In alkaline media, Fremy's salt oxidizes alkanols by a hydrogen atom transfer mechanism

Article abstract of DOI:10.1016/j.poly.2010.01.018

In aqueous alkali, Fremy's salt (potassium nitrosodisulfonate dimer), homolyses nearly exclusively to the monomer radical anion, nitrosodisulfonate (NDS). In this media, NDS almost quantitatively oxidizes benzyl alcohol (PhCH₂OH) to benzaldehyd

Full text of DOI:10.1016/j.poly.2010.01.018

Polyhedron 29 (2010) 1358–1362
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
In alkaline media, Fremy’s salt oxidizes alkanols by a hydrogen atom
transfer mechanism
Piyali De, Dhurjati Prasad Kumar, Amit Kumar Mondal, Pulak Chandra Mandal,
*
Subrata Mukhopadhyay, Rupendranath Banerjee
Department of Chemistry, Jadavpur University, Kolkata 700032, India
a r t i c l e i n f o
a b s t r a c t
Article history:
In aqueous alkali, Fremy’s salt (potassium nitrosodisulfonate dimer), homolyses nearly exclusively to the
monomer radical anion, nitrosodisulfonate (NDS). In this media, NDS almost quantitatively oxidizes ben-
Received 1 December 2009
Accepted 4 January 2010
Available online 18 January 2010
zyl alcohol (PhCH
2
OH) to benzaldehyde (PhCHO), itself being reduced to hydroxylamine disulfonate
ꢀ
(HNDS). The reaction is very nearly first-order in [NDS], [alkanol] and in [OH ]. However, with progres-
sive addition of HNDS, decay kinetics of NDS gradually deviates from first-order. Ultimately, with suffi-
cient excess of HNDS, the reaction becomes second-order in [NDS]. The consumption ratio, ( PhCH OH]/
[NDS]), is ꢁ2. PhCD OH manifests a large primary kinetic isotope effect (k /k = 11.6). Substituted ben-
zyl alcohols (RBzCH OH) with R-groups withdrawing electron density from the O–H bond accelerated the
reaction; those with R-groups donating electron density to the O–H bond retarded the reaction. The con-
version of 2-propanol to 2-propanone is much slower compared to that of benzyl alcohol to benzalde-
hyde. An alpha-H atom transfer mechanism seems logical.
Keywords:
Fremy’s salt
Kinetics
D
2
D
2
H
D
2
Benzyl alcohol
2
-Propanol
Nitroxyl radical
Hydroxylamine disulfonate
Ó 2010 Elsevier Ltd. All rights reserved.
1
. Introduction
Fremy’s salt (potassium nitrosodisulfonate dimer) is a useful
and fractionally distilled under reduced pressure [12]. 2-Propa-
nol (G.R, E. Merck) was purified by refluxing with CaO (200 g/
L) for several hours, followed by distillation. Acetone was
removed by treatment with and distillation from 2,4-dinitro-
phenylhydrazine. Peroxides re-form in 2-propanol if it is stood
for several days [13], hence only fresh stocks of purified 2-pro-
panol were used for the kinetic and stoichiometric studies. 2-,
source of the persistent nitroxyl radical anion, nitrosodisulfonate,
ꢀ
ONðSO Þ (NDS), a versatile reagent for organic conversions [1]
3
2
and a common reagent for determining the kinetics and mecha-
nisms of inorganic reactions [2–7]. The present work deals with
the mechanism for the near quantitatively base-catalyzed oxida-
tion of benzyl alcohol to benzaldehyde by NDS.
3- and 4-methoxybenzyl alcohol, 4-(trifluoromethyl)benzyl
0
alcohol (all 98%), benzyl-
a,a
-dideuterated alcohol (98 atom%
NDS is an aprotic oxidant. Prominent base-catalysis in its reac-
tions generally arises from the formation of the conjugate base of
the reducing agents, if they are sufficiently acidic, as with ascorbic
acid [8] and the hydrazinium ion [9]. But alkanols are too weak an
acid (pK > 15) [10,11] to form any significant amount of alkoxide to
account for the [H ] term seen in the rate. In this study, we have
traced the source of the overwhelming base-catalyzed path and
found the same to be associated with a hydrogen atom transfer
D) (Aldrich) and the other reagents used were of high purity
and were used without further purification. Solutions were
prepared from freshly boiled doubly distilled water, purged with
argon.
