l

w; r

(I
If

II

I
ll

. w i 1'
‘ w Ir
"H

I

y |

M

7‘
___

If
I

I

'H

fl

 

-H#-‘
Ioom
(0—4—8

KINETJCS AND MECHANESM OF THE BERTHELOT
REACTION WITH PARTECULAR REFERENCE '50 AMMONIA.
ANALYSIS

Thesis fer the Degree of M. S.

MECHJGAN STATE UNWERSETY

CHAREES JOHNSTON PAWON
3.975

lugma

 

 

 

\J:M., ”,. V'
-_.———q Qiigk-

- “wofw
gg* h‘fiq

7 ABSTRACT

KINETICS AND MECHANISM OF THE BERTHELOT REACTION
WITH PARTICULAR REFERENCE TO AMMONIA ANALYSIS

By
Charles Johnston Patton

An investigation of the mechanism of the Berthelot reaction
has led to an increased understanding of the many conflicts in the
literature concerning the reaction‘s use for the analytical deter—
mination of ammonia. Monochloramine is shown to be the first inter-
nediate, and the kinetics of its formation and decomposition is pre-
sented. The identical visible spectra obtained for a series of in-
dophenols synthesized in the course of this research and the blue
complexes formed when phenol or a substituted phenol was used in the
Berthelot reaction has confirmed that indophenol is the final prod-
uct of the reaction. Conditions which led to the formation of final
products other than indophenol have been observed and are also dis-
cussed. Finally, a new procedure for ammonia analysis using the
Berthelot reaction, which is superior in both sensitivity and pre-

cision to the best procedure found in the literature, is presented.

KINETICS AND MECHANISM OF THE BERTHELOT REACTION
WITH PARTICULAR REFERENCE TO AMMONIA ANALYSIS

By

Charles Johnston Patton

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE

Department of Chemistry
l975

DEDICATED TO
LEROY N. HAYNES

The blues ain't nothin' but a low-down shakin' chill.
If you ain't never had 'em, God knows I hope you never will.

Robert Johnson

ii

ACKNOWLEDGMENTS

The author wishes to express his appreciation to Dr. S. R.
Crouch for his guidance and encouragement during the course of this
investigation.

A large measure of thanks and gratitude are also due to
Philip Notz, who spent many hours modifying and computer interfacing
the stopped-flow spectrophotometer used in this research.

Dr. Eric R. Johnson also deserves mention for writing sever-
al computer programs used in this work.

Many thanks to Dr. Peter Kissinger for his guidance in Dr.
Crouch's absence; his stern, fatherly advice; and an endless supply
of gin and tonics. Thanks also to his research group for late night
comradship.

The author also wishes to thank Brother Dave, Mary, his
parents, and his parents-in-law for their support and encouragement.

Finally, to all those individuals never mentioned in "Acknow-
ledgment" sections, who made life a little less bearable by their ex-
istance, kindly observe the mistletoe pinned to the tails of the
author's coat as he walks down the corridors of the chemistry build-

ing for the last time.

1°11

TABLE OF CONTENTS

Page
LIST OF TABLES ....................... vii
LIST OF FIGURES ...................... viii
I. INTRODUCTION ..................... 1
II. HISTORICAL ...................... 3
Evolution of the Berthelot Reaction ........ 3
Kinetics and Mechanism of the Berthelot Reaction . . 4

Role of Pentacyano Iron Complexes in the Berthelot
Reaction ..................... 6

Use of Substituted Phenols in the Berthelot

Reaction ..................... 8
III. EXPERIMENTAL ..................... 10
Equipment and Instruments ............. 10
Chemicals ..................... 10
Standardization Procedures ............. 10
Data Analysis ................... 11
IV. SYNTHESIS ...................... 12
Preparation of Indophenol and Related Compounds . . 13
Synthesis of 4-nitroso alkylphenols ...... 14
Synthesis of 4-nitroso chlorophenols ...... 15
2-methyl-4-nitrosophenol ............ 15
2,6-dimethy1-4-nitrosophenol .......... l6
3,5-dimethyl-4-nitrosophenol .......... 16
3-methyl-6-isopropy1-4-nitrosophenol ...... 16
2,3,5,6-tetramethy1-4-nitrosophenol ...... 16
2-chloro-4-nitrosophenol ............ l6
2,6-dichloro-4-nitrosophenol .......... 17
Synthesis of 5-nitroso-N-methy1anthranilic acid 17

Synthesis of 2-hydroxy-5-nitrosobenzoic acid . . 17
Synthesis of substituted 4-aminophenols . . . . 18

2-methy1-4-aminophenol ............. 18
2,6-dimethy1-4-aminophenol ........... 18
3,5-dimethyl-4-aminophenol ........... 19
3-methy1-6-isopropy1-4-aminophenol ....... 19
2,3,5,6-tetramethyl-4-aminophenol ....... 19
2-chloro-4-aminophenol ............. 19

iv

2,6-dichloro-4-aminophenol ...........
2-hydroxy-5-amino benzoic acid .........
Synthesis of quinone-N-chloroimines ......
Quinone-N-chloroimine .............
Z-methylquinone-N-chloroimine .........
2,6-dimethquuinone-N-chloroimine .......
3,5-dimethy1quinone-N-chloroimine .......
3-methy1-6-isopropy1quinone-N-chloroimine
2,3,5,6-tetramethylquinone-N-chloroimine . . . .
2-chloroquinone-N-chloroimine .........
2,6-dichloroquinone-N-chloroimine .......
Synthesis of indophenols ............
Indophenol, sodium salt ............
2,2'-dimethylindophenol, sodium salt ......
2,6,2',6'-tetramethylindophenol, sodium salt . .
3,3'-dimethyl-6,6'-diisopropy1indophenol, sodium
salt ......................
3,5,3',5'-tetramethy1indophenol, sodium salt . .
2,3,5,6,2',3',5',6'-octamethylindophenol, sodium
salt .....................
2,2'-dich10roindophenol, sodium salt ......
2,6,2',6'-tetrachloroindophenol, sodium salt . .
Discussion ...................
Preparation of Other Materials ...........
Synthesis of sodium hypochlorite ........
Synthesis of monochloramine ..........
Synthesis of sodium aquopentacyanoferrate
Discussion ...................

V. NON-KINETICS EXPERIMENTS ...............

UV Spectrum of Hypochlorous Acid ..........
UV Spectrum of Monochloramine ...........
Visible Spectra of Pentacyano Iron Complexes . . . .
Visible Spectra of Indophenols ...........
Formation of Indophenols in Aqueous Solution . . . .
Identity of the "Catalyst" for the Berthelot
Reaction .....................

VI. KINETICS EXPERIMENTS .................

Formation of Monochloramine ............
Decomposition of Monochloramine ..........
Mechanism and Kinetics of Indophenol Formation .

Rates of Formation of Several Indophenols .....

VII. A NEW AMMONIA METHOD .................

Reagents ......................
Ammonia Standards ...............
Sodium hypochlorite ..............

Sodium aquopentacyanoferrate ...... ‘. . . .
2-chlorophenol .................
The Method .....................
Discussion ...................

LIST OF REFERENCES .....................

vi

LIST OF TABLES
Table Page

1. Absorption Maxima and Molar Absorptivities
of Some Pentacyano Iron Complexes ........ 32

2. Absorption Maxima and Molar Absorptivities
of Some Indophenols and Their Sodium Salts . . . . 34

3. Effect of Aquopentacyanoferrate Concentration
on the End Product of the Berthelot Reaction . . . 36

4. Effect of Aquopentacyanoferrate Concentration
on the End Product of the Berthelot Reaction
in the Presence of Hypochlorite Ions ....... 36

5. Data Obtained When Ammonia Was Analyzed by the
Proposed Method ................. 56

6. Data Obtained When Ammonia Was Analyzed by the
Solorzano Method ................. 57

vii

Figure

10.

11.

12.

13.

LIST OF FIGURES

UV Spectrum of Hypochlorous Acid as a Function
of pH ......................

UV Spectrum of Hypochlorite Ions .........
UV Spectrum of Monochloramine ...........

Visible Spectrum of 2,6,2',6'-tetramethy1indophenol
as a Function of pH ...............

Formation of Monochloramine from Ammonia and Hypo-
chlorous Acid in 1:1 Molar Ratio ........

Formation of Monochloramine from Ammonia and Hypo-
chlorous Acid in 1:5 Molar Ratio ........

Formation of Monochloramine from Ammonia and Hypo-
chlorous Acid in 1:10 Molar Ratio ........

Log of Observed Second-order Rate Constants for
the Formation of Monochloramine versus pH at
292 nm (o) and 243 nm (I) ............

Distribution Diagrams for Hypochlorous Acid-
Hypochlorite Ions and Ammonia-Ammonium Ions
as a Function of pH ...............

Decomposition of Monochloramine as a Function of
Absorbance at 243 nm and Time at pH values of
8.5 (I), 10.5 (0), and 12.0 (A) .........

Decomposition of Monochloramine as a Function of
Absorbance at 292 nm and Time at pH Values of
8.5 (I), 10.5 (o), and 12.0 (A) .........

Rate of Formation of 2,2'-dimethy1indophenol (A),
Indophenol (I), and 2,2'-chloroindophenol (0) . .

Calibration Curves for Ammonia as Determined by
the Solorzano (o) and Proposed (I) Methods

viii

Page

30
31
31

33

41

42

43

44

45

48

48

50

58

I. INTRODUCTION

Although a large number of analytical schemes for ammonia
analysis have appeared in the literature, the method finding the most
widespread application is the Berthelot reaction. In this reaction,
indophenol, an intensely blue charge-transfer complex, is formed
when an aqueous solution containing ammonia is treated with phenol
and hypochlorous acid. Under controlled conditions, the amount of
indophenol produced is proportional to the concentration of ammonia
originally present in the solution. The method is specific for am-
monia, and when sodium nitroprusside is used as a "catalyst," detec-

tion limits for ammonia concentrations as low at 10'7

M_have been
reported. The precision of such procedures is quite poor, however,
and despite the many papers which have appeared in the literature
describing "improved" procedures for the catalyzed Berthelot reac-
tion, a truly satisfactory method has yet to appear.

The trouble with most of the work to date on the Berthelot
reaction is that while reaction parameters have been wildly juggled
to "optimize" various procedures, little attention has been paid to
the mechanism which leads to indophenol formation. In this research,
a mechanistic approach to the problem of poor precision in the Ber-
thelot reaction was undertaken. The rate of formation of monochlor-

amine, the first intermediate in the Berthelot reaction, was deter-

mined as a function of pH. The rate was maximum at pH 8.0 with an

4 3 1

observed second—order rate constant of 2.4 x 10 i 0.4 x 10 a mole-

sec-1. It was also observed that dilute aqueous solutions of mono-
chloramine decomposed extensively in the presence of hypochlorous
acid if the pH was less than 10.0.

