Reactivity of the electrogenerated O2·-/CO2 system towards alcohols

Article abstract of DOI:10.1002/1099-0690(200105)2001:9<1689::AID-EJOC1689>3.0.CO;2-F

The reactivity towards alcohols of the system obtained from the reduction of dioxygen carried out in aprotic dipolar solvents and in the presence of carbon dioxide has been investigated. This reagent is able to convert primary and secondary alcohols into the corresponding aldehydes and ketones. The optimization of the experimental conditions as well as the advantages and the limitations of the method are reported. Wiley-VCH Verlag GmbH, 2001.

Full text of DOI:10.1002/1099-0690(200105)2001:9<1689::AID-EJOC1689>3.0.CO;2-F

FULL PAPER
Reactivity of the Electrogenerated O /CO System Towards Alcohols
·
؊
2
2
Maria Antonietta Casadei[a]
Keywords: Superoxide / Carbon dioxide / Alcohols / Oxidation / Electrochemistry
The reactivity towards alcohols of the system obtained from
the reduction of dioxygen carried out in aprotic dipolar solv-
ents and in the presence of carbon dioxide has been investig-
ated. This reagent is able to convert primary and secondary
alcohols into the corresponding aldehydes and ketones. The
optimization of the experimental conditions as well as the
advantages and the limitations of the method are reported.
Introduction
Previous studies from our group have shown that the sys-
tem obtained by one-electron reduction of dioxygen per-
formed in aprotic dipolar solvents and in the presence of
carbon dioxide is able to carboxylate different types of sub-
[
1]
strates. In particular, primary and secondary alcohols
bearing a good leaving group at the α or β position are
converted into cyclic carbonates, whereas unsubstituted
primary and secondary alcohols are transformed into the
corresponding linear carbonate after addition of an alkylat-
[
1b]
ing agent.
In the course of experiments carried out on
benzyl alcohol in order to optimize the reaction conditions,
benzaldehyde was detected in the electrolyzed mixture in an
amount roughly proportional to the concentration of carb-
oxylating reagent and reaction time.[ In the literature it
1b]
[2]
has been reported that the chemical system KO /CO /18-
2
2
crown-6 oxidizes olefins to oxiranes and sulfides to sulfox-
ides. Moreover, it is known that the electrochemically gener-
ated superoxide anion is able to convert primary and sec-
ondary alcohols into the corresponding carboxylic acids
and ketones, respectively.[ Therefore, the possibility to ox-
idize selectively primary alcohols to aldehydes by using a
safe reagent such as the electrogenerated superoxide/carbon
dioxide system seemed interesting, taking into account that
the conversion of alcohols to aldehydes is usually per-
formed with chromium complexes,[ which cannot be con-
sidered an environmentally friendly reagent. On this basis,
3]
Scheme 1
4]
function of the elapsed time before the addition of EtI are
reported in Table 1 (entries 1Ϫ5).
The maximum yield of benzyl ethyl carbonate is obtained
for reaction times Յ1 h. At the same time, the yield of
benzaldehyde 2a undergoes a moderate increase (from 6 to
·Ϫ
the reactivity of the electrogenerated O /CO system to-
2
2
wards representative primary alcohols was investigated. The
reaction was then extended to a series of secondary alcohols
to verify the limits and scopes of the method (Scheme 1).
