Selective oxidation of an electronically unsymmetrical distyrylbenzene at either of two sites

Article abstract of DOI:10.1149/2.013307jes

A p-distyrylbenzene bearing electron-withdrawing groups on one terminal ring and electron-supplying groups on the other was synthesized and one ring was found to be significantly more electron rich than the other. Electrochemical oxidative cleavage was sh

Full text of DOI:10.1149/2.013307jes

Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
G3091
0013-4651/2013/160(7)/G3091/6/$31.00 © The Electrochemical Society
JES FOCUS ISSUE ON ORGANIC AND BIOLOGICAL ELECTROCHEMISTRY
Selective Oxidation of an Electronically Unsymmetrical
Distyrylbenzene at Either of Two Sites
Anthony P. Davis∗ and Albert J. Fry∗∗,z
Chemistry Department, Wesleyan University, Middletown, Connecticut 06459, USA
A p-distyrylbenzene bearing electron-withdrawing groups on one terminal ring and electron-supplying groups on the other was
synthesized and one ring was found to be significantly more electron rich than the other. Electrochemical oxidative cleavage was
shown to take place predominantly at the electron-rich double bond. Selective protection of that double bond followed by anodic
oxidation and deprotection afforded the product of overall anodic cleavage of the electron-deficient double bond, thus expanding the
range of possible transformations of the distyrylbenzene.
© 2013 The Electrochemical Society. [DOI: 10.1149/2.013307jes] All rights reserved.
Manuscript submitted March 7, 2013; revised manuscript received April 16, 2013. Published April 25, 2013. This paper is part of
the JES Focus Issue on Organic and Biological Electrochemistry.
It is an article of faith in synthetic electrochemistry, amply justified
by precedent, that a substance with two electroactive sites of different
redox potential undergoes preferential electrochemical conversion at
the site of lower potential. This is an inherent advantage in opera-
tional electrochemistry over reduction or oxidation processes carried
out using chemical agents, which are often insufficiently selective. Yet
sometimes it might be desired to carry out an electrochemical trans-
formation at the site of higher potential without affecting the lower
potential site. We describe here the synthesis of an electronically un-
symmetric molecule and a strategy for selective anodic cleavage at
either of its two potentially electroactive sites.
The anodic oxidation of alkenes can produce a wide variety of
products, depending upon the alkene structure and the experimental
conditions.1–3 Depending upon conditions, a variety of products are
typically formed, e.g., cleavage of the carbon-carbon bond competes
with rearrangement of a phenyl group, affording diphenylacetalde-
hyde, which is then quickly oxidatively converted to benzophenone.4
However, a number of other products can also be produced including
dimeric and rearranged substances and products of nucleophilic attack
upon cationic intermediates. We have previously described the anodic
cleavage of stilbenes (1,2-diphenylethylenes) into two molecules of
aldehyde (Equation 1) in the presence of water.5,6
bis-[trifluoromethyl]benzyltriphenylphosphonium bromide (3) and an
equimolar amount of terephthaldehyde under phase transfer condi-
tions afforded stilbene 4 in 58% yield together with a smaller amount
(<9%) of 3,3,5ꢀ,5ꢀ-tetrakis[trifluoromethyl]stilbene. A second Wit-
tig condensation of 4 with 3,5-dimethylbenzyltriphenylphosphonium
bromide (5) afforded the desired unsymmetrical distyrylbenzene 6.
Both Wittig steps afforded mixtures of cis and trans isomers, which
could be converted completely to the trans form by overnight reflux in
toluene containing a trace of I2. This was particularly convenient with
the distyryl compound 6, which was initially produced as a mixture
of E,E, Z,E, E,Z, and Z,Z isomers by gc-mass spectrometric analy-
sis and immediately converted to the E,E isomer by iodine-catalyzed
equilibration.
Charge distribution in 6.— The first issue to be addressed was
whether the different substituents on the two terminal rings of 6 are
likely to result in significantly different reactivity between its two
double bonds. The electrostatic potential map of 6 (Figure 1) computed
at the DFT/6–31G(d) level with full geometry optimization, shows the
highest electron density to be in the dimethylated ring and its adjacent
double bond. A natural population analysis7 (NPA) computation of
the charges on each atom confirmed the electrostatic potential map in
showing that the carbon atoms in the dimethylated ring carry a higher
degree of negative charge than the corresponding ring atoms at the
other end of the molecule (Table I). This is borne out further in the
1H NMR spectra of 6, in which the proton resonances of the dimethyl
substituted ring as well as the adjacent double bond appear at higher
fields than those of the ring and double bond at the other end of the
molecule (Table II). In short, the data all lead to the conclusion that
the electron density in 6 is highly unsymmetrical and therefore that
anodic oxidation should take place at the dimethylated end.
