polymers

Article
PET Waste Recycling into BTX Fraction Using In Situ Obtained
Nickel Phosphide
Maria Golubeva * , Mariyam Mukhtarova, Alexey Sadovnikov and Anton Maximov

A.V.Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences (TIPS RAS),

Moscow 119991, Russia
* Correspondence: [email protected]

Abstract: The annual production of plastic waste is a serious ecological problem as it causes substan-
tial pollution of the environment. Polyethylene terephthalate, a material usually found in disposable
plastic bottles, is one of the most popular material used for packaging in the world. In this paper, it is
proposed to recycle polyethylene terephthalate waste bottles into benzene-toluene-xylene fraction
using a heterogeneous nickel phosphide catalyst formed in situ during the polyethylene tereph-
thalate recycling process. The catalyst obtained was characterized using powder X-ray diffraction,
high-resolution transmission electron microscopy, and X-ray photoelectron spectroscopy techniques.
The catalyst was shown to contain a Ni2 P phase. Its activity was studied in a temperature range of
250–400 ◦ C and a H2 pressure range of 5–9 MPa. The highest selectivity for benzene-toluene-xylene
fraction was 93% at quantitative conversion.

Keywords: PET; polyethylene terephthalate; plastic waste; waste recycling; nickel phosphide; Ni2 P;
BTX; dispersed catalysts

1. Introduction
More than 380 million tons of plastic are produced annually in the world [1]. However,
Citation: Golubeva, M.; Mukhtarova,
in 2015 it was reported that more than 50% of plastics are never recycled in any way. Some
M.; Sadovnikov, A.; Maximov, A. PET
plastic waste ends up in landfills, and some goes into the environment, thereby polluting it
Waste Recycling into BTX Fraction
and having a detrimental effect on the inhabitants. Under normal environmental conditions,
Using In Situ Obtained Nickel
depending on the type, plastic decomposes within 10–1000 years, and on average, its
Phosphide. Polymers 2023, 15, 2248.
decomposition time is measured in hundreds of years.
https://doi.org/10.3390/
polym15102248
One of the most common types of plastic is polyethylene terephthalate (PET), which
is widely used across the world to make bottles for water and soft drinks, textile fibers,
Academic Editors: Sylvie Dagréou sheets, and films [2,3]. Only a part of PET bottles is recycled into fiber, and is rarely reused
and George Z. Papageorgiou for bottle production [4]. This type of recycling is called mechanical, which is the most
Received: 2 April 2023 common type of plastic recycling. However, mechanical recycling requires careful sorting
Revised: 6 May 2023 of waste for the selection of the same composition plastics [5]. Moreover, a major problem
Accepted: 9 May 2023 in PET mechanical recycling is the fact that mechanical properties of recycled PET (rPET)
Published: 10 May 2023 decline significantly [6].
Chemical recycling is an alternative method for PET processing. Since PET con-
tains an aromatic structure, not only the PET monomer, and terephthalic acid (TPA), but
also arenes, such as benzene, toluene and xylenes, can be obtained. p-Xylene is known
Copyright: © 2023 by the authors. to be a raw material for TPA production [7]. Thus, PET can be recycled back into its
Licensee MDPI, Basel, Switzerland. raw material. Heterogeneous catalysts are effective for this goal [8]. For instance, Jing
This article is an open access article
et al. made an investigation on various noble metal-based catalyst activity in PET re-
distributed under the terms and
cycling into aromatics (toluene + p-xylene) [9]. Ru-based catalysts were shown to be
conditions of the Creative Commons
more active and selective than Pt- and Pd-based catalysts. Ru-based catalyst activity de-
Attribution (CC BY) license (https://
creased in a row: Ru/Nb2 O5 > Ru/ZrO2 > Ru/TiO2 > Ru/HZSM-5 (200 ◦ C, 0.3 MPa
creativecommons.org/licenses/by/
H2 , 12 h). The yield of aromatics over Ru/Nb2 O5 was 80.1%. In another work of this
4.0/).

Polymers 2023, 15, 2248. https://doi.org/10.3390/polym15102248 https://www.mdpi.com/journal/polymers

Polymers 2023, 15, 2248 2 of 11

research group, a comparison of Pt/NiAl2 O4 , Pd/NiAl2 O4 , Ru/NiAl2 O4 , and Ru/Nb2 O5

activity in H2 -free conversion of PET to benzene-toluene-xylenes (BTX) fraction was con-
ducted [10]. H2 was obtained using aqueous-phase reforming of ethylene glycol after
PET depolymerization. The BTX yield decreased after using catalysts in the following
order: Ru/Nb2 O5 > Ru/NiAl2 O4 > Pd/NiAl2 O4 > Pd/NiAl2 O4 (220 ◦ C, 2 MPa N2 , 12 h).
Ru/Nb2 O5 provided the BTX yield of 91.3%.
A three-step process of PET recycling was considered by Tang et al. [11]. In the first
step, thermal alcoholysis of PET was carried out. Methanol demonstrated the highest
effectiveness in alcoholysis compared to ethanol and butanol. Dimethyl terephthalate
(DMT) yield was 97.3% at 200 ◦ C, after 3.5 h of reaction. In the second step, DMT was
hydrogenated to dimethyl cyclohexane-1,4-dicarboxylate (DMCD) over Pt/C, Ru/C or
Pd/C. In the third step, DMCD was hydrodeoxygenated over Ru-Cu/SiO2 catalyst, with
C7 –C8 cycloalkanes and aromatics being observed. The yield of aromatics was 38% (400 ◦ C,
4 MPa H2 ; DMCD flow rate: 0.04 mL·min−1 , and H2 flow rate: 120 mL·min−1 ).
Hongkailers et al. carried out an investigation on Co/TiO2 activity in PET recycling
into arenes under H2 pressure [12]. At first, TPA was used as a substrate. The yield of
arenes, such as p-xylene and toluene, was 75.2 mol. %. (340 ◦ C, 3 MPa H2 , 4 h). Then, the
investigation of PET recycling was carried out. Co/TiO2 catalyst was shown to promote
both the depolymerization of PET and the hydrodeoxygenation (HDO) of monomers
obtained. Toluene and p-xylene were both observed with 78.9 mol. % yield. Thereafter,
the possibility of Co/TiO2 reusing was investigated. After the recycle, the total product
yield decreased from 90 mol. % to 32 mol. %, while arene yield decreased from 69 mol. %
to 4 mol. %. Catalyst deactivation was related to CoTiO3 phase formation instead of
individual phases of Co and TiO2 .
The present study sets out the recycling of waste PET bottle into BTX fraction using
in situ obtained nickel phosphide. Recently, nickel phosphide showed high activity in
the HDO of various oxygen-containing substrates [13]. Moreover, it is an approach to
the in situ synthesis of nickel phosphide-based catalysts in the reaction medium that we
suggested [14–16]. Nickel phosphide-based catalysts were obtained from water- and oil-
soluble precursors in polar or non-polar medium during the HDO of biomass-derived
compounds, such as furfural [14], guaiacol [15], and levulinic acid [16]. Partially and fully
deoxygenated products were obtained. This approach was applied in the present work. In
contrast to the previous works, which used two different precursors containing nickel and
phosphorus separately, a catalyst precursor used in the present work contained both nickel
and phosphorus in its composition. This allows to fix the Ni/P ratio throughout all catalytic
experiments. Nickel phosphide-based catalyst was obtained from the Ni-P-containing
precursor during the depolymerization and the HDO of PET. The highest selectivity of
BTX fraction was 93 wt. % at quantitative PET conversion over in situ generated nickel
phosphide-based catalyst (400 ◦ C, 9 MPa H2 , 6 h).

