Selective electrocatalytic oxidation of sorbitol to fructose and sorbose

Article abstract of DOI:10.1002/cssc.201402880

A new electrocatalytic method for the selective electrochemical oxidation of sorbitol to fructose and sorbose is demonstrated by using a platinum electrode promoted by p-block metal atoms. By the studying a range of C4, C5 and C6 polyols, it is found that the promoter interferes with the stereochemistry of the polyol and thereby modifies its reactivity.

Full text of DOI:10.1002/cssc.201402880

CHEMSUSCHEM
COMMUNICATIONS
DOI: 10.1002/cssc.201402880
Selective Electrocatalytic Oxidation of Sorbitol to Fructose
and Sorbose
Youngkook Kwon,[a] Ed de Jong,[b] Jan Kees van der Waal,[b] and Marc T. M. Koper*[a]
A new electrocatalytic method for the selective electrochemi-
cal oxidation of sorbitol to fructose and sorbose is demonstrat-
ed by using a platinum electrode promoted by p-block metal
atoms. By the studying a range of C4, C5 and C6 polyols, it is
found that the promoter interferes with the stereochemistry of
the polyol and thereby modifies its reactivity.
cesses for the direct selective oxidation of sorbitol to fructose
or sorbose are enzyme-catalyzed (by sorbitol dehydrogenase)
or obtained by a noncatalytic chemical route reported many
years ago.[8]
In this Communication, we report a new electrocatalytic
method for the selective oxidation of sorbitol to ketose iso-
mers using a platinum catalyst in combination with a p-block
metal promoter. Most previous works in electrochemistry have
studied sorbitol oxidation on platinum as a potential anode re-
action in fuel cells.[9] In situ infrared studies have reported that
COads poisons the platinum surface and that glucose, glucono-
d-lactone, gluconic acid, and CO2 are main non-adsorbed prod-
ucts.[10] A prolonged electrolysis study confirmed that glucose
is the predominant first oxidation product over the entire po-
tential range.[11] Here, we show that with suitable modification
of the platinum surface, sorbitol can be oxidized to four differ-
ent aldose/ketose intermediates/products: the aldoses d-glu-
cose and l-gulose via primary alcohol oxidation and the keto-
ses d-fructose and l-sorbose via secondary alcohol oxidation,
as illustrated in Scheme 1.
Given the depletion of oil resources and the continuously in-
creasing demand for energy, the use of efficient and environ-
mentally benign technologies for the utilization of renewable
resources such as biomass has become highly topical in recent
years.[1] Sorbitol is considered as one of the 12 potential bio-
mass-based platform chemicals as identified by the Depart-
ment of Energy in the USA.[2] Sorbitol can be obtained from
the hydrogenation of glucose,[1a,3] one of the main products of
cellulose hydrolysis, and may be further processed to com-
pounds such as alkanes, methanol, and hydrogen by aqueous-
phase reforming.[1b,4] Sorbitol is also an interesting starting
molecule for isosorbide, both as a building block for the poly-
mer industry (e.g., polyethylene isosorbate terephthalate, PEIT)
or as starting point for green solvents and fuel additives.[5] Two
particularly interesting selective oxidation products of sorbitol
are fructose and sorbose, the ketose isomers of the corre-
sponding aldoses glucose and gulose. Fructose is an important
starting material for the production of furanics, including hy-
droxymethylfurfural (HMF) and 2,5-furandicarboxylic acid
(FDCA), which are green building blocks for a large variety of
products including plastics, fuel additives, as well as fine chem-
icals.[6] It is well-documented that fructose is the preferred
starting material over glucose in dehydration reactions.[6] Inter-
estingly, the activity and selectivity of the different ketoses to-
wards HMF in acid-catalyzed dehydration is different, but the
difference between fructose and sorbose is minor.[7]
Industrially, fructose is produced from glucose by a xylose
isomerase enzyme. However, this process requires a highly
pure glucose source (i.e., starch) to avoid deactivation of the
enzyme. To the best of our knowledge, the only known pro-
Scheme 1. Possible d-sorbitol oxidation pathways.
