TitleIonic solutions under high pressures V : pressure effects on thewalden products and hydration of Et4N+ and ClO4- ions inwater
Author(s) Nakahara, Masaru; Osugi, Jiro
Citation The Review of Physical Chemistry of Japan (1974), 43(2): 71-81
Issue Date 1974-01-31
URL http://hdl.handle.net/2433/46989
Right
Type Departmental Bulletin Paper
Textversion publisher
Kyoto University
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
TaE REVIEW OF PHYSICAL CNE3IISTRY OF JAPAN, VOL, 43; NO. 2, 1973
IONIC SOLUTIONS UNDER HIGH PRESSURES V
Pressure Effects on The Walden Products and Hydration of
EtaN+ and CIOa Ions in Water
BY ~'fASARU N.4RAHAR.4 AND )IRO ~StiCI
The electrical conductivities of aqueous solutions at high pressures up to 5,000 atm have been measured in the concentration range from 10_4 to 10'sx for tetraethylammonium chloride, EtaNCI at 25 and 40'C and &or tetraethyl-ammonium perchlorate. EtINC10{ at 2~ C. The equivalent coaductances of [he electrolytes at i¢finite dilution have keen determined by means of the Onsager equation which was verified to be valid in the dilute solutions at high pressures in the previous papersl•r}. The limiting equfvalent conductances determined at high pressures were separated into the single-ion o¢es o¢ the basis of the same assumption as used in the previous paper~l. Although the limiting equi-valent conductance of EyN• has a maximum against pressure at 2YC ]ike other tetraalkylammonium ions, that of CIOa ,surprisingly and exceptionally, has no maximum even at 25'C where the viscosity of solvent water has a minimum at about 650 atms>. The Walden product of EtcN• decreases slightly with increasing pressure at 25'C and, probably, so at 40'C like that of ]SesN*. On [he other hand, the SValden product of CIO at 25'C dramatically decreases with increasing pressure. Thus, it is considered that the pressure dependence of the limiting equivalent conductance of the ion in water can not be explained merely in terms of such bulk properties of water as the viscosity and dielectric constant. These differences in the pressure coefficients of [he Walden products were ascribed to the differences between the density of the hydration shell and that of [he hull: water.
Introduction
Colladoo and S[urm41 are the first persons to. attempt to examine [he effect of pressure on the
electrical conductivities of electrolyte solutions, when it was not yet known what carries electricity
in solution. In Ig86, first of all, Finksl established that pressure (up to 500 atm) does decrease [he
electrical resistances of electrolyte solutions ; it was two years before the appearance of the Arrhenius
(Received September ]0, 1973) I) :1I. Nakahara, R. Shimizu and J. Osugi, This Journal, q2, 12 (1972) 2) b1. Nakahara, ibid., 42, 75 (1972) 3) J. B. Cappi, Ph. D, Thesis, London University (1964) 4) D. Colladoo and C. Sturm, Ann. Chim. Phys., 36, 23l (1827) 5) J• Fink, (Vied. Ann., 26, 481 (1865)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
72 M. Nakahara and J• Osugi
theory for the behavior of electrolyte solutions. In the 1890's, Rontgen5l, Fanjungn and Tammannsl
first found a parallelism between the pressure dependence ofthe electrical conductivities of aqueous
solutions and that of the soheat fluidity measured by Cohensl. Strictly speaking, however, those early
works were not so much accurate but only phenomenological, even alter the eminent Debye-Hiickel
theorylo> for the strong electrolyte solution was developed in 1923. It seems to the authors that 5rst
great efforts to determine accurately the limiting equivalent conductances at high pressures were made
in the 1950's by Hamann and his coworkers in particular for the purpose of examining the pressure
effect on the ionization of weak electrolytes, In the 1960's, the accourately determined limiting equi-
valent conducfances at high pressures began to be analyzed in terms of the transition state theoryrs)
by Bmmmer and Hills%~• ts). Osugi, Shimizu and Takizawa161, and Adams and Laidler~%>, and also is
terms of the dielectric relaxation effect by Skinner and Fuosslgl and Cussler and Fuosst9~.