In other experiments, the actual catalyst for the Berthelot
reaction was found to be sodium pentacyanoaquoferrate, rather than
sodium pentacyanonitroferrate or sodium pentacyanonitritoferrate as
reported by other workers. This observation has not been reported
elsewhere. Also, a series of substituted indophenols was synthesized,
and the details of their preparation, as well as their visible spec-
tra, have been included. These data confirmed that indophenol was
the final product of the Berthelot reaction.

The final outcome of this research was a new analytical pro-
cedure for ammonia which has a lower detection limit and higher pre-
cision than any published procedure for the catalyzed Berthelot re-
action. This method could be easily automated and, with minor modifi-

cations, should be well-suited to the analysis of ammonia in natural

waters.

II. HISTORICAL
Evolution of the Berthelot Reaction

The blue color produced in alkaline solutions containing
phenol, hypochlorite ions, and ammonia was first reported by Berthe-
lot (4). This reaction was first used analytically by Thomas (42)
to determine ammonia in blood. Although several other workers (6,43)
modified Thomas's procedure, the first real improvement in ammonia
analysis via the Berthelot reaction was reported by Russel (35).

She found that the reaction was catalyzed by Mn(II) ions, and ob-
served a three-fold increase in sensitivity relative to the uncata-
lyzed reaction.

Ten years later, two other catalysts more effective than
Mn(II) ions were reported: acetone (10) and sodium nitroferricya-
nide (22). The use of these new catalysts, especially sodium nitro-
pentacyanoferrate (nitroprusside), increased the sensitivity of the
Berthelot reaction to the extent that concentration of the ammonia
in biological fluids and natural waters prior to analysis was no
longer necessary. This, coupled with the reaction's relative insen-
sitivity to nitrogen-containing compounds other than ammonia, led to
its widespread application. Over the past twenty years, literally
scores of papers have appeared describing new applications or modi-

fied procedures for the catalyzed Berthelot reaction.

Unfortunately, few of these papers agree on such reaction
parameters as reagent concentrations, pH, or order and timing of
reagent additions. For example, a few papers (5,36) reported en-
hanced sensitivity when the chlorinating reagent was added first,
while most other papers reported virtually no sensitivity unless
the phenol reagent was added first. This led some researchers
(15,37) to the erroneous conclusion that the phenol-hypochlorite
and hypochlorite-phenol reactions were two distinct entities. Numer-
ous conflicts in the literature give the impression that no single
mechanism is operative in the catalyzed Berthelot reaction since the
only thing most procedures have in common is extremely poor preci-

sion.

Kinetics and Mechanism

of the Berthelot Reaction

Very little has been published concerning the kinetics and
mechanism of the Berthelot reaction. Russel (35) suggested that the
end product of this reaction was indophenol or a closely related com-
pound. Bolleter, gt;_al,, (5) proposed a generalized mechanism (out-
lined below) in which monochloramine and quinone-N-chloroimine were
intermediates and indophenol was the end product. The role of a

catalyst was not considered.

_ 11>
N33 + on L—s» H2NCI + ‘OH

monochloroimine

H2NCI + .an 4911—» o=.=Nc1

quinone-N-chloroimine

a... + cm‘o JO—H. oHN 0’ NJ

indophenol, sodium salt

There is substantial evidence in the literature (7,11,24)
which supports the first step of Bolleter's proposed mechanism. At
pH values greater than 8.5, monochloramine is the sole product in
the reaction between equimolar concentrations of ammonia and hypo-
chlorite ions. The fact that most papers in which reaction param-
eters for the Berthelot reaction were investigated reported a drastic
reduction in sensitivity when the hypochlorite reagent was added
first, however, led many workers to discount the proposed mechanism.

This conflict over the preferred order of reagent addition
seems to have been resolved by Horn and Squire (18). They noted that
all procedures in which the phenol reagent had to be added first in
order to obtain satisfactory sensitivity used hypochlorite reagents

in at least one thousand-fold excess of the stoichiometric amount

necessary to convert all the ammonia in the sample to monochloramine.
They found, however, that the order of reagent addition could be re-
versed, affecting an increase in sensitivity and precision in these
procedures if the concentration of the hypochlorite reagent was re-
duced by two orders of magnitude. This observation suggested that
monochloramine was unstable in the presence of hypochlorite ions.

It is interesting to note that Morris and Neil (27), who did exten-
sive research on the products formed when waste water was chlorinated,
reported that monochloramine was unstable in solutions containing a

large excess of hypochlorite ions.

Role of Pentacyano Iron Complexes

in the Berthelot Reaction

In a subsequent paper (19), Horn and Squire presented evi-
dence that the catalyst for the Berthelot reaction was sodium nitri-
topentacyanoferrate which formed spontaneously when sodium nitro-
ferricyanide, the assumed catalyst, was prepared in mildly alkaline
solutions. They reported that conversion of the nitro complex to the
nitrito complex prior to its use as a reagent increased both the
sensitivity and precision of the Berthelot reaction. Later, Mark
(23) confirmed this observation and suggested that the nitrito com-
plex catalyzed the formation of monochloramine.

There is evidence in the literature not pertaining directly
to the Berthelot reaction, however, which suggests that Mark's hy-
pothesis concerning the catalytic role of sodium nitritopentacyano-

ferrate is in error. First, work of Neil and Morris (47) and of

Yagil and Anbar (49) has shown that the formation of monochloramine
from ammonia and hypochlorous acid in the pH range of 7.5 to 11.5 is
too rapid to be studied by conventional techniques. It does not seem
likely that a reaction this rapid would require a catalyst. Second,
work of Ohkuma (30) and of Swinehart and Rock (41) has shown that
while sodium nitropentacyanoferrate is converted to sodium nitrito-
pentacyanoferrate, the nitrito complex is in equilibrium with sodium

aquopentacyanoferrate, as shown in the following scheme. For this
[Fe(CN)5NO]2- + 2'0H sé [Fe(CN)5N02]4' + H20
[Fe(CN)5N02]4' + H20 # [Fe(CN)5H20]3' + N02-

reason, it is possible that the real catalyst for the Berthelot re-
action is the aquo complex rather than either the nitro or the ni-
trito complexes.

In another paper (29), Ohkuma reported that sodium nitrofer-
ricyanide is converted directly to sodium aquopentacyanoferrate by
alkaline hydroxylamine. The aquoferrate thus formed incorporates a
second molecule of hydroxylamine as a ligand, and in the presence of

phenol, aminophenol was produced as shown below. The similarity be-

Na2[Fe(CN)5NO] + HZNOH + NaOH -———) Na3[Fe(CN)5H20] + N2 + -OH

' 'OH
Na3[Fe(CN)5H20] + HZNOH ——> Na3[Fe(CN)5H2NOH] + H20

Na3[Fe(CN)5H2NOH] + .ou -°—”+ HO—.—NH2 + Na3[Fe(CN)5H20]

tween monochloramine and hydroxylamine suggests first that an

analogous mechanism may explain the catalytic role of sodium nitro-
ferricyanide in the Berthelot reaction, and second that sodium aquo-
pentacyanoferrate may be a better catalyst than either sodium nitro-
pentacyanoferrate or sodium nitritopentacyanoferrate. To date,
there has been no mention in the literature concerning the relevance
of these works to the elucidation of the reaction mechanism for the

catalyzed Berthelot reaction.

Use of Substituted Phenols

in the Berthelot Reaction

Another interesting development in the evolution of the Ber-
thelot reaction as an analytical tool is the use of substituted
phenols. Roskam and de Langen (34) reported a two-fold increase in
sensitivity and more stable blanks when thymol (3-methyl-6-
isopropylphenol) was used in place of phenol in the acetone-
catalyzed Berthelot reaction. Procedures in which salicylic acid
(33), 2-chlorophenol (51), 3-methy1phenol (50), and l-naphthol (26)
were used have also been reported. Although no mention was made in
these references concerning the mechanistic or kinetics implications
of using substituted phenols in the Berthelot reaction, the work of
several dye chemists is relevant.

Corbett (9) reported a synthesis for indophenols in solution
as sodium salts by the ferricyanide oxidation of aminophenols to the
corresponding quinone imines. The imines thus formed coupled spon-

taneously with phenol to give indophenolate ions when the pH range

was between 8.5 and 11.5, as shown in the following scheme. Corbett

2[Fe(CN) 613' + H0—.—NH2 T—-———* O-.=NH + 2[Fe(CN) 614' + 2H+

quinone imine

O=.=NH + .OH :91» Na+'O—.—N‘.=O

found that in the preparation of methyl substituted indophenolate
ions, the use of the less methylated aminophenols and the more meth-
ylated phenols led to the fastest coupling rates. Thus when quinone
monoimine was reacted with 2- and 3-methylphenol, for example, there
was a thirty- and ten-fold increase, respectively, in the rate of in-
dophenolate ion formation relative to phenol. In the case of chloro-
indophenolate ion synthesis, the use of the more chlorinated amino-
phenol and the less chlorinated phenol led to the fastest coupling
rates. Gibbs, Hall, and Clark (14), in a much earlier work, made
this same observation regarding the coupling rates of quinone-N-

chloroimines and phenols although no quantitative data were given.

O‘NCl + H‘OH + 20H —) OHNO-

These observations suggest that there might be a kinetics advantage

in using certain substituted phenols in the Berthelot reaction.

III. EXPERIMENTAL
Equipment and Instruments

Melting point data were obtained with an oil bath immersion
apparatus (Thomas-Ashley). A11 pH data were obtained with a servo
digital pH/volt meter (Heath EU-302A) and a combination reference
and pH electrode (Beckman). All visible and ultraviolet spectra
were obtained with a scanning spectrophotometer (Cary 17). Infrared
spectra were obtained with a scanning infrared spectrophotometer
(Beckman AccuLab 2). Data from kinetics experiments were obtained
by using an automated stopped-flow spectrophotometer designed in

these laboratories (2,3).
Chemicals

The phenols used as starting materials for indophenol syn-
thesis were obtained from three sources: Aldrich, Eastman Kodak,
and K and K. All were marketed as = 97% purity and were used as re-
ceived. All other chemicals used in this research were reagent

grade, and no attempt was made to purify them further.
Standardization Procedures

Solutions of sodium hypochlorite and monochloramine were

standardized by thiosulfate titration. Two grams of potassium

10

ll

iodide and 5 ml of glacial acetic acid were added to an aliquot of
the solution to be analyzed. The sample was immediately titrated to
a starch end point with a 0.1 N_sodium thiosulfate, which had been
previously standardized against primary standard grade potassium
iodate.