1
3%) and significant amounts of starting material remain
in the mixture. Upon increasing the reaction time, the
amounts of benzyl ethyl carbonate and 1a decrease, and
after 24 h these products are no longer detectable in the
Results and Discussion
The nature and the yields of the products arising from solution. In parallel, the yield of 2a increases (up to 78%)
·
Ϫ
the reaction between benzyl alcohol 1a and O /CO as a and ethyl benzoate 3a is also formed. These data suggest
2
2
that the oxidation process is slower than the competitive
Dipartimento di Studi di Chimica e Tecnologia delle Sostanze carboxylation one, although the former becomes prevalent
[
a]
Biologicamente Attive Ϫ Universit a` degli Studi ‘‘La Sapienza’’,
by increasing the time elapsed before the addition of EtI,
possibly favoured by the decomposition of the initially
formed carboxylate anion. Other experiments have been
P. le Aldo Moro 5, 00185 Roma, Italy
Fax: (internat.) ϩ 39-06/4991-3133
E-mail: mariaantonietta.casadei@uniroma1.it
Eur. J. Org. Chem. 2001, 1689Ϫ1693

WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ193X/01/0509Ϫ1689 $ 17.50ϩ.50/0
1689

M. A. Casadei
FULL PAPER
Table 1. Reactivity of the electrogenerated superoxide/carbon dioxide system towards the primary alcohols 1aϪh (CH
TEAP; Hg cathode; Pt anode; E ϭ Ϫ1.0 V; substrate ϭ 2 mmol)
3
CN/0.1 mol·LϪ1
Entry
Substrate
n[a]
Time[b]
Products (yield %)[c]
1
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
1d
1e
1f
1.0
1.0
1.0
1.0
1.0
0.5
1.5
2.0
3 min.
10 min.
1 h
1a (58), 2a (6), PhCH
1a (51), 2a (6), PhCH
2
OCO
OCO
2
Et (25)
Et (32)
2
2
1a (52), 2a (13), PhCH
1a (32), 2a (38), PhCH
2a (78), 3a (8)
2
OCO
2
Et (33)
6 h
2
OCO
2
Et (11), 3a (5)
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
24 h
1a (45), 2a (40)
2a (85), 3a (10)
2a (62), 3a (22)
[d]
1.5
1.5
1.5
1.5
1.5
1.5
1a (54), 2a (35), 3a (2)
10
11
12
13
14
15
16
17
18
19
20
1b (15), 2b (61), 3b (11)
1c (21), 2c (54), 3c (17)
1d (28), 2d (61), 3d (2)
1e (28), 3e (67)
1f (2), 3f (97)
1f
0.25
0.5
1f (66), 3f (8)
1f
1f (55), 3f (19)
1f (48), 3f (25)
1f (17), 3f (70)
1g (97)
1f
0.75
1.0
1f
1g
1h
1.5
1.5
1h (91)
[
a]
Number of Faraday/mol of substrate. Ϫ Elapsed time between the current flow break and the addition of EtI. Ϫ [c] HPLC (entries
[b]
[d]
1
Ϫ19) or GC (entry 20) analysis. Ϫ DMF was employed as solvent.
performed in order to test the influence of both the current two reduction peaks, the first of which (at Ϫ0.74 V) is re-
·
Ϫ
amount and the nature of the solvent on the oxidation reac- lated to the one-electron reduction of O to O . It is gener-
2
2
tion. The yield of 2a increases by increasing the amount of ally accepted that differences of the reduction potential
current flowing through the cell and reaches a maximum values as small as those observed in the case of 2e,f and O2
value (85%) at 1.5F/mol of substrate (Table 1, entries 5Ϫ8). allow a fast homogeneous electron transfer to occur.[6]
On the other hand, if N,N-dimethylformamide (DMF) is Based on such a process, a mechanism has been proposed
substituted for CH CN as solvent under otherwise identical to explain the formation of carboxylates during the reduc-
3
conditions, the yield of 2a sharply decreases (Table 1, tion of α-diketones in the presence of dioxygen in the catho-
7]
entry 9).
lyte.[ On the assumption that superoxide anion can be re-
[8]
Under optimized conditions of time, current and solvent, leased in the solution by the oxidizing agent, a similar
the primary alcohols 1bϪh were subjected to the oxidation mechanism can be applied to the present case. It can ex-
process. Alcohols 1bϪd are converted into the correspond- plain the formation of the acids that always goes with the
ing aldehydes 2bϪd in good yield, even if the respective ac- aldehydes and that become prevalent when the difference
ids are also present in the reaction mixtures (Table 1, entries between the potential values allows a fast electron transfer.
