Polymer network formation using the phosphane-ene reaction: A thiol-ene analogue with diverse postpolymerization chemistry

Article abstract of DOI:10.1021/cm504784e

Air-stable primary phosphines were photopolymerized using phosphane-ene chemistry, the phosphorus analogue of the thiol-ene reaction, to fabricate a completely new class of polymer networks. It was demonstrated that the tunable thermal and physical properties accessible using thiol-ene chemistry could also be achieved using an analogous phosphane-ene reaction. At the same time, the presence of the ³¹P nucleus that is easily observed using NMR spectroscopy allowed the chemical structures of the networks to be directly probed using solid state NMR spectroscopy. Following its incorporation into the network, phosphorus offers the distinct difference and advantage of being able to undergo a diverse array of further derivatization to afford functional materials. For example, the networks were demonstrated to serve as effective oxygen scavengers and to bind transition metals (e.g., Pd). By using the air stable ferrocenyl phosphine (FcCH₂CH₂)PH₂, redox-active networks were produced and these materials could be pyrolyzed to yield magnetic ceramics. Overall, this demonstrates the promise of phosphane-ene chemistry as an alternative to thiol-ene systems for providing functional materials for a diverse range of applications.

Full text of DOI:10.1021/cm504784e

Article
pubs.acs.org/cm
Polymer Network Formation Using the Phosphane−ene Reaction: A
Thiol−ene Analogue with Diverse Postpolymerization Chemistry
†
†
†
†,‡
,†
Ryan Guterman, Amir Rabiee Kenaree, Joe B. Gilroy, Elizabeth R. Gillies, and Paul J. Ragogna*
†
‡
Department of Chemistry and the Centre for Advanced Materials and Biomaterials Research (CAMBR) and Department of
Chemical and Biochemical Engineering, The University of Western Ontario, 1151 Richmond Street, London, Ontario, Canada N6A
B7
5
*
S Supporting Information
ABSTRACT: Air-stable primary phosphines were photopolymerized using phos-
phane−ene chemistry, the phosphorus analogue of the thiol−ene reaction, to
fabricate a completely new class of polymer networks. It was demonstrated that the
tunable thermal and physical properties accessible using thiol−ene chemistry could
also be achieved using an analogous phosphane−ene reaction. At the same time, the
31
presence of the P nucleus that is easily observed using NMR spectroscopy allowed
the chemical structures of the networks to be directly probed using solid state NMR
spectroscopy. Following its incorporation into the network, phosphorus offers the
distinct difference and advantage of being able to undergo a diverse array of further
derivatization to afford functional materials. For example, the networks were
demonstrated to serve as effective oxygen scavengers and to bind transition metals
(
e.g., Pd). By using the air stable ferrocenyl phosphine (FcCH CH )PH , redox-
2 2 2
active networks were produced and these materials could be pyrolyzed to yield
magnetic ceramics. Overall, this demonstrates the promise of phosphane−ene chemistry as an alternative to thiol−ene systems
for providing functional materials for a diverse range of applications.
1
.0. INTRODUCTION
One element with the potential to benefit from the
fundamental chemical insights of the thiol−ene reaction is
phosphorus. In addition to its widespread utility in synthesis
Polymer networks are cross-linked polymers that have found
widespread use in actuators, microelectronics fabrication,
synthetic rubbers, coatings, tissue engineering scaffolds, and
other areas. The incredible diversity of these materials stems
from their highly tunable nature, ranging from soft hydrogels to
tough thermosets, with the properties dictated by the chemical
functionalities in the network as well as their cross-link
15−18
and materials science,
over the past several decades,
1
−6
phosphorus has increasingly been employed in polymer
1
7,19−23
chemistry.
In the 1960s, Pellon reported that thiols
and primary phosphines (RPH ) were similar with regard to
2
their radical mediated hydrothiolation (thiol−ene) and hydro-
24−26
7
,8
phosphination (phosphane−ene) chemistry.
As shown in
densities.
They can be synthesized thermally, but also
Figure 1A, upon hydrogen abstraction, phosphinyl and thiyl
radicals quickly add to CC bonds to form a carbon-centered
radical, which subsequently abstracts a hydrogen atom from a
nearby phosphine/thiol, leading to propagation. While hydro-
thiolation later became referred to as thiol−ene chemistry in
polymer and material science, phosphane−ene chemistry
remains entrenched in small molecule synthesis despite several
photochemically, where the photochemical approach is
generally referred to as photopolymerization or UV curing
and provides the possibility for both spatial and temporal
9
control over the polymerization process. While the vast
majority of polymer networks are synthesized using free-radical
polymerization of (meth)acrylates, thiol−ene polymerization
has garnered increasing popularity over the past decade as an
alternative approach. The network polymerization using thiol−
ene chemistry is relatively unique, as it is a radical mediated, air-
stable, step-growth process, combining the fast kinetics of
radical chemistry with reduced shrinkage, high conversion, and
delayed gelation, as described by Hoyle, Bowman, and Hawker
27−29
attempts to apply it to the production of macromolecules.
This likely stems from the relatively limited family of primary
phosphines as well as the difficulty in handling them, which
unlike thiols, are typically pyrophoric. However, recent
developments in fundamental phosphorus chemistry have
identified a systematic way to classify specific primary
phosphines as air-stable that possess remarkable oxidative
10−12
and their co-workers.
The key aspect to this approach is
the application of organosulphur chemistry for polymer science.
This represents one of relatively few widely used examples of a
30
resistance not explained simply by substituent steric effects.
13
polymerization reaction utilizing an inorganic element. It can,
therefore, serve as a template for the incorporation of other
inorganic elements into polymeric materials, thus expanding the
Received: December 29, 2014
Revised: January 20, 2015
14
scope and application beyond sulfur.
