Organophosphazenes. 26. Factors controlling the pathways observed in the reactions of ethynyl lithium reagents with hexafluorocyclotriphosphazene

Article abstract of DOI:10.1021/ic300481e

In contrast to previously studied reactions of ethynyl lithium reagents, the reactions of propynyl and hexynyl lithium with N₃P ₃F₆ lead to predominantly nongeminal isomers. A modest cis stereo selectivity was observed. Th

Full text of DOI:10.1021/ic300481e

Article
pubs.acs.org/IC
Organophosphazenes. 26. Factors Controlling the Pathways
Observed in the Reactions of Ethynyl Lithium Reagents with
Hexafluorocyclotriphosphazene
Maneesh Bahadur and Christopher W. Allen*
Department of Chemistry, University of Vermont, Burlington, Vermont 05405-0125, United States
ABSTRACT: In contrast to previously studied reactions of ethynyl lithium reagents, the
reactions of propynyl and hexynyl lithium with N3P3F6 lead to predominantly nongeminal
isomers. A modest cis stereo selectivity was observed. The sequential addition of a lithio
acetylene reagent which follows a predominately geminal pathway (lithiophenylacetylene)
and an aryl lithium reagent (p-propenylphenyl lithium) which follows a predominantly
nongeminal pathway were examined. The relative order of addition of the two reagents was
interchanged, resulting in a change of reaction pathway demonstrating ring substituent
control of the regio and stereo chemical pathways. Hydrogenation of the ethynyl unit in the
phosphazene derivatives provides a facile pathway to the difficult to prepare
alkylphosphazenes. The chemical shift of the organosubstituted phosphorus center undergoes a remarkably large change
(46−47 ppm) on going from the ethynyl to the alkyl derivatives, which reduces the complex 31P and 19F NMR spectra of the
ethynyl derivatives to easily interpretable first-order spectra, thus allowing for structure assignment. The 13C NMR data shows
that nongeminal regio selectivity increases with the amount of s character on the ethynyl carbon atom attached to the
phosphorus center. These observations allow for an understanding of the factors controlling regio and stereo chemical control in
the reactions of carbanionic nucleophiles with N3P3F6.
observed reaction pathways of N3P3F6 and to provide
additional ethynylphosphazenes for reactions with organo-
metallic reagents, we have examined the reactions of
alkylethynyl lithium reagents and mixed substituent ethynyl
lithium reagents and aryl lithium reagents.
INTRODUCTION
■
Reasonable models to rationalize the factors controlling the
nucleophilic substitution reactions of amines2 and oxyanions2−5
with cyclophosphazenes are available. By way of contrast, a
general understanding of the reactions with carbanionic
reagents with cyclophosphazenes is not complete. Allcock has
shown that the reactions of Grignard reagents with
hexachlorocyclotriphosphazene, N3P3Cl6, proceed by a halogen
abstraction pathway, leading to metalated phosphazene
intermediates.6 A similar pathway is followed in the reactions
of organolithium reagents.6,7 The corresponding reactions with
hexafluorocyclotriphosphazene, N3P3F6, are more typical of the
nucleophilic substitutions referred to above.2,6,8 A geminal
pathway is preferred for certain alkyl,2,9,10 alkenyl,2,11 phenyl-
ethynyl,12,13 and trimethylsilylethynyl13 lithium reagents, while
nongeminal products are obtained for a variety of aryllithium
reagents2,14 and t-butyl lithium.10 With the exception of the
region and stereospecific t-butyl lithium case,10 the observed
pathways are all regioselective with nongeminal products
accompanying geminal products.13,14 The resulting organo-
functional phosphazenes have proved to be useful synthetic
intermediates. The alkenylphosphazenes undergo addition
polymerization at the exocyclic olefin site.15−17 The ethynyl
center in phenylethynyl phosphazenes can react with metal
carbonyls to give the appropriate organometallic deriva-
tives11,12,18−20 or undergo cyclooligomerization reactions.18−23
Recent detailed work by Elias and co-workers has shown how
these processes may be used to create phosphazenes with
structurally complex organic substituents.21−23 In order to
understand better the role of the carbanionic species in the
EXPERIMENTAL SECTION
■
Materials. Hexachlorocyclotriphosphazene, N3P3Cl6 (Nippon Soda
Co., Ltd.), was converted to N3P3F6 (1),24 which in turn was
converted to N3P3F5CCPh (8)13 and N3P3F5C6H5CMeCH2
(9)14 by previously reported procedures. n-Butyl lithium (2.5 M in
hexane, Aldrich), phenylacetylene (Aldrich), propyne (Farchan
Laboratories), and hexyne (Farchan Laboratories) were used without
further purification. All solvents were distilled from appropriate drying
agents prior to use. All organo lithium reagents were handled under an
inert atmosphere.
