Advertisement

Journal of Thermal Analysis and Calorimetry

, Volume 133, Issue 1, pp 683–693 | Cite as

Thermal hazard analysis and combustion characteristics of four imidazolium nitrate ionic liquids

  • Wei-Cheng Lin
  • Wen-Lung Yu
  • Shang-Hao Liu
  • Shih-Yu Huang
  • Hung-Yi Hou
  • Chi-Min Shu
Article

Abstract

Thermogravimetry and differential scanning calorimetry (DSC) were used to investigate the thermal stability of four nitrate-based ionic liquids. The variations of thermal behavior for different numbers or lengths of alkyl substituents for imidazolium cations were analyzed systematically. Long-term stability and operating temperature for 1.0% mass loss during 10 h (T0.01, 10 h) were estimated using model-free perdition methodologies. The results of T0.01, 10 h were predicted to be 78.73 and 81.59 °C for [Bim][NO3] and [Mim][NO3], respectively. The apparent activation energy (Ea) was obtained using four isoconversional methods at various heating rates through DSC experiments. This study confirmed that [Bim][NO3] and [Mim][NO3] decomposed swiftly, and the gaseous products can result in ignition and continuous combustion.

Keywords

Ionic liquids Long-term stability Model-free perdition methodologies Isoconversional methods Combustion 

List of symbols

A(α)

Pre-exponential factor (s−1)

\(\alpha\)

Degree of conversion (dimensionless)

\(\beta\)

Heating rate (°C min−1)

Ea

Apparent activation energy (kJ mol−1)

f(α)

Most probable kinetic function (dimensionless)

\(\it {\text{G}}\left( \alpha \right)\)

Integral mechanism function (dimensionless)

ΔHd, avg

Average heat of decomposition (J g−1)

R

Universal gas constant (8.31415 J K−1 mol−1)

T

Temperature (°C)

T0.01, 10 h

Temperature at mass loss reaches 1.0% for 10 h (°C)

t

Time (min)

Introduction

Green chemistry is one of the widely discussed concerns for the design of safer chemicals and industrial processes and promises to considerably reduce the impact of hazardous substances on humans and the environment. Ionic liquids (ILs) are molten salts that typically have low melting points of less than 100 °C [1] and combine with organic cations and organic or inorganic anions. Because no vapor pressure is generated during their use and storage, ILs are considered as a new type of green solvent that can replace traditional organic solvents and reduce the physiological harmfulness of volatile organic compounds [2]. ILs have several unique physical properties, such as low vapor pressure, high thermal stability, nontoxicity, and nonflammability [3]. Chen et al. used a phosphorous IL as catalytic/synergistic agent to ameliorate the flame retardant effect on thermoplastic polyurethane, a high fire hazard material [4, 5]. ILs can be synthesized by distinct designs, such as adding functional groups on cation structures or combining with different types of anions [6]. Because of these advantages, ILs have been used for various applications, such as organic reaction solvents for high-temperature conditions [7, 8], solvents for cellulose [9, 10], electrolytes for solar cell [11, 12], and propellants [13].

Recent studies have focused on ILs because of their thermal safety and stability. Martyn et al. [14] indicated that some ILs exhibit violent thermal decomposition and can produce flammable products, resulting in ignition or spontaneous combustion. Liaw et al. [15] reported that vaporization of ILs engendered flammability. Cao et al. [16] investigated the thermal stability of 66 ILs by thermogravimetry and classified some ILs with nitrate-based anions as relatively unstable.

The physicochemical properties of ILs have relevance to either anions or cations. By incorporation of energetic functionalities, ILs can be designed as energetic materials [17]. Wellens attempted to synthesize pyrrolidinium nitrate ([Pyrr][NO3]) by removing excess water and nitric acid in a rotary evaporator with a drying temperature of approximately 80–100 °C; an explosion occurred during the heating process because of severe oxidative decomposition of nitrate [18]. Using an accelerating rate calorimeter, Smiglak et al. demonstrated that two nitrate-based ILs, [Bim][NO3] and [Mim][NO3], caused a violent exothermic reaction and an increase in pressure even under vacuum conditions without oxygen [19]. Several studies have indicated that oxidative decomposition of nitrate can lead to combustion and thermal hazards [20, 21, 22, 23]. The Classification, Labelling, and Packaging of Substances and Mixtures regulation established by the European Chemicals Agency has been defined [NO3] as a known explosive substance [24]. Therefore, the thermal properties of nitrate-based ILs should be further investigated.