−
−4e
ArCH = CHArꢀ −−−−−−−→ ArCH = O + O = CHArꢀ
[1]
99:1 CH CN:H O
3
2
We observed that oxidation potentials increase greatly as one or
more electron-withdrawing groups are attached to the stilbene nucleus.
Compound (1) containing a stilbene unit at one end and another at the
other end with the ends connected by a saturated chain (Scheme 1)
would surely undergo oxidation at the end bearing electron-donating
substituents. But it was not obvious whether selective oxidation might
be possible when the two units are connected such that the central
benzene ring is part of both two stilbene subunits, because such a
molecule, a so-called 1,4-distyrylbenzene (2), constitutes a single
extended electronic system. It was decided to construct such a system
and determine whether its electron density is unsymmetrical. If so, the
next questions would be (a) whether the more electron rich end could
be selectively protected from oxidation, permitting electrochemistry
at the other end and (b) the feasibility of removal of the protecting
group.
Results and Discussion
Synthesis of alkenes.— The synthesis of the desired distyrylben-
zene is straightforward (Scheme 2). A Wittig reaction between 3,5-
∗Electrochemical Society Active Member.
∗∗Electrochemical Society Fellow.
zE-mail: afry@wesleyan.edu
Figure 1. Electrostatic potential map of distyrylbenzene 6. Color code: red:
highly negative; yellow: slightly negative; green: slightly positive; blue: highly
positive.

G3092
Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
(CH2)n
(EWG)n
(EDG)n
(EDG)n
(EWG)n
1
2
Scheme 1. Distilbenes bearing both electron-withdrawing (EWG) and electron-donating (EDG) groups.
+
CHO
F3C
F3C
1. Na2CO3 (xs), Bu4N+I– (cat.)
1 : 1 CH2Cl2 : H2O
P(C6H3)3
Br–
CHO
CH3
+
2. I2 (trace), toluene, reflux
CF3
F3C
4
CHO
3
F3C
CH3
1. Na2CO3 (xs), Bu4N+I– (cat.)
1 : 1 CH2Cl2 : H2O
+
4
+
(C6H3)3P
Br–
CH
CH3
3 2. I2 (trace), toluene, reflux
F3C
6
5
CHO
H3C
1. Na2CO3 (xs), Bu4N+I– (cat.)
1 : 1 CH2Cl2 : H2O
CHO
+
5
2. I2 (trace), toluene, reflux
H3C
CHO
7
H3C
Br
CH2Cl2 , 0°C
CHO
–
7
+
PyrH+ Br3
Br
H3C
8
CH3
Br
F3C
1. Na2CO3 (xs), Bu4N+I– (cat.)
1 : 1 CH2Cl2 : H2O
3
+
8
Br
2. I2 (trace), toluene, reflux
CH3
F3C
9
Scheme 2. Synthesis of an electronically unsymmetrical distyrylbenzene and a derivative with the more electron-rich double bond protected.
Selective protection of the more electron rich double bond of 6.—
The next steps in the overall plan were to protect the more elec-
tron rich double bond of 6, carry out the electrolytic anodic cleav-
age of the remaining double bond, and finally remove the protecting
group. It was decided to synthesize 9, a derivative of 6 in which the
more electron rich double bond of unsymmetrical distyrylbenzene 6 is
protected against oxidation by dibromination, since the vic-dibromo
functionality could be easily removed cathodically. Synthesis of 9 was
straightforward: Wittig synthesis of stilbene aldehyde 7, followed by
Table II. NMR shifts of protons of unsymmetrical
distyrylbenzene 6.
Table I. Atomic charges of atoms in rings
distyrylbenzene 6.
