- Email: [email protected]
Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN)
Accepted Manuscript Title: Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN) Author: G. Santhosh P.B. Soumyamol M. Sreejith S. Reshmi PII: DOI: Reference:
S0040-6031(16)30041-7 http://dx.doi.org/doi:10.1016/j.tca.2016.03.019 TCA 77466
To appear in:
Thermochimica Acta
Received date: Revised date: Accepted date:
29-12-2015 9-3-2016 11-3-2016
Please cite this article as: G.Santhosh, P.B.Soumyamol, M.Sreejith, S.Reshmi, Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN), Thermochimica Acta http://dx.doi.org/10.1016/j.tca.2016.03.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN) G. Santhosha.*, P.B. Soumyamolb, M. Sreejitha, S. Reshmia a
Propellant Engineering Division, Polymers and Special Chemicals Group
b
Analytical and Spectroscopy Division, Analytical Spectroscopy and Ceramics Group
Vikram Sarabhai Space Centre, Trivandrum 695022, India
* Corresponding author E-mail: [email protected]
Highlights • The thermal decomposition of guanylurea dinitramide (GUDN) is studied by isoconversional method • The thermal decomposition products from GUDN are identified using TG-MS • Kinetic parameters are obtained from both ‘Isoconversional’ and classical ‘Kissinger’ methods • Isothermal kinetics of GUDN was predicted from isoconversional method and a decomposition mechanism for GUDN is proposed
Abstract The thermal decomposition of a high-energy oxidizer viz., guanylurea dinitramide (GUDN) has been studied by non-isothermal thermogravimetry (TG) and thermogravimetry combined with mass spectrometry (TG-MS). Isoconversional method of Vyazovkin (VZ) was used to investigate the dependence of activation energy (Ea) on conversion (α). A strong
dependence of Ea on α, where the Ea increases steadily for up to an α of ~0.6 followed by a marginal increase and reaches ~435.0 kJ mol-1 at the end of the reaction. Ea has also been determined for GUDN using Kissinger’s method. A comparison of the results from Kissinger and isoconversional method shows that the activation energies from these methods are comparable. Using the model free isoconversional method, the isothermal conversion as a function of time at two different temperatures was also computed. The evolved gases during the decomposition of GUDN were analyzed by TG-MS which revealed the formation of ions corresponding to N2, O2, NH3, H2O, N2, NO, methane diimine, isocyanic acid, N2O, NO2 and urea.
Keywords: Isoconversional methods; kinetics; guanylurea dinitramide; TGA; activation energy; Kissinger method.
1 Introduction Energetic materials find extensive use in composite propellants and explosives. Dinitramide salts occupy the class of new-high energy oxidizers and many of these were synthesized and characterized for military and space applications [1]. Among the dinitramide salts reported, the important ones that have attracted extensive interest are ammonium dinitramide (ADN), potassium dinitramide (KDN) and guanylurea dinitramide (GUDN). ADN is emerging as a promising candidate for replacing the conventional ammonium perchlorate (AP) used in composite solid propellants [2], while GUDN or FOX-12 is a nitrogen rich dinitramide salt (structure shown in Fig.1) which is intended for use as low-sensitivity oxidizer for propellants and explosives [3]. Its excellent thermal stability along with very low mechanical sensitivity, non hygroscopicity and fairly higher bulk density finds potential use as insensitive munitions and as gas generators [3-8]. Its synthesis [9-11], thermal decomposition kinetics [9, 12-14], dissolution properties [15, 16] toxicological aspects and photo degradation studies [17-18] have been reported. Limited studies have been carried on the thermal decomposition of GUDN and kinetics were derived using model fitting methods [9, 12-14]. Ostmark et al., have studied the thermal stability of GUDN by DSC and the sensitivity to thermal ignition was measured using Wood’s metal bath technique and the kinetic parameters were derived using ASTM E 698-79 [9].
Santhosh et al., have studied the decomposition of GUDN by DSC and the activation energy was calculated by Kissinger method [12]. Zhao et al., have studied the thermal decomposition of GUDN by DSC and the kinetic parameters were derived using both Kissinger and Ozawa methods [13]. The kinetic behaviour and decomposition of GUDN was studied by Gao et al., using a DSC and microcalorimeter and the decomposition was described using an autocatalytic model [14]. Since all the above model fitting methods derive a single activation energy (Ea) for the process, reliable kinetic analysis could not be made as thermal decomposition involve multistep kinetics.
