TEE BEHAVIOR OF THE GRAPHITE ELECTRODE IN
ELECTRO-CHEMICAL CELLS

I

The Determination of the OxidationReduction Potential of Sterile Culture
Media By Means of the Graphite Electrode

II

Oxidation-Reduction Studies in Relation
to the Growth and Differentiation of the
Species of the Genus Brucella

III

The Preparation and ITse of the PlatinizedGraphite Electrode

By
C. D # Tuttle

A THESIS
Submitted to the faculty of Michigan State
College in partial fulfillment of the
requirements for the degree
cf
Doctor of Philosophy

Departments of Chemistry and Bacteriology
East Lansing, Michigan
1933

ProQuest Number: 10008230

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note wilt indicate the deletion.

uest
ProQuest 10008230
Published by ProQuest LLC (2016). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6

43

JO

//
/£
/-i
(4CX. of K M h 0 ^

IS

(6

f7

Fig. 6. - A composite curve of triplicate electrometric
titrations on acidified 0.1W FeSO^
with
KMnO^t using a graphite electrode*

44
Results of study (continued)
Figures 1, 2, and 3 are electrometric titrations
on 20 cc. of FaOH with 0.1H KC1. These three titrations
were carried out in the same manner except for a differ­
ence in the electrodes that were used in following this
reaction. A study of Figures 1 and 2 showed that plati­
nized platinum and platinized graphite agree. The change
in voltage at the inflection point was a,bout 0.5 cf a volt.
The curve in Figure 3 shows that graphite electrodes do not
agree in alkaline solutions and that the voltage change at
the inflection point was only about 0,2 of a volt. This
observation would indicate that graphite electrodes do not
follow a pH change as well as platinised electrodes.
Figures,4, 5, and 6 are electrometric titrations of
20 cc. of acidified 0.1N FeS04 (hH4 )S04 with EMn 0 4 . The three
titrations were carried out in the same manner except for
a difference in electrodes. The curves in Figures 4, 5, and
6 show that platinized platinum, platinized graphite, and
graphite are in agreement. The only difference appeared in
the upper knee of the electrometric tit rat ?_on curve when
graphite was used as an electrode. The bend in the curve
is not so sharp. The graphite electrodes required a little
more time for equilibrium between readings than was requir­
ed for the platinized electrodes. All three of the eleo« •
trodes followed this oxidation-reduction change. It is evi­
dent from the titration curves that platinized platinum
and platinized graphite agree. Graphite electrodes do not
follow a change in pH# but do follow oxidation-reduction
changes.
Discussion
In preliminary experiments it was difficult to produce
platinized platinum and platinized graphite hydrogen revers­
ible electrodes that would agree with each other closer than
0.0002 of a volt in either acid or alkaline solutions. With­
out any change in procedure, the agreement was increased
from ten to a hundred fold by the addition of an electrical
shield (2) surrounding the cell. Then, with the elimination
of all errors due to contact potentials and heat effects
and the development of better technic, individual readings
were made accurately to one millionth of a voltcPlatinized
graphite electrodes were found to be a little slower in
reaching their equilibrium with the hydrogen ions than the
platinized platinum. This was especially tT»e in alkaline
solutions.
Four hours were required to reach equilibrium after
starting the flow of hydrogen. The length of time depended
somewhat upon the size of the cell and the rate of flow of
hydrogen.
The results of these experiments show that platinized
graphite may be substituted for platinized platinum in elec­
trometric titrations. In acid solutions and weak alkaline
solutions up to one normal, their agreement was within 50

45
millionths of a volt. In more alkaline solutions than one
normal, their agreement could not he relied upon within
0.2.of a millivolt. Without using the proper shielding,
the agreement was not greater than three or four tenths of
a millivolt in all solutions studied. The graphite elec­
trodes do not agree with these electrodes when used in ti­
trating an alkali with an acid. They do not even agree with
each other in alkaline solutions. The change in voltage at
the inflection point of an alkali-acid titration was small
compared with that of the platinized platinum and plati­
nized graphite electrodes. Graphite electrodes agree with
the platinized electrodes in oxidation-reduction titrations
if sufficient time is allowed for equilibrium. It was
found that graphite electrodes when treated as described,
were slightly more sluggish when compared with the plati­
nized electrodes. In view of these results, a platinumgraphite electrode couple for electrometric titrations
would be very satisfactory for alkali-acid titrations, but
unsuitable for oxidation-reduction titrations. Thus, it
would appear that the graphite electrode is very suitable
for following an oxidation-reduction change in a solution
where the pH change was slight as in buffered liquid cul­
ture medium inoculated with bacteria.
Summary
1. Platinized graphite electrodes may be substituted
for platinized platinum electrodes. They may be used in
acid solutions, weak alkaline solutions, and often in
stronger alkaline solutions.
2. -Graphite electrodes do not register a change in pH.
3. Graphite electrodes register oxidation-reduction
changes if sufficient time is allowed for equilibrium.
1.

References
Ralzton, Robert R.
Potentiometric titration of acidity in oils.
University of Michigan, Ann Arbor. Ana.lytic Industrial
and Engineering Chemistry, 4, 109, 1932.

2.

White, Walter P.
Leakage prevention by shielding especially in poten­
tiometer systems.
Journal of American Chemical Society, 36, 2011, 1914.

3.

Clark, William Mansfield
The determination of hydrogen ions.

4.

Poulk, C.W* and Marion Hollingsworth
A precision measurement of the composition of the
constant boiling mixture of hydrogen chloride and
water.
Journal of the American Chemical Society, 45, 1220,
1923.

THE BEHAVIOR OF THE GRAPHITE ELECTRODE IN
ELECTRO-CHEMICAL CELLS

I

The Determination of the OxidationReduction Potential of Sterile Culture
Media by Means of the Graphite Electrode

II

Oxidat 1 on-Reduction Studies in Relation
to the Growth and Differenbiation of the
Species of the Genus Brucella

III

The Preparation and Use of the PlatinizedGraphite Electrode

by
C* D. Tuttle

A THESIS
Submitted to the faculty of Michigan State
College in partial fulfillment of the
requirements for the degree
of
Doctor of Philosophy

Departments of Chemistry and Bacteriology
East Lansing, Michigan
1933

r m BEHAVIOR OP TEE GRAPHITE EIECTRODE IN
EIECTRO-CMEMICAL CEjITiS

I

II

The Determination of the OxidationReduction Potential of Sterile Culture
Media "by Means of the Graphite Electrode

p.l

Oxidation-Reduction Studies in Relation
to the Growth and Differentiation of the
Species of the Genus Bruce1la

p. 11

III The Preparation and Use of the PlatinizedGraphite Electrode
p.31

X - T h e Determination of the OxidationReduction Potential of Sterile Culture Media
by Means of the Graphite Electrode
It has already been demonstrated that the potential of
inorganic chemical oxidation-reduction systems and a few
organic ones may be measured at the surface of
bright no­
ble metal electrode with the conventional potentiometer
without difficulty. On the other hand in many organic oxidation-reduction systems, it is often difficult to make ox­
idation- reduction potential measurements with bright noble
metal electrodes by using the conventional potentiometer
due to the ease with which the electrodes are polarized.
It has been shown that this difficulty is increased in or­
ganic systems that are poorly poised (l). Bacteriological
culture media are examples of extremely poorly poised or­
ganic systems. Pour general methods have been employed to
overcome or to avoid polarization and its effects. They
are: the use of oxidation-reduction dyes to indicate the
extent of oxidation or reduction through their color changes
(2), the use of a vacuum tube voltmeter, or a potentiometer
with a vacuum tube arrangement or electrometer as the null
instrument (3), by making use of poising effects such as
poising substances, the passage of highly purified nitrogen
or the exclusion of air (4), and the use of an electrode
that is not easily polarized and permits of the use of the
conventional potentiometer (5), (6).
In the light of previous studies of this nature it ap­
pears desireable to find a suitable electrode which would
materially reduce the laborious technic -required in meas­
uring the oxidation-reduction potentials in systems of abiological nature. With this view in mi-nd, a study of elec­
trodes for use in measuring the oxidation-reduction poten­
tials of liquid culture media used for bacteriological pur­
poses was undertaken.
Experimental
A pparatus and Materials. The apparatus used in these
experiments w a s t h a t ordinarily employed in making elec­
tromotive force measurements in physical chemistry. Two
such sets of apparatus were used in connection with this
work; one at 25c C. and the other at 37DC. A small thermo­
stat was built to operate at 37°C £ O . 0 1 ° C . The thermostat
for 25° G was larger and could easily be maintained within
.005°C at 25°C for a long period of time.
A Leeds and Worthrup student type potentiometer with
accessories was used in connection with the work at 37°C.
The galvanometer was a high grade Leeds and Northrup
pointer type. The type K potentiometer and accessories
was used for the work done at 25°C. This included a pre-

