2681
Acidity Scales in Mixed HpO-CH&N Buffer Solutions
Acidity Scales in Mixed Water-Acetonitrile Buffer Solutions’ Frank Jordan2 The James Bryant Conant Laboratory of the Department of Chemistry, Harvard University, Cambridge, Massachusetts 02738 (Received May 24, 7973)
Spectrophotometric HO and potentiometric pHapp (with a glass electrode) measurements were performed in HzO-CH~CNmixtures containing acetate buffers and mineral acid (HC1). The results indicate that, to a good degree of approximation, the effective p H of the solutions can be calculated from the concentration of the acid in the aqueous portion of the solvent, neglecting the cosolvent altogether. The results with mineral acid indicate that the notion of the chemical hydration of the proton also extends to mixed solvent systems.
In a recent nmr study on the mechanism of the hydrolysis of the enamine ethyl P-cyanomethylaminocrotonate it became necessary to establish an approximate p H scale in aqueous buffer-acetonitrile (AN) mixture^.^ A brief survey of the literature indicated that there had not been any previous attempts to define an Ho scale in such mixtures of AN with pH 4 and 5 buffers. This note is concerned with such a study involving a comparison of hydronium ion activity measured by a glass electrode and uia the HO indicator m e t h ~ d . ~ The major conclusion of this work is that the effective p H of solutions of acids in aqueous acetonitrile can be calculated, to a good degree of approximation, from the concentration of the acid in the aqueous portion of the solvent, neglecting the cosolvent altogether.
According to the definition of activity coefficient ratio)
H,
= pK,
H04 (assuming
one for the
+ log ([I]/[IH+])
(1)
where [I] and [IH+] represent fractions of deprotonated and protonated forms of the indicator, respectively. Since the protonated form does not absorb at 290 nm, Ho for any CH3CX-HzO mixture can be obtained from the formula
H,= pK,
f log
‘CHbC\-hutfer fCHJC\-bdse
-
___ ( 2 )
tCH,C\
buffer
where c ( ~ ~ ~ z ~ is~ bthe~ fmolar f ~ ~ extinction ) coefficient for any mixture with the desired buffer and E(CH3CN-base) is the extinction coefficient for the totally Experimental Section deprotonated form in the same volume per cent mixture. pH and HODeterminations. The p H of all solutions was The pH of each solution was also measured with a conmeasured with a Radiometer pHM 4b p H meter employventional glass electrode and such apparent pH values are ing a combination glass electrode. The HZO-CH~CN quoted along with the Ho values in Table I and Figure 1. mixtures studied exhibited no drift and quick response to Distilled, deionized water was used in the preparation the electrode employed. of aqueous buffers. Commercial, reagent grade chemicals The spectrum of the indicator p-chloroaniline in the were employed without further purification. various solutions was recorded on a Cary E . p-ChloroaniFischer spectrograde CH3CN (stored over molecular line exhibits the following spectral properties:5 sieves) was used in the NO determinations, along with Solvent nm f nm f Eastman highest purity p-chloroaniline. 2 N HCI 0.1 N NaOH
215 5
263
9200
239
360
290
11,700 1,500
Under acidic conditions (0.1 or 1.0 N HCl) no absorption was detected a t 290 nm. This wavelength was chosen for all the subsequent No work. First the pKa of p-chloroanilinium ion was determined spectrophotometrically a t 290 nm a t 30” and 0.1 ionic strength using various aqueous buffers. Since the protonated form has no absorbance a t this wavelength any absorption will be due entirely to the basic form of the indicator. Measurement of the absorption at seven different pH’s between 2.2 and 6.0 gave an average value of 4.05 f 0.05 for the pK, in satisfactory agreement with the literature value of 4.15 a t 25°.6 Next the extinction coefficient of p-chloroaniline in aqueous alkali-CHsCN and dilute mineral acid-CH3CN mixtures was measured a t 290 nm to provide the values for totally protonated and totally deprotonated forms in the mixtures. Finally, the absorption a t 290 nm was measured in the aqueous buffer-CH3CN mixtures of interest.
