Small-scale screening of novel biobased monomers: the curious case of 1,3-cyclopentanediol

Article abstract of DOI:10.1039/C8RA08811J

In this work, we report on the small scale polycondensation and consecutive analysis of novel polyesters based on the potentially renewable 1,3-cyclopentanediol (CPdiol). To avoid evaporation of monomers during thin-film polymerization reactions, trimer pre-polyesters have been synthesized from the corresponding acid-chlorides with diol monomers. Polymerization of these trimers was explored by thermogravimetric analysis to identify potential side reactions, and to assess the ideal polymerization temperature. In general we observe that trans-1,3-cyclopentanediol exhibits good thermal stability up to 200 °C, whereas thermal dehydration of the alcohol end-groups occurs upon further heating. In contrast, for cis-1,3-cyclopentanediol, the ester bonds of the cyclopentane end-groups become labile, thereby generating carboxylic acid end-groups, and 3-cyclopentenol already at 180 °C. The thermal dehydration reactions yield double bond end-groups, which in turn facilitate cross-linking through cross-coupling and Diels-Alder reactions, leading to an increase in molecular weight. Despite the limited thermal stability of CPdiol, here we demonstrate that polymerization of CPdiol can successfully be achieved in thin-film polycondensation conditions at 180 °C, yielding molecular weights well above 10 kg mol^?1.

Full text of DOI:10.1039/C8RA08811J

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Small-scale screening of novel biobased
monomers: the curious case of 1,3-
Cite this: RSC Adv., 2018, 8, 39818
cyclopentanediol†
G. J. Noordzij,ab C. H. J. T. Dietz,c N. Leone,b C. H. R. M. Wilsens
b and S. Rastogib
*
´
In this work, we report on the small scale polycondensation and consecutive analysis of novel polyesters
based on the potentially renewable 1,3-cyclopentanediol (CPdiol). To avoid evaporation of monomers
during thin-film polymerization reactions, trimer pre-polyesters have been synthesized from the
corresponding acid-chlorides with diol monomers. Polymerization of these trimers was explored by
thermogravimetric analysis to identify potential side reactions, and to assess the ideal polymerization
temperature. In general we observe that trans-1,3-cyclopentanediol exhibits good thermal stability up to
200 ꢀC, whereas thermal dehydration of the alcohol end-groups occurs upon further heating. In
contrast, for cis-1,3-cyclopentanediol, the ester bonds of the cyclopentane end-groups become labile,
thereby generating carboxylic acid end-groups, and 3-cyclopentenol already at 180 ꢀC. The thermal
dehydration reactions yield double bond end-groups, which in turn facilitate cross-linking through
cross-coupling and Diels–Alder reactions, leading to an increase in molecular weight. Despite the limited
thermal stability of CPdiol, here we demonstrate that polymerization of CPdiol can successfully be
achieved in thin-film polycondensation conditions at 180 ꢀC, yielding molecular weights well above
Received 24th October 2018
Accepted 21st November 2018
DOI: 10.1039/c8ra08811j
rsc.li/rsc-advances
10 kg molꢁ1
.
overcome this mismatch in scale, a viable assessment method is
desired to polymerize and qualify monomers on a small scale.
High-throughput screening equipment is a well-known tool
in the fast screening of molecules in e.g. organic synthesis, drug-
discovery, and cell-experiments. Also in the eld of polymer
synthesis, semi-automated high-throughput techniques have
been used in solution polymerization,6,7 olen polymerization,8
and melt-polycondensation.9,10 For example, a small-scale high-
throughput melt-polycondensation experiment has been
described by Gruter et al.11 for the evaluation of catalyst effec-
tivity in the synthesis of poly(ethylene 2,5-furandicarboxylate)
(PEF), commonly considered to be the renewable replacement of
poly(ethylene terephthalate) (PET). In general, such screening
techniques can lead to faster catalyst development, but can also
provide systematic identication of structure–property relation
in polymers by varying monomer compositions. An added
benet of small-scale screening experiments is a minimal
requirement of monomer availability, making this an attractive
technique for assessing the potential of new building blocks.
Despite the advantages of high-throughput experimentation, full
automation for polymerization reactions can still be challenging
due to the time restrictions for polymer characterization. For
example, important polymer characterization techniques such
as gel-permeating chromatography (GPC) or differential scan-
ning calorimetry (DSC) require relatively long analysis times,
potentially becoming the bottleneck in high-throughput
screening processes.
Introduction
Worldwide, strenuous efforts are taking place to replace oil-
based monomers with renewable biomass-based monomers,
for both “drop-in” purposes and as new monomers.1–5 However,
in particular for new building blocks, it is not assured that these
monomers – or the resulting polymeric materials – will function
as desired. In order to fully assess such monomers and poly-
mers, generally large amounts (e.g. several kilos) are required to
conduct repeatable processing and characterization tests.
Unfortunately, academic research oen leads to new monomers
obtained via e.g. new (bio)catalytic processes having yields in
gram scale, or lower. Another major limitation for translations
of lab scale synthesis to large scale are the required revenues:
signicant investment in both time and money is needed to
develop new monomers and polymers on a larger scale. To
aChemelot InSciTe, Urmonderbaan 20F, NL-6167 RD Geleen, The Netherlands
bAachen-Maastricht Institute of Biobased Materials (AMIBM), Faculty of Science and
Engineering, Maastricht University, Brightlands Chemelot Campus, 6167 RD Geleen,
The Netherlands. E-mail: karel.wilsens@maastrichtuniversity.nl
cInorganic Membranes and Membrane Reactors, Dept. Chemical Engineering and
Chemistry, Eindhoven University of Technology, PO Box 513, Eindhoven, The
Netherlands
† Electronic supplementary information (ESI) available: ESI contains detailed
2D-NMR analysis of monomers, thin-lm polymerization details, and
MALDI-ToF-MS data. See DOI: 10.1039/c8ra08811j
39818 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
were obtained from ACROS. 2,5-Furan-dicarboxylic acid was ob-
tained from ABCR. 1,4-Cyclohexanediol (46/54 cis/trans) was ob-
tained from Alfa Aesar. Ethylene glycol was obtained from TCI.
Standard laboratory solvents were obtained from Biosolve.
Deuterated solvents were obtained from Buchem BV (Nether-
lands). The purchased compounds were used directly without
further purication, unless otherwise specied.