Most of the work on the thermal decomposition studies of GUDN was using classical Kissinger, Ozawa and ASTM methods. Heterogeneous solid state reaction can be better described using an isoconversional method. However, no work has been reported on the isoconversional kinetic studies of GUDN. In this work, the non-isothermal decomposition kinetics of GUDN has been studied by thermogravimetric analysis (TGA). Integral isoconversional method of Vyazovkin was used to investigate the dependence of activation energy (Ea) on conversion (α). Ea has also been determined for GUDN using Kissinger’s method. The conversion as a function of time at two different temperatures was also predicted by a model free isoconversional approach. The decomposition products were identified by TG-MS. A mechanistic aspect and a comparison of results from these methods are also given in the paper.
2 Experimental 2.1 Materials and Methods GUDN was synthesized by a reported procedure. The detailed synthesis of GUDN can be found elsewhere [9-11]. TG experiments were carried out using a TA instruments SDT 2960 TG-DTA. Nonisothermal runs were performed at variable heating rates of 2, 3, 4 and 5oC min-1 in the temperature range of 30 to 300oC under nitrogen flow (100mL/min). Samples were placed in open platinum pans. A sample mass of 1 to 2 mg were used for the nonisothermal experiments. The reproducibility of the TG signals was determined and there is practically no shift in the peak temperature when the heating rate and the mass of GUDN are kept the same. The mass effect was also studied,
however a sample mass >3mg produces a violent reaction generating a noise in the TG signal. Hence in our experiments the sample mass was kept below 2mg. GUDN showed a very sharp decomposition (narrow temperature range) without any melting. This has resulted in a sharp signal in a narrow temperature range as registered in the TGA. Several attempts have also been made to study the decomposition of this material at higher heating rates, however this has resulted in the sample being spurted out of the crucible at all times even though the sample mass is kept below 2 mg. TG-MS analysis was done at a heating rate of 5oC/min in helium atmosphere using PerkinElmer Pyris 1 TGA coupled with Clarus SQ 8T quadruple mass spectrometer through a heated fused silica transfer line. The transfer line temperature was maintained at 210 °C, and the evolved species were ionized in electron impact ionization mode with electron energy of 70 eV. Mass spectra of the evolved species were recorded for a mass range 10-500 amu. There is
practically no
time lag (time
lag is less than
20sec)
for
most of
the evolved species to reach the MS. But higher time lags are possible for species that may condense in the transfer line kept at 210°C. In other words species with boiling points higher than the transfer line temperature can have a time lag, which is not applicable in this case. Specific standards recommended by the supplier have been employed for the heat and temperature calibration of the instrument. The data analysis and kinetic calculations from the TG measurements were done by using Microsoft ® EXCEL, Microcal ORIGIN® or OCTAVE® software [19].
3 Results and Discussion A typical TG-DTG curve of GUDN at a heating rate of 2oC/min is shown in Figure 2. Similar curves were obtained for all other heating rates. It is seen that GUDN undergoes decomposition in the temperature range of 180 – 225oC leaving a residue of ~30% at 250oC.
The characteristic decomposition temperatures of GUDN viz., initial temperature (To), peak temperature (Tp) (from DTG), and final temperature (Tf) at various heating rates are given in
Table 1. The final temperature gives a clear indication of the completion of the decomposition reaction, and also the temperature difference between the initial and final values gives the temperature range of decomposition and table 1 provides a summary of these events. It is seen from Table 1, that as the heating rate is increased, all the parameters got shifted to higher temperature as expected.
3.1 Kinetic Calculations The basic equation for constant heating rate nonisothermal kinetic analysis is shown in Eq. (1).
dα A E = exp − a f (α ) dT β RT
(1)
where T is the temperature, β is the linear heating rate, A is the pre-exponential factor, R is the gas constant, f(α) is the differential function of conversion, and α is the degree of conversion which is a normalized measure of reaction progress ranging from 0 to 1 as a function of time or temperature. The kinetic analyses have been carried out using the Kissinger method as well as an isoconversional method. The VZ method was employed on nonisothermal decomposition studies of GUDN at different heating rates. This method assumes that at a constant extent of conversion, the reaction rate is dependent only on the temperature. This assumption is very useful as it allows equations to be derived that can calculate Ea without any prior knowledge of the analytical form of the conversion function f(α). As a result, the kinetic information calculated using isoconversional methods can provide more reliable insights into the kinetics and mechanism of complex reaction processes [20, 21] . 3.2
Kissinger Method The Kissinger method [22] has been used to estimate the Ea of GUDN. The Ea of GUDN
was evaluated from the Tpvalues at different heating rates β using Kissinger method given in Eq. (2).
