EPA/700-R-92-006
68-D9-0166
EACTIONS OF POTENTIAL ORGANIC WATER CONTAMINANTS
WITH AQUEOUS CHLORINE AND HONOCHLORAMINE
Versar Work Assignment No. 2-41
by
Dr. Frank E. Scully, Jr.
Department of Chemistry and Biochemistry
Old Dominion University
Norfolk, Virginia
and
Dr. William N. White
Department of Chemistry
University of Vermont
Burlington, Vermont
-------
TABLE OF CONTENTS
INTRODUCTION ... 1
WATER TREATMENT PROCESSES 1
NATURE OF REACTIONS OF CHLORINE-BASED DISINFECTANTS
WITH ORGANIC SUBSTRATES . 3
Reactive Species in Aqueous Chlorine and Chloramine . . 3
Aqueous Chlorine 4
Aqueous Chloramines 8
Nature of Reactions with Organic Substrates 12
Nucleophilic Substitution on Chlorine (Cl*
Transfer) 12
Relative Reactivity of Cl+ Donors (12) ;
Catalysis of Cl+ Transfer (14); Examples of
Cl+ Transfer Reactions (14)
Nucleophilic Substitution on N in NH2C1 (NH2+
Transfer) . 16
Nitrenes 16
Nucleophilic Reactivity of CIO" and NH2C1 17
Free Radical Reactions 17
Conclusion 18
SURVEY OF REACTIONS OF FUNCTIONAL GROUPS 18
Alkanes 19
Haloalkanes 19
Alcohols 19
Aqueous Chlorine 19
Chloramines 21
Monochloramine (22) ; Dichloramine (23);
Nitrogen Trichloride (24) ; Hypochlorous Acid
and Hypochlorite Ion (25)
Conclusion ..... 25
Ketones and Aldehydes 25
Aqueous Chlorine 25
Monochloramine 28
Conclusions 29
Amines 29
Aqueous Chlorine 29
Monochloramine 32
Conclusions 33
Amide Nitrogens ..... 33
Aqueous Chlorine 33
Monochloramine , ,-> 34
Conclusions .34
Aromatic Compounds j' 34
Aqueous Chlorine '34
Monochloramine 39
Conclusions 39
Heterocyclic Aromatic Compounds 40
Aqueous Chlorine 40
-------
Monochloramine 40
Conclusion 41
Alkenes and Alkynes 41
Aqueous Chlorine 41
Monochloramine 43
Carboxylic Acids 44
Aqueous Chlorine . 44
Simple Carboxylic Acids (44) ; Carboxylic
Acids with "Active Methylene" Groups (45)
Monochloramine 46
Conclusions 47
Carboxylic Esters 47
Aqueous Chlorine 47
Simple Carboxylic Esters (47); Carboxylic
Esters with "Active Methylene" Groups (48)
Monochloramine 50
Conclusions 5l'
Sulfur Compounds 51
1. Aqueous Chlorine 51
2. Monochloramine 53
Conclusion 54
Organophosphorus Compounds 54
Aqueous Chlorine 54
Monochloramine 56
Conclusion . 56
SUMMARY AND OVERALL CONCLUSIONS 57
REFERENCES 60
-------
REACTIONS OF POTENTIAL ORGANIC WATER CONTAMINANTS
WITH AQUEOUS CHLORINE AND MONOCHLORAMINE
I. INTRODUCTION
Unwarranted exposure of humans and animals to pesticides and
toxic substances can elicit adverse health effects. Of major
concern to the Office of Pollution Prevention and Toxics is the
presence of these substances in ground and surface waters used as
drinking water supplies. Based solely on available dose-response
^iata on adverse health effects of these compounds in humans or
animals, an exposure level can be identified that does not elicit
any significant long- or short-range harmful effects. However,
most drinking waters in the United States are prepared from natural
waters which have been subjected to a series of physical and
chemical processes designed to disinfect the water and make it
pleasing to drink. -Little is known of the effect these processes
may have on the removal or chemical transformation of these
compounds. Because of the especially reactive nature of chemical
disinfectants, many toxic substances may undergo transformations to
more toxic or less toxic substances.
Aqueous chlorine and monochloramine are the two most widely
used disinfectants of drinking waters in the United States. The
objective of this work is to review the available information on
the reactions of aqueous chlorine and monochloramine with various
organic functional groups in order to predict which classes of
chemical substrates might be most likely to undergo transformations
under drinking water disinfection conditions. For those reactions
for which kinetic data are available, the half-lives of substances
containing these functional groups will be estimated. Where
possible, known reaction mechanisms will be identified and
structure-reactivity relationships discussed. This overview of
organic functional group reactivity should enable the Office of
Pollution Prevention and Toxics to make assessments of the
likelihood of chemical transformations of various toxic substances
during drinking water treatment processes.
II. Disinfection Processes and the Nature of
Chemical Disinfectants
A. Water Treatment Processes
The major sources of drinking water in the United States are
groundwater, rivers, and freshwater lakes. Many of these waters
contain suspended particles, chemical components, and
microbiological agents which make them undesireable to drink
without prior treatment. Therefore municipal treatment systems
subject this raw water to combinations of physical and chemical
processes to make them acceptable to drink. A description of one
combination of these processes will illustrate the variety of
treatments which can affect the fate of a toxic substance in the
raw water. To one raw water a coagulating agent (either alum or
-------
ferric chloride) along with powdered activated charcoal is added to
the raw water as it enters the plant. The powdered activated
carbon adsorbs and removes a variety of organic chemicals. Alum
and ferric chloride hydrolyze in water to colloidal aluminum and
iron hydroxides. These hydroxides are charged and bind particles
(oppositely charged) such as clay particles and bacterial cells as
well as organic and inorganic ions in the water. At the same time
the hydroxides aggregate, become heavier and settle over several
hours in a settling basin. The clarified water is next filtered
through sand to remove particles that have not settled. At this
stage, the pH of the water is adjusted upward with lime, chlorine
is added as a disinfectant, and sometimes fluoride is added.
Utilities which employ monochloramine as the disinfectant add
chlorine and ammonia either simultaneously or sequentially.
Fpllowing disinfection, the treated water is generally held in a
clear well for several hours to allow time for the disinfectant to
take effect before it is allowed to enter the distribution system.
Not all utilities employ all of these processes, while others
utilize additional processes such as lime softening. The recent
Surface Water Treatment Rule [i] now requires all public water
systems that use surface waters and groundwaters under the direct
influence of surface water to disinfect the water. Some of these
systems may be required to practice filtration to ensure the
removal of Giardia lamblia and other protozoa, viruses, and
pathogenic bacteria. However, some utilities use only chlorine
disinfection.
Sufficient disinfectant is added to the water at the plant so
that there is at least a minimum concentration of disinfectant at
every tap in the distribution system. The actual concentration of
disinfectant in the water at the tap varies with the amount added
to the water at the treatment plant, the temperature of the water,
the reactions which might dissipate the disinfectant while it is in
the pipes, and the length of time the water has remained in the
pipes before it is withdrawn. Many states have established a
required minimum residual chlorine concentration of 0.2 mg/L free
chlorine or 1.0 mg/L combined chlorine (chloramines) at every point
in the distribution system. The U.S. Environmental Protection
Agency found that most of the drinking waters in a survey of 80
localities contained residual aqueous chlorine concentrations
between 0.4 and 2.8 mg/L (C12) with an average of approximately 2
x 10"5 M [2.]. In a recent position statement [1] the American Water
Works Association (AWWA) recommended a goal of a minimum of 0.5
mg/L free residual chlorine or a minimum of 2.0 mg/L combined
residual chlorine (as C12) maintained generally throughout the
distribution system.
Although a variety of disinfectants are used in drinking water
systems, aqueous chlorine and monochloramine are the most common.
For the purposes of this document, the reactions of these two
-------
disinfectants only will be considered.
Physical methods of disinfection, such as carbon adsorption,
coagulation, and filtration, may remove certain toxic substances.
However, it is the purpose of this document to evaluate only the
potential of chemical disinfectants to transform these substances.
To simplify the calculations in this document, a residual oxidant
concentration (either a free or a combined residual) of 0.71 mg/L
(as C12, 1 x 10"5 M) will be assumed.
Most utilities maintain a slightly basic pH throughout the
distribution system to avoid corrosion of pipes in the system. To
further simplify estimates of the reactivity of toxic substances
with disinfectants, the assumption will be made that a pH of 7.5 is
maintained throughout the distribution system.
The residence time of water in the pipes, the time between
when it leaves the plant and when it reaches the consumer, varies
with the size of the distribution system. Consumers living near a
treatment plant will receive water that has left the plant the same
day it is used. In large systems, the detention time may be
greater than five days. For the purposes of this document, the
typical distribution system will be assumed to have a 5-day
residence time. In any case, this time can affect the percent
conversion of a toxic substance in any reaction with a disinfectant
that has a half-life of days.
III. NATURE OF REACTIONS OF CHLORINE-BASED DISINFECTANTS
WITH ORGANIC SUBSTRATES
This review is concerned with how the water disinfectants,
chlorine and chloramine, react with organic compounds of various
functional types. The most common reactions result in substitution
(replacement of a hydrogen by chlorine), oxidation (usually
implying increased bonding of an atom to oxygen), and addition
(saturation of a multiple bond). Initial reactions are often
followed by others that result in carbon skeleton cleavage,
decarboxylation, elimination, etc. These follow-up reactions may
or may not involve additional disinfectant.
A. Reactive Species in Aqueous Chlorine and Chloramine
It is important in considering the nature of reaction of
organic compounds with aqueous chlorine and chloramine to have a
clear notion of the actual molecular or ionic species involved.
The reactive molecules in aqueous chlorine and chloramine seem
obvious: C1-C1 and C1-NH2. However, complex interactions of these
substances with water or with themselves provide a variety of
-------
entities in their solutions.
1. Aqueous Chlorine
When chlorine is added to water, it hydrolyzes:
C12 + H2O s x HOC1 + HC1 (1)
with an equilibrium constant K = 3.88 x 10"* M2 [4.]. However, the
product, HOCl, is a weak acid and the absolute concentration of
HOC1 is affected dramatically by changes in pH. HOCl has a pK. of
7.5 [5].
HOCl s ^ CIO' + H+ pKa = 7.5 (2)
At pH values above 7.5, the aqueous chlorine is primarily in the
form of CIO". At pH values below 7.5, HOCl predominates.
Consequently, there are a variety of chlorinating species in a
chlorinated drinking water, C12, HOCl, and CIO". The initial
amounts of each will depend on the initial concentration of
chlorine and the pH.
Typical chlorinated drinking water contains about 1 x 10"5 M
"free available chlorine", FAC, and 3 x 10"* M Cl" [6]. Under these
conditions and at pH 1, the actual C12 concentration is about 7% of
the total C12. As the pH increases to 7, this concentration falls
by tenfold per pH unit until it is 0.000006% of the total. Above
pH 7, the molarity of C12 decreases by 100-fold per pH unit. From
pH 1 to 7, hypochlorous acid accounts for almost all of the active
chlorine, but at higher pH's (>8) it molarity drops by about 10-
fold per pH unit. The concentration of hypochlorite ion is very
low at pH 1 (0.00003% of total chlorine). It increases 10-fold per
pH unit until pH 7, and above pH 8 almost all of the original C12
is present in this form.
Drinking water has a pH of about 7.5. At this pH, the
concentration of the various active chlorine species is given in
Table 1.
-------
Table l. Concentrations of Active Chlorine Species
in Typical Chlorinated Drinking Water
[Cl"] = 3.00 x 10"* M,
[FAC] = 1.00 X 10"5 M,
pH = 7.5
Species
C12
HOC1
ocr
cr
Concentration
I.
5.
5.
3.
27
00
00
05
X
X
X
X
10
10
10
10
-13
-6
-6
-4
(M)
When analyzing a water, a chemist measures the sum of all
these species and reports them as "free available chlorine"
(calculated as C12) . In this case, the "free available chlorine"
concentration, [FAC], which in a drinking water is approximately 1
x 10"5 M, may be represented as:
[FAC] = [C12] +' [HOC1] + [CIO"] = 1.0 X 10"5 M
(3)
At pH 7.5 the concentration of C12 is very low (though, as will be
shown below, not necessarily insignificant) and therefore
[HOC1] = [CIO"] = 5.0 x 10"B M
If a reactant ionizes in water, then its "effective
concentration" will change with pH. Consider, for example, the
reaction of HOCl with a toxic substance, TS:
HOC1 + TS
Products
where the rate expression for this reaction is:
rate = k[HOCl][TS]
and where k is the reaction rate constant. If TS reacts only with
HOCl and not with CIO", then the initial rate of reaction will not
depend on the [FAC], since at pH 7.5 [FAC] is twice that of the
actual concentration of HOCl. However, as HOCl reacts, the
equilibrium shifts and HOCl is rapidly regenerated. Consequently,
although the initial concentration of HOCl is less than [FAC], all
of the FAC may react as the equilibrium shifts to provide more HOCl
for reaction.
-------
To calculate the fraction of the protonated form of an acid
(e.g., the fraction of free available chlorine which is HOC1) the
following expression can be used:
fraction of protonated acid = a = 1 __— (4)
(1 + H-K"
At pH 7.5 the fraction of HOC1 (pKa 7.5) is
a = -1 . , . .— = 0.5
(1 + lo7-5'7-5)
Another way of representing the concentration of HOC1 is:
<*HOCI [FAC] = 5.0 x 10"6 M
and the rate expression becomes
rate = k btHocl [FAC] [TS]
Most reactions which take place in environmental systems do
not occur under conditions where the concentrations of reactants
are equal. Instead the concentration of one reactant is generally
much higher than that of the other. If the concentration of FAC is
at least ten times greater than that of TS, then [FAC] is essen-
tially unchanged from the beginning of the reaction to the end.
Since aHOC1 is constant at a given pH, the conditions are pseudo-
first order. Therefore,
rate = k'[TS]
where k1 = k aHOC1 [FAC]
The rate expression only describes the initial rate of reaction,
because after the reaction gets under way the concentrations of
reagents are lower than at the start. As it is, this expression is
not very useful. However, if it is integrated over time, then the
resulting expression can be used to calculate the concentration of
the limiting reagent [TS] at any time after the reaction is begun:
In [TS]0/[TS] = k't
In [TS] = -k't + In [TS]0
where k1 is the pseudo-first order rate constant. This last
equation describes a straight line (y = mx + b) where the slope is
-k1 and the intercept is In [TS]0.
A very useful equation relates the rate constant and the half-
life of a first order or pseudo-first order reaction:
-------
tI/2 = 0.693/k (5)
where k is the rate constant for a first order reaction or the
observed rate constant, k1, for a pseudo-first order reaction.
For an especially toxic substance, the half-life is not nearly
as significant a value as the time it takes for a disinfectant to
react with 90% of the compound. This is calculated from the
expression:
-90Z
= 2.3/k1 (6)
Until this point the reactions of C12 have been ignored
because rather insignificant amounts of it are present at pH 7.5
(only 0.0000013% of the total active chlorine). However, this does
not imply that C12 is not a reactive entity in these solutions.
Molecular chlorine is generally more reactive than HOC1.
Furthermore, since it can be produced from the other species by the
above equilibria, if it is consumed in a reaction, it will be
regenerated. If this happens fast enough, reactions of C12 could
explain the chemical behavior of aqueous chlorine solutions.
The mechanism of the conversion of HOC1 to C12 has been
studied by Eigen and Kustin [7].. They have proposed that an
intermediate, HOC12", is involved and that this intermediate is
transformed into C12 by two routes, one acid-catalyzed and the
other uncatalyzed.
HOC! + Cl" ^ x HOC12" Kx = 2 X 10"6 M"1 (7)
HOCV + H* s v H20 + C12 k2 = 2 X 1010 M^s'1 (8)
-2
k_2 = 11 s'1
K, = 1.8 x 109 M
HOC12" s s HO" + C12 k3 = 2 x 105 s'1 (9)
-3
k-3 = 1010
K3 = 2 x 10'5
The C12 produced in these reactions can react with a substrate TS.
TS + C12 > P k« (10)
Kustin and Eigen estimated the rate and equilibrium constants
for these reactions (noted above). These values can be used to
show that molecular chlorine may be an acceptable candidate for
some reactions of aqueous chlorine. Thus, a steady-state rate
expression for the above set of reactions can be derived. When the
conditions in Table 1 and the reaction constants are substituted
into this expression, it is found that the acid-catalyzed
decomposition of HOC12" (second reaction) is about 300 times slower
than the uncatalyzed path (third reaction). Therefore, the second
-------
step is not involved in typical chlorinated drinking water.
Interestingly, when both k4 and [TS] are large (k, [TS] > 10000
s"1) , the overall rate is independent of [TS]. This is because the
product-forming step is so fast that the total speed of the
reaction is controlled by how fast C12 is formed. The rate is
rate = K^HOCl] [Cl'J (11)
Inserting the rate and equilibrium constants and the raolarity of
chloride ion in drinking water gives
rate = k[HOCl] k = 1.2 x 10'* s'1 (12)
As an example, values of k* = 1010 M^s"1 and [TS] = 10"6 M would
result in kinetic behavior of this type. For this situation, the
half life would be about 15 min. (Note that, in calculating the
half-life of TS in situations where [FAC] > [TS] and the rate is
independent of [TS], much less than 50% of the active chlorine will
be consumed; in this example, it is 5%).
However, the concentrations of contaminants in drinking water
are small and so the product of k4 and [TS] is likely to be less
than 10000 s"1. Let us assume that the molarity of contaminants is
at least 10-times less than that of [FAC]. In other words, [TS] <
10"6 M. This reasonable assumption simplifies the kinetics by
making them pseudo-first order. Now, if k*[TS] < 1000 s"1, the rate
is dependent on [TS] and is
rate = K^kJHOCl] [Cl'] [TS]/[OH~] (13)
Using the reaction constants and the concentrations from Table 1
provides the equation
rate = (1.9 x 10'13)kJTS] (14)
As an example, if k4 = 107 M^s"1, the half-life for the reaction
would be about 100 hours. Thus, molecular chlorine can function as
a reactant in chlorinated drinking water.
2. Aqueous Chloramines
The situation with aqueous chloramine is even more complex.