The potassium nitrosodisulfonate dimer (Fremy’s salt) in
aqueous solutions undergoes nearly complete homolysis to
produce the monomer, nitrosodisulfonate radical anion,
+
ꢀ1
Å
2ꢀ
NOðSO
3
Þ
, NDS [14]. Solid Fremy’s salt was prepared and stored
2
(
HAT) mechanism.
below 5 °C in an ammonia environment [15]. Its purity was
checked by determining its oxidizing power from time to time
by iodimetry and by comparing its molar absorbance at 545 nm
with the literature value (20.8) [16]. Fresh preparations were
thus found to be sufficiently pure. Nevertheless, on storage, the
2
. Experimental
2.1. Materials
solid slowly lost its
e value, and fresh samples were prepared
regularly. In the stoichiometric studies, we used preparations
with an e value of at least 19.5 M cm (at 545 nm). Potassium
Benzyl alcohol was purified by shaking it with a solution of
Fe . The alcohol layer was then washed with distilled water
ꢀ1
ꢀ1
II
hydroxylamine
disulfonate
(HNDS)
was
prepared
as
*
Corresponding author. Tel.: +91 9432095474.
E-mail address: rupenju@yahoo.com (R. Banerjee).
reported [17].
0
277-5387/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2010.01.018

P. De et al. / Polyhedron 29 (2010) 1358–1362
1359
2.2. Stoichiometry and reaction products
3.2. Kinetics
The stoichiometry of the reaction with excess NDS in sodium
In alkali (>0.10 M), NDS suffers negligible bleaching [19] within
12 h. It, however, reacts rather fast with alkanols in alkali and the
decay kinetics closely adhere to first-order in [NDS], [NaOH]
(Fig. 1) and in [alcohol] (Fig. 2 and Table 1).
hydroxide (0.40 M) under an argon atmosphere was determined
by measuring the residual [NDS] spectrophotometrically.
To determine the stoichiometry under the kinetic conditions,
the product mixture of NDS with excess PhCH
2
OH was extracted
The primary alkanol, benzyl alcohol, is oxidized at a rate much
faster than that for the secondary alkanol, 2-propanol. It might be
noted here that the initial absorbance noted for benzyl alcohol oxi-
dation is somewhat lower (Fig. 1a) than that for the slow reacting
with dichloromethane and the extract thus obtained was subjected
to NMR and GC studies. Acetone produced in the reactions with 2-
propanol was spectrophotometrically quantified [18].
2
-propanol (Fig. 1b). Nevertheless, the initial absorbances for the
2
.3. Physical measurements and kinetics
benzyl alcohol oxidation may be extrapolated back with known
rate constants and the time lapsed between mixing and recording
of the first point.
Kinetics and absorbances were determined with a Shimadzu
(
UV1601) spectrophotometer, NMR spectra with a BRUKER, DPX-
00 spectrometer and gas chromatograms were recorded with an
3
Agilent, 5890N gas chromatograph. Kinetics were monitored
in situ at 25.0 (±0.2) °C, following the decrease in absorbance of
NDS with time at 545 nm, where only NDS absorbs. First-order
conditions were maintained using a large excess of [alcohol] over
3.3. Product inhibition
Kinetic order in [NDS] shifted from first-order to second-order
with progressive addition of one of the reaction products, [HNDS].
[
1
NDS] in the NaOH media. The ionic strength was maintained at
.0 M with NaCl. The absorbances (A ) versus time (t) data were
t
analyzed using non-linear least-squares regression routines to ex-
tract the rate constants. Some reactions were conducted under sec-
ond-order conditions with respect to [NDS]. For such reactions the
a
0.20
0.15
ꢀ1
(A )
t
versus time (t) data were straight lines and were used to
extract the second-order rate constants. For example, reactions
with substituted benzyl alcohols, [RBzOH], were carried out in 4%
(
2
v/v) acetonitrile-H O media under second-order conditions,
[
NDS] = [RBzOH]. We employed such second-order conditions be-
cause substituted benzyl alcohols have very poor solubility in
water. For comparison, a few reactions of benzyl alcohol were also
carried out in this medium under comparable conditions.
0
0
0
.10
.05
.00
3
. Results and discussion
3.1. Stoichiometry and reaction products
With excess NDS, spectrophotometric determination of the
residual [NDS] indicated a 1:2 stoichiometry (D[alkanol]:D[NDS])
0
.0
1.0
2.0
3.0
4.0
under an argon atmosphere. A similar stoichiometry was also
noted under kinetic conditions with excess benzyl alcohol. For
such stoichiometric determinations, the ratio of peak areas in the
gas chromatogram occupied by the residual benzyl alcohol to that
by benzaldehyde present in the dichloromethane-extracted prod-
uct mixture was measured and found to be 17.7 (Fig. S1). Assuming
-
3
10 time, s
b 0.20
complete consumption of NDS and a consumption ratio,
D[benzyl
alcohol]: [NDS] of 1:2, one expects the ratio to be 19.0. GC exper-
iments thus established a near quantitative (95.3%) conversion of
benzyl alcohol to benzaldehyde.