β E ln i 2 = Const − a (2) T RT p ,i p,i
where subscript i denotes different heating rates. Unlike the isoconversional methods, the Kissinger method takes a simplified approach that yields a single Ea value for the whole process. The Ea calculated for GUDN from the linear plots of ln(β/Tp2) vs. 1000/Tp is 444.61 ±24.1 kJ/mol and the plot is shown in Figure 3. The linear plot has an excellent correlation coefficient (R2) of 0.9945. Santhosh et al. reported an Ea value of 377.0 kJ/mol by Kissinger method for GUDN from DSC measurements [12]. However this value is different from the current Ea value, which could be attributed to the method and experimental conditions employed, while the present study used TG-DTG whereas the other one employed DSC. Since the nature and particle size of GUDN has influence on the decomposition, it will not be possible to have a one-to-one comparison of the Ea values from these methods. Ea values of 277 kJ/mol by DSC [9] and 237.7 kJ/mol by TG-DTG and DSC [13] were also reported in the literature for the decomposition of GUDN. The larger Ea for GUDN could be attributed to its high thermal stability and extensive stabilization by hydrogen bonding and delocalization of electron cloud compared to other dinitramide salts such as ADN and other proton bearing cations with dinitramide anions. Extensive hydrogen bonding in various dinitramide salts involving the hydrogen atoms and the dinitramide anions are reported from X-ray crystallographic studies [23, 24]. For e.g. all the four ammonium protons in ADN forms strong hydrogen bonds to the surrounding oxygen atoms of the dinitramide anion to set up a three dimensional arrangement [23]. Similarly in the case of guanidinium dinitramide (GDN), both the nitro groups forms a network of hydrogen bonds with crystallographically different independent guanidine cations [24]. In the case of GUDN, the protons from the guanylurea cation can form extensive hydrogen bonding framework with the dinitramide anion which is shown to be an important reason for the observed high thermal stability.
Simple models such as the Kissinger method, allow studying the decomposition process from a broad perspective. It provides with very little specific information on the various thermal
events happening during the decomposition reactions such as variation of Ea with α. It also fails to establish whether the process undergoes a single or multi-step kinetics. In this context, isoconversional methods provide more details as they determine the Ea dependence with respect to α. 3.3
Isoconversional analysis of decomposition of GUDN The isoconversional methods have the advantage of providing a comprehensive
description of the decomposition process of a heterogeneous solid-state reaction [25].The weight loss data from TGA can be converted to extent of conversion (α) using equation 3.
α=
Wi −Wa Wi −W f
(3)
Where, Wa, Wi & Wf are the sample mass at a given instant, initial and final respectively. The variation of conversion (α) with temperature (T) during the thermal decomposition of GUDN at various heating rates is shown in Figure 4. It is seen that the run at 2oC min-1 covers a temperature range of 200 to 217oC, whereas the run at 5oC min-1 covers about 205 to 223oC.
3.3.1
Vyazovkin’s method
Integral isoconversional method of Vyazovkin [26] was used to determine the apparent activation energy over conversion. The general assumption used in Vyazovkin’s method is that the pathway is independent of heating rate. For a set of experiments carried out at different heating rates, the activation energy Eaα can be determined at any particular desired value of α by finding the value of E aα for which the function Ω is minimum as shown in Eq. (4).
n
n
Ω = ∑∑ i =1 j≠i
β j I ( E aα , Tα i ) β i I ( E aα , Tα j )
(4)
The indexes i and j in Eq. (4) denote different heating rates, n is the total number of heating rates, and I is the temperature integral. The temperature integral I is given by Eq. (5).
I ( Eaα , Tαi ) = p ( x) =
Tαi
− E aα dT RT (5)
∫ exp 0
The temperature integral p(x), where x = Ea/RT in Eq. (5) is solved numerically using the in-built function ‘quadv’ in OCTAVE. The activation energy for any particular α is evaluated by minimizing Ω in OCTAVE using the medium scale quasi-Newton method with a mixed quadratic and cubic line search procedure using in-built unconstrained multivariable function ‘fminunc’. The minimization was done for each small intervals of conversion using the software and the Ea values were obtained with a relative error of ~8-15 kJ/mol. However a detailed determination of Ea error was not attempted. The plot of the dependence of Ea on α is shown in Fig. 5.
It can be seen from Figure 5 that the Ea steadily increases to a value of 419.5 kJ/mol at α=0.6 and then marginally increases over the rest of the conversion and reaches a maximum of 435 kJ/mol which indicates a multistep reaction mechanism. At low conversions, the Ea value is much closer to the reported Ea values of 149 kJ/mol (wood’s bath) and 277 kJ/mol (DSC) [9]. However it should be noted that the temperature range considered for the wood’s method is large i.e. from 190 to 240oC, while for the DSC it is from 200-225oC. The initial increase of Ea up to α=0.6 could be attributed predominantly to the rapid decomposition of the dinitramide anion of GUDN, whereas the marginal increase of Ea for the rest of the reaction is attributed to the formation of the guanylurea cation and its decomposition fragments.
