E2-elimination in the decomposition of N-bromoamino acid anions

Article abstract of DOI:10.1021/jo9506342

The kinetics of oxidation of eight structurally different amino acids by hypobromite ion (BrO^-) in the presence of hydroxide ion has been studied. The reactions proceed through the rapid formation of N-bromoamino acid anion which then decomposes in the rate-limiting step. The decomposition of N-bromoamino acid anions is found to proceed through an unimolecular and a base-catalyzed reaction. The large negative values of ΔS^? and the products formed suggest that the hydroxide ion-catalyzed reactions proceed through an E₂ mechanism. N-Bromoamino acid anion gives an absorption spectrum with λ_max at ≈290 nm.

Full text of DOI:10.1021/jo9506342

4336
J . Org. Chem. 1996, 61, 4336-4341
E2-Elim in a tion in th e Decom p osition of N-Br om oa m in o Acid
An ion s
M. S. Ramachandran* and D. Easwaramoorthy
School of Chemistry, Madurai Kamaraj University, Madurai 625 021, India
S. Vasanthkumar
Department of Chemistry, Thiagarajar College of Engineering, Madurai 625 015, India
Received April 3, 1995X
The kinetics of oxidation of eight structurally different amino acids by hypobromite ion (BrO-) in
the presence of hydroxide ion has been studied. The reactions proceed through the rapid formation
of N-bromoamino acid anion which then decomposes in the rate-limiting step. The decomposition
of N-bromoamino acid anions is found to proceed through an unimolecular and a base-catalyzed
reaction. The large negative values of ∆Sq and the products formed suggest that the hydroxide
ion-catalyzed reactions proceed through an E2 mechanism. N-Bromoamino acid anion gives an
absorption spectrum with λmax at ≈290 nm.
Nitrogen-containing organic molecules, especially amino
acids, can be oxidized to N-halo derivatives by a wide
variety of halogenating agents. The N-haloamino acids
themselves are halogenating agents, and hence, their
decomposition is of environmental interest. There has
been great progress in the past toward the quantitative
understanding of the kinetics and thermodynamics of the
reactions of N-chloroamino acids,1-7 since chlorine is
usually employed in the disinfection of potable water
and waste water.8 Even though the accepted mecha-
nism for the reaction of aqueous bromine with amino
acid is analogous to that of N-chloroamino acids,8-11
kinetics have received less attention probably because
of the complexity of the reactions.11 In this paper we
have given the details of the kinetics of oxidation of eight
structurally different amino acids (AA) by aqueous hy-
pobromite ion.
Resu lts
We have examined the rate of oxidation under the
condition [OH-]T g [AA], and we occassionally refer to
the term [OH-]f as free hydroxide ion concentration,
where [OH-]f ) [OH-]T - [AA]. All the kinetics were
carried out under pseudo-first-order conditions ([AA] .
[BrO-], i.e., at least in the ratio of 10:1) and usually at
[OH-]f ≈ 6.5 × 10-3 M.
Figure 1 is the semilog plot of the concentration of the
oxidizing species versus time. The semilog plots show
X Abstract published in Advance ACS Abstracts, May 15, 1996.
(1) Fox, S. W.; Bullock, M. W. J . Am. Chem. Soc. 1951, 73, 2754.
(2) Dennis, W. H.; Hull, L. A.; Rosenblott, D. H. J . Org. Chem. 1976,
32, 3783.
(3) Becker, K. B.; Grob, C. A. In Supplement A: The Chemistry of
Double-Bonded Functional Groups; Patai, S., Ed.; Wiley: New York,
1977; pp 653-723.
(4) Kaminski, J . J .; Bodor, N.; Higuchi, T. J . J . Pharm. Sci. 1976,
65, 553.
(5) Hand, V. C.; Snyder, M. P.; Margerum, D. W. J . Am. Chem. Soc.
1983, 105, 4022.
(6) Awad, R.; Hussain, A.; Crooks, P. A. J . Chem. Soc., Perkin Trans.
2 1990, 1233.
