Metal-catalysed Poly(Ethylene) terephthalate and polyurethane degradations by glycolysis

Article abstract of DOI:10.1016/j.jorganchem.2019.120972

The recovery of Polyol from PolyUretane (PU) and Bis(2-HydroxyEthyl) Terephthalate (BHET) from Poly(Ethylene) Terephthalate (PET) from plastic waste by transesterification reactions was achieved, by using a wide scope of simple air-stable metal salts, Lew

Full text of DOI:10.1016/j.jorganchem.2019.120972

Journal Pre-proof
Metal-catalysed Poly(Ethylene) terephthalate and polyurethane degradations by
glycolysis
Raymundo Esquer, Juventino J. García
PII:
S0022-328X(19)30415-2
DOI:
https://doi.org/10.1016/j.jorganchem.2019.120972
JOM 120972
Reference:
To appear in: Journal of Organometallic Chemistry
Received Date: 14 August 2019
Revised Date: 8 October 2019
Accepted Date: 9 October 2019
Please cite this article as: R. Esquer, J.J. García, Metal-catalysed Poly(Ethylene) terephthalate and
polyurethane degradations by glycolysis, Journal of Organometallic Chemistry (2019), doi: https://
doi.org/10.1016/j.jorganchem.2019.120972.
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 B.V.

Metal-catalysed Poly(Ethylene) Terephthalate and Polyurethane
degradations by glycolysis
Raymundo Esquer and Juventino J. García*
Facultad de Química. Universidad Nacional Autónoma de México. Ciudad
Universitaria. México City, 04510. México.
juvent@unam.mx
Abstract
The recovery of Polyol from PolyUretane (PU) and Bis(2-HydroxyEthyl) Terephthalate
(BHET) from Poly(Ethylene) Terephthalate (PET) from plastic waste by transesterification
reactions was achieved, by using a wide scope of simple air-stable metal salts, Lewis acids,
mixtures of commercially available phosphine/phosphite ligands with metal salts and low-
valent [Ni(COD)2] as catalytic precursors, in the presence of ethyleneglycol as a
transesterification agent. A kinetic study with cobalt chloride led to the conclusion that
the PET degradation with metal salts behaves as a consecutive reaction with an induction
period. The use of σ-donor or σ-donor/π-acceptor bidentate phosphine ligands, such as
dcype and dppe, along with CoCl2 or [Ni(COD)2] improved the PET degradation process.
For both rigid and flexible PU, FeCl3 was the most active catalyst precursor.
Keywords: glycolysis, polyurethane, polyethylene terephthalate, degradation, catalysis.
1

Introduction
Poly (Urethane) (PU) and Poly (Ethylene Terephthalate) (PET) are polymeric materials
widely used in a variety of applications. Their properties have made them very useful and
largely produced, making PU and PET-recycling an important task to be addressed. Besides
solving pollution problems, their chemical degradation is an attractive field because it can
recover the monomers to be re-used for the manufacture of products with the same
quality as the original, like new PET bottles, along with the reduction of waste
production.1-4
Glycolysis is a simple method for PET and PU chemical degradation with advantages such
as mild reaction conditions, low-volatility of solvents and their products, the final product
Bis(2-hydroxyethy) Terephthalate (BHET) and the recovered polyol, can be easily
recovered and then mixed with virgin raw materials.5-9 Many catalysts have been
developed for PET degradation by glycolysis in recent years, for instance metal
acetates,11,12,13 oxides,14,15 carbonates and sulfates,16,17,18 titanium phosphate,13 zeolites,19
ionic liquids (ILs)7,20-22 and MOFs.23 For catalysts used in PU degradation include
alkanolamines, metal hydroxide, alkoxides, acetates and octoates,9 mostly for flexible PU
foam glycolysis. Also relevant recent examples for PET degradation using organocatalysed
protocols have been disclosed with good yields and recyclability of the catalysts.29
Even when these catalysts are effective in the PET and PU glycolysis, they also present
drawbacks and are limited by disadvantages in their technical scope. Thus, their chemical
recycling is still far from being a real solution. Consequently, more efforts to develop new
2

catalysts and improve the chemical degradation processes to make them more profitable
are needed.
Herein, we describe the improved PU and PET degradation by transesterification reactions
for polyol and Bis (2-HydroxyEthyl) Terephthalate (BHET) recovery using a wide scope of
air-stable metal-salts and Lewis acids with ethyleneglycol as a low-cost transesterification
agent. We also describe the use of phosphine/phosphite ligands with air stable transition
metal salts and [Ni(COD)2] for the first time in this type of applications, as well as the use
of triethylborane to carry out the glycolysis process in good isolated yields.
Results and Discussion
1. PET Degradation
1.1. Metal Salts and Lewis Acids
A wide range of metal salts and Lewis acids were used to carry out the PET degradation
using the conditions shown in Table 1. In all cases (entries 2-16), the monomer bis (2-
hydroxyethyl) terephthalate was obtained. A blank experiment showed that the
degradation of PET without a catalytic precursor wasn´t observed (entry 1), not even PET
solubility was achieved at 190oC.
The highest yields were obtained with ZrCl4 (69%, entry 15), CoCl2·6H2O (67%, entry 13)
and BEt3 (56%, entry 16). To note, once full conversion of PET was reached, the reaction
mixture became a single homogeneous phase, with yields of BHET higher than 55%, which
indicates that soluble intermediates (oligomers), continue reacting to produce BHET.
3

