Advertisement

Journal of Thermal Analysis and Calorimetry

, Volume 133, Issue 1, pp 659–671 | Cite as

Thermal hazard evolution on guanidine nitrate

  • Yi Liu
  • Xuezhi Wang
  • Chi-Min Shu
  • Yu Wang
  • Dongfeng Zhao
  • Wanghua Chen
  • Jun Zhang
  • Jia Yin
Article
  • 152 Downloads

Abstract

Due to thermally reactive hazards, guanidine nitrate (GN) has caused numerous serious accidents involving manufacturing, storing, and transporting processes. Differential scanning calorimetry and thermogravimetry (TG) were used to study the thermal reactive hazards of GN and the influence of adding water, nitric acid, ammonium nitrate (AN), and urea on GN thermal decomposition in air. The results indicated that the exothermal onset temperature and peak temperature rose with the heating rates. According to the Kissinger method, Ozawa method, as well as Friedman method, the apparent activation energy of GN in the peak temperature is 121.1, 123.0, and 126.0 kJ mol−1, respectively. The thermal decomposition of GN consisted of four stages. The mass loss rate of GN reached the maximum at 300 °C, and this temperature was in line with the exothermic peak temperature. The TG curve of GN in nitrogen atmosphere has a certain hysteresis compared to the air atmosphere. The thermal hazard of GN was inversely proportional to the size of the particles. Water had little effect on the decomposition of GN. Furthermore, nitric acid can promote the decomposition of GN more vigorously and increase the hazards of thermal runaway of GN. With the growth of the amount of AN and urea, the decomposition reaction of GN was more likely to initiate and more difficult to govern. These results could be used as a reference guide to the actual manufacturing, storing, and transporting processes for GN.

Keywords

Guanidine nitrate DSC Thermal reactive hazards Thermal decomposition Thermal runaway 

List of symbols

A

Pre-exponential factor of Arrhenius equation, min−1

F(α)

Kinetic model, dimensionless

n

Reaction order, dimensionless

Ea

Apparent activation energy, kJ mol−1

M

Sample mass, mg

R

Gas constant, 8.314 J mol−1 K−1

T0

Exothermic onset temperature, °C

Tv−max

Temperature at the fastest change in the heat flow, °C

Tp

Peak temperature, °C

Ti

Corresponding decomposition temperature at different heating rates under the same reaction process, °C

r2

Correlation coefficient, dimensionless

α

Conversion degree, dimensionless

β

Heating rate, °C min−1

ΔHd

Heat of decomposition, kJ g−1

Introduction

Guanidine nitrate (GN) is widely applied to dyestuffs and pesticides, and also frequently as an important intermediate. It is widely used in disinfectants, explosives, paints, and other fields [1]. GN molecule constructor contains nitro and amino. GN is unstable because the reactive groups have less bond energy and possess strong activity, which is prone to decompose by heat, impact, and friction, and release a large amount of heat [2]. The released heat will promote thermal decomposition and bring hazards to the production, transportation, storage, and usage processes of GN [3]. Therefore, the thermal reactive hazards of GN are intractable to govern and lead to a thermal runaway accident when measures fail. A GN production plant exploded in Hebei Province of China on February 28, 2012. The explosion killed 25 people and injured 46, with four others missing. According to the statistics of the US Chemical Safety, and Hazard Investigation Board, 35% of the incidents related to thermal runaway events in 167 incidents occurred in the USA between 1980 and 2001 [4].

According to the relevant literature survey, reactive substances involved in the above thermal runaway accidents have been studied, but GN has rarely been studied. Damse studied the mechanism of thermal decomposition of GN by TG, differential thermal analysis, and DSC [1]. Oxley et al. [5] studied the apparent activation energy, pre-exponential factor, and decomposition mechanism of GN using DSC. Nakashima et al. [6] studied the thermal decomposition property of guanidine nitrate and basic copper nitrate mixture by the thermogravimetry–differential scanning calorimetry–Fourier transform infrared spectroscopy (TG/DSC/FTIR). However, they did not cover the effect of other impurities on thermal decomposition of GN in their research. GN has caused numerous serious accidents involving manufacturing, storing, and transporting processes. Accordingly, it is urgent to delve into the thermal hazards of GN. Our aim was to study GN in order to determine its thermal risk, which can provide a guide for the safety of GN in the process of production, transportation, storage, and usage.

