Waste-free electrochemical oxidation of alcohols in water

Article abstract of DOI:10.1002/adsc.200606199

We describe a new sol-gel molecular electrode made of a thin layer of organosilica doped with the nitroxyl radical TEMPO (2,2,6,6- tetramethylpiperidine-1-oxyl) electrodeposited on the surface of an ITO-coated glass and its employment as a selective and versatile oxidation catalyst in the electrochemical conversion of different alcohols to carbonyl compounds. Environmentally friendly water or a water/acetonitrile mixture buffered with bicarbonate is used as solvent. The electrode is highly stable and it can be reused for a prolonged period of time allowing easy separation from the products.

Full text of DOI:10.1002/adsc.200606199

COMMUNICATIONS
DOI: 10.1002/adsc.200606199
Waste-Free Electrochemical Oxidation of Alcohols in Water
Giovanni Palmisano,a,c Rosaria Ciriminna,a and Mario Pagliaroa,b,*
a
Istituto per lo Studio dei Materiali Nanostrutturati, CNR, via U. La Malfa 153, 90146 Palermo, Italy
Fax : (+39)-091-680-9247; e-mail: mario.pagliaro@ismn.cnr.it
Institute for Scientific Ethics and Methodology, CNR, via U. La Malfa 153, 90146 Palermo, Italy
Present address: Dipartimento di Ingegneria Chimica dei Processi e dei Materiali, Università degli Studi, viale delle
b
c
Scienze, 90128 Palermo, Italy
Received: April 24, 2006; Revised: July 31, 2006; Accepted: August 4, 2006
This article is dedicated to Professor Giuseppe De Rita on the occasion of his 75th birthday
Abstract: We describe a new sol-gel molecular elec-
trode made of a thin layer of organosilica doped
with the nitroxyl radical TEMPO (2,2,6,6-tetrame-
thylpiperidine-1-oxyl) electrodeposited on the sur-
face of an ITO-coated glass and its employment as
a selective and versatile oxidation catalyst in the
electrochemical conversion of different alcohols to
carbonyl compounds. Environmentally friendly
water or a water/acetonitrile mixture buffered with
bicarbonate is used as solvent. The electrode is
highly stable and it can be reused for a prolonged
period of time allowing easy separation from the
products.
tional chemical methods.[4] Accordingly, the use of
water as reaction medium is highly desirable as water
is a readily available, safe and environmentally friend-
ly solvent which is currently finding a number of sur-
prising applications in synthetic chemistry.[5]
Research efforts in the field of alcohol oxidation
have lately resulted in a number of aerobic heteroge-
neous metal catalysts of impressive selective activity,
mostly based on Ru,[6] Pd,[7] and, more recently, bi-
metallic Pd-Au[8] catalysts. These metal/O2 catalytic
systems are all atom-efficient;[9] and yet the ideal goal
is to avoid the use of any chemical oxidants prevent-
ing all risks associated with the manipulation of O2 or
H2O2 in the presence of organic compounds.
Keywords: alcohols; carbonyl compounds; electro-
chemistry; green chemistry; oxidation; TEMPO
In place of metal catalysts mentioned above, stable
di-tertiary alkylnitroxyl radicals such as 2,2,6,6-tetra-
methyl-1-piperidinyloxyl (TEMPO) are increasingly
finding use in industry as oxidation catalysts due to
their high selectivity and pronounced versatility
The design of clean, efficient and simple synthetic (Scheme 1).[10]
routes has become a central issue of chemical re-
search both in industry and in the academy. For exam- the cyclic nitrosonium ion generated in situ by an aux-
ple, in chemical oxidations for economic, ecological iliary oxidant. However, the species can be generated
and legislative reasons the chemical industry will cur-
rently not apply toxic transition metals as stoichio-
metric oxidants and will avoid environmentally unsafe
solvents.[1] Hence, the aerobic oxidation of alcohols to
carbonyl compounds – crucial precursors of drugs,
dyes and fine chemicals – has become a major goal of
contemporary chemical research because older indus-
trial processes made use of toxic heavy metals (nota-
bly Cr and Mn salts) in copious amounts generating
15–20 kg of hazardous heavy-metal waste per kg of
useful product.[2]
In these catalytic oxidations the active species is
Similarly, avoiding the use of carcinogenic and bio-
accumulating chlorinated (or aromatic) hydrocarbon
solvents in which these conversions are carried out is
also urgent[3] since solvents have a considerable life
cycle impact and continue to be used according to an
obsolete resource depletion model typical of tradi- mediated by TEMPO.
