Synthesis of energetic coordination polymers based on 4-nitropyrazole by solid-melt crystallization in non-ionization condition

Ting-wei Wng , Shu Bu , Kun Wng , Lu Zhng , Zhen-xin Yi , Shun-gun Zhu ,Jin-guo Zhng ,*

a State Key Laboratory of Explosion Science and Technology, Beijing Institute of Technology, Beijing,100081, China

b School of Chemical Engineering, Nanjing University of Science and Technology, 200 Xiaolingwei Street, Xuanwu, Nanjing, 210094, China

c Department of Chemistry,Laboratory of Structure and Functional Regulation of Hybrid Materials, Anhui University, Ministry of Education, Hefei, 230601,China

Keywords:4-Amino-1,2,5-oxadiazole-3-carbohydrazide Crystallization Coordination polymers Laser Primary explosives

ABSTRACT Based on the theory of crystallization, a solvent-free solid-liquid phase crystallization method called solid-melt crystallization was designed to prepare energetic coordination polymers.Two target compounds [Cu(NPyz)4NO3]∙NO3 (ECPs-1) and Cu(NPyz)4(ClO4)2 (ECCs-2) were prepared through programmed heating and cooling by using 4-nitropyrazole (NPyz), (Cu(NO3)2∙5H2O and Cu(ClO4)2∙5H2O as raw materials.In addition, crystallization pre-experiments and annealing experiments also verified the feasibility of the method.Their structures were confirmed by IR, elemental analysis, single-crystal X-ray diffraction and powder X-ray diffraction.The physicochemical properties and sensitivity test results showed that ECCs-2 has better thermal stability (Td = 221 °C), while ECPs-1 is less sensitive to mechanical stimuli (IS = 12 J, FS = 240 N).Calculations based on EXPLO5 and the Kamlet-Jacobs equation showed that ECCs-2 has more considerable detonation performance (P = 25.2 GPa, D = 7.5 km/s).In comparison,the more intuitive results from the HN test,flame test,thermal resistance test and lead plate explosion test revealed that ECCs-2 has an “ acceptable” detonation performance.The laser detonation test also showed that ECCs-2 is a promising excellent laser detonation material (E = 408 mJ, P = 24 W,τ = 17 ms).

1.Introduction

Oxidizers and combustibles are the cornerstones of energetic materials (EMs) because they can ensure that energetic materials undergo violent redox reactions spontaneously and produce specific chemical effects(light,smoke,color,heat,etc.)[1-3].Energetic groups are the heart of energetic materials,and it is because of their great chemical potential that EMs can play an incredible role[4-6].The common energetic groups,such as nitro(-NO2)and azide(-N3)[7-9], generally come from organic synthesis.However, complicated extraction and separation processes and dangerous reagents limit their popularization and application.A green and simple route is to introduce these energetic groups from inorganic salts such as nitrates and perchlorates[10,11].Obviously,energetic coordination polymers (ECPs) or energetic coordination compounds (ECCs)which are composed of energetic organic compounds and inorganic salts make excellent choices.It is undeniable that ECPs have been extensively studied in the field of primary explosives and pyrotechnics [12-14].

Scheme 1.An illustration of coordination site about skeletal and substituent in energetic materials.

Coordination bonds are the core of building ECPs, and we can roughly divide them into two types: “passive” coordination bonds and“active”coordination bonds(Scheme 1,path 2).The former can be understood as: organic ligands which have lost [H＋] form ECPs with metal cations, such as [Co(HTATT)]n (H3TATT: tris(5-amino tetrazole)triazine) [15], [Co(AzTO)(H2O)4∙2H2O]n (H2AzTO: 5,5′-Azotetrazole-1,1′-diol) [16] and [Cu(HBTI)(H2O)]n (H3BTI: 4,5-bistetrazole-imidazole) [17]; the coordination bond is not “pure”as there are other strong forces, including electrostatic force, van der Waals force,etc.;displacement reaction or acid-base reaction is an important driving force for the formation of such coordination bonds.In contrast,organic ligands without losing[H＋]can directly form ECPs with metal cations, and this type of coordination bond can be deemed to be“active”,such as[Zn(DAF)(H2O)4](NO3)2(DAF:3,4-diaminofurazan) [18], Ni(ANQ)2(N3)2(ANQ: 3-amino-1-nitroguanidine) [19] and Ag(H2ur)NO3(H2ur: 4-amino-1,2,4-triazole-3,5-dione)[20];the“active”coordination bond is only formed by a lone pair of electrons and an empty orbital,so it is purer;in order to keep the whole structure electrically neutral, there must be extra anions such as NO3-and ClO4-.From the perspective of preparation method, the construction of “passive” coordination bonds is relatively simple, only requiring the neutralization reaction of the ligand and the inorganic salt.However, ECPs constructed with“active”coordination bonds contain not only organic molecules and metals, but also inorganic anions, so they contain more energy.Therefore, it is more meaningful to prepare ECPs with “active” coordination bonds although the preparation process is complicated.