For the aforementioned reasons, we applied differential scanning calorimetry (DSC) for comparing the thermal properties in different types of cations and anions of six ILs, 1-butyl-3-methylimidazolium bromide ([Bmim][Br]), 1-butyl-3-methylimidazolium chloride ([Bmim][Cl]), 1-butyl-3-methylimidazolium nitrate ([Bmim][NO3]), 1-methyl-3-methylimidazolium nitrate ([Mmim][NO3]), 1-butylimidazolium nitrate ([Bim][NO3]), and 1-methylimidazolium nitrate ([Mim][NO3]). Then, we further explored nitrate-based ILs. The apparent activation energy (Ea) was calculated using Starink, Kissinger–Akahira–Sunose (KAS), Flynn–Wall–Ozawa (FWO), and differential isoconversional methods in DSC experiments [25, 26, 27]. Thermogravimetry (TG) was applied to obtain the pyrolysis kinetics for four nitrate-based ILs. The long-term isothermal reactions were predicted and described using a model-free methodology.

Experimental

Reagent

[Bmim][Br] (> 98 mass%) and [Bmim][Cl] (> 98 mass%) were purchased from Sigma-Aldrich Chemistry (St. Louis, Missouri, USA). 98 mass% of [Bmim][NO3], [Mmim][NO3], [Bim][NO3], and [Mim][NO3] were procured from Fusol Material Co., Ltd. (Tainan, Taiwan, ROC). The compounds’ abbreviations and structures are listed in Table 1. The chosen ILs are all imidazolium salts; anions include chlorine and bromine that belong to halogen and nitrate.
Table 1

List of experimental ionic liquids investigated with abbreviations and structures

Analytical methods

Mettler Toledo DSC 821e was used to determine the thermal curve through dynamic scanning experiments. In all DSC experiments, a gold-plated sample pan was used. The baseline was calibrated within a temperature range of 30–450 °C; the temperature and enthalpies were calibrated using the melting point and fusion heat (expected to be 156.60 °C and 28.45 J g−1, respectively). To compare the exothermic conditions of different types of anions (halogens or nitrate) and cations (long or short alkyl chain), the temperature ramp was 30.0–350.0 °C with a heating rate of 4.0 °C min−1. Four nitrate-based ILs were used within a scanning range of 30.0–400.0 °C at five heating rates (0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1).

TG analysis was performed using a PerkinElmer Pyris 1 to obtain the mass loss curves of [Bmim][NO3], [Mmim][NO3], [Bim][NO3], and [Mim][NO3]. The temperature range was set as 30.0–500.0 °C, and the heating rates were 0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1. Isothermal experiments were accomplished using distinct temperatures corresponding to the types of ILs. N2 was used as inert gas in all cases with a flow rate of 20.0 mL min−1.

Kinetic theory

The thermokinetics of [Bmim][NO3], [Mmim][NO3], [Bim][NO3], and [Mim][NO3] was studied with DSC and TG experiments. The Ea values were obtained using Starink, KAS, FWO, and differential isoconversional methods, and the results of each method were compared. The four kinetic analysis methods are explained as follows.

Starink method

$$\ln \left[ {\frac{\beta }{{T^{1.92} }}} \right] = C{-}1.0008\frac{{E_{\text{a}} }}{RT}$$
(1)
where β is the heating rate, T is the absolute temperature, R is the ideal gas constant, and C is a constant. Ea can be obtained from the slope in the linear regression trendline using ln[β (T−1.92)] versus 1000 T−1.

KAS method

$$\ln \left( {\frac{\beta }{{T^{2} }}} \right) = \ln \left[ {\frac{AR}{{E_{\text{a}} G\left( \alpha \right)}}} \right]{-}\frac{{E_{\text{a}} }}{RT}$$
(2)
where α is the conversion fraction and G(α) is the integral mechanism function. The terms of ln[AR (Ea G(a))−1] can be considered as a constant because of the coincident reaction degree (isoconversion) for different values of β. Thus, Ea can be obtained from the slope in the linear regression trend line using ln(βT−2) versus 1000 T−1.