2
cision type of- gadveenomeLter which was very sensitive o
smallr changes in current. Both potentiometers were shxelded'as described by 'White (7), The metal parts of the ther­
mostats were also connected to the main shield.
The cells used in these experiments were of simple
design, A IE KC1 calomel half cell with a long side arm
was used as the reference electrode for the studies at
37°C. The side arm dipped into a large intermediate ves­
sel which also received one side of the agar bridges from
several half cells containing the electrodes to be tested,
A similar arrangement was used for the work at 25°C. except
that a IP hydrochloric acid hydrogen half cell was used in
place of the calomel half cell as the reference electrode*
This cell was connected with the intermediate vessel by
use of a salt bridge containing a stopcock. The measure­
ments were made through the closed stopcock. The electrodes
were tested in half cells made from pyrex test tubes 25
mm, diameter and 100 mm. long. These tubes were fitted
with a rubber stopper-containing holes to carry the elec­
trode, an agar bridge, and a glass tube plugged with cot­
ton. A second type of half cell was also used. It was a
special flat bottom freezing point tube fitted with a rub­
ber stopper having holes for three electrodes, an agar
bridge, and a glass tube plugged with cotton,'
All chemicals used were Baker’s analyzed. The calo­
mel was prepared as described by Lewis, Brighton and Se­
bastian (8) from Baker1s analyzed chemicals without fur­
ther purification, The mercury was cleaned by running it
through a fine capillary several times. It was-dried in
vacuum over phosphorus pentoxide, The 1M KC1 solution
was made up by weighing the KC1 and dissolving it in the
required amount of distilled water. 13? H'Cl was prepared
as described by Foulk and Hollingsworth (9). The KC1 agar,
for agar bridges, was a 2 per cent agar in IE potassium
chloride solution.
The electrodes used in this study were made from onefourth inch sticks of special Acheson graphite cut to a '
length of four inches. After considerable study of methods
of preparing the electrode for satisfactory use the fol­
lowing procedure was adopted; The electrodes were first
cleaned with a fine grade of steel wool and polished with
a clean cheese cloth.They were then placed in distilled
water and boiled slowly from four to six hours. The boil­
ing under distilled water must be of such length as to
remove all traces of material left on the electrodes from
a previous experiment. Frequent changes of distilled wa­
ter hastened this process. They were cooled under distilled
water. Acheson graphite electrodes when prepared in this
manner may be used in measuring potentials with an ordi­
nary student type of potentiometer.
All of the experiments with electrodes were made on
two kinds of bacteriological culture media. A liquid

3
synthetic medium and liver infusion medium were used.
The synthetic medium was prepared by Hershey in this lab­
oratory and was composed of the following ingredients:
Water (distilled)
100 cc.
Yeast extract
0.3 grams
Ammonium citrate
0.5
Sodium chloride
0.5
Potassium dihydrogen phosphate
0.2
Ammonium sulphate
0.2
Ferric ammonium citrate
0.001
Magnesium sulphate
0-.05
Potassium nitrate
.0.05
The pH was adjusted to 6.6 with sodium bicarbonate.
The beef liver broth was made by extracting one lb,
of finely ground beef liver with 500 cc. of tap water in
flowing steam for two hours. To 500 cc* of the extract
was added 500 cc. of tap water, 10 grams of Bacto-peptone
and 5 grams of c. p. sodium chloride*. The mixture was
heated in flowing steam for 30 m i n ,f filtered and steri­
lized in the autoclave at 15 lb. pressure for 30 min-. The
pH was adjusted to 6.6 after sterilization.
Method of filling agar bridges and half cell. The
agar bridges were made by filling a H tube- made of soft
glass 2 mm. in diameter to within an inch- of the top of
the limbs with 2% agar in molar KC1. They were then steamed
for 20 min. and -cooled. The ends of the glass tubing, were
sealed off and steamed for 20 m-in.- to test- the- seals for
leakage. A mark was made- with, a file jUs-t below- the level
of the- agar on each arm of the bridges. The bridges we-re
wrapped in paper and sterilized in- the au-todave- at 15' -lb.
pressure f-or- 30 minutes-. The other- glassware was washed
and sterilized in h-ot-air. The assembled half c-ells minus
the agar bridges were placed in a large suitable vessel,
wrapped in paper and-sterilized- in -the'autoclave*- Follow­
ing. sterilization and cooling, the half cells were partly
filled with the sterile liquid medium and one limb- of the
agar bridges inserted through the holes in the rubber
stoppers.'Only the tip on the side of the agar bridge
which passed through the rubber stopper was broken off at
its file mark when each half cell was assembled. The half
cells were connected to the calomel -intermediate cell by
breaking off the tip of the other limb of the agar bridges
at' its file mark and inserting the arm in the hole of the
stopper on the intermediate vessel.
Standardizing the calomel half cell. The hydrogen for
the hydrogen half cell was produced by electrolyzing a 10
per cent solution of HaOE. The hydrogen generator wa.s a
modification of that described by Clark (10). The hydrogen
was purified by'passing it -over an electrically heated
platinum spiral. A piece of heavy capillary- glass tubing
about 1 meter long was placed between the generator and
the hot platinum wire as a safety device. Since the chem­
icals used in the calomel half cell were not repurified

4

03
CO

o
CO
00
CO
•

c\?
CO
03
CO
•

03
CO

o

o

o

o
H4
03
CO
•

o

03
o
03
to
0

o

o

4

4

4

4

4

4

H4
r00

03
rH
<J>

CO
02
0>

H4 in
c\2 (M c\2
a> O'. 03

c-

cO

•

to

to
3

to
o

CO

H*-

oto
03
to
•

O

o
•*

•

o

o

4

4

4

4

4

4

H4
03
z>
CO

CO
o
03
CO

H4
CO
03
CO

co
to
03
co

CO
C3
03
CO

o

o

to
03
03
CO
*

m

CO

O
<D

4

03.
Z>
03
CO
<»

C\2

&

4

02

CO
to
03
to

CO
CO
03
CO

03

£>
03
CO
•

o>
to

4>

■4

4

4

00

to

03
CO
•

CO

co
o

02

o

02

03

CO
•

03

in

03
CO
o

4

to

4

o-

4~

0

03
in
o
•
o

E**

•
o o

o

rH

4

o

o
4

o

o

4-

o

to

CO

4

o

to

to

CO
in
o
H4
•
O

03

I
—I

03

CO

+ 0.4063

+

rH

•

o

00

r~i
03
CO

00
to
a>
to

a

O

rH

rH

03

CO

.>

*

o

o

o

o

+

4

4

4

4

4-

o

CO
o

in
rH
O

to
rH
O
H4

to
rH
O

o

o

O

O

o

4

4

4

H
O
H*

CO

o

• •
o o o

«t?4

4

4

+

4

4

*+

o
H4
• *
o O
O
+ 4
4

CO
o
rH
H4
•
O

in
o
rH
H4
•
O

rH
O
rH
H4
•
O

H4
O
rH

to
O
rH
H4
*
O

03
H
m
H4

to
H
rH
H4

O
,-H
fw-j
H4

V»
O

H4
O
rH
H4
•
O

H4
03
O
H4

o

in
co
o
H4
♦
O

O

O

O

O

03
H
rH
H4
*
O

4

4

4

+

4

H

4

4

4

4

4

4

4

CO
to

03

00

03

CO

to

H4

H4
H4

H4

rH
rH
02

H4

H4
H4

02
H4
H4

03
C"

03
CO
rH
H4
•

CN2
03
rH
H4

00

to

02
H4

O

O

O

O

O

O

O

H4

LO

t

•

A

03

m

co

CO

C\2

to

CO
H4

H4
H4

o

O

H4
H4
•
O

•

0

4

'4-

O

rH

H4
H4
•

•

o

0

o

•

•

•

•
O

•

C2

H*
•

H4

o

o

O

•
• H4

•

•

•

o

•

•

4 0.3978

Hr

to
O
H4
•
O

t 0.3921

4

O
n
o
H4
•
o

+ 0.4009

Table
electrodes

4*

Ct]
to
o
H4
•
o

+ 0.4126

rO
o
u
+3
0)

