-
f. Indian Journal of ChemistryVol. 27A, November 1988, pp.
963-967
Kinetics of Oxidation of Lactic, Mandelic & Benzilic Acids
byTrichloroisocyanuric Acid in Aqueous Acetic Acid Media
P S RADHAKRISHNAMURTI*. NABEEN KUMAR RATH & RAM KRISHNA
PANDA
Chemistry Department, Berhampur University, Berhampur 760007
Received 15 September 1987; revised 21 December 1987; accepted
15 February 1988
Oxidation of lactic, mandelic and benzilic acids by
trichloroisocyanuric acid (TCICA) in aqueous acetic acid-acid
me-dia is pseudo-first order in [TC1CA]"both in the absence and
presence of added Cl- .The rate constants show a linear de-pendence
on [substrate]. Thereactiori remains unaffected at lowerjl-l +
](0.1 x 10-2-2 x 10- 2 mol dm -J); but increases lin-early at
higher [H +] ( lOx 10- 2-40 x 10 - 2mol dm - J). Increase in
dielectric constant of the medium has a negligible effecton the
reaction rate. Although kOh' consistently increases with increase
in [CI- lad' the magnitude of increase is rather small.The reaction
most probably proceeds through a concerted oxidative
decarboxylation mechanism leading to a carbonyl de-rivative and CO,
as the primary oxidation products.
Numerous reports are available on kinetics of oxi-dation of
a-hydroxy carboxylic acids (HA) by a varie-ty of oxidantsl". While
oxidation ofHA by one-elec-tron oxidants generally proceeds through
one of theseveral possible initial radical formation steps:',
oxi-dation by two-electron" oxidants usually proceedseither by a
ionic mechanism involving transfer of a-hydrogen or by a concerted
oxidative decarboxyla-tion involving cleavage of CI - C2 bond.
While a group of authors>"? suggested a hydrideion mechanism
leading to a-keto acid as the productof oxidation, parallel reports
I.H - II suggested alde-hyde or ketone as the product. The choice
of mecha-nistic route, envisaged to be operative, would thus
de-pend to a great deal on several factors, such as structu-ral
parameters of the substrates (HA), the nature andthe oxidising
capacity of the oxidant, the reaction con-ditions employed and the
actual products isolated.For quite sometime we have been interested
in thestudy of oxidation kinetics of organic substrates
bytrichloroisocyanuric acid (TCICA). The title investi-gation forms
a part of this broad programme.
Materials and MethodsThe experimental procedure was briefly
described
earlier'{. The substrates, viz lactic acid (LA), mandel-ic acid
(MA) and benzilic acid (BA) were of GR gradeand were recrystallised
or redistilled before use.Trichloroisocyanuric acid (TCICA) was of
Flukagrade. The kinetics were monitored by estimating
thedisappearance of TCICA iodometrically at regulartime
intervals.
Stoichiometry and product analysisStoichiometric runs with
[TCICAl > [HA]o at
[HCl04]=OA rnol dm " ', [C!-]=O.5 mol dm ":' in
30% acetic acid medium, did not give meaningful re-sults as the
self-decomposition of TCICA in thesecases was appreciably large.
However in the oxida-tion of mandelic acid and benzilic acid,
benzaldehydeand benzophenone were the oxidation products,
re-spectively. The product oflactic acid oxidation is
ace-taldehyde.
ResultsThe disappearance of TCICA followed a pseudo-
first order kinetics for more than four half-lives, asseen from
the perfect linearity of log [TCICAl, versustime plots. The
pseudo-first order rate constants( k"hs) showed a linear dependence
on [HAlo in therange studied. There was no perceptible change
ink"hs at lower [H+] range (0.1 x 10-2 to 2.0X 10-2mol dm - 3)
while change in kohs was linear at higher[H+] (10 x 10-2 to 40 x
10-2 mol dm-3).
Kate law and mechanismThe important steps connected with the
various
reactant species in the present reaction are shown inScheme
1.