FWO method

$$\ln \beta = \ln \left( {\frac{{AE_{\text{a}} }}{RG\left( \alpha \right)}} \right){-}5.3308{-}1.0516\frac{{E_{\text{a}} }}{RT}$$
(3)
The FWO method is similar to the KAS method. The Ea can be acquired from the slope in the linear regression trend line by ln(β) versus 1000 T−1.

Differential isoconversional method

$$\ln \left( {\frac{{{\text{d}}\alpha }}{{{\text{d}}t}}} \right) = \ln \left( {A'(\alpha )} \right){-}\left( {\frac{{E_{\text{a}} (\alpha )}}{RT(t)}} \right)$$
(4)
$$A'(\alpha ) = A(\alpha ) f(\alpha )$$
(5)
where \(A'(\alpha )\) can be considered as a constant for same a. Thus, by considering ln(dα dt−1) versus T−1 for different heating processes of the form T(t), the slope of linear regression trend line in each given conversion can be calculated as E(α).

These four isoconversional methods assume E(α) to be a variable (function of α), and all four methods do not formalize any precise reaction mechanism. In this study, a was divided into 100 equal parts from 0 to 1 for the four kinetic analysis methods.

Combustion test using thermograph and flash point analyzer

The apparatus of the combustion test is shown in Fig. 1. The temperature was recorded using a GR–3500 (Keyence, Osaka, Japan). Each IL sample weighed approximately 0.5 g; each IL sample was placed in a glass test tube. The ILs were continuously heated using an alcohol burner and ignited by an ignition source located on top of the tube.
Fig. 1

Apparatus for combustion test in conjunction with thermograph: a ignition source, b test sample, c thermograph, d temperature change of samples, e temperature curve of sample, and f ambient temperature

An HFP 360 Pensky–Martens flash point analyzer was used to measure the flash point temperature. The experiment was conducted according to the ASTM D 93 A standard test method, which can be used for a wide variety of viscous liquids [28].

Results and discussion

Thermal behavior of different ILs

The preliminary thermal behavior for different formations of ILs was investigated through DSC. Experimental results showed that anion structure has a considerable effect on the chemical and physical properties of ILs. Figure 2 represents the DSC results for six ILs at a heating rate of 4.0 °C min−1. The thermal curves primarily reflect the true characteristics of reactivity. As shown in Fig. 2, the thermal properties exhibited considerable variations among the three anions: [Br], [Cl], and [NO3]. ILs containing nitrate anions release heat at approximately 160–350 °C; however, thermal analysis curves showed that ILs combined with bromide or chloride ion have high thermal stability at 30–350 °C. The nitrate structure was strong oxidant which was readily decomposed by heat and produced oxygen under high-temperature condition [29]. Meanwhile, the oxidative capacity of [NO3] might actuate the exothermic reaction then incur the decomposition of ILs. The results indicated that instability was triggered by adding the nitrate structure; these findings are consistent with those found in the literature [16, 30]. Nonetheless, higher alkyl substituents in imidazolium cations advantageously promoted the thermal stability. However, a long alkyl chain used as a substituent in imidazolium-based cations presented an indistinct effect on the tendency of exothermic reaction.
Fig. 2

DSC thermal curves for a [Bmim][Br], b [Bmim][Cl], c [Bmim][NO3], d [Mmim][NO3], e [Bim][NO3], and f [Mim][NO3] at dynamic heating rate of 4.0 °C min−1

Figure 2 shows that [Bim][NO3] and [Mim][NO3] exhibited a salient exothermic reaction under relatively low temperature. All four nitrate-based ILs contained two unequal exothermic peaks. The main exothermic reaction for [Bim][NO3] and [Mim][NO3] occurred at the first peak; however, it occurred at the second peak for [Bmim][NO3] and [Mmim][NO3]. The results for [Bim][NO3] and [Mim][NO3] demonstrated a high hazard potential.

