Degradable precision polynorbornenes via ring-opening metathesis polymerization

Article abstract of DOI:10.1002/pola.27964

In an attempt to introduce monomer sequence control in a growing polynorbornene via ring-opening metathesis polymerization, we employ dioxepins to efficiently determine the location of the monomers on the macromolecule backbone. Owing to the acid-labile a

Full text of DOI:10.1002/pola.27964

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Degradable Precision Polynorbornenes via Ring-Opening Metathesis
Polymerization
Dafni Moatsou,1 Amit Nagarkar,2 Andreas F. M. Kilbinger,2 Rachel K. O’Reilly1
1Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry, CV4 7AL, United Kingdom
2
ꢀ
Department of Chemistry, University of Fribourg, Chemin Du Musee 9, Fribourg, CH-1700, Switzerland
Correspondence to: R. K. O’Reilly (E-mail: r.k.o-reilly@warwick.ac.uk)
Received 27 August 2015; accepted 19 October 2015; published online 22 November 2015
DOI: 10.1002/pola.27964
ABSTRACT: In an attempt to introduce monomer sequence con-
trol in a growing polynorbornene via ring-opening metathesis
polymerization, we employ dioxepins to efficiently determine
the location of the monomers on the macromolecule back-
bone. Owing to the acid-labile acetal group, dioxepins allow
scission of the polymer at the point of the dioxepin insertion
and thus provide an indirect way to determine the monomer
location. Additionally, dioxepins are used as spacers in the
synthesis of multiblock polynorbornenes that are readily cleav-
C
V
able to afford the individual polynorbornene blocks.
2015
Wiley Periodicals, Inc. J. Polym. Sci., Part A: Polym. Chem.
2016, 54, 1236–1242
KEYWORDS: degradable; dioxepins; norbornenes; ring-opening
metathesis polymerization
case of dienes and diacrylates.16–30 We previously studied
the use of exo norbornenes as functional monomers that
when added to the ROMP of the slower endo norbornenes,
were rapidly incorporated in the growing polymer chain,
thus resulting in copolymers with some control over the
monomer sequence.31 Nevertheless, when attempting the
introduction of a single exo norbornene moiety, the precision
was limited.
INTRODUCTION Our understanding on the way natural poly-
mers, such as proteins, inherit their properties from their
sequence and folding has led to a large body of research
being dedicated to the synthesis of polymers with control
over the monomer sequence. Step-growth-type polymeriza-
tions, such as solid-phase peptide synthesis, offer high preci-
sion and are widely used for the synthesis of such polymers,
although they are often limited by extensive purification
steps.1,2 Truly sequence-controlled polymers have also been
synthesized by the polymerization of macromonomers com-
prising of a short sequence.3,4 Nonetheless, relative precision
can also be achieved employing controlled chain-growth pol-
ymerizations for the synthesis of (multi)block copolymers,
such as in the cases of ring-opening polymerization,5 reversi-
ble addition-fragmentation chain transfer polymerization6,7
and transition metal-catalyzed radical polymerizations (such
as atom transfer radical polymerization, ATRP).8,9 Another
approach involves the exploitation of the kinetic parameters
that govern a copolymerization, thus allowing great precision
over the monomer sequence. A few such examples have
emerged based on the favored cross-polymerization of the
monomers, with the most frequently used ones being styr-
enics and maleimides in radical polymerizations.10–12 Kinetic
control of the monomer sequence has also been shown in
anionic polymerization,13–15 and ring-opening metathesis
polymerization (ROMP). A few examples in the literature
involve the synthesis of alternating copolymers using ROMP,
as a result of the catalyst-monomers pairing, such as in the
Apart from step-growth-type polymerizations, the introduction
of a single functionality within a synthetic polymer is usually
achieved at the polymer chain end, either the a-, the x- end,
or both, while in some polymerization methods the use of a
bifunctional initiating group yields a single functionality
within the polymer chain. Additionally, once a single function-
ality has been introduced to a polymer, post-polymerization
modifications broaden the range of accessible functional
groups.32,33 While this is readily feasible with most common
radical and ionic polymerizations,34–37 precise functionaliza-
tion in ROMP can be achieved with different pathways that
usually result in high yields, but often come with their own
challenges.38 Perhaps, the most common approach involves
the use of a chain transfer agent that allows the introduction
of a functional group onto the polymer chain end, however
often results in broad molecular weight distributions as such
reactions are dominated by chain transfer thus resulting in
secondary metathesis products.39–41 Typically, Ru-catalyzed
ROMP is quenched by the addition of a large excess of a vinyl
Additional Supporting Information may be found in the online version of this article.