Scheme 1 Reaction pathway for the synthesis of 1,3-cyclopentanediol
from renewable furfuryl alcohol.12
The group of Zhang et al.12 has developed a new industrially
scalable and cost-effective synthesis route to 1,3-cyclo-
pentanediol (CPdiol) from biomass. These authors reported
that CPdiol, usually obtained from non-renewable cyclo-
pentadiene, could be obtained from the aqueous phase rear-
rangement of furfuryl alcohol, followed by hydrogenation over
RANEY® (Scheme 1). CPdiol was obtained as a mixture of cis
and trans isomers, which could be further puried via fractional
distillation under vacuum.
Typically in racemic cycloaliphatic monomers the trans
isomer results in higher rigidity in the polymer, resulting in an
increase in glass transition temperature (Tg). Higher symmetry
of the trans isomer also results into more perfect crystals,
favoring higher crystallinity, as opposed to the cis isomer.
Indeed, this effect is well described for polyesters having
cycloaliphatic rings, for example 1,4-cyclohexanediol
(CHdiol),13,14 1,4-cyclohexanedimethanol (CHdm),15 and
others.14,16–19 In most cases the polyesters with a high trans ratio
are semi-crystalline, whereas crystallinity is oen lost when
reaching, or exceeding, a 50/50 cis/trans mixture. With this
literature precedent we expect that the properties of polymers
with CPdiol will vary with the cis/trans ratio. Additionally,
compared to six-membered rings, the ve membered cyclo-
aliphatic ring in CPdiol is expected to suppress rigidity and
symmetry.20 This is due to the steric conguration of the
cyclopentane ring, which cannot adopt the well-known ‘boat’-
conformation found in cyclohexane.21 These characteristic
differences make CPdiol an interesting candidate for polymer-
ization in order to provide a chemical and physical comparison
with CHdiol and CHdm.
Characterization methods
1H-NMR and 13C-NMR spectra were recorded with a Bruker
Ultrashield 300 spectrometer (300 MHz magnetic eld). NMR-
samples were prepared by dissolving ca. 10 mg of sample in
0.5 mL deuterated solvent, including dimethyl sulfoxide
(DMSO-d6), deuterated chloroform (CDCl3), and deuterated tri-
uoroacetic acid (d-TFA). All spectra were referenced against
tetramethylsilane (TMS), or residual solvent peak from the
deuterated solvent.
Molecular weight (Mn, Mw) and dispersity (Đ) of the polymers
were calculated aer gel permeation chromatography (GPC) on
a PSS SECcurity GPC system using Agilent 1260 Innity instru-
ment technology. The GPC system was equipped with two PFG
combination medium micro-columns with 7 mm particle size
(4.6 ꢂ 250 mm, separation range 100–1.000.000 Da), a PFG
combination medium pre-column with 7 mm particle size (4.6 ꢂ
30 mm), and a Refractive Index detector (RI). Distilled
1,1,1,3,3,3-hexauoroisopropanol (HFIP) containing 0.019%
sodium triuoroacetate was used as mobile phase at 40 ꢀC, with
a 0.3 mL minꢁ1 ow rate. The GPC apparatus was calibrated
with poly(methyl methacrylate) standards obtained from PSS.
GPC samples were prepared by dissolving 5 mg of polymer in
1.5 mL HFIP overnight under constant shaking, the samples
were ltered over a 0.2 mm PTFE syringe lter prior to injection.
Thermal stability of compounds and screening of polymeri-
zation conditions were performed via thermogravimetric anal-
ysis (TGA) using a TA Instruments Q500. Experiments were
performed under a nitrogen atmosphere with a heating rate of
10 ꢀC minꢁ1, from room temperature (RT) up to 700 ꢀC. For
polymerization screening experiments the sample was heated to
the desired reaction temperature (generally 180, 200, or 220 ꢀC)
at a rate of 10 ꢀC minꢁ1 and kept isothermal for the desired
reaction time. Thermal transition temperatures of the polymers
were analyzed via differential scanning calorimetry (DSC) using
a TA Instruments DSC Q2000. Typically, two heating and cool-
ing runs were performed at 10 ꢀC minꢁ1, where the rst heating
was used to erase any thermal history in the samples. The glass-
transition (Tg), melt-transition (Tm), crystallization (Tc), and
cold-crystallization (Tcc) temperatures were obtained from the
second heating and cooling run. DSC samples were prepared by
loading 3–5 mg oven-dried samples in Tzero Hermetic
Aluminium pans.
Zhang et al.12 polymerized 1,3-CPdiol into polyurethanes,
and several (enzymatic) transesterication reactions involving
various 1,2- and 1,3-CPdiol structures are known.22–24 However,
to the best of our knowledge, polyesters having 1,3-CPdiol have
not been reported in literature. This might be attributed to the
fact that CPdiol is only available on small scale.
To overcome the challenges of the limited availability of
CPdiol, and to investigate the physical properties of polyesters
with CPdiol compared to polymers having cyclohexane coun-
terparts, a small-scale polymerization screening method is re-
ported in this publication.
Experimental section
Materials
Polarized optical microscopy (POM) images were recorded
1,3-Cyclopentanediol (15/85 cis/trans), 1,4-cyclohexanedimethanol on an Olympus BX53DP 26, equipped with a Linkam HFSX350
(38/62 cis/trans), adipoyl chloride, sebacoyl chloride, terephthaloyl Hotstage. POM was used to determine melt-temperatures of
chloride, tin(II) 2-ethyl-hexanoate, and anhydrous chloroform, samples which could not be measured via DSC (e.g. because of
DMF, THF, and toluene were obtained from Sigma Aldrich. degradation upon melting). POM samples were loaded on
Magnesium(IV) sulfate, 4-dimethylaminopyridine and pyridine a microscopy slide and heated at a rate of 10 ꢀC minꢁ1. The
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 39818–39828 | 39819

View Article Online
RSC Advances
Paper
melt-temperature or range was determined by visual aid of the
samples becoming isotropic in nature.
Matrix assisted laser desorption/ionization-time of ight-
mass spectroscopy (MALDI-ToF-MS) analysis was recorded on
a Voyager DESTR from Applied Biosystems (laser frequency
20 Hz, 337 nm and a voltage of 25 kV). The matrix material was
DCTB (40 mg mLꢁ1 in HFIP). Potassium triuoroacetate was
added as cationic ionization agent (5 mg mLꢁ1 in HFIP). The
polymer sample was dissolved in HFIP (1 mg mLꢁ1), to which
Scheme 2 General synthesis route to obtain trimer pre-polyesters.
the matrix material and the ionization agent were added mixture overnight, the THF was removed in vacuo. The product
(5 : 1 : 5). If the samples were not fully dissolved, the mixture was extracted with CHCl3 (3 ꢂ 50 mL) and washed 10w% aq.
was ltered over a 0.2 mm PTFE syringe lter prior to placing on Cu2SO3 (3 ꢂ 50 mL), and with 0.01 M aq. HCl (3 ꢂ 50 mL). The
the target plate.