Investigation on pyrolysis reaction via TG and DSC experiments

DSC scanning results revealed that [Bim][NO3] and [Mim][NO3] decomposed readily at 150–250 °C. A detailed analysis of two active ILs using the DSC and TG techniques is required. Figures 3 and 4 present the DSC and DTG curves at dynamic heating rates of 0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1 for [Bim][NO3] and [Mim][NO3], respectively. The two ILs lost enormous mass (> 99%) at temperatures lower than 250 °C in TG experiments. The pyrolysis reactions for the two ILs were both determined in a single step because the DTG curve illustrated only one main peak. Figures 3 and 4 show that the DSC thermal curves contained two exothermic peaks for each IL, and the first exothermic peak appeared at a higher temperature than the peak of DTG. The aforementioned results indicated that the thermal decomposition reaction produced a gaseous product and engendered pressure accumulation in the sealed sample pan in DSC. The exothermic reaction was delayed because the high-pressure condition restrained gasification in the decomposition process. Figures 3 and 4 show no DTG peak relevant to the second peak of the DSC. The second exothermic peak might be ascribed to the decomposition from the gaseous products.
Fig. 3

DTG and DSC thermal curves for [Bim][NO3] at dynamic heating rates of 0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1

Fig. 4

DTG and DSC thermal curves for [Mim][NO3] at dynamic heating rates of 0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1

The thermal analysis results for four nitrate-based ILs are summarized in Table 2. Two exothermic reactions existed in four ILs, and the average heat (ΔHd, avg) of the two reactions is indicated in Table 2. [Mim][NO3] expressed the highest total ΔHd, avg of 1894.3 J g−1, and that for [Mmim][NO3] was 1630.4 J g−1. Because [Mim][NO3] and [Mmim][NO3] exhibited smaller molecular weights of 145 and 159 g gmol−1, respectively, the experimental results showed that [Mim][NO3] and [Mmim][NO3] possessed higher energy density than [Bim][NO3] and [Bmim][NO3] did. Based on the analysis results, some conclusions can be drawn as follows:
Table 2

Thermokinetic parameters for four nitrate-based ILs by DSC experiments

Substance

ΔHd, avg/J g−1

E a * /kJ mol−1

1st reaction

2nd reaction

1st reaction

2nd reaction

[Bim][NO3]

1339.9

105.7

109.5

148.0

[Mim][NO3]

1488.8

405.5

122.6

151.0

[Bmim][NO3]

143.9

814.0

75.5

147.4

[Mmim][NO3]

149.6

1480.8

97.3

115.2

* Calculated by differential isoconversional method

  • No other prominent validation existed for the change in ΔH when the alkyl group was augmented with imidazolium.

  • Thermal curves were similar if the system prolonged only the alkyl chain.

Investigation of apparent activation energy

Ea is the minimum energy required to initiate a chemical reaction. We evaluated Ea for [Bim][NO3] and [Mim][NO3] at five heating rates of 0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1 through DSC. Four isoconversional methods, expressed in Eqs. (1)–(5), were adopted to calculate the Ea in the conversion progress. The aim of this study was to confirm the relationship between the four methods. The first reaction part of [Bim][NO3] and both reaction peaks of [Mim][NO3] were analyzed in depth. The results of the Ea evolution trending with a are reported on the following, as demonstrated in Figs. 5 and 6. The average Ea obtained by each method is given in Table 3. Starink and KAS methods depicted similar developments and values of Ea in the three cases. In FWO method, the Ea translated upwards obviously from the results of the Starink method, however, possessed the parallel progresses. The differential isoconversional method could be used as a basis of model-free prediction. The peculiar curve of Ea trends with a was acquired, which presented more drastic amplitude. Nonetheless, the Ea average values were still very close to other methods.
Fig. 5

Apparent activation energy analysis for [Bim][NO3] by Starink, Kissinger–Akahira–Sunose, Flynn–Wall–Ozawa, and differential isoconversional methods

Fig. 6

Apparent activation energy analysis for two reaction steps of [Mim][NO3] by Starink, Kissinger–Akahira–Sunose, Flynn–Wall–Ozawa, and differential isoconversional methods

Table 3

Calculation of apparent activation energy for [Bim][NO3] and [Mim][NO3] by Starink, Kissinger–Akahira–Sunose, Flynn–Wall–Ozawa, and differential isoconversional methods

Substance

Ea, avg/kJ mol−1

Starink

KAS

FWO

Isoconversional method

[Bim][NO3]

104.8

104.6

106.9

109.5

[Mim][NO3]

124.4

124.2

125.5

122.6

150.7

150.4

152.2

151.0

Through the DSC experiments, the Ea for the four nitrate-based ILs was calculated by differential isoconversional methods, as given in Table 2. The first reaction step of [Bmim][NO3] had the lowest Ea, avg, namely 75.5 kJ mol−1; the calculation results indicated that [Bmim][NO3] should be active and decompose readily. However, ΔHd, was insignificant in the first reaction step; device designs could use this to promote thermal safety. Analogous characteristics were obtained for [Mmim][NO3].