Phosphate and borax buffers in the pH range of 4.0 to 12.5
were prepared as described in the "Handbook of Chemistry and Physics"
(46). These were referenced relative to commercial standard buffers

of pH 4.00, 7.00, and 10.00.
Data Analysis

General calculations, including standard deviation and
linear regression analysis, were performed on an SR-Sl calculator
(Texas Instruments). Control of the stopped-flow spectrophotometer,
data acquisition, and formating was performed by a POP/8e minicom-
puter (Digital Equipment Corporation). Data from kinetics runs
were analyzed with the aid of KINFIT, a generalized curve-fitting

routine developed by Dye and Nicely (13).

IV. SYNTHESIS

Many of the compounds and reagents used in this research
were not commercially available. The details of their preparation
are included in this chapter. The first section covers the synthe-
sis of symmetric indophenols, where the word symmetric implies that
the identity and position of substituents are identical on both rings
of the indophenol. The numbering system used for naming indophenols

is shown below. The second section covers the synthesis of sodium

 

hypochlorite, monochloramine, and sodium aquopentacyanoferrate, which
are participants in the Berthelot reaction.

All yields reported for compounds synthesized were calculated
on the basis of total dry weight of the crude product recovered.
Generally, only a small portion of the crude product was recrystal-
lized or sublimed for melting point and spectra determination, while
the rest was used as recovered for succeeding syntheses. Any signi-

ficant details for a particular synthesis not included in the general

12

13

procedures for each class of compounds can be found at the end of

each section.

Preparation of Indophenol

and Related Compounds

Since the end product of the Berthelot reaction is generally
assumed to be an indophenol dye, a series of these compounds was syn-
thesized. TWo pathways for indophenol synthesis were investigated:
the Liebermann nitroso reaction and the procedure outlined by Gibbs,
92.31..

In the Liebermann nitroso reaction, a phenol is reacted with
nitrous acid in the cold to produce the nitrosophenol. In the pre-
sence of excess phenol and strong mineral acid, the nitrosophenol
and phenol couple to form the acid sulfate of indophenol, as shown

below. Dilution with cold water precipitates the free indophenol.

‘+HONO —°°-> ‘e—S *3

14°44 ”HM: m,-

14

In the Gibbs procedure, a phenol is reacted with a quinone-
N—chloroimine in basic media to form an indophenol as the sodium

salt.

 

+ O 2NoOlL; _C N... NJ+N¢CI+2H20

 

 

 

Since the sodium salts of indophenols were found to be much
more stable than the free indophenols from the Liebermann nitroso re-
action, the Gibbs procedure proved to be the most satisfactory method
for indophenol synthesis.

In general, the 4-aminophenols necessary as starting material
for indophenol synthesis were not available, and so a series of com-
mercially available substituted phenols were 4-nitrosated and then
reduced to the desired 4-amino compounds. Nitration of phenols and
subsequent reduction to the amine proved less satisfactory for two
reasons. First, nitration gives several products which must be se-
parated, while nitrosation yields the 4-substituted products almost
exclusively. Second, the nitrosophenols are reduced under much

milder conditions than the corresponding nitrophenols.

Synthesis of 4-nitroso alkylphenols

 

For synthesis of 4-nitroso alkylphenols, 0.20 mole of the
alkylphenol was added to a 400 m1 beaker containing 50 ml of ethanol

and 50 ml of concentrated hydrochloric acid. An off-white slurry

15

formed after the reaction mixture was cooled to -2°C in a salt-ice
bath. Next, 0.25 mole of pulverized sodium nitrite was added grad-
ually while the reaction mixture was hand stirred with a glass rod.
After about half of the sodium nitrite was added, the reaction mix-
ture thickened considerably and a yellow-green sheen was observed

on the sides of the reaction vessel. Fifteen minutes after the ad-
dition of sodium nitrite was complete, the reaction mixture was
poured into a l e beaker containing 700 ml of ice cold water. A co-
pious, fluffy precipitate developed, which was suction filtered and

washed well with cold water.

Synthesis of 4-nitroso chlorophenols

A 500 ml beaker containing a solution of 0.03 mole of the
chlorophenol, 2 g of sodium hydroxide, and 10 g of sodium nitrite in
300 ml of water was cooled to 15°C in an ice bath. Next, 25 ml of
33% v/v sulfuric acid, cooled to 4°C, were added gradually. When
the addition of sulfuric acid was complete, the reaction mixture was
transferred to a refrigerator. After two hours, a large mass of
burnt-orange needles had formed, which were filtered and washed with
a little ice cold water. The mother liquor was returned to the re-
frigerator, and after eight hours, a second crop of needles was re-

covered.

2-metbyl-4-nitrosgphenol

 

The reaction of 22.0 g (0.20 mole) of 2-methylphenol gave
22.0 g (80.2%) of the crude nitroso-phenol. Recrystallization from
water gave white needles, MP = l34-135°C-d [lit (12):l34-135°C-d].

16

2,6-dimethyl-4-nitrosophenol

 

The reaction of 24.5 g (0.20 mole) of 2,6-dimethylphenol
gave 18.7 g (61.8%) of the crude nitrosophenol. Recrystallization
from benzene gave gold plates, MP = 17l-l73°C-d [lit (44): 171°C-d].

3,5-dimetby1-4-nitrosophenol

 

The reaction of 10.0 g (0.08 mole) of 3,5-dimethylphenol
gave 6.5 g (53.7%) of the crude nitrosophenol. Recrystallization
from benzene gave yellow plates, MP = 184-185°C-d [lit (28):
185-186°C-d].

3-metbyl-6-isopropy1:4-nitrosophenol

The reaction of 10.0 g (0.07 mole) of 3-methyl-6-
isopropylphenol gave 8.6 g (68.6%) of the crude nitrosophenol. Re-
crystallization from benzene gave white needles, MP = 162-163°C-d

[lit (12): l60-164°C-d].

2,3,5,6-tetramethyl-4-nitrosgphenol

The reaction of 30.0 g (0.20 mole) 2,3,5,6-tetramethy1phenol
gave 27.2 g (75.9%) of the nitrosophenol. Recrystallization from 50%
ethanol gave yellow needles, MP = 88-89°C-d [lit (40): 89°C-d].

2-chloro-4-nitrosophenol

 

The reaction of 4.0 g (0.03 mole) of 2-chlorophenol gave
2.5 g (52.9%) of the nitrosophenol. Recrystallization from 50% etha-
nol gave burnt-orange needles, MP = 147°C-d [lit (16): 148°C-d].

17

 

2,6—dichloro-4-nitrosgphenol

‘ The reaction of 4.0 g (0.02 mole) of 2,6-dichlorophenol gave
2.7 g (57.4%) of the crude nitrosophenol. Recrystallization from
benzene gave orange needles, MP = 162-164°C-d [lit (8): 160-164°C-d].

Synthesis of 5-nitroso-N-methylanthranilic acid

N—methylanthranilic acid (0.07 mole) was dissolved in 75 m1
of concentrated hydrochloric acid. The solution was cooled to 0°C
in a salt-ice bath, and 5 g of sodium nitrite were added gradually.
The reaction mixture was transferred to a refrigerator, and after
twelve hours the hydrochloride of 5-nitroso-N-methalanthranilic acid
had separated as a bright gold precipitate, which was filtered and
washed successively with 5 ml of cold ethanol and 40 m1 of ether.
The hydrochloride was transformed into the free base by trituration
with 0.05 mole of sodium bicarbonate and 300 ml of water. The blue-
green precipitate which formed was filtered and dried. The yield
was 9.5 g (75.4%). The crude product was recrystallized from 50%
ethanol to give emerald green crystals, MP = l9l-192°C-d [lit (20):
191°C-d].

Synthesis of 2-hydroxy-5-nitrosobenzoic acid

Crude S-nitroso-N-methylanthranilac acid (0.025 mole) was
added to 25 ml ice cold 10 N_sodium hydroxide. To this 150 m1 of ice
cold water were then added slowly. The temperature of the reaction
mixture was always less than 5°C. After twelve hours in a refrigera-

tor at 0°C, the reaction mixture had a deep redish-blue color. The

18

temperature was maintained at 0°C in a salt—ice bath, and ice cold
concentrated sulfuric acid was added drop-wise until a pH of 1 was
obtained. The blue-green precipitate which formed was filtered,
washed with several small portions of ice cold water, and dried in

a vacuum desiccator. The yield was 4.1 g (98.1%). The crude product
was recrystallized from benzene to give blue-green needles, MP =

161-162°C—d [lit (20): 162-163°C-d].

Synthesis of substituted 4-aminophenols

 

The 4-nitrosophenol (0.1 mole) was dissolved in 250 ml of
ammonium hydroxide solution prepared from 160 ml of water and 90 ml
of concentrated ammonium hydroxide. The deep yellow-brown solution
was suction filtered and transferred to a 400 ml beaker. Next, tank
hydrogen sulfide diffused through a sintered glass frit was vigorous-
ly passed through the solution. After five minutes, the reaction
mixture turned clear orange and a precipitate began to form. After
twenty minutes, the precipitate was copious and the hydrogen sulfide

was turned off. The reaction mixture was cooled to 0°C and filtered.

2-methyl-4-aminophenol

 

The reaction of 10 g (0.07 mole) of crude 2-methyl-4-
nitrosophenol gave 6.4 g (71.3%) of the amine. Recrystallization

from benzene gave white plates, MP = 174-176°C [lit (12): 175°C].

2,6-dimethyl—4-aminophenol

 

The reaction of 5 g (0.03 mole) of recrystallized 2,6-
dimethyl-4—nitrosophenol gave 3 g (66.3%) of the amine.

19

Recrystallization from benzene gave tan needles, MP = 138°C-d [lit

(44): 137-138°C-d].

3,5-dimethyl-4-aminophenol

 

The reaction of 6 g (0.04 mole) of crude 3,5-dimethyl-4-
nitrosophenol gave 5 g (91.1%) of the amine. Recrystallization from
benzene gave white plates, MP = 182-183°C—d [lit (44): 181.8-
183°C-d].

3-methyl-6-isopropy1-4-amingphenol

The reaction of 9 g (0.05 mole) of crude 3-methy1-6-
isopropyl-4-nitrosophenol gave 8 g (96.8%) of the amine. Recrystal-
lization from benzene gave white needles, MP = l78-l79°C [lit (12):
178-179°C].

2,3,5,6—tetramethyl-4-amin9phenol

 

The reaction of 6 g (0.03 mole) of crude 2,3,5,6-tetramethyl-
4-nitrosophenol gave 4.7 g (86.2%) of the amine. Sublimation gave a

white powder, MP = 179°C-d [lit (38): 177-178.5°C-d].