CHCl3 extract was dried with MgSO4, ltered, and reduced in
vacuo. Generally, the extracts could be used as is, however some
required further purication by precipitation from CHCl3 in
ether. Typically the trimers from FDCA and terephthalic acid
Synthesis methods
Enantiomeric enrichment of 1,3-cyclopentanediol via frac- were obtained as white solids. Purity was conrmed by NMR,
tional distillation. 1,3-Cyclopentanediol from Sigma Aldrich and/or LC-MS. The pre-polyesters were mostly isolated as
was obtained as an cis/trans enantiomeric mixture having >95% trimers, however also some pentamers and heptamers are
purity. The commercial batch was analyzed prior to use via NMR present. Detailed synthesis and characterization of the trimers
and chiral GC to determine the cis/trans ratio, and was found to used in this study are supplied in the ESI.†
be 15/85 cis/trans (see ESI for analysis†). Based on the different
boiling point of the cis/trans enantiomers, the commercial
batch was puried by fractional distillation under vacuum.
Polymerization screening methods
Various cis/trans ratios were obtained via batch-wise fractional
Small scale screening in TGA. Trimer pre-polyesters were
distillation; Generally, 1,3-cyclopentanediol was loaded in polymerized in the TGA with or without catalyst. Trimers with
a 50 mL 2-neck round-bottom ask, equipped with a 10 cm catalyst were prepared by dissolving the trimer in CHCl3 fol-
vigreux column, distillation 3-way adapter, condenser path, and lowed by the addition of 1 mol% catalyst solution (0.01 M in
a vacuum adapter to collect different fractions. Temperature of anhydrous CHCl3). Next, the CHCl3 was removed in vacuo, and
ꢀ
ꢀ
the oil-bath was varied from 130 C to 160 C. Temperature of the trimers with catalyst were stored in a desiccator until use.
the boiling phase was measured in the 3-way adapter, and the Polymerization in the TGA was performed by loading 5–10 mg of
depth of the vacuum was noted. All fractions were obtained sample, followed by heating at a rate of 10 ꢀC minꢁ1 to the
either as colorless liquids or as colorless solids: 7.6% cis (110 ꢀC, desired polymerization temperature. Next, the samples were
ꢀ
ꢀ
0.85 mbar), 14% cis ꢀ(125 C, 1.4 mbar), 20.7% cis (120 C, 1.6 kept isothermal for the desired polymerization time, aer
ꢀ
mbar), 30% cis (116 C, 1.2 mbar), 41% cis (111 C, 1.3 mbar).
which they were cooled to room temperature and used for
Synthesis of FDCA-Cl. 2,5-Furandicarboxylic acid (16.5 gr, analysis.
0.11 mol) was loaded in a 100 mL 3-neck round-bottom ask,
Small scale screening reactor design and validation. The
equipped with a N2-inlet and a condenser, and was purged 3 small-scale polycondensation reactions were carried out in
times with argon/vacuum cycles. Thionyl chloride (27.5 mL, a stainless-steel reactor. The design for this reactor (Fig. 1, le)
0.38 mol) and DMF (100 mL, catalytic) were added, and the is based on earlier work of Gruter et al.:11 the stainless steel
reaction mixture was reuxed for 4 hours at 85 ꢀC, until the reactor block is designed to t on top of a heating plate. To
reaction mixture was a clear solution. A N2-ow over the reac- either side an inlet/outlet switch is attached for vacuum/N2
tion ensured the formed gasses to pass through a gas-washing control. On the inside is an interchangeable reactor block with
bottle (250 mL 0.2 M aq. NaOH) attached via the top of the space for up to 57 HPLC vials. Reactor temperatures were
condenser. Aer cooling, the excess thionyl chloride was monitored via one temperature sensor in the bottom–middle of
removed in vacuo, the product was isolated in several recrys- the reactor and via a temperature sensor located in the side of
tallization steps from toluene as white crystals (15.2 gr, 75% the reactor wall. The reactor lid is closed with 4 screws, and
yield).
a rubber O-ring providing air-tight sealing. Due to the large
ꢀ
ꢀ
General synthesis of trimer pre-polyesters. A generally mass of the reactor block heating from 25 C to 200 C is ach-
applicable synthesis method for the preparation of trimer pre- ieved in approximately 30 minutes. Temperature calibration
polyesters has been developed (Scheme 2). The syntheses were experiments were performed using a temperature probe in oil
performed in anhydrous conditions. A solution of 9 mmol diol, inside the reactor and consistently showed a temperature in
7 mmol pyridine, and catalytic amount of DMAP in 5 mL between the set temperature, and the temperature measured on
anhydrous THF was stirred in an ice bath in a 25 mL 2-neck the side as is shown in the ESI.†
ꢀ
round-bottom ask equipped with a condenser. At 0 C, under
The reactor is designed for thin-lm polymerization reac-
N2-ow, a solution of 3 mmol di-acid chloride in 3 mL anhy- tions, as thin-lm polycondensation reactions do not require
drous THF was added dropwise. Aer stirring the reaction stirring for effective condensate removal. Therefore the lm
39820 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Fig. 1 The left image depicts the used stainless-steel reactor with a block for 57 HPLC vials for polycondensation screening reactions. The image
on the right depicts the effect of monomer loading (1–100 mg) on Mn (top) and Đ (bottom) for PET polymerized in the same reactor.
thickness should be low enough so that diffusion and mass
transfer do not limit the rate of polymerization. For the mate-
rials used in this study, placement of 4 mg of polymer material
Results and discussion
Trimer synthesis
Following previous work of Sweileh et al.,25 a general synthesis
method for the preparation of trimer pre-polyesters, shown in
Scheme 2, was developed. Although the use of acid chlorides is
not sustainable, it allows for synthesis of trimer pre-polyesters,
which do not suffer from evaporation and sublimation during
thin-lm polycondensation conditions. In other words, these
trimer pre-polyesters are excellent candidates for rapid
screening and optimization of polymerization conditions.
Please note, for bulk polymerizations the use of acid chlorides is
unnecessary as dimethyl esters can readily be used.
in standard HPLC vial results in a thin-lm with a thickness of
roughly 200 mm (taking into account an inner vial radius of r ¼
2.5 mm and assuming a polymer density of r ¼ 1.0 g mmꢁ3 (V ¼
(p ꢂ r2 ꢂ h)/r)).
