Subscriber access provided by UNIV OF NEBRASKA - LINCOLN
Article
Strong impact of platinum surface structure on primary and secondary alcohol oxidation during electro-oxidation of glycerol Amanda Cristina Garcia, Manuel J. Kolb, Chris van Nierop y Sanchez, Jan Vos, Yuvraj Y. Birdja, Youngkook Kwon, Germano Tremiliosi-Filho, and Marc T.M. Koper ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b00709 • Publication Date (Web): 02 Jun 2016 Downloaded from http://pubs.acs.org on June 5, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
Strong impact of platinum surface structure on primary and secondary alcohol oxidation during electro-oxidation of glycerol
1 2 3
Amanda C. Garciaa,b, Manuel J. Kolba, Chris van Nierop y Sancheza, Jan Vosa, Yuvraj Y. Birdjaa, Youngkook Kwona, Germano Tremiliosi-Filhob, Marc T. M. Kopera,‡
4 5 6 7
aLeiden
8 9
bInstituto
Institute of Chemistry, Leiden University, PO Box 9502, 2300 RA, Leiden, The Netherlands de Química de São Carlos, Universidade de São Paulo, Avenida Trabalhador São-Carlense 400, 13569-590 São Carlos, SP, Brazil
10 11 12 13
‡
E-mail:
[email protected] 14 15
Abstract
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Herein we describe a combined experimental and computational study of electrochemical glycerol oxidation in acidic media on Pt(111) and Pt(100) electrodes. Our results show that glycerol oxidation is a very structure sensitive reaction in terms of activity and, more surprisingly, in terms of selectivity. Using a combination of online HPLC and online electrochemical mass spectrometry, we show that on the Pt(111) electrode, glyceraldehyde, glyceric acid and dihydroxyacetone are products of glycerol oxidation, while on the Pt(100) electrode, only glyceraldehyde was detected as the main product of the reaction. Density Functional Theory calculations show that this difference in selectivity is explained by different binding modes of dehydrogenated glycerol to the two surfaces. On Pt(111), the dehydrogenated glycerol intermediate binds to the surface through two single Pt-C bonds, yielding an enediol-like intermediate, which serves as a precursor to both glyceraldehyde and dihydroxyacetone. On Pt(100), the dehydrogenated glycerol intermediate binds to the surface through one double Pt=C bond, yielding glyceraldehyde as the only product. Stripping and in situ FTIR measurements show that CO is not the only strongly bound adsorbed intermediate of the oxidation of glycerol, glyceraldehyde and dihydroxyacetone. Although the nature of this adsorbate is still unclear, this intermediate is highly resistant to oxidation and can only be removed from the Pt surface after multiple scans.
32 33
Keywords: glycerol, electrocatalysis, platinum, selective oxidation, structure sensitivity 1 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 27
1 2
INTRODUCTION
3
Glycerol is an important biomass-related compound, both as a model polyol compound and as an
4
abundant byproduct of biodiesel fabrication1,2. The application of glycerol is mainly related to its use
5
as a feedstock for polyether/polyols, drugs, pharmaceuticals, and food. However, glycerol has also
6
been considered as a possible fuel in direct alcohol fuel cells (DAFC)3 and as a compound to
7
cogenerate added-value chemicals in electrochemical energy conversion4,5.
8
Theoretically, the complete oxidation of glycerol to CO2 yields 14 electrons3. However, one of the
9
main challenges in the oxidation of glycerol is the difficulty of breaking both C-C bonds6,7, which
10
currently makes the use of glycerol in DAFC unattractive, since the various parallel reactions lead to
11
the production of several partial oxidation products, such as glyceraldehyde, 1,3-dihydroxyacetone,
12
glyceric acid, hydroxypyruvic acid, tartronic acid, mesoxalic acid, formic acid, oxalic acid and glycolic
13
acid. In addition, as with many other small organic molecules, the formation of poisoning adsorbed
14
intermediates (i.e. carbon monoxide) at the electrode surface at low overpotentials causes a decrease
15
in the overall system efficiency8. On the other hand, the production of soluble intermediates makes
16
this reaction very interesting since some of these products have a high added value and are
17
sometimes obtained by a high-cost process. Certain oxygenates derived from glycerol have important
18
practical applications, but unfortunately the selective functionalization of the triol molecule is a
19
difficult process9,10.
20
Of specific interest in this paper is the selectivity of the first step of glycerol electro-oxidation on
21
platinum, to the aldose glyceraldehyde (GALD) or to the ketose 1,3-dihydroxyacetone (DHA), i.e.
22
products obtained by oxidation of the primary and secondary hydroxyl group of glycerol,
23
respectively1,11. GALD can be oxidized to glyceric acid and tartronic acid, both commercially useful
24
compounds. DHA is an economically interesting product due to its application in the cosmetic 2 ACS Paragon Plus Environment
Page 3 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
industry as a self-tanning agent. Its further oxidation leads to mesoxalic acid. Usually, DHA is
2
produced by a fermentation process due to requirements of high yield, selectivity and quality.
3
However, microbial oxidation suffers from a low glycerol concentration and a long operating
4
time12,13.
5
Previous results on the electrochemical oxidation of glycerol have demonstrated that the activity and
6
selectivity of the reaction are strongly influenced by the pH6, the nature and structure of the
7
catalyst14,15 and the presence of p-block promoters in solution11,16. Especially promoters such as Bi
8
and Sb can have a dramatic effect on the selective formation of GALD vs. DHA. Whereas pure
9
platinum electrodes favor the oxidation of the primary alcohol, the presence of Bi and Sb can lead to
10
an almost 100% selectivity towards DHA. A similar effect has recently been observed with larger
11
polyols, such as sorbitol17. Although this effect of promoters on selective aldose vs. ketose formation
12
has been known from heterogeneous catalysis18,19, there is a remarkably little atomic-level insight into
13
the mechanism of selective polyol oxidation and the intermediates involved.
14
The present paper will present important new insights into the mechanism and selectivity of glycerol
15
oxidation on Pt single-crystal electrodes by showing that the selectivity towards GALD/DHA is
16
highly structure sensitive. By combining detailed spectro-electrochemical experiments with first-
17
principles density functional theory (DFT) calculations, we are able to show that the selectivity of
18
glycerol oxidation is determined by its initial mode of adsorption to the surface, suggesting that an
19
enediol-type surface intermediate plays a crucial role in DHA formation. Furthermore, our
20
experimental results prove the existence of an oxidation pathway involving an irreversibly polyol-
21
type intermediate that is very difficult to oxidize (and that is not carbon monoxide).