I(ITCICA+ H20
-
INDIAN J. CHEM., VOL. 27A, NOVEMBER 1988
H'A- + CI" y;:-- Products
H+HA+ HOCI k" Products
H+HA + Cl2 y;:-- Products
Schemel
If as a first approximation the reactivity of the TCI-CA species
is neglected in comparison to those of theother reacting species,
then the rate law (1) could bederived for the reaction in the
absence of added CI- .
Assuming [DCICAj in the denominator as negligi-ble compared to
the other terms, Eq. (1)would reduceto Eq. (2).
, r. +] r. ~]2 f j2 31~2'!S..QLH +1!.21LH +k3X~Q~4fr +l ~ [H+])
[HA]!obs'" 3. , ..·(2)
{H.jI1+~Q)(1+iS4 [H+}
where the concentration terms refer to those taken in-itially
and the k~x and ~x are termolecular reactionconstants involving one
unit each of [HA], [oxidant]and [H+].
Using the reportedl ' values of K.a at 35°C for var-ious
substrates and data in Tables 1-3 various con-stants could be
obtained the average values of whichare presented in Table 4. While
~x and k:Jx values areunambiguous, ~x and k;x values are the
estimates; thelatter were obtained from the mixed terms assumingthe
contribution of each of the two terms to be ap-proximately
equal.
In the presence of added CI- Eq. (3) can be used tocompute the
pertinent constants, the average valuesof which are also listed in
Table 4.
The average values of K4 and K4 K; obtained hereare in agreement
with those obtained earlier; e.g.K4 = 73 dm3mo\-1 (in 20% aq HOAc
at 35°C);
Table 1- Pseudo-first-order Rate Constants for Oxida-tion of
Lactic Acid (LA), Mandelic Acid (MA) andBenzilic Acid (BA) by TCICA
in Aqueous Acetic Acid-
Acid Media.
[TCICA],,= 5 x 10-.1 mol dm ',[H+]= I X 10-2 mol dm-3(ai,HOAc=
30%h
102[MA]() 10.1kOh'(s - I )(mol dm-3)
LA' MA" BA"
0.5 0.925 2.109 0.8071.0 1.806c 4.411d 1.618d
O.243d 5.442< 1.914<0.4221 7.6371 3.0391
1.5 2.7312.0 8.807 3.242.5 4.5144.0 9.03 18.18 6.3256.0 18.12
26.33 9.506
(~H~/kJmol-'l" 69 65 65( -l'1S ••/J K - I mol- I)g 27 25 27
(a) HCl04 was used;(b) acetic acid-water:
30%-70%(v/v);(c,d)at55· and 35·C respectively; (e, f) at 40· and
45·C respectively; and(g) values of the net activation
parameters.
Table 2-Pseudo-first Order Rate Constants for Oxida-tion of
Benzilic Acid by TCICA in Aqueous Acetic Acid-
Acid Media.[TCICA]u= 5 x 10-4 mol dm-3, HOAc = 30%b,
[benzilic
acid]" = 1 x 10 - 2 mol dm - 3; temp. = 35·C
102[Cl-l.d 104 kob,(S - I )(rnol dm=']
102[H +Jmol dm - 3(3) 4 10 40=1
Nil 1.618 1.806 2.506 11.31 1.703 2.025 3.037 13.085 1.834 2.431
3.718 15.11
10 1.927 2.604 4.108 17.34
(a) HCl04 was used; and (b) acetic acid-water: 30%-70% (v/v)
Table 3-Pseudo-first Order Rate Constants for Oxida-tion of LA,
MA and BA by TCICA in Aqueous Acetic
Acid-Acid Media[TCICAlo= 5 x 10-4 mol dm-3; [HA]u= Ix 10-2 mol
dm-3,
[H+] = 1 X 10-2 mol dm ":'; temp = 35·C
(S) 104 kobs (s - I) in medium containingHOAc(%)
15 30 45
1.8451.7284.614
60
1.7321.5273.907
Lactic acid" (LA) 1.309 1.806Benzilic Acid" (BA) 1.521
1.618Mandelic acid" (MA) 4.036 4.411
(a, b) At 55· and 35·C respectively.