Pyrazole skeleton is an important raw material for the preparation of secondary explosives, primary explosives, pyrotechnics and other EMs.In recent years, there have been reports about Pyrazole EMs, including TNP (3,4,5-trinitropyrazole) [21], NMethyl-4-azido-3,5-dinitropyrazole and N-Methyl-4-amino-3,5-dinitropyrazole [22].Although using these compounds as ligands,various ECPs can be prepared such as Dipotassium 3,5-dinitramino-4-nitropyrazolate [23] and Lithium 3, 5-Dinitropyrazolate Trihydrate [24].But their preparation has been very difficult due to the use of dangerous oxidizing reagents.Furthermore, none of these ECPs contain anions, greatly reducing their explosive power.Therefore, it is practically significant to select a commercially available pyrazole compound as a raw material to prepare anioncontaining ECPs.

At present,ECPs are prepared only by solvent methods(Scheme 1, path 1).The conventional process is dissolving ligands and inorganic salts in a solvent, and then adopting an appropriate crystallization method to prepare ECPs.The commonly used methods include hydrothermal method, solvent evaporation method and diffusion method.However, the presence of solvents can cause many problems: 1) Solvent compounds can be easily prepared,which affects the performance of ECPs;2) The solubility of ligands affects the preparation of ECPs, which is heavily dependent on the solvent;3)The poor volatility of solvents such as DMSO and DMF limits the crystallization of ECPs; 4) The ligand is easily affected by the solvent and ionized [H], so that the prepared ECPs do not contain anions, which greatly weakens their performance.Therefore, it is very important to design a new crystallization method to prepare ECPs that can avoid the solvent-induced disadvantages.

In this study,we selected 4-nitropyrazole as the raw material to prepare ECPs or ECCs,and developed a new crystallization method named solid-melt crystallization.The molten inorganic salt[Cu(NO3)2∙5H2O and Cu(ClO4)2∙5H2O]act as both a reactant and a solvent.Ligand protonation can be avoided in the solid-liquid reaction system,which ensures the ECPs prepared in this way contain anions.Through theoretical studies, pre-experiments and annealing experiments,the feasibility of preparing target compounds was judged and measured, and ECPs-1 ([Cu(NPyz)4NO3]∙NO3) and ECCs-2 (Cu(NPyz)4(ClO)2)were successfully prepared(see Scheme 2).Their structures were confirmed by X-ray diffraction analysis,their thermal stability was evaluated by TG-DSC, and their mechanical sensitivity and detonation performance were also analyzed.In particular,the mechanism of energy release of the two compounds were investigated.Based on the real data of this study,the existing methods of predicting the properties of energetic materials are questioned.

Scheme 2.The routes to synthesize the target compounds.

2.Experimental

2.1.Warning!

We used some potentially hazardous instruments and reagents to prepare and characterize our target products, which were also dangerous.Although no hazardous incidents have occurred, we strongly recommend that adequate precautions be taken when repeating our work.Gloves, baffles and face shields are very important.Except for some special tests,other operations must be carried out in the fume hood.

