Chemical and electrochemical oxidation and reduction of dithizone

Article abstract of DOI:10.1016/j.poly.2010.02.009

A non-aqueous electrochemical study of dithizone, H₂Dz, 1, is compared with the chemical oxidation and reduction profile of this versatile ligand. Chemical oxidation of 1 by I₂ initially leads to an isolatable disulfide-bridged species, (HDz)₂, 2₂, but ultimately monomeric dehydrodithizone, Dz, 3, is formed. Electrochemically, in CH₂Cl₂/0.1 mol dm^-3 [N(ⁿBu)₄][B(C₆F₅)₄], two oxidation processes are observed for 1. Evidence of the electrochemical formation of the dimer 2₂ was found, but on a CV timescale the fully oxidized species, 2_{2 oxidized}, did not convert to the chemically stable species 3. Regeneration of 1 during an irreversible electrochemical reduction of the electrochemically generated fully oxidized species, 2_{2 oxidized}, was detected. Two further one-electron electrochemical irreversible reduction steps were also identified to ultimately generate H₃Dz^-, 8, one of the synthetic precursors to 1. In contrast, resolution and identification of the electron transfer steps of 1 in both dimethylsulfoxide, DMSO, or in CH₂Cl₂/0.1 mol dm^-3 [N(ⁿBu)₄][PF₆] were hampered by solvation and ion paring of [PF₆]^- especially with the oxidized species of 1. A metathesis of water-soluble potassium dithizonate, KHDz, 4b, led to lipophilic [N(ⁿBu)₄][HDz], 4c.

Full text of DOI:10.1016/j.poly.2010.02.009

Polyhedron 29 (2010) 1727–1733
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Chemical and electrochemical oxidation and reduction of dithizone
*
Karel G. von Eschwege, Jannie C. Swarts
Department of Chemistry, University of the Free State, P.O. Box 339, Bloemfontein 9300, South Africa
a r t i c l e i n f o
a b s t r a c t
Article history:
A non-aqueous electrochemical study of dithizone, H2Dz, 1, is compared with the chemical oxidation and
reduction profile of this versatile ligand. Chemical oxidation of 1 by I2 initially leads to an isolatable
disulfide-bridged species, (HDz)2, 22, but ultimately monomeric dehydrodithizone, Dz, 3, is formed.
Electrochemically, in CH2Cl2/0.1 mol dmꢀ3 [N(nBu)4][B(C6F5)4], two oxidation processes are observed
for 1. Evidence of the electrochemical formation of the dimer 22 was found, but on a CV timescale the
fully oxidized species, 22 oxidized, did not convert to the chemically stable species 3. Regeneration of 1 dur-
ing an irreversible electrochemical reduction of the electrochemically generated fully oxidized species,
22 oxidized, was detected. Two further one-electron electrochemical irreversible reduction steps were also
identified to ultimately generate H3Dzꢀ, 8, one of the synthetic precursors to 1. In contrast, resolution and
identification of the electron transfer steps of 1 in both dimethylsulfoxide, DMSO, or in CH2Cl2/
0.1 mol dmꢀ3 [N(nBu)4][PF6] were hampered by solvation and ion paring of [PF6]ꢀ especially with the
oxidized species of 1. A metathesis of water-soluble potassium dithizonate, KHDz, 4b, led to lipophilic
[N(nBu)4][HDz], 4c.
Received 18 December 2009
Accepted 9 February 2010
Available online 13 February 2010
Keywords:
Dithizone
Cyclic voltammetry
Disulfide
Dehydrodithizone
Tetrabutylammonium dithizonate
Ó 2010 Elsevier Ltd. All rights reserved.
PhNHANHAðC@SÞANHANHPh
1. Introduction
H4Dz
ꢀOH; ꢀ2eꢀ
PhAN@NAðCASꢀÞ@NANHPh
In 1925, Helmuth Fischer illustrated the vast potential of dithiz-
one, H2Dz, 1, for the detection and determination of heavy metal
ions [1]. Through the years 1 has become more or less indispens-
able in trace metal analysis. Recently, spectrophotometric detec-
tion limits of down to 12 parts per billion for the analysis of lead
in plant materials were achieved by utilizing 1 as a complexing
agent [2]. Complex 1 exists in solution as well as in the solid state
in the highly delocalized (conjugated) thione form [3] (Scheme 1).
A convenient synthesis of 1 from aniline was reported by Pelkis
et al. [4]. In this synthesis, diazotised aniline is first coupled
with nitromethane to give a formazyl, PhN@NAC(NO2)@NANHPh.
This formazyl is reduced with ammonium hydrosulfide to dihydro-
formazylmercaptan, H4Dz = PhNHANHA(C@S)ANHANHPh, the
structure of which was solved [5]. A two-electron chemical oxida-
tion in basic media of H4Dz liberates the deprotonated dithizonato
anion, HDzꢀ, 4, Eq. (1), and 1 is recovered after acidification. The
structures of 1 [6] and 4 [7] as their potassium salt have been
solved. The deprotonated form of H4Dz, H3Dzꢀ, 8, was also ob-
tained electrochemically in this study after two one-electron
reduction steps of 1.
ƒƒƒƒƒ!