22
Our findings demonstrate the decisive role of surface crystallographic structure on selectivity of the
23
glycerol electro-oxidation reaction and based on this knowledge, eventually new practical catalysts
24
may be developed to improve the activity and selectivity of reaction to the desired products. 3 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 27
1 2
EXPERIMENTAL SECTION
3
Experimental conditions, electrodes and reactants
4
All experiments were performed at room temperature (20°± 1°C) in a conventional three-electrode
5
single cell, which was cleaned by a standard procedure to remove all organic traces20. The chemicals
6
used for solution preparation were high-purity perchloric acid (70% Merck Suprapur), glycerol
7
(GLY) (85%, analytical grade), glyceraldehyde (GALD) and 1,3-dihydroxyacetone (DHA) from
8
Sigma-Aldrich and ultrapure water from a Millipore system (18.2 MΩcm-1). The counter and
9
reference electrodes were a Pt wire and a reversible hydrogen electrode (RHE), respectively. The
10
working electrodes were Pt(111) and Pt(100) purchased from icryst, with geometric area 0.0402 and
11
0.0366 cm2, respectively.
12
Electrochemical measurements
13
Electrochemical measurements were controlled with a Potentiostat/galvanostat (µAutolab Type III).
14
In order to obtain ordered and clean surfaces, single crystal electrodes were prepared according to
15
the well-known Clavilier method21,22 which consists in annealing the working electrode in a hydrogen-
16
oxygen flame and cooling down in a reductive H2 / Ar atmosphere. Next, the electrode is protected
17
with a droplet of water saturated with the cooling gases, to prevent the contamination and
18
reconstruction of the electrode surface during transfer to the electrochemical cell. The working
19
electrode was introduced into the cell containing oxygen-free supporting electrolyte (0.5 M HClO4).
20
Prior to bringing the working electrode in contact with the solution, a potential of 0.1 V vs. RHE was
21
applied. Cyclic voltammograms (CV) were recorded to test the order of the Pt single crystals and the
22
cleanliness of the electrolyte by the stability of the characteristic blank voltammetric features of the
23
Pt single crystals in the potential range between 0.06 V to 0.9 V at 0.050 V s-1. Afterwards, the 4 ACS Paragon Plus Environment
Page 5 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
electrolyte was replaced to that containing the respective oxygen-free working solution with the
2
molecule under study (0.1 M GLY, GALD and DHA) in order to perform the electrochemical
3
oxidation experiments. In all experiments, CV profiles were obtained in meniscus configuration. The
4
upper potential limit was set to avoid oxide formation, which generates irreversible surface
5
disordering. In order to determine the involvement of the possible co-products and adsorbates from
6
glycerol oxidation with the Pt surface, we also studied the oxidation of GALD and DHA.
7
Electrochemical adsorption measurements
8
Experiments on adsorbate formation from adsorption of GLY, GALD and DHA on Pt(111) and
9
Pt(100) were performed by bringing the working electrode into contact with a 0.1 M solution of the
10
corresponding alcohol keeping the potential at 0.2 V vs. RHE for 1 min. The potential of 0.2 V was
11
chosen on the one hand high enough to enable adsorption and on the other hand low enough to
12
suppress oxidation reactions of the adsorbate. After adsorption, the electrode was rinsed in O2 free
13
ultrapure water to remove excess of alcohol and then transferred to the electrochemical cell, which
14
contained the pure electrolyte (0.5 M HClO4). CO stripping measurements were also performed on
15
Pt single crystals for comparison with the adsorbate stripping. For these experiments, CO gas was
16
bubbled into the pure electrolyte cell during 5 min with the potential kept at 0.1 V vs. RHE. Next,
17
excess CO was removed from the solution by bubbling argon by 10 min. The oxidative stripping of
18
the resulting adsorbates and CO were performed in a clean supporting electrolyte by sweeping
19
potential between 0.06 V to 0.9 V at 0.010 V s -1.
20
21
22 5 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 27
1
HPLC experiments
2
High Performance Liquid Chromatography (HPLC) was used to detect the liquid products
3
produced during the electrochemical oxidation of the polyol23. The reaction products were collected
4
during the cyclic voltammetry with a small Teflon tip (with an inner diameter of 0.38 mm) positioned
5
ca. 10 µm from the center of the electrode surface, and which was connected to a PEEK capillary
6
with inner/outer diameters of 0.13/1.59 mm. Before sample collection, the tip and capillary were
7
cleaned with ultrapure water and supporting electrolyte. The sample volume collected in each well
8
was 60 µL on a microtiter plate with 96 wells (270 µL/well, Screening Devices b.v.) using an
9
automatic fraction collector (FRC-10A, Shimadzu). The flow rate of sample collection was set at 60
10
µL/min with a Shimadzu pump (LC-20AT). After collection of the samples, the microtiter plate was
11
covered by a silicon mat to prevent evaporation of samples. Samples were taken every 60 mV, in the
12
potential range between 0.06 V to 0.9 V vs. RHE at 1.0 mV s-1 on Pt(111) and Pt(100) electrodes.
13
The microtiter plate with the collected samples was placed in an auto sampler (SIL-20A) holder and
14
the samples were analyzed by high-performance liquid chromatography (Prominence HPLC,
15
Shimadzu). 30 µL of sample was injected into the column configuration of an Aminex HPX 87-H
16
(BioRad) together with a Micro-Guard Cation H cartridge (BioRad) in series with a Sugar SH1011
17
(Shodex) column and diluted sulfuric acid (0.5 mM) was used as the eluent. The temperature of the
18
column was maintained at 70°C in a column oven (CTO-20A), and the separated compounds were
19
detected with a refractive index detector (RID-10A). The expected products were analyzed separately
20
to produce standard calibration curves at 70°C (i.e., glyceraldehyde, dihydroxyacetone, glyceric acid,
21
glycolic acid, formic acid, oxalic acid, mesoxalic acid, hydroxypyruvic acid and tartronic acid).
