A biography of potassium complexes as versatile, green energetic materials

Abstract

Potassium-based energetic complexes are environmentally friendly compared to their traditional heavy metal counterparts. They also have higher energy densities and better thermal stabilities than the parent organic compounds due to the deprotonation and introduction of the potassium cation. According to the anion ligands, they can be classified as nitrophenol-, furazan-, furoxan-, triazole-, tetrazole-, and their derivative-based potassium complexes. More than 20 potassium complexes have been prepared, with their crystal structures analyzed and properties characterized, demonstrating their superiority to traditional energetic materials. Multi-coordinated potassium cations facilitate the formation of energetic three-dimensional metal–organic frameworks, making them better candidates as energetic materials for both defense and civil applications.

---

Li-Yang Chen

Li-Yang Chen was born in 1994 in Sichuan, P. R. China. She enrolled the in Beijing Institute of Technology in 2013 as an undergraduate.

Jian-Guo Zhang

Jian-Guo Zhang was born in 1974 in Hebei, P. R. China. He received a B.S. in chemistry at the Hebei Normal University, an M. S. in physical chemistry and a Ph. D. in applied chemistry at the Beijing Institute of Technology. He has been a faculty member at the Beijing Institute of Technology since 2000, where he has served as the head and vice president of the energetics materials department for the school of mechatronical engineering. In 2012, he was named a University Distinguished Young Professor. His research interests include the molecular design, syntheses, characterization, properties, and application of energetic materials, and hydrogen storage materials.

Zun-Ning Zhou

Zun-Ning Zhou was born in 1969 in Shandong, P. R. China. He earned his B. S. in Chemistry from the Shandong University in 1994 and a Ph. D. in chemical engineering from Nanjing University of Science & Technology in 2004. He then spent two years as a postdoctoral fellow at China Ship Industry Co. As a visiting scholar, he completed a six-month research visit at the Department of Chemical and Materials Engineering University of Alberta, Canada in 2012. He is now an Associate Professor of State Key Laboratory of Explosion Science and Technology at Beijing Institute of Technology.

Tong-Lai Zhang

Tong-Lai Zhang was born in 1960 in Shandong, P. R. China. He received a B. S. and a Ph. D. in chemistry at the Nanijing Institute of Technology. He has been a faculty member at the Beijing Institute of Technology since 1995, after finishing his postdoctoral research at the University of Science and Technology, Beijing. He was given the title of Professor in 1999.

---

1. Introduction

For billions of years, the potassium cation has been indispensable for sustaining life on earth. Their main function is maintaining the balance of osmotic pressure in biological cells and living beings cannot survive once this balance is broken.¹ Potassium is abundant in both minerals and sea water. After benefiting from it unconsciously since the very beginning, human beings have learned to utilize potassium for both munitions and civil fields.^2–11

National defense is of vital importance for every country, and ammunitions are indispensable for the military. Since the invention of gunpowder, research on improving the properties of energetic materials has never stopped.^12–18 From fulminate (late 18th century) to tetrazene (1910), from diazodinitrophenol (1859) to lead azide (1890) and lead styphnate (1914), higher sensitivity and stronger explosions are the constant pursuit of energetic material research. However, defense applications also need to take environmental impacts into consideration. Traditional energetic materials containing lead or mercury have polluted the environment over the past decades. Mercury is highly toxic to humans, damaging the oral cavity, digestive tract, kidneys, and nervous system. It is readily absorbed via skin contact and also bioaccumulates along the food chain. Therefore, the impact of mercury poisoning in humans is significant. Lead, which is contained in lead azide (Pb(N₃)₂) and lead styphnate, mainly destroys the human digestive, hematologic, and nervous systems. Mental impairment caused by lead poisoning in children is particularly irreversible. As a relatively new type of substitute for traditional energetic materials, potassium energetic complexes have proven to be desirable and competitive in recent years. On the one hand, potassium is eco-friendly while the traditional heavy metal components are harmful to the planet.¹⁹ On the other hand, potassium cation substitution of hydrogen atoms also generally improves the performance of explosives, such as raising the stability and energy density of the system.

In this review, a number of potassium energetic complexes are discussed. They are generally classified according to their organic anions. Nitrophenol-based complexes are believed to have desirable performance due to their energetic aromatic rings with nitro groups. While the π-conjugated system contributes to the stability of the complex, the nitro group helps ameliorate the oxygen balance (OB) and improve the energy in spite of enhanced mechanical sensitivities at the same time. The concept of high energy density compounds (HEDCs) was proposed decades ago, and has been a goal of research since. HEDCs come in three typical categories: (1) furazan and furoxan compounds; (2) low sensitivity HEDC, represented by 1,1-diamino-2,2-dinitroethylene (FOX-7) and its derivatives; and (3) high-nitrogen-content high-energy materials (HNC-HEMs), represented by triazoles, tetrazoles, and their derivatives. The combination of several heterocycles is also found to be beneficial, owing to the vast possibilities of tailoring the energetic performance. Therefore, efforts have been made to coalesce heterocycles to produce novel properties. Potassium energetic complexes of these organics have shown excellent performance, as expected from their organic groups.

As to potassium energetic complexes for civil use, they could be generally divided into two kinds, organic and inorganic. The organic ones, based on energetic materials, can be used for civil applications, mainly in pyrotechnics. Traditional fireworks, while dazzling us with their brilliant and spectacular colors, also have the drawback of introducing significant pollution over a large area. The latter is attributable primarily to the toxic heavy metals used for coloring, such as barium, lead, copper, and strontium, which form heavy metal-containing aerosols.^20,21 The lead and barium aerosols are of special causes of concern due to their toxic soluble salts.²² In comparison, potassium complexes have good performance as well as lower toxicity. Their production is also cheaper. The inorganic ones, meanwhile, are basically used as oxidizers. Green potassium complexes, therefore, can play versatile and colorful roles. It is necessary to study them comprehensively in order to realize their many possibilities.

2. Highlights of potassium complexes

Although traditional energetic materials have served us for a long time and been continuously improved, their limitations are evident when compared to potassium energetic complexes. Generally speaking, potassium energetic materials surpass traditional ones in terms of both physico-chemical properties and explosive performances, and most importantly, their non-polluting nature.

The preparation of potassium complexes is usually facile and is based on basic acid–base neutralization reactions. Specifically, since most of their parent organics (parent organics refer to organics offering anion ligands to the potassium complex) are acidic, the potassium salts are often synthesized by adding KOH or K₂CO₃ into solutions of the respective parent organics. The number of potassium cations introduced per molecule is determined by the loss of (deprotonated) hydrogen atoms, and influenced by the acidity/basicity of the reactants as well as the structure of the organics. Sometimes, metathesis reactions (usually with sodium salts of the parent organics) and nitration reactions are involved, and more complicated synthesis methods are occasionally employed.

Improving the density of primary explosives has been persistently pursued for enhancing their explosive properties. A higher density usually brings higher explosive velocity and higher explosive pressure. The relatively low explosive velocity and explosive pressure can be a shortcoming for common primary explosives. While the density of presently used primary explosives are generally far below 2.000 g cm⁻³, potassium complexes usually have quite high densities, especially the multi-substituted ones. As we can see later from the tables, the densities of most potassium energetic materials are more than 1.900 g cm⁻³, and with nitro-rich anions this value can easily exceed 2.0 g cm⁻³.

Notably, in spite of the non-energy of the potassium cation itself, potassium energetic complexes can have desirable explosive properties, including high values of detonation pressure, detonation velocity, and heat of detonation, as well as good thermal stability. Their explosive velocity and explosive pressure can easily go above 8000 m s⁻¹ and 30.0 GPa, respectively, while other primary explosives have lower values. Furthermore, while the parent organics possess satisfactory thermal and explosive properties, their potassium salts usually exhibit even better performances, owing to deprotonation and introduction of potassium cation. In principle, the deprotonation of the organic ligands tends to enhance the thermal stability of the system, due to the overcompensation of the electron-withdrawing groups.²³ Besides, the ionic nature resulting from replacing hydrogen atoms by potassium cations can increase the lattice energies in the crystalline state along with the densities.

Another characteristic of potassium complexes is the three-dimensional (3D) network, which contributes to better thermal stability and lower sensitivity, making them suitable for next-generation energetic materials.^24–26 The presence of potassium cations results in the packing of anions and cations through coordinated bonds to form a network structure. The system benefits significantly from the potassium cation owing to its high coordination number. As an alkali metal ion, the potassium cation has a large ionic radius and permits more coordination ligands around it, thus making the molecular system much more compact. Meanwhile, the potassium cation has fewer d-electrons so that it exerts less repulsive force on the ligands, and the coordination atoms with lone-pair electrons could easily bond to it. All potassium compounds are coordination compounds or salts rather than covalent compounds because of the large ionic radius of the potassium cation.

Last but not least, apart from being primary explosives due to their relatively high sensitivity and rapid transition from deflagration to detonation (DDT) when subjected to heat or flame, potassium energetic complexes also play roles in ballistic modification, pyrotechnics, and as oxidizers.^27–30

3. Preparation and characterization of potassium energetic complexes

3.1. Nitrophenol-based potassium energetic complexes

Molecules containing nitro groups (NO₂) are good candidates for explosives because the nitro group provides the essential oxygen for combustion. It is often assumed that a compound with an OB closer to zero will be more brisant, powerful, or sensitive.³¹ A molecule's OB value and its potential usefulness as an energetic material can be increased by the introduction of nitro groups. On the other hand, the hydroxyl group is of great help in forming 3D networks, due to the intramolecular and intermolecular hydrogen bonds, thereby promoting the sensitivity and stability of the system.

The corresponding potassium complexes have been easily prepared by reacting 2,4-dinitrophenol (DNP), picric acid (PA), 2,4-dinitroresorcinol (DNR), trinitro-resorcinol (TNR), or trinitro-phloroglucinol (TNPG) with KOH or K₂CO₃ aqueous solutions (Scheme 1). Potassium picrate (KPA, 3) can also be prepared with K₂CO₃. Most of the synthesis reactions are carried out in aqueous solutions except those of [K(DNP)(H₂O)_0.5]_n (1) and [K(HTNR)(H₂0)]_n (5a), which are completed in ethanol. Yields of these nitrophenol-based potassium complexes are usually as high as above 80%, up to an 86% yield for [K₂(TNR)(H₂0)]_n (5b). However, the yield of [K(HDNR)(H₂DNR)(H₂O)]_n (2) can be quite low. There are three TNPG potassium complexes, namely mono- (4a), di- (4b), and tri- (4c) substituted TNPG salts. Likely, mono- (5a) and di- (5b) substituted salts of TNR are obtained by using different solvents and controlling the ratio of the reactants.

Scheme 1 Synthesis routes of 1–5.

The coordination numbers of potassium cations in these complexes are rather high, thereby contributing to better thermal stability and lower sensitivity through the formed 3D network structures. This can also change the coordination environment of the potassium cation. Interestingly, the potassium cations in these complexes only coordinate with O atoms from the phenolic hydroxyl groups, nitro groups, or coordinated water molecules. The coordination environment of the central potassium cation varies according to the ligands, even in the same complex and for the same ligand, due to the different ligand sources and coordination numbers.

The central ions K (1) and K (2) in complex 1 (Fig. 1) show different coordination numbers (10 and 7, respectively). The potassium cations are coordinated with the O atoms of phenolic hydroxyl and nitro groups rather than those of water molecules. The 10 coordination bonds are distributed with inversion symmetry, forming two steady-going aberrant six-membered rings and two aberrant four-membered rings. The 7 coordination bonds also have high symmetry, forming two four-membered rings.

