Ambient temperature ring-opening polymerisation (ROP) of cyclic chlorophosphazene trimer [N3P3Cl6] catalyzed by silylium ions

Article abstract of DOI:10.1039/b713933k

The temperature required for ring-opening polymerisation of cyclo-N ₃P₃Cl₆ can be dramatically lowered by employing trialkylsilylium carboranes [R₃Si(CHB₁₁X₁₁] as catalysts. The Royal Socie

Full text of DOI:10.1039/b713933k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Ambient temperature ring-opening polymerisation (ROP) of cyclic
chlorophosphazene trimer [N P Cl ] catalyzed by silylium ions
3
3
6
a
Yun Zhang, Keith Huynh, Ian Manners* and Christopher A. Reed*
b
bc
a
Received (in Berkeley, CA, USA) 12th September 2007, Accepted 21st November 2007
First published as an Advance Article on the web 11th December 2007
DOI: 10.1039/b713933k
The temperature required for ring-opening polymerisation of
cyclo-N Cl can be dramatically lowered by employing
trialkylsilylium carboranes [R Si(CHB X ] as catalysts.
to the mechanism of catalysis, but the generally accepted pathway
for the ROP of 2 involves initial formation of the cationic
3
P
3
6
+
+
3 3 5
intermediate [N P Cl ] ([3] ) via a thermally-induced or Lewis
3
11 11
acid-assisted loss of chloride ion, followed by electrophilic attack
+
of [3] on 2 (Scheme 1). There is no direct evidence for the
1
The chemistry of poly(dichlorophosphazene) [–PCl
2
LN–]
n
(1) has
played an integral role in the development of inorganic polymer
science over the past four decades, due to the inherent tunability of
the polymer’s properties via substitution of chloride by oxygen and
existence of the proposed coordinatively-unsaturated cationic
+
intermediate [3] , although a related cation has been implicated
5
in the PCl -catalyzed polymerisation of Cl PLNSiMe .
5
3
3
1
nitrogen nucleophiles at the phosphorus centre. The carbon-free
Reasoning that better routes to coordinatively-unsaturated
+
cation [3] would lead to better ROP catalysis of 2, we explored
backbone of these inorganic polymers provides a versatile scaffold
for widespread materials applications, in areas such as high
the reaction of 2 with various strong Lewis and Brønsted acids that
were potentially capable of producing cations via chloride ion
11
2
performance elastomers, polymer electrolytes and biomedical
3
4
membranes.
abstraction. However, consistent with studies involving neutral
1
2
+
+
Lewis acids, cationic electrophiles such as H , Ag , CH and
+
3
Due to substituent versatility, a wide variety of synthetic routes
to polyphosphazenes have been explored. For example, an
ambient temperature, living, cationic chain-growth polycondensa-
+
3
R Si formed adducts with an N atom of 2 rather than abstracting
a chloride ion from P, regardless of how weakly coordinating is the
+
5
11,13–15
accompanying anion.
+
+
tion approach to 1 has been developed in recent years. It involves
Nevertheless, H , CH and R Si
3 3
the PCl -catalyzed polymerisation of the N-silylphosphoranimine
5
adducts of 2 with carborane anions were newly isolated and
11
monomer Cl
3
PLNSiMe
3
, and gives high yields of 1 with
characterized chlorophosphazene cations, so we decided to
investigate their behaviour as ROP catalysts. Specifically, we
explored the catalytic potential of six salts, 4–9, shown in Fig. 1.
An important feature of these salts is the exceptional inertness of
moderately high molecular weight and narrow molecular weight
distributions. Substantial molecular weight control is possible by
5
adjustment of the catalyst (PCl ) to monomer ratio. Nevertheless,
1
6
the polyphosphazene field remains heavily reliant on the synthesis
6
of 1 via classical thermal ring-opening polymerization (ROP) of
their carborane anions. Without weakly coordinating counter-
ions, these cations cannot be prepared because of the low Lewis
17
7
the cyclic phosphazene trimer [N P Cl ] (2) in a melt at ca. 250 uC.
and Brønsted basicity of 2 (pK = 26 in nitrobenzene).
