An Addition-Isomerization Mechanism for the Anionic Polymerization of MesP=CPh2 and m -XylP=CPh2

Article abstract of DOI:10.1021/acs.macromol.8b00100

We report that the anionic polymerization of P-mesityl and m-xylyl-substituted phosphaalkenes follows an unusual addition-isomerization mechanism. Specifically, the polymerization of ArP=CPh₂ [Ar = Mes (1a), m-Xyl (1b)] involves the hindered nucleophilic anion intermediate, a -P(Ar)-CPh₂^-, which undergoes a proton migration from the o-CH₃ of the Mes/m-Xyl moiety to the -CPh₂ moiety to afford a propagating benzylic anion. This mechanism is supported by the preparation of model compounds MeP(CHPh₂)-4,6-Me₂C₆H₂-2-CH₂-CPh₃ (2a) or MeP(CHPh₂)-6-MeC₆H₃-2-CH₂-CPh₃ (2b), which were both crystallographically characterized. Polymerization of 1a or 1b in THF solution using n-BuLi (2 mol %) revealed ¹H and ¹³C NMR signals assigned to -CH₂- and -CHPh₂ groups consistent with an addition-isomerization polymerization mechanism to afford poly(methylenephosphine) 3a or 3b. A large kinetic isotope effect (≤23) was determined for the n-BuLi-initiated polymerization of 1a-d₉ compared to 1a in THF at 50 °C, consistent with C-H (or C-D) activation as the rate-determining step. This C-H activation step was modeled using DFT computations which revealed that the intramolecular proton transfer from the o-CH₃ of the Mes moiety to the -CPh₂ moiety has an activation energy (E_a = +18.5 kcal mol^-1). For comparison, this computational value was quite close to the experimentally measured activation energy of propagation ArP=CPh₂ in THF [E_a = 14.0 ± 0.9 kcal mol^-1 (1a), 15.6 ± 2.8 kcal mol^-1 (1b)].

Full text of DOI:10.1021/acs.macromol.8b00100

Article
Cite This: Macromolecules XXXX, XXX, XXX−XXX
An Addition−Isomerization Mechanism for the Anionic
Polymerization of MesPCPh2 and m‑XylPCPh2
Benjamin W. Rawe, Andrew M. Priegert, Shuai Wang, Carl Schiller, Sonja Gerke, and Derek P. Gates*
Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, British Columbia, Canada V6T 1Z1
S
* Supporting Information
ABSTRACT: We report that the anionic polymerization of P-
mesityl and m-xylyl-substituted phosphaalkenes follows an
unusual addition−isomerization mechanism. Specifically, the
polymerization of ArPCPh2 [Ar = Mes (1a), m-Xyl (1b)]
involves the hindered nucleophilic anion intermediate, Ⓟ−
−
P(Ar)−CPh2 , which undergoes a proton migration from the o-
CH3 of the Mes/m-Xyl moiety to the −CPh2 moiety to afford a
propagating benzylic anion. This mechanism is supported by the
preparation of model compounds MeP(CHPh2)-4,6-Me2C6H2−
2-CH2−CPh3 (2a) or MeP(CHPh2)-6-MeC6H3−2-CH2−CPh3
(2b), which were both crystallographically characterized. Polymerization of 1a or 1b in THF solution using n-BuLi (2 mol %)
1
revealed H and 13C NMR signals assigned to −CH2− and −CHPh2 groups consistent with an addition−isomerization
polymerization mechanism to afford poly(methylenephosphine) 3a or 3b. A large kinetic isotope effect (≤23) was determined
for the n-BuLi-initiated polymerization of 1a-d9 compared to 1a in THF at 50 °C, consistent with C−H (or C−D) activation as
the rate-determining step. This C−H activation step was modeled using DFT computations which revealed that the
intramolecular proton transfer from the o-CH3 of the Mes moiety to the −CPh2 moiety has an activation energy (Ea = +18.5 kcal
mol−1). For comparison, this computational value was quite close to the experimentally measured activation energy of
propagation ArPCPh2 in THF [Ea = 14.0 0.9 kcal mol−1 (1a), 15.6 2.8 kcal mol−1 (1b)].
highly regioregular polypeptides, [−CH2CH2C(O)N(R)−]n,
rather than the head-to-tail poly(vinylacrylamide), [CH2CH-
{C(O)NHR}]n.7−13
INTRODUCTION
■
The addition polymerization of the CC bond of an olefin is
perhaps the most important and general method for the
synthesis of macromolecules. Although propagation typically
occurs as a series of head-to-tail addition reactions, side
reactions such as termination, chain transfer, and branching can
greatly affect the resultant polymer architecture and properties.
Consider the polymerization of the simplest olefin, ethylene,
which can undergo irregular isomerization processes during
polymerization as a result of C−H bond activation. Thus,
depending on the initiator type and polymerization conditions,
linear, branched, hyperbranched, and cyclic microstructures are
possible for polyethylene (Scheme 1).1 Similar irregular H-
transfers resulting in complex microstructures can occur for the
anionic polymerizations of some vinyl organosilanes,2−4 α-
(aminomethyl)acrylates,5 and acrylonitriles.6 In contrast,
regioregular isomerizations during propagation are exceedingly
rare. Perhaps the most well-established is the anionic
polymerization of acrylamides, H2CCHC(O)NHR, to afford
The field of functional polymers featuring inorganic elements
in the main chain has grown dramatically due to their unique
properties and potential applications.14−18 As part of a broad
program to synthesize phosphorus-containing macromole-
cules,19 we have been interested in extending addition
polymerization to PC bonds. In particular, the close analogy
between phosphaalkenes and olefins in molecular chemistry20
has inspired us to develop the addition polymerization of the
monomer MesPCPh2 (A, R = Ph) and related phosphaal-
kenes.21−28 On the basis of NMR spectroscopic investigations,
the resultant poly(methylenephosphine)s were originally
assigned a microstructure that was consistent with a simple
head-to-tail propagation analogous to that observed for
polyolefins (i.e., B, x = 0 in Scheme 2). Recently, we discovered
that the nitroxide-mediated radical polymerization of A (R =
Ph) gives a polymer with a fascinating and unexpected
microstructure that is presumably derived from regioregular
C−H activation processes (i.e., B, x ≫ y).25 Follow-up studies
revealed that this novel addition−isomerization mechanism of
propagation for phosphaalkenes might be more general.21−24
Scheme 1. Polymerization of Ethylene To Give Polymers of
Many Different Structures through Side Reactions
Received: January 16, 2018
Revised: March 13, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 2. Addition−Isomerization Polymerization of
Phosphaalkene A (Mes =2,4,6-Trimethylphenyl) Affords
Poly(methylenephosphine) B with a C−H Activated
Microstructure
Herein, we report our detailed studies of the highly
regioregular addition−isomerization mechanism of anion-
initiated polymerization for the phosphaalkenes MesP=CPh2
(1a) and m-XylP=CPh2 (1b). Specifically, we describe
experimental and computational investigations of model
compounds and polymers derived from the aforementioned
monomers to show that the key step involves the unexpected
proton transfer from the o-CH3 of the P-aryl moiety to the
−CPh2 anion.
Figure 1. 31P NMR (162 MHz, toluene, 298 K) spectra of the reaction
mixture during the conversion of 3a from 1a; (a) after addition of
MeLi to 1a; (b) after addition of Trt-Cl; (c) after stirring at room
temperature for 72 h.