Characterization. Infrared spectra were recorded on a Perkin-
Elmer 1430 spectrophotometer. Mass spectrometry was carried out on
a Finnigan 4610 spectrometer operating at 70 eV. GC-MS studies
utilized a Finnegan GC equipped with a 30 m capillary column coated
with SE-30 (Restex). NMR spectra were recorded on a Bruker
WM250 spectrometer operating at 250.1 MHz (1H), 62.9 MHz (13C),
235.35 MHz (19F), and 101.2 MHz (13P) and using CDCl3 for a lock
compound. Tetramethylsilane (TMS) (1H and 13C) was used as an
internal reference, while 85% H3PO4 (31P) and CFCl3 (19F) were used
as external references. Chemical shifts upfield from the reference are
negative. Broad band 1H decoupling was used for the 13C, 19F, and 31
P
spectra. Elemental analyses were performed by Robertson Laboratory
Microlit Laboratories, Inc., Ledgewood, New Jersey.
Received: March 2, 2012
Published: April 12, 2012
© 2012 American Chemical Society
5465
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470

Inorganic Chemistry
Article
Preparation of N3P3F5CCMe (2). A 250 mL two-necked flask
equipped with magnetic stirrer and a dry ice condenser was flame-
dried and allowed to cool under an inert nitrogen atmosphere. The
apparatus was then evacuated and backfilled with nitrogen three times.
Dry ether (100 mL) was introduced into the flask via a syringe, and
the whole apparatus was immersed in a liquid N2−ethyl acetate slush
bath. After allowing the temperature to drop below −50 °C, excess
propyne was condensed in the reaction flask. A solution of 4.3 mL of
2.5 M n-butyl lithium (0.01 mol) was added dropwise via a syringe,
and the reaction mixture was stirred for 3 h. The temperature was then
raised to −20 °C, allowing the excess propyne to be removed. In a
separate 250 mL round-bottom flask, 2.5 g (0.01 mol) of P3N3F6 was
dissolved in 75 mL of Et2O, and the whole apparatus was cooled in an
ice bath. The lithiopropyne solution was then added via a cannula into
the hexafluorocyclotriphosphazene solution. The reaction mixture was
allowed to stir at room temperature for 12 h, during which time it
turned dark brown. The solvent was removed at room temperature,
and low boiling petroleum ether was added to precipitate the lithium
salts. After filtration through Celite, the crude product was sublimed,
yielding 1.0 g of white solid (37.1% yield) melting at 42−43 °C. Anal.
calcd. for N3P3F5C3H3: C, 13.35; H, 1.12; mol wt, 269. Found: C,
chromatographic analysis of the mixture showed that all three isomers
are present.
Preparation of N3P3F4(4-C6H4CMeCH2)CCPh, Method A
(6A). A 250 mL two-necked round-bottom flask was equipped with a
pressure equalizing addition funnel, magnetic stirrer, and nitrogen
bubbler. The whole apparatus was flame-dried and allowed to cool to
room temperature under positive pressure of nitrogen. The reaction
flask was charged with 100 mL of dry ether via cannula techniques, and
1.35 mL of phenylacetylene (12.0 mmol) was added via a syringe. The
whole apparatus was immersed in an ice bath, and 4.8 mL (12.0
mmol) of 2.5 M n-BuLi in hexane was added dropwise, during which
time the reaction mixture turned yellow. The reaction mixture was
stirred an additional 3 h after the addition was complete. A three-
necked 500 mL round-bottom flask, equipped with an oil bubbler and
magnetic stirrer, was flushed with nitrogen and charged with a solution
of 4. 2
g (12 mmol) of [4-(2-propenylphenyl)]-
pentafluorocyclophosphazene (9) in 75 mL of dry ether. The flask
was placed in an ice bath. The phenylethynyl lithium solution was
added dropwise at 0 °C. After the addition was complete, the ice bath
was removed, and the reaction mixture was allowed to stir overnight,
during which time it turned turbid brown. The ether was removed, and
low boiling petroleum ether was added to precipitate the lithium salts.
The salts were allowed to settle and removed by filtration through
diatomaceous earth. The petroleum ether was removed to yield a
yellowish oil, which was subjected to flash chromatography on silica
gel using petroleum ether−methylene chloride (80−20 v/v), to give
1.02 g (34.7% yield) of a mixture of isomers of [(propenyl-phenyl)][β-
phenylethynyl]tetrafluorocyclotriphosphazene, N3P3F4(4-
C6H4CMeCH2)(CCPh) (6A). A gas chromatographic analysis
in conjunction with 31P and 19F NMR revealed the isomer ratio to be
1.0:0.53:<0.10 for the cis, trans, and geminal isomers, respectively. The
isomers resisted further separation. Anal. calcd. for P3N3F4C17H14: mol
wt, 429. Found: mol wt, 429 (GC mass spectrum). 31P NMR (non
geminal cis and trans): 6.4 (t, JPF = 906, PF2); 5.2 (d, JPF = 896, 
PF[CCPh]); 33.2 (d, JPF = 974, PF(4-C6H4C(Me)CH2). 19F
1
13.51; H, 1.12; mol wt, 269 (mass spec). H NMR: 2.10 (multiplet
1
2
CH3). 13C NMR: 4.68 (s, CH3); 69.5 (d of d of t, JPC = 366, JFC
=
61.4, JPC = 15, CCCH3); 103.58 (d, JPC = 65.9, CCCH3). 31P
NMR: 4.68 (d, JPF = 862, PF[CCMe]); 9.12 (t, JPF = 888, 
PF2). 19F NMR: 73.4 (d, JPF = 861,  PF2[cis]; 72.1 (d, JPF = 902, 
PF2[trans]): 48.9 (d, JPF = 892,  PF[CCCH3]). IR: 2958, 2857
(w, νC−H), 2215 (s, νCC), 1271 (vs, νPN), 942 (s, νP−F (asym), 852 (s,
νP−F (sym)).