3
3
6
a
Although the thermal ROP of 2 provides access to many
derivatives of high molecular weight polyphosphazenes, this route
suffers from a number of drawbacks. High costs are associated
with the synthesis of 2, a high level of monomer purity is required
to achieve reproducible polymerization and a high temperature is
required to induce ROP. In the melt, only moderate yields of
soluble polymer can be obtained (ca. 40%) before crosslinking
reactions take place. Even when ROP is performed in solution, no
appreciable molecular weight control is possible because high
temperatures remain necessary. Broad molecular weight distribu-
Treatment of 2 with 10 mol% of either the N-protonated
phosphazene 4 or the N-methylated phosphazene 5 in 1,2-
dichlorobenzene did not trigger ROP at either room temperature
or when heated at 160 uC for 7 d. However, when 2 was treated
with 10 mol% of the N-silylated derivatives 6–9 (prepared in situ)
in 1,2-dichlorobenzene at 25 uC,{ complete conversion of
31
7
[
N P Cl ] (2) [d( P) = 20] to polydichlorophosphazene [NPCl ]
3 3 6 2 n
3
1
7
1) [d( P) = 218] was achieved at room temperature, as indicated
(
8
tions are typical.
The ROP of 2 can be catalysed by certain Lewis acids, such as
9
10
3
,
BCl
3
and AlCl
but the reduction in process temperature is
modest (to ca. 200 uC) and there are reproducibility issues. The
lack of generality to a wide range of Lewis acids lends uncertainty
a
Center for S and P Block Chemistry, University of California,
Riverside, 92521, USA. E-mail: chris.reed@ucr.edu;
Fax: +1 951 827 2027; Tel: +1 951 827 5197
Department of Chemistry, University of Toronto, 80 St. George Street,
b
Toronto, Canada M5S 3H6
School of Chemistry, University of Bristol, Cantock’s Close, Bristol,
c
UK BS8 1TS. E-mail: ian.manners@bristol.ac.uk;
Fax: +44 (0)117 929 0509; Tel: +44 (0)117 928 7650
Scheme 1
4
94 | Chem. Commun., 2008, 494–496
This journal is ß The Royal Society of Chemistry 2008

View Article Online
Scheme 2
stable poly(bistrifluoroethoxy)phosphazene [NP(OCH
10) as a white fibrous solid in 86% re-precipitated yield
Scheme 2). Gel permeation chromatography indicated a weight
2 3 2 n
CF ) ]
(
(
5
21
average molecular weight (M
polydispersity index (PDI) of 1.83 vs. polystyrene standards. The
value is encouragingly high and the polydispersity value
promisingly narrow.
w
) of 1.12 6 10 g mol and a
M
w
Fig. 1 The protonated, methylated and four silylated phosphazene
cations, together with the carborane counterions used in this study.
+
+
The finding that cations 6 –9 are potent catalysts for the
polymerization of 2 refocuses attention on the mechanism of ROP.
It is possible that trialkylsilylation on an N atom provides an
unexpectedly low energy pathway for the generation of the key
3
1
1
by P{ H} NMR spectroscopy. A summary of the polymerisation
conditions surveyed in this study is shown in Table 1.
Of the four silyl derivatives 6–9, compound 8 was the easiest to
prepare. Catalysts with hexabromo carborane anions (8 and 9)
were more active than those with the undecachloro carborane
anion (6 and 7), even though the latter is considered less
coordinating. Treatment of a 2 M 1,2-dichlorobenzene solution
of 2 with 10 mol% of 8 resulted in the complete conversion to
polymer 1 in just 90 min. Shorter reaction times were observed
with higher catalyst loadings. Higher concentrations of 2 led to the
formation of gels. Lower concentrations of 2 curtailed ROP,
probably in part because of the sensitivity of 6–9 to moisture and
errant nucleophiles in the solvent, and part because of the
thermodynamics of ROP. Collectively, these observations are
consistent with the energetics of ROP as recently deduced for the
+
coordinatively-unsaturated intermediate [3] via an intramolecular
3 3
elimination reaction of R SiCl. Indeed, Et SiCl is a major
byproduct of the thermal decomposition of 8 in the solid state
and in solution. An alternative ROP mechanism worthy of
consideration involves an electrophilically-assisted associative
chain propagation step. With soluble components and ambient
temperature catalysis, kinetic experiments to distinguish between
the different possible ROP mechanisms are now feasible. In
addition, the energetics of alternate mechanisms can be explored
computionally. A deeper insight into the ROP mechanism of
phosphazenes can be anticipated from the present findings, as well
as better control over polymer properties.