RESULTS AND DISCUSSION
■
Model Compounds. We have previously reported the
synthesis of several model compounds, prepared by the
addition of electrophiles to carbanion MesP(Me)−C(Li)Ph2
(made from the addition of MeLi + 1a).29 In each case, the
electrophile (E+) added selectively at the −CPh2 carbanion to
afford simple addition products, MesP(Me)−CPh2E [E = H,
Me, P(NEt2)2, SiMe3, and SiMe2H]. Although these results
provided support for the proposed head-to-tail addition
polymerization of MesP=CPh2 (1a), the electrophiles that
were used were relatively small compared to monomer 1a. In
light of our recent findings, we hypothesized that treating
ArP(Me)−C(Li)Ph2 (R = m-Xyl or Mes) with a far bulkier
electrophile (e.g., E = CPh3) may disfavor simple addition
chemistry and mimic a C−H activation−addition pathway that
has been observed for the polymerization of 1a (Scheme 3).
In accordance with our previously published procedure,29
MesP(Me)−C(Li)Ph2 was prepared in situ by treating
monomer 1a (1 equiv) with MeLi (1 equiv). The progress of
reaction was monitored by 31P{1H} NMR spectroscopy, which
revealed a major signal assigned to MesP(Me)−C(Li)Ph2 (δ =
−47, Figure 1a). Subsequently, the reaction mixture was cooled
to −78 °C, and a toluene solution of Trt-Cl (Trt =
triphenylmethyl, 1.02 equiv) was added. The initially red
solution immediately turned orange and was accompanied by a
white precipitate, presumably LiCl, as the reaction mixture was
warmed to room temperature. The 31P NMR spectrum of an
aliquot removed from the reaction mixture revealed that the
signal assigned to MesP(Me)−C(Li)Ph2 had been consumed
and was replaced by two new signals at −18 and −38 ppm (ca.
1:3 ratio; Figure 1b). Neither of these signals was consistent
with the protonated species Mes(Me)P−CHPh2 (δ = −24 in
CD2Cl2).29 Remarkably, upon standing for 3 days, these
intermediate signals were slowly consumed and quantitatively
replaced by a singlet resonance at −26 ppm (Figure 1c).
Virtually identical observations were made when monitoring
the reaction starting with m-xylyl-monomer 1b. In this case, the
final product was observed at −24 ppm.
The crude products for both reactions were isolated and
1
analyzed by H NMR spectroscopy. Importantly, a doublet
resonance was detected that was assigned to a ArP−CHPh2
proton [Ar = Mes: δ = 4.73, 2JHP = 4 Hz, 1H; Ar = m-Xyl: δ =
4.67, 2JHP = 4 Hz, 1H]. A similar resonance was observed in the
previously prepared model compound Mes(Me)P−CHPh2 [δ
a
Scheme 3. Synthesis of 2a,b from Phosphaalkenes 1a,b
= 4.87, JHP = 5 Hz] which aided this assignment.29
2
Importantly, this suggests that the trityl moiety was not
attached at the −CPh2 position. Two strongly coupled signals,
integrating for 1H each, were also observed in each spectrum.
These signals were indicative of diastereotopic Ar−CHaHb−R
2
protons [Ar = Mes: δ = 5.11 and 3.72, JHH = 15 Hz; Ar = m-
2
Xyl: 5.01 and 3.66, JHH = 15 Hz)]. In each case, the more
downfield resonance of the −CHaHb− moiety was further split
by coupling to phosphorus [4JHP = 10 Hz, in both cases] as
confirmed by recording the 1H{31P} NMR spectrum. The
appearance of resonances corresponding to CHPh2 and Ar−
CH2−R protons was consistent with the proposed structures 2a
and 2b.
This hypothesis was confirmed by an X-ray crystallographic
analysis of compounds of 2a and 2b (Figure 2). Remarkably,
each molecular structure revealed the presence of a −CH2−
a
Both 2a and 2b feature o-CH2−Trt linkages that are derived from the
C−H activation of an o-Me proton from the P−aryl moiety. The
simple addition product was not isolated in each case.
B
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
spectroscopy on a relatively low field spectrometer (300 MHz
27,28
1
for H NMR).
The present work was performed at higher
1
field (600 MHz for H NMR) and incorporated the results of
two-dimensional NMR experiments. Polymers 3a and 3b were
synthesized by the addition of n-BuLi (2.0 mol %, 1.6 M in
hexanes) to a THF solution of the corresponding phosphaal-
kene (Scheme 4). The progress of each reaction was monitored
Scheme 4. Synthesis of Polymers 3a,b from Phosphaalkenes
1a,b
by 31P NMR spectroscopy. In each case, the sharp signal
assigned to monomer (1a: δ = 234; 1b: δ = 231) was replaced
by a single broad signal at higher field, consistent with the
desired polymer (3a: δ = −10; 3b: δ = −11). Each polymer was
isolated as a pale-yellow powder after repeated precipitations
from concentrated THF solutions into hexanes.
The molecular weights of polymers 3a and 3b were
determined by gel permeation chromatography (GPC)
equipped with a multiangle light scattering detector (Table
1). Analogous to previous reports of the living anionic
Figure 2. Molecular structures of 2a (top) and 2b (bottom). Thermal
ellipsoids are shown at the 50% probability level; all hydrogen atoms
except H29 are omitted for clarity.
CPh3 linkage resulting from activation of the CH3 moiety of the
Mes and m-Xyl substituents. Although the two products
crystallized in different space groups (2a: P-1; 2b: P21/n), the
metrical parameters are in close agreement with one another.
The new C−C bond (C8−C9) lengths are slightly longer [2a:
1.576(6) Å; 2b: 1.587(2) Å] than typical carbon−carbon single
bonds (∼1.54 Å),30 indicative of the strain induced by the
bulky trityl group. Moreover, the C6−C8−C9 bond angle [2a:
117.2(3)°; 2b: 118.2(1)°] is significantly more obtuse than the
predicted sp3 geometry (109.5°), likely a consequence of steric
repulsion.
Table 1. Summary of Molecular Weight Data Obtained for
the Anionic Polymerization of Phosphaalkenes 1a, 1b, and
1a-d9
b
Mn calcd
Mp measd/Mn
a
c
entry monomer [M]:[I]
[g mol−1
]
measd [g mol−1
]
Đ
1
2
3
1a
50:1
50:1
50:1
15900
15200
16400
16700/13800
12000/8800
9000/10000
1.10
1.17
1.08
1b
1a-d9
a
b
[ArP=CPh2]:[n-BuLi]. Calculated using the monomer-to-initiator
Although its structure was not observed experimentally, we
were interested in using density functional theory (DFT) to
investigate whether the simple addition product from this
reaction, MesP(Me)−CPh2Trt, was plausible. A minimized
geometry was obtained; however the large substituents around
the −CPh2Trt carbon meant that the calculated −CPh2−Trt
bond length was found to be 1.707 Å, comparable to the range
of calculated bond lengths for the (currently) experimentally
elusive compound (Ph3C)2 (1.702−1.791 Å) that has also been
computed by DFT previously.31 Strikingly, the energy of the
DFT-minimized structure of MesP(Me)−CPh2Trt was 46.0
kcal mol−1 higher in energy than the computed energy of o-
CH3-activated 2a. This energy difference is rather substantial
c
ratio. Absolute molecular weights were determined using triple
detection GPC. The dn/dc values for 3a (0.239) and 3b (0.245) were
determined by direct measurement using a differential refractometer
(Figures S13 and S14).
