Chloramines VII: Chlorination of alanylphenylalanine in model solutions and in a wastewater

Article abstract of DOI:10.1021/es9607953

Products of the 30-min chlorination of the dipeptide alanylphenylalanine were determined at pH 7.0 in model solutions. At Cl₂/peptide mole ratios ≤ 1, N-chloroalanylphenylalanine (I) is the only product. 1 is very stable at 23 °C (t 1/4 = 111 ±

Full text of DOI:10.1021/es9607953

Environ. Sci. Technol. 1997, 31, 1979-1984
begun a systematic study of the chlorination reactions of
Chloramines VII: Chlorination of
Alanylphenylalanine in Model
Solutions and in a Wastewater
organic amino nitrogen compounds found in wastewaters.
We have reported studies of several amino acids and identified
a new class of unusually stable chlorination products,
N-chloroaldimines (17-21). However, dissolved free amino
acids represent only a small percentage of the total amount
of amino acids present in wastewaters (8) and, in fact, in
most natural waters as well. For that reason, we began a
systematic study of the reactions of polypeptides. In the
preceding paper in this series (22), we reported a study of the
dipeptide glycylphenylalanine, which reacts with 2 equiv of
aqueous chlorine to form a product that is dichlorinated on
the terminal amino nitrogen. This product loses 2 mol of
HCl in successive elimination steps to form a cyanoacyl
derivative of phenylalanine. However, compounds with
N-terminal residues other than glycine cannot lose 2 equiv
of HCl. Consequently, we studied the chlorination reactions
of alanylphenylalanine in model solutions and in a primary
wastewater and report the results of this work here.
T . C H R I S T O P H E R F O X ,
D A N I E L J . K E E F E , A N D
F R A N K E . S C U L L Y, J R . *
Department of Chemistry and Biochemistry, Old Dominion
University, Norfolk, Virginia 23529-0126
A . L A I K H T E R
Department of Chemistry, University of Virginia,
Charlottesville, Virginia 22901
Products of the 30-min chlorination of the dipeptide ala-
nylphenylalanine were determined at pH 7.0 in model solutions.
At Cl2/peptide mole ratios e1, N-chloroalanylphenylalanine
(I) is the only product. I is very stable at 23 °C (t1/2 ) 111
( 8.8 h). At mole ratios g2, N,N-dichloroalanylphenylalanine
(II) is the only product. II decomposes in model solutions
(t1/2 ) 4.1 ( 0.2 h) at pH 7.0 to form a compound identified
as the N-chloroketimine N-[2-(N ′-chloroimino)propanoyl]-
phenylalanine (III). The structure of III was identified by
converting it to N-pyruvylphenylalanine tert-butyl ester
by reduction, hydrolysis, and esterification and correlating
the mass spectrum and GC retention time of this derivative
with those of an authentic sample. III is unusually stable and
decomposes slowly (t1/2 ) 125 ( 6.5 h) to phenylalanine.
In order to monitor the reactions of the dipeptide at low
concentrations in a wastewater, alanyl-p-[3H]-phenylalanine
was synthesized. A primary wastewater (TKN ) 19.82 mg/
Experimental Methods
General. All chemicals, buffers, standard solutions, and all
syntheses and analyses, unless specifically described below,
are described in the preceding paper (22). All instrumentation
used in this work is also described in the preceding paper
(22). Separation of chlorination products was carried out on
a Whatman 5 µm Partisil ODS-3 analytical HPLC column with
a dual-solvent system described previously (17). The solvent
program (1 mL/ min) consisted of a 5-min isocratic elution
with 95% A/ 5% B, followed by a linear gradient to 55% A/ 45%
B over 20 min, isocratic elution for 3 min, and a final linear
gradient over 4 min to 10% A/ 90% B. Alanyl-p-bromophe-
nylalanine was converted to alanyl-p-[3H]-phenylalanine by
Moravek Biochemicals, Inc. by metal-catalyzed halogen
reduction in the presence of tritium gas. A solution of the
tritiated dipeptide (250 µL; 2 µCi/ µL) was obtained with a
specific activity of 25 Ci/ mmol.
Chlorination and Analysis of Model Solutions of Ala-
nylphenylalanine. Determination of the breakpoint curve
of model solutions of alanylphenylalanine as well as iden-
tification of chlorination products and studies of the de-
composition of these products were carried out on 1.43 ×
10-3 M solutions of the dipeptide in 0.025 M phosphate buffer
(pH 7.0) at 23 °C in the same manner as that described for
glycylphenylalanine (22).
