Superparamagnetic γ-Fe2O3 nanoparticles as an easily recoverable catalyst for the chemical recycling of PET

Article abstract of DOI:10.1039/c3gc41834k

There have been numerous studies to develop catalysts for the chemical recycling of poly(ethylene terephthalate) (PET) via glycolysis. However, in the field of PET glycolysis, only a few have attempted to recover and reuse the catalysts. This research utilized easily recoverable superparamagnetic γ-Fe₂O₃ nanoparticles as a reusable catalyst for PET glycolysis. γ-Fe₂O₃ nanoparticles were produced by calcining Fe₃O₄ nanoparticles prepared by the co-precipitation method. The produced γ-Fe₂O₃ nanoparticles had an average size of 10.5 ± 1.4 nm, and a very high surface area reaching 147 m² g^-1. Its superparamagnetic property was also confirmed. Glycolysis reactions were carried out, and the γ-Fe₂O₃ catalysts were recovered after the reactions by simple magnetic decantation. The use of magnetic iron oxide allowed the easy recovery of the catalyst from the glycolysis products. At 300 °C and a 0.05 catalyst/PET weight ratio, the maximum bis(2-hydroxyethlyl) terephthalate (BHET) monomer yield reached more than 90% in 60 min. At 255 °C and a 0.10 catalyst/PET weight ratio, the BHET yield reached more than 80% in 80 min. The catalyst was reused 10 times, giving almost the same BHET yield each time.

Full text of DOI:10.1039/c3gc41834k

View Article Online
View Journal
Green Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: L. D. C. Bartolome, M. Imran, K. G. Lee,
A. A. Sangalang, J. K. Ahn and D. H. Kim, Green Chem., 2013, DOI: 10.1039/C3GC41834K.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Accepted Manuscripts are published online shortly after acceptance, which is prior
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Volume 12 | Number 9 | September 2010 | Pages 1481–1676
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 1463-9262
COMMUNICATION
CRITICAL REVIEW
Luque, Varma and Baruwati
Dumesic et al.
Magnetically seperable organocatalyst Catalytic conversion of biomass
for homocoupling of arylboronic acids to biofuels
1463-9262(2010)12:9;1-U
www.rsc.org/greenchem
Registered Charity Number 207890

Page 1 of 8
Green Chemistry
View Article Online
DOI: 10.1039/C3GC41834K
Superparamagnetic γ-Fe2O3 Nanoparticles as Easily Recoverable
Catalyst for the Chemical Recycling of PET
Leian dC Bartolome,a Muhammad Imran,b Kyoung G. Lee,c Arvin Sangalang,a Jeong Keun
Ahn,d and Do Hyun Kim*, a
Superparamagnetic γ-Fe2O3 is applied as efficient, easily recoverable, and reusable catalyst for the
chemical recycling of PET via glycolysis.
aDepartment of Chemical & Biomolecular Engineering (BK21 Program), KAIST, 291 Daehak-ro, Yuseong-gu,
Daejeon 305-701, Republic of Korea. E-mail: DoHyun.Kim@kaist.ac.kr; Fax: +82 42 350 3910; Tel: +82 42
350 3929
bChemical Engineering Department, King Saud University, Saudi Arabia
cDepartment of Chemical Engineering, University of Michigan, Ann Arbor, MI, 48109-2136, USA
dDepartment of Microbiology, School of Bioscience & Biotechnology, Chungnam National University, Daejeon,
305-764, Republic of Korea

View Article Online
DOI: 10.1039/C3GC41834K
Green Chemistry
Page 2 of 8
Dynamic Article Links ►
Green Chemistry
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
PAPER
Superparamagnetic γ-Fe2O3 Nanoparticles as Easily Recoverable
Catalyst for the Chemical Recycling of PET
Leian dC Bartolome,a Muhammad Imran,b Kyoung G. Lee,c Arvin Sangalang,a Jeong Keun Ahn,d and
Do Hyun Kim*, a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x5
There have been numerous studies to develop catalysts for the chemical recycling of poly (ethylene terephthalate) (PET) via glycolysis.