Fig. 1 The extended coordination scheme of the potassium cation in complex 1. Reproduced from ref. 37 with permission from WILEY-VCH, copyright [2004].

Complex 5b has a similar structure. The potassium cation K (1) and K (2) have coordination numbers of 9 and 8, respectively (Fig. 2). While both coordinate with O atoms of phenolic hydroxyl and nitro groups, K (1) also coordinates with water. Four six-membered rings are formed in a single molecule.³²

Fig. 2 The extended coordination scheme of the potassium cation in 5b. Reproduced from ref. 32 with permission from the Journal of inorganic chemistry, copyright [2003].

Complex 4c also has potassium cations in two different coordination modes (Fig. 3). However, both are 8-coordinated with water molecules and TNPG³⁻, although their specific coordination conditions are different. It is noticeable that K (1) and K (2) are coordinated with each other in complex 4c, which is rare among such compounds.^33,34

Fig. 3 The extended coordination scheme of the potassium cation in 4c. Reproduced from ref. 33 with permission from WILEY-VCH, copyright [2007].

The coordination environments of complexes 2 and 5a are relatively simple. The central potassium cations are octa- (Fig. 4) and hepta- (Fig. 5) coordinated by the O atoms of the nitro/phenolic hydroxyl groups and water molecules, respectively. Multi-membered rings are also formed in complexes 2 and 5a, yet significantly less than those of other nitrophenol-based analogues due to the simplicity of coordination. As a result, they have lower thermal stability. On the other hand, these anions present multi-dentate ligands, and thus, form an infinite cross-linked 3D network in the lattice. These coordination bonds have coactions and higher strength thereby providing high stability and good heat resistance.

Fig. 4 The extended coordination scheme of the potassium cation in 2. Reproduced from ref. J. G. Zhang, Z. M. Li, J. W. Liu, T. L. Zhang, X. Q. Niu, C. C. Chen, L. Yang and K. B. Yu, synthesis and structure investigation of a novel 3-dimensional potassium super-molecular compound based on 2,4-dinitro resorcinol, Chin. J. Chem., 2011, 29, 913–918 with permission from WILEY-VCH, copyright [2011].

Fig. 5 The extended coordination scheme of the potassium cation in 5a. Reproduced from ref. 41 with permission from Chinese Chemical Society, copyright [2003].

The energetic and structural properties of complex 3 have been studied with periodic density functional theory (DFT) calculations in order to obtain insight into its chemical properties and detonation mechanism. The results reveal that complex 3 belongs to the family of ionic crystals, and the Mulliken populations indicate the tendency of the C–N bond to rupture upon impact or other compression stimuli.³⁵ Other calculated details, such as the natural bond orbital (NBO) charges, optimized geometries, vibrational frequencies and intensities, and the standard thermodynamic functions, were obtained via the statistical thermodynamic method.³⁶

As mentioned earlier, potassium salts usually exhibit better performances than their organic counterparts, including higher thermal stability and lower sensitivity, owing to the deprotonation of hydrogen atoms and the introduction of potassium cation. On the one hand, the deprotonation of the organic ligands enhances the thermal stability of the system due to the overcompensation of the electron-withdrawing groups. On the other hand, the ionic nature resulting from replacing hydrogen with potassium can cause higher lattice energies in the crystalline state along with higher densities. By comparing the properties of TNPG substitutes with different K-substituted numbers (4a, 4b, and 4c), more potassium cation substitution clearly corresponds to a more stable complex (Table 1). The sensitivity of 4a, 4b, and 4c, as shown in Table 1, is negatively correlated with the degree of substitution: the more potassium cations in the molecule, the less sensitive the complex. However, all these compounds are regarded as sensitive to mechanical stimuli (impact and friction). There is an explanation for the higher sensitivity of the potassium salts than the parent TNPG: the existence of metal atoms makes the phenol oxygen atoms more reactive, which is in accordance to the “principle of the easiest transition” concerning monomer explosives. It can also be explained in terms of decreased band gaps.

Table 1 Kinetic parameters and enthalpies of thermal decomposition, results of sensitivity tests of 4^a

Complexes	Ek (kJ mol^−1)	Eo (kJ mol^−1)	ln AK (s^−1)	IS (J)	FS (%)	Flame sensitivity (h50, cm)	Color
a Subscripts k and o denote calculation results by Kissinger's method and Ozawa–Doyle's method, respectively. FS was determined on a model MGY-1 pendulum friction sensitivity apparatus by applying a standard method. Flame sensitivity was tested on a model HGY-1 flame sensitivity apparatus.
4a	249.9	244.9	26.62	1.14	88	29.21	Yellow
4b	381.4	371.2	35.76	1.77	26	34.56	Orange-yellow
4c	585.9	566.2	52.42	2.72	12	42.87	Red

---

Except for 3 and 4, all these complexes contain coordinated water molecules. Coordinated water contributes to the stability of molecular systems because it contains no explosive groups or nitrogen atoms (i.e., not an energetic ligand). It also increases the hydrogen bonding in the system, thereby reducing the mechanical sensitivity. The O atoms of water molecules provide electrons to form more coordination bonds, which further enhance the stability.

Thermal analysis confirms that the thermal decomposition process features high-temperature dehydration before further decomposition. The dehydration of complex 1 takes place between 126–156 °C, with a dehydration enthalpy of 34 kJ mol⁻¹. The single exothermic step at the high temperature of 323 °C has a large enthalpy of −945 kJ mol⁻¹, contributing to the wide application of 1 in ammunition and civil fields as an energetic material.³⁷ The decomposition process of complex 5b resembles that of 1. The endothermic dehydration occurs at 131.8 °C, and the exothermic decomposition shows a top temperature of 319.0 °C. However, its analogue 5a decomposes in more stages. The second endothermic peak with the highest temperature at 115.7 °C corresponds to dehydration. There are two exothermic stages at 234.1 °C and 327.3 °C.

The thermal behaviors of 4a, 4b and 4c are similar, characterized by decomposition in two major stages: endothermic, complete dehydration in the temperature range of 78.3–118.6 °C, and another exothermic process.³⁸ Exothermic decomposition of 4a (156–310 °C) and 4b (235–360 °C) both come with two apparent peaks while that of 4c (270–320 °C) possesses only one at 295 °C, as observed by differential scanning calorimetry (DSC). Obviously, their thermal resistance decreases in the sequence: 4c > 4b > 4a.³⁹

The final decomposition residues of complexes 1, 4a, 5a, and 5b after heating to above 350 °C are all KCN, while those of the complexes of 4b and 4c also contain K₂CO₃.⁴⁰ The residue rates of 1 and 5b are rather low (21.35% and 11.62%, respectively) and comparable to that of 3. Yet, complex 5a has a much lower rate of just 2.9%; this might be attributed to its two exothermic decomposition stages.⁴¹ It is worth noting that during the decomposition of complex 3, volatile potassium isocyanate (KNCO) appears and is ultimately converted to K₂O and K₃N. A hypothetical application of these volatile metal isocyanates is the damping of high-frequency acoustic oscillations in solid rocket motors due to their continued reaction to form metal oxide particles. Flash pyrolysis of complex 4 produces much more KNCO than 3, making 4 a ballistic addictive candidate while 3 produces too little KNCO for use as a ballistic modifier.²⁷

3.2. Energetic potassium complexes of furazan and furoxan compounds

Energetic materials with a zero or positive OB are assumed to be “greener” than those with a negative OB because the former converts all the carbon into carbon dioxide and all the hydrogen into water with a minimum amount of liberated toxic gases. While introducing nitro groups is known as an efficient way to improve the OB, furoxan can function as a “hidden” nitro group that possesses a high oxygen content (37.2%) among the N-heterocyclic rings. Furthermore, both the density and velocity of detonation of the compound can be improved through the replacement of nitro groups by furoxan groups. The introduction of amino groups increases both the density and heat resistance, as well as decreases impact sensitivity.^42,43 Derivatives of furazan and furoxan, therefore, deserve further investigation. The three potassium complexes of furazan and furoxan compounds we are about to discuss, namely the potassium complex of 4,5-bis(dinitromethyl)furoxanate (6), potassium 4,6-dinitro-benzofuroxan (KDNBF, 7), and potassium 5,7-diamino-4,6-dinitrobenzofuroxan (K(CL-14), 8), are all competitive candidates for new green primary explosives with excellent properties.

The preparations of 6, 7, and 8 (Scheme 2) are relatively complicated compared with those of nitrophenol-based potassium complexes. The potassium complex of 6 is synthesized in five steps from commercially available cyanoacetic acid by means of a more accessible nitration method with a mixture of trifluoroacetic acid anhydride and 100% HNO₃, finally affording yellow prismatic crystals. Fewer synthesis steps are used for the potassium complexes of 7 and 8. 7 can be prepared via reactions of aqueous potassium nitrate or sulfate and aqueous NaDNBF, exhibiting small plate-shaped crystals that have poor pouring and mixing properties with a high yield of 84%.

Scheme 2 Synthesis routes of 6–8.

The potassium complex of 8, as a derivative of 7, is prepared by reacting potassium nitrate with the sodium salt (which is always the precursor for other alkali metal salts) of CL-14 originating from DNBF. The yellow product can reach a yield of 81%.

It is accepted that molecules containing dinitromethyl groups show enhanced OB and density, which contribute to improved explosive performances. Hence, the combination of the dinitromethyl group with a furoxan ring dramatically improves the OB. For example, 6 exhibits a positive OB and properties competitive with those of Pb(N₃)₂.⁴⁴ It could be seen from Table 2 that 6 possesses a noticeable density, up to 2.130 g cm⁻³ at room temperature. The detonation performance of 6 is also superior: its detonation velocity (7759.0 m s⁻¹) is much higher than that of the traditional primary explosive Pb(N₃)₂ (5876.8 m s⁻¹), and has a comparable detonation pressure (27.3 GPa). On the other hand, the planarity of the dinitromethyl group makes the molecule more thermally stable than trinitromethyl-containing molecules, which have non-planar geometries. The carbon atoms of dinitromethyl and all the atoms in the furoxan ring are coplanar while the two dinitromethane moieties are almost planar, thereby contributing to the increased stability of 6. It is insensitive to light and very stable upon storage under ambient conditions, though it is quite sensitive to impact and friction stimuli. These desirable explosive properties make 6 a competitive green primary explosive, although its mechanical sensitivity needs amelioration.

Table 2 Physico-chemical and explosive properties of 6–8 and Pb(N₃)₂

	ρ (g cm^−3)	Mp (°C)	Tdec (°C)	Tdet (K)	Tdef (°C)	IS (J)	FS (N)	D (m s^−1)	P (GPa)
6	2.130		218.3	3574		2	5	7759.0	27.3
7		215			220	7	38
8		265	265		264	13	324
Pb(N3)2	4.800		315	3353		2.5–4.0	0.1–1.0	5876.8	33.4

---

On account of the deprotonation and introduction of the potassium cation, 7 is more stable and less impact-sensitive than its parent DNBF.⁴⁵ It behaves as a typical primary explosive considering its desirable heat resistance and high mechanical sensitivity, as shown in Table 2.^45,46 Compound 8 has a better performance compared to 7. The deflagration temperature of 8 (264 °C) is higher than that of 7 (220 °C), and so are the melting points. Both of them explode when heated to deflagrate. With respect to sensitivity, 8 tends to be much less sensitive, with its impact sensitivity twice and friction sensitivity more than eight times that of 7 (7 J and 38 N, respectively). Although 8 has advantages over 7 for being safer towards thermal and mechanical stimuli, 7 stands out for its electrostatic initiation at 0.045 J (ref. 47) while 8 failed to give off any flame or sound in the test of suitability in electro explosive devices (EED).