2-chloro-4-aminophenol

 

The reaction of 5.1 g (0.03 mole) of crude 2-chloro-4-
nitrOSOphenol gave 2.9 g (67.3%) of the amine. Recrystallization

from benzene gave white needles, MP = 151°C-d [lit (12): 153°C-d].

2,6-dichloro-4-aminophenol

 

The reaction of 2.4 g (0.01 mole) of crude 2,6-dichloro-4-

nitrosophenol gave 1.6 g (71.9%) of the amine. Recrystallization

20

from benzene gave tan needles, MP = l68-l70°C-d [lit (12): 165-
166°C-d].

2-hydroxyr5-amino benzoic acid
When 2.3 g (0.02 mole) of 2-hydroxy-5-nitroso benzoic acid

was reacted, no product was isolated.

Synthesis of quinone-N-chloroimines

The aminophenol to be reacted was dissolved in 100 m1 of 5%
v/v hydrochloric acid (2 0.6 N) and cooled to )°C. This solution
was then added gradually to a three- to five-fold excess of sodium
hypochlorite solution also cooled to 0°C and containing a few pieces
of crushed ice. A bright yellow to orange precipitate formed im-
mediately. After five minutes, this precipitate was filtered,

washed with cold water, and dried in a vacuum desiccator.

Quinone-N-chloroimine

The reaction of 5.5 g (0.05 mole) of commercial 4-
aminophenol gave 5.7 g (81.4%) of the N-chloroimine. Recrystalliza-
tion from ether gave yellow needles, MP = 87-89°C-d [lit (45):
88°C-d].

2-methy1quinone-N-chloroimine
The reaction of 1.2 g (0.01 mole) of 2-methyl-4-aminophenol
gave 1.4 g (90%) of the N-chloroimine. Recrystallization from ether

gave yellow needles, MP = 99-100°C-d [litz no data].

21

2,6-dimethylquinone-N-chloroimine

 

The reaction of 2.4 g (0.2 mole) of 2,6-dimethy1-4-
aminophenol gave 2.1 g (61.9%) of the N-chloroimine. Recrystalliza-
tion from benzene gave bright yellow needles, MP = 69.5-7l°C-d [litz
no data].

3,5-dimethquuinone-N-ch1oroimine

The reaction of 4.6 g (0.03 mole) of 3,5-dimethyl-4-
aminophenol gave 3.9 g (69.7%) of the N-chloroimine. Recrystalliza-
tion from ether gave yellow needles, MP = 175-177°C-d [litz no data].

3-methyl~6-isopropylquinone-N-chloroimine
The reaction of 5.0 g (0.03 mole) of 3-methy1-6-isopropy1-4-
aminophenol gave a bright orange oil. Repeated attempts to crystal-

lize the product failed.

2,3,5,6-tetramethylquinone-N-chloroimine

The reaction of l g (0.06 mole) of 2,3,5,6-tetramethyl-4-
aminophenol gave 0.7 g (59.0%) of the N-chloroimine. The product ap-
peared dirty and no suitable recrystallization solvent was found,

MP = 77-80°C-d [litz no data].

2-chloroquinone-N-chloroimine

The reaction of 5.5 g (0.04 mole) of 2-chloro-4-aminophenol
gave 3.5 g (52.3%) of the N-chloroimine. Recrystallization from
ether gave orange needles, MP = 85-86°C-d [lit (24): 86-87°C-d].

22

2,6-dichloroquinone-N-chloroimine

 

The reaction of 8 g (0.04 mole) of commercial 2,6-dichloro-
4-aminophenol gave 3.3 g (35.2%) of the N-chloroimine. Recrystal-
lization from ether gave gold needles, MP = 64-45°C-d [lit (25):
66-67°C-d].

Synthesis of indophenols

A 100 ml beaker containing 30 ml of water and a few pieces
of ice was cooled to 0°C in a salt-ice bath. Next equimolar quanti-
ties of phenol and quinone-N-chloroimine were added. Then enough
3 N_sodium hydroxide to provide a 2:1 molar ratio plus 10% was added
slowly while the temperature of the reaction mixture was maintained
at 0°C. The reaction mixture immediately darkened and passed
through several shades of green before finally attaining a deep blue
color. The calculated amount of sodium chloride necessary to give a
near saturated solution was added, and the reaction mixture was
transferred to a refrigerator. After approximately four hours, the
irradescent sodium salt of the indophenol had precipitated. The
precipitate was suction filtered and dried in a vacuum desiccator

over calcium chloride.

Indophenol, sodium salt

 

The reaction of 4.7 g (0.05 mole) of phenol, 7.1 g (0.05
mole) of quinone-N-chloroimine, and 38 ml of 3 N_sodium hydroxide

solution gave 12.2 9 (109.4%) of the crude indophenol salt.

23

2,2'-dimethylindophenol, sodium salt

 

The reaction of 1.0 g (0.009 mole) of 2-methylphenol, 1.4 g
(0.009 mole) of 2-methquuinone-N-chlorimine, and 7 m1 of 3 N_sodium

hydroxide solution gave 3.5 9 (156.1%) of the crude indophenol salt.

2,6,21,6'-tetramethylindgphenol,,sodium salt

 

The reaction of 1.4 g (0.012 mole) of 2,6-dimethy1phenol,
2.0 g (0.012 mole) of 2,6-dimethquuinone-N-ch1oroimine, and 8 m1 of
3 N_sodium hydroxide solution gave 3.6 g (108.3%) of the crude in-
dophenol‘salt.

3,3'-dimethyl-6,6'-diisgpropylindophenol, sodium salt

 

The reaction of 0.8 g (0.005 mole) of thymol, 1.0 g (0.005
mole) of the orange oil thought to be 3-methyl-6-isopropy1quinone-
N-chloroimine, and 4 ml of 3 N_sodium hydroxide solution gave a

sticky, purple solid which could not be further purified.

3,5,3',5'-tetramethylindophenol,_sodium salt

When 1.4 g (0.012 mole) of 3,5-dimethylphenol, 2.0 g (0.012
mole) of 3,5-dimethquuinone-N-ch1oroimine and 8 ml of 3 N_sodium
hydroxide were reacted, the charactersitic blue color did not de-

velop and no product was recovered.

2,3,5,6,2',3',5',6'-octamethylindophenol

When 0.46 g (0.003 mole) of 2,3,5,6-tetramethylphenol, 0.60
g (0.003 mole) of the crude 2,3,5,6-tetramethquuinone-N-chloroimine
and 2 ml of 3 N_sodium hydroxide solution were reacted, the charac-

teristic blue color did not develop and no product was recovered.

24

2,2'-dichloroindgphenol, sodium salt

 

The reaction of 2.3 g (0.018 mole) of 2-chlorophenol, 3.2 g
(0.018 mole) of 2-chloroquinone-N-chloroimine, and 13 ml of 3 N_so-
dium hydroxide solution gave 6.1 9 (116.8%) of the crude indophenol
salt.

2,6,2',6'-tetrachloroindophenol, sodium salt
The reaction of 2.8 g (0.013 mole) of 2,6-dichloroquinone-N-
chloroimine and 8.5 m1 of 3 N_sodium hydroxide solution gave 4.9 g

(105.10%) of the crude indophenol salt.

Discussion

4-nitroso alkylphenols and 4-aminophenols were prepared by
the Kremers and Wakeman procedure (31), and 4-nitroso chlorophenols
by the Hodgson procedure (16). The synthesis of 5-nitroso-N-
methylanthranilic acid and 2-hydroxy-5-nitrosobenzoic acid followed
the procedure of Houben (20). Quinone-N-chloroimines and sodium
salts of indophenols were prepared by the procedures of Gibbs, gt,
21, (14).

Generally, reduction of nitrosophenols to aminophenols with
hydrogen sulfide proceded smoothly. It proved essential, however,
to filter the ammoniacal nitrosophenol solutions prior to hydrogen
sulfide reduction in order to obtain the aminophenols in high purity.
Due to the insolubility of 2,3,5,6-tetramethyl#4-nitrosophenol in
water, it was necessary to dilute the concentrated ammonium hydrox-

ide with ethanol rather than with water. After this reaction was

25

complete, addition of 50 ml of water enhanced the precipitation of
the aminophenol.

The oxidation of aminophenols to quinone-N-chloroimines also
proceded smoothly. Precooling both the hypochlorite and acidic
aminophenol solutions to 0°C and maintaining that temperature in the
reaction mixture was essential. The fact that some of the amino-
phenols were not completely soluble in 5% v/v hydrochloric acid so-

lution did not seem to hinder their complete reaction. For all
4-amino alkylphenols reacted, a commercial sodium hypochlorite solu-
tion, ChloroxR, was a satisfactory reagent. For 4-amino chloro-
phenols, however, Rashig's hypochlorite, prepared as described in
the next section, was a better reagent.

The difficulty in synthesizing symmetrical indophenols arose
in their separation and purification. In all cases, sodium chloride
had to be added to "salt out" the product, and invariably the sodium
salts of indophenols thus obtained were contaminated with sodium
chloride. This accounts for the greater than 100% yields reported
for the indophenols synthesized. Two purification procedures were
investigated. In one procedure, the crude product was dissolved in
water and resalted with about half the calculated amount of sodium
chloride necessary to give a saturated solution. This procedure
was fairly efficient, but of course some sodium chloride was still
a contaminant. In a second procedure, the crude product was dis-
solved in a small amount of anhydrous methanol and filtered to re-

move any insoluble material, including sodium chloride. A

26

fifteen-fold volume of ether was then added to reprecipitate the in-
dophenol. This procedure was not satisfactory, however, because in
many cases decomposition of the indophenol occurred in non-aqueous
solutions. No truly satisfactory method for isolating high purity
sodium salts of indophenols was devised. In the only published
values for the visible spectra of indophenols as their sodium salts,
the indophenols were prepared in solution by the reduction of a 4-
‘ aminophenol with ferricyanide to the quinone monoimine, which then
coupled with phenol in basic media to form the indophenol salt. In-
dophenols prepared in this research have absorption maxima in good
agreement with the published values, but their molar absorptivities
are 10-15% lower. This difference nay well be accounted for by
contamination of the indophenols with sodium chloride.

It was observed that when both the quinone-N-chloroimine and
phenol were 3,5-disubstituted, no indophenol was produced. Both 3,5-
dimethyl and 2,3,5,6-tetramethylquinone-N-chloroimines produced the
blue color charactersitic of indophenols when added to alkaline
phenol, so steric hinderance, rather than a lack of reactivity, was

assumed to inhibit the formation of these symmetric indophenols.
Preparation of Other Materials

Synthesis of sodiumghypochlorite
Sixty-three grams of sodium hydroxide were dissolved in
100 m1 of water, cooled, and poured onto enough crushed ice to give

a total weight of 1 kg. Next, tank chlorine diffused through a

27

glass frit was bubbled through the solution until the weight had in-
creased by 75 g. The resulting solution was pale yellow and still

contained a large amount of ice.