Experimental

Materials and instruments

GN is a type of particle with different sizes, at which purity exceeds 99 mass%, which was produced by Shandong Western Asia Chemical Industry Co., Ltd. (Shandong Province, China) in Western Shandong Province. It was dried at 45 °C for 1 h prior to the experiment. Table 1 lists the basic information of GN.
Table 1

Basic information of GN

Material

Chemical formula

CAS

Molecular weight

Specific gravity

Melting point/°C

Purity

Guanidine nitrate

CH6 N4O3

506–93–4

122.08

1.44

217.0

> 99%

First, we measured the thermal decomposition curves of GN at different heating rates (β) by DSC. The mass loss curves of GN in the process of thermal decomposition in air and nitrogen atmospheres were investigated by TG. Then, we divided the GN into different particle sizes (161–212, 213–270, and 271–550 μm) by the griddle and tested the influence of particle size on the thermal decomposition by DSC. Finally, we studied the effects of impurity (water, nitric acid, AN, and urea) on the thermal hazards of GN by DSC. The content of HNO3 ranged from 65 to 68 mass%, and the content of AN and urea was both more than 99 mass%.

The DSC instrument applied in this study was developed by the Beijing Scientific Instrument. Its model was HSC-1/2 which had DSC function. Instrument of thermogravimetry (Q500, TA Corp., USA) was also used. The process of the reaction was in a dynamic atmosphere of nitrogen, and gas flow rate was 50.0 mL min−1. We used standard objects to calibrate the instrument before the experiments and repeated experiments when the results showed abnormalities.

Results and discussion

Different heating rates

Our aim was to study the influence of different heating rates for thermal decomposition of GN. The influence can be presented by changes in exothermic peak, endothermic peak, T0, Tv−max, Tp, and ΔHd. As shown in Fig. 1, the DSC curves included an endothermic peak and an exothermic peak. The curves kept the same tendency at different β, which indicated the reaction mechanism of GN was the same in the decomposition process. As shown in Fig. 2, with the increase of β, the exothermic peak moved to the upper limit. The thermal efficiency increased due to the growth of β, which rendered rising to a large temperature difference. Therefore, the exothermic peak moved toward to the high-temperature region.
Fig. 1

Heat flow temperature curves of GN at different β values

Fig. 2

Disposed exothermic peak of the GN DSC curves by AKTS

In addition to the qualitative description of the heat flow curve, AKTS can also obtain quantitative information of the curve. Through the analysis, the exothermic peak parameters of GN at various β are listed in Table 2.
Table 2

Exothermic peak parameters of GN at various β values

Mass/mg

β/°C min−1

To/°C

T v−max

Tp/°C

− ΔHd/J mg−1

5.0

2.0

223.1

251.7

276.6

610.0

5.1

4.0

242.8

272.2

289.5

583.0

5.1

8.0

253.1

281.5

303.4

501.0

5.2

12.0

256.8

289.6

312.9

428.0

With the increase of β, T0, Tv−max, and Tp increased gradually. It is difficult to dictate the decomposition reaction. It can provide a guide for the thermal safety of GN in the process of production, transportation, storage, and usage. ΔHd can be used to measure the risk of a material’s thermal decomposition. ΔHd in Table 2 was the heat released in the exothermic peak. Here, − ΔHd decreased with the increase of β because GN has several exothermic peaks. The medium heat because of GN decomposition was 522.5 J g−1, which exceeded 200.0 J g−1, indicating GN probably produced thermal runaway reactions [7, 8]. When the β was 8.0 and 12.0 °C min−1, the decomposition of GN presented too fast and could not decompose completely at once due to the rapid heating. A second exothermic peak appeared in the heat flow temperature curve.