Potential equilibria in these solutions are [£, 2, 10]
NH2C1 + H20 v s HOC1 + NH3 K = 6.7 x 10'12 M'1 (15)
2 NH2C1 + H* ^ ^ NHC12 + NH44 K = 1 X 109 M'1 (16)
8
-------
NHC12 + HOCl
HOCl
NCI 3 + H20
ocr
K = 1.6 X 10" M'1 (17)
K = 3.16 X 10"8 M (18)
All of these chlorine-containing molecules (NH2C1, NHC12, NC13,
HOC1, and OCl") have been shown to be reactive species. The
relative amounts of each of these in a particular solution will
depend on the concentrations and the pH.
If a drinking water contains low initial concentrations of
both FAC and NH3 (about 10~5 M) , NC13 will predominate at very low
pH's when most of the NH3 will be tied up as NH4+, but the amount
drops rapidly as the pH increases. The concentration of NHC12 is
low at low pH's, but rises to a maximum at pH 5, where it accounts
for most of the active chlorine, and then drops off again. NH2C1
shows similar behavior except that it peaks at pH 10. As might be
expected, HOCl is important only at low pH's and OC1" at high pH's.
Table 2 gives the concentrations of the various species in
"chloramine"-treated drinking water.
Table 2. Concentrations of Active Chlorine Species
in NH3 and C12 Treated Drinking Water
[FAC] = [NH3]lnltlal = 1.00 X lO'5 M
pH = 7.5
Species
NH2C1
NHC12
NC13
HOCl
OCl"
NH3
NHA+
NH3C1*
Concentration (M)
0.80 x 10'6
4.53 x 10'6
6.14 x 10"8
9.21 X 10'11
9.21 x 10'11
5.96 X 10'8
4.53 X 10'6
2.5 X 10"13
The concentrations of NH3C1+, NC13, HOCl, and OCl" are rather
low and it might be questioned whether they can serve as effective
-------
reactants in drinking water. The various species in aqueous
chloramine generally interconvert much more slowly than those in
aqueous chlorine. However, a reaction may still depend on a highly
reactive species that is present in very low concentration if it
can be generated rapidly enough.
The rate constants for the hydrolysis of NH2C1 to HOC1 are
available [fJ] and can be used to determine if HOC1 or OCl" are
kinetically competent as reactants in chloramine-treated drinking
water.
NH2C1
TS
HO
HOC1
NH3
+ HOC1
= 1.9 X 10"5 s"1 (19)
= 2.9 X 106 M'V1
Using the steady-state approximation, it is found that if kz[TS] >
0.8 s"1, then the overall rate of the reaction is independent of the
concentration of TS and is given by
rate =
On the other hand, if k2[TS] < 0.008 s"1, the rate becomes
rate =
[TS]/[NH3]
(20)
(21)
If the concentration of drinking water contaminant (TS) is more
than 10-fold less than that of NH2C1 and the amounts of NH2C1 and
NH3 are as stated in Table 2, the reaction becomes pseudo-first
order
rate = (1.4 x 10"9)k2[TS]
(22)
Consider the case in which kz = 10000 M^s"1. The half-life of the
reaction is about 14 hours. This illustrates that HOCl is a
potential reactant under these conditions.
NC13 is also present in the chloramine-treated drinking water
solution at tiny concentrations. Thus, it might not be a
significant reactant. Hand and Margerum [£] have shown that the
most efficient route for the generation of NC13 is tha general-
base-catalyzed reaction of HOCl with NHC12. The most effective
base in chloraminated water at pH 7.5 is OH"
NHC12
HOCl
OH"
NC1
OH"
HO
where
3.3 x 109 M'V1
0.14 M'V1
10
-------
A water contaminant, TS, can react with the NC13 formed:
NC13 + TS > P (23)
rate = k2[NCl3][TS]
However, this latter transformation competes with the destruction
of NC13 by NHC12.
NC13 + NHC12 + 3 OH' - * N2 + 2 HOCl + 3 Cl' + H2O
In chloramine-disinfected drinking water (Table 2) , this is
catalyzed by OH" [ill :
rate = k3[NC!3] [NHC12] [OH"] (24)
where k3 = 1 x 108 M'V1. According to Table 2 [NHC12] =2.12 x 10"6
M and [OH"] = 3.16 x 10"7 M so the product k3[NHC!2] [OH"] = 7 x 10"5
s"1. If TS is to compete successfully with NHC12, then k2[TS] must
be comparable to or greater than 7 x 10"5 s"1. A steady state rate
law may be derived for NC13 formation followed by reaction with TS.
At the concentrations in Table 2, if k2[TS] > 6 x 10~6 s"1, then the
initial rate is independent of [TS] and is given by
rate = 1.7 x 10"6 [NHC12] (25)
As an example, if [TS] is about 2 x 10~7 M, its half life would be
about 8 hours.
Inorganic monochloramine also has a protonated form, the N-
chlorammonium ion, which is a more potent chlorinating agent.
, NH2C1 + H+ -v s NH3C1+ (26)
The best estimates suggest that the pKa of NH2C1 is approximately
1 [8., 37] . As the pH of a solution approaches 1 the concentration
of NH3C1* increases.
The rate expression for the second order reaction of NH3C1+
with a toxic substance TS would be
rate = k oNHci* [NH2C1] [TS] , (27)
where <*NH cl+ is the fraction of CRC present as NH3C1*. At pH 7.5 the
fraction of NH3C1+ is 3.2 x 10"7, which yields a concentration of 2.5
x 10"13 M. As with C12, this number appears deceptively small as
model calculations will show.
Since all terms in the rate expression above are constant
11
-------
except [TS], the amount of decomposition of the toxic substance on
reaction with NH3C1+ would be represented by
t1/2 = 0.693/(k a«ci* [NH2C1]) (28)
o
t90, = 2.3/(k am [NH2C1]) ... (29)
In general, it appears that any of the active chlorine species
(NC13/ NHC12, NH2C1, HOC1, or OC1") in chloramine-treated drinking
waters may be kinetically significant chlorinating agents either
directly or through hydrolysis to HOCl.
B. Nature of Reactions with Organic Substrates
1. Nucleophilic Substitution on Chlorine (Cl* Transfer)
Reaction of organic compounds with chlorinated or
chloraminated water can result in substitution, addition,
oxidation, etc. Despite this diversity of reaction results, the
initiating step in most of the reactions has a common mechanistic
characteristic. It is a nucleophilic substitution by an electron-
rich grouping (the nucleophile) on a halogen,
Nu:~ + Cl-X > Nu-Cl + X:"
H-Nu: + Cl-X -*> :Nu-Cl 4- H* + X:'
There is a transfer of an electrophilic (electron-deficient)
chlorine from Cl-X to the nucleophile. In essence, a Cl+ is
exchanged.- The second reaction above can also be described as an
electrophilic substitution or a chlorination. Both reactions are
usually oxidations in the "theoretical" sense since they most often
result in an increase in the oxidation state of the atom or group
Nu. Cl-X may be C1-C1, Cl-OH, Cl-CT, C1-NH2, C1-NHC1, or C1-NC12.
a. Relative Reactivity of Cl* Donors. The reactivity of the
different electrophilic chlorine donors which have been mentioned
varies greatly. The reaction of molecular chlorine with ammonia is
diffusion-controlled [12],
NH3 + C1-C1 * NH2-C1 + H-C1
while reaction of HOCl and ammonia is 1000 times slower [12].
NH3 -f HO-C1 > NH2-C1 + H-OH
12
-------
Chlorine seems to be considerably more reactive than HOC1.
The reaction of most of the other species are acid-catalyzed
which in some instances complicates a direct comparison of their
reactivities. Thus, the H+-catalyzed reaction of OC1"
H* •• + OCl" 4- Br" * HO" + Cl-Br
is about 30,000 times faster than that of HOC1 F131
H* + HOC1 4 Br" > HO-H + Cl-Br
However, when water supplies the proton as in
HO-H -I- OC1 4 Br" * 2 HO" 4 Cl-Br
HO-H 4 HOC1 4 Br" > HO" + H-OH 4 Cl-Br
the rate for OCl" is 2,000,000 times slower than for HOC1. Thus,
there is some ambiguity with respect to the relative reactivity of
HOC1 and OCl".
The rates of the general-acid-catalyzed reactions of I" with
OCl", NH2C1, and NHC12 [14]
A-H + "OCl 4 I" > A" 4 HO" 4 Cl-I
A-H + NH2C1 + I" • * A" + NH3 + Cl-I
A-H 4 NHC12 + I" > A" + NH2C1 + Cl-I
differ from each other by about 10,000-fold (OCl" > NH2C1 > NHC12) .
This relative order is maintained over a series of catalyzing acids
with widely ranging pKa's although the difference in rates
decreases somewhat as acid strength diminishes.
The reaction of NC13 with I" [15] is not catalyzed by acid so
it is difficult to rank its activity with the other chlorine-
transfer reagents. However, it is reported that NC13 reacts a
little bit slower with I" at pH 7 than HOCl does.
A slightly different reactivity ordering is observed when
sulfite ion is the nucleophile [16].
X-C1 -f SO3= > X" 4 C1SO3"
C1SO3" 4 H20 > S04" 4 Cl" 4 2 H*
In this case, reaction with OCl* and NH2C1 is acid-assisted while
that with NHC12 and NC13 is not. NC13 is about 1000 times as active
as NHC12. The latter has a rate 100 times larger than that of NH2C1
13
-------
at pH 7 but almost 1,000,000 times larger than the water-catalyzed
Cl* transfer from NH2C1. Chloramine and OC1" have their usual
reactivity ordering.
These considerations lead to the following "general" order of
reactivity for these species
C12 > HOC1, OC1', NCI 3 > NH2C1 > NHC12
b. Catalysis of Cl+ Transfer. The reactions of nucleophiles
with electrophilic chlorine donors is often acid- or base-
catalyzed. This catalysis serves to assist the departure of the
leaving group attached to the electrophilic chlorine or to remove
a proton from the nucleophilic atom to avoid the formation of a
positive charge.
Nu:~ + Cl-X + H-A - * Nu-Cl + X-H + A'
B:" + H-Nu: + Cl-X - * B-H + :Nu-Cl + X:'
Either specific or general catalysis may occur and are freguently
useful for diagnosing the specifics of the mechanism. Specific
catalysis often suggests that proton transfer is occurring in the
rate-determining transition state, while general catalysis may
indicate an equilibrium proton transfer before this transition
state is reached.
c. Examples of Cl+ Transfer Reactions. When Cl+ donors such
as Cl2r HOC1, OC1~, NH2C1, NHC12, NC13, etc. react with a
nucleophile, a substitution or association product is formed [17.
18, 19].
Me2N-H + HO-C1 - * Me2N-Cl + H-OH
H-C6H«-OH -I- HO-C1 - * o- and p-Cl-C6H4-OH + H-OH .
ph2s: + ci-ci - > pn2s*-ci cr
In some cases, the initially formed product is not stable and
undergoes some further reaction. One simple reaction of this type
is the transfer of the electrophilic chlorine to another
nucleophile [20] .
Et2N-Cl + H-CgH^-OH - > Et2N-H -f C1-C6H4-OH
In another type of reaction, the chlorine intially introduced
into the substrate can be replaced by a nucleophilic substitution
on the atom to which it is attached. One example is the reaction
used to "sweeten" petroleum fractions [21]
R-S-H + HO-C1 - * R-S-C1 -f HO-H
14
-------
R-S-C1 4- "S-R > R-S-S-R 4- Cr
Another illustration is the oxidation of sulfides [22]
Bu2S + HO-C1 * Bu2S+-Cl + HO" >• Bu2S*-CT 4- H-C1
Some products of chlorination may undergo rearrangement.
Thus, nuclear chlorination of anilines appears to proceed through
the N-chloro compounds [23]
C6H5-NR-H + HO-C1 > HO-H 4- C6H5-NR-Cl > ortho and para
C1-C6H4-NHR
Other instances of rearrangement of chlorinated compounds are the
Orton and Hofmann rearrangements of N-chloroamides.
The product of Cl* transfer may also suffer elimination.
Primary and secondary alcohols are oxidized in this way [24]
Me2C-O-H 4- HO-C1 * H-OH + Me2C-0-Cl > Me2C=0 + H-C1
I I
H H
N,N-Dichloroamino acids apparently undergo a decarboxylative
elimination [25]
s-Bu-CH-COOH * s-Bu-CH=N-Cl 4- CO2 + HCl
NC12
The latter reaction is also an illustration of another fate of
chlorinated substances. Under the proper conditions, they undergo
carbon skeleton cleavage. A simple example is the haloform
reaction [26]
C10H7-C-CH3 + 3 HO-C1 > 3 HO-H 4- C10H7-C-CC13
II I •
o o
C10H7-COOH + CHC13
Frequently, this type of transformation involves highly chlorinated
intermediates.
Finally, the addition of Cl* donors to carbon-carbon multiple
bonds must be mentioned. The pi electrons serve as an electron-
rich center to combine with the putative Cl+ to form a cyclic
chloronium ion or an open carbocation. The intermediate reacts
with a nucleophile to complete the reaction [271
15
-------
R2C=CR2 + HO-C1 ' » R2C-CR2 * R2C-CR2
V I I
Cl HO Cl
2. Nucleophilic Substitution on N in NH2C1 (NH2* Transfer)
Reactions in which an amino group is transferred from the
chloramine to a nucleophilic organic substrate are well-known and
useful in synthesis. Some examples are [28. 29]
Et-0" + NH2C1 * Et-O-NH2 + Cl'
Me2N-H + NH2C1 > Me2N-NH2 + H+ + Cl'
+
Me2S -f- NH2C1 > Me2S-NH2 + Cl"
(The last of these reactions may actually be a nucleophilic
substitution by sulfide sulfur on chlorine followed by replacement
of the chlorine on sulfur by amino) . In a somewhat more
complicated example, breakpoint chlorination involves a base-
catalyzed nucleophilic substitution by NHC12 on the N of NC13 [10] .
In most cases, these reactions are relatively slow as compared
to substitutions on chlorine (Cl+ transfer). Thus, the reaction
[30]
HO" + NH2-C1 * HO-NH2 + Cl" , (30)
where k = 6.3 x 10'5 M'V1, has a half-life of 1100 years at pH 7.5
while the competing Cl+ transfer [9]
HO-H + C1-NH2 > HO-C1 + H-NH2 k =1.9 X 10"5 s"1
is 50% complete in 10 hours.
Because of the relative slowness of the substitutions on
nitrogen vs. chlorine, reactions of this type are usually not
important in drinking water chemistry.
3. Nitrenes
Nitrenes (e.g., NH and NCI) have been suggested [35. 36] as
intermediates in some of the nucleophilic reactions on nitrogen
shown in the section directly above . These nitrenes are formed by
base-catalyzed dehydrochlorination of NH2C1 and NHC12. However,
16
-------
careful kinetic examination [35. £, 10] of the suspect
transformations has shown these intermediates are not involved.
4. Nucleophilic Reactivity of ClO" and NH2C1
The oxygen of CIO" and the nitrogen of NH2C1 are nucleophilic
centers. The pKa's of ClO" and NH2C1 (7.5 and 1.4, respectively)
would prognosticate a relatively low reactivity of this type.
However, these entities fall into a special class known as alpha-
effect nucleophiles. Systems in this class have an atom with one
or more unshared electron pairs associated with the atom bonded to
the nucleophilic center. For some reason, this results in enhanced
reactivity of the nucleophile.
For example, OC1" (pKa = 7.5) hydrolyzes p-nitrophenyl acetate
faster than OH" (pKa = 15.7) [3JL].. This reactivity may be involved
in the oxidation of aldehydes by hypochlorite solutions at high pH.
Similarly, NH2C1 reacts readily with aromatic aldehydes at 0°
to form imines (Schiff bases) [28]
Ar-CH=0 + NH2C1 * Ar-CH=N-Cl + H-OH
Amines with pKa's similar to NH2C1 (e.g., p-nitroaniline) do not
exhibit similar reactivity. A similar reactivity enhancement was
noted. in the -SH2 reaction on carbon using N-chlorobenzene-
sulfonamide anion as nucleophile and methyl methanesulfonate as
substrate [32]. The rate was 10 times greater than expected.
5. Free Radical Reactions
Molecular chlorine has been implicated in numerous free
radical reactions. Most of these are chain processes. They occur
in non-polar media which do not provide a favorable environment for
ionic mechanisms. These homolytic reactions do not seem to be of
much significance in polar aqueous solutions.
Some two phase systems containing aqueous hypochlorite have.
been postulated to react by free radical pathways [3_3]. However,
the homolytic process occurs in the nonpolar phase. The
contaminants in drinking water seldom form a separate phase and so
reactions of this type are unlikely.
Numerous free radical reactions of chlorinated ammonia or
amine species have been reported [3_4 ]. These conversions are not
restricted to nonpolar solvents. However, they often involve high
concentrations of acid, light, and/or transition metal ions for
initiation. Except for the transition metal ions, these conditions
are not characteristic of drinking water. Indeed, in the synthesis
of hydrazine from NH2C1 and NH3 in aqueous solution it is necessary
to add gelatin to complex these transition metal ions to obtain
17
-------
better yields.
In certain very favorable circumstances, the species which are
found in aqueous chlorine or chloramine may also engage in single
electron transfer (SET) processes. This might be the situation if
the substrate was a very easily oxidized compound. A possible
example of this is the oxidation of DPD to the Wurster's red cation
radical (a common analytical procedure for active chlorine in
drinking water [37.]). However, this may not correspond to SET from
the chlorinating agent, but rather the initial formation of the N-
chloro derivative of DPD followed by its solvolysis to the
quinonediiminium cation and the reaction of the latter with DPD by
a SET process to form two Wurster's red cation radicals (the
kinetics of the reaction do not allow a discrimination between
these possibilities [38]).
6. Conclusion
In general, "ionic" reactions are favored for the reactive
species found in aqueous chlorine and chloramine. Almost always,
these involve the transfer of an electrophilic Cl+ from the
chlorine donor to a nucleophile.
Nur" + X-C1 > Nu-Cl + X:"
Less commonly, nucleophilic substitution on the X of X-C1 will
expel Cl" and tranfer X to the substrate.