·Ϫ
1
0Ϫ12). Surprisingly, both 4-nitrobenzyl alcohol 1e and 3-
Alcohols 1g,h are stable towards the O /CO system
2 2
nitrobenzyl alcohol 1f yield almost quantitatively carb- (Table 1, entries 19,20). It is likely that the acidity of the
oxylates 3e and 3f, respectively (Table 1, entries 13,14). Al- substrate plays a role in the oxidation process whose first
dehyde 2f is never detectable in the reaction mixture, even step, according to the literature,[ involves a deprotonation
after decreasing the current amount (Table 1, entries reaction. Previous studies report that electrogenerated su-
3]
1
5Ϫ18). Although the formation of aldehydic intermediates peroxide anion oxidizes primary alcohols to the corres-
in the sequence from alcohols to acids can be easily put ponding carboxylic acids independently of their nature.[
forward, separate experiments have been carried out to con- In order to compare the reactivity of the electrogenerated
firm this hypothesis.[ Under reaction conditions identical superoxide anion with the system under study in the same
3a]
5]
·
Ϫ
to those used in the oxidation of the alcohols, O /CO con- experimental conditions, O was reduced at a mercury cath-
2
2
2
verts 2f into 3f quantitatively but converts 2a into 3a only ode and 1a added to the solution. As expected, HPLC ana-
to a limited extent (13%). In order to understand this beha- lysis showed that 3a is the main oxidation product (see Ex-
viour, cyclic voltammetry measurements have been carried perimental Section). If compared with the data reported in
out for solutions of aldehydes 2aϪf in CH CN/0.1  TEAP Table 1 (entry 9), these data indicate that the reactivity of
3
·
Ϫ
at a Hg cathode. The voltammograms of 2aϪc show only the O /CO system as an oxidant is quite different from
2
2
one reduction peak (at Ϫ1.84, Ϫ1.73 and Ϫ1.52 V, respect- that of O· alone. In contrast with the behaviour of this
Ϫ
2
ively), whereas that of 2d presents two reduction peaks at latter (see before), the system under study oxidizes only
Ϫ1.78 and Ϫ1.93 V. Finally, in the voltammograms of 2e primary alcohols of the benzylic or allylic types. Assuming
and 2f three peaks are present at Ϫ0.70, Ϫ1.12, Ϫ2.25 V that their reduction potentials are not too close to that of
and Ϫ1.00, Ϫ1.78, Ϫ2.25 V, respectively. Under identical dioxygen, the corresponding aldehydes are formed in good
experimental conditions, the voltammogram of O shows yields.
2
1690
Eur. J. Org. Chem. 2001, 1689Ϫ1693

Reactivity of the Electrogenerated O·Ϫ
/CO System Towards Alcohols
2 2
FULL PAPER
·
Ϫ
The reactivity of the secondary alcohols 4aϪd towards O /CO were allowed to react under otherwise identical
2
2
·
Ϫ
O /CO has also been tested. At first, 1-phenylethanol (4a) conditions: in this case compound 6 and ethyl benzoate
2
2
was checked with solutions containing different concentra- were isolated (Table 3, entry 3). The formation of 6 from 5d
tions of the reagent. The results of the analyses of the reac- involves a reduction step to the hydroxy ketone 4d, which
tion mixtures are reported in Table 2 (entries 1Ϫ8) and can occur through an electron transfer in solution between
·
Ϫ
[5Ϫ7]
show 1.5F/mol of alcohol as the optimum current amount. O /CO and 5d.
The resulting radical anion evolves to
2
2
Once again, if DMF is employed as solvent instead of ace- 6, probably through a complex, multistep sequence requir-
tonitrile the ketone yield decreases (Table 2, entry 9). Under ing protonation and acylation. To verify that a protonation
·
Ϫ
these conditions, alcohol 4b yields the corresponding ke- step occurs (probably from the solvent), O /CO was gen-
2
2
tone 5b in high yield (Table 2, entry 10), whereas the con- erated in DMF and 5d added to the solution. After 24 h,
version of cyclohexanol 4c into ketone 5c takes place in only 5d and ethyl benzoate were present in the reaction mix-
lower yields, significant amounts of starting 4c being reco- ture (Table 3, entry 4). However, if a proton donor [10%
vered from the reaction mixture (Table 2, entry 11).