3
2
Preparation of N3P3F4(CCMe)2 (3). This procedure is
equivalent to that used for the preparation of 2 using the following
amounts of reagents: 2.5 g (0.01 mol) of P3N3F6, 8.6 mL (0.02 mol) of
2.5 M n-BuLi in hexane, and excess propyne gas. The crude reaction
mixture was sublimed to remove the monosubstituted product (0.56 g,
20.9% yield). The residue was mixed with petroleum ether, filtered to
remove insoluble materials, and distilled at reduced pressure to isolate
the product as the fraction boiling at 71−73 °C (1.23 g, 42.5% yield).
Mass spectrum: M+ ion, 289 (calcd., 289). A gas chromatographic
analysis of the mixture showed that all three isomers are present.
Preparation of N3P3F5CCC4H9 (4). A dry 250 mL flask
equipped with a magnetic stirrer was purged with nitrogen and
charged with 0.83 g (0.01 mol) of hexyne in 100 mL of dry ether at 0
°C. A solution of 4.0 mL of 2.5 M (0.01 mol) n-BuLi was added
dropwise via a syringe, and the solution was stirred for 5 h. The
lithiohexyne solution was added via cannula technique to a solution of
2.5 g (0.01 mol) of P3N3F6 in 50 mL of ether at 0 °C, and the reaction
was stirred overnight at room temperature, during which time it turned
brown. The solvent was removed under reduced pressure, and light
petroleum ether was added to precipitate the lithium salts. The
solution was filtered through Celite to remove the lithium salts. The
crude product was distilled under reduced pressure (boiling point 44−
45 0 °C at 0.02 mmHg) to yield 1.76 g (56% yield) of product. Anal.
Calcd for P3N3F5C6H9: C, 23.15; H, 2.89; mol wt, 311. Found: C,
23.18; H, 2.85, mol wt, 311 (mass spec). 31P NMR: 9.0 (t, JPF = 896,
PF2); 4.7 (d, JPF = 865.84, PF[CCC4H9]). 19F NMR: 49.0 (d,
JPF = 893, PF[CCC4H9]; 72.5 (d, JPF = 897, PF2[trans]), 74.0
NMR (non geminal cis): 69.8 (d, JPF = 918, PF2[cis]; 72.8 (d, JPF
=
918, PF2[trans]); 54.6 (d, JPF = 982, PF[4-C6H4C(Me)CH2]);
44.1 (d, JPF = 915, PFCCPh). 19F NMR (non geminal trans):
71.5 (d, JPF = 907; PF2); 56.1 (d, JPF = 982, PF[p-C6H4C(Me)
CH2]); 46.3 (d, JPF = 915 PF[CCPh]). 13C NMR: 21.5 (CH3);
103.3 (d, JPC =46.7; CCPh); 119.2 (CCPh); 115.4 (C(CH3)
CH2); 141.5 (C(CH3)CH2), 124−133 (overlapping resonances
from the two phenyl groups). IR: 2920, 2950, 3030, 3060 (m, νC−H),
2250 (s, νCC), 1633 (m, νCC), 1280 (vs, νPN), 920 (s, νP−F(asym)),
810 (s, νP−F(sym)).
Preparation of N3P3F4(4-C6H4CMeCH2)CCPh, Method B
(6B). In this reaction, 2-(4-lithophenyl)propene was allowed to react
with P3N3F5[CCPh]. The details of the synthetic procedure are
similar to the previous reaction using the following amounts of
reagents: 8.0 mL (20 mmol) of 2.5 M n-BuLi in hexane was added
dropwise to a solution of 3.8 g (20 mmol) of 4-bromophenylpropene
in 100 mL of ether at 0 °C. The yellow solution was stirred for 4 h at 0
°C and then added dropwise to a solution of 6.62 (20 mmol) of
P3N3F5[CCPh] in 50 mL of ether at 0 °C. The reaction was
monitored using TLC and required 18 h of stirring at room
temperature for completion. The workup was similar to that for
method A. After flash chromatography, 1.8 g (25% yield) of the
disubstituted material was obtained. A gas chromatographic analysis in
conjunction with 31P and 19F NMR revealed the geminal isomer to be
the major product, and only small amounts of nongeminal products
were observed. Anal. calcd. for N3P3F4C17H14: mol wt, 429. Found:
429 (GC mass spectrum). 31P NMR (geminal): 6.4 (t, JPF = 906, 
(d, JPF = 875, PF2[cis]). 13C NMR: 13.03 (s, CH3); 18.98 (s, CH2);
2
21.82 (s, CH2); 28.95 (s, CH2); 70.1 (d of d, of t, JPC = 371, JFC
=
3
60.4, Jpc = 14.7, CCC4H9); 107.6 (d, JPC = 65.8, CCC4H9). IR:
2953, 2869 (w, νC−H), 2207 (s, νCC), 1274 (s, νPN).