Results a n d Discussion The HO measurements indicate that addition of acetonitrile decreases the acidity of acetic acid-acetate type buffer systems (Table I). Qualitatively, this agrees with the effect on proton activity of AX addition (AN is a cosolvent of lower basicity and dielectric constant than HzO). The results also parallel those of Bates and Schwarzenbach7 in water-ethanol mixtures. The pKa of anilinium type indicators has been shown to be essentially independent of AX concentration to perhaps 60-7070 (v/v) AN added,s which finding would justify the choice of the indicator in the present study. Vedel studied the behavior of two different functions: a ferrocene-ferrocinium couple and a colorimetric indicator in HzO-CH~CNmixtures with varying amount of HC104 added.g In the mole fraction AN region here employed both functions indicated much lowered acidities with increased AN concentration as is observed in the present study. It was of some interest to compare the No values with The Journal o f f h y s i c a l Chemistry, Vol. 77, No. 22, 1973
Frank Jordan
2682 TABLE I : Ho and p H Scales in HpO-CH$N-Acetate
Buffersa
% C H & N in Mixture
v/v % mol YO t
0
0.0
basic
(0.1 N
0.730
10 3.83 0.727
20 7.93 0.730
0.454 4.12 4.18 4.30
0.595 4.34 4.42 4.69
0.096 3.05 3.11 3.24
0.210 3.25 3.33 3.67
30 12.87 0.730
40 18.68 0.724
50 25.63 0.718
60 34.07 0.718
0.706 4.80 4.99 5.62
0,716 5.06 5.34 5.94
0.716 5.31 5.74
0.555
0.650 4.03 4.31 4.93
0.700 4.28 4.71 5.33
NaOH) pH 3.95 Bufferh
e buffer PHobsd
3.94
PaH* Hoe
0.665 4.58 4.71 5.07
pH 2.92 Bufferd t
buffer
PHobsd
2.92
patr* Hoe
0.371 3.54 3.67 4.06
3.79 3.98 4.53
a Ionic strength in aqueous layer maintained at 0.1; pHobad measured with a glass electrode. Ho as defined in text based on indicator ratio. Acetic acid-acetate buffers. pHobsd corrected with 6 values as given in ref 11. Chloroacetic acid-chloroacetate buffers. e Calculated with a pK, of 4.05 for the o-chloroanilinium ion.
TABLE II. Acidity
Scales in H20-CH3CN-HCI Mixturesa % CH&N in Mixture
vjv % mol % 6b
0 0.0 -0.00
10 3.83 -0.06
20 7.93 -0.08
30 12.87 -0.13
50
40 18.68 -0.19
25.63 -0.28
60 34.07 -0.43
2.500 2.69 2.477 -0.20
2.400 2.68 2.400 -0.30
2.270 2.70 2.300 -0.43
1.500 1.69 1.490 -0.21
1.485 1.76 1.420 -0.23
1.380 1.81 1.320 -0.33
0.002 M HCI
PHcalcd
2.715 2.715 2.700
A C
0.00
-0.05
2.600 2.68 2.60 -0.10
PHobsd
1.725 1.725 1.710
1.680 1.74 1.660 -0.03
1.650 1.73 1.610 -0.06
PHobsd
PaH*
2.650 2.71 2.65
0.02
paH” PHcalcd A C
-0.01
2.544 2.68 2.544 -0.15
M HCI 1.605 1.73 1.560 -0.11
a PHobsd measured with glass electrode; paH* operational pH as defined in text; pHcalcd molar concentration of HCI per volume of HzO added. ref 11. A values equai to pHobsd - (-log ( H + ) ) .