Synthesis of monochloramine

Twenty grams of ice, 70 m1 of ChloroxR (2 0.7 M_Na0Cl) and
100 m1 of ether were combined in a 500 m1 separatory funnel. After
about ten minutes, the contents of the separatory funnel had cooled
to 2°C, and 7.5 m1 of concentrated (2 7.4 M) ammonium hydroxide were
added. The reaction mixture was gently shaken, after which the
aqueous phase was discarded. Then 100 ml of ice cold distilled
water were added. After gentle shaking, the aqueous phase was trans-

ferred to a brown bottle and refrigerated.

Synthesis of sodium aguopentagyanoferrate

Solutions containing 20 g of sodium nitroferricyanide in 60 m1
of distilled water, 8 g of sodium hydroxide in 20 ml of distilled
water, and 7 g of hydroxylamine hydrochloride in 20 ml distilled
water were prepared individually and cooled to -2°C in a salt-ice
bath. The sodium hydroxide solution was added to the sodium nitro-
prusside solution, and then the hydroxylamine solution was added
drop-wise. The reaction mixture turned brownish-green and gas was
evolved. At no time did the temperature of the reaction mixture
rise above 0°C. After one hour, 300 m1 of ice cold ethanol was
added, and the aquoferrate precipitated as a brown resinous mass.

The liquid was decanted, and the solid was redissolved in about 60

28

m1 of ice cold distilled water. The aquoferrate was again pre-
cipitated by the addition of 200 m1 of ice cold methanol and suction
filtered through a medium glass frit. After several more dissolu-
tions with distilled water and reprecipitations with methanol, the
aquoferrate was filtered, washed with a little ether, and trans-
ferred to a vacuum desiccator over concentrated sulfuric acid. The

dry product was a gold powder.

Discussion

Sodium hypochlorite was prepared by the procedure of Raschig
(32). The procedure was straightforward but required a rather cum-
bersome, large-capacity, double-pan balance to monitor the weight
changes in the reaction mixture. When used for qunione-N-chloroimine
synthesis, this reagent was prepared just prior to its use and still
contained some crushed ice.

Monochloramine was prepared by the procedure of Kleinberg,
gt,_a1, (21). The concentration of each batch, as determined by
thiosulfate titration, varied slightly but was usually about 0.06 M,
Due to the instability of this solution, it had to be prepared
freshly every few days.

Sodium aquopentacyanoferrate was prepared by the procedure
of Hofmann (17) as modified by Asperger, gt,_al, (l). The tempera-
ture had to be maintained at or below 0°C throughout the synthesis
and for the first few recrystallizations to prevent the formation of
iron oxides. Due to the resinous nature of the product in prelimi-
nary precipitations, decantation, rather than filtration, for the

removal of the supernatant liquid was more expedient.

V. NON-KINETICS EXPERIMENTS
UV Spectrum of Hypochlorous Acid

The ultraviolet absorption spectrum of hypochlorous acid as
a function of pH is shown in Figure 1. Free hypochlorous acid is

characterized by a single band at 233 nm with a molar absorptivity

of 134 2 mole-1cm']. Hypochlorite ions display a broad band with

maximum absorption at 293 nm and a molar absorptivity (above pH 11)
1cm-l

of 358 2 mole" Under conditions of constant pH, solutions of

hypochlorite ions obey Beer's Law in the concentration range of 10'5
to 10"3 molar. The ultraviolet absorption spectrum of sodium hypo-

chlorite buffered at pH 11.5 is shown in Figure 2.
UV Spectrum of Monochloramine

Solutions of monochloramine display a broad absorption band
between 200 and 300 nm with Amax at 243 nm and a molar absorptivity
of 460 z mole'1cm']. Dilute solutions of monochloramine decompose
slowly but are relatively stable in the pH range of 7.5 to 11.5.
The ultraviolet absorption spectrum of monochloramine buffered at

pH 10.0 is shown in Figure 3.

29

30

 

0-5 F no

0.4

0.3

0.2

O.|

 

 

 

 

0-0 1 1 l 1 n I
220 240 260 280 300 320 340

Mnm)

Figure 1: UV Spectrum of Hypochlorous Acid as a
Function of pH.

LC)
(>13
(3J3
()3?
CXGB
(345
(All

(3J3

(X23.

Ct!

()1)

 

 

 

 

 

IIIIIIIIIII

220 250 280 310 340

Figure 2:

Mnm)

UV Spectrum of
Hypochlorite Ions

31

 

LC)

()1? -

CRIB-

(337 ~

(LES

(315

(3J4 -
(3J3 -
10.2?-

(D.l -

 

 

 

 

()1)

220 250 280

Figure 3:

 

I l' I I] I

3IO 34o
Mnrn)

UV Spectrum of
Monochloramine

32

Visible Spectra of

Pentacyano Iron Complexes

The absorption maxima of the pentacyano iron complexes used
in this research are listed in Table 1. When monochloramine is
added to aqueous solutions of any of these complexes, a rapid shift
in the wavelength of maximum absorption to 392 nm occurs. An iden-
tical shift occurs when hypochlorous acid is added to any of these
complexes. Literature values for Amax and log Emax are shown in

parentheses.

TABLE 1
Absorption Maxima and Molar Absorptivities

of Some Pentacyano Iron Complexes

 

 

Compound Xmax(nm) log emax
NaZEFe(CN)5N0] 398 1.30
(398) (1.31)
Na4[Fe(CN)5N02] 398 2.92
(398) (2.99)
Na3[Fe(CN)5H20] 415 2.29
(428) (2.40)

Visible Spectra of Indophenols

The visible absorption spectrum of 2,6,2',6'-

tetramethylindophenol as a function of pH is shown in Figure 4 and

33

 

 

 

 

 

 

0.9 P "'°
0.8 '-
07 L
0.6 *-
90
0.5 -'
A
0.4-
0.3 -
7.0
0.1 *-
6.0
5.0\.
0-0 In I I I T I I I I I.
460 500 540 580 620 660 700

X (nm)

Figure 4: Visible Spectrum of 2,6,2',6'—tetramethylindophenol as a
Function of pH.

34

is typical of indophenols. The free indophenol which predominates
in acidic solutions is characterized by a broad band between 400 and
500 nm, while the indophenolate ion, which predominates in basic so-
lutions, displays a more intense band between 600 and 700 nm. The
wavelengths of maximum absorption and the molar absorptivities of
indophenols synthesized in this research are listed in Table 2 along
with the literature values, which are shown in parentheses. Values
of Amax agree quite favorably with the literature, and the approxi-

mately 10% lower values for Emax are attributable to sodium chloride

impurities, as discussed in Chapter IV.

TABLE 2
Absorption Maxima and Molar Absorptivities of

Some Indophenols and Their Sodium Salts

 

 

Indophenol Indophenolate

Substituents Amax log emax Amax log emax
none 495 3.40 635 3.95
(496) (3.83) (637) (4.50)
2,2'-dimethyl 495 3.64 648 4.15
(495) (3.97) (648) (4.52)
2,6,2',6'-tetramethyl 493 3.50 658 4.14
~ (494) (3.82) (662) (4.47)
2,2'-dichloro 508 3.04 650 4.41
(no data) (no data) (662) (4.47)
2,6,2',6'-tetrachloro 515 3.15 668 4.02

(no data) (no data) (no data) (no data)

35

Formation of Indophenols

in Aqueous Solution

When solutions of monochloramine, pentacyanoaquoferrate, and
phenol buffered at pH 11.5 were mixed, the blue complex which formed
displayed a visible absorption spectrum identical to that of indo-
phenol. As the molar ratio of pentacyanoaquoferrate to monochlor-
amine approached l:l, the wavelength of maximum absorption remained
constant while the absorbance approached a maximum. Molar ratios of
pentacyanoaquoferrate to monochloramine exceeding 1:1, however, re-
sulted in progressively increasing hypsochromic shifts in the wave-
lengths of maximum absorption which approached 700 nm. The same
trends were observed when phenol in the reaction mixture was re-
placed with methyl- or chloro-substituted phenols. The data from
this experiment are presented in Table 3. A similar experiment was
performed in which ammonia and a five-fold molar excess of sodium
hypochlorite was used in place of the monochloramine solution. Data
from this experiment are presented in Table 4. Both data sets are in
excellent agreement.

These data indicate that only one mole of pentacyanoaquofer-
rate is required for each mole of indophenol produced. Furthermore,
pentacyanoaquoferrate in excess of this stoichiometric concentration
reacts with indophenol or one of its intermediates to form an uni-
dentified complex. This complex is not stable, however, and after
several days its green color fades into blue, and the visible ab-

sorption spectrum again corresponds to that of indophenol. The fate

36

TABLE 3
Effect of Aquopentacyanoferrate Concentration on

the End Product of the Berthelot Reaction

- -fi—

 

 

 

3..
_ [Fe(CN) H 0]
[Fe(CN)5H20]3 [HZNCl] 5 2 max A* at Amax
[HZNCl]
1.05 x 10'5 8.52 x 10"5 1.23 x 10" 635 1.076
5.26 x 10"5 8.52 x 10'5 6.17 x 10'1 635 1.680
1.05 x 10‘4 8.52 x 10'5 1.23 641 1.520
5.26 x 10"4 8.52 x 10'5 6.17 688 1.348
1.05 x 10"3 3.52 x 10‘5 1.23 x 101 700 1.426
*1 cm cells
TABLE 4

Effect of Aquopentacyanoferrate Concentration on
the End Product of the Berthelot Reaction

in the Presence of Hypochlorite Ions

 

-_.._.-..—.._. .-.._...~-.—._ “—_. —.—_-._ 7...... - ._____——_.._ _._
m.“ .-——..-—.--—. :— _.—._. -———_—.—.._..— -—. .——__-...—_ --— . .—......—.-.-._._ ._ .. ___

 

 

Fe CN H 0 3' A+ at
[Fe(CN)5H20]3- [NH3] [-OCIJ E EH2;:]§*] max Amax
1.05 x 10'5 2.47 x 10'4 1.49 x 10'3 4.25 x 10'2 633 1.596
5.26 x 10'5 2.47 x 10"4 1.49 x 10‘3 2.13 x 10" 635 3.266
1.05 x 10'4 2.47 x 10'4 1.49 x 10'3 4.25 x 10"3 635 3.560
5.26 x 10'4 2.47 x 10"4 1.49 x 10'3 2.13 675 2.614
1.05 x 10'3 2.47 x 10'4 1.49 x 10'3 4.25 695 2.926

*Total conversion of ammonia in solution to monochloramine is as-
sumed.

