mm... M».

.2 .0.)

.V5 , NJ: 0 o I

. . . mt... S *5.-

s o. _ 2.

F... 33 1W é." sue ....

1... 3h 43w”. bwvs R T . . .4

... . V... a .. a... a; 5.

R i .m. a an 2....
.. - J .3.

AN
“:53
\-
"§
.
:5
' E:, n“
5.:
‘.
.‘v
..
7*

«a
. .. H... .
r... hm». WM “HAS .».... Wm .s .
”Hm an. in a.“ a.“ .x... .rfiv
W,” up...» We“ .7 J“. r5. AU)
.a. . .2. ”w MM .1.” :3

P'
OE
C
l

t
2‘:
i;

49’»
on.

‘G

a}:

* b

312-:

: t'a.

. :U'
I!”

fi‘sué‘

2...... .»...L
. aw». ......
8
a.“ u, m 8
C. .. am
.2.‘

MSC?ANCE AM}

C

 

L I B R A R Y
Michigan Stave
University

  

  

A;

 

 

MICHIGAN 37A" WWI!!!"

RECEIVED
JUN 1 1959

DI". 0' CIWNIY

 

 

 

/' r. r {11' —._ :1" ‘0 -vf~ ~ -rfl:‘ 9;; 3-4
‘uJ-:..)'. ;L.. ;CE [xAEJ’ .1. lf-A4x.:‘- .2: ;‘J44 .L'v' .2... -14. I...)
" '3. . ‘
(a; A'q.vboud x3UuuLiU1iL) 01“

f» a-; 1. L" .. ‘9’" \.‘;_'-.v _._ $:£_.'~‘-r,~\ rt: F\-_""f\"'
‘JU‘-’J.‘Jsl l‘.‘ .31 ‘1-Ji£,..\ 2:1. 44.14...) -.'..“" M. '; 4.5.7,

.-v 9“ Mn}. mm
A“; ‘I- .JU..!511L«J.

.-bm1tted to the Collcge of Science and Arts
nichiflan tate Lniversi.ty of A; riculture anu
ip=Mlie balance in partial fulfillnzent of
t1.e reQLirezents fer tie de; rea of

Vf‘ry 7*“ “.3 av-

..izd‘dl 11‘ {u ”GIL“.- bun

{cgartgent of Chemistry

 

Approved:

195’)

FOflSEIN PASAIFEH

anti- “9‘

,5
r)
*3

Recent theoretical developments have prompteo an
investigation of the behavior of strong 2-2 electrolytesin
water. Certain anomalies which oopeerofi using zinc sulfate,
resulted in a search for another 2-2 electrolyte which would
not undergo hydrolysis and whose ions would not be capable
of forming covalent bonds with the solvent or with Oppositely
charéec ions. ZLe strong electrolyte, barium metetenzeneci-
salfonnte sooned to fit these requirements. In oréer to
' aid in the theoretical treatment, an inccpendent value of
the limiting equivalent conflictance of the metebenzonofiiSWI-
fonato ion was needed. This thesis describes the meeewre-
cent of the conductance of sodium metabenzenedisculfoneto.
The value of Al9 obtained is 59.66 ohm"1 equiv.“1 one.

It was found that in dilute solutions, the conflic-
tanoe data fit the limiting slope preoicted by the Oncoger
equation. At higner concentrations. the deviations were in
the proper éirection, and had the correct magnitude to be
described by higher order terms of the electrOpLoretic
equation.

In view of recent deviations of some 1-2 and 2-1
electrolytes from the Oneeger countion in dilute solutions,
it wonld be of interest to determine the effect of cation
size on the conductance. This measurement on sodium meta-

benzenedisulfonnte is one step in this seriee.

. ’ ‘ ’5‘ v = ~: 'v-~ F'- -*». . 1"“
v _ an; [a :13» fi- 4 ' ,v-"Hw ‘77,. r
~ .o.. t-_ V . _. . . _ , .

\v-1 . I . (4- -'\ I" 9' if.“ ‘3 .. I"?
(w- 1C..¢...'_.‘ .4 40u4¢1 44 C);
v V "

T s f.

: ‘§.v~_},<,v.~~,v.‘y-qg. .4. ‘ ‘COV‘F‘Y-‘v

. ‘1 -'.: . ‘ .- . | - " ’

5".L:‘.J-~Jll.--.li_\'-lLJi..3-J—.‘f' ;uf\1 U
T3

3 r! . u'hfno:

7:4...‘5 -— I’T“ 7'7. . ‘ ‘
1.99.4 Ubl*:.5 r!)¢-..:‘3.i L.) .4

I’ New ”A H!
A . .. .‘
s A i. (.13.; a)

Sibmittefi to the College of Science and Arts
Fichigan State University of Agriculture and
Applied Science in partial fAlflllwent of
the requirements for the degree of

v anw ~n uCI ”C“
.' (14.1141; .1‘ a $.13.- '3‘...

Department of chemistry

\
K L:

’2 / ‘x.
.,/ r9”

A C i135 021.811? FEE:"Z’S

The author wishes to express his most sincere thanks
to Dr. J. L. Dye whose interest, patienca, and counsel
greatly facilitated the completien of this study. fie also
wishes to thank Hr. G. Gordon and other members of the

Chemistry Department for their helpful suggestions.

