Experimental and theoretical characterization of the hexaazidophosphate(V) ion

Article abstract of DOI:10.1021/ic801536z

(PPN)[P(N₃)₆] (2) was synthesized by the metathesis of Na[P(N₃)₆] (1) and (PPN)N₃ (PPN⁺ = {(Ph₃P)₂N)⁺), allowing for the isolation and full characterization of a stable hexaazidophosphate(V) salt by ³¹P and ¹⁴N NMR, UV absorption, IR and Raman spectroscopy, elemental and thermal analyses, X-ray diffraction, and Hartree-Fock and density functional theory calculations. The colorless single crystals of 2 are triclinic, space group P1, a = 9.6296(9), b = 9.8158(9), c = 10.1414(10) A, α = 92.635(5)°, β = 93.437(5)°, γ = 92.105(4)°, and Z = 1. The [P(N₃)₆]^- ion of 2 is isolated in the solid state and adopts S₆ symmetry both in the crystal and in solution. Thermogravimetric analysis and differential scanning calorimetry measurements reveal a surprising thermal stability of 2 (T_dec ca. 200 °C). No friction sensitivity was encountered.

Full text of DOI:10.1021/ic801536z

Inorg. Chem. 2008, 47, 12004-12009
Experimental and Theoretical Characterization of the
Hexaazidophosphate(V) Ion
P. Portius,* P. W. Fowler, H. Adams, and T. Z. Todorova
Department of Chemistry, UniVersity of Sheffield, Brook Hill, Sheffield, S3 7HF, United Kingdom
Received August 12, 2008
(PPN)[P(N3)6] (2) was synthesized by the metathesis of Na[P(N3)6] (1) and (PPN)N3 (PPN+ ) {(Ph3P)2N}+), allowing
for the isolation and full characterization of a stable hexaazidophosphate(V) salt by 31P and 14N NMR, UV absorption,
IR and Raman spectroscopy, elemental and thermal analyses, X-ray diffraction, and Hartree-Fock and density
¯
functional theory calculations. The colorless single crystals of 2 are triclinic, space group P1, a ) 9.6296(9), b )
9.8158(9), c ) 10.1414(10) Å, R ) 92.635(5)°, ꢀ ) 93.437(5)°, γ ) 92.105(4)°, and Z ) 1. The [P(N3)6]- ion
of 2 is isolated in the solid state and adopts S6 symmetry both in the crystal and in solution. Thermogravimetric
analysis and differential scanning calorimetry measurements reveal a surprising thermal stability of 2 (Tdec ca. 200
°C). No friction sensitivity was encountered.
Introduction
energetic salt [N5][P(N3)6]10,11 is extremely shock-sensitive
and explodes upon the slightest warming toward room
temperature. The introduction of larger, weakly coordinating
R4N+and Ph4E+ counterions, R ) CH3 and C2H5,6 E ) P
and As,8 leads to somewhat less-sensitive compounds which,
however, still easily and violently explode upon slight
warming. These characteristics hamper the isolation and
characterization of [P(N3)6]-, which is thus far limited to
vibrational6,10 and 31P NMR9 spectroscopy. The recent rapid
progress in the quest for new energetic molecules and
ongoing research in the area of homoleptic main group azides
prompted the investigation of [P(N3)6]- reported in this paper.
The synthesis, isolation, and characterization of the first stable
hexaazidophosphate(V) salt was achieved by using the
nonelectrophilic and bulky bis(triphenylphosphoranylide-
ne)ammonium cation {(Ph3P)2N}+ (PPN+).12,13
Covalent phosphorus-nitrogen molecules are difficult to
isolate and handle, owing to their highly endothermic
character and very low energy barriers, which often cause
uncontrolled explosive decomposition.1,2 Despite recent
success with the shock-sensitive binary cyclic P-N molecule
{NP(N3)2}3,2,3 the related homoleptic phosphorus azides
P(N3)3 and P(N3)5 could be neither isolated nor structurally
characterized.4-7 Nevertheless, studies of the preparation of
the ionic complex [P(N3)6]-
lenging, accessibility of related homoleptic phosphorus
azides. According to these studies, complete removal of the
solvent from an acetonitrile solution of Na[P(N3)6] (1) results
in explosion at ambient temperature,6,8 and the highly
6,8-10
revealed the, albeit chal-
* Author to whom correspondence should be addressed. Fax: (+44)114-
22-29385. E-mail: p.portius@sheffield.ac.uk.
(1) Tornieporth-Oetting, I. C.; Klapoetke, T. M. Angew. Chem., Int. Ed.
