U. S.

DEPARTMENT OF COMMERCE NATIONAL BUREAU OF STANDARDS

RESEARCH PAPER RP1558

Part of Journal of Research of the N.ational Bureau of Standards, Volume 31,
October 1943

THE ASSAY OF POTASSIUM p~PHENOLSULFONATE, ITS

pH RANGE, AND ITS ULTRAVIOLET ABSORPTION SPEC~
TRUM
By Elizabeth E. Sager, Marjorie R. Schooley, and S. F. Acree

ABSTRACT
Potassium p-phenolsulfonate is a good buffer for the pH range of 8.4 to 9.2
and for spectrophotometric studies of metacresolsulfonphthalein and thymolsul-
fonphthalein. Its useful pH range lies between those for borates and secondary
phosphates and therefore fills an important gap. The purified product is not
commercially available and quantitative tests for indicating its purity have not
been reported. It was found t hat a pure product may be obtained after only
three recrystallizations. A quantitative method of analysis by bromometric
titration was developed. Two molecules of bromine react quantitatively with
1 molecule of p-phenolsulfonate in molar hydrochloric acid at 0° C within 5
minutes.
Ultraviolet absorption spectra were obtained and showed differences between
the primary and the secondary salt. The spectrophotometric data indicate that
the sulfonate group is almost completely ionized in dilute solutions.

CONTENTS Page
I. Introduction ___ - - - - - - - __ _- __ ____ __ __ ____ __ __ ____________________ 197
II. Purification of potassium p-phenolsulfonate__ __ __ ___ ___ __ _ ____ _ __ __ _ 198
III. Bromometric assay of potassium p-phenolsulfonate _ __ ___ __ _ ___ __ _ __ _ 198
1. Reagents _ _ _ ___ __ ___ ___ ___ __ __ ___ ___ __ ___ ___ __ _ _ ___ __ ___ _ 199
2. Experimental procedure ___ ________________________________ 199
IV. Spectral absorption in the ultraviolet by potassium p-phenolsulfonate __ 200
1. INTRODUCTION
In spectrophotometric studies of indicators it is sometimes neces-
sary to use buffers to stabilize the solutions which cover the pH and
colorimetric transformation ranges. It is especially important that
the compounds used as buffers and the indicators do not react or
have overlapping absorption bands. The phenolsulfonates and some
of their halogenated derivatives have been suggested 1 as components
of a stable buffer system over a wide pH range.
Highly purified phenolsulfonates are not available commercially.
However, some of the compounds can be easily recrystallized from
water and hence can be purified in the laboratory. After preliminary
investigations of the potassium, sodium, calcium, and zinc salts of
p-phenolsulfonic acid and of several salts of halogenated derivatives,
potassium p-phenolsulfonate was chosen for more detailed study. The
fact that it is anhydrous makes it more desirable than the sodium salt.
The latter crystallizes with 2 molecules of water, which can be re-
moved by drying but are taken up again at room temperature at
1 S. F. Acree, R. R. Mellon, B. M. Avery, and E . A. Slagle, A stable single buffer solution pH 1 to 12.
J. Infectious Diseases 29, 7 to 10 (1921).
197
198 Jou1'nal of Research of the National Bureau of Standards

