Selective production of dihydroxyacetone and glyceraldehyde by photo-assisted oxidation of glycerol

Article abstract of DOI:10.1016/j.cattod.2019.09.035

Glycerol is a by-product during biodiesel production and represents a potential low-cost raw material for obtaining high-cost products like Dihydroxyacetone (DHA) and glyceraldehyde (GCD) amongst others. In this work, Fe-Pillared clay (Fe-PILC) was assessed as catalyst of the selective photo-oxidation of glycerol to obtain DHA and GCD at moderate conditions (298 K and atmospheric pressure). This was conducted in a 100 mL Pyrex glass batch reactor where a Pen-Ray lamp of mercury of 5.5 W UV light (UVP) was placed at the centre. The Fe-PILC was prepared by ion exchange. The pillaring was confirmed by XRD, and a 17% w/w of Fe was determined by Atomic Absorption Spectroscopy. The active phases were established by XPS and found to be FeO and Fe₃O₄. The specific surface area of the clay (bentonite), determined by N₂ physisorption, increased from 34 m²/g to 227 m²/g and the pore volume increased from 0.058 cm³/g to 0.106 cm³/g. The studied variables were catalyst loading and glycerol initial concentration. An experiment with TiO₂ Degussa P25 was also performed as reference. It was found that by adding Fe-PILC to the glycerol oxidation system, selectivity towards DHA or GCD can be tuned. A selectivity towards DHA was found to be 87% with 0.1 g/L of Fe-Pillared after 8 h reaction. The in situ production of H₂O₂ was observed and therefore concluded that the glycerol oxidation occurs via a fenton process, i.e. via free radicals.

Full text of DOI:10.1016/j.cattod.2019.09.035

Journal Pre-proof
Selective production of Dihydroxyacetone and Glyceraldehyde by
Photo-assisted Oxidation of Glycerol
´
Arisbeht Mendoza, Rub´ı Romero, Galilea P. Gutierrez-Cedillo,
´
´
Gustavo Lopez-Tellez, Omar Lorenzo-Gonzalez, Rosa Mar´ıa
´
Gomez-Espinosa, Reyna Natividad
PII:
S0920-5861(19)30532-2
DOI:
https://doi.org/10.1016/j.cattod.2019.09.035
CATTOD 12488
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
5 March 2019
12 September 2019
23 September 2019
´
´
Please cite this article as: Mendoza A, Romero R, Gutierrez-Cedillo GP, Lopez-Tellez G,
´
´
Lorenzo-Gonzalez O, Gomez-Espinosa RM, Natividad R, Selective production of
Dihydroxyacetone and Glyceraldehyde by Photo-assisted Oxidation of Glycerol, Catalysis
Today (2019), doi: https://doi.org/10.1016/j.cattod.2019.09.035
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2019 Published by Elsevier.

Selective production of Dihydroxyacetone and Glyceraldehyde by Photo-assisted Oxidation of
Glycerol
Arisbeht Mendoza, Rubí Romero*, Galilea P. Gutiérrez-Cedillo, Gustavo López-Tellez, Omar
Lorenzo-González, Rosa María Gómez-Espinosa, Reyna Natividad*
Centro Conjunto de Investigación en Química Sustentable, UAEM-UNAM, Universidad Autónoma
del Estado de México, Mexico, km 14.5 Toluca-Atlacomulco-Road, CP 50200
Corresponding authors e-mail: rnatividadr@uaemex.mx, rromeror@uaemex.mx, +52 722 2766610,
Ext.7723.
Graphical abstract

Photolysis
0.05 g/L Fe-PILC
0.1 g/L Fe-PILC
0.2 g/L Fe-PILC
0.2 g/L TiO2
100
80
60
40
20
0
0.1 g/L TiO2
0
1
2
3
4
5
6
7
8
9
10
Time reaction (h)
(a)
HIGHLIGHTS



Fe-PILC enhances photo-oxidation of glycerol towards didhydroxyacetone
Fe-PILC aids to conduct selective oxidations
H2O2 is produced during glycerol photo-oxidation
Abstract
Glycerol is a by-product during biodiesel production and represents a potential low-cost raw
material for obtaining high-cost products like Dihydroxyacetone (DHA) and glyceraldehyde (GCD)
amongst others. In this work, Fe-Pillared clay (Fe-PILC) was assessed as catalyst of the selective
photo-oxidation of glycerol to obtain DHA and GCD at moderate conditions (298 K and
atmospheric pressure). This was conducted in a 100 mL Pyrex glass batch reactor where a Pen-Ray