Prediction of long-term thermal stability

Although ILs are asserted to have high thermal stability, the factors of long-term thermal stability such as temperature that leads to a = 1.0% (mass loss in 1.0%) for a 10-h isothermal reaction (T0.01, 10 h) are still significant for estimating the operating temperature. Therefore, TG was used to observe the mass loss of four [NO3]-based ILs under constant heating rate conditions (0.5, 1.0, 2.0, 4.0, and 8.0 °C min−1).

Before conducting the prediction, thermokinetic parameters acquired by TG data for four ILs were stressed. The average Ea is displayed in Table 2. Meanwhile, an order from lower to higher value was [Bim][NO3] < [Mim][NO3] < [Bmim][NO3] < [Mmim][NO3]. The aforementioned Ea analysis indicated inconsistent consequences from the DSC experiments. Figure 7 shows the DTG and DSC curves for the four ILs at a heating rate of 4.0 °C min−1. In TG analysis, the mechanism of pyrolysis for the four ILs was described in a single-step degradation procedure. The following peculiar outcomes were obtained: [Bim][NO3] and [Mim][NO3] had higher Ea in DSC than in TG. However, [Bmim][NO3] and [Mmim][NO3] showed lower Ea values in DSC than in TG. The test results may lead to overestimation of long-term thermal stability using TG analysis. Most tests were conducted in atmospheric pressure, implying no pressure accumulation. Therefore, the investigation of long-term thermal stability was conducted through TG experiments.
Fig. 7

DTG and DSC analysis for four ILs with the heating rate of 4.0 °C min−1

The pyrolysis kinetics were obtained using the differential isoconversional method, which was used to validate the efficacy of calculation of the thermokinetic parameters and predict the long-term isothermal reactions. The essence of the model-free prediction is expressed in Eq. (6):
$$t_{{\upalpha{\text{i}}}} = \int\limits_{\alpha 0}^{\alpha i} {\frac{{{\text{d}}\alpha }}{{A'(\alpha ) \exp \left( {\frac{{{-}E(\alpha )}}{RT(t)}} \right)}}}$$
(6)
where tαi is the reaction time for the corresponding α and T(t) is the temperature treatment relevant to time. T(t) can be a constant under isothermal conditions. Equation (6) presents the isothermal reactions at various isothermal temperatures. The prediction interval for the temperature was from 70 to 268 °C for [Bim][NO3] and [Mim][NO3] and from 150 to 400 °C for [Bmim][NO3] and [Mmim][NO3], investigated every 2 °C. The time–temperature conversion plot is shown in Fig. 8. Isothermal TG experiments for [Bim][NO3] and [Mim][NO3] were conducted to validate the accuracy of the prediction method. Table 4 presents the experimental values and an overview of the prediction method. The T0.01,10 h values were predicted as 78.73, 81.59, 181.11 (182.9 °C in [29]), and 182.84 °C for [Bim][NO3], [Mim][NO3], [Bmim][NO3], and [Mmim][NO3], respectively.
Fig. 8

Time–temperature conversion diagram for a [Bim][NO3], b [Bim][NO3], c [Bmim][NO3], and d [Mmim][NO3] by model-free isothermal TG prediction

Table 4

Calculation of long-term stability, experimental and prediction values

Substance

Isothermal temp./°C

Time (α = 0.1)/h

Time (α = 0.9)/h

Exp.

Pred.

Exp.

Pred.

[Bim][NO3]

120

14.27

22.06

NA

NA

130

6.85

10.99

138.33

127.20

140

3.83

5.75

69.00

63.65

150

2.83

3.14

36.15

33.00

160

1.58

1.78

17.93

17.69

170

0.97

1.04

11.57

9.78

230

0.12

0.07

0.48

0.48

[Mim][NO3]