11

II.

IV.

- ..-.- 4.-. . , 4.. .4-..
Ay£qu¢LLuUI-Luw;d . g

.."_.h. flu .‘ '*g.~I '1."
”8A ~‘1Vi—v’ JD 4. Jafi .
rsl‘ I " gfl"
IL. L‘..-‘ . . . .

1". -
4. 03158;

C. {ytcngioms of Onsager's Theory

L. 18Vi&t10n8 frog All Zheoriss

E” ‘4" I ~‘. .« r
c'" “~v Q‘s-J-v-A- it‘d .

A. ?ur1fication and Preparation

8. Conductance fleasuremcnt

0

~ “‘ 9m I "H
b {J V A
O O C
I O
O O

C. Tranafarence Number .

7“. v'r‘

3"

me" gv-a

A. Fegults . o

1"

I? I ‘ ‘fi'f‘ " ' .-:‘ " 37'. "_ '\
..~ . . .l :
Ll.» Ike-Hail... 4.1... LlJ-uu

. k"

;:h 'x l 1-11 h '-
“M,V~-U ‘3‘!“ L ‘Ev nL;

a}

I ’31“;

m. Biscuuslon of Errors

111

gar Theory and Conductance

n. chye-flfickel Thcury of Ion-Interaction

kn \a we >-

C")

10
11
13
13

17
19
19
21
32

TABLE

I.

II.

III.
IV.

LIbT G? TABng

Analysis of stock solution and metathcc
CEICUIBthUB o o o I o o o o o o o o

Conductance Results, Series 1 . . . .
Conductance Besults. Scriss 2 . . . o
Conductance Results, beries 3 . . . a

Cell Constant Determination . . . . .

Transference Number of Ba 113-?(503)2

iv

Q
43
0

PAGE

r1
P4
f

. 1
f3
”-11
“J
H
t‘
'53
L 1
[u

£13933 PAGE

L. Conductance oi sociuc metabcnzcncisuiforate
Versus square root of concentration . . . . . . 29

2. fluviation from the Cnsafer thccry cf coacuctance 30

\A

v-
. .

. ?r9ns?erence number of cerium-matchen7erdisnlfov 31
sate

I. INTRODJCTION

Ever since Arrhenius”J in 1383, from studies of the
oonéuotances of aqueous solutions of acids, postulated that
an electrolyte solution contains free ions, solutions of
these ions have been studied extensively. Kohlrauscn in
studying the conductances of such solutions noticed that
oxygen and hydrOgon which tend to be absorbed at the sur-
face of the electrodes can be dissolved when the polarity
is reversed. This led him to apply an alternating potential
when making conductance measurements. As the adsorption is
also affected by the surface area, be coated the platinum
with platinum black. Hohlransch‘z), who also emphasized the
importance of good temperature control, made quite precise
measurenents; his cats on potassium chloride in water are
still valid. since specific conductance of the solution
refers to that of one cubic centimeter, it is necessary to
measure the cell constant 2. (The area of electrodes is A
and the cistanco which separates the electrodes is d.)

In 1923 Parker‘B) observed that the cell "constants”
are apparently not really constant, but vary with the re-
sistance being measured. nhedlovsky, Jones, and Bollinger‘”),
showed that this effect is due to the capacitance of the

cell, and designed new types of cells which minimized the

1

2

effect. Jones and his co-workers, with some refineoents in

the tech.iqoe, 1.5,, using an oil bath, a special triage and

a sensitive amplifier with a telephone headset for cetection,

were able to get results of high precision (0.02;).

Any modern theory of connectivity mist be concerned

with the concept of the interaction between thermal motion

of the ions and their electrical attractions and repulsions.

Using the results of Lebye-fiuckel‘5) theory of ionic inter-

action (1923), Onseger‘e) (1929) was able to calculate from

solvent properties the limiting behavior of 1-1 electrolytes,

alkaline earth halides (2-1) and rare earth halides (3-1).

The disagreement of the conductance of 1-4 and 2-2 electrolytes

from this theory was attributed to ion-pair formation. In

a study of theoretical ans experimental behavior of :nsoh by

J. L. Dye and co-workers(7), it was desired to measure
‘xzn+* from data on Zn(ClOu)2. These workers found a large

deviation from the Oncoger equation even in dilute solution.
Probably hydroysis effects or covalent tendencies of Ln++
are responsible. In the case of K2Pt(CN)u for which similar
deviation are observeC‘B), the spatial configuration (square
coplanar) could allow attachment of water molecules in the
octahedral positions. The observed deviations promptea a
search for another 2-1 or 1-2 electrolyte for which hydrolysis

could be definitely ruled out and for which the formation

of covalent bonds was unlikely.

The purpose of this study of sodium metabenzenedisulfonate,

flazaQRSO , is a basic step in the interpretation of 2-2

3’2
electrolyte behavior. 'I‘nia involves the study of Ba :22 9(SOBJ2:
which has already been begun(9). The measurements on
0

Nez 3-9(503L, were done to evaluate A- for m~?(303)2 .

