This work was funded by the Defense Advanced Research Projects Agency, the Air Force Office of Scientific Research and the National Science Foundation. R.H. thanks the Deutsche Forschungsgemeinschaft for a postdoctoral fellowship. We thank Prof. G. A. Olah, and Drs. A. Morrish, D. Woodbury, and M. Berman, for their steady support, and Dr. R. Wagner for his help and stimulating discussions.
High-Energy-Density Materials: Synthesis and Characterization of N5+[P(N3)6]−, N5+ [B(N3)4]−, N5+ [HF2]−⋅n HF, N5+ [BF4]−, N5+ [PF6]−, and N5+ [SO3F]−†
Article first published online: 16 SEP 2004
Copyright © 2004 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Angewandte Chemie International Edition
Volume 43, Issue 37, pages 4919–4924, September 20, 2004
How to Cite
Haiges, R., Schneider, S., Schroer, T. and Christe, K. O. (2004), High-Energy-Density Materials: Synthesis and Characterization of N5+[P(N3)6]−, N5+ [B(N3)4]−, N5+ [HF2]−⋅n HF, N5+ [BF4]−, N5+ [PF6]−, and N5+ [SO3F]−. Angew. Chem. Int. Ed., 43: 4919–4924. doi: 10.1002/anie.200454242
- Issue published online: 16 SEP 2004
- Article first published online: 16 SEP 2004
- Manuscript Received: 11 MAR 2004
- high energy-density materials (HEDM);
- polynitrogen chemistry;
- vibrational spectroscopy
During the past two decades, polynitrogen containing compounds have received increasing attention as promising candidates for high energy-density materials (HEDM).1–17 While most of the efforts were devoted to theoretical studies, the long-known existence of the stable azide anion (N3−)18 and the recent syntheses of stable salts of the pentanitrogen cation (N5+)1–3 have demonstrated the feasibility of experimentally pursuing polynitrogen-containing materials. The only known direct method for preparing N5+ compounds is their synthesis from an [N2F]+ salt with HN3 in HF solution according to Equation (1)((1)).1, 2
This direct synthesis route is restricted by the small number of [N2F]+ salts available. Except for N2FAsF6 and N2FSbF6 and reports on unstable N2FBF419 and N2FPF620 salts, no other [N2F]+ compounds have been reported.
For a successful metathetical reaction, each ion must be compatible with the solvent, and both starting materials and one of the products must be highly soluble, while the second reaction product must have low solubility. Because of its highly oxidizing nature, N5SbF6 is compatible with only a limited number of solvents, for example, HF, SO2 and CHF3, thus severely restricting the general usefulness of the metathetical approach. Because SbF5 is among the strongest known Lewis acids,21 the displacement of SbF5 in N5+ [SbF6]− by a stronger Lewis acid is also rarely feasible. Therefore, the development of a more general method for the syntheses of N5+ compounds is desirable. Furthermore, in the interest of preparing N5+ salts of higher energy content, the combination of N5+ with highly energetic counterions was pursued. Previous attempts to combine N5+ with either N3−, [ClO4]−, [NO3]−, or [N(NO2)2]− had been unsuccessful.22
While in theory, F− abstraction from FN5 by a strong Lewis acid, such as SbF5, could provide a general synthesis for N5+ salts [Eq. (3)((3))], the required FN5 precursor is unknown.
Theoretical studies23, 24 identified at least six vibrationally stable isomers of FN5 but, in accordance with experimental results, the predicted lifetimes of these species are only in the nanosecond range.23
During attempts to prepare N5+ [N(CF3)2]− by metathesis from N5+ [SbF6]− and Cs+ [N(CF3)2]− in HF solution at −78 °C [Eq. (4)((4))], the expected CsSbF6 precipitate was formed and removed by filtration.