D
0
0
0
0
.15
.10
.05
.00
The ratio of the integrated peak area for the benzylic hydroxyl
proton to that of the –CHO proton in the NMR spectrum of the
product was found to be 1.00/0.057 = 17.86 (Fig. S2). The ratio ex-
pected for a 1:2 stoichiometry is 19.00. The ratio (17.86) found
therefore indicates a 94% yield.
Passage of oxygen gas retarded the reaction and decreased the
yield of benzaldehyde. A plausible reason shall be discussed later.
The acetone produced in the reaction of excess 2-propanol with
NDS also indicated a 1:2 stoichiometry. Therefore, the reaction of
alkanols with NDS may be presented by Eqs. (1) and (2).
0
.0
2.0
4.0
6.0
8.0
10.0
Å
2ꢀ
2ꢀ
PhCH
2
OH þ 2ð ONðSO
3
Þ
Þ ! PhCHO þ 2HONðSO Þ
3
2
2
-
3
1
0
time, s
NDS
HNDS
ð1Þ
ð2Þ
Fig. 1. First-order decay in absorbance of NDS. [NDS] = 0.01 M; T = 25.0 °C;
I = 1.0 M. The solid line is the least-squares fit of absorbance – time data (black
points) to first-order decay. (a) [BzOH] = 0.06 M, [NaOH] = 0.40 M (b) [2-propa-
nol] = 0.08 M, [NaOH] = 1.0 M.
Å
2ꢀ
2ꢀ
CH
3
CHOHCH
3
þ 2ð ONðSO
3
Þ
Þ ! CH
3
COCH
3
þ 2HONðSO Þ
3
2
2
NDS
HNDS

1
360
P. De et al. / Polyhedron 29 (2010) 1358–1362
ꢀ
1
With sufficient excess of HNDS, the second-order plots of (A
versus t plots turned into good straight lines (Fig. S3), whereas A
t data gave an excellent fit to first-order exponential decay in
the absence of HNDS.
t
)
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
BzOH
t
ꢀ
Figure 2a
3.4. Primary deuterium kinetic isotope effect
0
Benzyl-
a
,a
-dideuterated alcohol (C
6 5 2
H CD OH) reduced NDS
/k = 11.6, (Table 1)) for the kinetic isotope
effect. A rate determining hydrogen atom abstraction is thus
indicated.
k
with a large value (k
H
D
2
-propanol
3.5. Reactivities of substituted benzyl alcohols
Substitution at different positions of the benzyl group by differ-
0
.0
0.2
0.4
0.6
0.8
1.0
ent groups changed the reaction rate significantly. Compared to the
unsubstituted benzyl alcohol, groups withdrawing electron density
from the O–H bond accelerated the rate (Table 2). On the other
hand, groups donating electron density to the O–H bond, and thus
strengthening the bond, retarded the reaction rate. These observa-
[
alcohol], M
2
2
1
1
0
0
.5
.0
.5
.0
.5
.0
tions support an
step.
a
-C–H bond breaking at the rate determining
BzOH
Figure 2b
ꢀ
3
.6. Origin of the [OH ] dependence on rate
k
Alkanols themselves are too weak an acid to undergo kinetically
significant deprotonation. However, formation of an N-bound pre-
cursor of alkanol with NDS may assist deprotonation of the alkanol.
Similar precursors of NDS with water and hydroperoxide have
been reported earlier [2,6,20]. When rapid equilibrium formation
of the N-bound precursor is incorporated (Eq. (7)) in the reaction
scheme (Scheme 1) framed with benzyl alcohol as the paradigm,
the base-catalysis is accounted for.
2
-propanol
0.6
0
.0
0.2
0.4
NaOH], M
0.8
1.0
[
.
-
A steady state assumption for the PhCHO radical leads to the
rate law Eq. (7).