In addition it would be desirable to stuoy tie effect
of cation type on the conductivity. Some other metabenzene-

m“ we}

032 3 a ... ‘VV
and [26(3u)u]2 m-¢P(303)2 will be sticiied to determine the

disulfoneteesuch as Liz m-‘Pwo ,[N(CH)
effect of cation size on the conductance. Such an effect
has been recently found to be very significant for h-l
electrolytee(ls).

The observation by Deniel‘ll) of electroanalysie in
a three compartment cell showed that positive and negative
ions do not carry equal amounts of electricity. In other
‘uords. different ions have different case of moveeent unfier
the influence of an electrical field. This is contrary to
the earlier ioca of Arrhenius silo erroneously ascribed the
decrease of equivalent conductance to a decrease in the
number of free ions and assumed the mobilities to be con»
stout.

Ionic conductivity and transference number are in-
ternally related by the equation t+ - .1: and t =- 22... .

.J\ ’ -‘fb

Where hand t__are transference numbers, A + and A - are the
1€Inic oonductancee of cation end anion respectively, and /\

18 the equivalent conductance.

n,

Three types cf exueriuental sethods o; Catepmlbation of
transference n qur Lgvu Lten agglojedz

(1) ’he hittorf method of exgari eat. which was
frequently used in the vast, regulrus acourata anal;tica1
work and is quite t1;e consuming.

(2) She emf method, involving JCJéqPQLLdtS of calls
w1th and witho;t transference, requires electrodes revers-
ible to bots ions, and even then is of lluibei accuracy.

(3) The moving boundar' cettod, with wligh these
experiments are concerncé, inuolves the observgtion of the
ve10¢1ty of a boundary between two solutions. 2;15 method
18 relatively rapid and provides the best accnracy attain-
able at present.

In addition, transference numbers afforg an 1ndepen~
cent experimcabal test of theon3.11 happens that tPfiHSTBP-
ence ngmbers of unsymetrical electrolytea are Lore sensitive

to deviation fr m Gnaager theory than are canJACtanoes(12).

 

€115:

CO";

0111

00h;

 

It has been shown that the Debye-ankcl limiting law
gives theoretical prcéictionn which fit experiment at very
low concentrations. Bonever the theory is :not valid at high
concentrations.

Debgaand Huckol mode cone assumptions in develOping
the theory. which in moot cases are not valid at high con-
centrations. These assumptions are the following:

1. long behave as hard non-polarizablo Spheres. This
anonnption gives to each electrolyte a mean distance of
closest approach. Ions are asnuned unable to penetrate
within this ointanco and the charges are assumed to be
spherically distributed over the ions. It is BXpected that
this assumption will be invalid for large ions whose charge
distribution is somewhat distorted.

2. The solution is assumed to have a uniform dielec-
tric constant - actually solvent molecules are about the
same size as ions and the assumption of uniform dielectric
constant cannot be valid if a few solvent molecules are
on the average between each pair of ions. The dielectric
constant of the bound water is not the same as that of free

water in the pore state. Furthermore the work of Hosted,

5

 

Eitscr.

I 901..

‘O
LG! 1:

0f tin

the l

6
Eitson and Collie‘13) showsthat the dielectric constant of
a solution is less than that of the pure solvent.

3. The time-average charge distribution can replace
the instantaneous distributions.

In using the time-average distribution of ions in
the POisson equation, instantaneous interaction forces were
ignored. These forces give rise to tre so called "fluctu-
ation term”, whose magnitude has been the subject of much study
over the years. A semi-empirical treatment by Ejerrur‘lu),
Fuoss, and Krono‘15) took care of this situation reasonably
well for symmetrical electrolytes through the introduction
of the ion-pair concept. This does not completely exolein
the behavior of unsymmetrical electrolytes however. Recent
advances using statistical mechanics by fiayer and Foirier‘lé),
Kirkuood(17) and Keeron(13), have sought the answer to this
problem.

The interaction between two ions of given charge
depends upon the sign, structure and polarizability of the
ions, and also on the size, structure, dipole moment, and
pclerizability of the solvent molecules with which the
ions likewise interact. As the interaction between ion and
solvent increases, the ion-ion interaction decreases and
when ion-solvent interaction decreases, ion-ion interaction
increases and ion-pairs will form. For weak ion-solvent
interaction there is a likelihood of the formation of ion-

triples and higher clusters.

'. l
:75;
U
.yT..
'vCLtJ

5:1 2‘
m 1
tea-'11

U61.

h. coliticns are dilute-—the theory of bebye and
Huckel accounts satisfactorily for the “ehnvior of normal
l-l salts, such as KCl, in nqneons sclition up to concentra-
tions of about 0.01 x. Above this concentration both the

physical and the nethenotical approxiustions creek ccwn.

The essential feature of pebye-finckel theory is the
calculation of electrostatic potential W at a point in the
solution in terms of concentrations anfi charges of the ions
and tre properties of the solvent. Tris is sotieved by the
device of combining the Poisson equation of elestrostntic
theory with a statistical mechanical distribution formula.