3.3.2 Prediction of isothermal kinetics The model free isoconversional method proposed by Vyazovkin et al., can also be used to predict the isothermal conversion as a function of time from arbitrary temperature,T0 [26]. The time at which the given conversion will be reached at an arbitary temperature, can be computed using equation 6.
=
(6)
Where Tα is the temperature of the nonisothermal process at a given conversion α, with constant heating rate β, and at a fixed temperature, T0. Equation 6 was applied on the decomposition of GUDN to predict the isothermal conversion as a function of reaction time at two different temperatures of 205 and 210oC and the results are shown in figure 6.
It can be observed from fig.6, that the predicted time(t) and conversion(α) data for the decomposition of GUDN shows that the decomposition proceeds very fast at 210oC than at 205oC. As the decomposition of GUDN proceeds in a narrow temperature range of 200-220oC, the time required for 90% conversion at isothermal temperatures of 205oC and 210oC were predicted to be ~150 and ~50mins respectively. The results show that the thermal stability of GUDN is less at temperatures above 200oC and is thermally stable below 200oC.
3.4
Identification and analysis of evolved gases by mass spectrometry Analysis of the evolved gases by TG-MS during the decomposition of GUDN revealed
the formation of ions corresponding to m/z 14, 16, 18, 28, 30, 42, 43, 44, 46 and 60amu. The complete mass spectra showing the reaction products are given in figure 7.
The ions are identified as 14 (N2), 16 (O2), 17 (NH3), 18 (H2O), N2 (28), 30 (NO), 42 (HN=C=NH, methane diimine), 43 (HNCO, isocyanic acid), 44 (N2O), 46 (NO2) and 60 (NH2CO-NH2, urea). The temperature range of the evolution of products along with the corresponding TG data is shown in figure 8.
It is seen from figure 8, that the gaseous products start to appear from temperature as early as 195oC until 220oC. Though all the products were identified in this narrow decomposition range, the products detected are N2O and NO2 which appeared in the mass spectrum initially
when the time of appearance of mass fragments was considered, and are followed by HNCO, HN=C=NH. Whereas, the presence of urea (m/z 60) was detected in trace levels.
3.5 Mechanistic aspects and comparison of results Among the many dinitramide salts, the decomposition of ADN has only been studied by TG-MS so far. The mass spectrometric study on the decomposition of ADN revealed the presence of NH3, N2O and NO in the earlier stages (<100oC), while at temperatures above 150oC, the concentration of NO, H2O and NO2 increases [27]. In the case of GUDN decomposition, most of the products appeared in the mass spectra are evolved during the narrow temperature regime between 200-220oC predominantly forming N2O, NO2, isocyanic acid and methane diimine. The mechanism of decomposition and the fractions from GUDN are represented in scheme 1.
In the isoconversional methods, the effective activation energy computed for a multi-step process varies with the extent of conversion. Revealing the dependence of Ea with α helps to understand the complexity of the process and also to shed some light to identify the kinetic scheme. Vyazovkin et al. have given a detailed account on the shapes of the dependence of Ea with α to identify the kinetic scheme of the process [28]. An increasing dependence of Ea on α is attributed to the competing reactions. These dependencies were observed for the thermal decomposition of many linear polymers [28]. In the present study, the observed dependencies (Ea increases with α) by VZ method for the thermal decomposition of GUDN could be attributed to the competing reaction mechanism. The mean Ea and Ea(max) values could be used for comparing the activation energy from other methods which a gives single activation energy. However this is only a qualitative exercise in knowing the general trend as there is no direct method of correlation. The computed mean Ea for the decomposition of GUDN is 358 kJ/mol by VZ method. The Ea(max) by VZ method ~ 435 kJ/mol which is also closer to the Ea value obtained by Kissinger method. However, a higher value of ~80 kJ/mol by the Kissinger method to the mean Ea value obtained by VZ method could be because Kissinger method relies on the relative positions of Tp rather than considering the whole decomposition profile of GUDN.