F igu r e 1. Plot of log [OBr-] vs. time at 35 °C: (A) [glycine]
) 0.05 M; [OH-]f ) 8.46 × 10-3 M; (B) [alanine] ) 0.05 M;
[OH-]f ) 6.46 × 10-3 M; (C) [â-alanine] ) 0.05 M; [OH-]f )
0.1065 M.
very good correlation between log [OBr-] and time as
evidenced by the high correlation coefficients (usually r
> 0.99). The pseudo-first-order rate constants, kobs
,
calculated from the semilog plots are independent of the
initial concentrations of the oxidant used and, thus,
(7) Losada, M.; Santaballa, J . A.; Armesto, X. L.; Antelo, J . M. Acta
Chim. Hung. 1992, 129, 535.
(8) Isaac, R. A.; Morris, J . C. In Water chlorination, Environmental
Impact and Health Effects; J olley, R. L., Brungs, W. A., Cummings,
R. B., Eds.; Ann Arbor Science: Ann Arbor, MI, 1980; Vol. III, Chapter
17.
(9) Friedman, A. H.; Morgulis, S. J . Am. Chem. Soc. 1936, 58, 909.
(10) Stanbro, W. D.; Smith, W. D. Environ. Sci. Technol. 1979, 13,
446.
(11) Stanbro, W. D.; Lenkevich, M. J . Int. J . Chem. Kinet. 1983, 15,
1321.
S0022-3263(95)00634-7 CCC: $12.00 © 1996 American Chemical Society

Decomposition of N-Bromoamino Acid Anions
J . Org. Chem., Vol. 61, No. 13, 1996 4337
F igu r e 2. Plot of kobs vs. [OH-]f at 35 °C: (A) [glycine] ) 0.05
M; [OBr-] ) 3.56 × 10-3 M; (B) [alanine] ) 0.05 M; [OBr-] )
3.54 × 10-3 M; (C) [N-methylglycine] ) 0.05 M; [OBr-] ) 3.50
× 10-3 M.
F igu r e 3. Plot of kobs vs. [amino acid] at 35 °C ([OH-]f ) 6.44
× 10-3 M; [OBr-] ) 3.56 × 10-3 M): (A) glycine, (B) alanine,
(C) N-methylglycine.
confirm that the rate of oxidation is first order in [oxi-
dant]. Moreover, the initial concentration of the oxidant,
[OBr-]0 obtained by the extrapolation of the semilog plot,
agrees well to the analytical value within the limits of
experimental error ((5%). An interesting observation is
that, of all the amino acids studied, the rate of oxidation
of R-aminoisobutyric acid alone is immeasurably fast
(half-life is less than 1 min) under all conditions used in
this study.
Ta ble 1. Ra te Con sta n ts for th e Decom p osition of
N-Br om o-r-a m in o Acid An ion a t 35 °Ca
103k1
103k2
103k3
bromoamine of
(s-1
)
(M-1 s-1
)
(M-1 s-1
)
glycine H2NCH2COO-
324.0
186.2
86.7
460.2
169.0
1.7
14.7
10.7
alanine NH2CH(CH3)COO-
4.0
2.9
5.9
1.9
valine NH2CH(CH(CH3)2)COO-
phenylalanine NH2CH(CH2C6H5)COO-
N-methylglycine CH3NHCH2COO-
â-alanine NH2CH2CH2COO-
24.0
The influence of [OH-]f on kobs is found to be a simple
first order reaction (Figure 2) as kobs ) k1 + k2[OH-]f in
all the amino acids studied. The pseudo-first-order rate
constant shows a linear dependence on the amino acid
concentrations (Figure 3) in glycine, N-methylglycine,
and alanine only. Perusal of the results shows that the
pseudo-first-order rate constant for the decomposition of
N-bromoamino acid anion satisfies the equation (1). The
a
Average of the values obtained from different plots are given.
N-Bromothreonine gives kobs ) 6.18 × 10-3 at 6.5 × 10-3 M of
[OH-]f, and k1 and k2 could not be separated due to a very large
increase in kobs with [OH-]f. N-Bromo-R-aminoisobutyric acid
decomposes rapidly, and we could not measure the k values.
new absorption peaks appeared immediately on mixing
amino acid and OBr-. The spectra of the intermediate
showed two peaks, one at ∼290 nm and the other at
∼225-230 nm (Figure 4). The peak at the shorter
wavelength (∼225-230 nm) was highly asymmetric and
the absorbance came down sharply on the shorter wave-
length side and so the λmax value may involve some error.