1
8
cyclic thionylphosphazene [NSOCl(NPCl
2
)
2
] , and are character-
This work was supported by a Creativity Extension Award from
the National Science Foundation (CHE 0349878 to CAR). K. H.
thanks the NSERC for a Postgraduate Fellowship (2005–2008).
I. M. acknowledges the support of a Marie Curie Chair from the
European Union and a Royal Society Research Merit Award at
Bristol. We also thank Vivienne Blackstone for GPC measure-
ments at Bristol.
istic of the ROP of cyclic monomers with low degrees of ring
strain. A weakly coordinating anion is crucial for successful
catalysis. The presence of triflate anion quenches catalysis by 8,
presumably via desilylation of 6–9 to form R
3
Si(OTf) (trialkylsilyl
triflates do not silylate 2).
In order to determine the molecular weight of a representative
polymer obtained from an ambient temperature ROP reaction, a
standard P–Cl bond substitution reaction was carried out.
Treatment of a sample of polymer 1, obtained from a 1 M 1,2-
dichlorobenzene solution of monomer 2 and 10 mol% catalyst 8,
with sodium trifluoroethoxide in dioxane gave the hydrolytically
Notes and references
{
Typical ROP reaction of [N
Et Si(CHB11 Br ) (149 mg, 0.20 mmol) was combined with [N
702 mg, 2.0 mmol) and then charged with 2.04 mL of dry 1,2-
dichlorobenzene at room temperature (1 M solution). The solution was
3
P
3
Cl
6
] (2). In an inert atmosphere glovebox,
3
H
5
6
3
P
3
Cl
6
]
(
31
1
stirred magnetically and the formation of polymer 1 monitored by P{ H}
NMR spectroscopy at 122 MHz (d = 217.7) after 2 (0% conversion), 30
(3% conversion), 60 (35% conversion), 90 (75% conversion), 120 (95%
conversion) and 150 (100% conversion) minutes.
Table 1 Conditions of polymerisation
Conc. of 2 in
1,2-dichlorobenzene/M
Mol% of
Catalyst
Reaction
time/h
Catalyst
Preparation of [NP(OCH
polymerization mixture was diluted with 50 mL of dioxane and cooled to
12 uC. An excess of NaOCH CF (freshly prepared from Na and
CF CH OH in dioxane) was added dropwise to the cooled polymer
2 3 2 n
CF ) ] (10). In a typical experiment, the
0.3
0.5
1.0
1.5
2.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
8
8
8
8
8
8
8
8
7
7
9
9
6
10
10
10
10
10
5
20
33
10
2
36
5
2.5
2
1.5
48
,1
,1
48
144
2
2
3
3
2
solution. The resulting white suspension was stirred at 100 uC for 2 h, and
then at room temperature overnight. The volatiles were removed from the
reaction mixture, which was then cooled to 0 uC and quenched with
copious amounts of water. The crude polymer was filtered off, dried,
redissolved in a minimal amount of dioxane and reprecipitated with hexane
3
1
1
in 84% yield. P{ H} NMR (122 MHz, d, CD
= 61 000, M = 112 000, PDI = 1.83.
3
CN, 25 uC): 26.9 (s). GPC:
M
n
w
10
5
10
24
72
1
H. R. Allcock, Chemistry and Applications of Polyphosphazenes, John
Wiley & Sons, New York, 2003.
This journal is ß The Royal Society of Chemistry 2008
Chem. Commun., 2008, 494–496 | 495

View Article Online
2
3
(a) D. P. Tate, J. Polym. Sci., Polym. Symp., 1974, 48, 33; (b)
R. E. Singler, N. S. Schneider and G. L. Hagnauter, Polym. Eng. Sci.,
7 (a) R. L. Kugel and H. R. Allcock, J. Am. Chem. Soc., 1965, 87, 4216;
(b) H. R. Allcock, R. L. Kugel and K. J. Valan, Inorg. Chem., 1966, 5,
1709.