lamp of mercury of 5.5 Watts UV light (UVP) was placed at the centre. The Fe-PILC was prepared
by ion exchange. The pillaring was confirmed by XRD, and a 17% w/w of Fe was determined by
Atomic Absorption Spectroscopy. The active phases were established by XPS and found to be FeO
and Fe3O4. The specific surface area of the clay (bentonite), determined by N2 physisorption,
increased from 34 m2/g to 227 m2/g and the pore volume increased from 0.058 cm3/g to 0.106
cm3/g. The studied variables were catalyst loading and glycerol initial concentration. An experiment
with TiO2 Degussa P25 was also performed as reference. It was found that by adding Fe-PILC to
the glycerol oxidation system, selectivity towards DHA or GCD can be tuned. A selectivity towards
DHA was found to be 87% with 0.1 g/L of Fe-Pillared after 8 h reaction. The in situ production of
H2O2 was observed and therefore concluded that the glycerol oxidation occurs via a fenton process,
i.e. via free radicals.
Keywords
Photocatalysis, Fe-PILC, selective oxidation, Fenton, H2O2 production, H2O2 dissociation
1. Introduction
In recent years, the energy demand has increased due to urbanization and rapid population growth.
As a result of this potential demand and to the problem of climate change due to the use of fossil
fuels, it becomes imperative to find sustainable solutions in the short and medium term, such as
biofuels (biodiesel). Biodiesel is considered an important alternative source of energy and it is the
best candidate to be used in diesel engines for transport and heavy machinery without having to
make major changes in them, in addition to advantages such as high biodegradability, reduction of
carbon monoxide and oxide emissions of nitrogen, in its combustion organic aromatic products are
not generated, decreases the emission of solid particles and the greenhouse effect, does not contain
sulfur and it is highly biodegradable in water [1]. The glycerol or glycerin is one by-product from
production of biodiesel (10 wt % of biodiesel is glycerol with a purity between 80-88%) and the
estimated global production of biodiesel for 2018 is 35405 millions of liters and forecast value for
2023 is 40259 millions of liters [2], which translates to 4025 millions of liters of glycerol produced

in the market. With this potential increase in the amount of crude glycerol is necessary to have
various methods for the elimination and/or use of the crude glycerol. To cope with excess glycerol,
it is necessary to develop innovative and ecological catalytic processes to convert glycerol into
higher economic value products, which is essential for the present and future of the crude glycerol
market. The new applications of crude glycerol can indirectly help reduce the production costs of
biodiesel and turn glycerol into a fundamental part of renewable energy. Nowadays, glycerol has
been used in livestock feed, biogas, thermo-chemical products, combustion and as additive of
pharmaceutics and cosmetics products [3]. Nevertheless, the crude glycerol obtained from the
production of biodiesel contains many impurities (alcohol, organic or inorganic salts, water,
glycerides traces, soap and catalysts). Therefore, the purification of crude glycerol is required if it is
to be used in the pharmaceutical, cosmetic and food industries. These industries require purities up
to 95 and 99% [4]. The purification of crude glycerol to above 99% for use in the pharmaceutical,
food or cosmetic industries, is an expensive process [3]. Therefore, it is necessary to find new
applications where the unrefined glycerol can be used, and in this way, obtain added value products.
This could notably increase the economic aspect of overall biodiesel production, providing new
applications and outlets for large quantities of crude glycerol obtained from biodiesel industry.
The interest on obtaining value-added products have motivated to study several pathways of
degradation using selective catalysts towards some specific products. Some promising routes for the
valorization of glycerol are: dehydration, hydrogenolysis, dehydroxylation, polycondensation,
selective oxidation and fermentation [5–7].
Oxidation processes are the most commonly assessed to convert glycerol in a large number of high-
value added products such as: glyceraldehyde (GAD), dihydroxyacetone (DHA), hydroxyl pyruvic
acid, glyceric acid, tartronic acid, glycolic acid, glyoxylic acid, oxalic acid and formic acid [8,9].
Within these compounds, Dihydroxyacetone (DHA) occupies a very important place in the market,
due to its use in the pharmaceutical and cosmetic industry. Also, it is used as a monomer in
polymeric biomaterials [10]. Because of its use in the food industry, another important compound is