However, in the field of PET glycolysis, only a few have attempted to recover and reuse the catalysts. This research utilized easily
recoverable superparamagnetic γ-Fe2O3 nanoparticles as reusable catalyst for PET glycolysis. γ-Fe2O3 nanoparticles were produced by
10 calcining Fe3O4 nanoparticles prepared by co-precipitation method. The produced γ-Fe2O3 nanoparticles had average size of 10.5 ± 1.4
nm, and had very high surface area reaching 147 m2/g. Its superparamagnetic property was also confirmed. Glycolysis reactions were
carried out, and γ-Fe2O3 catalysts were recovered after the reaction by simple magnetic decantation. The use of magnetic iron oxide
allowed the easy recovery of the catalyst from the glycolysis products. At 300 °C and 0.05 catalyst/PET weight ratio, the maximum
bis(2–hydroxyethlyl) terephthalate (BHET) monomer yield reached more than 90% in 60 min. At 255 °C and 0.10 catalyst/PET weight
15 ratio, the BHET yield reached more than 80% in 80 min. The catalyst was reused 10 times, giving almost the same BHET yield each
time.
oxide16,17 and graphene oxide-manganese oxide18 catalysts for
melt-phase PET glycolysis. Catalysts fabricated at nanoscale have
increased catalytic efficiency, due to the increased number of
1. Introduction
20 Poly(ethylene terephthalate), more commonly known as PET, is an
indispensable material due to its excellent physical and chemical
properties. It is an important material in the textile industry, and in
food packaging where it has noticeably become the choice for
beverage containers.1 Due to its extensive use, the increasing
25 consumption of PET has led to the creation of large amounts of
waste, inspiring research for recycling. In view of sustainability,
chemical recycling, which leads to the formation of the raw
materials from which PET was originally made, remains the most
viable option.2
55 active sites. In addition, the change of intrinsic properties of the
catalysts at nanoscale usually leads to better catalytic
performance.19 As a result, the metal oxide catalysts gave higher
monomer yield, exceeding 90 %, at a shorter reaction time
compared to the other previously studied catalysts.16
60
In spite of previous research efforts, there is still a problem as
the toxic metals in these catalysts are harmful when released to the
environment. The difficulty of separation of nanomaterials after
the reaction makes the problem even more challenging. The failure
to separate the catalyst from the glycolysis products does not only
30
Various processes have been developed and utilized in PET
chemical recycling. Among them, glycolysis is the most
established and popular, because of its simplicity and low cost. In
addition, glycolysis produces the monomer BHET, which can be
used in both dimethyl terephthalate (DMT) and terephthalic acid
65 pose environmental risks, but may also change the properties of
the products. Thus, there has been a need to find a catalyst that is
both efficient and can easily be recovered.
There have been efforts recently to utilize recoverable catalysts
for PET glycolysis, yet there are still issues that need to be
70 addressed. Wang et al. have reported that various ionic liquid
catalysts for PET glycolysis could be recovered, then used
repeatedly.20-22 In a separate study, Yue et al. used basic ionic
liquid.23 These ionic liquid catalysts had less than 80% BHET yield,
which is evidently lower than the yield from metal oxide
35 (TPA) production lines to make virgin PET, or other advanced
materials.2
Studies on the kinetics show that without catalysts, PET
glycolysis is very slow, and complete conversion of PET to
monomer BHET is almost impossible.3, 4 Metal acetates were the
40 first reported catalysts for PET glycolysis.5-8 Later on, researchers
developed more environment-friendly alternatives like mild
alkalies,9,10 sulfates,10,11 metal chlorides,12 and zeolites.13 However,
even with these aforementioned catalysts, the PET glycolysis still
required long reaction times, and gave low BHET monomer yields.
45 These catalysts could only increase the glycolysis rate at
temperatures below the melting point of PET, and had no catalytic
activity in melt-phase glycolysis due to mass transfer limitations. 14
A list of all the studied catalysts for PET glycolysis can be found
in previous work.15
-
75 catalysts.16 18 More recently, Wang et al. reported urea to be a
reusable catalyst with good low-temperature catalytic activity for
the glycolysis of PET.24 However, the catalyst was recovered using
vacuum distillation, which is an energy intensive process. Thus, to
date, the issues of catalyst efficiency and environmental
80 sustainability have not yet been resolved simultaneously. Eco-
friendly catalysts, such as mild alkalies and chlorides, show less
efficiency compared to the not-so-eco-friendly ones (e.g. metal
oxides) usually containing toxic metals.
In this paper, we present superparamagnetic γ-Fe2O3
85 nanoparticles as an efficient and environment-friendly catalyst for
a sustainable PET chemical recycling. As catalyst, γ-Fe2O3
50
Recently, our group has developed thermally stable and highly
selective silica nanoparticle–supported metal (Mn, Zn, Ce)
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

View Article Online
DOI: 10.1039/C3GC41834K
Page 3 of 8
Green Chemistry
nanoparticles are easily recoverable and reusable. By recovering
nature, a Lakeshore EM4-HVA Vibrating Sample Magnetometer
(VSM) was used to obtain the magnetization curve of the
nanoparticles. Thermogravimetric curve was obtained by a
NETCH TG209F3 Thermogravimetric Analyzer (TGA).
the catalyst, the danger of releasing harmful catalyst components
to the environment is reduced. By reusing the catalyst, the process
becomes more cost-effective and sustainable.