[a] Dr. Y. Kwon, Prof. Dr. M. T. M. Koper
Leiden Institute of Chemistry
Leiden University
PO Box 9502, 2300 RA Leiden (The Netherlands)
Fax: (+31)71-5274451
E-mail: m.koper@chem.leidenuniv.nl
Homepage: http://casc.lic.leidenuniv.nl
Figure 1 shows the linear-sweep voltammogram and corre-
sponding product distribution of sorbitol oxidation on
a carbon-supported platinum catalyst (“Pt/C”) obtained at
1 mVsÀ1. The online product analysis was performed using our
combined voltammetry–HPLC setup.[12] As can be seen from
the product distribution, sorbitol is predominantly oxidized
through the primary alcohol oxidation pathway to generate
glucose and gulose in a 1:1 ratio at low potential ranges and
[b] Dr. E. de Jong, Dr. J. K. v. d. . Waal
Avantium Chemicals
Zekeringstraat 29, 1014 BV Amsterdam (The Netherlands)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201402880.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
&
1
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Figure 2. Sorbitol oxidation (0.1m) in 0.5m H2SO4 on a Pt/C electrode in bis-
muth-saturated solution and a antimony-modified Pt/C (irrev., qSb =90%):
(a) current density profile at scan rate of 1 mVsÀ1, (b) concentration changes
of reaction products, and (c) selectivity (%) of products as a function of po-
tential.
hance the catalytic activity during oxidation of small organic
molecules,[13d,14] but the mechanism through which the ada-
toms alter the selectivity during polyol oxidation has not yet
been well established. Figure 2 shows how in a bismuth-satu-
rated solution the Pt/C electrode enhances the sorbitol oxida-
tion activity to ca. 180% compared to Pt/C in terms of maxi-
mum current density (which is proportional to the turnover
frequency; TOF), and, more importantly, steers the product se-
lectivity in favor of sorbose and fructose at potentials below
0.7 VRHE. A similar effect is observed with an antimony-modified
Pt/C: an antimony coverage of ca. 90% on Pt/C surface re-
duces the current of sorbitol oxidation with respect to Pt/C,
but also in the presence of antimony, sorbitol is selectively oxi-
dized to sorbose and fructose up to 0.9 VRHE, with a correspond-
ing significant suppression of the primary alcohol oxidation.
Underpotential deposited bismuth and antimony on the plati-
num surface are in the metallic state and the stability of the
adatoms has been demonstrated in direct formic acid fuel cells
(DFAFC).[14] For both bismuth and antimony, at higher poten-
tials, voltammetric peaks may correspond to the adsorption of
oxygenated species and finally to oxidative desorption of the
adsorbed bismuth and antimony, as a similar product distribu-
tion as on unmodified Pt/C is obtained.[15a,b,16] Note that, in
a real application, the promoter ions would be present in solu-
tion, so that the adsorption/desorption process would be re-
versible. In addition, lead, tin, and indium on Pt/C also enhance
the catalytic activity and change the selectivity of sorbitol oxi-
dation (see Supporting Information, Figure S1), whilst these
adatoms do not alter the selectivity during glycerol oxida-
tion.[13c]
Figure 1. Sorbitol oxidation (0.1m) on Pt/C electrode in 0.5m H2SO4: (a) cur-
rent density profile at a scan rate of 1 mVsÀ1, (b) concentration changes of
reaction products, and (c) selectivity (%) of products as a function of poten-
tial.
a somewhat higher selectivity to glucose at higher potentials.
This suggests that there is no strong stereoselective effect for
the electrocatalytic oxidation of sorbitol on the unmodified
platinum catalyst, whilst the secondary alcohol oxidation is
strongly suppressed. Interestingly, at the potentials of platinum
surface oxide formation (>1.0 V), the selectivity to formic acid
increases via CÀC bond cleavage as commonly seen in polyol
oxidation[11,13] in combination with an increased selectivity to-
wards fructose; an effect that remains to be further under-
stood.