Preparation of N3P3F4(CCC4H9)2 (5). This procedure is
equivalent to that used for the preparation of 4 using the following
amounts of reagents: 1.64 g (0.02 mol) of hexyne in 100 mL of ether,
8.0 mL of 2.5 M n-BuLi solution (0.02 mol), and 2.5 g (0.01 mol) of
P3N3F6. After completion, the solvent was removed, and petroleum
ether was added to precipitate the lithium salts, which were removed
by filtration through Celite. The crude product was distilled at reduced
pressure (0.02 mmHg) to remove the monosubstituted derivative
(boiling point 44−45 0 °C, 0.47 g, 15.3% yield). The fraction boiling
at 91 −93 °C was the desired product (1.70 g, 45.6% yield). A gas
2
PF2); 4.68 (t, JPP = 61.7, P[4-C6H4CMe)CH2][CCPh]). 19F
NMR (geminal): 71.8 (d, JPF = 857, PF2). 13C NMR: 21.5 (CH3);
103.3 (d, JPC = 46.7; CCPh); 119.2 (CCPh); 115.4 (C(CH3)
CH2); 141.5 (C(CH3)CH2); 124−133 (overlapping resonances
from the two phenyl groups). IR: 2920, 2950, 3030, 3060 (m, νC−H),
2250 (s, νCC), 1635 (m, νCC), 1280 (vs, νP−N), 920 (m, νP−F(sym)).
Preparation of 2,4-N3P3(4-C6H4CH(CH3)2)(CH2CH2Ph) (7A). A
solution of 0.25 g (0.58 mmol.) of the cis/trans mixture of
5466
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470

Inorganic Chemistry
Article
Scheme 1. Reactions of Selected Ethynyl Lithium Reagents
N3P3F4C6H5 C(CH3)CH2)(CCPh) (6A) in 50 mL of methanol
was hydrogenated using palladium/charcoal as a catalyst. The reaction
flask was connected to a balloon filled with hydrogen gas. The flask
was evacuated and backfilled with hydrogen gas, and the procedure
was repeated three times to ensure complete exclusion of air in the
reaction flask. The reaction mixture was stirred at room temperature,
and the reaction was allowed to proceed until the uptake of hydrogen
gas had ceased. After completion, the reaction mixture was filtered
twice from Celite to remove the catalyst, and the solvent was removed
under reduced pressure. The crude product was subjected to
chromatography (petroleum ether as eluant) on activated silica
through a short path (∼10 cm) column to isolate 0.18 g (82%
3H); 1.58 (2H); 1.9 (2H). 13C NMR: 32.1 (m, 1JPC = 145, 2JPC = 16.5,
Cα); 25.7 (d, JPC = 6.0, Cβ); 18.5 (CH3).
Preparation of N3P3F4(C3H7)2 (9). A solution of 8.0 mg of
P3N3F4(CCCH3)2 (3) in 5 mL of methanol was hydrogenated using
the same procedure as above. No further purification was done. Mol
wt. calcd.: 297. Found: 297 (mass spec). 31P NMR (cis): 8.0 (d of t,
2
JPF = 899, JPF = 916, 2JPP = 49.0, PF2); 48.6 (d of d, JPF = 998, JPP
=
=
2
49.0, PF[C3H7]). 31P NMR (trans): 6.8 (d of t, JPF = 899, JPP
57.2, PF2); 48.6 (d of d, JPF = 998, JPP = 57.2, PF[C3H7]). 31P
2
2
NMR (gem): 10.1 (t, JPF = 891, PF2); 48.0 (t, JPP = 32.7, 
2
P(C3H7)2). 19F NMR (cis): 69.6 (d of d, JPF = 918, JFF = 75.0, 
PF2[cis]); 73.5 (d, JPF = 918, PF2[trans]); 60.3 (d, JPF = 1010, 
PF[C3H7]). 19F NMR (trans): 72.0 (d, JPF = 900, PF2); 62.1 (d, JPF
= 992, PF[C3H7]). 19F NMR (gem): 72.6 (d, JPF = 881, PF2).