2. Materials and Methods

2.1. Synthesis of Nickel Phosphide Precursor
A total of 5 g of nickel acetate tetrahydrate (98%, Sigma-Aldrich, St. Louis, MO, USA),
3.9 g of hypophosphorous acid (50 wt. % in water, Sigma-Aldrich, St. Louis, MO, USA),
and 100 mL of deionized water were mixed in a 150 mL glass cup. The mixture was stirred
while heating to dissolve the reagents and then slowly evaporate water. A greenish-yellow
precipitate was formed. The water residues were removed in an oven at 120 ◦ C for 3 h.
After that, the solid was ground in a porcelain mortar to obtain a powder.

2.2. Catalytic Tests

The polyethylene terephthalate bottle used for storing sparkling mineral water was
milled. The fraction containing particles of ~0.04 cm2 in size was selected for the experiment.
A total of 150 mg of polyethylene terephthalate, 90.7 mg of the precursor, 1.5 g of dodecane
(Sigma-Aldrich, St. Louis, MO, USA, >99%), and a magnetic stirrer were put in a batch
Polymers 2023, 15, 2248 3 of 11

autoclave reactor of 45 cm3 in volume. The reactor was sealed, filled with hydrogen
(≥98%, Air Liquide, Paris, France) up to 5–9 MPa and heated to 250–400 ◦ C while stirring.
The catalytic reaction was carried out for 6 h. Then, the reactor was cooled to room
temperature and depressurized. The reaction products were separated from the catalyst via
centrifugation (5000 rpm). The catalyst was washed several times using dimethyl sulfoxide
(DMSO; Component-reaktiv, Moscow, Russia, >99.8%) to remove terephthalic acid, and
also washing using petroleum ether 40/70 (Component-reaktiv, Moscow, Russia, tech.).
Then, the catalyst was dried at room temperature in atmospheric argon (≥98%, Air Liquide,
Paris, France) until the ether completed evaporation, then ground in a porcelain mortar
and passed through the sieve (mesh size of 0.2 mm) to remove PET leftovers.

2.3. Characterization
The phase composition of the catalysts obtained was determined using powder X-ray
diffraction (XRD). The diffractograms were obtained in a range of 10−60◦ 2θ using a Rigaku
Rotaflex D/MAX-RC and a Bruker D8 Advance diffractometers containing CuKα radiation.
Qualitative phase analysis of the samples was carried out using the PDF-2 ICDD database
of powder diffraction patterns. The average sizes of crystallites were calculated using the
Scherrer equation.
Transmission electron microscopy (TEM), high-resolution transmission electron mi-
croscopy (HRTEM) images and elemental maps of the catalyst samples were obtained
using a FEI Tecnai Osiris transmission electron microscope equipped with a field emission
electron gun operated at 200 kV and an attachment for energy dispersive X-ray (EDX)
spectroscopy analysis. Fast Fourier transformation (FFT) of the HRTEM images obtained
was carried out using Gatan DigitalMicrograph software.
X-ray photoelectron spectra (XPS) of the catalyst were obtained using a PREVAC
EA15 electronic spectrometer and AlKα radiation source (hν = 1486.74 eV, 150 W). Spectra
deconvolution was carried out using a PeakFit software.
Gaseous carbon-containing products were identified using a Chromos GH-1000 chro-
matograph equipped with a thermal conductivity detector, and two packed columns: a
NaX/3X molecular sieve column (2 m × 3 mm), and a HayeSep R column (3 m × 3 mm),
with helium used as a carrier gas.
Terephthalic acid obtained after the catalytic reaction was identified using NMR. 1 H
13
and C NMR-spectra of TPA were recorded on a Bruker AVANCE III HD (400 MHz)
spectrometer, with a solvent signal used as internal reference. DMSO-d6 (Solvex-D, 99.8%)
was applied to dissolve TPA.
The qualitative analysis of liquid reaction products was carried out via gas chromato-
graphy–mass spectrometry (GC–MS) using a Thermo Scientific ISQ 7000 GC–MS equipped
with a Restek 5XI-17SIL MS CAP capillary column (30 m × 0.25 mm × 0.25 µm), with helium
as a carrier gas. The quantitative analysis of liquid reaction products was carried out via
gas–liquid chromatography (GLC) using a Crystallux 4000 M gas chromatograph equipped
with a flame ionization detector, an Optima-1 capillary column (25 m × 0.32 mm × 0.35 µm),
and with helium used as a carrier gas. PET conversion (wt. %) and product selectivity (%) were
calculated using the following equations:

initial mass o f PET − mass o f unreacted PET

PET conversion (wt.%) = × 100%, (1)
initial mass o f PET

mass o f product f ormed

Product selectivity (%) = × 100%. (2)
∑ mass o f all products

3. Results and Discussion

3.1. Nickel Phosphide In Situ Formation
Recently, we studied the in situ formation of nickel phosphide catalysts during the
hydrodeoxygenation of oxygen-containing bio-based substrates using hypophosphorous
 above. It was conducted in order to avoid even a minimal variation in the acid/salt ratio,
and the presence of free acid and water in the reaction medium at the initial moment of
reaction.
As a result of nickel (II) acetate and hypophosphorous acid mixture heating in an
aqueous solution, nickel (II) acetate was hydrolyzed to form nickel (II) hydroxide, which
Polymers 2023, 15, 2248
was dissolved in hypophosphorous acid to form nickel hypophosphite. Then, the4 offor- 11

mation of nickel phosphide from nickel hypophosphite during the decomposition of PET
under H2 pressure has been investigated.
acid and nickelof(II)
The effect theacetate
reactionas temperature
catalyst precursors [14–16].
(300, 340, and 380In°C) this
andwork, a solid
pressure (5, catalyst
7, and 9
precursor,
MPa H2) on nickel
the (II) hypophosphite,
formation of nickelwas easily obtained
phosphide from the
was studied using compounds
XRD (Figure mentioned
1). The
above. It was conducted
nickel phosphide phase of inNi
order to avoid
2P (PDF even awas
№74-1385) minimal
shown variation
to be formedin theinacid/salt
all cases.ratio,
Ni2P
and the presence of free acid and water in the reaction medium at the
reflections were identified in the samples at 2θ = 40.6°, 44.5°, 47.2°, 54.0°, 54.8° and corre- initial moment
of reaction.
sponded to the (111), (201), (210), (300), (211) planes, respectively. In addition to the Ni2P
phase,Asphosphate
a result of phases
nickel (II)
were acetate andNiH
formed. hypophosphorous
2P2O7 (PDF №34-617) acid mixture heating in
was identified in the
an
aqueous
samples solution,
obtained nickel
at 300 (II)
°C, acetate
9 MPa was hydrolyzed
H2; and to 5,
at 340 °C, form
7, 9nickel
MPa H (II) hydroxide, which
2. Phosphates can be
was dissolved in hypophosphorous acid to form nickel hypophosphite.
formed as a result of hypophosphite disproportionation. Guan et al. [17] supposed Then, the formationthat
of nickel phosphide from
the following reaction occurs: nickel hypophosphite during the decomposition of PET under
H2 pressure has been investigated.
Ni(H2PO2)2 =(300,
The effect of the reaction temperature ↑ + NiHPO
PH3340, and 380 4. ◦ C) and pressure (5, 7, and (3)
9 MPa H
Then, ) on the formation of nickel phosphide was studied using
2 NiHPO4 can be transformed into NiH2P2O7 when heated [18,19]. In the sample XRD (Figure 1). The
nickel phosphide
of the catalyst obtained phase atof380
Ni2°C under№74-1385)
P (PDF 9 MPa H2,was shownoftoNi(PO
the phase◦ ◦
be formed in all cases.
3)2 (PDF №28-708)
◦ ◦
Ni
was2 P found
reflections were identified
as a result of NiH2P2in O7the samples at 2θat=higher
decomposition 40.6 , temperature
44.5 , 47.2 , 54.0 54.8◦ and
[19].,Moreover,
corresponded to the (111), (201), (210), (300), (211) planes, respectively. In addition to the
in the sample obtained at 340 °C under 9 MPa H2, terephthalic acid (PDF №31-1916) was
Ni2 P phase, phosphate phases were formed. NiH2 P2 O7 (PDF №34-617) was identified in
identified. This indicates the formation of TPA during the decomposition of PET over
the samples obtained at 300 ◦ C, 9 MPa H2 ; and at 340 ◦ C, 5, 7, 9 MPa H2 . Phosphates can
nickel phosphide, and also that the sample was not completely washed. Thus, the reaction
be formed as a result of hypophosphite disproportionation. Guan et al. [17] supposed that
conditions do not significantly affect the phase composition of the catalysts; however, with
the following reaction occurs:
an increase in the reaction temperature, NiH2P2O7 decomposes with the formation of
Ni(PO3)2. Ni(H PO ) = PH ↑ + NiHPO . (3)
2 2 2 3 4

Figure1.
Figure 1. XRD
XRDpatterns
patternsofofthe
thecatalyst
catalyst samples
samples obtained
obtained at (a)
at (a) 300,300,
340,340,
380 380 °C,
◦ C, 9 9 MPa
MPa H2 , H 2, 6 h; (b) 340
6 h; (b) 340 ◦ C,
°C, 5, 7, 9 MPa H 2, 6 h.
5, 7, 9 MPa H2 , 6 h.

The average
Then, NiHPOcrystallite sizes of the phases obtained are presented in Table S1. With
4 can be transformed into NiH2 P2 O7 when heated [18,19]. In the sample
of the catalyst obtained at 380 ◦ C under
an increase in reaction temperature, crystallite
9 MPasizeH2 ,enlargement
the phase ofof (PDF №28-708)
Ni2P 3is)2observed,
Ni(PO which
may be due to particle sintering at a higher temperature. For phosphate phases,
was found as a result of NiH2 P2 O7 decomposition at higher temperature [19]. Moreover, a similar
dependence
in the sampleisobtained
observed.atThe ◦ C under 9 MPa
340 dependence of the
H2crystallite sizesacid
, terephthalic on the
(PDF №31-1916)
initial H2 pressure
was
is not found;
identified. theindicates
This differencethein formation
sizes is within
of TPAthe during
error forthe
phosphide as wellof
decomposition asPET
for phos-
over
phate.phosphide, and also that the sample was not completely washed. Thus, the reaction
nickel
conditions do not significantly affect the phase composition of the catalysts; however,
with an increase in the reaction temperature, NiH2 P2 O7 decomposes with the formation
of Ni(PO3 )2 .
The average crystallite sizes of the phases obtained are presented in Table S1. With an
increase in reaction temperature, crystallite size enlargement of Ni2 P is observed, which
may be due to particle sintering at a higher temperature. For phosphate phases, a similar
dependence is observed. The dependence of the crystallite sizes on the initial H2 pressure is
not found; the difference in sizes is within the error for phosphide as well as for phosphate.
Since the effect of initial pressure on the catalyst phase composition and crystallite
sizes was negligible, only the samples obtained at different temperatures were further
Polymers 2023, 15, x FOR PEER REVIEW 5 of 11