CH CN or CH CH(CO Et) in equimolar amounts with
3 3 2 2
·Ϫ
respect to the substrate] was added to O /CO and 5d in
2
2
Table 2. Reactivity of the electrogenerated superoxide/carbon diox-
DMF, compound 6 was formed in concentrations propor-
3
ide system towards the secondary alcohols 4aϪc (CH CN/0.1
Ϫ1
mol·L TEAP; Hg cathode; Pt anode; E ϭ Ϫ1.0 V; reaction tional to the acid strength (Table 3, entries 5,6), in agree-
time ϭ 24 h; substrate ϭ 2 mmol)
ment with the hypothesis of the solvent involvement in the
formation of 6.
Entry
Substrate
n[a]
Products (yield %)[b]
1
2
3
4
5
6
7
8
9
4a
4a
4a
4a
4a
4a
4a
4a
4a
4b
4c
0.2
0.4
0.6
0.8
1.0
1.2
1.5
2.0
5a (31)
5a (63)
5a (71)
5a (84)
5a (89)
5a (95)
5a (97)
5a (97)
5a (41)
5b (90)
5c (31)[
[c]
1.5
1.5
1.5
1
1
0
1
d]
[
a]
[b]
Number of Faraday/mol of substrate. Ϫ HPLC (entries 1Ϫ10)
[c]
or GC (entry 11) analysis. Ϫ DMF was employed as solvent. Ϫ
Scheme 2
[
d]
4c (66%) is also present.
It should be pointed out that 2-hydroxy-1,2-diphenyl-
ethanone (4d) showed a different behaviour. From the
workup of the reaction mixture, 1,2-diphenylethan-2-one-1- Conclusion
yl benzoate (6) and benzoic acid were isolated together with
minor amounts of benzil (5d) (Scheme 2, Table 3, entry 1).
The electrogenerated superoxide-carbon dioxide system
However, if DMF is employed as solvent instead of aceto- behaves as a selective oxidizing agent. Starting from prim-
·
Ϫ
nitrile the reaction of O /CO with 4d gives rise to 5d and ary alcohols of the benzylic or allylic types this system al-
2
2
benzoic acid (Table 3, entry 2). The nature of the products lows the formation of aldehydes as the main products of
obtained in acetonitrile suggest a reaction pathway invol- the reaction, provided that the reduction potential of the
ving a partial oxidation of 4d to 5d and a further trans- aldehyde is not too close to that of dioxygen. Benzylic sec-
formation of this latter to an acylating species capable of ondary alcohols are converted into the corresponding ke-
converting 4d into 6. Benzoic acid is probably formed as a tones in excellent yield. Other types of alcohols are stable
secondary product both from the ester synthesis and the or only partially oxidized. Within the limits of reactivity
·
Ϫ
acylating agent decomposition. To confirm this hypothesis, discussed above, the use of O /CO as an oxidant presents
2
2
and with the aim of isolating the acylating species, 5d and some advantages: compared with the classical methods
Table 3. Reactivity of the electrogenerated superoxide/carbon dioxide system towards 4d and 5d (Hg cathode; Pt anode; supporting
Ϫ1
electrolyte ϭ 0.1 mol·L TEAP; E ϭ Ϫ1.0 V; n ϭ 1.5 F/mol of substrate; reaction time ϭ 24 h; substrate ϭ 2 mmol)
Entry
Substrate
Solvent
Products (yield %)[a][b]
1
2
3
4
5
6
4d
4d
5d
5d
5d
5d
CH
3
CN
CN
5d (4), 6 (38), PhCO
5d (83), PhCO H (16)
6 (42), PhCO Et (57)
5d (86), PhCO Et (27)
5d (69), 6 (5), PhCO Et (29)
5d (17), 6 (35), PhCO
2
H (58)
DMF
2
CH
DMF
DMF/CH
DMF/CH
3
2
2
CN[
c]
3
3
2
Et) [
d]
Et (67)
CH(CO
2
2
2
[
a]
mol of product/mol of substrate) ϫ 100. Ϫ [b] HPLC analysis. Ϫ [c] 10% (v/v). Ϫ [d] Equimolar with the substrate.