22
23 6 ACS Paragon Plus Environment
Page 7 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
OLEMS experiments
2
The volatile products of the electro-oxidation of GLY, GALD and DHA were detected using online
3
electrochemical mass spectroscopy (OLEMS) with an evolution mass spectrometer system
4
(European Spectrometry systems Ltd)24. The porous Teflon tip (inner diameter, 0.5 mm) was
5
positioned close (≈10 µm) to the center of the electrode. Before the experiments, the tip was dipped
6
into a 0.2 M K2Cr2O7 in 2 M H2SO4 solution for 15 min and rinsed with ultrapure water thoroughly.
7
The gas products were collected through a polyether ether ketone (PEEK) capillary into the mass
8
spectrometer. A 1500-V secondary electron multiplier (SEM) voltage was applied for all the
9
fragments. The OLEMS measurement was conducted while the CV was scanned from 0.06 to 0.9 V
10
and back at a scan rate of 1.0 mV s-1. The OLEMS measurements were not calibrated and therefore
11
provide qualitative information only.
12
FTIR experiments
13
To provide information about the surface-adsorbed intermediates of the electrochemical glycerol
14
oxidation reaction, in situ FTIR spectra were collected. The FTIR instrument was a Bruker Vertex
15
80V IR spectrometer equipped with liquid nitrogen cooled detector. In situ FTIR experiments were
16
performed in a three electrode spectro-electrochemical cell with a CaF2 prism attached to the bottom
17
of the cell. Details concerning the cell are described in the literature25. FTIR spectra were obtained in
18
the wavenumber range between 2500 cm-1 to 1000 cm-1. Spectra were collected in 0.1 V steps from
19
0.2 V to 0.9 V vs. RHE. They were computed from the average of 100 interferograms with spectral
20
resolution set to 8 cm-1. Spectra are presented as absorbance, according to A = - log (R / R0), where
21
R and R0 are the reflectance corresponding to the single beam spectra obtained at the sample and
22
reference potentials, respectively.
7 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 27
1
COMPUTATIONAL METHODS
2
The computations of the interaction of GLY, GALD and DHA with the Pt surface were performed
3
using the ab-initio density functional code VASP26–29 using the PBE functional30 and PAW
4
projectors31,32. The unit cell was chosen as a 4x4 cell on Pt(111) and a 4x5 cell on Pt(100) such that
5
the repeated images of the adsorbates are far enough apart to minimize interaction. The k-point
6
sampling was chosen so the density of k-points is kept as constant as possible, which corresponds to
7
a 3x3x1 sampling for both surfaces. A basis set with a cutoff energy of 400eV was chosen. To
8
calculate onset potentials for oxidation, we employ the computational hydrogen electrode scheme
9
suggested by Norskov et al.33.
10
Adsorption energies are given as DFT energies and do not include solvation energies to compute
11
free adsorption energies of the products, since those were not available for some species, e.g. the
12
enediol species. The reported DFT energies include a correction term for the zero point energy. In
13
discussing the computational results, we will employ a naming scheme for the individual hydrogen
14
atoms that will subsequently be removed from the glycerol, which is shown in Figure 1. The scheme
15
is built so that odd numbers correspond to the hydrogen atoms that are connected to the main
16
carbon chain, while even numbers are part of the OH groups. At first glance the hydrogen atoms 1
17
and 3 / 7 and 9 appear to be equivalent, and in gas-phase or liquid phase glycerol they certainly are,
18
however when adsorbed the rotational symmetry is broken due to the presence of the surface that
19
inhibits the free movement and rotation of the adsorbed species. This means that both 1-3 and 7-9
20
are no longer equivalent, due to the interaction of the OH groups with the surface. This has quite a
21
large influence on the adsorption energies of these states as will be discussed later on.
8 ACS Paragon Plus Environment
Page 9 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1 2 3
Figure 1: Hydrogen naming procedure, even numbers denote hydrogen that are part of OH groups, odd numbers hydrogen atoms that are bound to carbon atoms.
4 5
RESULTS AND DISCUSSION
6
Experimental Results
7
Electro-oxidation of polyol on Pt(111) and Pt(100)
8
Figure 2 shows the cyclic voltammetry profile obtained at 1.0 mV s-1 in 0.5 M HClO4 + 0.1 M polyol,
9
as well as the corresponding main liquid products of the electrochemical oxidation of GLY, GALD
10
and DHA detected by HPLC and the ionic current related to the mass signal (m/z= 44) from the
11
CO2 production, as detected by OLEMS measurements, on both Pt(111) and Pt(100) electrodes.
12
The base voltammetry of Pt(111) and Pt(100) in pure HClO4 electrolyte are shown in the
13
Supplementary Information SI (Fig.S1), showing the typical features of well-ordered Pt single
14
crystals, as reported by Clavilier and co-authors34,35.
9 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 27
1 2 3 4
Figure 2. (a, d and g) cyclic voltammograms for 0.1 M of glycerol, glyceraldehyde and dihydroxyacetone oxidation on Pt(111) (black line) and Pt(100) (red line) electrodes in HClO4, respectively. (b, e and h) the corresponding liquid product concentration as measured by online HPLC and (c, f and i) the corresponding mass signal (m/z= 44) related to the production of CO2 as measured by OLEMS.
5
Figure 2a shows the CV for the glycerol oxidation reaction on both Pt(111) and Pt(100) electrodes.
6
In the forward scan, the onset potential for the electrochemical oxidation of glycerol on Pt(111)
7
electrode is at 0.5 V. The current peak at 0.60 V is followed by a sharp drop. On the Pt(100)
8
electrode a similar CV profile is observed, only shifted to higher potential by ca. 0.1 V. Figure 2b
9
displays the corresponding main liquid products of the electrochemical oxidation of glycerol on the 10 ACS Paragon Plus Environment
Page 11 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
two surfaces. Concerning the first step of glycerol electro-oxidation, glyceraldehyde (GALD) and
2
dihydroxyacetone (DHA) are the products obtained by oxidation of the primary and secondary
3
hydroxyl group of glycerol, respectively1,11. According to the HPLC data, on the Pt(111) electrode the
4
first product of glycerol oxidation is glyceraldehyde, which is detected from 0.48 V with its
5
concentration increasing until 0.85 V, after which it decreases again. The second product detected,
6
albeit in a lower amount, is DHA, as well as glyceric acid, a product of the further glyceraldehyde
7
oxidation, which both appear from 0.6 V. Although Pt(111) forms more than one product, only a
8
single anodic wave is observed in the anodic scan. Remarkably, on the Pt(100) electrode, GALD was
9
detected as the only product of the glycerol electro-oxidation reaction, which increases its
10
concentration from 0.6 V until around 0.75 V, after which the concentration decreases . The results
11
suggests that the faradaic current of the electrochemical oxidation of glycerol on Pt(100) (Figure 2a)
12
should be ascribed primarily to the conversion of glycerol into glyceraldehyde.