4; HDzꢀ
Hþ
! PhN@NACðSHÞ@NANHPh
ð1Þ
1; H2Dz
In its interaction with iodine, 1 may be partially oxidized and the
disulfide, (HDz)2, 22 isolated, while with K3Fe(CN)6, the completely
oxidized compound dehydrodithizone, Dz, 3 (Scheme 1) forms. Ir-
ving synthesized, isolated and characterized 22 [8–10], while the
single crystal X-ray structure of the ortho-fluorophenyl derivative
of 22 was recently solved by one of us [11]. Irving clearly demon-
strated that 22 disproportionates into 3 and 1 in solution, the rate
of which is highly dependent on solvent polarity [10]. At 25 °C,
the half life of the disulfide increases from 10.5 min in acetone to
77 days in n-hexane. The structure of 3 was also solved by means
of X-ray crystallography [12]. Ogilvie and Corwin showed that 3
can be chemically reduced back to 1 [13]. The ability of dithizone
to act as a weak reducing agent has been described by Zhang et
al. [14].
The aqueous electrochemistry of 1 was investigated in the sev-
enties [15]. Two weak oxidation processes and two strong reduc-
tion processes were identified utilizing a hanging mercury drop
electrode in basic aqueous solution and the mechanisms of electro-
chemical and aerial oxidation of dithizone were considered. We
* Corresponding author. Tel.: +27 (0)51 4012781; fax: +27 (0)51 4446384.
E-mail address: swartsjc.sci@ufs.ac.za (J.C. Swarts).
0277-5387/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2010.02.009

1728
K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
H
N
S
H
I2
C
N
2
N
N
-2H+
N
N
N
N
N
N
N
N
H
C
S
S
C
H
1
1. Dextrose,
1. Na or KOH,
2. [N(nBu)4]Br
(for 4c)
EtOH & OH-,
H+
2. H+
22
K3[Fe(CN)6]
disproportionation
into 1 and 3
(giving 3 only)
S
M
S
H
N
C
N
N
C
2
N
N
K3[Fe(CN)6]
H
N
S
H
N
N
N
N
+
C
(giving 3 only)
N
N
4a: M+ = Na+
4b: M+ = K+
1
3
4c: M+ = [N(nBu)4]+
Scheme 1. Complete oxidation of dithizone, 1, to dehydrodithizone, 3, is best achieved with K3[Fe(CN)6]. Iodine oxidation generates 22 which disproportionates into 1 and 3
as described by Irving and co-worker [10]. The monomeric radical precursor to the dimer 22 i.e. 2, is shown in the electrochemical Scheme 2. Reaction with NaOH, KOH or
sodium wire ionizes dithizone to the water-soluble dithizonate salts, 4a and 4b, which may be converted into the N(nBu)4 salt soluble in non-polar solvents, 4c.
have recently embarked on studies of mercury [16], tin and cobalt
dithizone complexes. During the course of this research it became
necessary to investigate the non-aqueous electrochemistry of 1.
Towards this goal, the supporting electrolyte [N(nBu)4][B(C6F5)4]
in CH2Cl2 [17] was employed with beneficial effect. Acetonitrile,
THF and DMSO are known to interact with positively charged oxi-
dized species [18,19]. Traditional supporting electrolytes such as
[N(nBu)4][PF6] are also known to interact with positively charged
oxidized species via ion-pair formation. In contrast, the solvent/
electrolyte system CH2Cl2/[N(nBu)4][B(C6F5)4], first developed by
Geiger and co-workers [17], is known not to interact with posi-
tively charged oxidized species [20].
In this study, the full redox cycle of dithizone in CH2Cl2/
0.1 mol dmꢀ3 [N(nBu)4][B(C6F5)4] is for the first time highlighted
and related to the chemical oxidation profile of this versatile li-
gand. Comparisons of the electrochemical results obtained in
DMSO and CH2Cl2/0.1 mol dmꢀ3 [N(nBu)4][PF6] are made. The syn-
thesis of tetrabutylammonium dithizonate, 4c, an ionic salt of 1,
soluble in common organic solvents, is also described. Since potas-
sium dithizonate, 4b, the other easily available ionic salt of 1, is not
soluble in common organic solvents, the availability of 4c will
simplify experimental procedures for complexation reactions with
a variety of transition metals.
[N(nBu)4][B(C6F5)4] was synthesized according to published proce-
dures [17]. Melting points were determined on a Reichert Thermo-
pan microscope with a Koffler hot-stage, and were uncorrected.
2.2. Syntheses
2.2.1. Disulfide, (HDz)2, 22
The following procedure is based on that developed by Irving
and co-worker [10]: dithizone (0.5 g, 2 mmol) was dissolved in
dichloromethane (50 ml), and oxidized by sonication (5 min) to-
gether with a dichloromethane solution (10 ml) of iodine (0.51 g,
4 mmol) and water (25 ml). The organic layer was removed and
the solvent was removed under reduced pressure at room temper-
ature. The residue was dissolved in ether (leaving I2 behind) and
chromatographed through silica gel on a short column packed in
a sintered glass funnel of dimensions 50 ꢁ 40 mm. The first red
fraction was collected. It was finally purified by the addition of
cyclohexane (30 ml) to ether solution (60 ml), whereupon the
disulfide separated as a dark precipitate. Filtration and air-drying
gave 0.2 g (40%) of 22; m.p.: 107 °C; UV–Vis kmax (diethyl ether,
nm): 247, 299, 409; 1H NMR (300 MHz, CDCl3, 25 °C): d 7.10–
8.05 (m, 10H, C6H5).