2.2.Synthesis of ECPs-1 ([Cu(NPyz)4NO3]·NO3)

Method 1: Mixing 4-Nitropyrazole (6 mmol) with Cu(NO3)2∙5H2O (1 mmol) and placing the mixture in a glass Petri dish(without grinding or other pretreatments);heating it to 110°C using an oven and holding for 20 min (heating rate: 10°C/min);continuing heating to 167°C(purification temperature)and holding for 20 min; then, slowly cooling it down to room temperature(cooling rate: 2°C/min); washing the obtained product with icecold ethanol and drying it to obtain the target product; yield:88.4% (based on 4-Nitropyrazole).

Method 2 (purification process): Redissolving the powder solid obtained from the preparation by Method 1 in anhydrous ethanol(5-10 ml);placing the clear reaction solution in a fume hood for the solvent to slowly evaporate,and finally obtaining a large amount of blue solid after 3 days.Yield: 77.6% (based on 4-Nitropyrazole).

IR (KBr, ν/cm-1): 3131(s), 1553(s), 1519(s), 1412(m), 1317(s),1293(s), 1201(s), 1146(s), 1119(s), 1053(s), 883(s), 819(s), 750(s).Elemental analysis (%) for C12H12N14O14Cu (Mr = 639.861 g/mol):calcd.C 22.50, H 1.88, N 30.63; found C 22.37, H 1.75, N 30.87.

2.3.Synthesis of ECCs-2 (Cu(NPyz)4(ClO)2)

The synthesis method of ECCs-2 is basically the same as that of ECPs-1, and the only difference is that Cu(NO3)2∙5H2O is replaced by Cu(ClO4)2∙5H2O(in Method 1,the purification temperature was changed to 190°C.).The yields of the target products prepared by Methods 1 and 2 were 87.4%and 74.6%,respectively.More details of the two preparation methods are in the supporting information.IR(KBr, ν/cm-1): 3255(m), 3142(s),1561(s),1529(s),1413(s),1356(s),1289(s),1107(s),1052(s),1032(s), 895(s), 819(s), 666(s).Elemental analysis(%) for C12H12N12O16Cl2Cu (Mr = 714.751 g/mol): calcd.C 20.15, H 1.68, N 23.50; found C 20.06, H 1.59, N 23.61.

3.Results and discussion

3.1.Feasibility analysis of synthetic ECPs

Here, we must emphasize that this “real” coordination bond is formed directly with the metals and organic compounds which have not been protonated.Coordination bonds formed through acid-base reactions or metathesis reactions are not within our scope of consideration.Therefore, we performed electrostatic surface potential analysis (ESP) on six compounds including 4-nitropyrazole.The numerical value and position of the extreme point are the judgment criteria since the two are indispensable.

As can be seen from Fig.1, the “nitration” of C1 and C3 has resulted in a severe shift of the lone pair electrons in N1,and only 4-nitropyrazole still maintains strong coordination ability, because the nitro group at the C2 position does not affect the intensity and extreme point position of the lone pair of electrons in N1.In addition,we performed atomic charge analysis on the six pyrazole compounds(Fig.S1),and found that only the N1 of 4-nitropyrazole has strong coordination ability,which is consistent with the result of ESP.This conclusion is also consistent with that of some current articles.The reason is that the ligands of the prepared polynitropyrazole ECPs are all protonated, such as KDNANP (Potassium 3,5-Dinitro-4-aminopyrazolate) [25] and Ag (3,4-DNP) (3,4-DNP: 3,4-dinitropyrazole) [26], and thus only the N1 of 4-nitropyrazole has the real ability to form a coordination bond.This conclusion lays the foundation for our work.

Cu(II) has strong coordination ability, and meanwhile Cu(NO3)2∙5H2O and Cu(ClO4)2∙5H2O have very low melting points,making them suitable for the role of “molten solvent”.In order to find the appropriate reaction temperature, we conducted thermal analysis tests on all the three raw materials.As illustrated in Fig.S2,Cu(NO3)2∙5H2O starts to melt from 90°C.Cu(ClO4)2∙5H2O begins to lose weight at room temperature due to its strong hygroscopicity,its melting point is 82°C,and there is a violent decomposition peak at 95°C(Fig.S3),The melting point of 4-nitropyrazole is very high,reaching 152°C(Fig.S4).Moreover,the melting process of the three raw materials is accompanied by vigorous sublimation.Therefore,we believe that the crystallization reaction is suitable to be carried out around 110°C.