Table 1. PET degradation with metal salts and Lewis acids
Entry
Catalytic
Precursor
---
PET Conversion
Yield of BHET
TON
(%)
0
(%)
0
1
2
---
3
Ni(NO3)2∙6H2O
FeSO4∙7H2O
CuSO4∙2H2O
CuCl2∙2H2O
Zn (OAc)2∙2H2O
FeCl3∙6H2O
AlCl3
13
4
3
44
12
17
19
23
42
44
45
48
55
55
57
65
69
56
9
4
49
15
12
22
31
29
29
31
42
38
48
46
80
37
5
53
6
55
7
57
8
59
9
ZnCl2
60
10
11
12
13
14
15
16a
MnCl2∙4H2O
NiSO4∙7H2O
NiCl2∙6H2O
ZnSO4∙7H2O
CoCl2∙6H2O
ZrCl4
62
100
100
100
100
100
100
BEt3
*All reactions were carried out in a Parr reactor (25 mL) and heated at 190oC per 3 hours,
with 1 g of PET, 10 mL of ethylene glycol and 1 weight % of catalytic precursor. a Prepared
inside a glovebox (<1 ppm H2O and O2) and heated in a Schlenk Flask with a reflux
condenser, connected to an inert-gas/vacuum double manifold.
The activity trend found for metal chlorides Co2+ > Mn2+ > Zn2+ was different from that
reported for the acetates by Ghaemy and Mossaddegh24, Zn2+ > Mn2+ > Co2+ using
ethylene glycol. Likewise, with metal sulphates, the obtained trend was the following: Zn2+
> Ni2+ > Cu2+ > Fe2+. No clear relation between cation and anion has been established to
explain the improved yield of BHET, but probably is related to the cation interaction with
polyester, thus, related to its hardness, charge density and anion softness. The
4

combination of all these properties may allow an interaction with the carbonyl groups in
the polyester chains in PET that favour its degradation. Furthermore, the recyclability of
metal salts/ethylene glycol system was studied, having very good results in terms of yield
after more than one cycle (see SI section S18).
The PET degradation kinetics was studied using CoCl2·6H2O as a catalyst precursor at
200oC. A consecutive reaction behaviour with an initial induction period was identified in
Scheme 1, which was also confirmed with a PET pseudo-first order plot (S10). The
induction period was required to allow the metal-carbonyl interaction at the ester group
with the small amount of soluble PET. This Lewis acid type interaction allowed for a
nucleophilic attack at the carbonyl group by EG for the PET degradation.
Scheme 1. PET degradation at 200oC with ethylene glycol, using 1 weight % of CoCl2·6H2O
as a catalyst precursor. The mol % of oligomers was obtained by the difference between
the undepolimerized PET and BHET yield.
5

As seen in Scheme 1, as the polymer size decreases, the transesterification reactions are
slower. Furthermore, the presence of a soluble small quantity of the dimer was identified
by 1H-NMR in the reaction mixture (Figure S4), confirming that the degradation occurs via
a transesterification mechanism with the formation of a dimer prior to the monomer
BHET. No significant changes were observed between 1.5 and 3 hours; therefore, the last
step could be described as a slow equilibrium between the dimer and BHET, as described
in Scheme 2.
Scheme 2. Mechanistic proposal for PET degradation trough transesterification reactions
with CoCl2·6H2O.
The addition of triethylborane increased the yield of BHET between 1-3 hours and 170-
200oC (Scheme 3), giving the highest yields even compared with an increase of cobalt
chloride, probably due the synergic effect of both precursors.
6

CoCl2 (anh) + BEt3
CoCl2·6H2O
170
190
200
1
2
3
Temperature (oC)
Time (h)
Scheme 3. Use of triethylborane in PET degradation with cobalt chloride. a) Effect of
temperature at 3 hours; b) Effect of time at 200oC. *All reactions were carried out in a
Schlenk flask (50 mL), with 1 g of PET, 10 mL of ethylene glycol, 1 weight % of CoCl2 (anh)
and 1 weight % of BEt3.
1.2. Use of Metal salts with phosphine/phosphite ligands
In order to improve the PET degradation, we studied the effect of phosphine/phosphite
ligands with CoCl2 and NiCl2·6H2O, using ethylene glycol as solvent and transesterification
agent as well and an excess of ligand (Scheme 4).
7