2.2.2. Dehydrodithizone, 3: K3Fe(CN)6 method
A modification of a published procedure [8] was followed: a
solution of dithizone (1 g, 3.9 mmol) in dichloromethane (300 ml)
was stirred for 2 h with a solution of potassium hexacyanofer-
rate(III) (3.2 g, 9.7 mmol) and potassium carbonate (3 g,
21.7 mmol) in water (100 ml). The organic layer was removed
and the solvent was evaporated under reduced pressure. The prod-
uct residue, on recrystallization from minimum hot acetone and
30% water, gave 0.52 g (52%) of orange-red crystallographic quality
crystals of pure, water-insoluble 3 (52%); m.p.: 173 °C (explode);
UV–Vis k (acetone, nm): 456; 1H NMR (300 MHz, DMSO-d6,
25 °C): d 7.52–7.67 (m, 6H, m, p-C6H5), 7.72 (2 ꢁ d, 4H, o-C6H5).
2. Experimental
2.1. Materials and apparatus
Unless otherwise stated, dithizone, other solids and solvents
(Sigma–Aldrich, Merck) were used without further purification.
CH2Cl2 was dried by refluxing under nitrogen over calcium hydride
and redistilled immediately prior to use. Water was double dis-
tilled. 1H NMR spectra at 20 °C were recorded on a Bruker Avance
DPX 300 NMR spectrometer at 300 MHz with chemical shifts pre-
sented as d values referenced to SiMe4 at 0.00 ppm utilizing acid-
free CDCl3 as solvent. Acid traces were removed from CDCl3 by
passing the deuterated solvent through basic alumina immediately
prior to use. The potassium content, against a series of standard
solutions, was determined with a Jenway PFP7 Flame Photometer.
2.2.3. Attempted synthesis via the Na metal method
Dithizone (0.5 g, 2 mmol) was added to dry tetrahydrofurane
(200 ml) containing sodium wire, and stirred for 1 h at room tem-
perature. The resulting orange-red solution was decanted from any

K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
1729
solids and the solvent was removed under reduced pressure. The
residue was recrystallized from acetone/hexane to give 0.49 g
(91%) of pure NaHDz, 4a, characterization data as for KHDz, 4b, be-
low. Recrystallization under a pure oxygen atmosphere also failed
to generate 3. Addition of DDQ (2,3-dichloro-5,6-dicyanobenzoqui-
none or oxone (a registered trade mark for potassium peroxy-
analyses. All potentials in this study were manipulated on Excel to
be referenced against Fc/Fc+ as recommended by IUPAC [21]. In dry
CH2Cl2/0.100 M [N(nBu)4][B(C6F5)4] at 25 °C, the Fc/Fc+ couple
exhibited ipc/ipa = 0.94,
D
Ep = 81 mV, and E°0 = 412 mV versus Ag/
AgCl. Caution must, however, be exercised in utilizing this E°0 value
as a universal constant. Changes in electrodes or cell construction
from experiment to experiment caused this E°0 value to drift be-
tween 362 and 426 mV. To report accurate potentials utilizing
the Ag/AgCl reference electrode, each set of experiments was con-
ducted in the absence and presence of ferrocene. Each individual
set of data was thereafter manipulated to give the H2Dz potentials
referenced to the Fc/Fc+ internal standard couple at 0 V.
monosulfate, an oxidant, by Aldrich) did give
3 but it so
contaminated with side products that after several attempts, this
procedure was discarded as unsuitable for the synthesis of 3.
2.2.4. Potassium dithizonate, 4b
A modification of a published method [7] was as follows: potas-
sium hydroxide (0.4 g, 7 mmol) dissolved in hot methanol (100 ml)
was added dropwise to dithizone (1.0 g, 3.9 mmol) dissolved in
dichloromethane (300 ml) while stirring. The solution changed
from dark green to deep orange-red. The solvents were removed
under reduced pressure, followed by two extractions of the prod-
uct residue, first with acetone (80 ml), and then with methanol
(40 ml). After every extraction the solution was gravity-filtered
through filter paper and the combined solvent fractions evaporated
under reduced pressure. The remaining orange residue was finely
ground and dried overnight at 100 °C to give 1.12 g (98%) of 4b.
Flame photometry showed the K+:HDzꢀ ratio to be 1:1; m.p.:
200 °C (decomposed); UV–Vis k (acetone, nm): 501; 1H NMR
(300 MHz, DMSO-d6, 25 °C): d 6.70–7.75 (m, 10H, C6H5).
3. Results and discussion
3.1. Chemical oxidation
Irving reported eight different methods to oxidize 1 to the fully
deprotonated compound, dehydrodithizone, 3 [8]. Two of the re-
ported methods which gave the highest yields were repeated here,
namely oxidation of dithizone with potassium hexacyanofer-
rate(III) and iodine, Scheme 1.
The iodine-base oxidation method can be tuned to give either 22
or 3, depending on the reaction conditions and time. Iodine oxida-
tion first generates dithizone disulfide, 22, which spontaneously
fissions into equimolar amounts of 3 and 1 as described by Irving
and co-worker [10], Scheme 1. If given enough reaction time, high
yields of 3 may be obtained. However, the intermediate dimer 22
can be isolated from non-polar solvents – here we used dichloro-
methane and isolated 22 within 5 min – and stored for a longer
time (months) in the solid state. In contrast, oxidation of 1 with
the iron(III) salt, K3[Fe(CN)6], yields under basic conditions pure
3 via the potassium dithizonate salt, KHDz, 4b, but no 22 is found.
Care must be taken against the exposure of the tetrazolium salt 3
to high heat, as it explodes violently at 173 °C.