Scheme 1. Electrochemical oxidation of primary alcohol
Adv. Synth. Catal. 2006, 348, 2033 – 2037
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2033

Giovanni Palmisano et al.
COMMUNICATIONS
also electrochemically by applying a small electric po-
tential (0.7 V vs. Ag/AgCl) to a solution of nitroxyl
radicals. Hence, recently a number of different alco-
hols were oxidised in good yield and selectivity either
to carbonyl compounds (in organic solvent mixed
with 5% water) or to carboxylic acids in H2O.[11]
Similarly, the electrochemical oxidation of several
alcohols can be heterogeneously carried out either (in
organic solvent) using a graphite electrode coated
with a thin poly(acrylic acid) (PAA) layer grafted
with 4-amino-TEMPO;[12] or (in water) using silica-
entrapped TEMPO coupled to catalytic Br+ anodical-
ly regenerated from bromide.[13]
Now, we report the preparation of a molecular elec-
trode able to afford high yields of commercially
valued carbonyl compounds in water with complete
selectivity and marked stability as shown by repeated
use of the same electrode in consecutive reaction
runs. Indeed, only one and the same electrode was
used in all the experiments discussed in the following.
The electrode is made of a thin film of sol-gel orga-
nosilica doped with nitroxyl radicals deposited on the
surface of an indium-tin oxide (ITO) electrode ac-
cording to a recent method[14] that exploits the pH in-
crease at the electrode surface by reduction of pro-
tons near such a surface [Eq. (1)].
Scheme 2. Synthesis of the TEMPO@DE electrode.
Such proton consumption in the proximity of the
electrodeꢁs surface catalyses the sol-gel polycondensa-
tion of an organosilane such as methyltrimethoxysi-
lane [MTMS, Eq. (2)] and thus ensures the electrode-
position of a thin layer of organosilica [Eq. (3)]:
Figure 1. . TEMPO@DE is obtained by electrodeposition of
a thin layer of organosilica doped with TEMPO upon appli-
cation of ꢀ1.1 V (vs. Ag/AgCl) for 15 min to a solution of
Carrying out this organosilica electrodeposition in
the presence of aminopropyl(trimethoxy)silane deri- suitable organosilanes (left). The electrocatalytic film there-
by obtained selectively converts benzyl alcohol dissolved in
0.2M NaHCO3 to pure benzaldehyde by simply applying a
1.4 V voltage (right).
vatised with the TEMPO moiety (via reductive ami-
nation with 4-oxo-TEMPO), a thin film of TEMPO-
entrapped ORMOSIL (organically modified silicate)
is obtained (Scheme 2 and Figure 1) which shows ex-
traordinary performance in terms of both selective ac- setting the voltage (relative to the reference elec-
tivity and stability. trode) at 1.4 V and in the presence of 0.2M hydrogen
Voltammetry experiments recorded in a single cell carbonate acting both as the electrolyte and as the
(Figure 2) comparing the electrode thereby obtained base needed for proton abstraction (Scheme 1). A
(TEMPO@DE) with the CV of the bare ITO elec- typical CV had the shape shown in Figure 4.
trode (Figure 3) clearly show the powerful electro-
For control, under the same reaction conditions no
chemical activity imparted to the electrodeposited oxidation of benzyl alcohol occurred both at a bare
electrode by the entrapped radical (the negative sign ITO electrode and at an ORMOSIL-modified ITO
of current at positive potential is due to the instru- electrode without entrapped TEMPO even after
mentꢁs software).
200 h. On the other hand, when the protocol was ap-
The scope of the method is illustrated by the range plied to benzyl alcohol using TEMPO@DE, after
of primary, secondary, allylic and benzylic alcohols 180 h reaction was complete (Figure 5) and all the
that are oxidised in high conversion and selectivity benzaldehyde formed – which partly separated from
(Table 1) to form either their corresponding alde- solution due to its low water solubility (0.1gL ꢀ1 at
hydes or ketones. All oxidative runs were carried out 208C) – was recovered from the reaction mixture by
2034
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2033 – 2037

Waste-Free Electrochemical Oxidation of Alcohols in Water
COMMUNICATIONS
Table 1. Conversion of primary and secondary alcohols.[a]
Substrate
Product
Time Conversion Selec-
[h]
[%]
tivity
1-phenyletha- acetophenone
nol
12
98
99
benzyl alco-
hol
benzaldehyde
180
100
88
99
geraniol[b]
cinnamyl al-
cohol
citral
cinnamaldehyde 90
85
87
99
99
[a]
Conditions: 0.5 mmol alcohol, TEMPO@DE (1cm 2),
water (10 mL), NaHCO3 (0.2M), V=1.4 V vs. Ag/AgCl.