CoCl2 (anh)
NiCl2·6H2O
a
a
---
PPh3
P(OiPr)3
dcype b
dppe b
dppf b
Ligand
Scheme 4. PET degradation with metal salts and phosphine/phosphite ligands at 190oC
and 3 hours. amol ratio of 1:6 of metal salt and monodentate ligand, respectively (x=6);
bmol ratio of 1:3 of metal salt and bidentate ligand, respectively (x=3).*All reactions were
carried out in a Schlenk flask (50 mL), with 1 g of PET, 10 mL of ethylene glycol and 1
weight % of the corresponding metal salts.
The use of monodentate ligands led to the complete inhibition of the degradation process,
while bidentate ligands showed to favour the process but only with ꢀ-donors like dcype.
Thus, compounds formed between cobalt and bidentate ligands were more stable in
ethylene glycol under the reaction conditions compared with those with monodentate
ligands where decomposition was observed. Additional experiments to assess the
influence of the amount of ligand in the degradation process with NiCl2·6H2O were made,
they confirmed that monodentate ligands did not facilitated the degradation process
(Scheme S11), while with the use of bidentate ligands the yield of BHET increased as the
ligand amount decreased (Table S11).
8

The reactivity of the mixture CoCl2:dcype in PET degradation was further optimised (Table
2). The yield of BHET recovered with different amounts of ligand (entries 1-2) and the use
of complex [Co(dcype)Cl2] (
) may be formed in the reaction mixture. The use of an increased load of complex
increased the BHET recovery.
1) (entry 3) was practically the same, leading to conclude that
(1
1
Table 2. Reactivity of CoCl2:dcype system in PET degradation using ethylene glycol.
Entry
Cat. Precursor Load (mol %) Temperature (oC)
Time (h) BHET Yield (%)
1
2
3
4
5
6
9
CoCl2 1:3 dcype
CoCl2 1:1 dcype
[Co(dcype)Cl2]
[Co(dcype)Cl2]
[Co(dcype)Cl2]
[Co(dcype)Cl2]
[Co(dcype)Cl2]
1.5
1.5
1.5
1.5
1.5
1.5
3.0
190
190
170
190
190
190
190
3
1
3
1
2
3
3
10
58
10
58
75
75 / 70a / 5b
82
*All reactions were carried out in a Schlenk flask (50 mL), with 1 g of PET, 10 mL of
a
ethylene glycol and 1.5 or 3.0 mol % of CoCl2 (anh) or catalyst precursor. Mercury drop
test. bAddition of DMSO drops.
A mercury drop test showed no significant change in BHET yield (Table 2, entry 6).
Additionally, the initial addition of DMSO decreased the monomer recovery almost
completely, suggesting that the presence of a highly coordinating solvent modifies the
catalytically active species by decreasing the vacant sites.
Based on the above experimental observations, two mechanisms can be envisaged:
9

1) A Lewis acid-catalysed mechanism, where ligands with higher ꢁ-acidity generate acid
metal centres that interact with donor moieties in the intermediates and the final product
(BHET), leading to product inhibition.
2) A Co(I)-Co(III)-Co(I) cycle, where Co(I) is generated in situ from Co(II), which might be
reduced in the presence of phosphines and ethylene glycol.
To shed some light in order to identify the possible mechanism, we carried out
experiments using initial amounts of BHET at the same reaction conditions, as shown in
Scheme 5.
0
2.5
5.0
x equiv. [BHET]0
Scheme 5. Inhibition by product with BHET. *All reactions were carried out in a Schlenk
flask (50 mL), with 1 g of PET, 10 mL of ethylene glycol and 1.5 mol % of catalyst precursor.
10

As a result, the addition of initial amounts of monomer decreased the final yield of BHET,
probably due to a coordination of this monomer to the Lewis acids involved in the
reaction mechanism. Additionally, the difference between cobalt and nickel could lie in
the possibility of creating vacant sites in solution, d7 Cobalt complexes could generate
vacant sites easier by octahedral-tetrahedral coordination equilibrium, while d8 square
nickel complexes create them through a high-energy associative mechanism or ligand
dissociation instead.
1.3. [Ni(COD)2].
In sharp contrast, we decided to assess the reactivity of an electron rich metal centre
using [Ni(COD)2] with and without phosphine/phosphite ancillary ligands for PET
degradation, as shown in Table 3. Surprisingly, the reactivity trend was similar to the one
previously observed with Co(II): the use of monodentate ligands gave low yields (entries 2-
3), while bidentate ligands gave higher yields (entries 4-6). Again, the intermediates
formed with monodentate ligands could be unstable in ethylene glycol under the reaction
conditions, while bidentate ligands gave good results.
To note, the use of [Ni(COD)2] without ligands was highly active in PET degradation (entry
1), possibly due the formation of nickel (COD)-solvent compounds or nickel nanoparticles
(NiNPs) (vide infra). The highest yield of BHET monomer, between bidentate ligands, was
achieved with dppe (67%, entry 5), compared to dcype (59%, entry 4) and dppf (15%,
entry 6).
11