Glyceraldehyde (GAD) and is produced by the oxidation of the primary hydroxyl groups of
glycerol. In order to improve the selectivity towards either of these compounds and the
sustainability of the whole process, different strategies have been assessed. In this context, a
promising approach has been heterogeneous photocatalysis since allows the use of atmospheric
conditions and relative low-cost oxidizing agents such as air, oxygen and peroxide to conduct the
selective oxidation of glycerol [11]. The most commonly assessed catalyst has been TiO2 and the
main studied variables are the type and the percentage of doping elements such as Au, Pt and Pd,
with the disadvantage that they are deactivated at high reaction times, as a consequence of oxygen
poisoning [10,12]. It has been shown that these variables and reaction conditions have a significant
influence on the selectivity attained towards a specific product.
Thus, it is the aim of this work to assess the plausibility and effect of using iron-pillared clays (Fe-
PILC) to conduct the photo-assisted selective oxidation of glycerol towards DHA. This catalyst has
been proven to be effective on enhancing oxidation processes conducted by photo-fenton [13].
Interestingly enough, in this work, the addition of hydrogen peroxide was not necessary to conduct
the partial oxidation of glycerol, but it was produced in-situ and its dissociation was catalyzed by
Fe-PILC.
2. Experimental Section
2.1 Reagents
The reagents used for the synthesis of pillared clays were: Bentonite (pure-grade) supplied by
Fisher Scientific with a >2 μm particle size and a cation exchange capacity of 94 meq/100g, Ferric
Chloride Hexahydrate (FeCl3·6H2O with purity of 98.7%, Fermont), Sodium hydroxide (NaOH,
98.6%, Fermont), hydrochloric acid (37%) and deionized water (HYCEL). For the atomic
absorption analysis, a Fe standard solution obtained from Spectro pure was utilized. To conduct the
photocatalytic experiments, anhydrous glycerol (C3H5(OH)3, 99.9%) supplied by J.T.Baker was

employed and for the analysis of products, standards provided by Sigma-Aldrich (glyceraldehyde
and dihydroxyacetone) were utilized.
2.2 Synthesis and catalyst characterization
The method used for the catalyst synthesis has been described in detail by Valverde and Martín del
Campo [13,14] and is summarized here as follows: 300 mL of aqueous solution of ferric chloride
Hexahydrated (0.2M) was dropwise added to 600 mL of aqueous sodium hydroxide solution (0.2
M) under continuous stirring, after that the new solution was kept for further 4 hours under
continuous stirring and pH of 1.8  0.02 by using hydrochloric acid (5 M). The resulting solution
was dropwise added to 1000 mL clay suspension (0.1% wt) under stirring, the final solution was
aged at room temperature and the stirring continued for further 18 hours. Afterwards, the solid was
obtained by filtration and washed with distilled and deionized water, dried overnight at 75 °C and
calcined for 2h at 400°C. The catalyst was then labeled as Fe-PILC and characterized by X-ray
Diffraction (XRD), Atomic Absorption Spectroscopy (AAS), N2 Physisorption and X-ray
Photoelectron Spectroscopy (XPS).
The XPS spectra of the Fe-PILC were obtained in a Photoelectron spectrometer Jeol JPS 9200
(standard MG-Kα excitation source (1253.6 eV) and binding energies were calibrated with
regarding to carbon signal (C 1s) at 285 eV). XRD analysis was conducted in a Bruker Advance 8
equipment (Cu-Kα radiation at 35 kV and 30 mA and collection was conducted from 0 to 40° (2θ)
with a step of 0.004°/min). For the AAS analysis, a catalyst sample was diluted with a HF solution
and analyzed with an AA240FS VARIAN spectrometer. This was conducted in order to establish
the real iron content in the prepared material. The specific surface areas of bentonite and Fe-PILC
were calculated by nitrogen physisorption and this was carried out in a Quantachrome Autosorb
analyzer (adsorption relative pressure P/Po=0.99 and 77° K, degassing conditions at 250°C for 2h
under vacuum of 6.6 x10-9 bar).