Nu:' + X-C1 3» Nu-X + Cl'
IV. SURVEY OF REACTIONS OF FUNCTIONAL GROUPS
In this section a survey of the known reactions of aqueous
chlorine and monochlorine with various organic functional groups
will be made. The purpose of this survey is to provide a
compendium of reactions and their rates which might be used to
assess the relative susceptibility of various toxic substances to
chemical transformation in a drinking water treatment system.
Theory suggests that the reactivity of the different functional
groups which comprise a toxic substance can be treated separately.
For example, the aromatic ring and the carbamate moiety (half ester
and half amide) present in the pesticide carbaryl can undergo
reactions independent of one another. Thus the primary reactions
of carbaryl with aqueous chlorine or monochloramine can be modeled
by the reactions of these disinfectants with aromatic compounds,
with esters, and with amides.
18
-------
A. Alkanes
Alkanes are a comparatively inert class of compounds.
Although halogens react with alkanes, their reaction requires that
either light or heat initiate the formation of halogen free
radicals and that the concentrations of substrates and other
halogen molecules is sufficient to sustain a chain reaction. In a
water distribution system it is unlikely that the necessary
conditions of light or heat are present for radical initiation or
that the concentration of Cl2(aq) is high enough to sustain a chain
reaction.
B. Haloalkanes
This class of compounds contains a number of pesticides such
as dieldrin, aldrin, chlordane, endrin, heptachlor, lindane,
toxaphene, mirex, and kepone. Several of these, like dieldrin,
aldrin, chlordane, endrin, heptachlor, and toxaphene also contain
alkene moieties, while kepone contains a ketone functional group.
In general, haloalkanes are electrophilic species, as HOC1 and
Cl2(aq) are, and are not chlorinated further. For instance,
chloroform is produced in drinking water treatment systems when
methyl ketones and polyhydroxylated aromatic residues of natural
humic substances are chlorinated. However, chloroform is not
further chlorinated to carbon tetrachloride in a treatment system.
Thus, pesticides such as dieldrin, aldrin, DDT, lindane, mirex and
lindane would not be alterred by the addition of a disinfectant to
a water supply.
C. Alcohols
1. Aqueous Chlorine
Alcohols have a nucleophilic oxygen atom. This is the site at
which electrophilic chlorine donors react with these compounds.
The hydrogen bonded to the oxygen atom is replaced by a chlorine
and an alkyl hypochlorite is formed.
R-O-H -f Cl-X > R-0-C1 + H-X
Anbar and Dostrovsky [39] have investigated the formation and
hydrolysis of t-butyl hypochlorite
•"•;'
t-Bu-O-H + HO-Cl s N t-Bu-O-Cl + HO-H
They found that labelled oxygen was not exchanged between the
alcohol and water during the reaction. This shows that the
19
-------
reaction does not involve the cleavage of the C-O bond of the
alcohol, and therefore the transformation is an substitution of the
H of the OH group by a Cl.
The equilibrium constant for the formation reaction is
K = [t-BuOCl][H2OJ/[t-BuOH][HOCl] = 42
In drinking water (see Table 1) , if [t-BuOH] < io'6 M, only 0.00038%
of the alcohol is converted to the hypochlorite.
The kinetics were also examined. The reaction is subject to
general acid and general base catalysis and thus, its rate changes
with pH. By extrapolation of Anbar and Dostrovsky's data, it is
apparent that at pH 7.5 in the absence of general acids and bases
that the reaction is dominated by OH" catalysis. The rates for the
forward and reverse reactions at this pH are
kf(formation) = 0.053 Vl'ls'1
kr(hydrolysis) = 0.00127 M'V1
If [HOC1] > [t-BuOH] as would be likely in drinking water where the
alcohol is a contaminant, the rate of approach to equilibrium is
k = k£[HOCl] + kr[H2O]
At the concentrations in drinking water, kr[H2O] > kf[HOCl] so k =
kr[H2°] = 0.070 s"1. The half-life for reaching equilibrium is about
10 seconds.
Churganova and Lopyrev [40] have carried out a similar study
for the formation and hydrolysis of ethyl hypochlorite. They
report rate constants for approach to equilibrium about 2-3 times
as great as those found by Anbar and Dostrovsky.
Primary and secondary alkyl hypochlorites are very unstable
[11]. They explode when exposed to light and decompose quickly and
exothermically when warmed to room temperature or when diluted with
a polar solvent. The tertiary compounds are somewhat more stable
and can be prepared and stored successfully if they are not left in
the light for an extended period. However, explosions have been
reported in the preparation and use of t-butyl hypochlorite. When
the primary and secondary compounds decompose thermally, they
produce mainly aldehydes and ketones, although some of alpha-
halogenated compounds may also be formed. Reactions of secondary
alcohols with hypochlorite solutions can be used to prepare ketones
[42, 43].
Me2CH-OH + HOC1 > Me2C=O + HC1
20
-------
The oxidation is quite specific for secondary and benzylic
alcohols. Diols having both primary and secondary hydroxyl groups
provide good yields of the corresponding ketones having the primary
alcohol group intact [43]. This reaction probably involves the
alkyl hypochlorite as an intermediate.
The kinetics of this reaction have been studied at low pH's
(0.5-2) by Kudesia and Mukherjee [4_4]. From the effect of changes
in the concentrations of HOCl, Cl~, and H* on the rate, they
established that C12 was involved in the reaction. The rate is
given by
rate = kclcl[HOCl] [H+] [d~] [iPrOH] kclcl = 125 M'V1
At higher pH's (7-8) [4_5], the rate of the reaction is independent
of the pH and the amount of Cl" suggesting the following rate law
rate = kHocl[HOCl] [OH~] [iPrOH] kHOC1 = 422 M^s'1
Reaction in this intermediate pH range probably involves the OH"
catalyzed formation of the alkyl hypochlorite. Thus, either C12 or
HOCl can .serve as the Cl+ donor with alcohols.
However, neither of these reaction paths will yield
significant amounts of ketone under concentration conditions
prevailing in chlorinated drinking water during reasonable
residence times. Using the concentrations typical of drinking
water and assuming that [Cl2]total > [iPrOH], the rate for the
oxidation by C12 becomes
rate = (5.9 x 10'15 s'1) [iPrOH]
The half-life of the reaction would be almost 4 million years! The
expression for oxidation by HOCl is
rate = (6.7 x 10'10 s'^fiPrOH]
In this case 50% conversion would take 33 years and 1% would
require 174 days.
The selectivity of the reaction shows that primary alcohols
react at least ten tiroes slower than secondary ones. Thus, their
transformation would be even slower. Benzyl alcohols react faster.
Obviously, based on this analysis, the oxidation of alcohols
by the chlorine in drinking water does not occur to any practical
extent.
2. Chloramines
The variety of species in drinking water treated with ammonia
21
-------
and chlorine and the different types and levels of reactivity that
these molecules exhibit leads to an assortment of potential
reactions for alcohols in this medium.
a. Monochloramine. Chloramine might function as a Cl* donor
and convert an alcohol, such as t-butyl alcohol, to its
hypochlorite
t-Bu-O-H + NH2C1 ^ "* t-Bu-0-Cl + NH3 K,
BA
The equilibrium constant for this reaction is not available in the
literature but it may be calculated from published information.
The equilibrium and rate constants for the following reactions are
reported (the values listed below include the concentration of H2O
as 55.5 M) [39_, 12]
H-O-H + NH2C1 s ^ H-0-C1 + NH3 KM = 1.2 X 10'13
k«f = 3.4 x 10'7 Vl'ls'1
knr = 2.9 x 106 Vl~ls~l
t-Bu-O-H + H-0-C1 s v t-Bu-0-Cl + H-O-H K = 42
The equilibrium constant for the t-BuOH + NH2C1 reaction is given
by
KBA = KKw = 5.0 X 10'12
Thus, in typical drinking water treated with chlorine and ammonia
(Table 2) only 6.7 x 10"9% fo the t-BuOH is present as t-BuOCl.
Note that the equilibrium constant for t-BuOH (KBA) is about 42
times larger than that for the similar reaction of H20 (K^,) . This
observation can be used to estimate the forward and reverse rate
constants for the t-BuOH + NH2C1 reaction. If the difference is
equally distributed between these rate constants, they would be
kBAf = 2.2 x 10'6 M'V1
kBAr = 4.5 x 105 M'V1
These speculative rate constants can be used to estimate the rate
of approach to equilibrium. The half-life turns out to be about 26
seconds.
Of more significance is the potential for oxidation of primary
and secondary alcohols via their hypochlorites. We discovered that
this is not an important outcome in aqueous chlorine. With
chloramine, the equilibrium constant for the formation of the
hypochlorite is over a trillion times smaller and the rate constant
is over 20000 times slower than was found for HOC1 in aqueous
chlorine. Therefore, the oxidation of alcohols by chloramine is
not significant.
22
-------
Chloramine undergoes another reaction with alcohols—a
nucleophilic substitution of the anion of the alcohol on the N of
NH2C1 [.28.]. This results in the formation of an alkylhydroxyl-
amine
R-0" + NH2C1 *• R-0-NH2 + Cl~
This reaction has been suggested as a preparative method for
generating these compounds. However, the yields are not good (30-
50%) because of the incursion of side reactions. However, the
question remains whether this a likely reaction in chloramine-
treated water. The rate constant for the similar transformation
involving HO" has been determined [30]
H-CT + NH2C1 3" H-0-NH2 + Cl" kHO = 6.3 X 10'5 M^s'1
An alcohol contaminant in water would have to dissociate before it
could participate in such a change. As an example,
Et-O-H v k- Et-O" + H* K. = 1.3 X 10'16 M
The overall rate of ethylhydroxylamine formation would be given by
rate = kETOKJEtOH] [NH2C1]/[H+]
We can use the kHO for the HO" ion as an approximation for kETO of
the EtO" ion. Then, substituting kHO, the Ka for ethanol, and the
pH of drinking water (7.5)
rate = (2.6 x 10'13 M'V1) [EtOH] [NH2C1]
With the amount of NH2C1 in drinking water (0.80 x 10"6 M) and at
least a 10-fold smaller concentration of alcohol, the half-life for
alkylhydroxylamine formation would be about 100 billion years.
b. Dichloramine. The possibility of transfer of Cl+ from NHC12
to an alcohol can be evaluated in the same manner as we used for
NH2C1. The critical rate and equilibrium constants for NHC12 are
known [12]
H-O-H + NHC12 s v H-0-C1 + NH2C1 K« = 7.8 X 10'n
k«f = 1.2 x 10"8 M'V1
kwr = 1.5 x 102 M"^"1
Use of these values leads to the following equilibrium and
speculative rate constants for reaction of t-BuOH:
23
-------
t-BuOH + NHC12 v, ^ t-BuOCl + NH2Cl KBA = 3.3 X 10"9
kBAf = 7.8 X 10"8
BA
BAr
kBAr = 2.3 X 101 M'V1
In this case, the equilibrium constant for alkyl hypochlorite
formation is about 10 billion times less favorable than for aqueous
chlorine and the rate constant is almost a million times slower
than in aqueous chlorine. The amount of reaction in the latter
system was found to be insignificant. Thus, the reaction of
dichloramine with alcohols to form alkyl hypochlorites and their
possible further reaction products are not of significance under
the conditions expressed in Table 2.
c. Nitrogen Trichloride. The same treatment can be applied to
the possibility of reaction of alcohols with NC13. The necessary
equilibrium constant is [10]
H-O-H + NC13 s ^ H-O-C1 + NHC12 KW = 1.13 X 10'10
When this is combined with the similar constant for the formation
of t-BuOCl from t-BuOH and HOCl, the equilibrium constant for the
following reaction is obtained
t-BuOH + NC13 s s t-BuOCl + NHC12 KBA = 4.7 X 10
-9
8.4 X 107
The rate behavior of the reaction of NC13 and H2O is more
complicated. Apparently, the reaction involves a nucleophilic
substitution of HO" on the Cl of NC13 while some acid donates a
proton to the N. This is a so-called general acid-specific base
catalyzed reaction.
HO' + C13N + H30+ - •»> HOCl -f C12NH + H2O
HO" + C13N + H20 - •» HOCl + C12NH + OH"
1.4 X 10"1 M^S"1
The similar transformations with t-BuOH would involve t-BuO" as
nucleophile
t-BuO" + C13N + H3O+ - * t-BuOCl + C12NH + H2O kBAH
t-BuO" + C13N + H2O - * t-BuOCl + C12NH + OH' kBAW
This anion is formed by the acid dissociation of the alcohol
t-BuO-H N ^ t-BuO" + H+ Ka = 6.3 X 10"20 M
The rates for the H3O+ and H2O assisted reactions of t-BuOH would
24
-------
become
rateBAH = kBAHKa[t-BuOH][NCl3]
rateBAW = kBAWKa[t-BuOH] [NC13] [H2O]/[H+]
Assuming the increase in the equilibrium constant for the reaction
of t-BuOH as compared to water is equally proportioned between the
forward and reverse rates of the reactions of t-BuOH
rateBAH = (3.4 x 10'11 M'V1) [t-BuOH] [NC13]
rateBAW = (3.2 x 10'18 s'1) [t-BuOH] [NC13]/[H+]
At the pH of drinking water (7.5), the last equation becomes
rateBAW7-5 = (1.0 x 10'10 M'V1) [t-BuOH] [NC13]
These estimated rate and equilibrium constants for the formation of
alkyl hypochlorite are almost 10 billion times smaller than the
corresponding values for HOCl (the principal reactive species in
aqueous chlorine). Since reaction of alcohols with HOCl was not an
important process in aqueous chlorine, the same conclusion applies
to reaction of alcohols with NC13.
d. Hypochlorpus Acid and Hypochlorite Ion. HOCl and OC1" are
also present in chloramine-treated drinking water. However, their
concentrations are relatively low, being about 9 x 10"11 M as
compared to 5 x 10"6 M in chlorinated drinking water. Since
interaction of the latter with alcohols was negligible, the
presence of HOCl and OCl" in drinking water disinfected with
ammonia and chlorine is not a source of concern.
3. Conclusion
It does not appear probable that drinking water that has been
treated with chlorine or with ammonia and chlorine at the
concentration levels expressed in Tables 1 and 2 will undergo
significant reaction with alcohols that may be present as
contaminants in the water.
D. Ketones and Aldehydes
1. Aqueous Chlorine
Ketones and aldehydes are halogenated at their a-carbon atoms
on reaction with aqueous chlorine. The halogenation of methyl
ketones results in the elimination of a trihalomethane molecule.
This reaction is frequently referred to as the haloform reaction.
25
-------
o o o
1 HOC1 or 1 S
R-C-CH3 * R-C-CH2C1 * * R-C-O" + CHC13
CIO'
This reaction has been thought to be partially responsible for
formation of chloroform on chlorination of natural humic substances
during water treatment processes.
The o-halogenation of ketones and aldehydes is acid or base
catalyzed [4J5-51] • Each mechanism will be considered separately.
At pH values greater than 11 the reaction is bimolecular [49.50]
and the rate law can be written as:
rate = k [CIO"] [ketone] . (31)
The pH of drinking water is not high enough for this process to be
of any significance. Therefore it will be ignored.
At pH values less than 9 the rate is independent of the
concentration of halogen and at higher concentrations of halogen
independent of the reacting halogen [4_6-4_8,5_l]. The mechanism
which explains these results involves initial slow enolization of
the ketone. For instance, the mechanism for the acid catalyzed
reaction of acetone is:
O HO
n k, I
H+ + CH3-C-CH3 v s CH3-C=CH2
HO O
C12 + CH3-C==CH2 >> CH3-C-CH2C1 + Cl"
The mechanism for the base-catalyzed reaction is:
HO' + CH3-C-CH3
O'
I
C12 + CH3-C=CH2
26
-------
Bell and Yates [51] have suggested that both C12 and HOC1 are
reactants. Using the steady-state approximation the rate
expression for the reaction of acetone with C12 can be written:
d[acetone] k2 Jq [H+] [C12] [acetone]
rate = - = - (33)
dt k+
where kt = 2.9 x 10~7 M"1 sec'1, k.t = 0.116, and k2 = 7.3 x 10s M"1
sec"1. Since the term k2[Clz] » k-ifH*] in the denominator above,
the rate expression reduces to:
rate = k^H*] [acetone]
This suggests that the rate is dependent only on the rate of
ionization of acetone. Since the pH is constant, the equation
above is in the form of a first order equation:
rate = k1 [acetone] (34)
where the rate constant
k' = MH+]
Using equation 5 a half life of 2.4 million years is calculated.
Estimation of the rate of base-catalyzed halogenation of
acetone requires the assumption that the reaction of the enolate
anion with halogen (k«) is much faster than k.j even at low
concentrations of halogen and ketone. This assumption is necessary
since the value of k« is unknown. Studies at higher concentrations
of reactants where the reaction is zero order in halogen concentra-
tion suggest this assumption is valid. In this case, the rate law
reduces to an expression similar to that for the acid-catalyzed
chlorination:
rate = k3[HO"] [acetone]
At constant pH, k3[HCT] is a constant. Using the value of k3 = 0.25
M"1 sec"1 [52] and equation 5 a half life of 102 days is calculated.
These calculations suggest that the a-halogenation of simple
aliphatic ketones and aldehydes is not a significant reaction of
these compounds under water treatment conditions.
By contrast, 1,3-diketones, which enolize to a much greater
extent than acetone, are much more reactive with aqueous chlorine
and produce one molecule of chloroform per molecule of ketone [53].
Using values of the rate constants (k3) for the reaction of
27
-------
hydroxide ion with ketones of varying structure and other compounds
which readily form enolates, half-lives at pH 7.5 can be calculated
[.52., 541. A more detailed discussion of the chlorination of
malonic ester is given below in Section K.b.
Compound
Acetone
Hexan-2 , 5-dione
Chloroacetone
1 , 1-Dichloroacetone
Malonic ester
Acetylacetone
*3
0.25
1.67
9.3
450.
10*
2 X 106
k3[HO-]
7.9 X 10'8
5.2 x 10"7
2.9 X 10'6
1.4 x 10'*
3.2 X 10'*
6.3 X 10'1
^1/2
102 days
15.4 days
2.7 days
1.4 hr
3 . 6 min
1.1 sec
It is obvious from the Table that under drinking water
treatment conditions structure can dramatically affect the half
lives of ketones and other compounds that readily form enolates.