Table 3. Reactivity of [Ni(COD)2] in PET degradation.
Entry
Ligand
PET Conversion
BHET Yield
(%)
(%)
1
---
PPh3
100
55
2a
3a
4b
5b
6b
27
0
13
0
P(OiPr)3
dcype
dppe
100
100
30
59
67
15
dppf
*All reactions were carried out in a Schlenk flask (50 mL) and heated at 190oC per 3 hours,
with 1 g of PET, 10 mL of ethylene glycol and 1 weight % of [Ni(COD)2]. amol ratio of 1:4 of
nickel precursor and monodentate ligand, respectively (x=4); bmol ratio of 1:2 of nickel
precursor and bidentate ligand, respectively (x=2).
The effect of the variation in the amount of ligand dppe, temperature and time were
examined, the results are summarised in Table 4. The highest yield was obtained with 2
equivalents of dppe, compared to the addition of 1 equivalent of dppe only or the use of
the complex [(COD)Ni(dppe)], where catalyst precursor decomposition was observed in
both cases. Additionally, mercury drop tests were also performed for entries 1, 3 and 6,
with no decrease in BHET yield, from which can be concluded the homogeneity of the
catalytic reaction is likely. In entry 1, mercury drop test is not totally conclusive, if we
consider that PET degradation occurs through reversible transesterification reactions,
both glycolysis and polymerisation processes could be mediated by NiNPs and be just
partially inhibited.
12

Table 4. Reactivity of [Ni(COD)2]: dppe system in PET degradation.
Entry
System
Temperature (ºC)
Time (h)
BHET Yield (%)
1
2
3
4
5
6
7
8
[Ni(COD)2]
190
190
190
190
190
190
190
200
3
3
3
1
2
3
5
3
55 / 60a
[(COD)Ni(dppe)]
[Ni(COD)2] 1:1 dppe
[Ni(COD)2] 1:2 dppe
[Ni(COD)2] 1:2 dppe
[Ni(COD)2] 1:2 dppe
[Ni(COD)2] 1:2 dppe
[Ni(COD)2] 1:2 dppe
44
45 / 48a
30
57
67 / 64a
59
71
*All reactions were carried out in a Schlenk flask (50 mL), with 1 g of PET, 10 mL of
ethylene glycol and 1 weight % of [Ni(COD)2] or the same amount (in mol%) of catalyst
precursor. a Mercury drop test.
Considering the current results and previous reports related to ester C-O activation with
nickel and the fact that low valent nickel compounds coordinates side-on in the C=O
moiety in carbonylic groups,25-27 a mechanistic proposal for the PET degradation is
presented in Scheme 6. The first step involves the ꢂꢃ-C,O coordination (side-on) of the
nickel centre at the carbonyl moiety with ligand dissociation (1), followed by the oxidative
addition of the ester C-O bond (2). Later on a nucleophilic attack by ethylene glycol (3)
generates a hydride complex, then a reductive elimination releases the hydrogenated
polymer chain (4). This cycle is repeated until the formation of low-molecular weight
oligomers and BHET monomer.
13

Scheme 6. Mechanistic proposal for PET degradation mediated by Ni(0).
While the first steps are favoured with
species in solution at high temperatures gets compromised. Thus, the highest yield was
obtained with an intermediate bearing a donor/ acceptor bidentate ligand, as dppe.
σ-donor bidentate ligands, the stability of formed
σ
π
To sum up, the best catalytic system in terms of activity and BHET yield was found to be
the use of CoCl2.6H2O as a cheap catalyst, compared with the already reported catalytic
systems.11-22 Noteworthy, the addition of BEt3 improved the BHET yield and reaction time
up to 84% yield in 3h with cobalt; also the use of an ancillary ligand can help to increase
the activity but a careful selection of it must be considered.
14

2. Polyurethane degradation
2.1. Flexible PU foam
The degradation of flexible PU foam was carried out also using metal salts and Lewis acids
as catalytic precursors and foam wastes in ethylene glycol. The GPC results (Table 5)
showed the recovery of high-molecular weight polymers or oligomers with Mn values
between (13 - 19) x 103; the polyol, with Mn values between (4-6) x 103, which is expected
for flexible polyol and other by-products, including additives and the final structure of Di-
Isocyanate, with molecular weights lower than 300.
Table 5. Flexible PU degradation with metal salts and Lewis acids.
Entry
Catalytic Precursor
ZrCl4
PO Mn (g/mol)
5,901
IP
Polyol (%)
1
2
3
4
5
6
1.115
1.690
1.067
1.324
1.384
1.394
23
91
96
92
92
92
MnCl2·6H2O
FeCl3·6H2O
CoCl2·6H2O
NiCl2·6H2O
AlCl3
5,278
5,438
6,336
5,337
5,298
*All reactions were carried out in a Parr Reactor (25 mL), with 0.5 g of PU, 10 mL of
ethylene glycol and 1 weight % of catalytic precursor. Polyol % was determined by GPC.
In almost all cases, the catalytic precursors used were found to be quite active, since more
than 90% of polyol was obtained, except for ZrCl4. However, highly polarising metal
cations, as Zr4+, could favour both degradation and re-polymerisation processes, while
intermediate hard metal cations and Lewis acids promote the PU foam degradation.
15