humidities as low as 40 and 50 percent. Two samples of the recrystal-

lized potassium salt lost 0.02 and 0.03 percent in weight when heated
at 110 0 C for 24 hours and regained about the same amount over a
period of 10 days, during which they were exposed from time to time in
air of 40- to 50-percent humidity. Bromometric assay, by the proce-
dure described later, showed no change in purity of the samples
heated at 110 0 C for 24 hours. The same was true of a 10-2 molar
solution which stood on the laboratory bench for 2 months. After
heating at about 167 0 C for 20 hours, however, three samples lost 0.30
to 0.36 percent in weight and were found to contain sulfate. Bromo-
metric analysis showed these samples to be 99.7 percent pure.
There apparently is no oxidation-reduction or other adverse reaction
with the sulfonphthalein indicator series, as indicated by spectropho-
tometric curves, using 10-2 molar potassium p-phenolsulfonate
buffers with metacresolsulfonphthalein and thymolsulfonphthalein.
II. PURIFICATION OF POTASSIUM p-PHENOLSULFONATE
The p-phenolsulfonate received from the manufacturer was a pink-
ish-white crystalline powder with a distinct phenolic odor. Impurities
were removed before the first crystallization by treating a hot solution
for about 10 minutes with an activated carbon. To remove all the
carbon, the solution was filtered several times while hot. Each re-
crystallization was made by adding just enough hot water (about 80°
to 90 0 C) to dissolve the crystals. The solution was then cooled to
about 10 0 C, and the crystals were filtered with suction but not allowed
to dry between recrystallizations. A yield of about 40 percent for
each recrystallization was obtained. The first mother liquor was
pinkish and contained a small amount of sulfate. The subsequent
mother liquors were colorless and sulfate t ests were negative.
Two separate lots of potassium p-phenolsulfonate were recrystal-
lized four and six times, respectively. Some material from each
recrystallization was reserved for experimentation and dried for 2
hours in a vacuum oven at 2-mm pressure and 60° C. Analyses of
the recrystallized product by micromethods for hydrogen, carbon, and
sulfur, made by K. D. Fleischer of this Bureau, gave the following
results: Found: C, 34.02, 34.03; H, 2.40, 2.57; S, 14.92, 15.04. Theo-
retical: C, 33.95; H, 2.37; S, 15.10.
III. BROMOMETRIC ASSAY OF POTASSIUM
p-PHENOLSULFONATE
Previous work on furfural and related compounds 2 showed that
under controlled conditions of acidity and temperature the reaction
of bromine with some organic compounds can be used as a method for
their quantitative determination. Preliminary experiments indicated
that the reaction of bromine with phenolsulfonate gives slightly high
results at room temperature. However, using essentially the method
previously reported for furfural, it was found that at 0 0 C 2 molecules
of bromine react within 5 to 10 minutes with 1 molecule of p-phe-
nolsulfonate in normal hydrochloric acid to form 2,6-dibromophenol-
4-sulfonate and 2 molecules of hydrobromic acid. The reagents used
and experimental procedure are given below.
'Elizabeth E . Hugbes and S. F. Acree, Quantitative determination oj JurJu ral at O· C with bromine, Ind
Eng. Chern. 6, 123 (1934); Volumetric estimation oj 5-bromo-f-Juroic acid with bromine, Ind. Eng. Chern. 6, 292
(1934).
 Purification and A88ay of Pota88iwIn p-Phenol8ulfonate 199
1. REAGENTS

Potassium bromate was recrystallized three times from water, and

5.567 g was used to make a solution approximately 0.2 N, which con-
tained 50 g of potassium bromide in 1 kg of water. The bromate-
bromide solution was used on a weight basis to standardize in normal
hydrochloric acid through iodine the 0.1 N thiosulfate solution used
for the titrations. The weighed amount of the bromate-bromide
delivered by a special 25-ml pipet with curved tip was reproducible
within 0.05 percent. A solution of potassium iodide, about 10 per-
cent by weight, was prepared freshly before use. Hydrochloric acid
was used to make the reacting solution about normal in acidity. The
potassium bromide, potassium iodide, and hydrochloric acid conformed
to specifications of the American Chemical Society.3 As an additional
check, the thiosulfate was standardized against potassium dichromate,
recrystallized twice, using the same procedure described below. The
two methods of standardization agreed within 0.1 percent.
2. EXPERIMENTAL PROCEDURE
Specially designed titration flasks, described previously (see foot-
note 2), were fitted with two side-arms for the bromate and iodide
solutions, respectively. By means of pipets with curved tips, 25 ml
of the 0.2 Nbroma te-bromide solution was placed in one side-arm and 10
ml of the potassium iodide solution in the other. A known amount of
phenolsulfonate was transferred to the body of the flask and diluted
to about 200 ml. The solution was made normal in acidity by the
addition of hydrochloric acid, using extreme care that the fumes did
not reach the bromate in the side-arm . Blanks of the reagents were
prepared in a similar manner, substituting water for the phenolsul-
fonate. Each flask was closed with a ground-glass stopper and placed
in an icebox to attain a temperature of 0° to 2° C.
The reaction was started by tilting the flask so that the bromate-
bromide solution ran into the body of the flask, and the contents were
gently swirled to mix the liberated bromine with the phenolsulfonate
without disturbing the potassium iodide solution in the other sIde-arm.
A stop watch was used to time the reaction, which was stopped by
the addition of the potassium iodide. The contents were thoroughly
shaken to absorb any bromine vapors, and the iodine equivalent to the
residual bromine was titrated with 0.1 N thiosulfate, using starch as an
indicator. At 0° C the discharge of the blue color at the end point
was very definite, but it was found that the titration must be made
as rapidly as possible. The blue color returns after several minutes,
and consequently error may be introduced if the titration is prolonged .
Experiments over a period of several hours, in which the blue color
was discharged with thiosulfate as soon as it returned, showed that this
error was of the order of 0.01 ml per minute at low temperatures
and twice this amount at room t emperatures.
The amount of bromine consumed by the phenolsulfonate is deter-
mined from the difference between the amounts of standard thiosul-
fate solution used in titrating the blank and the solution containing
the phenolsulfonate. The figures in table 1 for "percent reaction"
are calculated on the assumption that in the desired reaction 1 mole
of the phenolsulfonate consumes 2 moles of bromine.
3 ACS Analytical Reagents. (American Chemical Society, Washington, D . C., March 1941).
200 Journal of R esearch of the National Bureau of Standards
TABLE I.- Reaction of potassium p-phenolsulfonate with bromine at 0° C in normal
hydrochloric acid