Engl. 1995, 34, 511–520.
Results and Discussion
(2) Goebel, M.; Karaghiosoff, K.; Klapoetke, T. M. Angew. Chem., Int.
Ed. 2006, 118, 6183–6186.
(3) Muralidharan, K.; Omotowa, B. A.; Twamley, B.; Piekarski, C.;
Shreeve, J. M. Chem. Commun. 2005, 5193–5195.
(4) Buder, W.; Schmidt, A. Z. Anorg. Allg. Chem. 1975, 415, 263–267.
(5) Zeng, X.; Wang, W.; Liu, F.; Ge, M.; Sun, Z.; Wang, D. Eur. J. Inorg.
Chem. 2006, 416–421.
(6) Volgnandt, P.; Schmidt, A. Z. Anorg. Allg. Chem. 1976, 425, 189–
192.
(7) Knapp, C.; Passmore, J. Angew. Chem., Int. Ed. 2004, 43, 4834–4836.
(8) Roesky, H. W. Angew. Chem., Int. Ed. Engl. 1967, 6, 637.
(9) Dillon, K. B.; Platt, A. W. G.; Waddington, T. C. J. Chem. Soc. Chem.
Comm. 1979, 889.
Na[P(N3)6] (1) selectively forms within 1 h under strict
exclusion of air and water in acetonitrile by the addition of
a PCl5 solution to a cold suspension containing sodium azide
in a more than 2-fold excess, and subsequent warming up
to room temperature. Spectroscopic analysis of the brownish
reaction solution, which can be stored indefinitely at -30
°C, shows only one singlet 31P NMR resonance at -178.8
(11) Singh, R. P.; Verma, R. D.; Meshri, D. T.; Shreeve, J. M. Angew.
Chem., Int. Ed. 2006, 45, 3584–3601.
(10) Haiges, R.; Schneider, S.; Schroer, T.; Christe, K. O. Angew. Chem.,
Int. Ed. 2004, 43, 4919–4924.
(12) Appel, R.; Hauss, A. Z. Anorg. Allg. Chem. 1961, 311, 290–301.
(13) Ruff, J. K.; Schlientz, W. J. Inorg. Synth. 1974, 15, 84–90.
12004 Inorganic Chemistry, Vol. 47, No. 24, 2008
10.1021/ic801536z CCC: $40.75  2008 American Chemical Society
Published on Web 11/13/2008

Characterization of the Hexaazidophosphate(V) Ion
Scheme 1
Table 1. Results of Thermal Analysesa
fusion
step 1
step 2
method
DSC
(PPN)[P(N3)6] (2)
(PPN)2[Si(N3)6] (3)
Ton
∆H
Ton
∆m
Ton
∆H
Ton
∆m
Ton
∆H
Ton
191
+68
242
-946
192
23
256
-824
263
6.1
256
297
-140
300
TGA
DTA
TGA
DSC
TGA
ppm and the complete consumption of PCl5. Furthermore, a
very intense infrared absorption appeared in the region of
the asymmetric azide stretching vibrations centered at 2116
cm-1 next to very weak bands of HN3 and an unknown
decomposition product at 2138 cm-1 and 2172 cm-1,
respectively. Addition of the solution containing 1 to
(PPN)N3 causes precipitation of NaN3. The resulting (PP-
N)[P(N3)6] salt (2; Scheme 1) forms in high yield from the
supernatant solution as white crystals, which show neither
noticeable signs of decomposition upon several minutes of
exposure to the air nor sensitivity toward friction and static
electricity. However, solutions of 2 are highly air-sensitive.
Compound 2 is moderately soluble in CH3CN and very
soluble in THF14 and stable in both solvents under strict
exclusion of air at ambient temperature, whereas rapid
decomposition occurs in CH2Cl2. The very intense band of
N3- at 1993 or 2005 cm-1 is absent from IR spectra recorded
in THF and CH3CN, respectively, indicating that the
[P(N3)6]- ion does not dissociate in these solvents.