When we deal with a transport property in solution, [here is one fundamental question as to how
much the transport property might reflect the equilibrium property of ions in solution. In the case
of ionic conductance, however, there would be fairly good correspondence beriveen them for [he tol•
lowing reasons. The ions in solution are originally moving very rapidly in a random way colliding
with the solvent molecules or with each other, as well-known as Brownian motion. When an external
electric. field is applied to the system, the ions begin to move preferentially in the direction of the
applied Seld. At this time, if the external perturbation is so weak, as often in the conductance
measurement, that it may not disturb the internal field exerted by the ions themselves and polar
solvent molecules, the limiting ionic equivalent conductance would minutely reflect the ion-solvent
interaction in the equilibrium state. In consequence, the limiting ionic equivalent conductance could
be used as a useful probe to study ionic hydration. 9s a matter of fact, it was previously reported~f
that there are linear correlations between the hydration number calculated from the limiting ionic
equivalent conductance and the hydration enthalpy for the alkali metal ions and for the halogen ions
at normal pressure.
6) 7) 8) 9)
10) I1) 12) 13) 14) 13) lfi) 17) 18) 19) 20)
W. C. R"ontgea, A'achrichlen de> k. Gesellsch. zu Gdttingen, 509 (1893) I. Faajuag, Z. Pkys. Chenf., 14. 673 (1894) G. Tammana, ibid., 17, 725 (1895) E. Cohen, R4ed. Ann., 45, 666 (1892) P. Debye and E. Huckel, Phys. Z., 24, 185 (1923) S. D. Hamnn, "Physico-Chemical Effects of Pressure", Butternrorths, Loadoa (1957) S. D. Hamana aced N. Strauss, Trance Faraday Soc., 51, 1684 (1955) S. Glasstone, %. J. Laidler aced H. Eyring, "The Theory of Rate Processes", \fcGraw-Hill (1941) 5. B. Brummer and G. J• Hills, Trans. Faraday Soc., 37, 1816 (196t) S. B. Brummer and G. J. Hills, ibid., 57. 1823 (1961)
J. Osugi, K. Shimizu and H. Takizawa, This lonrnal, 34, 55 (1964) W. A. Adams and %. J. Laidler, Can. L Chen+, 46, 1989 (1966)
J. %. Skinner and R. \f. Fuoss, J. Phys. Chem,. 71, 4455 (1967) E. L. Cussler and R. M. Fuoss, ibid., 71, 4459 (1967) M. fiakahara, R. Shimizu and J. Osugi, Nippon Kagaku Zasslri (!. Chem. Soc, Japan, Pure Chem. Sec.), 92, 785 (1971)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
Ionic Solutions under High Pressures V 73
Expermiental
Tetraethylammonium chloride, EC,NCI was obtained by using anion exchange resin from its
bromide which was first synthesized by the bienschutkin reaction. Et,NCI salt was three times
recrystaBized from its methanol solution by adding chilled ether, and dried in vacuum a[ about 50°C
for a week. Tetraethylammonium perchlorate was precipitated by adding excess amount of perchloric
acid to the aqueous solution of EgVBr*, recrystallized twice from its aqueous solution, and dried in
vacuum at room temperature for 3 days after heated up to 100-C for 2 hours with great care to avoid
explosion. The dilute sample solutions were prepazed from stock solutions just in the same way as
beforet.zU.
The high-pressure apparatus and the conductivity cell were already described elsewhere.
Results and Discussion
Determination of d°t Pa
The equivalent conductances..hPl of Et,NCI and EyNC10, in dilute aqueous solution a[ pres
sure P were determined, after the corrections for the solvent conductivity and changes in concentration
and cell constant with pressure had been made just in the same manner as the previous s[udytl.