+l cm cells

37

of pentacyanoaquoferrate in this process was not investigated. It
should be noted, however, that when solutions containing synthetic
indophenol and a large excess of pentacyanoaquoferrate were mixed,

no shift in wavelength was observed.

Identity of the "Catalyst"

for the Berthelot Reaction

Methods of ammonia analysis in which sodium nitropentacyano-
ferrate (nitroprusside) or sodium nitritopentacyanoferrate is used
to catalyze the Berthelot reaction often produce colors with a
greenish tint, rather than the clear blues characteristic of indo-
phenols. No clear-cut correlation for the stoichiometric amounts of
these compounds necessary to produce the hypsochromic shift observed
for sodium aquopentacyanoferrate could be made, but in general, 100
to 150 times as much was required. This led to the hypothesis that
sodium nitropentacyanoferrate and sodium nitritopentacyanoferrate
were in equilibrium with sodium aquopentacyanoferrate. This hypoth-
esis was confirmed by the work of Swinehart and Rock (41), who re-
ported rapid conversion of the nitro complex to the nitrito complex
in alkaline solutions. After this initial reaction (I), they ob-

served that a second equilibrium (II) was established between the

[Fe(CN)5N0]2' + 20H" # [Fe(CN)5N02]4' + H20, req = 1.5 x 106 (I)
[Fe(CN)5N02]4' + H20 ‘7: [Fe(CN)5H20]3' + N02", Keq = 3 x 10'4 (II)

 

. .ll.
[II-I'll"!

38

nitrito and the aquo complexes. These observations compled with the
simple 1:1 stoichiometry observed between monochloramine and sodium
aquopentacyanoferrate in indophenol formation support the hypothesis
that the aquo complex is the actual catalyst for indophenol forma-

tion in the Berthelot reaction.

VI. KINETICS EXPERIMENTS

Formation of Monochloramine

The rate of formation of monochloramine as a function of pH
were obtained by mixing buffered, equimolar aqueous solutions of am-
monium chloride and sodium hypochlorite in the stopped-flow spectro-
photometer and recording the change in absorbance with time at
either 243 nm or 292 nm, which are the wavelengths of maximum ab-
sorption for monochloramine and hypochlorite ions, respectively. At
both wavelengths, the reaction was found to be second-order overall
and first-order each in hypochlorous acid and ammonia. The max-
imum observed rate constant occurred at pH 8.0. The magnitude of
the observed rate constants calculated from data collected at 243 nm
was approximately twice as large as those calculated from data col-
lected at 292 nm for pH 3 10.0. This is because the band for mono-
chloramine at 243 nm is overlapped by the hypochlorous acid band at
233 nm, the dichloramine band at 230 nm, and the tailing edge of the
hypochlorite ion band, while the hypochlorite ion band at 292 nm is
not overlapped by any other species present in the reaction mixture.
For this reason, the data collected at 292 nm should give the best
estimate for observed rate constants. Data for the formation of
monochloramine when the molar ratio of ammonia to hypochlorous acid

was 1:5 and 1:10 were also collected at 292 nm over the same pH

39

40

range. Computer plots of data collected at pH 8.0 when solutions
of ammonia and hypochlorous acid with molar ratios of 1:1, 1:5, and
1:10 were mixed in the stopped-flow spectrophotometer are presented
in Figures 5, 6, and 7, respectively. On the plots, "x" represents
experimental points, "0" represents points calculated from the data
assuming general second-order kinetics, and "=" represents the over-
lap of experimental and calculated points. Each point plotted is
the average of ten separate determinations. Values for the observed
second-order rate constant agree within 10%. Data from these ex-
periments are presented in Figure 8, where the log of the observed
second-order rate constant is plotted against pH.

As indicated in Figure 8, the observed second-order rate
constants for monochloramine formation are highly pH dependent.
The fact that the maximum rate occurs at pH 8.0 indicates that the
species involved in monochloramine formation are ammonia and hypo-
chlorous acid rather than ammonium ions and hypochlorite ions. This
is most easily visualized with the aid of a distribution diagram for
ammonia, ammonium ions, hypochlorous acid, and hypochlorite ions as
a function of pH, such as the one presented in Figure 9. As can be
_seen in Figure 9, pH 8.0 corresponds to the maximum concentrations
of ammonia and hypochlorous acid relative to each other. If the re-
action involved ammonium ions and hypochlorite ions, the maximum ob-

served rate constant would occur at pH 9.5.

41

. -o_ x mo.F u “cmemaucw , -o_ x ¢S.N n 268088 “6 8=_8> , -o_ x so.“
u no» we maps» .5: «mm um wocmagomam mw mvxm Pmueggm> .m-o_ x em.m acmemcucp
m72 x em.m mipcm_2 pm m=_8> , -o_ Xem.m Hipemp 88 6=_8> .mmv as?“ m_ 85x8 P8p=o~waoz
.E 3.0. x me._ u .m-_ooU ecu .2 e-o_ x me._ u . mzzH .e: Nam u x .o.m n :8
.owpmm LwFOZ Fu— cw ku< maogopguoaxz ucm mwcoss< sage mcwEmgopgoocoz we cowpmsgom ”m wgzmwm

HF-

:uTa UU—wur_u upmfi..t:nn>r\q_?:$.¢ e oeseaaaoooereeoen-
_Uiii1411113111IHIIIIJIIIIXIIIIUIIIImIIIIUIIIIUIIIIUIIIIw11IINIIII!JJIIUIIII- 11111 xiiiouililwiIIIUIIII.
”in HLVr n
» >Lhnhnr» .r
— u,IHIH.I,. H
r pH!» 5
p u nun L. n
U stun“ u
w "MHCX .u.
_ 22." 1 ..
w ‘ Hun .
.— HHH .
U "all" u i.
» re” 5 .
b "C" o.
w H" p
» nun _
U nu - 1 1 11 -1-..M
F Hh.‘ w
p .H O
u H" o
_ up - w
w "Cy u
» ycx —
— cS. »
F I - - 1 xcn :11 - .~
» HS »
u Tu «
» u »
. n - m
w H" .
p u- h
n C u
h - I I 1.1m1 111 - 111 L.
_ u M
r us N
— .g h
v - 1 - p u
_ en. M
b c h
p C u
c - i - i 11- x0 - - w
v u u
_ x w m
2 p2 w
2 - 1 c- - a
h re p
... x 0
. u m
. .l,1; _ . . -1.-- ".2
p r»
2 on
.m1111w11110111101i118 11111 01-11811inviiiaviiIIJiiiiwiiiimiiiixiiiimiiiimiiiiuiiii0111101111m1111m1111~

42

. -33 x 33.3 - pcmsmaoc3 m -33 x “3.3 1 533333 38 3:333 ,_-o_ x 33.3
n no» we m=_mw .Ec Nmm an mucmagomam we mpxm pmuwugm> . 103 x mm.~ u pamEmLucp
"Niop x Fo.~ m1ugmwg um mspm> m io_ x om.o m1u$m3 pm o33m> .mmv wavy m3 mpxm FmpcoNPLo:
.E 3-33 x 33.3 n .m-333U .3:3 .E 3-33 x 33.3 - . mzzH .2: N33 u x .o.m u :3
.oppum LmFoz may :3 u30< mzoLo—gooaxz use wwcoes< seem mcpemgo350ocoz yo cowmeLom no mgzmwm

9.0 «110400! mu each m muchnumh Juc1\ntz tattooeeeeeectoeoee-

 

—
O cox
Cnx
1-1 Cnxx
— ccuxx
OCIx _-
an
.-- I 1 ouli -
3 x13 .
w xuoc ..
_ - xxpc
— xxcc
u xqu
» xllc
.Wi xlo :
w xnc
xfln
— xnc .
W xnn ;
an r
u"
"no r
u 70 .
h “H V
m- il." _-
m
— KuC
XHC _-
W e
* "NX m
en _
V nux w
3 on 3
* "C W
u xmo m
w 3" u
— "C n
_ W
b "n
~ "u M
V "x. U
— Cux u
— Onxxu
. c _
» O~
-m-U'UWUIUUWUUUUm-UUUUUUUUwUI-UUWUUUUm'UUUVUUUUUUU-IIWUUUUWUUUUVD'5'“-nuUUmUDUDUUUn-UWUI-UUUUUUWUUUUWUUUUn

43

u no» pm «33¢»-
m_-o_ x No.N m.3;mmg 33 m=_3>
2 x 33.3 a

.m

o_ x Nm.3 u 333233333 ”w-op x 33.3 n
.5: «am pm mucmagomnm 3

op x omwm mlpmwp no mspm>

mwxm Fuowpgm>

soupon pm m33m> «3.03 x mo.m
.muop x mm.3 u ucmsmgocw
.«mv mswp m3 mvxm Fmpcongo:

ommzzU .5: Nam u 3 .o.m u =3

.3 -
.owpmmgmmo

~m----m----

Mu now""tubululuuunuuuunuuuuuuunuCOOIXxx
xxx Cccuuuuflnuunno x

I
l
l
S
l
I

'— u-b-o— h-m DID-Inna“ Hunt-Hr unto-Ir- tuba—any our- Int-cu onuo-om‘fu-nut-o—nm I-lO-rt-IO-iu‘ Hun-hut-
1
I
V

w-1--m----m--'-w-'-lr'--‘V'-’-r-..-'m-'--w-|.II-all-"-u---aim-'"W'---m'--'m--'-w----w-"Im---'m"--m'-.

v--1.xm----m---‘W'---m----m----m---.llmu.jul' .111'm---”

- FUOH 3:3 .3 -33 x 33.3 n
z opu_ 33 3333 3333333333>=.w

cm upcoea< sogw mcwsmgopguocoz mo cowumELom

.0 watt mg

I.-- -'--m-'--

quCInII

”n mgsmwm

CJOW c— Whhnmwh Aomz\n12 tn...0¢o¢¢.t¢toot¢¢o

I---m---'m...,"m----m--1-M

ha Int-«pa ”U" *O-O-IU‘O-‘O-ONHUY nwwmI—Ho-o— mo—amWo—ounnm—u—Ifiwu‘ mumwu

IO!

44

 

 

log kObS

 

 

 

l I l I

7.5 8.0 9.0 I0.0 l LO l2.0

 

pH

Figure 8: Log of observed Second-order Rate Constants for the
Formation of MonochIoramine versus pH at 292 nm (.)
and 243 nm (I).

45

 

 

 

 

 

 

 

Figure 9: Distribution Diagrams for Hypochlorous Acid-Hypochlorite
Ions and Ammonia-Ammonium Ions as a Function of pH.

