Synthesis of poly(alkyl/arylphosphazenes) via the ambient temperature phosphite-mediated chain-growth polycondensation of (N-Silyl) bromophosphoranimines

Article abstract of DOI:10.1021/ma100876z

The room temperature addition of stoichiometric amounts of trimethyl phosphite, P(OMe)₃, to N-silyl(halogeno)organophosphoranimines BrRR'PdNSiMe₃ in chlorinated solvents led to the direct formation of high molecular weight polyphosph

Full text of DOI:10.1021/ma100876z

7446 Macromolecules 2010, 43, 7446–7452
DOI: 10.1021/ma100876z
Synthesis of Poly(alkyl/arylphosphazenes) via the Ambient
Temperature Phosphite-Mediated Chain-Growth Polycondensation of
(N-Silyl)bromophosphoranimines
Thomas J. Taylor,† Alejandro Presa Soto,† Keith Huynh,‡ Alan J. Lough,‡ Anthony C. Swain,§
Nicholas C. Norman,† Christopher A. Russell,† and Ian Manners*,†
†School of Chemistry, University of Bristol, Cantocks Close, Bristol, BS8 1TS, U.K.,
‡Department of Chemistry, University of Toronto, 80 St. George Street, Toronto M5S 1A1, Ontario,
Canada, and §AWE, Aldermaston, Reading, Berkshire RG7 4PR, U.K.
Received April 20, 2010; Revised Manuscript Received July 15, 2010
ABSTRACT: The room temperature addition of stoichiometric amounts of trimethyl phosphite, P(OMe)3,
to N-silyl(halogeno)organophosphoranimines BrRR0PdNSiMe3 in chlorinated solvents led to the direct forma-
tion of high molecular weight polyphosphazenes [RR0PdN]n. The majority of polymerizations were complete
within 18 h. The polymers prepared include poly(dialkylphosphazenes) (e.g., [nBu2PdN]n 1b), poly(alkylaryl-
phosphazenes) (e.g., [PhMePdN]n 1d), new materials featuring unsaturated substituents (e.g., [nHex-
{H2CdC(H)CH2}PdN]n 1n), and random copolymers (e.g., {[PhMePdN]x-[PhnBuPdN]y} where x:y = 2:1,
7a). The precursor silylaminophosphines RR0P-N(SiMe3)2 (5a-q) and bromo(silylamino)phosphoranimines
BrRR0PdNSiMe3 (4a-q) were synthesized and fully characterized prior to polymerization studies. The presence
of alkoxy, carboranyl, and/or phenyl substituents on the N-silylbromophosphoranimines, as found in BrEt-
[CF3CH2O]PdNSiMe3 (4g), Br(2-[Me]-o-C2B10H10)EtPdNSiMe3 (4h), or BrPh2PdNSiMe3 (4i), respectively,
was found either to severely retard or to preclude polymerization altogether. The mild reaction conditions enabled
the preparation of polyphosphazenes that are substituted with reactive alkyne groups (e.g., R = Et, R0 = -CH2-
CtCSiMe3 1p), materials that have not been accessible using high-temperature thermal routes. These moieties
undergo further convenient chemical transformations as illustrated by deprotection of 1p by TBAF 3H2O
3
(TBAF = tetra(n-butyl)ammonium fluoride) as well as chemical cross-linking with the disiloxane HMe2SiOSi-
Me2H in the presence of Karstedt’s catalyst. The polyphosphazene materials were characterized by a variety of
techniques including 1H, 13C, and 31P NMR spectroscopy and GPC and, in selected cases, by IR, DLS, TGA,
DSC, and WAXS.
Introduction
attractive properties that complement those of polyphosphazenes
accessed via the ROP route.3,4
In the field of inorganic polymers, polyphosphazenes are re-
garded as promising candidates for a wide variety of different
applications.1-3 The most well-established route to these interesting
materials involves thermal ring-opening polymerization (ROP) of
[NPCl2]3 at 200-250 ꢀC. This leads to the formation of polydichlo-
rophosphazene [Cl2PdN]n which, following subsequent macro-
molecular halogen replacement reactions with oxygen- or nitro-
gen-based nucleophiles, provides access to polyphosphazenes with
alkoxy, aryloxy, and amino substitutents.1 Poly(alkyl/arylphos-
phazenes) (1, [RR0PdN]n) with pendant organic groups directly
attached through phosphorus-carbon bonds to the backbone are
structural analogues of the isoelectronic silicones such as [Me2Si-
O]n and are also an attractive target.3 However, the formation
of polyphosphazenes with side groups bound by P-C bonds via
substitution chemistry on poly(dichlorophosphazene) is generally
not feasible with Grignard and organolithium reagents as they
degrade the polyphosphazene backbone, especially under the for-
cing conditions often required to ensure complete substitution.3
The key breakthrough in the preparation of poly(alkyl/arylphos-
phazenes) involved the discovery of the high-temperature (180-
200 ꢀC) chain growth polycondensation of phosphoranimines of
type (CF3CH2O)RR0PdNSiMe3 (2), which was reported in 1980
by Neilson and Wisian-Neilson (Scheme 1).3 Materials formed
by this route and by subsequent postpolymerization modifica-
tion involving deprotonation and electrophilic strategies possess
Recently, two examples of the preparation of poly(dichloro-
phosphazene) via an ambient temperature route have been des-
cribed. These methodsinvolveROP of [NPCl2]3 in the presence of
silylcarborane initiators5 and chain growth polycondensation6 of
the phosphoranimine Cl3PdNSiMe3 in the presence of PCl5. The
latter route can also be adapted to organo-substituted phosphor-
animines RCl2PdNSiMe3 and RR0ClPdNSiMe3 to yield poly-
phosphazenes with substituents bound by direct P-C bonds,
although the molecular weights are generally low (Scheme 2).7
In a preliminary communication we reported that the chain
growth polycondensation of N-silyl-(bromo)organophosphorani-
mines can be performed at room temperature in the presence of
stoichiometric or catalytic amounts of phosphites P(OR)3 (e.g.,
P(OMe)3) to give high molecular weight poly(alkyl/arylphospha-
zenes) (Scheme 3).8 This mild new procedure offers opportunities
to prepare poly(alkyl/arylphosphazenes) with unsaturated side
groups such as vinyl, allyl, and acetylide which are difficult or
impossible to introduce via the elevated temperature procedure.9 In
this paper we examine the scope and limitations of this new and
promising ambient temperature approach.