100

147.6

148.6

NA

NA

130

5.50

12.13

95.40

125.76

140

4.57

5.89

54.65

57.91

150

1.08

3.00

28.35

27.74

160

1.97

1.59

16.95

13.79

170

1.48

0.87

7.20

7.09

180

1.42

0.49

4.82

3.76

190

0.35

0.29

2.55

2.06

NA Not applicable

Combustion tests

Previous studies have reported on the flammability of ILs. However, we tested the combustion characteristics of decomposed products. A comparison of the relationship between mass loss and heat release of [Bim][NO3] and [Mim][NO3] shows that a decomposition reaction could have occurred promptly at less than 200 °C. The results of the combustion tests are summarized in Table 5; [Mim][NO3] was dissolved from solid to liquid, and then [Bim][NO3] and [Mim][NO3] in the liquid state began boiling. The color, in turn, changed from colorless to yellow, accompanied by yellowish brown gas generation. As the yellow gas accumulated swiftly, a flash fire phenomenon appeared because of the igniter. Above the flash point, the decomposed gas products continued to be produced and started to burn until yellow ILs decomposed and turned into black coke. The flammability was caused by the gaseous products of decomposition; meanwhile, the occurring exothermic reaction may trigger gaseous pyrolysis further, leading to continuous combustion.
Table 5

Thermal decomposition phenomenography and temperature of [Bim][NO3] and [Mim][NO3]

Tests by flash point analyzer

In the flash point phenomenon, flammable liquids produce sufficient vapor after heating and result in momentary ignition on application of a flame. The results of thermal analysis show that the combustion properties of ILs may be generally different from those of flammable liquids. Nonetheless, testing using flash point analyzers can produce standardized results for judging the flammability of different ILs.

After analyzing the four nitrate-based ILs, we further compared two considerably unstable ILs, [Bim][NO3] and [Mim][NO3]. The flash point values of [Bim][NO3] and [Mim][NO3] were measured at 207 and 217 °C because the heating rate of the standard test method ASTM D 93 A was set at 5.0–6.0 °C min−1. The mass loss and mass loss derivative versus temperature are shown in Fig. 9; the flash points of the two ILs corresponded to the TG curve (obtained at heating rate 5.0 °C min−1) and had a mass loss ratio of approximately 95%.
Fig. 9

Mass loss and derivative mass loss versus temperature combined with flash points of a [Bim][NO3] and b [Mim][NO3]

The practical test results were different from the theoretical predictions regarding flash points. The flash points of ILs are mainly relevant to the combustion of their decomposed products rather than that of their evolved vapors. For the aforementioned behaviors, we analyzed the flammable hazards and loss prevention in the manufacturing process.

Conclusions

This study corroborated that [NO3] in ILs can enhance exothermic performance. ILs with [NO3] structure began thermal decomposition at approximately 160–350 °C; however, ILs combined with [Br] and [Cl] displayed exceptional thermal stability until 350 °C. Further investigation of the thermal properties indicated that ILs can exhibit higher activity by reducing the alkyl number from imidazolium of the IL’s cation. For example, the pyrolysis temperature of [Bim][NO3] obtained by TG was reduced by more than 100 °C than [Bmim][NO3]. In the TG and DSC analyses, the consequents demonstrated consistent reaction behavior when the length of the alkyl chain was increased. For example, the cation of imidazolium substituted butyl for methyl exhibited similar reaction, such as the correlation of [Mim][NO3] and [Bim][NO3] or [Mmim][NO3] and [Bmim][NO3].

[Bim][NO3] and [Mim][NO3] were observed to have extremely low Ea values in the TG investigation. The T0.01, 10 h values were predicted as 78.73 and 81.59 °C, which were unexpected results because the temperature was lower than 100 °C.

In conclusion, the main characteristics of the detrimental effects for four nitrate-based ILs were exothermic reaction and gaseous products after thermal decomposition. Combustion test and flash point examination of the ILs proved that the gaseous product could be ignited and cause continuous burning. Thus, the recommended operating temperature for [Bim][NO3] and [Mim][NO3] in the inherently safer process should not exceed 100 °C to avoid serious thermal decomposition.

Notes

Acknowledgements

The authors are indebted to the Ministry of Science and Technology (MOST) in Taiwan under the Contract Number 104-2622-E-224-009-CC2 for financial support, as well as the Department of Natural Sciences Key Fund, Bureau of Education, Anhui Province, China, for its financial support under Contract Number KJ2017A078.