5
As a reusable catalyst, γ-Fe2O3 nanoparticles have shown good 65 Temperature Profile Desorption with NH3 as molecular probe
-
performance a number of reactions. 25 27 However, to the best of
(NH3-TPD) was also done to investigate the acidity of the
produced material.
our knowledge, superparamagnetic γ-Fe2O3 nanoparticles have
never been used in any glycolysis reactions particularly in PET. In
fact, to date, no magnetic nanomaterial has been applied as a
PET Glycolysis using γ-Fe2O3
10 recoverable and reusable solid catalyst in PET glycolysis. Among
the magnetic materials, we selected nano-γ-Fe2O3, because of its
stability, high catalytic activity and magnetic susceptibility.25 In
addition, it is cheap, nontoxic, environment-friendly and abundant.
Most importantly, due to its superparamagnetic property, it can be
15 easily recovered by simple magnetic decantation.
17, 18
70 The glycolysis procedure was based on previous works.
To
compare the performance of γ-Fe2O3 with other previously
reported metal oxides, glycolysis reactions were carried out in a
10-mL stainless steel batch-type reactor at 300 °C, and
approximately 1.1 MPa. A mixture of 0.3 g of virgin PET, 1 mL of
75 99.9 % anhydrous EG, and a designated amount of catalyst (3, 4.5,
6, 15 mg) were loaded into the reactor, which was then placed in a
furnace. After the desired reaction time (40-80 min), the reactor
was removed from the furnace and quenched with cold water.
The performance of γ-Fe2O3 was also compared with urea. The
80 glycolysis reactions were carried out at 255 °C, with 0.1
catalyst/PET weight ratio.
2. Experimental
Materials
For the preparation of γ-Fe2O3 nanoparticles, the FeCl2 and FeCl3
20 precursors were purchased from Sigma-Aldrich. HCl, Citric acid
granules, and NaOH pellets were from Junsei Chemical. For the
PET glycolysis, the PET pellets were provided by HUVIS. The
pellets were ground to fine powder with mean diameter of 150 μm.
Ethylene glycol (EG) and the standard BHET were purchased from
25 Sigma-Aldrich, and the THF used as solvent was purchased from
J.T. Baker. The chemicals were used without any further
purification. Deoxygenated deionized (DI) water was used all
throughout the experiment.
The BHET monomer yield was measured with Agilent 1000
series Reverse Phase High Performance Liquid Chromatography
(HPLC). The use of HPLC for the qualitative and quantitative
85 analysis of the BHET monomer was discussed in detail in a
28
previous work.
After quenching the reaction, the glycolysis
product was dissolved in 20 mL THF. From this solution, the
catalysts were recovered via magnetic decantation and reused. For
the HPLC sample, 0.1 mL was taken out from the solution and
90 diluted to 24 mL 50/50 (v/v) THF/H2O solution. This solution was
used in the HPLC analysis. The UV detector was set to 254 nm,
and the flow rate of the mobile phase was set to 1.0 mL/min. The
mobile phase used was 50/50 (v/v) THF/H2O solution. The BHET
monomer yield was calculated by the equation
30 Synthesis of γ-Fe2O3 as catalyst and its characterization
The γ-Fe2O3 nanoparticles were produced by calcining Fe3O4
95
moles of BHET produced
nanoparticles prepared through the traditional co-precipitation
=
Yield of BHET mol%
× 100 (1)
28,29
moles of PET units
method
with slight modifications. To make the Fe3O4
nanoparticles, an aqueous solution of Fe2+/Fe3+ with molar ratio of
35 1:2 was prepared by dissolving pre-determined amounts of FeCl2
and FeCl3 precursors in an HCl solution (0.4 M, 25 mL). To this
Fe2+/Fe3+ aqueous solution, NaOH (1.5 M, 250 mL) was added
rapidly with vigorous stirring. The addition of NaOH solution
instantly produced black precipitate, a clear indication that Fe3O4
40 formed. Citric acid (0.2 M, 50 mL) was added as surfactant to
control the nanoparticle size. This procedure was done under
nitrogen atmosphere.
The black Fe3O4 precipitate was then separated via magnetic
decantation. It was washed with water, and centrifuged at 4000
45 rpm for 4 min. This was done four times. Finally, the precipitate
was washed with 0.01 M HCl solution to neutralize the anionic
charges on the surface of the nanoparticles.30 The obtained Fe3O4
precipitate was dried, and then calcined at 210 °C for 3 hours, to
produce γ-Fe2O3.