8 (a) A. N. Mujumdar, S. G. Young, R. L. Merker and J. H. Magill,
Macromolecules, 1990, 23, 14; (b) G. D’Halluin, R. De Jaeger,
J. P. Chambrette and P. Potin, Macromolecules, 1992, 25, 1254.
9 M. S. Sennett, G. L. Hagnauer, R. E. Singler and G. Davies,
Macromolecules, 1986, 19, 959.
10 Y. S. Sohn, Y. H. Cho, H. Baek and O.-S. Jung, Macromolecules, 1995,
28, 2566.
11 Y. Zhang, F. S. Tham and C. A. Reed, Inorg. Chem., 2006, 45,
10446.
1975, 15, 321; (c) G. S. Kyker and T. A. Antkowiak, Rubber Chem.
Technol., 1974, 47, 32; (d) J. K. Valaitis and G. S. Kyker, J. Appl. Polym.
Sci., 1979, 23, 765.
(a) P. M. Blonsky, D. F. Shriver, P. Austin and H. R. Allcock, J. Am.
Chem. Soc., 1984, 106, 68854; (b) H. R. Allcock, C. J. Nelson,
W. D. Coggio, I. Manners, W. J. Koros, D. R. B. Walker and
L. A. Pessan, Macromolecules, 1993, 26, 1493; (c) J. Paulsdorf,
M. Burjanadze, K. Hagelschur and H.-D. Wiemh o¨ fer, Solid State
Ionics, 2004, 169, 25; (d) H. R. Allcock and R. M. Wood, J. Polym. Sci.,
Part B: Polym. Phys., 2006, 44, 2358.
4
5
(a) H. R. Allcock, M. Gebura, S. Kwon and T. X. Neenan,
Biomaterials, 1988, 19, 500; (b) H. R. Allcock and A. M. A. Ambrosio,
Biomaterials, 1996, 17, 2295; (c) A. Singh, N. R. Krogman,
S. Sethurman, L. S. Nair, J. L. Sturgeon, P. W. Brown,
C. T. Laurencin and H. R. Allcock, Biomacromolecules, 2006, 7, 914.
(a) C. H. Honeyman, I. Manners, C. T. Morrissey and H. R. Allcock,
J. Am. Chem. Soc., 1995, 117, 7035; (b) H. R. Allcock, J. M. Nelson,
S. D. Reeves, C. H. Honeyman and I. Manners, Macromolecules, 1997,
12 (a) A. L. Macdonald and J. Trotter, Can. J. Chem., 1974, 52, 734; (b)
A. J. Heston, M. J. Panzner, W. J. Youngs and C. A. Tessier, Inorg.
Chem., 2005, 44, 6518.
13 H. Bode, K. Butow and G. Lienau, Chem. Ber., 1948, 81, 547.
14 A. J. Heston, M. Panzner, W. J. Youngs and C. A. Tessier, Phosphorus,
Sulfur Silicon Relat. Elem., 2004, 179, 831.
15 M. Gonsior, S. Antonijevic and I. Krossing, Chem.–Eur. J., 2006, 12,
1997.
3
0, 50; (c) H. R. Allcock, S. D. Reeves, C. R. de Denus and C. A. Crane,
16 (a) C. A. Reed, Acc. Chem. Res., 1998, 31, 133; (b) C. A. Reed, Chem.
Commun., 2005, 1669.
17 D. Feakins, W. A. Last, N. Neemuchwala and R. A. Shaw, J. Chem.
Soc., 1965, 2804.
Macromolecules, 2001, 34, 748; (d) B. Wang, E. Rivard and I. Manners,
Inorg. Chem., 2002, 41, 1690; (e) H. R. Allcock, E. S. Powell,
A. E. Maher, R. L. Prange and C. R. de Denus, Macromolecules, 2004,
3
4
7, 3635; (f) E. Rivard, A. J. Lough and I. Manners, Inorg. Chem., 2004,
3, 2765.
H. N. Stokes, Am. Chem. J., 1897, 19, 782.
18 A. R. McWilliams, D. P. Gates, M. Edwards, L. M. Liable-Sands,
I. Guzei, A. L. Rheingold and I. Manners, J. Am. Chem. Soc., 2000,
122, 8848.
6
4
96 | Chem. Commun., 2008, 494–496
This journal is ß The Royal Society of Chemistry 2008