polymerization of 1a,26,27 a narrow molecular weight
distribution was observed, and the calculated and measured
values of Mn for 3a were in close agreement (Mn calcd = 15 900
g mol−1; Mn measd = 13 800 g mol−1; Đ = 1.10). In contrast,
the measured Mn value obtained for 3b was somewhat lower
than expected from the [M]:[I] ratio (Mn calcd = 15 200 g
mol−1; Mn measd = 8800 g mol−1). The peak molecular weight
value (Mp) measured for polymer 3b was considerably closer to
that predicted for given the [M]:[I] ratio of 50:1 used for the
polymerization (Mp measd = 12 000 g mol−1). In addition, the
dispersity (Đ = 1.17) was slightly higher than are generally
accepted for a living polymerization, and tailing at the low MW
end of the GPC chromatogram was observed. These
observations suggest that some quenching of the propagating
anion, perhaps due to trace impurities in the monomer or
solvent, may have occurred during the slow polymerization of
1b. Nonetheless, the polymer sample obtained was suitable for
NMR spectroscopic analysis.
3
considering that the dissociation energy of Csp3−Csp bonds
range from 77.8 to 90.2 kcal mol−1.32 Thus, we conclude that
extreme steric hindrance of MesP(Me)−CPh2Trt prevents its
formation in favor of the observed TrtCH2(Ar)P(Me)−CPh2H
(2a). The following sections will probe whether an analogous
mechanism is observed during the anionic polymerization of 1a
and 1b.
Polymer Microstructure. Given the aforementioned
results, we embarked on a detailed spectroscopic examination
of the microstructure of polymers 3a and 3b, respectively. We
1
had previously investigated 3a by 31P, H, and 13C{1H} NMR
C
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Recording several 1D and 2D NMR spectra of each polymer
in CDCl3 solution helped to elucidate the microstructures of
polymers 3a,b. For example, the 1H−13C HSQC NMR
spectrum of 3a is shown in Figure 3. Importantly, cross-
than predicted from the [M]:[I] ratio (Mn calcd = 16 400 g
mol−1; Mn measd = 9000 g mol−1; Đ = 1.08).
As expected, the 31P{1H} and 13C{1H} NMR spectra of 3a-d9
closely resembled those obtained for 3a. The 1H NMR
spectrum of 3a-d9 lacked the prominent resonances attributable
1
to aliphatic protons (Figure 4c) which are prevalent in the H
1
1
Figure 3. H−13C HSQC NMR spectrum (600 MHz for H, CDCl3,
298 K) of polymer 3a. The ordinate axis shows the 13C APT NMR
spectrum, and the abscissa axis shows the H NMR spectrum.
1
correlations were observed that permitted the assignment of
signals to aromatic C−H groups and Ar−CH3 groups. In
addition, cross-correlations were observed that could only be
assigned to the presence of −CHPh2 [ δH = 4.8 (3a), 5.0 (3b);
δC = 52 (3a), 52 (3b)] and −CH2− moieties [ δH = 3.6 (3a),
3.5 (3b); δC = 32 (3a), 32 (3b)]. The 13C APT NMR spectrum
was also consistent with these assignments. It should be noted
that the appearance of the NMR spectra recorded for 3a and 3b
closely matched the spectra of polymers previously obtained by
the radical initiated polymerization of 1a and 1b.25
To further interrogate the structure of these polymers, we
sought to prepare a polymer from phosphaalkene 1a-d9. This
monomer had an identical structure to 1a but featured ortho-
and para-CD3 substituents on the mesityl ring (rather than CH3
in 1a) and was synthesized from mesitylene-d9.33
Figure 4. (a) 1H NMR spectrum (600 MHz, CDCl3, 298 K) of
1
polymer 3a. (b) H NMR spectrum (600 MHz, CDCl3, 298 K) of
1
polymer 3b. (c) H NMR spectrum (600 MHz, CDCl3, 298 K) of
polymer 3a-d9. (d) 2H{1H} NMR spectrum (92 MHz, CHCl3, 298 K)
of polymer 3a-d9. ∗ = CDCl3
Our initial attempts to polymerize 1a-d9 at room temperature
using 2 mol % of n-BuLi as an anionic initiator (1.6 M in
hexanes, reaction solvent = glyme) were unsuccessful. Upon
addition of the anionic initiator the reaction mixture turned
dark red immediately, and this color persisted for several hours.
When an aliquot removed from the reaction mixture was
analyzed by 31P{1H} NMR spectroscopy after 3.5 days, no
broad resonances indicative of polymer formation were
observed. Nevertheless, the dark red color of the reaction
suggested that initiation had been successful to form the
anionic species, [BuMesP−CPh2]−. In an effort to increase the
rate of propagation, the polymerization was repeated, under
otherwise identical conditions, at 50 °C, and the reaction
progress was monitored by 31P{1H} NMR spectroscopy. After
7 days, the spectrum suggested that 90% of 1a-d9 had been
consumed to afford 3a-d9. The reaction was subsequently
terminated with methanol (3 drops), and the polymer was
isolated as an off-white powder after precipitation from
hexanes. The isolated 3a-d9 (yield = 10%) was analyzed by
GPC to reveal a narrow dispersity but a lower molecular weight
NMR spectrum for protonated polymers 3a,b (Figures 4a and
4b, respectively). Instead, only resonances corresponding to the
2
aromatic protons are observed for 3a-d9. A H{1H} NMR
spectrum (Figure 4d) of 3a-d9 was recorded which showed a
prominent resonance centered at 2.1 ppm which was assigned
to CD3 deuterium atoms. A shoulder, centered at 2.9 ppm, was
assigned to Ar−CD2−P moieties. In addition, very broad
resonance was observed centered at ca. 4.4 ppm that was
assigned to CDPh2 deuterons within the polymer. The
integrated ratio of the resonances at 2.1 and 2.9 ppm compared
to that at 4.4 ppm was ca. 8:1. Importantly, no signals for
aromatic deuterons were observed.
Overall, these results suggest that a C−D activation step
occurred in the polymerization reaction and that 3a-d9 is best
represented by the microstructure shown in Scheme 5 (where x
≫ y). Polymers 3a, 3b, and 3a-d9 likely result from a
regioregular addition−isomerization mechanism involving a C−
H activation of ortho-methyl protons of the mesitylene group.
The following section will investigate the energetics of such a
transformation.
D
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 5. Anionic Polymerization of the Deuterated
Phosphaalkene 1a-d9 To Yield Polymer 3a-d9
Activation Energies (Kinetics and DFT). The polymer-
ization of 1a and 1b could be performed at room temperature,
but 1a-d9 would only polymerize at reasonable rates at ca. 50
°C. Thus, it was hypothesized that C−H/D activation was the
rate-limiting step in the anionic polymerization of 1a/b. A first-
order rate plot showing the data collected for the anionic
polymerization of 1a, 1b, and 1a-d9 at ca. 50 °C is shown in
Figure 5. Clearly, the rate of propagation is considerably slower
Figure 6. Results of DFT computations, showing computed
mechanisms to dimeric structures III and IV. Energies are electronic
a
energies/free energies in kcal mol−1. Only electronic energies are
provided due to computational expense of frequency calculation on a
large system.
kcal mol−1) which suggests that microstructure y of polymer 3
is less probable than microstructure x.