Recovery of γ-Fe2O3 catalyst and BHET
To evaluate the recoverability of the catalyst, another set of
glycolysis reactions were carried out. After the reaction, the
100 glycolysis product was dissolved in water at 90 °C, in which
BHET, its dimers, and EG are soluble. From this solution, the
catalyst was recovered via simple magnetic decantation. The
recovered catalyst was washed with deoxygenated DI water three
times, dried, and finally weighed to measure the amount recovered.
105
BHET from the glycolysis product was also recovered for
further characterization. After the catalyst was recovered, the
solution was kept at 90 °C for 45 min under vigorous stirring. The
PET residual was separated from the mixture through filtration.
When the resulting filtrate reached 60 oC, it was re-filtered to
110 remove the dimers from the BHET monomers. The resulting
filtrate from the second filtration was refrigerated for 24 h, and
then filtered to recover the monomer BHET. The recovered BHET
crystals were dried, and characterized by Nuclear Magnetic
Resonance Spectroscopy (NMR).
50
The morphology of the produced nanoparticles was observed
using JEOL JEM-2100F 200 kV transmission electron
a
microscope (TEM). Its specific surface area was estimated using
the N2 adsorption isotherm obtained by Micrometrics ASAP 2020
BET surface area and porosimetry analyzer. To evaluate the
115
55 crystallinity of the material, X-ray diffraction (XRD) spectra with
2θ ranging from 10° to 80⁰ were obtained using RIGAKU
D/MAX-2500(18kW) Micro-Area X-ray Diffractometer. X-ray
photoelectron spectroscopy (XPS) measurements were carried out
using Thermo VG Scientific Sigma Probe Multi Purpose X-Ray
60 Photoelectron Spectrometer. To confirm its superparamagnetic
Reuse of γ-Fe2O3 catalyst
The reusability of the γ-Fe2O3 catalyst is another key parameter in
PET recycling. The recovered catalyst was reused 10 times under
the same reaction conditions.
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
Green Chemistry
DOI: 10.1039/C3GC41P83a4gKe 4 of 8
3. Results and Discussion
γ-Fe2O3 catalyst characterization
For the application of γ-Fe2O3 as catalyst, it is important to use
superparamagnetic nanoparticles. The superparamagnetic property
of the catalyst is essential for the reversible control of dispersion
and flocculation. Small crystallites of ferromagnetic and
ferrimagnetic materials (ranging from 1 to 30 nm) exhibit
superparamagnetism.31,32 In this research, slight modifications of
the conventional co-precipitation method were made in order to
5
10 obtain γ-Fe2O3 nanoparticles small enough to exhibit
superparamagnetism. The TEM images and the particle size
histogram of the produced γ-Fe2O3 in Fig. 1 show that the size of
the nanoparticles ranges from 8 to 14 nm, with an average size of
10.5 nm ± 1.4 nm. The small size and monodispersity of the
15 produced γ-Fe2O3 nanoparticles resulted from a) the initiation of
rapid nucleation by adding the base to the Fe2+/Fe3+ solution
instead of adding the latter dropwise to the former as in the
traditional co-precipitation method, and b) the control of crystal
growth through the addition of citric acid surfactant.
20
35
Fig. 2 Magnetization curve of the synthesized γ-Fe2O3 confirming its
superparamagnetic nature
Fig.1 TEM images of (a, b) pristine and (c, d) used γ-Fe2O3, and (e) the
40
TEM histogram for the pristine catalyst
Fig. 3 XRD pattern of (a) pristine γ-Fe2O3, and after the (b) first, (c) third,
(d) sixth, (e) and ninth time of use with diffraction peaks corresponding to
γ-Fe2O3 (JCPDS 39-1346, space group P4132)
25
Due to the small size of the γ-Fe2O3 nanoparticles produced, it
exhibited superparamagnetism, as confirmed by the magnetization
curve from the VSM. It can be seen in Fig. 2 that when a magnetic
field was applied, the material showed a strong response, and the
saturation magnetization reached 47 emu/g. More importantly, the
45
The XRD pattern in Fig. 3a shows sharp peaks, with major
peaks at (311) and (440), indicating good crystallinity of the
obtained product. The peaks in the XRD pattern coincided with
those of standard γ-Fe2O3 (JCPDS 39-1346, space group P4132).
The XRD pattern, suggests that the material is γ-Fe2O3. However,
30 curve did not have any hysteresis loop, a unique property of
superparamagnetic materials. The absence of hysteresis loop
implies that the catalyst will re-disperse when used again in the
subsequent reactions.