Kinetic analysis

  1. (1)

    Kissinger method [9]

     
Equation (1) expresses the reaction kinetics equation of the Kissinger method. Ln(β/T p 2 ) versus 1/Tp has a linear function. If the reaction obeys Arrhenius kinetics, Ea, A can be obtained from the slope and intercept of the function. Figure 3 shows the linear function of ln(β/T p 2 ) versus 1/Tp which was expressed by Eq. (1). According to Fig. 3, Ea, A, and lnA were 121.1 kJ mol−1, 8.7597 × 104, and 11.38 s−1.
$$\ln \left( {\frac{\beta }{{T_{\text{p}}^{2} }}} \right) = \ln \left( {\frac{RA}{{E_{\text{a}} }}} \right) - \frac{{E_{\text{a}} }}{R}\frac{1}{{T_{\text{p}} }}$$
(1)
Fig. 3

Apparent activation energy of GN fitted by Kissinger method

  1. (2)

    Ozawa method [10]

     
The principle of Ozawa method is shown in Eq. (2).
$${ \lg }\beta_{\text{i}} = \ln \frac{{AE_{\text{a}} }}{Rf\left( \alpha \right)} - 2.315 - 0.4567\frac{{E_{\text{a}} }}{{RT_{\text{i}} }}$$
(2)
Figure 4 shows the GN activation energy fitted by Ozawa method.
Fig. 4

Apparent activation energy of GN fitted by Ozawa method

  1. (3)

    Friedman method [11]

     
The Friedman method can use several DSC curves data at the same conversion degree (α) to calculate a more reliable activation energy value without assuming a dynamic mode function. Ea is presented as a curve that varies with the severity of the reaction. We selected four DSC curves under 2.0, 4.0, 8.0, and 12.0 °C min−1 and applied AKTS to analyze Ea. Figures 57 show the GN decomposition reaction process, the reaction rate of GN, and Ea of GN decomposition process.
Fig. 5

Decomposition reaction processes of GN at different β values by DSC

Fig. 6

Reaction rate and profile of GN at different β values by DSC

Fig. 7

Curves of Ea and the correlation coefficient (r2) by Friedman method. a Mass loss curve of GN in N2, b mass loss curve of GN in air

The Ea calculated by Kissinger method corresponded to the exothermic peak temperature. It was a constant and independent of the reaction process. However, the apparent activation energy calculated by Ozawa method and Friedman method was related to the reaction process. Figures 4 and 7 show that the apparent activation energy of GN curves was consistent; the apparent activation energy of the peak temperature was 123 and 126 kJ mol−1, respectively. As a result, the apparent activation energies calculated by the three methods are similar. The results were smaller than that of Oxley’s study [6].

Thermogravimetric analysis

To further study the thermal decomposition of the GN, we investigated the mass loss curve of GN in the process of thermal decomposition in air and nitrogen atmosphere by TG. Table 3 lists the TG experimental schemes of GN. Figure 8 shows the experimental results.
Table 3

TG test results of GN under air and N2 atmospheres

Sample

Atmosphere

Flow/mL min−1

Mass/mg (start–end)

β/°C min−1

Temperature/°C (start–end)

GN

Air

50.0

5.4–0

4.0

30.0–800.0

N2

50.0

5.3–0

4.0

30.0–800.0

Fig. 8

Mass loss curves and mass loss derivative of GN by TG tests with 4.0 °C min−1. a Mass loss curve of GN in N2, b Mass loss curve of GN in air

The mass loss rate reached the maximum value at 300 °C, which was in line with the exothermic peak temperature. In the third stage, the mass loss rate of GN decreased significantly. At this stage, the undecomposed GN or the product of the second stage decomposed. As shown in Fig. 8a, b, GN can be completely decomposed at 595 °C in air, but it needs 610 °C in nitrogen. Since oxygen can promote the reaction of GN when GN in the air, the TG curve of GN in nitrogen atmosphere had a certain hysteresis compared to the air atmosphere.