of different

in
H4
o
H1
e
O

*

CD

CO
to
o
H4
0
o
*

o
H4
O
H4
0
o

O
O
H4

to
in
o
Kjf
•
o

to
z>
03
to
•
o

0.4208

I
in synthetic

medium

to

rH
in
o
•
O

02
tD
CO
cO
*
o

4

6

•H W
0) Ct3

CO

O

O

C

O

O

rH
•H

H

rH

C\2

CO

H4

lO

rH

rH

rH

rH

infusion medium

0

of different

Table II
electrodes in liver

0
Xt
O

fH
-P
O
0
rH
W
<H CO
O
fj
M
0

Jj
02

£>
rH
CO
CO
to

m
co
02
CO

o

CO
rCM
CO
•
o

o

z>
c02
to
•
o

<0
CO
to
to
•
o

CO
o
CO
to
•
o

in
00
02
CO
•
o

o
c&
02
co
•
o

o
£02
CO
•
o

02
o
to
•
o

CO
CO
CO
to
O
to

CO
o
CO
co
to
o
to

CM
rH
CO
CO
•
o
to

m
rH
CO
CO
•
o
to

o>
CM
CO
«
o
+-

CM
O
OcO
•
o
-to

in
vo
CO
0
o
■to­

CO
CO
VO
to
•
o
to

CM
CO
VO
cO
•
o
to

rco
vo
CO
to
o
to­

o
Cco
to
o

rn
vO
CO
to
o

cr>
CM
c0
•
o

E'­
en
CO
CO
•
o

ol
vo
to
CO
•
o

CO
rH
CO
•
O

O
toi*
CO
•
o

o
ro
CO•
o

in

CO•
o

o
H
co
•
o

rH '

in.
CO
•
o

o
c-

*
o

o

r02
CO
to
•
o
*

02
02
in
CO
•
o
*♦

CO
02
LO
CO
«
o
to-

o
VO
in
CO
•
o
•to

xj*
cO
in
co
■to
o
to

in
CO
to
o
•to

rH
O
CO
CO
to
o
4

00
in
z>
co
to
O
+

rH
in
2>
to
to
o
to*

in
CO
tco
•
o
+

02
o
CO
to
o
■to

in
VO
o
rH
£>
rCO
CO
to
to
o
o
+■ . Hr

00
CT>

o
CO
o
to
o

o
CO
cr>
CO
•
o

in
02
CO
CO
•
o

oo
£>
to
to
o

CO
rH
ID
CO
to
o

rH

•
O

CM

o
■

CO
•
o

O

CO

to
CO
CO
CO

lO
in
co
cO
•
o

CM
o
to
VO
in
Kt«
to , CO
•
•
o
o

to

o

VO
CO
CO
•

-

cr>
lO
CO
to
to
o
+

CM
CO
•
o

O o
rH in
GO
C
CO•
to CO• O
to
O o
o
o
-to to to to-

co
rH O
o
in
vo
CM O
V
O
C- o- VO
VO
to
co
co
CO co
• • to to •
o
O o
o
o
to- to
to­
to
♦

o

vo

00
o

o
in
CO

to '

.

Cfl
02
CO
•
o

CO

CM o>
CO rH
VO VO
CO CO co
• • •
o
vo

o
to-

o

o

to-

to-

rH

rH

G
•H CO
0 d

•H
tr»

CO

in

t-

o
rH

rH
rH

CM
H

CO
rH

in

6

0
0
o

I
P

0 «H

FJP

0

I
o

0 0
rH 0
0 *d

0 rH O
Ft O
0 > Ft
o
0 fcl0 F
c«

era
p
iH

Ch C P F! -P

O
>

*H H Ch H
d P
*C(

to m
to CM
CO
CM
o• o•
o o

o> CO in
CM
00
to CM to
o• o• o•
o o o

to

CO

CO

Ft

*rH
(D
O
C
0

Jh
0
Cm
C*H
•H
tJ
§
X
d
0
bD
C
•H
*5

0
Ch 0

O

'd
O0

Ft Ft P
0 p P
p O *H

0
0 0
3 rH O

CM

fe 0

♦rH

fH
H

H
0

t
—I
.0
ClJ
§H

O

CO
CO
0
<d

•r-t !>

■H
0
o
c
0

m
H
to
O•
o

o>
in
o
•
o

co to
to rH
co CO
o• o
•
o o

00

rH
CM

H

CO

to
cr>
o*
o

o

Ft
P
o
0

I—I
0

p

fl

0

Ft

<0

rH

00

rH

0

«R

CH
•H
'd

Ch

o

P3

C\J

in

?
and the concentration in the agar bridges varied slightly,
it was necessary to standardize the calomel half cell.
Standardization was accomplished by substituting a hydro­
gen reversible half cell for one of the half cells con­
taining an electrode to be tested and measuring this cell.
The resulting E is corrected for temperature and pressure.
All values can now be calculated into values as though
the readings were made with a hydrogen reversible elec­
trode (EjJ by a simple algebraic addition.
Results of Study
In table I are tabulated the corrected readings for
six graphite electrodes in the synthetic medium measured
against the hydrogen reversible electrode over a period
of fifteen days. The cell measured is represented as fol­
lows: graphite electrode, synthetic medium, KC1 agar
bridge, IF HC1, 1%, Pt at 25°C. These readings are repre­
sentative of many similar experiments*
In table II are tabulated the corrected readings for
six graphite electrodes in the liver infusion medium
measured against the hydrogen reversible electrode over a
period of fifteen days. The cell measured is represented
as follows: Graphite electrode, liver infusion medium,
KC1 agar bridge, IF HC1, ET^, Ft at 25°C. These readings
are representative of many similar experiments.
The values given in the table I show that the elec­
trodes in the synthetic medium changed slightly from their
original values after a period of fifteen days incubation.
As the period of incubation increases, there was a tenden­
cy for the electrodes to drift more positive. The elec­
trodes in liver infusion medium moved from their original
values more than those in the synthetic medium. The change
in potential of the electrodes, in this case, was in the
negative direction. But even at that the negative drift
was small in comparison to that which obtains for metal
electrodes.
The tabulated E^ readings in the two tables indicate
that there is a much closer agreement between electrodes
as the period of incubation increases. The electrodes ap­
proach a close agreement after a period of seven to ten
days incubation in either media. A longer period of incu­
bation does not bring about any closer agreement.
In continuing this study it was decided to hold the
complete half-cell in an incubator for seven days at
37°C before mounting in the thermostat. The E^ readings
of several electrodes in liver infusion broth held in this
manner are tabulated in table III.
Table III contains condensed data from five experi­
ments using 87 electrodes. It shows the number of elec­
trodes used in each experiment, the maximum variation of
voltage between all the electrodes and the maximum differ­
ence in voltage between all electrodes after removing the
most erratic ones for each of the five experiments. The

8
data in the table reveal that only 18,4 per cent of the
electrodes deviate to a considerable degree from the aver­
age potential of the graphite electrode in liver infusion
medium. If large noble metal electrodes had been used in
the liver infusion medium from one-fourth to one-half of
the electrodes would have been even more irregular than
the graphite electrodes,
^
If a small amount of IF acid or base is added to the
medium in one of the half cells, the value of the oxida­
tion-reduction potential is not affected as much as one
would expect.
Discussion
The results of these experiments indicate that Ache­
son graphite electrode is a suitable electrode for oxida­
tion-reduction potential measurements in liquid culture
media used for growing bacteria. In view of the fa,ct that
it was not necessary to exclude the air from the half cell
or to make use of a vacuum tube arrangement or electrom­
eter in making the measurements, it is quite evident that
the electrode was very well poised. The measurements were
easily made with the ordinary student type potentiometer
using a high grade pointer galvanometer. The use of this
type of electrode has a distinct advantage over electrodes
made of noble metals in that it is often necessary to ex­
clude the air and to use a vacuum tube arrangement or
electrometer in making the measurements. It was necessary
to use a potentiometer shield and to connect all the met­
al parts of the thermostat to this shield in order to ob­
tain very close agreement between several electrodes in
the same cell. The agreement was within 35 millivolts, A
slight change in pH had only a negligible effect upon the
oxidation-reduction potential.
The possibility of using a common metal as an oxida­
tion-reduction electrode in the synthetic media was exam­
ined. Several of the metals, after standing in the medium
a few days, were discovered to be dissolving slowly. As a
result of this finding the more common metals were exam­
ined for solubility in the synthetic medium.
It was found that silver, nickel, mercury, lead,
magnesium, manganese, aluminum, antimony, cobalt, cadmi­
um, tin, zinc, and copper were slightly soluble. Chromium
was the only metal of the less noble ones which was not
soluble. The solubility of the metal was increased by
autoclaving in the synthetic medium at fifteen pounds
steam pressure for 45 minutes, it was possible to detect
the metals in solution by qualitative analysis.
Summary
An electrode made of special Acheson graphite is
satisfactory for oxidation-reduction studies of linuid me,
dia used in growing bacteria.