I 0.1698 g of 0.21?4 g of
potassium p-pbeuolsulfonate potassium p-pbenolsuHonate
Time
allowed
for lVJilliequiva- MiIIiequiva-
reaction lents of Percentage lents of Percentage
bromine of reaction bromine of reaction
consumed consumed
----
Minutes
2 3.937 98.4
- ---------
~ -----------
{ 3.922 98.0
5 3.195 99.9 4.001 100.0
10 3. 199 100.0 4.006 100.1
{ 3.200 100.0 4.002 100.0
15 3.199 100.0 4.004 100.0
{ 4.009 100.1
20
{ 3.205 100.1 ----------- --- ---- ---.
30
3.207
3.210
100.2
100.2
----- ----.- .---- ------
-.-.- - ----- - ----------
120 3.256 101. 8 4.025 JOO.6
180 ----------- - ---- - ---- - 4.038 JOO.9

The values in table 1 show the variation with time in the consump-
tion of bromine at 0° C by two concentrations of phenolsulfonate.
In all cases there were approximately 5.0 milliequivalents of bromine
available, requiring 50 ml of 0.1 N thiosulfate for titration of the
blank. For the above amounts of phenolsulfonate, the excess bromine
required about 18 and 10 ml of thiosulfate, respectively. The table
indicates that the reaction of 1 molecule of phenolsulfonate with 2
molecules of bromine is complete after about 5 minutes, and little
additional reaction is noted until after 30 minutes. The figures for
percentage of reaction show that the method is accurate to about 0.1
percent. A difference of 0.02 ml in thiosulfate titration, representing
in the above instances less than 0.1 percent, is possible due to the
difference in time taken to add the 50 ml for titration of the blank and
that taken to add the smaller volume in the titration of the sample.
The values in table 2 show the analyses of the original material and
the recrystallized fractions.
TABLE 2.-Bromometl·ic assay of samples of potassium p-phenolsu~fonate sub-
jected to a series of recrystallizations

First Second
lot lot

Percent Fercent
Original materiaL ______________________ _ 92.8 92.7
Recrystallization 1. ____________________ _ 9S.4 99.6
Recrystallization 11. ____________________ _ 99.7 100.0
Recrystallization III. _________ ___ ______ _ 100.0 100.1
Recrystallization IV ___ _________ ________ _ 100.0 100.0
Recrystallization V ___ __________________ _ 100.1
Recrystallization V1. ___________________ _ 100.1

IV. SPECTRAL ABSORPTION IN THE ULTRAVIOLET BY

POTASSIUM p-PHENOLSULFON ATE
In the spectrophotometric studies of indicators stabilized with
10- 2 molar phenolsulfonate buffer, it was found that ultraviolet
absorption spectra could not be obtained below 280 mj.L, owing to the
absorption bands of the buffer itself. A study therefore was made of
the buffer without the indicator. Repeated dilutions of the phenol-
 PU1'ijication and Assay of Potassiwin p-Phenolsulfonate 201
sulfonate solutions were made and their transmittancy values measured
until the main absorption bands were located. The series of spectral
transmittancy values given in figure 1 represent concentrations of
the phenolsulfonate in water from 10- 2 molar (curve A) to 10- 5 molar
(curve J). The main absorption band, whose peak is at 230 mJ.!, is
not sharply defined except in the more dilute solutions (curves E
through J). The transmittancy curves show bands similar to those
of indicators.
Dilution of the phenolsulfonate with water immediately introduces
the question of dissociation. At any wavelength the molar absorp-