(
Eur. J. Org. Chem. 2001, 1689Ϫ1693
1691

M. A. Casadei
FULL PAPER
based on chromium complexes this procedure does not re- at room temperature for a further 24 h and then analyzed. The
results of the analysis are reported in Table 1 (entry 10).
quire toxic or polluting reagents, and shows a much greater
selectivity, in terms of both substrate and product selectiv-
ity, when compared to superoxide anion only.
Cinnamyl Alcohol (1c): The reaction was carried out as described
for 1b. The results of the HPLC analysis are reported in Table 1
(
entry 11).
4
-Methoxybenzyl Alcohol (1d): The reaction was carried out as de-
Experimental Section
scribed for 1b. The results of the HPLC analysis are reported in
Table 1 (entry 12).
General: The electrochemical apparatus, the cells, and the reference
electrode have already been described.[ The potential values are
reported relative to SCE. Acetonitrile (Riedel-de Ha e¨ n), N,N-di-
methylformamide (DMF; Riedel-de Ha e¨ n) and tetraethylammo-
nium perchlorate (TEAP; Fluka) were purified as previously de-
9]
4-Nitrobenzyl Alcohol (1e): The reaction was carried out as de-
scribed for 1b. The results of the HPLC analysis are reported in
Table 1 (entry 13).
[
10]
3-Nitrobenzyl Alcohol (1f): The reaction was carried out as de-
scribed for 1b. The results of the HPLC analysis are reported in
Table 1 (entry 14). A set of experiments was carried out using dif-
ferent amount of current. The current amount values together with
the results of the HPLC analyses performed on these solutions are
reported in Table 1 (entries 15Ϫ18).
scribed.
IR, NMR, HPLC, GC and melting point apparatus
were already described.[
10] 1
H NMR spectra were recorded as solu-
3 4
tions in CDCl , using Me Si as internal standard. HPLC analyses
were carried out using a Merck Hibar LiChrocart (250Ϫ4, 5 µm)
RP-18 column. A CH CN/H O mixture in a linear gradient from
5:65 to absolute CH CN over 20 min., followed by an isocratic
step at this composition for 5 min., was used as eluent. A CH CN/
0.01 mol L mixture in a linear gradient from 0% to 100%
CN over 15 min., followed by an isocratic step at this com-
3
2
3
3
3
3
-Phenyl-1-propanol (1g): The reaction was carried out as described
for 1b. The results of the HPLC analysis are reported in Table 1
entry 19).
Ϫ1
H
3
PO
4
of CH
3
(
position for 5 min., was used for the analysis of benzoic acid. The
Ϫ1
flow rate was always 1 mL min. . GC analyses were carried out
1
-Heptanol (1h): The reaction was carried out as described for 1b.
using a Supelco Porapack PS 100 packed column (6 feet ϫ 1/8
inch) in the temperature range 50Ϫ200 °C. All the quantitative ana-
lyses were performed with the internal standard method. In the
HPLC analyses N-benzylpropanamide was used in the case of
The results of the GC analysis are reported in Table 1 (entry 20).
1-Phenyl-1-ethanol (4a): A set of experiments was carried out in
order to define the optimum current amount to employ for the
secondary alcohols. After the flow of 120 C a sample (5 mL) was
1
2
2
aϪb and 1d, N-phenylpropanamide in the case of 1c, 1f, 2aϪd,
f, 3bϪd, 3f, 4d, 5b, 5d and 6, N-phenylacetamide in the case of 1e, taken out and added to 4a (0.25 mmol). The mixture was stirred at
e, 3e and 5a, N-benzylacetamide in the case of 1g, 2g, 3g and 4a, room temperature for 24 h; 2 mL of 0.01 mol·LϪ1
H
3
PO were ad-
4
and N-benzyl-2-methylacrylamide in the case of 3a and 4b. Di-
methyl carbonate was used as internal standard for the GC ana-
lyses.
ded to the solution, which was then analyzed. Other samples were
taken out at different current amounts. The current amount values
together with the results of the HPLC analyses carried out on these
solutions are reported in Table 2 (entries 1Ϫ8). Under the best ex-
perimental conditions of current, DMF was employed as solvent.