However, after pumping off all volatile material from the filtrate at −64 °C, the low-temperature Raman spectrum of the resulting clear liquid residue exhibited only bands attributable to N5+ (Figure 1). This finding reminded us of a situation encountered 24 years ago with the metathetical reaction of NF4SbF6 and CsF in HF [Eq. (5)((5))].
This reaction resulted in the formation of thermally unstable, liquid NF4HF2⋅n HF,25 which exhibited characteristics very similar to those observed in the above N5+ reaction, that is, a failure to observe anion bands because a polybifluoride anion is an extremely weak Raman scatterer. The additional formation of N5HF2⋅n HF in the reaction in Equation (4) can be explained if liquid HF is capable of displacing HN(CF3)2 from its [N(CF3)2]− salts according to Equation (6)((6)).
The above assumptions were confirmed by carrying out a reaction of N5SbF6 with CsF in anhydrous HF at −64 °C which resulted in the expected precipitation of CsSbF6 and the formation of a polybifluoride of N5+ [Eq. (7)((7))].
N5HF2⋅n HF was isolated as a clear, colorless liquid after filtering off the CsSbF6 precipitate and removing all volatiles at −64 °C from the filtrate. The observed low-temperature Raman spectrum was identical to that shown in Figure 1. It exhibits, in addition to some weak bands due to the Teflon-FEP sample container and a trace of [SbF6]− from the starting material, only bands due to N5+. The experimental Raman frequencies and assignments are listed in Table 1. On warming to room temperature, the N5HF2⋅n HF salt decomposed under formation of trans-N2F2, NF3, and N2, which were identified by checking for noncondensible gas at −196 °C and FT-IR spectroscopy.
|Observed frequency (cm−1) and relative intensity||Assignments|
|N5HF2⋅n HF||N5SbF62||N5PF6||N5BF4||N5SO3F||N5+ (C2v)||MF6− (Oh)||[BF4]− (Td)||[SO3F]− (Td)|
|3357 vw||3364 w||(ν1 + ν3 + ν9)(B2)=3358|
|3334 vw||3337 w||(ν1 + ν8)(B2)=3323|
|3079 vw||3082 mw||(ν2 + ν7)(B2)=3077|
|2681 vw||2685 w||(ν1 + ν9)(B2)=2682|
|2279 (10)||2270 m||2268 (9.4)||2273 ms||2269 (10)||2283 (10)||2271 (10)||ν1(A1)|
|2218 (2.2)||2205 s||2205 (2.0)||2219 s||2209 (1.3)||2221 (3.0)||2210 (2.2)||ν7(B2)|
|1921 vw||1926 w||(ν3 + 3ν9)(B2)=1914|
|1891 vw||1891 w||(ν8 +2ν9)(B2)=1883|
|1240 vw||comb. bands|
|1092 ms||1099 s||(ν3 + ν9)(B2)=1086[a]|
|1064 s||1072 s||ν8(B2)|
|902 vvw||(ν5 + ν6)(B2)=903|
|877 (1.3)||871 w||872 (0.6)||869 (0.6)||880 (1.5)||871 (2.3)||ν2(A1)|
|840 (0.9)||835 vw||837 (0+)||826 (0+)||837 (0.7)||829 (1.7)||(2ν9)(A1)=828[b]|
|672 (2.7)||672 (1)||672 s||668 (2.2)||674 (2.7)||669 (3.2)||ν3(A1)|
|655 vs||881 s, 839 vs}||ν3(F1u)|
|652 (10)||750 m||747 (3.8)||ν1(A1g)|
|574 (2.4), 564 (2.4)}||ν3(A1)|
|582 w||571 (0.8)||563 vs||578 (0.3)||ν2(Eg)|
|481 (0.7)||478 (0+)||473 w||476 (0.7)||477 (1.4)||ν5(A2)|
|447 w||447 w||?|
|422 (0.6)||425 ms||426 (0.4)||420 (1.9)||ν6(B1)|
|413 (0.6)||412 mw||416 (0+)||416 (0+)||412 (0.5)||407 (2.0)||ν9(B2)|
|284 vs||563 vs||ν4(F1u)|
|282 (2.8)||473 w||474 (0.6)||ν5(F2g)|
|202 (5.8)||204 (5.0)||211 (2.5)||202 (4.8)||203 (5.7)||ν4(A1)|
|107 (5.0)||120 (1.6)||113 (2.0)||111 (4.5)||lattice vibrations|
The usefulness of the N5HF2⋅n HF salt as a reagent for the synthesis of other N5+ salts by displacement reactions with Lewis acids stronger than HF was explored by treating it with PF5, BF3, and HSO3F,26, 27 resulting in the formation of N5PF6, N5BF4, and N5SO3F, respectively, according to Equations (8)((8))–(10)((9)), ((10)).