Our previous work regarding the electrocatalytic oxidation
of glycerol has revealed that Pt/C electrodes modified by bis-
muth or antimony enhance the catalytic activity and change
the reaction selectivity toward secondary alcohol oxidation to
generate dihydroxyacetone.[13c,d] Previous work has shown that
the product distribution obtained from voltammetric studies
with online HPLC was consistent with a prolonged electrolysis
at a chosen potential.[13d] It is known that bismuth and antimo-
ny on the platinum surface block the CO formation and en-
Interestingly, on all Pt/C electrodes modified by adatoms, we
observe that the selectivity towards each primary (glucose vs.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
&
2
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
Arabitol is selectively oxidized to arabinose compared to
lyxose regardless of the presence of bismuth (a scheme of the
possible pathways and the results of ribitol oxidation are pre-
sented in the Supporting Information, Scheme S1 and Fig-
ure S2). We note that the presence of bismuth on Pt/C signifi-
cantly enhances the oxidation activity of arabitol to the secon-
dary oxidation product xylulose as the dominant product, with
a lower selectivity to ribulose. The same observation was made
for the oxidation of C4 polyols (see Supporting Information,
Scheme S2, Figures S3 and S4): the placement of the primary
and secondary OH on the same side of the carbon chain in the
Fischer projection renders the polyol oxidizable on platinum;
if, however, the OHs are placed on opposite sides, the polyol
has a very low oxidation activity. A similar stereoselective oxi-
dation activity has been observed before by Enea and Ango[17]
and Parpot et al.[18] for polyol oxidation in alkaline media. Addi-
tion of bismuth to the reactive system removes this stereose-
lective oxidation activity (though some preference of one
isomer over the other still remains), and leads to the selective
oxidation of the secondary alcohol. This strongly suggests that
the p-block-type promoter atoms interfere with the intrinsic
stereochemistry of the polyol in such a way that the secondary
alcohol is oxidized selectively. It has suggested before that
metal cation can act as a Lewis acid capable of catalysing the
ketose–aldose isomerisation.[19]
Figure 3. Arabitol oxidation (0.1m) in 0.5m H2SO4 on a Pt/C electrode (blank)
in comparison with sorbitol (0.1m) and ribitol (0.1m) oxidation and arabitol
oxidation on a Pt/C in Bi-saturated solution: (a) current density profile at
scan rate of 1 mVsÀ1, (b) concentration of reaction products of arabitol oxi-
dation, and (c) corresponding selectivity (%) of products as a function of po-
tential.
The above experiments were designed to demonstrate the
principle possibility of selectively oxidizing the secondary alco-
holic group on sorbitol by choosing carefully controlled elec-
trochemical conditions in a half cell. It shows that by promot-
ing the platinum surface by p-block metal atoms, one may
steer the selectivity of polyol oxidation in a fundamental way.
Application of this model approach to an electrolysis cell
based on a carbon cloth diffusion layer would require the con-
sideration and optimization of several parameters (e.g., catalyst
loading, membrane, concentration of sorbitol, operating tem-
perature). Our work here demonstrates the “proof-of-principle”
that the p-block modified platinum electrode is the most
promising catalyst for obtaining a high yield and selectivity for
such an application. In fact, the actual practical application of
this result may or may not involve an electrochemical setup,
but would likely still use the catalytic system suggested here.