Results and Discussion
1. Synthesis and Characterization of the Phosphoranimine
Monomers (4). a. Synthesis and Characterization of N-Bis-
(dimethylsilyl)phosphines (5a-q). The precursors to the phos-
phoranimine monomers 4, the N-bis(trimethylsilyl)phosphines
*Corresponding author. E-mail: ian.manners@bris.ac.uk.
pubs.acs.org/Macromolecules
Published on Web 08/17/2010
r 2010 American Chemical Society

Article
Macromolecules, Vol. 43, No. 18, 2010 7447
Scheme 1
Scheme 2
Scheme 3
Scheme 4
Figure 1. Molecular structure of N-bistrimethylsilylamino-([Me]-o-do-
decacarborane)ethylphosphine (5h) using 50% probability thermal
ellipsoids. All hydrogen atoms have been removed for clarity.
Scheme 6
Scheme 7
The carborane/ethylphosphine species 5h was made in a
straightforward manner by lithiating Me-o-dodecacarborane
and reacting the product with ClEtPN(SiMe3)2 to give an air-
stable phosphine that could be recrystallized from ethanol. This
species was notably resistant to oxidation to the phosphine
oxide compared to the other phosphines prepared. The struc-
ture of 5h was also confirmed by single crystal X-ray diffrac-
tion (Figure 1).13 Even with two sterically demanding pendant
Scheme 5
groups, access to the P atom does not appear to be signifi-
P
cantly blocked from potential oxidants ( —P=315.1ꢀ). The
31P NMR resonance at 72.6 ppm is removed from those of the
other silylaminophosphines in this study (40-55 ppm). This is
in good agreement with analysis of a similar neutral dodecar-
borane disopropylamino indenyl phosphine ligand (6) synthe-
sized by Xie et al. which possesses a 31P NMR resonance at 74.1
ppm, and the respective P-C/P-N distances all fall within 0.03
(5, Scheme 4), were prepared by reacting the chlorophosphines
(Cl3P, ClRR0P or Cl2RP) with lithium bis(trimethylsilyl)amide
followed by the appropriate nucleophile(s) in Et2O, THF, or a
combination thereof (Scheme 5).
It was found that the route favored in published syntheses,10
which involves the direct reaction of Grignard reagents with the
chlorophosphine, in our hands generally led to lower yields.
Whenever feasible, the corresponding lithium reagent was cho-
sen for the synthesis of new phosphines from chlorinated species
in the nucleophilic substitution step. The n-hexyl/allyl variant 5n
required the protection and deprotection of the allylphos-
phine via the diisopropylamine phosphine (Scheme 6).11 For
the allenic/phenyl species 5o, the phosphine was synthesized by
lithiating but-2-yne12 and reacting with ClPhPN(SiMe3)2.
Using the protected 1-trimethylsilylpropyne gave the alkyne-
containing product 5q exclusively (Scheme 7). The distilled
phosphines 5 were in most cases quite sensitive to oxidation and
were characterized by multinuclear NMR spectroscopy prior
to conversion to the phosphoranimine (vide infra).
14
˚
A of those in the structure of 5h.
b. Synthesis and Characterization of N-Silylbromopho-
sphoranimines (4a-q). The N-silylhalogenophosphorani-
mine monomers for the polymerizations were synthesized
from the respective phosphines by direct bromination

7448 Macromolecules, Vol. 43, No. 18, 2010
Taylor et al.
Table 1. Synthesis of Poly(alkyl/arylphosphazene) Homopolymers [RR0PdN]n (1) via the Phosphite-Mediated Condensation of the Respective
Monomers (4) (in CDCl3, 1 equiv of P(OMe)3, 25 ꢀC)
monomer
R
R0
polymer product
yield, % (time, h)
Mw (ꢀ10-5),b g mol-1
PDI
a
4a
4b
4c
4d
4e
4f
4g
4h
4i
Me
Me
nBu
nHex
Ph
1a
1b
1c
1d
1e
90 (3)
85 (12)
80 (12)
87 (3)
87 (12)
no reaction
60 (240)
no reaction
80 (60)
nBu
nHex
Me
(insoluble)
(insoluble)
4.80
1.70
3.40
nBu
-OCH2CF3
-OCH2CF3
o-(Me)C2B10H11
Ph
Ph
-OCH2CF3
Et
1.28
1g
1i
5.34
1.41
Et
Ph
(insoluble)
4j
4k
4l
4m
4n
4o
4p
4q
-CtC-Me
Ph
Ph
no reaction
no reaction
90 (36)
90 (6)
95 (4)
no reaction
95 (4)
17 (50)
-CtC-Ph
-CH2C(H)dCH2
-CH2C(H)dCH2
-CH2C(H)dCH2
(Me)CdCdCH2
-CH2CtC-SiMe3
-CH2CtC-SiMe3
-CH2C(H)dCH2
Et
nHex
Ph
Et
1l
1m
1n
3.07
4.85
2.24
1.81
1.23
3.32
1p
1q
1.06
2.33
5.20
4.01
Ph
a Could not be analyzed by our GPC instrument due to insolubility in THF. b GPC data were collected in THF containing 0.1 wt % [nBu4N]Br using
polystryene standards for column calibration.