References

  1. 1.
    Lei Z, Chen B, Koo YM. Introduction: ionic liquids. Chem Rev. 2017;117:6633–5.CrossRefPubMedGoogle Scholar
  2. 2.
    Rogers RD, Seddon KR. Ionic liquids–solvents of the future? Science. 2003;302:792–3.CrossRefPubMedGoogle Scholar
  3. 3.
    Maton C, Vos ND, Stevens CV. Ionic liquid thermal stabilities: decomposition mechanisms and analysis tools. Chem Soc Rev. 2013;42:5963–77.CrossRefPubMedGoogle Scholar
  4. 4.
    Chen X, Ma C, Jiao C. Synergistic effects between [Emim]PF6 and aluminum hypophosphite on flame retardant thermoplastic polyurethane. RSC Adv. 2016;6:67409–17.CrossRefGoogle Scholar
  5. 5.
    Jiao C, Wang H, Li S, Chen X. Fire hazard reduction of hollow glass microspheres in thermoplastic polyurethane composites. J Hazard Mater. 2017;332:176–84.CrossRefPubMedGoogle Scholar
  6. 6.
    Kamavaram V, Reddy RG. Thermal stabilities of di-alkylimidazolium chloride ionic liquids. Int J Therm Sci. 2008;47:773–7.CrossRefGoogle Scholar
  7. 7.
    Wang R, Jin CM, Twamley B, Shreeve JM. Syntheses and characterization of unsymmetric dicationic salts incorporating imidazolium and triazolium functionalities. Inorg Chem. 2006;45:6396–403.CrossRefPubMedGoogle Scholar
  8. 8.
    Xiao JC, Shreeve JM. Synthesis of 2, 2′-biimidazolium-based ionic liquids: use as a new reaction medium and ligand for palladium-catalyzed Suzuki cross-coupling reactions. J Org Chem. 2005;70:3072–8.CrossRefPubMedGoogle Scholar
  9. 9.
    Swatloski RP, Spear SK, Holbrey JD, Rogers RD. Dissolution of cellose with ionic liquids. J Am Chem Soc. 2002;124:4974–5.CrossRefPubMedGoogle Scholar
  10. 10.
    Tao F, Song H, Chou L. Catalytic conversion of cellulose to chemicals in ionic liquid. Carbohydr Res. 2011;346:58–63.CrossRefPubMedGoogle Scholar
  11. 11.
    Latip NAA, Ng HM, Farah N, Ramesh K, Ramesh S. Novel development towards preparation of highly efficient ionic liquid based co-polymer electrolytes and its application in dye-sensitized solar cells. Org Electron. 2017;41:33–41.CrossRefGoogle Scholar
  12. 12.
    Lin HH, Peng JD, Suryanarayanan V, Velayutham D, Ho KC. Perfluoro anion based binary and ternary ionic liquids as electrolytes for dye-sensitized solar cells. J Power Sources. 2016;311:167–74.CrossRefGoogle Scholar
  13. 13.
    Zhang Y, Gao H, Joo YH, Shreeve JNM. Ionic liquids as hypergolic fuels. Angew Chem Int Ed. 2011;50:9554–62.CrossRefGoogle Scholar
  14. 14.
    Earle MJ, Esperanca JMSS, Gilea MA, Lopes JNC, Rebelo LPN, Magee JW, Seddon KR, Widegren JA. Lithium-ion batteries: advanced materials and technologies. Nature. 2006;439:831–4.CrossRefPubMedGoogle Scholar
  15. 15.
    Liaw HJ, Huang SK, Chen HY, Liu SN. Reason for ionic liquids to be combustible. Procedia Eng. 2012;45:502–6.CrossRefGoogle Scholar
  16. 16.
    Cao Y, Mu T. Comprehensive investigation on the thermal stability of 66 ionic liquids by thermogravimetric analysis. Ind Eng Chem Res. 2014;53:8651–64.CrossRefGoogle Scholar
  17. 17.
    Katritzky AR, Singh S, Kirichenko K, Smiglak M, Holbrey JD, Reichert WM, Spear SK, Rogers RD. In search of ionic liquids incorporating azolate anions. Chem Eur J. 2006;17:4630–41.CrossRefGoogle Scholar
  18. 18.
    Wellens S, Thijs B, Binnemans K. How safe are protic ionic liquids? Explosion of pyrrolidinium nitrate. Green Chem. 2013;15:3484–5.CrossRefGoogle Scholar
  19. 19.