Both the density and detonation velocity of 7 and 8 are significantly improved compared to those of 3 through the replacement of nitro groups by furoxan groups. Comparing the properties of 7 and 8, we find that the introduction of amino groups can markedly improve the thermal stability and reduce the sensitivity to mechanical stimuli.

As an initiating composition for both military and commercial applications since the early 1950s, there is more in-depth research on 7. The thermal analysis indicates that 7 decomposes via a multi-step exothermic process in both inert gases or air, with a rapid and strong initial exotherm at a peak of 216 °C, followed by a much weaker subsequent peak. There is also an extra third exothermic peak attributable to the oxidization of the solid residues at temperatures above 300 °C and in air. In fact, the solid residues at 230 °C are RCOOK, KNCO, RNO₂, and KNO₃; at 306 °C, the solid residues are KNC, RCOOK, and KNO₃. This, together with experiments under various pressures, suggests that the decomposition is not determined by the nature of the gas environment. Its kinetic parameters were obtained through various test methods (ARC (accelerating rate calorimetry), HFC (heat flux calorimetry), DSC, and TG (thermal gravity analysis)) showing that the results depend on the experimental conditions like the nature of the system, sample size, and heating rate. The average values of E_a and lg(A min⁻¹) are 163 kJ mol⁻¹ and 37, respectively.

Experiments verified that the morphology of energetic complexes has a significant impact on their density, thermal stability, energetic properties, and free-flowing properties.⁴⁸ The morphology study of 7 confirmed that granular spherical crystals with a smooth surface tend to have better performance. This rule is readily applied to other complexes. 7 with various morphologies could be obtained by adding different surfactants to the reaction system.⁴⁹ Studies of five samples of varied morphologies concluded that the spherical sample, a result of mixing with surfactants, has the lowest decomposition temperature (peak at 208.1 °C) and the highest activation energy (E_a = 195.4 kJ mol⁻¹) of all.⁵⁰ This dependence on morphology can be explained by the hot spot initiation theory, which claims that sheet samples will engender more hot spots to make the decomposition reaction easier than that of the spherical samples. However, the surfactants did not affect the result of decomposition, except for changing the rate of weight loss. Additionally, sensitivity tests demonstrated that the spherical 7 sample had the highest impact and friction sensitivity and lower flame sensitivity than other forms, and its mechanical sensitivity was between those of mercury fulminate and lead azide. In contrast, 7 synthesized without any surfactants shows much more inferior properties than the surfactant-ameliorated ones.

3.3. Potassium energetic complexes of FOX-7 and its derivatives

“Push–pull” nitro-enamines are unusual in that their typical structure contains a highly polarized carbon–carbon double bond with positive and negative charges stabilized by the amino group and nitro group, respectively, and they share many tautomers and resonances.^51–53 FOX-7 was first synthesized in 1998 by Latypov et al.⁵⁴ Since then, quite a few studies have been carried out on its synthesis, reactivity, reaction mechanism, molecule structure, thermal behavior, explosive performance, and applications.^54–66 It is a typical “push–pull” nitro-enamine compound with acidic properties, and can react with some nucleophiles to afford new energetic derivatives.^67–69 As the primitive source of compounds 9–12, FOX-7 is favored for its high-energy nature with high thermal stability and low sensitivity to impact and friction, which allow it to exhibit excellent performance in ammunitions and solid propellants. The theoretically computed explosive and ballistic parameters are close to those of RDX, while its mechanical sensitivity could be as low as that of TATB (Table 3).

Table 3 Explosive properties of FOX-7, RDX, and TATB

Compounds	D (m s^−1)	P (GPa)	IS (J)	FS (N)
FOX-7	9090	36.6	25.2	>360
RDX	8800	34.7	7.6	120
TATB	8108	31.1	34	>360

---

The potassium salts of FOX-7 (KFOX-7, 9a,⁷⁰ KFOX-7·H₂O, 9b⁷¹), 1-amino-1-hydrazino-2,2-dinitroethylene (K(AHDNE), 10),^72,73 2,3-dihydro-4-nitro-3-(dinitromethylene)-1H-pyrazol-5-amine (K(NNMPA), 11),⁷⁴ and 2-(dinitromethylene)-1,3-diazepentane (KDNDZ, 12)⁷⁵ also belong to the “push–pull” nitro-enamine family of compounds, thus they are similar in many aspects and exist in many tautomers and resonances like their parents (Fig. 6).

Fig. 6 Tautomers and resonances of FOX-7, AHDNE, and DNDZ.

Preparations of 9–12 are based on the acid–base neutralization reactions of KOH and the corresponding organics (Scheme 3). Reactions of FOX-7 with KOH in different solvents leads to different products – 9a in water and 9b in methanol. 10, 11, and 12 are all derivatives of 9 and can be synthesized from FOX-7. Interestingly, 10 and 11 are actually obtained in one step through the reaction in methanol with yields of 62% and 18%, respectively. However, the yield of 10 can be dramatically improved to 95% if the solvent is changed to absolute ethanol. Compound 12 is obtained in water with a relatively high yield of 71%. The reaction mechanism for the AHDNE anion transferring to the NNMPA anion is determined to occur in two steps (Scheme 4), supported by the slight bubbling observed in the synthesis process. In fact, the formation of the NNMPA anion typically exemplifies the high reactivity of the adjacent amino-hydrazino group in AHDNE.

Scheme 3 Synthesis routes of 9–12.

Scheme 4 Conjectural reaction mechanism for the formation of the NNMPA anion in KOH–methanol solution. Reproduced from ref. 74 with permission from Elsevier, copyright [2013].

While 9 and 10 are determined to be simple salts featuring hydrogen-bonding interactions between the cations and anions, 11 and 12 are both coordination complexes with multi-coordinated potassium cations and infinite 3D networks. The potassium cation of 11 is connected with five adjacent NNMPA anions through seven K–O and one K–N coordination bonds (Fig. 7), forming a very distorted dodecahedron structure with the potassium cation being the coordination center. Considering that the five-membered heterocyclic ring, presenting as an irregular pentagon, is more stable than the adjacent amino-hydrazino group of AHDNE, a big π bond is formed and supported by the conjugated C–C bonds. Therefore, it is reasonable to believe that NNMPA is more stable than AHDNE, which easily self-ignites or explodes at room temperature. Sensitivity tests confirmed that 11 is less sensitive than 10, with impact sensitivity values of 16.7 and 5 J, respectively (Table 3). The potassium cation coordination number of 12 was found to be 8 and is comprised of 6 K–O bonds and 2 K–N bonds (Fig. 8).

Fig. 7 The extended coordination scheme of the potassium cation in 11. Reproduced from ref. 74 with permission from Elsevier, copyright [2013].

Fig. 8 The extended coordination scheme of the potassium cation in 12. Reproduced from ref. 79 with permission from WILEY-VCH, copyright [2009].

Plate/cubic shaped crystals of 9a feature relatively high mechanical sensitivity and good energetic and thermal properties. The impact and friction sensitivities of 9a (6.8 J and 160 N, respectively) indicate its sensitive nature towards mechanical stimuli. The DSC curve of 9a reveals an exothermic decomposition step, with a peak temperature of 225 °C (ΔH = −386 J g⁻¹). The coordinated water in 9b strongly interacts with the cation. Hence, it is difficult for the molecule to lose water in thermal decomposition (>180 °C), making 9b more thermally stable.

The excellent detonation pressure (37.3 GPa) and velocity (8973 m s⁻¹) of 10 prove it to be a competitive energetic material.⁷⁶ Although 11 is less thermally stable than 10, it is relatively insensitive to friction as its explosion probability from friction stimuli is 32% (tested 25 times). Its impact sensitivity (16.7 J) is much lower than that of 10 (5 J), yet higher than that of FOX-7 (25.2 J).

In fact, 10 serves as a flame suppressant in solid propellants, substituting inorganic potassium salts such as KCl, K₂SO₄, KNO₃, and K₃AlF₆, in order to generate more energy and cleaner gas wastes. Besides, 10 can raise the combustion rate of the propellant because of its intense thermal decomposition.⁷⁷ The gaseous products of 11 decomposition were determined mainly to be CO₂, N₂O, NO, and H₂O. Surprisingly, there was noNH₃, although there is an amino group in the NNMPA anion.⁷⁸ Compound 12 is also applied as a flame suppressant in propellants to substitute the inorganic potassium salts for the same purposes.

The thermal behaviors of 10–12 show different stages of exothermic decomposition. There are three obvious exothermic decomposition stages for 10. The first one is the most intense and determined to be the main decomposition process without melting. 11 and 12 decompose in two stages. The second stage is more intense (ΔH_dec = −1384 J g⁻¹) in 11, while the first stage of 12 seems to be more conspicuous (ΔH_dec = 1833 J g⁻¹). It is worth noting that there is a very inconspicuous crystal phase transition in 12 at a peak temperature and enthalpy of 131.4 °C and 1.923 J g⁻¹, respectively. The kinetic parameters of the first stage of 10 and 11, and both stages of 12, are given in Table 4. Comparing the kinetic parameters, we could rank their thermal stability as 12 > 10 > 11. The same ordering is suggested by data in Table 5, in which the critical temperature of thermal explosion (T_b) and adiabatic time-to-explosion (t_a) are important parameters that can be easily used to directly evaluate the thermal stability. The T_b values of 12, 10, and 11 are 208.6 °C, 171.4 °C, and 146.3 °C, respectively, and their detected values of t_a are 158, 90, and 50–90 s, respectively. They suggest that the thermal stability sequence is 12 > 10 > 11 as well. With respect to T_b and t_a in Table 5, it is clear that 12 has a lower thermal stability than its parent DNDZ.⁷⁹

Table 4 Kinetic parameters of 10–12^a

Complexes	Exothermic stage	Ek (kJ mol^−1)	lg A (s^−1)	rk	Eo (kJ mol^−1)	ro
a Subscript k is data obtained by Kissinger's method. Subscript o is data obtained by Ozawa's method.
10	1	186.2	19.63	0.9985	184.2	0.9986
11	1	166.5	18.42	0.9972	165.1	0.9974
12	1	152.0	13.93	0.9985	152.4	0.9987
	2	181.5	13.80	0.9993	182.0	0.9994

---

Table 5 Physico-chemical and explosive properties of 9–12 and DNDZ

Compounds	ρ (g cm^−3)	Tdec (°C)	Tb (°C)	ta (s)	ΔHf (kJ mol^−1)	IS (J)	FS (N)
9a		225				6.8	160
10	1.922	181.6	171.4	50–90	437.8	5
11			146.3	40		16.7
12		227.7	208.6	158
DNDZ			261.04	>158

---

3.4. Potassium energetic complexes of triazoles, tetrazoles, and their derivatives

There are many reports to date concerning triazoles and their derivatives.^80–84 With their high nitrogen content and the resulting high heats of formation, triazoles (and especially tetrazoles) are very useful heterocyclic backbones for the preparation of high-performance energetic materials.^85–88

3.4.1. Potassium energetic complexes of triazole derivatives. Triazoles are a kind of unsaturated five-membered aromatic heterocycle that contains three nitrogen atoms located at the 1,2,3- or 1,2,4-positions in the ring. They show higher thermal stability than tetrazoles due to the less catenated nitrogen atoms in one row, and feature two carbon atoms that allow more energetic groups for tailoring the material characteristics.

While the potassium complex of 5-azido-3-nitro-1H-1,2,4-triazole (13) is favored for its facile preparation, most other derivatives of triazoles are difficult to prepare or have low reaction yields. The potassium complex of 3-nitro-1,2,4-triazol-5-one (KNTO·H₂O, 14) gathered attention for its good performance deriving from its parent NTO, whose performance properties are comparable to those of RDX with an insensitivity similar to TATB (Scheme 5).⁸⁹

Scheme 5 Synthesis routes of 13 and 14.