9
The electrode in liquid media is stable enough to
peTmit the use of a potentiometer.
Graphite electrodes when treated as described above
can be made to agree with each other within 35 milli­
volts .
The oxidation-reduction potential of graphite elec­
trodes is not affected by small changes in pH of the
medium.
1.

References
Michaelis, L.
Oxidation-reduction potentials.
Translated by Louis B. Elexner.
J.B.Lippincott Co., 1930.

2.

Lubos, Rene
Observations on the oxidation-reduction properties
of sterile bacteriological media.
The Journal of Experimental Medicine, vol. XLIX,
Ho.3, p. 507-523, 1929.

3.

Allyn, W. P. and I. L. Baldwin
Oxidation-reduction potentials in relation to the
growth of an aerobic form of ba.cteria.
Journal of Bacteriology, vol. XXIII, Ho. 5, 1932.

4.

Knight, B.C.J.G.
Oxidation-reduction studies in relation to bacterial
growth. IX. A method of poising the oxidation-reduc­
tion potential of bacteriological culture media.
The Biochemical Journal, vol. XXIV, Ho.4 ,p.1075,1930.

5.

Coulter, Calvin B.
Oxida.ti on-reduct ion equilibria in biological systems.
1. Reduction potentials of sterile culture bouillon.
The Journal of General Physiology, vol. XII, Ho.l,
p. 139-146, 1928.

6.

Elexner, Louis B, , and E. S, Guzman Barron
Oxidation-reduction potentials at carbon and tungsten
electrodes.
Journal of American Chemical Society, vol. 52,
P. 2773-76, 1930.

7.

7/hite, Walter P.
Leakage prevention of shielding especially in
potentiometer systems.
Journal of American Chemical Society, vol. 36,
p. 2011, 1914.

8.

Lewis, Gilbert H . , Thomas Brighton and Ruben L. Se­
bastian, A study of hydrogen and ca.lomel electrodes.
Jour, of American Chemical Society, v o l .39 ,p. 2245 ,lfl"7.

10
9.

Foulk, C, W . , and Marion Hollingsworth
A precision measurement of the composition of the
constant hoiling mixture of hydrogen* chloride and
water*
Journal of American Chemical Society, vol. 45,
p. 1220, 1923.

10.

Clark, W. M.
The determination of hydrogen ions,
Baltimore,, 2nd edition, 1922.

II - Oxidation-Reduction Studies in Relation to the
Growth and Rifferentiation of the Species of
the Genus Brucella
The object of this investigation has been to deter­
mine the oxidation-reduction time potential associated
with the growth of strains of Bruce 11a me litensis, Bru­
cella abortus t/ and Brucella suds in a suitable liquid- culture medium T n the piesence of and without the addition of
certain growth inhibiting dyes. It was thought that the
results of such a study might furnish further informa­
tion on the biological relationship of the three species
and enlightenment as to the mechanism of the selective
action of the dyes toward the growth of B ruce 11a (l). It
is obvious that in making a comparative study of this na ­
ture of closely related aerobic bacteria, the growth con­
ditions that are to be maintained should closely approx­
imate those normally used in growing the organism. In oth­
er words, if there are metabolic differences in intensity
and capacity in the three species of Brucella when grown
on culture medium under aerobic conditions, the differences
should be measurable in an oxidation-reduction system if
the same growth conditions are maintained. The measure­
ment of electrode potentials of culture media under aero­
bic conditions requires the employment of a system in
which the E^ of the cell will remain at or near the same
level for a period before inoculation. The maintenance of
electrode potentials at a constant level in liquids con­
taining organic materials has been a difficult problem
without the introduction of some dynamic method to pre­
vent the drift of potential and polarization of the elec­
trode, These troublesome factors may be overcome by deaerating the liquid in the electrode vessel and measuring
the E.1T.F, changes with a vacuum-tube potentiometer or anelectrometer.
By making use of an electrode made of Acheson special
graphite.as in this study, the troublesome drift of poten­
tial and the polarizing tendency so characteristic of
metal electrodes was avoided. Further, one does not en­
counter marked potential differences in different graphite
electrodes that are commonly found when different platinum
or gold electrodes are used in the same cell.
Another important advantage of using the graphite
electrode for measuring oxidation-reduction potentials
is, that a large number of cells may be studied simultane­
ously with a high degree of accuracy.
The experimental procedure and results of this study
follows
Experimental Procedure
Apparatus. A thermostat was used to maintain the

12

V'V'ai

/- L/veh infusion medium.

Aaar* hrM#* UN.KCt).
3- /M KCI.
4 ~ fffzClz puste.

S-'Mer c u>~y.
6~-Gir*phiie e/eettode

7- InecaiGt-jng tube,

8~*$t opcoct\«

Figure A - Schematic Arrangement of Complete Cell

13
cells at a constant temperature* An insulated w o o d e n b o x
lined with galvanized sheet iron was used as a container.
The thexmoregulator was of the mercury in glass expansion
type and was made in our laboratory. A Bernell polarized
telegraphic relay was used in conjunction with the thermo­
regulator for controlling the current in a 250 watt elec­
tric heater. The stirring device was constructed from an
ordinary laboratory stirring pulley and was driven by a
small electric motor. The temperature was maintained at
37 C + .01 C.
A Leeds and Forthrup student type potentiometer and
accessories were used throughout these experiments. A
high grade Leeds and Forthrup pointer galvanometer was
found satisfactory for this work.
The potentiometer assembly and working battery were
placed on sheets of metal and all were connected to the
metal case of the thermostat, thus forming a shield
against stray currents (2). Care was taken to see that each
post was well insulated from the shield. The shielding
helps to prevent drifting of the potential, due to stray
currents, and to stabilize the galvanometer during damp or
stormy weather.
The cell used in these experiments is shown in Figure
A. A calomel cell with a long side tube leading into the
intermediate vessel was used as a reference electrode for
ten oxidation-reduction half cells. The intermediate ves­
sel was a small jar fitted with a cork stopper containing
ten holes for receiving the agar bridges from the oxida­
tion-reduction half cells. A center hole received the cal­
omel side tube* The oxidation-reduction half cells were
pyrex test tubes 25 mm, by 100 mm. fitted with rubber
stoppers containing holes for the electrode, agar bridge,
and an inoculation tube. The' inoculation tube was a short
piece of small bore glass tubing passing through the rub­
ber stopper and projecting above the stopper. It was cov­
ered with a short piece of larger bore glass tubing
plugged at one end with cotton.
Inoculating pipettes were made by drawing out pieces
of soft glass tubing 5 mm, in diameter to a capillary end
of a sufficient diameter to pass through the inoculation
tube in the cell*. The large end of the pipettes was
plugged with cotton. They were sterilized in a dry oven.
Materials. All the chemicals used in filling the
calomel half cell were prepared from "Baker*s Analyzed"
chemicals. The mercury was cleaned by running it through
a fine capillary several times. It was dried in vacuum
over phosphorus pent oxide. The calomel was prepared as
described by Lewis, Brighton, and Sebastian (3j without
further purification of the chemicals. 1M potassium chlor­
ide solution was made by weighing the potassium chloride
and dissolving it in the required amount of distilled
wat e r .
The agar bridges contained two per cent washed agar

14
in 3JT potassium chloride solution# ^
The electrode used in an oxidation-reduction hall
cell is of great importance. After repeated trials with
many metal electrodes, it was found that they could no
e
used in liver infusion broth under aerobic conditions
the measurements were made with a potentiometer system# T e
current required by the galvanometer can easily polarize
poorly poised electrodes and make
measurements un­
certain. Many investigators studying oxidation-reduction po
tentials in bacteriological culture media have overcome
this difficulty in a limited way by using some form of an
electrometer or vacuum tube amplifier in place of the gal­
vanometer. Oftentimes a special form of vacuum tube volta­
meter is used. The electrometer, vacuum tube amplifier, and
vacuum tube voltmeter require elaborate shielding, skill
in making the measurements and oftentimes the control of
the humidity of the air if accurate readings are to be ob­
tained for each individual potential reading* klexner and
Barron (4) make the following statement in their conclu­
sions on carbon and tungsten electrodes when used as oxida­
tion-reduction electrodes, "Graphite gives the same results
as platinum in a dyestuff system, be it well or poorly
poised, whereas tungsten gives reliable results only when
the system is well poised." Flexner and Barron recommended
the use of lead pencil graphite for electrodes. While this
type of electrode was found more satisfactory than metal
ones in liver infusion medium, it was inferior to those
made of Acheson special graphite. Large electrodes 1/4 inch
by 4 inches of this material gave surprisingly good re­
sults, This type of electrode does not become polarized
easily and when used in the potentiometer system described
herein, each individual reading can be made with an accur-.
acy of .3 or .4 of a millivolt.
The Oxidation-Reduction Half Cell; The pyrex glass
containers are thoroughly cleansed, rinsed in distilled wa­
ter and sterilized in the dry oven, hew electrodes are
cleaned with 00 steel wool and polished with a clean cotton
cloth. Fext, they are rinsed in distilled water and insert­
ed into rubber stoppers. One should avoid touching that
part of the surface of the electrode which projects below
the bottom of the rubber stopper after it is cleaned and
rinsed in distilled water. The glass cell is fitted with
the rubber stopper containing the electrode and glass tube
through which the medium is inoculated, covered with paper
and sterilized in the autoclave at 15 pounds pressure for
30 minutes. The sterile cells are stored until ready for
use,
The agar bridges are made by filling 2 mm. diameter
soft glass tubes bent in a IT shape to within one inch of
the openings with two per cent agar containing m KC1 Thev
are heated in flowing steam for 20 minutes and cooled’ The
ends of the glass tubing are sealed off, heated again’in
flowing steam for 20 minutes to test the seals for leakage.