(Scheme 4) which proceeded at -78 ꢀC in dichloromethane.
This reaction modification was found to increase the yield
over the published method using benzene at 0 ꢀC for
brominations.3b Even when unsaturated groups were pre-
sent, no significant bromination of the alkene or alkyne
moieties was observed. For example, the allenic species 5o
was cleanly oxidized to phosphoranimine 4o, readily identi-
fied in particular by the diagnostic 13C NMR resonance for
the internal allenic carbon at 210 ppm.
identified by comparison to literature by 1Hand31P NMR. The
poly(dialkylphosphazenes) [nBu2PdN]n (1b) and [nHex2Pd
N]n (1c) were prepared in a similar manner. These polymers
remained soluble in the reaction media, and their solution 31
P
NMR spectra showed signals at 20.5 and 21.0 ppm, respec-
tively.3 Further analysis was hindered on the purified materials
as they became increasingly insoluble in common solvents after
multiple precipitations. The unsymmetrical polyphosphazene
[PhnBuPdN]n (1e) was also synthesized. This mixed-substitu-
ent material was much more soluble in a variety of solvents and
was fully characterized.
The more volatile (bp<100 ꢀC, ca. 10-2 mbar) or robust
phosphoranimines were purified by distillation as clear
colorless liquids; many of the less volatile phosphoranimines,
particularly those with trimethylsilyl-capped alkyne moieties
(4q and 4p), decomposed under the same conditions. The
brown residues that remained gave broad signals in their 31P
NMR spectra, and a series of low molecular weight species
were indicated by MALDI-TOF measurements. The partial
purification for these phosphoranimines was achieved by
removal of the volatile side products (e.g., solvent, BrSiMe3),
leaving the product in 95-99% purity (by 1H and 31P NMR
spectroscopy). This level of purification was found to be
sufficient for the following polymerizations.
2. P(OMe)3-Mediated Polycondensation of Phosphoranimines
4a-e, 4g, 4i, 4l-n, 4p, and 4q. a. Homopolymerizations. To
examine the polymerization behavior of the bromophosphor-
animines, these species were treated with a stoichiometric
amount of P(OMe)3 in chloroform, and the resulting solution
was allowed to stir at room temperature under an inert atmo-
sphere. The reaction was monitored by 31P NMR spectroscopy.
Substoichiometric amounts of P(OMe)3 also initiate the poly-
merization effectively, but the reaction is significantly slower.8
Completion of the successful reactions was suggested by the
increased viscosity of the solutions and confirmed by the con-
sumption of the starting phosphoranimine monomer and the
growth of broad signals significantly shifted upfield by 31PNMR
spectroscopy which were assigned to the resulting polymer.
Workup was carried out by concentration of the reaction
solution followed by multiple precipitations into n-pentane
and/or hexanes. Molecular weights of the resulting polyphos-
phazenes were estimated by GPC using polystyrene standards,
and for soluble materials Mw values were around or larger than
105 g mol-1. The results are tabulated in Table 1.
The chloro-substituted phosphoranimine ClPhMePdNSi-
15
Me3 (3, R=Me, R0=Ph, Scheme 2) was also shown to under-
go polymerization to afford [PhMePdN]n (1d) under similar
conditions to the bromo analogue 4d. Interestingly, the quan-
titative generation of 1d16 was observed after 18 h, substantially
longer than the polymerization of BrMePhPdNSiMe3 (4d)
(3 h). The shorter reaction time for the polymerization of 4d
may be the due to the weaker P-Br bond compared to the
P-Cl bond present in ClMePhPdNSiMe3.
The presence of electron-withdrawing groups at phosphorus
was found to significantly affect the propensity of N-silylhalo-
genophosphoranimines to polymerize in the presence of a phos-
phite. Fully halogenated species (i.e., Cl3PNSiMe3) as well as
those with two alkoxides directly bound to the phosphorus
center such as Br(CF3CH2O)2PdNSiMe3 (4f) were resistant to
polymerization in the presence of P(OMe)3. Increasing the
reaction temperature did not prove to be a viable alternative:
heating 4f with P(OMe)3 at 60 ꢀC did not result in polymeri-
zation, and instead the formation of an unidentified mixture of
products was indicated by 31P NMR spectroscopy.
The mixed alkyl/alkoxy species, Br(CF3CH2O)EtPdNSi-
Me3 (4g), showed no noticeable polymerization after stirring
for 2 days in CDCl3 with1equivofP(OMe)3. However, after10
days the solution was found to be slightly viscous, and 31P
NMR spectroscopy showed a broad peak centered at 20.2 ppm.
A polymer(1g) was isolated in 60% yield as a white powder and
possessed a Mw of 534 000 Da with a PDI value of 1.41. The
presence of even a single alkoxy substituent group in 4g appears
to severely hinder the polymerization process.