A
and B of
CH3
D'
B
F3C
C
E
G
CH3
7
A
3
F3C 2
1
6
C'
CH3
E'
F3C
4
D
8
F
5
B
A
Proton
Chemical shift (δ)a
CH3
A
B
7.76
7.93
7.24
7.15
7.55
7.11
7.18
6.95
F3C
C
a
a
Ring A carbon AtomicChargeA
Atomic ChargeB
Ring B carbon
Cꢀ
D
1
2
3
4
−0.188
−0.160
−0.176
−0.046
−0.234
−0.160
−0.228
−0.058
8
7
6
5
E,Eꢀb
F
G
aAtomic charges computed by the Natural Population Analysis proce-
dure at the density functional B3LYP/6–31G(d) level.
aChemical shifts measured at 400 MHz in CDCl3.
bProtons E and Eꢀ appear as a slightly broadened singlet.

Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
G3093
Figure 4. Cyclic voltammogram of 1,2-diphenylethylene (stilbene).
that the stronger carbon-chlorine bond and lesser leaving group ability
of chlorine under electrophilic conditions12 would confer greater sta-
bility upon the cation radical of 10c. The oxidation potential of 10c is
+0.73 V13 and lower than that of 6 (+0.90 V). In the event, controlled
potential anodic oxidation of a solution of 6 in acetonitrile/water solu-
tion containing 10 mol% of 10c was carried out at +0.75 V vs Ag/Ag+.
The initial blue solution of the cation radical of 10c immediately be-
gan to stream from the anode as expected, but the solution quickly
began to develop a purple color, suggesting that 10c was decompos-
ing. Well before 6 had been consumed, the current passing through
the cell had dropped to zero and the color had become light pink,
indicating that the cation radical of 10c had completely decomposed.
Analysis by GC-MS indicated the presence of two major components:
unreacted 6 (48%) and 4-formyl-3ꢀ,5ꢀ-bis[trifluoromethyl]stilbene (4)
(36% crude yield of a 5:1 mixture of the stilbene aldehydes 4 and 7;
72% yield based upon unreacted starting material). Workup and flash
chromatography afforded a mixture of 4 and 7, which was separated
by preparative thin layer chromatography resulted in isolation of pure
4 and 7 in 5:1 ratio, confirming our expectation that the double bond
nearer the dimethyl substituted ring would be the preferred site of
cleavage of 6 (Scheme 4). The rapid decrease in current, coupled with
the change in color, suggested that the catalyst was decomposing dur-
ing electrolysis. Peters has reported similar decomposition of metal
salen complexes during attempted electrocatalytic reduction of alkyl
halides.14 Addition of fresh catalyst to the solution was unsuccessful
because decomposition of the catalyst appeared to take place faster
as electrolysis proceeded. The nature of the decomposition of 10c is
under investigation.
Figure 2. Cyclic voltammogram of unsymmetrical distyrylbenzene 6.
bromination of the double bond of 7 using pyridinium perbromide8
(Scheme 2) and a final Wittig reaction to produce 9.
Voltammetry.— The oxidation potentials of compounds 6 (+0.90
V), 9 (+1.44 V), and the parent molecule stilbene (+1.45 V) were
measured by cyclic voltammetry at a glassy carbon electrode in ace-
tonitrile/0.1 M Bu4NBF4 relative to Ag/0.1 M AgNO3 in acetonitrile
(Figures 2–4). As a consequence of its greater degree of conjuga-
tion, distyrylbenzene 6 is about 0.5 V easier to reduce than 9, whose
reduction potential (ca. 1.45 V) is close to that of stilbene itself.
Electrocatalytic oxidation of 6.— We had previously employed
tris[4-methyl-2-nitro]triphenylamine (10a) as an electrocatalyst
(more properly, mediator of electron transfer) for the anodic cleav-
age of stilbenes bearing as many as four electron-withdrawing sub-
stituents (Scheme 3).4,5 However, the oxidation potential of 10a is too
high to be used in similar fashion for the anodic oxidation of the dis-
tyrylbenzene 6. The most widely used anodic mediator is tri-4-bromo-
triphenylamine (10b), introduced for this purpose by Steckhan.9 How-
ever, we were aware that the cation radical of 10b has since been
found to undergo slow decomposition.10 We therefore decided to use
the corresponding trichloro derivative 10c, which we synthesized (by
a modification of an earlier literature procedure11) in the expectation
Anodic oxidation and deprotection of 9.— Because of the decom-
position of mediator 10c during the anodic oxidation of 6, it was
decided to carry out the oxidation of 9 directly, i.e., without mediator.
Because of the readily reducible vicinal dibromide functionality of 9,
the oxidation was carried out in a divided cell at a platinum anode.