All these new salts are white, marginally stable solids that were characterized by NMR and vibrational spectroscopy. The 14N NMR spectrum of N5PF6 was recorded in HF at −40 °C. It showed a strong resonance at δ=−165.1 ppm for the Nβ atoms and a very broad line at about δ=−101 ppm for the terminal Nα atoms, and is in good agreement with previously published values for N5+ salts.1–3 In the 14N NMR spectra of N5BF4 and N5SO3F in HF at −40 °C, the resonances for the Nβ atoms were observed at δ=−164.3 ppm and δ=−164.7 ppm, respectively. The experimental vibrational frequencies and assignments of the three salts and, for comparison, of N5SbF6 are listed in Table 1. The observed Raman and IR spectra of N5PF6 are shown in Figure 2, and the Raman spectra of N5BF4 and N5SO3F are shown in Figure 3 and Figure 4, respectively. They establish beyond any doubt the composition of these salts1–3, 28–30 and their ionic nature.
Whereas the N5+ ion is a highly energetic ion with a calculated endothermicity of 351.6 kcal mol−1,22 all of its known salts contained non-energetic counterions.1–3 Although a significant advance in potential performance was achieved by successfully doubling the number of polynitrogen ions in a salt by formation of a 2:1 salt [N5+]2[SnF6]2−,3 salts containing energetic counterions were still missing. Attempts to combine the N5+ ion with the energetic anions, [ClO4]−, [NO3]− and N3− by metathetical reactions failed, and a recent theoretical analysis showed that, after inclusion of entropy corrections, N5+ N3− is unstable by 76 kcal mol−1 with respect to spontaneous decomposition to N3 and N2.22 In spite of these challenges, we have now successfully synthesized two highly energetic N5+ salts.
The metathetical reaction between N5SbF6 and NaP(N3)6 in SO2 proceeded with the expected precipitation of NaSbF6 and the combination of the N5+ ion with the energetic ion P(N3)6−30 to form N5P(N3)6 [Eq. (11)((11))].
However, the compound is extremely shock sensitive and violently explodes upon the slightest provocation or warming towards room temperature (see Figure 5). In addition to its very high energy content, this salt is remarkable for its high nitrogen content of 91.2 wt %.
Again, the salt is extremely shock-sensitive and explodes on warming towards room temperature. Its nitrogen content of 95.7 wt % significantly exceeds even that of N5P(N3)6 and any other known, solid high-nitrogen compound. There are only five other compounds whose nitrogen content exceeds 90 wt %. These are: [NH4]+ N3− (93.3 %), [N2H5]+ N3− (93.3 %), [N2H5]+ N3−⋅N2H4 (91.6 %), 2H-tetrazolylpentazole (90.6 %), and Li+ [B(N3)4]− (90.4 %).33 Attempts to carry out the above metathetical reactions with CsP(N3)6 and CsB(N3)4 in HF solution were unsuccessful because HF reacts with the polyazido anions to give [PF6]− and [BF4]−, and lead to the isolation of N5PF6 and N5BF4, respectively. Both polyazido salts were identified and characterized by low-temperature Raman spectroscopy.