13
Figure 2c displays the mass signal (m/z = 44) related to the production of CO2 from glycerol
14
oxidation. On Pt(111), in the positive-going scan, the production of CO2 from glycerol oxidation
15
starts at about 0.5 V, passes through a maximum at around 0.65 V followed by a decrease to zero
16
current. In the negative-going scan, the CO2 formation was detected between 0.85 and 0.5 V, but the
17
maximum ionic current is somewhat lower than the maximum ionic current detected in the positive-
18
going scan. Comparing these results with those obtained by HPLC (Figure 2b), the results show that
19
in the potential range considered, glycerol is oxidized to incomplete reaction products such as
20
GALD, DHA and glyceric acid, but the C-C bond breaking and further CO2 formation also take
21
place, presumably through glycolic acid and formic acid intermediates, as suggested by the
22
mechanism presented by Kwon et al.15.
23
On Pt(100), the ionic current for the production of CO2 follows the same trend as the voltammetric
24
profile showed in the Figure 2a. According to HPLC (Figure 2b), glyceraldehyde is the only product 11 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 27
1
on Pt(100), but based on the OLEMS results (Figure 2c) we can affirm that Pt(100) electrode also
2
leads to the C-C bond breaking of the glycerol, suggesting the formation of other intermediate
3
products such as glyceric acid and formic acid and its subsequent oxidation to CO2.
4
The results in Figure 2a-c confirm that glycerol oxidation is a very structure sensitive reaction14, not
5
only in terms of activity, but, more surprisingly, also in terms of reaction selectivity. Whereas Pt(100)
6
is selective for primary alcohol oxidation, Pt(111) leads to both primary and secondary alcohol
7
oxidation. In the section on the computational results, we will show that this difference in selectivity
8
is likely due to a fundamentally different binding mode of dehydrogenated glycerol on the two
9
surfaces.
10
Since glyceraldehyde and dihydroxyacetone are the first products from electrochemical glycerol
11
oxidation and they can diffuse into the bulk of the solution or can be further oxidized, we also
12
investigated the reactivity of these alcohols on Pt(111) and Pt(100) electrodes by combined HPLC
13
and OLEMS voltammetry measurements.
14
Figure 2d-f compare the cyclic voltammetry profiles for GALD oxidation on Pt(111) and Pt(100)
15
electrodes and the corresponding liquid products and CO2 production, respectively. The onset
16
potential for GALD oxidation is almost the same on Pt(111) and Pt(100) electrodes (Figure 2d) at ca.
17
0.6 V, however the faradaic current at higher potential is much higher for Pt(100) than for Pt(111).
18
On the other hand, in the return scan, Pt(111) shows a higher activity.
19
The results displayed in Figure 2e show that both Pt(111) and Pt(100) electrodes oxidize GALD to
20
glyceric acid (GEA), glycolic acid (GOA) and formic acid (FA). The ionic currents related to the CO2
21
production from GALD oxidation are shown in Figure 2f, showing that the electrochemical
22
oxidation of GALD on Pt(111) and Pt(100) electrode are both accompanied by production of CO2.
23
In the positive-going scan, the production of CO2 (Figure 2e) starts at around 0.6 V on both Pt(111)
24
and Pt(100) electrodes, the same potential at which GALD starts to be oxidized. However, on 12 ACS Paragon Plus Environment
Page 13 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
Pt(100) the CO2 production is higher than on Pt(111). These results reinforce our believe that
2
glyceric acid is also a product from glycerol electro-oxidation on Pt(100), though it was not detected
3
by HPLC (Figure 2b), probably because on Pt(100), glyceric acid is formed in a potential region in
4
which it is quickly oxidized to other products like glycolic acid and formic acid, as shown in Figure
5
2e. Formic acid is further oxidized to CO2, explaining the higher current obtained for Pt(100) in
6
comparison to the Pt(111).
7
The reactivity for DHA electro-oxidation on Pt(111) and Pt(100) electrodes is shown in Figure 2(g-i).
8
Figure 2g shows that Pt(100) is much more active for the oxidation of DHA than Pt(111). The
9
results obtained by HPLC (Figure 2h) show that hydroxypyruvic acid (HPA) is one of the products
10
of the DHA oxidation reaction, but, even though the oxidation of DHA is more pronounced on
11
Pt(100), the concentrations of HPA are similar for Pt(111) and Pt(100) in the studied potential range.
12
More surprising results were obtained by OLEMS measurements (Figure 2i) in which a very low
13
signal related to the CO2 production was detected on Pt(111), while on Pt(100) no mass signal (m/z=
14
44) from CO2 was detected. It means that the oxidation of dihydroxyacetone and possible by-
15
products on Pt(100) are not accompanied by the cleavage of the C-C bond. The concentrations of
16
HPA and CO2 are very low for both electrodes, meaning that on Pt(100), which shows a relatively
17
high current, neither of them can be a major product.
18
In order to characterize possible irreversibly adsorbed intermediates during the electrochemical
19
oxidation of glycerol, glyceraldehyde and dihydroxyacetone in acid solution, adsorbate-stripping
20
measurements were performed. Figure 3a and b show the stripping of the adspecies from GLY on Pt
21
(111) and Pt(100), respectively. For both electrodes, an oxidation-stripping peak is observed starting
22
in the double-layer region, which disappears only after five scans, suggesting very slow oxidation
23
kinetics of the corresponding adsorbate. Interestingly, the typical peaks in the H-region, between
24
0.06 to 0.4 V, related to the H adsorption/desorption34,35, show incomplete blockage of the hydrogen 13 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 27
1
sites by adsorption of this species, and from the second scan the hydrogen region is characterized by
2
the presence of free hydrogen-adsorption states, though oxidative stripping of the adsorbate is still
3
observed.