Next, a pathway was considered to account for the formation
of microstructure x of polymer 3 (Scheme 4). Benzyllic anion
II, the postulated intermediate, is only slightly higher in energy
than −CPh2 anion I (ΔE° = +9.5 kcal mol−1, ΔG° = +12.1 kcal
mol−1) but is significantly less sterically hindered. In contrast to
the formation of dimer III from I and 1a, the formation of
dimer IV from II and 1a is thermodynamically favorable (ΔE°
= −25.3 kcal mol−1). Moreover, there is a strong thermody-
namic preference for product IV compared to head-to-tail
product III (ΔE° = −20.2 kcal mol−1), and the overall
formation of IV from I and 1a is favorable (ΔE° = −15.8 kcal
mol−1). On the basis of the aforementioned experimental data
and this computational data, we conclude that the addition−
isomerization mechanism of propagation for the anionic
polymerization of 1a is favored over a simple addition
mechanism on both energetic and steric grounds. In other
words, the −CPh2 group is simply too sterically encumbered to
permit significant head-to-tail enchainment (i.e., I + 1a → III).
With the energy of the most plausible intermediates
determined, it remained to computationally examine the
transition states in the two-step transformation of I to IV.
The deuterium labeling study described above suggests that the
intramolecular proton transfer step (I → II) would be rate-
determining with a transition state (TS‡I−II). This is consistent
with the fact that n-BuLi adds rapidly to 1a to afford I (i.e., low
activation energy), and the same would be expected for the
addition of the benzylic anion II to 1a. The activation barrier
was found to be moderate (TS‡I−II: ΔE° = +21.2 kcal mol−1,
ΔG° = +17.3 kcal mol−1, and Ea = +18.5 kcal mol−1). This
activation barrier is similar to that determined experimentally
for 1a (Ea = +14.0 0.9 kcal mol−1).26
Figure 5. Plot showing the linear portion of ln [M]0/[M] vs time plots
for the anionic polymerization of 1a,b and 1a-d9 with 2% n-BuLi at
∼50 °C in glyme. Rates of propagation kp (L mol−1 h−1) = 153 20
(1a), 152 23 (1b), 6.6 1.2 (1a-d9).
for the deuterated monomer relative to the protonated
monomers. These measurements suggest a kinetic isotope
effect, kH/kD, of ≤23, consistent with the postulated addition−
isomerization mechanism to form 3a (microstructure x). Given
the extremely slow rate of propagation for 1a-d9, the rate
constant is likely underestimated given that trace impurities will
have a great effect on the measured kp. Therefore, the kinetic
isotope effect is likely overestimated. Nonetheless, the data
does show that C−H/C−D bond cleavage is likely a major
factor in the rate-determining step of polymerization.
Additional insight into the reaction mechanism was gained
by developing a model for the first propagation step in the
polymerization of 1a using DFT (Figure 6). The geometries of
all species were optimized using the B3LYP functional and 6-
31+G(d) basis set. Single point energy calculations on the
optimized structures were performed using the B97D3
functional and 6-311++G(d,p) basis set, in addition to a
solvation correction (in THF, ε = 7.52). Further computational
details and coordinates of minimized geometries can be found
in the Supporting Information. Two reaction pathways were
considered. The simplest was the head-to-tail addition of
−CPh2 anion I to monomer 1a to afford dimer III. This
mechanism was thermodynamically unfavorable (ΔE° = +4.4
For comparison, kinetic experiments on the polymerization
of 1b with n-BuLi (2 mol %) in glyme were performed at five
different temperatures. From the Arrhenius plot (Figure S15)
the activation energy was determined to be +15.6 2.8 kcal
mol−1, which is similar to the barrier obtained for the
E
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
103 Å (1000−75 000 g mol−1), a Wyatt Optilab T-rEx (refractive index
detector, λ = 658 nm, 40 °C), a Wyatt miniDAWN (laser light
scattering detector, λ = 690 nm), and a Wyatt ViscoStar viscometer. A
flow rate of 0.5 mL min−1 was used, and samples were dissolved in
THF (ca. 1.5 mg mL−1). The dn/dc values for 3a (0.239) and 3b
(0.254) were determined by direct measurement using a Wyatt
Optilab T-rEx refractive index detector (λ = 658 nm). Hexanes and
toluene were deoxygenated with nitrogen and dried by passing
through a column containing activated basic alumina. THF and glyme
were freshly distilled from sodium/benzophenone ketyl before use.
Methanol was degassed prior to use. Benzophenone (Aldrich) was
sublimed prior to use. MeLi (1.6 M in diethyl ether) and n-BuLi (1.6
M in hexanes) were purchased from Aldrich, and the exact
concentration was determined by titration with N-benzylbenzamide
(Aldrich) before use.34 Chlorotrimethylsilane and trityl chloride were
purchased from Aldrich and were used as received. 1a35,36 and
polymerization of 1a and to that determined computationally.
Furthermore, Eyring plots for the polymerization of 1a and 1b
(Figure S16) were used to determine the enthalpy and entropy
of activation for the polymerization. High enthalpies of
activation were obtained for both polymerizations (1a: ΔH‡
= +13.4 kcal mol−1; 1b: ΔH‡ = +18.8 kcal mol−1), whereas the
entropies of activation were close to zero (ΔS‡ = −0.0072 kcal
K−1 mol−1; 1b ΔS‡ = +0.0111 kcal K−1 mol−1). These
experimentally determined values are well matched with those
obtained computationally (ΔH‡ = +17.9 kcal mol−1 and ΔS‡ =
+0.0192 kcal K−1 mol−1). Hence, we conclude that the near-
zero entropy of activation reflects the fact that the rate-
determining C−H activation step is intramolecular rather than
intermolecular. We note that this experimental result is
consistent with the proposed addition−isomerization mecha-
37
XylP(SiMe3)2 were prepared according to adapted literature
nism involving TS‡
.
I−II
procedures. 1a-d9 was prepared in the same manner, however, starting
from mesitylene-d9 which was prepared according to a literature
procedure.33 Kinetic studies were performed following a literature
procedure.26
SUMMARY
■
We have employed a multifaceted approach to elucidate an
unexpected microstructure and a highly unusual addition−
isomerization mechanism for the anion-initiated polymerization
of phosphaalkenes (ArPCPh2, 1a,b). Specifically, 1- and 2-
dimensional NMR spectroscopic experiments provided spec-
troscopic evidence for the presence of −CHPh2 and Ar−CH2−
moieties within the resultant polymers 3a and 3b. Their
formation requires that the o-CH3 group of the P−Ar
substituent be activated and a proton transferred in a highly
regioregular fashion during propagation. These data were
complemented through the preparation of model compounds
and deuterium-labeling experiments that demonstrated a C−
H/C−D activation of the o-CH3 moiety is feasible. Kinetic
studies of the n-BuLi-initiated polymerization of 1a and 1a-d9 in
THF revealed a large kinetic isotope effect (≤23). These
measurements confirmed that C−H/D activation of the o−Me
group of the P−Ar substituent is the rate-determining step.
Moreover, the experimentally determined activation energy for
propagation was close to that modeled using DFT
computations. Therefore, we conclude that the anionic
polymerization of P-Mes and P-Xyl phosphaalkenes (1a and
1b, respectively) occurs via a unique addition−isomerization
mechanism of propagation. Future studies will explore the
potential fine-tuning of this novel mechanism, perhaps
uncovering methods to favor microstructure x or micro-
structure y, and provide access to unprecedented hybrid
functional homo- and copolymers.