Polymers 2023, 15, 2248 5 of 11

Since the effect of initial pressure on the catalyst phase composition and crystallite
sizes was negligible, only the samples obtained at different temperatures were further
studied. The
studied. The morphology
morphologyof ofNi
Ni2 P-containing
2P-containingcatalysts
catalystsobtained
obtainedinin situ
situ at at 300–380
300–380 ◦ C,°C, un-
under
9der
MPa9 MPa H2, during
H2 , during 6 hstudied
6 h was was studied using(Figure
using TEM TEM (Figure 2a–c). HRTEM
2a–c). HRTEM images of images of the
the catalysts
(Figure
catalysts2d–f) show
(Figure a crystal
2d–f) showlattice withlattice
a crystal interplanar spacings corresponding
with interplanar to Ni2 P. Due
spacings corresponding to
to
Ni2aP.reduced
Due to aFFT of HRTEM
reduced FFT ofimages,
HRTEM diffraction patterns ofpatterns
images, diffraction the catalysts
of thewere obtained.
catalysts were
There are reflections
obtained. of (111) andof(212)
There are reflections (111)planes in Niplanes
and (212) 2 P withininterplanar spacings of 0.221
Ni2P with interplanar nm
spacings
and 0.128 nm, respectively, in the diffraction pattern of the sample obtained at 300 ◦C
of 0.221 nm and 0.128 nm, respectively, in the diffraction pattern of the sample obtained
(Figure
at 300 °C 2d). The diffraction
(Figure pattern ofpattern
2d). The diffraction the sample
of theobtained at 340 ◦ Catshows
sample obtained 340 °Creflections
shows re-
corresponding to (111) plane
flections corresponding of Ni
to (111) 2 P (Figure
plane of Ni2P2e). The 2e).
(Figure reflections in the diffraction
The reflections pattern
in the diffraction
of the sample
pattern obtained
of the sample at 380 ◦at
obtained C 380
are °C
attributed to (110)
are attributed toand
(110)(211)
and planes in Ni2in
(211) planes P with
Ni2P
interplanar spacings of 0.294 nm and 0.167 nm, respectively (Figure 2f).
with interplanar spacings of 0.294 nm and 0.167 nm, respectively (Figure 2f). Thus, the Thus, the presence
of the Ni2of
presence P phase
the Ni2in the catalyst
P phase in the samples, identified
catalyst samples, using the
identified XRD,
using is XRD,
the confirmed via the
is confirmed
HRTEM technique.
via the HRTEM technique.

Figure 2. TEM, HRTEM images and corresponding reduced FFT of Ni2P-containing catalysts ob-
Figure 2. TEM, HRTEM images and corresponding reduced FFT of Ni2 P-containing catalysts obtained
tained in situ under 9 MPa H2 at (a,d) 300 °C; (b,e) 340 °C; (c,f) 380 °C.
in situ under 9 MPa H2 at (a,d) 300 ◦ C; (b,e) 340 ◦ C; (c,f) 380 ◦ C.
Ni and
Ni and P P elemental
elemental maps
maps ofof the
the catalysts
catalysts (Figure
(Figure S1)
S1) indicate
indicate the
the uniform
uniform distribution
distribution
of these
of these elements
elements inin the
the samples.
samples. Oxygen
Oxygen as as aa part
part of
of phosphates
phosphates was was also
also identified.
identified.
The electronic
The electronic states
states on
onthetheNi
Ni22PP surface
surface were
werestudied
studiedusing
usingXPS.
XPS.The
TheNi2p
Ni2p3/2
3/2 spectra
spectra
are shown in Figure 3a,c,e. Each spectrum was deconvoluted into three peaks.peaks
are shown in Figure 3a,c,e. Each spectrum was deconvoluted into three peaks. The The
with awith
peaks maximum at 853.1–853.8
a maximum eV areeVascribed
at 853.1–853.8 to Nito
are ascribed δ+ species and associated
Niδ+ species with with
and associated Ni2P
state
Ni 2 P [20].
state The
[20].peaks at 856.7–857.0
The peaks eV areeV
at 856.7–857.0 corresponded
are correspondedto Ni2+tospecies in nickel
Ni2+ species phos-
in nickel
phate. The The
phosphate. binding energies
binding of 861.7–862.3
energies of 861.7–862.3eVeV areare
related
relatedtotothe
theshake-up
shake-upsatellite
satellite peaks
accompanying the Ni2+2+main
the Ni mainpeaks.
peaks.The content
The content ofof
NiNi 2+ species
2+ species on on
thethe
catalyst surface
catalyst ex-
surface
ceeds the δ+δ+
exceeds thecontent
contentofofNi Ni , which
, whichcan canbebeassociated
associatedwith withthe
the surface
surface oxidation
oxidation of the
catalysts, as well as the incomplete reduction of phosphate formed by the decomposition
of the precursor. The P2p spectra are presented in Figure 3b,d,f. The binding energy values
Polymers 2023, 15, x FOR PEER REVIEW 6 of 11

Polymers 2023, 15, 2248 6 of 11

catalysts, as well as the incomplete reduction of phosphate formed by the decomposition
of the precursor. The P2p spectra are presented in Figure 3b,d,f. The binding energy values
of 134.2–135.2 eV
eV and
and 135.7–136.4
135.7–136.4eVeVare correspondedtotoP5+
arecorresponded P5+inin2p2p and 2p
3/2 and
3/2 2p1/2
1/2 regions,
regions,
respectively. δ
δ−−
respectively. P species
specieswere
werenot
notidentified
identifiedon
onthe
thecatalyst
catalystsurface.
surface.