Fig.1.The surface potential analysis (ESP) on six compounds.

Besides theoretical analysis, we also used TG-DSC to further make the actual pre-experimental judgment: the ligand and Cu(NO3)2∙5H2O are put in an Al2O3crucible and tested by heating(both have a mass of 0.2 mg), and if new peaks appear, they are most likely generated by the decomposition of new substances.The preliminary experimental results of ECPs-1 (a mixture of 4-nitropyrazole and copper nitrate) are shown in Fig.3(a).According to the figure,significant weight loss begins to appear at 152°C(the weight loss rate exceeds 60%), and obvious melting of the ligands in the DSC spectra can be observed; surprisingly, a clear exotherm appears at 338°C(green box in Fig.3(a)),which can not be found on the respective TG-DSC data of the raw materials.So,we can boldly guess that this must be the special thermal behavior of the new matter.

To prove our idea, we performed annealing experiments, in which the mixture was heated to 167°C and then naturally cooled to room temperature, and the TG-DSC test was conducted again using the residual solid (Fig.3(b)).As expected, the annealed samples did not show any thermal behavior before 167°C.An obvious melting phenomenon appeared at 172°C (green part in Fig.3(b)),but was not caused by the ligand.Moreover,this melting was accompanied by vigorous sublimation, which involved an insignificant exothermic peak.Perhaps due to the large weight loss rate of the ligand, we did not observe this masked detail in the preliminary experiment.But the thermal decomposition behavior at 311°C/314°C is completely consistent with the pre-experiment.If the exothermic peak was the result of a chemical reaction based on the respective decomposition products of the feedstock, then the melting peak should not have been present.Therefore,it is judged that the new substance ECPs-1 has been obtained,and ECPs-1 has evident melting and sublimation behaviors.

To further verify the correctness of our idea, we replaced Cu(NO3)2∙5H2O with Cu(ClO4)2∙5H2O and explored the feasibility of using this method to prepare ECCs-2.Likewise, we performed both pre-experiments and annealing experiments.The preexperimental results are shown in Fig.3(d), and three very characteristic thermal decomposition behaviors can be observed:1)No exotherm of Cu(ClO4)2∙5H2O occurs at 95°C.2)152°C is the melting point of the ligand.3)A violent exothermic peak appears at 231°C.In addition, the results of annealing experiments (190°C) showed that ECCs-2 were prepared successfully because of the retention of a unique and intense exothermic peak (Fig.3(e)).In summary, all the experimental results show that it is feasible to prepare ECPs using the solid-melt crystallization method.

3.2.Single-crystal structure analysis

The crystals of ECPs-1 and ECCs-2 can be prepared by“Method 1′′, but the cooling rate must be strictly controlled.The second method is also an option, in which with the powder dissolved in absolute ethanol, a large number of crystals can also be prepared.The IR and powder X-ray diffraction of the samples prepared by different methods as well as additional crystallographic data are presented in the supporting information (Figs.S5-S8).

In these two Cu(II)-based ECPs, although Cu(II) exhibits only a single six-coordination mode, the dimensions of the two ECPs are different,which may be caused by the different steric hindrance of anions.The independent unit cell structure of ECPs-1 (crystal system:monoclinic;space group:C2/c(Z=4);crystal density:1.799 g/cm3) contains three components: ligand, Cu(II) and [NO3-] (the simplest ratio is 4:1:2).It can be seen from Fig.2(a) that 4-nitropyrazole is the simplest monodentate ligand, in which N4/N1 is a coordination element (Cu - N1 = 2.005 Å, Cu -N4=2.012 Å),and forms a coordination bond with Cu in the form of "μ2-η1".Surprisingly, the [NO3-] has completely different coordination ability and coordination environment.Both the two [O](O5 and O7)from one[NO3-]form coordination bonds with Cu(II)and form the 1D chain (Cu - O5 = 2.324 Å, Cu - O7 = 2.400 Å).Although the other[NO3-]does not form a coordination bond with the metal, it builds abundant hydrogen bonds with the ligand(N2 - H2…O6 = 2.133 Å, N5 - H5…O8 = 2.336 Å).Besides, there are other hydrogen bonds between the molecules(N2 - H2…O6 = 2.157 Å, N5 - H5…O5 = 2.413 Å), which greatly increase the stability of ECPs-1(as shown in Fig.2(b)).Fig.2(c)is a 3D visual image of ECPs-1, where the LSHE crystals are crossstacked, making the molecular structure more compact.