C
V
2015 Wiley Periodicals, Inc.
1236
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1236–1242

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
ether, thus simultaneously rendering the catalyst inactive and
adding a functional group onto the polymer chain end. High
chain-end functionalization efficiency is also achieved by using
a purposely synthesized initiator, which results in the intro-
duction of a single functionality at the a-chain end. Neverthe-
less, such syntheses are challenging, which is far from ideal
given that a new initiator needs to be synthesized for each
desired functionality.42–44
Herein, we seek to introduce degradable monomers, func-
tional dioxepins, with precision in a polynorbornene chain,
while their potential use as single monomer inserts in the
ROMP of norbornenes is examined.
EXPERIMENTAL
Instrumentation
Nuclear magnetic resonance (1H and 13C NMR) spectra were
recorded in CDCl3 or CD2Cl2 solutions on a Bruker AC-250, a
Bruker DPX-300, a Bruker AV-400 or a Bruker DPX- 400, a
Bruker DRX-500, and a Bruker AV II-700 spectrometer.
Chemical shifts are reported as d in parts per million (ppm)
and referenced to the chemical shift of the residual solvent
resonances (CDCl31H: d 5 7.26 ppm; 13C: d 5 77.16 ppm)
and/or internal standards (TMS 1H: d 5 0.00 ppm; 13C:
d 5 0.00 ppm). High resolution mass spectra (HRMS) were
collected using a Bruker MaXis UHR-ESI TOF. Matrix-assisted
laser desorption/ionization time of flight (MALDI-ToF) mass
spectra were acquired on a Bruker Daltonics Ultraflex and
an Autoflex MALDI-ToF mass spectrometer in positive-ion
ToF detection performed using an accelerating voltage of 25
kV. Solutions in tetrahydrofuran (THF) of dithranol as matrix
(30 mg/mL), sodium or potassium trifluoroacetate as ioniza-
tion agent (2 mg/mL) and analyte (1 mg/mL) were mixed
before being spotted on the MALDI plate and air-dried. The
samples were measured in reflector ion mode and calibrated
by comparison to SpheriCal (Polymer Factory) single molecu-
lar weight standards (1200–8000 Da). Size exclusion chro-
matography (SEC) measurements were performed on an
Agilent 390-MDS equipped with differential refractive index
(DRI) and ultraviolet wavelength (UV) detectors. The separa-
tion was achieved by a guard column (Varian PLGel 5 lm)
and two mixed-D columns (Varian PLGel 5 lm) using THF
(2% Et3N mixture) or chloroform (2% Et3N mixture) as the
eluent at a flow rate of 1 mL/min. Data analysis was per-
formed using Cirrus v3.3 with calibration curves produced
using Varian Polymer laboratories Easi-Vials linear poly(sty-
rene) standards with molecular weights ranging from 162 to
2.4 3 105 Da.
A
new approach to chain-end functionalization is the
“sacrificial copolymerization” approach whereby a removable
monomer (typically adds an acid-labile group to the polymer
backbone) is copolymerized and is then “sacrificed,” remov-
ing it from the polymer, while its by-product becomes the
chain-end functionality. Grubbs et al. introduced the concept
of “sacrificial” polymerization in ROMP, whereby the copoly-
merization of cyclooctadiene and dioxepin, the latter being
the sacrificial monomer, is reported.45 Polymers synthesized
from dioxepins have backbones that contain acetal groups
which can be hydrolyzed by exposure to acidic conditions.