2.3 Selective photo-oxidation of glycerol
A jacketed Pyrex glass batch photoreactor was used for evaluation of Fe-PILC in the photo-
oxidation of glycerol. The reactor dimensions were 20 cm of height and 2.5 cm of inner diameter.
Inside, at the center of the reactor, an UVP Pen-Ray lamp of mercury of 5.5 Watts UV light was
placed. This lamp emits primary energy at 254 nm with a typical intensity of 4400 μW/cm2 and uses
a power supply of 115 V/60 Hz. All experiments were conducted under continuous stirring and
temperature was kept constant at 298 K by means of a cooling jacket.
For the reaction, firstly, 100 mL of solution of glycerol were loaded into the photoreactor along
with a specific amount of Fe-PILC, immediately the lamp was turned on and the reaction started.
The investigated variables were catalyst loading and type of catalyst. When catalyst loading was
assessed, the experiments were performed with a 0.1M glycerol solution and four different Fe-PILC
loadings were assessed: 0, 0.05, 0.1, and 0.2 g/L under T= 298±2 K and atmospheric pressure of
0.77 atm. With a catalyst loading of 0.2 g/L, the effect of type of catalyst (Fe-PILC and TiO2
Degussa P25) was studied. An experiment with TiO2 Degussa 25 was conducted in order to have as
reference results obtained with a typical photocalyst.
1 mL samples were taken at different reaction times and centrifuged to remove the catalyst, after
that they were filtered with a Thermo Scientific filter of 30 mm Nylon of 0.2µm for analysis by
Ultra high-Performance Liquid Chromatography (UHPLC). Once the concentration of products was
determined, the selectivity was calculated by means of Eq. 1,
푚푀
푖
푆푖 =
∗ 100
푗
Eq. 1
푚푀 +푚푀
푖
where Si= selectivity of compound i (dihydroxyacetone or glyceraldehyde), mM
2.4 Analytical methods
The products generated by photo-oxidation of glycerol using Fe-PILC were analyzed by ultrahigh
performance liquid chromatography (UHPLC). The equipment used for this purpose was a Thermo

Scientific Vanquish with a refractive index detector (RefractoMax520, Thermo Scientific). A
Carbomix H NP 5: 8% (4.6 x 300, 5µm Non-Porous) column was used at 353 K. A mobile phase
consisting of H2SO4 solution (10 mM) at constant flow rate of 0.26 mL/ min was employed. The
reaction products were identified and quantified with the corresponding calibration curves. All
measurements were conducted by triplicate and an error of 2% was established.
3. Results
3.1 Fe-PILC characterization
Figure 1 shows the diffractogram of the synthesized Fe-PILC. The diffractogram of the bentonite
clay prior the pillaring process was also obtained. This is not shown here since has already been
published somewhere else [13]. Figure 1 is the evidence that pillaring process was successfully
conducted. The resulting basal spacing was 21.43 A. This is specifically indicated by the reflection
d001 that appears at low angles (2θ=3°-4°) in figure 1, this is the evidence of the basal spacing
enlargement characteristic when the metal becomes a pillar [13–15]. On the other hand, the
presence of peaks at around 6° to 10° could give evidence that the species are intercalated in an
irregular manner [16]. Reflections at around 21° and 27° are characteristic of bentonite [15]. There
can also be observed in figure 1 some reflections that might be ascribed to magnetite and goethite.
The results show an increase in surface area and pore volume after the pillaring process. Regarding
surface area, this increased from 34 m2/g to 227 m2/g and the pore volume increased from 0.058
cm3/g to 0.106 cm3/g. This can be ascribed to the pillars formed among the clay layers [17].
Figure 2 depicts the XPS spectra and two deconvolution curves are shown. They were obtained
from the binding energy corresponding to Fe 2p3/2. According to the National Institute of Standards
and Technology (NIST), the first deconvolution curve corresponds to Fe3O4 (709.2 eV) and the
second one can be ascribed to Fe2+ in FeO (709.6 eV).
3.2. Selective photo-oxidation of glycerol

3.2.1. Effect of catalyst loading and catalyst type
The study of this variable is rather important since allows to elucidate whether or not a
heterogeneous process is being under the control of transport phenomena or under the control of the
surface reactivity. It also dictates the product distribution of a consecutive/parallel reaction system
[18–20]. In this context, figure 3 shows the DHA and GCD concentration profiles depending on
time and Fe-PILC loading. Photolysis was included as reference and is considered as zero catalyst
loading. The obtained results with TiO2 at two catalyst loadings, 0.1 and 0.2 g/L, are also presented
in figure 3.
Within glycerol photo-catalyzed oxidation related literature [21], it seems that the effect of UV light
on glycerol molecule has been underestimated. In this work, however, we observed that both, DHA
and GCD, were generated by only irradiating the 0.1 M glycerol solution without the addition of
any catalyst. These results are shown in Fig. 3 and it can be observed that the accumulated
concentration of DHA over 8 hours of reaction was 0.22 mM (20.04 mg/L). GCD, however,
exhibits a rather different behavior since its concentration reaches a stationary state after only 1
hour of reaction at 0.08 mM (7.3 mg/L). This indicates that at this point, the rate of GCD
accumulation becomes negligible and this implies that GCD is being consumed at the same rate that
is being produced. GCD might be converting to other oxidation products like glycolic and glyceric
acid. It can also be observed that during the first hour of reaction, GCD production rate (0.07 x10-3
mol/Lh) is about twice faster than DHA one (0.035 x 10-3 mol/Lh). This is expected since the
reactivity of the OH in C1 is higher than in C2 and thus the first hydrogen abstraction is light
driven. It is also observed that the temporal DHA and GCD profiles obtained with TiO2 are rather
similar to those obtained by photolysis (figure 3). Actually, the DHA temporal profile is practically
the same with TiO2 than with light only. This was observed only when a TiO2 loading of 0.2 g/L is
employed, however, when a lower titania loading is used (0.1 g/L) a lower DHA concentration is
achieved. This suggests that at low concentrations, titania competes for photons with glycerol and
this negatively affects glycerol selective oxidation towards DHA (figure 3a).