Conversion and selectivity determined by GC.
Reaction in H2O/CH3CN (7:3).
[b]
Figure 2. Scheme of the oxidation cell. TEMPO@DE is the
working electrode (WE) while the reference electrode (RE)
is made of Ag/AgCl and the counter electrode (CE) is a
platinum thin wire.
Figure 4. Cyclic voltammetry of sol-gel doped electrode
TEMPO@DE in 0.2M NaHCO3 and 0.05M 1-phenylethanol
at voltage between 0 and 1.7 V (vs. Ag/AgCl).
Figure 3. Cyclic voltammetry of bare ITO and sol-gel doped
electrodes in 0.2M KNO3.
final extraction with ethyl acetate. No traces of benzo-
ic acid were detected (by GC) throughout the reac-
tion using a highly sensitive FID detector connected
to a column for separation of carbonyl compounds
and organic acids.
Figure 5. Benzyl alcohol oxidation in water mediated by the
electrode TEMPO@DE (V=1.4 V vs. Ag/AgCl).
the other hand, its production by the method de-
A brief analysis shows the potential applications of scribed herein starting from benzyl alcohol[16] might
this new technology. Benzaldehyde is industrially ob- be desirable for the fragrance and pharmaceutical in-
tained by as a by-product in the oxidation of toluene dustries where this aromatic aldehyde is employed in
directed towards the production of phenol or directly large amounts as an intermediate in the manufacture
by the catalytic aerobic oxidation of toluene.[15] On of flavours, perfumery chemicals and pharmaceuticals.
Adv. Synth. Catal. 2006, 348, 2033 – 2037
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2035

Giovanni Palmisano et al.
COMMUNICATIONS
The use of water instead than organic solvent and the surface of a similar, markedly alkylated ORMO-
the absence of transition metals and traditional chem- SIL not only is highly hydrophobic but also shows a
ical oxidants would indeed afford a product of unpre- pronounced low degree of hydrophilicity due to de-
cedented purity.
pletion of silanol groups at the organosilica matrix
Despite being highly desirable, electroorganic syn- surface.[21]
theses traditionally have been limited by the low se-
Finally, in contrast with the immobilisation of ni-
lectivity typical of the electrochemical reactivity of troxyl radicals by surface derivatisation of an organic
most conductive surfaces (metals or semi-metals) polymer resulting in a material whose stability was
where reactions take place.[17] Indeed, a few synthetic not reported,[12] sol-gel encapsulation in the inner po-
electrochemical processes are carried out on an indus- rosity of the ORMOSIL film protects the radicals
trial scale including the 340,000 t/year production of from chemical degradation which is due to intermo-
adiponitrile[18] [it is of relevance to this report that the lecular quenching of the radicals tethered at the sur-
success of this process was also due to the use of face.[22] As a result, we were able to use the same cat-
water as reaction solvent and to the relatively low alytic electrode in tens of experiments with different
voltage (3.8 V) employed]; and the production (with- substrates without any treatment between consecutive
out mediator) of aryl aldehydes on a 5000 t/year scale oxidative runs besides sonication with ultrasound in
from toluenes.[19]
distilled water.
On the other hand, functionalised sol-gel electrodes
Clearly, reaction rates that are here limited by the
have a rich and varied electrochemistry due to their electrodeꢁs low area for the experiments described in
vast accessible inner porosity that allows the oxidant this report, need to be improved by using larger elec-
and reducing reactant molecules to diffuse through trodes. Moreover, far less amounts of water can be
the material and eventually to the surface of a con- successfully employed which here was prevented by
ducting electrode that can be modified and tailored to the limited versatility of the reaction cell employed.
selectively target specific applications.[20] For instance, Finally, we are currently investigating the reason why
a major advantage of the sol-gel technique herein de- encapsulation of the TEMPO moiety within a thin
scribed is the ability to prepare electrodes with the layer of methylsilica results in an alteration of the
desired surface hydrophilicity-lipophilicity balance electrochemistry of the dopant TEMPO molecule
(HLB) in order to render them suitable for the oxida- which in solution has an oxidation potential at about
tion of alcohols in water and prevent overoxidation to 0.7 V (vs. Ag/AgCl) and upon entrapment shows a
acids.
relatively high current occurring already at 0.2 V
Hence, whereas in water solution with TEMPO, (Figure 3).
benzyl alcohol is anodically oxidised to benzoic
However, considering both the ease and reproduci-
acid,[11] the use of the hydrophobic sol-gel molecular bility of electrochemical TEMPO-mediated oxida-
electrode TEMPO@DE affords uniquely benzalde- tions[23] and of the sol-gel electrodeposition process
hyde. We make the hypothesis that such a selectivity for the preparation of organosilica films,[14] the find-
is due to the HLB of the organosilica surface which ings described in this report may open the route to
prevents diffusion of the hydrated aldehyde molecules the commercial electrochemical oxidation of alcohols
(gem-diol) to the hydrophobic surface of the elec- in the fine chemicals industry, especially in the case of
trode (Scheme 3). Indeed, supporting this hypothesis, small scale reactions.