After passage of the calculated4 4 Faradays/mol of current, gc-mass
spectrometric analysis indicated that the expected cleavage of the
Ar
Ar
N
Ar
10a, Ar = 4-Methyl-2-nitrophenyl
b, Ar = 4-Bromophenyl
c, Ar = 4-Chlorophenyl
Scheme 3. Triphenylamines bearing electron-withdrawing groups.
Figure 3. Cyclic voltammogram of dibromodihydrodistyrylbenzene 9.

G3094
Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
CH3
H3C
H
F3C
H
F3C
[O]
6
+
O
+
O
H3C
CH3
F3C
7
F3C
4
48%
6
15%
53%
Scheme 4. Anodic oxidation of an unsymmetrical distyrylbenzene at the more electron-rich double bond.
double bond had taken place. Workup afforded the expected cleavage
product (8) together with a small amount of debrominated material (7)
(Scheme 5). Finally, cathodic deprotection of 8 occurred smoothly at
a carbon cathode under controlled potential to afford the deprotected
aldehyde 7 in 75% yield, identical with a previously synthesized (see
Scheme 2) sample.
equiv)and the mixture was stirred vigorously overnight. 25 ML of
water was added and the organic layer was removed by evaporation.
The aqueous layer was extracted with 3 15-mL portions of CH2Cl2.
The combined organic layers were dried over MgSO4 and evaporated
to afford a yellow liquid, that mostly crystallized upon standing
overnight. Analysis by GC-MS showed a 45:55 mixture of stereoiso-
mers. The mixture was purified by flash chromatography with
toluene as eluent. A trace amount of I2 was added to the combined
product fractions and the solution stirred overnight at reflux to obtain
(E)-isomer (99%) in toluene. The solvent was evaporated and the
residue recrystallized from 95% EtOH to afford 1.17 g of fine white
Experimental
Tri-4-chloro-triphenylamine (10c).— was prepared using a pro-
cedure described by Patil et al.11 1-Chloro-4-iodobenzene (2.73 g,
11.5 mmol) and 4-chloroaniline (0.696 g, 5.5 mmol) were dissolved
in dry toluene (35 mL). Copper(I) iodide (38.1 mg, 3.6 mol%) and
2,2ꢀ-bipyridyl (31.5 mg, 3.6 mol%) were added to the solution, and the
reaction vessel was placed in an oil bath heated to 115◦C. Potassium
tert-butoxide (1.85 g, 16.5 mmol) was added, and then the vessel was
sealed and purged with N2. After 5 hr, the reaction was allowed to
cool to RT. Solids were removed by vacuum filtration, and solvent was
removed by rotary evaporation. The crude product was dissolved in
CH2Cl2 (30 mL) and washed with 0.1 M aqueous NaHSO3 (2.3 mL),
followed with H2O (1.3 mL). Purification by flash chromatography
over silica with hexane as the eluent, followed by recrystallization
from 95% EtOH, yielded 853 mg of 10c (44% yield): mp 144–146◦C
(lit.16 mp 148–149◦C).
1
needles (58% yield): mp 140–142◦; H NMR (400 MHz, CH3CN):
9.96 (s, 1H), 7.87 (d, J = 8.0 Hz, 2H), 7.72 (d, J = 8.0 Hz, 2H),
7.30/7.24 (qAB, J = 12.3 Hz, 2H), 7.24 (s, 2H), 6.98 (s, 1H), 2.32 (s,
6H); 13C NMR (CDCl3): 191.7, 143.8, 138.5, 136.6, 135.4, 132.6,
130.5, 130.4, 127.1, 127.0, 125.0, 21.5; MS m/z (relative intensity):
236 (100), 192 (38), 178 (31).
3,5-Dimethylbenzyltriphenylphosphonium bromide (5).— 3,5-
Dimethylbenzyl bromide (1.99 g, 10.0 mmol) and triphenylphosphine
(3.70 g, 14.0 mmol) were dissolved in 100 mL of benzene and heated
to reflux overnight to produce a white precipitate. The mixture was
cooled in ice and the precipitate was removed by vacuum filtration
and dried at 90◦C to yield 4.40 g (95% yield) of the phosphonium salt,
mp (dec) 313–315◦C; 1H NMR (400 MHz, DMSO-d6): 7.88–7.93 (m,
3H), 7.60–7.8 (m, 12H), 6.93 (s, 1H), 6.50 (s, 2H), 5.03 (d, J = 15.6
Hz, 2H), 2.06 (s, 6H).