Fig.2.(a) Preliminary experiments for the preparation of ECPs-1.(b) Thermal behavior of ECPs-1 after annealing at 110 °C.(c) TG-DSC curves of ECPs-1 (5 °C∙min-1).(d) Preliminary experiments for the preparation of ECCs-2.(e) Thermal behavior of ECCs-2 after annealing at 110 °C.(f) TG-DSC curves of ECCs-2 (5 °C∙min-1).

Fig.3.(a)The coordination environment of ECPs-1;(b)Hydrogen bonds in ECPs-1;(c)3D packing diagram of ECPs-1;(d)Coordination environment of ECCs-2;(e)Hydrogen bonds in ECCs-2; (f) 3D packing diagram of ECCs-2.

As shown in Fig.2(d),ECCs-2 also exhibits the six-coordination mode.Although it is consistent with ECPs-1 in terms of structural components(three components),ligand coordination sites(N1,N4,N7 and N10), and bridging methods (μ2-η1), its space group is different from that of ECPs-1 (P21/n, Z = 4).Since each [ClO4-]participates in the formation of coordination bonds, ECCs-2 are only 0D structures.And surprisingly, each ligand and each [ClO4-]are in a different crystal environment,molecular arrangement,and lattice structure, which also results in different Cu-O (Cu -O15=2.490 Å,Cu-O9=2.401 Å)and Cu-N(Cu-N1=1.989 Å,Cu-N4=1.999 Å,Cu-N7=2.008 Å,Cu-N10=1.998 Å)bonds.Of course, as shown in Fig.2(e), [ClO4-] also builds a variety of hydrogen bonds with ligands (N11 - H11…O13 = 2.303 Å, N8 -H8…O12 = 2.166 Å, N11 - H11…O9 = 2.133 Å, N2 -H2…O15 = 2.457 Å), which greatly contributes to increasing the stability and density of the entire molecule.Fig.2(f) is a 3D visual diagram of ECCs-2,and similarly,its LSHE crystal stacking is also a crossing type.

3.3.Physical and chemical stability analysis

For ECPs,the physicochemical stability generally tested includes thermal stability and mechanical sensitivity.Often used as primary explosives and pyrotechnics, ECPs have different properties and requirements.For example, primary explosives and pyrotechnics are generally detonated by thermal or mechanical stimulation.Therefore,the thermal decomposition temperature of ECPs should be kept within an appropriate range.If it is too low, there will be hazards during storage or transportation; when too high, it will lead to difficult detonation.For mechanical sensitivity, there are also have some regulations and requirements.In order to avoid the introduction of impurities in the preparation process affecting the thermal behavior of ECPs, the samples we tested were all crystalline.TG-DSC was adopted to characterize the thermal decomposition behavior (the heating rate was 5°C/min1and the atmosphere was N2).The final test results are shown in Figs.3(c)and 3(f).

The initial decomposition temperature of ECPs-1 is not high,and it exhibits a two-stage thermal decomposition behavior.ECPs-1 began to melt at 172°C, with the melting process being accompanied by vigorous sublimation.The weight loss rate reached 45%due to sublimation, which is very rare for ECPs.Due to the higher purity of the test sample and the more severe sublimation effect,the exothermic peak was not observed in the first segment (but observed in the annealing experiment).A considerable exothermic peak did not appear until 314°C, and the weight loss rate of this process was about 27%.

Compared with the particularity of ECPs-1,the thermal behavior of ECCs-2 is simple and crude.The initial decomposition temperature of ECPs-1 is 220°C,which is an anomaly because the general rule is that the decomposition temperature of[NO3-]-based ECPs is higher than that of[ClO4-]-based ECPs.Even more incredible is that ECCs-2 has no melting point,and its decomposition characteristic is one-step decomposition.It is possible that some solids were ejected due to an explosion during the decomposition process, resulting in a weight loss rate of 93%.It can be seen that the decomposition processes of ECPs-1 and ECCs-2 are completely different(we have studied them in detail and show them in Section 3.6).