The phosphoranimines BrPh(MeCtC)PdNSiMe3 (4j) and
BrPh(PhCtC)PdNSiMe3 (4k) were also treated with equimo-
lar amounts of P(OMe)3 in CDCl3 at room temperature, and no
reaction was observed. The phosphoranimine BrPh2PdNSi-
Me3 (4i) was surveyed as well, and while the insoluble polymer
[Ph2PdN]n (1i) precipitated out of solution as it formed, the
All of the poly(alkyl/arylphosphazenes) 1a-e, 1g, 1i, 1l-n,
1p, and 1q were isolated as white/off-white fibrous materials or
powders. The previously reported and well-known polymers
[Me2PdN]n (1a)3 and [MePhPdN]n (1d)3 were synthesized and

Article
Macromolecules, Vol. 43, No. 18, 2010 7449
3 days required to fully consume the monomer indicated that
this was one of the slower reactions surveyed in this study.
Other phosphoranimines with electron-withdrawing and
sterically demanding groups were synthesized, and their pro-
pensity for polymerization was investigated. One such example
was the (o-methyldodecacarboranyl)phosphoranimine (4h),
for which no conversion to the polymer was detected after 2
days on exposure to stoichiometric amounts of P(OMe)3 in
to hinder the polymerization of the phosphoranimine. Polyphos-
phazene 1p was stable under ambient conditions and was readily
purified by standard precipitation techniques. The purified
material formed a hard light-yellow glass upon standing but
was very soluble in chlorinated solvents and moderately so in
THF. The material was even moderately soluble in non-coordi-
nating solvents such as toluene.
Two of the phosphoranimines featuring alkyne groups, 4q and
4p, were also reacted with small amounts of PCl5 at room temp-
erature in CH2Cl2 and monitored by 31P and 1H NMR spectro-
scopy to determine if the alternative cationic route (see Scheme 2),
known to be compatible with the presence of saturated alkyl and
phenyl groups,7 was also viable with unsaturated alkyne moieties.
However, a variety of partially chlorinated species, insoluble mate-
rial, and starting material were recovered after 2 h of reaction time.
b. Synthesis and Characterization of Poly(alkyl/arylphospha-
zene) Random Copolymers. This room temperature phosphite-
mediated polymerization route also allowed for the preparation
of different poly(alkyl/arylphosphazene) random copolymers
by using a mixture of two different monomers, as demon-
strated previously3 at elevated temperature (180-200 ꢀC). For
example, treatment of different molar ratios of BrPhMePd
NSiMe3 (4d) [δ(31P)=4.0 ppm] and BrPhnBuPdNSiMe3 (4e)
[δ(31P)=13.3 ppm] in the presence of a stoichiometric amount of
P(OMe)3 in CH2Cl2 at 25 ꢀC afforded random copolymers 7a
and 7b in good yields as white fibrous materials. 31P NMR
spectroscopy confirmed the random nature of these polymers by
showing a single and very broad signal centered at 10.3 ppm (7a)
and 13.3 ppm (7b), respectively, unlike the two signals expected
in case of polymerization of both monomers separately ([Ph-
MePdN]n, δ(31P) = 1.4 ppm; [Ph(n-Bu)PdN]n, δ(31P) = 6.5
ppm). The polymers were shown to possess high molecular
weights by GPC with a Mw values of ca. 1.5 ꢀ 105 g/mol and
a PDI of ca. 2.0. The presence of only one single peak by GPC is
also consistent with the formation of a random copolymer rather
than a mixture of the two respective homopolymers. Moreover,
this result raised the question of whether a random copolymer
with [PhMePdN] and [R2PdN] (R = Me, nBu, and nHex)
repeat units would also be a soluble material. Thus, the random
copolymers (7c-h) were prepared in an analogous manner to 7a,
b and were indeed found to be soluble materials with a ratio of
10-20% of [R2PdN] (R = Me, nBu, and nHex) units in the
main chain. Analysis by 31P NMR spectroscopy again confirmed a
random copolymer structure for 7c-h. The polyphosphazene
copolymers 7c,d showed a single very broad signal centered at
6.3 ppm. However, two sets of broad signals were found for the
copolymers 7e,f (-[NPMePh]-, δ(31P) = 8.1 ppm; -[NP(n-
Bu)2]-, δ(31P) = 29.9 ppm) and for 7g,h ([NPMePh], δ(31P) =
ca. 5.1 ppm; [NP(n-Bu)2], δ(31P) = ca. 27.2 ppm). The relative
integration of the two signals was consistent with the ratio of repeat
units proposed for the copolymers 7e-h. These organopolypho-
sphazenes with saturated alkyl substituents and aryl groups,
possessed high molecular weights by GPC (Table 2).
6
CDCl3. Further attempts to polymerize the material with PCl5
led only to the salt [Br(Et)(o-MeC2B10H10)PdNdPCl3][PCl6]
which was characterized by 1H and 31P NMR spectroscopy.
These results suggest that unsaturated and/or bulky organic
substituents directly attached to the phosphorus center in the
bromophosphoranimines either hinders or prevents polymeri-
zation. The reason for this change in reactivity is not known at
present but may be clarified in the future upon the elucidation of
the mechanism of the polymerization. Nevertheless, the success-
ful synthesis of polyphosphazenes with alkyne and alkene
groups (1l-n, 1p,q) was accomplished by the introduction of a
methylene spacer between the unsaturated moiety and the
backbone (see below). Presumably the presence of the spacer
reduces the influence of the unfavorable electronic and/or steric
factors that otherwise hinder the polymerization.
The polymers [{H2CdC(H)CH2}2PdN]n (1l) and [Et{H2Cd
C(H)CH2}PdN]n (1m) were both readily synthesized, although
the polymerization was much slower in the former case. After
two precipitations into hexanes these materials were found to be
insoluble in hydrocarbons, sparingly soluble in THF, and only
slightly soluble in halogenated solvents. Freshly sonicated sam-
ples of 1m in CH2Cl2 showed a major peak by dynamic light
scattering (DLS) assigned to be unimers with a hydrodynamic
radius, Rh, of 4.7 nm. The material then began to form aggre-
gates of size greater than 250 nm within a period of minutes,
giving a visible precipitate within hours. Addition of aqueous
strong acids to suspensions of 1l and 1m actually decreased their
solubility, and introducing anhydrous HCl to THF suspensions
of the materials did not solubilize the polymers to any degree.