Figure 3. XPS in Ni2p3/2 and P2p regions for the Ni2 P-containing catalysts obtained in situ under
Figure 3. XPS in Ni2p3/2 and P2p regions for the Ni2P-containing catalysts obtained in situ under 9
9 MPa H2 at (a,b) 380 ◦ C; (c,d) 340 ◦ C; (e,f) 300 ◦ C.
MPa H2 at (a,b) 380 °C; (c,d) 340 °C; (e,f) 300 °C.
The presence of Niδ+ species leads to the formation of metal active sites, while the pres-
ence of Nipresence
The of Ni
2+ and P–OH δ+ species leads to the formation of metal active sites, while the
leads to the formation of Lewis and Brønsted active sites, respectively.
presence of Ni and P–OH leads to the formation of Lewis and Brønsted active sites, re-
2+

3.2. Catalytic Activity

spectively.
The catalytic activity of the Ni2 P obtained in situ was investigated in PET hydropro-
3.2. Catalytic Activity◦ C, under 5–9 MPa H2 , during 6 h of reaction. PET conversion was
cessing at 250–400
quantitative in all cases,
The catalytic except
activity of thefor
Nitwo. At 250 ◦in
2P obtained C situ
underwas 5 MPa H2 , the conversion
investigated was 44
in PET hydropro-
wt. %, and ◦
at 250 C°C, under 7 MPa
cessing at 250–400 under 5–9HMPa
2 , theHconversion
2, during 6 was
h of52reaction.
wt. %. AnPETincrease in pressure
conversion was
as well as anin
quantitative increase in temperature
all cases, except for two. promotes
At 250 °Ctheunder
quantitative
5 MPa H depolymerization of PET.
2, the conversion was 44
As a result of PET depolymerization, monomers such as terephthalic acid
wt. %, and at 250 °C under 7 MPa H2, the conversion was 52 wt. %. An increase in pressure and ethane are
obtained.
as TPA
well as an production
increase was confirmed
in temperature via 1the
promotes H, 13 C NMR (Figures
quantitative S2 and S3). of
depolymerization Ethane
PET.
formation was confirmed using gas chromatography (Table S3). TPA,
As a result of PET depolymerization, monomers such as terephthalic acid and ethylene in turn, undergoes
subsequent transformations: direct deoxygenation (DDO) and decarboxylation (DCO2 ).
Figure 4 shows the effect of temperature and pressure on the product selectivity. BTX
 are obtained. TPA production was confirmed via 1H, 13C NMR (Figures S2 and S3). Eth-
ylene formation was confirmed using gas chromatography (Table S3). Since a non-polar
Polymers 2023, 15, 2248 solvent dodecane is used in this reaction, ethylene hydrolysis does not occur. TPA, in 7turn, of 11
undergoes subsequent transformations: direct deoxygenation (DDO) and decarboxylation
(DCO2). Figure 4 shows the effect of temperature and pressure on the product selectivity.
BTX fraction,
fraction, and oxygen-containing
and oxygen-containing compounds,
compounds, such such as TPA,
as TPA, p-methylbenzoic
p-methylbenzoic (p-MBA)
(p-MBA) and
and benzoic (BA) acids were obtained among the reaction products. It has
benzoic (BA) acids were obtained among the reaction products. It has been found that both been found that
both
an an increase
increase in temperature
in temperature and
and in in pressure
pressure contribute
contribute to antoincrease
an increase in selectivity
in the the selectivity
for
the target BTX fraction and a decrease in the selectivity for terephthalic acid. TPATPA
for the target BTX fraction and a decrease in the selectivity for terephthalic acid. waswas
not
not detected
detected at allatamong
all among the reaction
the reaction products
products at 380at◦ C380
and °C400
and◦ C400 °C under
under 9 MPa.9 Following
MPa. Fol-
lowing
the the dynamics
dynamics of changesof changes in the selectivity
in the selectivity for p-MBA for p-MBA
and BA, inand BA,
most in most
cases, withcases, with
increasing
temperature, the selectivity for them first increases and then decreases. It is relatedIt to
increasing temperature, the selectivity for them first increases and then decreases. is
related to the TPA conversion into these substances, and to their following
the TPA conversion into these substances, and to their following transformation into BTX transformation
into BTX All
fraction. fraction. All transformations
transformations occurringoccurring
with PETwith PET
and its and its derivatives
derivatives proceed overproceed over
metal or
metalactive
acid or acid active
sites sites
of the of the catalyst.
catalyst.

Figure 4.
Figure 4. The
The effect
effectofofreaction
reactiontemperature
temperature
onon
thethe product
product selectivity
selectivity under
under H2 pressure
H2 pressure of (a)of (a) 5
5 MPa;
MPa; (b) 7 MPa; (c) 9 MPa.
(b) 7 MPa; (c) 9 MPa.

Based on
Based on the
theexperimental
experimentaldata dataobtained,
obtained,reaction
reactionpathways
pathways were
were proposed
proposed (Figure
(Figure 5).
5). First,
First, PETPET depolymerization
depolymerization withwith the formation
the formation of TPA
of TPA and ethane
and ethane occurred.
occurred. Acidic Acidic
cata-
catalysts
lysts are known
are known to catalyze
to catalyze PETPET depolymerization
depolymerization withwith the TPA
the TPA obtained
obtained [21].[21]. There-
Therefore,
fore, it can be concluded that acid sites of the Ni P-based catalyst catalyze
it can be concluded that acid sites of the Ni2 P-based catalyst catalyze depolymerization
2 depolymeriza-
tion reaction.
reaction. Then,Then,
TPA TPA can
can be be transformed
transformed via three
via three pathways.
pathways. Through
Through DDO,DDO,
p-MBA p-MBA
acid
acid is obtained. Benzoic acid and benzene are formed by
is obtained. Benzoic acid and benzene are formed by DCO2 reaction. DDO occurs DCO 2 reaction. DDO occurs
over
over sites,
acid acid sites,
while while
DCO DCO 2 occurs
2 occurs overover metal
metal sites
sites [22,23].Aldehydes
[22,23]. Aldehydeswerewerenotnot identified
identified
among the reaction products; thus, we suggest that BA and benzene are formed directly
from TPA.
TPA. From
FromFigure
Figure4,4,it itcan bebe
can concluded
concluded thatthat
DCO 2 prevails
DCO overover
2 prevails DDO, sincesince
DDO, in most
in
cases,cases,
most the selectivity for BAfor
the selectivity exceeds the selectivity
BA exceeds for p-MBA.
the selectivity Benzene
for p-MBA. also canalso
Benzene be formed
can be
formed as a result of BA DCO2 ; toluene can be formed as a result of p-MBA DCO2 and BA
DDO; and p-xylene is formed only as a result of p-MBA DDO. Figure S4 shows that the
total selectivity for toluene and benzene prevails over the selectivity for p-xylene, which
indicates the predominance of DCO2 pathway over DDO. With an increase in temperature,
Polymers 2023, 15, x FOR PEER REVIEW 8 of 11

as a result of BA DCO2; toluene can be formed as a result of p-MBA DCO2 and BA DDO;
Polymers 2023, 15, 2248 and p-xylene is formed only as a result of p-MBA DDO. Figure S4 shows that the 8total of 11
selectivity for toluene and benzene prevails over the selectivity for p-xylene, which indi-
cates the predominance of DCO2 pathway over DDO. With an increase in temperature,
the selectivity for
the selectivity for DCO
DCO2 products increases, which correlates with an increase in the Niδ+
δ+
2 products increases, which correlates with an increase in the Ni
species content
species content onon the
the catalyst
catalyst surface.
surface.