4 Conclusions Non-isothermal kinetic investigations of the decomposition of GUDN were carried out using nonlinear isoconversional method of Vyazovkin. The variation of Ea with α at various conversions was evaluated. The Ea values increases steadily in the first stage (0 ≤ α ≤ 0.6) and in the later stage (0.6 ≤ α ≤ 1) it increases only marginally. The observed Ea(max) of ~435 kJ/mol by VZ method is close to the Ea value of 441 kJ/mol obtained from Kissinger method. Isothermal kinetic prediction at two different temperatures was also made by a model free isoconversional method and the results show that GUDN is thermally stable below 200oC. The major decomposition products from GUDN were identified from TG-MS as N2O, NO2, N2, isocyanic acid and methane diimine and a decomposition mechanism was proposed.
Acknowledgements Deepthi Thomas, Analytical and Spectroscopy Division, Vikram Sarabhai Space Centre (VSSC), Trivandrum is thanked for her help in TG-MS analysis and also for the inputs on decomposition mechanism. Dr. C.P. Reghunadhan Nair, Deputy Director, PCM Entity, VSSC and Dr. Dona Mathew, Polymers and Special Chemicals Division, VSSC is thanked for critically going through the manuscript and for the comments.
5 References [1]
Ang How Ghee, G. Santhosh, Advances in Energetic Dinitramides – An Emerging Class of Inorganic Oxidizers, World Scientific, Singapore, 2008.
[2]
S. Venkatachalam, G. Santhosh, K.N. Ninan, An overview on the synthetic routes and properties of ammonium dinitramide (ADN) and other dinitramide salts, Prop. Expl.Pyro. 29(3) (2004) 178-187.
[3]
H. Östmark, A. Helte, T. Carlsson, R. Adolfsson, L. Bodin, C. Eldsäter, H. Edvinsson, J. Lundgren, H. Örnhed, N-guanylurea dinitramide (FOX-12): Properties, FOI-R--2312—SE, 2007.
[4]
J. Dahlberg, New low-sensitivity modular charge propellant based on GUDN, Int. Annu. Conf. ICT, 46/1-46/5, 2006.
[5]
P. Sjoberg, Gas-generating material for gas-actuated car safety devices, PCT Int. Patent Appl. No. WO 00/40523, 2000.
[6]
Menke Klaus, Eisele
Siegfried, Gas generators and propellants based on guanylurea
dinitramide with additional energetic components, DE102011100113A1, 2012. [7]
Michienzi, Christine Marie, Knott, Christine Dianne, Insensitive primers and munitions containing low-sensitivity, low-vulnerability compositions, US Patent No. 8425703B1, 2013.
[8]
Siegfried Eisele, Klaus Menke, Gas generator propellants based on GUDN, 41st Int. Annu. Conf. ICT, 43/1 to 43/10, 2010.
[9]
H.
Ostmark, U. Bemm, H. Bergman, A. Langlet, N-guanylurea-dinitramide: a new
energetic material with low sensitivity for propellants and explosives applications, Thermochim. Acta 384 (2002) 253-259. [10] Vorde Carin, Skifs Henrik, Method of producing salts of dinitramidic acid, US Patent No. 7981393 B2, 2011. [11] D.M. Badgujar, Rashmi M. Wagh, S.J. Pawar, A.K. Sikder, Process optimization for synthesis of guanylurea dinitramide (GUDN), Prop. Expl. Pyro. 39 (2014) 658-61. [12] G. Santhosh, Robbin Poh Cheng Tien, Ang How Ghee, Thermal decomposition kinetics of ammonium dinitramide-guanylurea dinitramide mixture analyzed by isoconversional methods, Thermochim. Acta 480(2008) 43-48. [13] F.Q. Zhao, P. Chen, H.A. Yuan, S.L. Gao, R.Z. Hu, Q.Z. Shi, Thermochemical properties and non-isothermal decomposition reaction kinetics of N-guanylurea dinitramide (GUDN),Chin. J. Chem. 22 (2004) 136-141. [14] H.X Gao, H. Zhang, F.Q. Zhao, R.Z. Hu, H.X. Ma, K.Z. Xu, J.H.Yi, S.Y. Xu, Y. Gao, Kinetic behaviour of the exothermic decomposition reaction of N-guanylurea dinitramide, Acta Phys. Chim. Sin.24(3) (2008) 453–8. [15] Jiao-qiang Zhang, Wei-wei Zhao, Tie-zheng Ji,Hong-xu Gao, Rong-zu Hu, The dissolution properties of N-guanylurea-dinitramide (FOX-12) in dimethyl sulfoxide (DMSO), J. Therm. Anal. Calorim. 115 (2014) 641-646.