It is also observed in figure 3, that both, DHA and GCD concentration profiles depend on Fe-PILC
loading. The DHA concentration profile obtained at low catalyst loading, i.e. 0.05 g/L, shown in
figure 3 a), suggests that the Fe-PILC does not compete with the effect of light and apparently does
not exhibit any photo-activity. Furthermore, the effect of this solid seems to worsen GCD
production compared to photolysis (Fig. 3b). However, a further increase in the catalyst loading
leads to an increase in DHA concentration and selectivity towards this compound (Figure 4a). The
highest DHA concentration (0.54mM) is observed with the catalyst loading of 0.1 g/L and this can
be ascribed to the generation of hydroxyl radicals by means of reaction of eq.1 [22],
퐹푒3+ + 퐻2푂 + ℎ푣 → 퐹푒2+ +∙ 푂퐻 + 퐻+ − − − (1)
Fig 3b) shows the concentration profiles of GCD obtained with different catalyst loading. In this
graph it can be observed that the catalyst loading of 0.2 g/L does not follow the same trend with
respect to the others. This behavior can be attributed to the prevailing reaction mechanism, when
glycerol is in contact with Fe may be forming a ligand with the ·OH of the third carbon of glycerol,
which by photons absorption produces GCD [23,24], where the limiting step can be the desorption
of the products to the bulk fluid. The production of GCD was 0.15mM after 8 h of reaction. In the
other reactions shown in Fig 3 b), it is observed that the production of GCD is kept constant and this
behavior could be due to the transformation of GCD to other products such as glycolic acid which
was identified but not quantified due to the very low observed signal.
It can also be observed that with 0.05 g/L, the formation of GCD is lower than under photolysis.
This is suggesting that the addition of the Fe-PILC is limiting the photons usage by glycerol. This
supports the statement that the initial stage of the glycerol oxidation is light driven and consists in
the hydrogen abstraction (Scheme 1). This can subsequently react with oxygen from the medium
(reaction 2) and form the hydroperoxyl radical that may lead to form hydrogen peroxide by reaction
3,
∙ 퐻 + 푂2 → 퐻푂2∙ − − − − − − − −(2)

∙
∙
( )
퐻푂2 + 퐻푂2 → 퐻2푂2 + 푂2 − − − − 3
Actually, the in-situ production of hydrogen peroxide was verified and its evolution was established
when using 0.2 g/L of Fe-PILC. This was plotted in figure 4. The generation of Hydrogen Peroxide
during the photo-oxidation of glycerol was evaluated using a colorimetric method [25], that
consisted on the formation of pertitanic acid (410 nm). Once H2O2 is produced, its dissociation is
expected to readily occur via UV light action (reaction 4) and by the catalytic action of Fe2+
(reaction 5) [26],
H2O2 + h  2OH
(4)
Fe2+ + H2O2  Fe3+ + OH- + OH
(5)
Thus, as depicted in scheme 1, the production of GCD and DHA will be limited by the generated
hydroxyl radicals, i.e. produced H2O2 (Fig. 4).
Regarding selectivity, it can be observed in figure 5 that selectivity towards DHA or GCD is not
only a function of time but also of catalyst loading and catalyst type. In any case, it can be
concluded that Fe-PILC, at the studied conditions, outperforms Degussa P25 since a higher
selectivity towards DHA and glyceraldehyde is obtained via the Fenton process. In this sense, it is
worth noticing that the selectivity towards DHA or GCD depends on the catalyst loading and this
suggests that at the end what is influencing the product distribution is the concentration of reactive
oxidant species. This is in concordance with that previously reported [27]. On the other hand, the
results obtained with any of the TiO2 loadings, showed less intermediaries production than Fe-
PILC, and this can be due to a different oxidation mechanism [11,27].
3.2.2 Effect of initial glycerol concentration.
Two initial concentration of glycerol, 0.1M and 0.05M were evaluated with 0.1 g/L of Fe-PILC
(Fig. 6). It can be observed that the production rate of both, DHA and GCD, directly depends on