Then. the equivalent conductances at infinite dilution at pressure P, ,frP> were obtained with the aid
of Onsager's equation for conductance,
[ha[ is,
1-rzt~C '
from which A°tl'~ of EyKC] at 25 and 40`C and EyNC10. at 2S'C were calculated and given in
Tables I^•3. The adequacy of this method to obtain d°tPa from Acl~ fn the dilute concentration
range would be supported by the approximate constancy of d`~ around their mean value within the
experimental error, which was already found in the previous study on \Ie,NCI and Bu,NCIz>. How•
ever, i[ was reported~•zz> that the direct graphicai extrapolation of the conductance data of Eq\CI to
infinite dilution with the aid of the theoretical limiting slope was not satisfactory exceptionally.
R'hen the conductances of EtaNCI shown in Table 1 are compared with those measured at a higher
concentration by Horne and Y'oungz+l, there are rather large discrepancies between them as found
* This salt was kindly supplied by Mr. T. Hori, Laboratory of Analytical Chemistry of our Depart ment.
21) ~I, Nakahara, K. Shimizu and J. Osugi, Tlais Journal, 40, 1 (1970) 22) A. B. Gancy and S. B. Stammer, J. Chern. Eng, Data, 16, 1763 (1968)
23) S. B. Stammer and A. B. Gancy, "Water and Aqueous Solutions", Chap. 19, Part I, ed. by R. A. Horne, Wiley-Interscience, New York ([912)
24) R. A, Horne and R. P. Young, L Phyr. Chem., 72, 1763 (1968)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
74 .\l. Nakabara and J. Osugi
in the cases of Me,NCI and Bu,NClzl.
Table 1 A'CPl (ohm'l~cm2~equiv-1) of Eta NCI in Hs0 at 2SC
Sample*
Pressure, atmA B C D Average
1
500
1000
1,500
2,000
1,500
3,000
3,500
4,000
4,500
5,000
! 08.9
110.9
111.4
111.2
109.9
108.3
106.2
103.3
100.6
98.3
93_fi
108.9
110.6
1111
110.6
109a
107.6
101.3
102.9
100.3
97.6
950
108.8
110.7
Ill.4
110.9
109.7
107.8
105.7
103.1
100.3
97.7
94.9
(06.8
110.8
I1Ll
110.2
109.1
107.1
105.1
102.8
100.1
91.5
94.6
108.9
110.8
IIL3
110.7
109.5
107.7
105.6
103.0
100.4
97.6
95.0
* A: 3.866 x 10-~ n, B: 6.201.x 10-+ r, C: 8.536 x 10'1 s, D: 11.598 x 10'~ s at 1 a[m
Table 2 A'p•) (ohm lmmZ•equis~ 1) of EtgNC] in Hr0 a[ 40'C
~- _ Sample*
Pressure, atm-- -_
1
500
1,000
1,500
2,000
2, 500
3,000
3,500
4,000
4,300
5,000
A B C D Average
143.9
143.6
141.6
141.2
139.6
137.2
134.6
131.4
118.2
124.9
121.5
143.9
143.7
141,6
141.2
139.1
I3fi5
133.8
1305
127.2
123.6
120.1
143.8
143.7
142,6
1410
138.9
136.4
133.5
130.3
126.9
123.4
1 ] 9.9
143.9
143.6
142.5
I4L1
139.2
136.6
133.7
13o.a
127.0
123.4
119.8
143.9
143.7
142.6
141.1
139.2
136.6
133.9
130.7
127.3
123.8
120.4
* A: 3.847 x t0-~ ti, B: 5.992 x 10_a ~, C: 8.495 x IO-~ v, D : 11.542 x 10'~ a at 1 atm
Obtaining of 1°<rl from if°« The single-ion equivalent conductances at pressure P, ,)°tPl were calculated on the basis of the
same postulate used in Ref. (2) ; it was assumed that the Walden product of Bu~N' is approximately
independent of pressure. In the present calculation, the previous~> interpolations of the values of
water viscosity measured by Cappi3l have been corrected as shown in Table 4, because the interpo•