Since solubility was important for the reaction success the temperature and time of
reaction were optimised and the type of diol used in the PU degradation (Table 6). Thus,
no reaction was observed using temperatures lower than 160oC. Between 160-200oC, the
amount of recovered polyol was greater as the temperature increased (entries 2-4 for
FeCl3 and entries 9-10 for CoCl2). For longer reaction times at 200oC, no polyol was
recovered (entries 5-6), due to the thermal decomposition of products. Comparing both
glycols, the use of EG (entry 4) at 200oC, generated a higher polyol content product,
compared with 1,4-butanediol (entry 8).
Table 6. Additional studies of PU foam degradation.
Entry
Catalytic
Precursor
Time Temperature Glycol
(h)
PO Mn
(g/mol)
---
IP
Polyol
(%)
---
1
2
---
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
FeCl3·6H2O
CoCl2·6H2O
CoCl2·6H2O
CoCl2·6H2O
1
1
140
160
180
200
200
200
200
200
180
200
200
EG
EG
6,047
4,062
5,438
---
1.043
1.319
1.067
---
50
59
96
---
---
84
60
46
92
85
3
1
EG
4
1
EG
5
2
EG
6
7a
3
EG
---
---
1
EG
4,814
4,990
6,120
6,336
6,161
1.166
1.124
1.113
1.324
1.056
8
1
1,4-BD
EG
9
1
10
11
1
EG
0.5
EG
*All reactions were carried out in a Parr Reactor (25 mL), with 0.5 g of PU, 10 mL of glycol
and 1 weight % of catalytic precursor. Polyol % was determined by GPC. a 5 mL of glycol.
16

2.2. Rigid PU
The degradation of rigid PU was carried out using Polyurethane SelectophoreTM in ethylene
glycol (Table 7). The GPC results showed the recovery of high-molecular weight polymers
or oligomers with Mn values between (4-23) x 103; the polyol, with Mn values between (2-
11) x 102, as expected for rigid PU and other by-products, including additives and the final
di-Isocyanate.
Table 7. Rigid PU degradation with several catalyst precursors.
Entry
Catalytic Precursor
GPC Product Composition (%)
High-Molecular Weight
Polyol
Products
56
1
2
3
4
5
6
7
8
FeCl3
FeCl2
44
17
2
83
98
77
99
99
95
96
Zn(OAc)2
ZnCl2
23
1
CoCl2
NiCl2
1
[Ni(COD)2] 1:2 dppe
CoCl2 1:1 dcype
0
0
*All reactions were carried out in a Schlenk Flask (50 mL), with 0.3 g of PU, 2 mL of
ethylene glycol and 1 weight % of catalyst precursor. Products composition determined by
GPC.
The reactivity of metal chlorides seems to be related with their corresponding cation
charge density and hardness, according to the following order: Fe3+ (232 C/mm3) > Zn2+
17

(112 C/mm3) > Co2+ (108 C/mm3). This can be seen on comparing the results obtained with
both iron chlorides (entries 1-2). Therefore, more electron-rich metal precursors (entries
7-8), led to a lower polyol recovery, in sharp contrast with the results obtained for flexible
foam.
Further optimisation studies with FeCl3 showed that yield increased with temperature
reaching a maximum of polyol recovery at 200oC, after 5 h of reaction, giving an increase
in polyol molecular weight (Table 8, entries 3-4), probably due to a re-polymerisation
process at longer reaction times.
Table 8. Rigid PU degradation with FeCl3
Entry Time (h) Temperature (OC)
PO Mn (g/mol)
440
Ip
Polyol (%)
1
2
3
4
1
1
1
5
160
180
200
200
1.221
1.509
2.299
2.460
2
361
27
44
95
1,113
4,996
*All reactions were carried out in a Schlenk Flask (50 mL), with 0.3 g of PU, 2 mL of
ethylene glycol and 1 weight % of anhydrous FeCl3. Products composition was determined
by GPC.
A simplified cartoon-type representation of flexible and rigid PU is depicted in Scheme 7
and Figure 1, showing the types of transesterification reactions during PU degradation:
Type 1) those where the polymer is only degraded without polyol release via k1 (see
Scheme 7 and Figure 1).
Type 2) those where polyol is released via k2, making easier to recover selectively the
18

polyol (IP = 1.0 - 1.4), while PU degrades if k2 > k1. On using rigid PU foam the last is less
favoured due to its high cross-linked structure and the high functionality of the polyol,
consequently a stronger Lewis acid is needed to increase the possibility of a type 2
transformation. Most of this transesterification reactions led mainly to a molecular weight
decrease instead polyol recovery (k1 > k2), as can be seen in Table 9 using a non-highly
active catalytic precursor (CoCl2) for rigid PU degradation, where the molecular weight of
the polyurethane main structure decreases with a low release of polyol, as time increases.
Scheme 7. Transesterification reactions in PU degradation
a)
b)
Figure 1. Transesterification reactions in PU degradation. a) Flexible PU foam; b) Rigid PU
foam
19

Table 9. Rigid PU degradation with CoCl2
Entry
Time (h)
Polymer Mn
(g/mol)
8,753
IP
Polyol (%)
1
2
3
4
1
3
1.732
2.235
2.464
2.030
1
4
7
8
7,099
5,608
4,700
5
10
*All reactions were carried out in a Schlenk Flask (50 mL), with 0.3 g of PU, 2 mL of
ethylene glycol and 1 weight % of CoCl2. Products composition was determined by GPC.
Conclusions
The metal-catalysed degradation by glycolysis or transesterification for PET and PU (rigid
and flexible) was achieved to yield monomeric BHET and polyol respectively, in good
yields. The use of ancillary
σ donor/π acceptor bidentate phosphine ligands allowed the
PET degradation process with Ni(0) and Co(II) systems. Furthermore, the use of soft Lewis
acids in flexible PU degradation, led to the recovery of low-dispersity polyol with high
polyol recovery. In contrast, catalytic precursors with metals in high oxidation state were
more efficient in rigid PU degradation compared with those of low oxidation state. These
results may be of interest for further design of new heterogeneous catalysts that provide
important advantages such as catalyst recyclability.
Experimental Section
Unless otherwise mentioned, all manipulations were performed using the standard
20