Compound 2 shows surprising thermal stability, which can
be directly compared with those of the related silicon and
germanium salts (PPN)2[E(N3)6] (3, E ) Si; 4, E ) Ge),
which differ solely in the anion charge, the nature of the
anions’ central atom, and the anion/cation ratio. Melting of
2 occurs at 200-202 °C, 10-20 °C below that of 3 (225
°C) and 4 (211 °C). Rapid heating of sealed capillaries
containing 1-2 mg of 2 causes detonation at ca. 263 °C. A
combined thermogravimetric (TGA) and differential scanning
calorimetry (DSC) analysis reveals that melting at an
extrapolated onset temperature (Ton) of 191 °C, ∆Hf ) +68
kJ mol-1, is followed by a two-step decomposition at Ton1
) 242 °C and Ton2 ) 297 °C, accompanied by molar enthalpy
changes of -946 and -140 kJ mol-1, respectively. These
properties indicate 2 as thermally slightly less-stable than
the related silicon and germanium salts 3 and 4 (see Table
1) while releasing significantly more energy per mole upon
decomposition during the first step. The second decomposi-
tion step at Ton2 ) 300 °C (TGA) coincides with the onset
of decomposition of (PPN)N3. The thermal behavior of the
latter was independently determined in a TGA experiment
by subjecting a genuine sample of (PPN)N3 to the same
conditions (Ton ) 304 °C). Hence, the substantially lower
molar enthalpy change of the second step in comparison with
3 and 4 is largely due to the reduced PPN cation content of
2. Furthermore, the TGA analysis of 2 shows the onset of
an extensive 23% mass loss (-192 g mol-1) associated with
the first step already at 192 °C, whereas mass loss in 3 is
much smaller (6.1%, -82 g mol-1) and sets in at a
considerably higher temperature (263 °C).
204
321
-430
325
(PPN)2[Ge(N3)6] (4)
194
+64
312
-482
304
-705
(PPN)N3
a Extrapolated onset temperature Ton/°C, molar enthalpy change ∆H/kJ
mol-1, mass loss ∆m/%. Data obtained from differential scanning calo-
rimetry (DSC), thermogravimetric analysis (TGA) and differential thermal
analysis (DTA) of the salts 2 and (PPN)N3 (this paper), 3,15 and 4.16
onset of the current of positive ions was observed under
electron impact, with the masses m/z ) 199 (100%) and 115
(92%) being most abundant. An accurate mass measurement
allows these masses to be assigned to [P(N3)4]+ (199.009927
amu, +3.6 ppm) and [P(N3)2]+ (tentatively). Chemical
ionization mass spectra (NH3) show masses at m/z ) 199
(100%) and 242 (10%), corresponding to [P(N3)4]+ and
[P(N3)5H]+, respectively. Furthermore, prolonged heating at
210 °C in Vacuo turns 2 into brownish oil, containing some
CH3CN insoluble material. Fourier transform infrared (FTIR)
spectra recorded for a CH3CN solution of the oil revealed a
selective growth of the νasym(N3) absorption band of unco-
ordinated N3 at 2005 cm-1 at the expense of the band of
-
[P(N3)6]-. Thus, it can be concluded that heating causes the
dissociation of [P(N3)6]- into N3 and P(N3)5, which partly
-
evaporates in Vacuo or decomposes under ambient pressure.
An understanding of the relative stability of the [P(N3)6]-
salts can be approached through semiquantitative estimates
of the energetics of the proposed initial dissociation reactions
(PPN)[P(N3)6](s) f (PPN)N3(s) + P(N3)5 (i) and Na[P(N3)6](s)
f NaN3(s) + P(N3)5 (ii) based on lattice energy.17 The gain
in lattice energy of reaction i is close to zero, whereas
reaction ii leads to a net gain of ca. 240 kJ mol-1. This
finding indicates Na[P(N3)6] to be thermodynamically less
stable than (PPN)[P(N3)6], irrespective of the unknown state
of P(N3)5. Furthermore, an analogous reaction involving
the salts 3 and 4, (PPN)2[E(N3)6](s) f (PPN)N3(s)
+
(PPN)[E(N3)5](s) (iii), leads to a net loss of lattice energy of
ca. 170 kJ mol-1 each, which contributes to the higher
stability of the latter salts in comparison to 2 (see the
Supporting Information for further details).
Data for the [P(N3)6]- ion obtained from IR, Raman, and
31P NMR spectra of 2 recorded in solution or the solid state
are largely consistent with the literature.6,9,18 The 14N NMR
spectrum of 2 shows three singlet resonances at δ ) -268,
-141, and -170 ppm due to NR, Nꢀ, and Nγ atoms of
magnetically equivalent azido ligands. [The Nꢀ resonance
(15) Filippou, A. C.; Portius, P.; Schnakenburg, G. J. Am. Chem. Soc. 2002,
124, 12396–12397.
(16) Filippou, A. C.; Portius, P.; Neumann, D. U.; Wehrstedt, K.-D. Angew.
Chem., Int. Ed. 2000, 39, 4333–4336.
Mass spectrometry was used to analyze the evolved gas
during the thermal decomposition of 2. At ca. 200 °C, the
(17) Jenkins, H. D. B; Roobottom, H. K.; Passmore, J.; Glasser, L. Inorg.