4 5
Figure 3: Adsorbate stripping from glycerol adsorption on (a) Pt(111) and (b) Pt(100) electrodes, in 0.5 M HClO4; scan rate 0.010 V s-1. First scan (dashed line) and subsequence scan (solid line).
6 7
Table 1: Values of the oxidation charges Q(Ox) involved in each process QOx(CO) QOx(GLY) QOx(GALD) Pt(100) Pt(111)
QOx(DHA)
µC cm-2
µC cm-2
µC cm-2
µC cm-2
280 208
390 680
120 330
300 223
8 9
The oxidation of the adsorbates from glycerol on Pt(111) and Pt(100) starts at around 0.4 and 0.5 V,
10
respectively. On Pt(111) (Figure 3a), the oxidative current gives a broader peak with a maximum at
11
0.68 V, followed by a small and sharp peak at around 0.8 V, which is related to the OHads region of
12
the Pt(111)36. A sharper oxidation peak was observed for Pt(100) with a maximum at 0.63 V (Figure
13
3b). These findings imply that on Pt(111) the glycerol adsorbate is oxidized at a lower potential than
14
on Pt(100). The oxidation charges QOx(GLY) involved in the process, calculated by the sum of the
14 ACS Paragon Plus Environment
Page 15 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
stripping charges of all cycles, is much higher for Pt(111) than for Pt(100) (see Table 1), i.e. 680 vs.
2
390 µC cm-2, respectively.
3
Results shown in Figure S2 of the SI show the same features for the oxidation of glyceraldehyde and
4
dihydroxyacetone adsorbates on Pt(111) and Pt(100). By comparison with the glycerol adsorbate
5
(Figure 3), the profiles are very similar, however the relative charges involved in the oxidative process
6
are significantly lower for GALD and DHA, as presented in the fourth and fifth column in Table 1,
7
where QOx(CO), QOx(GLY), QOx(GALD) and QOx(DHA) are the oxidation charges calculated from oxidation of
8
adsorbates formed from CO, glycerol, glyceraldehyde and dihydroxyacetone, respectively.
9
To compare the results obtained from the stripping of the polyol adsorbate, CO stripping
10
measurements were also carried out and the results are included in Figure S2. The voltammetric
11
stripping profiles of the CO adlayers on Pt (111) and Pt(100) are very similar to those reported by
12
Cuesta et al.36, characterized by a complete blocking of hydrogen adsorption region, since no peak
13
was observed in the range potential between 0.06 - 0.4 V, and by a main CO stripping peak at around
14
0.74 V and 0.72 V, for Pt(111) and Pt(100), respectively.
15
A comparison between the stripping profiles obtained from oxidation of the polyol adsorbates and
16
CO shows that the profiles are very different. As previously reported by several authors36–39, CO
17
strongly adsorbs on the Pt surface, blocking completely the hydrogen region, and at higher potential
18
CO is completely oxidized in a single scan. However, the polyol adsorbate exhibits limited blockage
19
of the hydrogen region and requires several scans for complete oxidation.
20
These results suggests that when the electrode is held at 0.2 V in GLY, GALD and DHA containing
21
solution, an adsorbed intermediate is formed that is difficult to oxidize, but that is not CO. In situ
22
FTIR spectra taken after adsorption of species are shown in SI3. According to the set of spectra, the
23
oxidation of species coming from GLY, GALD and DHA adsorption seems to be accompanied by
24
CO2 production as the applied potential increases but no band related to adsorbed CO or any other 15 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 27
1
surface adsorbate was detected in the studied wavenumber range. This suggests that the coverage of
2
the adsorbate is low, as also indicated by the limited blockage of the hydrogen region. Previous ATR-
3
SEIRAS experiments by Schnaidt et al.40 on glycerol electro-oxidation on Pt thin-film electrodes have
4
provided evidence for the existence of adsorbed glyceroyl and adsorbed glycerate intermediates. The
5
authors found that glycerate/glyceric acid is rather inert to further oxidation on platinum, but also
6
that adsorbed glycerate is in a fast adsorption/desorption equilibrium with glyceric acid in solution.
7
Therefore, the strongly adsorbed intermediate observed in our work should probably not be
8
identified with adsorbed glycerate. Also, adsorbed glycerate cannot be an intermediate from DHA
9
oxidation.
10 11
In situ FTIR measurement of the glycerol oxidation on Pt(111) and Pt(100)
12
Figure 4 shows the in situ FTIR spectra obtained during oxidation of 0.1 M glycerol on Pt(111) and
13
Pt(100), in the wavenumber region ranging between 2500 cm-1 to 1000 cm-1 and the corresponding
14
integrated absorbance signal as a function of potential for the species detected. The set of spectra
15
was acquired by increasing the potential in steps from 0.1 V to 0.9 V vs. RHE. The reference
16
spectrum was taken at 0.1 V. Different bands can be observed depending on the surface orientation
17
and on the applied potential. The most noticeable spectral features are the appearance of bimodal
18
bands near 2045 cm-1 and 1840 cm-1, ascribable to the build-up of linearly41 and bridge-bonded42
19
adsorbed carbon monoxide, respectively, and the appearance of a band at 2345 cm-1 assigned to the
20
O-C-O asymmetric stretching mode of the CO2 molecule43.
16 ACS Paragon Plus Environment
Page 17 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2
ACS Catalysis
Figure 4: In situ FTIR spectra (a and c) and the corresponding integrated absorbance signal (b and d) as a function of potential, (100 scans, resolution 8 cm-1) of (a and b) Pt(111) and (c and d) Pt(100), in 17 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
1 2
Page 18 of 27
0.5 M HClO4 + 0.1 M glycerol. The reference spectra were collected at 0.1 V vs. RHE and the sample spectra were acquired by applying potential steps.
3 4
Bridge-bonded CO at 1830-1870 cm-1 was observed only on Pt(100) (Figure 4c), whereas linearly-
5
bonded CO was detected on both Pt(111) and Pt(100). According to Figure 4 a and b, linearly-
6
bonded CO on Pt(111) appears at 0.3 V. The production of CO2 starts at 0.7 V. At 0.8 V linearly-
7
bonded CO has been completely oxidized to CO2, since no band at 2045 cm-1 is observed anymore
8
(Figure 4a).