Iation from the direct plot of water viscosity, r!° against pressure was found to be less accurate than
23) Al. Nakabara, K. Shimizu and J. Osugi, TFis Journal, 40, 12 (1970)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
Table
Ionic Solutions under High Pressures 4
3 d°tP7 (ohm ~•cmt•equiv r) of Eta1`rClO. in Hz0 at 25'C
1i
~'~
Pressure,
Sample*
atm~~A H C n Average
1
500
1,000
1;500
2,000
2,100
3,000
3.500
4,000
4,500
s.ooo
I 99.86
98.37
91.94
93.55
9D.45
87.34
84.40
81.11
7 i.89
14,80
71.93
99.95
95,71
9fi.49
93.82
90.69
87.4fi
84.38
81:11
i 8.02
74.99
72.08
99.82
98.61
96.27
93.41
90.41
87.20
84.11
80.90
77.85
74.82
71.96
99.95
98.89
96.43
93.60
90.53.
57.48
84.16
80.94
i 7.88
74.88
72.02
99.9
98.6
96.3
93.b
90.5
87.4
84.3
81.0
i7.9
14.9
i 2.0
k 4: 4.147 x IO_q N, B: 7.276 x 10-+ x-, C: 9.093 x 10-~ x, D: 10919 x 10-~ v
that from the plot of log >7U against pressure a[ the . high pressures due to the steep dependency of r`
upon pressure and the scarcity of the measured points (The corrected values of r` in Table 4 does not
make the curve of d-(I{CL)•r,` vr. pressure at 25-C cross [hat at 40`C; see Fig. 4 in Ref. (I)) The
obtained values of ),'cl~ at 25`C are summarized in Table i, where the values of A'<r.' of other ions so
far investigated are also given for comparison after the above correction has been made. In Table 5,
it is seen that the correction does not cause so large alteration in the values of x`l~ and the discussion
and conclusion in the previous paper might not be amended.
Table 4 The viscosity of wateq 7 (cP) at 25'C
(interpolated by plotting Cappf's data against pressure)
Pressure, atm Present Previous~7
1
500
l,ooo
lsoo
z.ooo
zsoo 3,000
3,500
4.000
4,500
s.ooo
0.8937
0.8865
0:8905
0.9033
0.9260
0.9532
0:9853
L0135
1.0589
1.0895
1.10.57
0.9266
0.9534
0.9709
1.0022
1.0392
1.0756
1.1163
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
76
Table
M. Nakahara and J. Osugi
5 2°tr) (ohm ~•cmz•equiv'~) of We ions in H2O at 25°C
_` ns Pressure. atm~
1
500 1,000
1,500
z,ooo
2,500
3,000
3,500 4,000
4,500
s,ooo
Buahr: Bu~N"" EtaN• Me4N• K* CI- CIO,
19.4
19.6
19.5
19.2
I s.7
18.2
1 i.6
17.1
16.4
15.9
151
19.4
19.6
19.5
19.2
1s.7
I8.2
17.9
17.3
16.7
16.1
us
32.5
32.9
32.5
31.8
30.9
30.0
29.0
27.8
26.8
26.0
24.8
44.7
45.0
44.4
43.8
42.8
41.6
40.3
38.9
37.6
36.1
34.5
73.5
74.7
74.6
74.0
73.0
71.9
70:1
68.4
66.6
64.9
62.4
i 6.4
77.9
i8.8
78.9
i S.6
77.7
76,6
75.2
73.6
71.8
70.2
67.4
65.7
63.8
61.8
59.6
57.4
55.3
53.2
51.1
48.9
47.2
~ Fram Ref . (2)
The postulate introduced to obtain the single-ion equivalent conductances at infinite dilution at
high pressures has been justified in Ref. (2) by comparing the calculated transference numbers of K*
in KCI,
it°<P~(KCl)
with those duectly measured up to 1.000 a[m at 25°C. In order to estimate [he values of d-<~ at 40'C,
it was additionally assumed Chat
Table 6 Walden products of the ions in H.,O at 40'C (ohm ~•cmr•equiv ~•cP)
-Ions Pressure, atm -~~
EtaN• Me~N' K• C1-
1
500
1,000
1,500
2,000
2500
3,000
3,500