©
XXXX American Chemical Society
A
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 1. (A) Mechanism for hydrophosphination/phosphane−ene reaction and for hydrothiolation/thiol−ene reaction. (B) Schematic
represenation of polymer network formation using 1 and photoinitiator Irgacure 819. Soft gel-like materials are produced when TEGDAE is used,
while firm polymers capable of holding 1 kg of weight is produced when TTT is used.
This has initiated what some refer to as “The Primary
Phosphine Renaissance”.
2.0. RESULTS AND DISCUSSION
31
2
.1. Phosphane−ene Network Synthesis and Proper-
Katti et al. showed that a variety of air-stable phosphines
could be synthesized and incorporated into peptides using
standard methods without any oxidation or byproducts, a feat
not easily accomplished using typical phosphines containing a
ties. To successfully photopolymerize primary phosphines
using phosphane−ene chemistry, the molecule must possess
sterically unhindered R-PH
functionality, low molecular
2
weight, transparency to UV light, and facile olefin addition
chemistry. While aryl phosphines have previously been the
focus of phosphane−ene reactions of small molecules, the high
stability of the arylphosphinyl radical leads to inefficient
hydrophosphination. This arises from a reverse-addition
reaction as described by Pellon, resulting in sluggish or absent
3
2−34
P−H functionality.
These discoveries allow for the
widespread exploration of these interesting yet simple func-
tional groups beyond constructing small molecules. In this
context, we bring new perspective to the long established P−H
bond by harnessing its chemistry to develop for the first time
the phosphorus analogues of thiol−ene networks. This allows
for the generation of polymeric materials with a wide variety of
properties and functions including oxygen scavenging, redox
activity, polyelectrolytes, and ceramics. Our approach exploits
the concepts established for thiol−ene chemistry, such as
appropriate monomer choice, polymerization methodologies,
and inhibitor additives, to fabricate these new materials, while at
the same time expanding on their utility by exploiting the
diverse chemistry of phosphorus in derivatizing the network
2
5,28,35
polymerization.
A more reactive alkylphosphine should
be more efficient. Considering also that tetra-alkylthiols are
used in thiol−ene networks, we deemed primary phosphine 1
as the most suitable candidate for our studies because of its
resistance to oxidation, high P−H functionality, alkyl character,
and ease of synthesis (Figure 1B: see Experimental Section for
36
synthetic details).
Thiol−ene networks are typically synthesized by reacting
multifunctional thiols with olefins in a step-growth fashion,
where the monomer compositions and functional group
stoichiometry enables tuning of the chemical, thermal, and
3,37
31
mechanical properties of the material.
Depending on the
after polymerization. These materials can be analyzed using P
solid-state NMR spectroscopy (SS-NMR), providing a unique
handle not available in thiol−ene systems to probe these
materials and thus investigate the chemistry within the
networks.
bias imposed on the original formulation (i.e., excess olefin or
thiol), residual functional groups are supported by the polymer
network after polymerization and then exploited for further
38
chemistry. It follows then that trends in both the thermal and
chemical properties of phosphane−ene networks should mimic
B
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
a
Table 1. Effect of Changing Olefin Composition and PH /Olefin Ratios on Network Properties
2
polymer
PH :olefin ratio
swelling ratio
gel content (%)
Tg (°C)
Tg−ox (°C)
Oxcalc (%)
Oxexp (%)
2
SPN-1
SPN-2
SPN-3
FPN-1
FPN-2
FPN-3
0.75:1
0.5:1
0.92 ± 0.01
0.62 ± 0.01
1.52 ± 0.26
0.29 ± 0.01
0.05 ± 0.01
0.05 ± 0.01
89 ± 1
>98
−61
−52
−64
56
−39
−37
−54
-
-
-
5.1
4.0
3.0
6.6
5.4
4.6
5.2
4.1
3.4
6.5
1.6
1.6
0.38:1
0.75:1
0.5:1
81 ± 2
>98
>98
101
106
0.38:1
>98
a
The % mass increase upon oxidation, Ox, was measured and compared to the calculated value, Oxcalc = 100((nphosMOxy + W )/W ) − 1), where
0
0
nphos is the moles of phosphorus in the sample, MOxy is the atomic mass of oxygen, and W is the weight of the sample.
0
those of the thiol−ene systems upon changing cross-link
densities and monomer composition. To evaluate this
hypothesis, bisphosphine 1 was polymerized in a Teflon mold
under an inert atmosphere with either tetra(ethylene glycol)
diallyether (TEGDAE) to create soft polymer networks (SPN-
1
to SPN-3) or 1,3,5-triallyl-1,3,5-triazine-2,4,6(1H,3H,5H)-
trione (TTT) to create firm networks (FPN-1 to FPN-3)
Figure 1B, Table 1). Formulations consisting of a RPH /olefin
Figure 2. 31P SS-NMR of SPN-1 showing the presence of only three
(
2
phosphorus environments within the polymer network.
molar ratio of 0.5:1 should result in complete conversion of
both functional groups to form a network, while formulations
with excess RPH (0.75:1) or olefin (0.38:1) should form
polymer networks with supported P−H bonds or olefins,
respectively.
2
these sites, a unique feature not possible with thiol−ene
reactions. To our knowledge, this is the first example of a
network-supported primary/secondary phosphine.