Schlenk techniques in an inert-gas/vacuum double manifold or under an argon
atmosphere (Praxair 99.998) in an MBraun glovebox (<1 ppm H2O and O2). All liquid
reagents were purchased as reagent grade and were degassed before use. The solvents
were dried using standard techniques and stored in the glovebox before use. Deuterated
solvents were purchased from Cambridge Isotope Laboratories and were stored under 4 Å
molecular sieves for 24 h before use. NMR spectra were recorded at room temperature on
a 300 MHz Varian Unity spectrometer unless otherwise mentioned. 1H-NMR spectra are
reported relative to the residual protio-solvent. 13C{1H}-NMR spectra give the
characteristic carbon signal of each solvent. 31P{1H}-NMR chemical shifts are reported
relative to external 85% H3PO4. 1H-NMR, 13C{1H}-NMR and 31P{1H}-NMR spectra were
obtained in DMSO-d6, CDCl3 and/or THF-d8. [Ni(COD)2] was purchased from Aldrich and
purified on a Celite 545 plug, using THF or hexane. Anhydrous salts were obtained by
heating above 140oC and drying for at least 6 hours in an inert-gas/vacuum double
manifold below 0.005 mmHg and stored inside a glove box. IR (FTIR) spectra were
recorded at room temperature on a Spectrum Two FTIR Bruker Spectrometer. GPC
analyses were performed on a Waters 2695 ALLIANCE Separation Module. Post-consumer
PET provided from a single source of plastic bottles, was cut until 1x1 mm2 pieces, washed
with organic solvents, heated at 100oC and dried for at least 4 hours. Flexible PU provided
from commercial foams, was cut until 0.125 cm3 pieces, washed with organic solvents and
dried for at least 4 hours. Rigid PU (Polyurethane SelectophoreTM) was purchased from
Aldrich.
General Procedure for PET Degradation studies. All reactions were made in a 25 mL Parr
21

reactor or a 50 mL Schlenk Flask, typically charged with 1 weight % of catalyst precursor, 1
g of PET and 10 mL of ethylene glycol. If phosphine/phosphite ligands, triethylborane
and/or [Ni(COD)2] were used, all reactants were added to the Schlenk Flask in a MBraun
glovebox. The solution was heated with vigorous stirring at the selected reaction
temperature in an oil bath for the corresponding reaction time. If triethylborane was
added to the reaction mixture, the solution was heated in a Schlenk Flash ensambled with
a condenser, also using an inert-gas/vacuum double manifold. The reactor or Schlenk flask
was opened after the reaction time in a well-vented hood prior to workup. Then, two
parts of water were added to the reaction mixture, heated to boiling point. The under
polymerised PET and other by-products, like oligomers and/or ligand residues, are
separated by hot filtration. The under polymerised PET can be recovered from the residue,
washed with water, organic solvents, and then dried. The conversion of PET is defined as
WPET,0 − WPET,1
Conversion of PET (%) =
× 100%
WPET,0
where ꢄꢅꢆꢇ,ꢈ is the initial weight of PET, ꢄꢅꢆꢇ,ꢉ is the weight of undepolymerized PET.
The collected filtrate was concentrated under vacuum and then stored in a fridge at 0oC
for at least 12 hours. White crystalline flakes of BHET were obtained and separated by
filtration and dried. The yield of BHET is defined as
⁄
WBHET MWBHET
Yield of BHET (%) =
× 100%
⁄
WPET,0 MWPET
22

where ꢄꢅꢆꢇ,ꢈ is the initial weight of PET and ꢄꢊꢋꢆꢇ is the weight of BHET. ꢌꢄꢅꢆꢇ and
ꢌꢄꢊꢋꢆꢇ are the molecular weight of BHET (254 g/mol) and the PET repeating unit (192
g/mol), respectively.
General Procedure of flexible PU degradation studies. All reactions were made in a 25 mL
Parr reactor, typically charged with 1 weight % of catalyst precursor, 0.5 g of PET and 10
mL of ethylene glycol. The solution was heated with vigorous stirring to the reaction
temperature for the corresponding reaction time. After this, the reactor was opened in a
well-vented hood prior to workup. One part of water was added to the reaction mixture
and then transferred to 4 test tubes and centrifuged for 4 hours. The product mixture was
recovered from the liquid phase by liquid-liquid extraction with two parts of CH2Cl2 or
CHCl3. The solvent was evaporated in an inert-gas/vacuum double manifold and the
product mixture was dried for at least 4 hours.
General Procedure of rigid PU degradation studies. All reactions were made in a 50 mL
Schlenk flash, typically charged with 1 weight % of catalyst precursor, 0.3 g of PET and 2
mL of ethylene glycol. If phosphine/phosphite ligands and/or [Ni(COD)2] were used, all
reagents were added to the Schlenk Flask in a MBraun glovebox. The solution was heated
with vigorous stirring in an oil bath at reaction temperature for the corresponding
reaction time. After this, the Schlenk flask was opened in a well-vented hood prior to
workup. Two parts of water was added to the reaction mixture and then transferred to 4
test tubes and centrifuged for 4 hours. The product mixture was recovered from the liquid
23