By doing so, the alcohol product remains on the polymer
and thus a hydroxyl end-group polymer is obtained.45
Kilbinger et al. used this concept in the ROMP of norbornenes,
synthesizing diblock copolymers whereby dioxepins were
used to form the second block that was then sacrificed.46 This
method showed high efficiency and further reaction of the
alcohol with a range of functional moieties was demonstrated,
indicating the versatility of this method as a single functional-
ity insertion approach.47–49 This concept was further
developed using vinyl lactones, dithiepins, and diazaphosphe-
pins.49–51 It should be noted that in most cases the sacrificial
monomer was observed to only form oligomers.
Extending their study on the kinetics of the dioxepin ROMP,
the efficiency of the dioxepin addition was investigated.
Three functional dioxepins were synthesized and used for
the chain extension of polynorbornene, while monitoring the
rate of the reaction. It was found that the polynorbornene
carbene was quickly substituted by the dioxepin carbene (in
under 20 min) using Grubbs’ first-generation catalyst (G1)
and isopropyl dioxepin as the monomer. It should be noted
that the homopolymerization of functional dioxepins was
unachievable, unless the catalyst benzylidene was substi-
tuted with an alkylidene, obtained by initiation with norbor-
nene.52 In a later report, the successful homopolymerization
of methyl dioxepin was shown, albeit in a noncontrolled
manner.53 Further sequential chain extensions with alternat-
ing batches of norbornene and dioxepin allowed the synthe-
sis of multiblock copolymers that, upon hydrolysis of the
acetals in the backbone, resulted in polynorbornenes with
both chain ends being hydroxyl-functionalized.
Materials
Synthesis of 2-Phenyl-4,7-dihydro-2H-1,3-Dioxepin
(DxpPhe)
For the synthesis of DxpPhe a modified literature proce-
dure46 was followed. Benzaldehyde (9.8 g, 92.5 mmol, 1
equiv.), cis22-butene-1,4-diol (8.6 g, 97.3 mmol, 1.05 equiv.),
and p-toluenesulfonic acid (170 mg, 0.9 mmol, 0.01 equiv.)
were dissolved in CH2Cl2 (30 mL). Anhydrous MgSO4 was
added until the supernatant was clear and the mixture was
stirred overnight at room temperature. The solvent was
removed under reduced pressure and the remaining oil was
passed through a basic alumina plug to yield the monomer
as a clear oil (10.33 g, 60% isolated yield).
Roberts et al. explored the potential single monomer inser-
tion in the ROMP of a peptide-bearing norbornene by first
reacting a functional norbornene with an equimolar amount
of the ROMP catalyst before the peptide-bearing norbornene
was added. It should, however, be noted that the success of
the single monomer addition was not discussed and a statis-
tical distribution was most likely obtained.54
1H NMR (500 MHz, CD2Cl2, d): 7.57 (d, 3J 5 6.5 Hz, 2H, Ar),
7.41 (m, Ar, 3H), 5.88 (s, acetal, 1H), 5.83 (m, CH5CH, 2H),
4.35 (m, -CH2, 4H); 13C NMR (125 MHz, CD2Cl2, d): 139.2,
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1236–1242
1237

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
129.9, 128.3, 128.1, 126.4, 102.1, 64.6; HRMS (m/z):
[M1Na]1 calcd for C11H12O2, 199.0730; found, 199.0732.
Synthesis of 2-Methyl-4,7-dihydro-2H-1,3-Dioxepin
(DxpMe)
A similar procedure to the synthesis of DxpPhe was followed
for the preparation of DxpMe, using acetaldehyde (3.9 g,
89.1 mmol, 1.3 equiv.) and cis22-butene-1,4-diol (9.2 g, 86.5
mmol, 1 equiv.) in a THF/CH2Cl2 (1:4) mixture. The product
was isolated in 98% yield (9.7 g).
FIGURE 1 Schematic representation of the addition of DxpPyr
to the polymerization of endoNbHex.
1H NMR (500 MHz, CD2Cl2, d): 5.73 (m, CH5CH, 2H), 4.99
(q, 3J 5 5.2 Hz, 1H, acetal), 4.25 (m, 4H, -CH2), 1.32 (d,
3J 5 5.2 Hz, 3H, -CH3); 13C NMR (125 MHz, CD2Cl2, d): 129.8,
101.1, 64.6, 19.8; HRMS (m/z): [M1H]1 calcd for C6H10O2,
115.0754; found, 115.0755.