    Smiglak M, Reichert WM, Holbrey JD, Wilkes JS, Sun L, Thrasher JS, Kirichenko K, Singh S, Katritzky AR, Rogers RD. Combustible ionic liquids by design: is laboratory safety another ionic liquid myth? Chem Commun. 2006;24:2554–6.CrossRefGoogle Scholar
  20. 20.
    Gelfand BE, Khomik SV, Eremenko LT, Tsiganov SA. Basic features of the self-ignition of atomized liquid nitro/nitrate/nitrite compounds in a gaseous medium. Proc Combust Inst. 2000;28:879–83.CrossRefGoogle Scholar
  21. 21.
    Hainer RM. The application of kinetics to the hazardous behavior of ammonium nitrate. Symp Int Combust. 1955;5:224–30.CrossRefGoogle Scholar
  22. 22.
    Olivares RI. The thermal stability of molten nitrite/nitrates salt for solar thermal energy storage in different atmospheres. Sol Energy. 2012;86:2576–83.CrossRefGoogle Scholar
  23. 23.
    Wang C, Du Y, Che D. Study on N2O reduction with synthetic coal char and high concentration CO during oxy-fuel combustion. Proc Combust Inst. 2015;35:2323–30.CrossRefGoogle Scholar
  24. 24.
    Classification, Labelling and Packaging (CLP) of Substances and Mixtures, European Chemicals Agency, Helsinki. ISBN: 978–92–9247–413–3.Google Scholar
  25. 25.
    Grigiante M, Brighenti M, Antolini D. Analysis of the impact of TG data sets on activation energy (E a). J Therm Anal Calorim. 2017:129:553–65.CrossRefGoogle Scholar
  26. 26.
    Quraishi KS, Bustam MA, Krishnan S, Khan MI, Wilfred CD, Lévêque JM. Thermokinetics of alkyl methylpyrrolidinium [NTf2] ionic liquids. J Therm Anal Calorim. 2017;129:261–70.CrossRefGoogle Scholar
  27. 27.
    Sinapour H, Damiri S, Pouretedal HR. The study of RDX impurity and wax effects on the thermal decomposition kinetics of HMX explosive using DSC/TG and accelerated aging methods. J Therm Anal Calorim. 2017;129:271–9.CrossRefGoogle Scholar
  28. 28.
    ASTM D 93, Standard test methods for flash-point by Pensky-Martens closed cup tester, American Society for Testing and Materials, West Conshohocken, PA, USA (2016).Google Scholar
  29. 29.
    Geyer A, Alicke B, Konrad S, Schmitz T, Stutz J, Platt U. Chemistry and oxidation capacity of the nitrate radical in the continental boundary layer near Berlin. J Geophys Res-Atmos. 2001;106:8013–25.CrossRefGoogle Scholar
  30. 30.
    Smiglak M, Hines CC, Reichert WM, Vincek AS, Katritzky AR, Thrasher JS, Sun L, McCrary PD, Beasley PA, Kelley SP, Rogers RD. Synthesis, limitations, and thermal properties of energetically-substituted, protonated imidazolium picrate and nitrate salts and further comparison with their methylated analogs. New J Chem. 2012;36:702–22.CrossRefGoogle Scholar

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2018

Authors and Affiliations

  • Wei-Cheng Lin
    • 1
  • Wen-Lung Yu
    • 2
  • Shang-Hao Liu
    • 3
  • Shih-Yu Huang
    • 4
  • Hung-Yi Hou
    • 5
  • Chi-Min Shu
    • 4
    • 6
  1. 1.Graduate School of Engineering Science and TechnologyNational Yunlin University of Science and Technology (YunTech)YunlinTaiwan, ROC
  2. 2.Department of Industrial Education and TechnologyNational Changhua University of EducationChanghuaTaiwan, ROC
  3. 3.Department of Ammunition Engineering and Explosion TechnologyAnhui University of Science and TechnologyHuainanChina
  4. 4.Department of Safety, Health, and Environmental EngineeringYunTechYunlinTaiwan, ROC
  5. 5.Department of Occupational Safety and Health, Jen-Teh Junior College of MedicineNursing and ManagementMiaoliTaiwan, ROC
  6. 6.Center for Process Safety and Industrial Disaster PreventionYunTechYunlinTaiwan, ROC

Personalised recommendations