Analysis of the spectra showed that the lower energy
transition appeared at the same wavelength for all amino
acids irrespective of the substituents at the amino carbon.
A red shift of ∼10 nm was observed for a methyl
substitution at the amino nitrogen. This longer wave-
length λmax is listed for the intermediate from various
amino acids in Table 3. The time history of the absorp-
kobs ) k1 + k2[OH-]f + k3[AA]
(1)
values of k1, k2, and k3 for various amino acids are
tabulated in Table 1. Ionic strength (0.1-0.3 M) and
bromide ion (0.01-0.075 M) have no effect on kobs. From
the effect of temperature on kobs, the enthalpy of activa-
tion and entropy of activation for the rate constants k1,
k2, and k3 are calculated for glycine, N-methylglycine, and
alanine and given in Table 2.
The UV-vis absorption spectra of the reaction inter-
mediates were also recorded. The absorption peak12 of
OBr- was observed at 330 nm with ꢀmax 350 dm3 mol-1
cm-1. The peak due to OBr- completely disappeared, and
(12) Soulard, M.; Block, F.; Hatterer, A. J . Chem. Soc., Dalton Trans.
1981, 2300.

4338 J . Org. Chem., Vol. 61, No. 13, 1996
Ramachandran et al.
Ta ble 2. Th er m od yn a m ic P a r a m eter s for th e
Decom p osition of N-Br om oglycin e,
i
N-Br om o-N-m eth ylglycin e, a n d N-Br om oa la n in ea
N-bromoamino
acid anion
∆Hq
∆Sq
∆Gq
(kcal/mol) (cal‚deg-1/mol) (kcal/mol)
glycine
k2
k3
15
15
29 (27.5)
10
13
-13
19
21
22
19
20
22
19
-17
N-methylglycine k1
+23 (14.6)
-29
k2
k3
k1
k2
-22
alanine
25 (25.6)
11
+11 (10.7)
-28
a
k1, k2, and k3 represent the process of decomposition as shown
in Scheme 1. Values in parentheses correspond to N-chloroamino
acid (from ref 5). ∆Gq value at 35 °C.
ii
tion spectrum showed that the absorbance of both peaks
are decreasing.
The stoichiometry of the reaction may be represented
as
1 amino acid + OBr- f products
(2)
The oxidative product of amino acid was identified as the
corresponding carbonyl compounds, ammonia and carbon
dioxide. Glycine and N-methylglycine were oxidized to
only glyoxalic acid, and the percentage yield was >80%.
Alanine gave a mixture of acetaldehyde (70 ( 5%) and
pyruvic acid (∼25 ( 5%). The yield of acetone in R-amino
isobutyric acid was 90%.
iii
Discu ssion
In the presence of alkali, the amino acids would be in
the form of an amino acid anion according to the following
reaction.
RCH(NH3+)COO- + OH-
f
RCH(NH2)COO- + H2O (3)
The conversion into amino acid anion may be quantita-
tive since the pKa values for most of the amino acids are
pKa(1) ) 2.1 ( 0.3 and pKa(2) ) 9.6 ( 0.7. Therefore, it is
reasonable that any excess hydroxide ion concentrations
over the stoichiometry of the reaction (3) will be available
for interaction with intermediates, and we can define this
as free hydroxide ion concentration [OH-]f; that is
F igu r e 4. Time history of absorption spectrum of the inter-
mediate ([amino acid] ) 0.10 M; [OH-]f ) 6.5 × 10-3 M; [OBr-]
) 3.5 × 10-3 M): (i) glycine A ) 1 min; B ) 3 min; C ) 6 min;
D ) 10 min; (ii) alanine A ) 1 min; B ) 3 min; C ) 5 min; D
) 7 min; E ) 10 min; (iii) N-methylglycine A ) 1 min; B ) 2.5
min; C ) 3.5 min; D ) 5 min; E ) 7 min.
practically does not change with the concentration of
amino acid anion, while a Beer’s law type of relationship
is obtained with [OBr-]. Further, we could not observe
any initial increase in the absorbance at ∼290 nm; that
is, the peak appears instantaneously (i.e., <1 min) in all
the amino acids studied. This suggests that the reaction
between amino acid anion and hypobromite ion is very
fast and can be represented by the following equation.