L; [NH ] ) 19.79 mg/L) was inoculated with the
3
radiolabeled dipeptide and chlorinated to seven different
chlorine concentrations spanning the breakpoint curve
of the wastewater. Products identical to those observed
in model solutions were formed. The stabilities of the
tritiated analogs of II and III in the wastewater were similar
to those determined in model solutions. Time studies of
the decomposition of N,N-dichloroalanyl-p-[3H]-phenylalanine
revealed the formation of an intermediate (A) not previously
recognized. Modeling of the reactions of II suggested that
A was a decomposition product of III in the formation
of phenylalanine and was probably either an isocyanate or
a carbamic acid formed from hydrolysis of an isocyanate
intermediate.
Determining the amount of each chlorination product
formed was made possible using alanyl-p-[3H]-phenylalanine
(0.7 µCi/ mL) in the same manner used for determination of
the products of the tritiated analog of glycylphenylalanine
(22.).
Synthesis and Identification of N-[2-(N ′-Chloroim ino)-
propanoyl]phenylalanine (III). With constant stirring, 5.0
mmol of aqueous chlorine was added to 50 mL of buffered
(0.55 g of NaH2PO4, pH 7.0) Milli-Q water containing 2.5 mmol
(0.59 g) of alanylphenylalanine. The solution was incubated
in the dark with periodic removal of 10-µL aliquots for HPLC
analysis. After all of the dichlorinated alanylphenylalanine
had decomposed to the chloroketimine, the solution was
acidified to pH 2.0 with concentrated HCl. The oily product
was extracted with diethyl ether and dried over anhydrous
Na2SO4, and the solvent was removed in vacuo. 13C-NMR
(acetone-d6) (ppm): 176 (s, CdO), 173 (s, CdO), 161 (s, CdN),
127-136 (aromatic), 54 (d, NH-CH), 38 (t, C6H5-CH2), 18 (q,
CH3-C); 1H-NMR (acetone-d6) (ppm): 7.4 (1, m, NH), 7.3 (5,
d, aromatic), 4.8 (1, m, NH-CH), 3.32-3.15 (2, m, CH2-C6H5),
2.3 (3, s, CH3-CdN); IR (thin film): 2500-3500 (m, O-H),
Introduction
Chloramines formed on disinfection of wastewaters are
believed to be responsible for toxic responses of some fish
to the discharge of chlorinated wastewaters and for stronger
fish avoidance reactions than free residual chlorine (1-7).
Organic chloramines in particular are believed to be respon-
sible for disinfection interferences in certain types of waste-
waters containing high concentrations of organic amino
nitrogen compounds (8-16). Because organic chloramines
in chlorinated wastewaters are poorly characterized, we have
* Corresponding
author;
e-mail
address:
FES100U@
hydrogen.chem.odu.edu.
1750 (s, CdO), 1690 (s, CdO) cm-1
.
9
S0013-936X(96)00795-X CCC: $14.00
1997 Am erican Chem ical Society
VOL. 31, NO. 7, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 7 9

The structure of III was confirmed by converting it to
N-pyruvylphenylalanine tert-butyl ester and comparing its
GC/ MS retention time and spectrum with an authentic
sample. Alanylphenylalanine (1.0 mmol, 0.24 g) was dissolved
in 50 mL of buffered (0.41 g of NaH2PO4, pH 7.0) Milli-Q
water. Standardized aqueous chlorine (2.0 mmol) was added
to the solution with continuous stirring to form the dichlo-
rinated dipeptide. This solution was incubated in the dark,
and 10-µL aliquots were periodically removed for analysis by
HPLC. When all of the dichlorinated dipeptide had decom-
posed to the chloroketimine, portions of Na2S were added to
reduce the chloroketimine. The progress of this reaction was
also followed by HPLC. When reduction was complete, the
pH of the solution was reduced to approximately 2.0 with
concentrated HCl and extracted with diethyl ether. The
extract was dried over Na2SO4 and the solvent removed in
vacuo. 13C-NMR (CDCl3) (ppm): 197 (s, CdO); 172 (s, CdO);
161 (s, CdO); 127-136 (aromatic); 54 (d, NH-CH); 38 (t,
C6H5CH2); 24 (q, CH3-CdO).
An ether extract of the reaction mixture was concentrated,
and the residue (0.18 g) was dissolved in 15 mL of tert-butyl
acetate. To this solution 0.18 mL of HClO4 (60% aqueous
solution) was added. The reaction was stirred for 10 min and
allowed to react for 4 days (23). After the incubation time,
the solution was cooled in an ice bath and extracted with 0.5
M HCl (3 × 10 mL). The volume of tert-butyl acetate was
reduced in vacuo, and the remaining solution was analyzed
by GC/ MS, m/z (relative abundance): 235 (47, M - C4H8),
190 (47, M - C4H8 - COOH), 148 (87, 190 - H2CdCdO), 120
(100, C6H5-CH2-CHdNH2+), 91 (37, C7H7+), 57 (83, C4H9+).