50 since Fe3O4 and γ-Fe2O3 have the same crystalline structure, they
give the same XRD pattern. Hence, further analysis was done to
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

View Article Online
DOI: 10.1039/C3GC41834K
Page 5 of 8
Green Chemistry
35
Fig. 6 NH3-TPD pattern of γ-Fe2O3
Fig. 4 Fe2p XPS patterns of the material (a) before and (b) after
calcinations
from this, there were no other transformations observed, indicating
that the material did not degrade, and maintained its oxidation state
40 at temperatures up to 800 °C. Thus, the thermal stability of the
catalyst was confirmed. Although transformation from γ-Fe2O3 to
α-Fe2O3 is also possible, it did not occur during the reactions as
confirmed by the XRD spectrum of the used catalyst shown in Fig.
3b-3e.
45
Reaction mechanism of γ-Fe2O3-catalyzed glycolysis
Previous studies show that in a transesterification reaction as in
PET glycolysis, metal oxides can act as either base catalysts or
acid catalysts. The positive metal ions in the structure of metal
50 oxides possess Lewis acidity and the negative oxygen ions behave
as Brönsted bases. When acting as a base catalyst, the metal oxide
provides sufficient adsorption sites for the alcohol (i.e. ethylene
glycol) to transform it into catalytically active (and more
nucleophilic) species RO-. The catalytically active species RO-
55 then attacks the carbonyl group in the ester. In the case of PET
glycolysis, the RO- attack will lead to the breaking of the PET
chain, producing oligomers and dimers, then finally to BHET
monomers. On the other hand, when acting as an acidic catalyst,
the metal oxide protonates the carbonyl group of the ester making
60 the carbonyl group more electrophilic. When the carbonyl group is
more electrophilic, it is more vulnerable to nucleophilic attack.35 In
the case of PET glycolysis, the carbonyl carbon will be more
vulnerable to the attack of nucleophile EG, resulting in the
breaking of the PET chain. 36 The reaction mechanism of γ-Fe2O3
65 catalyzed PET glycolysis can therefore be postulated by
determining if γ-Fe2O3 acts as acidic or basic catalyst.
The acidity of the γ-Fe2O3 catalyst was evaluated using NH3-
TPD analysis, where NH3 is adsorbed on the surface of the
material at low temperature, and desorbed at higher temperature.
70 The temperature at which NH3 desorbs depends on the strength of
the acid sites in the material.37 The TPD profile in Fig. 6 shows
three distinct peaks. By deconvolution, the peaks were determined
to be at 177.0, 282.0, and 384.8 °C. The peak at 177.0 °C
corresponds to weak acid sites, while the intense peaks at 282.0 °C
75 and 306.0 °C correspond to high concentrations of medium
strength acid sites. Thus, the γ-Fe2O3 nanoparticles act
predominantly as mildly acidic catalyst. From this, the glycolysis
5
Fig. 5 Thermogravimetric analysis for γ-Fe2O3 showing weight loss only at
temperature below 200 °C
confirm the nature of the material. The confirmation of the
10 transformation of Fe3O4 during the calcinations to γ-Fe2O3 is
important, since Fe3O4 is unstable at high temperatures, and may
affect the glycolysis reaction; starting at 200 °C, Fe3O4 transforms
to γ-Fe2O3, and at 375 °C, γ-Fe2O3 begins to transform to α-
Fe2O3.33 The unstable nature of Fe3O4 was also one of the reasons
15 why γ-Fe2O3 was used as a catalyst instead of Fe3O4. To confirm
that the produced nanoparticles after calcination were indeed γ-
Fe2O3 and not Fe3O4, XPS was done. The Fe2p XPS patterns of the
material before and after calcinations are presented in Fig. 4.
Before calcination, the Fe2p3/2 and Fe2p1/2 were located at
20 approximately 711.6 eV and 724.8 eV, respectively. After
calcination, they were located at approximately 710.9 eV and
724.2 eV, respectively. The set of higher binding energies before
.
calcination (Fig. 4a) is a characteristic of Fe3O4 The shift to
relatively lower binding energies after the calcination implies the
25 transformation from Fe3O4 to γ-Fe2O3. Additionally, the satellite
peak at approximately 720 eV in Fig. 4b is characteristic of γ-
Fe2O3.34 Thus, the Fe3O4 nanoparticles produced from the
precipitation method did transform to γ-Fe2O3 during calcination.