To test the actual effect of monomer loading on the poly-
merization results, a PET trimer – 1,4-bis(ethyleneglycol-
terephthalate) – was polymerized with a loading varying from
1 mg to 100 mg (Fig. 1, right). With increasing monomer
loading from 1 mg to 100 mg, the Mn values drops from >25 kg
molꢁ1 to <5 kg molꢁ1, respectively. For a monomer loading
around 10 mg, Mn values around 15 kg molꢁ1 were consistently
obtained on all runs. Therefore, to ensure repeatability of the
polymerization experiments, in combination with ease in
weighing, further work with this reactor is based on a monomer
loading of 10 ꢃ 1 mg. A detailed overview of the polymerization
reactions performed for assessment of the repeatability of the
thin-lm polymerization reactions is provided in the ESI.†
General small-scale screening reactor polymerization.
Trimer pre-polyesters were polymerized in a stainless-steel
reactor block with or without catalyst. All the vials were
weighed before and aer reaction to monitor the weight loss
during polymerization. Generally, 10 ꢃ 1 mg of trimer was
loaded in each vial, and 1 mol% of catalyst solution (0.01 M in
anhydrous CHCl3) was added when required. The reactor block
was placed in the reactor and next the reactor was closed aer
which it was purged 3ꢂ with a N2/vacuum cycle. The reactor was
heated under a low N2-ow to the desired polymerization
temperature, and kept at isothermal conditions for the desired
polymerization time. Reduced pressure was applied during the
polymerization when required. Polymerization was stopped by
switching from vacuum to N2, followed by cooling of the reactor.
Aer cooling down, the vials were weighed and the visible color
and state of the polymer materials were noted.
To evaluate the effect of cis/trans ratio in CPdiol, 1,3-
cyclopentanediol-furanoate (CP-F) trimers with varying cis
content (8%, 20%, 30%, and 40%) were prepared. In spite of the
addition of a 3-fold excess of free diol in the reaction mixture,
some traces of pentamer and heptamer oligomers were found in
addition to the expected trimer oligomers. The presence of
1
trimers, observed from H-NMR, LC-MS, and GPC analysis, is
exemplied in Fig. 2 for CP-F trimer having a 8/92 cis/trans ratio.
A detailed overview of the analysis of the other trimers used in
this work is provided in the ESI.† In general, with the used
synthesis method, the ratio of trimer/pentamer/heptamer olig-
omers was 1/0.15/0.01. Nevertheless, all oligomers could be
synthesized successfully using the reported synthesis method.
Interestingly, we observed that the CP-F trimers with the
trimers with increasing cis content displayed increased discol-
oration aer drying at 110 ꢀC under vacuum. To investigate the
cause of the increased discoloration the oven-dried samples
were analyzed with LC-MS. In addition to the trimer/pentamer/
heptamer distribution, new peaks with a mass difference of
ꢁ18, and ꢁ84 g molꢁ1 were identied. The weight loss of 18 g
molꢁ1 likely corresponds to the occurrence of a dehydration
reaction, whereas the weight loss of 84 g molꢁ1 suggests the
liberation of a 3-cyclopentenol (CPol) group.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 39818–39828 | 39821

View Article Online
RSC Advances
Paper
1
Fig. 2 The image on the left depicts the H-NMR spectrum of the CP-F trimer with 8/92 cis/trans ratio. The image on the right displays the
elution of the same trimer in LC-MS, highlighting the presence of small amounts of pentamer and heptamer.
To obtain the products in a pure form, the samples were
To this end, all trimers were mixed with 1 mol% of Sn(II)
dissolved in chloroform, resulting in a clear solution in octanoate catalyst and the same TGA experiments were per-
combination with black insoluble particles, which were ltered formed aer extensive drying (Fig. 3, right). Indeed, upon the
out. Precipitation in heptane yielded the desired products, addiꢀtion of the tin catalyst weight loss is found to start around
ranging from off-white to light brown in color. Interestingly, 100 C, followed by a stable plateau. Further heating results in
ꢀ
with increasing cis content, the yield dropped from ꢄ80% to degradation, starting above 220 C. The amount of weight-loss
ꢄ50%. Overall, these observations suggest that CPdiol exhibits between 100 ꢀC and 200 ꢀC increases with increasing cis-
limited thermal stability, where the cis isomer seems less ther- CPdiol content (respectively 3.9, 6.8, 9.1, and 13.5%, Fig. 3,
mally stable than the trans isomer. An image of the discolored right). These ndings indicate that trimers having a higher cis-
products and the analysis of the trimers before and aer puri- CPdiol content generate volatiles more rapidly during poly-
cation are provided in the ESI.†
merization, potentially indicating an increased rate of
polycondensation.
To identify the optimal polymerization temperature, all
trimers were polymerized in the TGA for one hour at 180, 200,
and 220 ꢀC (Table 1). As observed previously, an increased
weight-loss is observed with increasing cis-fraction, combined
with an increased discoloration of the nal product. Due to the
absence of a vacuum-step in the polymerization in the TGA, the
molecular weight of the resulting oligomers remained low, with
Mn varying between 500 and 3000 g molꢁ1. The melting
behavior of the polymers was determined via polarized optical
microscopy (POM) equipped with a Linkam hot-stage, rather
than in DSC due to the limited thermal stability of the samples.
The hot-stage was heated at 10 ꢀC minꢁ1, and the visual off-set
of melting was noted.
As expected, all samples showed rapid signs of (further)
discoloration with the evolution of gasses directly aer melting,
indicating that the degradation temperature (Td) is very close to
the melting temperature (Tm). Additionally, increasing of the
cis-fraction resulted in an increase in Mw and a decrease in Tm,
until a fully amorphous material is obtained for the polymer
having 40% cis-CPdiol. These ndings are in coherence with
literature on the cis/trans effect of other cycloaliphatic diols.
The difference in polymerization behavior of varying cis-
CPdiol content becomes apparent by analyzing the GPC traces
(Fig. 4). In general, theꢀtrimer containing 8% cis-CPdiol poly-
merized rapidly at 180 C, indicated by the narrow Đ and the
relatively high molecular weight compared to the polymer with
higher cis-CPdiol content. When increasing the reaction
temperature to 200 ꢀC, the rise of a high molecular weight tail is
Evaluation and optimization of polymerization conditions
TGA studies were performed to ascertain the right polymeriza-
tion conditions for each trimer and preventing potential
degradation resulting from the thermal instability of the CP-F
trimers. From the le image in Fig. 3, it can be observed that
the CP-F trimers are all thermally stable up to 200 ꢀC, irre-
spective of the cis/trans ratio. Considering that no evaporation,
ꢀ
sublimation, or degradation occurs below 200 C, polymeriza-
tion can be performed in TGA. In this case, all observed mass-
loss should result from the generation of the condensate
CPdiol: the mass loss at the theoretical maximum conversion
equals the mass-loss of one CPdiol group of the trimer, which
contributes to 31% of the weight.