Synthesis of N-Pyruvylphenylalanine tert-Butyl Ester.
Dicyclohexylcarbodiimide was added to a solution of 40 mg
(0.45 mmol) of freshly distilled pyruvic acid in 1 mL of dry
THF under argon at 0-5 °C. After 1-2 min, a solution of 100
mg of phenylalanine tert-butyl ester in 0.5 mL of dry THF was
added dropwise. After stirring 1 h at room temperature, the
solution was filtered to remove dicyclohexylurea. The solvent
was evaporated, and the residue was purified by flash
chromatography on 10-15 g of silica gel using an ethyl acetate:
hexane gradient (1-50%) to yield 46 mg (35% yield) of product.
Exact mass measurements on the parent ion found m/ z
291.14848 (calcd mass 291.14706). 1H NMR (DMSO-d6)
(ppm): 7.35 (1, m, NH), 7.2 (5, d of d, aromatic), 4.67 (1, m,
CH-CH2-C6H5), 3.1 (2, d, CH2-C6H5), 2.45 (3, s, CH3), 1.4 (9,
s, C(CH3)3).
Description of Wastewater. A primary wastewater was
obtained from the Army Base Treatment Plant located in
Norfolk, VA, at a point following settling. The facility has
been described previously as Plant 2 (24). Handling and
storage of the wastewater has been described previously (24).
The total Kjeldahl nitrogen (TKN) and ammonia concentra-
tions were determined by the Hampton Roads Sanitation
District according to published procedures (25). Amino acid
concentrations in the wastewater were determined before
and after hydrolysis (26) by precolumn derivatization with
o-phthalaldehyde (OPA) and HPLC analysis (18, 19).
Analysis of Chlorination Products in Wastewater. Waste-
water was buffered at pH 7.0 as described previously. The
breakpoint curve was determined by chlorinating 100-mL
aliquots to concentrations of 40, 80, 120, 160, 200, 240, and
280 mg of Cl2/ L and treating them as described previously
(17).
Aliquots (15.0 mL) of wastewater were inoculated with 1.0
mL of a stock solution of alanyl-p-[3H]-phenylalanine (7.0
µCi/ 10 mL; specific activity 25 Ci/ mmol; 2.8 × 10-8 M),
chlorinated to each of the seven levels used in determining
the breakpoint curve. The volumes of each solution were
normalized to 17.0 mL, and the reaction mixtures were
incubated, fractionated, and assayed as described previously
(17). The rates of decomposition of N,N-dichloroalanylphe-
nylalanine (II) and N-[2-(N ′-chloroimino)propanoyl]phe-
nylalanine (III) were determined in a similar manner over a
65-h time period by HPLC analysis of buffered wastewater
chlorinated to 240 mg/ L.
Results and Discussion
Because wastewaters are not typically discharged until 30
min after they are chlorinated, the reactions that take place
within the first 30 min are more likely to have the greatest
initial environmental impact on receiving waters. However,
depending on the byproducts formed, decomposition of the
primary chlorination products over longer time periods may
have even greater long-term environmental impact. Con-
sequently, in this study the chlorination products of ala-
nylphenylalanine formed in 30 min were identified, and their
subsequent decomposition reactions were studied to estimate
the potential environmental impact of the chlorination
reactions of polypeptides on the discharge of chlorinated
wastewater.
Identification of Prim ary Alanylphenylalanine Chlori-
nation Products in Model Solutions. The 30-min chlorine
demand curve of an alanylphenylalanine model solution is
similar to that of glycylphenylalanine (22). Neither a notice-
able chloramine maximum nor a breakpoint is observed. At
Cl2/ peptide mole ratios e1.0, the 30-min residual chlorine
concentration equals the amount of chlorine added to the
solution. A slight inflection point is seen at a mole ratio of
approximately 1.5, which indicates some decomposition of
the oxidant formed at the higher concentration levels. At
mole ratios >2.0, free residual chlorine is measured, and the
total residual chlorine concentration increases linearly with
increases in the amount of aqueous chlorine added.
When model solutions of alanylphenylalanine (Ala-Phe)
were chlorinated to Cl2/ peptide mole ratios e1 and analyzed
by HPLC, only one chlorination product was observed in the
chromatogram. Alanylphenylalanine itself elutes in two peaks
because it exists as a mixture of diastereomers. The one
chlorination product elutes as a single peak and is believed
to be N-chloroalanylphenylalanine (I) for the same reasons
N-chloroglycylphenylalanine was identified as such (22). First,
as it eluted from the column, it oxidized iodide to iodine.
Second, as the molar ratio of Cl2/ Ala-Phe increased up to 1.0,
the peak area increased to a maximum and began to decrease
at mole ratios >1.0.