The thermal stability of the catalyst is also essential in melt-
30 phase glycolysis, because it affects catalyst performance and
reusablity. The thermo-gravimetric curve in Fig. 5 shows only one
stage of weight loss. The drop of weight below 200 °C is a result
of the evaporation of the adsorbed water from the surface. Aside
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
Green Chemistry
DOI: 10.1039/C3GC41P83a4gKe 6 of 8
25
Fig. 9 BHET yield with time for different catalyst/PET weight ratios
Fig. 7 Proposed mechanism for PET glycolysis with γ-Fe2O3 as catalyst
30
35
40
Fig. 10 BHET yield with time for glycolysis with γ-Fe2O3 and various
silica nanoparticle-supported metal oxides 16, 17 at 0.01 catalyst/PET weight
ratio and 300 °C
5
Fig. 8 HPLC spectra of (a) standard BHET and (b) the glycolysis product
45 γ-Fe2O3 catalyst performance
reaction mechanism for PET glycolysis with γ-Fe2O3 acting as
acidic catalyst may be postulated (Fig. 7). The carbonyl group of
To evaluate the performance of the catalyst, the BHET monomer
yield was measured via HPLC. HPLC, together with the NMR
10 the polyester is more vulnerable to attack by any foreign species.
The metal (Fe+3) interacts with the oxygen of the carbonyl group
and drags more electrons towards itself, creating highly partial
positive charge on the carbon. A free lone pair on the EG oxygen
then attacks the carbonyl carbon of the ester group of the polyester,
15 resulting in the formation of C-O bond with the oxygen of the EG
while breaking the C-O bond of the polyester. By this way, the
longer PET chains break down to shorter chain oligomers, dimer,
and finally to monomer BHET.36 The transformation of dimer to
BHET monomer is a reversible process. Longer reaction after the
20 equilibration of the two is attained will cause the reaction to shift
backwards, increasing the amount of dimer and oligomers at the
expense of the BHET monomer.30 It is thus important to determine
the optimum reaction time for the glycolysis reaction.
spectra (See Fig. S1†), also confirmed that the main product of the
reaction was BHET. The elution time of the main peak (2.8 min)
50 for the monomer product matched with that of the standard BHET
(Fig. 8). Fig. 9 shows the BHET yield from the HPLC analysis for
different amounts of catalyst. The significant difference between
the monomer yield of the reaction with and without γ-Fe2O3 shows
that the nanoparticles successfully catalyzed the reaction.
55 Moreover, under the same reaction condition and catalyst/PET
weight ratio with those of the previously reported silica
nanoparticles-supported metal oxide catalysts such as ZnO, Mn3O4,
16,17
and CeO2,
γ-Fe2O3 yielded the same degree of performance
(Fig. 10). At 0.010 catalyst/PET weight ratio, the BHET yield
60
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

View Article Online
DOI: 10.1039/C3GC41834K
Page 7 of 8
Green Chemistry
35 Table 1 Catalyst recovery rate
Amount
of
catalyst
(g)
catalyst/PET
weight
Amount of
recovered
catalyst (g)
Amount of
%
recovery
catalyst (g)
lost
ratio
0.01
0.015
0.02
3
4.5
6
2.6
4.1
0.4
0.4
0.8
0.9
87
91
87
94
5.2
0.05
15
14.1
the higher catalyst/PET weight ratio permitted the reaction to reach
40 equilibrium earlier. The glycolysis reaction with 0.05 catalyst/PET
weight ratio only required 60 min to achieve BHET yield greater
than 90%. The shorter reaction time translates to a more energy-
efficient and cost-effective process compared to the other metal
oxide nanocatalysts studied.
45
We also evaluated the performance of γ-Fe2O3 at milder
temperature, and compared its performance with that of another
recently reported recyclable catalyst, urea (Fig. 11). It is apparent
that γ-Fe2O3 increased the glycolysis reaction rate. At 255 °C, the
γ-Fe2O3-catalyzed reaction gave >80% BHET yield in only 80
Fig. 11 BHET yield with time for non-catalyzed, γ-Fe2O3 catalyzed, and
urea-catalyzed glycolysis at 255 °C, and 0.1 catalyst/PET weight ratio
5
50 min. It also showed better performance compared to urea, which
attained the maximum BHET yield in 120 min. Thus, Fe2O3 is
effective even at mild temperatures.
For industrial applications, the catalyst should be easily
separated from the product mixture. The superparamagnetic
55 property of γ-Fe2O3 nanoparticles provides a unique condition to
serve as catalyst for PET glycolysis as well as easy separation
from the product using a magnet. The separation process does not
require any energy-intensive or complicated separation method.