Fig. 3 TGA experiments for CP-F trimer, with varying cis content, with
a heating ramp of 10 ꢀC minꢁ1, non-catalyzed (left) and catalyzed
(right).
39822 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Table 1 Results of small-scale polymerizations performed in TGA
Polymer
Cis
8%
8%
8%
20%
20%
20%
30%
30%
30%
40%
40%
40%
Pol. T.
W. loss
Color
Mn (g molꢁ1
)
Mw (g molꢁ1
)
Đ
Tm (ꢀC)
Td (ꢀC)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
Poly(C-PF)
Poly(CP-F)
Poly(CP-F)
Poly(CP-F)
180 ꢀ
200 ꢀ
220 ꢀ
180 ꢀ
C
C
C
C
11.1%
19.1%
26.1%
9.2%
20.3%
26.8%
11.8%
23.3%
28.9%
14.2%
23.5%
29.1%
Light yellow
Light orange
Light brown
Light orange
Light orange
Light brown
Light brown
Light brown
Brown
1224
1846
2769
546
1602
2680
525
1660
2342
663
2071
2330
2448
3628
6401
1206
3484
6772
1161
3724
7169
1414
4503
8129
2.0
260
300
285
290
285
267
245–280
245–282
254–274
—
285
308
295
296
296
281
280
280
274
n.d.
n.d.
n.d.
1.97
2.31
2.21
2.17
2.53
2.21
2.24
3.06
2.13
2.18
3.49
200 ꢀC
220 ꢀ
180 ꢀ
200 ꢀ
C
C
C
220 ꢀC
180 ꢀ
200 ꢀ
C
C
Transp. brown
Transp. brown
Dark brown
—
—
220 ꢀC
Fig. 4 GPC traces of poly(CP-F) polymerized at 180 ꢀC (left), 200 ꢀC (middle), and 220 ꢀC (right), with varying cis-1,3-CPdiol content from 8%
(black), 20% (red), 30% (blue) to 40% (purple).
observed in the molecular weight distribution. In fact, this is
true for all samples, and is particularly dominant in the sample
having 40% cis-CPdiol. Further heating to 220 ꢀC broadens the
molecular weight distribution even further, and this effect again
becomes more apparent with increasing amount of cis-CPdiol.
It should be noted that the oligomers with increasing cis-CPdiol
content became increasingly difficulty to dissolve in the HFIP
solvent. In fact, especially for the oligomers containing 30 and
40% cis-CPdiol, black insoluble components remained in the
HFIP solvent, which were ltered off prior to GPC analysis. This
indicates that the GPC traces shown in Fig. 4 might not reect
the true polymer content as not all material has dissolved.
Nevertheless, the formation of insoluble polymer particles
suggests the presence of a cross-linking side reaction: the
presence of a bimodal distribution is highly unlikely in
polycondensation reactions considering the continuously
occurring transesterication reaction in combination with
Flory's equireactivity theory. Instead, considering the
increasing discoloration and decreasing solubility of poly-
mers with increasing cis-CPdiol content, we consider it more
likely that cis-CPdiol undergoes a degradative cross-linking
reaction in addition to the polycondensation reaction,
thereby generating high molecular weight components. To
assess whether such a side reaction proceeds during poly-
merization, the obtained oligomers have been analyzed using
MALDI-TOF-MS.
Oligomer analysis using MALDI-ToF-MS
Assuming that the CP-F polycondensation reaction proceeds
without side-reactions, only the presence of linear chains with
cyclopentanol (CPol) groups on both chain ends is expected
(structure 1, Fig. 5). As is observed from LC-MS analysis,
thermal dehydration (2, 3), and loss of 1,3-CPol end-group (4, 5)
can be expected to occur as a side-reaction during polymeriza-
tion. The mass distributions of the oligomers aer polymeri-
ꢀ
ꢀ
ꢀ
zation at 180 C (blue, top), 200 C (red, middle), and 220 C
(black, bottom) of 1,3-CP-F trimers having 8% (Fig. 6, le), 20%
Fig. 5 Expected products for poly(CP-F) before and after thermal
stability experiments.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 39818–39828 | 39823

View Article Online
RSC Advances
Paper
Fig. 6 MALDI-ToF-MS spectra of poly(CP-F) polymerized at 180 ꢀC (blue, top), 200 ꢀC (red, middle), and 220 ꢀC (black, bottom), with cis
fractions of 8% (left), and 20% (right). Zoomed in on 750–1050 dalton, between repeating units n ¼ 2 (m/z ¼ 806.82) and n ¼ 3 (m/z ¼ 1029.02) of
linear diol 1.
(Fig. 6, right), 30% (Fig. 7, le), and 40% cis-CPdiol (Fig. 7, right) with 20%, 30%, and 40% cis-CPdiol polymerized at 180 ꢀC
are reported. The depicted region is selected between 800 and contain products corresponding to chains with one (structure 4)
1050 dalton, in order to show the distribution between repeat or two (structure 5) carboxylic acid end-groups. Since the
units n ¼ 2 (m/z ¼ 806.82) and n ¼ 3 (m/z ¼ 1029.02) of expected carboxylic acid moiety is not present as starting material, they
main product 1. A more detailed list, the calculated masses of have to be generated during polymerization. Generally,
possible chain distributions, and the full MALDI-TOF-MS carboxylic acid groups can be generated through ester hydro-
spectra of the trimers are provided in the ESI section.†
lysis by water, (degradative) chain-scission of ester bonds, or
As can be seen from Fig. 6, le, the polymer having 8% cis- complete removal of a 3-cyclopentenol end-group.26,27 Hydro-
CPdiol polymerized at 180 ꢀC shows a single distribution of the lysis by water would lead to a distribution of chains with a CPol
expected diol product 1 (as is noted on top of the gure). end-group on one side, and a carboxylic acid on the other side.