As the Cl2/ Ala-Phe mole ratio was increased between 1
and 2, three new chromatographic peaks appeared within 30
min with HPLC retention times of 22.3, 22.6, and 22.9 min
(Figure 1). The latter two peaks formed immediately after
chlorination, and their intensity increased in proportion to
increases in the Cl2/ Ala-Phe mole ratio and also in proportion
to decreases in the amount of I present. The peak that eluted
at 22.3 min increased over time as the other two peaks
decreased (Figure 1). All three compounds oxidized iodide
as they eluted from the column, suggesting that they contained
oxidizing moieties like N-chlorinated amines.
In the preceding study of glycylphenylalanine (22), the
addition of 2 equiv of aqueous chlorine produced N,N-
dichloroglycylphenylalanine, a crystalline solid that could be
precipitated from solution by the addition of acid, purified
by recrystallization, and characterized by elemental and
spectral analysis. When a solution of alanylphenylalanine
was chlorinated with 2 equiv and acidified, an oil was formed
that did not lend itself to further purification and elemental
analysis. However, based on the study of glycylphenylalanine,
the peaks that eluted at 22.6 and 22.9 min were believed to
be diastereomers of N,N-dichloroalanylphenylalanine (II).
Spectral Identification of N-[2-(N ′-Chloroim ino)pro-
panoyl]phenylalanine (III). Based on its spectral properties,
the product that eluted at 22.3 min was believed to be the
N-chloroketimine N-[2-(N ′-chloroimino)propanoyl]phenyl-
alanine (III). III was isolated by extraction after an aqueous
solution of N,N-dichloroalanylphenylalanine had been al-
9
1 9 8 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 7, 1997

not be identified. The compound eluting at 17.6 min was
found to make up approximately 8% of the total decomposi-
tion products and has been labeled unknown A.
The quantitation of both the primary and secondary
chlorination products was made possible by inoculating
model solutions with the radiotracer alanyl-p-[3H]-pheny-
lalanine. The solution was chlorinated to a Cl2/ Ala-Phe mole
ratio of 2.0/ 1 and analyzed periodically by HPLC and liquid
scintillation counting of collected fractions. The retention
times of all peaks in the radiochromatograms corresponded
to those observed in the UV chromatograms. Although the
structure of the compound eluting at 17.6 min was not
confirmed by spectral analysis, modeling studies discussed
below strongly suggested it is an isocyanate (or its hydrolysis
product, a carbamic acid) formed by loss of HCl and
acetonitrile from III.
Stabilities of Chlorination Products in Model Solutions.
The shape of the breakpoint curve of alanylphenylalanine
can be explained by the unusual stability of the chloramines
formed. When alanylphenylalanine is chlorinated to Cl2/
Ala-Phe molar ratios e1.0, N-chloroalanylphenylalanine is
the only product formed and, because it is so stable (t1/ 2
)
111 ( 8.8 h in model solutions), a linear relationship exists
in this range between the amount of chlorine applied and the
total residual chlorine measured 30 min later. The stabilities
of monochlorinated dipeptides can vary greatly. At 22 °C,
N-chloroglycylphenylalanine has a half-life of 127 h (22). At
37 °C, the half-lives of the monochlorinated dipeptides
examined by Stelmaszynska et al. (27) ranged from 6.6 h for
monochlorinated Gly-Leu to 25 h for monochlorinated Leu-
Gly (pH 6.6). At temperatures 15 °C lower, these half-lives
would be much longer.
As the Cl2/ Ala-Phe molar ratio increases to g1.0, N,N-
dichloroalanylphenylalanine (II) begins to form and an
inflection point is observed in the breakpoint curve because
of its decomposition (t1/ 2 ) 4.1 ( 0.2 h, k1m ) 2.85 × 10-3
min-1; r 2 ) 0.997, where k1m is the first-order decomposition
rate constant in the model solution at 23 °C). However, within
30 min at 23 °C only about 9% of the dichlorinated dipeptide
decomposes. The decomposition product itself contains
oxidizing chlorine and, consequently, the decrease in the
oxidizing power of the solution is not large. Keefe et al. (22)
reported that the half-life of N,N-dichloroglycylphenylalanine
is slightly longer (t1/ 2 ) 6.4 h at 22 °C) than the dichlorinated
dipeptide described here.
FIGURE 1. HPLC chromatograms of an alanylphenylalanine model
solution (1.43 × 10-3 M, pH 7.0) chlorinated with 2 equiv and analyzed
after 1.5, 10, and 96 h.
lowed to decompose by >90% and acidified. All NMR and
IR spectra were consistent with the proposed structure.
Distinctive spectral characteristics that supported the as-
signment of structure included the presence of three carbonyl
carbons (161, 173, and 176 ppm) in the decoupled 13C-NMR,
a proper off-resonance 13C spectrum, and the fact that the
1H-NMR spectrum demonstrated that the carbon adjacent to
the alanyl methyl protons in the parent compound had lost
a proton to form the CH3-CdN group.