I-
Z POTASSIUM
W
o p-PHEN OLSULF -
0::
ONATE
~ 60L-~~~~--~~4+----~------~4+~+
A - 10- 2 M
~
B - 10- 3 M
o>- C - 5 X 10- 4 M
z o- 3 X 10- 4 M
~ E - 2 X 10- 4 M
I-
~ F - 1.0- 4 M
(f)
G - 5 X 10- 5 M
Z
<t: H - 2.5 X 10- 5 M
0::
I- I - 10- 5 M

260
WAVELENGTH IN MILLIMICRONS
FIGU RE I.-Ultraviolet spectra of several concentrations of purified potassium
p-phenolsulfonate.
tion index is determined by dividing the index ( -loglo transmittancy)
for that solution by the concentration multiplied by the depth of the
solution (cell length in centimeters) through which the light beam
passes, according to Beer's law. If the molar absorption indices do
not agree well within experimental error for several wavelengths,
then dissociation, association, or some chemical change is suspected.
For this purpose the absorption indices of solutions of phenolsulfonate
10-4, 2XIO- 4, and 3XlO- 4 molar in water were first compared.
Additional solutions of the same concentrations were made in 10- 1
molar hydrochloric acid. These data agreed well within experimen tal
error for practically the entire absorption curve, except on the stef'p
portions of the bands where the errors are large, showing that the
sulfonate gTOUp is completely ionized and the phenolic group non-
ionized in all the above solutions. As an additional check, a 10- 4
molar solution in the 5-cm cells was compared with a solution in the
1- and 2-cm cells. The agreement was well within experimental error.
The spectrophotometric data were obtained by use of a Beckman
quartz photoelectric spectrophotometer.4 It was improved by the
• Made by the National Technical Laboratories, Pasadena, Calif. Described in J . Opt. Soc. Am. 31,
682 (1941).
202 Jowrnal of Research of the National Bureau of Standards

replacement of the cell compartment 5 by a small light-tight box with

a slide for holding two 38-mm (i. d.) cylindrical metal cell holders in
which Pyrex cylinders of 1-, 2-, and 5-cm lengths could be fitted with
quartz end plates to hold the solutions. One cell was filled with the
solution under investigation. The solvent, or 100-percent cell, con-
tained the same components of the solution except the compound
under investigation. The cells were placed alternately in the light
beam for the transmittancy measurements. They were calibrated
each day at the beginning and end of the test by filling both cells with
distilled water, using the solvent cell for 100-percent transmission and
reading the value for the solution cell at lO-mJ.L intervals from about
210 to 400 mJ.L. The quartz end plates used in this work were well-
matched and gave a satisfactory calibration curve over the entire
spectral range. The transmittancy values obtained for the solutions
were corrected by use of the calibration curve. Reproducibility of
transmittancy values for a given concentration of solution prepared
at different times was usually within ±0.2 percent, except at the
lowest wavelengths (210 to 230 mJ.L), where discrepancies of ±0.5
percent transmittancy occur. The measurements were made in a
room kept at 25° ±2° C. Solutions were prepared in a room also
kept at a constant temperature of 25° C. This control of tempera-
ture is important in spectrophotometric studies of solutions, especially
when the pH value of the solution and ionization of the buffer change
with temperature, and cause corresponding differences in the trans-
mittancy of the solution.
The ultraviolet transmittancy values of the potassium p-phenol-
sulfonate subjected to the recrystallization steps described above
were also obtained and are given in figure 2. The original material
is represented by curve A. The first recrystallization apparently
removes most of the impurities which absorb light6 between 250 and
300 mJ.L, as shown by curve B, whereas the third, fourth, fifth, ang
sixth crops of crystals give practically the same transmittancy values,
curve O. The material successively recrystallized from the first lot of
phenolsulfonate showed a systematic change in the transmittancy
values, as in the bromine analyses (given in table 2 but not shown in
fig. 2).
Studies of a number of buffers have shown them to be highly
resonant and clu:omophoric in the ultraviolet, and also that their
absorption index bands are altered greatly by pH changes, just as
in the case of indicators in the visible, ultraviolet, and infrared
regions. Some measurements therefore were made of the ultraviolet
absorption of solutions of the secondary p-phenolsulfonate salt, which
is formed when alkali is added to a solution of the primary salt. The
spectrophotometric curves shown in figure 3 represent the p-phenol-
sulfonate, 5 X 10-5 molar, in 10- 1 molar hydrochloric acid (curve A)
and the same concentration in 10-3 molar sodium hydroxide (curve D).
The two intermediate curves, Band 0, were obtained by using dilute
borax-boric-acid buffers of pH 8.4 and 9.0, which do not have absorp-
tion Dands in this region. In these dilute solutions there may have
been some error due to the absorption of carbon dioxide during the
• The authors are indebted to Jurg. A. Senn for help in designing the metal box, as well as other improve-
ments in the electric circuit, and calibration of the instrument. Extensive data, which have not yet been
reported, were obtained previously with filters and indicator solutions to test the performance of the in-
strument. The precision is about ±0.5 on the scale from 0.0- to 100.0-percent transmittancy.
• For example, figure 3 shows that any secondary salt present as impnrity absorbs strongly between 240
and 300m".
 Purification and Assay of Potassiwrn p-Phenolsulfonate 203
100 r-~-------'-------:------'---~-=~~