The mechanical sensitivity of ECPs can be tested by the standard BAM method (Bundesanstaltfür Materialforschung).The values of impact sensitivity and friction sensitivity obtained from the test are presented in Table 1.The test results reveal that ECPs-1 is less sensitive to mechanical stimulation than ECCs-2, which is consistent with conventional cognition.ECPs-1 showed good tolerance to frictional stimuli up to 240 N, which is much lower than that of many pyrazole-based ECPs, such as Ag (3,4-DNP) (IS = 7.5 J,FS = 115 N) [26].Besides, it is also insensitive to impact stimuli(IS = 12 J), which is just slightly inferior to KDNANP (IS = 18 J,FS = 160 N) [25], and even more insensitive than the Fe(II)-basedtetrazine ECPs [(TriTzPyr)3Fe][ClO4]2(IS = 8.5 J, FS = 183 N) [27].Due to the instability of[ClO4-],ECCs-2(IS=7 J,FS=144 N)is more sensitive to mechanical stimulation than ECPs-1.However,ECCs-2 still has good mechanical stability as it is comparable to Ag(II)-based ECPs Ag(3,4-DNP) [26], and even better than Cu(H2ur)2(-ClO4)2(FS=2 N,IS<1 J)[28]and Ag(ATCA)ClO4(FS=72 N,IS=5 J)[29] prepared by us earlier.Combined with previous research works and current experimental data,ECCs-2 seems to have greater development potential and value as a primary explosive.

Table 1Performance parameters of some selected compounds.

3.4.Energetic properties

The detonation capability of ECPs includes detonation velocity(D) and detonation pressure (p).Currently, only the “Kamlet-Jacobs”(K-J)equation and“EXPLO5”are employed to calculate these two parameters.The standard molar enthalpy of formation(ΔfHθm)is the basic parameter used to calculate the detonation performance, which can be calculated theoretically, or obtained by converting the heat of reaction at constant pressure(ΔCU,measured by the oxygen bomb calorimetry) by using Hess's law.To more accurately evaluate the detonation performance of ECPs, both calculation schemes are adopted in this work.The specific calculation process is shown in the supporting information.

The predicted results are shown in Table 1.According to the table, the P and D values calculated by the EXPLO 5 software are higher than those by the K-J method, but the difference between the two is not large and within an acceptable range.Due to the introduction of the explosive groups [NO3-] and [ClO4-] from the molecular level, the detonation performance of the ligands is greatly increased.The power of [ClO4-] is greater than that of[NO3-], so the detonation performance of ECCs-2 (P = 25.17 GPa,D = 7.45 km/s) is higher than that of ECPs-1 (P = 21.43 GPa,D=7.01 km/s),and this rule is also in line with common sense.The detonation performance of ECPs-1 and ECCs-2 with active coordination bonds is higher than that of Ag (3,4-DNP) (P = 15.88 GPa,D = 5.532 km/s) [26], and can even be compared with KDNANP(P = 22.52 GPa, D = 7.299 km/s) [25].Moreover, since the 4-nitropyrazoles we used is EMs, the powers of ECPs-1 and ECCs-2 are much stronger than the compound we previously prepared Ag(ATCA)ClO4(D = 6.8 km/s) [29].Judging by common sense, we believe that our data are reasonable compared to some ECPs which have the incredible detonation performance but do not even have explosive groups.

3.5.Explosion performance evaluation: non-continuous heating and continuous heating

ECPs are unique in that they continue to burn or explode independently, and do not require additional energy from the external environment.Because of the special physical and chemical behavior of ECPs, their explosive power has been weakened, but this is not considered by the current theoretical calculation methods.Although the predicted results show that both ECPs-1 and ECCs-2 have good detonation performance, there is a big gap between their performances as the special melting and sublimation properties of ECPs-1 greatly weaken its explosive power.This conclusion can be drawn from the TG-DSC data.Theoretical predictions have not paid attention to this particular behavior, so we need more accurate experimental means to measure their power.We find and judge more practical evaluation methods by using continuous and non-continuous heating.