phase by liquid-liquid extraction with two parts of CH2Cl2 or CHCl3. The solvent was
evaporated in an inert-gas/vacuum double manifold and the product mixture was dried
for at least 4 hours.
General Procedure for GPC. A 3-5 mg/mL solution of recovered product mixture in THF
(HPLC-grade) was prepared for each sample, cleaned with a porous filter and transferred
to HPLC vials. All the analyses were carried out on a Waters 2695 ALLIANCE Separation
Module, with an integrated refractive index detector (Waters 2414), in-line de-gasifier;
column furnace and a two-column bench (Waters HPLC): Styragel HR 5E with molecular
weight range from 50 to 1 x 105, and Styragel HR 5E with molecular weight range from 2 x
103 to 4 x 106. THF is used as eluent at a flow rate of 1.0 mL/min at 35oC. The calibration
curve was performed with 18 polystyrene standards, with molecular weight range of 3.7 x
102 to 4.3 x 106.
Mercury Drop Test. Following the procedure described above for PET degradation studies,
the very same reactions were prepared and two drops of elemental Hg were added to the
reaction mixture.
Preparation of [(dppe)Ni(COD)]: The nickel (0) complex [(dppe)Ni(COD)] was prepared
following the method of Freud et al28 for the dcype complex. Solid dppe (48 mg, 0.12
mmol) was added in one portion to a yellow solution of [Ni(COD)2] (33 mg, 0.12 mmol) in
THF (3 mL) and was stirred at room temperature. After the dppe was dissolved, the
24

solution was stirred for 30 minutes at room temperature. The solvent was evaporated and
further dried for six hours under vacuum to yield a yellow solid. 31P{1H}-NMR (121.32 MHz,
THF-d8, r.t.) 43.74 ppm (s).
ASSOCIATED CONTENT
Supporting Information available: Includes complete experimental procedures, additional
experiments, selected IR, NMR spectra and GPC determinations.
AUTHOR INFORMATION
Email: juvent@unam.mx
Conflict of interests
There are no conflicts to declare
ACKNOWLEDGMENTS
We thank CONACyT (A1-S-7657) and DGAPA-UNAM (IN-200119) for financial support;
CONACYT (607672) for a graduate study grant. Also, we thank M.S. Salvador López
Morales (IIE-UNAM) for technical support in GPC determinations. We also thank Dr. Alma
Arévalo for her technical assistance.
REFERENCES
[1] K. Hamad, M. Kaseem, F. Deri, Recycling of waste from polymer materials: An overview
of the recent Works. Polym. Degrad. Stab. 98 (2013) 2801-2812.
25

[2] A. Rahimi, J. M. García, Chemical recycling of waste plastics for new materials
production. Nat. Rev. Chem. 1 (2017) 1-11.
[3] N. George, T. Kurian, Recent Developments in the Chemical Recycling of Postconsumer
Poly(ethylene terephthalate) Waste. Ind. Eng. Chem. Res. 53 (2014) 14185-14198.
[4] G.F. Lu, Y.B. Ding, C.S. Zhao, D. S. Cui, Research development of chemical method
recovery waste polyurethane at home and aboard. Chem. Engineer. 109 (2004) 10, 45-51.
[5] H. Grunschke, W. Hammerschick, B. Nauheim, H. Medem, Process for depolymerizing
polyethylene-terephthalate to terephthalic acid dimethyl ester. U.S. Patent 3,403,115
(1968).
[6] Datye, K. V., Vaidya, A. A. Chemical Processing of Synthetic Fibers and Blends. John
Wiley and Sons: New York, (1984).
[7] H. Wang, Y. Liu, Z. Li, X. Zhang, S. J. Zhang, Y. Q. Zhang, Glycolysis of poly(ethylene
terephthalate) catalyzed by ionic liquids. Eur. Polym. J. 45 (2009) 1535−1544.
[8] K.M. Zia, H. N. Bhatti, I. A. Bhatti, Methods for polyurethane and polyurethane
composites, recycling and recovery: A review. React. Funct. Polym. 67 (2007) 8, 675-692.
[9] C. Molero, A, de Lucas, J. F. Rodríguez, Recovery of polyols from flexible polyurethane
foam by “split-phase” glycolysis with new catalysts. Polym. Degrad. Stab. 91 (2006) 894-
901.
[10] L. Bartolome, M. Imran, B. Gyoo Cho, W. A. Al-Masry, D. Hyun Kim. (2012) Recent
Developments in the Chemical Recycling of PET, Material Recycling - Trends and
Perspectives. Dimitris S. Achilias. INTECH.
26