Experimental Section
All chemicals were purchased from Aldrich and used with-
out further purification. Ultrapure water (Milli-Q Millipore
System) was used in all the experiments. Cyclic voltammetry
and amperometric curves were performed by using a CH In-
struments (Austin, TX) potentiostat CHI630B Electrochem-
ical Workstation. A standard three-electrode cell was used
for the experiments. The volume of the reaction vessel was
10 mL and an Ag/AgCl electrode in 3M KCl was used as
reference electrode. The counter electrode was a platinum
wire (d=0.5 mm) whereas the working electrodes were
indium-tin oxide (ITO) plates obtained from Delta Technol-
Scheme 3. Alcohols are oxidised at the inner surface of
TEMPO@DE (1 ! 2) but not so the hydrophilic hydrated
aldehydes (3 ! 4) which cannot enter the pores due to the
HLB of the material (see text).
ogies (Stillwater, MN) with a coated surface area of
0.36 cm2. The ITO-coated glasses were washed prior to film
deposition with water, 80% (v/v) ethanol and water after
which they were dried at room temperature for 2 h.
2036
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 2033 – 2037

Waste-Free Electrochemical Oxidation of Alcohols in Water
COMMUNICATIONS
[5] For carrying out reactions in aqueous suspension: S. Nar-
ayan, J. Muldoon, M. G. Finn, V. V. Fokin, H. C. Kolb,
K. B. Sharpless, Angew. Chem. Int. Ed. 2005, 44, 3275.
[6] For a recent review, see: S. Campestrini, R. Ciriminna,
M. Pagliaro, Chem. Soc. Rev. 2005, 347, 825.
General Procedure for the Preparation of the Doped
Electrode
The organosilane precursor solution (solution A) was ob-
tained adding 5 mL of 3-aminopropyl(trimethoxy)silane to a
solution of 1.32 g of 4-oxo-TEMPO (4-oxo-2,2,6,6-tetrame-
thylpiperidine-1-oxyl) and 221.2 mg of NaBH3CN (95%) in
MeOH (18 mL) in a reaction flask kept closed under a ni-
trogen atmosphere which results in the required compound
as described elsewhere.[24] The resulting mixture was stirred
for 48 h, after which the excess of sodium cyanoborohydride
was quenched with 7M HCl (50 mL).
The solution for the electrodeposition of sol-gel film con-
sisted of 0.612 mL of MeTMS (methyltrimethoxysilane),
0.612 mL of solution A, 5 mL of ethanol and 5 mL of a solu-
tion buffered at pH 4 (Fisherbrand) and 0.4M in KNO3.
Upon stirring this solution for 1h at room temperature to
promote hydrolysis in the presence of the ITO electrode, a
constant negative potential of ꢀ1.1 V vs. Ag/AgCl was ap-
plied to the solution for 15 min under fast stirring. The re-
sulting coated electrode was washed with water and absolute
ethanol and then dried at room temperature for 6 h.
[7] G. J. ten Brink, I. W. C. E. Arends, R. A. Sheldon, Sci-
ence 2000, 287, 1636.
[8] D. I. Enache, J. K. Edwards, P. Landon, B. Solsona-
Espriu, A. F. Carley, A. A. Herzing, M. Watanabe, C. J.
Kiely, D. W. Knight, G. J. Hutchings, Science 2006, 311,
362.
[9] B. M. Trost, Angew. Chem. Int. Ed. 1995, 34, 259.
[10] A. E. J. de Nooy, A. C. Besemer, H. van Bekkum, Syn-
thesis 1996, 1153. TEMPO-mediated industrial oxida-
tions are for instance carried out by DSM using poly-
oxometalate/O2 as primary oxidant: R. Ben-Daniel, P.
Alsters, R. Neumann, J. Org. Chem. 2001, 66, 8650.
[11] D. Liaigre, T. Breton, E. M. Belgsir, Electrochem.
Commun. 2005, 7, 312.