Engfeldt [55] has reported the oxidation of formaldehyde to
formic acid, presumably by the mechanism shown below:
H-C-H
CIO'
H-C-O-C1
I
H
H-C-Q-
HC1
However, larger aliphatic aldehydes undergo the haloform reaction.
2. Monochloramine
Since monochloramine is a much slower chlorinating agent as
discussed above, it would not be expected to chlorinate the a-
carbon of ketones or aldehydes to any significant extent.
Monochloramine reacts with aldehydes to form N-chloroaldimines
[5J5-60]. The mechanism involves an addition of the monochloramine
nitrogen to the carbonyl carbon followed by dehydration of the
carbonolamine intermediate:
28
-------
O OH N-C1
» I -H20 -|
R-C-H + NH2C1 - > R-C-NH-C1 - - *> R-C-H
I
H
Le Cloirec and Martin have reported that inorganic
monochloramine can react with acetaldehyde to produce acetonitrile
[ 61 } . Presumably the reaction involves the intermediacy of an N-
chloroaldimine which is dehydrohalogenated to a nitrile.
N-C1
II -HC1
R-C-H - * R-CsN
Scully [62] has reported the results of a preliminary study of
the kinetics of this reaction. When the reaction of NH2C1 (0.004
M) with an excess of acetaldehyde (0.038 M) in 0.025 M phosphate
buffer and 0.5 M sodium perchlorate (20 °C) was followed by UV
analysis, the loss of the absorption of monochloramine was found to
follow first order kinetics and increase with decreasing pH. At pH
6 . 5 the observed rate constant was 0.022 min"1. By assuming a
second order reaction at constant pH and extrapolating to
concentrations of reactants found in drinking water ([NH2C1] = 10"5
M and [acetaldehyde] < 10"6 M) a half -life for the aldehyde at pH
6.5 of 82 days is estimated. The half-life would be longer at
higher pH-. .- In any_case, the direct reaction of monochloramine with
aldehydes is believed to be too slow to be of significance in a
drinking water system.
3. Conclusions
Reactions of simple ketones and aldehydes are too slow to be
of significance in a drinking water. Only ketones and aldehydes
which contain active methylene groups like acetylacetone or or-
halogenated ketones can undergo significant reaction in a drinking
water disinfected with aqueous chlorine. There is no evidence that
monochloramine chlorinates ketones and aldehydes to any significant
extent .
E. Amines
1. Aqueous Chlorine
Primary and secondary aliphatic amines react rapidly with
aqueous chlorine over a wide range of pH values. The mechanism
involves the reaction of a free amino group with hypochlorous acid
as illustrated below for the reaction of glycine, Gly:
H2N-CH2-COO" + HOC1 *• C1NH-CH2-COO" + H2O
29
-------
The concentration of free amino acid is dependent on pH and the
fraction of unprotonated amino acid present at pH 7.5 is
considerably lower than the total amino acid concentration.
However, as unprotonated amino acid is consumed, the equilibrium
shifts and the unprotonated form is regenerated.
The fraction of unprotonated form is given by:
(1 - «) = - - - v — 5— (35)
1 • K"-H v '
Since the pKa of glycine of 9.77 [63.], at pH 7.5 the fraction of
the total glycine concentration, [Gly]T, which is unprotonated is:
[Gly]T = [protonated Gly] + [unprotonated Gly]
At pH 7.5 the concentration of free unprotonated glycine is given
by the expression
[unprotonated Gly] = (l-a)Gly [Gly]T
The rate expression for the reaction is
rate = k [unprotonated Gly] [HOCl]
Consequently, the rate is dependent on the fraction of free
available chlorine which is present as HOCl at a given pH and the
fraction of free amino groups present at that pH. The rate is
fastest at pH values between the pKa value of HOCl and that of the
amino nitrogen of glycine. The rate expression can be written:
rate = k oHocl [FAC] (l-o)Gly [Gly]T (36)
where k is the reaction rate constant, [FAC]T and [Gly]T are the
total concentrations of free available chlorine and glycine,
respectively. The terms aHocl and (l-a)Gly are the fractions of the
protonated acid (HOCl) and unprotonated amine (glycine) , respec-
tively, at a given pH. Since [FAC] » [Gly]T the conditions are
pseudo-first order. Therefore,
rate = k'[Gly]T (37)
where k1 = k aHOC1 (l-a)Gly[FAC] (38)
The value of k for the reaction of HOCl with glycine is 1.1 x
108 M^sec"1 f64. 651 . The value of orHOC1 at pH 7.5 is 0.50, [FAC] is
1 x 10"5 M (as C12) , and (l-o)Gly is 0.0053. Therefore, k1 = 2.9
30
-------
sec"1. The half-life of the reaction would be 0.23 sec and 90%
would react in 0.8 sec.
Margerum [12] has shown that free amines react with C12 at
diffusion-controlled rates according to the equation:
R-NH2 + C12 > R-NHC1 + HC1 (39)
The rate expression (as discussed earlier, equation 14) for the
reaction of C12 in drinking water when the product of the rate
constant and the concentration of toxic substance is < 10* is:
rate = (1.9 x 10"13) k [TS]
where the rate constant k is 1.6 x 109 M"1 sec"1. From this a
pseudo-first order half-life of 38 min can be calculated.
The chlorination of aliphatic amino nitrogens to form N-
chloramino compounds during drinking water treatment is an
extremely rapid and significant reaction. Although the reaction of
C12 is fast, HOC1 is a more rapid chlorinating agent.
In general N-chlorinated secondary aliphatic amines without a
good leaving group attached to the a-carbon appear to be stable and
do not decompose rapidly at pH 7.5 [.66.]. They decompose by
dehydrohalogenation on irradiation with UV light to form imines
[66]-
Cl
| hv
CH3-CH2-N-CH2-CH3 * CH3-CH2-N=CH-CH3 + HC1
Aromatic amines react with aqueous chlorine to form ring
chlorinated products. The mechanism involves electrophilic
aromatic substitution and will be discussed below.
Because there is an excess of free available chlorine under
drinking water conditions, primary amines can be chlorinated
further to form N,N-dichloramines:
C1-HN-CH2-CH2-COO" + HOC1 * C12N-CH2-CH2-COO" + H2O
The rate for this reaction is considerably slower than the rate of
the first chlorination yet is sufficiently fast that N,N-dichlor-
amines are the major products of the chlorination of primary amines
under drinking water conditions. The rate constants for the second
chlorination of N-chloro-/3-alanine is 2.8 x 102 M^sec"1 [12.]. If N-
chloro^/3-alanine (1 x 10"6 M or less) is formed in the presence of
1 x 10"5 M free available chlorine (as C12) at pH 7.5, it would form
31
-------
N,N-dichloroglycine with a half-life of about 8 min. Greater than
90% will react in about 27 min.
N,N-dichlorinated a-amino acids are unstable and decay rapidly
to N-chloroaldimines and nitriles [25_,67., <>8_] :
N-C1
-co2, -cr i
Cl2N-CH-COCr *• R-C-H + R-C=N
R
At neutral pH N-chloroaldimines like N-chloroisobutyraldimine, N-
chlorophenylacetaldimine, and N-chloro-2-methylbutyraldimine have
half-lives of about 40 hours and decompose to the corresponding
aldehydes presumably by hydrolysis [69].
2. Monochloramine
Monochloramine has been shown to transfer its chlorine
directly to the amino group of organic amines and amino acids [63.
70] according to the equation:
NH3C14 + H2N-CH2-COCf > NH/ + C1NH-CH2-COO"
Here the rate constant is 9.36 x 108 M^sec"1 [(53.]. A calculation
similar to that in the above section can be carried out for the
chlorine transfer reaction of monochloramine (0.80 x 10"6 M, see
Table 2) with amino acids. Since the fraction of NH3C1* at pH 7.5
is 3.2 x 10"7, the pseudo-first order half-life is found to be 48
min and 90% will react in 2.7 hr. This reaction will be
significant in a drinking water distribution system.
Unlike hypochlorous acid, monochloramine is not likely to
transfer two chlorine atoms to the same amine to form a dichlor-
amine. The stability of organic N-chloramines depends on the
structure of the compound. As mentioned earlier, N-chlorinated
secondary aliphatic amines appear to be stable and do not decompose
rapidly at pH 7.5 [66]. However, if a toxic substance contains a
good leaving group attached to the a-carbon, it may undergo an
elimination reaction to form an unstable imine which will hydrolyze
to an aldehyde [25, 71, 72]. For example, a-amino acids can
decompose by the following scheme:
32
-------
N-H O
-co2, -cr | +H2o I
C1NH-CH-COO" »• R-C-H > R-C-H
I ~NH3
R
Except for N-chloroglycine which has a half-life of over 5 days, N-
chlorinated amino acids have comparatively short half-lives
[Z2,7_J3], between 30 and 100 min. The dipeptide N-chloroglycyl-
glycine has a long half-life (9.7 days at pH 8.5), but N-chloro-
glycine ethyl ester has a half-life of only 50 min at pH 8.38 [73].
Presumably the products are aldehydes which form by elimination of
HCl and hydrolysis of the resulting imine.
3. Conclusions
Chlorination of aliphatic amino nitrogens by either HOCl or
monochloramine is one of the most rapid and significant reactions
which can take place in a disinfected drinking water.
F. Amide Nitrogens
1. Aqueous Chlorine
Mauger and Soper [74] have studied the kinetics and mechanism
of the.reaction of amides with aqueous chlorine. They suggest that
the uncatalyzed reaction as illustrated for N-acetylglycine:
CH3CO-NH-CH2COO" + CIO" >• CH3CO-N-CH2COO" + HO"
I
Cl
follows the rate law:
,-d[Amide]/dt = k [amide] [CIO"]
= k orhyp[amide][FAC]T (40)
where ahyp is the fraction of unprotonated hypochlorite at a given
pH and [FAC]T is the free available chlorine concentration. The
value of k for the chlorination of amides is small. For N-acetyl-
glycine k - 5 x 10"2 M"1sec"1. Consequently, at 25 °C and at pH 7.5
the half-life for chlorination of an amide with a concentration of
< 10"6 M is 32 days and it would take 106 days for 90% to react.
This is not a significant pathway for decomposition of toxic
substances containing amide functional groups. Thus, carbamates,
like the pesticide carbaryl, which contain the amide functional
group would not decompose significantly in a chlorinated drinking
water.
Mauger and Soper [74] also describe an acetic acid-catalyzed
33
-------
chlorination of the amide bond. However, the concentration of
carboxylic acid in a finished drinking water would have to be as
high as 1.3 x 10~3 M before the catalytic process would double the
rate of chlorination of amides by the uncatalyzed process. Morris
[64] discusses the possibility that the reaction is general acid
catalyzed, but the concentrations of other acids in water is still
not significant. The reaction of amides with C12 may have much
greater significance, but this reaction has not been studied.
2. Monochloramine
No reactions of monochloramine with amides has been reported,
probably because there is no reaction.
3. Conclusions
Reactions of amides (and probably carbamates as well) are too
slow to be of significance in a drinking water disinfected with
aqueous chlorine or monochloramine.
G. Aromatic Compounds
1. Aqueous Chlorine
Aqueous chlorine reacts with aromatic compounds to form ring-
chlorinated substitution products. The reaction rate increases
with an increase in the number of activating (i.e., electron-
donating) groups attached to the ring. For instance, de la Mare et
al. [75] have shown that toluene is chlorinated by HOCl 60 times
faster than benzene.
Studies of the acid-catalyzed chlorination of aromatic
compounds have been the subject of several studies over a number of
years. Early work by de la Mare et al. [76] suggested that the
reaction rate was zero order in the aromatic substrate and first
order in the HOCl concentration in acid solution. They proposed
the intermediacy of the chlorinium ion, Cl*, as the active
chlorinating species. However, for many years there appeared to be
a discrepancy between the kinetics which suggested the formation of
Cl+ was fast and the thermodynamics which suggested that the
equilibrium constant for Cl+ formation (Keq = 10"*°) would make the
concentration of Cl+ too low to be kinetically significant. Work
by Swain and Crist [77] resolved this conflict by studying the
reaction over a much wider range of concentrations of acid and
reactants. They showed that anisole, AnH, was chlorinated in acid
solution by three parallel pathways and found the complete rate law
to be:
- d[AnH]/dt = k'[HOC1]2 -I- k" [H3O+] [HOCl]2 + k" • [AnH] [H3O+] [HOC1(]I4)
34
-------
where
[HOC1] = aHocl [FAC]
They interpreted the first two terms of the expression as
indicative of rate-determining formation of C12O by independent
pathways:
k1
2 HOC1
or
2 HOC1 + H,0+ -
k"
C12O + H2O (slow) (41)
C120 + H3O* + H20 (slow) (42)
followed by
C12O + AnH
AnCl -f HOC1
(fast)
(43)
They interpreted the third term of the rate expression as a
possible pre-equilibrium formation of H2OC1+ followed by rate-
determining chlorination of the anisole:
HOC1
H,OC1+
AnH
H2OC1+
AnCl
H20
H30+
(fast)
(slow)
(44)
(45)
The problem with modelling this chemical reaction in dilute
aqueous solution is lack of knowledge of the rate constant for the
reaction of C12O with AnH, k*. In the millimolar concentration
range used by Swain and Crist the rate of the final chlorination
step involving C120 was sufficiently fast for the reaction to
depend solely on the rate of C12O formation because the term k*[AnH]
was sufficiently large (see the earlier general discussion of the
mechanism of the formation and reaction of C12 with toxic
substances, TS) . However, at submicromolar concentrations this
term may not be large enough for the AnH to react with the C12O as
fast as it is formed. In this case, the rate is
where
rate = k*K*[HOCl]2[AnH]
[C120]
[HOC1]'
(46)
(47)
K* = 8.7 x 10'3 [78.]. Since k* is unknown, the true rate cannot be
predicted. However, the high reactivity of C12O will be assumed
[7_8.-80_] and a minimum half-life can be predicted if the formation
of C12O is assumed to be the rate determining step even in dilute
solution.
35
-------
If the pH is constant at 7.5, [FAC] = 1 x 10'5 M, and the
concentration of substrate is 1 x 10"7 M, the concentration of HOC1
is constant over the course of the reaction and the first two terms
of the rate expression are pseudo-zero order expressions. The
half-life of a zero order reaction is dependent on its initial
concentration.
1/2 [Substrate]lnltlal
Half-life = = (48)
k"[HOCl]2
For zero-order processes the half-life of the reactant decreases by
a factor of 10 for each ten-fold decrease in initial concentration.
If the rate expression for each pathway is evaluated separately,
the relative significance of each can be determined. For the first
term
- d[AnH]/dt = k'[HOCl]2 (49)
Swain and Crist found k1 = 0.124 M"1sec"1. By this pathway, half the
anisole would react in 4.5 hours and 90% would react in 8.1 hours.
For the second term
- d[AnH]/dt = k"[H30+] [HOC1]2 (50)
they found k" = 3.06 M"2sec"1. From this expression a half-life of
656 years is calculated. For the third term which is first order
in anisole
- d[AnH]/dt = k1 ' ' [AnH] [H30+] [HOC1] (51)
they found k111 = 0.478 M^sec"1. From this a half-life of > 260,000
years is calculated.
The first term of the expression appears to dominate the
kinetics at pH 7.5 and provides a pathway for significant
decomposition of reactive aromatics in chlorinated drinking water,
while the second two terms of the rate expression are comparatively
unimportant.
Reinhard and Stumm have elucidated the complete rate law for
the reaction of p-xylene [79]. The rate of formation of
chloroxylene is given by:
k,"
rate = — [XH] [HOC1] [H+] [Cl'] + k2"K2[XH] [HOC1]2 (52)
Ki
where [XH] is the concentration of xylene, kj" is the second order
rate constant (150 M^sec'1) for the reaction of the aromatic with
C12, Kj is the equilibrium constant (3.94 x 10"A M2) for the
36
-------
hydrolysis of C12 to HOC1 and HCl, k2" is the rate constant (33 M"
1sec"1) for the reaction of the xylene with C12O, and K2 is the
equilibrium constant (8.7 x 10"3 M"1) for formation of C12O from two
molecules of HOCl. Where Swain and Crist assumed the reaction of
H2OCl+f Reinhard and Stumm suggest that the reactive species is C12.
Based on this rate law, they estimate that less than 0.001% of
xylene (approx. 10"9 M) would react in one hour under drinking water
disinfection conditions ([FAC] = 10"* M) . The [FAC] used in the
present study is ten times less, which would suggest an even lower
level of chlorination of xylene in a drinking water.
Carlson et al. [80] examined the uptake of aqueous chlorine (7
x 10"* M) by a series of aromatic compounds (9.5 x 10"* M) at pH
values of 3, 7, and 10.1. Phenol showed extensive uptake at all pH
values. Anisole was less reactive, but still showed 11 % uptake in
20 min at pH 7. Snider and Alley [JLl,iL2] observed chlorina-tion of
biphenyl over 12 hr at 40 °C at pH 7. The reaction was studied
over a variety of pH values (pH 6.78 - 9.17) and aqueous chlorine
concentrations which showed that lower pH values gave higher
yields. The chlorine concentrations used were high (between 2 x
10"* M and 4 x 10"3 M [ref. 82 lists concentrations 1000 times lower
than this, but these appear to be incorrect]). Burleson et al.
[83] found no ring substitution products of the chlorination of
phenylalanine on superchlorination of wastewater, only breakdown
products of the chlorination of the terminal amino group. Carlson
and Caple [841 found that the model compound toluene was
chlorinated at least 40 times slower than phenol at either pH 3 or
at pH 10. This observation would suggest that compounds like the
PCBs and DDT are unreactive. Because of the deactivating nature of
halogen substituents, it is difficult to predict the contrasting
effects of alkoxy substituents and halogens on reaction of the
aromatic ring in phenoxyacetic acid pesticides like 2,4-D. Such an
aromatic ring is probably too deactivated to react significantly in
a water treatment and distribution system. Carlson et al. [80]
have studied the aqueous chlorination of a series of polynuclear
aromatic compounds under varying conditions of time (0.5 to 3 hr),
chlorine dosage (3 x 10"5 M to 3 x 10"* M) , and pH (4-7) that might
exist during actual drinking water conditions. Both chlorination
and oxidation products were observed.