3,5-Bis[trifluoromethyl]benzyltriphenylphosphonium
bromide
(3).— 3,5-bis[trifluoromethyl]-benzyl bromide (3.07 g, 10.0 mmol)
and triphenylphosphine (3.70 g, 14.0 mmol) were dissolved in
100 mL of benzene and heated to reflux overnight to produce a
white precipitate. The mixture was cooled in ice and the precipitate
was removed by vacuum filtration and dried at 90◦C to yield 5.60 g
(98% yield) of the phosphonium salt, mp (dec) 314–320◦C; 1H NMR
(400 MHz, DMSO-d6): 8.10 (s, 1H), 7.92–7.96 (m, 3H), 7.67–7.85
(m, 12H), 7.58 (s, 2H), 5.39 (d, J = 15.6 Hz, 2H).
(E,E)-1-(3,5-Dimethylstyryl)-2–4-(3ꢀ5ꢀbis[trifluoromethyl]styryl)-
benzene (6).— The crude product from the synthesis of 4a was
dissolved in 25 mL CH2Cl2. Tetrabutylammonium iodide (124 mg,
10 mole%) and Na2CO3 (12.9 g, 40 equiv) were added, followed
by yy (1.28 g, 2.78 mmol) and 25 mL H2O. The flask was capped
and stirred vigorously overnight. Water (100 mL) was added and
the organic layer separated by separatory funnel. The aqueous
layer was extracted with CH2Cl2 (3 × 30 mL) and the combined
organic layers were dried over Na2SO4 and evaporated to afford
2.97 g of crude product. The product mixture was analyzed by
GC-MS and found to consist of unreacted 4a, three stereoisomers
of 1,4-bis[trifluoromethyl)styrylbenzene, and four stereoisomers
(E)-4-Formyl-3ꢀ,5ꢀ-bis[trifluoromethyl]stilbene
(4).— 3,5-bis
[trifluoromethyl]benzyltriphenyl-phosphonium bromide (3.42 g,
6.0 mmol) and terephthalaldehyde (0.87 g, 6.5 mmol) were reacted
in 120 mL of 50/50 vol% CH2Cl2/H2O in the presence of Na2CO3
(30.0 g, 40 equiv) and tetrabutylammonium iodide (0.30 g, 10%
H3C
H3C
CH3
Br
– 4 e–
CHO
Br
F3C
+
CHO
CH3CN:H2O)
0.1 M LiBF4
Br
H3C
H3C
Br
CH3
F3C
7
8
9
2 e–
H3C
H3C
Br
c.p.e., Pt cathode
CHO
CHO
CH3CN:H2O)
0.1 M LiBF4
Br
H3C
H3C
7
8
Scheme 5. Anodic oxidation of the protected distyrylbenzene at the more electron-deficient double bond.

Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
G3095
of 6. The crude product and a few crystals of I2 were dissolved
in 50 mL xylene and refluxed for three days. GC-MS showed that
the bis[trifluoromethyl)styrylbenzene and 6 were now present as
single stereoisomers, presumably E,E. The mixture was cooled,
extracted with thiosulfate solution, separated, dried over Na2SO4
and evaporated to afford 3.0 g crude product. Purification by flash
chromatography over silica gel beginning with hexanes followed by
2% EtOAc in hexanes as eluent, followed by recrystallization from
EtOH, yielded 600 mg (48% yield) of 6, mp 177–181◦C; IR ν (cm−1):