X-ray Crystallography. All single crystals were immersed in oil
and mounted on a glass fiber. Data were collected on a Bruker X8
APEX II diffractometer with graphite-monochromated Mo Kα
radiation. All structures were solved by direct methods and subsequent
Fourier difference techniques. All non-hydrogen atoms were refined
anisotropically with hydrogen atoms being included and refined using
the riding model. All data sets were corrected for absorption effects
(SADABS), Lorentz, and polarization effects. All calculations were
performed using the SHELXT crystallographic software package from
Bruker AXS.38 Additional crystal data and details of data collection and
structure refinement are included in the Supporting Information
(Table S1). All crystallographic data have been deposited with the
Cambridge Structural Database: 1811958 (2a) and 1058778 (2b).
Computational Details. Density functional theory calculations
were performed in Gaussian 09 (Revision D.01).39 Initial geometry
optimizations were performed using the 6-31+G(d) basis set and the
B3LYP functional (denoted as BS1).40,41 Frequency calculations were
used to confirm the identity of minima (no negative eigenvalues) and
transition states (one negative eigenvalue) using BS1. Single point
energy calculations were used to calculate enthalpy for the geometries
using the 6-311++G(d,p) basis set on all atoms and the B97D3
functional. Additionally, solvation corrections were implemented using
the polarizable continuum model42 using tetrahydrofuran (ε = 7.43) as
solvent. Frequency calculations were used to obtain free energies (G°)
where possible (systems fewer than 70 atoms). Single point energy
calculations were rerun with the wB97XD and CAM-B3LYP43
functionals but showed a marginally higher transition state energy
for TS‡I−II. Optimizing the ground state geometries with these
functionals using the 6-31+G(d) basis set had little to no change in
substrate conformation.
Synthetic Details. Preparation of 1b. This compound has been
reported previously from the dehydrochlorination of m-XylP(Cl)-
CHPh2 (m-Xyl = 2,6-dimethylphenyl).44 The present work employed
a different procedure. To a stirred solution of m-XylP(SiMe3)2 (6.35 g,
22.5 mmol) in THF (30 mL), MeLi (15 mL, 24 mmol, 1.6 M in Et2O)
was added. The resulting yellow solution was heated to 55 °C for 1.5
h. Upon cooling to −78 °C, a solution of benzophenone (4.10 g, 22.5
mmol) in THF (20 mL) was added, and the reaction mixture turned
dark red. After stirring for 30 min the reaction mixture was warmed to
room temperature. At this time an aliquot was removed from the
reaction mixture and analyzed by 31P NMR spectroscopy, which
showed a single resonance at 231 ppm. The reaction was stirred for an
additional 30 min, over which time it was cooled to −78 °C, and
chlorotrimethylsilane (3.7 mL, 29 mmol, ρ = 0.86 g mL−1) was added.
Upon warming of the reaction vessel to room temperature the solution
turned from a dark red to a yellow color over a period of 30 min. All
volatiles were removed in vacuo, and to the resulting oil, hexanes (3 ×
30 mL) was added, followed by filtration and solvent removal in vacuo.
The crude product was distilled under vacuum and subsequently
EXPERIMENTAL SECTION
■
Materials and Methods. All manipulations of air- and/or water-
sensitive compounds were performed under a nitrogen atmosphere
using standard Schlenk or glovebox techniques. 1H, 31P, and 13C NMR
spectra were recorded on Bruker Avance 300, 400, or 600 MHz
spectrometers at room temperature unless otherwise specified.
Chemical shifts are reported relative to residual CHCl3 (δ = 7.26
1
1
for H) and C6D5H (δ = 7.16 for H); 85% H3PO4 as an external
standard (δ = 0.0 for 31P) or CDCl3 (δ = 77.0 for 13C, δ = 7.26 for
2H). Mass spectra were acquired using a Kratos MS 50 instrument in
EI mode (70 eV). Elemental analyses were performed in the
University of British Columbia Department of Chemistry Micro-
analysis Facility. Polymer molecular weights were determined by triple
detection gel permeation chromatography (GPC-MALS) using an
Agilent liquid chromatograph equipped with an Agilent 1200 series
isocratic pump, an Agilent 1200 series standard autosampler,
Phenomenex Phenogel 5 μm narrow bore columns (4.6 × 300 mm)
104 Å (5000−500 000 g mol−1), 500 Å (1000−15 000 g mol−1), and
F
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
recrystallized from hexanes to afford a crystalline yellow solid. Yield:
2.86 g (42%). 31P{1H} NMR (162 MHz, CDCl3): δ 231 (s); 1H NMR
(400 MHz, CDCl3): δ 7.62−6.91 (m, 13H; Ar−H), 2.36 (s, 6H, Ar−
Preparation of 3a. Procedure followed an adapted literature
procedure.26 To a stirred solution of 1a (5.00 g, 15.8 mmol) in THF
(20 mL) at room temperature, n-BuLi (0.21 mL, 0.32 mmol, 1.5 M in
hexanes) was added. The conversion of 1a (234 ppm) to 3a (−10
ppm) was monitored by 31P{1H} NMR. Upon complete conversion
(after ∼16 h), the reaction mixture was quenched with 3−4 drops of
degassed methanol and added dropwise to hexanes (3 × 100 mL) to
precipitate the product. After filtration, the product was isolated as a
1
CH3). 13C{1H} NMR (101 MHz, CDCl3): δ 193.8 (d, JCP = 43 Hz,
PC), 144.8 (d, JCP = 24 Hz), 143.2 (d, JCP = 14 Hz), 140.6 (d, JCP
=
7 Hz), 140.0 (d, JCP = 44 Hz), 129.1 (s), 129.0 (s), 128.7 (m), 128.6
(s), 128.4 (s), 127.9 (s), 127.7 (s), 127.6 (s), 127.5 (s), 127.4 (s), 22.5
(d, JCP = 9 Hz, Ar-CH3). Elemental analysis calcd (%) for C21H19P: C
83.44; H 6.29; found: C 83.65, H 6.36.
pale-yellow powder. Yield: 3.85 g (77%). GPC-LLS (THF): Mn
=
Preparation of 2a. To a stirred solution of 1a (0.25 g, 0.79 mmol)
in toluene (3 mL) at −78 °C, MeLi (1.6 M in diethyl ether, 0.50 mL,
0.80 mmol) was added. Upon warming, the solution became a deep
red color. 31P{1H} NMR analysis of this solution showed complete
consumption of the phosphaalkene starting material indicated by the
absence of a resonance corresponding to the starting material (234
ppm). Instead, a broad singlet (−46 ppm) could be observed, assigned
to Mes(Me)P−CPh2Li. The reaction mixture was once again cooled to
−78 °C, at which time a solution of trityl chloride (0.18 g, 0.81 mmol)
in toluene (1 mL) was added. Upon warming, the reaction became
cloudy and orange. 31P{1H} NMR analysis of this solution showed an
absence of the resonance assigned to Mes(Me)P−CPh2Li and instead
two broad singlet resonances (−18 and −38 ppm) in approximately a
1:3 ratio. After 3 days of stirring at room temperature, 31P{1H} NMR
analysis showed complete conversion to a new broad singlet resonance
(−26 ppm). The reaction mixture was filtered through glass microfiber
paper, and all volatiles were removed under vacuum to give the crude
product as an orange-red solid. Crude yield: 0.32 g (87% when taking
into consideration amounts removed for NMR analysis). In a separate
reaction done on a larger scale (1.52 g, 4.8 mmol of MesPCPh2),
immediately after the addition of trityl chloride and warming the
reaction was filtered, and an aliquot was removed. From this aliquot,
after slow evaporation of the toluene, crystals suitable for X-ray
analysis were isolated. 31P{1H} NMR (121 MHz, CD2Cl2): δ −26 (s).