-
RADHAKRISHNAMURTI et al. : KINETICS OF OXIDATION OF LACTIC,
MANDELIC & BENZILIC ACIDS
Table 4-Average Values of Resolved Constants' Asso-ciated with
Oxidation of LA, BA and MA by TCICA in
Aqueous Acetic Acid-AcidMedia.[HOAc=30%j
Constant Lactic Mandelic Benzilicacid" acid* acid"(LA) (MA)
(BA)
K4 drrr'mol ' I b 73 73 73K4KS dm" mol- 2 b 1.8 X 10J 1.8 x 1(j3
1.8 X to3K. mol dm-3 b 10-.10 10-34 to-3.11(at 35°C)k;x dm'tmol :
2S- I 1.8 X lO2 1.5 X 102 0.24 X 102k2x dm-rnol : 2S- I 0.4 1.0
0.3k;x dm-mol : 2S- I 40 30 2.7Isx dm"mol- 2S- I 0.28 0.52 0.27k~x
dm'tmol : 2S- I 2.0k4' dmsrnol : 2S- I 0.2
(a) cfEqs (2) and (3): (b) temperature dependence in the range
35°-55°C is assumed to be negligible.
K4KS = 1.8 X 103 dm-mol" 2 (in 20% aq HOAc at35°C);K4Ks = 1.4 x
103dm6mol-2(in 15%aqHOAcat 35°C). The average value of 1I (K4 K,
)in the presentstudy works outto be 5.5 x 10-4 mol-dm -6)in
aque-ous medium for the equilibrium: Cl2 + H20 4't H ++CI- +
HOCl.
The pseudo-first order rate constants calculatedusing the
various constants of Table 4 are in goodagreement with the
experimental kobs'
Inspite of the apparent complexity of the ratelawis), some
reasonably simple inferences on thereactivity pattern emerge from
the study. A glance atTable 4 suggests that for all the reaction
steps the reac-tivities of the substrates follow the orders:
mandelicacid> lactic acid and mandelic acid> benzilic
acid.For all the reducing substrates studied, k~x > kux
im-plying that anion A-reacts significantly faster thanthe
undissociated hydroxy acid (HA). Moreover, thereactivity orders:
k;x> k;x ~ «: and ~x ~ ~x ~ k4xsuggest that the relative
oxidising capacity of the oxi-dant species decreases in the
order:HOCI > HzOCI + ~ Cl., It may be recalled that
whenchlorination of substrates is considered the
relativechlorinating ability of these species follows a
reverseorder; Cl, > HzOCI + ~ HOG, in aqueous acetic ac-id-acid
media. The reported redox potentials (EO), in-volving two-electron
processes for the couplesHOClICl- andCI?12Cl- are - 1.49Vand -'
1.36 V,respectively I 2. Sin-ce H20Cl + + CI- reaction givesCl,
(e.g. K; ~ 25 dm ' mol- I), the redox potential forthe couple H2OCl
+ ICl- would not be much differentfrom that of Cl/2Cl- (viz. -l.36
V). Thus with anestimated value of EO = - 1.38 V for the
coupleH20CI + ICI .: the reactivity pattern k;x> k;x ~ k~x
agrees in magnitude as well as in reaction order withthe
relation, ~ log kobs = ~E 0 + constant, for the reac-tion ofthe A"
with HOCI,H20Cl + and C12• However,this agreement is not so
strictly satisfied for the reacti-vities of HA species with HOCI,
HzOCI + and Cl2which are approximately the same (since~x ~ k3x ~
k4x)· The latter observation leads to the as-sumption that in the
medium of [H + 1where HA is thepredominant species, viz. HzOCl +
and Cl, (in thepresence of added CI-) are also predominant
andprobably contribute as significantly as is possible by aless
abundant (but more reactive) HOCI species.