[16] Na Li , Feng Qi Zhao, Yang Luo, Hong Xu Gao, Li Bai Xiao, Rong Zu Hu, Rong Hui Ju, Dissolution properties of N-guanylurea dinitramide (GUDN) in dimethyl sulfoxide and Nmethyl pyrrolidone, J. Therm. Anal. Calorim. 115 (2014) 869-873. [17] Nancy N. Perreault, Annamaria Halasz, Sonia Thiboutot, Guy Ampleman, Jalal Hawari, Joint Photomicrobial Process for the Degradation of the Insensitive Munition Nguanylurea-dinitramide (FOX-12), Environ. Sci. Technol. 47 (2013) 5193-5198. [18] Helen Stenmark, Sune Sjökvist, Toxicological Testing of ADN, GUDN and FOX-7, 41st Int. Annu. Conf. ICT, 56/1 to 56/3, 2010. [19] John W. Eaton, David Bateman, Soren Hauberg, GNU Octave version 3.6.4, A high level interactive language for numerical computations, http://www.octave.org. [20] S. Vyazovkin, A.K. Burnham, Hose M. Criado, Luis A. Perez-Maqueda, Crisan Popescu, Nicolas Sbirrazzuoli, ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data, Thermochim. Acta 520 (2011) 1-19. [21] S. Vyazovkin, A unified approach to kinetic processing of nonisothermal data, Int. J. Chem. Kinetics 28 (1996) 95-101. [22] H.E. Kissinger, Reaction kinetics in differential thermal analysis, Anal. Chem. 29 (1957) 1702-1706. [23] R. Gilardi, J. Flippen-Anderson, Clifford George, Ray J. Butcher, A new class of flexible energetic salts: The crystal structures of the ammonium, lithium, potassium and cesium salts of dinitramide, J. Am. Chem. Soc 119 (1997) 9411-9416. [24] R. Gilardi, Ray J. Butcher, A new class of flexible energetic salts: Part 7 – The structures of the guanidinium and hydroxyguanidinium salts of dinitramide, J. Chem. Cryst 32 (2002) 477-484. [25] S. Vyazovkin, C.A. Wight, Model-free and model-fitting approaches to kinetic analysis of isothermal and nonisothermal data, Thermochim.Acta340-341 (1999) 53-68. [26] S. Vyazovkin, C.A. Wight, Isothermal and non-isothermal kinetics of thermally stimulated reactions of solids, Int. Rev.Phys. Chem. 17 (3) (1998) 407-433. [27] S. Vyazovkin, C.A. Wight, Ammonium Dinitramide: Kinetics and Mechanism of Thermal Decomposition, J. Phys. Chem. A.101(31) (1997) 5653-5658. [28] S. Vyazovkin, C.A. Wight, Kinetics in solids, Annu. Rev. Phys. Chem. 48 (1997) 125-149.
Figure legends
Figure 1: Chemical structure of GUDN Figure 2: TG-DTG curve of GUDN at a heating rate of 2oC/min Figure 3: Kissinger plot for GUDN Figure 4: Conversion (α) as a function of temperature (T) at various heating rates for GUDN Figure 5: Dependence of Ea with α by Vyazovkin method for the decomposition of GUDN. Figure 6:Isothermal kinetic prediction at 205 and 210oC using eqn.6 for the decomposition of GUDN Figure 7: TG-MS spectrum of decomposition products from GUDN Figure 8: Mass-to-charge ratio of major ions detected during the decomposition of GUDN from TG-MS experiment at 5oC/min Scheme 1: Decomposition mechanism and products from GUDN
Table 1: Characteristic temperature of decomposition for GUDN from TG H.R
To (oC)
Tp (oC )
Tf (oC)
2
179.8
216.5
227.1
3
182.4
218.0
230.0
4
184.8
219.5
233.2
5
187.3
220.5
235.5
(oC min-1)
Figure2 .
Figure3 .
Figure4 .
Figure5 .
Figure6 .
Figure7 .
Figure8 .