1H NMR (400 MHz, CD2Cl2): δ 7.51 (d, JHH = 7 Hz, 2H, Ar−H),
7.34−7.11 (m, 23H, Ar−H), 6.76 (s, 1H, Ar−H), 6.43 (m, 1H, Ar−
H), 5.10 (dd, 2JHH = 15 Hz, 4JHP = 10 Hz, 1H, −CHH−), 4.73 (d, 2JHP
13 800 g mol−1, Đ = 1.10, dn/dc = 0.239. 31P{1H} NMR (162 MHz,
1
CDCl3): δ −10 ppm (br). H NMR (600 MHz, CDCl3) δ 7.74−5.94
(br, Ar−H), 5.33−4.27 (br, P−CHPh2), 3.78−2.98 (br, −CH2−),
2.86−1.54 ppm (br, Ar−CH3). 13C{1H} NMR (151 MHz, CDCl3) δ
148.0−146.2 (br), 145.1−141.6 (br), 140.3−136.9 (br), 132.2−127.4
(br), 127.1−125.1 (br), 54.1−49.0 (br, P−CHPh2), 33.8−26.3 (br,
−CH2−), 23.6−21.6 (br, Ar−CH3), 21.3−20.0 (br, Ar−CH3).
Preparation of 3b. To a stirred solution of 1b (0.273 g, 0.904
mmol) in THF (1.5 mL), n-BuLi (11.5 μL, 18.1 μmol, 1.57 M in
hexanes), 16 μL, was added. The conversion of 1b (231 ppm) to 3b
(−11 ppm) was monitored by 31P{1H} NMR. Upon complete
conversion (after ∼22 h), the reaction mixture was quenched with 3−4
drops of degassed methanol and added dropwise to hexanes (3 × 50
mL) to precipitate the product. After filtration, the product was
isolated as a pale yellow powder. Yield: 0.199 g (72%). GPC-LLS
(THF): Mn = 8800 g mol−1, Đ = 1.17. dn/dc = 0.254. 31P{1H} NMR
(162 MHz, CDCl3): δ − 11 ppm (br). 1H NMR (600 MHz, CDCl3):
δ 7.91−6.41 (br, Ar−H), 5.37−4.57 (br, P−CHPh2), 4.11−3.39 (br,
−CH2−), 3.22−1.54 (br, Ar−CH3). 13C{1H} NMR (151 MHz,
CDCl3): δ 148.8−145.2 (br), 144.9−140.4 (br), 130.9−126.9 (br),
126.5−125.0 (br), 56.1−55.9, 53.0−50.0 (br, P−CHPh2), 36.3−28.5
(br, −CH2−), 25.9−24.1 (br, Ar−CH3), 23.9−21.4 (br, Ar−CH3).
Representative Procedure for the Kinetic Studies of
Polymerization. All kinetic measurements followed an identical
procedure, and a representative example for 1b will be given here. All
reactions were setup in a predried vial in the glovebox. To a stirred
solution of 1b (0.364 g, 1.21 mmol) in glyme (3.0 mL) was added n-
BuLi (15.4 μL, 1.57 M in hexanes, 0.0236 mmol). The reaction was
stirred for 1 min, and after this time an aliquot of the reaction mixture
was transferred to a NMR tube. The sample was loaded into the NMR
spectrometer (T = 298.6 K), and a 31P NMR spectrum was recorded
every 15 min (72 scans per NMR experiment) until the reaction was
deemed complete. In order to permit reliable integration of the
obtained 31P NMR spectra, a relaxation time of 3 s and a 30° tip angle
were used. When the reaction was deemed complete by NMR
spectroscopy, the reaction mixture was removed from the glovebox,
and 5 drops of degassed methanol was added to the reaction mixture
to terminate the polymerization reaction. The polymer was purified by
precipitation of a THF solution into hexanes (3 × 50 mL), yield 30%.
The isolated polymers had identical features in their 31P and 1H NMR
spectra to that reported above. Each polymerization was repeated at
least once more at each temperature to ensure reproducible data.
Preparation of 3a-d9. To a stirred solution of 1a-d9 (0.250 g, 0.790
mmol) in glyme (2 mL) at 50 °C, n-BuLi (10 μL, 0.016 mmol, 1.6 M
in hexanes) was added. The conversion of 1a-d9 (234 ppm) to 3a-d9
(−11 ppm) was monitored by 31P{1H} NMR over the course of 7
days. Once the reaction appeared to reach maximum conversion
(∼90%), the mixture was quenched with 3 drops of degassed MeOH
and precipitated with hexanes (3 × 30 mL). After filtration, the
product was isolated as a pale yellow powder. Yield: 25 mg (10%).
GPC-LLS (THF): Mn = 11 000 g mol−1, Đ = 1.09. 31P{1H} NMR
2
= 5 Hz, 1H, P−CHPh2), 3.72 (d, JHH = 15 Hz, 1H, −CHH−), 2.56
2
(s, 3H, o-CH3), 1.90 (s, 3H, p-CH3), 0.89 (d, 3H, JHP = 6 Hz, P−
CH3). 13C{1H} NMR (101 MHz, CD2Cl2): δ 147.7 (s), 146.7 (s),
143.7 (s), 143.6 (d, JCP = 13 Hz), 143.0 (d, JCP = 12 Hz), 138.1 (s),
132.7 (d, JCP = 24 Hz), 130.9 (s), 130.7 (s), 130.0 (s), 129.7 (d, JCP
=
5 Hz), 129.5 (s), 129.4 (s), 129.3 (s), 129.2 (s), 129.2 (s), 129.0 (s),
128.9 (s), 128.9 (s), 128.6 (s), 128.0 (s), 127.0 (d, JCP = 2 Hz), 126.7
3
(d, JCP = 2 Hz), 126.4 (s), 59.0 (s, CPh3), 52.2 (d, JCP = 16 Hz;
3
−CH2−), 44.3 (d, JCP = 34 Hz, P−CHPh2), 23.5 (d, JCP = 5 Hz; o-
CH3), 21.1 (s, p-CH3), 9.7 (d, 3JCP = 10 Hz, P−CH3). MS (ESI-TOF):
m/z: 575.5 [M + H]+. Elemental analysis calcd (%) for C42H39P: C,
87.77; H, 6.84. Found: C, 87.52; H, 6.80.
Preparation of 2b. Procedure adapted from that for 2a, using
phosphaalkene 1b (1.51 g, 5.00 mmol) in toluene (30 mL), MeLi (1.6
M in diethyl ether, 3.4 mL, 5.4 mmol), and trityl chloride (1.62 g, 5.81
mmol) in toluene (5 mL). A notable difference from the preparation
of 2a is that immediately after addition of the trityl chloride solution
just one intermediate is observed (−36 ppm) when monitoring the
reaction progress by 31P{1H} NMR spectroscopy. Yield: 2.21 g (79%).