Complex 13 appeared in the literature as early as in 1974; however, its preparation and characterization did not appear until 2013.⁹⁰ 13 is prepared by mixing KOH with 5-azido-3-nitro-1H-1,2,4-triazole in ethanol with a satisfactory density of 1.933 g cm⁻³ at −100 °C. 14 is prepared by the reaction between KOH and NTO in water and was initially synthesized by Manchot and Noll in 1905.

Both complexes 13 and 14 form 3D networks. Actually, the potassium cation in 13 is nine-coordinated, consisting of 6 K–N coordination bonds and 3 K–O bonds, resembling a distorted singly capped tetragonal prism (Fig. 9).⁹¹ The potassium cation in 14 is eight-coordinated through 7 K–O coordination bonds and 1 K–N bond. Additionally, the bonds between potassium cations and water molecules are weaker than those connecting potassium cations and NTO anions, suggesting the dehydration of 14 during pyrolysis.

Fig. 9 The extended coordination scheme of the potassium cation in 13. Reproduced from ref. 91 with permission from MDPI, copyright [2012].

Complex 13 is quite sensitive to impact stimulus and its impact sensitivity is 15 mJ. Experiments confirmed its capability of igniting RDX when subjected to heat or flame, identifying it as a new primary explosive. Furthermore, the toxicity of 13 is slightly lower than that of RDX from comparing the EC₅₀ values, which is the effective concentration needed to decrease the bioluminescence of the bacteria strain Aliivibrio fischeri by 50% after a defined period of exposure.

The comparison results between 14 and its parent NTO are similar to those between 10 and its parent AHDNE. Compared to NTO, compound 14 possesses lower sensitivity towards impact and friction, with reported impact sensitivity lower than 34 J and friction sensitivity up to 360 N. In contrast, its thermal stability is worse than NTO, which is explained by the small energy difference between the frontier orbitals (ΔE = E_LUMO − E_HOMO).⁹²

13 shows a decomposition temperature of 175 °C and a thermal stability for at least 48 h at 75 °C, in isoperibolic long term measurements in an open glass vessel. On the other hand, according to the frontier orbital theory, the chemical properties of 14 are dominated by the parent, NTO, since the highest occupied molecular orbital and lowest unoccupied molecular orbital (LUMO and HOMO, respectively) are both located in the NTO rings. This matches quite well with the experimental results as the thermal decomposition process and products of 14 are congruent with those of NTO. Compound 14 exhibits three-stage decomposition in DSC.²⁸ The former two separate stages are endothermic in the temperature range of 135–200 °C, corresponding to the weight loss of water molecules, as observed by TG. Exothermal decomposition occurs in the temperature range of 247–268 °C, peaking at 257 °C. The products of the thermal decomposition of 14, containing gases released by NTO decomposition, are characterized by KNCO vapor upon flash pyrolysis. In fact, compound 14 enhances the burning rates up to 12% according to the best catalytic effect. The produced volatile metal isocyanate or metal nitrite, along with the burning rate study results of ammonium perchlorate-hydroxyl terminated polybutadiene (AP-HTPB) composite propellant incorporating 14, indicates that 14 is a ballistic modifier candidate, as its continued reaction can form metal oxide particles that produce subsequent damping in high-frequency acoustic oscillations in solid rocket motors. Particle damping is desirable here because the acoustic modes can couple with the combustion pressure wave, which, in turn, leads to unsteady combustion and possibly failure of the device.⁹³

3.4.2. Potassium energetic complexes of tetrazole and its derivatives. Tetrazole (HTZ) is an unsaturated five-membered heterocycle with four nitrogen atoms in the ring. HTZ is the most basic tetrazole structure, whose two tautomeric forms (Fig. 10) have nearly equal calculated energies. However, the 1H-isomer is favored in the crystalline state and in solution; while in the gas phase, the second one has lower energy. HEDM derives most of its energy from either oxidation of the carbon backbone or its high positive heat of formation. Tetrazoles and their derivatives have attracted broad interest due to their very high nitrogen content related to a high density, and thus favorable explosive performance, as well as good thermal stabilities owing to their aromatic ring system.

Fig. 10 Tautomers and resonances of HTZ.

While the tetrazole ring itself shows a basicity lower than that of aniline, the 5-substituted derivatives are weak acids with pK_a values in the range of 1.1–6.3 (similar to those of carbonic acids), which facilitates the preparation of their potassium complexes. The combination of nitro functionalities with tetrazoles is known from substances like 5-nitro-2H-tetrazole and its derivatives.⁹⁴

Protonation takes place preferentially on the 4-nitrogen atom. While the acylation of 5-monosubstituted tetrazoles selectively occurs on the 2-nitrogen atom in most cases, the alkylation is not selective and yields mixtures of 1,5- and 2,5-disubstituted tetrazoles. Electrophiles usually attack tetrazoles at one of the ring nitrogen atoms, and the attack position strongly depends on the substituent on the carbon atom, reaction conditions, and reagents. While electron withdrawing groups like NO₂ at C5 lead to substituents at nitrogen N2, electron donating groups, such as NH₂, lead to alkylation at nitrogen N1.

The potassium salt of HTZ (KTZ, 15) is the most basic complex based on tetrazole. There are several derivatives, namely the potassium salts of 5-aminotetrazole (16), 5-nitrotetrazole (KNT, 17), 5-(dinitromethylene)-4,5-dihydro-1H-tetrazole (KHDNTz, 18), 5-azido-1H-tetrazole (KCN₇, 19), and azidotetrazole 2-oxide (CN₇OK, 20).⁹⁵ The variation of their properties generally complies with the universal rules mentioned before, corresponding to the anion ligands. In addition, we observed that by comparing 19 and 20, the oxidation of a tetrazole ring to its 2-oxide could improve the OB and reduce the mechanical sensitivity.

Interestingly, the coordination condition of the potassium cation seems to follow certain rules. While electron donating groups like NH₂ at C5 lead to coordination with the nitrogen atom N2 (16), electron withdrawing groups, such as NO₂ and N₃, lead to potassium coordination with all the nitrogen atoms on the ring (17 and 19). Besides, once there are oxygen atoms, potassium cations must coordinate with oxygen (17, 18, and 20).

Particular attention should be paid to 19 due to its excellent explosive properties. N-rich molecules are considered as prime candidates for “green” energetic materials since the materials exhibit desirable performance characteristics in high explosives or propellant formulations, while the main combustion product is non-toxic N₂ gas. The CN₇⁻ anion represents a milestone in the development of these nitrogen-rich “green” energetic materials. Nitrogen-rich heterocyclic compounds as ligands tend to lead to high density, high heat of detonation, and good thermal stability.^96,97 Nevertheless, 19 is extremely sensitive, so it has no current application and is only of academic interest. Fortunately, the oxidation of a tetrazole ring to its 2-oxide (20) has been shown to be effective for reducing the sensitivity of tetrazole-based energetic materials towards mechanical stimuli.⁹⁸

The construction of infinite metal–organic frameworks through a systematic change of the organic ligands and metal centers has provided an impetus to further research metal–organic super-molecular architectures.^99–102 These architectures have attracted extensive interest due to their unique physico-chemical and unusual topological properties that lead to applications in adsorption, gas storage, catalysis, optics, and energetic materials. The potassium complex of 4-amino-3,5-dinitropyrazole (ADNPK·H₂O, 21) is based on a relatively new high-energy material LLM-116, which is a small molecule initially synthesized in 2001 with excellent performance.^103,104

Nitraminotetrazoles are of specific interest as HEDCs because of their large positive heats of formation and good thermal performance. The potassium salt of 1,5-di(nitramino)tetrazole (CK₂N₈O₄, 22), possessing a C-nitramino moiety and an N-nitramino moiety, has very powerful energetics.

There are many ways to prepare complexes 15–22 (Scheme 6). 15 and 16 can be obtained by the deprotonation of the respective parent (HTZ and HAT) with K₂CO₃ in water. 17 can be synthesized by two different methods, both starting from 5-amino-1H-tetrazole. One involves sodium nitrite, copper(II) sulfate pentahydrate, KOH, and 5-amino-1H-tetrazole; the other uses KOH and synthesized ANT hemihydrate to enable a larger and more practical synthesis scale (∼1 g), and to avoid the direct manipulation of the highly sensitive, acidic copper salt of 5-nitrotetrazole. 18 is synthesized by treating 5-(dinitromethylene)-4,5-dihydro-1H-tetrazole, or its dehydrate, with K₂CO₃ in acetone solution, and is isolated as pale yellow crystals with over 90% yield. 19 can also be synthesized by the acid–base neutralization reaction of KOH with 5-azido-1H-tetrazole in methanol, considering the strong acidity of the latter resulting from strong π-delocalization in the anion. 20 is prepared by heating KOH with ammonium azidotetrazolate 2-oxide to reflux. By the reaction of LLM-116 and KOH solution, 21 is obtained as bright red crystals. The synthesis of 22 starts with commercially available dimethylcarbonate, and nitration in acetonitrile with N₂O₅ is required before the nitramide is decomposed in solution with aqueous KOH to produce 22 as a white precipitate.

Scheme 6 Synthesis routes of 15–22.

The crystal structures of 15–22 also vary. Some are simple crystals, some form 2D chains or expand to 2D planes, and some establish 3D frameworks. The common factor is that all potassium cations are coordinated with several atoms, thus making the structures more stable.

Among the simple crystals, 17 crystallizes in the monoclinic crystal system without crystal water. Interestingly, the determined lattice parameters and the packing forms are remarkably similar to those of ANT, suggesting strong structural similarities that can be explained by the similar ion radii of potassium and ammonium. Each cation coordinates with six anions and three of them are nearly co-planar with the cation (“co-planar” anions), two of which have additional weaker contacts to the cation by chelation through a nitro-group oxygen atom. The result is an eight-coordinated potassium center (Fig. 11). The strongest cation–anion interactions are observed to occur between the cation and the three nitrogen atoms of the three “co-planar” anions. Last but not least, highly efficient packing is indicated by the utilization of all available electron donor atoms in the anion to form contacts with the cation. In 19, the coordination of potassium cation, surrounded by six nitrogen atoms (Fig. 12), is comparable to that of 16, and can be described as a distorted octahedral. The azido group is bent, a result of the hyper-conjugation effects.¹⁰⁵ 22 crystallizes anhydrously in the monoclinic space group with four molecules in a unit cell. Its molecular unit is asymmetric and the nitramino moiety attached to the carbon is almost coplanar with the ring (Fig. 13), which may account for its good thermal stability.

Fig. 11 The extended coordination scheme of the potassium cation in 17. Reproduced from ref. 108 with permission from the Royal Society of Chemistry.

Fig. 12 The extended coordination scheme of the potassium cation in 19. Reproduced from ref. 105 with permission from American Chemical Society, copyright [2009].

Fig. 13 The extended coordination scheme of the potassium cation in 22. Reproduced from ref. 114 with permission from WILEY-VCH, copyright [2015].

Compounds 16 and 20 are determined to be 2D, formed by chains and planes, respectively. 16 crystallizes in the monoclinic crystal system with four molecules in the unit cell. The infinite 2D layers are connected to each other via distorted trigonal coordination of the potassium cations (Fig. 14).¹⁰⁶ 20 crystallizes in the monoclinic crystal system. X-ray data reveals that the crystal structure of the potassium salt has asymmetric units, and within the structure, two flight chains along the a-axis are formed (Fig. 15).¹⁰⁷

Fig. 14 The extended coordination scheme of the potassium cation in 16. Reproduced from ref. 106 with permission from WILEY-VCH, copyright [2007].