Chem. 1999, 38, 3609–3620.
(14) The related PPN salts (PPN)2[E(N3)6] (E ) Si, Ge) are all soluble in
(18) In CH2Cl2, the anion resonates at -180.0 ppm: Dillon, K. B.; Platt,
CH2Cl2 and CH3CN but insoluble or sparingly soluble in THF.
A. W. G. J. Chem. Soc., Dalton Trans. 1983, 1159–1164.
Inorganic Chemistry, Vol. 47, No. 24, 2008 12005

Portius et al.
overlaps with the solvent 14N resonance at -136.6 ppm and
was determined by Lorentzian curve fitting. A 14N resonance
of the PPN cation could not be found, which is in accord
with the absence of the resonance of (PPN)N3 in the same
solvent.] These chemical shifts are within the spectral region
defined by the related [As(N3)6]- and [Sb(N3)6]- complexes
(As: -253, -142, -163;19 Sb: -287, -141, -185 ppm20)
and are clearly distinct from those of doubly charged anions
[E(N3)6]2- (E ) Si,15 Ge,16 Se,21 Te:22 -297 to -289 ppm,
NR; -139 ppm, Nꢀ; -248 to -208 ppm, Nγ). Near-UV
spectra recorded on the low-energy side of the PPN absorp-
tion peaks in THF and MeCN solution are nearly identical
and show two weak, broad, and overlapped bands assignable
to [P(N3)6]-. Gaussian deconvolution shows these bands to
be centered at 34 700 cm-1 (εmax ∼ 700 L mol-1 cm-1, fwhm
∼ 3600 cm-1) and 31 400 cm-1 (εmax; ∼ 100 L mol-1 cm-1,
fwhm ∼ 1900 cm-1). No signs of self-aggregation have been
found.
Raman, IR, and 14N NMR data indicate a highly symmetric
structure of the [P(N3)6]- ion; however, no unambiguous
assignment among candidate D3, D3d, and S6 structures can
be made relying on these methods alone.
The solid-state structure of 2 was determined by X-ray
diffraction of a single crystal, which was obtained by slow
cooling of a warm saturated solution in CH2CN.23 Compound
Figure 1. Packing arrangement in the crystal of compound 2 with the
vertices of the unit cell occupied by [P(N3)6]- ions, while the center is
occupied by the PPN cation; hatched, dotted, and empty circles depict P,
N, and C atoms, respectively; H atoms not shown.
j
2 crystallized in the triclinic crystal system P1 with one
formula unit per unit cell and consists of discrete anions,
each of which is surrounded by eight PPN+ cations (Figure
1). The complex anion has a strict Ci symmetric structure in
(19) Karaghiosoff, K.; Klapoetke, T. M.; Krumm, B.; Noeth, H.; Schuett,
T.; Suter, M. Inorg. Chem. 2002, 41, 170–179.
(20) Haiges, R.; Boatz, J. A.; Vij, A.; Vij, V.; Gerken, M.; Schneider, S.;
Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem., Int. Ed.
2004, 43, 6676–6680.
(21) Klapoetke, T. M.; Krumm, B.; Scherr, M.; Haiges, R.; Christe, K. O.
Angew. Chem., Int. Ed. 2007, 46, 8686–8690.
(22) Haiges, R.; Boatz, J. A.; Vij, A.; Gerken, M.; Schneider, S.; Schroer,
T.; Christe, K. O. Angew. Chem., Int. Ed. 2003, 42, 5847–5851.
(23) Crystal data for C36H30N19P3 are as follows: M ) 821.70; crystallizes
from acetonitrile as colorless prisms; crystal dimensions, 0.32 × 0.14
× 0.14 mm3; triclinic; a ) 9.6296(9), b ) 9.8158(9), c ) 10.1414(10)
Å; R ) 92.635(5)°; ꢀ ) 93.437(5)°; γ ) 92.105(4)°; U ) 954.83(16)
Figure 2. Projection of the [P(N3)6]- anion in 2 down the S2 axis. The P2
atom is located at a center of inversion. Thermal ellipsoids are set at the
50% probability level. Bond lengths (Å) and angles (deg): P2-N2
1.8121(12), P2-N5 1.8040(12), P2-N8 1.8071(12), N2-N3 1.2290(18),
N5-N6 1.2254(17), N8-N9 1.2282(17), N3-N4 1.1325(18), N6-N7
1.1341(18), N9-N10 1.1266(18), P2-N2-N3 117.44(9), P2-N5-N6
118.03(10), P2-N8-N9 117.95(10), N2-N3-N4 175.12(14), N5-N6-N7
175.02(15), N8-N9-N10 174.65(15), N2-P2-N5 89.14(6), N5-P2-N8
91.04(6), N8-P2-N2 89.08(6).