31P{1H} NMR (162 MHz, CDCl3): δ −24. 1H NMR (400 MHz,
CDCl3): δ 7.49 (d, JHH = 8 Hz, 2H, Ar−H), 7.34−7.05 (m, 23H, Ar−
H), 6.90 (d, JHH = 8 Hz, 1H, Ar−H), 6.75 (t, JHH = 8 Hz, 1H, Ar−H),
2
4
6.61 (m, 1H, Ar−H), 5.01 (dd, JHH = 15 Hz, JHP = 10 Hz, 1H,
2
2
−CHH−), 4.67 (d, JHP = 3 Hz, 1H, P−CHPh2), 3.66 (d, JHH = 15
2
Hz, 1H, −CHH−), 2.56 (s, 3H, Ar−CH3), 0.90 (d, JHP = 6 Hz, 3H;
P−CH3). 13C{1H} NMR (101 MHz, CDCl3): δ 147.1 (s), 142.1 (d,
JCP = 26 Hz), 143.3 (s), 142.8 (s), 142.7 (s), 142.1 (d, J = 12 Hz),
136.1 (s), 135.9 (s), 130.7 (s), 130.3 (s), 129.6 (s) 129.5 (s), 129.2
(s), 129.1 (s), 128.7 (s), 128.5 (s), 128.5 (s), 128.3 (d, JCP = 6 Hz),
128.1 (s), 127.5 (s), 126.6 (s), 126.5 (s), 126.2 (s), 125.9 (s), 58.6 (s,
1
(162 MHz, CDCl3): δ −11 (br). H NMR (600 MHz, CDCl3): δ
7.92−6.23 (br, Ar−H). 2H{1H} NMR (92 MHz, CHCl3): δ 6.21−3.68
(br, P−CDPh2), 3.53−0.46 (br, Ar−CD3). 13C{1H} NMR (151 MHz,
CDCl3): δ 148.1−146.7 (br), 144.9−141.7 (br), 140.2−136.7 (br),
131.2−129.1 (br), 127.3−126.1 (br), 53.9−49.8 (br, P−CDPh2),
32.6−26.3 (br, −CD2−), 23.4−21.5 (br, Ar−CD3), 21.1−19.7 (br,
Ar−CD3).
1
3
−CPh3), 51.8 (d, JCP = 34 Hz, P−CHPh2), 44.0 (d, JCP = 17 Hz,
3
1
−CH2−), 23.4 (d, JCP = 5 Hz, Ar−CH3), 9.7 (d, JCP = 21 Hz, P−
CH3). MS (EI, 70 eV): m/z: 560.0 [M+]. Elemental analysis calcd (%)
for C41H37P: C, 87.86 H, 6.61. Found: C, 87.94; H, 6.75.
G
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
isocyanate and acrylamide derivatives. J. Polym. Sci., Part A: Polym.
Chem. 1999, 37, 465−472.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.8b00100.
■
S
(10) Guaita, M.; Camino, G.; Trossarelli, L. Initiation of the Sodium
tert-Butoxide Catalyzed Hydrogen Transfer Polymerization of N-
substituted Acryamides in Aprotic Solvents: N-Methylacrylamide and
N-Phenylacrylamide. Makromol. Chem. 1970, 131, 309−311.
(11) Wexler, H. Migrational Polymerization of Methacrylamide.
Makromol. Chem. 1968, 115, 262.
NMR spectra of synthesized compounds, Arrhenius and
Eyring plots for polymerization of 1a and 1b; X-ray
crystallographic parameters for 2a and 2b; coordinates of
computational structures; kinetics data for polymer-
ization of 1a and 1b (PDF)
Crystal data for 2a (CCDC 1811958) and 2b (CCDC
1058778) (CIF)
(12) Yokota, K.; Shimizu, M.; Yamashita, Y.; Ishii, Y. Hydrogen
Migration Polymerization of N-Substituted Acrylamides. Makromol.
Chem. 1964, 77, 1−6.
(13) Breslow, D. S.; Hulse, G. E.; Matlack, A. S. Synthesis of Poly-
Beta-Alanine from Acrylamide - a Novel Synthesis of Beta-Alanine. J.
Am. Chem. Soc. 1957, 79, 3760−3763.
(14) Priegert, A. M.; Rawe, B. W.; Serin, S. C.; Gates, D. P. Polymers
and the p-block elements. Chem. Soc. Rev. 2016, 45, 922−953.
(15) Allcock, H. R. The expanding field of polyphosphazene high
polymers. Dalton Trans. 2016, 45, 1856−1862.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: dgates@chem.ubc.ca (D.P.G).
ORCID
Derek P. Gates: 0000-0002-0025-0486
Notes
The authors declare no competing financial interest.
(16) Carrera, E. I.; Seferos, D. S. Semiconducting Polymers
Containing Tellurium: Perspectives Toward Obtaining High-Perform-
ance Materials. Macromolecules 2015, 48, 297−308.
́
(17) Jakle, F.; Fernandez, E.; Whiting, A. Recent Advances in the
̈
Synthesis and Applications of Organoborane Polymers. Top. Organo-
met. Chem. 2015, 49, 297−325.
(18) Zhou, J.; Whittell, G. R.; Manners, I. Metalloblock Copolymers:
New Functional Nanomaterials. Macromolecules 2014, 47, 3529−3543.
(19) Bates, J. I.; Dugal-Tessier, J.; Gates, D. P. Phospha-organic
chemistry: from molecules to polymers. Dalton Trans. 2010, 39,
3151−3159.
ACKNOWLEDGMENTS
■
We are grateful to the Natural Sciences and Engineering
Research Council (NSERC) of Canada for support of this work
in the form of Discovery and Research Tools grants to D.G.
The authors also thank Prof. Pierre Kennepohl for valuable
discussions and guidance in performing the DFT calculations,
(20) Dillon, K. B.; Mathey, F.; Nixon, J. F. Phosphorus: The Carbon
Copy; Wiley: West Sussex, England, 1998.
(21) Rawe, B. W.; Brown, C. M.; MacKinnon, M. R.; Patrick, B. O.;
Bodwell, G. J.; Gates, D. P. A C-Pyrenyl Poly(methylenephosphine):
Oxidation “Turns On” Blue Photoluminescence in Solution and the
Solid State. Organometallics 2017, 36, 2520−2526.
(22) Serin, S. C.; Dake, G. R.; Gates, D. P. Addition-Isomerization
Polymerization of Chiral Phosphaalkenes: Observation of Styrene-
Phosphaalkene Linkages in a Random Copolymer. Macromolecules
2016, 49, 4067−4075.
(23) Serin, S. C.; Dake, G. R.; Gates, D. P. Phosphaalkene-Oxazoline
Copolymers with Styrene as Chiral Ligands for Rhodium(I). Dalton
Trans. 2016, 45, 5659−5666.
(24) Rawe, B. W.; Chun, C. P.; Gates, D. P. Anionic Polymerisation
of Phosphaalkenes Bearing Polyaromatic Chromophores: Phosphine
Polymers Showing “Turn-On” Emission Selectively with Peroxide.
Chem. Sci. 2014, 5, 4928−4938.
(25) Siu, P. W.; Serin, S. C.; Krummenacher, I.; Hey, T. W.; Gates, D.
P. Isomerization Polymerization of the Phosphaalkene MesP=CPh2:
An Alternative Microstructure for Poly(methylenephosphine)s. Angew.