While the lowered solubility of 1l may be rationalized by com-
parison of the material to the intractable poly(di-n-propylphos-
phazene) ([nPr2PdN]n),17 the similar behavior in the mixed sub-
stituent material 1m species was more surprising. The origin of
this limited solubility may be the insufficient length of the side
groups to promote solubility on organic solvents. Indeed, sub-
stantially changing the size of the aliphatic side group was found
to lead to soluble materials. The [nHex{H2CdC(H)CH2}PdN]n
variant (1n), made specifically due to the poor solubilities of the
other species, was isolated as an off-white fibrous material and
was soluble in a broad range of polar and nonpolar solvents.
Phosphoranimines with unsaturated carbon centers attached
directly to phosphorus were found to either hinder polymeriza-
tion or to be completely unreactive. The phenyl/allene phosphor-
animine 4o did not react after stirring 7 days at room temperature
in the presence of P(OMe)3. Increasing the temperature to 65 ꢀC
resulted in the formation of multiple products by 31PNMRspectro-
scopy, none of which corresponded to the corresponding polymer.
We also investigated whether phosphoranimines with the
acetylenic group -CH2CtCSiMe3 would polymerize efficiently.
If successful, this would show that monomers with an internal
alkyne functionality were compatible with this new polymeriza-
tion route and that a saturated methylene group spacer circum-
vents the problems detected when unsaturated organic substi-
tuents are attached directly to phosphorus. A comparison of the
reactivities of the ethyl and phenyl variants of the methylene-
alkynated phosphoranimines (4p and 4q, respectively) clearly
indicated conversion to 1p after 4 h in near quantative yield by
31P NMR spectroscopy while after 2 days only a 17% yield of 1q
was attained. As noted previously, the bulky phenyl group seems
3. Morphology of [nBu2PdN]n (1b) and [nHex2PdN]n (1c).
The crystallinity of the insoluble n-alkyl-substituted poly-
phosphazenes 1b and 1c was probed by DSC and WAXS

7450 Macromolecules, Vol. 43, No. 18, 2010
Taylor et al.
Table 2. Synthesis of Poly(alkyl/arylphosphazene) Random Copolymers [(RR0PdN)-(RR0PdN)]n (7a-h) via the Phosphite-Mediated
Condensation of the Respective Phosphoranimine Monomers 4 (in CDCl3, 1 equiv of P(OMe)3, 25 ꢀC)
RR0 groups from
monomer A (equiv)
RR0 groups from
monomer B (equiv)
polymer
product
yield, %
(time, h)
Mw (ꢀ10-5),
monomer A
monomer B
g mol-1
PDI
4d
4d
4d
4d
4d
4d
4d
4d
PhMe (0.67)
PhMe (0.33)
PhMe (0.9)
PhMe (0.8)
PhMe (0.9)
PhMe (0.8)
PhMe (0.9)
PhMe (0.8)
4e
4e
4a
4a
4b
4b
4c
4c
PhnBu (0.33)
PhnBu (0.67)
Me2 (0.1)
7a
7b
7c
7d
7e
7f
92 (12)
84 (12)
77 (12)
73 (12)
86 (12)
69 (12)
89 (12)
76 (12)
1.48
1.66
1.52
2.17
1.20
0.74
1.05
0.59
2.0
1.8
1.8
1.6
2.4
3.7
2.3
3.2
Me2 (0.2)
nBu2 (0.1)
nBu2 (0.2)
nHex2 (0.1)
nHex2 (0.2)
7g
7h
Figure 2. WAXS pattern (2θ vs counts [arbitrary]) for polymer 1c.
Figure 3. WAXS patterns (2θ vs counts [arbitrary]) for polymers 1b (top) and 1e (bottom) at 25 ꢀC.
measurements, as reported previously for the known inso-
the diffraction peaks and conformational information. For
comparison, the WAXS pattern of the unsymmetically sub-
stituted material [PhnBuPdN]n (1e), which would be expected
to be amorphous, was collected as well. The pattern for the
latter polymer showed only a broad peak and an amorphous
halo (Figure 3, lower trace). The results suggest that the inso-
lubility of 1b and 1c may also be attributed to their crystallinity,
which provides an additional enthalpic barrier to dissolution
due to the lattice energy. However, the insolubility of other
mixed-substituent materials such as 1m suggests that short (C4
or less) substituents also impart insufficient hydrophobicity to
the polymer to counter the effect of the highly polar polymer
backbone, and even such amorphous materials show very low
solubility in common organic solvents.
18
luble polyphosphazenes [Et2PdN]n and [nPr2PdN]n.17 In
the latter two cases the insolubility of the material has been at
least partially attributed to their crystallinity (Tm = 217 and
248 ꢀC, respectively). Polymer 1c displayed a clear Tm at 155
ꢀC that was consistent over multiple heating/cooling cycles.
In contrast, 1b had no detectable Tm below 175 ꢀC (the
maximum temperature investigated).