The result for the potential qu is given by:

 

VJ CAI— . . 2....-
li-Kd r
where is the electrical charge on the control J-ion, "e"

represents the limit within which no other ion can op.roech

the central ion and

2 22 i
Kn [#170 E31 1]
w J

 

Onseeorilg) has shown that the correct time-uverofie ionic

A
hwy

distribution function is given by the equation
.11 ~41ka

Where wji represents the time-average energy to charge the

i-ion at a cistence of r free 3-ion, less the tine average energy

 

re .1

tic

3.

P11

ca.

t} g.

to

required to charge the i-icn at an infinite distance from
the 5-103, (tit still retaining in solution). "he corres-
;onding value of tji from the equation of Lebye and Tuckol

18 3131 I: 31??

 

The ficbye-Euckel theory. in its general form, accounted
satisfactorily for the thermodyneaic properties of dilute
solutions. A sicple conductance treetcecf of dilute solu-
tiona by Bongo-Huckel was not coupletely satisfactory.
Lowever. the necessary refineicnts were given by Occsgcr(6).
who studied the theory of concuotance and diffusion in
electrolyte colcticnc. Onseger develop d a theory consider-
ing the quantitative cepects of the conductance problem
which gives the indiviéaal ionic condectivity of dilute
solutions as a fcnction of concentration. At the some time
the results are applicable to the change of transference
number with concentration.

The motion of an ion under an external field is con-
plioated by two effects arising from interionio forces; the
relaxation and the electrophoretic efifect. Each of these
acts as a drag Operating between Oppositely chargedions
moving in orpoeite directions.

id The time of relaxation effect - Onsager in the
calculation of retarfiation by tre time of relaxation effect used

the equation of motion and the eqcetionsoi'continuity to

 

set 1
pm

tric

9
set up an expression for the asymmetry of the ionic atmos-
phere. This treatment involves the use of espreeeions for
frictional forces and ionic interaction and results in a
complex differential equation. Approximate solution of the
differential equation gives a retarding force which is pro-
portional to the square root of the concentration and to
the mobility of the ion in question. The expression also
includes the dielectric constant of the solvent and the
temperature.

2.) The electrophoretic effect - Another electro-
static action which lowers the mobility of the ion is the
electrophoretic effect. The ions comprising the atmosphere
around the central ion are moving themselves in the Opposite
direction. As these ions are usualy solvated they tend to
carry with them their associated solvent molecules, so that
there is a not flow of solvent in a direction Opposite to
the motion of the central ion. which 18 thus forced to
“swim upstream” against this current. The electrOpnoretic
effect normally makes the larger contribution to the con-
ductance. The basic eXpression used in the treatment of
the electrOphoretic effect is Stokes‘ Law of the velocity
of a Sphere moving through a viscous medium. It is also
assumed that a steady state is quickly reacted in which
momentum is transferred from the ions to the solvent and

vice-verse.

 

an

lit:

I. T:

rt.

00!

10
According to the Cnsager theory, the conductance of

an electrolyte is given by the equation:
A a: A. .. (NA.+4355’

whereld anal? are constants depending on ‘he dielectric con-
stant, temperature, and viscosity, and A. is the limiting;
conductance. This equation is a limiting for-n. If A is
plotted versus 0%, than in dilute solutions the points
approach a straight line whose slope is equal to “ A.+ ’3 o
For salts of a given charge type, the slopes differ little.
but the magnituée of the slepe increases markedly as the
number of unit charges on the ions increases. Onsager's
equation is very useful as a limiting form for purposes of

extrapolation.

 

l. Faces-Onsnger Theory - Recently Fucss and Onsager(2°)
re-examined the equations of continuity and the boundary
conciticn of the differential equation and incluced higher
order terms than before as well as the idea of a finite
ionic dianeter. These workers obtained a new eXpression
for the relaxation field. The resulting equation for con-

ductance has the form
. 1 1
A uA...(“/\.+4) c71+DC 1n: 4» (he +ch’/2) (l - c3?)

'3 -
where the terms higher than 83/“ are neglectec. The terms

 

aqua
Test
60115

T981

tea
tits
I 14::
an.
far
He:
met

c211

ll

Jl and 32 are eXplicit functions of ionic size, Agent: some
advent properties; D is indepenoent of ionic size. This
equation was derived only for symmetrical electrolytes.
Tests of this eguation in solvents of various Gielectric
constants give consistent results for the ionic size and
reasonable values of the association constant.(12).

2. Higher terms of the electrophoretic'efrect - Ex-
tension of Oncoger's theory by Eye and Spooning involved
the inclusion of higher-order terms in the distribution
function useé in the electrOphcretic effect. Tris treatment
gave satisfactory agreement for both confinotenees end trans-
ference numbers of alkaline earth and rare earth halides.
Recently, the integrals have been evaluated by nunsrical

methods using machine calculations(21) for a variety of

charge types, ion sizes, and concentrations.

 

l. Solvent structure - The ions in solution are
subject to strong electrical fields due to solvent molecules.
The intensity of this field depends on the value of the
dipole moment of the solvent molecules. In considering this
interaction the smaller the molecules of solute, the greater
is the force of interaction with the solvent. In general
the results of solventaicn interaction will manifest them-
selves in three different ways: effect on the notion of ions;

solvetion properties; and interaction of ions with each other.