glycerol initial concentration. This is expected since a higher amount of glycerol will also lead to a
higher concentration of hydrogen peroxide and if there is the right amount of catalyst, such
hydrogen peroxide will be readily dissociated thus originating the oxidant species to increase the
glycerol oxidation rate which is translated into a higher production of DHA.
4. Conclusions
Iron pillared clay (Fe-PILC) can be used to enhance the selectivity of the glycerol photo-oxidation
towards dihydroxyacetone (DHA) or glyceraldehyde (GCD). Hydrogen peroxide is produced during
such a reaction. Therefore, the production of DHA is plausible to be due to the oxidation of glycerol
via the generated hydroxyl radicals produced by the dissociation of the in situ produced H2O2. The
production rate and selectivity towards DHA is higher with Fe-PILC than with the typical
photocatalyst TiO2.
Acknowledgements
Arisbeht Mendoza acknowledges CONACYT for financial support to conduct postgraduate studies,
CONAYCT project 269093 and UAEM project 4373 are also acknowledged. Authors are grateful
to Alejandra Núñez Pineda, Citlalit Martínez Soto, Uvaldo Hernandez Balderas, Lourdes Hurtado
and Deysi Amado Piña for technical support.

References
[1]
[2]
[3]
A. Demirbas, Biodieselꢀ: a realistic fuel alternative for diesel engines, Springer-V, London,
2008.
OECD-FAO, OECD-FAO Agricultural Outlook 2014–2023-Commodity Database (OECD,
Paris, 2015)., 2014. doi:dx.doi.org/10.1787/agr_outlook-2012-en.
M. Ayoub, A.Z. Abdullah, Critical review on the current scenario and significance of crude
glycerol resulting from biodiesel industry towards more sustainable renewable energy
industry, Renew. Sustain. Energy Rev. 16 (2012) 2671–2686.
doi:10.1016/j.rser.2012.01.054.
[4]
[5]
Farm Energy, New Uses for Crude Glycerin from Biodiesel Production, EXtension. (2012)
20–23. doi:10.3390/su9101734.
L. GONG, Y. LÜ, Y. DING, R. LIN, J. LI, W. DONG, T. WANG, W. CHEN, Solvent
Effect on Selective Dehydroxylation of Glycerol to 1,3-Propanediol over a Pt/WO3/ZrO2
Catalyst, Chinese J. Catal. 30 (2009) 1189–1191. doi:10.1016/S1872-2067(08)60142-4.
J. Xu, Y. Zhao, H. Xu, H. Zhang, B. Yu, L. Hao, Z. Liu, Selective oxidation of glycerol to
formic acid catalyzed by Ru(OH)4/r-GO in the presence of FeCl3, Appl. Catal. B Environ.
154–155 (2014) 267–273. doi:10.1016/j.apcatb.2014.02.034.
Y. Jun Choi, J. Hwan Park, T. Yong Kim, S. Yup Lee, Metabolic engineering of Escherichia
coli for the production of 1-propanol, Metab. Eng. 14 (2012) 477–486.
doi:10.1016/j.ymben.2012.07.006.
F.D. Elziebeta Skrzynska, Anna Wondolowska-Grabowska, Mickaël Capron, Crude glycerol
as a raw material for the liquid phase oxidation reaction, 482 (2014) 245–257.
doi:10.1016/j.apcata.2014.06.005.
P.U. Okoye, B.H. Hameed, Review on recent progress in catalytic carboxylation and
acetylation of glycerol as a byproduct of biodiesel production, Renew. Sustain. Energy Rev.
53 (2016) 558–574. doi:10.1016/j.rser.2015.08.064.
[6]
[7]
[8]
[9]
[10] A. Behr, J. Eilting, K. Irawadi, J. Leschinski, F. Lindner, Improved utilisation of renewable
resources: New important derivatives of glycerol, Green Chem. 10 (2008) 13–30.
doi:10.1039/b710561d.
[11] T. Jedsukontorn, V. Meeyoo, N. Saito, M. Hunsom, Effect of electron acceptors H2O2and
O2on the generated reactive oxygen species1O2and OH·in TiO2-catalyzed photocatalytic
oxidation of glycerol, Cuihua Xuebao/Chinese J. Catal. 37 (2016) 1975–1981.
doi:10.1016/S1872-2067(16)62519-6.
[12] J.A. Sullivan, S. Burnham, The selective oxidation of glycerol over model Au / TiO 2
catalysts — The in fl uence of glycerol purity on conversion and product selectivity,
CATCOM. 56 (2014) 72–75. doi:10.1016/j.catcom.2014.06.026.
[13] E. Martin Del Campo, R. Romero, G. Roa, E. Peralta-Reyes, J. Espino-Valencia, R.
Natividad, Photo-Fenton oxidation of phenolic compounds catalyzed by iron-PILC, Fuel.
138 (2014) 149–155. doi:10.1016/j.fuel.2014.06.014.
[14] J.L. Valverde, A. Romero, R. Romero, P.B. García, M.L. Sánchez, I. Asencio, Preparation
and characterization of Fe-PILCS. Influence of the synthesis parameters, Clays Clay Miner.
53 (2005) 613–621. doi:10.1346/CCMN.2005.0530607.
[15] M. Luo, D. Bowden, P. Brimblecombe, Applied Catalysis Bꢀ: Environmental Catalytic
property of Fe-Al pillared clay for Fenton oxidation of phenol by H 2 O 2, 85 (2009) 201–
206. doi:10.1016/j.apcatb.2008.07.013.
[16] C. Catrinescu, D. Arsene, P. Apopei, C. Teodosiu, Degradation of 4-chlorophenol from
wastewater through heterogeneous Fenton and photo-Fenton process , catalyzed by Al – Fe
PILC, Appl. Clay Sci. 58 (2012) 96–101. doi:10.1016/j.clay.2012.01.019.
[17] T. Undabeytia, M.C. Galán-jiménez, E. Gómez-pantoja, J. Vázquez, B. Casal, F. Bergaya, E.
Morillo, Fe-pillared clay mineral-based formulations of imazaquin for reduced leaching in