Preparation of N3P3F5C6H13 (10). The hydrogenation reaction of
P3N3F5[CCC4H9] (4) was performed as above. No purification was
necessary. Mol wt. calcd.: 315. Found: 314 (mass spec). 31P NMR: 6.8
(t, JPF = 1,000, PF2); 52.3 (d, JPF = 898, PF[C6H13]). 19F NMR:
72.3 (d, JPF = 878, PF2[cis]); 71.8 (d, JPF = 925, PF21[trans]; 63.7
(d, JPF = 967, PF[C6H13]). 13C NMR: 30.1 (complex, JPC = 150.2,
yield) of the product. 31P NMR: 7.7 (d of d of d of d, JPF = 937, JPF
881, 2JPP = 59.8, 2JPP = 56.4, 3JPF = 5.6, PF2[cis]; 6.5 (t of t of t, JPF
=
=
2
2
3
903, JPP = 62.1, JPP = 63.5, JPF = 3.7, PF2[trans]); 32.4
(overlapping multiplets, JPF = 999, 2J = 56.5, PF[4-C6H4CH-
2
(CH3)2]); 47.0 (overlapping multiplets, JPF = 1005, JPP = 45.1, 
PF[CH2CH2Ph]). 19F NMR (non geminal cis): 69.4 (d of d, JPF = 917,
2JFF = 73.0, PF2[trans]; 73.1 (d of d, JPF = 893, JFF = 73, 
2
2
2JFC = 18.9, C1); 24.6 (d, JPC = 9.8, C2); 21.6, 20.5, 18.1, 12.6.
PF2[cis]); 53.2 (d, JPF = 974, PF[4-C6H4CH(CH3)2]); 60.1 (d, JPF
= 1006, PF[CH2CH2Ph]). 19F (non geminal trans): 71.3 (d, JPF
=
Preparation of N3P3F4(C6H13)2 (11). The hydrogenation of
P3N3F4(CCC4H9)2 (5) was carried out as described in the previous
sections. Mol. wt. calcd.: 381. Found: 380 (mass spec). 31P NMR (cis):
5.5 (d of d of d, JPF = 893, JPF = 917, 2JPP = 48.7, PF2); 45.7 (d of d,
JPF = 1003, 2Jpp = 48.7, PF(C6H13). 31P NMR (trans): 5.2 (d of d of
909, PF2); 56.1 (d, JPF = 974, PF[4-C6H4CH(CH3)2]); 61.9 (d,
1
JPF = 1006, PF[CH2CH2Ph]). H NMR (mixture of isomers): 1.28
(d, JHH = 3.85, CH(CH3(2); 2.3 (complex multiplets, CH(CH3)2);
2.8−3.1 (overlapping multiplets, CH2CH2Ph); 7.1−7.9 (complex
multiplets, phenyl groups).
2
3
t, JPF = 893, JPP = 56.8, JPP = 6.5, PF2); 45.7 (d of d, JPF = 1003,
2JPP = 56.8, PF[C6H13]). 31P NMR (gem): 5.7 (d, JPF = 909, 
PF2); 45.7 (t, 2JPP = 34.1, P[C6H13]2). 19F NMR: 69.6 (d, JPF = 927,
PF2[cis]); 73.4 (d, JPF = 898, PF2[trans]; 60.5 (d, JPF = 1004, 
PF[C6H13]). 19F NMR: 72.5 (d, JPF = 888, PF2); 62.4 (d, JPF = 994,
PF[C6H13]). 19F NMR (gem): 71.8 (d, JPF = 908, PF2).
Preparation of (2,2-N3P3F4[4-C6H4CH(CH3)2](CH2CH2Ph) (7B).
A solution of 0.15 g (0.35 mmol) of geminal N3P3F4(4-C6H4C-
(CH3)CH2)(CCPh)(6B) was hydrogenated in 50 mL of
methanol using palladium/charcoal as a catalyst. The procedure for
the reaction was similar to that in the previous section. After work up
and chromatography, 0.11 g (73% yield) of the product was isolated.
2
31P NMR: 9.1 (t, JPF = 891, PF2); 36.7 (t, JPP= 41 0.5, P[4-
RESULTS AND DISCUSSION
■
C6H4CH(CH3)2][CH2CH2Ph]). 19F NMR: 71.7 (d of d, JPF = 882, 
PF2). 1H NMR: 1.28 (d, JHH = 3.85, CH(CH3)2); 2.3 (complex
multiplets, CH(CH3)2); 2.8−3.1 (overlapping multiplets,
CH2CH2Ph); 7.1−7.9 (complex multiplets, phenyl groups). 13C
NMR: 23.59 (CH(CH3)2); 27.24 and 27.28 (CH2CH2); 34.29
((CH(CH3)2); 126−132 (phenyl groups).
The reactions of N3P3F6 (1) with various lithioacetylene
reagents were carried out to produce the mono- and
disubstituted derivatives. In general, it was found that the
substituted ethynyl derivatives (2−5) could be synthesized in
moderate yields (40−60%). In the reactions of 1 with two
equivalents of the lithioacetylene reagents, significant amounts
of the monosubstituted product were obtained. Additionally,
the sequential addition of a lithioacetylene reagent which
follows a predominately geminal pathway (lithiophenylacety-
lene)13 and an aryl lithium reagent (p-propenylphenyl lithium)
which follows a predominately nongeminal pathway14 were
examined. The relative order of addition of the two reagents
was interchanged, resulting in a change of reaction pathway. We
Preparation of N3P3F5C3H7 (8). A solution of 0.02 g of
P3N3F5[CCMe] (2) in 15 mL of methanol was hydrogenated
using the same procedure as above. No further purification was
required as evidenced by the TLC experiments on silica gel using
petroleum ether and GC. Mol wt. calcd.: 273. Found: 273 (mass spec).