3031 (w), 2919 (w), 1599)m), 1511 (w), 1466 (w), 1379 (s), 1278 (s),
1119 (s), 999 (w), 952 (s), 941 (s), 887 (s), 850 (s), 836 (m) 802 (w),
stereoisomers of 9. Recrystallization from ether afforded 219 mg
1
(28% yield) of 9 as fine white crystals, mm 164–166◦C; H NMR
(300 MHz, CD3CN): 7.83 (s, 1H), 7.77 (s, 2H), 7.50 (d, J = 8.3 Hz,
2H), 7.27 (J = 8.3 Hz, 2H), 7.22 (s, 2H), 7.03 (s, 1H), 6.93 /6.78
(qAB, J = 12.3 Hz, 2H), 5.75 (d, J = 12.2 Hz, 1H), 5.66 (d, J = 12.2
Hz, 1H), 2.33 (s, 6H). Anal. Calcd for C26H20Br2F6: C, 51.51; H,
3.33. Found: C, 52.17; H, 3.40.
Electrocatalytic oxidation of 6.— Controlled potential anodic ox-
idation of a solution of 58 mg (0.13 mmol) of 6 was carried out in
25 mL of 84:16 v/v acetonitrile/water solution containing 10 mol% of
10c at +0.75 V vs Ag/Ag+ reference. The initial blue solution of the
cation radical of 10c immediately began to stream from the anode, but
the solution quickly began to develop a purple color, suggesting that
it was decomposing to a different substance. Well before 6 had been
consumed, the current passing through the cell had dropped to zero
and the color had become light pink, indicating rapid decomposition
of 10c under the electrolysis conditions. Analysis by GC-MS indi-
cated the presence of two major components: unreacted 6 (48%) and
4-formyl-3ꢀ,5ꢀ-bis[trifluoromethyl]stilbene (4) (36% crude yield of a
mixture of the stilbene aldehydes 4 and 7; 72% yield based upon un-
reacted starting material). Workup and flash chromatography afforded
a mixture of 4 and 7, which was separated by preparative thin layer
chromatography resulted in isolation of pure 4 (24 mg; 50%) and 7
(4.5 mg; 10%). The rapid decrease in current, coupled with the change
in color, suggested that the catalyst was decomposing throughout the
electrolysis. Addition of fresh catalyst to the solution was unsuccess-
ful because decomposition of the catalyst appeared to take place faster
as electrolysis proceeded.
1
682 (s); H NMR (300 MHz, CD3CN, δ): 8.11 (s, 2H),7.85 (s, 1H),
7.59 (s, 4H), 7.44/7.32 (qAB, J = 16.8 Hz, 2H), 7.20 (s, 2H), 7.18
(s, 2H), 6.94 (s, 1H), 2.30 (s, 6H); 13C NMR (CDCl3, ppm): 139.7,
138.4, 138.3, 137.2, 135.3, 132.3, 132.2 (q, J = 131.3 Hz), 129.9,
129.8, 127.8, 127.5, 127.1, 126.3 (q), 125.4, 124.8, 121.8, 120.9 (m),
21.5; MS m/z (% relative intensity): 447 (29), 446 (100); Anal. Calcd
for C26H20F6: C, 69.95; H, 4.51. Found: C, 69.72; H, 4.50.
(E)-4-Formyl-3,5-dimethylstilbene (7).— 3,5-Dimethylbenzyl-
triphenylphosphonium bromide (1.85 g, 4.0 mmol) and terephtha-
laldehyde (0.58 g, 4.3 mmol) were reacted in 80 mL of 50/50 vol%
CH2Cl2/H2O in the presence of Na2CO3 (20.0 g, 0.19 mmol) and
tetrabutylammonium iodide (0.20 g, 0.5 mmol). After 20 hr, 25 mL
of water was added and the organic layer was removed. The aqueous
layer was extracted with 3 15-mL portions of CH2Cl2. The combined
organic layers were dried over MgSO4 and evaporated to afford
a yellow liquid that mostly crystallized upon standing overnight.
Analysis by GC-MS showed a 45:55 mixture of stereoisomers. The
mixture was purified by flash chromatography with toluene as eluent.
A trace amount of I2 was added to the combined product fractions and
the solution stirred overnight at reflux to obtain the pure (E)-isomer in
toluene. The solution was decolorized with NaHSO3 and the solvent
evaporated to afford 0.829 g of tan solid (83% crude yield), which
was recrystallized from 95% EtOH to afford 0.617 g of tan crystals
(62% yield): mp 76–77◦; 1H NMR (400 MHz, CH3CN): 9.96 (s, 1H),
Direct electrochemical oxidation of protected alkene 9.— Methyl-
cellulose (1 gm) was dissolved in 20 mL of hot DMF containing
0.1 M LiBF4. A small amount of the solution was added to the cath-
ode side of the frit and allowed to set to produce gel, thus prevent-
ing mass transit from the two sides of the cell. Alkene 9 (55 mg;
0.09 mmol) was dissolved in 20 mL of 95:5 acetonitrile:water con-
taining 0.1 M containing 0.1 M LiBF4 and added to the the anode side
of the cell. An additional amount of the electrolyte solution was added
to the cathode side of the cell until the liquid levels on both sides of the
cell were approximately equal. A graphite carbon cloth (6 cm2) was
placed in the cathode compartment and a platinum mesh electrode
(2 cm2) was added on the anode side. The solution was purged of
oxygen by passing a slow stream of acetonitrile-saturated nitrogen
through the cell for 20 min. A fixed current of 10 mA was then passed
through the cell until an amount of current equal to 4 Faradays/mol of
9 had passed. The contents of the anode compartment were washed
with 10 mL of brine, then evaporated to dryness. The remaining solids
were extracted with ether, leaving the supporting electrolyte behind.