~ 80 1----+-------~-------4--------++~----~--~
Z
W
o
0::
W
~ 60 ~--+_------_r--·~---4~~~r~

o>- POTASSIUM I> - PHENOLSULFONATE

z r----+---------+--I-------4- ~ 10-4~'u_______~----1
«
~ i I

t=
~
A - ORIGINAL MATERIAL I
(j) B - RECRYSTALLIZATION I I
z
« C- i n, m, f,Y,Jll
0:: I
~

260 280 300

IN MILLIMICRONS ( M.u)
FIGURE 2.-Ultraviolet spectra of impure and recrystallized potassium p-phenol-
sulfonate.
101~-''--------'------~------~--~~--· ~~

~801 ~-+---------r~-----4---~---=~----~~--~
z
w
o
0::
~601~~~----~-4~--~~+-~~---+---------~--~
z
POTASSIUM I> - PHENOL-
o>- SULFONATE (5 X 10.5 M)
z40·~--~~.---~~~------~~

~ A - IN 10- 1M HCI (pH 1.0)

~
B-
. BUFFER ( .. 8.4)
~
(/)
~ 201r-~---------+--~------1---- C -
. ( . 9 .0)

0:: 0-
. 10-3 M Na.OH
~

0'~~2~2~0~--~2~4~0----~2~6~0----~2*8=0-----=30kO~­
WAVELENGTH IN MILLIMICRONS - (M,u)
FIGURE 3.-Ultraviolet spectra which represent the transformation of potassium
p-phenolsulfonate from the primary to the secondary salt.
204 Journal of Research of the National Bureau of Standard8

filling of the spectrophotometric cells. Curve D, in which 20 moles

of alkali are added to the phenolsulfonate, probably represents its
complete conversion into the secondary salt. A series of intermediate
curves has since b een obtained, and calculations are under way in
correlating the spectral curves with the neutralization and ionization
of the phenol group. The preliminary curves given here show that
such data furnish valuable information in the study of the spectral
properties, ionization constant, activity coefficients, and general
behavior of such polar buffers (see fig. 4). The curves show that
the secondary salt has an entirely different resonance structure,
leading to an absorption index curve which is shifted 25 m~ toward
the red. Two isobestic points were observed at 215 and 238 m~.
Of the two resonant structures 1 and 2, figure 4, for the primary
salt, 2 is probably predominant and in equilibrium with only a small

2 3 4
FIGURE 4.-Structural and resonant forms of p-phenolsulfonates. 1 and 2, resonant
forms of the univalent salt; 3 and 4, resonant forms of the bivalent salt .

amount of the dipolar quinoidal form of 1. The resonance is indicated

by the dotted lines. From similar data on other sulfonates and
buffers, it is believed that the bands in figure 1 correspond chiefly to
the absorption of the sulfonate group, with peaks at about 230, 271,
and 278 m~ and the indices decreasing in the same order. There is
apparently also a weak band with a peak at about 266 m~. Because
of the marked change in the character of the absorption and the shift
of the band toward the red upon increase in pH and formation of the
secondary salt, it seems likely that the resonance path is lengthened
and hence extended considerably outside the sulfonate group. This
would be represented by the equilibrium between 3 and 4, figure 4,
with 4 probably predominating. The peak of the dominant band is
at about 254 m~, and there is apparently a weaker band with a peak
at about 280 to 285 m~. Such spectrophotometric data help in
giving a clearer understanding of the ionic and polar structures of
buffers and their ions and the activity coefficients to be assigned to
them. Such data also show that in securing very accurate spectro-
photometric values for the absorptions by indicators and organic
compounds, especially in purification procedures, it is important to
control the pH values and the concentrations of each component.
WASHINGTON, August 7, 1943.