For the WAXS measurements, the polymers were analyzed
in their as prepared state and also after annealing at 145 ꢀC for
12 h; no significant differences in the WAXS traces were
detected. The WAXS patterns of 1b and 1c (Figure 2 and
upper trace of Figure 3), showed several relatively sharp peaks
characteristic of a semicrystalline material. Although beyond
the scope of this work, the WAXS traces are of sufficient quality
to suggest that further studies of highly oriented materials such
as fibers are worthwhile and may allow successful indexing of
4. Postpolymerization Reactions of Acetylide Polymer 1p.
Proof-of-concept experiments to stimulate further work with
these materials involved a brief study of the postpolymerization

Article
Macromolecules, Vol. 43, No. 18, 2010 7451
Scheme 8
access to living systems with their associated molecular weight
control and narrow polydispersities. Understanding the various
steps in the polymerization may also allow us to expand the range
of polymerizable monomers for this new, mild, and versatile
method.
Experimental Section
Materials. Acetonitrile, diethyl ether, THF (for bulk reac-
tions), and toluene were collected from Grubbs type solvent
purification systems. For sensitive reactions, THF was dried over
sodium benzophenyl ketyl and distilled prior to use. P(OMe)3
(97%) was bought from Aldrich and used directly under glovebox
conditions as well as being further purified by distillation from
sodium wire. It was found that the polymerizations with freshly
distilled phosphite proceeded similarly but were up to 1.5 times
slower. The purchased chloro- and dichlorophosphines were dis-
tilled prior to use. All other reagents were used as purchased. Me2-
PN(SiMe3)2 [5a],20 BrMe2PdNSiMe3 [4a],20 MePhPN(SiMe3)2
[5d],20 BrMePhPdNSiMe3 [4d],20 (CF3CH2O)2PN(SiMe3)2 [5f],20
Br(CF3CH2O)2PdNSiMe3 [4f],20 Ph2PN(SiMe3)2 [5i],20 BrPh2Pd
NSiMe3 [4i],20 (CH2dCHCH2)2PN(SiMe3)2 [5l],21 and Br(CH2d
CHCH2)2PdNSiMe3 [4l]22 were prepared according to the litera-
ture. Preliminary details of the phosphite-mediated synthesis of
[Me2PdN]n (1a) and [MePhPdN]n (1d) were previously reported.8
All synthetic steps were carried out under an atmosphere of nitrogen
or argon. Precipitation of polymers usually involved the formation
of a viscous solution from a good solvent by careful solvent removal
via rotary evaporation followed by the addition of the sample into
the vortex of a stirred nonsolvent.
Equipment. NMR experiments were performed on a JEOL
Lambda 300 and Eclipse 300 spectrometers in dry solvents.
NMR chemical shifts were referenced to tetramethylsilane or
residual protonated solvent peaks. Conventional gel permeation
chromatography (GPC) was carried out on a Viscotek GPCmax
chromatograph equipped with a refractometer. Two columns
(with lower and upper exclusion limits 100-10 000 and 100-
1 000 000) were packed with a porous styrene divinylbenzene
copolymer packing material (5 μm particle size). A flow rate
of 1 mL/min in THF with 0.1% w/w nBu4NBr was used,
and calibration was performed using polystyrene standards.
Dynamic light scattering experiments were performed using a
Malvern spectrometer in a quartz cuvette using THF at 20 ꢀC.
MALDI-TOF mass spectra were collected on a 4700 Proteomics
Analyzer (Applied Biosystems). Samples were prepared in THF
using 2,5-dihydroxybenzoic acid as a matrix (1:5, sample:
matrix) and drop-cast by micropipet into sample wells. TGA
measurements were run on a TGA Q500 apparatus at 10 ꢀC/
min. DSC was run on a DSC Q100 apparatus at 10 ꢀC/min under
a N2 atmosphere. The instrument was calibrated using indium
for heat flow and sapphire for heat capacity. Tg and Tm values
were calculated from the inflection point and the endothermic
maximum, respectively. IR spectra were measured using a Perkin-
Elmer FT-IR spectrometer. Powder X-ray diffraction was mea-
sured on a Bruker D8 Advance diffractometer with Cu KR
radiation, and the samples were scanned at step widths of 0.02ꢀ
with 1.0 s/step in the range of 5ꢀ-50ꢀ 2θ. Details of the single-
crystal X-ray diffraction are in the Supporting Information.
Representative Syntheses. Full experimental details and char-
acterization data can be found in the Supporting Information.
The following are representative syntheses.
derivitatization of 1p. The good solubility of the material
allowed two selected reactions, a deprotection and cross-link-
ing, to be attempted (Scheme 8).
The silyl group of 1p can be readily cleaved with TBAF
3
3H2O (TBAF = tetra(n-butyl)ammonium fluoride) in THF
(carefully avoiding excess fluoride which is capable of cleav-
ing the phosphazene backbone), and the process was mon-
itored by the disappearance of the silyl peak and the growth
of a broad acetylenic proton signal at 2.1 ppm by 1H NMR
spectroscopy. A fine white fibrous polymer (1r) was collected
after precipitations into distilled water and n-pentane. Com-
parison of the parent polymer with the deprotected material
in THF solution through DLS showed similar peaks for the
˚
˚
two species (Rh values of 12.5 A [1p] and 9.5 A [1r]) with the
expected contraction of the hydrodynamic radius for the
deprotected material and with no suggestion of degradation
of the polymer backbone.
For the cross-linking reaction, the protected polymer 1p was
found to undergo direct hydrosilylation with 5 mol % of the
bifunctional siloxane O(SiMe2H)2 in the presence of Karstedt’s
catalyst in toluene. A white precipitate formed rapidly on the
side of the reaction flask, and the cross-linked material was found
to be insoluble in all standard solvents. Further analysis was
carried out using IR spectroscopy; the relative intensity of
alkyne C-C stretch (2171 cm-1) to the starting material
weakened while a strong band in the alkene C-C stretching
region (1696-1670 cm-1) appeared, as expected. Clearly
many applications of the facile modification chemistry possi-
ble with materials such as 1p can be envisaged.