(1.8 g, 20.7 mmol, 1 equiv.) in CH2Cl2 (20 mL). After removal
of the volatiles, the product was isolated via recrystallization
from ethyl acetate as yellow crystals in 17% yield (1 g).
Synthesis of N-Hexyl-endo-norbornene-5,6-dicarboximide
(endoNbHex)
In a round-bottom flask equipped with a magnetic stirrer
1H NMR (500 MHz, CD2Cl2, d): 7.90–8.40 (m, 9H, pyrene),
6.61 (s, 1H, acetal), 5.78 (m, 2H, CH5CH), 4.28–4.45 (m, 4H,
-CH2); 13C NMR (125 MHz, CD2Cl2, d): 131.8, 131.6, 131.2,
130.6, 128.6, 127.6, 127.5, 127.4, 126.0, 125.4, 125.3, 124.9,
124.6, 124.2, 123.9, 123.6, 101.0, 65.2; HRMS (m/z):
[M1Na]1 calcd for C21H16O2, 323.1043; found, 323.1043.
bar,
cis25-norbornene-endo22,3-dicarboxylic
anhydride
(10 g, 60.92 mmol, 1 equiv.) was dissolved in toluene
(200 mL) before addition of hexylamine (8.21 mL, 62.13
mmol, 1.02 equiv.). The reaction mixture was stirred under
reflux overnight. The solvent was then removed under
reduced pressure and the crude product was dissolved in
CH2Cl2 and passed through a short silica plug to remove
unreacted hexylamine. The pure product was collected as
off-yellow viscous oil (11 g, 73% isolated yield).
RESULTS AND DISCUSSION
Use of Dioxepins as Functional Monomers in the ROMP
of Endo Norbornenes
Initially, we sought to evaluate the use of dioxepins as
degradable building blocks in their copolymers with endo
norbornenes. We hypothesized that as a result of the degrad-
ability of the dioxepin moiety, its insertion at a precise loca-
tion on the polymer backbone could be determined by
scission of the polymer and characterization of the products.
This would further add to our understanding of the preci-
sion of the method. As such, endoNbHex was initiated by G1
catalyst before the addition of a pyrene-functional dioxepin
(DxpPyr) (Fig. 1).
1H NMR (400 MHz, CDCl3, d): 6.09 (m, 2H, CH5CH), 3.39
(m, 2H, 5CH-CH), 3.31 (t, 3J 5 7.5 Hz, 2H, N-CH2), 3.24 (dd,
3J 5 1.3, 1.5 Hz, 2H, 5CH-CH-CH), 1.55–1.74 (m, 2H, CH2
bridge), 1.42 (tt, 3J 5 8.3, 6.5 Hz, 2H, N-CH2-CH2-), 1.26 (m,
6H, (CH2)3), 0.87 (t, 3J 5 6.8 Hz, 3H, CH3); 13C NMR (100
MHz, CDCl3, d): 177.7, 134.4, 52.2, 45.7, 44.9, 38.4, 31.3,
27.7, 26.5, 22.5, 14.0; HRMS (m/z): [M1Na]1 calcd for
C15H21NO2, 270.1470; found, 270.1468.
The pyrene moiety permits characterization of the polymers
by SEC using both a refractive index and a UV detector (Fig.
2).
Synthesis of N-Hexyl-exo-norbornene-5,6-dicarboximide
(exoNbHex)
ExoNbHex was synthesized using the same procedure fol-
lowed for the synthesis of endoNbHex using cis25-norbor-
nene-exo22,3-dicarboxylic anhydride as a starting material.