4,000
4.500
5,000
za.i
z 7.a
27.9
27.8
27.i
27.5
27.5
27.5
27.2
27.2
27.0
41.1
39.6
40.2
40.9
39.8
39.8
39.8
39.8
39.i
39.5
39.1
63.0
63.7
64.4
65.4
66.2
67.1
67.9
68.7
69.4
70.0
70.3
b6.0
67.9
68.i
70.3
7 2.0
73.2
74.3
75.5
76.8
77.5
79.0
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
Ionic Solutions under High Pressures V i7
is nearly independent of temperature, because in our laboratory~l it has been recently shown that
r°cP%(K*) decreases by only 0.7% with the increase in temperature from 15 to 25~C or from 15 to
40`C both at 1,000 and at 1,500 atm. By using the values of c't~ at 25'C calculated from the ionic
equivalent conductance values in Table 5; d`tP~ (EtaNCI) in Table 2, d°Ctn (KCl) in Ref. (1), d-<~
(Me.NCI) in Ref. (2) and E`p> (K*) at 40-C in Ref. (27}, the single-ion values of the limiting equi~ valent conductances at 40°C a[ high pressures were obtained and used for the calculation of the R'a4
den Products in Table 6.
Pressure dependence of .i°<r~
The relative variation of the limiting equivalent conductances of the ions with pressure is shown
in Fig. 1. It is to be noted that the pressure dependence of J,'tP~ of tetraalkylammonium ions such
as Bu,V*, Et,N° and DIesN* are somewhat similar [o each other and. moreover, to that of the viscosity
of solvent water. On the other hand, the pressure dependence of d't~ of K*,. CI- and, above all,
C10. are quite different with each other. Furthermore, it is surprising that CIO,' ion has no maximum
conductance against pressure at 25'C, although all other ions so far studied hate a maximum con-
ductance against pressure at [he same temperature. These facts would suggest that the limiting ionic
equivalent conductance at high pressure can not be interpreted only by such a macroscopic property
of the solvent as aiscosit, in spite of the early finding and statement by RBntgea, Fanjung and Tam-
mann.
L1
0.9
0.8
o.io
/` C10,-
e,
Cl-
K•
i~ t,N•
u,N•
1 2 3 4
Pressure, 103 atm5
Fig. 1 d`CP)~,y°(U vs. pressure at 2s'C
Variation of the Walden product with pressure
In order to deduct the piessure influence on the macroscopic viscosity of water from that on the
ionic conductance, the Walden products, IV=d`1P)•n'<~ of the ions at 25`C are calculated by using
the values of d'tP' in Table 5 and r't~ is Table 4 and given in Table 7, and their relative variations
with pressureare shown in Fig. 2. There, we see that the pressure coefficient of the Walden product,
8W/r7P at 1 a[m and 2SC is positive, slightly negative and remazkably negative for G- and K* ions,
26) Y. lfatsubara. K. Shimizu and J. Osugi. This JonrnaI, 43, 24 (1973) 27) R. W. Allgood,.D. J. LeRoy and a. R. Gordoq J. CGenr. Phys.. g. 418 (1940)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
78 \I. Nakahara and J. Osugi
Table 7 Walden products of the ions in Hz0 at 2SC
(ohm ~•cmZ~equiv r•cP)
Ious
Pressure, atm
I
500
1,000
1,500
2,000
2.500
3.000
3,500
4,000
4,500
5,000
EtrN• \fe4N• K• CI' CIOa
29.0
29.2
28.9
28.8
28.fi
28,6
28,fi
28.2
25.4
28.3
28.4
39.9
39.9
39.5
39.7
39.7
39.7
39.)