The reactivity of phenols is markedly different from that of
other aromatics, because it can ionize at higher pH values. Lee
[85] and Soper and Smith [86] studied the rate of reaction of
phenol with aqueous chlorine at pH 5-12 and found the reaction
mechanism involves chlorination of the phenoxide ion by HOCl. It
appears to follow the rate law:
, -d[phenol]/dt = kz [HOCl][phenoxide] (53)
(l-a)^ [Phenol]T
37
-------
where k is the second order rate constant, FAC is the concentration
of free available chlorine, (l-a)pi,o is the fraction of ionized
phenol and [Phenol]T is the total concentration of phenol. Soper
and Smith [86] report a value of kz for phenol of 3.3 x 10s Heroin"1.
Since phenol has a pKa of 9.89, (l-a)^ = 4.1 x 10~3 at pH 7.5. If
it is assumed that the concentration of phenol is less than 1CT6 M,
then the conditions are pseudo-first order and a pseudo-first order
half-life of phenol in a drinking water at pH 7.5 would be 103 min.
It would require 5.7 hr for 90% to react.
Values of k2 calculated from the data of Lee [85] are three to
five times greater than that reported by Soper and Smith. In
addition they observed a doubling of the value of k2 from pH 12 to
pH 6 which attribute to a small amount of acid catalysis [86]. The
data of Lee simply suggests that the half-life of phenol would be
shorter and hence its reaction more significant.
The products of the chlorination of phenol, primarily o- and
E-chlorophenol, can react further in a drinking water to form
dichlorinated products. Lee [86] has shown that the observed rates
of chlorination of a series of chlorinated phenols varies widely.
Thus, g-chlorophenol is 4 times less reactive than phenol, but the
maximum observed rate constant for the reaction o-chlorophenol is
approximately equal to that of phenol. However, because of the
lower pKa of the chlorinated phenols, the pH at which the maximum
rate is observed is lower than that of phenol. Thus, o-
chlorophenol will react with aqueous chlorine about 50% faster at
pH 7 than phenol will. Therefore, it may be concluded that the
reactions of phenols and their chlorination products at pH 7.5 are
sufficiently fast to be of significanec in a drinking water. In
fact, the presence of chlorinated phenols is frequently associated
with an undesireable taste and odor of chlorinated waters.
Nowell and Crosby [87] have reported the reaction of
nitrophenol and 3-hydroxybenzoic acid with aqueous chlorine in both
light-induced and dark reactions.
Larson and Rockwell [88] have demonstrated that ring
substitution in some phenols can mean replacement of other
functional groups on the ring. Thus chlorination of phenolic
acids, such as p-hydroxybenzoic acid, gives replacement of the
carboxyl group by a chlorine atom.
The reactions of aqueous chlorine with polyhydroxylated
aromatics can give numerous products. 1,3-Dihydroxylated aromatics
residues are believed to be one of the main sources of chloroform
produced on chlorination of humic substances in drinking water
[89.]. A mechanism has been proposed.
Grimley and Gordon [90] have shown that at low pH and in high
chloride concentrations the rate of chlorination of phenol depends
38
-------
on the concentration of Cl2(aq). However, this reaction is not
significant in drinking water.
2. Monochloramine
Brown and Soper [911 have studied the reaction of N-chloro-
diethylamine with phenol at vaying pH values. The rate expression
is:
rate = k0[chloramine] + kgfl-cOpho aR2NHC1[R2NCl]T[PhOH]T (54)
where k0 is the rate constant for the hydrolysis of the chloramine
to HOC1 and amine, k3 = 3.93 x 108 M^min'1 for phenol. At pH 7.5
aR NHCI = 3.16 x 10"7 and (l-a)ph0 = 4.1 x 10"3. If monochloramine with
a Concentration of 5.7 x 10"6 M (Table 2) reacts as rapidly as N-
chlorodiethylamine, then each term can be evaluated separately.
Brown and Soper report that the first term of the expression
is due to the hydrolysis reaction followed by reaction of the HOC1
with phenol:
NH2C1 + H20 v v NH3 + HOC1
PhOH + HOC1 * Cl-PhOH + H2O
However, with the comparatively high concentrations of phenol they
used, the reaction appears to be zero order in phenol. However, at
much lower concentrations (e.g., 5 x 10~7 M), the reaction becomes
second order and is dependent on the equilibrium concentration of
HOC1. However, the hydrolysis is so thermodynamically unfavorable
(Keq = 6.7 x 10"12 [H]) that the reaction cannot be significant
unless the substrate being chlorinated can react fast enough with
the HOC1 as it is formed. In dilute solution, this reaction is not
fast enough and chlorination of phenol by hydrolysis would have a
half-life of over 7000 years.
The second term in the rate expression is pseudo-first order
in [PhOH]. Under these conditions, the half-life of the reaction
is approximately 175 days. Activated phenols, such as p-cresol
react about three times faster than phenol, while deactivated
phenols, like p-chlorophenol react about 7 times slower [91].
3. Conclusions
Ther reactivity of aromatic compounds varies widely and depends
on the type of substituents attached to the ring. Aromatic
compounds substituted with simple alkyl groups like p-xylene do not
react to any significant extent in a chlorinated or chloraminated
drinking water. Aromatic compounds substituted with more strongly
39
-------
electron-donating groups like methoxy groups (anisole) will undergo
significant transformation in water disinfected with aqueous
chlorine. Phenolic compounds, which can ionize to even a small
extent at pH 7.5, will react to a significant extent in chlorinated
water. In fact, the chlorination products may undergo further
chlorination at this pH. In general, aromatic compounds will not
undergo significant transformation in chloraminated drinking water.
H. Heterocyclic Aromatic Compounds
1. Aqueous Chlorine
Gould et al. [92. 93. 94] have studied the chlorination of
purines and pyrimidines. Products have been identified in several
cases. Seven products of the chlorination of caffeine have been
identified and 5-chlorouracil was identified in the chlorination of
uracil. At high chlorine-to-substrate ratios dichlorouracil and
oxidation products were formed. The aminopyrimidines, cytosine and
5-methylcytosine, form remarkably stable N-chloramino products.
Preliminary kinetic studies by Gould revealed a complex reaction
mechanism for the chlorination of caffeine and uracil, but the data
is not sufficient to extrapolate their studies to a drinking water
system. Rosenblatt [ 95 ] has discussed the results of Jolley who
has identified several chlorinated heterocyclic compounds with
significant concentrations in a chlorinated wastewater: 5-chloro-
uracil (4.3 ng/L) , 5-chlorouridine (1.7 /jg/L) , 8-chlorocaffeine
(1.7 /ig/L) , 6-chloroguanine (0.9 jzg/L) , 8-chloroxanthine (1.5
Lin and Carlson [96] have examined the reactivity of a group
of environmentally important heterocyclic compounds under
disinfection conditions. Indole, 3-methylindole, 2-phenylindole,
N-phenylpyrrole, dibenzothiophene, and carbazole at concentrations
of 10"5-10"6 M reacted to greater than 95% with aqueous chlorine (10"
* M) within 15 min at pH 7. Similar solutions of other
heterocyclic compounds reacted (95%) over somewhat longer times:
acridine (45 min), phenanthridine (1 hr) , benzo[b] thiophene (1.5
hr) , and isoquinoline (4 hr) . Ring chlorination and oxidation
products were identified.
2 . Monochloramine
Lin and Carlson [96] have also examined the reactivity of
heterocyclic compounds (10"5-10~6 M) with monochloramine (10"* M) .
Reaction times at pH 7 for >95% reactivity were: indole (21.5 hr) ,
3-methylindole (21.5 hr) , 2-phenylindole (3.5 hr) , and N-
phenylpyrrole (50 hr) . Products were identified.
40
-------
3. Conclusion
No systematic mechanistic rate studies of the reactions of
heterocyclic aromatic compounds have been conducted to predict the
half lives of these compounds. Chlorination products of
heterocyclic aromatics have been identified in chlorinated
wastewaters and studies designed to model chlorinated and
chloraminated drinking waters suggest that some of these undergo
significant transformation under drinking water conditions. The
half lives of the heterocyclic aromatic compounds studied by Lin
and Carlson [9(5] appear to be considerably longer in chloraminated
water than in chlorinated water.
I. Alkenes and Alkvnes
1. Aqueous Chlorine
Alkenes react with HOCl to form halohydrins and these
compounds can react further to form epoxides. This reaction is
illustrated below for the reaction of ethylene:
HO Cl
I I
H2C==CH2 + HOCl * H2C—CH2
The mechanism of the addition of HOCl to a double bond is believed
to involve electrophilic attack by the positive end of the HOCl
dipole and formation of a cyclic chloronium ion, followed by
reaction of this intermediate with H2O or HO".
Cl* HO Cl
/ \ H2O or HO' | |
H2C==CH2 + HOCl * H2C CH2 * H2C—CH2
In a study of the addition of HOCl across the double bond of
unsaturated fatty acids and natural products at pH values from 2 to
10, Carlson and Caple r84] identified chlorohydrin products and
trace quantities of epoxides.
HO Cl 0
II HO' /\
H2C—CH2 > H2C CH2 + HC1
However, experimental details were not reported and, therefore, it
41
-------
is difficult to extrapolate these data to concentrations found in
a drinking water.
The review by Fukayama et al. [97] of the work of Leopold and
Mutton reports that "aqueous chlorine completely saturates
triolein" (glyceryl trioleate) . Ghanbari et al. [98.,21] found that
approximately 10-15 mole percent of the 36C1 was incorporated per
double bond into fatty acids and their methyl esters on reaction
with aqueous HO36C1 (0.1 M phosphate pH 6, 60 min). For instance,
linolenic acid, which contains three double bonds, incorporated 2.5
times more chlorine than did oleic acid which contains only one.
Interpretation of their rate measurements is impeded by their use
of oily suspensions and the probable presence of the acids in
micelles. Rates in these cases do not follow simple diffusion
laws. There work also raises the reverse problem of extrapolating
reaction rates measured in a homogeneous system to an aqueous
system in which a lipophilic toxic substance may be bound to
colloidal organic matter in the water.
Israel et al. [100] have studied the reaction of HOCl with
allyl alcohol at pH values ranging from 3.8-4.5. The observed rate
of disappearance of the alkene was dependent on two competing
reactions:
rate = k1[alkene][HOCl] + k"[HOCl]2 (55)
where the first term referred to the direct reaction of HOCl with
the alkene, while the second term referred to the rate of formation
of chlorine monoxide (C120) which reacted much faster with the
alkene. At pH 3.8 the observed reaction rate constants k' equalled
1.29 M'hnin'1 (2.15 x 10"2 M^sec"1) while k" equalled 10.21 M'hnin'1
(0.17 M^sec"1) . If the competing pathways are evaluated
independently, the half-life of the reaction at pH 3.8 between HOCl
(1 x 10"5 M) and the alkene (1 x 10"7 M) would be 37 days.
Two mechanisms for the formation of C12O are discussed above
in the section on the reactions of aromatic compounds with aqueous
chlorine [77]. One of of the pathways is acid catalyzed. Israel
et al. [100] state that the rate of formation of chlorine monoxide
increases with decreasing pH. Nevertheless, using the mechanisms
and the rate constants of Swain and Crist [77], the acid catalyzed
reaction only contributes a small amount to the overall reaction
rate, even at pH 3.8. The most important unknown for predicting
the reactions of alkenes in drinking water treatment is what the
rate constant is for the reaction of C12O with an alkene. If it is
sufficiently high, the reaction is dependent only on the rate of
formation of C12O. However, if it is not fast the overall reaction
rate could be insignificant.
Assuming that the rate of the reaction of C120 with an alkene
in dilute solution is so fast that the rate is independent of the
42
-------
substrate concentration and only on the rate of formation of C12O,
then the second term can be evaluated. Since it is assumed that
[HOC1] is constant over the course of the reaction, k"[HOCl]2 is a
constant, this pathway is a zero order process, and the half-life
is determined by:
1/2 [Substrate]lnltlal
Half-life = ; (56)
k"[HOCl]2 v '
The half-life of a zero order reaction is dependent on its initial
concentration. If the initial concentration of alkene is 1 x 10~7
M, then from equation 56 a half-life of 3.3 hr is calculated. If
the initial concentration is 1 x 10"8 M, the half-life is 19.6 min.
The reactivity of an alkene depends on the substituents
attached to the double bond. In general, alkenes are considered
electron-rich species and they react with electrophilic reagents.
If electron-withdrawing groups are attached to the double bond,
their reactivity is greatly diminished by the pathway outlined by
Israel et al. [1001 above. In fact, if an alkene is substituted
with a sufficient number of electron-withdrawing groups, its
reactivity is completely reversed. Rosenblatt [95] has found that
electron-rich hypochlorite epoxidizes the following electron-poor
double bond:
0
CIO' / \
o-Cl-Ph-CH=C-CsN > o-Cl-Ph-CH C-C=N
I I
C=N C=N
2. Monochloramine
No - studies of the products or rates of reaction of
monochloramine with alkenes or alkynes are known. However, if
reactions of HOC1 are slow, reactions of monochloramine are
probably even slower.
3. Conclusions
There are no mechanistic rate studies of the reactions of
alkenes with aqueous chlorine at neutral pH and experimental
details of model studies are not sufficient to properly evaluate
their significance. The rate of reaction of the intermediate C120
is unknown. Therefore, predictions of the rate of chlorohydrin
formation can range from a few hours to 37 days. More extensive
rate studies are needed before the true reactivity of alkenes can
be evaluated.
43
-------
J. Carboxvlic Acids
1. Aqueous Chlorine
a. Simple Carboxylic Acids. Carboxylic acids like alcohols
have a hydroxyl group with a nucleophilic oxygen atom. Therefore,
it might be expected that in analogy with alcohols that organic
hypochlorites could be formed. This reaction apparently does occur
with acetic acid (HOAc).
Me-C-O-H + HO-C1 =F=^ Me-C-O-Cl + HO-H
The product is variously known as acetyl hypochlorite or chlorine
acetate. Anbar and Dostrovsky [101] have detected a strong
anomalous ultraviolet absorption in acetic acid solutions of this
purported species—an absorption different than that of any other
molecule likely to be present under the reaction conditions (e.g.,
C120, HOC1, etc.).
The equilibrium constant for AcOCl formation has been measured
[102] and found to be 0.0025 (this includes the concentration of
water). This value is 17000 times less than the similar constant
for the formation of t-butyl hypochlorite [101]. In drinking water
(Table 1) containing less than 10"6 M HOAc (mainly present as its
conjugate base, AcO", at pH 7.5), only 2.3 x 10"9% of the original
HOAc is converted to its hypochlorite (at 10"6 M initial HOAc, the
concentration of AcOCl is 2.3 x 10"17 M) . This is obviously not an
important component of the solution.
Interestingly, acyl hypochlorites are postulated to be
intermediates in the Hunsdiecker reaction which is used to form
alkyl halides from carboxylic acids. Furthermore, some of the
procedures used to prepare solutions of acetyl hypochlorite are
reminiscent of those used in the Hunsdiecker reaction.
The carbon-bound hydrogens in carboxylic acids can be replaced
by chlorine. The reaction is photochemically induced and the major
products are substituted in the 2- and 3- positions ri03]. It is
a free radical reaction and does not appear to occur in aqueous
solution. Hydrogens on the alpha-carbon atoms in ketones and
aldehydes are easily replaced by halogens in aqueous solution.
These reactions occur through the formation of a nucleophilic enol
or enolate. It might be anticipated that the alpha-carbon in
carboxylic acids would be a site for a similar substitution.
However, this does not seem to be the case for simple carboxylic
acids. They do not react easily with halogens. In fact, acetic
acid is frequently used as a solvent for halogens in electrophilic
halogenations of other compounds.
44
-------
There is a report of the oxidation of cyclohexanecarboxylic
acid to succinic, glutaric, and adipic acids in unspecified yields
by heating at 60-70° C with a 10- molar excess of 1.5 M NaOCl (pH
12-13) for 3-5 hours [104]. These are vigorous conditions and ones
likely to lead to the disproportionation of CIO" ion to the
powerful oxidant C1Q3" ion. Similar treatment of hydrocarbons led
to the formation of chlorinated substances suggestive of free
radical reactions.
b. Carboxylic Acids with "Active Methylene" Groups. Although
simple carboxylic acids are unreactive towards aqueous chlorine and
"chloramine", the same is not true of those acids that have keto,
carboxyl, cyano, nitro, and similar groups attached to the alpha-
or 2-carbon atom. These compounds do react fairly readily with
aqueous solutions of halogens. Studies of a few of these compounds
have been carried out.
Malonic acid has been the subject of several of these
investigations. The interest in this substance stems from the fact
that the bromination of malonic acid is one of the steps in the
much studied Belousov-Zhabotinsky oscillating reaction ri05]. The
reaction produces bromomalonic acid and under the conditions of the
reaction ([H2SO«]_ = about 0.8 M) is first-order in both malonic acid
and H+ ion, but does not depend on the concentration of halogen
[ 105]. The rate constant is 0.013 M^.s"1. This suggests that the
rate-controlling step in the reaction is the acid-catalyzed
enolization of malonic acid and that the enol is subsequently
rapidly halogenated.
HOOC-CH2-COOH > HOOC-CH=C(OH)2 slow
HOOC-CH=C(OH)2 + X2 * (HOOC)2CH-X fast
In a different study, the bromination of malonic acid between pH 1
and 2 turned out to be first order in malonic acid, but independent
of H+ ion and Br2 [106] with k =0.002 s"1. This result apparently
conflicts with the previous one. It seems likely that chlorination
in the same pH range (0 to 3) proceeds in a similar way to
bromination. However, these conditions are far removed from those
prevailing in drinking water.
At pH 7.5, [C12]total - 0.0002-0.002 M, and [malonic acid]total =
0.0004-0.002 M, the rate is dependent on the concentrations of both
malonic acid and active chlorine [45]
rate = k[malonic acid]total[Cl2]total
These kinetics imply that the rate-determining step is no longer
enolization of malonic acid but the reaction of the enol with the
chlorinating agent. The rate constant for water catalysis is 0.13
45
-------
M"1s"1. Application of this constant to the situation in chlorinated
drinking water (Table 1) with a limiting concentration of malonic
acid provides the rate equation
rate = (1.3 x 10"6 s"1) [malonic acid]total
This would lead to a half-life of 148 hours or 6.2 days. Thus,
this reaction may occur to a small extent in drinking water.