15
A mark is made with a file just below the agar level so
that the sealed ends may be broken off easily. They are
wrapped in paper and sterilized in the autoclave.
A half-cell is assembled for study by using every
possible aseptic precaution. First, 25 cc. of sterile liv­
er infusion broth pH 6.6 are added to the- sterile glass
cell by removing the stopper. Next, the end of one arm of
the agar bridge is broken off and inserted through a hole
in the stopper until it is well below the surface of the
broth. The other sealed end of the bridge is not broken
off until the cell is placed in the water bath and connect­
ed to the calomel intermediate cell. The assembled half­
cell is now placed in an incubator at 37°C for four to
seven days to permit the electrode potential to become sta­
ble. It is now placed in the constant temperature water
bath and connected to the calomel intermediate cell by
means of the agar bridge. After a period of 24 hours the
initial potential of the half-cell is taken and recorded-.
Inoculation of the Malf-cell. The inoculum is prepared by suspending the growth of a 48 hour agar slant in
sterile physiological salt solution a few minutes before
use. Approximately 100,000 living cells are introduced
into the medium by means of a sterile capillary pipette
through the glass inoculating tube in the rubber stopper.
Preparation of Solutions of D y e s . The dyes used in
this study were thionin (certified) and basic fuchsin
(certified) manufactured by the national Aniline and Chem­
ical Co. They were made up in one per cent suspensions in
sterile distilled water. The suspensions are heated in
flowing steam for 20 minutes before use to obtain uniform
suspension of the dyes. They must be heated each time they
are used as the dyes do not remain in suspension in this
concentration. The hot dye suspensions are added to the
broth just' before the inoculum is added. The amount of
thionin solution that was found satisfactory was 0.2 cc.,
of basic fuchsin 0.3 cc.
Procedure of Cleaning Electrode at th e Completion of
an Experiment. The cells were ^aken^apart and the elec­
trode placed in a container of distilled water. The con­
tainer was then placed in flowing steam for 24 hours. The
water was then poured off, fresh distilled water added/
and the container placed in flowing steam again. The water
is changed about every two hours until no more of the dye
is present. The electrode is cleaned with 00 steel wool
and polished with a clean cotton cloth. It is placed in a
clean glass tube or beaker and stored until ready for use.
Method of Standardizing the Calomel Kalf-cell and
Calculating
values. Since the chemicals used in i’he
calomel half-cell were not repurified and the concentra­
tions in the agar bridge vary slightly, it was necessary
to standardize the calomel half-cell. This can be accom­
plished by measuring the following cell and applying the
proper corrections for temperature and pressure:

16
(1) (-) Pt, Ha (1 atm.), W HC1; 1ST KC1 2% agar
bridge; 1STKC1, Hg.Cla,Hg (+ ). at 37 C.
The E-.K.E. of the following cell was measurea:
(2) C, liver infusion medium; Iff KG1 2% agar bridge;
1M KC1, Hg«Ola, Hg at 37°C. Our results are reported in
terms of E*^ values which are the same as though the 1o±lowing celT had been measured:
.
*
(3) (+) C, liver infusion medium; Iff HC1, Hgll atm.j,
Pt (-} at 37°C•
If we place cell (l) in series and in opposition to
cell
(2 )we have cell (3).
.
(4 ) (+ ) C, liver
infusion medium; Iff KC1 2%
agar
bridge; Iff KOI, Hg^Cl *, Hg - Hg, Hg^Cla , Iff K C 1 5 Iff KC1
2% agar bridge; Iff HCl,
(l atm.), Pt (-) at 37 C.
In order to calculate the values of the cells meas­
ured in terms of Eh cell (3) we simply add the values of
cell
(1) and cell (2)algebraically. The sign on the elec­
trode in these experiments has
always been taken as thaton
the potentiometer. This is a convenient convention to fol­
low since a + E^ value represents a degree of oxidation.
Results
The Eh values ofcells containing sterile liver infusion broth of a pH of 6.6 without and with dyes are set
forth in table I on the following page. The time potential
curves of the broth without and with dyes are illustrated
in figure 1. The E^ values and curves obtained on these
cells are examples taken from determinations made on a
large number of cells with different batches of the medium.
It is interesting to note that the addition of the dyes
to the medium in the cells causes no appreciable differ­
ences between the E^ readings of the cells.
In table II is set forth a list of the
values of
cells containing medium, without and with dyes, after inoc­
ulating one set with Br, melitensis strain 344, and another
set with strain 608. Both strains are of human origin. The
E^ values of cells inoculated with strain 344 plotted against time are compared in figure 2. There is very little
difference between the speed and degree of reduction of
the medium in each of the three cells as a.result of the
growth of the organism.
The time Eh values obtained after inoculating liver
infusion broth without and with dyes with Br. abortus
strain 109 and strain 93 are present in taFIe TTl. The
time potential curves for strain 109 are shown in figure 3.
Strain 109 is of human origin and strain 93 is of bovine
origin.
There is only a slight difference in the direction of
the curve and final E^ value of the inoculated medium with­
out dye and the one containing basic fuchsin. The time
potential curve of the inoculated medium containing thionin
takes an entirely different direction from that of the oth­
er two curves. The change in the E h value of the medium

17
containing thionin would indicate that a slight reduction
of the medium was in progress during the period of incu­
bation.
Two sets of time E^ values of the medium inoculated
with two strains of Br, suis are presented in table IV.
Figure 4 shows the ETme potential curves of the medium
without and with dyes after inoculating with strain 533,
Strain 533 is of porcine origin a.nd strain 570 is of
human origin. The successive E^ values in the table and
position of the curve plainly show that the organism
causes little if any reduction in the medium in the
presence of basic fuchsin. The reduction of the medium
in the presence of thionin progresses just as rapidly af­
ter inoculation with Br. suis as does the medium without
the d y e .

Table I, - Time E^ of Sterile Liver Infusion Broth
with and without dyes .
Broth without
dye
Time
Hours

Eh

Broth with
Thionin
Time
Hours

Eh

Broth with
B. Fuchsin
Time
Hours

Eh

0

0.3225

0

0.3179

0

♦ 0.3156

20

♦ 0,3189

30

+ 0.3117

20

0.3165

54

♦ 0.3162

46

0.3131

54

+ 0.3224

68

+ 0.3128

70

0,3137

68

+ 0.3275

92

+ 0,3098

94

0.3136

92

+ 0.3296

116

+ 0.3090

118

+ 0.3165

116

■f 0.3248

140

+ 0.3102

142

t 0.3162

140

0.3186

164

♦ 0.3108

166

+ 0.3166

164

+ 0.3176

188

+ 0.3116

188

0.3229

18
Table II. - Time Eh of Liver Infusion Broth ^jjoculated
with Br. melitensis Strain 344 5 Br. melitensis Strain
Inoculated wit strain 344
Broth with
Broth without
Broth wi
B. Puchsin
dye
Thionin
Time
Time
Bh
Time
Eh
Eh
Hours
Hours
Hours
+ 0.3026
0
0
t 0.3073
4 0.3153
0
24