Dehydration of the alcohol end-group takes place when However, such a distribution was not detected in MALDI-TOF-
increasing the polymerization temperature, leading to a rise of MS analysis (data provided in ESI†). Instead, (degradative)
products 2 and 3 at the expense of product 1, as is clearly visible chain-scission of ester bonds would lead to product 4 and 5,
ꢀ
from the MALDI-ToF-MS spectrum taken at 220 C. Similar to however would also inherently lead to a drop in molecular
the samples having 8% cis-CPdiol, an increase in polymeriza- weight, which was not observed in GPC analysis (Fig. 4). Thus it
tion temperature for the sample having 20% cis-CPdiol results seems likely that only the CPol end-groups have been removed
in dehydration of the CPdiol end-groups, as is indicated by the from the product, leading to the observed drop in product 1,
rise in reaction products 2 and 3.
Interestingly, the presence of product 1 becomes less abun-
dant at 180 ꢀC as the cis content increases. Instead, the samples 200 ꢀC and 220 ꢀC for CP-F samples having 20%, 30%, and 40%
and rise in the carboxylic acid terminated products 4 and 5.
An increase in polymerization temperature from 180 ꢀC to
Fig. 7 MALDI-ToF-MS spectra of poly(CP-F) polymerized at 180 ꢀC (blue, top), 200 ꢀC (red, middle), and 220 ꢀC (black, bottom), with cis fractions
of 30% (left), and 40% (right). Zoomed in on 750–1050 dalton, between repeating units n ¼ 2 (m/z ¼ 806.82) and n ¼ 3 (m/z ¼ 1029.02) of linear
diol 1.
39824 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Scheme 3 Anticipated degradation process of oligomers. A cyclopentanol end-group can undergo thermal dehydration, generating a cyclo-
pentene end-group, which in turn can undergo a cross-coupling reaction.
cis-CPdiol results in a decrease in distributions 4 and 5. This Alder side reaction would result into the formation of distri-
could potentially be resulting from thermal decarboxylation of bution 9, as shown in Scheme 4. Indeed, such a distribution is
the generated FDCA moieties,26,27 which yields a mono- found in the polymer having 30% cis-CPdiol, (highlighted by
substituted furan based diene, which can undergo a Diels– dotted line 9 in Fig. 7, le). With increased polymerization
Alder reaction with CPene end-groups, also providing a poten- temperature to 220 ꢀC, distribution 9 decreases, and a distri-
tial route towards branched or cross-linked species. We will bution with a mass loss of 18 g molꢁ1 arises (dotted line 10,
elaborate further on this hypothesis later in this work.
Fig. 7). It is likely that distribution 10 corresponds to the
To recall, both LC-MS (provided in the ESI†) and the MALDI- dehydrated product of 9. In fact, this distribution is observed in
TOF-MS analysis depicted in Fig. 6 suggest that thermal dehy- both polymers having 20% and 40% cis-CPdiol, polymerized at
dration takes place on the external OH group of CPdiol, thereby 220 ꢀC. These ndings provide further proof for the occurrence
generating cyclopentene end-groups. These cyclopentene groups of a Diels–Alder reaction between the degradation products.
are likely to undergo cross-coupling reactions (Scheme 3),
It should be noted that the obtained mass-distributions in
leading to products 6, 7, and 8 (Fig. 8), thereby providing these MALDI-ToF-MS analysis seem limited to t only linear
a means to build up high molecular weight materials, in addi- chains that are ending with either carboxylic acid, CPol, or
tion to the polycondensation reaction. Indeed, a rise in products CPene end-groups. Although cross-linked or branched mass-
6, 7, and 8 can be observed in the samples with increasing cis- distributions are expected given the presence of insoluble
CPdiol content, conrming that CPene cross-coupling reactions components in GPC analysis, such cross-linked or branched
are occurring. This CPene–CPene coupling reaction also explain fractions are not detected in these MALDI-ToF-MS experiments.
the rise in molecular weight observed in GPC analysis of samples Most likely these branched or cross-linked materials are ltered
having higher cis-CPdiol content (Fig. 4).
off during sample preparation or are too low in concentration to
As indicated earlier, an additional chain-extension reaction measure. Nevertheless, the current MALDI-ToF-MS data
that can take place is the Diels–Alder addition reaction between provides valuable insights in the polymerizability and thermal
decarboxylated furan moieties and CPene groups. This Diels– stability of CPdiol as monomer in polycondensation reactions.
Stability difference cis and trans 1,3-cyclopentanediol
The obtained results in the polymerization stability study
suggest that there is an inherent difference in thermal stability
between the cis and trans isomer of 1,3-cyclopentanediol:
structures with mostly trans-CPdiol end-groups undergo
thermal dehydration at temperatures of 200 ꢀC and higher,
thereby generating CPene end-groups. Structures with cis-
CPdiol end-groups rather undergo degradative ester bond
cleavage at 180 ꢀC instead, thereby generating free carboxylic
acid groups in addition to CPene end-groups. It is possible that
Fig. 8 Expected products for poly(CP-F) after CPene cross-coupling
reaction and thermal dehydration.
the origin of this stability difference can be found in the
Scheme 4 Anticipated Diels–Alder side reaction between a furan group and a CPene group leading to a polymer distribution 9.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 39818–39828 | 39825

View Article Online
RSC Advances
Paper
Table 2 Overview of properties of prepared polymers of cycloali-
phatic diols 1,3-cyclopentanediol (CP), 1,4-cyclohexanediol (CH), and
1,4-cyclohexanedimethanol (CHdm), bridged by terephthalate (T),
adipate (A), and sebacate (S) moieties
different conformations of these two isomers. The conforma-
tions of cis- and trans-1,3-cyclopentanediol have been evaluated
using theoretical and NMR investigations in the work of
Koniotou et al.21 These authors reported that the most stable
conformers for cis-CPdiol have the hydroxyl groups in axial
positions, allowing for intramolecular hydrogen bonding.28 In
contrast, the most stable conformers for trans-CPdiol have the
hydroxyl groups in equatorial positions, where intramolecular
hydrogen bonding is not observed. When these ndings are
extrapolated to esters of 1,3-cyclopentanediol, preliminary
minimized energy calculations in ChemDraw 3D soware show
similar axial and equatorial positioning (Fig. 9). These confor-
mations possibly allow cis-CPdiol to undergo intramolecular
hydrogen bonding of the external hydroxyl group with the ester
bond on the same CP-ring (Fig. 9, le).