The average amounts of recovered catalysts are presented in Table
60 1. There is a minimal amount of lost catalyst with increased
amount of catalyst used. Since the 15 mg catalyst delivered the no
highest glycolysis and recovery rate, this amount was used for the
evaluation of the reusability of the catalyst. Each time the catalyst
was reused, the yield was almost the same (Fig. 12), with
65 significant drop. Statistical analysis using analysis of variance
(ANOVA) at 99.9% confidence level showed that the number of
times the catalyst was reused did not significantly affect the BHET
Fig. 12 BHET yield for re-used catalyst
10 reached higher than 90% in 80 min. The remarkable catalytic
performance of the γ-Fe2O3 nanoparticles may be credited to the
high surface area, which renders greater accessibility to the active
sites. The measured surface area of the nanoparticles in this
research is 147 m2/g; bulk γ-Fe2O3 has about only 9 m2/g.24 In
15 comparison, The high catalytic efficiency (high monomer yield in
our case) with metal oxide catalysts may also be attributed to the
nature of metal cation, its oxidation state, reduction potential, its
distribution in the crystal structure, and the metal oxide spinel
geometry. The γ-Fe2O3 (maghemite) possesses inverse cubic spinel
monomer yield (See Table S1†). This result confirmed the
reusability of the catalyst, which may be attributed to its stability,
70 as evidenced by the TGA (Fig. 5), and the XRD spectra of the used
catalyst in Fig. 3b. Thus, γ-Fe2O3 the repeated use of the catalyst
for PET glycolysis at high temperature does not affect its
performance as catalyst.
20 structure with formula (Fe+3)[Fe+3
□
40/3
1/8] O32, where (Fe+3) and
[Fe+3] represent the tetrahedral and octahedral coordination of Fe+3
in the spinel structure. The □ presents the iron vacancy sites on the
octahedral.38 Cheng et al.39 reported that the type of spinel
structure (cubic, tetragonal) explained the significant effect of the
25 oxygen binding capability on the catalyst surface. The faster attack
of Fe+3 cations on the oxygen of carbonyl group of polyester due to
their distribution on tetrahedral and octahedral sites combined with
the inverse cubic spinel structure is therefore anticipated.
75 Conclusions
The application of superparamagnetic γ-Fe2O3 nanoparticles as a
easily recoverable and reusable catalyst for PET glycolysis was
evaluated. The use of magnetic iron oxide catalyst facilitated the
easy recovery. Results showed that γ-Fe2O3 delivered excellent
80 performance in catalytically depolymerizing PET back to
monomer BHET. This may be attributed to its ability to catalyze
the glycolysis through redox reaction, high surface area rendering
more active sites, thermal stability, and good crystallinity. Also,
the easy recoverability and reusability of the γ-Fe2O3 permitted the
85 use of larger amount of catalyst for the glycolysis reaction. The
It is also observed from Fig. 9 that for the reactions with
30 different amounts of catalyst, maximum BHET yield was reached
at different times. The possibility of the catalyst’s easy recovery
and reuse allowed the use of higher amount of catalyst. As a result,
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
Green Chemistry
DOI: 10.1039/C3GC41P83a4gKe 8 of 8
25 N. Koukabi, E. Kolvari, A. Khazaei, M. Zolfigol, B. Shirmadi-
higher catalyst/PET weight ratio resulted in a better performance,
speeding up the reaction, and thereby reaching maximum
monomer yield (>90% BHET) at an earlier time (60 min), which
translates to more energy-efficient and cost-effective process. The
catalyst also proved to be reusable, making the process more
sustainable and environment-friendly.
Shaghamesi, and H. Khavasi, Chem. Commun., 47, 9230.
70 26 A. Garade, M. Bharadwaj, S. Bhagwat, A. Athawale, and C. Rode,
Catal. Commun., 2009, 10, 485.
27 G. Picasso, A. Quintilla, M. Pina, and J. Herguido, Applied Catal. B,
2003, 46, 133.
5
28 Y. Kang, S. Risbud, J. Rabolt, and P. Stroeve, Chem. Mater., 1996, 8,
75
80
2209.
29 J. Jeong, S. Lee, J. Kim, and S. Shin, Phys. Stat. Sol., 2004, 241, 1593.
30 M. Imran, B. Kim, M. Ha, B. Cho, and D. Kim, Polym. Degrad.
Stab., 2010, 95, 1686.
31 K. Troev, G. Grancharov, R. Tsevi, and I. Gitsov, J. Appl. Polym.
Sci., 2003, 90, 1148.
32 N. Burke, H. Stover, and F. Dawson, Chem. Mater., 2002, 14, 4752.
33 H. Lepp, The American Minerologist, 1957, 42, 679.
34 J. Lu, X. Jiao, D. Chen, and W. Li, J. Phys. Chem., 2009, 113, 4012.
35 A.A. Refaat, Int. J. Environ. Sci. Tech., 2011, 8, 203.
Acknowledgements
This work was supported by the Basic Science Research Program
through a National Research Foundation of Korea (NRF) grant
10 funded by the Ministry of Education, Science and Technology
(2010-0025671).