[12] a) Y. Kashiwagi, H. Ono, T. Osa, Chem. Lett. 1993, 257;
the method was recently extended to oxidative synthe-
sis yields of enantiopure lactones from optically active
diols by grafting of an optically active cyclic nitroxyl
radical: b) Y. Kashiwagi, F. Kurashima, S. Chiba, J.
Anzai, T. Osa, J. M. Bobbitt, Chem. Commun. 2003,
114.
General Procedure for the Oxidation of Alcohols
Cyclic voltammetry experiments were recorded in 0.2M
KNO3 at potentials between ꢁ0.2 V vs. Ag/AgCl without
stirring. The oxidation of alcohols (0.5 mmol) was carried
out in 10 mL of buffered water (0.2M in NaHCO3) although
for alcohols with low water solubility (geraniol, for instance)
a CH3CN/H2O solution 30:70 (v/v) was employed. Initial
substrate concentration was 0.05M. For all the substrates
the selected potential was 1.4 V and the electrolysis was car-
ried out under fast stirring (900 rpm). Substrate and product
concentrations were assessed by a Shimadzu GC-17 A gas
chromatograph equipped with a Supelcowax 10 capillary
column (30 m, 0.25 mm ID) by the internal standard method
(with previously calculated response factors) using 10 mL
decane as internal standard.
[13] H. Tanaka, Y. Kawakami, K. Goto, M. Kuroboshi, Tet-
rahedron Lett. 2001, 42, 445.
[14] R. Shacham, D. Avnir, D. Mandler, Adv. Mater. 1999,
11, 384.
[15] a) R. V. Chaudhari, V. H. Rane, A. A. Deshmukh, S. S.
Divekar, US Patent Application 20050215827, 2005;
b) F. Wang, J. Xu, X. Li, J. Gao, L. Zhou, R. Ohnishi,
Adv. Synth. Catal. 2005, 347, 1987.
[16] Yet, as one referee rightly remarks, benzyl alcohol that
is here proposed as the starting material for the pro-
duction of benzaldehyde is still almost universally man-
ufactured from toluene by chlorination/hydrolysis, with
the usual environmental problems of traditional chemi-
cal processes that are being addressed by chemical re-
search inspired by the “green chemistry” principles.
[17] Organic Electrochemistry, Vol. 8 in the series: Encyclo-
paedia of Electrochemistry, (Eds.: A. Bard, M. Strat-
mann), Wiley-VCH, Weinheim, 2004.
Acknowledgements
We thank the Quality College del CNR and Fox Petroli
(Pesaro, Italy) for financial support.
[18] H. Lund, J. Electrochem. Soc., 2002, 149, 21 .
[19] H. Pꢂtter, in: Organic Electrochemistry, (Eds.: H.
Lund, O. Hammerich), 4th edn., Marcel Dekker, New
York, 2001.
[20] A. Walcarius, D. Mandler, J. A. Cox, M. Collinson, O.
References
Lev, J. Mater. Chem. 2005, 15, 3663.
[21] A. Fidalgo, R. Ciriminna, L. M. Ilharco, M. Pagliaro,
Chem. Mater. 2005, 17, 6686.
[22] T. Fey, H. Fischer, S. Bachmann, K. Albert, C. Bolm, J.
[1] The potential waste coproduced with active pharma-
ceutical ingredients (APIs) is in the range of 500 mil-
lion to 2 billion kg per year. Even at a nominal disposal
cost of $1per kg, the potential savings just in waste
avoidance are significant faced to the pharmaceutical
industry annual sales (almost $500 billion in 2003):
B. W. Cue, Chem. Eng. News 2005, 83 (39), 46.
Org. Chem. 2001, 66, 8254.
[23] Partly immobilized TEMPO in similar oxidations was
described in: E. M. Belgsir, H. J. Schꢃfer, Electrochem.
Commun. 2001, 3, 32. In these oxidations, hydrogen
carbonate buffer electrolyte can be completely re-
moved by a cation-exchange resin membrane: M. Schꢃ-
mann, H. J. Schꢃfer, Eur. J. Org. Chem. 2003, 351.
[24] R. Ciriminna, J. Blum, D. Avnir, M. Pagliaro, Chem.
Commun. 2000, 1441.
[2] R. A. Sheldon, I. W. C. E. Arends, G. J. ten Brink, A.
Dijksman, Acc. Chem. Res. 2002, 35, 774.
[3] J. M. DeSimone, Science 2002, 297, 799.
[4] S. J. Broadwater, S. L. Roth, K. E. Price, M. Kobaslija,
ˇ
D. Tyler McQuade, Org. Biomol. Chem. 2005, 3, 2899.
Adv. Synth. Catal. 2006, 348, 2033 – 2037
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
2037