31P NMR: 10.0 (t, JPF = 991, PF2); 50.3 (d, JPF = 899, PF[C3H7]).
19F NMR: 73.7 (d, JPF = 901, PF2 [cis]); 71.2 (d, JPF = 915, 
PF2[trans]]); 65.3 (d, JPF = 977, PF[C3H7]). 1H NMR: 0.98 (broad,
5467
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470

Inorganic Chemistry
Article
have previously shown that one of the resulting mixed
substituent derivatives, 6A, can be transformed into hybrid
organic−inorganic polymers.20 The results of all of these
reactions are summarized in Scheme 1. The disubstituted
ethynylphosphazenes (3, 5) were mixtures of the geminal and
nongeminal isomers which resisted further bulk scale
separation. All of the bis derivatives showed three peaks in
the GC-MS with the same parent ion peaks, indicating that all
of the isomers were present. Previous work has shown that the
31P and 19F NMR spectra can be used for definitive structural
assignment for fluorophosphazene derivatives.1,4,10,11,14 How-
ever, in the case of the ethynylphosphazenes, the spectra were
complicated by second order effects and resonance overlap due
to the small difference in chemical shifts between the various
phosphorus centers to such a degree that it is difficult to make
more than a rough estimate of the major NMR parameters and
impossible to assign the various isomers in the disubstituted
derivatives. This problem was overcome by hydrogenation of
the ethynyl center (and also the olefin in 6A and 6B) in the
ethynyl phosphazenes (Scheme 2). The chemical shift of the
organosubstituted phosphorus center undergoes a remarkably
large change (46−47 ppm) on going from the ethynyl to the
alkyl derivatives, while the PF2 resonances only undergo
minimal change. Similar changes were observed in the 19F
NMR spectra. The changes are sufficient to transform the
complex spectra to readily interpretable ones, as demonstrated
in Figure 1. There are two possibilities for the origins of the
change in phosphorus chemical shifts in the organo-
fluorocyclotriphosphazenes. The observed trend shown in
Figure 2 is a linear decrease in the phosphorus chemical shift
with increased s character (hence, increased orbital electro-
negativity) of the α carbon atom of the organic substituent.
This is the direction one would predict if the shift were
controlled by the local diamagnetic term (σd). However, σd
control is not expected in heavier nuclei. Alternatively, the
increased number of low lying excited states available with the
unsaturated organic substituents allows for larger local
paramagnetic term (σp) contributions as the cause of the
trend in the phosphorus shifts. It is important to note that all
evidence available argues against specific phosphorus−carbon
multiple bonds,25−27 but the global MOs of the molecules will
contain contributions from the unsaturated carbon centers. A
general understanding of substituent effects in phosphazene
chemical shifts is lacking and would be of interest if it were to
become available. In addition to ruling out rearrangement to an
allene,18 further insight into the electronic effects operative in
these systems comes from a consideration of the 13C NMR data
(Table 1) of alknylphosphazenes. There is a progressive
deshielding of the C2 (carbon attached to the phosphazene)
carbon atom on going from the alkyl to aryl substituted alkyne,
Scheme 2. Hydrogenation of Ethynyllphosphazenes
Figure 1. Effect of hydrogenation on 31P NMR spectra.
5468
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470

Inorganic Chemistry
Article
allowed for assignment of the gas chromatograms and
quantification of the relative amount of each isomer. The
mass spectral fragmentation patterns conform to previously
observed pathways with the nongeminal isomers proceeding
through ring cleavage to produce linear ions, while the geminal
isomers exhibit cleavage of the phosphorus−carbon bonds,
leaving the phosphazene ring intact.32,33 Structural assignments
of the mixed substituent derivatives, 6A and 6B, were
accomplished by an analogous process from the hydrogenated
derivatives 7A and 7B. The isomers obtained in all of the
reactions are summarized in Scheme 1.
There are several intriguing observations concerning regio
and stereo control in the reactions of cyclophosphazenes which
have arisen from this work. The reactions leading to the mixed
substituent derivatives, 6A and 6B, show clearly that the
reaction pathway is controlled by the substituent on the
phosphazene. Thus, the isomeric composition of products of
the reaction of 9 with lithio phenyl acetylene is essentially the
same as found in the reaction of 9 with 2-(4-lithophenyl)-
propene, and the reaction of 8 with 2-(4-lithophenyl)propene
gives a similar isomeric distribution to that observed in the
reaction of 8 with lithio phenyl acetylene. In each case, the
incoming nucleophile does not influence the reaction pathway.