p H meter readings. In mixtures of HzO with cosolvent one needs to define a reference point for apparent hydrogen ion activity. Bates approached this problem by defining an “operational pH” quantitylO paH* where aH* “is the hydrogen ion activity referred to the standard state in the mixed solvent” (Bates);with paH* = pHobqd -
ES
+ log
yH = pHobso - 6
(3) where pHobsd is measured with a glass electrode employing standard aqueous buffers for calibration, & is the liquid junction potential, and YH is the activity coefficient of the proton. 6 then is the difference between the liquid junction potential effect and the medium effect on hydrogen ions. Values of 6 have been tabulated by Douheretll for aqueous mixtures of AN and some other solvents. Table I and I1 and Figures 1 and 2 provide p H (measured with glass electrode), paH* (corrected with the 6 values extracted from ref l l ) , along with the 6 values for aqueous acetate buffer-AN mixtures. A study with strong acid (HC1) added t o H20-AN mixtures is included. In this case pHobsd decreases with added AN. again as suggested by Bates and SchwarzenThe Journal of Physical Chemistry, Voi. 77. No. 22, 7973
* From
b a ~ h the ; ~ pa^*, on the other hand, remains essentially constant as expected, irrespective of AN concentration. The pHobsd in HzO-AN solutions of HC1 is, within experimental error limits, identical with the pHcaicd a quantity equal (ir? each mixture) to molar concentration of H + per liters of HzO (only) present. One can also define a quantity, A , the difference between pHobsd and -log ( H + molar concentration). Such A values in HC1 solutions are essentially identical with Douheret’s 6 valuesll (the latter arrived a t by a much more rigorous process) and thus their application to any buffer solution leads immediately to paH* values for any H20-AN mixture, i,e., an “operational p H scale.” As Table I indicates such acidities ( p H * ) are uniformly below HOvalues for acetate type buffers. It should be emphasized that the phenomenon concerning the behavior of the pHobsd in H20-CH3CN solutions of HC1 is precedented. Bascombe and Be1112 and Wyatt13 advanced the notion of the chemical hydration of proton14 based on the fact that very similar Ho values are obtained for similar acid molalities of strong acid solutions independent of the anion present. Ho depends only
2683
Acidity Scales in M i x e d HzO-CH3CN Buffer Solutions
28-
_*
10 I I N HCI,
pa;
:
\
: PH 6 2.4 2
5m'
pHobs
22
2.0-
A
I
P
I
Ho
161
pHobs
\
100 A I N HCll 5m'
14
Figure 2. Response of the glass electrode to mineral acid in HzO-CH3CN.
24
Ib
20
30
Figure 1. Behavior of the in H z O - C H ~ C Nmixtures.
40 5 0 60 "loCH3CN ( V / V )
glass
electrode and the indicator.(Ho)
on the ratio of the number of hydrogen ions to the number of water molecules. The present finding extends the proton hydration theory to mixed solvent systems. The fact that the pHobsd for any mixture with HCI corresponds to the PHcalcd values can be explained with the above theory since the increased acidity is accompanied by a decreased water activity.
Acknowledgment. Helpful advice and encouragement provided by Professor F. H. Westheimer during the course of this study is much appreciated. Financial support of the author in the form of an NIH postdoctoral fellowship is acknowledged with pleasure.
References and Notes (1) Supported by GM 04712 from the Institute of General Medical Sciences of the National Institute of Health. (2) NIH Postdoctoral Fellow; Present address, Department of Chemistry, Newark College of Arts and Sciences, Rutgers University, Newark, N. J. 07102. (3) F. Jordan, submitted for publication. ( 4 ) M. A. Paul and F. A. Long, Chem. Rev., 57, 1 (1957). (5) L. Doub and J. M. Vandenbelt, J. Amer. Chem. SOC.. 69, 2715 (1947). (6) "Handbook of Chemistry and Physics," The Chemical Rubber Publishing Co., 50th ed, Cleveland, Ohio, 1969, p D-115. ( 7 ) As discussed in R. G. Bates, "Determination of pH, Theory and Practice," Wiley, New York, N. Y., 1964, p 21 1. (8) J. Desbarres, Bull. SOC.Chim. Fr., 3240 (1966). (9) J. Vedel,Ann. Chim. (Paris), 2, 335 (1967). (10) Reference 7, p 223. (11) G. Douheret, Bull. SOC.Chim. Fr., 3122 (1968). (12) K. N. Bascombe and R. P. Bell, Discuss. Faraday Soc., 24, 158 (1957). (13) P. A. H. Wyatt, Discuss. FaradaySoc., 24, 163 (1957). (14) The author wishes to express his appreciation to a referee for calling his attention to M . Liler's, "Reaction Mechanisms in Sulphuric Acid," Academic Press, New York. N. Y., 1971, p 52-56, which presents a concise summary of this theory.
The Journal of Physical Chemistry, Vol. 77, No. 22, 1973