S0040-6031(16)30041-7 http://dx.doi.org/doi:10.1016/j.tca.2016.03.019 TCA 77466
To appear in:
Thermochimica Acta
Received date: Revised date: Accepted date:
29-12-2015 9-3-2016 11-3-2016
Please cite this article as: G.Santhosh, P.B.Soumyamol, M.Sreejith, S.Reshmi, Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN), Thermochimica Acta http://dx.doi.org/10.1016/j.tca.2016.03.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Isoconversional approach for the non-isothermal decomposition kinetics of guanylurea dinitramide (GUDN) G. Santhosha.*, P.B. Soumyamolb, M. Sreejitha, S. Reshmia a
Propellant Engineering Division, Polymers and Special Chemicals Group
b
Analytical and Spectroscopy Division, Analytical Spectroscopy and Ceramics Group
Vikram Sarabhai Space Centre, Trivandrum 695022, India
* Corresponding author E-mail: [email protected]
Highlights • The thermal decomposition of guanylurea dinitramide (GUDN) is studied by isoconversional method • The thermal decomposition products from GUDN are identified using TG-MS • Kinetic parameters are obtained from both ‘Isoconversional’ and classical ‘Kissinger’ methods • Isothermal kinetics of GUDN was predicted from isoconversional method and a decomposition mechanism for GUDN is proposed
Abstract The thermal decomposition of a high-energy oxidizer viz., guanylurea dinitramide (GUDN) has been studied by non-isothermal thermogravimetry (TG) and thermogravimetry combined with mass spectrometry (TG-MS). Isoconversional method of Vyazovkin (VZ) was used to investigate the dependence of activation energy (Ea) on conversion (α). A strong
dependence of Ea on α, where the Ea increases steadily for up to an α of ~0.6 followed by a marginal increase and reaches ~435.0 kJ mol-1 at the end of the reaction. Ea has also been determined for GUDN using Kissinger’s method. A comparison of the results from Kissinger and isoconversional method shows that the activation energies from these methods are comparable. Using the model free isoconversional method, the isothermal conversion as a function of time at two different temperatures was also computed. The evolved gases during the decomposition of GUDN were analyzed by TG-MS which revealed the formation of ions corresponding to N2, O2, NH3, H2O, N2, NO, methane diimine, isocyanic acid, N2O, NO2 and urea.
Keywords: Isoconversional methods; kinetics; guanylurea dinitramide; TGA; activation energy; Kissinger method.
1 Introduction Energetic materials find extensive use in composite propellants and explosives. Dinitramide salts occupy the class of new-high energy oxidizers and many of these were synthesized and characterized for military and space applications [1]. Among the dinitramide salts reported, the important ones that have attracted extensive interest are ammonium dinitramide (ADN), potassium dinitramide (KDN) and guanylurea dinitramide (GUDN). ADN is emerging as a promising candidate for replacing the conventional ammonium perchlorate (AP) used in composite solid propellants [2], while GUDN or FOX-12 is a nitrogen rich dinitramide salt (structure shown in Fig.1) which is intended for use as low-sensitivity oxidizer for propellants and explosives [3]. Its excellent thermal stability along with very low mechanical sensitivity, non hygroscopicity and fairly higher bulk density finds potential use as insensitive munitions and as gas generators [3-8]. Its synthesis [9-11], thermal decomposition kinetics [9, 12-14], dissolution properties [15, 16] toxicological aspects and photo degradation studies [17-18] have been reported. Limited studies have been carried on the thermal decomposition of GUDN and kinetics were derived using model fitting methods [9, 12-14]. Ostmark et al., have studied the thermal stability of GUDN by DSC and the sensitivity to thermal ignition was measured using Wood’s metal bath technique and the kinetic parameters were derived using ASTM E 698-79 [9].
Santhosh et al., have studied the decomposition of GUDN by DSC and the activation energy was calculated by Kissinger method [12]. Zhao et al., have studied the thermal decomposition of GUDN by DSC and the kinetic parameters were derived using both Kissinger and Ozawa methods [13]. The kinetic behaviour and decomposition of GUDN was studied by Gao et al., using a DSC and microcalorimeter and the decomposition was described using an autocatalytic model [14]. Since all the above model fitting methods derive a single activation energy (Ea) for the process, reliable kinetic analysis could not be made as thermal decomposition involve multistep kinetics.
Most of the work on the thermal decomposition studies of GUDN was using classical Kissinger, Ozawa and ASTM methods. Heterogeneous solid state reaction can be better described using an isoconversional method. However, no work has been reported on the isoconversional kinetic studies of GUDN. In this work, the non-isothermal decomposition kinetics of GUDN has been studied by thermogravimetric analysis (TGA). Integral isoconversional method of Vyazovkin was used to investigate the dependence of activation energy (Ea) on conversion (α). Ea has also been determined for GUDN using Kissinger’s method. The conversion as a function of time at two different temperatures was also predicted by a model free isoconversional approach. The decomposition products were identified by TG-MS. A mechanistic aspect and a comparison of results from these methods are also given in the paper.