Fig. 15 The extended coordination scheme of the potassium cation in 20. Reproduced from ref. 107 with permission from WILEY-VCH, copyright [2011].

3D networks are formed during the crystallization of complexes 15 and 21. 15 crystallizes in the hexagonal crystal system with six molecules in the unit cell. Its crystal structure is composed of K₂TZ₃ units, in which two potassium cations are coordinated trigonally by the same three tetrazolate rings involving equivalent nitrogen atoms (Fig. 16).⁹⁵ The delocalized ring is suggested by the analysis of bond lengths. The units are symmetrical and arranged along the c-axis at different levels, leading to a 3D structure. In the determined crystal structure of 21, each potassium cation coordinates with five anion ligands and one water molecule while each ligand is bonded to four potassium cations, forming an infinite 3D framework by self-assembly. This structure might contribute to the thermal stability.

Fig. 16 The extended coordination scheme of the potassium cation in 15. Reproduced from ref. 95 with permission from WILEY-VCH, copyright [2008].

The impact and friction sensitivities of 15 are >100 J and >360 N, respectively, indicating its insensitive nature. However, the density of 15 (1.774 g cm⁻³) is relatively low among these complexes. It combusts nearly smokelessly, but shows a brilliant flame color of purple-red due to the potassium cation. Similarly, 16 is insensitive towards impact and friction stimuli and shows no reaction under electrical discharge. It is a common intermediate in the synthesis of alkylated amino-tetrazoles and their derivatives. It can also be used as coloring agents in modern pyrotechnics due to the high nitrogen content.

The other complexes in this series, however, are rather sensitive. Anhydrous complex 17 is extremely sensitive to impact (10 J) and friction stimuli (less than 5 N).¹⁰⁸ It has sensitivities similar to lead azide and significantly higher¹⁰⁹ than 16 as well as potassium 5-nitroimino-1,4H-tetrazolate,¹¹⁰ but lower than 19. When placed in flame, even a small quantity (∼0.5–1 mg) of 17 always exhibits a loud explosion. It is a prospective primary explosive and initiator, in spite of the need for further amelioration.

19 is classified as extremely sensitive to friction (less than 5 N). Moreover, its impact sensitivity cannot be tested because this extremely energetic and explosive compound cannot be handled without violent explosions. This complex is so explosive that it even explodes spontaneously when dry. Therefore, it is probably of only academic interest right now. Nevertheless, sensitivity tests of the N-oxide counterpart of 19 were carried out and the following data were obtained: impact sensitivity 1.5 J, friction sensitivity less than 5 N, and electrostatic discharge sensitivity 1.3 mJ, thereby verifying the less sensitive property of 20. Actually, the addition of the single oxygen atom to the ring not only decreases the heat of formation, but also allows more intermolecular interactions (considering the increased density), both of which contribute to reducing the sensitivity of an energetic material. Simultaneously, the increased density as well as the increased OB makes transforming a tetrazole to the N-oxide a useful strategy for improving the energetic performance. The density of 20 (up to 2.073 g cm⁻³) is comparable with other salts of azido-tetrazolate 2-oxides, and is the highest among all these HNC-HEM complexes. Its only downside is the slightly lowered decomposition temperature compared to other N-oxide based systems with increased thermal stability.^111,112 Overall, 20 has the characteristics of lower sensitivity, especially to impact stimulus, and higher performances compared to its non-oxide analogue. However, anhydrous 20 crystals are still highly sensitive whereas some salts of azido-tetrazolate 2-oxide crystallize as monohydrates that are completely insensitive towards impact.

21 serves as a flame suppressant in solid rocket propellant, replacing current inorganic inert potassium salts like KCl, K₂SO₄, KNO₃, and K₃AlF₆. It can inhibit “post-combustion” in solid rocket motor exhaust, and at the same time, increase the energy density because of the higher number of energetic groups and potassium cations.¹¹³

22 is very sensitive as its impact and friction sensitivities are 1 J and less than 5 N, respectively. However, it shows promise as a primary explosive with a detonation velocity of 10 011 m s⁻¹, detonation pressure of 52.2 GPa at the CJ point (at the CJ point, the explosive velocity equals to the velocity of explosive product particles adding local sound velocity), and the highest density (2.137 g cm⁻³ at 25 °C) among these potassium salts of tetrazole and its derivatives. Experiments have been conducted to show that 22 could easily detonate RDX with its shockwave when initiated by a standard pyrotechnical igniter. Therefore, 22 is recommended to be used as an environmentally benign and thermally stable sensitizer in place of tetracene.¹¹⁴

Comparing the decomposition temperatures of 15–22 (Table 6), we could see that 18, 19, and 20 have much lower decomposition temperatures than others, with that of 20 being possibly the lowest one. 20 decomposes at 138 °C with no prior melting, 10 °C lower than its non-N-oxide counterpart (19). 18 undergoes violent explosive decomposition at 188 °C and is classified as a promising candidate for primary explosives.³¹ Although the decomposition temperature of 22 (240 °C) is just in the middle of 15–22, it is much higher than that of its parent 1,5-di(nitramino)tetrazole (110 °C), thus making 22 more suitable for use in primary compositions.

Table 6 Physico-chemical and explosive properties of 15–22

Complexes	ρ (g cm^−3)	Tdec (°C)	IS (J)	FS (N)
15	1.774	308	>100	>360
16	1.961	350
17			10	<5
18		188	3.5	10
19	1.917	148		<5
20	2.073	138	1.5	<5
21	1.907	328
22	2.137	240	1	<5

---

Thermal analysis shows a small melting endotherm in 17 prior to the onset of decomposition, and a detonation temperature of 195 °C. Although 17 shows a rather low decomposition point compared to the detonation temperatures of lead azide (315 °C) and lead styphnate (282 °C), it still meets the requirement for potential initiators.¹¹⁵

The decomposition temperatures of 15, 16, and 21 are all above 300 °C, thus making them possible initiators. 15 has better thermal stability than its parent HTZ, a finding congruent with theoretic predictions. In experiments, 15 shows a mild melting point of 210 °C before decomposing at 308 °C while HTZ exhibits a sharp melting point at 154 °C and a decomposition temperature of 188 °C. Not surprisingly, 15 possesses a high positive enthalpy of formation (174 kJ mol⁻¹). 16 possesses the highest decomposition temperature (350 °C). 16 is all “green” with end products of KO₂, CO₂, H₂O, and N₂; and it combusts with a purple flame. Thermal behavior analysis reveals that there are two decomposition stages for 21. The first is attributable to melting and dehydration, and the second is an obvious exothermic decomposition process with a peak temperature at 328 °C and a decomposition enthalpy of 2036 J g⁻¹. In addition, a multiple heating method was employed in order to obtain the kinetic parameters, showing that the apparent activation energy for the exothermic decomposition of 21 is 237.3 kJ mol⁻¹.¹¹³

3.5. Potassium energetic complexes of nitro-rich poly-heterocycles

The combination of two different heterocycles is beneficial owing to the vast possibilities of tailoring energetic performances. Due to its two acidic heterocycles, the potassium complex of 5-(5-azido-1H-1,2,4-triazol-3-yl)tetrazol-1-ol (KAzTTO, 23) can form salts with different numbers of potassium cations, and thus, present different properties. The potassium complex of 2,4,6-trinitroanilino benzoic acid (K(TABA)(H₂O)₂, 24) tends to be insensitive, as expected from the two opposite benzene rings that make the system quite balanced and stable. Potassium 1,1′-dinitramino-5-5′-bistetrazolate (K₂DNABT, 25) and potassium dihydridobis(4-nitroimidazolyl)borate (26) also possess desirable stabilities with their symmetric anions.

Boron-based formulations could be a promising alternative for a light emitter and burn-rate modifier in pyrotechnics by serving as barium substitutes.¹¹⁶ However, most of the boron compounds currently tested possess no energetic character. Efforts are being made to create new energetic boron compounds, among which the nitrogen-rich heterocycles combining both thermal and physical stability are worthy of investigation. A variety of poly(azolyl)borates were studied in the last decades, especially the scorpionate ligands.¹¹⁷ The high-yield synthesis methods^118–121 of poly(azolyl)borates bearing insensitive groups like alkyl- or aryl-substituents are based on Trofimenko's thermolytic approach, which typically involves reactions of mixed corresponding azole and alkali metal borohydrides at high temperatures. In coordination, organometallic, and bioinorganic chemistry, pyroazoliborate has exhibited versatility as a mono-anionic, nitrogen-based, and multi-dentate ligand. Apart from this, tri(imidazolyl)borates^122,123 and tetra(imidazolyl)borates¹²⁴ have been of interest in the construction of new metal–organic framework structures. In order to increase the energetic character of the borate, nitrated azoles are introduced in 26. Potassium 4,4′-bis(dinitromethyl)-3,3′-azofurazanate (27), comprised of azofurazan moieties, dinitromethyl groups, and a nontoxic potassium cation, presents excellent performance properties. In the first place, the furazan rings enhance the density and OB while the introduced azo increases the heat of formation. Secondly, the dinitromethyl moiety could serve as a superior energetic functional group with good thermal stability.

The preparations of 23–27 vary significantly (Scheme 7). Potassium and dipotassium salts of AzTTO, namely KAzTTO H₂O (23a) and K₂AzTTO 3H₂O (23b), are easily prepared using AzTTO and K₂CO₃, in water for 23a and in ethanol for 23b. This is because AzTTO features a second nitrogen-rich heterocycle that can be deprotonated, thereby leading to the formation of 1 : 1 and 2 : 1 salts with cations. Besides, both rings carry an acidic nitrogen-bound proton, thus enabling the formation of energetic salts by Brønsted acid–base or salt synthesis reactions. However, 23b is contaminated by 23a and could not be obtained in pure form. Furthermore, the synthesis of AzTTO needs further optimization owing to its sophistication. 24 is easily synthesized by adding KOH solution to the aqueous solution of TABA and the yellow product is obtained in 90% yield. The synthesis of 25 is feasible but difficult, as the starting material 1-dinitroamino-tetrazole is hard to obtain. Having compared several methods of synthesis, researchers picked out the best one. Although still sophisticated, this method, based on dimethylcarbonate and glyoxal, uses commercially available starting materials and has a facile route with a relatively high yield. The low water solubility of 25 facilitates the isolation and purification of the crystals. 26 is synthesized by reacting 4-nitroimidazole with potassium borohydride in freshly distilled acetonitrile at reflux. The target complex is obtained as a beige solid with a yield of 80%. This pathway can be easily scaled up and industrialized thanks to its simplicity and high yield. 27 is synthesized through a complex pathway from 3-amino-4-cyanofurazan and via reaction of KI instead of a common acid–base neutralization reaction.

Scheme 7 Synthesis routes of 23–27.

23a crystallizes from water as a colorless monohydrate in the triclinic crystal system featuring an interesting cation-water framework. The crystal structure consists of symmetric [K₂(H₂O)₂]²⁺ units where each potassium cation is coordinated with 2 oxygen and 2 nitrogen atoms (Fig. 17). It shows a very broad and almost invisible signal in the ¹H NMR spectra due to fast proton exchange in dimethyl sulfoxide (DMSO). 23b crystallizes as a trihydrate in the triclinic crystal system. The structure confirms the deprotonation of both rings. The most striking structural motif is the formation of a cation–water framework, similar to 23a but strongly enhanced, by the two potassium cations and all three water molecules. 25 crystallizes in the triclinic crystal system with irregularities in coordination (Fig. 18). Three possible isomeric compounds for 26 have been predicted (Fig. 19) amongst which the first one is the most reasonable. However, NMR spectroscopic data has not identified any of the isomers, although their existence has not been ruled out either.²⁹ 27 crystallizes in the triclinic crystal system. The molecular structure is symmetric and each potassium cation is chelated by five oxygen atoms from the nitro groups (Fig. 20). The potassium centers are connected by dinitromethyl groups into chains, and these chains are further linked by azofurazan moieties to form a 3D network.