soil, Appl. Clay Sci. 80–81 (2013) 382–389. doi:10.1016/j.clay.2013.07.001.
[18] Q. Chen, P. Wu, Z. Dang, N. Zhu, P. Li, J. Wu, X. Wang, Iron pillared vermiculite as a
heterogeneous photo-Fenton catalyst for photocatalytic degradation of azo dye reactive
brilliant orange X-GN, Sep. Purif. Technol. 71 (2010) 315–323.
doi:10.1016/j.seppur.2009.12.017.
[19] O.B. Ayodele, J.K. Lim, B.H. Hameed, Pillared montmorillonite supported ferric oxalate as
heterogeneous photo-Fenton catalyst for degradation of amoxicillin, Appl. Catal. A Gen.
413–414 (2012) 301–309. doi:10.1016/j.apcata.2011.11.023.
[20] B. Iurascu, I. Siminiceanu, D. Vione, M.A. Vicente, A. Gil, Phenol degradation in water
through a heterogeneous photo-Fenton process catalyzed by Fe-treated laponite, Water Res.
43 (2009) 1313–1322. doi:10.1016/j.watres.2008.12.032.
[21] T. Jedsukontorn, N. Saito, M. Hunsom, Photocatalytic behavior of metal-decorated TiO2and
their catalytic activity for transformation of glycerol to value added compounds, Mol. Catal.
432 (2017) 160–171. doi:10.1016/j.mcat.2017.02.022.
[22] L.F. González-bahamón, D. Fernando, N. Benítez, C. Pulgarín, Chemosphere New Fe-
immobilized natural bentonite plate used as photo-Fenton catalyst for organic pollutant
degradation, 82 (2011) 1185–1189. doi:10.1016/j.chemosphere.2010.11.071.
[23] N. Klamerth, S. Malato, A. Agüera, A. Fernández-Alba, Photo-Fenton and modified photo-
Fenton at neutral pH for the treatment of emerging contaminants in wastewater treatment
plant effluents: A comparison, Water Res. 47 (2013) 833–840.
doi:10.1016/j.watres.2012.11.008.
[24] M.N. Timofeeva, S.T. Khankhasaeva, Y.A. Chesalov, S. V. Tsybulya, V.N. Panchenko, E.T.
Dashinamzhilova, Synthesis of Fe,Al-pillared clays starting from the Al,Fe-polymeric
precursor: Effect of synthesis parameters on textural and catalytic properties, Appl. Catal. B
Environ. 88 (2009) 127–134. doi:10.1016/j.apcatb.2008.09.013.
[25] G.M. Eisenberg, Colorimetric Determination of Hydrogen Peroxide, Ind. Energineering
Chem. 15. No.5 (1943) 327–328.
[26] L. Hurtado, R. Romero, A. Mendoza, S. Brewer, K. Donkor, R.M. Gómez-Espinosa, R.
Natividad, Paracetamol mineralization by Photo Fenton process catalyzed by a Cu/Fe-PILC
under circumneutral pH conditions, J. Photochem. Photobiol. A Chem. 373 (2019) 162–170.
doi:10.1016/j.jphotochem.2019.01.012.
[27] C. Minero, A. Bedini, V. Maurino, Glycerol as a probe molecule to uncover oxidation
mechanism in photocatalysis, Appl. Catal. B Environ. 128 (2012) 135–143.
doi:10.1016/j.apcatb.2012.02.014.