Different particle sizes

As shown in Figs. 9 and 10, the height of the GN exothermic peak increased with the decrease in the particle sizes. It also indicated that the thermal decomposition reaction of GN became more intense with the decline of the particle size. Because specific surface area of particles increased with the reduced particle size, the heat transfer efficiency rose. Therefore, the heat cannot release to the environment in a timely manner and as a positive feedback to promote the increase in decomposition intense. Table 4 shows that the thermal decomposition reaction of GN was more readily initiated with the decrease in GN particle sizes. Furthermore, the released heat increased with the decrease in the GN particle size. Therefore, the thermal hazard of guanidine nitrate was inversely proportional to the sizes of the particles.
Fig. 9

Heat flow versus temperature curve of GN with different particle sizes under 6.0 °C min−1

Fig. 10

Heat flow versus temperature local curve of GN with different particle sizes under 6.0 °C min−1

Table 4

Exothermic peak parameters of GN with different particle sizes with 10.0 °C min−1

Mass/mg

D/μm

To/°C

Tp/°C

− ΔHd/J mg−1

5.2

271.0–550.0

265.9

310.7

449.0

5.1

213.0–270.0

249.2

308.6

465.0

5.4

161.0–212.0

245.4

307.5

487.0

Effect of impurity

According to the production progress of GN, the influence of water, nitric acid, AN, and urea to the decomposition reaction of GN was studied by DSC. Figure 11 shows that the exothermic peak of the GN with different amounts of H2O remained the same trend. In Table 5, the parameters of the exothermic peak GN changed little. Because the DSC instrument furnace used in the study was partly closed, H2O evaporated and was lost to the outside, making water not affect the decomposition of GN. If the instrument was closed, with the increase in temperature, the H2O vapor became adiabatic pressure steam. It might influence the thermal decomposition of GN.
Fig. 11

Heat flow versus temperature curve of the mixture of GN and H2O by DSC with 6.0 °C min−1

Table 5

Exothermic peak parameters of GN with H2O with 6.0 °C min−1

Mass/mg

To/°C

Tp/°C

− ΔHd/J mg−1

GN 8.5 mg

249.9

301.2

270.0

GN 8.5 mg + H2O 2.5 mg

242.9

300.0

280.0

GN 8.5 mg + H2O 4.0 mg

248.2

300.3

330.0

GN 8.5 mg + H2O 7.5 mg

249.2

304.2

200.0

Figure 12 shows the heat flow versus temperature curve of nitric acid. Figure 13 demonstrates the exothermic peak gradually moved the lower temperature area with the increase in the addition of nitric acid. It indicates nitric acid can promote the decomposition of guanidine nitrate more vigorously. Table 6 shows the decrease of T0 with the increase in adding nitric acid, which demonstrated that nitric acid can contribute to the thermal decomposition of GN and increase the hazard of thermal runaway of GN.
Fig. 12

Heat flow versus temperature curve of the solution of nitric acid by DSC with 6.0 °C min−1

Fig. 13

Heat flow versus temperature of the mixture of GN and nitric acid by DSC with 6.0 °C min−1