DiscussionThe present reactions do not appear to involve a
radical mechanism (lack of polymerisation of acry-lonitrile).
Moreover, in predominantly aqueous acidconditions employed, all the
oxidant species containpositive chlorine and under this condition
irreversi-ble transfer of two electrons from the substrate will bea
facile and thermodynamically feasible process.
A two-electron transfer might occur through thetwo pathways
shown in Scheme 2 (1) A rate-deter-mining formation of a hypo
halite ester to give the pro-duct (formed in two pathways) 14.
Scheme 2
(2) A concerted mechanism leading to carbonylderivative as
products. It appears to us that the con-certed mechanism is
possibly the best pathway forthese TCICA oxidations.
Benzilic acid (BA) with no a-hydrogen shouldreact much slower
than those HAs containing a-hydrogen, since the energy required for
the C1 - C2bond cleavage would obviously be relatively higherthan
that required for the C2 - H transfer.
965
-
INDIAN 1. CHEM .. VOL. 27 A, NOVEMBER 1988
Table S-Comparison of Rates of Oxidation of a-Hydroxy Acids with
Those of Other Re1avant Com-
pounds
[TCICA]" = 5 x 10-· mol dm-J, [S]"= 1 x 10-2 mol d~-J,[H 'I= 1 x
10-2 mol dm-3, HOAc= 30%, temp. = 35 C
Substrates 10· kn~' Substrates 10· k"b,(s -\) (s - \)
Lactic acid 0.243 Benzaldehyde 0.22(0.4)"Pyruvic acid 14.03
Benzyl alcoholAcetaldehyde Fast Benzilic acid 1.618Ethyl alcohol
1.105 Benzhydrol 1.047Mandelic acid 4.411 Benzophenone
NilPhenylglyoxalic > 200acid
(a) HClO. was used
Evidences in favour of a rate-determining
oxidativedecarboxylation involving heterolytic cleavage ofC1- C2
bond . .
A rate-determining oxidative decarboxylation in-volving
heterolytic cleavage of C 1 - C2 bond is sup-ported on the basis of
the following arguments. .
(a) The predominant product of LA or MA oXId~-tion isolated in
this work is not an a-keto carboxyhcacid, but a carboxyl derivative
containing one carbonatom less than the number of carbons in the
substrateHA, this carbon being lost as CO2 (e.g.acetic acid
pro-duced rapidly from acetaldehyde in the reaction ofLA;
benzaldeyde, and probably some benzoic acid byslow oxidation of
benzaldehyde, in the reaction ofMA; benzophenone in the reaction of
BA: Table.5).We believe that even in the oxidations by oxidants
hkeCAT, CAB, NCS etc. (which effectively pro~uceHOCl H OCI + and/or
CI2 by the acid hydrolysis ofthe re~ge~t), the predominant products
of oxida~ionof a-hydroxy acids are those formed through OXIda-tive
decarboxylation, but not the keto acids as report-ed earlier.
(b) A pointer to similarity in mechanism of oxida-tion ofBA, LA
and MA is that the same kinetic patternand the same range of net
activation parameters havebeen observed.
(c) The data in Table 5 show that keto acids likepyruvic acid
and phenylglyoxalic acid react ~uch ~as-ter than LA and MA,
respectively, under identicalconditions; and hence keto acids can
not be thought ofas predominant products of oxidation of
a-hydroxyacids. The magnitudes of the measurably faster ratesof
oxidation of pyruvic acid and phenylglyoxalic aciddo not permit the
assumption that keto acids are pr~-duced in the initial step. If
this were true then the OXI-dation would have a complicated
kinetics and notsmooth one as observed presently.