[OH-]f ) [OH-]T - [AA]
Hereafter, by amino acid (AA) we refer only to amino acid
anion.
To have insight into the reaction mechanism, it is
necessary that the intermediate(s) indicated by the
absorbance at ∼225 nm and ∼290 nm be identified. The
generally accepted mechanism for the reaction between
amino acids and hypobromite, analogous to the hypochlo-
rite, is the formation of N-bromoamino acids,8-11 as
shown below.
RCH(COO-)NH2 + OBr- fast8
RCH(COO-)NHBr + OH- (5)
RCH(COO-)NH2 + HOBr fast8
RCH(COO-)NHBr + H2O (4)
Therefore, the species responsible for the absorption
around 290 nm may be N-bromoamino acid anion which
can be considered the λmax of simple N-bromamine (λmax
280 nm)13 influenced by the alkyl substituent at the
nitrogen atom. This is also supported by the fact that
the λmax of N-bromoamino acid anion is not affected either
by the nature of the alkyl substituent at the R-carbon or
It was observed that the intermediate in â-alanine is
quite stable relative to the intermediates in all the other
R-amino acids. As a result, we find out the relationship
between the absorbance of the longer wavelength peak
(at 290 nm), [â-alanine] and [OBr-]. The absorbance
(13) Galal-Gorchev, H. A.; Morris, J . C. Inorg. Chem. 1965, 4, 899.

Decomposition of N-Bromoamino Acid Anions
J . Org. Chem., Vol. 61, No. 13, 1996 4339
Ta ble 3. Lon ger Wa velen gth Absor p tion Ma xim u m of
Ta ble 4. Rela tive Ra te of Decom p osition of
N-Br om oa m in o Acid An ion a t 35 °C
N-Br om oa m in o Acid An ion a t 35 °C
N-bromo-R-amino acid aniona
λmax (nm)
relative rate
a
N-bromoglycine
N-bromoalanine
N-bromo-N-methylglycine
N-bromo-â-alanine
290
287
305
290
N-bromo-R-amino acid anion
kobs
k2
k1
N-bromoglycine
N-bromoalanine
N-bromovaline
N-bromophenylalanine
N-bromo-N-methylglycine
1.00 (1.00)
1.95 (64.0)
1.30
1.00
0.58 1.00
0.27 0.73 (0.75)b
1.42 1.49
Spectra were recorded at [OH-]f ) 6.5 × 10-3 M.
a
3.68
1.60 (12.0)
0.52 0.49 (0.31)b
(0.19)c
by moving the carboxylate group away from the amino
carbon (e.g., â-alanine). However, the substitution of a
methyl group at the amino nitrogen, which can be
considered as the dialkyl-substituted N-bromamine, shifts
this maximum peak to a longer wavelength.
N-bromo-â-alanine
N-bromothreonine
N-bromo-R-aminoisobutyric
acid
∼0.02
∼0
2.28 (50.0)
very high
(3100.0)
[OH-]f ) 6.5 × 10-3 M. Values given in the parantheses
correspond to N-chloroamino acid anion from ref 5. The relative
rate for k1 is determined with respect to N-bromoalanine since k1
a
The oxidative decarboxylation of R-amino acids by
other oxidants such as HSO5 in alkaline medium also
-
b
could not be determined for N-bromoglycine. Value calculated
produces an intermediate/product which gives a highly
asymmetric peak around 220 nm.14 Therefore, in com-
parison with other oxidants, the intermediate species
responsible for the λmax at ∼230 nm can be assigned to
the imine.
On the basis of the experimental results, we can pro-
pose the following reaction scheme for the oxidation of
amino acids by hypobromite ion as follows in Scheme 1.
from ref 6. c Value calculated from ref 5.