Free-standing pucks were formed after exposure to UV light
on a conveyor assembly and were analyzed by attenuated total
reflectance Fourier transform infrared (ATR-FTIR) spectros-
copy, which revealed a decrease in both the P−H bond (∼2300
2
.2. Solid-Supported Chemistry. A defining feature of
thiol−ene networks that has been critical for their successful
implementation in materials science is the utilization of either
excess thiol or olefin functionality in the network to further
−1
−1
cm ) and olefin (∼3050 cm ) signal intensities (Supporting
Information Figures 1 and 2). For SPN-2 and FPN-2, complete
conversion of all P−H and olefin functional groups was
observed, while SPN-1/FPN-1 and SPN-3/FPN-3 samples
contained unreacted P−H bonds and olefins, respectively, after
polymerization. The physical and thermal properties of these
materials followed an identical trend to that of thiol−ene
systems, as all samples possessed a very high gel content, where
TEGDAE networks (SPN-1 to SPN-3) were the softest and
38
modify the material. While phosphane−ene networks should
be capable of analogous chemistry, the presence of tertiary alkyl
phosphines throughout the polymer backbone provides an
additional synthetic handle to exploit. These functional groups
possess significantly greater reactivity toward electrophiles and
oxidants relative to thioethers and also have potential utility in
chemical synthesis and catalysis. In most cases, functional
groups become unreactive or less reactive after polymerization;
however, the nucleophilic and basic character of phosphines
increases upon each successive alkylation (polymerization)
step, thus creating a reactive polymer network primed to
perform further chemistry throughout its backbone.
most swellable, possessing very low glass transitions (T ), while
g
those comprised of TTT (FPN-1 to FPN-3) were firmer and
less swellable because of higher cross-link densities and the
37
rigid olefin structure (Table 1). These data show that highly
tunable phosphane−ene networks could be made using a
similar approach used in thiol−ene network synthesis with ease,
providing a convenient route for the fabrication of network-
supported phosphines. The internal composition of the
network remained elusive, however, as IR spectroscopy alone
To evaluate this, a series of chemical reactions were
performed on SPN-1 to utilize R P, R PH, and RPH
2
3
2
functional groups (Figure 3A), and the outcomes of these
31
1
reactions were probed by P{ H} SS-NMR spectroscopy. This
chemistry was also performed in solution on the small molecule
1
(
1) as a model to compare 31P{ H} chemical shifts with those
cannot differentiate between primary (RPH ) or secondary
2
obtained in reactions of SPN-1 (Figure 3B). Compound 1
(blue) was reacted with 1-hexene until a spectroscopic profile
similar to that of SPN-1 (red) was obtained. Treatment with
ethyl iodide under mild reaction conditions selectively
quaternized the highly nucleophilic tertiary phosphines (at δP
= −33), providing a phosphonium salt/polyeletrolyte network
(δ = 37; SPN-1a). This significantly increased the T because
(
R PH) phosphine P−H bonds, which has implications for
2
33 31
further chemistry of the networks. Unlike S, P is an ideal
nucleus for NMR spectroscopy in solution and in the solid-
state, and upon examination of SPN-1 by P SS-NMR
spectroscopy, we observed three very well resolved, sharp
signals that corresponded to the primary, secondary, and
3
1
P
g
tertiary phosphines (Figure 2), with spin−spin coupling in the
of the resulting cation−anion interactions, while leaving all
primary and secondary phosphines untouched. The remaining
P−H bonds present at this stage were reacted with 1-hexene,
1
solid-state identical to primary phosphine 1 ( J
= 192 Hz),
P−H
even at temperatures well below the polymer T (−100 °C,
g
Supporting Information Figure 3). These data suggest that
these networks are homogeneous in composition with discrete
phosphorus environments and that the polymerization process
proceeds with no phosphorus-containing byproducts. The
capability to detect each phosphorus signal with such precision
should provide the means to monitor chemistry performed at
forming additional R P moieties (SPN-1b), which were then
3
quickly oxidized using elemental sulfur (δ = 50; SPN-1c),
P
increasing the T once more (Supporting Information Figure
g
S4). These results show the diversity of the R P functionality
3
and its orthogonal chemistry when compared to both primary/
secondary phosphines within the network. Furthermore, these
C
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 3. (A) Schematic representation of the solid-supported reactions illustrates three separate functionalization steps and their effect on glass
transition. (B) 31P{ H} NMR spectra comparing identical chemistry (shown in A) performed on the small molecule 1 (blue spectra) in solution and
SPN-1 (red spectra) in the solid-state.
1
chemical reactions could be monitored using 31P SS-NMR
spectroscopy to observe the distribution of phosphorus
environments with identical precision to that found in solution,
a feature not shared with thiol−ene systems. These two aspects
combined illustrate how phosphane−ene networks provide a
platform to exploit organophosphine chemistry in a novel
manner, allowing for the elucidation of more complex solid-
supported phosphine chemistry that would otherwise be
impossible to observe.
2
.3. Thermoresponsive Oxygen Scavenging. The high
phosphine loading and tunable nature of these polymer
networks suggests that they may be promising oxygen
scavenging materials for packaging of oxygen sensitive
3
9
substances. We examined the oxidation behavior of these
polymers gravimetrically by placing samples (SPN-1 to SPN-3)
in a thermogravimetric analysis (TGA) instrument under a flow
of air at 25 °C for 30 min followed by heating to 100 °C to
hasten the oxidation process (Figure 4A, green). Assuming a
Figure 4. (A) TGA thermographs for SPN-1 → SPN-3 (green lines
and axis) and FPN-1 → FPN-3 (red lines and axis). All polymers were
first held at 25 °C for 30 min to ascertain their stability at room
temperature, followed by heating to either 100 °C for soft networks or
1:1 ratio between phosphorus and oxygen, the calculated %
mass increase (Oxcalc) was compared to the experimentally
determined value, Oxexp (Table 1). For SPN-1 to SPN-3, these
values were in close agreement, demonstrating that the mass
increase depends on phosphorus content. Following the
31
1
1
20 °C for firm networks. (B) P{ H} NMR spectra of SPN-2 before
and after oxidation in air.