Chem., Int. Ed. 2013, 52, 6967−6970.
(26) Noonan, K. J. T.; Gates, D. P. Studying a Slow Polymerization:
A Kinetic Investigation of the Living Anionic Polymerization of P=C
Bonds. Macromolecules 2008, 41, 1961−1965.
(27) Noonan, K. J. T.; Gates, D. P. Ambient-Temperature Living
Anionic Polymerization of Phosphaalkenes: Homopolymers and Block
Copolymers with Controlled Chain Lengths. Angew. Chem., Int. Ed.
2006, 45, 7271−7274.
(28) Tsang, C. W.; Yam, M.; Gates, D. P. The Addition
Polymerization of a P=C Bond: A Route to New Phosphine Polymers.
J. Am. Chem. Soc. 2003, 125, 1480−1481.
2
Dr. Paul Xia for help with H NMR experiments, and Dr.
Spencer Serin for performing X-ray diffraction experiments on
crystals of 2a and 2b.
REFERENCES
■
(1) Odian, G. G. Principles of Polymerization, 4th ed.; Wiley-
Interscience: Hoboken, NJ, 2004.
(2) Yang, J. X.; Liu, S. C.; Zhu, F. H.; Huang, Y. W.; Li, B.; Zhang, L.
New Polymers Derived from 4-Vinylsilylbenzocyclobutene Monomer
with Good Thermal Stability, Excellent Film-Forming Property, and
Low-Dielectric Constant. J. Polym. Sci., Part A: Polym. Chem. 2011, 49,
381−391.
(3) Ganicz, T.; Stanczyk, W. A.; Gladkova, N. K.; Sledzinska, I.
Vinylsilanes as monomers for side chain polymer liquid crystals.
Macromolecules 2000, 33, 289−293.
(4) Gan, Y. D.; Prakash, S.; Olah, G. A.; Weber, W. P.; HogenEsch,
T. E. Anionic synthesis of narrow molecular weight distribution
poly(trimethylvinylsilane) (PTMVS), polystyrene-PTMVS block
copolymers, and poly(phenyldimethylvinylsilane). Conversion of
poly(phenyldimethylvinylsilane) into poly(fluorodimethylvinylsilane).
Macromolecules 1996, 29, 8285−8288.
(5) Baraki, H.; Habaue, S.; Okamoto, Y. Stereospecific Anionic
Polymerization and Novel Hydrogen-Transfer Polymerization of α-
(Aminomethyl)acrylates Having Unprotected Amino Group. Polym. J.
1999, 31, 1260−1266.
(6) Ono, H.; Hisatani, K.; Kamide, K. NMR Spectroscopic Study of
Side Reactions in Anionic Polymerization of Acrylonitrile. Polym. J.
1993, 25, 245−265.
(29) Gillon, B. H.; Noonan, K. J. T.; Feldscher, B.; Wissenz, J. M.;
Kam, Z. M.; Hsieh, T.; Kingsley, J. J.; Bates, J. I.; Gates, D. P.
Molecular Studies of the Initiation and Termination Steps of the
Anionic Polymerization of P=C Bonds. Can. J. Chem. 2007, 85, 1045−
1052.
(30) Zavitsas, A. A. The Relation Between Bond Lengths and
Dissociation Energies of Carbon-Carbon Bonds. J. Phys. Chem. A
2003, 107, 897−898.
(7) Iwamura, T.; Adachi, K.; Chujo, Y. Synthesis of organic-inorganic
polymer hybrids utilizing in-situ anionic hydrogen-transfer polymer-
ization of acrylamide. Polymer 2016, 92, 13−17.
(8) Iwamura, T.; Tomita, I.; Suzuki, M.; Endo, T. Hydrogen-transfer
polymerization behavior of N-acylacrylamide. J. Polym. Sci., Part A:
Polym. Chem. 2000, 38, 430−435.
(9) Iwamura, T.; Tomita, I.; Suzuki, M.; Endo, T. Hydrogen-transfer
polymerization of vinyl monomers derived from 4-methylbenzoyl
H
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(31) Schreiner, P. R.; Chernish, L. V.; Gunchenko, P. A.;
Tikhonchuk, E. Y.; Hausmann, H.; Serafin, M.; Schlecht, S.; Dahl, J.
E. P.; Carlson, R. M. K.; Fokin, A. A. Overcoming Lability of
Extremely Long Alkane Carbon-Carbon Bonds Through Dispersion
Forces. Nature 2011, 477, 308−311.
(32) Characteristic Bond Lengths in Free Molecules. CRC Handbook
of Chemistry and Physics, 84th ed.; CRC: Boca Raton, FL, 2003.
(33) Chen, T.-S.; Wolinska-Mocydlarz, J.; Leitch, L. C. Synthesis of
Deuteriomethyl Aromatic Hydrocarbons by Exchange with Dimethyl-
sulfoxide-D6. J. Labelled Compd. 1970, 6, 285−288.
(34) Burchat, A. F.; Chong, J. M.; Nielsen, N. Titration of
alkyllithiums with a simple reagent to a blue endpoint. J. Organomet.
Chem. 1997, 542, 281−283.
(35) Becker, G.; Uhl, W.; Wessely, H.-J. Acyl- und Alkylidenphos-
phane. XVI. (Dimethylaminomethyliden)- und (Diphenylmethyliden)-
phosphane. Z. Anorg. Allg. Chem. 1981, 479, 41−56.
(36) Yam, M.; Chong, J. H.; Tsang, C. W.; Patrick, B. O.; Lam, A. E.;
Gates, D. P. Scope and Limitations of the Base-Catalyzed Phospha-
Peterson P=C Bond-Forming Reaction. Inorg. Chem. 2006, 45, 5225−
5234.
(37) Becker, G.; Mundt, O.; Rossler, M.; Schneider, E. Bildung und
Eigenschaften von Acylphosphanen. VI. Synthese von Alkyl- und
Arylbis (trimethylsilyl)- sowie Alkyl und Aryltrimethylsilyphosphanen.
Z. Anorg. Allg. Chem. 1978, 443, 42−52.
(38) Sheldrick, G. SHELXT - Integrated space-group and crystal-
structure determination. Acta Crystallogr., Sect. A: Found. Adv. 2015,
71, 3−8.
(39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers, E. N.;
Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.;
Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
̈
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(40) Becke, A. D. Density-functional thermochemistry. III. The role
of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652.
(41) Becke, A. D. Density-Functional Exchange-Energy Approx-
imation with Correct Asymptotic Behavior. Phys. Rev. A: At., Mol., Opt.
Phys. 1988, 38, 3098−3100.
(42) Tomasi, J.; Mennucci, B.; Cammi, R. Quantum Mechanical
Continuum Solvation Models. Chem. Rev. 2005, 105, 2999−3094.
(43) Yanai, T.; Tew, D. P.; Handy, N. C. A new hybrid exchange−
correlation functional using the Coulomb-attenuating method (CAM-
B3LYP). Chem. Phys. Lett. 2004, 393, 51−57.
(44) van der Knaap, T. A.; Klebach, T. C.; Visser, F.; Bickelhaupt, F.;
Ros, P.; Baerends, E. J.; Stam, C. H.; Konijn, M. Synthesis and
Structure of Aryl-Substituted Phospha-Alkenes. Tetrahedron 1984, 40,
765−776.
I
DOI: 10.1021/acs.macromol.8b00100
Macromolecules XXXX, XXX, XXX−XXX