The N-chloroketimine was converted to N-pyruvylphe-
nylalanine tert-butyl ester, and its GC/ MS retention time and
mass spectrum were compared to those of a sample inde-
pendently synthesized. The retention times of the two were
identical. The fragmentation patterns for both samples were
also identical and consistent with the proposed structure.
Decomposition of II produces the N-chloroketimine III,
which is even more stable (t1/ 2 ) 125 ( 6.5 h, k2m ) 9.25 ×
2
10-5 min-1; r ) 0.989, where k2m is the first-order decom-
position rate constant in the model solution at 23 °C) than
I or II. It is also significantly more stable than the corre-
sponding N-chloroaldimine (t1/ 2 ) 36 h at 22 °C) formed from
chlorination of Gly-Phe (22). At molar ratios g2.0, all of the
dipeptide is dichlorinated, and free residual chlorine is
detected.
The product distribution of a tritiated alanylphenylalanine
model solution chlorinated with 2 equiv of aqueous chlorine
and analyzed periodically over 164 h is shown in Figure 2.
The major decomposition product of N,N-dichloroalanylphe-
nylalanine (II) is the corresponding N-chloroketimine (III).
After the concentration of N-chloroketimine (III) reaches a
maximum within 24 h, the concentration of phenylalanine
(IV) appears to rise. However, N-chloroketimine (III) and
phenylalanine (IV) do not account for 100% of the decom-
position products of N,N-dichloroalanylphenylalanine (II).
Unknown A accounts for the difference. Its concentration
rises within the first 24 h and appears to approach a steady-
state concentration at the same time the concentration of
N-chloroketimine reaches a maximum. Modeling studies
described below suggest that this compound is an interme-
diate such as an isocyanate or a carbamic acid in the formation
of phenylalanine (IV) from the N-chloroketimine (III).
Identification of Secondary Alanylphenylalanine Chlo-
rination Products in Model Solutions. Over time in aqueous
solution (23 °C, pH 7.0) three new decomposition products
of N,N-dichloroalanylphenylalanine (II) appeared in the
chromatograms (Figure 1) with retention times of 5.1, 12.3,
and 17.6 min. The peak eluting at 5.1 min was identified as
phenylalanine both by its retention time and by derivatizing
the solution with o-phthalaldehyde (OPA) and comparing
the retention time of the OPA derivative with that of an
authentic sample. In radiotracer studies discussed below,
the compound eluting at 12.3 min was found to represent
less than 1% of the decomposition products formed and could
9
VOL. 31, NO. 7, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 8 1

then hydrolyzes to phenylalanine and HCN through possible
isocyanate or carbamic acid intermediates. It is conceivable
that the N-chloroaldimine also decomposes more directly to
an isocyanate or carbamic acid in the same way the
N-chloroketimine does in Figure 3. However, this pathway,
if it exists, is minor and no evidence for it was apparent.
Stelmaszynska et al. (27) also suggested that an N-
chloroketimine and an N-chloroaldimine would decompose
to form a nitrile, carbon dioxide, and the C-terminal amino
acid. This has been supported, as both dichlorinated Ala-
Phe and dichlorinated Gly-Phe decompose to form phenyl-
alanine. The overall mechanism for the decomposition
appears to be dependent on the N-terminal residue.
Synthesis of Alanyl-p-[3H]-phenylalanine. In order to
observe the reactions of alanylphenylalanine in a wastewater
and to measure the quantities of each chlorination product
formed, alanyl-p-[3H]-phenylalanine was synthesized. 1H-
and 13C-NMR and IR spectra for all intermediates were
consistent with proposed structures. Alanyl-p-bromophe-
nylalanine also gave correct elemental analysis. It was
converted to alanyl-p-[3H]-phenylalanine by metal-catalyzed
halogen reduction with tritium gas.
Chlorination Products and Their Rates of Decom position
in a Wastewater. The primary wastewater used in this work
had an ammonia concentration of 19.79 mg/ L and a TKN
concentration of 19.82 mg/ L. The breakpoint curve for the
wastewater exhibited a chloramine maximum at approxi-
mately 120 mg/ L and a breakpoint at approximately 210 mg/
L. The concentrations of the dissolved free amino acids
alanine and phenylalanine were found to be 5.87 × 10-7 and
1.04 × 10-7 M, respectively. The concentration of a hydro-
lyzable amino acid is a measure of the concentration of that
amino acid which is bound in proteins. The concentrations
of hydrolyzable alanine and phenylalanine in the wastewater
were found to be 1.04 × 10-5 and 2.92 × 10-6 M, respectively.