The experimental vibrational frequencies and tentative assignments for N5P(N3)6 and N5B(N3)4 are given in the Experimental Section. The observed Raman spectra of N5P(N3)6 and N5B(N3)4 are shown in Figure 6 and Figure 7, respectively. In addition to high energy densities of about 2 kcal gram−1 and extremely high sensitivities, these compounds exhibit the typical high detonation velocities of covalent azides which render the handling and further characterization of these compounds particularly difficult.
Caution! Azides and N5+compounds are highly endothermic and can decompose explosively under various conditions! N5+compounds are highly energetic oxidizers.1–3Contact with potential fuels must be avoided. These materials should be handled only on a scale of less than 2 mmol. The polyazides in this work are extremely shock-sensitive. Because of the high energy content and high detonation velocities of these azides, their explosions are particularly violent and can cause, even on a one mmol scale, significant damage. The use of appropriate safety precautions, such as face shields, heavy leather welding suits, leather gloves, and ear plugs is mandatory.34Teflon containers should be used, whenever possible, to avoid hazardous fragmentation. Ignoring safety precautions can lead to serious injuries!
All reactions were carried out in Teflon-FEP (FEP=perfluoroethylenepropylene polymer) ampules that were closed by stainless steel valves. Volatile materials were handled in stainless steel/Teflon-FEP or grease-free Pyrex-glass vacuum lines.35 Nonvolatile solids were handled in the dry argon atmosphere of a glove box. All reaction vessels and the stainless steel line were passivated with ClF3 prior to use.
Infrared spectra were recorded in the range 4000–400 cm−1 on a Midac FT-IR model 1720 at a resolution of 1 cm−1. Spectra of solids were obtained by using dry powders pressed between AgCl windows in an Econo press (Barnes Engineering Co.). Raman spectra were recorded in the range 4000–80 cm−1 on a Bruker Equinox 55 FT-RA spectrophotometer using a Nd:YAG laser at 1064 nm with power levels of 200 mW or less. Pyrex melting point tubes that were baked out at 300 °C for 48 h at 10 mTorr vacuum or 9-mm o.d. Teflon-FEP tubes with stainless steel valves that were passivated with ClF3 were used as sample containers. 14N NMR spectra were recorded unlocked at 36.13 MHz on a Bruker AMX 500 spectrometer using solutions of the compounds in DMSO in sealed standard glass tubes. Neat CH3NO2 (0.00 ppm) was used as the external reference.
The N2FSbF6 starting material was prepared from cis-N2F2 and SbF5 in anhydrous HF solution.19, 20, 36–39 N5SbF6 was prepared from N2FSbF6 and HN3 in HF,2 NaP(N3)6 was prepared from PCl5 and NaN3,31 and NaB(N3)4 from NaBH4 and HN3.32 The HF (Matheson Co.) was dried by storage over BiF5 (Ozark Mahoning).40 PCl5 (Aldrich) was purified by sublimation in a dynamic vacuum. The CsF (KBI) was fused in a platinum crucible, transferred while hot to the dry box, and finely powdered. BF3 (Matheson), PF5 (Ozark Mahoning), NaN3 (Aldrich), NaBH4 (Aldrich), and HSO3F (Aldrich) were used without further purification.
N5HF2⋅n HF: A solution of CsF (1.00 mmol) in HF (2 mL) was siphoned through a Teflon–FEP tube into a Teflon–FEP ampule containing a solution of N5SbF6 (1.00 mmol) in HF (3 mL) at −64 °C. Immediately, a white precipitate was formed. The reaction mixture was stirred for 10 min to ensure complete reaction. The mixture was allowed to settle, and the supernatant liquid was siphoned into a second Teflon–FEP ampule kept at −64 °C. The CsSbF6 residue was washed twice with HF (about 1 mL each time). The HF was pumped off from the combined liquids at −64 °C, leaving behind a colorless liquid (0.156 g; weight calculated for 1.00 mmol of N5HF2⋅2.5 HF: 0.159 g).