In another series of experiments, in an attempt to obtain Na2Dz,
dithizone was added to dried tetrahydrofuran containing sodium
wire. After stirring the solution for about 1 h, the green colour
changed into deep orange-red colour. The UV–Vis absorbance
spectrum of the THF solution showed kmax = 480 nm, possibly
including the unstable Na2Dz, as opposed to kmax = 506 nm for
the singly deprotonated form, 4a, and kmax = 456 nm in acetone,
for pure 3. We were, however, unsuccessful in recovering and puri-
fying Na2Dz. All attempts to do this resulted in the regeneration of
NaHDz, 4a, with characterization data as for 4b.
2.2.5. Tetrabutylammonium dithizonate, 4c
Tetrabutylammonium bromide (0.05 g, 0.155 mmol) and potas-
sium dithizonate (0.046 g, 0.155 mmol) were dissolved in acetone
(50 ml) and stirred for 1 h. The solvent was removed under re-
duced pressure, after which the residue was dissolved in dichloro-
methane and filtered. This solvent was again removed under
reduced pressure. The extraction procedure was repeated twice
more before the final residue was dried overnight at 50 °C to liber-
ate the pure product, 4c, quantitatively; m.p.: 71 °C; UV–Vis k (ace-
tone, nm): 501; 1H NMR (300 MHz, CDCl3, 25 °C): d 0.92–1.02 (t,
12H CH2CH2CH2CH3), 1.33–1.47 (m, 8H, CH2CH2CH2CH3), 1.51–
1.65 (m, 8H, CH2CH2CH2CH3), 3.15–3.35 (t, 8H, CH2CH2CH2CH3),
7.10–8.05 (m, 10H, C6H5).
2.3. Electrochemistry
Cyclic voltammetry (CV) experiments were performed on ca.
1 mM solutions of 1,
2
and
3
in dry CH2Cl2/0.100 M
[N(nBu)4][B(C6F5)4], and/or in DMSO/0.100 M [N(nBu)4][PF6] and
CH2Cl2/0.100 M [N(nBu)4][PF6] utilizing a standard three-electrode
cell with glassy carbon electrode of surface area 7.07 mm2 pre-
KHDz, 4b, as a water-soluble ionic source of dithizone, finds
extensive application in the field of spectrophotometric trace me-
tal analyses in aqueous media. To isolate 4b, the amount of KOH
dissolved in methanol required for the complete neutralization of
1 was determined titrimetrically with dithizone acting as a self-
indicator. The endpoint of the titration, however, was not easily
identifiable. The colour of an acetone solution of dithizone changed
in colour from green through grey to orange-red with the addition
of a methanolic solution of potassium hydroxide (0.38 mol dmꢀ3).
The appearance of a translucent solution with an orange-red colour
was taken to be indicative of the complete conversion of 1 to 4b. A
37% excess of potassium hydroxide was required for this purpose.
For the synthesis of 4b in this study, a 40% excess of KOH was
continuously used. After the solvent was removed under vacuum,
the dried salt 4b was redissolved twice in acetone and filtered, to
separate it from the excess acetone-insoluble KOH. Flame photom-
etry, to determine the potassium content of 4b, confirmed a 1:1 ra-
tio of K+:HDzꢀ, thereby showing that the excess of KOH used did
not convert H2Dz partially to K2Dz. As opposed to 1 and 3, which
is soluble in most of the common organic solvents, 4a and 4b is
treated by polishing on a Buehler microcloth, first with 1
lm and
then 1/4 m diamond paste, a Pt-wire counter electrode and an
l
Ag/AgCl reference electrode. The silver wire was coated with AgCl
by connecting two silver wires to the positive and negative poles of
a DC power source and submersed into a dilute HCl (0.1 M) solu-
tion. 700 mA of current was passed through the solution with the
evolution of H2 gas at the cathode. After the anode had acquired
a uniform dark-colored AgCl coating, the wire was removed and
washed with water, methanol and acetone, and dried at 110 °C di-
rectly before use. A fresh AgCl coating can be prepared on the Ag
wire by first washing away the old AgCl coating with concentrated
ammonia followed by rinsing of the now clean Ag wire with copi-
ous amounts of water. A fresh AgCl deposition is then achieved by
following the same coating procedure as described above. All elec-
trochemical measurements were conducted under a blanket of ar-
gon at 20 °C in a Faraday cage connected to a BAS 100 B/W
electrochemical workstation interfaced with a personal computer.
Data, uncorrected for junction potentials, were collected with stan-
dard BAS 100 software and exported to Excel for manipulation and

1730
K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
Table 1
only soluble in very polar solvents such as acetone, DMSO and also
water.
Absorption maxima and corresponding extinction coefficients of dithizone derivatives
in acetone.
Availability of the mono-sodium and potassium salts 4a and 4b
of dithizone salts eliminates the use of bases in the follow-up reac-
tions that are often required in metal coordination procedures [22].
For many metal coordination reactions, availability of the HDzꢀ an-
ion in a form that is soluble in non-polar organic solvents, like
dichloromethane, would be very advantageous. To achieve this,
4b may conveniently be reacted with tetrabutylammonium bro-
mide to form tetrabutylammonium dithizonate, 4c, in near quanti-
tative yield with kmax = 501 nm in acetone, Scheme 1.
UV–Vis spectra (Fig. 1 and Table 1) can clearly differentiate be-
tween 1, 22, 3 and 4a–c.