Thus the concentrations of these amino acids bound to
proteins are between 20 and 30 times greater than their free
dissolved concentrations.
Wastewater was inoculated with alanyl-p-[3H]-phenyla-
lanine, chlorinated for 30 min to seven points along the
breakpoint curve, and analyzed. Product distributions in the
wastewater 30 min after chlorination were similar to those
in model solutions except for a large amount of radioactivity
(almost 20% of the total after 167 h), which eluted in the void
volume. Throughout the reaction period, the radioactivity
eluting in the void volume was equivalent to about half of the
amount of phenylalanine expected to form in the wastewater
at any given time as predicted by the model studies.
Phenylalanine is a nonpolar amino acid and may bind
reversibly to wastewater organic compounds and elute with
them. On the other hand, it is possible that a radioactive
intermediate, such as the proposed isocyanate precursor of
phenylalanine discussed above, may react with -OH or -NH2
moieties in polar wastewater humic substances to bind
irreversibly.
The stability of N,N-dichloroalanylphenylalanine (II) was
slightly less in wastewater than in model solutions with t1/ 2
) 3.6 ( 0.1 h (k1w ) 3.19 × 10-3 min-1, r 2 ) 0.999, where k1w
is the first-order decomposition rate constant in the waste-
water at 23 °C). The stability of the N-chloroketimine was
similar in both model solutions and in the wastewater with
t1/ 2 ) 138 ( 6.8 h (k2w ) 8.36 × 10-5 min-1, r 2 ) 0.990, where
k2w is the first-order decomposition rate constant in the
wastewater at 23 °C).
FIGURE 2. Measured and calculated chlorination product distribu-
tions of an alanyl-phenylalanine model solution (1.43 × 10-3 M, pH
7.0) chlorinated with 2 equiv and analyzed over 164 h.
Modeling the Decom position of Chlorination Products.
To demonstrate the relationship between the different
chlorination products II, III, unknown A, and IV, a model
based on three consecutive first-order reactions was devel-
oped. It was assumed that unknown A is an intermediate in
the decomposition of the N-chloroketimine (III) to phenyl-
alanine (IV). Initially, the rate constants k1m and k2m were
used to model the decomposition of II and corresponding
formation of III and the decomposition of III and corre-
sponding formation of unknown A. The rate constant for the
decomposition of unknown A (k3m), which corresponded to
the formation of phenylalanine (IV) was initially estimated.
Once the model was developed, all three rate constants were
manipulated so that calculated concentrations mirrored
experimental values. The measured and calculated (from
model) rate constants along with the corresponding half-
lives are shown in Table 1. While the calculated and measured
rate constants for the decomposition of the dichlorinated
dipeptide (k1m) are similar, the calculated rate constant for
the decomposition of N-chloroketimine (III) appears to be
significantly lower than the measured value. The measured
value of k2m is probably more accurate than the calculated
value because k2m had to be manipulated to model the
appearance of unknown A. However, since the concentration
of unknown A remains small throughout the 164 h of the
study, its measurement (and consequently the calculated
value of k2m) contains a much greater error than the
measurement of III. Figure 2 shows the measured and
calculated concentrations of chlorination products over a
period of 164 h. The correlation between the measured and
calculated concentrations of II, III, and unknown A strongly
suggests that unknown A is an intermediate in the formation
of phenylalanine, which supports the model. Measured
changes in the concentration of phenylalanine appear to
deviate from calculated concentrations. This appears to be
caused by degradation of phenylalanine in the solution or its
adsorption on the walls of the reaction vessel.
Mechanism s. When alanylphenylalanine is chlorinated
with 2 equiv, N,N-dichloroalanylphenylalanine (II) is formed,
which then dehydrohalogenates to form an N-chloroketimine
(III). The N-chloroketimine is then believed to undergo
â-elimination to form an unstable isocyanate intermediate.
Hydrolysis of the isocyanate leads to the formation of a
carbamic acid which, in turn, would lose CO2 to form
phenylalanine (IV). Figure 3 shows the proposed mechanism
for the decomposition of N,N-dichloroalanylphenylalanine
(II).
Im plications for Wastewater Treatm ent. Figure 4 is a
plot of the chlorination products found after 30 min in a
wastewater that had been chlorinated to various points along
the breakpoint curve. The tritiated dipeptide was used as a
tracer to follow the product formation. N-Chloroalanylphe-
nylalanine (I), N,N-dichloroalanylphenylalanine (II), and the
N-chloroketimine (III) are all observed at various chlorination
Scully and Keefe (22) showed that N,N-dichloroglycylphe-
nylalanine decomposes by two successive dehydrohaloge-
nation steps to form an N-chloroaldimine and an N-cyanoa-
cylphenylalanine intermediate. N-Cyanoacylphenylalanine
9
1 9 8 2 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 7, 1997

TABLE 1. Measured and Calculated Rate Constants and Half-Lives of Alanylphenylalanine Chlorination Products in a Model
Solution
rate constants (min-1 a
)
half-lives (h)a
compound
constant
measured
calculated
measured
calculated
N,N-dichloroalanylphenylalanine (II)
N-chloroketim ine (III)
unknown A
k1m
k2m
k3m
2.85 × 10-3
9.25 × 10-5
N/A
2.6 × 10-3
1.2 × 10-4
8.0 × 10-4
4.1 ( 0.2
125 ( 6.5
N/A
4.4
100
14.4
a
At 23 °C.