Fig. 17 The extended coordination scheme of the potassium cation in 23a. Reproduced from ref. 90 with permission from WILEY-VCH, copyright [2013].

Fig. 18 The extended coordination scheme of the potassium cation in 25. Reproduced from ref. 125 with permission from WILEY-VCH, copyright [2014].

Fig. 19 Three possible isomers of 26.

Fig. 20 The extended coordination scheme of the potassium cation in 27. Reproduced from ref. 128 with permission from WILEY-VCH, copyright [2016].

23a, 23b, 25, and 27 are all extremely sensitive towards mechanical stimuli (Table 7). They have similar impact sensitivities (less than 1 J for 23a and 23b, and 1 J for 25). The friction sensitivities of 23a and 23b are similar, with high values of under 10 N, while that of 25 is much higher (less than 1 N). In addition, they have all been determined to be nontoxic using the luminous bacteria method.

Table 7 Physico-chemical and explosive properties of 23–27

Complexes	ρ (g cm^−3)	Tdec (°C)	ΔHf (kJ mol^−1)	IS (J)	FS (N)	D (m s^−1)	P (GPa)
23a				<1	<10
23b				<1	<10
24		274		>34	>360
25	2.172		326	1	<1	8330	31.1
26		189		>40	>360
27	2.039	229	110	2	20	8138	30.1

---

Because of their stable performance in detonation, 23a and 23b were tested as true primary explosives instead of just highly sensitive compounds. However, when fixed to a surface with a bit of transparent tape and poked with a preheated needle, only 23a showed a very fast deflagration. Thus, 23a was tested for its ability to initiate the commonly used secondary explosive RDX. The positive results suggest 23a is a potential candidate for a primary explosive, and it may require further investigation.

25 has a high density (2.172 g cm⁻³ at −173 °C), impressive detonation velocity (8330 m s⁻¹) and detonation pressure (31.1 GPa), and high heat of formation (326 kJ mol⁻¹). Apart from sensitivity towards mechanical stimuli, its electrostatic discharge sensitivity was determined to be high (0.003 J) and comparable to that of lead azide. The desirable ignition capacity and the instant detonation of the material in contact with a flame or a hot metal needle ensures the suitability of 25 as a primary explosive. Thorough comparisons of the energetic properties and calculated performance data of lead azide and 25, it was concluded that all calculated critical detonation parameters of 25 are superior to those of lead azide, especially the excellent thermal stability with a detonation temperature above 200 °C. Therefore, it is reasonable to research 25 as an alternative to lead azide.¹²⁵

Sensitivity tests reveal that 24 and 26 can both be classified as insensitive materials, with impact sensitivities of more than 34 and 40 J, respectively, and friction sensitivities of more than 360 N in both. In order to assess the performance of 24 as an energetic ballistic modifier, the salt was incorporated into AP-HTPB composite propellants for experiments. Interestingly, although alkali metal salts are widely known as burning rate suppressants, 23 produced a 81% burning rate enhancement in the entire pressure region studied (2–9 MPa) as well as brought down the pressure exponent of burning rate (n) from 0.33 to 0.15 according to the best catalytic effect.²⁸ Furthermore, the incorporation of 14 and 24 decreased the peak decomposition temperature of the exotherm, suggesting that their overall catalytic effect on the burning rates of the condensed/near surface gas phase of the propellant (with 23 having a more pronounced catalytic effect on the decomposition of propellant). In fact, 24 was found superior in this regard in view of its remarkable effects on enhancing the burning rate and reducing the pressure index value.

The insensitivity of 26 was confirmed for possible application in pyrotechnical mixtures according to BAM standards (impact sensitivity > 40 J, friction sensitivity > 360 N, and ESD > 0.85 J). Its solvolytic stability was also studied since the general stability of a complex decreases with more hydrogen atoms attached to the central boron atom.¹²⁶ Although tetra(imidazolyl)borate is completely stable against hydrolysis, it was even investigated as a gravimetric reagent for proton.¹²⁷ Experiments show the rather high stability of 26 against solvolysis in methanol or methanol/acetonitrile solution (0.6 : 1.0). Furthermore, the complex exhibits moderate stability when subjected to metathesis reactions. Considering its desirable properties, 26 can be a promising candidate for new environmentally benign coloring agents in pyrotechnics.

27 exhibits a high density of 2.039 g cm⁻³, mild sensitivity to mechanical stimuli (with an impact sensitivity of 2 J and friction sensitivity of 20 N), high heat of formation (110 kJ mol⁻¹), and benign detonation performance with a detonation velocity of 8138 m s⁻¹ and detonation pressure of 30.1 GPa. All these properties make 27 a very promising candidate for primary explosives.¹²⁸

Thermal analysis shows that 23a has much better thermal stability than 23b due to the deprotonation of the tetrazol-1-ol with the C-azido-1,2,4-triazole left protonated.⁹⁰ As an energetic material, 24 has better thermal performance with a high decomposition temperature peaking at 274 °C. 24 undergoes two-stage decomposition. The first endothermic process occurs in the temperature range of 115–140 °C, a result of the loss of water molecules. This is followed by an exothermic decomposition during 250–350 °C with a weight loss of 31%, suggesting its energetic nature. The gaseous products were determined to be NO₂, CO₂, and NH. 26 and 27 are also thermally stable with their respective decomposition temperatures being 189 °C and 229 °C.

3.6. Inorganic potassium energetic complexes

Compared with the organic potassium complexes, the inorganic ones usually have little to do with explosives directly. However, potassium dinitramide (KDN, 28) and potassium nitroformate (KNF, 29) could both serve as benign oxidizers.

3.6.1. Potassium dinitramide. 28, which is based on the exotic dinitramide anion (DN⁻), is an energetic oxidizer with possibilities in future solid rocket propellant formulations. Its preparation is kind of complicated (Scheme 8). Experimental and calculated structural parameters of 28 were both obtained, showing that it has a monoclinic unit cell.¹²⁹ A combination of theoretical calculations (by DFT) and surface sensitive spectroscopy was embarked on in order to reveal the interfacial molecular structure of 28. The vibrational sum frequency spectroscopy (VSFS) experiments of monocrystalline 28 crystals reveal a surface consisting of a thin layer of the decomposition product KNO₃ (which is undetectable by IR and Raman spectroscopies because they only detect the bulk composition), along with the possible presence of distorted 28 molecules.¹³⁰

Scheme 8 Synthesis route of 28. Reproduced from ref. 129 with permission from Elsevier Science, copyright [2001].

The thermal behavior of 28 has been researched as well. On a microscopic level, the anisotropic thermal expansion, which may relate to the shock sensitivity, was analyzed by experimentally detecting the thermal motions of the dinitramide anion (which is well represented by the rigid-body model) and the potassium cation.¹³¹ Macroscopically, the decomposition of 28 is anomalous in that it is governed by surface chemical processes involving polarized (twisted) dinitramide anions of reduced stability.³⁰ The larger the surface area, the faster the decomposition. In contrast to most other inorganic salts, 28 decomposes faster in the solid state than in the molten state with autocatalytic characteristics. It decomposes directly in the solid state, primarily into KNO₃ and N₂O accompanied by NO and NO₂. However, DSC studies show that 28 exhibits complicated eutectic, fusion, and liquefaction processes in the solid state. The melting point of the eutectic mixture of 28 and the attached KNO₃ is 109 °C. A decomposition mechanism of 28 was proposed including two main stages. The initial step is the dissociation of a polarized (distorted) dinitramide anion into NO₂ and NNO₂ radical anions, with an activation enthalpy of 150.7 kJ mol⁻¹. The second step is the hemolytic cleavage of the NNO₂ radical anion, producing N₂O and leaving an anionic oxygen radical bound to the metallic surface with an activation enthalpy of 74.1 kJ mol⁻¹. Interestingly, moisture, as well as KNO₃, helps to stabilize 28, which is substantially destabilized under dry and vacuum conditions. Tests show that the initial activation energy for 28 varies with the experimental conditions. Experimental data claims 28 to be stable and it can be classified as an oxidant.

3.6.2. Potassium nitroformate. Salts of nitroformate (NF) have drawn extensive attention due to their superior combustion or explosive performance, as well as being environmentally benign oxidizers. 29 was first reported in 1967 and three polymorphs have been obtained to date. The first two polymorphs, one needle-like and the other rhombohedral, were prepared by the reaction of tetra-nitromethane with KOH and glycerol. The second form is too reactive to be directly measured.¹³² The third polymorph crystallized from acetone at room temperature was found to be suitable for X-ray powder diffraction (XRD) analysis.¹³³

29 crystallizes into a highly symmetric tetragonal crystal system with a space group of I4̄2d and eight formula units per unit cell, though the unit is asymmetric. Electronic structure and Mayer bond order analyses, as well as the deformation electron density iso-surface plots, indicate that 29 is an ionic solid.¹³⁴

While the first two polymorphs of 29 have rather high densities (2.216 and 2.217 g cm⁻³, respectively), the last one possesses an even significantly higher density of 2.325 g cm⁻³. Sensitivity tests prove that 29 meets the United Nation (UN) recommendations for the transport of dangerous goods, with its friction sensitivity greater than 360 N and impact sensitivity greater than 29.6 J.¹³⁵

In terms of thermal behavior, three exothermic peaks were observed in the decomposition of 29 with an onset temperature of 80 °C. The final gaseous products of decomposition were determined to be CO₂, N₂, NO, N₂O, and CO.

4. Conclusions

We have discussed the progress of potassium energetic complexes with a series of examples. In many applications, potassium energetic complexes have numerous advantages over traditional ones, such as being free of pollution, broad applicability, ready accessibility, reasonable price, high coordination numbers and consequently stable 3D networks, high density, and superior explosive properties. The safety and soundness of using green potassium complexes in place of traditional heavy metals in primary explosives have been discussed. Also, their high density (around 2.000 g cm⁻³) and high explosive velocities and pressures (comparable to secondary explosives) are extremely conspicuous, while improvement in traditional materials seems inefficient. They also usually exhibit better performances compared to their parent organics because deprotonation of the organic ligands gives rise to enhanced thermal stability of the system.

However, some challenges still remain. Some complexes are obtained via rather complicated pathways or with quite low yields, both of which apparently restrict further investigations. Simplification of the preparation process and higher yields may rely on changes of the reaction environment or using novel synthetic paths. For complexes that are too sensitive to be extensively applied, we propose two countermeasures. First, the N-oxidation of a tetrazole ring is an effective method to reduce the sensitivity towards mechanical stimuli due to the decreased heat of formation and more intermolecular interactions. Second, coordination water molecules are not energetic ligands as they contain no explosive groups or nitrogen atoms. They increase the hydrogen bonding in the system, thereby reducing the mechanical sensitivity as well as enhancing the stability of the molecular system. However, coordinating water may negatively influence the explosive performance.

The exploration of potassium energetic complexes has been ongoing for decades and seems to be progressing and fruitful. Future research will certainly focus on both the optimization of existing materials and synthesis of new ones, thus developing more applications for cutting-edge science and technology. To produce novel properties in potassium energetic complexes, improved synthesis of new energetic materials is of primary importance since their performance is largely determined by the parent molecule. Another avenue is modifying the properties of existing potassium energetic materials by surface adjustments through choosing surfactants or screening for better solvents and exploring the reaction system environment.