The reactions of n-alkylethynyl lithium reagents with 1 were
surprisingly found to exhibit nongeminal regioselectivity, which
is more pronounced for the propynyl derivatives (3, cis > trans
> gem; 5, (cis > trans) ∼ gem). Previously reported reactions of
substituted ethynyl lithium reagents with 1 follow a primary
geminal pathway with greater geminal selectivity shown for the
phenyl versus the trimethylsilyl substituted entity.12,13 An
examination of the 13C NMR data (Table 1) shows a
remarkable correlation between increasing removal of π
electron density (by polarization, conjugation, or hyper-
conjugation) from the ethynyl center to the substituent and
increased regio selectivity. The n-alkylethynyl derivatives also
show some degree of contra steric cis stereo selectivity. These
observations provide added insight into the factors which
control the reaction pathways followed by carbanionic
nucleophiles in reactions with cyclophosphazenes. We have
previously shown that the halogen abstraction mechanism
which controls the reactions of Grignard and organolithium
reagents with chlorophosphazenes6,7,18 plays, at best, a minor
role in the reactions of 1,34 so one can assume that nucleophilic
substitution is the dominant mechanism. It is a general
observation in the reactions of chlorophosphazenes that
geminally substituted products arise from dissociative pro-
cesses.2 If one assumes that the same fundamental mechanistic
factors are operative in the corresponding fluorophosphazenes,
then the geminal products observed arise via a dissociative
mechanism, and the balance of geminal to nongeminal product
distribution reflects the relative rates of the dissociative and
associative pathways. The ethynyl lithium reagents which favor
geminal substitution are those in which the terminal carbon
atom has a reduced s character and hence more σ electron
releasing. This in turn reduces the effective positive charge on
the phosphorus atom allowing for more facile fluoride
dissociation. Conversely, the alkyl substituted ethynyl lithium
reagents have a higher terminal carbon s character (see above)
and thus a concomitantly higher phosphorus formal charge,
thereby strengthening the phosphorus fluorine bond and
favoring an associative process. Steric factors, such as in the
case of t-butyl lithium,10 can also result in the nongeminal
pathway. In the associative process, the nongeminal site has a
Figure 2. Plot of 31P NMR chemical shifts for N3P3F5R (R = alkyl,
alkenyl, alkynyl) vs percentage of s character of the carbon atom
attached to the phosphorus center.
a
b
Table 1. Seleted 13C NMR and IR Data for N3P3F5CCR
R
δ C2
1JPC
δ C1
2JPC
νCC
c
C6H5
102.7
92.6
69.5
70.1
69.2
332.8
366
118.0
112.7
103.6
107.6
61
2180
2145
2215
2207
c
SiMe3
CH3
51.5
61.4
60.4
n-C4H9
371
a
Chemical shifts in ppm; C2 is carbon attached to the phosphorus
b
c
center, coupling constants in Hz Frequencies in cm−1 Data from ref
11.
indicating increased removal of electron density from C2.28−30
The mechanisms involved include hyperconjugation in the
trimethylsilyl derivative31 and resonance delocalization from the
alkyne to the phenyl group in the phenyl acetylene
derivative.28−30 This proposal is consistent with the parallel
decrease in the carbon−carbon triple bond stretching frequency
1
(Table 1). The remarkable decrease in Jpc on going from the
alkyl to the other derivatives is based on the higher percentage
of s character in C2 when the alkyne is unperturbed. The fact
that the 13C NMR parameters do not change significantly on
going from the monosubstituted to the geminal disubstituted
phenyl acetylene derivatives11 indicates that the phosphazene
does not contribute to changes in the alkyne electronic
structure. This, in turn, is consistent with the proposal of
absence of mesomeric interactions between unsaturated
substituents and the phosphazene.25−27
In addition to spectral simplification, the hydrogenation
reactions described above provide a high yield route to a variety
of primary and secondary alkyl-substituted cyclophosphazenes
which are not available, with the exception of the methyl
derivatives,9 from direct reactions with alkyl lithium reagents
due to the facile formation of stabilized α anions at the α
carbon atom.10
The 31P NMR of the hydrogenated materials clearly shows
the presence of all three disubstituted products. The PF2 center
in the cis-2,4 isomer appears as a doublet of doublets due to the
nonequivalence of the fluorine atoms, whereas in the trans
isomer the equivalent fluorine atoms give rise to a triplet in the
31P NMR. The geminal isomer was identified from the PR2
resonance without the 1JPF coupling. The qualitative estimation
of the mixture composition available from the NMR spectra
5469
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470

Inorganic Chemistry
Article
higher positive formal charge27,35 and thus is the site of
nucleophilic attack. In consideration of the stereochemistry of
the associative reactions, it is important to recognize that the
expectation in reactions of any ring system is that the second
reagent will approach from the opposite side of the ring, giving
a trans product.2,10 The cis stereo selectivity observed in the
formation of 3 and 5 follows the general pattern of a reagent
containing an unsaturated organic center2,5 and has been
attributed to the attraction of the positive counterion to the
distributed negative charge on the ring substituent and thus
directs the incoming entity to the cis position.5 Counter ion
directive effects have also been established in other cyclo-
phosphazene substitution reactions.36,37
(25) Allen, C. W.; Brown, D. E.; Hayes, R. F.; Tooze, R.; Poyser, G.
L. ACS Symp. Ser. 1994, 572, 389.
(26) Allen, C. W.; Worley, S. D. Inorg. Chem. 1999, 38, 5187.
(27) Calichman, M.; Hernandez-Rubio, D.; Allen, C. W. Phosphorus,
Sulfur Silicon Relat. Elem. 2002, 177, 1811.
(28) Dawson, D. A.; Reynolds, W. F. Can. J. Chem. 1975, 53, 373.