FIGURE 3. Proposed mechanism for the decomposition of N,N-dichloroalanylphenylalanine (II).
FIGURE 5. Product distribution in a wastewater (pH 7.0) inoculated
with alanyl-p-[3H]-phenylalanine that was chlorinated to 240 mg/L
and analyzed over 168 h.
FIGURE 4. Product distribution in a wastewater (pH 7.0) inoculated
with alanyl-p-[3H]-phenylalanine and chlorinated along the break-
point curve (40-280 mg/L).
levels. The breakpoint curve for the wastewater is also shown
in the figure.
appear. At the breakpoint (210 mg/ L chlorine dose) and
beyond, all of the dipeptide had been converted to N,N-
dichloroalanylphenylalanine (II). This was to be expected as
free chlorine was now present in the wastewater.
At a chlorine dose of 40 mg/ L all of the dipeptide had been
converted to N-chloroalanylphenylalanine. This may seem
unusual because the residual chlorine concentration suggests
that all amino nitrogens were not fully chlorinated until a
dose of 120 mg/ L (chloramine maximum) had been reached.
However, this type of observation has been observed in a
previous study involving chlorination of the amino acid
phenylalanine (19). Two possible factors could account for
this: the more rapid direct chlorination of the dipeptide by
free chlorine (hypochlorous acid) (28) and chlorine transfer
to the organic nitrogen compound from inorganic chloram-
ines (29) within 30 min. Beyond the chloramine maximum
(120 mg/ L), the amount of N-chloroalanylphenylalanine (I)
decreased as N,N-dichloroalanylphenylalanine (II) and its
decomposition product, an N-chloroketimine (III), began to
Figure 5 shows the product distributions in a wastewater
inoculated with the tritiated dipeptide and chlorinated to
240 mg/ L. At this chlorine level all of the dipeptide is
converted to the dichlorinated analog. The plot traces the
decomposition of the dichlorinated dipeptide and its de-
composition products over a period of 168 h. The decom-
position rates of N,N-dichloroalanylphenylalanine (t1/ 2 ) 3.6
( 0.1 h) and the N-chloroketimine (t1/ 2 ) 138 ( 6.8 h) are very
similar to those found in the model solutions. However, there
are some differences in the product distribution between the
two solutions. In the wastewater there appeared to be
significant binding of the chlorinated dipeptide analogs to
organic matter in the solution incubated at room temperature.
9
VOL. 31, NO. 7, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 9 8 3

(10) Marks, H. C.; Strandskov, F. B. Ann. N.Y. Acad. Sci. 1950, 53,
163-171.
(11) Wolfe, R. L.; Ward, N. R.; Olson, B. H. J. Am. Water Works Assoc.
1984, 76 (5), 74-88.
(12) Wolfe, R. L.; Olson, B. H. In Water Chlorination: Chemistry,
Environmental Impact and Health Effects; Jolley, R. L., et al.,
Eds.; Lewis Publishers: Chelsea, MI, 1985; Vol. 5, pp 555-573.
(13) Wolfe, R. L.; Ward, N. R.; Olson, B. H. Environ. Sci. Technol.
1985, 19, 1192-1195.
(14) Johnson, J. D. In Water Chlorination: Environmental Impact
and Health Effects; Jolley, R. L., Ed.; Ann Arbor Science: Ann
Arbor, MI, 1978; Vol. 1, pp 37-63.
(15) Jensen, J. N.; Johnson, J. D. Environ. Sci. Technol. 1990, 24 (7),
981-985.
(16) Jensen, J. N.; Johnson, J. D. Environ. Sci. Technol. 1990, 24 (7),
985-990.
(17) Nweke, A.; Scully, F. E., Jr. Environ. Sci. Technol. 1989, 23, 989-
994.
(18) McCormick, E. F.; Conyers, B.; Scully, F. E., Jr. Environ. Sci.
Technol. 1993, 27, 255-261.
As a result, a large amount of radioactivity eluted in the void
volume. Chlorination products of phenylalanine have also
been found to bind to organic matter in wastewater (19).