Abbreviations

DNP	2,4-Dinitrophenol
H2DNR	2,4-Dinitroresorcinol
PA	Picric acid
TNPG	Trinitrophloroglucinol
TNR	2,4,6-Trinitroresorcinol
DNBF	4,6-Dinireo-7-hydrobenzofurozan
Cl-14	5,7-Diamino-4,6-dinitrobenzofuroxan
RDX	1,3,5-Trinitro hexahydros-triazine
TATB	1,3,5-Triamino-2,4,6-trinitro benzene
FOX-7	1,1-Diamino-2,2-dinitroethylene
AHDNE	1-Amino-1-hydrazino-2,2-dinitroethylene
NNMPA	2,3-Dihydro-4-nitro-3-(dinitromethylene)-1H-pyrazol-5-amine
DNDZ	2-(Dinitromethylene)-1,3-diazepentane
NTO	3-Nitro-1,2,4-triazol-5-one
HTZ	1H-1,2,3,4-Tetrazole
ANT	Ammonium 5-nitrotetrazolate
HAT	5-Aminotetrazole
ADNP	4-Amino-3,5-dinitropyrazole
LLM-116	4-Amino-3,5-dinitro-1H-pyrazole
AzTTO	5-(5-Azido-1H-1,2,4-triazol-3-yl)tetrazol-1-ol
TABA	2,4,6-Trinitroanilino benzoic acid
DNABT	1,1′-Dinitramino-5-5′-bistetrazolate
KN	Potassium nitrate
KDN	Potassium dinitramide
KNF	Potassium nitroformate
IS	Impact sensitivity
FS	Friction sensitivity
ρ	Density
Mp	Melting point
Tdec	Decomposition temperature (onset temperature at a heating rate of 5 °C min^−1)
Tdet	Detonation temperature
Tdef	Deflagration temperature
IS	Impact sensitivity
FS	Friction sensitivity
ΔHf	Heat of formation
D	Detonation velocity
P	Detonation pressure
Ea	Reaction activation energy
A	Pre-exponential factor
Tb	Critical temperature of thermal explosion
ta	Adiabatic time-to-explosion
ΔHdec	Decomposition enthalpy
DSC	Differential scanning calorimetry
TG	Thermal gravity analysis
DTG	Differential thermal gravity analysis
ARC	Accelerating rate calorimetry
HFC	Heat flux calorimetry

Acknowledgements

We gratefully acknowledge the support of the National Natural Science Foundation of China (10776002) and Project YBKT 1604 from State Key Laboratory of Explosion Science and Technology.