3
1
1
oxidation experiments, the P{ H} SS-NMR spectrum of
D
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
SPN-2 was collected and revealed a new resonance at δ = 51,
secondary phosphine oxides, supporting our postulation that
oxidation facilitates polymerization. Photopolymerization of
aerated samples containing 0.5 wt % Irgacure 819 and 0.67 wt
% diphenylamine was investigated; however, greater photo-
initiator content (1.15 wt %) was required to obtain complete
conversion (Supporting Information Figure S7). This demon-
strates that phosphane−ene systems have the potential to be
handled and exploited in air, although phosphines possessing
greater resistance to oxidation in addition to inhibitor
optimization should increase shelf life of these formulations.
P
consistent with the formation of R PO exclusively, with no
3
evidence for polymer backbone degradation (Figure 4B). Glass
transition temperatures of the oxidized SPN-1 to SPN-3
(Tg−ox) were measured and showed a significant increase
relative to their nonoxidized states, with this effect being more
pronounced with greater phosphine content (Supporting
Information Figure S5). The capability for the polymer to
oxidize appeared to depend on T , with firmer polymers
g
displaying greater oxidative resistance (SPN-2 > SPN-1 > SPN-
2
.5. Metal-Containing Phosphane−ene Networks.
3). As a comparison we analyzed polymers comprised of TTT
Given the strong Lewis basic character of the alkylphosphines
in the network, they are ideally suited to bind transition metals
and are particularly attractive for metals used in solid-supported
(
FPN-1 to FPN-3), which possess higher T values (Figure 4A,
g
red). Collectively, these networks were significantly more
stable, with FPN-2 and FPN-3 (T = 101 and 106 °C,
g
15
catalysis (e.g., Pd; PdCl or Pd(OAc) ). As a proof of
respectively) exhibiting minimal oxidation over 2 days at 120
2
2
principle, we have achieved palladium loadings of 1.6 mmol/g
of SPN-2, which corresponds to 65% of the number of the
available alkylphosphines binding to a metal center. To expand
the scope of phosphane−ene polymerization, compound 2, an
air-stable ferrocenyl-containing primary phosphine, was in-
corporated into a polymer network (Figure 6A). The redox
°
2
C, while FPN-1 (T = 56 °C) displayed excellent stability at
5 °C (0.0016%/min) but oxidized rapidly upon heating (120
g
°
C), resulting in an increase in its mass to the calculated value.
From these results it was concluded that oxidative stability
depends on the stiffness of the material, with oxidation
occurring rapidly when phosphines are present in a soft
environment and to a much lesser extent when in a firm
environment. Thus, by simply adjusting the composition and
cross-link density of these phosphane−ene polymers, the Tg
and therefore the oxidation temperature can be tuned. This
system is unique in a sense that the element responsible for the
polymerization reaction is also responsible for oxygen
scavenging, as opposed to other systems which require fillers,
catalysts, and/or UV irradiation, thus providing a simpler
40
means to prepare oxygen scavenging materials.
.4. Formulation Air Stability. Autopolymerization of
2
thiol−ene formulations in the absence of light has been
identified as a significant issue hindering potential commerci-
alization of the technology and is likely a result of impurities or
a charge-transfer complex between the thiol and olefin resulting
4
1,42
in S−C bond formation.
In this context, we examined the
3
1
stability of our phosphane−ene systems using P NMR
spectroscopy. While formulation of SPN-2 under a nitrogen
atmosphere was stable for over 30 days (δ = −139), polymer
Figure 6. (A) Synthesis of a ferrocene-containing polymer network
P
(
FcPN). (B) Shaped ceramics can be formed from the pyrolysis of
quickly formed after exposure to air (10 min), suggesting that
oxygen may have a role in the autopolymerization. Radical
inhibitors were then added as phosphine oxidation is believed
to occur via a radical process forming peroxy radicals, which
may in turn facilitate phosphane−ene chemistry prematurely,
FcPN. A thin strip of polymer film was manually formed in to a ring,
dried in an oven, and retained its shape upon pyrolysis. (C) Cyclic
voltammetry of a thin film of FcPN can undergo pseudo-reversible one-
electron oxidation, demonstrating the redox activity of these materials.
4
3,44
even at very low concentrations.
Indeed the addition of
activity and ceramic forming capabilities of ferrocenyl polymer
networks have been of significant interest in recent years, but
relatively few fabrication methods exist, prompting us to
explore our phosphane−ene chemistry as a potential synthetic
either butylated hydroxytoluene (Supporting Information
Figure S6) or diphenylamine (Figure 5) significantly increased
stability (1 and 5 h, respectively) after which secondary and
3
1
tertiary phosphines were detected, while P NMR spectrosco-
45
route. The intense color of 2 prevented photopolymerization
py revealed the presence of new signals (δ = 4 (t) and 31 (d),
P
using previous methods, so instead ferrocenyl-containing
polymer networks (FcPN) were prepared by mixing a 1.5:1
molar ratio of 2 and TTT in air followed by dissolution in
dimethylformamide (60 wt % solutions) and irradiation in a
glass mold (200 μm thick). It is noteworthy that this
formulation was observed to be air-stable in ambient light for
months without any signs of autopolymerization. This can
likely be attributed to the high oxidative stability of 2. FcPN
films swelled slightly in toluene, possessed high gel content
1
JP−H = 452 Hz), consistent with the formation of primary and
(
swell ratio = 0.30, ∼93%), and preserved the characteristic
orange color of 2. Upon pyrolysis of the sample (800 °C), a
black magnetic ceramic was isolated, which retained its original
shape (60 wt % char yield, Figure 6B) and contained carbon,
Figure 5. (A) 31P{ H} NMR spectra of formulation SPN-2 with
1
2
wt % diphenylamine in air, which imparts increased stability in air.