The results of the HPLC analyses are reported in Table 2 (entry 9).
Electrochemistry: The controlled-potential electrolyses were carried
Ϫ1
3 2 2
out in CH CN/0.1 mol·L TEAP (50 mL) where O and CO were
simultaneously bubbling. After the flow of current was stopped, 40
mL of this solution were withdrawn and added to the substrate
Diphenylmethanol (4b): The electrolysis was stopped after the flow
(
2 mmol). The mixture was stirred at room temperature for 24 h.
of 360 C, corresponding to 1.5F/mol of alcohol. After 24 h, 15 mL
Ϫ1
of 0.01 mol·L
3 4
H PO were added to the solution, which was then
Benzyl Alcohol (1a): After the flow of 240 C, five aliquots of 5 mL
each were withdrawn and added to 1a (0.25 mmol for each aliquot,
in order to obtain 1.0F/mol of alcohol). At different times, a five-
analyzed. The results of the HPLC analyses are reported in Table 2
entry 10).
(
fold molar excess of EtI was added, the mixtures were stirred at Cyclohexanol (4c): The reaction was carried out as described for 4b.
room temperature for 24 h, and then analyzed. The results of
HPLC analyses carried out on these solutions are reported in
Table 1 (entries 1Ϫ5). Another set of experiments was planned in
order to define the optimum current amount to employ. After the
flow of 120 C, a sample (5 mL) was taken out and added to 1a
The results of the GC analyses are reported in Table 2 (entries 11).
2
-Hydroxy-1,2-diphenylethanone (4d): The electrolysis was stopped
after the flow of 360 C, corresponding to 1.5F/mol of alcohol.
After 24 h, a sample (2 mL) was taken out for the HPLC analysis.
The remaining solution was diluted with water (150 mL) and ex-
(0.25 mmol, in order to obtain 0.5F/mol of alcohol). Other samples
tracted with Et
were dried (Na SO
pressure. On the basis of IR and H NMR spectra the residue
2
O (3 ϫ 50 mL). The combined organic extracts
were taken out at different current amounts and treated as above.
The solutions were stirred at room temperature 24 h, a fivefold
molar excess of EtI was added and after a further 24 h analyzed
by HPLC. The current amount values together with the results of
the analyses carried out on each solution are reported in Table 1
2
4
) and the solvent was evaporated under reduced
1
(
(
0.21 g) was identified as 1,2-diphenylethan-1-one-2-yl benzoate
6): m.p. 120Ϫ121 °C (ref.[ m.p.119 °C). The aqueous solution
11]
was acidified (HCl) and extracted with CHCl
3
(3 ϫ 50 mL). The
4
) and the solvent
(entries 6Ϫ8). Under the best experimental conditions of current
combined organic extracts were dried (Na SO
2
and reaction time, DMF was employed as solvent. The results of
the HPLC analyses are reported in Table 1 (entry 9).
was removed under reduced pressure. The residue (0.12 g) was ben-
zoic acid. The results of the HPLC analysis are reported in Table 3
(entry 1). Under otherwise identical conditions, the reaction was
even carried out with DMF as solvent. The results of the analysis
2-Furfuryl Alcohol (1b): The electrolysis was stopped after the flow
of 360 C, corresponding to 1.5F/mol of alcohol. After 24 h, a five-
fold molar excess of EtI was added to the mixture that was stirred are reported in Table 3 (entry 2).
692 Eur. J. Org. Chem. 2001, 1689Ϫ1693
1

Reactivity of the Electrogenerated O·Ϫ
/CO System Towards Alcohols
2 2
FULL PAPER
Benzil (5d): The reaction was carried out as described for 1b. The
results of the HPLC analysis are reported in Table 3 (entry 3). The
[1] [1a]
M. A. Casadei, S. Cesa, F. Micheletti Moracci, A. Inesi, M.
[1b]
Feroci, J. Org. Chem. 1996, 61, 380Ϫ383. Ϫ
S. Cesa, M. Feroci, A. Inesi, L. Rossi, F. Micheletti Moracci,
M. A. Casadei,
reaction was also performed with DMF, a DMF/CH
3
CN mixture
[1c]
Tetrahedron 1997, 53, 167Ϫ176. Ϫ
eletti Moracci, G. Zappia, A. Inesi, L. Rossi, J. Org. Chem.