(29) Amass, A. J.; Brough, P. E.; Colcough, M. E.; Philbin, I. M.;
Perry, M. C. Des. Monomers Polym. 2004, 7, 413.
(30) Lin, S. T.; Lee, C. C.; Liang, D. W. Tetrahedron 2000, 56, 919.
(31) Szewczyk, J.; Gryff-Keller, A; Starowieyski, K. J. Chem. Res. (M)
1992, 26.
(32) Allen, C. W.; Toch, P. L. J. Chem. Soc., Dalton Trans. 1974, 1685.
(33) Allen, C. W.; Diek, R. L.; Brown, P.; Moeller, T.; Schmulbach,
C. D.; Cook, A. G. J. Chem. Soc., Dalton Trans. 1978, 173.
(34) Allen, C. W.; Toch, P. L. Inorg. Chem. 1981, 20, 8.
(35) Calichman, M.; Derecskei-Kovacs, A.; Allen, C. W. Inorg. Chem.
2007, 46, 2011.
AUTHOR INFORMATION
Corresponding Author
*E-mail: christopher.allen@uvm.edu.
■
(36) Belsi
̧ , S.; Coles, J. S.; Davies, D. B.; Kilic,̧ A.; Okutan, E.; Eci̧ k, E.
T.; Ciftci, G. Y. J. Chem. Soc., Dalton Trans. 2011, 40, 4959.
̧
̧
(37) Brandt,K. In Applicative Aspects of Cyclophosphazenes; Geleria,
M., DeJager, R., Eds; Nova Science Publishers, Inc: NewYork, 2004;
Chapter 5, pp 97−138.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Partial support of this work by the Office of Naval Research and
Vermont EPSCoR is appreciated.
■
REFERENCES
■
(1) Part 25: Bahadur, M.; Allen, C. W.; Geiger, W. E.; Bridges, A.
Can. J. Chem. 2002, 80, 1393.
(2) Allen, C. W. Chem. Rev. 1991, 91, 119.
(3) Allen, C. W.; Freund, A. S.; Calichman, M. Phosphorus, Sulfur
Silicon Relat. Elem. 2008, 183, 479.
(4) Freund, A. S.; Calichman, M.; Allen, C. W. Z. Anorg. Allg. Chem.
2004, 630, 2059.
(5) Carter, K. C.; Calichman, M.; Allen, C. W. Inorg. Chem. 2009, 48,
7476.
(6) Allcock, H. R. Chemistry and Applications of Polyphosphazenes;
Wiley Interscience: Hoboken, NJ, 2003; pp 62−68.
(7) Winter, H.; Van de Grampel, J. C. J. Chem. Soc., Dalton Trans.
1986, 1269.
(8) Cristau, H.-J.; Fruchier, A.; Plenat, F.; Taillefer, M.; Vincent, V. In
Applicative Aspects of Cyclophosphazenes; Gleria, M., DeJager, R., Eds.;
Nova Science: New York, 2004; Chapter 11, p 277.
(9) Paddock, N. L.; Rangunathan, T. N.; Todd, S. M. Can. J. Chem.
1971, 49, 164.
(10) Ramachandran, K.; Allen, C. W. J. Am. Chem. Soc. 1982, 104,
2396.
(11) Allen, C. W.; Bright, R. P. Inorg. Chem. 1983, 22, 1291.
(12) Chivers, T. Inorg. Nucl. Chem. Lett. 1971, 7, 827.
(13) Allen, C. W.; Desorcie, J. L.; Ramachandran, K. J. Chem. Soc.,
Dalton Trans. 1984, 2843.
(14) Shaw, J. C.; Allen, C. W. Inorg. Chem. 1986, 25, 4632.
(15) Allen, C. W. Trends Polym. Sci. 1994, 2, 342.
(16) Van de Grampel,J. C. In Synthesis and Characterizations of
Poly(organophosphazens); Gleria, M., DeJager, R., Eds.; Nova Science:
New York, 2004; Chapter 7, p143.
(17) Allen, C. W. Inorg. Chim. Acta 2011, 372, 32.
(18) Allcock, H. R.; Descorcie, J. L.; Riding, G. H. Polyhedron 1987,
6, 119.
(19) Allen, C. W.; Malik, P.; Bridges, A.; Descorcie, J.; Pellon, B.
Phosphorus, Sulfur Silicon Relat. Elem. 1990, 50, 443.
(20) Allen, C. W.; Badahur, M. J. Inorg. Organomet. Polym. 1998, 8,
23.
(21) Kumar, M. S.; Upreti, S.; Elias, A. J. Inorg. Chem. 2006, 45, 7835.
(22) Kumar, M. S.; Gupta, R. P.; Elias, A. J. Inorg. Chem. 2008, 47,
3433.
(23) Keshav, K.; Singh, N.; Elias, A. J. Inorg. Chem. 2010, 49, 5753.
(24) Moeller, T.; John., K.; Tsang, F. Chem. Ind. (London) 1961, 347.
5470
dx.doi.org/10.1021/ic300481e | Inorg. Chem. 2012, 51, 5465−5470