Aqueous-phase heterogeneously catalyzed oxidation of sorbi-
tol by a PtBi catalyst may in fact an industrially more conven-
ient option (see, e.g., refs. [20]) for the selective oxidation of
glycerol by PtBi using such a setup), but those experiments
will not easily give the very detailed information collected in
Figure 1–3. Still, it is useful to comment on the implications of
reactivity that can be extracted from Figure 2. The experiment
illustrated in Figure 2(a)–(c) took ca. 20 min, which is by far
not long enough to report on conversion factors. The currents
are stable on this time scale, but we have not tested the stabil-
ity on the time scale of hours. The maximum current density in
Figure 2 is ca. 0.1 mAcmÀ2 at 0.5 V at room temperature, equiv-
alent to a conversion of ca. 5ꢁ10À10 mol of sorbitol molecules
per second per cm2 of catalyst, from a 0.1m solution. This im-
plies that 1 g of platinum of 200 m2 gÀ1 surface area will yield
10À3 mol fructose plus sorbose in 1 second. This yield will be
gulose) and secondary (fructose vs. sorbose) alcohol oxidation
product is not identical. To study why the selectivity to glucose
and sorbose is superior to that to gulose and fructose, we in-
vestigated how the molecular structure, especially the stereo-
chemistry of the 2-carbon position, is related to the selectivity
of the oxidation products. Figure 3 compares the electrocata-
lytic oxidation of two C5 polyols that is, d-arabitol and d-ribitol
(C5H12O5), on a Pt/C electrode in acidic solution, in absence
and presence of bismuth in solution. The possible oxidation
pathways of arabitol are shown in Scheme 2.
The difference between the structure of arabitol and ribitol
is in the steric position of -OH on C2. Remarkably, on an un-
modified Pt/C electrode, the oxidation activity of arabitol is
strongly suppressed compared to that of sorbitol and ribitol.
Scheme 2. Possible oxidation pathways of d-arabitol and its structural com-
parison with d-ribitol.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
&
3
&
ÞÞ
These are not the final page numbers!

CHEMSUSCHEM
COMMUNICATIONS
www.chemsuschem.org
higher at higher temperature, though selectivity may be im-
paired by further oxidation of the ketoses.
Acknowledgements
In conclusion, we demonstrate the fundamental possibility
of the selective electrochemical oxidation of sorbitol to fruc-
tose and sorbose by using a platinum electrode promoted by
p-block metal atoms. This simple and robust approach would
be easily scalable or translated into aqueous-phase heteroge-
neous catalysis, but the high selectivity and yield towards
a single product as in conventional enzymatic synthesis is still
challenging. Given the possibility of the electrochemical reduc-
tion of glucose to sorbitol,[1a,3] the approach would also open
up the way to (electro)catalytically convert glucose to its ke-
tonic isomers. Furthermore, by studying a range of C4, C5, and
C6 polyols, we provide evidence for the idea that the promoter
interferes with the stereochemistry of the polyol and thereby
modifies its reactivity. Still, the exact mechanism of promotion
remains to be understood in detail.
This research has been performed within the framework of the
CatchBio program. The authors gratefully acknowledge the sup-
port of the Smart Mix Program of the Netherlands Ministry of
Economic Affairs and the Netherlands Ministry of Education, Cul-
ture and Science.
Keywords: electrocatalysis
·
fructose
·
heterogeneous
catalysis · promotion · sorbitol
[1] a) A. M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 2012,
51, 2564–2601; Angew. Chem. 2012, 124, 2614–2654; b) R. D. Cortright,
R. R. Davda, J. A. Dumesic, Nature 2002, 418, 964–967.
[2] T. Werpy, G. Petersen, in Top Value Added Chemicals from Biomass. Re-
sults of Screening for Potential Candidates from Sugars and Synthesis Gas,
Vol. 1, US Department of Energy, 2004.
[3] Y. Kwon, M. T. M. Koper, ChemSusChem 2013, 6, 455–462.
[4] G. W. Huber, J. W. Shabaker, J. A. Dumesic, Science 2003, 300, 2075–
2077.
[5] R. M. de Almeida, J. Li, C. Nederlof, P. O’Connor, M. Makkee, J. A. Moulijn,
ChemSusChem 2010, 3, 325–328.