Mn (g molꢁ1
)
Mw (g molꢁ1
)
Tg (ꢀC)
Tm (ꢀC)
CP-T
CP-A
CP-S
CH-T
CH-A
CH-S
CHdm-T
CHdm-A
CHdm-S
5k
11k
57k
92k
1.4k
37k
45k
7k
65 ꢀ
C
240–300a ꢀC
Amorphous
17k
25k
1.1k
13k
16k
4k
ꢁ30 ꢀC
ꢁ38 ꢀ
C
45 ꢀ
C
n.o.
300+b ꢀC
117 ꢀ
75 ꢀ
5
ꢀC
C
C
ꢁ22 ꢀC
n.o.
240+b ꢀC
12k
18k
27k
42k
ꢁ30 ꢀ
C
92 ꢀ
44 ꢀ
C
C
ꢁ38 ꢀC
a
Degradation upon melt. b Obtained via polarized optical microscopy.
Combining our experimental ndings with the preferred
conformations of CPdiol esters, suggests that the origin in
thermal stability between cis- and trans-CPdiol might be related
to the absence or presence of intramolecular hydrogen bonding.
Even though in-depth mechanistic studies of the degradation
are beyond the scope of this research, a possible degradation
mechanism based on the conformations of CPdiol is postulated
in Scheme 5. In both cases, b-hydrogen elimination takes place,
followed by the breaking of a s-bond: either C–OH (le), or C–
OR (right). Without intramolecular hydrogen bonding it
appears that the weakest s-bond to break is C–OH, leading to
dehydration in the trans isomers (Scheme 5, le). However, with
intramolecular hydrogen bonding it appears that C–OR
becomes the weaker s-bond, leading to removal of the CPol
group, generation an acid-terminated chain (Scheme 5, right).
Furthermore, dehydration via b-hydrogen elimination is acid
catalyzed, and indeed it is observed that with increasing acid-
terminated species 5, dehydrated species (2 and 3) start to
form at the expense of diol terminated chains (1) (Fig. 6 and 7).
Therefore, it appears that the thermal degradation mechanism
of cis-1,3-CPdiol leads to free acid species, which in turn cata-
lyzes the dehydration of the remaining trans 1,3-CPdiol end-
groups, leading to an even lower thermal stability of the polymer.
Please note that there is no increase in the low molecular
weight fractions observed at the evaluated polymerization
temperatures (Fig. 4), indicating that the difference in thermal
stability at these temperatures seems to originate solely from
the difference in cis or trans end-groups. The thermal stability
seems enhanced once the 1,3-CPdiol end-groups are completely
polymerized even though ‘normal’ b-hydrogen chain scission
still occurs at elevated temperatures.27 Unfortunately, we were
unable to study the rate difference in b-hydrogen chain scission
between the cis and trans conformers of CPdiol. This is related
to the inability of the poly(CP-F) to build up high molecular
weight under the explored polymerization conditions.
Assessment of 1,3-cyclopentanediol as building block
In the previous section we reported on the thermal lability of the
cis-CPdiol, and the corresponding thermal limitations. Even
though presence of 15% cis content in the commercial batch of
1,3-cyclopentanediol limits the polymerization temperature to
180 ꢀC, polyesters could be prepared in small-scale thin-lm
polycondensation reactions. Apart from poly(CP-F), trimers of
CPdiol with adipic acid (A), sebacic acid (S), and terephthalic
acid (T) have been synthesized. Furthermore, trimers with
structurally related 1,4-cyclohexanediol (CH) (50/50 cis/trans
ratio) and 1,4-cyclohexanedimethanol (CHdm) (30/70 cis/trans
ratio) have been prepared and polymerized. The polymeriza-
ꢀ
tions were performed at 180 C with 1 mol% of catalyst for 3
Fig. 9 Minimized energy calculations of the mono-benzoate ester of
cis-CPdiol (left) and trans-CPdiol (right). Calculated with Chemdraw
3D v17, via MM2 molecular dynamics and minimize energy.
hours under vacuum. The resulting polyesters were analyzed for
their molecular weight and thermal properties, as is
Scheme 5 Proposed thermal degradation mechanisms of trans (left) and cis (right) 1,3-CPdiol.
39826 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Fig. 10 DSC curves of prepared polymers of cycloaliphatic diols 1,3-cyclopentanediol (CP, bottom), 1,4-cyclohexanedimethanol (CHdm,
middle), and 1,4-cyclohexanediol (CH, top) with soft blocks adipate (A, left) and sebacate (S, right). Heating and cooling performed with
10 ꢀC minꢁ1, 2nd heating run is shown.
summarized in Table 2. Note, the full details of all polymeri-
zation experiments are available in the ESI.† The polymers are
Conclusions
abbreviated according to the diol and diacid used, e.g. poly(1,3-
cyclopentane-terephthalate) is abbreviated as poly(CP-T).
Poly(CP-T) was obtained as semi-crystalline polymer with an
unstable melt above 240 ꢀC, which is likely a result from the
previously observed instability of CPdiol based materials.
Furthermore, the melting temperature of poly(CP-S) is signi-
cantly lower compared to its 1,4-cyclohexanediol based coun-
terpart, i.e. 45 ꢀC versus 75 ꢀC. Instead, the melting behavior of
poly(CP-S) is comparable to the polymer containing its 1,4-
cyclohexanedimethanol counterpart, as they have similar
melting temperatures. Furthermore, poly(CP-A) is fully amor-
phous, whereas poly(CH-A) and poly(CHdm-A) are both semi-
crystalline in nature (Fig. 10, le) unfortunately, the Tg of
poly(CH-T) and poly (CHdm-T) could not be observed in DSC
analysis, likely due to their high crystallinity. With respect to the
glass transition temperature (Tg), polymers containing CPdiol
are rather comparable to materials containing CHdmol: both
monomers yield polymers with signicantly lowered Tg
compared to polymers containing CHdiol. These results indi-
cate that polyesters with CPdiol have a reduced rigidity
compared to polyesters with CHdiol, and should be compared
to the more exible CHdmol instead. However, although the
thermal transitions are comparable, the rate of crystallization of
polymers containing CPdiol is signicantly faster than for
polymers containing CHdmol, as is indicated by the absence of
cold-crystallization for poly(CP-S), shown in Fig. 10, right.