Notes and references
85 36 J. Scheirs and T. E. Long, Modern Polyesters: Chemistry and
Technology of Polyesters and Copolyesters. 2003, John Wiley and
Sons, Ltd, West Sussex.
37 Dahlan, D. Marsih, I. Makertihartha, P. Praserthda, S. Parpranot, and
Ismunandar, Chemistry and Materials Research, 2012, 2, 31.
90 38 V. Hiremath and A. Venkataraman, Bull. Mater. Sci., 2003, 26, 391.
39 F. Cheng, J. Shen, B. Peng, Y. Pan, Z. Tao, and J. Chen, Nature
Chemistry, 2011, 3, 79.
aDepartment of Chemical & Biomolecular Engineering (BK21 Program),
KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon 305-701, Republic of
15 Korea. E-mail: DoHyun.Kim@kaist.ac.kr; Fax: +82 42 350 3910; Tel:
+82 42 350 3929
bChemical Engineering Department, King Saud University, Saudi Arabia
cDepartment of Chemical Engineering, University of Michigan, Ann Arbor,
MI, 48109-2136, USA
20 dDepartment of Microbiology, School of Bioscience & Biotechnology,
Chungnam National University, Daejeon, 305-764, Republic of Korea
†Electronic Supplementary Information (ESI) available: ANOVA for
BHET yield, 13C and 1H NMR spectra of the BHET obtained from the
glycolysis reaction. See DOI: 10.1039/b000000x/
25
1
2
F. Welle, Conservation and Recycling, 2011, 55, 865.
D. Achilias and G. Karayannidis, Water, Air, & Soil Pollution:
Focus, 2004, 4, 385.
3
J. Campanelli, M. Kamal and D. Cooper, J. Appl. Polym. Sci., 1994
54, 1731.
30
35
40
45
4
5
J. Chen and L. Chen, J. Appl. Polym. Sci., 1999, 73, 35.
S. Baliga and W. Wong, J. Polym. Sci. Part A: Polym. Chem., 1989,
27, 2071.
M. Ghaemy and K. Mossaddegh, Polym. Degrad. Stabil., 2005, 90,
570.
C. Chen, J. Appl. Polym. Sci., 2003, 87, 2004.
G. Xi, M. Lu, and C. Sun, Poym. Degrad. Stabil., 2005, 87,117.
S. Shukla and K. Kulkarni, J. Appl. Polym. Sci., 2002, 85, 1765.
6
7
8
9
10 R. López-Fonseca, I. Duque-Ingunza, B. de Rivas,S. Arnaiz, and J.
Gutiérrez-Ortiz, Polym. Degrad. Stabil., 2010, 95, 1022.
11 S. Shukla, and A. Harad, J. Appl. Polym. Sci., 2005, 97, 513.
12 N. Pingale, V. Palekar,and S. Shukla, J. Appl. Polym. Sci., 2010, 115,
249.
13 S. Shukla, V. Palekar,and N. Pingale, J. Appl. Polym. Sci., 2008, 110,
501.
14 C. Y. Kao, W. H. Cheng, and B. Z.Wan, Thermochimica Acta, 1997,
292, 95.
15 L. Bartolome, M. Imran, B. G. Cho, W. Al-Masry and D. H. Kim,
Material Recycling - Trends and Perspectives, InTech,, 2012
50 16 M. Imran, K. G.. Lee, Q. Imtiaz, B. Kim, M. Han, B. Cho, and D.
Kim, J. Nanosci. Nanotechnol, 2011, 11,824.
17 R. Wi, M. Imran, K. G. Lee, S. Yoon, B. Cho, and D. Kim, J.
Nanosci. Nanotechno.l, 2011, 11, 6544.
18 G. Park, L. Bartolome, K. G. Lee, S. J. Lee, D. H. Kim, and T. J. Park,
55
Nanoscale, 2012, 4, 3879.
19 U, Heiz and U. Landman, Nanocatalysis, 2007, Springer-Verlag,
New York.
20 H. Wang, Z. Li, Y. Liu, X. Zhang, and S. Zhang, Green Chem., 2009,
11,1568.
60 21 H. Wang, Z. Li, Y. Liu, X. Zhang, S. Zhang, and Y. Zhang, Eur.
Polym. J., 2009, 45, 1535.
22 H. Wang, R. Yan, Z. Li, X. Zhang, and S. Zhang, Catal. Commun.,
2010, 11, 763.
23 Q. Yue, Z. Wang, L. Zhang, Y. Ni, and Y. Jin, Polym. Degrad.
65
Stabil.,2011, 96, 399.
24 Q. Wang, X. Yao, S. Tang, X. Lu, X. Zhang, and S. Zhang, Green
Chemistry, 2012, 14, 2559.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7