Pedersen [107] has examined the bromination of acetoacetic
acid. This reaction has no dependence of the amount of Br2 present
and is catalyzed by general bases, but not by acids (variations in
the concentrations of HC1 had no influence on the rate) . He
postulates that the rate-determining step is the base-catalyzed
enolization of undissociated acetoacetic acid. In this case, the
water-catalyzed rate constant extrapolated to pH 7.5 and 25° C is
6.1 x 10"8 s"1 which is about 21 times less than the corresponding
constant for malonic acid in these circumstances. Thus, the water-
catalyzed reaction of acetoacetic acid with chlorinated water has
a half-life of about 130 days.
However, Bell and Lidwell [52] extrapolated Pedersen's data to
25° C and derived a Bronsted equation for the reaction. This
equation predicts a catalytic constant for OH" ion (koH) of about 3
x 106 M"1s"1. If the reaction in chlorinated water (conditions of
Table 1) is similar to bromination, OH" catalyzed reaction would
predominate over the water- or general base-assisted reaction. The
rate expression for limiting concentrations of acetoacetic acid is
rate = (1.2 x 10"* s"1) [acetoacetic acid]total
The half-life of the reaction would be 1.6 hours. On the basis of
this analysis, reaction of beta-keto acids with chlorinated water
is well within a reasonable time frame. However, it must be
emphasized that this analysis is highly speculative since it
involves an extrapolation over a wide range and depends upon HO"
conforming to a Bronsted relation which it sometimes does not.
Cyanp groups have an activating effect on alpha hydrogens
similar to that of ketones [108] and thus cyanoacetic acid may also
be substituted in chlorinated water. The same conclusion applies
to nitroacetic acids since the nitro group is even more effective
in labilizing hydrogens than is the cyano or keto group.
2. Monochloramine
The species present in water treated with ammonia and chlorine
generally seem to be less reactive than those in ordinary
chlorinated water. Since simple carboxylic acids appear to be
inert to the latter, it seems probable that chloramine-treated
water will not induce reactions with these acids either.
46
-------
The various molecules in "chloramine"-treated drinking water
might also interact with the more highly activated carboxylic acids
having keto, carboxyl, cyano, nitro, etc. groups attached to their
alpha- or 2-carbon atoms. Unfortunately, however, the information
necessary to make such a judgment does not appear to be available.
3. Conclusions
Reactions of carboxylic acids with drinking water treated with
chlorine or with ammonia and chlorine do not appear to occur very
rapidly and therefore, are probably not too significant. However,
some activated alpha-substituted acids may be reactive enough to be
partially chlorinated under these conditions.
K. Carboxylic Esters
1. Aqueous Chlorine
a. Simple Carboxylic Esters. Unlike the carboxylic acids,
the esters do not have a hydroxyl hydrogen that can be replaced by
a halogen to yield a hypochlorite. However, there are carbon-
linked hydrogens in both the alkyl and acyl groups of most esters.
These are replaced by photochemical free radical chlorination in
either the vapor, or liquid phase [108]. Chlorination occurs
principally in the 2- and 3-positions of the acyl moiety. Similar
reactions in polar media do not seem to have been reported.
Aqueous chlorine does not appear to substitute hydrogens in
carboxylic esters.
The acyl group of esters is subject to nucleophilic addition.
The resulting intermediate can decompose by loss of the alkoxyl
group and formation of a new acid derivative.
O O" O
I | II
R-C + Nu:' »> R-C-NU ! * R-C + RO'
I II
OR OR NU
The principal nucleophiles in aqueous chlorine are water and HO'and
CIO" ions. Jencks and Carriuolo [31] have compared the reactivity
of these species with the ester, p-nitrophenyl acetate (PNPA) (R'=
CH3-; R = p-O2N-C6H4-) . The second-order rate constants are
lr — 9fi fi M'^c;"1
JX£JQ ~~ £• \J . \J 11 O.
koH = 14.8 M'V1
kHOH = 1 x 1(T8 M'V1
At the concentrations in chlorinated drinking water (Table 1), the
47
-------
ClO'-catalyzed hydrolysis of the ester is about 30 tiroes faster
than HO" catalysis, and about 250 times faster than the water
reaction. The rate is given by
rate = (1.4 x 10'* s"1) (PNPA)
This would result in a reaction half-life of about 1.4 hours which
is well within the residence time of drinking water.
p-Nitrophenyl acetate is very reactive hydrolytically because
p-nitrophenoxide is an excellent leaving group. More common esters
such as ethyl acetate have lower reactivities. Thus, the koH for
ethyl acetate is 0.12 M"1s"1 [109] or about 120 times less reactive
than p-nitrophenyl acetate. Ethyl benzoate is more than ten-fold
less reactive than ethyl acetate [110]. In these less reactive
esters, the leaving group—an alkoxide ion—is more basic than the
nucleophile in contrast to the situation with p-nitro-phenyl
acetate where the leaving group—a phenoxide ion—is less basic
than the attacking reagent. Jencks and Gilchrist [111] have found
that when the leaving group is more basic than the nucleophile that
the reactivity difference between nucleophiles is about five times
as great as when the leaving group is less basic. Thus, with these
more common esters, the reactivity ratio of ClCT and HO' should be
about 9.0 instead of only 1.8 as it is for the p-nitrophenyl ester.
Thus, hydrolysis catalyzed by CIO" ion should be even more
dominant. Based on these approximations and koH of ethyl acetate
(0.12 M^s"1) , kdo for this ester should be about 1.1 M^s"1. In
chlorinated drinking water (Table 1) the rate of reaction of the
ester if its concentration is limiting would be
rate = (5.5 x 10"6 s"1) [ethyl acetate]
The half-life of the reaction is about 35 hours. Ethyl benzoate is
at least 10 times less reactive. Ester hydrolysis catalyzed by
CIO" ion will occur to a considerable extent in aqueous chlorine.
b. Carboxylic Esters with "Active Methylene" Groups. The
esters derived from beta-ketoacids, 2-cyanoacids, 2-nitroacids, and
malonic acids may possess hydrogens on their alpha-or 2-carbon
atoms that can be easily replaced by halogens just as was the case
with the corresponding acids.
Diethyl malonate (malonic ester) reacts with halogens to
undergo displacement of the methylene H's by halogens. Either one
or both H's may be substituted. Bell and coworkers [112-114] have
examined the reaction of malonic ester with bromine and chlorine.
The reactions are catalyzed by bases and at high halogen
concentrations they are independent of the amount of halogen, but
at low concentrations dependence on the halogen is noted. This
information suggests that the ester is converted to the enolate
and/or enol by the base, and that these intermediates then react
48
-------
with the halogen.
R-CH2-C-OEt + B" v N R-CH=C-OEt + HB N ^ R-CH=C-OEt + B'
1 I I
O O' O-H
R-CH=C-OEt + X2 > R-CH-C-OEt + H-X
I I I ' '
O-H X O
R-CH=C-OEt + X2 * R-CH-C-OEt + X'
I I I
O' X O
If the halogen concentration is high, the halogenation will be fast
and the enolization will be rate-determining; but if [X2] is small,
then the halogenation steps become slow and rate-limiting. Bell
and Yates [114] studied chlorination in dilute solutions where most
of the active chlorine is present as HOC1. They concluded that
"the reactivities of C12 and HOCl...do not differ much." Using
their constants, it can be shown that under the conditions
prevailing in drinking water (Table 1) that the water-catalyzed
reaction" of the ester is independent of active chlorine. Their
rate constant for water catalysis is 2.8 x 10"5 s"1 so the rate
equation is
rate = (2.8 x 10"5 s"1) [malonic ester]
This would result in a half-life for the reaction of about 6.9
hours. The reaction is fast enough to be of importance in drinking
water.
This group of workers did not determine the catalytic
efficiency of HO" ion. If this ion fits on a Bronsted plot for the
buffer bases that they did use, koH would be about 1 x 107 M"1s"1.
Calculation indicates that HO'-catalyzed chlorination by drinking
water (Table 1) would depend on the amount of active chlorine and
the final rate expression would be
rate = (0.01 s"1) [malonic ester]
The half-life for this reaction is only about a minute. However,
the calculations leading to this equations are very speculative—
they involve a long extrapolation and the assumption that HO" ion
behaves like a normal Bronsted base, which is sometimes not true.
The second halogen is more easily introduced than the first in
base-catalyzed halogenation. This is because enolate anion of the
monohalo 'compound is stabilized by the electronegative halogen and
thus, it is easier to form. In the chlorination of malonic ester,
the rate constant for the substitution of the second chlorine is
49
-------
about 13 times as large as for the first one.
Ethyl acetoacetate reacts with halogens in the same way that
malonic ester does—the methylene H's are replaced in consecutive
steps by halogen. The bromination has been studied by Pedersen
[115] and at the experimental concentrations he used, the rate was
independent of Br2, first-order in ester, and base-catalyzed.
These similarities to the halogenation of diethyl malonate indicate
that a similar mechanism is involved, namely base-catalyzed, rate-
controlling enolization of the ester followed by fast reaction of
the enol and/or enolate with the halogen. Since the rate is
independent of the amount of Br2, the reaction of acetoacetic ester
with C12 should be the same kinetically. The water-catalyzed
bromination of the ester has a rate constant of 5.2 x 10"* s"1.
Since this simply represents enolization, chlorination should have
the same rate constant and the rate would be
rate = (5.2 x 10"* s"1) [ethyl acetoacetate]
This translates into a half-life of about 22 minutes.
Ethyl cyanoacetate and ethyl nitroacetate have water-catalyzed
bromination rates of 1.2 x 10"3 s"1 and 6.3 x 10"3 s"1, respectively
C1081. These are larger than the values for ethyl acetoacetate and
thus they will undergo chlorination even more readily.
In conclusion, esters with activated methylene groups are
fairly easily halogenated by chlorinated drinking water. The
resulting chlorinated compounds are quite reactive—halogens
situated on carbons next to carbonyl groups are readily replaced by
nucleophiles. Thus, these compounds are good alkylatihg agents.
Many lachrymators (e.g., tear gas) are substances of this type.
2. Monochloramine
The chlorinating species in drinking water treated with
ammonia and chlorine are usually less reactive than those in
agueous chlorine. Agueous chlorine did not exhibit significant
activity towards simple carboxylic esters and so it is unlikely
that these esters will be affected to any important degree by
agueous chloramine.
The same may not be true for carboxylic esters with active
methylene groups. There are reports of the formation of
aminomalonates by reaction of diethyl malonate anion with NH2C1 in
ether [116, 1171.
(EtOOC)2CH" + NH2-C1 * (EtOOC)2CH-NH2 + Cl"
This may be an example of a nucleophilic substitution on the
nitrogen of chloramine or alternatively, on the chlorine to form a
50
-------
chloromalonic ester which then undergoes replacement of the
chlorine by NH3. Similar reactions in aqueous solution have not
been reported.
The same research group has reported a somewhat different type
of reaction of chloramine with ethyl acetoacetate in ether in the
absence of base [118].
CH3-CO-CH2-COOEt + NH2C1 >• CH3-CO-NH2 + Cl2CH-COOEt
Apparently, the methylene group is first fully chlorinated by the
chloramine and then the keto-ester is cleaved by NH3 into an amide
and an ester. This suggests that chloramine might chlorinate these
active methylene compounds in aqueous solution. However,
information on this is lacking and as mentioned above, the
chlorinating molecules in aqueous chloramine are generally less
reactive than those in chlorinated water.
Unfortunately, not enough information is available on the
behavior of aqueous chloramine to make any quantitative conclusions
about the fate of carboxylic esters in this medium.
3. Conclusions
The hydrolysis of simple carboxylic esters is enhanced in
aqueous chlorine because of the presence of hypochlorite ion.
Otherwise, these compounds are lacking in reactivity in drinking
water treated with chlorine alone or with chlorine and ammonia.
Carboxylic esters in which the alpha-carbon atom is attached
to another ester group or a keto, nitrile, or nitro group are much
more reactive and generally will be chlorinated on the alpha-carbon
atom in chlorinated drinking water. Whether, the same type of
reaction *can occur in chloramine-treated water is not certain.
L. Sulfur Compounds
1. Aqueous Chlorine
Some toxic substances, like the pesticides Aldicarb and
Demeton, contain reduced sulfur compounds which can be oxidized to
several different oxidized products. For instance, sulfides can be
oxidized as shown below:
O O
10] I (0) 1
R-S-R *> R-S-R > R-S-R
51
-------
Dialkyl disulfides can be oxidized on both sulfur atoms with four
oxidizing equivalents in a similar fashion. Thiols can be oxidized
as follows:
O O O
(0) i [0] II [01 II
R-S-H > R-S-H > R-S-H > R-S-CT
I fl
O O
It is probably because of this complexity that rate constants
for the reactions of aqueous chlorine with thiols and sulfides are
unknown. Nevertheless, the reaction of aqueous chlorine is used as
an effective reagent for the rapid and quantitative removal of
hydrogen sulfide from sulfurous water. Model compounds will
therefore be used to estimate rate constants.
Thiocyanate may be a good model for the reaction of thiols,
although definitive studies of the reaction of thiocyanate ion with
aqueous chlorine have not been performed. Kobayashi and Okuda
[119] have found that thiocyanate requires 4 equivalents of aqueous
chlorine to fully oxidize it at low pH. The products are sulfate
and cyanide ions. Several studies indicate that the rate-
'S-C=U + 4 HOC1 *• S
-------
range 106 to 108 M^sec"1. If a toxic thiol or sulfide has a
concentration of 5 x 10"7 M and if its reaction with all free
available chlorine species follows simple second order kinetics
(one sulfur reduces one oxidant molecule) and if a rate constant of
106 M^sec"1 is assumed, then half of it would be oxidized by aqueous
chlorine under pseudo-first order conditions in a drinking water at
pH 7.5 in 0.14 sec, while 90% will react in 0.46 sec.
The sulfenic acid (R-SOH) and dialkyl sulfoxide (R2SO)
oxidation products can be expected to be oxidized further.
However, the more oxidized the sulfur, the less rapid it would be
expected to react with aqueous chlorine. This supposition is
supported by the work of Kice et al. f 125] who studied the reaction
of aqueous chlorine with benzenesulfinic acid sodium salt
(Ph-SO2~, Na+) at pH values ranging from 5.2 to 9.0 at 25 °C.
Ph-S' + CIO" > Ph-S-CT + Cl~
II I
o o
The reaction is pH dependent with the fastest rates observed at
higher pH values. This suggests that the more active form of
chlorine is hypochlorite. The second order rate constant for the
reaction of hypochlorite is 1.05 x 103 M^sec"1, while the reaction
of HOCl occurs with a rate constant 350 times smaller (3 M^sec"1) .
Using these rate constants, a half-life in drinking water of 2.2
min is predicted for the more highly oxidized analogues of the
organic thiols and sulfides. Greater than 90% would be reacted in
7.3 min. For this reason, it is safe to say that reduced
organosulfur compounds will be completely oxidized under drinking
water conditions.
2. Monochloramine
Work of Jacangelo and Olivieri F1261 suggests that a 5-fold
excess of the sulfur-containing amino acids cysteine, cystine, or
methionine completely reduced monochloramine within 2 min. They
determined that each monochloramine oxidized two cysteine residues
to one cystine which contains the -S-S- bond. In the presence of
excess oxidant cystine can be oxidized further to thiolsulfinates,
sulfinyl sulfones, sulfinic and sulfonic acids [127]. They
describe these reactions as a critical part of the mechanism by
which bacteria are inactivated by disinfectants.
Yuki et al. [127] studied the reaction of N-chlorosuccinimide
and chloramine-T with the amino acid cysteine under physiological
conditions to model the anti-viral activity of chloramines.
Cysteine was oxidized to cystine, cystine disulfoxide, and cysteine
sulfinic acid. The anti-viral activity of chloramines was
53
-------
completely inhibited by 50% calf serum, which they attributed to
reduction by cysteine residues in proteins.
Because monochloramine is a less potent oxidizing agent than
free available chlorine, its reduction by sulfur-containing amino
acids is likely to be considerably slower than reduction of free
available chlorine. Stanbro and Lenkovitch [128] have shown that
bisulfite reduces organic N-chloramines at much slower rates than
previously expected. They demonstrated that iodimetric methods for
the measurement of residual oxidant concentration gave erroneous
results. Several groups [129-130] have studied the rate of
reduction of chloramine-T (an N-chlorosulfonamide) by dialkyl
sulfides. The second order rate constant for the first oxidation
of dimethylsulfide by the chloramine is 2.7 x 10* M^sec"1 [129]. If
inorganic monochloramine reacts as rapily, a toxic sulfide (5 x 10"7
M) would have a half-life in a drinking water of 5.1 sec and 90%
would react in 17 sec. The rate of oxidation of thiols by
monochloramine may be sensitive to pH due to ionization of the
sulfhydryl group.
3. Conclusion
Mechanistic studies of the rates of reaction of alkyl sulfides
with aqueous chlorine and monochloramine are unknown. By
extrapolation from model compounds, the rates are believed to be
extremely rapid with half-lives of less than a second. More highly
oxidized sulfur compounds are slower to react with aqueous
chlorine, but their reactions are still very rapid with predicted
half-lives of a few minutes. Model studies with chloramine-T
suggest that even monochloramine would react extremely rapidly with
alkyl sulfides.
Lin and Carlson [96] have studied the reactivity of several
heterocyclic aromatic sulfur compounds. Their reactivity is
discussed, in an earlier section. The reactivity of S-alkyl
thiocarbamates is probably less than that of simple alkyl sulfides,
because of the deactivation of the carbonyl. The S-alkylphosphoro-
thioate, TETP, discussed below may be a suitable model, in which
case, the reaction is too slow to be of significance. There are no
good models of O-alkylthiocarbamates. More work is needed to
determine the half-lives of these compounds in a drinking water
disinfected with aqueous chlorine or monochloramine.
M. Organophosphorus Compounds
1. Aqueous Chlorine
Organophosphorous compounds of varying structure form an
important class of toxic substances. The kinetics and mechanisms
54
-------
of the reactions of several compounds of this class have been
studied in detail.
Hypochlorite solutions have been used to detoxify equipment
contaminated by the phosphoramide poison schradan or dimefox,
( (CH3)2N)2POF, a compound with potent anticholinesterase activity
[131].