Figure 5. Proposed pathways of PET recycling into BTX fraction over Ni22P-containing catalyst.

In the
the present
presentstudy,
study,ring-hydrogenated
ring-hydrogenatedproducts productswere were notnot observed;
observed; thisthis
is anis ad-
an
advantage
vantage of Ni of2P-containing
Ni2 P-containing catalyst
catalyst usedused
compared compared
to thetoones
thedescribed
ones described in the lit-
in the literature.
erature.
Hongkailers Hongkailers
et al. found et that
al. methylcyclohexane
found that methylcyclohexane and dimethylcyclohexane
and dimethylcyclohexane are obtained
are obtained in low amounts over
in low amounts over Co/TiO2 catalyst [12]. Jing Co/TiO 2 catalyst [12]. Jing
et al. made a comparisonet al. made
of noblea compar-
metal-
ison of noble metal-containing catalysts [9]. Ring-hydrogenated
containing catalysts [9]. Ring-hydrogenated products, such as cyclohexane, methylcyclo- products, such as cy-
clohexane, methylcyclohexane,cyclohexanecarboxylic
hexane, dimethylcyclohexane, dimethylcyclohexane, acid, cyclohexanecarboxylic acid, and 4-
and 4-methylcyclohexanecar-
methylcyclohexanecarboxylic
boxylic acid were obtained acid overwere
Ru/Nbobtained
2O5, Pd/Nbover 2Ru/Nb
O5, Pt/Nb 2 O52,OPd/Nb 2 O5 ,2Pt/Nb
5, Ru/ZrO , Ru/TiO 2 O52,,
Ru/ZrO , Ru/TiO , Ru/HZSM-5. Moreover, only ring-hydrogenated
Ru/HZSM-5. Moreover, only ring-hydrogenated products were obtained over Pt/Nb2O
2 2 products were ob-
5.
tained over Pt/Nb
In other works, NiAl O
2 2O . In other works, NiAl
5 4 also was used as a catalyst O also was used as a catalyst
2 4 for noble metals [10]. Cyclohexanecar- for noble
metals
boxylic[10].
acid,Cyclohexanecarboxylic acid, and 4-methylcyclohexanecarboxylic
and 4-methylcyclohexanecarboxylic acid were obtained over Pd/NiAl acid were
2O4 and
obtained over Pd/NiAl
Ru/NiAl2O4. Thus, Ni2P-containing O
2 4 and Ru/NiAl
catalyst O . Thus,
2 obtained
4 Ni P-containing catalyst obtained
in 2situ is an effective catalyst for PET in
situ is an effective catalyst
recycling into BTX fraction. for PET recycling into BTX fraction.

3.3. Recycling Tests

3.3. Recycling Tests
The possibility of catalyst reuse has been investigated within 5 runs (Figure 6). The
The possibility of catalyst reuse has been investigated within 5 runs (Figure 6). The
recycling test runs of Ni2 P were carried out under moderate PET conversion (250 ◦ C,
recycling test runs of Ni2P were carried out under moderate PET conversion (250 °C, 7
7 MPa H2 , 6 h). The choice of reaction conditions for the recycling test is determined by the
MPa H2, 6 h). The choice of reaction conditions for the recycling test is determined by the
fact that, at quantitative conversion or close to it, a possible deactivation of the catalyst may
fact that, at quantitative conversion or close to it, a possible deactivation of the catalyst
not be noticed. Hence, the conditions when the BTX selectivity was the highest at moderate
may not be were
conversion noticed. Hence,
chosen. thethe
After conditions when
1st run, the the BTX was
conversion selectivity
52 wt. was thealmost
% and highestdid
at
moderate
not changeconversion were
after the 2nd chosen.
run. Afterafter
However, the 1st
the run, the of
3rd run conversion wasthe
the reaction, 52 conversion
wt. % and
almost did not change after the 2nd run. However, after the 3rd run of the
gradually decreased and after the 5th run it was 18 wt. %. Product selectivity changedreaction, the
conversion gradually decreased
slightly during test runs. and after the 5th run it was 18 wt. %. Product selectivity
changed slightly during test runs.
Polymers 2023, 15, x FOR PEER REVIEW 9 of 11
Polymers 2023, 15, x FOR PEER REVIEW 9 of 11
Polymers 2023, 15, 2248 9 of 11

Figure 6. The recycling test of Ni2P-containing catalyst.