 

1-1
11:!
Gem

I311

12

2. High concentrations - In experimental work on
1-1 electrolytes up to concentrationsof 0.01 normal, the
limiting law of Onsager can be applies. Beyond this con-
centration many varied and complex effects come into play
which cannot be quantitatively predicted. For crample,
the viscosity of the solutions is altered, and as the con—
centration increases, a quasi—lattice ferns anfl certainly
in very concentrated solution the ionic distribution must
alternate as in a crystal.

3. Unsymmetrical electrolytes - Snsymretricsl elec-
trolytes are complicated by the fact that the theory of
Fuoss and Oncoger(22) is applicable only to the synretricol
type. -Fnrther complications arise from the fact that any
ion-pair which might form still carries a net charge and

this species has on unknown mobility.

 

 

fa“

? .
vii

rlf

A.

 

Since pure Kaz n - <P (503,2 wee not available, and be-
cause of uncertainties in weighing due to hyfiration of the
salt, the stock solutions were prepared as follows:
Eenzenedisulphonio acid was neutralized with barium
hydroxide according to E2 1:: ~<P(1302)3 4- Ba(0;‘3)2—-9 Be In -?(SO3)2
+ £20. and the end-point was checked with pH paper indicator.
After purification, a solution of known concentration of

Ba :13 -¢(oo and also a solution of Na 52.0 was metathesized.

3’2 2 a
After removal of barium sulfate by filtration. the solution
was weighed and its molelity calculatea. Other solutions
were prepared from the stock by weight and stored in Pyrex
brand flasks with rubber cars to prevent loss by
evaporation.

l. Farina metabonzenedisulfonete - The crate
Ba m -Q(:303)2 obtained from neutralization tees dissolved in
a small amount of confiuctance water enfl recrystallizefi with
ethyl alcohol five or six tines.

2. Sodium sulfate - Analytical reagent arade sodiln
sulfate which might still contain ecte soéium hydroren cul-
rnno

fete and moisture wee heated for three toere at 14] C.,

cooled in a deesicotor and used to prepare etcck colitione.

13

lb
These solitions were analyzed for sulfate by bariun sulfate
precipitation.

3. Water - Preparation of a good conductivity water
free from CO2 and other impurities which may contaminate
the water is of great importance. To provide a good water
for the experimental work of this thesis, the following
essential points have been observed:

s. The cistilletion flasks end columns were thorough-
ly washed with cleaning solution and soaked in good water
for a period of time before using.

b. For the purpose of preparation of water free
from organic substances and contaminant guess, a two-step
distillation was performed. First deninerelized water was
distilled from an alkaline permanganate solution to remove
organic materials by oxidation. Secondly, the water thus
produced was introouced into a separate distillation flask,
and distilled under a stream of nitrogen. Before the nitro-
gen was introduced into this system, it was purified by

passage through gas-washing towers containing H250 , K05

I.
-

solution and water containing phenolphtnalein reopectively.

The average conductivity of the water used was 0.3?
l

C. -
x 10'5. chm cm

B.

 

The apparatus used for measurement of conductivity

is described below:

15
1. Bridge - The bridge used is a modification of the
Jones brifigc fiescribcd by Rogers and Thompson(23). It con-
sisted of an audio regocncy oscilator, an amplifier and a
cathcie ray tube. This bridge allows the use of frequencies

of #03, C00, 1000, an& £000 ops. The balance point can be

{.4

reproioce; to,; 0.10 i. The resistance showed e slight frequency
dcpenfienco which was corrected to infitc frequency in the

usual fashion. For high resistance a 20,000 onn shunt was

used in parallel with tie cell.

2. Oil Loin - in oil bath was used instead of a water
bath in rcer to decrease the capacitance effect. $cmner-
ature control to‘; 0.010 C. was obteincd using a mercury
therrorcgilator to control a heat lamp. The both was stirred
with a German - Kipp centrifugal pump and tenperctnre gradi-
onto within the both were less than 2° c.

3. Conductance Cell - The conductance cells for low
concentr.ticns wore constricted from leeos and Northrop Type
A cells. A 53C 31. Erlenmeyer flask was attachefi to the
electrofle chatter. This makes thorosgh mixing of the
solution possible. The platinum electrodes were lightly
platinizcd before using (cc. to sec. at 40 ma.). The steps
involved in a typical conductivity run are:

a.) water transfer and measurement - Before intro-
ducing the water into the cell, it was clecnco with hot,

fuming nitric acid and rinsed 15-20 times with water and

finally alloked to stand overnight filled with conéuctsnce

16
water. After drying the cell, the stopcock and standard
taper cap were lubricated with Apiczon ”E" grease and the
cell was weighcé. Water from the final distillation under
mitrocen was transferred into the cell undor nitrogen
pressure. Before introducicg any water, the trapped air
inside the cell was swept out with nitrogen for 15 minutes.
Only about 300 ml. or loos of water was used so that
efficient mixing could be obtained. after coogletion of

the transfer, the cell and water were weighcé to 1,0.01 g.

b. deacoroment of water conductance. - The cell with
water was placed irto the oil bath for at least 1/2 hour
before measuring tho resistance. flue to the high resistance
of water, it was necessary to measure the resistance in
parallel with a 20,000 ohm shunt. The cell was removed
and too contents mixed gently until reproouoiblo resistance

readings were obtained.