Ta ble 5. F r a gm en ta tion a of N-Br om oa m in o Acid s
% fragmentation through
unimolecular OH- catalyzed AA catalyzed
bromoamine of
(k1)
(k2)
(k3)
glycine
74
21
16
34
34
26
9
alanine
valine
phenylalanine
N-methylglycine
70
84
66
59
Sch em e 1
7
RCH(COO-)NH2 + OBr- fast8
RCH(COO-)NHBr + OH- (5)
a
Values are calculated from the k values at [AA] ) 0.05 M,
[OH-]f ) 6.5 × 10-3 M, and [OBr-] ) 3.5 × 10-3 M at 35 °C.
process (k1) is almost identical in N-bromovaline and
N-chlorovaline. This suggests that the mechanism for
the decomposition of N-bromo amino acids through the
process represented by k1 is unimolecular as its chlorine
analogue. This is also supported by the large positive
entropy value observed for k1 in N-methylglycine and
alanine (Table 2).
k1
9
RCH(COO-)NHBr
8 RCHdNH + Br- + CO2
(6)
k2
RCH(COO-)NHBr + OH- 9
8
RC(COO-)dNH + H2O + Br- (7)
k3
The percentage of fragmentations through unimolecu-
lar hydroxide ion and amino acid anion catalyzed pro-
cesses are calculated for various N-bromoamino acids
from the rate constant values, and they are given in Table
5. Comparison of the values in Table 5 with product
yields (e.g., alanine) supports the calculated fragmenta-
tion percentage. Analysis of the results in Table 5
indicates that the major fraction of the alkyl side chain
R-amino acids decompose through the unimolecular
process (k1). To explain this, we can first consider a
reaction scheme in which the breaking up of the C-COO-
bond preceeds the carbon nitrogen bond formation result-
ing in a carbanion intermediate. This mechanism is very
similar to the one proposed by Awad et al.6 for the
unimolecular decomposition of N-chloroamino acid anion.
If carbanion intermediate is formed in the rate-limiting
step, we should expect that N-bromoglycine would de-
compose faster than N-bromoalanine since the former
compound forms a more stable carbanion intermediate.
On the other hand, in the disappearance of carbanion,
the more stable the anion the slower the reaction would
be, which is consistent with the observation that N-
bromoalanine decomposes faster than N-bromoglycine.
This will also explain the fact that the oxidation of
R-aminoisobutyric acid is very fast since it forms a more
unstable (more substituted) carbanion. Though the
mechanism of disappearance of the carbanion intermedi-
ate explains the observed variation of the unimolecular
decomposition k1, it does not rationalize the thermody-
namic parameters which are independent of the nitrogen-
RCH(COO-)NHBr + RCH(NH2)COO- 9
RC(COO-)dNH + RCH(NH2)COO- + H+ + Br- (8)
8
The imines will hydrolyze to give the corresponding
carbonyl compounds. The observed pseudo-first-order
rate constant for the disappearance of [OBr-] from the
reactions 5-8 is given as
kobs ) k1 + k2[OH-] + k3[AA]
(9)
The values of k1, k2, and k3 calculated from various plots
such as kobs vs [OH-] and kobs vs [AA] are agreeable within
the limits of experimental error, and the average values
are given in Table 1.
The unimolecular decomposition of N-chloro-R-amino
acids in alkaline medium has been studied extensively,
and several mechanisms have been proposed.1-6 The
accepted mechanism is the one in which the chlorine and
carboxylate groups are in antiperiplanar configuration
and the N-chloro compound decomposes via a slow
concerted fragmentation, i.e., an E2-like mechanism. This
mechanism explains the observed changes in the rates
with change of substituents. The relative rates of
decomposition of N-bromoamino acids along with their
chlorine analogues are given in Table 4. Analysis of the
results shows that the relative rate for unimolecular
(14) Ramachandran, M. S.; Easwaramoorthy, D.; Sureshkumar, D.
Tetrahedron 1994, 50, 9495.