Figure
Figure 6.
6. The
The recycling
recycling test
test of
of Ni
Ni22P-containing
P-containingcatalyst.
catalyst.
The catalysts obtained after the 1st and the 5th run were investigated using the XRD
The catalysts
The
technique catalysts
(Figure 7). obtained
obtained
After the after
after the
1stthe
run, 1st
1st and
Niand the 5th
the
2P phase 5th
(PDFrun№74-1385)
run were investigated
were investigated using
with theusing
average thecrys-
the XRD
XRD
technique
technique
tallite (Figure
size (Figure
of 30 ± 27).nm7). After
After
wasthe the 1st
1st run,In
obtained. run, Ni P
Niaddition,
2P phase
2 phase
(PDF (PDF №74-1385)
№74-1385)phases,
two phosphate with
with the the
Niaverage average
3(PO4)2 (PDF crys-
crystallite
tallite
№35-987) size
size and
of of± 30
30Ni(PO2 nm±3)22was
nmobtained.
(PDF was obtained.
№28-708), In addition,
In addition,
with twocrystallite
two phosphate
the average phosphate
phases,phases,
sizes 64Ni±43)214
of3(PO
Ni (PO
(PDF4 )2
nm
(PDF
№35-987)
and №35-987)
32 ± 5 and and
nm, Ni(PO Ni(PO
3)2 (PDF
respectively, ) (PDF №28-708),
№28-708),
3 2 were with
withAfter
identified. the average
the average
the 5th run, crystallite
crystallite sizes
sizesNiH
Ni2P and of
of 64 64
2P2O
±
± 71414 nm
nm
(PDF
and 32 ±±55phases
and 32
№34-617) nm,respectively,
nm, respectively,
with average were
were identified.
identified.
crystallite After
of 29the
sizesAfter ± 25th
the nmrun,
5th andNi
run, 6122P
Ni ± and
P NiH
andnm,
11 NiH 22P
P22OO77 (PDF
respectively,(PDF
№34-617)
№34-617)
were phases
phases
observed. with
withthe
Thus, average
average
average crystallite
crystallite
crystallite sizes
sizes
sizeof 29
ofof29Ni± 2 nm
± 22P nm and 61
and
remained ±±11
61unchanged.
11nm,
nm,respectively,
respectively,
However,
were observed.
werecomposition
the observed. Thus, Thus, the
of thethe average crystallite
average crystallite
phosphate phases has size
size of Ni 2 P remained
of Ni2P remained
changed. unchanged.
Moreover,unchanged. However,
However,
the peak intensity of
the
each composition
the composition
phase in theof of the
XRD phosphate
the pattern
phosphate phases has
phasestohas
is known changed.
changed.
depend Moreover,
on itsMoreover, the peak intensity
the peak
volume content intensity
in the of each
mixture. of
phase
each
As in thein
a phase
result, itXRD
the pattern
can XRD
be assumed is known
pattern is known
that to depend
the on
to depend
content of its volume
on
the content
its2Pvolume
Ni phase in theinmixture.
content
decreases theAs
the mixture.
after 5tha
result,
As
testarun itcompared
canitbecan
result, assumed
be that
assumed
to the the
1st one, content
that of thebe
the content
which could Ni
ofthe P phase
2the Ni2P decreases
reason phase after the
decreases
for a decrease 5th catalyst
inafter
the test run
the 5th
compared to
test run compared
activity. the
However, to 1st one,
the 1st
other which could
one, which
reasons be the reason for
could bedeactivation,
of the catalyst a decrease
the reason forand in the
a decrease catalyst activity.
in theimprove-
the stability catalyst
However, other reasons of the catalyst deactivation, and the stability improvement remain
activity.
ment However,
remain the goalsother reasonsstudy.
of further of the catalyst deactivation, and the stability improve-
the goals of further study.
ment remain the goals of further study.

Figure 7. XRD patterns of the spent catalysts obtained at 250 ◦ C, 7 MPa H2 , 6 h after the 1st and 5th
Figure 7. XRD
recycling patterns of the spent catalysts obtained at 250 °C, 7 MPa H2, 6 h after the 1st and 5th
test run.
recycling test run.
Figure 7. XRD patterns of the spent catalysts obtained at 250 °C, 7 MPa H2, 6 h after the 1st and 5th
4. Conclusions
recycling test run.
Nickel phosphide catalysts were obtained in situ in the reaction medium during PET
waste recycling. The presence of Ni2 P phase in the catalysts was confirmed using XRD and
HRTEM. It was shown via XRD that an increase in reaction temperature and pressure has
almost no effect on the phase composition. However, an increase in temperature promotes
Polymers 2023, 15, 2248 10 of 11

crystallite growth, which can be a result of sintering. Nickel phosphate was also obtained
using XRD and XPS. The Ni2 P-containing catalyst obtained was applied in the PET to
BTX thermocatalytic conversion for the first time. An increase in reaction temperature
and pressure promotes an increase in BTX selectivity. The highest selectivity of 93% for
BTX fraction was shown to be obtained at quantitative conversion. Then, the possibility
of catalyst reuse was investigated. The PET conversion was found to decrease after the
3rd run, which might be related to the decrease in the Ni2 P phase content in the catalyst
sample. The investigation of other reasons for Ni2 P-based catalyst deactivation, and the
catalyst stability improvement are the goals of further studies.