9
Carbonyl compounds, like aldehydes, ketones and carboxylic acid are formed as intermediates of the
10
glycerol oxidation reaction on Pt(111) electrode in acidic media, as seen by the bands at 1740 cm-1,
11
1420 cm-1 and 1250 cm-1 which appear at 0.5 – 0.6 V (Figure 4b), and they are related to the C=O
12
stretching of carbonyl groups, the C-O stretching / OH deformation of carboxylic acids, and to
13
symmetric bending vibration (O-C-O), respectively44. These carbonyl compounds can be assigned to
14
glyceraldehyde, dihydroxyacetone, glyceric acid and formic acid, in good agreement with HPLC data
15
(Figure 2b).
16
Bridge-bonded CO is formed on Pt(100) (Figure 4c and d) at ca. 0.2 V as evidenced by a well-
17
defined negative band at 1870 cm-1 (Figure 4c). The positive weak band at 2045 cm-1, which appears
18
at 0.2 V, is related to the appearance of linearly-bonded CO. Note that bridge-bonded CO shows a
19
negative peak from 0.2 V, implying that it is being converted into linearly-bonded CO with
20
increasing potential. This band grows until 0.6 V, the same potential at which CO2, assigned at 2345
21
cm-1, starts to form (Figure 4d). At potentials higher than 0.6 V, the CO2 band increases its intensity
22
and at 0.9 V the band at 2045 cm-1 disappears completely. It suggests that the production of CO2 on
23
Pt(100) is more related to the oxidation of linearly-bonded CO than to the oxidation of bridge-
24
bonded CO. 18 ACS Paragon Plus Environment
Page 19 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1 2
Computational results
3 4
In order to gain more insight into the glycerol oxidation reaction pathway, we have performed DFT
5
calculations on Pt(111) and Pt(100) surfaces. Figure 5a shows the key intermediates and the onset
6
potentials for their formation using the computation hydrogen electrode scheme. We only show the
7
most stable intermediates found. It should be noted that configurationally ”similar” configurations,
8
e.g. adsorbates with the hydrogen numbered 1 or 9 and 1 or 7 removed can differ by as much as 0.2
9
eV in energy, due to the different interaction of the OH groups with the platinum surface.
10
In our calculations we found that the first dehydrogenation step of glycerol on Pt(111) preferentially
11
abstracts the hydrogens that are connected to the carbon atoms, i.e. number 1,3 and 5, similar to the
12
findings of Liu and Greeley45. The second dehydrogenation step also involves the breaking of a C-H
13
bond. All intermediates that adsorb through O-H bond breaking are found to be at least 0.5 eV less
14
favorable than those adsorbing through carbon. The most stable intermediates on Pt(111) were
15
found to bind through the two primary carbons/atoms (i.e. 1-7, 1-9, 3-7 and 3-9), with carbons (C)
16
coordinating in an on-top position to a surface platinum atom, again in good agreement with
17
previous calculations of Liu and Greeley. It seems however that the intermediate can only dissociate,
18
rather than come off the surface, since there is no possibility for it to exist in liquid phase without
19
undergoing significant atom rearrangement. Therefore, we postulate that the actual reaction to form
20
GALD and DHA must proceed via the other stable intermediate shown in figure 5a, that binds
21
through the primary and secondary carbon (i.e. 1-5 and 3-5). This intermediate is similar to an
22
enediol species and can therefore leave the surface. Since this enediol species is the intermediate of
23
the acid-base catalyzed isomerization reaction between DHA and GALD, also known as the Lobry
19 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 27
1
de Bruyn-Alberda van Ekenstein transformation46, we assume that after the enediol desorbs from the
2
surface it will further react to GALD and DHA.
3
Figure 5b shows the adsorption energies of the key intermediates on Pt(100). We find that glycerol
4
oxidation on the Pt(100) surface shows similar trends for the first dehydrogenation step as on
5
Pt(111). Carbon-bound intermediates are always significantly more stable than oxygen-bound
6
intermediates, by about 0.2 eV. The first dehydrogenation step shows the same trend as on Pt(111)
7
with a preference for the removal of the hydrogens numbered 1,3 and 5. The second
8
dehydrogenation step again shows mostly the same trends as before, with a preference for binding
9
through carbon to the surface. However, here the main qualitative difference between Pt(111) and
10
Pt(100) appears. The intermediate 1-3, which corresponds to a doubly dehydrogenated carbon bound
11
to a surface platinum atom, is significantly more stable on the Pt(100) surface. This intermediate,
12
which is not favorable on Pt(111), can desorb after transferring 1 hydrogen from the primary OH
13
group to the primary carbon itself, forming only GALD and no DHA.
14
The higher stability of the doubly-bound glyceraldehyde precursor on Pt(100) may be qualitatively
15
explained by the lower coordination number of the platinum atoms in the Pt(100) surface, making
16
the formation of stronger double bonds to a single surface atom more favorable than on the close-
17
packed Pt(111) surface. The close-packed Pt(111) surface, on the other hand, will favor the
18
formation of single bonds, and hence intermediates coordinating to the surface through two C-Pt
19
single bonds will be more stable.
20 ACS Paragon Plus Environment
Page 21 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
(a)
(b)
1 2
Figure 5: Glycerol oxidation reaction pathways on (a) Pt(111) and (b) Pt(100) as suggested by the DFT calculations. 21 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 27
1
GENERAL DISCUSSION
2
Based on the results presented in the previous sections, we can suggest a detailed new mechanism for
3
the glycerol electro-oxidation reaction on Pt(111) and Pt(100) electrodes in acid solution, as shown
4
schematically in Figure 6. Our combined experimental and computational results suggest two
5
different active intermediates of initial glycerol dehydrogenation on the Pt surface: an enediol-type
6
intermediate binding with two adjacent C atoms to the Pt(111) surface, and an intermediate binding
7
through a single primary C to the Pt(100) surface. We attribute the different bonding modes to the
8
Pt(111) and Pt(100) surface to the different coordination numbers of their Pt surface atoms, allowing
9
only single Pt-C bonds on Pt(111) but favoring a double Pt=C bond on Pt(100). The enediol-type
10
intermediate is similar to the intermediate of the glyceraldehyde-dihydroxyacetone isomerization
11
reaction and serves as a precursor to the formation of both glyceraldehyde and dihydroxyacetone on
12
Pt(111). However, the intermediate favored on Pt(100) yields only glyceraldehyde as soluble product.