6. Addition of a solute — A weight burotte contain-
ing; stock solution was used to introduce 5:01th into the
water wtict woo previously weighed and whose conductance was
measured. 23 kcowing tre weight of the burette before and
after aidition the amount of solute introduced into the cell
coulfi be obtained. The addition was made with nitrogen
flowing throogh the cell. Care was exercised in introducing
the solute into the cell. to make sure that the irops fell

directly into the bulk of the liquid.

17
d. {cacarement of resistance ~ Tho resistance measure-
ment was carriea out as for the water except that the shunt
was not necessary. If the room temperature was lower than
the bath, some conicncatic: coooared on the cop and neck
of the cell. it was then necessary to heat the portion of

the cell above the bath with a hoot lacy.

n

e. ror measurement of high concentrations (.007 N.),
3 Leeds and Eortkrup type B cell was usoé. Quintions made
from the stock were prepared in Pyrex flocks onc measured
directly. The water used in this preparation was soparately

stored and its concuctanoe measured.

 

Transference numbers were measurefl by tta torin:
boundary mothol using visual observation of the boundary.
This detection is based upon the diffarencc of refractive
indices between the leading solution and the incicctor
solution. The instruments used in this exporizont are:

1. Call - Tho cell consists of cocoa and cathoée
compartments and a U tube with a calibrated pipette forming
one of the arms. The cathode onfl anode cocoartments are
separated by two hollow stOpcooks which allow measurement
of either rising or falling boundaries. Eton the stopcock
between the leaoing and indicator solutions is opened and
an adjusted potential is applied.the boundary will form and

begin to move through the calibrated tube. The rate of

13
motion of the boundary is observed visually. The anode is
mafia of a clenimally pure oadniom plug and the cetlofle 13

a silver-silver chloride electrofie made from silver sheet.

2. Current Control Apparatus - Current for the measure-
ment was supplied by an electronic conetont-current device
fiescribed by Speeding, Porter, and Wright(2”) and mofiified
by Dye. The current through the cell was also passed through
a standard 100 ohm resistor and the out developed woo
measured with a Loofie and Korthrup Type E potentiometer.

The unbalance free the potentiometer was fed to a Brown
'Electronik" amplifier which was used to (rive a synchronous
motor attached to the control Helipot" of the constant-
0urrent circuit. The current could be maintainefi constant

at 0.5 to 10 me to i 0.053.

3. Light and Telescope - The position of the boundary
was detected by moane of 8 narrow slit of light passed

through the tronei,rence tube from the back side with a

teleecoDe focused on the tube from the front.

a. Stopwatch - The time required for the boundary
to pass tetwoen the gaps that were marked on the pipette
was measured with stopwatchce which too been calibrated

against the standard KEV signal.

1V. REQJLES AND SISCLEoIOT

 

9.
Tko analyses of solutionsosofi are given in Table I.
The conéuctunco results are given in Tublc IE anfi are alone

I
traphioally in Figures 1 and 2. Fiqxre 2 is a grern of l\,

versus the square root of normality, where

/~:='4A»* 0‘ /\0»'+"’) ’2?

The values of K and 4 used were taken frog} Earned and

GK
Owen(“/ one are

I
30.5227 “o a 0.32.79 w G

 

W2"

4‘ no :1 "
= .w 4' :- 60.19 w
r: 4

Table III gives the data used for determination of

 

a'.

the cell constant. For the equivalent oomd:ctew.e cf K61,

the equation

A " A.- 3111:; + {5C 105.: C + EJC
was used.
The truneference cell was (foetal-r?l e niwbor of times

‘

using KCl solution. The values obtaineu for'f+ a;reed with

19

20
the literature values to within 0.1%.

Five runs were made with Ba~m¢(SO )7 solution which

3 ..
were combined wi-h earlier data of Eye and Cordon(‘h). The
conbined results are given in Table lV’and shown in Figure
3. The llmitiqw value of T+, denoted bz'T+° is of particu-
lar 1r1tC‘-3:3t mince it fixes a value chf-En' m- 9 (303)2‘.‘
which can be co:.;pared wit}1 conductance result-s. T+:o§fig
this give; wit}; A‘3:* a value of ‘1- ‘93? .

863.5

Tiree etiewpts were made to measure the traneference
number of eat in neiabenzenefiisulfonate without 5100888.
Only for the most concentratea solution coulé a boundary be
seen and even for this case it was too faint to follow.
This 00:11 be caused by close values of the refractive index
Of tVe lenéirg and following solutions or by tke presence

~ - .- i . \r . ++ f“- : '7'”

of a slitqt exceba cl Be or ooh ions. inis will need

firther investigation.
0

fire valxo of obtained in this work is 109.76 onm'l
- 2 n
eQJiv 1 cm . using
. l 9 +
A... a: 59.10 013:. equiv on" for Zia gives
A' a
- a 53.56 chm”l equiv'l cm“ for m- (503).
0
"£1115 1:3 to be :30 i:‘-'..»1.;I‘C3& with the amine of A- " ‘0-7

obtaiien from trunnfereuce measurements on tke beriwn snlt.
Application of the extenfied electrophoretic calculat-

e ,_ o ‘
ions of [ye an& oncoming gives an a value cf event 7

El
Angstrome. Enis is to be considered tentative, however since

transference numbers are also necéec for this calculation.