Epstein et al. [132] have studied the reaction of the nerve
gas, Sarin (isopropyl methylphosphonofluoridate) with hypochlorite.
Sarin hydrolyzes much more rapidly in slightly acidic or slightly
basic solutions of chlorine than in water. The tentative reaction
mechanism used to explain the first order dependence of the
reaction on both Sarin and hypochlorite is shown below:
O O O
1 slow II 1
i-Pro-P-F + cio" - * i-PrO-p-oci - *• i-PrO-p-cr + Hoci
I I I
CH3 CH3 CH3
An unstable phosphoryl hypochlorite intermediate is proposed which
rapidly hydrolyzes. Hypochlorite acts like a true catalyst and is
regenerated in the reaction. The tentative rate law can be written
as:
rate = k[C!O~] [Sarin]
= k orclo [FAC] [Sarin] (57)
where k = 570 M'hnin"1 or 9.5 M^sec"1, an average of eleven
determinations at pH values from 5.0 to 9.0. Because hypochlorite
is not consumed in the reaction, the observed kinetics are first
order, no matter what the Sarin concentration is. The half -life of
Sarin in.;-a drinking water would be 4.05 hours.
Aqueous chlorine reacts with S-alkylphosphorothioates. Lordi
and Epstein have studied the hydrolysis of triethylphosphorothio-
ate (TETP) by aqueous chlorine [133 ] .
O O
I slow I
ETO-P-SEt + 3 C12 - »> EtO-P-OH + EtSO3H + 6 HC1
I H20 |
OEt OEt
One problem with interpreting their rate data is their measurement
of the loss of FAC rather than the loss of the organophosphate and
there are three consecutive processes in the reaction described
55
-------
below which consume FAC. Nevertheless, the mechanism of the
reaction between pH 5 and 8 appears to depend primarily on the
reaction of TETP with C12. Studies in which the pH and the
chloride ion concentration were varied suggested that HOCl is not
a reactant. The detailed mechanism for the formation of C12 in
aqueous solution as discussed in an earlier section had not been
elucidated by Eigen and Kustin until after the work of Lordi and
Epstein, but their data for the reactivity of TETP appears to
support the simple rate law for reactivity between pH 5 and 8:
rate = k[C!2] [TETP] (58)
The proposed rate-determining step of the chlorination of TETP is:
O 0 Cl
i k | |
ETO-P-SEt + C12 * EtO-P-S-Et
I I
OEt OEt
where k, the specific rate constant for the chlorination of TETP,
is 1.5 x 108 M^min"1 or 2.5 x 106 M^sec"1. Since [TETP] is small,
the reaction is first order in TETP concentration and equations 14
and 5 discussed above apply. From this a half-life of 16.9 days is
calculated.
While this compound may be considered a good model for the
reaction of the pesticide malathion which is an S-alkylated
thiophosphate, it is not clear whether this compound is a good
model for parathion, an O-alkyl thiophosphate.
2. Monochloramine
Organic chloramines are reported to undergo reaction with a
wide variety of organophosphorus compounds [34.]. However, most of
the reported reactions have been carried out in non-aqueous media
and few rate studies are available.
Epstein et al. [132] have also examined the rate of
detoxification of Sarin with chloramine-T. They found no
enhancement of the rate of aqueous hydrolysis in the presence of 3
x 10"3 M chloramine-T. For this reason, it is expected that
monochloramine has . no effect on the decomposition of phosphono-
fluoridates in a drinking water.
3. Conclusion
Kinetic studies of the reactions of organophosphorus compounds
with disinfectants are scarce. Although hypochlorite solutions
have been used to detoxify surfaces exposed to such poisons as the
phosphoramide schradan or dimefox, mechanistic rate studies of only
56
-------
two organophosphorus compounds are known. S-Alkylphosphorothioates
do not appear to undergo significant transformation under drinking
water conditions. On the other hand, the nerve gas, Sarin
(isopropyl methylphosphonofluoridate) does react with hypochlorite
at a rate sufficient to make it significant in a drinking water.
V. SUMMARY AND OVERALL CONCLUSIONS
A treated drinking water will spend several hours in a
treatment plant and possibly 5 days in a typical municipal
distribution system. Based on this time frame the various
functional groups discussed above can be placed in various
tentative classes of reactivity. These will be:
Very Reactive >50% reacted in <5 min
Reactive >50% reacted in <1 day
Somewhat Reactive >50% reacted in <5 days
Slightly Reactive >50% reacted in <1 year
Unreactive
Unknown compounds for which reactions with
disinfectants are known, but for
which information about their
reactivity in dilute aqueous
solution is unknown.
57
-------
TABLE 3. OVERALL REACTIVITIES OF FUNCTIONAL GROUPS
Aqueous Chlorine
Very Reactive Amines
Alkyl sulfides
/3-diketones
/?-ketoesters
malonic ester analogs
some nitrogen hetero-
cyclic aromatics
Reactive
phenols
p-nitrophenylacetate
alkoxybenzenes
/3-ketoacids
phosphonofluoridates
Monochloramine
Alkyl Sulfides
amines
some nitrogen hetero-
cyclic aromatics
Somewhat
Reactive
a,a-diketones
aliphatic esters
S-alkylphosphonofluoridates
Slightly
Reactive
Unreactive
Unknown
aromatic esters
amides
carbamates (?)
alcohols
simple ketones and aldehydes
alkanes
alkyl halides
xylene
simple carboxylic acids
alkenes
alcohols
alkanes
alkyl halides
aromatic compounds
phenols
alkenes
amides
carboxylic acids
esters
aldehydes & ketones
58
-------
Table 3 contains only the functional groups for which kinetic
information is known. It does not cover all the functional groups
found in pesticides and toxic substances. Examples of classes of
functional groups for which little kinetic information is known
include phosphorus- and sulfur-containing compounds. The
discussion in the sections describing individual functional groups
was designed to describe the individual structural factors which
can affect the reactivity of individual toxic substances.
59
-------
VI. REFERENCES
1. Pontius, F.W. "Complying with the New Drinking Water Quality
Regulations," J. Amer. Water Works. Assoc. 82(2). 32-52
(1990).
2. Symons, J.M. et al. J. Am. Water Works Assoc. 1975, 67, 634-
47.
3. American Water Works Assoc. "New Position Statements Cover
Water Quality Parameters," Mainstream. 34(5). 8 and 11 (1990).
4. Connick, R. E. ; Chia, Y. "The Hydrolysis of Chlorine and Its
Variation with Temperature," J. Am. Chem. Soc. 81. 1280-1284
(1959).
5. Jencks, W. P.; Regenstein, J. "lonization Constants of Acids
and Bases," in Handbook of Biochemistry; Sober, H. A., Ed.;
Chemical Rubber Co.: Cleveland, OH, 1970; p J-190.
6. Snoeyink, V.L.; Jenkins, D. Water Chemistry (New York: John
Wiley & Sons, 1980), p.6.
7. Eigen, M.; Kustin, K. "The Kinetics of Halogen Hydrolysis,"
J. Am. Chem. Soc. 84., 1355-1361 (1962).
8. Gray, E. T. ; Margerum, D. W. ; Huffman, R. P. "Chloramine
Equilibria and the Kinetics of Disproportionation in Aqueous
Solution," in Organometals and Oraanometalloids. Occurrence
and Their Fate in the Environment; Brinckman, F. E. and
Belloma, J. M., Eds. Washington, DC: American Chemical
Society, ACS Symposium Series 82, 1978; pp 264-277.
9. Hand, V. C.; Margerum, D. W. "Kinetics and Mechanism of the
Decomposition of Dichloramine in Aqueous Solution," Inorq.
Chem. 22. 1449-1456 (1983).
10. Kumar, K. ; Shinness, R. W. ; Margerum, D. W. "Kinetics and
Mechanism of the Base Decomposition of Nitrogen Trichloride in
Aqueous Solution," Inorq. Chem. 26. 3430-3434 (1987)
11. Yiin, B. S.; Margerum, D. W. "Non-metal Redox Kinetics:
Reactions of Trichloramine with Ammonia and with
Dichloramine," Inora. Chem. 29. 2135-2141 (1990).
12. Margerum, D. W. ; Gray, E. T. ; Huffman, R. P. "Chlorination
and the Formation of N-Chloro Compounds in Water Treatment."
In: Organometals and Oroanometalloids Occurrence and Their
Fate in the Environment; Brinckman, F. E. and Belloma, J. M.,
Eds. Washington, DC: American Chemical Society, ACS Symposium
Series 82, 1978; pp 278-291.
60
-------
13. Kumar, K. ; Margerum, D. W. "Kinetics and Mechanism of General
Acid-Assisted Oxidation of Bromide by Hypochlorite and
Hypochlorous Acid," Inorg. Chem. 26. 2706-2711 (1987).
14. Kumar, K.; Day, R. A.; Margerum, D. W. "Atom Transfer
Kinetics: General-Acid-Assisted Oxidation of Iodide by
Chloramines and Hypochlorite," Inorg. Chem. 25f 4344-4350
(1986).
15. Nagy, J. C. ; Kumar, K. ; Margerum, D. W. "Non-Metal Redox
Kinetics: Oxidation of Iodide by Hypochlorous Acid and by
Nitrogen Trichloride Measured by the Pulsed-Accelerated-Flow
Method," Inora. Chem. 27. 2773-2780 (1988).
16. Yiin, B. S.; Margerum, D. W. "Non-metal Redox Kinetics:
Reaction of Sulfite with Dichloramines and Trichloramine,"
Inora. Chem. 29. 1942-1948 (1990).
17. Coleman, G. H. "The Reaction of Alkylchloramines with Grignard
Reagents," J. Am. Chem. Soc. 55. 3001-3005 (1933).
18. Ingold, C. K. Structure and Mechanism in Organic Chemistry.
2nd ed. Ithaca, NY: Cornell University Press, 1969; pp 341-
344.
19. Fries, K. ; Vogt, W. "Chlorides and Bromides of Diphenyl
Sulfide," Liebias Ann. Chem. 381. 337-346 (1911).
20. Brown, L. O. ; Soper, F. G. "Studies of N-Halogenocompounds.
Part V. The Interaction of Dialkylchloramines with Phenols,"
J. Chem. Soc. 1953. 3576-3580 (1953).
21. Reid, E. E. Organic Chemistry of Bivalent Sulfur (New York:
Chemical Publishing Co., 1958), pp 125-126.
22. Ramsden, J. H.; Drago, R. S.; Riley, R. "A Kinetic Study of
Sulfide Oxidation by Sodium Hypochlorite Using Phase Transfer
Catalysis," J. Am. Chem. Soc. 111. 3958-3961 (1989).
23. Gassman, P. G.; Campbell, G. A.; Frederick, R. C. "Thermal
Rearrangement of N-Chloroanilines. Evidence for the
Intermediacy of Nitreniura Ions," J. Am. Chem. Soc. 94. 3891-
3896 (1972).
24. Kudesia, V. P.; Mukherjee, S. K. "Kinetics of Reaction of
Isopropyl Alcohol with Agueous Chlorine," Indian J. Chem. ISA,
513-.516 (1977).
25. Nweke, A.; Scully, F.E. "Stable N-Chloraldimines and Other
Products of the Chlorination of Isoleucine in Model Solutions
and in a Wastewater," Environ. Sci. Technol. 23. 989-994
(1989).
61
-------
26. Newman, M. S.; Holmes, H. L. "beta-Naphthoic Acid," in Organic
Syntheses; Blatt, A. H., Ed.; Wiley: New York, 1943; Coll.
Vol. II, pp 428-430.
27. de la Mare, P. D. B. Electrophilic Halogenation (London:
Cambridge University Press, 1976), p 100-101.
28. Theilhacker, W. ; Wegner, E. "Organic Syntheses Using
Chloramine.11 In Newer Methods of Preparative Organic
Chemistry. Vol. 3, Foerst, W., Ed. New York: Academic Press,
1964; pp 303-317.
29. Appel, R.; Haenssgen, D. Nonmetallic Imino Compounds. XXXVIII.
Simplified Chloramination of Amines and Thioethers," Chem.
Ber. 103. 3733-3734 (1970).
30. Anbar, M.; Yagil, G. "The Hydrolysis of Chloramine in Aqueous
Solution," J. Am. Chem. Soc. 84. 1790-1796 (1962).
31. Jencks, W. P.; Carriuolo, J. "Reactivity of Nucleophilic
Reagents toward Esters," J. Am. Chem. Soc. 82. 1778-1786
(I960).
32. Beale, J. H. "The Reactivity of N-Chloro- and N-Methyl-
benzenesulfonamide Anions with Methyl Methanesulfonate in
Methanol," J. Org. Chem. 37. 3871-3872 (1972).
33. Fonouni, H. E.; Krishnan, S.; Kuhn, D. G. ; Hamilton, G. A.
"Mechanisms of Epoxidations and Chlorinations of Hydrocarbons
by Inorganic Hypochlorite in the Presence of Phase Transfer
Catalysts." J. Am. Chem. Soc. 105. 7672-7676 (1983).
34. Kovacic, P.; Lowery, M. K. ; Field, K. W. "Chemistry of N-
Bromamines and N-Chloramines," Chem. Rev. 70. 646-647, 649-
651, 658-661 (1970).
35. Yagil, G.; Anbar, M. "The Kinetics of Hydrazine Formation from
Chloramine and Ammonia," J. Am. Chem. Soc. 84. 1797-1803
(1962).
36. White, R. E., and Kovacic, P. "Reaction of N,N-
Dichlorourethane and of Diethyl Azodicarboxylate with
Alkoxide," J. Am. Chem. Soc. 96, 7284-7288 (1974).
37. Standard Methods for Examination of Water and Wastewater. 17th
ed.; Clesceri, L. S.; Greenberg, a. E.; Trussell, R. R. , Eds.
Washington, DC: American Public Health Assn., 1989; pp 4-62 to
4-64.
38. Moore, H. E. ; Garmendia, M. J. ; Cooper, W. J. "Kinetics of
Monochloramine Oxidation of N,N-Diethyl-p-phenylenediamine,"
Environ. Sci. Technol. 18. 348-353 (1984).
62
-------
39. Anbar, M. Dostrovsky, I. "A Kinetic Study of the Formation
and Hydrolysis of tert.-Butyl Hypochlorite," J. Chem. Soc.
1954. 1094-1104 (1954).
40. Churganova, S. S.; Lopyrev, V. A. "Acid-base Catalysis in the
Formation and Hydrolysis of Ethyl Hypochlorite," Zh. Obshch.
Khim. 44. 2758-2762 (1974).
41. Chattaway, D. C.; Backeberg, O. G. "Alkyl Hypochlorites,"
J.Chem. Soc. 123. 2999-3003 (1923).
42. Mohrig, D. M.; Nienhuis, C. F.; Linck, C. V. Z.; Fox, B. G.
"The Design of Laboratory Experiments in the 1980's. A Case
Study of the Oxidation of Alcohols with Household Bleach," J.
Chem. Ed. 62. 519-521 (1985).
43. Stevens, R. V.; Chapman, K. T.; Weller, H. N. "Convenient and
Inexpensive Procedure for Oxidation of Secondary Alcohols to
Ketones," J. Ora. Chem. 45. 2030-2032 (1980).
44. Kudesia, V. P.; Mukherjee, S. K. "Kinetics of Reaction of
Isopropyl Alcohol with Aqueous Chlorine," Indian J. Chem. 15A.
513-516 (1977).
45. White, H. S. private communication.
46. Lapworth, J. Chem, Soc. 35. 30 (1904).
47. Rice and Fryling, J. Am Chem. Soc. 47. 382 (1925).
48. P.O. Bartlett, and J.R. Vincent, "Enolization as Directed by
Acid and Basic Catalysts. I. The Acid-Catalyzed Enolization of
Menthone, J. Am. Chem. Soc. . 5_5, 4992-4997 (1933).
49. P-D-s Bartlett, "Enolization as Directed by Acid and Basic
Catalysts. II. Enol Mechanism of the Haloform Reaction, J. Am.
Chem. Soc.. 56, 967-969 (1934).
50. P.O. Bartlett and J.R. Vincent, "The Rate of the Alkaline
Chlorination of Ketones," J. Am. Chem. Soc.. 57. 1596-1600
(1935).
51. R.P. Bell and K. Yates, "Kinetics of the Chlorination of
Acetone in Aqueous Solution," J. Chem. Soc.. 1962. 1927-1933
(1962).
52. Bell, R.P.; Lidwell, O.M. "The Base Catalyzed Prototropy of
Substituted Acetones," Proc. Rov. Soc. A. 176. 88-113 (1940).
53. Boyce, S.D., Hornig, J.F. "Formation of Chloroform from the
Chlorination of Diketones and Polyhydroxybenzenes in Dilute
63
-------
Aqueous Solution." In: Water Chlorination; Environmental
Impact, and Health Effects. Vol. 3, R.L. Jolley et al., Eds.
Ann Arbor, MI: Ann Arbor Science, 1980; pp. 131-140.
54. Morris, J.C. "Reaction Dynamics in Water Chlorination." In:
Water Chlorination: Chemistry. Environmental Impact, and
Health Effects. Vol. 5, R.L. Jolley et al., Eds. Chelsea, MI:
Lewis Publishers, 1985; pp. 701-711.
55. Engfeldt, N.O. "Die Wirkung der Dakinschen Hypochloritlosung
auf gewisse organische Substanzen, " Z. Physiol. Chem. 121.
18-61 (1922).
56. Lindsay, M. and Soper, F.G. J. Chem. Soc.. 1946f 791-???.
57. Hauser, C.R. and Mauser, M.L. "Researches on Chlorimines. I.
ortho-Chlorobenzalchlorimine and Anisalchlorimine," J. Am.
Chem. Soc.. 52, 2050-2054 (1930).
58. Hauser, C.R. and Hauser, M.L. J. Am. Chem. Soc.. 52. 4158-
???? (1930).
59. Hauser, C.R. The Preparation of Primary Amines from Aldehydes
and Monochloro-Amine," J. Am. Chem. Soc.. 52. 1108-1111
(1930).
60. Poziomek, E.J.; Kramer, D.N.; Mosher, W.A. "Pyridine Ald-
Chlorimines," J. Pro. Chem.. 25, 2135-???? (1960).