E
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
oxygen, iron, and phosphorus (Supporting Information Figures
S8 and S9). The electrochemical properties of a thin film of
FcPN on a glassy carbon electrode were examined using cyclic
voltammetry, and in each case, a chemically-reversible one-
electron oxidation event was observed, confirming that the
organometallic networks created using our phosphane−ene
chemistry possess redox activity that could be exploited for
future applications (Figure 6C).
under a nitrogen atmosphere for 10 min to remove residual solvent
and then cooled to 25 °C. At this point, data acquisition began and air
(100 mL/min) was introduced for 30 min, after which the sample was
heated at 2 °C/min until 120 °C and held for 50 h, and the rate of
oxidation for FPN-1 was determined from the first 30 min of air
exposure at 25 °C. The calculated % mass increase (Oxcalc) was
determined using the equation Oxcalc = 100((nphosMOxy + W )/W ) −
0
0
1
), where nphos is the moles of phosphorus in the sample, MOxy is the
atomic mass of oxygen, and W is the weight of the sample, and was
0
compared to the experimentally determined value, Oxexp. Glass
transition temperatures were determined using differential scanning
calorimetry (DSC) on a DSC Q20 TA Instruments and analyzed by
TA Universal Analysis. A sample of approximately 5 mg was placed in
an aluminum Tzero pan and underwent a heat/cool/heat profile at 10
°C/min under nitrogen atmosphere (50 mL/min). The transitions
were determined from the final heat cycle of the heat/cool/heat
profile. Cyclic voltammetry experiments were performed with a
Bioanalytical Systems Inc. (BASi) Epsilon potentiostat and analyzed
using BASi Epsilon software. SEM/EDX analysis was performed on a
Hitachi S4500 FESEM equipped with a Quartz XOne EDX system.
Synthesis of Tetraethyleneglyocol Diallylether (TEGDAE).
3
.0. CONCLUSIONS
In summary, we have demonstrated for the first time the
feasibility of synthesizing functional polymer networks using
phosphane−ene chemistry. The properties of the networks was
readily tuned by the choice of alkene monomer and the
pendant groups supported by the resulting network could be
selected based on the ratios of the monomers in the
3
1
formulation. The NMR-active P nucleus allowed for the
31
chemical structures of the networks to be directly probed by P
SS-NMR spectroscopy, a distinct advantage relative to the 32
S
TEGDAE was synthesized according to a modified literature
nucleus employed in the widely used thiol−ene network
formation. In addition, the chemical versatility of the resulting
phosphines in the networks was exploited to demonstrate the
potential utility of these materials (e.g., scavenge oxygen; bind
transition metals). The incorporation of a ferrocenyl phosphine
afforded a redox-active polymer network that could be
pyrolyzed to obtain a magnetic ceramic. Overall, this work
demonstrates the immense potential of phosphane−ene
chemistry for the production of functional materials.
46,47
procedure.
Briefly, tetraethylene glycol (5.0 g, 25.7 mmol) and
allyl bromide (10 g, 83 mmol) were dissolved in toluene (40 mL)
followed by the addition of sodium hydroxide (3.0 g, 75 mmol). The
mixture was heated to 45 °C and stirred for 18 h. The mixture was
decanted and volatiles removed in vacuo. The product was dissolved in
DCM and extracted with water (100 mL × 3), dried with MgSO , and
4
decanted. Volatiles were removed in vacuo followed by distillation
under reduced pressure (0.2 Torr). A colorless oil was isolated and
identified as tetraethyleneglyocol diallylether (5.0 g, 71%).
Synthesis of H P(CH ) S(CH ) S(CH ) PH (1). Diethyl (3-
2
2 3
2 3
2 3
2
bromopropyl)phosphonate and 1 were synthesized according to
literature procedures with minor changes for the synthesis of
4
.0. EXPERIMENTAL SECTION
Formulations were prepared under a N atmosphere and prepared in a
(OEt) (O)P(CH ) S(CH ) S(CH ) P(O)(OEt) (see Supporting
2 2 3 2 3 2 3 2
48
2
nitrogen-filled MBraun Labmaster 130 glovebox unless otherwise
noted (See Supporting Information for synthetic procedures). Dried
solvents were collected under vacuum in a flame-dried Strauss flask
and stored over 4 Å molecular sieves in the glovebox. Photoinitiator
Information Scheme 1). Briefly, 1,3-propanedithiol (5.90 g, 52.90
t
mmol) and KOBu (12.23 g, 105.8 mmol) were added to THF (350
mL) and heated to reflux for 30 min. Diethyl (3-bromopropyl)-
phosphonate (27.01 g, 111.0 mmol) was added dropwise and stirred
for 2 h. Volatiles were removed in vacuo at 40 °C, and the residual
material was dissolved in diethyl ether (500 mL). The mixture was
filtered through Celite and volatiles removed in vacuo, leaving a cloudy
oil. Filtration through 200 nm syringe filters yielded a colorless oil
(
1
Irgacure 819) was purchased from Ciba Chemicals and 1,3,5-triallyl-
,3,5-triazine-2,4,6(1H,3H,5H)-trione was purchased from Sigma-
Aldrich and used as received. Solution phase nuclear magnetic
resonance (NMR) spectroscopy was conducted on a Varian INOVA
1
13
31
1
4
00 MHz spectrometer ( H 400 MHz, C 158 MHz P{ H} 162
identified as (OEt)
Synthesis of H
modified literature procedure, vinylferrocene (2.00 g, 9.43 mmol)
and azobisisobutylnitrile (AIBN) (0.06 g, 0.4 mmol) were dissolved in
dry and degassed toluene (180 mL), added to a Parr reactor, and
2
(O)P(CH
2
)
3
S(CH
)Fe(C H
5 5
2
)
3
S(CH
2 3 2
) P(O)(OEt) .