The hot-needle (HN) test is a discontinuous heating method.If the sample can be exploded and detonate the surrounding samples smoothly, it means that the sample has the power of “practical significance”[14].As the exploding sample generates a lot of heat,and this heat can be passed on smoothly and detonate the surrounding samples,indicating it is a persistent process.The HN test results show that only ECCs-2 can produce violent deflagration(see Fig.4).Initially, a bright blue flame was produced.As the burn progressed, the flames became more intense, accompanied by bright lights and a small amount of smoke due to carbon formation.The entire combustion process lasted 0.6 s.As we expected,due to the specificity of ECPs-1, its combustion performance is poor(Fig.S9).

Significant differences between the two ECPs were demonstrated using the discontinuous heating test, and then we wondered whether there were still differences using the continuous heating method.So we used the flame test to see the results.The samples were placed on metal spoons and slowly heated with an alcohol lamp, and then the difference in performance was judged by observing their burning behavior.Before its burning,we clearly saw the melting of ECPs-1 (Fig.5(a)), which is a perfect match with TG-DSC.As the heating progressed, a yellow-green flame was produced, but the burning was gentle.In contrast,ECCs-2 did not melt(Fig.5(e)),and deflagration behavior occurred,with the violent flame area covering the entire shooting window.

In addition, the method of “continuous heating of electric heating wire” was also adopted to further confirm the difference between the two.The sample was wrapped around a resistance wire to make a match head(Figs.6(a)and 6(e)).Direct current was used for continuous heating, and the difference in power was judged by observing the combustion behavior of the samples(0-10 V,2.8 Ω).ECPs-1 was successfully ignited at a voltage of 10 V,and only showed normal combustion performance, while ECCs-2 was ignited smoothly at just 3.5 V, and produced a brighter, more dazzling flame.

3.6.Explosion performance evaluation: detonation methods

Fig.4.The HN experiment of ECCs-2: (a) 0.017s; (b) 0.167 s; (c) 0.223 s; (d) 0.367 s; (e) 0.417 s; (f) 0.533 s.

Fig.5.Flame test of ECPs-1 (a)-(d): (a) 0 s; (b) 0.45 s; (c) 0 0.9 s; (d) 1.8 s, and ECCs-2 (e)-(h): (e) 0 s; (f) 0.22 s; (g) 0.4 s; (h) 0.6 s.

Fig.6.Thermal resistance tests of ECPs-1 (a)-(d): (b) 0.2 s; (c) 1.33 s; (d) 2.67 s, and ECCs-2 (e)-(h): (f) 0.2 s; (g) 0.9 s; (h) 1.02 s.

Laser is a new tool to detonate ECPs, so we also used laser to ignite these the two ECPs (the specific operation method and instrument parameters are shown in the supporting information).ECCs-2 has excellent detonation performance(P=24 W,τ=17 ms,E=408 mJ),and this conclusion is consistent with the previous test results.Under the irradiation of the laser, ECCs-2 was ignited and detonated, and a bright blue flame illuminated the entire field of view(see Fig.7).However,none of the ECPs-1 samples changed in the maximum range of our instrument(Pmax=30 W,τmax=50 ms,Emax= 1500 mJ).

Therefore,it is concluded that ECCs-2 possesses stronger power.Then we further measured their power by using the lead plate explosion test (percussion cap: 6#; lead plate: 3 mm).The experimental results showed that ECPs-1 cannot detonate RDX,but ECCs-2 (6.4 mm, Fig.8(a)) can fully detonate RDX under the same conditions.

Fig.8.Damage of lead plates of(a)ECCs-2 and(b)ECPs-1;(c)Pressure trace of ECPs-1 and ECCs-2.

Fig.7.Deflagration of ECCs-2 during laser irradiation: (a) 0 s; (b) 0.03 s; (c) 0.05 s; (d) 0.12 s.

Fig.9.The initial decomposition of (a) ECPs-1 and (b) ECCs-2 with the corresponding time.

Table 2Typical structural parameters of ECPs-1 and ECCs-2.