[11] C. Cheng-Ho, C. Chuh-Yeahn, L. Yu-Wen, M. Ching-Feng, L. Wei-Tung. Studies of
glycolysis of poly(ethylene terephthalate) recycled from postconsumer soft drink
‐
bottles. I. Influences of glycolysis conditions. J. Appl. Polym. Sci. 80 (2001) 943–948.
[12] G. Xi, M. Lu, and C. Sun. Study on depolymerization of waste polyethylene
terephthalate into monomer of bis(2-hydroxyethyl terephthalate). Polym. Degrad. Stab.
87 (2005) 1, 117-120.
[13] K. Troev, G. Grancharov, R. Tsevi, I. Gitsov. A novel catalyst for the glycolysis of
poly(ethylene terephthalate). J. Appl. Polym. Sci. 90 (2003) 1148-1152.
[14] M. Imran, D. H. Kim, W. A. Al-Masry, A. Mahmood, A. Hassan, S. Haider, S. M. Ramay.
Manganese-, cobalt-, and zinc-based mixed-oxide spinels as novel catalysts for the
chemical recycling of poly(ethylene terephthalate) via glycolysis. Polym. Degrad. Stab. 98
(2013) 4, 904-915.
[15] F. Chen, F. Yang, G. Wang, W. Li. Calcined Zn/Al hydrotalcites as solid base catalysts
for glycolysis of poly(ethylene terephthalate). J. Appl. Polym. Sci. 131 (2014) 41053.
[16] R. Lopez-Fonseca, I. Duque-Ingunza, B. de Rivas, L. Flores-Giraldo, J. I. Gutierrez-Ortiz,
Kinetics of catalytic glycolysis of PET wastes with sodium carbonate. Chem. Eng. J. 168
(2011) 312-320.
[17] S. R. Shukla and K. S. Kulkarni. Depolymerization of poly(ethylene terephthalate)
waste. J. Appl. Polym. Sci. 85 (2002) 8, 1765-1770.
[18] S. R. Shukla and A. M. Harad. Glycolysis of polyethylene terephthalate waste fibers. J.
Appl. Polym. Sci. 97 (2005) 513-517.
27

[19] S. R. Shukla, V. Palekar, N. Pingale. Zeolite catalyzed glycolysis of poly(ethylene
terephthalate) bottle waste. J. Appl. Polym. Sci. 110 (2008) 501-506.
[20] Q. F. Yue, C. X. Wang, L. N. Zhang, Y. Ni, Y. X. Jin Glycolysis of poly(ethylene
terephthalate)(PET) using basic ionic liquids as catalysts. Polym. Degrad. Stab. 96 (2011)
399−403.
[21] A. M. Al-Sabagh, F. Z. Yehia, A. M. M. F. Eissa, M. E. Moustafa, G. Eshaq, A. R. M.
Rabie, A. E. ElMetwally, Glycolysis of Poly(ethylene terephthalate) Catalyzed by the Lewis
Base Ionic Liquid [Bmim][OAc]. Ind. Eng. Chem. Res. 53 (2014) 18443-18451.
[22] A. M. Al-Sabagh, F. Z. Yehia, A. M. M. F. Eissa, M. E. Moustafa, G. Eshaq, A. M. Rabie,
A. E. ElMetwally. Cu- and Zn-acetate-containing ionic liquids as catalysts for the glycolysis
of poly(ethylene terephthalate). Polym. Degrad. Stab. 110 (2014) 364-377.
[23] Q. Suo, J. Zi, Z. Bai, S. Qi. The Glycolysis of Poly(ethylene terephthalate) Promoted by
Metal Organic Framework (MOF) Catalysts. Catal. Lett. 147 (2017) 1, 240-252.
[24] M. Ghaemy, K. Mossaddegh. Depolymerisation of poly(ethylene terephthalate) fibre
wastes using ethylene glycol. Polym. Degrad. Stab. 90 (2005) 3, 570-576.
[25] F. Delbecq, P. Sautet. eta-2 versus .eta-1 coordination of aldehydes and ketones in
organometallic complexes. A semiempirical theoretical study. J. Am. Chem. Soc. 114
(1992) 7, 2446-2455.
[26] K. Muto, J. Yamaguchi, A. Lei, K. Itami. Isolation, Structure, and Reactivity of an
Arylnickel(II) Pivalate Complex in Catalytic C–H/C–O Biaryl Coupling. J. Am. Chem. Soc. 135
(2013) 16384-16387.
28

[27] N. Castellanos-Blanco, A. Arévalo, J. J. García. Nickel-Catalyzed Transfer
Hydrogenation of Ketones Using Ethanol as Solvent and Hydrogen Donor. Dalton Trans. 45
(2016) 13604-13614.
[28] R. R. A. Freund, H. Görls, J. Langer. Nickelalactones with an allyl subunit – the effect
of penta-coordination on structures and stability. Dalton Trans. 43 (2014) 13988–14000.
[29] C. Jehanno, I. Flores, A. P. Dove, A. J. Muller, F. Ruipérez, H. Sardon. Green Chem. 20
(2018) 1205-1212.
29