1
MHz) unless otherwise noted. All H NMR spectra were referenced
P(CH (C
2
2
)
2
5
H
4
) (2). According to a
4
9
1
relative to tetramethylsilane (CDCl ; H δH = 7.26 ppm). The
3
chemical shifts for 31P{H} NMR spectroscopy were referenced using
an external standard (85% H PO ; δ = 0). Solid-state NMR
3
4
P
spectroscopic experiments were conducted on Varian INOVA 400
MHz at a spin rate of 9300 Hz and referenced using an external
standard ((NH ) PO ; δ = 0). High resolution mass spectrometry
purged with N gas. The vessel was then charged with PH gas (80
2
3
PSI) and heated to 45 °C for 16 h. CAUTIONphosphine gas is toxic
and pyrophoric. The excess phosphine gas was burned off under
controlled conditions in a fume hood before the Parr reactor was
opened inside a glovebox. The reaction mixture was transferred to a
Schlenk flask before the volatiles were removed in vacuo to afford a
light orange residue. The residue was dissolved in a minimum amount
of DCM and purified by column chromatography (hexanes) using
4
3
4
P
was performed on a MASPEC II system, and ESI ± mass spectrometry
was performed on a Micromass LCT spectrometer. Infrared spectra
were recorded using a Bruker Tensor 27 spectrometer with an
attenuated total reflectance (ATR) attachment using a ZnSe crystal.
Photopolymerization was performed using a modified UV curing
conveyor system purchased from UV Process and Supply Inc. with a
deactivated silica containing 4% deionized water (R ≈ 0.2) as the
f
2
2
2
mercury bulb (UVA: 154 mW/cm , 189 mJ/cm ; UVB: 74 mW/cm ,
mobile phase. The solution was concentrated in vacuo, and the
resulting yellow solid was recrystallized from ethanol, identified as 5
(1.62 g, 70%). Yield = 1.62 g, 70%.
2
9
0 mJ/cm ), except for the synthesis of FcPN which was photo-
2
polymerized using a static light source (UVA: 175 mW/cm ). Lamp
power was determined by a PP2-H-U Power Puck II purchased from
EIT Instrument Markets. All polymer samples were milled by hand
and then immersed in toluene for at least 24 h and dried to remove
unpolymerized material prior to thermal analysis. Oxidation experi-
ments were conducted using thermal gravimetric analysis (TGA) on a
Q600 SDT TA Instruments and analyzed using TA Universal Analysis.
For samples SPN-1 → SPN-3, a 5 mg milled sample was placed in an
alumina cup and exposed to medical grade air (100 mL/min) for 30
min at 25 °C, following by heating at 2 °C/min until 100 °C and held
for 210 min. Samples FPN-1 → FPN-3 were first heated to 120 °C
Photopolymerization. The appropriate molar ratios of phos-
phine, olefin, and photoinitiator (0.5 wt %) were combined and mixed
in vials within a nitrogen filled glovebox. Solutions (∼75 μL) were
pipetted in to a circular Teflon mold (1 mm deep, 1 cm in diameter)
and placed in a sealed vessel possessing a glass window. The entire
assembly was carefully removed from the glovebox and irradiated
2
2
2
2
(UVA: 154 mW/cm , 189 mJ/cm ; UVB 74 mW/cm , 90 mJ/cm for
each exposure, at a conveyor speed of 0.1 m/s) The polymer was
quickly removed from the mold and stored in the glovebox to
minimize oxidation. Polymer FcPN was prepared as a 60 wt % solution
F
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
in DMF with 5 wt % photoinitiator and irradiated (UVA: 175 mW/
cm , 6 h) through a glass mold (200 μm). For stabilized systems,
inhibitor (butylated hydroxytoluene or diphenylamine) was added to
the formulation within the glovebox and mixed until complete
dissolution. Samples were then exposed to an air atmosphere and
periodically stirred for 10 min prior to casting and photopolymeriza-
tion through a glass window.
followed by the addition of S (75 mg). The three possible products
from this reaction series were confirmed by ESI± mass spectrometry
(Supporting Information Figure S10).
8
2
ASSOCIATED CONTENT
Supporting Information
■
*
S
IR/NMR spectra, TGA, SEM, EDX, and DSC data. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Swelling Experiments and Gel Content. Samples were weighed
∼70 mg) and immersed in toluene for 24 h. Samples were then
weighed again, and a swelling ratio was determined using the equation
(
Q = (m − m )/m where m = swelled mass, m = dry mass, and Q =
t
0
0
t
0
swelling ratio. Samples were then dried in a vacuum oven at 80 °C
over 24 h and weighed again. By comparing the original mass to the
polymer mass after toluene extraction, gel content was obtained.
Cyclic Voltammetry. A 1.5:1 molar ratio of 2 and TTT in air were
dissolved in dimethylformamide (60 wt % solutions), of which a
portion (20 μL) was diluted in DCM (400 μL) and drop cast on a 3
mm glass carbon electrode. Following the evaporation of DCM the
AUTHOR INFORMATION
Corresponding Author
E-mail: pragogna@uwo.ca.
Funding
The Natural Sciences and Engineering Research Council of
Canada through Discovery Grant and Strategic Grant programs
(NSERC-DG; NSERC-SPG) and the Ontario Ministry of
Research and Innovation (OMRI).