3
1
Å3; Z ) 1; Dcalcd ) 1.429 Mg/m ; space group P1; Ci ; No. 2; λ )
0.71073 Å (Mo KR); µ (Mo KR) ) 0.213 mm-1; F(000) ) 424; T )
150 K. Data collected were measured on a Bruker Smart CCD area
detector with an Oxford Cryosystems low-temperature system. Cell
parameters were refined from the setting angles of 7906 reflections
(θ range 2.08 < 28.31°). Reflections were measured from a hemisphere
of data collected of frames, each covering 0.5° in Ω. Of the 16811
reflections measured, all of which were corrected for Lorentz and
polarization effects and for absorption by semi-empirical methods
based on symmetry-equivalent and repeated reflections (minimum and
maximum transmission coefficients 0.9349 and 0.9708), 4070 inde-
pendent reflections exceeded the significance level |F|/σ(|F|) > 4.0.
The structure was solved by direct methods and refined by full matrix
least-squares methods on F2. Hydrogen atoms were placed geo-
metrically and refined with a riding model and with Uiso constrained
to be 1.2 times the Ueq of the carrier atom. Refinement converged at
a final R ) 0.0385 (wR2 ) 0.1082, for all 4674 data, 265 parameters,
mean and maximum δ/σ 0.000, 0.000) with allowance for the thermal
anisotropy of all non-hydrogen atoms. Minimum and maximum final
electron densities were -0.316 and 0.610 e Å-3. A weighting scheme,
j
which a central phosphorus atom is bound in an octahedral
fashion to the NR atoms of six surrounding azido ligands,
creating an unprecedented nitrogen hexa-coordinated PN6
framework within the resulting [P(N3)6]- complex (Figure
2). The P-N bonds and the P-N-N and N-N-N angles
of the individual azido ligands deviate only marginally from
each other (Figure 2 caption), causing a slight distortion to
a structure that otherwise would have perfect S6 point-group
symmetry. This structural motive appears in all main group
element hexaazides bearing a central atom without a stere-
ochemically active lone pair reported to date (E ) As, Sb,
Si, Ge, Sn, Pb, Se; for refs, see Table 2).
2
2
2
w ) 1/[σ2(Fo ) + (0.0675P)2 + 0.2521P] where P ) (Fo + 2Fc )/3
was used in the latter stages of refinement. Complex scattering factors
were taken from the program package SHELXTL (an integrated system
for solving and refining crystal structures from diffraction data (revision
5.1), Bruker AXS LTD) as implemented on a Viglen Pentium
computer.
Structural data on hexa-coordinate phosphorus azides are
unavailable. However, a comparison with the pentagonal-
12006 Inorganic Chemistry, Vol. 47, No. 24, 2008

Characterization of the Hexaazidophosphate(V) Ion
Table 2. Selected Average Bond Lengths and Angles of the [E(N3)6]n- Complexes of Group 14, 15, and 16 Elementsa
bond lengths [Å]b
bond angles [deg]b
E-NR
NR-Nꢀ
Nꢀ-Nγ
∆ΝΝc
E-NR-Nꢀ
NR-Nꢀ-Nγ
(PPN)[P(N3)6] (2)
1.81
1.94
2.08
1.87
1.97
2.13
1.23
1.23
1.22
1.20
1.21
1.20
1.13
1.13
1.13
1.15
1.15
1.14
0.10
0.10
0.10
0.06
0.07
0.07
118
116
117
124
120
115
175
175
175
176
176
176
(py-H)[As(N3)6] (9)
(Ph4P)[Sb(N3)6] (10)
(PPN)2[Si(N3)6] (3)
(PPN)2[Ge(N3)6] (4)
(Ph4P)2[Se(N3)6] (11)
a E ) P, As,19 Sb,20 n ) 1; Si,15 Ge,16 Se,21 n ) 2. Data were excluded for the ions [Sn(N3)6]2-
,
29 [Pb(N3)6]2- 30 and those which possess a stereochemically
,
active lone pair. b Rounded mean of the crystallographically independent parameters; uncertainty of the bond lengths is generally smaller than the last digit.
c ∆ΝΝ ) Do(NR-Nꢀ) - Do(Nꢀ-Nγ).