The ether solution was dried over Na2SO4 and evaporated. Preparative
thin layer chromatographic separation of the residue resulted in isola-
tion of the cleavage product, aldehyde 8 (11 mg, 30% yield) together
with a small amount of debrominated aldehyde (4); the yield is 33%
yield if the isolated 4 is included.
7.87 (d, J = 8.0 Hz, 2H), 7.72 (d, J = 8.0 Hz, 2H), 7.30/7.24 (qAB
,
J = 12.3 Hz, 2H), 7.24 (s, 2H), 6.98 (s, 1H), 2.32 (s, 6H); 13C NMR
(CDCl3): 191.7, 143.8, 138.5, 136.6, 135.4, 132.6, 130.5, 130.4,
127.1, 127.0, 125.0, 21.5; MS m/z (relative intensity): 236 (100), 192
(38), 178 (31).
1,2-Dibromo-1-(4-formylphenyl)-2-(3ꢀ,5ꢀ-dimethylphenyl)ethane
(8).— (E)-4-Formyl-3ꢀ,5ꢀ-dimethylstilbene (7) (100 mg, 0.4 mmol)
was dissolved in CH2Cl2 and cooled to 0◦C. Pyridinium tribromide
(186 mg, 1.3% molar xs) was added and the solution was allowed to
stir overnight while gradually warming to room temperature, during
which time it turned from light orange to yellow. It was washed with
35 mL aq. sodium thiosulfate. The colorless organic layer was washed
with water, dried over Na2SO4, and evaporated to afford 164 mg
(97% yield): mp 168–172◦C; 1H NMR (400 MHz, CH3CN): 10.03 (s,
1H), 7.96 (d, J = 7.8 Hz, 2H), 7.80 (d, J = 7.8 Hz, 2H), 7.25 (s, 2H),
7.06 (s, 1H), 5.85/5.73 (qAB, J = 9.0 Hz, 2H), 2.34 (s, 6H). Anal.
Calcd for C17H16OBr2: C, 51.55; H, 4.07. Found: C, 51.66: H, 3.90.
(E)-1-(1,2-Dibromo-2–3,5-dimethylphenyl)ethyl-4-(3ꢀ5ꢀ-bis[tri-
fluoromethyl]styryl)-benzene (9).— Dibromide 8 (ca. 1.3 mmol) was
dissolved in 15 mL CH2Cl2 and 15 mL was added. Na2CO3 (5.5 g)
and TBAI (335 mg) were added and stirred to dissolve. Phosphonium
bromide 3 was added and the mixture was stirred vigorously for
two days. the orange solution was then transferred to a separatory
funnel with 40 mL CH2Cl2, decolorized by washing with 15 mL aq.
sodium thiosulfate, and washed with H2O. The organic layer was
separated, the aqueous layer was extracted with CH2Cl2, and the
combined organic layers dried over Na2SO4. Evaporation afforded
900 mg of yellow-orange waxy solid, which was purified by flash
chromatography, beginning with hexane and gradually increasing to
5% EtOAc as eluents. TLC analysis indicated the presence of four
products in the eluate; two were stereoisomers of 6 and the other were
Cathodic debromination of 8.— Dibromoaldehyde (100 mg;
0.25 mmol) was dissolved in 25 mL of 0.1 Bu4NBF4 in acetonitrile in
the cathode compartment of the divided electrochemical cell used in
the preceding experiment. The solution was degassed by passing ni-
trogen through the solution for 20 minutes while stirring magnetically.
Controlled potential electrolysis was carried out at −1.75 V, during
which time the current decreased from 40 mA to less than 5 mA.