Compound
kmax/nm; (e )
/dm3 molꢀ1 cmꢀ1
H2Dz, 1
444 (19 560); 610 (32 590)
412 (33 710)
456 (1130)
(HDz)2, 22
Dz, 3
HDzꢀ, 4a–4c
501 (11 780)
Fc
2
1
10 µA
d
4
d
2
3.2. Electrochemical studies
IV
From the above description of the chemical behaviour of 1, it is
evident that dithizone is readily involved in electron transfer reac-
tions. An electrochemical study was performed to establish some
of the fundamental non-aqueous electrochemical properties of this
system. Fig. 2 gives a comparison between the cyclic voltammo-
grams of 1 obtained in DMSO – a solvent that readily solvates elec-
tron-deficient species – and CH2Cl2 in the presence of [N(nBu)4]
[PF6], and also in CH2Cl2 in the presence of [N(nBu)4][B(C6F5)4] as
a supporting electrolyte. The [PF6]ꢀ anion is known to form ion
pairs with especially positively charged oxidized species [18,19]
and frequently prevents detection of electrochemical couples
involving labile species, such as the ruthenocene/ruthenocenium
couple where ruthenocenium is the unstable radical and seventeen
electron species (Cp)2RuIII [17]. The labeling of observed oxidation
processes in Fig. 2 is sequential and starts at the resting state of 1,
while reduction processes are labeled I through IV. Anodic and
cathodic peak potentials for these processes are listed in Table 2.
The effect of the coordinating polar solvent, DMSO, was to lower
the anodic (oxidation) peak potential 1 of 1 from 705 mV in CH2Cl2
to 152 mV versus Fc/Fc+ in DMSO. Peak potential 2 was also shifted
to substantially smaller values. This is indicative of DMSO interac-
tion (solvation) with the oxidized forms of 1. CH2Cl2 is a non-coor-
dinating solvent and no interaction is expected between this
solvent and 1. However, the reduction waves III and IV moved to
only slightly higher values (ꢀ1058 and ꢀ1284 mV) in DMSO as
compared to CH2Cl2 (ꢀ1083 and ꢀ1341 mV). The result of
DMSO–dithizone interactions is therefore to narrow down the po-
tential range over which oxidation wave 2 and reduction wave IV
span, by 610 mV, in comparison with the analogous cyclic voltam-
mogram obtained in CH2Cl2.
1
4
SW-R
3
3
4
c
LSV-R
c
2
1
I
IIb
IIa
4
III
2
1
IV
SW-L
I
III
IV
LSV-L
3
-
CH2Cl2, PF6
b
1
2
I, II
III
IV
DMSO
a
I, II
III
IV
-2000 -1500 -1000
-500
0
500
1000
1500
+
Potential / mV vs Fc/Fc
Fig. 2. Cyclic voltammograms of 1 mmol dmꢀ3 solutions of H2Dz, 1, at scan rates of
100, 200, 300, 400 and 500 mV sꢀ1 and 20 °C, at a glassy carbon working electrode,
in
(a)
(CH3)2SO/0.1 mol dmꢀ3
[N(nBu)4][PF6];
(b)
CH2Cl2/0.1 mol dmꢀ3
[N(nBu)4][PF6]; (c) and (d) CH2Cl2/0.1 mol dmꢀ3 [N(nBu)4][B(C6F5)4], (c) also shows
the square wave voltammograms at 100 Hz and linear sweep voltammograms at
2 mV sꢀ1. The trace LSV-L has the special feature of being initiated at potentials
where oxidation processes 1 and 2 have not been invoked. As a consequence, peaks
I and II were absent. (d) Free ferrocene (Fc) has been added as internal standard.
Assigned oxidation (waves 1–4; numbering starts from the resting state of 1) and
reduction peaks (waves I–IV) correspond to reactions shown in Schemes 2 and 3.
4
3
2
1
0
Fig. 2 also clearly shows the enhanced resolution and observa-
tion of processes that can be observed in experiments conducted
in the presence of [B(C6F5)4]ꢀ anions, as compared to those that
were done in the presence of [PF6]ꢀ. The improved results in the
presence of [N(nBu)4][B(C6F5)4] as a supporting electrolyte is the
consequence of the lower charge density on [B(C6F5)4]ꢀ anions
compared to [PF6]ꢀ which leads to a much less tendency to form
ion pairs of the type HDz+ꢂ ꢂ ꢂ[B(C6F5)4]ꢀ. With respect to the elec-
trochemistry of 1, use of [N(nBu)4][B(C6F5)4] clearly allowed for
the observation of the otherwise completely absent cathodic peaks,
350
450
550
650
Wavelength / nm
Fig. 1. UV–Vis spectra of 3 (yellow), 4b (orange), 22 (red) and 1 (green) in acetone.
Line colours correspond to observed solution colours; concentrations of acetone
stock solutions were ca. 0.04 mM (or 0.2 mM for 3) to give absorbance values that
never exceeded 1.4. (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)

K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
1731
Table 2
before the second oxidation wave, wave 2 at 1026 mV versus Fc/
Fc+, sets in. At wave 2, the dimeric species 22 is expected to oxidize
at the relatively labile SAS bond, thereby creating a positive charge
on the sulfur atom in one electron per repeating unit during the
formation of monomeric 22 oxidized (Scheme 2).
Cyclic voltammetry peak potentials vs. Fc/Fc+ of 1, in DMSO and CH2Cl2 in the
presence of 0.1 mol dmꢀ3 [N(nBu)4][PF6], or in CH2Cl2 containing 0.1 mol dmꢀ3
[N(nBu)4][B(C6F5)4] as supporting electrolyte at a scan rate of 200 or 500 mV sꢀ1 and
20 °C, Fc = ferrocene.