a
Table 3. Results of HF and DFT Calculations on [P(N3)6]-
bond lengths [Å]
bond angles [deg]
P-NR
NR-Nꢀ
Nꢀ-Nγ
∆ΝΝ
0.111
P-NR-Nꢀ
NR-Nꢀ-Nγ
S6 RHF/6-311G*
1.804
1.208
1.097
119.4
176.1
S6 B3LYP/6-311G*
1.834
1.808(4)
1.217
1.228(2)
1.137
1.131(4)
0.080
0.097(6)
120.5
117.8(3)
175.1
174.9(2)
b
X-ray
a Details of the theoretical calculations are found in the Supporting Information. b Unweighted mean xu of the three crystallographically independent bond
lengths and angles of 2 listed.31
bipyramidal PX4(N3) azido phosphoranes [P(O-R-O)-
(O-R′-O)(N3)],24 [P(O-R-O)(O-R′-NR′′)(N3)],25 {cyclo-
(P(N3)3(NPh)}2,26 and [P{N(CH2CH2NR)3}(N3)]+ (5-8)27
and the monoanionic group 15 hexaazides [As(N3)6]- (9)
and [Sb(N3)6]- (10) offers intriguing insight. For instance,
the bond lengths and angles of the axial azido group of
complexes 5-8 adopt values within narrow ranges and fully
encompass those of [P(N3)6]-, with the exception of slightly
longer P-NR bonds (1.8040(12)-1.8121(12) Å) in the latter.
Thus, the overall negative charge and increased coordination
number of 2 lengthen the coordinative P-NR bonds by 3-4
pm with respect to the unstrained phosphoranes 5 and 6.
Furthermore, a close relationship with the hexaazides 9 and
10 can be established on the basis of the range of average
NR-Nꢀ and Nꢀ-Nγ bond lengths, resulting in similar ∆NN
bond length differences of ca. 10 pm, as in 2 (Table 2). The
doubly charged group 14 and 16 hexaazido anions 3, 4, and
[Se(N3)6]2- (11), however, on average exhibit slightly shorter
NR-Nꢀ bonds and slightly longer Nꢀ-Nγ bonds, resulting
in smaller ∆NN values (6-7 pm). A comparison of the ∆NN
parameter as an indicator for π delocalization along the azido
ligand and E-N3 bond ionicity suggests that the ionic
character of the coordinative bonds in [P(N3)6]- is similar
to that in the As and Sb congeners 9 and 10 but weaker
than those of the dianionic complexes 3, 4, and 11. The fact
that, unlike [Si(N3)6]2- and [Ge(N3)6]2-, [P(N3)6]- does not
react with 2,2′-bipyridine under identical conditions at
ambient temperature in acetonitrile solution supports this
interpretation.
structure of S6 symmetry, which was checked by an evalu-
ation of the Hessian (energy, geometry, and vibrational
frequencies are given in the Supporting Information). The
flexibility of this structure is indicated by the presence of
angular-distortion modes with extremely low frequencies, the
lowest of which was found at 53 cm-1 (RHF) and 38 cm-1
(B3LYP), respectively. The calculated structural parameters
are close to those obtained from X-ray diffraction (see Table
3).
Furthermore, the near neighborhood of the S6 minimum
was explored for other stationary points. Imposition of D3d
symmetry yielded only a high-order stationary point on the
energy hypersurface, lying 201 and 139 kJ mol-1 above the
S6 minimum at RHF and DFT levels, respectively, and six
imaginary-frequency modes, A2g + A1u + Eg + Eu at i61 to
i167 cm-1 (RHF/6-311G*) and i50 to i138 cm-1 (B3LYP/
6-311G*), respectively. Thus, D3d cannot be a candidate for
the symmetry of the “free” anion. Relaxation of this structure
within the D3 subgroup of D3d gave a new structure predicted
(24) Kumaraswamy, S.; Muthiah, C.; Swamy, K. C. K. J. Am. Chem. Soc.
2000, 122, 964–965.
(25) Kumar, N. S.; Kommana, P.; Vittal, J. J. S.; Swamy, K. C. K. J. Org.
Chem. 2002, 67, 6653–6658.
(26) Aubauer, C.; Klapoetke, T. M.; Noth, H.; Schulz, A.; Suter, M.;
Weigand, J. Chem. Commun. 2000, 2491–2492.
(27) Thirupathi, N.; Liu, X.; Verkade, J. G. Inorg. Chem. 2003, 42, 389–
397.
(28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzales, C.; Pople, J. A. Gaussian 03, revision
B.01; Gaussian, Inc.: Pittsburgh, PA, 2003.
The results of the X-ray investigation of the structure of
the [P(N3)6]- ion are in contradiction with the previously
proposed6 D3d point-group symmetry for the ion in solution.