References

1. F. Kukita, J. Membr. Biol., 2011, 242, 119–135.
2. K. Bouzazi, H. Boughzala and H. Zamali, J. Alloys Compd., 2016, 680, 512–521.
3. Z. Gao, A. Sierra, Z. Y. Zhu, S. R. K. Koganti, E. Subbotina, A. Maheshwari, M. E. Anderson, L. V. Zingman and D. M. Hodgson-Zingman, PLoS One, 2016, 11, 17.
4. S. Gondal, H. F. Svendsen and H. K. Knuutila, Chem. Eng. Sci., 2016, 151, 1–6.
5. Y. H. Liu, T. Takasaki, K. Nishimura, M. Yanagida and T. Sakai, Mater. Lett., 2016, 177, 132–134.
6. S. Sehati and M. H. Entezari, Ultrason. Sonochem., 2016, 32, 348–356.
7. Y. X. Tang, C. L. He, L. A. Mitchell, D. A. Parrish and J. M. Shreeve, Angew. Chem., Int. Ed., 2016, 55, 5565–5567.
8. I. T. Tsialtas, S. Shabala, D. Baxevanos and T. Matsi, Field Crop Res., 2016, 193, 94–103.
9. B. D. Wang, C. M. Zhu, Z. Z. Zhang, W. L. Zhang, X. Q. Chen, N. N. Sun, W. Wei, Y. H. Sun and H. F. Ji, Fuel, 2016, 179, 274–280.
10. L. J. Zhai, X. Z. Fan, B. Z. Wang, F. Q. Bi, Y. Li and Y. L. Zhu, RSC Adv., 2015, 5, 57833–57841.
11. F. L. Zhang, H. S. Huang and X. H. Xie, Chin. J. Struct. Chem., 2016, 35, 514–520.
12. D. M. Badgujar, M. B. Talawar, S. N. Asthana and P. P. Mahulikar, J. Hazard. Mater., 2008, 151, 289–305.
13. F. M. Gallant, H. A. Bruck, S. E. Prickett and M. Cesarec, Composites, Part A, 2006, 37, 957–969.
14. A. Mehrkesh and A. T. Karunanithi, ACS Sustainable Chem. Eng., 2013, 1, 448–455.
15. M. B. Talawar, R. Sivabalan, T. Mukundan, H. Muthurajan, A. K. Sikder, B. R. Gandhe and A. S. Rao, J. Hazard. Mater., 2009, 161, 589–607.
16. J. E. Um, T. Yeo, W. Choi, J. S. Chae, H. S. Kim and W. J. Kim, Sci. Adv. Mater., 2016, 8, 164–170.
17. A. L. Winfrey, M. A. Abd Al-Halim, S. Mittal and M. A. Bourham, IEEE Trans. Plasma Sci., 2015, 43, 2195–2200.
18. J. C. Zhang, Y. Du, K. Dong, H. Su, S. W. Zhang, S. H. Li and S. P. Pang, Chem. Mater., 2016, 28, 1472–1480.
19. T. M. Klapötke and N. Mehta, Propellants, Explos., Pyrotech., 2014, 39, 7–8.
20. A. Dutschke, C. Lohrer, S. Seeger and L. Kurth, Chem. Ing. Tech., 2009, 81, 167–176.
21. G. Croteau, R. Dills, M. Beaudreau and M. Davis, Atmos. Environ., 2010, 44, 3295–3303.
22. G. Steinhauser and T. M. Klapötke, Angew. Chem., 2008, 47, 3330–3347.
23. D. Izsák, T. M. Klapötke, R. Scharf and J. Stierstorfer, Z. Anorg. Allg. Chem., 2013, 639, 1746–1755.
24. Q. H. Zhang and J. M. Shreeve, Angew. Chem., Int. Ed., 2014, 53, 2540–2542.
25. S. Zhang, Q. Yang, X. Y. Liu, X. N. Qu, Q. Wei, G. Xie, S. P. Chen and S. L. Gao, Coord. Chem. Rev., 2016, 307, 292–312.
26. J. H. Zhang and J. M. Shreeve, Dalton Trans., 2016, 45, 2363–2368.
27. T. B. Brill, T. L. Zhang and B. C. Tappan, Combust. Flame, 2000, 121, 662–670.
28. P. B. Kulkarni, G. N. Purandare, J. K. Nair, M. B. Talawar, T. Mukundan and S. N. Asthana, J. Hazard. Mater., 2005, 119, 53–61.
29. T. M. Klapötke, M. Rusan and V. Sproll, Z. Anorg. Allg. Chem., 2013, 639, 2433–2443.
30. M. Rahm and T. Brinck, J. Phys. Chem. A, 2010, 114, 2845–2854.
31. R. Haiges and K. O. Christe, Inorg. Chem., 2013, 52, 7249–7260.
32. Y. F. Li, T. L. Zhang, J. G. Zhang, G. X. Ma, J. C. Song, Y. H. Sun and K. B. Yu, Chin. J. Inorg. Chem., 2003, 19, 861–864.
33. H. Y. Chen, T. L. Zhang, J. G. Zhang, L. Yang and X. J. Qiao, Chin. J. Chem., 2007, 25, 59–62.
34. H. S. Huang, T. L. Zhang, J. G. Zhang, X. C. Hu, L. Yang and X. J. Qiao, Acta Chim. Sin., 2008, 66, 847–852.
35. X. H. Ju, Y. L. Lu, X. F. Ma and H. M. Xiao, Propellants, Explos., Pyrotech., 2006, 31, 290–293.
36. X. H. Ju, X. J. Xu and H. M. Xiao, J. Energ. Mater., 2005, 23, 121–130.
37. G. X. Ma, T. L. Zhang, J. G. Zhang, B. Shao, Y. F. Li, J. C. Song and K. B. Yu, Chin. J. Chem., 2004, 22, 131–135.
38. L. Li, T. Zhang, J. Zhang and C. Sun, Energ. Mater., 2008, 16, 9–11.
39. Q. Zhang, X. J. Qiao, J. G. Zhang and X. L. Zuo, Acta Phys.-Chim. Sin., 2009, 25, 1081–1087.
40. P. M. Boorman, P. J. Craig and T. W. Swaddle, Can. J. Chem., 1970, 48, 838–844.
41. Y. F. Li, T. L. Zhang, J. G. Zhang, G. X. Ma, J. C. Song and K. B. Yu, Acta Chim. Sin., 2003, 61, 1020–1024.
42. N. X. Wang, C. Boren and Y. X. Ou, Propellants, Explos., Pyrotech., 1994, 19, 255–256.
43. W. P. Norris, R. J. Spear and R. W. Read, Aust. J. Chem., 1983, 36, 297–309.
44. C. He and J. M. Shreeve, Angew. Chem., 2016, 55, 772–775.
45. D. E. G. Jones, H. T. Feng and R. C. Fouchard, J. Therm. Anal. Calorim., 2000, 60, 917–926.
46. R. J. Spear and W. P. Norris, Propellants, Explos., Pyrotech., 1983, 8, 85–88.
47. Mehilal, N. Sikder, S. K. Chougule, A. K. Sikder and B. R. Gandhe, J. Energ. Mater., 2004, 22, 117–126.
48. Mehilal, A. K. Sikder, S. Pawar and N. Sikder, J. Hazard. Mater., 2002, 90, 221–227.
49. J. L. Feng, J. G. Zhang, T. L. Zhang, Z. M. Li, L. Yang and S. Z. Wang, Acta Phys.-Chim. Sin., 2010, 26, 2613–2618.
50. Y. F. Li, T. L. Zhang and J. G. Zhang, Energ. Mater., 2004, 12, 203–206.
51. G. Herve and G. Jacob, Tetrahedron, 2007, 63, 953–959.
52. G. Herve, G. Jacob and N. Latypov, Tetrahedron, 2005, 61, 6743–6748.
53. S. Rajappa, Tetrahedron, 1981, 37, 1453–1480.
54. N. V. Latypov, J. Bergman, A. Langlet, U. Wellmar and U. Bemm, Tetrahedron, 1998, 54, 11525–11536.
55. B. Buszewski, M. Michel, S. Cudzilo and Z. Chylek, J. Hazard. Mater., 2009, 164, 1051–1058.
56. H. Cai, Y. Shu, W. Yu, Y. Tian, G. Zeng and B. Cheng, Chem. Res. Appl., 2004, 16, 264–265.
57. H. Q. Cai, Y. J. Shu, W. F. Yu, J. S. Li and B. B. Cheng, Acta Chim. Sin., 2004, 62, 295–301.
58. I. G. Cullis and R. Townsley, J. Appl. Mech., 2011, 78, 1–8.
59. W. P. C. de Klerk, C. Popescu and A. van der Heijden, J. Therm. Anal. Calorim., 2003, 72, 955–966.
60. Z. A. Dreger, A. I. Stash, Z.-G. Yu, Y.-S. Chen, Y. Tao and Y. M. Gupta, J. Phys. Chem. C, 2016, 120, 1218–1224.
61. V. G. Kiselev and N. P. Gritsan, J. Phys. Chem. A, 2014, 118, 8002–8008.
62. A. Meents, B. Dittrich, S. K. J. Johnas, V. Thome and E. F. Weckert, Acta Crystallogr., Sect. B: Struct. Sci., 2008, 64, 42–49.
63. V. S. Mishra, S. R. Vadali, R. K. Garg, V. S. Joshi, R. D. Wasnik and S. Asthana, Cent. Eur. J. Energ. Mater., 2013, 10, 569–580.
64. C. C. Ye, F. Q. Zhao, S. Y. Xu and X. H. Ju, Can. J. Chem., 2013, 91, 1207–1212.
65. Y. Zhang, Q. Sun, K. Xu, J. Song and F. Zhao, Propellants, Explos., Pyrotech., 2016, 41, 35–52.
66. C. Zhou, Y. Zhou, X. Huang and X. Wang, Chin. J. Explos. Propellants, 2005, 28, 65–67.
67. G. Herve, Propellants, Explos., Pyrotech., 2009, 34, 444–451.
68. K. Z. Xu, J. R. Song, X. Yang, C. R. Chang, X. K. Yang, H. X. Ma, J. Huang and F. Q. Zhao, J. Mol. Struct., 2008, 891, 340–345.
69. K. Z. Xu, Y. S. Chen, M. Wang, J. A. Luo, J. R. Song, F. Q. Zhao and R. Z. Hu, J. Therm. Anal. Calorim., 2011, 105, 293–300.
70. M. Anniyappan, M. B. Talawar, G. M. Gore, S. Venugopalan and B. R. Gandhe, J. Hazard. Mater., 2006, 137, 812–819.
71. Y. Zhang, Q. Sun, K. Xu, J. Song and F. Zhao, Propellants, Explos., Pyrotech., 2016, 41, 35–52.
72. A. J. Bellamy, A. E. Contini and N. V. Latypov, Propellants, Explos., Pyrotech., 2008, 33, 87–88.
73. C. R. Chang, K. Z. Xu, J. R. Song, B. Yan, H. X. Ma, H. X. Gao and F. Q. Zhao, Acta Chim. Sin., 2008, 66, 1399–1404.
74. Q. Q. Qiu, X. Yang, K. Z. Xu, Z. Gao, J. R. Song, S. H. Yang and F. Q. Zhao, Inorg. Chim. Acta, 2013, 405, 356–361.
75. L. Y. Zhang, K. Z. Xu, J. R. Song, F. Q. Zhao, S. Y. Heng and Y. Y. Wang, Chin. J. Inorg. Chem., 2010, 26, 899–904.
76. H. Gao, Y. H. Joo, D. A. Parrish, T. Vo and J. M. Shreeve, Chem. - Eur. J., 2011, 17, 4613–4618.
77. K. Z. Xu, X. G. Zuo, H. Zhang, B. Yan, J. Huang, H. X. Ma, B. Z. Wang and F. Q. Zhao, J. Therm. Anal. Calorim., 2011, 110, 585–591.
78. W. Zhang, J. Huang, K. Xu, T. An, S. Yang, J. Song and F. Zhao, J. Anal. Appl. Pyrolysis, 2013, 104, 703–706.
79. K. Z. Xu, F. Q. Zhao, J. R. Song, X. L. Ren, H. X. Gao, S. Y. Xu and R. Z. Hu, Bull. Korean Chem. Soc., 2009, 30, 2259–2264.
80. A. A. Dippold and T. M. Klapötke, Chem.–Eur. J., 2012, 18, 16742–16753.
81. N. Kommu, A. S. Kumar, J. Raveendra, V. D. Ghule and A. K. Sahoo, Asian J. Org. Chem., 2016, 5, 138–146.
82. S. Nimesh and A. G. Rajendran, Propellants, Explos., Pyrotech., 2012, 37, 267–274.
83. Y. L. Shao, Y. Pan, Q. Wu, W. H. Zhu, J. S. Li, B. B. Cheng and H. M. Xiao, Struct. Chem., 2013, 24, 1429–1442.
84. A. S. Kumar, V. D. Ghule, S. Subrahmanyam and A. K. Sahoo, Chem.–Eur. J., 2013, 19, 508–517.
85. T. M. Klapötke and T. G. Witkowski, Propellants, Explos., Pyrotech., 2015, 40, 366–373.
86. Y. C. Li, W. Liu and S. P. Pang, Molecules, 2012, 17, 5040–5049.
87. J. G. Zhang, X. Q. Niu, S. W. Zhang, T. L. Zhang, H. S. Huang and Z. N. Zhou, Comput. Theor. Chem., 2011, 964, 291–297.
88. Z. H. Zhao, Z. M. Du, Z. Y. Han, Y. H. Zhang and C. L. He, J. Energ. Mater., 2016, 34, 183–196.
89. T. Mukundan, G. N. Purandare, J. K. Nair, S. M. Pansare, R. K. Sinha and H. Singh, Def. Sci. J., 2002, 52, 127–133.
90. D. Izsák, T. M. Klapötke and S. Reuter, Eur. J. Inorg. Chem., 2013, 2013, 5641–5651.
91. D. Izsák and T. M. Klapötke, Crystals, 2012, 2, 294–305.
92. J. R. Song, Z. X. Chen, H. M. Xiao, R. Z. Hu, F. P. Li and K. B. Yu, Acta Chim. Sin., 1998, 56, 270–277.
93. T. B. Brill, T. L. Zhang and B. C. Tappan, Combust. Flame, 2000, 121, 662–670.
94. T. M. Klapötke and F. X. Steemann, Propellants, Explos., Pyrotech., 2010, 35, 114–129.
95. T. M. Klapötke, M. Stein and J. Stierstorfer, Z. Anorg. Allg. Chem., 2008, 634, 1711–1723.
96. H. B. Zhang, M. J. Zhang, P. Lin, V. Malgras, J. Tang, S. M. Alshehri, Y. Yamauchi, S. W. Du and J. Zhang, Chem.–Eur. J., 2016, 22, 1141–1145.
97. X. Y. Liu, W. J. Gao, P. P. Sun, Z. Y. Su, S. P. Chen, Q. Wei, G. Xie and S. L. Gao, Green Chem., 2015, 17, 831–836.
98. R. A. D. Marco and J. M. Shreeve, Inorg. Chem., 1973, 12, 1896–1899.
99. D. Cave, J. M. Gascon, A. D. Bond, S. J. Teat and P. T. Wood, Chem. Commun., 2002, 1050–1051.
100. H. S. Lee and M. Niemeyer, Inorg. Chim. Acta, 2011, 374, 163–170.
101. J. A. Perez, V. Montoya, J. A. Ayllon, M. Font-Bardia, T. Calvet and J. Pons, Inorg. Chim. Acta, 2013, 394, 21–30.
102. G. F. Swiegers and T. J. Malefetse, Chem. Rev., 2000, 100, 3483–3537.
103. A. R. Mitchell, M. D. Coburn, R. D. Schmidt, P. F. Pagoria and G. S. Lee, Thermochim. Acta, 2002, 384, 205–217.
104. R. D. Schmidt, G. S. Lee, P. R. Pagoria, A. R. Mitchell and R. Gilardi, J. Heterocycl. Chem., 2001, 38, 1227–1230.
105. T. M. Klapötke and J. Stierstorfer, J. Am. Chem. Soc., 2009, 131, 1122–1134.
106. V. Ernst, T. M. Klapötke and J. Stierstorfer, Z. Anorg. Allg. Chem., 2007, 633, 879–887.
107. T. M. Klapötke, D. G. Piercey and J. Stierstorfer, Chem.–Eur. J., 2011, 17, 13068–13077.
108. T. M. Klapötke, C. M. Sabate and J. M. Welch, Dalton Trans., 2008, 45, 6372–6380.
109. T. M. Klapötke and J. Stierstorfer, Helv. Chim. Acta, 2007, 90, 2132–2150.
110. T. M. Klapötke, H. Radies and J. Stierstorfer, Z. Naturforsch., B: J. Chem. Sci., 2007, 62, 1343–1352.
111. A. M. Churakov and V. A. Tartakovsky, Chem. Rev., 2004, 104, 2601–2616.
112. M. Noyman, S. Zilberg and Y. Haas, J. Phys. Chem. A, 2009, 113, 7376–7382.
113. Y. L. Wang, F. Q. Zhao, K. Z. Xu, Y. P. Ji, J. H. Yi, B. Chen and T. An, Inorg. Chim. Acta, 2013, 405, 505–510.
114. D. Fischer, T. M. Klapötke and J. Stierstorfer, Angew. Chem., 2015, 54, 10299–10302.
115. A. Hammerl, G. Holl, T. M. Klapötke, P. Mayer, H. Noth, H. Piotrowski and M. Warchhold, Eur. J. Inorg. Chem., 2002, 834–845.
116. N. I. Golovina and L. O. Atovmyan, Zh. Strukt. Khim., 1967, 8, 307.
117. A. L. Rheingold, B. S. Haggerty, L. M. Liable-Sands and S. Trofimenko, Inorg. Chem., 1999, 38, 6306–6308.
118. C. Di Nicola, F. Marchetti, M. Monari, L. Pandolfo and C. Pettinari, Inorg. Chem. Commun., 2008, 11, 665–668.
119. M. M. Diaz-Requejo and P. J. Perez, J. Organomet. Chem., 2005, 690, 5441–5450.
120. T. Tietz, C. Limberg, R. Stoesser and B. Ziemer, Chem.–Eur. J., 2011, 17, 10010–10020.
121. S. Trofimen, J. Am. Chem. Soc., 1966, 88, 1842–1844.
122. C. Janiak, T. G. Scharmann, P. Albrecht, F. Marlow and R. Macdonald, J. Am. Chem. Soc., 1996, 118, 6307–6308.
123. C. Janiak, S. Temizdemir and S. Dechert, Inorg. Chem. Commun., 2000, 3, 271–275.
124. B. H. Hamilton, K. A. Kelly, W. Malasi and C. J. Ziegler, Inorg. Chem., 2003, 42, 3067–3073.
125. D. Fischer, T. M. Klapötke and J. Stierstorfer, Angew. Chem., Int. Ed., 2014, 53, 8172–8175.
126. S. Trofimen, J. Am. Chem. Soc., 1967, 89, 3170–3177.
127. S. Chao and C. E. Moore, Anal. Chim. Acta, 1978, 100, 457–467.
128. Y. Tang, C. He, L. A. Mitchell, D. A. Parrish and J. M. Shreeve, Angew. Chem., Int. Ed., 2016, 55, 5565–5567.
129. V. A. Shlyapochnikova, M. A. Tafipolskya, I. V. Tokmakovb, E. S. Baskira, O. V. Anikina, Y. A. Strelenkoa, O. A. Luk'yanova and V. A. Tartakovskya, J. Mol. Struct., 2001, 559, 147–166.
130. M. Rahm, E. Tyrode, T. Brinck and C. M. Johnson, J. Phys. Chem. C, 2011, 115, 10588–10596.
131. M. J. Hardie, A. Martin, A. A. Pinkerton and E. A. Zhurova, Acta Crystallogr., 2001, 57, 113–118.
132. N. I. Golovina and L. O. Atovmyan, Nanoscale Res. Lett., 1967, 8, 79.
133. M. Goebel and T. M. Klapötke, Z. Anorg. Allg. Chem., 2007, 633, 1006–1017.
134. H. S. Huang, S. T. Zhang, T. L. Zhang, J. G. Zhang, W. H. Hu and F. L. Zhang, Comput. Theor. Chem., 2013, 1004, 1–4.
135. M. Gobel, T. M. Klapötke and P. Mayer, Z. Anorg. Allg. Chem., 2006, 632, 1043–1050.

---