4340 J . Org. Chem., Vol. 61, No. 13, 1996
Ramachandran et al.
halogen bond (Table 2). The mechanism that would be
more consistent with the observed results may be the
concerted E2-like mechanism in which decarboxylation
and debromination occur in a single step via an imine-
like transition state as suggested by Hand and Margerum
in N-chloroamino acids.5 In the activated state carbon-
nitrogen bond formation and C-COO- bond fission may
occur to a large extent while the change in the nitrogen-
halogen bond would be relatively small. Since the -I
effects of Cl and Br are almost equal (as measured from
the σ* values),15 one cannot expect much energy differ-
ence between the activated states of N-chloro- and
N-bromoamino acids. Since the bonded halogen atom
would be solvated appreciably,5 the change in the degree
of solvation of halogen, as it gains (partial) negative
charge, is negligible. Thus, the contribution for ∆Sq from
the N-Cl and N-Br bonds is almost identical. This will
explain why the ∆Sq values for N-chloro- and N-bro-
moalanine are identical.
The influence of substituents on the stability of sub-
stituted imines may be same as that of the stability of
olefins.16 Thus, the transition state with highly substi-
tuted imines may be more stable and more reactive.
Each methyl substitution at the amino carbon increases
the unimolecular decomposition of N-chloroamino acids5
by a factor of ∼60. The similar trend may hold true for
N-bromoamino acid also. This will explain why N-
bromoglycine disintegrates very slowly while the N-bromo-
R-amino isobutyric acid decomposes rapidly.
For unimolecular decomposition, the halogen element
effect (k1Br/k1Cl) is calculated for alanine and N-meth-
ylgycine and the values are 3.6 and 7.6,17 respectively.
Thus, the halogen atom effect is more pronounced in
N-substituted amino acid. Comparison of the results for
glycine and N-methylglycine in Table 5 indicates that the
methyl substitution at the amino nitrogen enhances the
fragmentation through the unimolecular process. These
two observations indicate that the mechanism of unimo-
lecular fragmentation of N-bromo-N-methylglycine may
be different from other N-bromoamino acids. Some
authors7,18 have suggested the formation of nitrenium ion
as an alternative for the E2-like unimolecular mechanism
in N-chloroamino acids. Therefore, N-bromo-N-methyl-
glycine may disintegrate through the formation of a
nitrenium ion intermediate which would be stabilized by
the electron-donating methyl group. Moreover, a nitre-
nium ion would be expected to yield substitution and/or
elimination products. This will explain why the major
reaction product in sarcosine is glyoxalic acid rather than
formaldehyde.
N-Bromoglycine and to a smaller extent (∼30-40%)
other N-bromoamino acids disintegrate through the
hydroxide ion- and R-amino acid anion-catalyzed mech-
anism. We can suggest a E2 mechanism for the base-
catalyzed reaction in which the R-hydrogen and bromine
atom would be in an antiperiplanar configuration, thereby
eliminating hydrogen bromide (Figure 5). The resultant
intermediate R-imino acid may rapidly hydrolize to give
R-carbonyl carboxylate. The large negative entropies of
activation observed for k2 and k3 are also consistent with
the reactions involving E2 elimination of hydrogen halide
from N-chloro compounds.19
F igu r e 5. Mechanistic scheme.
The factors that influence the carbon-heteroatom
multiple bond formation through â-elimination are (i)
lower stability of the heteroatom-leaving group bond and
(ii) increased basicity of â-hydrogen.20 Thus, the observed
variation in the base-catalyzed reaction contants (k2 and
k3) with the structure of amino acid should come from
the relative basicity of the hydrogen atom at the R-car-
bon.
Earlier reports on the oxidation of R-amino acids
suggest that the hydrogen atom at the R-carbon is
slightly acidic and can interact with hydroxide ion.1,14
This interaction is influenced by the substituent R at the
R-carbon, and the observed order14 is H > CH3 > (CH3)2-
CH. Perusal of the k2 values in Table 1 shows that the
observed values are also in the above order. Thus, the
inductive effect (+I) of alkyl substituent R is an impor-
tant factor in the base-catalyzed elimination. A simple
linear free energy relationship (LFER) analysis21 of k2
with Taft’s constant σ* shows a reasonably good correla-
tion (r ) 0.967) even with limited amino acid anions such
as glycine, alanine, and valine. The observed relation-
ship is log k2 ) -0.86 + 0.79σ*. Phenylalanine shows a
positive vertical deviation (to this correlation line), and
this could be assigned to the R effect.22 Similar deviation
is exhibited by phenylalanine in the reaction with N-
bromosuccinimide also.23 Even though great reliance
cannot be placed on the absolute value of the reaction
constant F*, the positive value indicates the presence of
the carbanion character at the R-carbon in the transition
state; i.e., the transition state has appreciable scission
of the C-H bond. We would expect that an electron-
donating/releasing substituent at the R-position (relative
to the halogen) may slightly increase the base-catalyzed
reaction constant k2 or at least should remain the same
as that of the unsubstituted one.24 Comparison of the k2
values of N-bromo-N-methylglycine and N-bromoglycine
shows that the methyl substitution at the nitrogen atom
actually decreases the rate. This means that the electron-
donating/releasing substituent at the nitrogen, i.e., R-po-
sition, probably destabilizes the incipient negative charge
or double bond.25 This is in accordance with the proposed
(18) Armesto, X. L.; Canle, M.; Losada, M.; Santaballa, J . A. Int. J .