[6] R.-J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H. J.
Heeres, J. G. de Vries, Chem. Rev. 2013, 113, 1499–1597.
[7] R.-J. van Putten, J. N. M. Soetedjo, E. A. Pidko, J. C. van der Waal, E. J. M.
Hensen, E. de Jong, H. J. Heeres, ChemSusChem 2013, 6, 1681–1687.
[8] W. R. Sullivan, J. Am. Chem. Soc. 1945, 67, 837–840.
[9] A. Oyarce, C. Gonzalez, R. B. Lima, R. W. Lindstrçm, C. Lagergren, G.
Lindbergh, Electrochim. Acta 2014, 116, 379–387.
[10] a) L. ProenÅa, M. I. S. Lopes, I. Fonseca, A. Rodes, R. Gomez, A. Aldaz,
Electrochim. Acta 1998, 44, 735–743; b) L. ProenÅa, M. I. S. Lopes, I. Fon-
seca, J. M. Leger, C. Lamy, J. Electroanal. Chem. 1997, 432, 193–198; c) L.
ProenÅa, M. I. S. Lopes, I. Fonseca, F. Hahn, C. Lamy, Electrochim. Acta
1998, 44, 1423–1430; d) J. M. Lꢂger, F. Hahn, B. Beden, C. Lamy, M. F.
Bento, I. Fonseca, M. I. S. Lopes, J. Electroanal. Chem. 1993, 356, 255–
267.
Experimental Section
Electrochemical measurements were carried out in a standard
three-electrode cell controlled by a potentiostat/galvanostat (m-Au-
tolab Type III). A thin-film electrode with 3 nm Pt/C nanoparticles
(50 wt%, Tanaka) was fabricated by loading defined amounts of
a nanoparticle suspension (10 mL) in water (1 mgmLÀ1) onto a pol-
ished glassy carbon substrate, subsequently dried by purging Ar at
room temperature. In order to confirm the electrochemically active
surface area of the loaded catalyst, a blank voltammogram was re-
corded before each experiment. A large platinum plate was em-
ployed as a counter electrode and a reversible hydrogen electrode
(RHE) as a reference. Oxidation of polyols (sorbitol and arabitol)
was performed in a mixture of polyols (0.1m, analytical grade) and
0.5m H2SO4 under deaerated conditions by purging Ar.
[11] L. ProenÅa, M. I. S. Lopes, I. Fonseca, K. B. Kokoh, J. M. Leger, C. Lamy, J.
Electroanal. Chem. 1997, 432, 237–242.
For the Pt/C surface modification, analytical grades of bismuth
(Bi2O3, Sigma–Aldrich), antimony (Sb2O3, Sigma–Aldrich), lead
(Pb(ClO4)2, Acros), tin (SnCl2·2H2O, Merck), and indium (In2(SO4)3,
Sigma–Aldrich) were used. Irreversible adsorption of adatoms was
performed by placing the freshly prepared Pt/C electrode in con-
tact with antimony-saturated 0.5m H2SO4 solutions and 1 mm solu-
tion of tin for 3 min.[13c] After rinsing the electrode with water, each
coverage is determined by recording the suppression of the hydro-
gen region,[21] and then transferred to the electrochemical cell. For
reversible adatom adsorption/desorption, 1 mm adatoms (lead and
indium) and a sufficiently large amount of Bi2O3, which corre-
sponds to a concentration of ca. 10À5–10À4 m, were introduced di-
rectly into the electrochemical cell. A freshly prepared Pt/C elec-
trode was kept at a constant potential of 0.1 V for 3 min,[13c] after
which the surface coverage is measured by the suppression of the
hydrogen region in a blank solution, before applying it for sorbitol
oxidation.