13
Besides these two active intermediates, our stripping experiments suggest the existence of a third
14
“inactive” intermediate, which is formed from glycerol, glyceraldehyde as well as dihydroxyacetone
15
(though its coverage is lower when formed from the latter two). This inactive intermediate has a low
16
coverage (whence our inability to identify it with FTIR spectroscopy) and it is very resistant to
17
oxidation, since it takes multiple oxidation scans to remove it from the surface. The only oxidation
18
product observed from the stripping of this adsorbate is CO2, but the adsorbate is clearly not CO. At
19
this moment, we can only speculate on the nature of this species. The species bound to the Pt
20
surface through its two primary C atoms (see Figure 6) is a potential candidate. DFT calculations
21
suggest that this species is the most stable adsorbate on Pt(111), but it does not have a soluble
22
counterpart, in contrast to the two “active” intermediates.
23
In the mechanism shown in Figure 6, glyceraldehyde is oxidized to glyceric acid in a subsequent two-
24
electrons transfer step. As a next two-electron transfer step, glyceric acid is oxidized by cleavage of 22 ACS Paragon Plus Environment
Page 23 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1
C-C bond to formic acid and glycolic acid15. The production of formic acid is considered based on
2
the OLEMS results, which showed that CO2 is a product of the glycerol electro-oxidation, and on
3
the basis of the HPLC results, in which formic acid was identified as a product of glyceraldehyde
4
oxidation on both electrodes. In the pathway of the secondary alcohol oxidation, dihydroxyacetone is
5
further oxidized to hydroxypyruvic acid via a four-electron transfer reaction. Dihydroxyacetone is
6
very difficult to oxidize to CO2, explaining the very small amount of CO2 observed in the OLEMS
7
experiments of dihydroxyacetone oxidation.
8 9
Figure 6: Glycerol oxidation mechanism on Pt(111) and Pt(100) electrodes in acidic solution.
10
The mechanism shown in Figure 6 adds important details to a mechanism suggested previously by
11
Kwon et al.15. In particular, it highlights the unique importance of the initial bonding mode of
12
glycerol and its role in the first oxidation product formed, i.e. glyceraldehyde (aldose) vs.
13
dihydroxyacetone (ketose). From the Tafel slope of 120 mV dec-1 measured by Kwon et al.15 we can 23 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 24 of 27
1
conclude that the formation of the first adsorbate is rate determining. In addition, it identifies a new
2
adsorbed intermediate that is strongly adsorbed and very difficult to oxidize. We expect these
3
mechanistic features to apply also to higher polyols with four or more carbon atoms, to be
4
investigated in future work.
5
CONCLUSIONS
6
The elucidation of the mechanism of the electrochemical glycerol oxidation reaction is complex since
7
it involves several intermediates and reaction steps including the cleavage of C-C bonds. An
8
important new aspect identified in this paper is the high structure sensitivity of the selectivity of the
9
initial glycerol oxidation on platinum towards primary or secondary alcohol oxidation. By a
10
combination of detailed electrochemical experiments, spectroscopy and density functional theory
11
calculations, we have shown that the selectivity of primary vs. secondary oxidation is determined by
12
the initial bonding mode of the dehydrogenated glycerol adsorbate. On the Pt(111) electrode,
13
doubly dehydrogenated glycerol binds through the primary and secondary carbon atoms, yielding an
14
enediol intermediate. This intermediate yields glyceraldehyde and dihydroxyacetone, i.e products of
15
the oxidation of the primary and secondary hydroxyl group of glycerol, respectively, as the first stable
16
soluble intermediates of the reaction. On the Pt(100) electrode, on the other hand, dehydrogenated
17
glycerol binds through a single primary carbon atom, so that only GALD is the first stable
18
intermediate of the reaction. In addition, our results also suggest the formation of another strongly
19
bonded intermediate during glycerol oxidation, that is not CO, and that is very difficult to oxidize.
20
Our results have provided essential new information that help elucidate the factors determining the
21
selectivity of the glycerol oxidation reaction. We expect that these results will impact on our
22
understanding of selective glycerol, including the effect of promoters, as well as on the selective
23
oxidation of higher poly-ols and other sugar molecules. 24 ACS Paragon Plus Environment
Page 25 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1 2 3
Associated Content
4
Supporting Information
5
Further details are given in Figures S1-S3
6
This material is available free of charge via the Internet at http://pubs.acs.org.
7
Acknowledgments
8
ACG and GTF acknowledge financial support from the São Paulo Research Foundation from Brazil
9
(Project number 2014/24438-8). This work was sponsored by the Netherlands Organization for
10
Scientific Research (NWO) and also by the NWO Exacte Wetenschappen, EW (NWO Physical
11
Sciences Division) for the use of supercomputer facilities, with financial support from the
12
Nederlandse Organisatie voor Wetenschappelijk Onderzoek (Netherlands Organisation for Scientific
13
Research, NWO).
14
References
15 16
(1)
Pagliaro, M.; Ciriminna, R.; Kimura, H.; Rossi, M.; Della Pina, C. Angew. Chemie Int. Ed. 2007, 4434–4440.
17 18
(2)
Zhou, C.-H. (Clayton); Beltramini, J. N.; Fan, Y.-X.; Lu, G. Q. (Max). Chem. Soc. Rev. 2008, 37, 527–549.
19
(3)
Simões, M.; Baranton, S.; Coutanceau, C. Appl. Catal. B Environ. 2010, 93, 354–362.
20 21
(4)
Mougenot, M.; Caillard, A.; Simoes, M.; Baranton, S.; Coutanceau, C.; Brault, P. Appl. Catal. B Environ. 2011, 107, 372–379.
22
(5)
Ketchie, W. Murayama, M. Davis, R. J. Catal. 2007, 250, 264–273.
23
(6)
Roquet, L.; Belgsir, E. M.; Léger, J.-M.; Lamy, C. Electrochim. Acta 1994, 39, 2387–2394.