Chem. Kinet. 1993, 25, 1.
(19) Bartsch, R. A.; Cho, B. R. J . Am. Chem. Soc. 1979, 101, 3587.
(20) Hine, J . Physical Organic Chemistry; McGraw-Hill: New York,
1962; p 207.
(15) Lange’s Hand Book of Chemistry; Dean, J . A., Ed.; McGraw-
Hill: New York, 1979; Section 3, pp 134-135.
(21) Pavelich, W. A.; Taft, R. W. J . Am. Chem. Soc. 1957, 79, 4935.
(22) Edwards, J . O.; Pearson, R. G. J . Am. Chem. Soc. 1962, 84, 16.
(23) Ramachandran, M. S.; Easwaramoorthy, D.; Rajasingh, V.;
Vivekanandam, T. S. Bull. Chem. Soc. J pn. 1990, 63, 2393.
(24) Reference 16, p 883.
(16) March, J . Advanced Organic Chemistry, 3rd ed.; Wiley East-
ern: New Delhi, 1986; p 889.
Cl
(17) The k1 values at 35 °C are calculated from the data in ref 5.

Decomposition of N-Bromoamino Acid Anions
J . Org. Chem., Vol. 61, No. 13, 1996 4341
polyatomic molecules29 suggest that the nitrogen-
bromine bond is less stable by 14.0 kcal/mol than the
nitrogen-chlorine bond. Therefore, the lower stability
of heteroatom-leaving group bond (i.e., >N-Br) is
mainly responsible for the observed difference between
N-bromoamino acid and N-chloroamino acids toward the
base-catalyzed elimination.
Exp er im en ta l Section
All the chemicals used were of highest purity and used as
received unless stated otherwise. R-Aminoisobutyric acid (Aib)
was from Sigma (USA), and all other amino acids were from
Loba chemie (India). The ionic strength was adjusted to 0.15
M with sodium perchlorate produced from the neutralization
of Na2CO3 with HClO4.
Aqueous solutions of hypobromite ion were prepared by
dissolving bromine in an aqueous solution of sodium hydroxide.
The formation of hypobromite ion is rapid as shown in eq 10.
F igu r e 6.
activated state in which the C-H bond fission and C-N
bond formation occur to a large extent.
The amino acid anion can be expected to act as a base
since the pKa and pKb of classical amino acid RCH(NH2)-
COOH is ≈9.8.26 The glycine anion-catalyzed reaction
is also reported in the oxidation by N-bromosuccinimide
in alkaline medium.23 Therefore, the k3 values tabulated
(Table 1) for glycine, N-methylglycine, and alanine
represent the amino acid anion-catalyzed elimination
reactions.
Br2 + 2OH- f BrO- + Br- + H2O
(10)
Since the rate of disproportionation of BrO- to BrO3- is at least
at higher pH (∼13.0),30 an excess of OH- over Br2 (usually
2[OH-] - [Br2] ≈ 0.035 M) was kept in the stock solution.
Hypobromite solutions were prepared fresh daily just before
starting the experiments. The OBr- concentration was esti-
mated by iodometry, while its purity was ascertained by
spectrometry at 330 nm.13
The kinetics of decomposition were followed by measuring
the concentration of active bromine by iodometry. Hypobro-
mite solution was rapidly injected into the reaction mixture
with an excess (usually ∼10-fold) of amino acid anion. The
[OBr-] remaining at different times was measured by arresting
the reaction through the addition of acidified KI solution (10%
w/v) into the reaction mixture. Rate constants were deter-
mined from the slope of the least-squares regression of log
[OBr-]t vs time. For the calculation of [OH-]f in the kinetic
expression, the concentration of OH- ion produced as in eq 5
was also taken into account.