Figure Captions
Fig.1 X-Ray Diffractogram of Fe-PILC.
5
10
15
20
25
30
35
40
2
Fig. 1. XRD pattern of Fe-Pilc´s. (•Fe3O4 (magnetite) [JCPDS #89-0688], FeO(OH) (goethite)
[JCPDS #81-0463])
Fig.2. XPS Spectra of Fe-PILC at Fe 2p3/2 region.

709.2
709.6
710.2 710.0 709.8 709.6 709.4 709.2 709.0 708.8 708.6
Binding Energy (eV)
Figure 3. Effect of catalyst loading and catalyst type on a) Dihydroxiacetone, b) Glyceraldehyde
evolution with time. Reaction conditions: [Glycerol]o=0.1 M, T=298 K, stirring speed= 1000 rpm.
Fig. 3 a) Effect of catalyst loading on Dihydroxyacetone (DHA) evolution with time. Reaction
conditions: [Glycerol]o= 0.1 M, T= 298 K, UV wavelength= 254 nm, 1000 r/min.

6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0.2 g/L TiO2
0.1 g/L TiO2
0.2 g/L Fe-PILC
0.1 g/L Fe-PILC
0.05 g/L Fe-PILC
Photolysis
0
1
2
3
4
5
6
7
8
9
10
Reaction Time (h)

2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
0.2 g/L TiO2
0.1 g/L TiO2
0.2g/L Fe-PILC
0.1 g/L Fe-PILC
0.05 g/L Fe-PILC
Photolysis
0
1
2
3
4
5
6
7
8
9
10
Reaction Time (h)
Fig. 3. Effect of catalyst loading on a) Dihydroxiacetone (DHA), b) Glyceraldehyde (GCD)
evolution with time. Reaction conditions: [Glycerol]o=0.1 M, T= 298 K, UV wavelength= 254 nm,
1000 rpm.
Fig. 4. H2O2 evolution. Experimental conditions: [Glycerol]0= 0.1M, stirring speed=1000 rpm, UV
wavelength=254nm, T=298 K, WFe-PILC=0.2 g/L.

2.00
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
0
15
30
45
60
Time (minutes)
Figure 5. Effect of Fe-PILC loading, catalyst type and reaction time on selectivity towards a) DHA
and b) GCD. Reaction conditions: [Glycerol]o=0.1 M, T=298 K, stirring speed= 1000 rpm.

Photolysis
0.05 g/L Fe-PILC
0.1 g/L Fe-PILC
0.2 g/L Fe-PILC
0.2 g/L TiO2
100
80
60
40
20
0
0.1 g/L TiO2
0
1
2
3
4
5
6
7
8
9
10
Time reaction (h)
(a)

120
100
80
60
40
20
0
Photolysis
0.05 g/L Fe-PILC
0.1 g/L Fe-PILC
0.2 g/L Fe-PILC
0.2 g/L TiO2
0.1 g/L TiO2
0
1
2
3
4
5
6
7
8
9
10
Reaction Time (h)
(b)
Fig. 5 Effect of catalyst loading and type, and reaction time on selectivity towards a)
Dydroxyacetone (DHA), b) Glyceraldehyde (GCD). Reaction conditions: [Glycerol]o = 0.1 M, UV
wavelength= 254 nm, 1000 rpm, T= 298 K ± 2.
Fig. 6. Effect of initial glycerol concentration on GCD and DHA temporal profiles. Reaction
conditions: stirring speed=1000 rpm, UV wavelength=254nm, Fe-PILC loading= (0.1 g/L) and
T=298 K.

6
5
4
3
2
1
0
DHA (0.05M Glycerol)
DHA (0.1M Glycerol)
GCD (0.05M Glycerol)
GCD (0.1M Glycerol)
0
1
2
3
4
5
6
7
8
9
10
Reaction Time (h)
Fig. 6. Concentration of GCD and DHA as a function of initial concentration of Glycerol catalyzed
with Fe-PILC. Reaction conditions: UV wavelength= 254 nm, T=298 K ±2, 1000 rpm, Wcat=0.1
g/L
Scheme 1. Glycerol photo-oxidation reaction scheme.

H
H
OH
OH
HO
Glycerol
+ OH
+ OH
+
UV (254 nm)
O
OH
OH
HO
O
HO
Glyceraldehyde
Dihydroxyacetone
+ OH
+ OH
O
HO
OH
Glycolic acid