Summary
We have explored and expanded the scope of the promising
room-temperature polycondensation of N-silylbromophosphora-
nimines (4) in the presence of P(OMe)3 by synthesizing a number
of previously known and new polyphosphazenes, often in near-
quantitative yields. Both poly(alkyl/arylphosphazene) homopoly-
mers (1) and copolymers (7) are accessible with this methodology.
Notably, the soluble polymers such as 1n and 1p containing side-
group unsaturation were made efficiently as a result of the mild
conditions. The latter species was shown to be capable of further
reactivity, as illustrated by deprotection and cross-linking experi-
ments. The potential for this new and mild polymerization route
will to a large extent be dependent on the feasibility of preparing
the N-silylbromophosphoranimines monomers, although some
electron-withdrawing and/or bulky groups (alkoxy, phenyl) do
appear to hinder the phosphite-mediated polycondensation. In
addition, chlorophosphoranimines appear to polymerize more
slowly than their bromo analogues.
a. Polyphosphazene Homopolymer: [nHex(H2CdCHCH2)-
PdN]n (1n). A 1 mL CDCl3 solution containing 1.24 g (10
mmol) of P(OMe)3 was added to a rapidly stirring 15 mL CHCl3
solution of 4n (3.2 g, 10 mmol). The resulting solution became
very viscous after 18 h. All volatiles were removed from the
polymer syrup that yielded a white solid. The polymer was
redissolved in CHCl3 and then added dropwise into 200 mL of
n-pentane. The resultant white polymer was filtered and dried
under vacuum for 24 h (yield: 1.63 g, 95%). 31P{1H} NMR
We are currently investigating the mechanism of the phos-
phite-mediated phosphoranimine polycondensation reaction in
detail. We have recently prepared P-donor-stabilized phosphor-
animine cations that may be related to the proposed propagating
site.19 Detailed understanding of the mechanism may allow

7452 Macromolecules, Vol. 43, No. 18, 2010
Taylor et al.
(CDCl3): δ=19.8 ppm. 1H NMR (CD2Cl2): δ=0.80 (t, 3JHH
=
C. H.; St. John, J. V.; Wisian-Neilson, P. J. Am. Chem. Soc. 2001, 123,
3846–3847. (f) Kmecko, T.; Wang, X.; Wisian-Neilson, P. J. Inorg.
Organomet. Polym. Mater. 2007, 17, 413–421.
6.2 Hz, CH3); 1.20-1.27 (other CH2s); 1.48 (br, PCH2CH2);
1.79 (m, PCH2CH2); 2.75 (br, CH2CHdCH2); 5.10 (br, CH2-
CHdCH2); and 5.81 ppm (br, CH2CHdCH2). Mw =224 000
(PDI=3.32).
(5) Zhang, Y.; Huynh, K.; Manners, I.; Reed, C. A. Chem. Commun.
2008, 494–496.
(6) (a) Honeyman, C. H.; Manners, I.; Morrissey, C. T.; Allcock, H. R.
J. Am. Chem. Soc. 1995, 117, 7035–7036. (b) Allcock, H. R.; Reeves,
S. D.; Nelson, J. M.; Crane, C. A.; Manners, I. Macromolecules 1997
30, 2213–2215. (c) Wang, B. Macromolecules 2005, 38, 643–645.
(d) Allcock, H. R.; Reeves, S. D.; de Denus, C. R.; Crane, C. A.
Macromolecules 2001, 34, 748–754. (e) Wang, B.; Rivard, E.; Manners,
I. Inorg. Chem. 2002, 41, 1690–1691. (f) Krogman, N. R.; Steely, L. B.;
Hindenlang, M. D.; Nair, L. S.; Laurencin, C. T.; Allcock, H. R. Macro-
molecules 2008, 41, 1126–1130. (g) Blackstone, V. A.; Lough, A. J.;
Murray, M.; Manners, I. J. Am. Chem. Soc. 2009, 131, 3658–3667.
(7) Allcock, H. R.; Nelson, J. M.; Reeves, S. D.; Honeyman, C. H.;
Manners, I. Macromolecules 1997, 30, 50–56.
(8) For a preliminary communication, see: Huynh, K.; Lough, A. J.;
Manners, I. J. Am. Chem. Soc. 2006, 128, 14002–14003.
(9) For a preliminary report that a copolymer containing substituted
vinyl groups (-CHdCHR) could be made directly from (p-
trifluoroethoxy)(N-silyl)phosphoranimines without cross-linking,
see ref 3 and Neilson, R. H.; Hani, R.; Scheide, G. M.; Wettermark,
U. G.; Wisian-Neilson, P.; Ford, R. R.; Roy, A. K. ACS Symp. Ser.
1988, 360, 283–289.
(10) (a) Wilburn, J. C.; Neilson, R. H. Inorg. Chem. 1978, 18, 347–351.
(b) Neilson, R. H.; Wisian-Neilson, P. Inorg. Chem. 1982, 21, 3568–
3569.
(11) Diemert, K.; Kottwitz, B.; Kuchen, W. Phosphorus Sulfur Silicon
Relat. Elem. 1986, 26, 307–320.
(12) De Jong, R. L. P.; Brandsma, L. J. Organomet. Chem. 1986, 312,
277–282.
(13) Details of the refinement are provided in the Supporting Informa-
tion.