UV-vis absorption spectra were recorded (200-360 nm) on
a Schimatzu UV-160 spectrophotometer using 1 cm matched
cells. Corrections for the absorbance of amino acid, if any,
were made.
Glyoxalic acid was detected by the reaction with phenylhy-
drazine and H2O2.31 The presence of pyruvic acid as a minor
oxidation product was established by the color reaction with
â-naphthylamine.32 The concentration of carbonyls were
determined spectrophotometrically by their (2,4-dinitrophen-
yl)hydrazones as given by Wells.33 Estimation of glyoxalic acid
was optimized by generating a calibration curve from a stock
solution. Acetaldehyde and pyruvic acid were estimated from
the (standardized) ꢀ values at 425 and 550 nm.
The functional group -CH2COO- with a σ* value of
-0.06 is a poor electron donor compared to -COO- (σ*
) -1.06).15 Perusal of the structure of amino acids in
Table 1 suggests that â-alanine should be more reactive
than glycine toward E2 elimination, contrary to what is
observed in this study. The unusual stability of N-bromo-
â-alanine suggests that some other force is also operating
additionally. The more probable reason may be that the
conformation with bromine and carboxylate on the same
side of the carbon-nitrogen bond may reduce the reactiv-
ity of the bromine by the interaction with carboxylate
oxygen as in (Figure 6). If it is so, one would expect that
N-bromo-γ-aminobutyric acid anion, BrNHCH2CH2CH2-
COO-, should also be stable toward the base-catalyzed
elimination of HBr. This is found to be true from the
rate constant value for k2 which is of the same order as
that of â-alanine.27 This observation also supports the
fact that the Br and H atoms are in an antiperiplanar
configuration, which is favorable for E2 elimination.
Finally, we have to explain the important difference
in the decomposition of N-chloroamino acid anions and
N-bromoamino acid anions. The N-bromoamino acid
anions undergo a base-catalyzed hydrogen halide elimi-
nation resulting in carbon-nitrogen double bond forma-
tion. The -I effect of chlorine (σ* ) 2.96) is slightly
greater than bromine (σ* ) 2.84).15 Therefore, we can
only expect the acidity of the hydrogen atom at the
R-carbon in N-bromoamino acid to be approximately
equal to (should be slightly less than) that of N-chloro-
amino acid. The actual bond energies of RCH(COO-)-
HNCl and RCH(COO-)HNBr or of the related compounds
H2NCl and H2NBr are not known. However, the calcu-
lated bond energies of NCl and NBr in diatomic28 and
Ack n ow led gm en t. D.E. acknowledges CSIR, New
Delhi, for the associateship. S.V.K. acknowledges the
authorities of Thiagarajar College of Engineering, Ma-
durai 15, for encouragement.
J O9506342
(28) CRC Handbook of Chemistry and Physics; Lide, D. R., Fred-
erikse, H. P. R., Eds.; CRC Press: Boca Raton, 1993; Section 9, p 123.
(29) CRC Handbook of Chemistry and Physics; Weast, R. C., Ed.;
CRC Press: Cleveland, OH, 1977; Section F, p 240.
(30) Engel, P.; Oplatka, A.; Perlmutter-Hayman, B. J . Am. Chem.
Soc. 1954, 76, 2010.
(25) Reference 16, p 893.
(26) Reference 15, Section 5, pp 17-41.
(31) Feigl, F. Spot Tests in Organic Analysis, 7th English ed.;
Elsevier: Amsterdam, 1966; pp 482-484.
(27) The rate constant for the base-catalyzed disintegration in
N-bromo-γ-aminobutyric acid is calculated as 3.4 × 10-3 M-1 s-1 at 35
°C from the oxidation by Br2 in aqueous alkaline medium.
(32) Reference 31, p 485.
(33) Wells, C. F. Tetrahedron 1966, 22, 2685.