Most of the catholyte was evaporated and the residual oily mixture
was filtered through a silica plug using CH2Cl2 to remove most of the
electrolyte. The filtrate was then separated by flash chromatography
over silica using CH2Cl2 as eluent to afford aldehyde 7 (47 mg, 80%
yield), identical with a previously synthesized sample.

G3096
Journal of The Electrochemical Society, 160 (7) G3091-G3096 (2013)
Conclusions As we found in the present study, the higher redox reactivity aimed
for in such substances often means that their cation radicals are corre-
spondingly more reactive toward chemical decomposition. We suspect
that the same may be true of mediators intended for cathodic use. In
fact, as we noted above, Peters et al., have found that metal salen
complexes intended to mediate the reduction of alkyl halides often
are destroyed by reaction with the free radical products of the electron
transfer that they mediate.14
These experiments demonstrate the feasibility of effecting elec-
trochemical oxidation at the site of higher oxidation potential of a
substance containing two oxidizable sites by first protecting the lower
potential site. The site of lower potential, the dimethylbenzene end of
the molecule, was shown to be more electron rich by spectroscopic
and computational methods. Because it is more electron rich, the dou-
ble bond nearer the dimethylbenzene group reacts selectively with
pyridinium bromide, protecting it from anodic transformation, leav-
ing the remaining double bond as the only site susceptible to anodic
cleavage. On the other hand, anodic cleavage was only selective by a
factor of 5:1.
Acknowledgment
Financial support for this research was provided by Wesleyan Uni-
versity. Computations on compound 6 were carried on the Wesleyan
High Performance Computer Cluster, financed in part by the National
Science Foundation.
Searches for useful electrocatalysts (more properly, electron trans-
fer mediators) frequently begin with cyclic voltammetric studies. Cri-
teria often applied in such studies are:
1. The mediator must have a potential lower than that of the substrate
of interest, but not too much lower: a difference of 400–600 mV
is often quoted.15
References
2. The mediator must exhibit reversible cyclic voltammetric
behavior.
3. Electron transfer from the substrate (for oxidations, the cation
radical of the mediator must exchange more or less rapidly with
the substrate).
1. E. Steckhan, J. Am. Chem. Soc., 100, 3526 (1978).
2. E. Steckhan and H. J. Scha¨fer, Angew. Chem. Int. Ed. Engl., 13, 472 (1974).
3. U.-St. Ba¨umer and H. J. Scha¨fer, Electrochim. Acta, 42, 1971 (1996).
4. R. L. Merzel and A. J. Fry, J. Electrochem. Soc., 159, G117 (2012).
5. X. Wu, A. P. Davis, and A. J. Fry, Org. Lett., 9, 5633 (2007).
6. S. M. Halas and K. Okyne, Electrochim. Acta, 48, 1837 (2003).
7. A. E. Reed, R. B. Weinstock, and F. Weinhold, J. Chem. Phys., 83, 735 (1985).
8. U. Husstedt and H. J. Schafer, Synthesis, 964 (1979).
9. W. Schmidt and E. Steckhan, Chem. Ber., 113, 577 (1980).
10. L. Eberson and B. Olofsson, Acta Chem. Scand., 23, 2355 (1969).
11. N. M. Patil, A. A. Kelkar, and R. V. Chaudhari, J. Mol. Catal. A Chem., 223, 45
(2004).
12. C. L. Perrin and G. A. Skinner, J. Am. Chem. Soc., 93, 3389 (1971).
13. X. Wu, A. P. Davis, P. C. Lambert, K. L. Steffen, O. Toy, and A. J. Fry, Tetrahedron,
65, 2408 (2009).
14. D. M. Goken, D. G. Peters, J. A. Karty, and J. P. Reilly, J. Electroanal. Chem., 564,
123 (2004).
However, a point brought out clearly by these studies is the fact
that a mediator that satisfies all of these criteria may still not be ad-
equate if it is unstable electrochemically on the longer time scale
of a preparative electrolysis. A less obvious problem, in our expe-
rience, takes place when one synthesizes a redox catalyst to have
the high reactivity toward substrates of high oxidation potential. For
example, we recently reported the synthesis of several high redox
potential mediators5,13 intended for catalyzing the even higher oxida-
tion potentials we encountered in a series of stilbenes bearing several
electron-withdrawing groups.5,6
15. E. Steckhan, Top. Curr. Chem., 142, 1 (1987).
16. R. I. Walter, J. Amer. Chem. Soc., 77, 5999 (1955).