 

Errors due to actual measurenent of resistance were
certainly negligible. This is evident in retest runs of
conductivity using the high cell. Sofieat measurements
éiffered by not more than 3,0.Czfl.

Different analysis for the stock coliticns used to
make the final solution egreefi with each other to mithin
.i,9011 and the netethized solitlon was checkej for extra
ion of each relatent. ieverthelces, the solution certainly
contained sore extra Ea++ and 30# 2 because of the sol.—
bility of BaQQu. This shoulfi not have nnoe an appreciable
error in concoctauce, howeverbpn menSnregcnic anoweo that
the acidity of tie in 12.2- ? (L203 )2 etc-3k oclmtion after several
months standing has essentially unchanged fron that of the
water used. Clock determinations of t?e conductance showed
no ciange during this periofi. This infiicstoe to oeuooposition
or evaporation of the SOIJthfl. As the ireph or conductance
zgnfiga,the square root of concentration shows, at concentra-
tione below about 0.005 K the changes in conflnctunce aro’in
accordance with Oneager'c theory. et Litier concentrations
where the theory mathematically breaks down, egsteaetic
deviations set in. At least some of theJe chiations can

be attributed to higher terms in the oleotrorhoretic equation.

22

_‘ I
It will be iz-itcrcsting to see whether other salts of m—¢(2203)2
with univsiont cationseleo give agreement kits ‘heory or

whether a pronounced ion-size effect occurs.

T BLE I

AEAJYEIb CF ETQCK SOL¢TICN #19 fiKTATHFJIf CALCTLATIOES

- hXfierirfintal Run.
:3 m -¢(.;C )5
U 3 ‘I c 'fT

LO. 1 N0. 2 L00 3

 

weight of sciatica 5.7h53 5.11u6 5.23u1
keight or ranch 0.2199 0.2uoo 0.2h55
o

Rolallty (Calculated) .2176 0.2173 0.2172

Height of solution 5.8061 5.3113 5./791
W91; t Cf FalOu 0.5563 0.5159 0,5n51
Folallty (caloulateé) O.h3€3 0.h335 0.h3€6

 

FiotaticagAs - (33.2.0 5. Ba :2 q? (.2:03)2 31.001: = 0.2145 equivalent

" \ 3::

29?.52 g. fig no, stock a 0.2fif equivalent of

fa,i0
2c-

C.D’2;? m rolallty of final Bolition

Series 1 - Cell F-3; constant - 1.0357

 

 

 

, -I
1'00 conductance = 0.3143 x 10"“ cm’1 ,ohm .
Initial weight cf water = 305.0?
Folality of stock a c.0025?3
(at 25° c)
_ . ,, : :-::: : =:: ::::::::=- g: _.._.,_.__§1.:_.: ,1 3
Folulity of Nov ality of ’
solut on _ solutign R0. A A0
g. Stock x 107* x 10
1.0635 1.4665 2.9203 3:h77.6 106.32 109.30
1.7953 4.163;; 14.9021 13730.0 105.12 109.311
2.8191 3.3u75 7.6723 12773.5 105.23 109.33
3.9156 5.325% 10.6193 9315.0 10L.L3 109.90
5.1654 (.9972 13.9530 7137.0 103.32 110.09

 

veries 2 - C811 “-3; CORSt-Snt a 1.0 357

.70 C02010t0000 = 0.3? x 10‘"6 chin"3 on'l

Initial 001. t of water a 235: .62

101211ty of Stock . 0.0L2578

 

 

 

(at 25° c)
z: ”’ “I * 5 * W w 2*“tg==:
Eolility Rorflality /
8013:1011 sciftion R” A A.
5. Stock x 10’4 x 104
3 03h1 0.2609 8.0955 11567 105.05 109.90
0.0930 6.2310 12.525 7925 100.10 110.0%
5.9071 8.2722 16.0955 606? 103.30 110.13
0.3291 11.0903 22.9206 0006.6 102.03 110.52
10.3203 10.1736 23.2730 3600.1 101.72 110.65
12.5 37 17.0531 30.0050 3010 101.00 110.30
14.1500 19.1573 33. 222 2593.? 100.59 110.97
10.2504 21.37;:>7' 03.6300 2370.1 100.17 111.16
21.2936 23.2031 56.2395 1337.0 99.20

 

W? w-
m

Series 3 - Cell E—l; constant = 29.19
(at 25° C)

1.— _1_¢A4_L

Iolality
g. of
a. stock holation Solution Zenslty Eornulity

197.58 39d.30 0.0209? 1.000“ 0.0h172
275.7 010.32 0.02322 1.0015 0.05626
.stock stock 0.0h2578 1.0039 0.0?Lh?