The reaction products during voltammetry were collected in a mi-
crotiter plate with a fraction collector and collected samples were
analysed with an HPLC system (online HPLC), which provides quali-
tative and quantitative information as a function of potential as de-
scribed in our previous work.[12,,13d,22] Sample volumes of 30 mL
were injected into the four columns in series of an Aminex HPX
87-H (Bio-Rad) and three columns of Sugar SH1011 (Shodex) with
diluted sulfuric acid (0.5 mm) as eluent. The selected temperature
of column oven was 858C. Details of system configuration are de-
scribed elsewhere.[12,13]
[12] Y. Kwon, M. T. M. Koper, Anal. Chem. 2010, 82, 5420–5424.
[13] a) Y. Kwon, K. J. P. Schouten, M. T. M. Koper, ChemCatChem 2011, 3,
1176–1185; b) Y. Kwon, S. E. F. Kleijn, K. J. P. Schouten, M. T. M. Koper,
ChemSusChem 2012, 5, 1935–1943; c) Y. Kwon, T. J. P. Hersbach, M. T. M.
Koper, Top. Catal. 2014, 57, 1272–1276; d) Y. Kwon, Y. Birdja, I. Spanos,
P. Rodriguez, M. T. M. Koper, ACS Catal. 2012, 2, 759–764.
[14] a) J. K. Lee, H. Jeon, S. Uhm, J. Lee, Electrochim. Acta 2008, 53, 6089–
6092; b) S. Uhm, H. J. Lee, Y. Kwon, J. Lee, Angew. Chem. Int. Ed. 2008,
47, 10163–10166; Angew. Chem. 2008, 120, 10317–10320.
[15] a) E. Herrero, J. M. Feliu, A. Aldaz, J. Electroanal. Chem. 1994, 368, 101–
108; b) A. Fernandez-Vega, J. M. Feliu, A. Aldaz, J. Clavilier, J. Electroanal.
Chem. 1989, 258, 101–113; c) S. P. E. Smith, K. F. Ben-Dor, H. D. Abruna,
Langmuir 2000, 16, 787–794; d) S. Uhm, Y. Yun, Y. Tak, J. Lee, Electro-
chem. Commun. 2005, 7, 1375–1379.
[16] S. H. Cadle, S. Bruckenstein, Anal. Chem. 1972, 44, 1993–2001.
[17] O. Enea, J. P. Ango, Electrochim. Acta 1989, 34, 391–397.
[18] P. Parpot, N. Nunes, A. P. Bettencourt, J. Electroanal. Chem. 2006, 596,
65–73.
[19] a) B. E. Tilley, D. W. Porter, R. W. Gracy, Carbohydr. Res. 1973, 27, 289–
296; b) R. W. Nagorski, J. P. Richard, J. Am. Chem. Soc. 2001, 123, 794–
802.
[20] a) H. Kimura, Appl. Catal. A 1993, 105, 147–158; b) W. Hu, D. Knight, B.
Lowry, A. Varma, Ind. Eng. Chem. Res. 2010, 49, 10876–10882.
[21] P. Rodrꢃguez, J. Solla-Gullon, F. J. Vidal-Iglesias, E. Herrero, A. Aldaz, J. M.
Feliu, Anal. Chem. 2005, 77, 5317–5323.
[22] Y. Kwon, S. J. Raaijman, M. T. M. Koper, ChemCatChem 2014, 6, 79–81.
Received: August 22, 2014
Published online on && &&, 0000
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
&
4
&
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
New cats on the block: A new electro-
catalytic method for the selective elec-
trochemical oxidation of sorbitol to fruc-
tose and sorbose is developed by using
a platinum electrode promoted by p-
block metal atoms. By studying a range
of C4, C5, and C6 polyols, it is found
that the promoter interferes with the
stereochemistry of the polyol and there-
by modifies its reactivity.
Y. Kwon, E. de Jong, J. K. v. d. . Waal,
M. T. M. Koper*
&& – &&
Selective Electrocatalytic Oxidation of
Sorbitol to Fructose and Sorbose
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 0000, 00, 1 – 4
&
5
&
ÞÞ
These are not the final page numbers!