Overall, this preliminary polymerization screening data
shows that CPdiol can readily be polymerized under thin-lm
conditions with various dicarboxylic acids, despite the previ-
ously observed degradation reactions. We expect that the
application of reduced pressure promotes the polycondensation
reaction, and thereby limits the molecular weight build-up by
the formed degradation products reported earlier. In fact, under
A generic synthesis route has been developed to prepare trimer
pre-polyesters from various diols, in particular for 1,3-CPdiol,
and di-acid-chloride monomers. These trimers were directly
used in small-scale thin-lm polycondensation reactions as
they do not have the drawback of monomer evaporation. Using
this approach, TGA was used as a screening technique to
identify the thermal stability, and to determine optimal poly-
merization conditions of the CPdiol based trimers. In general
we observed that esters having the trans-CPdiol isomer are
ꢀ
found to be stable up to 180 C, and undergo dehydration at
200 ꢀC and higher. In contrast, esters having the cis-1,3-CPdiol
isomer are thermally labile, as they undergo rapid degradation
already at 180 ꢀC, which is attributed to the ability to form
intramolecular hydrogen bonding. Using the optimal poly-
merization temperature of 180 ꢀC, high molecular weight
polymers (>15 kg molꢁ1) having 1,3-CPdiol and sebacic acid or
adipic acid were successfully made. A comparative study
showed that polyesters having CPdiol exhibit lower rigidity and
crystallinity than their CHdiol counterpart. Instead, the CPdiol
moiety is found to be comparable CHdmol, which can likely be
attributed to the less stable half-boat conformation of CPdiol
compared to the stable boat-conformation of cyclohexane
structures.
Conflicts of interest
There are no conicts to declare.
Acknowledgements
these conditions, Mn values > 15 kg molꢁ1 are readily obtained This work was performed under the framework of Chemelot
for polymers based on CPdiol and sebacic acid or adipic acid. InSciTe and is supported by contributions from the European
However, the use of high-melting polymers with CPdiol should Regional Development Fund (ERDF) within the framework of
be avoided due to the inherently unstable melt of CPdiol, pre- OP-Zuid and with contributions from the province of Brabant
venting melt processing.
and Limburg and the Dutch Ministry of Economy.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 39818–39828 | 39827

View Article Online
RSC Advances
Paper
16 M. Colonna, C. Berti, E. Binassi, A. Celli, M. Fiorini,
P. Marchese, M. Messori and D. J. Brunelle, Polym. Int.,
2011, 60, 1607–1613.
17 C. Berti, A. Celli, P. Marchese, E. Marianucci, S. Sullalti and
G. Barbiroli, Macromol. Chem. Phys., 2010, 211, 1559–1571.
18 M. Zhang, R. B. Moore and T. E. Long, J. Polym. Sci., Part A-1:
Polym. Chem., 2012, 50, 3710–3718.
19 D. R. Kelsey, B. M. Scardino, J. S. Grebowicz and
H. H. Chuah, Macromolecules, 2000, 33, 5810–5818.
20 T. E. Sandhya, C. Ramesh and S. Sivaram, Macromolecules,
2007, 40, 6906–6915.
21 R. J. Abraham and R. Koniotou, Magn. Reson. Chem., 2003,
41, 1000–1008.
References
1 M. Sara, S. K. Brar and J. F. Blais, in Platform Chemical
Biorenery, Elsevier, 2016, pp. 249–283.
2 H. Nakajima, P. Dijkstra and K. Loos, Polymers, 2017, 9, 523.
3 M. A. Hillmyer, Science, 2017, 358, 868–870.
4 Y. Zhu, C. Romain and C. K. Williams, Nature, 2016, 540,
354–362.
5 J.-V. Bomtempo, F. Chaves Alves and F. de Almeida Oroski,
Faraday Discuss., 2017, 202, 213–225.
6 R. Hoogenboom and U. S. Schubert, Rev. Sci. Instrum., 2005,
76, 062202.
7 M. A. R. Meier, R. Hoogenboom and U. S. Schubert,
Macromol. Rapid Commun., 2004, 25, 21–33.
¨
22 A. Mattson, C. Orrenius, N. Ohrner, C. R. Unelius, K. Hult,
T. Norin, M. N. Homsi, F. K. H. Kuske, M. Haugg,
¨
8 V. Murphy, X. Bei, T. R. Boussie, O. Brummer,
¨
N. Trabesinger-Ruf and E. G. Weinhold, Acta Chem. Scand.,
G. M. Diamond, C. Goh, K. A. Hall, A. M. Lapointe,
M. Leclerc, J. M. Longmire, J. A. W. Shoemaker, H. Turner
and W. H. Weinberg, Chem. Rec., 2002, 2, 278–289.
9 R. A. Potyrailo, J. P. Lemmon and T. K. Leib, Anal. Chem.,
2003, 75, 4676–4681.
1996, 50, 918–921.
23 A.-B. L. Fransson, Y. Xu, K. Leijondahl and J.-E. Backvall,
¨
J. Org. Chem., 2006, 71, 6309–6316.
24 T. Agou, R. Ohata, Y. Mizuhata, N. Tokitoh, H. Fukumoto
and T. Kubota, J. Fluorine Chem., 2018, 210, 78–82.
25 B. A. Sweileh, H. R. Al-Qalawi and H. A. Mohammad, J. Appl.
Polym. Sci., 2014, 131, 39904.
26 V. Tsanaktsis, E. Vouvoudi, G. Z. Papageorgiou,
D. G. Papageorgiou, K. Chrissas and D. N. Bikiaris,
J. Anal. Appl. Pyrolysis, 2015, 112, 369–378.
27 Z. Terzopoulou, V. Tsanaktsis, M. Nerantzaki, D. S. Achilias,
T. Vaimakis, G. Z. Papageorgiou and D. N. Bikiaris, J. Anal.
Appl. Pyrolysis, 2016, 117, 162–175.
10 S. Brocchini, K. James, V. Tangpasuthadol and J. Kohn,
J. Biomed. Mater. Res., 1998, 42, 66–75.
11 G.-J. M. Gruter, L. Sipos and M. Adrianus Dam, Comb. Chem.
High Throughput Screening, 2012, 15, 180–188.
12 G. Li, N. Li, M. Zheng, S. Li, A.-Q. Wang, Y. Cong, X. Wang
and T. Zhang, Green Chem., 2016, 3607–3613.
13 J. Wang, X. Liu, Z. Jia, L. Sun, Y. Zhang and J. Zhu, Polymer,
2018, 137, 173–185.
14 A. Celli, P. Marchese, L. Sisti, D. Dumand, S. Sullalti and
G. Totaro, Polym. Int., 2013, 1210–1217.
28 K. Saegebarth, J. Org. Chem., 1960, 25, 2212–2213.
15 C. J. Kibler, A. Bell and J. G. Smith, J. Polym. Sci., Part A: Gen.
Pap., 1964, 2, 2115–2125.
39828 | RSC Adv., 2018, 8, 39818–39828
This journal is © The Royal Society of Chemistry 2018