N5PF6 and N5BF4: Excess PF5 or BF3 (2.0 mmol) was condensed at −196 °C into an ampule containing a frozen solution of N5HF2⋅n HF (1.00 mmol) in HF (1 mL). The temperature was raised to −64 °C and the reaction mixture kept at this temperature for 1 h to ensure complete reaction. All volatile material was pumped off at −64 °C, leaving behind a white solid (N5PF6: 0.220 g, weight calculated for 1.00 mmol of N5PF6: 0.215 g; N5BF4: 0.167 g; weight calculated for 1.00 mmol of N5BF4: 0.157 g).
N5SO3F: At −64 °C, a solution of HSO3F (1.00 mmol) in HF (2 mL) was added to a solution of N5HF2⋅n HF (1.00 mmol) in HF (1 mL). The reaction mixture was stirred for 30 min at this temperature to ensure complete reaction. All volatiles were pumped off at −64 °C leaving behind a white solid (0.175 g; weight calculated for 1.00 mmol of N5SO3F: 0.169 g).
N5P(N3)6 and N5B(N3)4: At −64 °C, a solution of N5SbF6 (0.50 mmol) in SO2 (3 mL) was added to a solution of NaB(N3)4 or NaP(N3)6 (0.50 mmol) in SO2 (3 mL), respectively. After the mixture had settled, the liquid phase was transferred into another Teflon–FEP ampule that had been cooled to −64 °C, and the remaining NaSbF6 was washed twice with about SO2 (1 mL). Pumping on the collected liquid phase at −64 °C gave a white solid. N5P(N3)6: 0.184 g, expected for 0.50 mmol: 0.177 g; Raman (50 mW, −80 °C): =2266(10.0) (N5+ν1), 2203(7.5) (N5+ν7), 2182(5.4)/2074(2.9) (P(N3)6−νasN3), 1302(4.7) (P(N3)6−νsN3), 873(3.9) (N5+ν2), 730(7.4) (P(N3)6−νPN), 666(8.0) (N5+ν3), 522(5.0) ((P(N3)6−δN3), 483(4.6) (N5+ν5), 419(4.7) (N5+ν9), 458(4.7) ((P(N3)6−δPNN), 327(4.9) ((P(N3)6−δPNN), 203(9.1) (N5+ν4)
N5B(N3)4: 0.137 g; expected for 0.50 mmol: 0.124 g; Raman (50 mW, −80 °C): =2269(1.9) (N5+ν1), 2207(1.2) (N5+ν7), 2172(5.4)/2148(2.0) (B(N3)4−νasN3), 1334(2.9)/1292(3.7) (B(N3)4−νsN3), 875(3.1) (N5+ν2), 664(3.6) (N5+ν3), 581(3.0)/532(4.7) (B(N3)4−), 483(2.3) (N5+ν5), 421(2.1) (N5+ν9), 293(2.4) (B(N3)4−), 203(2.6) (N5+ν4), 189(5.0)/165(6.8)/123(10.0) (B(N3)4−).
Dedicated to Professor Herbert Roesky on the occasion of his 70th birthday
- 7Chem. Ind. 2000, 140, and references therein; a compilation of data for N2 to N10 can be found at http://www.qtp.ufl.edu/∼bartlett/downloads/polynitrogen.pdf.,
- 19Russ. J. Inorg. Chem. 1968, 13, 1345., ,
- 30Anwendungen der Schwingungsspektroskopie in der Anorganischen Chemie, Anorganische und Allgemeine Chemie in Einzeldarstellungen, VII, Springer, Heidelberg, 1996.,
- 32aZ. Naturforsch. 1954, 96, 497;, ,
- 34cR. Haiges, J. A. Boatz, S. Schneider, T. Schroer, M. Yousufuddin, K. O. Christe, Angew. Chem. 2004, 116, 3210; Angew. Chem. Int. Ed. 2004, 43, 3148.
- 37Inorg. Chem. 1966, 5, 1971.,