Peak
no.
E/mV in DMSO/
PF6
E/mV in CH2Cl2/
PF6
E/mV CH2Cl2/
Our results are consistent with a fast electrochemical and
chemical conversion between 22 oxidized, 22 and 2. It is easy to
ꢀ
ꢀ
ꢀ
BðC6F5Þ4
1
2
I
IIa
IIb
III
IV
3
152
- - -
- - -
- - -
- - -
ꢀ1058
ꢀ1284
- - -
705
991
- - -
558; 624b
understand how 22
should chemically cyclise into mono-
oxidized
990; 1026b
ꢀ130; ꢀ210b
ꢀ572; ꢀ662b
- - -; ꢀ778b
meric 3 (see Supplementary data), but this does not happen on
the CV time scale, because the CV of authentic 3 does not repro-
duce the same CV fingerprint that 1 or 22 did.
- - -
ꢀ841
ꢀ1083
ꢀ1341
ꢀ850a
ꢀ303
ꢀ996; ꢀ1048b
ꢀ1342; ꢀ1382b
ꢀ850a; ꢀ850a,b
ꢀ344; ꢀ304b
A complete reduction reaction sequence for 1 in the cathodic
(reduction) half-cycle of CV ‘‘c” of Fig. 2 is proposed in Scheme 3.
The first reduction wave, wave I, was observed at ꢀ210 mV. Here,
monomeric 22 oxidized is expected to be reduced with the formation
of radical 2 upon the addition of an electron. Again, 2 should chem-
ically dimerise to give 22. However, reduction wave II manifested
as a twin peak with subwaves IIa and IIb at scan rates faster than
200 mV sꢀ1. This is consistent with two separate species which
were reduced. One explanation for this is simply that the kinetics
of dimerisation of 2 is too slow to allow complete dimerisation be-
fore the onset of wave II. Hence waves IIa at ꢀ662 mV and IIb at
ꢀ778 mV, both at 500 mV sꢀ1, may represent the reduction of dif-
ferent mixtures of 22 and 2, the composition of which is deter-
mined by the scan rate. A close inspection of CV ‘‘c” in Fig. 1
supports this conclusion. To clarify, at a scan rate of 100 mV sꢀ1
waves I and II were not clearly observed, because at this slow scan
rate, by the time the potentials were reached where 22 oxidized, 2
and 22 should be reduced, diffusion effects to homogenize the solu-
tion would have depleted the concentration of these species to the
point that detection would be difficult. However, at a scan rate of
200 mV sꢀ1, wave IIb is barely observable, but wave IIa can be
identified. This is consistent with the enough time that has elapsed
to allow the substantial conversion of 2 to 22 as well as the assign-
ment of wave IIa to the reduction of dimer 22. Strikingly, by
increasing the scan rate to 500 mV sꢀ1, wave IIb becomes more
dominant. This is consistent with the too little available time to al-
low 2 to dimerise quantitatively before the potentials associated
with reduction waves IIa and IIb are reached. The reduction of both
4
- - -
- - -, No redox process observed.
Supplementary data contains data at scan rates of 100, 200, 300, 400 and
500 mV sꢀ1 in this solvent electrolyte system.
a
Estimated value.
b
All the second values are for a scan rate of 500 mV sꢀ1
.
I, IIa and IIb. Without these peaks, an accurate description of the
redox cycle would have been impossible.
With all the above evidence at hand, Schemes 2 and 3 for the
complete redox cycle of H2Dz may now be proposed. Cyclic vol-
tammetry of 1 (Fig. 2) reveals two oxidation and two reduction
electrochemically irreversible one-electron transfer processes.
The oxidation sequence of 1 from the resting state is a consequence
of free H2Dz having two acidic protons that may be removed chem-
ically or electrochemically. The first oxidation step at 624 mV
(wave 1 at 500 mV sꢀ1, CV ‘‘c”, Fig. 2) is consistent with 1 losing
one electron and one proton to form the sulfide radical
2
(Scheme 2). Although 2 can in principle be oxidized with the loss
of a proton to give immediately a monomeric doubly oxidized spe-
cies, a separate investigation in the electrochemistry of the dimeric
species 22 revealed that this dimer in essence gives exactly the
same CV’s as 1 (see Supplementary data). This leads us to conclude
that the sulfide radical 2 predominantly chemically dimerises fast
enough under oxidizing conditions to form 22 on the CV timescale
H
S
H
N
S
H
N
N
C
N
C
-2e-, -2H+
2
2
N
N
N
N
Wave 1
2
1
N
C
N
H
N
C
N
H
N
N
N
N
N
N
-2e-
Wave 2
chemical
dimerization
S
S
S
C
N
2
N
H
22 oxidized
22
Scheme 2. Proposed oxidation of 1. Chemical dimerisation of 2 to 22 prior to the second oxidation step is feasible because a solution of authentic 22 showed the same CV as 1.
The final oxidation product is proposed to be 22 oxidized because on CV timescale, formation of the tetrazolium dimer, 3, was not observed. For clarity, only one isomer of 1 is
shown. Electrochemistry wave numbers correspond to wave numbers shown on the CV’s of Fig. 2.