To shed further light on the [P(N3)6]- structure in solution,
ab initio calculations were carried out.28 Assignment of
idealized S6 point-group symmetry to the [P(N3)6]- ion was
found to be consistent with the results of calculations for
the “free” ion. Unconstrained optimization at both RHF/6-
311G* and B3LYP/6-311G* levels gave a minimum for a
Inorganic Chemistry, Vol. 47, No. 24, 2008 12007

Portius et al.
to be a secondary local minimum (34 kJ mol-1 above S6) at
the RHF/6-311G* level, again with ultrasoft vibrational
modes (g 31 cm-1). However, the B3LYP/6-311G* level
of theory predicts the same structure to be a doubly
degenerate high-order stationary point (i14 cm-1), rather than
a true minimum. Relaxation of the D3d structure within the
S6 subgroup of D3d led back to the previous minimum.
Intriguingly, a comparison of the molecular shape of the D3
and S6 molecules shows that the S6 structure has adjacent
azide groups in closer vicinity and more efficiently avoiding
opposing positive partial charges of Nꢀ atoms of adjacent
azide groups. The results of frequency calculations were used
to distinguish between the D3 and the S6 structures using
data from FTIR and Raman spectra. The match of relative
intensity and position between predicted and observed bands
was found to be imperfect with both structures. However,
the D3 structure gives noticeably poorer agreement, as
predicted bands due to Raman active vibrations centered at
530 and 750 cm-1 were not observed and the predicted
relative intensity of an IR band at 575 cm-1 is several orders
of magnitude higher than was observed. The combination
of experimentally determined solid-state structure and the
absence of other minima in ab initio quantum mechanical
calculation strongly suggests that [P(N3)6]- has in fact S6
symmetry in solution at room temperature.
In summary, the structure of the energetic ion [P(N3)6]-
has been established beyond reasonable doubt in solution
and in the solid state. The introduction of the bulky PPN
cation has a stabilizing effect both kinetically (spatial
separation and phlegmatization of the energetic ions) and
thermodynamically (by reduction of the azide nitrogen
content and reducing the net gain of lattice energy in an azide
dissociation prior to decomposition into dinitrogen). The
P-N bonds in [P(N3)6]- are comparably weak as a result of
phosphorus hexacoordination. These factors contribute to the
marked stability of salt 2 in comparison to the highly
explosive and sensitive phosphorus polyazides P(N3)3 and
{NP(N3)2}3. The hexaazidophosphate has now found a firm
place on the landscape of energetic molecules.
Schlenk and glovebox techniques were employed throughout. NMR
spectra were recorded on a Bruker 250 MHz (1H, 13C{1H}, 31P{1H})
or 400 MHz (14N) spectrometer run by TopSpin software. Mass
spectra, TGA, and DSC measurements were obtained at the Centre
for Chemical Instrumentation and Analytical Services at Sheffield,
using a VG AutoSpec spectrometer, a Perkin-Elmer Pyris 1 station
(under argon, heating rate 10 °C min-1), and a DSC7. Samples for
Raman spectroscopy were sealed in glass capillary tubes and spectra
recorded between 3800 and 300 cm-1 on a Renishaw Ramascope
System 2000 spectrometer equipped with a holographic notch filter
and a 785 nm diode laser. Only those bands are quoted which are
above 2% of the intensity of the main peak. FTIR spectra were
recorded between NaCl windows using a Bruker FTIR spectrometer.
The spectra of 2 in Nujol suspension contain a number of very
weak bands, which are not quoted.
Caution! Appropriate safety measures need to be taken in
attempts to prepare phosphorus polyazides. Filter residues other
than 2 should not be allowed to dry in the air or in Vacuo and
should be discarded into dilute aqueous sodium hydroxide.
In Situ Generation of Na[P(N3)6] (1). A Schlenk tube was
charged with PCl5 (124 mg, 0.595 mmol), and CH3CN (ca. 10
mL) was added, resulting in a clear and colorless solution. At
-30 °C, the solution of the phosphorus chloride was added
dropwise to a second Schlenk tube containing a stirred suspen-
sion of 14 equiv of NaN3 (549 mg, 8.44 mmol) in ca. 15 mL of
CH3CN, immersed in a cold bath. The cold bath was then
removed and stirring continued for 1 h, after which a brownish
suspension had formed, which was left to settle down. Spectra
recorded of the orange supernatant solution revealed the forma-
tion of the [P(N3)6]- anion, νj ) 3376vvw, 2561vvw, 2116vs,
1288w, 730w, 543w cm-1. NMR 31P{1H} (CD3CN/CH3CN,
ppm): δ -178.8 (s, ∆ν1/2 ) 56 Hz). The reaction solution was
filtered, and the IR-active part of the pale-rose filter residue (493
mg) was identified by FTIR spectroscopy in a nujol mull as being
mainly NaN3 (3391vw, 3300vvw, 2122vs, 640w cm-1). Under
the assumption that all SiCl4 has converted, the concentration
of Na[P(N3)6] in the filter solution is 2.4 × 10-2 mol dm-3
.