966
The discussion on the exact site of electron transferand the
nature of the transition state would rest onwhether the
decarboxylation is slightly more adv-anced than oxidation (or
vice-versa). A clear descrip-tion of this is difficult on the basis
of kinetic data alone;nevertheless, a reasonable mechanistic
pathway canbe suggested, which is shown in Scheme 3. We envi-sage a
concerted oxidative decarboxylation processinvolving most probably
a cyclic transition state of thetype shown in Scheme 3 in which
there is simultane-ous development of partial charges at C~- and
Cf+before the actual transfer of electrons.
H++ - + 2+
H-O 0 II ["'OCL"- ~"_O~' ~+ 1 1 l!.3x h \:.-H20C1 + 1'1- C- c••O
-- R - t:.....c: °
+ - ,H20 +3H +CI +R-CO-P.+C02 (4)
I R:CH3' R;Hin LA;R: C6HS,k.HinMA;R :R~C6H5 in BA)
Scheme 3
That the C2 - OH group in the HA substrate is in-volved in the
product formation suggests that thetransition state should have a
considerable productcharacter and that it is the C2 - OH group (or
ratherthe OH bond at C-2) which is effectively involved inthe
two-electron transfer process. This appears to becompatible with
the observed reactivity patterns: (i)HOCI> H20Cl + ~ Cl2 (ii) A-
> HA, (iii) MA> LA;(iv) BA < MA. The transition state
depicted inScheme 3 also explains the need for extra H + in
thereaction even though HOCI and A-are the most ef-fective reacting
species. We have not been able to getthe activation parameters for
the various reactionsteps. However, the net !J.H+ values, although
notvery high for a termolecular transition state are not solow to
rule out the concerted process!", The higherreactivity of a-hydroxy
acid than that of the corre-sponding primary alcohol (Table 5 lis
also compatiblewith this transition state as the hydroxy acid
containsan electron withdrawing (- COOH) group in theplace of an H
in the simple primary alcohol.
AcknowledgementOne of the authors (NKP) is grateful to the
authorit-
ies of Berhampur University and Khallikote Collegefor permission
and facilities to carry out the work.
References1 Banerji K K,lndiunJ Chem, 15A (1977)675; 17A(
1979)90.2 Comprehensive chemical kinetics, Vol. 7 edited by C H
Bam-
ford & C F H Tipper {Elsevier, London) 1972.3 Waters W A
& Kemp, J chem Sac, (1964) 339, 1192,3\01.
3193.4 Wiberg K B. Oxidation in organic chemistry, Part A (1965)
p.
65.282.
-
RADHAKRISHNAMURTI et al. : KINETICS OF OXIDATION OF LACTIC,
MANDELIC & BENZILIC ACIDS
5 Mushran S P,J chern Soc(8)(1971 ) 1712,JhaSailendra,Shar-maP D
& Gupta Y K, InorgChern, 22(9)(1983) 1393.
6 Jain A L & Banerji K K,] Indian chern Soc, 59 (1982) 654.7
Mathur AK & Banerji K K, Indian] Chern, 208(1981) 529.8
Venketasubramanian N, Indian J Chern, 22A(1983) 292.9 Barakat M Z
& Abdul Wahab M F,] Am chern Soc, 75 (1953)
5731. Venkatasubramanian N & Natarajan R, Indian JChern, 13
(1975) 261.
10 PinkJ M & Stewart R, Can] Chern, 49 (1971) 649, 654,
Bark-er & Aukett, J chern Soc, Perkin /I, (1973) 965.
11 Beebe T R, Adkins R L, Ba1cher A I, Choy T, Fuller A E,
Mor-gan V L, Senchery B B, Russell L J & Yates S W, ] orgChern,
47 (1982) 3006.
12 Radhakrishnamurti P S, Rath N K & Panda R K,] chern
SocPerkin /I, (1987) 517.
13 Hand book of analytical chemistry, edited by L Meites(McGraw
Hill, London) 1963; Sec. I, p. 8, 20.
14 Bishnoi M L, Negi S C & Banerji K K, Indian] Chern,
2SA(1986)660.
967