46
Decomposition of Monochloramine

Monochloramine and hypochlorous acid are mutually unstable,
as evidenced by the decrease in absorbance at 243 nm and 292 nm when
their dilute aqueous solutions are mixed. The effect of pH on the
stability of these species was determined by mixing 1.45 x 10'4 ll

monochloramine and 1.49 x 10'3

F hypochlorous acid, buffered at the
same pH, in the stopped-flow spectrophotometer. The pH values
chosen for these experiments were 8.5, 10.5, and l2.0. All experi-
mental data presented are the average of four runs.

In the first set of experiments, the change in absorbance at
243 nm with time was recorded. A "blank" in which buffered hypo-
chlorous acid was mixed with buffer containing no monochloramine was
also performed at this wavelength so that the contribution of hypo-
chlorous acid and hypochlorite ions to the absorbance could be sub-
tracted. In the second set of experiments, the change in absorbance
at 292 nm with time was recorded for identical solutions. No blank
was necessary for these experiments since hypochlorite is the only
species in solution which absorbs significantly at this wavelength.

Neither set of data could be fitted to overall second-order
kinetics. Furthermore, the data could not be fitted to either
second-order in monochloramine, or second-order in monochloramine
and first-order in hypochlorous acid. This indicates that the de-
composition of monochloramine is a multi-step process. Some possi-

ble pathways for monochloramine are outlined below. Reaction (I) is

not expected to occur above pH 7.5 (7). Reactions (II) and (III),

47

HZNCl + HOCl ———> HNCl2 + H20 (1)
HZNCl + HOCl ——) HZNOH + 2c1' (II)
'OH - (III)

however, were reported by Yagil and Anbar (49) as side-reactions in
synthesis of hydrazine when the concentration of monochloramine was

less than 1043

H, The ammonia formed in reaction (III) would react
with any hypochlorous acid remaining in the solution to form addi-
tional monochloramine, which would further complicate the kinetics
of the overall decomposition reaction.

Plots of absorbance at 243 nm (corrected for absorbance of
hypochlorous acid and hypochlorite ions) and 292 nm versus time are
presented in Figures l0 and ll, respectively. Both data sets are in
excellent agreement. The sharp rise in absorbance in Figure 10 with
a maximum occurring at 35, 65, and 305 for pH values of 8.5, 10.5,
and l2.0, respectively, are corresponded by sharp decreases in ab-
sorbance in Figure l0 at the same time intervals. Initial changes
in absorbance indicate the formation of monochloramine from small
quantities of ammonia in the monochloramine solution. The steepness
of these changes reflects the effect of pH on the rate of monochlor-
amine formation reported in the previous section. Furthermore, the

data show that monochloramine and hypochlorous acid are relatively

stable in the pH range of 10.5 to 12.0.

48

 

 

 

p : i d J—M;_¢
.IO-
A -
.08-
.06-
-04 l I l I l

 

 

 

20 so IOO ' I40 ' Ie'o ' 22'o ' 260 300

Time (s)

Figure 10: Decomposition of Monochloramine as a Function of Absor-
bance at 243 nm and Time at pH Values of 8.5 (I),
l0.5 (0), and l2.0 (A).

 

 

 

 

 

0.56 ‘-

o.54 - — fl
A .J
0.52 -
0.50-

I I I I I I I T I r I I 7 I T

20 so IOO I40 :60 220 260 zoo

Time (s)

Figure ll: Decomposition of Monochloramine as a Function of Absor-
bance at 292 nm and Time at pH Values of 8.5 (I),
l0.5 (0), and l2.0 (A).

49

Mechanism and Kinetics

of Indophenol Formation

The mechanism and kinetics of indophenol formation from am-
monia, hypochlorous acid, sodium aquopentacyanoferrate, and phenol
are complex. Experiments in which buffered solutions containing
phenol, 2—methylphenol, or 2-chlorophenol plus sodium aquopentacy-
anoferrate and monochloramine were mixed in the stopped-flow spec-
trophotometer were performed. The change in absorbance with time
was recorded at 635 nm and 650 nm, which corresponds to the wave-
lengths of maximum absorption for indophenol, 2,2'-dimethyl-, and
2,2'-dichloraoindophenol. Representative data are presented in
Figure 12. Brief attempts to fit these data to rate laws were un-
successful. Despite this fact, the following mechanism is proposed.
It is based on the results of this research, related research of

other workers, and a fair measure of intuition and speculation.

NH3 + HOCl ——+ HZNCl + H20 (IV)

HZNCl + [Fe(CN)5H20]3' 3:2 [Fe(CN)5H2NCl]3' + H20 (V)
91

[Fe(CN)5H2NCl]3' + .014 3-=-*—— [Fe(CN)5-N-.—0H] + 3H20 (VI)

Cl

C
[Fe(CN)5-l!l—.—OH]3' + H20 x—"—‘—— [Fe(CN)5H20]2" + :Il

Cl
l -
H0-.-N: + .-0H + 30H —)
0=”=N--O- + C1- + 3H20 (VIII)

1

‘ou

(VII)

III IIIIIIIIIIII III I .

I I IIIIll

50

 

.400

.300

.200

..|00

 

.000

Figure l2:

/

2!) EH)

/

J

KDC)

l

h4()

 

l l I

ISO 220 260 300

seconds

Rate of Formation of 2,2'-dimethylindophenol (A),
Indophenol (I), and 2,2'-chloroindophenol (0).

51

Reaction (IV) corresponds to the formation of the first in-
termediate in the overall reaction. The rate of this step is gov-
erned primarily by the pH of the reaction medium, as shown in the
previous section. Reaction (V), which corresponds to the formation
of the second intermediate, is supported by experiments discussed in
Chapter V. Reactions (VI) and (VII) are speculative, but similar
reactions between sodium nitropentacyanoferrate and acetophenone re-
ported by Holfe and Swinehart (48) lend some support. Reaction
(VIII) corresponds to the formation of indophenol and is supported

by the work of Corbett (9) and of Gibbs and Clark (l4).

Rates of Formation

of Several Indophenols

Referring again to Figure 12, it can be seen that the rate
of formation of 2,2'-dimethylindophenol is greater than that of in-
dophenol and that the rate of formation of 2,2'-dichloroindophenol
is less than that of indophenol. These observations are easily
rationalized in terms of the inductive effects discussed in Chapter
II. Despite the fact that the rate of formation for 2,2'-
dichloroindophenol was the slowest, its equilibrium value for ab-
sorbance was always the highest. This is probably because its molar
absorptivity is the highest of the three. As an aside, 2,6-
dimethylphenol coupled much faster than 2-methylphenol, and 2,6-
dichlorophenol coupled much slower than 2-chlorophenol. The use of

these phenols was hindered, however, due to their low solubility in

52

water. Furthermore, 2,6-dimethylphenol was rapidly chlorinated by
hypochlorous acid, which resulted in yellow-orange oils. It was
interesting to note that two ortho substituents on a phenol exerted
effects greater than would be predicted by simply doubling the ef-
fect of one ortho substiuent. The relative rates of coupling for

these phenols are as follows:

CH3 CH3 Cl CI
‘OH >> ‘OH > .OH > ‘OH >> ‘OH
CH3 Cl

VII. A NEW AMMONIA METHOD

In this chapter, a method for ammonia analysis which evolved
from a mechanistic consideration of the Berthelot reaction is pre-
sented. The order of reagent addition has been reversed relative to
most procedures, and 2-chlorophenol was used in place of phenol.

Sodium aquopentacyanoferrate was used to catalyze the reaction.
Reagents

Ammonia standards

 

(A) 7.44 x 10'1 M; Dissolve 3.9797 g of ammonium chloride
(dried for two hours at l20°C) in lOO ml deionized distilled water.

(B) 7.44 x 10'4 11: Dilute 1 m1 of A to 1 9. with deionized
distilled water.

(C) 7.44 x 10'6 _11: Dilute 1 m1 of 13 to 100 m1 with deion-
ized distilled water.

(D) 7.44 x lO'5 M; Dilute 10 ml of B to lOO ml with deion-
ized distilled water.

When l ml, 3 ml, and 5 ml of solution C were diluted to 25

-7

ml, the concentrations of ammonia were 2.98 x 10'7 H, 8.93 x 10 fl,

and 1.49 x 10'6 fl, respectively. When 1 ml, 3 ml, and 5 ml of solu-

tion D were diluted to 25 ml, the final concentrations of ammonia

6 6 5

were 2.98 x 10' M, 8.93 x 10' M, and l.49 x 10' M, respectively.

53

54

Sodiumghypochlorite

 

(E) 7.44 x 10"2

R)

M; Dilute l0 ml of 0.744 M_sodium hypo-

to 100 ml with deionized distilled water.
2

chlorite (Chlorox
(F) l.49 x 10' M; Dilute l0 ml of solution E to 50 ml

with deionized distilled water.

Sodium aquopentacyanoferrate

(e) 4.14. x 10'3

 

M; Dissolve 0.03 g of Na3[Fe(CN)5H20]'H20
25 ml of deionized distilled water.

4

(H) l.66 x lO' M; Dilute l ml of solution G to 25 ml with

deionized distilled water.

2-chlorophenol

 

(I) 1.06 M; Dissolve 2.3 g of 2-chlorophenol in 20 ml of
95% ethanol.
(J) 1.05 x 10“ 1_1_: Dilute 5 m1 of solution I to so ml with

deionized distilled water.

When l ml each of reagents F, H, and J were diluted to 25 ml,

their final concentrations were 5.95 x 10'4'M, 6.62 x lo-6

3

M, and

4.24 x lO' M, respectively.

The Method

Into twelve 25 ml volumetric flasks, which had been rinsed
with l0% HCl and deionized distilled water just prior to use, were
pipetted the required amount of standards C and D to give ammonia
7 7 11.1.49 x 10'6 n.

concentrations of 2.98 x lO- M, 8.93 x l0"

55

—6 6 5

2.98 x 10 M, 8.93 x 10' M, and l.49 x 10' M_when diluted to 25
ml. Next, 15 ml of pH l0.5 buffer was pipetted into each flask, as
well as two other 25 ml volumetric flasks containing no ammonia,
which served as reagent blanks. Then 1 ml of reagent F was added
and the flask was swirled. After one minute, 1 ml each of H and J
were added in rapid succession, and the mixture was swirled, diluted
to the mark with deionized distilled water, and shaken. This pro-
cedure was repeated for each flask. The flasks were allowed to
stand for two hours at room temperature after which the absorbance
of each solution was measured in 5 cm cells at 650 nm against pH
l0.5 buffer. The results are presented in Table 5.