+ 0.2530

22

0.2459

24

♦ 0.2979

48

+ 0.2062

46

♦ 0.2139

48

4 0.2559

72

f 0.1772

70

f 0.1862

74

♦ 0.2261

96

4- 0.1667

94

f 0.1692

96

+ 0.2085

120

♦ 0.1596

118

* 0.1599

120

4 0.1890

156

f 0.1537

154

t 0.1566

146

♦ 0.1761

168

+ 0.1400

170

4 0.1651

192

4- 0.1335

192

4 0•1538

216

t 0.1330

216

4 0.1397

Inoculated with strain 608
0

+ 0.2778

0

t 0.2798

0

4 0.2691

24

t 0.2431

24

■f 0.2305

24

4 0.2628

48

♦ 0.2014

48

0.2173

48

4 0.2445

72

+ 0.1817

72

+ 0.2052

72

f 0.2168

96

♦ 0.1752

96

♦ 0.1951

96

4 0.2034

120

+ 0.1751

120

+ 0.1942

120

4

156

* 0.1617

156

* 0.1920

156

4 0.1600

168

t 0.1511

168

4 0.1816

168

4 0.1602

192

♦ 0.1522

192

t 0.1746

192

4 0.1666

216

t 0.1556

216

+ 0.1624

216

4- 0.1489

0.1863

19
Table III. - Time Eh of Liver Infusion Broth Inoculated
with Br. abortus Strain 109; Br, abortus Strain 93
Inoculated with Strain 109
Broth without
dye
Fime
Tours

Broth with
Thionin
Time
Kours

n

*®h

Broth with
B* Fuchsin
Time
Hours

Eh

0

4 0.3130

0

0.2957

0

4 0.3218

20

4 0.2634

24

0,2897

20

4 0.3164

54

4 0,1995

48

0.2758

54

4 0.2531

92

4 0.1731

74

'0.2568

68

4 0.2248

116

4 0.1648

96

0.2520

92

4 0,2006

140

+ 0.1596

120

0.2423

116

4 0.1855

164

4 0.1541

146

0.2225

140

+ 0.1747

188

4 0.1509

170

0.2291

164

4 0.1676

192

0.2199

188

t 0.1638

216

0 .2 2 0 1

Inoculated with Strain 93
0

4 0.3099

0

0.2893

0

4 0.2810

30

4 0.2349

30

0.2696

24

4 0.2721

46

4 0.2305

46

0,2732

48

4 0.2277

70

f 0.2048

70

0.2703

72

4 0.2067

94

4 0.1833

94

0.2635

96

4 0.1923

118

+ 0,1694

118

0.2570

120

4 0.1661

142

4 0.1558

142

0.2532

170

4 0.1518

166

4. 0.1438

166

0.2488

192

4 0,1534

20
Table XV. - Time

of liver Infusion Broth Inoculated

with B r . suls Strain 533;.Br. suis Strain 570
Broth without
dye
Time
Eh
Hours

Inoculated with Strain 533
Broth with
Broth with
Thionin
B, Fuchs in
Time
Time
Eh
Eh
Hours
Hours

0

0 o3132

0

4 .0.2497

0

4 0.2715

.24

0.2632

24,

4 0.1481

24

4 0.2585

48

0.2035

48

4 0.1285

48

4 0.2482

72

0.1662

72

+ 0.1207

72

4 0.2500

96

0.1465

96

4 0^1084

96

+ 0.2514

120

0.1264

120

4 0.0897

120

4 0.2517

146

0.1320

146

4 0.0982

170

4 0.2437

168

0.1215

168

+ 0.0897

192

4 0,2510

192

0.1148

192

4 0.0866

216

0.0939

216

4 0.0857

Inoculated with Strain 570
0

0.3300

0

f 0.2940

0

4 0.3076

30

0.2509

30

4 0.1864

24

4 0.2946

46

0.2345

46

4 0.2062

48

4 0.2794

70

0 .2 1 2 0

70

4 0.1768

74

4 0,2694

94

0.1975

94

4

0.1539

96

4 0.2699

118

0.1795

118

4 0.1525

120

4 0,2698

142

0.1719

142

4 0.1362

146

4 0.2689

166

0 *1 6 1 6

166

4

0.1340

170

4 0.2704

192

4 0,2722

216

4 0.2748

21

..—

&

Fig. 1. - Time potential curves of sterile liver infusion
broth; ■*-*- without dye, -SMt- with thionin, -e— e-witli basic
fuchsin.

22

2+

4a

72

36

12,0

/44

Time in hours,
Pig, 2. - Comparison of time potential curves of liver
infusion broth inoculated with Br, melitensis strain
344;
without dye,
witJTthionin, -e— e— with
basic fuchsin.

23

4

«

t

/

O

24

43

72

96

/SO

/4 4

t&S

/$ 2

240

Time in hours.
Fig* 3* - Comparison of time potential curves of liver
infusion "broth inoculated with Br. abortus strain 109;
without dye 9 -x--x - with thionin,
-Q~ with basic
fuchsin*

24

2.+

48

72

86
/SO M
Time in hours<

Fig. - Comparison of time potential curves of liver
infusion broth inoculated with Br. suis strain 533;
without dye,
with thionin, -o- -o with basic
fuchsin*

25

24

48

72

96
tzo }4
Time in hours.

Fig,5. - Comparison of time potential curves of liver
infusion broth without dye; -++. inoculated with Br,
abortus strain 109, X—x inoculated with Br. suis strain
533,
inoculated with Br. melitensis strain"S44.

26

4

3

. <•

^.2

9

/

0

£4

46

/£

$e> /ao 14 +
Time ih heats*

i£&

!$£.

QlI

34-0

Fig. 6. - Comparsion of time potential curves of liver
infusion "broth containing thionin; ■■«■■-» inoculated with
Br* abortus strain 109,*rjh inoculated with Br* suia
strain 533, -o ...q - inoculated with Br. melitensis strain 344*

21

Fig. 7. - Comparison of time potential curves of liver
infusion broth containing basic fuchsin; '■t—+ ■inoculated
with Br. abortus strain 109,
inoculated with Br.
suis strain 533,
inoculated with Br. melitensis
strain 344.

28
A comparison of the reducing capacity of the three spe­
cies of Brucella in liver infusion "broth without and with
the dyes is illustrated more clearly in figures 5, 6, and
7. A comparison of the time potential curves resulting^
from the growth of each of the three species in liver in­
fusion broth is seen in figure 5. There is little variation
between the direction of the three curves or individual
values. The final E, value for each of the three spe­
cies varies to only a slight degree. The addition of thionin to the medium considerably retards the growth of Br*
abortus and as a consequence checks the reduction of the
medium as illustrated in figure 6* While Br, melitensis
does not reduce the medium in the presence of this dye as
rapidly or to the same degree as does Br. suds, there is
on the other hand, a noticeable difference in its reduc­
ing capacity over that of Br, abortus.
The differences in the reducing"”capacities of the
three species in the presence of basic fuchsin is well il­
lustrated in figure 7. The reducing capacity of Br. suis
is held almost completely in check while that of the other
two species is not affected by the dye.
Discussion
The foregoing data indicate that sticks of special
graphite (Acheson) when properly prepared are suitable as
electrodes for measuring the oxidation-reduction intensity
of bacteria when grown in a sterile liquid culture medium
maintained under aerobic conditions. The potential of
sterile beef liver broth after a preliminary incubation
period of 4 to 7 days, lies between - 0.31 and - 0.34
volt.
The time potential curves of the medium during the
growth of the three species of Bruce 11a do not vary to a
great degree. During the first 48 hour growing period
there is a, decline in potential to + 0,2 volt. At this
point the curves separate and as the time of incubation
increases the separation becomes more ma.rked, Br. suis
grows more rapidly than the other two species of Brucella
and as the curve indicates, has much greater reducing ca­
pacity than the other two species. The rate of reduction
due to the growth of Brucella proceeds at a much slower
pace than that which has been observed for many bacteria
when measured in oxidation-reduction systems under aerobic
conditi ons,
Tn view of the fact that there has not been found a
marked variation in the reducing capacity of the three
species of B ruce 11a after seven days of incubation in liver
infusion broth, one would presume that there is not a
marked difference in the metabolism of the organisms in
the same medium.
The differences in metabolism of the organisms become
apparent when bacteriostatic dyes are added to the liquid
medium. The addition of a suitable amount of thionin to the

29
medium retards or prevents its reduction when inoculated
with Br, abortus . The presence of basic fuchsin holds the
reduction of the medium in check when inoculated with Br.
suis, Neither dye, to any extent, prevents the medium from
Toeing reduced by the growth’"of Br. m e litensis.
Each of the cells were examined at the end of the
growth period for viable organisms, but no attempt was
made to determine the number on a quantitative basis. It
was rare to find Br. abortus alive at the end of the in­
cubation growth period in the cells to which thionin had
been added. Br. suis was always recovered from the cells
containing basic fuchsin although the potentials of the
cells indicated the absence of growth activity.
In those cells in which growth occurred, the pH shift­
ed from 6.6 to 7. There would be very little difference in
the time potential curves were they corrected for the change
in pH of the medium which occurred during the growth of
the bacteria.
The differences recorded in the metabolism of the spe­
cies of Brucella in a liquid medium in the presence of
bacteriostatic dyes by means of oxidation-reduction poten­
tial measurements, serve to confirm that which is already
known of the differences in their growth behavior on a sol­
id medium in the presence of the same dyes (l).
Summary
1, Electrodes made of graphite (Acheson special) are
suitable for measuring the oxidation-reduction reactions
of aerobic bacteria in liquid media under aerobic condi­
tions. ‘
2. The reduction potentials of Brucella in beef liv­
er infusion broth under aerobic conditions show a negative
drift that attains the value: E^ + 0.15 to 0.1 volt at
the end of an incubation period of eight days, Br. suis
shows a slightly more negative drift than the other two
species.
3*
abortus in the presence of thionin is unable
to reduce the potential of the medium,
4* Br* suis in the presence of basic fuchsin is un­
able to reduce the potential of the medium.
5. The presence of neither thionin or basic fuchsin
retards the negative potential drift of the medium caused
by the growth of Br. melitensis.