(14) Wang, H.; Wang, H.; Li, H.; Xie, H. Organometallics 2004, 23, 875–
885.
b. Polyphosphazene Random Copolymer: {[PhMePdN]0.67
-
[PhnBuPdN]0.33}n (7a). In the glovebox, 0.30 g (1.0 mmol) of 4d
and 0.10 g (0.3 mmol) of 4e were dissolved in 3 mL of CH2Cl2,
and 0.15 mL (1.3 mmol) of P(OMe)3 was added. The reaction
was stirred at room temperature overnight. The viscous solution
was precipitated into hexanes, and the white solid obtained was
purified by two more precipitations (CH2Cl2/hexanes) to afford
a white fibrous material. The polymer was dried at 40 ꢀC for
2 days (yield: 0.185 g, 92%). 31P{1H} NMR (CD2Cl2): δ=10.3
ppm (br, 7-17 ppm). 1H NMR (CD2Cl2): δ = 0.6 (br, CH3), 1.1
(br, CH2), 1.3-1.5 (PCH3), 1.6-2.1 (PCH2CH2), and 7.1-7.7
ppm (PC6H5). Mn=147 600 (PDI = 2.0).
Acknowledgment. We thank AWE for financial support.
A.P.S. thanks the EU for a Marie Curie postdoctoral fellowship,
and I.M. thanks the EU for a Marie Curie Chair.
Supporting Information Available: Full Experimental de-
tails and TGA, DSC, WAXS, and single crystal X-ray diffrac-
tion data. This material is available free of charge via the
Internet at http://pubs.acs.org.
References and Notes
(1) (a) Allcock, H. R.; Kugel, R. L. J. Am. Chem. Soc. 1965, 87, 4216–
4217. (b) Allcock, H. R. Adv. Mater. 1994, 6, 106–115. (c) Chemistry
and Applications of Polyphosphazenes; Allcock, H. R., Ed.; Wiley-
Interscience: Hoboken, NJ, 2003. (d) Mark, J. E.; Allcock, H. R.; West,
R. Inorganic Polymers, 2nd ed.; Oxford University Press: New York,
2005.(e) Cho, S.-Y.; Allcock, H. R. Chem. Mater. 2007, 19, 6338–
6344. (f) Phosphazenes: A World Wide Insight; Gleria, M., De Jaeger,
R., Eds.; Nova Science: Hauppauge, NY, 2004.(g) Mujumdar, A. N.;
Young, S. G.; Merker, R. L.; Magill, J. H. Macromolecules 1990, 23,
14–21. (h) D'Halliun, G.; De Jaeger, R.; Chambrette, J. P.; Potin, P.
Macromolecules 1992, 25, 1254–1258. (i) Carriedo, G. A.; Garcia-
Alonso, F. J.; Gomez-Elipe, P.; Ignacio Fidalgo, J.; Garcia-Alvarez,
J. L.; Presa-Soto, A. Chem.;Eur. J. 2003, 9, 3833–3836.
(15) Neilson, R. H.; Jinkerson, D. L.; Kucera, W. R.; Longlet, J.;
Samual, R. C.; Wood, C. E. In Inorganic and Organometallic
Polymers II; Wisian-Neilson, P., Allcock, H. R., Wynne, K. J., Eds.;
ACS Symposium Series 572; American Chemical Society: Washington,
DC, 1994; Chapter 18, pp 232-245.
(16) [PhMePN]n (1d) prepared from ClPhMePdNSiMe3 was shown to
possess Mw = 1.14 ꢀ 106 g/mol and a PDI = 1.88.
€
(17) Corradi, E.; Farina, A.; Gallazzi, M. C.; Bruckner, S.; Meille, S. V.
(2) (a) Montague, R. A.; Matyjaszewski, K. J. Am. Chem. Soc. 1990,
112, 6721–6723. (b) Ding, J. H.; Wang, L.; Yu, H. J.; Yang, Q. A.; Deng,
L. B. Des. Monomers Polym. 2008, 11, 215–222. (c) Couffin-Hoarau,
A. C.; Leroux, J. C. Biomacromolecules 2004, 5, 2082–2087.
(3) (a) Wisian-Neilson, P.; Neilson, R. H. J. Am. Chem. Soc. 1980, 102,
2848–2849. (b) Neilson, R. H.; Wisian-Neilson, P. Chem. Rev. 1988,
88, 541–562.
(4) (a) Wisian-Neilson, P.; Schaefer, M. A. Macromolecules 1989, 22,
2003–2007. (b) Wisian-Neilson, P.; Safiqul Islam, M. Macromolecules
1989, 22, 2026–2028. (c) Wisian-Neilson, P.; Garcia-Alonso, F. J.
Macromolecules 1993, 26, 7156–7160. (d) Wisian-Neilson, P.; Zhang,
C.; Koch, K. A. Macromolecules 1998, 31, 1808–1813. (e) Walker,
Polymer 1999, 40, 4473–4481.
(18) Neilson, R. H.; Hani, R.; Wisian-Neilson, P.; Meister, J. J.; Roy,
A. K.; Hagnauer, G. L. Macromolecules 1987, 20, 910–916.
(19) Huynh, K.; Lough, A. J.; Forgeron, M. A. M.; Bendle, M.; Presa
Soto, A.; Wasylishen, R. E.; Manners, I. J. Am. Chem. Soc. 2009,
131, 7905–7916.
(20) Wisian-Neilson, P.; Neilson, R. H. Inorg. Chem. 1980, 19, 1875–1878.
(21) Ford, R. R.; Goodman, M. A.; Neilson, R. H.; Roy, A. K.;
Wettermark, U. G.; Wisian-Neilson, P. Inorg. Chem. 1984, 23,
2063–2068.
(22) Roy, A. K.; Hani, R.; Neilson, R. H.; Wisian-Neilson, P. Organo-
metallics 1987, 6, 378–384.