■
*
2
wet film was irradiated (UVA: 175 mW/cm ) for 6 h to induce cross-
linking. Typical electrochemical cells consisted of a three-electrode
setup including a polymer network film-covered glassy carbon working
electrode, platinum wire counter electrode, and silver wire pseudo-
reference electrode. Experiments were run at variable scan rates in
degassed acetonitrile solutions containing 0.1 M tetrabutylammonium
hexafluorophosphate ([Bu N][PF ]) as supporting electrolyte. Cyclic
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
4
6
■
voltammograms were referenced against an internal standard (∼1 mM
ferrocene) after a small portion of the film had been removed from the
glassy carbon electrode surface. Internal cell resistance was corrected
using the BASi Epsilon software.
The authors would like to thank, NSERC, CFI, ORF, Cytec
Industries, and The University of Western Ontario for their
support.
Palladium Loading. Polymer SPN-2 (150 mg) and a stoichio-
REFERENCES
metric amount of PdCl or Pd(OAc) (0.37 mmol) relative to the
■
2
2
(
1) Zhao, Q.; Dunlop, J. W. C.; Qiu, X.; Huang, F.; Zhang, Z.; Heyda,
J.; Dzubiella, J.; Antonietti, M.; Yuan, J. Nat. Commun. 2014, 5, 4293.
2) Kloosterboer, J. G. Network formation by chain crosslinking
photopolymerization and its applications in electronics; Advances in
Polymer Science; Springer-Verlag: Berlin/Heidelberg, 1988; Vol. 84.
number of phosphines in the network were combined in benzene (15
mL) and stirred for 3 days, after which the polymer was centrifuged
and rinsed with benzene (3 × 15 mL) and pentane (3 × 15 mL) and
dried in vacuo.
(
Reaction Series on SPN-1 and Model Compound 1. 31P{ H}
NMR spectroscopy and DSC were used to monitor reaction progress.
Polymer SPN-1 was milled by hand, immersed in toluene for 24 h,
isolated, and dried to remove unpolymerized material. Samples (0.20
g) were swelled in a 1:1 (v/v) ratio of toluene and acetonitrile (16
mL) followed by the addition of ethyl iodide (0.40 mL). The mixture
was stirred for 1 h and then centrifuged. The polymer was then rinsed
with pentane (2 × 15 mL) and dried in vacuo. Hydrophosphination of
the primary and secondary phosphines proceeded by swelling the
quaternized polymer (0.20 g) in a 1:1:1 (v/v/v) ratio of toluene,
acetonitrile, and 1-hexene (15 mL) followed by the addition of
Irgacure 819 (17.5 mg). After 24 h, IR spectroscopic analysis revealed
that most of the available P−H bonds were consumed. An additional
1
̈
(
3) Carlborg, C. F.; Haraldsson, T.; Oberg, K.; Malkoch, M.; van der
Wijngaart, W. Lab Chip 2011, 11, 3136.
4) Leterrier, Y.; Singh, B.; Bouchet, J.; Man
Fayet, P. Surf. Coat. Technol. 2009, 203, 3398.
5) Zhu, J.; Marchant, R. E. Expert Rev. Med. Devices 2011, 8, 607.
6) Wang, J.; He, R.; Che, Q. J. Colloid Interface Sci. 2011, 361, 219.
7) Xu, J.; Feng, E.; Song, J. J. Am. Chem. Soc. 2014, 136, 4105.
(8) De, B.; Karak, N. J. Appl. Polym. Sci. 2014, 131, n/a.
(9) Revzin, a; Russell, R. J.; Yadavalli, V. K.; Koh, W. G.; Deister, C.;
Hile, D. D.; Mellott, M. B.; Pishko, M. V. Langmuir 2001, 17, 5440.
(10) Hoyle, C. E.; Lee, T. Y.; Roper, T. J. Polym. Sci., Part A: Polym.
Chem. 2004, 42, 5301.
(11) Bowman, C. N.; Kloxin, C. J. AIChE J. 2008, 54, 2775.
(12) Kade, M. J.; Burke, D. J.; Hawker, C. J. J. Polym. Sci., Part A:
Polym. Chem. 2010, 48, 743.
(
̊
son, J. E.; Rochat, G.;
(
(
(
1
7.5 mg of photoinitiator was then added and irradiated for 5 h to
complete the reaction. Polymer samples were centrifuged, rinsed with
toluene (4 × 15 mL) and pentane (2 × 15 mL), and then dried in
vacuo. Sulphurization of these polymer samples proceeded by swelling
(13) Mark, J. E.; Allcock, H. R.; Wester, R. Inorganic Polymers; 2nd
ed.; Oxford University Press: 2005; pp 3−50.
the polymer in a saturated solution of S (12 mL) and stirring for 24 h.
8
Polymer was centrifuged, rinsed with toluene (4 × 15 mL) and
pentane (2 × 15 mL), and dried in vacuo.
́
(14) El-Roz, M.; Lalevee, J.; Allonas, X.; Fouassier, J.-P. Macromol.
Rapid Commun. 2008, 29, 804.
1
was used as a model to compare the chemistry performed on
(15) Bai, L.; Zhang, Y.; Wang, J.-X. QSAR Comb. Sci. 2004, 23, 875.
(16) Welch, G. C.; San Juan, R. R.; Masuda, J. D.; Stephan, D. W.
Science 2006, 314, 1124.
(17) Hemp, S. T.; Zhang, M.; Tamami, M.; Long, T. E. Polym. Chem.
2013, 4, 3582.
(18) Wu, B.; Wang, Y.-Z.; Wang, X.-L.; Yang, K.-K.; Jin, Y.-D.; Zhao,
H. Polym. Degrad. Stab. 2002, 76, 401.
(19) Tsang, C.-W.; Yam, M.; Gates, D. P. J. Am. Chem. Soc. 2003,
125, 1480.
(20) Bates, J. I.; Dugal-Tessier, J.; Gates, D. P. Dalton Trans. 2010,
39, 3151.