The most intuitive time-stress test was performed to further prove our idea,and the experimental results(time-pressure curve)are shown in Fig.8(c).As can be seen from the figure,although the maximum gas pressure generated by ECCs-2(15.15 GPa)was higher than that by ECCs-2(4.16 GPa),it could be reached in 0.048 ms,and the rate reached 3.16×105MPa/s,which was higher that of ECPs-6(7.14 × 102MPa/s).This is a good illustration of the reasonable performance of ECCs-2.Therefore, analyses and conclusions made through tests are the core of determining the detonation capability of ECPs or EMs.

3.7.Decomposition mechanism

Only the difference in anions causes the differences in the physicochemical properties, stability and output energy of these two compounds.To further investigate the source of their differences, we investigated their decomposition mechanisms.Detailed information on research methods,procedures and software can be obtained from the SI.The CPMD methods were applied to understand the relationship between their structures and stabilities in the decomposition processes.The simulation curves of time(step)-potential energy(a.u)relationship are presented in Fig.S11.All the initial decomposition pathways for (1) and (2) are summarized in Fig.9.The typical structure parameters are listed in Table 2.

For ECPs-1,although O-Cu(Cu48-O49)has the lowest WBI and MBO values,they are not the first coordination bond to be broken.It seems that the N-Cu bond is less stable,so the cleavage of ECPs-1 starts from here, and its decomposition process consists of two parts (Fig.9(a)).First, the N-Cu bonds were broken to form two molecule ligands (at 0.642 ps).Since the temperature at this moment was greater than the melting point of the ligand, an obvious endothermic peak appeared on the DSC.The ligands generated by the decomposition were sublimated under the action of heat, so the weight loss of TG was significantly reduced.Subsequently, with the increase of temperature, the remaining O-Cu bonds started to break and form [NO3] radicals (at 0.721 ps).At the same time,a violent redox reaction occurred,leading to a clear exothermic phenomenon.

However, the decomposition mechanism of ECCs-2 differ significantly.Although N-Cu and O-Cu coordination bonds have the smallest WBI and MBO values, they seem to be more stable(Fig.9(b)).In ECCs-2, the [ClO4-] appears to be more reactive, so it breaks with less thermal stimulation.First, the Cl-O bond was broken and[ClO3]radical was produced at 0.472 ps.Since there was no ligand generation,there was no endothermic peak in DSC and no weight loss in TG.As the temperature rose, more and more [ClO3]free radicals were produced (at 0.474 ps).The high-activity free radicals reacted with the reduction components, and a fierce chemical reaction occurred.So only one violent exothermic behavior was observed in ECCs-2.Through the “melting point” to find those special decomposition phenomena, we can find these clues in some of Klap¨otke's research works [30,31].Unfortunately,the author did not provide specific explanations.The theoretical calculation results are in complete agreement with our experimental results.Due to the different decomposition mechanisms of the two ECPs,their mechanical sensitivities also vary.All the above conclusions prove that our actual operation and characterization method is very intuitive and effective,which also indirectly shows that the theoretical calculation method is flawed.We do not deny the contribution of theoretical calculation to this study, but when using it, we must consider its usability and accuracy.

4.Conclusions

This work innovatively utilized the melt-solid crystallization method to prepare two compounds:[Cu(NPyz)4NO3]∙NO3(ECPs-1)and Cu(NPyz)4(ClO4)2(ECCs-2).The prepared ECPs by this new method was preliminarily judged by pre-experiment and annealing experiment.The results of detonation performance and physicochemical tests showed thatECCs-2has better performance:excellent thermal stability (Td= 221°C), good mechanical sensitivity (FS = 144 N, IS = 7 J), and decent detonation performance(P=25.2 GPa,D=7.5 km/s).Since the current calculation methods for predicting detonation performance are flawed, we chose the more intuitive and accurate HN test, flame test and thermal resistance test to prove our idea.As we expected, all test results consistently showed thatECCs-2performed better.And this law was also proved by the lead plate detonation experiment and laser detonation experiment (E = 408 mJ, P = 24 W, τ = 17 ms).It is important to point out that the decomposition model equation obtained from theoretical calculation fit perfectly with the DSC data.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We are thankful to the projects of National Natural Science Foundation of China(Grant Nos.22175025 and 21905023)for their generous financial support.

Appendix A.Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.dt.2022.12.015.