Table 6

Exothermic peak parameters of GN with nitric acid with 6.0 °C min−1

Mass/mg

To/°C

Tp/°C

− ΔHd/J mg−1

GN 8.0 mg

249.9

301.2

260.0

GN 8.0 mg + HNO3 4.9 mg

235.5

304.7

390.0

GN 8.0 mg + HNO3 7.2 mg

221.6

303.8

370.0

GN 8.0 mg + HNO3 11.5 mg

212.2

292.4

410.0

Figure 14 shows that AN absorbed and released heat quickly. The endothermic and the exothermic processes were divided into four and one stages, respectively. The exothermic peak temperature was 285.0 °C. Because the boundary between the endothermic and exothermic phases was very short, it could readily lead to thermal decomposition. Figure 15 indicates that, with the increase in the amount of AN, the exothermic peaks gradually appeared two shoulder peaks, and the height of the exothermic peak decreased, which illustrates that the reaction intensity decreased. When two consecutive peaks appeared in Fig. 15, we used the average of the TP of the two peaks as the results presented in Table 7. Table 7 indicates T0 decreased with the increase in the amount of AN, which indicated that the decomposition reaction of GN was more likely to initiate.
Fig. 14

Heat flow temperature curve of the AN with 6.0 °C min−1

Fig. 15

Heat flow versus temperature curve of the mixture of GN and AN with 6.0 °C min−1

Table 7

Exothermic peak parameters of GN with AN with 6.0 °C min−1

Mass/mg

To/°C

Tp/°C

− ΔHd/J mg−1

GN 8.0 mg

249.9

301.2

260.0

NH4NO3 8.5 mg

GN 8.0 mg + NH4NO3 5.0 mg

245.2

299.6

220.0

GN 8.0 mg + NH4NO3 7.8 mg

236.5

294.4

140.0

GN 8.0 mg + NH4NO3 9.8 mg

225.9

286.2

110.0

Figure 16 shows the range from 410.0 to 430.0 °C was the exothermic stage of urea. The heat flow curve presented that the urea absorbed heat rapidly and released slowly. As shown in Fig. 17, compared with the pure GN, the endothermic and the exothermic processes were divided into several sections, and the peak changed moderately and the reaction rate decreased.
Fig. 16

Heat flow versus temperature curve of urea with 6.0 °C min−1

Fig. 17

Heat flow versus temperature curve of the mixture of GN and urea with 6.0 °C min−1

In Table 8, T0 decreased with the increase in the amount of urea, so the decomposition reaction of GN was more ready to initiate. Therefore, with the increase in the amount of urea in GN, its decomposition reaction intensity weakened, but the reaction was more likely to be triggered.
Table 8

Exothermic peak parameters of GN with urea with 6.0 °C min−1

Mass/mg

To/°C

Tp/°C

− ΔHd/J mg−1

GN 8.0 mg

249.9

301.2

260.0

Urea 8.7 mg

GN 8.0 mg + urea 4.5 mg

245.7

290.3

90.0

GN 8.0 mg + urea 5.7 mg

241.9

292.4

100.0

GN 8.0 mg + urea 8.4 mg

232.6

293.6

50.0

Conclusions

The thermal decomposition characteristics were studied of GN and the influence of adding water, nitric acid, AN, and urea on GN thermal decomposition in the air by DSC and TG. The main conclusions are as follows:
  1. (1)

    With the increase of β, the exothermic peak moved to the high-temperature area; then, T0, Tv−max, and Tp increased prominently. The moderate heat released of GN decomposition was 522.5 J g−1, which indicated that GN probably produced thermal runaway reactions. According to the Kissinger, Ozawa, and Friedman methods, the Ea of GN in the peak temperature was 121.1, 123.0, and 126.0 kJ mol−1, respectively.

     
  2. (2)

    The thermal decomposition of GN consisted of four stages. The second stage was the fastest stage in terms of mass reduction. The mass loss rate reached the maximum value at 300.0 °C, which was in line with the exothermic peak temperature. The TG curve of GN in nitrogen atmosphere shows a certain hysteresis compared to the air atmosphere.

     
  3. (3)

    With the reduction in particle size, the thermal decomposition reaction of GN was more readily initiated. Furthermore, the released heat increased with the decrease in the GN particle size. Therefore, the thermal hazard of guanidine nitrate was inversely proportional to the particle size.