(21) Allcock, H. R.; Kugel, R. L. J. Am. Chem. Soc. 1965, 87, 4216.
(22) Naka, K.; Umeyama, T.; Nakahashi, A.; Chujo, Y. Macro-
molecules 2007, 40, 4854.
31
1
SPN-1. P{ H} NMR spectroscopy was used to monitor the reaction
progress. No purification steps were performed aside from removal of
volatiles in vacuo. A solution of 1 containing toluene, 1-hexene, and
Irgacure 819 was irradiated until a comparable distribution of primary,
secondary, and tertiary phosphines to that of SPN-1 were obtained (30
s). Quaternization of the tertiary phosphines proceeded by dissolving
the product in a 1:1 (v/v) mixture of toluene and acetonitrile (0.8
mL), followed by the addition of 6.5 equiv of ethyl iodide (200 μL, 2.5
mmol) and left for 1 h. Hydrophosphination of the remaining primary
and secondary phosphines proceeded by dissolving the quaternized
product and photoinitiator (2 mg) in a 1:1 (v/v) mixture of toluene,
acetonitrile, and 1-hexene (1.5 mL). The solution was irradiated for 24
h to ensure complete P−H bond conversion. Sulphurization of the
mixture proceeded in a 1:1 (v/v) toluene and acetonitrile (1 mL) ratio
(23) Hodgson, J. L.; Coote, M. L. Macromolecules 2005, 38, 8902.
G
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(
(
(
24) Pellon, J. J. Polym. Sci. 1960, 43, 537.
25) Pellon, J. J. Am. Chem. Soc. 1961, 83, 1915.
26) Stiles, A. R.; Rust, F. F.; Vaughan, W. E. J. Am. Chem. Soc. 1952,
7
4, 3282.
27) Alvey, L. J.; Rutherford, D.; Juliette, J. J. J.; Gladysz, J. a. J. Org.
Chem. 1998, 63, 6302.
28) Obata, T.; Kobayashi, E.; Aoshima, S.; Furukawa, J. J. Polym. Sci.,
Part A: Polym. Chem. 1994, 32, 475.
(
(
(
(
29) Greenberg, S.; Stephan, D. W. Inorg. Chem. 2009, 48, 8623.
30) Stewart, B.; Harriman, A.; Higham, L. J. Organometallics 2011,
3
0, 5338.
(
31) Higham, L. J. Phosphorus Compounds; Peruzzini, M., Gonsalvi,
L., Eds.; Catalysis by Metal Complexes; Springer: Netherlands,
Dordrecht, 2011; Vol. 37, pp 1−19.
(
32) Gali, H.; Karra, S. R.; Reddy, V. S.; Katti, K. V. Angew. Chem.,
Int. Ed. 1999, 38, 2020.
33) Prabhu, K. R.; Pillarsetty, N.; Gali, H.; Katti, K. V. J. Am. Chem.
Soc. 2000, 122, 1554.
34) Smith, C. J.; Reddy, V. S.; Karra, S. R.; Katti, K. V.; Barbour, L. J.
Inorg. Chem. 1997, 36, 1786.
(
(
(
35) Greenberg, S.; Stephan, D. W. Inorg. Chem. 2009, 48, 8623.
(
36) Cramer, N. B.; Couch, C. L.; Schreck, K. M.; Carioscia, J. A.;
Boulden, J. E.; Stansbury, J. W.; Bowman, C. N. Dent. Mater. 2010, 26,
1.
37) Mongkhontreerat, S.; Oberg, K.; Erixon, L.; Lo
Hult, A.; Malkoch, M. J. Mater. Chem. A 2013, 1, 13732.
38) Gupta, N.; Lin, B. F.; Campos, L. M.; Dimitriou, M. D.; Hikita,
2
(
̈
̈
wenhielm, P.;
(
S. T.; Treat, N. D.; Tirrell, M. V.; Clegg, D. O.; Kramer, E. J.; Hawker,
C. J. Nat. Chem. 2010, 2, 138.
(
39) Ferrari, M. C.; Carranza, S.; Bonnecaze, R. T.; Tung, K. K.;
Freeman, B. D.; Paul, D. R. J. Membr. Sci. 2009, 329, 183.
40) Tung, K. K.; Bonnecaze, R. T.; Freeman, B. D.; Paul, D. R.
Polymer (Guildf). 2012, 53, 4211.
41) Cook, W. D.; Chen, F.; Pattison, D. W.; Hopson, P.; Beaujon,
M. Polym. Int. 2007, 56, 1572.
42) Kuhne, G.; Diesen, J. S.; Klemm, E. Angew. Makromol. Chem.
996, 242, 139.
(
(
(
̈
1
(
43) Floyd, M. B.; Boozer, C. E. J. Am. Chem. Soc. 1963, 85, 984.
(
44) Buckler, S. A. J. Am. Chem. Soc. 1962, 84, 3093.
(
45) Hempenius, M. A.; Cirmi, C.; Savio, F. L.; Song, J.; Vancso, G. J.
Macromol. Rapid Commun. 2010, 31, 772.
46) Kurian, P.; Zschoche, S.; Kennedy, J. P. J. Polym. Sci., Part A:
Polym. Chem. 2000, 38, 3200.
(
(
47) Zona, C.; D’Orazio, G.; La Ferla, B. Synlett 2013, 24, 709.
(
48) Smith, C. J.; Katti, K. V.; Volkert, W. A.; Barbour, L. J. Inorg.
Chem. 1997, 36, 3928.
49) Rabiee Kenaree, A.; Berven, B. M.; Ragogna, P. J.; Gilroy, J. B.
Chem. Commun. 2014, 50, 10714.
(
H
DOI: 10.1021/cm504784e
Chem. Mater. XXXX, XXX, XXX−XXX