39.4
39.8
39.3
39.5
fi3.7
66.2
66.4
67.0
67.6
68.5
69.1
69.3
70.5
70,7
i 1.5
6.83
69.1
70.2
71.4
72.8
74.1
i 5.5
76.2
77.9
78.2
80.4
fi0.2
58.2
56.9
55.9
55.2
54.7
54.5
53.9
54.1
53.3
54.1
tetraalkylammonium ions and CIOa ion, respectively. Comparing Fig. 2 with Fig. 3~•~>, we notice
[hat 81~V18P at 25'C and I atm does not correlate with 8TJ'/17T at 1 atm 25'C for these ions. The
possible view-points for the interpretation of the pressure and temperature coe&cients of the ionic Walden product sre summarized in Table S, where al] the view-points but the first one are relevant
Table 8 Prediction o[ the sign of 8 [I%BP
Point of view Ref.
C)
D)
E)
P)
5)
Compression effect
Dielectric friction theory
Electrostrittioacheory
Pressure-induced dehydration
Structural change of water
i ~I
ao, n
18, 19
30
28
Sign of 8lY/BP
}
- . 0, +
[o ion-solvent interaction at any rate. Before we correlate the pressure coefficient of the ionic Walden
product with the ion-solvent interaction, we now consider the theoretical background of W or .i°.
Although in very dilute electrolyte solutions. +i has been successfully represented in some quantitative
forms by Oasager, Fuoss and others, no satisfactory theory for ionic conductance has been established
at the two extremes, at in5nite dilution and at high concentration. Owing to this undeveloped stage
of the theory for i°, we could not make any completely quantitative explanation of the pressure
28) R. L. Bay and D. F. Evans, !. Phyr. Chen+., 70, 2321 (1966) 29) R. A. Robinson and R H. Stokes, "Elettcolyte Solutions", Butterworths, London (1957)
30) R. A. Horne, "Advances in High Pressure Research", Vol. 2, Chap. 3, ed. by R. S. Bradley, Academic Press, London (1969)
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
Ionic Solutions under High Pressures V 79
s E
1.20e
1.1 ih a
CC
L10; v
K+
LOi
,Bu~N+1.00 o-o
n
o -~
0.9i TSe~N' Et~N+
C10,-0.90
0 1 2 3 4 5 Pressure, 103 atm
Fig. 2 iV(r')/NP) vs, pressure at 2SC
1.1
U
N j.0 ~F
V °' 0 .9 of
0.8
Bu,N'
EC,N•
Me.N*
K'CIO,
Fig. 3
0 20 •W 6D 80 100 Temperature,'C
Variation of the~i'alden products with temperature at 1 atm .t° of Bu,N*, E t,N• and Me,N• are cited from Ref. (28), and x' of R', CI- and CIOa irom Ref, (29).
coefficient of the ionic Walden product as yet. Then, we want to try some qualitative discussion by
using the modified~•211 Stokes equation,
CT re
where a, e, F, C, d` and r. are the ionic valence, protonic charge, Faraday constant, hydrodynamic
parameter being a fucction of r., solvent viscosity and effective radius of a hydrated ioa, respectively.
From Eq. (5) we hate
W-~.r-- GeF (0) When Eq. (6) is differentiated with respect to pressure, it follows that
8W __ air ~ t 1 a'C ~ ~ where the third factor in [be right-hand side has a positive value because oo^C/8r. is positive~•ul.