Supplementary Materials: The following supporting information can be downloaded at: https://
www.mdpi.com/article/10.3390/polym15102248/s1, Table S1: The average crystallite sizes of the
catalyst sample phases calculated using the Scherrer equation; Table S2: Oxidation states (Ni2p,
and P2p regions) on the Ni2 P surface; Table S3: The results of gaseous carbon-containing product
analysis; Figure S1: High-angle annular dark field–scanning transmission electron microscopy
(HAADF–STEM) images and EDX elemental mapping of the Ni2 P-containig catalyst obtained in
situ at (a) 300 ◦ C; (b) 340 ◦ C; (c) 380 ◦ C; Figure S2: Part of the 1 H NMR spectrum of terephthalic
acid. 400 MHz, DMSO-d6 ; Figure S3: Part of the 13 C NMR spectrum of terephthalic acid. 100 MHz,
DMSO-d6 ; Figure S4: The effect of reaction temperature on the product selectivity under H2 pressure
of (a) 5 MPa; (b) 7 MPa; (c) 9 MPa.
Author Contributions: Conceptualization, M.G. and A.M.; validation, M.G. and M.M.; formal
analysis, M.G. and M.M.; investigation, M.G., M.M. and A.S.; resources, A.S.; writing—original
draft preparation, M.G.; writing—review and editing, M.G. and A.M.; visualization, M.G. and M.M.;
supervision, M.G. and A.M.; project administration, M.G. and A.M. All authors have read and agreed
to the published version of the manuscript.
Funding: The research was funded by Russian Science Foundation, grant number 22-79-00255;
https://rscf.ru/en/project/22-79-00255/ (accessed on 8 May 2023).
Institutional Review Board Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: The authors thank Ivan Levin from TIPS RAS for XRD analysis, Mikhail Ivantsov
from TIPS RAS for chromatographic analysis of gaseous products, and Yury Grigoriev from A.V.
Shubnikov Institute of Crystallography RAS for HRTEM analysis. This work was conducted using the
equipment of the Shared Research Center Analytical center of deep oil processing and petrochemistry
of the A.V.Topchiev Institute of Petrochemical Synthesis RAS and the JRC PMR IGIC RAS.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ritchie, H.; Roser, M. Plastic Pollution. Our World Data. 2018. Available online: https://ourworldindata.org/plastic-pollution
(accessed on 8 May 2023).
2. Welle, F. Twenty years of PET bottle to bottle recycling—An overview. Resour. Conserv. Recycl. 2011, 55, 865–875. [CrossRef]
3. Jiang, J.; Shi, K.; Zhang, X.; Yu, K.; Zhang, H.; He, J.; Ju, Y.; Liu, J. From plastic waste to wealth using chemical recycling: A review.
J. Environ. Chem. Eng. 2022, 10, 106867. [CrossRef]
4. García, J.M. Catalyst: Design challenges for the future of plastics recycling. Chem 2016, 1, 813–815. [CrossRef]
5. Yaroslavov, A.A.; Arzhakov, M.S.; Khokhlov, A.R. Disposable polymer packaging: A problem without a solution? Her. Russ. Acad.
Sci. 2022, 92, 600–608. [CrossRef]
6. Thiounn, T.; Smith, R.C. Advances and approaches for chemical recycling of plastic waste. J. Polym. Sci. 2020, 58, 1347–1364.
[CrossRef]
7. Tomás, R.A.F.; Bordado, J.C.M.; Gomes, J.F.P. p-Xylene oxidation to terephthalic acid: A literature review oriented toward process
optimization and development. Chem. Rev. 2013, 113, 7421–7469. [CrossRef]
8. Tan, T.; Wang, W.; Zhang, K.; Zhan, Z.; Deng, W.; Zhang, Q.; Wang, Y. Upcycling plastic wastes into value-added products by
heterogeneous catalysis. ChemSusChem 2022, 15, e202200522. [CrossRef]
9. Jing, Y.; Wang, Y.; Furukawa, S.; Xia, J.; Sun, C.; Hülsey, M.J.; Wang, H.; Guo, Y.; Liu, X.; Yan, N. Towards the circular economy:
Converting aromatic plastic waste back to arenes over a Ru/Nb2 O5 catalyst. Angew. Chem. Int. Ed. 2021, 60, 5527–5535. [CrossRef]
Polymers 2023, 15, 2248 11 of 11

10. Lu, S.; Jing, Y.; Feng, B.; Guo, Y.; Liu, X.; Wang, Y. H2 -free plastic conversion: Converting PET back to BTX by unlocking hidden
hydrogen. ChemSusChem 2021, 14, 4242–4250. [CrossRef]
11. Tang, H.; Li, N.; Li, G.; Wang, A.; Cong, Y.; Xu, G.; Wang, X.; Zhang, T. Synthesis of gasoline and jet fuel range cycloalkanes and
aromatics from poly(ethylene terephthalate) waste. Green Chem. 2019, 21, 2709–2719. [CrossRef]
12. Hongkailers, S.; Jing, Y.; Wang, Y.; Hinchiranan, N.; Yan, N. Recovery of arenes from polyethylene terephthalate (PET) over a
Co/TiO2 Catalyst. ChemSusChem 2021, 14, 4330–4339. [CrossRef]
13. Golubeva, M.A.; Zakharyan, E.M.; Maximov, A.L. Transition metal phosphides (Ni, Co, Mo, W) for hydrodeoxygenation of
biorefinery products (a review). Pet. Chem. 2020, 60, 1109–1128. [CrossRef]
14. Golubeva, M.A.; Maximov, A.L. Hydroprocessing of furfural over in situ generated nickel phosphide based catalysts in different
solvents. Appl. Catal. A 2020, 608, 117890. [CrossRef]
15. Golubeva, M.A.; Maximov, A.L. Catalytic system based on nickel(II) acetate and hypophosphorous acid for the selective
hydrodeoxygenation of guaiacol. Mendeleev Commun. 2019, 29, 550–552. [CrossRef]
16. Golubeva, M.A. In situ generated nickel phosphide based catalysts for hydroprocessing of levulinic acid. Pet. Chem. 2021, 61,
670–675. [CrossRef]
17. Guan, Q.; Li, W.; Zhang, M.; Tao, K. Alternative synthesis of bulk and supported nickel phosphide from the thermal decomposition
of hypophosphites. J. Catal. 2009, 263, 1–3. [CrossRef]
18. Gopalakrishnan, J.; Pandey, S.; Rangan, K.K. Convenient route for the synthesis of transition-metal pnictides by direct reduction
of phosphate, arsenate, and antimonate precursors. Chem. Mater. 1997, 9, 2113–2116. [CrossRef]
19. De Andrade, J.B.; Nunes, G.S.; Veiga, M.P.; Costa, A.C.S.; Ferreira, S.L.C.; Amorim, A.M.M.; Reis, S.T. Spectrophotometric and
inductively coupled plasma atomic emission spectrometric determination of titanium in ilmenites after rapid dissolution with
phosphoric acid. Talanta 1997, 44, 165–168. [CrossRef] [PubMed]
20. Naumkin, A.V.; Kraut-Vass, A.; Gaarenstroom, S.W.; Powell, C.J. NIST X-ray Photoelectron Spectroscopy Database, Version
4.1, National Institute of Standards and Technology, Gaithersburg MD. 2012. Available online: https://srdata.nist.gov/xps/
(accessed on 8 May 2023).
21. Sinha, V.; Patel, M.R.; Patel, J.V. PET waste management by chemical recycling: A review. J. Polym. Environ. 2010, 18, 8–25.
[CrossRef]
22. Peng, B.; Zhao, C.; Kasakov, S.; Foraita, S.; Lercher, J.A. Manipulating catalytic pathways: Deoxygenation of palmitic acid on
multifunctional catalysts. Chem. Eur. J. 2013, 19, 4732–4741. [CrossRef]
23. Ameen, M.; Azizan, M.T.; Ramli, A.; Yusup, S.; Abdullah, B. The effect of metal loading over Ni/γ-Al2 O3 and Mo/γ-Al2 O3
catalysts on reaction routes of hydrodeoxygenation of rubber seed oil for green diesel production. Catal. Today 2020, 355, 51–64.
[CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.