1732
K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
H
S
N
C
N
H
S
N
C
N
N
C
N
H
S
N
C
N
N
N
N
N
N
N
N
N
2e-
chemical
2
S
2
dimerization
Wave I
H
22
2
22
oxidized
H
S
N
H
S
N
H
S
N
N
C
N
N
N
N
H
N
N
H
2e-, 2H+
2e-, 2H+
2e-
C
N
2
2
2
C
N
Wave III
Wave IV
Wave IIa, b
H
H
H
1
8
7
neutral resting state
Scheme 3. The complete reduction half-cycle of 22
involves four one-electron transfer processes to first produce 2. Acid–base equilibria are anticipated between
oxidized
protonated and deprotonated species resulting from all reduction steps, but for clarity are not shown. Also for clarity, just one isomer of 1 is shown. Electrochemistry wave
numbers correspond to wave numbers on the CV’s shown in Fig. 2. Reoxidation of 8 involves the reverse reactions associated with waves IV and III, but are labeled 3 and 4 in
Fig. 2.
2 and 22 ultimately leads to the formation of 1 in a one-electron
transfer process per dithizonyl fragment after the uptake of a pro-
ton. It is, however, reasonable to assume that a shortage of H+ ex-
ists at the surface of the electrode during reduction of 22 and 2
because the only freely available source of H+ in the reaction med-
ium is that which was generated during oxidation wave 1, and that
which can be abstracted from free 1 (pKa = 5.77 [23]) in the bulk of
the solution. Diffusion effects to homogenize the solution imply
free generated H+ diffused away from the surface during oxidation
and 1 was depleted of the diffusion rate steady state concentration
during oxidation. Since diffusion effects imply that neither of these
proton sources will be instantaneously available at the surface of
the electrode, a kinetic effect related to the availability of protons
makes the formation of the dithizonato anion 4 together with 1 en-
tirely possible during the processes associated with waves IIa and
IIb.
The fact that waves I (and II) really are associated with oxida-
tion waves 1 and 2 was proved by initiating CV and LSV scans in
the negative direction from 0 V versus Fc/Fc+. Upon scanning in this
direction, no oxidation processes were initiated and hence peaks I
or IIa and IIb should not be present. Fig. 2 c shows this for the LSV
scan labeled LSV-L. The square wave curve labeled SW-L allows
better resolution of waves IIa and IIb (Fig. 2).
through-bond conjugated paths [24,25], which implies that if both
1 and 4 were present, wave III should have been split into two
components IIIa and IIIb. Radical formation on the indicated sulfur
atom of 7 must dominate over radical formation on the nitrogen
atoms (Scheme 3) but sterical hindrance of radical 7 must be larger
than that in 2 to prevent 7 from dimerising because no evidence for
a dimer of 7 could be found.
In the last reduction step, the H3Dzꢀ anion 8 forms during wave
IV at ꢀ1382 mV. The proposed structure 8 is the deprotonated
form of the four-proton diphenylthiocarbodihydrazide, H4Dz. Crys-
tallographic evidence [5] showed H4Dz, one of the isolatable pre-
cursors in the chemical synthesis of 1 according to Eq. (1), to
exist with the fourth H-atom being bonded to the fourth nitrogen
of 8 and a C@S bond formed by isomerisation, but the sulfide anio-
nic form, 8, shown in Scheme 3 is regarded to form during CV in
view of the shortage of available protons in the non-protic solvent
system used at the surface of the electrode.
The last part of the dithizone cyclic voltammogram, namely the
two one-electron re-oxidations from 8 to 1 via 7 as shown by
waves 3 and 4 in Fig. 2 is simply the reverse of the last two steps
shown in Scheme 3. Loss of an electron from 8 is estimated to take
place at about ꢀ850 mV (wave 3) while the parent compound 1 is
generated during wave 4 at ꢀ304 mV. The square wave curve la-
beled SW-R highlights the enhanced detection of reduction wave 3.
Reduction wave III (Fig. 2), at ꢀ1048 mV, represents the forma-
tion of the trihydrogen-containing radical species, H3DzÅ, 7,
(Scheme 3) upon one-electron reduction of 1. Since wave III does
not manifest as a twin peak possessing sub-peaks a and b, it is con-
cluded that the source of 7 is not a mixture of 4 and 1 but only 1.
Different formal reduction potentials for complexes in which
mixed-valent intermediates are generated are well known in sys-
tems that allow electron delocalisation via through-space or
4. Conclusions
The chemical and electrochemical electron transfer behaviour
of
1
correlates well. Electrochemical results in CH2Cl2/
[N(nBu)4][B(C6F5)4] are consistent with the identification of six

K.G. von Eschwege, J.C. Swarts / Polyhedron 29 (2010) 1727–1733
1733
species: H3Dzꢀ, 8; H3DzÅ, 7; H2Dz, 1; HDzÅ, 2; (HDz)2, 22 and
(HDz+)2, 22 oxidized. The fully oxidized electrochemically generated
complex, 22 oxidized, can chemically collapse into the explosive
monomeric tetrazolium salt 3, dehydrodithizone, just as 1 can do
so under the influence of I2 or K3[Fe(CN)6, but does not do so on
a CV timescale. Of these six electrochemically generated species,
four, namely H3Dzꢀ, 8 (as H4Dz); H2Dz, 1, HDzꢀ, 4, (HDz)2, 22 as
well as the chemically oxidized species 3 are all verifiable by
means of chemical synthetic procedures and X-ray crystallographic
structure determinations. DMSO solvation and ion pairing between
[8] H.M.N.H. Irving, A.M. Kiwan, D.C. Rupainwar, S.S. Sahota, Anal. Chim. Acta 56
(1971) 205.