Then, it may be approximated that water exists in the following two states, -
stateI (standard state) K stateII
water ~ water. (8)
(in the bulk) (in the hydration shell)
For the above equilibrium, we can write
K= ai ~n>nru (al = 1, to ¢ l), (D) where K is the equilibrium constant, a the activity of wateq h the hydration number of an ioa at
infinite dilution, m the concentration of the ion waich is arbitrarily very small, and rg the activity
The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
50 LI. Nakaha:a and J. Osugi
coefficient of water in the state IL More than ten years ago, the kinetic aspect of the hydration of
ions was discussed by Samoilov3l•~3 especially from the view-point of energetics. A'ow, we attempt
to discuss the hydration equilibrium in terms bf density. By dffierentiatiag the logarithmic form of Eq. (9) with respect to pressure, neglecting the pressure coefficient of Yg and Considering the basic
thermodynamic relztionship, we have
t7P h 8P RT (10)
where V°1 and V°q are the molal volumes of water in each state. Since rve can neglect the com-
pression effectll for such weakly hydrated (bulky monovalent) ions as RrN* and C1Og ,
sign of~ 8P ~ =sign of~ 8P ~. (Il) From Eqs. (7) and (10), we have
sign of( 8P ) =sign of (V°n-V'1), (12) if /~~0.
Eq. (12) means that the density of the hydration shell is larger than that of the bulk water if the
Walden product of Che ion has a negative pressure coefficient and vice ve>sn. As shown in Fig. 2,
o W(CIOs')/oP is strongly negative at L atm and comes to be nearly zero at ahout 5,000 atm. There-fore, rye could say the density of the water molecules in the vicinity of CIO. ion is higher than that
of the bulk water at the lox•er pressures and the difference becomes very small at about 5,000 atm.
This higher density around CIO9 ion could be accounted for by its breaking effect on the water
structure which would becomeweaker at high pressures because pressure would break down the water
structure. Furthermore, concerning the two types of molecular models~•~1 proposed for the orient-
ation of a water molecule with respect to an anion, Buckingham's one that seems to result in the
higher density of the hydration shell would be preferred especially for C1O~ ion. Judging from Fig.
2 and Table 6, oa the other hand, the hydration shells of the tetraalkylammonium ions have slightly
higher densities than that of the bulk water ; 81F{Btt~N*)/aP really becomes slightly negative, if the
directly measured transference number data~l are used for the estimation of the limiting equivalent
coaductances of the ions at high pressures instead of the postulate, d'<tl(BuaN*)•r~<3)=1.`tP~(Bu>N*)•
r`tP'. Although it is not sufficiently known in terms of both energy and density what kind of struc-ture of water is formed about alkyl chains, the above conclusion drawn from the pressure effect on the
\'r'alden products of Me.N*, EtcN* and BurN* ions seems to be conformed with the following voles
metric results :the negative contribution3s> to the partial molal volumes of the hydrophobic hydration
31) O. Ya. Samoilov, `structure of Aqueous Electrolytic Solutions and the Hydration of Ions", Chap. 3, Consultant Bureau, New York (1965)
32) 0. Ya. Samoilov, INrcussion Faraday Soc., 24, 141 (1957) 33) J. D. Bernal and R.H. Fowler, J. Chem. Plrys„ I, 535 (1933) 34) A. D. Buckingham, Discussion Pa>aday Soc.,.24, I51 (1957) 35) E J. ~Iillero, "Water and Aqueous Solutions", Chap. 13, ed. by R. A. Horne, Wiley-Interscience
(1972)
t The Review of Physical Chemistry of Japan Vol. 43 No. 2 (1973)
Ionic Solutions under High Pressures V 81
of the tetrnalkylammonium ions and the depressing effectss> of the tetrnalkylammonium ions on the
temperature of maximum density of water.
Laboratory of Physical Cheuaistry
Depar6neut of Chemistry
Faculty of Science
Xyoto University
Xyoto, 606
Japan
36) A. J. Darnell and J. Greyson, L Phys. Chene. 72, 302 (1968)