The method for ammonia analysis described by Solorzano (39),
which is in widespread use in water chemistry laboratories in this
country because of its high sensitivity, was used for comparison
with the proposed method. Solorzano's procedure used sodium nitro-
pentacyanoferrate as the catalyst. Standard solutions were prepared
identically to those used for the proposed method except l5 ml of
deionized distilled water was added to each flask instead of buffer.
After addition of reagents, the solutions were allowed to stand for

two hours, after which the absorbance of each solution was measured

in 5 cm cells at 640 nm. The results are presented in Table 6.

Discussion

 

The calibration curves for the Solorzano and the proposed

method are presented in Figure 13. Both obey Beer's Law in the

7 5

ammonia concentration range of 10' M;to 10' M, The slope and

56

TABLE 5
Data Obtained When Ammonia Was Analyzed

by the Proposed Method

 

 

[NH3] A650 AStd - Ab‘k s so (% REL. so)

0.00 .032 -

0.00 .034 -

2.98 x 10'7 .070 0.034 e .002 (5.98)
2.98 x 10‘7 .067

8.93 x 10'7 .126 0.096 s .005 (4.77)
8.93 x 10'7 .128

1.49 x 10'6 .191 0.162 e .006 (3.92)
1.49 x 10"6 .200

2.98 x 10‘6 .382 0.344 s .007 (2.06)
2.98 x 10’6 .372

8.93 x 10'6 1.022 0.086 e .004 (0.36)
8.93 x 10'6 1.017

1.49 x 10'5 1.533 1.494 e .008 (0.52)

-5

1.49 x 10 1.522

Data Obtained When Ammonia Was Analyzed

57

TABLE 6

by the Solorzano Method

 

 

wfi .- m -—.-——-——.-- -“1

 

[NH3] A640 .Astd - Ablk 3 so (% REL. so)

0.00 .048 -

0.00 .048 -

2.98 x 10'7 .073 0.024 e .002 (9.03)
2.98 x 10"7 .070

8.93 x 10‘7 .139 0.093 s .003 (3.04)
8.93 x 10'7 .143

1.49 x 10‘6 .202 0.144 e .014 (9.82)
1.49 x 10"6 .182

2.98 x 10‘6 .306 0.274 e .023 (8.39)
2.98 x 10'6 .339

8.93 x 10‘6 .885 0.865 e .040 (4.58)
8.93 x 10'6 .941

1.49 x 10‘5 1.326 1.344 e .093 (6.89)
1.49 x 10‘5 1.456

58

 

I.4 - / /
1.2 - /

1.0 e
0.8 —
0.6 -
0.4 —

()1? -

 

1 I 1 l 1 I 1 l 1 P 1 I 1 l

o 20 4o 60 80 ICC 120 140
7
[NH3] x l0

Figure l3: Calibration Curves for Ammonia as Determined by the
Solorzano (0) and Proposed (I) Methods

 

59

intercept obtained from a linear least squares fit of the data from

4 and -2.43 x 10'3,

the Solorzano and proposed methods are 9.69 x 10
and 1.11 x 105 and 2.28 x 10'3, respectively. This corresponds to a
12.7% increase in sensitivity of the proposed method over the Solor-
zano method. An examination of Tables 5 and 6, in which the aver-
age and standard deviation of absorbance for each set of standards,
shows that the precision of the proposed method is as much as ten
times better than the Solorzano method, eSpecially at higher con-
centration ranges. The precision of the proposed method increases
with increasing ammonia concentration, while the precision of the
Solorzano method is low and random for all concentrations. It is
interesting to note that the color produced in the Solorzano method
fades to green or yellow after twelve hours, while the color pro-
duced in the proposed method is stable for at least four days.

The small increase in sensitivity for the proposed method
relative to the Solorzano method may be largely due to the fact that
the molar absorptivity of 2,2‘-dichloroindophenol is slightly higher
than that of indophenol. The large increase in precision of the
proposed method relative to the Solorzano method, however, is at-
tributable to the fact that side reactions have been minimized in
the proposed method. For example, in the Solorzano method, reagents
used are in several thousand-fold excess of their stoichiometric
concentrations, and the hypochlorite reagent is added last. Under
these conditions, ammonia must compete for hypochlorous acid with

phenol and catalyst molecules, which are present in very large

60

excess. In addition, both the first intermediate and the final pro-
duct of the Berthelot reaction, monochloramine and indophenol, are
unstable in the large excess of sodium hypochlorite recommended by
Solorzano. In the proposed method, the hypochlorite reagent is
added first and in only five-fold excess of the stoichiometric
amount necessary to convert all the ammonia in the highest standard

to monochloramine.

7 5

In the concentration range of 10' M_to 10' M, the time re-
quired to convert 99.5% of ammonia present in solution to monochlor-
amine was calculated as forty to fifty seconds when an observed
second-order rate constant of 2 x 103 2 mole-lsec-1 was assumed and
the initial concentration of the hypochlorite reagent was 5.95 x 10"4
M, This means that the first intermediate in the Berthelot reaction
is formed before the other reagents are added so side reactions are
minimized.

Although the proposed method offers increased sensitivity
and precision, it is important to stress that real samples were not
analyzed. Calcium, iron, and magnesium present in fresh- and sea-
water would be precipitated at pH 10.5 causing serious interference.
The addition of sodium citrate or EDTA to the buffer, however,
should alleviate this problem. The proposed method could be easily

automated and after minor modification could become the preferred

method for trace ammonia analysis.

LIST OF REFERENCES

\OCDNQU'l-bw

10.
11.

12.

13.
14.

15.
16.
17.
18.

LIST OF REFERENCES
Asperger, S., Murati, I., and Pavlovic, D., J. Chem. Soc., 730
(1960).

Beckwith, P. M., Ph.D. Thesis, Michigan State University, East
Lansing, Michigan, 1972.

Beckwith, P. M., and Crouch, S. R., Anal. Chem. 44, 221 (1972).
Berthelot, M., Repertoire de Chemie Applique 1, 284 (1959).
Bolleter, N. T., et. al., Anal. Chem. 34, 592 (1961).

Borsook, H., J. Biol. Chem. 119, 481 (1935).

Chapin, R. M., J. Am. Chem. Soc. 51, 2112 (1929).

Chem. Abs. 44, 5213c (1950).

Corbett, J. F., J. Chem. Soc., 1502 (1970).

Crowther, A. B., and Large, R. 5., Analyst g1, 64 (1956).

Czech, F. W., Fuchs, R. S., and Antczak, H. F., Anal. Chem. 33,
705 (1961).

"Dictionary of Organic Compounds," 4th ed., Oxford University
Press, New York, 1965.

Dye, J. L, and Nicely, V. A., J. Chem. Ed. 48, 443 (1971).
Gibbs, H. D., Hall, W. L., and Clark, W. M., "Studies in Oxida-
tion Reduction: XIII. Preparation of Indophenols Which May Be
Used as Oxidation Indicators," U.S. Public Health Reports,
Suppl. No. 69 (1929).

Harwood, J. W., and Huyser, D. J., Water Res. 4, 501 (1970).
Hodgson, W., J. Chem. Soc., 866 (1932).

Hofmann, K. A., Ann. 312, l (1900).

Horn, 0. B., and Squire, C. R., Clin. Chim. Acta. 14, 185 (1966).

61

19.
20.
21.

22.

23.
24.
25.

26.
27.

28.
29.
30.
31.

32.
33.

34.

35.
36.
37.
38.

39.

62

Horn, D. B., and Squire, C. R., Clin. Chim. Acta. 12, 99 (1967).
Houben, J., et. al., Ber. 48, 2750 (1909).

Kleinberg, J., Tecotyky, M., and Audrieth, L. F., Anal. Chem.
88, 1388 (1954).

%uboc?insky, B., and Zalta, J., Bull. Soc. Chem. Biol. 88, 1363
1954 .

Mark, H. B., et. al., Anal. Chem. 41, 848 (1969).
Metcalf, W. S., J. Chem. Soc., 148 (1942).

Mikhailov, G. I., Inst. Pure Chem. Reagents, No. 16, 83 (1937)
translated in Chem. Abs. 84, 37078 (1940).

Morita, Y., and Ogure, K., Ippon Kagaku Zasshi 84, 816 (1963).
Morris, J. C., and Neil, I., "The Formation of Monochloramine
and Dichloramine in Water Chlorination," American Chemical
Society Meeting, Detroit, Mich., April 17, 1951.

Norris, R. K., and Sternhell, S., Aust. J. Chem. 18, 859 (1966).
Ohkuma, S., Yakugaku Zasshi 88, 493 (1960).
Ohkuma, S., Yakugaku Zasshi 88, 505 (1960).

"Organic Synthesis," Collective Vol I, p. 511, Wiley, New York,
1941.

Raschig, F. E., Angew. Chem. 88, 2065 (1907).

Reardon, J., Foreman, J. A., and Searcy, R. L., Clin. Chim.
Acta. 14, 403 (1966).

Roskam, R. T., and de Langen, 0., Anal. Chim. Acta. 88, 56
(1964).

Russel, J. A., J. Biol. Chem. 1§§, 457 (1944).
Scheurer, P. G., and Smith, F., Anal. Chem. 81, 1616 (1944).
Searcy, R. L., et. al., Clin. Chim. Acta. 18, 170 (1965).

Smith, L. I., and Irwin, W. B., J. Am. Chem. Soc. 88, 1041
(1941).

Solorzano, L., L. & 0. 8, 799 (1969).

40.

41.
42.
43.

44.
45.
46.

47.
48.
49.
50.

51.

63
Summerford, W. T., and Dalton, D. N., J. Am. Chem. Soc. 88,
1330 (1944).
Swinehart, J. H., and Rock, P. A., Inorg. Chem. 8, 573 (1966).
Thomas, P., Bull. Soc. Chim. 11, 797 (1912).

Van Slyke, D. D., and Hiller, A., J. Biol. Chem. 188, 499
(1933). '

Vaughn, W. R., and Finch, G. K., J. Org. Chem. 81, 1201 (1956).
Vemura, P. 1., and Abe, M., Bull. Chem. Soc. Jap. 18, 59 (1937).

Weast, Robert C., Ed., "Handbook of Chemistry and Physics,"
48th ed., Chemical Rubber Co., Cleveland, 1967.

Neil, W., and Morris, J. C., J. Am. Chem. Soc. 11, 1664 (1949).
Wolfe, S. K., and Swinehart, J. H., Inorg. Chem. 1, 1855 (1968).
Yagil, G., and Anbar, M., J. Am. Chem. Soc. 84, 1790 (1962).

Yamaguchi, R., Macheda, T., and Veki, M., Yakugaku Zasshi 88,
1534 (1969).

Yamaguchi, R., Suzuki, H., and Hirashima, H., Chem. Pharm. Bull.
18, 1866 (1970).