 - 43.145 ramers,6 for they obtained apparent molecular weights for the gas as large as 87.4. Further, they Although the data could be correlated with the showed that one way of interpreting the available existence of a larger number of species in equilibrium, the simpler hypothesis seemed reasonable in (1) Part of these results were presented to the American Chemical view of the fact that the molecular weight apparSociety at Milwaukee, September, 1938. (2) National Research Fellow in Chemistry. ently approached 120 as the temperature was (3) Two reviews have appeared: J. H. Simons, Chcm. Rev., 8, lowered and that within the experimental error 213 (1931); K.Fredenhagen, 2. Elekfrochem., 87, 884 (1931). (4) Mallet, A m . Chem. J . , 8, 189 (1881). agreement was found with the even whole number ( 5 ) 3. H . Simons and J. H . Hildebrand, THISJOURNAL, 46, 2183 (1924). six for the degree of polymerization. Latimer