Fig. 2. Effect of reagent variations: [NDS] = 0.01 M, I = 1.0 M, T = 25.0 °C. Experi-
mental first-order rate constants are shown in black squares, triangles or in circles
where the solid lines represent linear least-squares fit: (a) [alcohol] variation;
NaOH] = 0.40 M for [BzOH], 1.0 M for [2-propanol]. 10 slope = (22 ± 0.2) M
for benzyl alcohol and (0.6 ± 0.01) M
2
½NDSꢂg½Hþꢂ
Rate ¼ k
1
k
2
K½NDSꢂ ½BzOHꢂ=fkꢀ1½HNDSꢂ þ k
2
ð7Þ
3
ꢀ1 ꢀ1
[
s
ꢀ
1 ꢀ1
s
for [2-propanol]; (b) [NaOH] variation;
Under the condition, k [NDS] ꢃ kꢀ1[HNDS], Eq. (7) is reduced to
Eq. (8) and the reactions follow a first-order kinetics in [NDS] when
HNDS] is relatively small.
2
[
2-propanol] = 1.0 M, [BzOH] = 0.10 M.
[
Rate ¼ k
1
K½NDSꢂ½BzOHꢂ=½Hþꢂ
ð8Þ
Table 1
a
The experiments well verified this rate law (Fig. S3).
[NDS], the rate
Representative first-order rate constants for the oxidation of alcohols by NDS in
aqueous alkaline media. [NDS] = 0.010 M, T = 25.0 °C, I = 1.0 M.
Further, under the conditions kꢀ1[HNDS] ꢃ k
2
equation reduces to Eq. (9) and a second-order kinetics is observed
at a sufficiently high [HNDS].
103
k
o
(s
ꢀ1
)
[
NaOH] (M)
[
BzOH] (M)
0
0
0
0
0
0
0
0
.10
.10
.10
.10
.10
.04
.06
.08
0.10
0.20
0.30
0.40
0.50
0.40
0.40
0.40
0.5
1.0
1.4
2.2
2.5
0.9
1.3
2.0
b
Table 2
a
0
Second-order rate constants (k ) for the oxidation of substituted benzyl alcohols.
[NDS] = 0.01 M, [alcohol] = 0.01 (M), [NaOH] = 1.0 M, T = 25.0 °C, I = 1.0 M, 4% (v/v)
CH CN–H O media.
3
2
2
0
ꢀ1 ꢀ1
Substitution
k
(M
s )
2
[
2-Propanol] (M)
4-(Trifluoro methyl)
3-Methoxy
Benzyl alcohol
4-Methoxy
2-Methoxy
0.26b
0.12b
0.03
0.02
0.01
1.0
1.0
1.0
1.0
1.0
0.6
0.8
0.9
1.1
0.50
0.60
0.70
0.80
1.0
1.0
1.0
1.0
1.0
0.3
0.4
0.5
0.5
0.6
0.4
0.5
0.6
0.7
c
d
d
a
Average of at least four parallel experiments. Individual runs reproducible
within 2–4%. The regression coefficient is higher than 0.99 for individual absor-
bance (A ) versus time (t) least-squares fit to the first-order equation.
t
b
Groups withdrawing electron density from the O–H bond accelerate the rate in
comparison to benzyl alcohol.
a
Average of at least four parallel experiments. Individual runs reproducible
within 2–4%. The regression coefficient is higher than 0.99 for an individual least-
squares fit to the first-order equation.
c
Experiments conducted under the same conditions as for others.
Groups donating electron density towards the O–H bond, and thus strength-
d
b
3
ꢀ1
1
0
k
0
(s ) = 0.19 for C
6
H
5
CD
2
OH.
ening the bond, accelerate the rate in comparison to benzyl alcohol.

P. De et al. / Polyhedron 29 (2010) 1358–1362
1361
2
-
CH2OH
3
-
SO3
.
O3S
O
SO3
+
K
+
.
+
H
O
N
(3)
(4)
N
SO3
CH2
O
[
NDS]
[BzOH]
.
-
2-
COH
3
-
SO3
SO3
O S
SO3
k1
3
O
.
OH
N
N
+
CH2
k-1
O
[
HNDS]
.
2
-
-
3-
CHO
COH
SO3
SO3
.
k2
+
-
O
N
+
(5)
(6)
O
N
SO3
SO3
[
NDS]
3
-
2-
SO3
SO3
SO3
SO3
fast
-
-
O
N
+ H2O
HO
N
+
OH
[
HNDS]
Scheme 1. Proposed mechanism for the oxidation of benzyl alcohol by NDS.
K½NDSꢂ ½BzOHꢂ=fkꢀ1½HNDSꢂg½Hþꢂ
2
ð9Þ
Rate ¼ k
1
k
2
Further, the standard potential for the conversion HNDS ? NDS
is high (E versus NHE in water is 1.36 V) [21]. This may be an addi-
tional barrier to the operation of an NDS/HNDS catalytic cycle.