[9] A.M. Kiwan, H.M.N.H. Irving, Chem. Commun. (1970) 928.
[10] A.M. Kiwan, H.M.N.H. Irving, J. Chem. Soc. B (1971) 901.
[11] A. Kuhn, K.G. von Eschwege, Inorg. Chem. Acta, to be submitted.
[12] (a) Y. Kushi, Q. Fernando, J. Am. Chem. Soc. 92 (1970) 1965;
(b) T.V. Koksharova, Zh. Strukt. Khim. (Russ.) (J. Struct. Chem.) 45 (2004) 361;
(c) F. Jian, P. Zhao, L. Zhang, Y. Hou, J. Org. Chem. 70 (2005) 8322.
[13] J.W. Ogilvie, A.H. Corwin, J. Am. Chem. Soc. 83 (1961) 5023.
[14] W. Zhang, L. Xu, G. Xi, W. Yu, Y. Qian, Chem. Lett. 33 (2004) 1476.
[15] (a) L. Tomcsanyi, Anal. Chim. Acta 70 (1974) 411;
(b) L. Tomcsanyi, Anal. Chim. Acta 88 (1977) 371;
(c) L. Tomcsanyi, Anal. Chim. Acta 89 (1977) 409.
ꢀ
[16] K.G. von Eschwege, J. Conradie, J.C. Swarts, J. Phys. Chem. A 112 (2008) 2211.
[17] (a) R.J. LeSuer, C. Buttolph, W.E. Geiger, Anal. Chem. 76 (2004) 6395;
(b) J.C. Swarts, A. Nafady, J.H. Roudebush, S. Trupia, W.E. Geiger, Inorg. Chem.
48 (2009) 2156.
[18] (a) A. Auger, J.C. Swarts, Organometallics 26 (2007) 102;
(b) A. Auger, A.J. Muller, J.C. Swarts, Dalton Trans. (2007) 3623;
(c) M.J. Cook, I. Chambrier, G. White, E. Fourie, J.C. Swarts, Dalton Trans. (2009)
1136;
PF6 and especially oxidized dithizone species have a detrimental
effect on the resolution and observation of electrochemical
processes. By reacting potassium dithizonate with tetrabutylam-
monium bromide, a dithizone salt that is also soluble in non-polar
solvents, such as CH2Cl2, can be obtained.
Acknowledgement
(d) E. Fourie, J.C. Swarts, I. Chambrier, M.J. Cook, Dalton Trans. (2009) 1145.
[19] (a) K.C. Kemp, E. Fourie, J. Conradie, J.C. Swarts, Organometallics 27 (2008)
357;
The authors acknowledge the Central Research Fund of the Uni-
versity of the Free State for financial assistance.
(b) H.J. Gericke, N.I. Barnard, L. Erasmus, J.C. Swarts, M.J. Cook, M.A.S. Aquino,
Inorg. Chem. Acta (2010), in press.
[20] (a) F. Barrière, N. Camire, W.E. Geiger, U.T. Mueller-Westerhoff, R. Sanders, J.
Am. Chem. Soc. 124 (2002) 7262;
(b) F. Barrière, W.E. Geiger, J. Am. Chem. Soc. 128 (2006) 3890.
[21] G. Gritzner, J. Kuta, Pure Appl. Chem. 56 (1984) 461.
[22] A.T. Hutton, J. Chem. Ed. 63 (1986) 888.
[23] A.R. Al-Salihy, H. Freiser, Talanta 17 (1970) 182.
[24] (a) C. Creutz, H. Taube, J. Am. Chem. Soc. 91 (1969) 3988;
(b) W.E. Geiger, N. Van Order, D.T. Pierce, T.E. Bitterwolf, A.L. Reingold, N.D.
Chasteen, Organometallics 10 (1991) 2403;
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.poly.2010.02.009.
References
(c) N. Van Order, W.E. Geiger, T.E. Bitterwolf, A.L. Reingold, J. Am. Chem. Soc.
109 (1987) 5680;
(d) D.T. Pierce, W.E. Geiger, Inorg. Chem. 33 (1994) 373;
(e) W.L. Davis, R.F. Shago, E.H.G. Langner, J.C. Swarts, Polyhedron 24 (2005)
1611.
[1] H. Fischer, Wiss. Veröffentlich. Siemens-Werken 4 (1925) 158.
[2] A.L.D. Comitre, B.F. Reis, Talanta 65 (2005) 846.
[3] A.T. Hutton, Polyhedron 6 (1987) 13.
[4] P.S. Pelkis, R.G. Dubenko, L.S. Pupko, J. Org. Chem. USSR 27 (1957) 190.
[5] M. Harding, M.J. Adams, P.A. Alsop, H.M.N.H. Irving, Anal. Chim. Acta 67 (1973) 204.
[6] F.H. Herbstein, W. Schwotzer, J. Am. Chem. Soc. 106 (1984) 2367.
[7] H. Noda, L.M. Engelhardt, J.M. Harrowfield, C. Pakawatchai, J.M. Patrick, A.H.
White, Bull. Chem. Soc. Jpn. 58 (1985) 2385.
[25] (a) W.C. Du Plessis, J.J.C. Erasmus, G.J. Lamprecht, J. Conradie, T.S. Cameron,
M.A.S. Aquino, J.C. Swarts, Can. J. Chem. 77 (1999) 378;
(b) J. March, Advanced Organic Chemistry, 4th ed., John Wiley and Sons, New
York, 1992. p. 17, 263, 273.