30
References
1* Huddleson, I. Forest
The differentiation of the species of the Genus
Brucella.
Technical Bulletin No. 100, Michigan State College,
East Lansing, Michigan.
2» White, Walter P.
Leakage prevention by shielding especially in poten­
tiometer systems.
Journal of American Chemical Society, p. 2011,
vol. 36, 1914.
3. Lewis, Gilbert F . , Thomas Brighton and Reuben L. S e - ■
bastian. A study of hydrogen and calomel electrodes.**
Journal of American Chemical Society, p. 2245, vol.39,
1917.
4. Flexner, Louis B,, and E. S. Guzman Barron
Oxidation-reduction potentials at carbon and tungsten
electrodes.
• ‘
■■
**
Journal of American Chemical Society, p. 2773, vol.52,
1930.

Ill - The Preparation and Use of the PlatinizedGraphite Electrode
In a search for a more suitable electrode for meas­
uring the oxidation-reduction potential of liquid culture
media for the growth of bacteria, considerable time was
devoted to the study of the behavior of graphite electrodes.
In the course of this work it was decided to determine the
suitability of this type of electrode, platinized as a re­
versible electrode. In addition to this study a more exten­
sive comparative study was formulated of platinized plati­
num, platinized graphite, and graphite as electrodes for
electrometric titrations (l).
Apparatus . The apparatus that was employed in connec­
tion with the work on hydrogen reversible cells was of
precision type and quality. A water thermostat'was used
that maintained the temperature within 0.005 C. of the
working temperature, 25 C.
A type K 9 Leeds and Northrup potentiometer was used.
This instrument and the galvanometer were very carefully
shielded according to the method described by White (2).
The metal tank of the thermostat and the copper shields
which surrounded the cells were connected to the main
shield. All alternating current circuits were removed as
far as possible from the immediate vicinity of the poten­
tiometer system. The galvanometer was a Leeds and Forthrup
precision type. The distance between it and the reading
telescope and scale was about 5 ft. The sensitivity of the
potentiometer system was such that one millionth of a volt
difference in potential would cause a deflection in the
galvanometer system of 1 mm. This sensitivity was obtained
on the 16 millivolt scale of the potentiometer and in read­
ing the cell of highest resistance.
The cell finally used in these experiments was of very
simple design. A special flat bottom freezing point tube
which was closed with a number 8 rubber stopper was found
satisfactory, holes were made in the stopper for gas inlet
and exit tubes and also for four electrodes. The four
electrodes, two platinized platinum and two platinized
graphite, were conveniently spaced around the gas inlet
tube which was placed in the center of the rubber stopper.
A hole was also provided for the exit tube which led into
a trap.
The hydrogen was produced by electrolyzing a 10% so­
lution of FaOF in a generator similar to that described by
Clark (3). The hydrogen was purified by passing it over an
electrically heated platinum spiral. A piece of heavy cap­
illary glass tubing about 1 meter long was placed between
the generator and the hot platinum wire as a safety device.
The hydrogen was brought into the cells from the generator
and purifier through glass tubing coupled together with
short rubber connections.

32
The apparatus used in connection with the electro­
metric titrations was that commonly employed for this Kina
of work. It consisted of a heeds and Uorthrup student^po­
tentiometer, motor driven stirrer, and a burette. A sil­
ver, silver chloride electrode in 1 molar potassium chlor­
ide solution served as the reference electrode*
Materials. The hydrochloric acid solutions were made
from Baker’s chemical pure acid as described by Foulk and
Hollingworth (4). The solutions were IF, 0.1F, and 0.01F
in hydrogen chloride. The sodium hydroxide solutions were
made up in the usual way and titrated with the above acids
using phenolphthalein as the endpoint indicator. The plat­
inum electrodes were made of platinum foil# The graphite
electrodes were Acheson special graphite rods 1/4 inch in
diameter and 6 inches long. This grade of graphite is
free from interfering impurities and is more dense than
other grades.
The electrodes were platinized by electrolyzing in a
7>% chloroplatinic acid solution containing a trace of
lead acetate (about 0.02^), This solution was prepared as
follows: Scraps or pieces of platinum were dissolved in
concentrated chemically pure hydrochloric acid by vigor­
ously electrolyzing the platinum as the anode. Once the
platinum begins to dissolve it will continue slowly with­
out further electrelyzing if kept tightly stoppered. The
platinum black on platinized electrodes was rapidly re­
moved by electrolyzing in concentrated hydrochloric acid.
In one case the platinum was dissolved in the usual way
in aqua regia. The solution of platinum in concentrated
hydrochloric acid was placed in a casserole and boiled
down to a syrupy consistency over a Bunsen burner. From
time to time small amounts of concentrated hydrochloric
acid were added to prevent precipitation. If precipita­
tion did occur during this process, sufficient concen­
trated hydrochloric acid was added to the hot solution to
redissolve the precipitate. The rich orange solution was
reduced to dryness over a steam bath. The resulting mass
had a very deep orange color. The residue was taken up in
the required amount of distilled water containing a trace
of lead acetate and brought near the boiling point over a
flame. The precipitate thus formed was redissolved by
adding concentrated hydrochloric acid ( a drop at a time
from a capillary pipette) to the solution until nearly all
the platinum was back in solution. The addition of acid
was stopped just before the lead precipitated as the
chloride. The cool solution was filtered, decanting as
much as possible. The small amount of precipitate which
was not redissolved was dissolved in concentrated hydro­
chloric acid and used as an electrolytic strip for plati­
nized electrodes. Cleanliness was found to be of great
importance throughout the above procedure. At first de­
posits from this solution were of a d.ark gray color in­
stead of black. However, it was found that such deposits

33
were as satisfactory as the black*
Procedure; The platinum black was removed from the
electrodes by electrolyzing in chemically pure hydrochlor­
ic acid. The platinum electrodes were heated to a white
heat in the reducing cone of a Bunsen flame preparatory
to platinizing. The platinum electrodes were platinized by
making them the cathode and electrolyzing for 3 or 4 min­
utes in the platinizing solution. The graphite electrodes
were heated to a dull red and immediately placed in cold
distilled water. They were further cleaned with a fine
grade of steel wool and polished with a new cheese cloth.
They were placed in distilled water and slowly boiled
from four to six hours. At the end of this time they were
allowed to cool under distilled water, wrapped in clean
paper, and dried over night at 110°C. The graphite elec­
trodes were platinized from 5 to 10 minutes. The anode
was a spiral of platinum wire around the graphite elec­
trode and one-half inch distance from its surface. The
current passed was sufficient to produce a vigorous evo­
lution of hydrogen. Since platinized electrodes should
never be allowed to dry before using, they were immedi­
ately rinsed with and placed under distilled water. The
electrodes and gas inlet tube were very carefully placed
in the stopper used in closing the top of the cell.. The
electrodes were rinsed several times with the solution to
be used in the cell. The cell was filled to the proper
level and the stopper carrying the electrodes and inlet
tube was carefully fitted to the cell which was placed in
its shield in the thermostat. The trap assembly was placed
on the gas exit of the cell and the flow of hydrogen
started. Pure hydrogen was continuously passed during the
life of the cell. No measurements were made until the gas
had passed for 6 hours. Daily readings were made on all
the combinations between the four electrodes (two plati­
nized platinum and two platinized graphite) until check
readings were obtained.
The electrometric titrations were conducted in the
usual manner. No special precautions were taken outside
of that ordinarily employed in such work. The cell was
open to atmospheric air. A 20 ml, sample of the reagent
to be titrated was pipetted into a 250 ml. beaker and di­
luted to 50 ml. for all titrations reported.
Results of Study
The tabulated results of the experiments on the
hydrogen reversible electrode are given in Tables I, II,
III, IV, V, and VI which follow. The readings from one
cell for four days are given for each concentration of
hydrochloric acid and sodium hydroxide studied..
Theoretically, the potential difference between the
electrodes in all cases represented in Tables I, II, III,
IV, V, and VI is exactly zero. That this is true comes
from the following consideration. The electrodes of these