Preparation of (PPN)[P(N3)6] (2). A Schlenk tube was charged
with (PPN)N3 (322 mg, 0.555 mmol) and immersed into a cold
bath at -35 °C. A total of 19 mL of the solution of Na[P(N3)6]
(ca. 0.45 mmol) was added dropwise via a transfer cannula while
the reaction mixture was stirred. The cold bath was removed, and
after several hours, a colorless suspension had formed. The
suspension was filtered, which afforded ca. 29 mg of a white solid
(according to FTIR mainly NaN3) that was discarded. At ambient
temperature, the volume of the clear and colorless filter solution
was diminished until precipitation commenced. The precipitate was
redissolved by gentle warming and the resulting solution kept at
-30 °C overnight, which resulted in the formation of large colorless
crystals. The supernatant solution, containing the excessive (PP-
N)N3, was filtered off and discarded and the residue washed with
approximately 3 mL of CH3CN and dried in vacuo. This procedure
afforded 279 mg of white, crystalline 2 (0.340 mmol), ca. 76%
yield with respect to PCl5, mp ) 200-202 °C (sealed glass
capillary). Elem. anal. for C36H30N19P3, 821.67 g mol-1, calcd: C,
Crystallographic data for the crystal structure reported in
this paper have been deposited with the Cambridge Crystal-
lographic Database (CCDC No. 697529). This material can
be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html (or from the CCDC, 12 Union Road, Cam-
bridge CB2 1EZ, U.K.; fax: +44 1223 336033; e-mail:
deposit@ccdc.cam.ac.uk).
Experimental Section
PCl5 (Fluka) was sublimed prior to use. NaN3 (Acros) was dried
at 110 °C overnight under high vacuum conditions. Predried
acetonitrile (Grubbs) was stored over CaH2 and added to reaction
vessels by vacuum transfer. Filtration was carried out using a
stainless steel cannula fitted with a fiberglass filter (Whatman).
(PPN)N3 was prepared by a published procedure.32 Standard
(31) Standard deviation given in parentheses (n ) 3). Thermal motion in
the crystal may produce an apparent shrinkage of the molecular
dimensions and is considered by calculating the instantaneous bond
lengths Di, see Fundamentals of Crystallography; Giacovacco, C., Ed.;
IUCr, Oxford University Press: New York, 1992, resulting in almost
no change for the P-NR and NR-Nꢀ, distances, whereas Nꢀ-Nγ is
affected (Di ca. 1.140(2) Å).
(29) Fenske, D.; Doerner, H. D.; Dehnicke, K. Z. Naturforsch., B: Chem.
Sci. 1983, 38, 1301–1303.
(30) Polborn, K.; Leidl, E.; Beck, W. Z. Naturforsch., B: Chem. Sci. 1988,
43, 1206–1208.
(32) Martinsen, A.; Songstad, J. Acta Chem. Scand., Ser. A 1977, 31, 645–
650.
12008 Inorganic Chemistry, Vol. 47, No. 24, 2008

Characterization of the Hexaazidophosphate(V) Ion
52.62; H, 3.68; N, 32.39%; found: C, 52.67; H, 3.48; N, 32.42%.
Supporting Information Available: Spectroscopic data of 2,
crystal data and structure refinement, projections of the crystal
structure, details of the density functional theory and Hartree-Fock
calculations including results of frequency calculations of the S6
and D3 minimum structures, IR and Raman intensities and activities,
TG and DSC graphs, UV absorption spectra of 2 and (PPN)N3,
and empirical lattice energy calculations. This material is available
free of charge via the Internet at http://pubs.acs.org.
IR (cm-1) νj ) 2113 (nujol), 2116 (MeCN), 2115 (CH2Cl2), 2112
1
(THF). Details of IR; Raman; and H, 13C, 31P, and 14N NMR
spectra can be found in the Supporting Information.
Acknowledgment. We thank the EPSRC (RA114873) for
funding and Dr. R. Devonshire and Dr. V. Ramos for
recording Raman spectra.
IC801536Z
Inorganic Chemistry, Vol. 47, No. 24, 2008 12009