Of more importance to current wastewater disinfection
practices are the reactions occurring at the higher chlorination
levels. In order to minimize nutrient loading into rivers and
bays, many treatment plants have begun to nitrify their
wastewaters. As a result, ammonia concentrations in these
effluents are generally low as compared to organic nitrogen
compounds. Consequently, compounds such as dipeptides
and proteins comprise a larger fraction of the total amino
nitrogen. This can lead to difficulty in determining if a
wastewater has been adequately disinfected, by interfering
with the measurement of the bactericidal potential in the
wastewater. As discussed earlier, the organic chloramines
formed have little or no bactericidal activity, and current
techniques used to measure the oxidant level of the waste-
water cannot distinguish between organic and inorganic
chloramines (8-16). The toxicological impact of an increase
of these chlorination products being released to a receiving
body is also unknown.
(19) Conyers, B.; Scully, F. E., Jr. Environ. Sci. Technol. 1993, 27,
261-266.
(20) Scully, F. E., Jr.; Winters, D. S.; Conyers, B. By-Products of the
Chlorination of Methionine: Products and Implications for
Disinfection of Wastewater. Presented at the 204th National
Meeting of the American Chemical Society, Washington, DC,
Aug 23-28, 1992.
Acknowledgments
We are grateful to the Hampton Roads Sanitation District
Commission for their cooperation in providing total Kjeldahl
nitrogen and ammonia analyses on wastewater samples. This
research was supported by the National Science Foundation
Grant BCS-9218565, Dr. Edward Bryan, project manager. Any
opinions, findings, and conclusions or recommendations
expressed in this publication are those of the authors and do
not necessarily reflect the views of the National Science
Foundation.
(21) Conyers, B.; Scully, F. E., Jr. Environ. Sci. Technol. 1997, 31,
1680-1685.
(22) Keefe, D. J.; Fox, T. C.; Conyers, B.; Scully, F. E., Jr. Environ. Sci.
Technol. 1997, 31, 1973-1978.
(23) Bodanszky, M.; Bodanszky, A. In The Practice of Peptide Synthesis;
Springer-Verlag: New York, 1994; p 39.
(24) Scully, F. E., Jr.; Howell, G. D.; Penn, H. H.; Mazina, K.; Johnson,
J. D. Environ. Sci. Technol. 1988, 22, 1186-1190.
(25) Franson, M. A., Ed. Standard Methods for the Examination of
Water and Wastewater, 18th ed.; American Public Health
Assoc.: Washington, DC, 1992.
(26) Nickelson, M. G.; Nweke, A.; Scully, F. E., Jr.; Ringhand, H. P.
Chem. Res. Toxicol. 1991, 4, 94-101.
(27) Stelmaszynska, T.; Zgliczynski, Eur. J. Biochem. 1978, 89, 301-
308.
(28) Morris, J. C. In Principles and Applications of Water Chemistry;
Faust, S. D., Hunter, J. V., Eds.; Wiley: New York, 1967; pp 23-
53.
(29) Isaac, R. A.; Morris, J. C. In Water Chlorination:Environmental
Impact and Health Effects; Jolley, R. L., et al., Eds.; Ann Arbor
Science: Ann Arbor, MI, 1983; Vol. 4, pp 63-75.
Literature Cited
(1) Brungs, W. A. J. Water Pollut. Control Fed. 1973, 45, 2180-2193.
(2) Brungs, W. A. Effects of Wastewater and Cooling Water Chlorina-
tion on Aquatic Life; Ecological Research Series; U.S. EPA:
Washington, DC, 1976; EPA-600/ 3-76-078.
(3) Tomkins, J. A.; Tsai, C. Trans. Am. Fish. Soc. 1976, 105, 313-321.
(4) Merkins, J. C. Water Waste Treat. J. 1958, 7, 150.
(5) Doudoroff, P.; Katz, M. Sewage Ind. Wastes 1950, 22 (11), 1432-
1458.
(6) McKee, J. E.; Wolf, H. W. Water Quality Criteria, 2nd ed.; California
State Water Quality Control Board Publication 3-A; U.S. Gov-
ernment Printing Office: Washinton, DC, 1978; 548 pp.
(7) Capuzzo, J. M., Davidson, J. A., Lawrence, S. A.; Libni, M. Estuarine
Coastal Mar. Sci. 1977, 5, 733-741.
Received for review September 16, 1996. Revised manuscript
received February 10, 1997. Accepted February 21, 1997.X
ES9607953
(8) Scully, F. E., Jr.; Hartman, A. C.; Rule, A.; LaBlanc, N. Environ.
Sci. Technol. 1996, 30, 1465-1471.
(9) Feng, T. H. J. Water Pollut. Control Fed. 1966, 38, 614-628.
X Abstract published in Advance ACS Abstracts, May 1, 1997.
9
1 9 8 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 7, 1997