24
(7)
Farias, M. J. S.; Camara, G. A.; Tanaka, A. A. J. Solid State Electrochem. 2007, 11, 1465–1469.
25
(8)
Lamy, C.; Belgsir, E. M.; Léger, J. M. J. Appl. Electrochem. 2001, 31, 799–809.
26
(9)
Mcmorn, P.; Roberts, G.; Hutchings, G. J. Catal. Letters 1999, 63, 193–197. 25 ACS Paragon Plus Environment
ACS Catalysis
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 26 of 27
1
(10)
Behr, A.; Eilting, J.; Irawadi, K.; Leschinski, J.; Lindner, F. Green Chem. 2008, 10, 13–30.
2
(11)
Kwon, Y.; Birdja, Y.; Spanos, I.; Rodriguez, P.; Koper, M. T. M. ACS Catal. 2012, 2, 759–764.
3
(12)
Fongarland, P.; Mimura, N. Green Chem. 2011, 13, 1960–1979.
4
(13)
Svitel, J.; Sturdik, E. J. Ferment. Bioeng. 1994, 1783, 351–355.
5 6
(14)
Gomes, J. F.; de Paula, F. B. C.; Gasparotto, L. H. S.; Tremiliosi-Filho, G. Electrochim. Acta 2012, 76, 88–93.
7
(15)
Kwon, Y.; Schouten, K. J. P.; Koper, M. T. M. ChemCatChem 2011, 3, 1176–1185.
8
(16)
Know, Y.; Hersbach, T. J. P. .; Koper, M. T. M. Top. Catal. 2014, 57, 1272–1276.
9
(17)
Kwon, Y.; Jong, E. De; Waal, K. Van Der; Koper, M. T. M. ChemSusChem 2015, 8, 970–973.
10
(18)
Fiaty, K.; Laurent, P.; Bardot, C. Catal. Today 2000, 56, 173–178.
11
(19)
Science, B.; Surgery, D.; Road, I. J. Appl. Electrochem. 1981, 11, 261–267.
12
(20)
Lai, S. C. S.; Koper, M. T. M. Faraday Discuss. 2009, 140, 399.
13 14
(21)
Clavilier, J.; Faure, R.; Guinet, G.; Durand, R. J. Electroanal. Chem. Interfacial Electrochem. 1979, 107, 205–209.
15
(22)
Climent, V.; Feliu, J. M. J. Solid State Electrochem. 2011, 15, 1297–1315.
16
(23)
Kwon, Y.; Koper, M. T. M. Anal. Chem. 2010, 82, 5420–5424.
17 18
(24)
Wonders, A. H.; Housmans, T. H. M.; Rosca, V.; Koper, M. T. M. J. Appl. Electrochem. 2006, 36, 1215–1221.
19
(25)
Garcıa, G.; Rodrıguez, P.; Rosca, V.; Koper, M. T. M. Langmuir 2009, 25, 13661–13666.
20
(26)
Kresse, G.; Hafner, J. Phys. Rev. B 1993, 47, 558–561.
21
(27)
Kresse, G.; Hafner, J. Phys. Rev. B 1994, 49, 14251–14269.
22
(28)
Kresse, G.; Furthmüller, J. Phys. Rev. B 1996, 54, 11169–11186.
23
(29)
Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6, 15–50.
24
(30)
Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865–3868.
25
(31)
Blöchl, P. E. Phys. Rev. B 1994, 50, 17953–17979.
26
(32)
Kresse, G. Phys. Rev. B 1999, 59, 1758–1775.
27 28
(33)
Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.; Jónsson, H. J. Phys. Chem. B 2004, 108, 17886–17892.
29 30
(34)
Rodes, A.; Zamakhchari, M. A.; El Achi, K.; Clavilier, J. J. Electroanal. Chem. Interfacial Electrochem. 1991, 305, 115–129. 26 ACS Paragon Plus Environment
Page 27 of 27
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Catalysis
1 2
(35)
Clavilier, J.; Albalat, R.; Gomez, R.; Orts, J. M.; Feliu, J. M.; Aldaz, A. J. Electroanal. Chem. 1992, 330, 489–497.
3 4
(36)
Cuesta, A.; Couto, A.; Rincón, A.; Pérez, M. C.; López-Cudero, A.; Gutiérrez, C. J. Electroanal. Chem. 2006, 586, 184–195.
5
(37)
Markovic, N. M.; Ross, P. N. J. Surf. Sci. Rep. 2002, 45, 117–229.
6 7
(38)
Lebedeva, N. P.; Koper, M. T. M.; Feliu, J. M.; Van Santen, R. A.; . J. Electroanal. Chem. 2002, 524, 242–251.
8 9
(39)
Lebedeva, N. P.; Rodes, A.; Feliu, J. M.; Koper, M. T. M.; Van Santen, R. A.; . J. Phys. Chem. B 2002, 106, 9863–9872.
10 11
(40)
Schnaidt, J.; Heinen, M.; Denot, D.; Jusys, Z.; Jürgen Behm, R. J. Electroanal. Chem. 2011, 661 (1), 250–264.
12 13
(41)
Avramov-Ivic, M.; Léger, J.-M.; Beden, B.; Hahn, F.; Lamy, C. J. Electroanal. Chem. 1993, 351, 285–297.
14
(42)
Iwasita, T.; Nart, F. C.; Lopez, B.; Vielstich, W. Electrochim. Acta 1992, 37, 2361–2367.
15
(43)
Jeffery, D. Z.; Camara, G. A. Electrochem. commun. 2010, 12, 1129–1132.
16 17
(44)
Bott-Neto, J. L.; Garcia, A. C.; Oliveira, V. L.; de Souza, N. E.; Tremiliosi-Filho, G. J. Electroanal. Chem. 2014, 735, 57–62.
18
(45)
Liu, B.; Greeley, J. J. Phys. Chem. C 2011, 115, 19702–19709.
19 20 21
(46)
Angyal, S. J. The Lobry de Bruyn–Alberda van Ekenstein transformation and related reactions, in: Glycoscience: epimerisation, isomerisation and rearrangement reactions of carbohydrates. Springer-Verlag, Berlin 2001, 125, 1–14.
22 23
TOC Graphic
24 27 ACS Paragon Plus Environment