     
  4. (4)

    (a) Water had little effect on the decomposition of GN. (b) Nitric acid can promote the decomposition of GN more strongly, and increase the hazards of thermal runaway of GN. (c) AN and urea had the same influence on the decomposition reaction of GN. With the increase in the amount of AN and urea, T0 decreased, which indicated that the decomposition reaction of GN was more easy to initiate and more difficult to control.

     

Notes

Acknowledgements

The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (Grant No. 5100-6122) and the National Key Research and Development Program of China (Grant No. 2016-YFC080-1500).

References

  1. 1.
    Damse RS. Studies on the decomposition chemistry of triaminoguanidine azide and guanidine nitrate. J Hazard Mater. 2009;172(2):1383–7.CrossRefPubMedGoogle Scholar
  2. 2.
    Xu W, Li J, Liu F, Jiang Y, Li Z, Li L. Study on the thermal decomposition kinetics and flammability performance of a flame-retardant leather. J Therm Anal Calorim. 2017;128(2):1107–16.CrossRefGoogle Scholar
  3. 3.
    Charmas B, Skubiszewska-Zięba J. Application of differential scanning calorimetry to study porous structure of hydrothermally modified silica. J Therm Anal Calorim. 2017;129(1):23–32.CrossRefGoogle Scholar
  4. 4.
    Gu Y. Leakage risk assessment method for the sulfur recovery unit and its application. China University of Geosciences (Beijing, PR China) [Doctoral dissertation]; 2015.Google Scholar
  5. 5.
    Oxley J, Smith J, Naik S, Moran J. Decompositions of urea and guanidine nitrates. J Energ Mater. 2008;27(1):17–39.CrossRefGoogle Scholar
  6. 6.
    Nakashima M, Itaura T, Matsunaga H, Higashi E, Takagi S, Katoh K. A fundamental study on the thermal decomposition and combustion behaviors of guanidine nitrate and basic copper nitrate mixture. J Thermal Anal Calorim. 2018;131(1):95–100.CrossRefGoogle Scholar
  7. 7.
    Zhang X, Yang S. The effect of high ground temperature environment on risk of coal self-heating. China Min Mag. 2016;09:159–65.Google Scholar
  8. 8.
    Stoessel F. Thermal safety of chemical processes. Weinheim: Wiley; 2008.CrossRefGoogle Scholar
  9. 9.
    Yin RL, Chen LP, Chen WH, Wu K, Zhang CX, Li H, Cheng Y. Kinetic compensation effect under two different calorimetric modes for thermal decomposition. Acta Phys Chim Sin. 2016;32(02):391–8.Google Scholar
  10. 10.
    Lee MH, Chen JR, Shiue GY, Lin YF, Shu CM. Simulation approach to benzoyl peroxide decomposition kinetics by thermal calorimetric technique. J Taiwan Inst Chem E. 2014;45(1):115–20.CrossRefGoogle Scholar
  11. 11.
    Cui HW, Jiu JT, Sugahara T, Nagao S, Suganuma K, Uchida H. Using the Friedman method to study the thermal degradation kinetics of photonically cured electrically conductive adhesives. J Thermal Anal Calorim. 2015;119(1):425–33.CrossRefGoogle Scholar

Copyright information

© Akadémiai Kiadó, Budapest, Hungary 2018

Authors and Affiliations

  1. 1.State Key Laboratory of Heavy Oil ProcessingChina University of Petroleum (East China)QingdaoChina
  2. 2.College of Chemical EngineeringChina University of Petroleum (East China)QingdaoChina
  3. 3.Center for Safety, Environmental, and Energy Conservation Technology of China University of Petroleum (East China)QingdaoChina
  4. 4.Department of Safety, Health, and Environmental EngineeringNational Yunlin University of Science and TechnologyYunlinTaiwan, ROC
  5. 5.Department of Safety Engineering, School of Chemical EngineeringNanjing University of Science and TechnologyNanjingChina

Personalised recommendations