In the presence of dissolved oxygen the reaction course changes
Fig. 3). So, we determined the kinetics under an argon atmosphere
o
Fig. S3 illustrates how progressive addition of HNDS changes
the kinetics from first-order to second-order in [NDS] and (A )
versus time plots exhibit good linearity in the presence of 0.15 M
HNDS. Moreover, Fig. S3 also includes an excellent non-linear fit
ꢀ1
t
(
to minimize the effect of dissolved oxygen on the reaction rate.
2
(
r = 0.999) of absorption versus time data confirming the second-
order decay of [NDS] at high [HNDS].
Acknowledgements
3.7. Effect of oxygen
This work was carried out under financial assistance from DST
(
(
Sanction No. SR/S1/IC-40/2007) and CSIR (Sanction No. 01
2085)/06/EMR-II). The award of a Junior Research Fellowship to
NDS does not catalyze the auto-oxidation of benzyl alcohol. Dis-
solved oxygen prolongs reaction time and lowers the yield of benz-
DPK (by DST) and to PCM (by CSIR) are also thankfully
acknowledged.
ꢀ
aldehyde. Possibly, dissolved oxygen consumes PhCOHÅ and thus
the reaction goes astray.
Appendix A. Supplementary data
0
0
0
0
0
0
.20
.16
.12
.08
.04
.00
A without Ar
B with Ar
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.poly.2010.01.018.
References
A
[1] H. Zimmer, D.C. Lankin, S.W. Horgan, Chem. Rev. 71 (1971) 229.
[
2] E.S. Gould, Acc. Chem. Res. 18 (1985) 22.
[3] S.K. Ghosh, E.S. Gould, Inorg. Chem. 28 (1989) 3651.
[4] S.K. Saha, M.C. Ghosh, E.S. Gould, Inorg. Chem. 31 (1992) 5439.
B
[
5] S.P. Ghosh, M.C. Ghosh, E.S. Gould, Int. J. Chem. Kinet. 26 (1994) 665.
[6] P.N. Balasubramanian, E.S. Gould, Inorg. Chem. 22 (1983) 1100.
[
[
[
7] Z. Yang, E.S. Gould, J. Chem. Soc., Dalton Trans. (2002) 3840.
8] Z. Yang, E.S. Gould, Res. Chem. Intermed. 29 (2003) 181.
9] G. Gualtieri, S. Colacicchi, C. Corvaja, J. Chem. Soc., Perkin Trans. 2 (2002) 1917.
[
[
[
[
10] F.M. Menger, M. Ladika, J. Am. Chem. Soc. 109 (1987) 3145.
11] See also G.O. Andrés, R.H. de Rossi, ARKIVOC X (2003) 127.
12] J.H. Mathews, J. Am. Chem. Soc. 48 (1926) 562.
0
2000
4000
6000
Time, s
8000 10000 12000
13] L.E. Gilson, J. Am. Chem. Soc. 54 (1932) 1445.
Fig. 3. Effect of dissolved oxygen on the kinetics: [2-propanol] = 1.0 M,
NDS] = 0.01 M, [NaOH] = 0.50 M, T = 25.0 °C, I = 1.0 M, (A): ambient atmosphere,
B): argon-saturated media.
[14] (a) R.W. Z Asmussen, Anorg. Allg. Chem. 212 (1933) 317;
(b) B.J. Wilson, J.M. Hayes, J.A. Durbin, Inorg. Chem. 15 (1976) 1703.
[15] S. Yamada, R. Tsuchida, Bull. Chem. Soc. Jpn. 32 (1959) 721.
[
(

1
362
P. De et al. / Polyhedron 29 (2010) 1358–1362
[
[
[
16] J.H. Murib, D.M. Ritter, J. Am. Chem. Soc. 74 (1952) 3394.
[19] B.J. Wilson, D.L. Fillmore, Inorg. Chem. 16 (1977) 1404.
[20] B.J. Wilson, D.M. Ritter, Inorg. Chem. 2 (1963) 974.
[21] V.F. Toropova, V.N. Degtyareva, Izv. Vyssh. Uchebn. Zaved. 17 (1974) 175
(Chem. Abs. 80 (1974) 152439q).
17] G.K. Rollefsoann, C.F. Oldershaw, J. Am. Chem. Soc. 54 (1932) 977.
18] Vogel’s Textbook of Quantitative Chemical Analysis, fifth Ed., Longmans
Scientific and Technical, New York, 1989, p. 706.