61. Le Cloirec, C. and Martin, G. "Evolution of Amino Acids in
Water Treatment Plants and the Effect of Chlorination on Amino
Acids." In: Water Chlorination; Chemistry. Environmental
Impact, and Health Effects. Vol. 5, R.L. Jolley et al., Eds.
Chelsea, MI: Lewis Publishers, Inc., 1985; pp. 821-834.
62. Presented before the Division of Environmental Chemistry at
the 200th National Meeting of the American Chemical Society,
Washington, DC, August 25-31, 1990.
63. Isaac, R.A. and Morris, J. C. "Transfer of Active Chlorine to
Nitrogenous Organic Compounds. 2. Mechanism. Environ. Sci.
Technol. 19, 810-814 (1985).
64. Morris, J. C. "Kinetics of Reactions Between Aqueous Chlorine
and Nitrogen Compounds." In: Principles and Applications of
Water Chemistry. Faust, S.D. and Hunter, J. V., Eds. New
York: John Wiley & Sons, Inc., 1967; pp. 23-53.
65. Margerum in Ref. 11 reports a rate constant of 5 x 107 M"1
sec"1-
64
-------
66. Scully, F.E., Jr. and Bempong, M.A. "Stability of Aqueous
Solutions of N-Chloropiperidine and N-Chlorodiethylamine with
Varying pH." In: Water Chlorination; Environmental Impact.
and Health Effects. Vol. 3, R.L. Jolley et al., Eds. Ann
Arbor, MI: Ann Arbor Science, 1980; pp. 203-208.
67. Stelmaszynska, T. and Zgliczynski, J. M. "N-(2-Oxoacyl)amino
Acids and Nitriles as Final Products of Dipeptide Chlorination
Mediated by the Myeloperoxidase/H2O2/Cl" System," Eur. J.
Biochem. 92., 301-308 (1978).
68. Stanbro, W. D. and Lenkovitch, M. J. "Kinetics and Mechanism
of the Decomposition of N,N-Dihalopeptides," Int. J. Chem.
Kinetics 17, 401-411 (1985).
69. Conyers, B. and Scully, Frank E., Jr., unpublished results.
70. Snyder, M. P. and Margerum, D. W. "Kinetics of Chlorine
Transfer from Chloramine to Amines, Amino Acids, and Pep-
tides," Inora. Chem. 21, 2545-2550 (1982).
71. Stanbro, W.D. and Smith, W.D. "Kinetics and mechanism of. the
decomposition of N-Chloroalanine in Aqueous Solution,"
Environ. Sci. Technol. .13, 446-451 (1979).
72. Hand, V.C.; Snyder, M.P.; Margerum, D.W. "Concerted
Fragmentation of N-Chloro-cr-amino Acid Anions," J. Am. Chem.
SOC. 105. 4022-4025 (1983).
73. Isaac, R.A. and Morris, J.C. "Modeling of Reactions Between
Aqueous Chlorine and Nitrogenous Compounds." In: Water
Chlorination; Environmental Impact and Health Effects. Vol. 4,
Jolley, R.L. et al., Eds. Ann Arbor, MI: Ann Arbor Science,
19835; pp. 63-75.
74. Mauger, R. P. and Soper, F. G. "Acid Catalysts in the
Formation of Chloramides from Hypochlorous Acid," J. Chem.
SOC. 1946. 71-75 (1946).
75. de la Mare, P.B.D., Harvey, J.T.; Hassan, M.; Varma, S. "The
Kinetics and Mechanisms of Aromatic Halogen Substitution.
Part VII. Some Experiments Relating to the Halogenation of
Toluene and tert-Butylbenzene," J. Chem. Soc. 2756-2759
(1958).
76. de la Mare, P.B.D.; Ketley, A.D.; Vernon, C.A. "The Kinetics
and Mechanisms of Aromatic Halogen Substitution. Part I.
Acid-Catalyzed Chlorination by Aqueous Solutions of
Hypochlorous Acid," J. Chero. Soc. 1954. 1290-1297 (1954).
77. Swain, C.G.; Crist, D.R. "Mechanisms of Chlorination by
65
-------
Hypochlorous Acid. The Last of Chlorinium Ion, Cl+," J. Am.
Chem. Soc. 94., 3195-3200 (1972).
78. Roth, W.A. "On the Thermochemistry of Chlorine and
Hypochlorous Acid," Z. Phvsik. Chero. A 145. 289-297 (1929).
79. Reinhard, M., Stumm, W. "Kinetics of Chlorination of j>-Xylene
in Aqueous Solution." In: Water Chlorination: Environmental
Impact and Health Effects. Vol. 3, Jolley, R.L. et al., Eds.
Ann Arbor, MI: Ann Arbor Science Publishers, Inc./ 1980; pp
209-218.
80. Carlson, R..M. , Carlson, R.E., Kopperman, H.L., Caple, R.
"Facile Incorporation of Chlorine into Aromatic Systems During
Aqueous Chlorination Processes," Environ. Sci. Technol. j),
674-675 (1975).
81. Snider, E.H. and Alley, F.C. "Kinetics of the Chlorination of
Biphenyl under Conditions of Waste Treatment Processes,"
Environ. Sci. Technol. 13_, 1244-1248 (1979).
82. Snider, E.H., Alley, F.C. "Kinetics of Biphenyl Chlorination
in Aqueous Systems in the Neutral and Alkaline pH Ranges."
In: Water Chlorination; Environmental Impact and Health
Effects, Vol. 3, Jolley, R.L. et al., Eds. Ann Arbor, MI: Ann
Arbor Science Publishers, Inc., 1980; pp 219-225.
83. Burleson, J. L. ; Peyton, G. R.; and Glaze, W. H. "Gas-
Chromatographic/Mass Spectrometric Analysis of Derivatized
Amino Acids in Municipal Wastewater Products," Environ. Sci.
Technol. 14(H) , 1354-1359 (1980).
84. Carlson, R. M. and Caple, R. "Organochemical Implications of
Water Chlorination." In: Water Chlorination; Environmental
Impact and Health Effects. Vol. 1, Jolley, R. L. et al., Eds.
Ann Arbor, MI: Ann Arbor Science Publishers, Inc., 1978; pp
65-75.
85. Lee, C. F. "Kinetics of Reactions Between Chlorine and
Phenolic Compounds." In: Principles and Applications of Water
Chemistry. Faust, S.D. and Hunetr, J. V., Eds. New York: John
Wiley & Sons, Inc., 1967; pp. 55-74.
86. Soper, F. G. and Smith, G. F. J. Chem. Soc. 1582-1591
(1926).
87. Nowell, L.H.; Crosby, D.G. "Photodegradation of Water
Pollutants in Chlorinated Water," In: Water Chlorination:
Chemistry. Environmental Impact and Health Effects. Vol. 5,
Jolley, R.L. et al., Eds. Chelsea, MI: Lewis Publishers,
Inc., 1985; Chapter 83, pp. 1055-1062.
66
-------
88. Larson, R.A., Rockwell, A.L. "Chloroform and Chlorophenol
Production by Decarboxylation of Natural Acids During Aqueous
Chlorination." In: Water Chlorination; Environmental Impact
and Health Effects. Vol. 2, Jolley, R.L. et al., Eds. Ann
Arbor, MI: Ann Arbor Science Publishers, Inc., 1978? pp 67-74.
89. Rook, J.J. "Possible Pathways for the Formation of
Chlorinated Degradation Products During Chlorination of Humic
Acids and Resorcinol." In; Water Chlorination; Environmental
Impact and Health Effects. Vol. 3, Jolley, R.L. et al., Eds.
Ann Arbor, Ml: Ann Arbor Science Publishers, Inc., 1980; pp
85-98.
90. Grimley, E. and Gordon, G. "Kinetics and Mechanism of the
Reaction Between Chlorine and Phenol in Acidic Aqueous
Solution," J. Phvs. Chem. 77, 973-978 (1973).
91. Brown, L.O. and Soper, F.G. "Studies of N-Halogeno-Compounds.
Part V. The Interaction of Dialkylchloramines with Phenols,"
J. Chem. SQC. 3576-3580 (1953).
92. Gould, J.P., Hay, T.R. "The Nature of the Reactions Between
Chlorine and Purine and Pyrimidine Bases: Products and
Kinetics," Water Sci. Technol. 14', 629-640 (1982).
93. Gould, J.P.; Richards, J.T. ', Miles, M.G. "The Kinetics and
Primary Products of Uracil Chlorination," WaterRes. 18(2),
205-212 (1984).
94. Gould, J.P.; Richards, J.T.; Miles, M.G. "The Formation of
Stable Organic Chloramines During the Aqueous Chlorination of
Cytosine and 5-Methylcytosine," Water Res. 18(8). 991-999
(1984).
95. Rosenblatt, David H. "Chlorine and Oxychlorine Species
Reactivity with Organic Substances." In: Disinfection, Water
and Wastewater. Johnson, J.D., Ed. (Ann Arbor, MI: Ann Arbor
Science, 1975), pp. 249-276.
96. Lin, S.; Carlson, R.M. "Susceptibility of Environmentally
Important Heterocycles to Chemical Disinfection: Reactions
with Aqueous Chlorine, Chlorine Dioxide, and Chloramine,"
Environ. Sci. Technol. 18. 743-748 (1984).
97. Fukayama, M. Y. ; Tan, H.; Wheeler, W. B. ; and Wei, C.182.
"Reactions of Aqueous Chlorine and Chlorine Dioxide with Model
Food Compounds." Environ. Health Perspect. 6_9, 267-274 (1986).
98. Ghanbari, H. A.; Wheeler, W. B.; Kirk, J. R. "Reactions of
Aqueous Chlorine and Chlorine Dioxide with Lipids, Chlorine
Incorporation," J. Food Sci. 47. 482-485 (1982).
67
-------
99. Ghanbari, H. A.; Wheeler, W. B.; Kirk, J. R. "Reactions of
Chlorine and Chlorine Dioxide with Free Fatty Acids, Fatty
Acid Esters, and Triglycerides." In: Water Chlorination;
Environmental Impact and Health Effects. Vol. 4, Jolley, R. L.
et al., Eds. Ann Arbor, MI: Ann Arbor Science Publishers,
Inc., 1983; pp 167-177.
100. Israel, G. C.; Martin, J. K.; Soper, F. G. "The Kinetics of
Chlorohydrin Formation. Part I. The Reaction Between Hypochlo-
rous Acid and Allyl Alcohol in Aqueous Solution," J. Chem.
Soc.. 1950. 1282-1285 (1950).
101. Anbar, M. Dostrovsky, I. "Ultra-violet Absorption Spectra of
Some Organic Hypohalites," J. Chem. Soc. 1954f 1105-1108
(1954) .
102. de la Mare, P. B. D. ; Hilton, I. C. ; Vernon, C. A. "The
Kinetics and Mechanism of Aromatic Halogen Substitution. Part
VIII. Chlorine Acetate in Slightly Aqueous Acetic Acid," J.
Chem. Soc. 1960. 4039-4044 (1960).
103. Walling, C. Free Radicals in Solution; J. Wiley: New York,
1957; pp 363-364.
104. Chakrabarty, S. K. ; Kretschmer, H. O. "Sodium Hypochlorite as
a Selective Oxidant for Organic Compounds," J. Chem. Soc..
Perkin Trans. 1 1974. 222-228 (1974).
105. Field, R. J.; Koros, E. ; Noyes, R. M. "Oscillation in Chemical
Systems. II. Thorough Analysis of Temporal Oscillation in the
Bromate-Cerium-Malonic Acid System," J. Am. Chem. Soc. 94.
8649-8664 (1972) .
106. Leopold, K. R. ; Haim, A. "Equilibrium, Kinetics, and Mechanism
of the Malonic Acid-Iodine Reaction," Int. J. Chem. Kinet. 9_,
83-95 (1977).
107. Pedersen, K. J. "Velocity of Bromination of Acetoacetic Acid,"
J. Phvs. Chem. 38. 999-1022 (1934).
108. Pearson, R. G.; Dillon, R. L. "Rates of lonization of Pseudo
Acids. IV. Relation Between Rates and Equilibria," J. Am. Chem
Soc. 75, 2439 (1953) .
109. Potts, J. E.; Amis, E. S. "The Alkaline Hydrolysis of Ethyl
Acetate from the Standpoint of lon-Dipole Theory," J. Am.
Chem. Soc. 71. 2112-2116 (1949).
110. Tommila, E. ; Ketonen, L. "Kinetic Studies on Ester Hydrolysis.
X. The Alkaline Hydrolysis of Ethyl Benzoate and Certain
Substituted Ethyl Benzoates in Methanol-Water Mixtures,'J
Suomen Kemistilehti 18B. 24-28 (1945) quoted in Tables of
68
-------
Chemical Kinetics. Homogenous Reactions. Supplement 1;
Stauffer, C. H., Project Director; U. S. Govt Printing Office:
Washington, DC, 1956; p 136.
111. Jencks, W. P.; Gilchrist, M. "Nonlinear Structure-Reactivity
Correlations. The Reactivity of Nucleophilic Reagents toward
Esters," J. Am. Chem. Soc. 90. 2622-2637 (1968)
112. Bell, R. P.; Spiro, M. "The Mechanism of Bromination of Ethyl
Malonate," J. Chem. Soc. 1953. 429-435 (1953).
113. Bell, R. P.; Rawlinson, D. J. "Kinetics of the Bromination of
Ethyl Malonate and Methyl Methanetricarboxylate," J. Chem.
Soc. 1961. 726-729 (1961)
114. Bell, R. P.; Yates, K. "Kinetics of the Chlorination of
Diethyl Malonate in Aqueous Solutions," J. Chem. Soc. 1962.
2285-2290 (1962).
115. Pedersen, K. J. "The Velocity of Bromination of Acetoacetic
Ethyl Ester. II. The General Base Catalysis," J. Phvs. Chem.
31, 601-621 (1934).
116. Horiike, M. ; Oda, J. ; Inouye, Y. ; Ohno, M. "Synthesis of
Substituted Aminomalonates by the Reaction of Chloramine with
Malonate Carbanions," Agr. Biol. Chem. (Tokyo) 33. 292-293
(1969); Chem. Abstr. 71. 2929z (1969).
117. Horiike, M. ; Oda, J. ; Inouye, Y. ; Ohno, M. ; Matsumoto, K.
"Amination of Malonate Esters with Chloramine," Japanese
Patent 7100165; Chem. Abstr. 74. 124863f (1974).
118. Oda, J.; Horiike, M. ; Inouye, Y. "Reaction of Chloramines. IV.
Reaction of Chloramine with beta-Keto Esters and beta-
Diketones," Bull. Inst. Chem. Res.. Kyoto Univ. 50. 183-191
(1972') ; Chem. Abstr. 78. 15728C (1973).
119. Kobayashi, T. ; Okuda, T. "Continuous Method of Monitoring
Water Quality by Chlorine Consumption," Water Res.. 6, 197-209
(1972).
120. Wilson, I. R.; Harris, G. M.; "The Oxidation of Thiocyanate
Ion by Hydrogen Peroxide. I. The pH Independent Reaction," J.
Am. Chem. Soc.. 82. 4515-4517 (1960).
121. Wilson, I. R.; Harris, G. M.; "The Oxidation of Thiocyanate
Ion By Hydrogen Peroxide. II. The pH Independent Reaction," J.
Am. Chem. Soc.. 83. 286-289 (1961).
122. Briot, G. T. ; Smith, R. H. "Kinetics of Oxidation of
Thiocyanate by Aqueous Iodine," Austr. J. Chem.. 26. 1863-1869
(1973).
69
-------
123. Curci, R. ; Edwards, J. O. "Peroxide Reaction Mechanisms -
Polar." In: Organic Peroxides. Vol. I, Swern, D., Ed. New
York: Wiley-Interscience, Chapter IV, pp 199-264.
124. Pearson, D. G. ; Sobel, H.; Songstad, J. "Nucleophilic
Reactivity Constants toward Methyl Iodide and trans-[Pt(py) 2C-
lz]r" J. Am. Chem. Soc.. 90r 319-326 (1960).
125. Kice, J. L. and Puls, A. R. "The Reaction of Hypochlorite
with Various Oxidized Derivatives of Disulfides and with
Sulfinate Ions," J. Am. Chem. Soc. 99., 3455-3460 (1977).
126. Jacangelo, J. G. and Olivieri, V. P. "Aspects of the Mode of
Action of Monochloramine." In: Water Chlorination; Chemistry.
Environmental Impact and Health Effects. Vol. 5, Jolley, R.L.
et al., Eds. Chelsea, MI: Lewis Publishers, Inc., 1985;
Chapter 45, pp. 575-586.
127. Yuki, H. ; Kishikawa, T. ; Tohira, Y. ; Suzuki, S.; Sano, F. ;
Hayakawa, S. "Antiviral Agents. V. Synthesis and Antiviral
Activity of N-Chloro Compounds," Chem. Pharm. Bull. 15. 1123-
1128 (1967) .
128. Stanbro, W. D. and Lenkovitch, M. J. "Slowly Dechlorinated
Organic Chloramines," Science. 215. 967-968 (1982).
129. Ruff, F. and Kucsman, A. "Mechanism of the Reaction of
Sulfides with N-Chloroarenesulphonamides," J. Chem. Soc. .
Perkin II. 1975. 509-519.
130. Mahadevappa, D. S.; Gowda, B. T.; Gowda, N. M. N. "Kinetics
and Mechanism of Oxidation of Thiocyanate Ion by Sodium N-
Chloro-4-methylbenzenesulfonamide in Alkaline Solution," Z.
Naturforsch.. B; Anorq. Chem.. Orq. Chem.. 34B. 52-57 (1979).
131. Tsuyuki, H., Stahmann, M.A., Casida, J.E. "Preparation,
Purification, Isomerization, and Biological Properties of
Octamethylpyrophosphoramide N-Oxide," J. Agric. Food Chem.
1, 922-930 (1955).
132. Epstein, J., Bauer, V.E., Saxe, M., Demek, M.M. "The
Chlorine-catalyzed Hydrolysis of Isopropyl
Methylphosphonofluoridate (Sarin) in Aqueous Solution," J. Am.
Chem. Soc. 78. 4068-4071 (1956).
133. Lordi, N., Epstein, J. "Kinetics and Mechanism of Chlorination
of Triethylphosphorothiolate in Dilute Aqueous Media at 25 °,"
J. Am. Chem. Soc. 80. 509-515 (1958).
70
------- |