1-3
5‘...
t‘l
{'11
H
H
H

CELL CCFQTAET PET??VIN0TIQQ
Low C611 - 5-1
Nolality of stock - 0,04913

N20 Confiuotanoe n 0.83 x 10'6 0111!"1 om'l

 

 

 

(at 25° c)
ermv¢W --:;-~ —~ vw
Holality Nartality
. 5035011011 5013:1311 R” 'A
g. 0tccu x 10' x 10
1.6776 2.6320 2.67h1 25503 1h3.3h
2.5520 n.0631 0.0502 170&0 103.00 1.0370
3.5499 5.6009 5.620% 12363 147.66 1.0370
h.L3b3 7.1050 7.03b2 93b9.3 1&7.h0 1.0366
5.5271 3.7270 0.7010 0003.3 107.15 1.0366
6.8912 10.933 10.301 (£99.u 105.3. 1.0352
8.2%?) 12.?09 12.571 5h?2.0 lb€.5? 1.0363

'i

{7
J

" I

*r 7 1-
.11.3111: 1V

15.731513112110110 100112503 CF Ba 0MP (003)2

 

28,

 

 

 

 

 

(at 25° c)

W Y 7“ 7 .......»...fi * 1 :3
C’ J? T.
0.02036 .1026 0.5123
0.00100 .2025 0.5137
0.06102 .2070 0.5102
0.06182 .2070 0.5106
0.06090 0.25h6 0-5153
0.10906 0.3302 0.5100
0.10900 0.3302 0.5102
0.10906 0.3302 0.5107

 

1.4—~ _—

rrFII

29

. «83933280
ho noon unaud- osnuop ougunsuuvouougopofl I532. uo 3.3926900 .a 9.3:."

 

 

.2. 0»
«a nu 0N ON 0. N. O Q
q . fi

q -

 

1 No.

 

O.

 

 

o:

 

30

505303280 Mo F323 house 23 song 83325

.N ouawdm

 

 

‘

u

q

0.x

.0
a

bk.

«'0

‘

 

l

o:

 

 

N:

 

31

.cannoun«donounonuuoaulsdham Mo hopes: conchouunuua

.n unawam

 

IQ;

Ayn. .uN. nz.

 

 

- - -

fine». 0-E0m no. oz

muzuzuumzth

.m.
0;.

 

 

 

5.
6.

7.

9.
10.
ll.
12.

16.
17.
18.

1'21""738'11 .1” 3". C319"?- f0? 0 I 12.511314 (1911‘) '

Iffi’"1?3Ui§"-h End'- I.:. Half-DPT), "Le-ltvermrjrvev} “er
electrolrtE” Tefibner. L010210 (133.).

-._. Bf r\ I h-
C. I‘BY‘LJET. U0 3"“. 93'9ng $00.; 33,, 13""? U (192-; )0
J(fi€3, arc C. K. Bollinger, J. Am. Chem. 500.. 510
011 (1931).
Cchye an! F. ?Jokel, th8., 2£,(1923) 135.
{
cfifitjzr, i: 19?? 3
,pr--

 

 

0. “yo, 1. P. Faber and P. J. rarl to be published.
Covfion, C. H. firgbakcr and J. L. 130. to be nublisked.
L. Ty0.' anublished paper.

Gorgon, C. H. £r1baker and J. L. Eye. to be published.

1". Daniel. Fir-110 Trans-9 12.2. 97 (1:3)).

4

D3“) &n(1 if. H. Spefidlng, J. Am. Cren. 3°C.. EE:, CA“?

a-
.-

rn*tbx, D. r. $11000, 0n& C. h. Collie, J. Chem.
Pf‘vS.’ II" (19’1“?) 10

BIerrIm, tr _1. DansLe Vlcnegk. Selskeb., math-fys
’ca...;. 20. 9 (ISP’).

rn

.--~-«~? A Y T .fl ._ “
.. S - - .. fi. 37232, 0. rm. Cher. 00C., 7‘, 330L

(195?.

a..-

9‘)

c. $01.13r, J. Chem. rhys.,I;;, 965. 972 (1773).
3, ll'IOO'o J, Chea. 11w ys.. 24 ?€? (133’).

Kreron, J. Cfiet. rhys., 3;, 630 (193:).

32

19.
20.
21.

f)
(—0

23.

211.

25.

J.

33
Ocsager, J. Chem. Phys., 3, 599 (193M).

’1'!

L. Dye and co-workers. Unpublished.

H. ?1053 and L. Onanger, J. Phys. Chem.. £1. 668 (1957).

""f‘,

.h.‘ 7:", 1 ~ ‘ . ’\ -_n .‘ ~ ‘ ‘ I . ‘ - ,4 I
b. LHOVLJQUI"! £12721. 1. 1115);..9Ib, PM? If)??? (.L‘}_‘)C:)._

E. gpeddind, I. 3. F_ tar, and J. 1. hfil Kt. J. km.
Che1. 103., ya I 1'71 (193:).

3. turned and h. J. Chen ”1he ELJS. Chew. 3f “lestro-
lyric Sol1tiuns," (3rd ed.), Reinrolé, N.Y., 1958.

L. P33 533 G. Gordon. to be publiaheé.

E‘FQ'Py J. P'v’iY‘r‘F. cre'noo (l: 66’} (1957).

 

R

I!

 

”1111711 11111 111 11111111111

IES