997, 62, 6754Ϫ6759.
T. Nagano, H. Yamamoto, M. Hirobe, J. Am. Chem. Soc. 1990,
12, 3529Ϫ3535 and references cited therein.
M. A. Casadei, F. Mich-
(
9:1, v/v), and DMF containing CH CH(CO Et) (2 mmol) as pro-
3
2
2
ton donor. The results of the HPLC analyses are reported in
Table 3 (entries 4Ϫ6).
1
[2]
1
Reaction Between O·
؊
and 1a: The controlled-potential electrolyses
2
[
[
3] [3a]
[3b]
M. Sing, R. A. Misra, Synthesis 1989, 403Ϫ404. Ϫ
M.
Ϫ1
were carried out at Ϫ1.0 V in DMF/0.1 mol·L TEAP (50 mL)
where O was bubbling. The electrolyses were stopped after the
Singh, K. N. Singh, S. Dwivedi, R. A. Misra, Synthesis 1991,
2
291Ϫ293.
4] [4a]
flow of 360 C, corresponding to 1.5F/mol of alcohol. At the end
of the electrolysis 40 mL of the solution was withdrawn and added
to 1a (2 mmol). The mixture was stirred at room temperature for
J. C. Collins, W. W. Hess, F. J. Frank, Tetrahedron Lett.
[4b]
1968, 3363Ϫ3366. Ϫ
E. J. Corey, J. W. Suggs, Tetrahedron
[4c]
Lett. 1975, 2647Ϫ2650. Ϫ
R. W. Ratcliffe, R. J. Rodehorst,
[4d]
J. Org. Chem. 1970, 35, 4000Ϫ4003. Ϫ
J. I. De Graw, J. O.
24 h. EtI (10 mmol) was added to the solution, which was stirred
[4e]
Rodin, J. Org. Chem. 1971, 36, 2902Ϫ2903. Ϫ
W. W. Hess, Org. Synth. 1972, 52, 5Ϫ10.
The author wishes to thank a referee who suggested these ex-
periments.
J. Simonet, in M. M. Baizer, H. Lund, Organic Electrochem-
istry, Marcel Dekker, New York, 1983.
K. Boujel, J. Simonet, Tetrahedron Lett. 1979, 1063Ϫ1066.
The nature of the superoxide-carbon dioxide system is not yet
well established: cf. reference 2, and: J. L. Roberts, S. Calder-
wood, D. T. Sawyer, J. Am. Chem. Soc. 1984, 106, 4667Ϫ4670.
J. C. Collins,
for a further 24 h and then analyzed. The results of the HPLC
analyses showed the presence of 1a (60%) together with 2a (5%)
and 3a (25%).
[
[
5]
6]
Reaction Between O·
؊
2
/CO
and 2a,f: The reactions were carried out
2
as described for 1b. HPLC analyses showed for the reaction of 2a
the presence of starting 2a (80%) and of 3a (13%) and for 2f the
quantitative formation of 3f.
[7]
[8]
[
9]
M. A. Casadei, S. Cesa, M. Feroci, A. Inesi, New J. Chem.
1999, 23, 433Ϫ436.
M. A. Casadei, A. Gessner, A. Inesi, W. Jugelt, F. Micheletti
Moracci, J. Chem. Soc., Perkin Trans. 1 1992, 2001Ϫ2004.
Acknowledgments
[
[
10]
11]
The author wishes to thank Prof. Franco Micheletti Moracci and
Prof. Achille Inesi for the helpful discussions, and Mrs. Sonia
Renzetti for typing the manuscript. Financial support from CNR
and Universit a` La Sapienza, Roma, Italy are gratefully acknow-
ledged.
G. Schening, A. Hensle, Justus Liebigs Ann. Chem. 1924, 440,
7
2Ϫ88.
Received April 20, 2000
[O00202]
Eur. J. Org. Chem. 2001, 1689Ϫ1693
1693