While the incorporation of the pyrene moiety is apparent
from the increase of the intensity of the polymer trace as
detected by absorption at k 5 343 nm, corresponding to the
absorption of pyrene, the overall molecular weight does not
increase, even 5 h after the addition of dioxepin (Supporting
Information Table S1). The isolated polymers were also char-
acterized by MALDI-ToF (Fig. 3) where it was observed that
although dioxepin is capable of reacting with the alkylidene
living polymer end, as evidenced by the appearance of a sec-
ond distribution 30 min after the addition, the acetal in the
polymer backbone appears hydrolyzed and thus further
chain extension would not be possible. This conclusion was
based on the appearance of a third population corresponding
to hydroxyl terminal poly(NbHex), although it is most likely
a result of the sample preparation and the high voltage used
1H NMR (400 MHz, CDCl3, d): 6.28 (m, 2H, CH5CH), 3.45
(m, 2H, 5CH-CH), 3.26 (m, 2H, N-CH2), 2.65 (d, 3J 5 1.1 Hz,
2H, 5CH-CH-CH), 1.52 (m, 2H, N-CH2-CH2-), 1.24–1.47 (m,
2H, CH2 bridge), 1.24–1.37 (m, 6H, (CH2)3), 0.86 (t, 3J 5 6.6
Hz, 3H, CH3); 13C NMR (75 MHz, CDCl3, d): 177.5, 137.2,
17.2, 44.5, 42.1, 38.1, 30.7, 27.1, 26.0, 21.8, 13.4; HRMS (m/
z): [M1Na]1 calcd for C15H21NO2, 270.1470; found,
270.1467.
Synthesis of 2-(1-Pyrenyl)24,7-dihydro-2H-1,3-dioxepin
(DxpPyr)
A similar procedure to the synthesis of DxpPhe was followed
for the preparation of DxpPyr, using 1-pyrenecarboxaldehyde
(5 g, 21.7 mmol, 1.05 equiv.) and cis22-butene-1,4-diol
1
for the MALDI as by H NMR the acetal was found to be to a
1238
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1236–1242

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 2 Chromatograms from the ROMP of endoNbHex
before and after the addition of DxpPyr at different time inter-
vals, detected by DRI (top) and UV (k 5 343 nm) (bottom) (SEC
in THF).
great extent intact (Supporting Information Fig. S1). It
should also be noted that the peaks corresponding to the
addition of DxpPyr on the copolymer suggest single mono-
mer addition.
In order to circumvent the problematic characterization of
DxpPyr, we then sought to determine the ability of the less
bulky phenyl-functional dioxepin, DxpPhe, to react with the
alkylidene of the poly(NbHex) living chain end. To assess
this, we first determined the reactivity ratios of endoNbHex
and DxpPhe employing conventional Mayo-Lewis equations
(Supporting Information Figs. S2 and S3) and found them to
be 3.48 and 0.19, respectively. Although this suggests a
greater propensity for homopolymerization of endoNbHex, the
successful copolymerization with DxpPhe is also anticipated.
FIGURE 3 Top: MALDI-ToF mass spectra for poly(NbHex)
before and after addition of DxpPyr, at different time intervals.
Bottom: Expanded region of the spectrum with the correspond-
ing masses assigned to the sodium adducts of the reported
formulae.
It is a well-known fact that endo norbornenes suffer from
low polymerization rates due to steric interactions with the
propagating species and those steric interactions are possi-
bly the underlying reason behind the inability of dioxepins
to copolymerize efficiently with endo norbornenes. As such,
we sought to copolymerize the exo isomer of NbHex with
DxpPhe, as well as with the methyl functional DxpMe. While
establishing the reactivity ratios was not possible (see Sup-
porting Information), we set out to examine the ability of
dioxepins to chain extend a living polynorbornene. Previ-
ously, Kilbinger et al. reported the slow addition of dioxepins
to living polynorbornenes with a high dependence on the
dioxepin substituents;52 however, we found that using
DxpMe and long polymerization times (16 h) little change in
the polymer molecular weight was observed (Fig. 4). Further
investigation into the success of the chain extension revealed
that although dioxepin was successfully added to the grow-
ing polymer chains, only a single repeat unit was observed
(Fig. 5). It should be noted that, while unfunctionalized pol-
y(NbHex) still remains as the secondary population, no
FIGURE 4 Molecular weight distributions of the polymers
obtained from the ROMP of exoNbHex before and 16 h after the
addition of DxpMe, obtained by SEC in THF using a DRI detector.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 54, 1236–1242
1239