34
cells are hydrogen gas electrodes* The reversible^reaction
at the surface of a hydrogen electrode is
2e.
The finely divided platinum, covering the surface of the
electrode, adsorbs the gaseous hydrogen and acts^as a cat­
alyst for this reaction,, The only differences which re­
main between platinized platinum and platinized graphite,
when used as a gas electrode, are the differences in the
adsorptive properties of platinum and graphite* Tables I,
IX, rxr, V, and VI show that very little adsorption of
either a positive or negative ion had taken place because
all the values are very close to the theoretical value.
Table IV shows that in all probability some adsorption of
the sodium ion on the graphite had taken place because
there is a tendency for the platinized graphite to become
more positive than platinized platinum. This adsorption
was aided by the removal of minute particles of platinum
black due to the soapy action generated by the bubbling of
the gas through the alkali*
Key to symbols used in table. Pt-Pt difference in
volts between the two platinized platinum electrodes*
Ptp-CPtjj difference in volts between the right plat­
inized platinum electrode and the right platinized graph­
ite electrode.
Pt^-CPt^ difference in volts between the right plat­
inized platinum electrode and the left platinized graphite
electrode,
CPt-CPt difference in volts between the two plat­
inized graphite electrodes.
Pt^--CFtpj difference in volts between the left plati­
nized platinum electrode and the right platinized graph­
ite electrode.
Ptr-CFtj^ difference in volts between the left plati­
nized platinum electrode and the left platinized graphite
electrode.
R -*' right electrode connected to the positive terminal
of the potentiometer,
L* left electrode connected to the positive terminal
of the potentiometer.
C+ platinized graphite electrode connected to the
positive terminal of the potentiometer.
C- platinized graphite electrode connected to the
negative terminal of the potentiometer.
The results of the electrometric titrations are shown
graphically in Figures 1 to 6 inclusive which follow.
These curves are composites of triplicate electrometric
titrations. A silver, silver chloride electrode in one mo­
lar potassium chloride solution served as the reference
electrode. The same standardized reagents were used in
making all the electrometric titrations represented in
Figures 1, 2, and 3. The experiments represented in Fig­
ures 4, 5, and 6 were conducted in a similar manner.

35
Table I
Hydrogen Cell of IF HC1Hours in
Thermostat

Pt-Pt
Volts

PtR -CPtR

PtR -CPti

*8

JU 0*000002

volts
C+ 0.000015

96

0.000000

C* 0.000021

C + 0.000004

120

0.000000

C+ 0.000030

C+ 0.000004

144

0.000000

C+ 0.000039

G+ 0.000006

CPt - CPt

PtL - CPtR

volts
0+ 0.000006

« r

CPti

volts
C + 0.000019

volts
C- 0.000007

96

HC + 0.000021

c+ 0.000024

C + 0.000004

120

RCt 0.000034

c+ 0.000030

Cf

144

RC + 0.000032

c+ 0.000039

c+ 0.000006

•

O
o
o
o
o
o

72

volts
KC + 0 <,000021

Table II
Hydrogen Cell of 0.1F HC1
Hours in
Thermostat

Pt-Pt

PtR -CPtR

PtR -CPtt

24

volts
It* 0.000010

volts
c- 0.000010

volts
c+ 0.000030

48

L+ 0.000004

0.000000

c+ 0.000004

72

I>+ 0.000003

c+ 0.000002

c+ 0.000005

96

L+ 0.000004

0.000000

c+ 0.000004

CPt - CPt

PtL - CPtR

PtL - CPtL

24

volts
LC+ 0.000039

volts
C- 0.000020

volts
C+ 0.000018

48

LC+ 0.000004

C- 0.000004

0.000000

72

I.C+ 0.000002

0.000000

C+ 0.000002

96

LC+ 0.000004

C- 0.000004

0.000000

36
Table III
Hydrogen cell of 0.01P MCI
Hours in
Thermostat

Pt -Pt

Pt^-CPtpj

PtR-CPtj^

24

volts
0.000000

volts
C- 0.000002

volts
C+ 0.000002

48

R+ 0.000001

C+ 0.000004

C* 0.000004

72

0.000000

C+ 0.000004

C+ 0.000004

96

0.000000

C+ 0.000004

C+ 0.000004

CPt - CPt

PtL-CPtR

PtL-CPtL

24

volts
LC+ 0.000004

volts
C- 0.000002

volts
C+ 0.000002

48

0.000000

Cf 0.000005

C+ 0.000005

72

0.000000

C* 0.000004

C+ 0.000004

96

0.000000

C+ 0.000004

C* 0.000004

Table IV
Hydrogen cell of 1.004P MaOH
Hours in
Thermostat

Pt - Pt

24

volts
R* 0,000002

volts
C- 0.000005

Pt -CPtT
R
volts
C- 0.000015

48

R+ 0.000003

C+ 0.000017

04 0.000073

72

R+ 0.000005

C + 0.000040

C ■+ 0.000121

96

R4 0.000005

c+ 0.000054

c+ 0.000174

CPt - CPt

Pt^-CPt^

Pt’I-CPtK

Pt,£-CPt-jr

24

volts
RC + 0.000010

volts
C- 0.000007

volts
C- 0.000017

48

LC+ 0.000052

c* 0.000020

c+ 0.000078

72

LC + 0.000081

c+. 0.000045

C4 0.000126

96

I.C4 0.000120

C4 0.000051

CU 0.000170

37
Table V
Hydrogen cell of 0.0998F HaOH
Hours in
Thermostat

Pt -Pt

PtR -CPtR

PtR -CPtL

24

volts
0.000000

volts
C* 0.000035

volts
C+ 0.000102

48

0.0QQ000

C* 0.000021

C4 0.000045

72

0.000000

0.000000

C+ 0.000005

96

0.000000

C+ 0.000010

C4 0.000010

CPt - CPt

PtT~CPt„
L
R

Pt^-CPtj,

24

volts
L C + 0.000066

volts
C4 0.000035

volts
C4 0.000102

48

LC +

0.000024

C4 0.000021

C4 0.000045

72

XC+ 0.000005

0.000000

C+ 0.000005

96

0.000000

C4 0.000010

C4- 0.000010

Table VI
Hydrogen cell of 0.00992F HaOH
Hours in
Thermostat

Pt - Pt

PtR -CPtR

PtR -CPtL

24

volts
R* 0.000004

volts
C*f 0.000094

volts
C- 0.000036

48

0.000000

C- 0.000006

C- 0.000009

72

0.000000

C- 0.000010

C- 0.000005

96

0.000000

C- 0.000020

C- 0.000010

CPt - CPt

PI^-CPtR

Ptj.-CPtj.

24

volts
RCt 0.000130

volts
C4 0.000100

volts
C- 0.000030

48

RC4 0.000003

C- 0.000006

0.000009

72

LC+ 0.000005

C- 0.000010

C- 0.000005

96

LC4 0.000010

C- 0.000020

C- 0.000010

38

*

8

1
€

,4-

4

I

/r
le
CX. Of O j f H C h

2a

Z3

Pig* 1. - A composite curve of triplicate electroraetric
titrations on 20 cc. of NaOH with O.lff HC1, using a
platinized platinum electrode.

39

■y*>--~-

7

Vo/ts.

5

3
2

*6

tS

20

Zt

23

24

25

CX. Of OJf HCf0
Fig.. 2* - A composite curve of triplicate electrometric
titrations on 20 cc. of IfraOH with '0.1JT HC1, using a
platinized graphite electrode#

is

is

tr
ib
i9
zo
C.C. Of o .l f HCI.

a/

zz

&

£ 4.

&s

-

3. - A composite curve of triplicate electrometsic
tit rati oris on. 20 cc. of HaOH with O . M B©1,. using a : .
graphite electrode.'
' .
,

41

1.0

3
8
»

,6

s
8

10
C. C.

U

iZ

of

ft M ti O4

18

/4

IS

le

17

*8

FIg. 4 # - A composite curve of triplicate electrometric
titrations on acidified 0.11T FeS04 {HH4 ) 2 S04 with
KMh045 using a platinized platinum electrode.

^ .7

iO
CX. Of K M 11O 4,
Pig. 5. - A composite curve of triplicate electrometric
titrations on acidified O.ltf FeSO^KH^JSO^ with KE&1O4 ,
using a platinized graphite electrode.