2 Experimental 2.1 Materials and Methods GUDN was synthesized by a reported procedure. The detailed synthesis of GUDN can be found elsewhere [9-11]. TG experiments were carried out using a TA instruments SDT 2960 TG-DTA. Nonisothermal runs were performed at variable heating rates of 2, 3, 4 and 5oC min-1 in the temperature range of 30 to 300oC under nitrogen flow (100mL/min). Samples were placed in open platinum pans. A sample mass of 1 to 2 mg were used for the nonisothermal experiments. The reproducibility of the TG signals was determined and there is practically no shift in the peak temperature when the heating rate and the mass of GUDN are kept the same. The mass effect was also studied,
however a sample mass >3mg produces a violent reaction generating a noise in the TG signal. Hence in our experiments the sample mass was kept below 2mg. GUDN showed a very sharp decomposition (narrow temperature range) without any melting. This has resulted in a sharp signal in a narrow temperature range as registered in the TGA. Several attempts have also been made to study the decomposition of this material at higher heating rates, however this has resulted in the sample being spurted out of the crucible at all times even though the sample mass is kept below 2 mg. TG-MS analysis was done at a heating rate of 5oC/min in helium atmosphere using PerkinElmer Pyris 1 TGA coupled with Clarus SQ 8T quadruple mass spectrometer through a heated fused silica transfer line. The transfer line temperature was maintained at 210 °C, and the evolved species were ionized in electron impact ionization mode with electron energy of 70 eV. Mass spectra of the evolved species were recorded for a mass range 10-500 amu. There is
practically no
time lag (time
lag is less than
20sec)
for
most of
the evolved species to reach the MS. But higher time lags are possible for species that may condense in the transfer line kept at 210°C. In other words species with boiling points higher than the transfer line temperature can have a time lag, which is not applicable in this case. Specific standards recommended by the supplier have been employed for the heat and temperature calibration of the instrument. The data analysis and kinetic calculations from the TG measurements were done by using Microsoft ® EXCEL, Microcal ORIGIN® or OCTAVE® software [19].
3 Results and Discussion A typical TG-DTG curve of GUDN at a heating rate of 2oC/min is shown in Figure 2. Similar curves were obtained for all other heating rates. It is seen that GUDN undergoes decomposition in the temperature range of 180 – 225oC leaving a residue of ~30% at 250oC.
The characteristic decomposition temperatures of GUDN viz., initial temperature (To), peak temperature (Tp) (from DTG), and final temperature (Tf) at various heating rates are given in
Table 1. The final temperature gives a clear indication of the completion of the decomposition reaction, and also the temperature difference between the initial and final values gives the temperature range of decomposition and table 1 provides a summary of these events. It is seen from Table 1, that as the heating rate is increased, all the parameters got shifted to higher temperature as expected.
3.1 Kinetic Calculations The basic equation for constant heating rate nonisothermal kinetic analysis is shown in Eq. (1).
dα A E = exp − a f (α ) dT β RT
(1)
where T is the temperature, β is the linear heating rate, A is the pre-exponential factor, R is the gas constant, f(α) is the differential function of conversion, and α is the degree of conversion which is a normalized measure of reaction progress ranging from 0 to 1 as a function of time or temperature. The kinetic analyses have been carried out using the Kissinger method as well as an isoconversional method. The VZ method was employed on nonisothermal decomposition studies of GUDN at different heating rates. This method assumes that at a constant extent of conversion, the reaction rate is dependent only on the temperature. This assumption is very useful as it allows equations to be derived that can calculate Ea without any prior knowledge of the analytical form of the conversion function f(α). As a result, the kinetic information calculated using isoconversional methods can provide more reliable insights into the kinetics and mechanism of complex reaction processes [20, 21] . 3.2
Kissinger Method The Kissinger method [22] has been used to estimate the Ea of GUDN. The Ea of GUDN
was evaluated from the Tpvalues at different heating rates β using Kissinger method given in Eq. (2).
β E ln i 2 = Const − a (2) T RT p ,i p,i
where subscript i denotes different heating rates. Unlike the isoconversional methods, the Kissinger method takes a simplified approach that yields a single Ea value for the whole process. The Ea calculated for GUDN from the linear plots of ln(β/Tp2) vs. 1000/Tp is 444.61 ±24.1 kJ/mol and the plot is shown in Figure 3. The linear plot has an excellent correlation coefficient (R2) of 0.9945. Santhosh et al. reported an Ea value of 377.0 kJ/mol by Kissinger method for GUDN from DSC measurements [12]. However this value is different from the current Ea value, which could be attributed to the method and experimental conditions employed, while the present study used TG-DTG whereas the other one employed DSC. Since the nature and particle size of GUDN has influence on the decomposition, it will not be possible to have a one-to-one comparison of the Ea values from these methods. Ea values of 277 kJ/mol by DSC [9] and 237.7 kJ/mol by TG-DTG and DSC [13] were also reported in the literature for the decomposition of GUDN. The larger Ea for GUDN could be attributed to its high thermal stability and extensive stabilization by hydrogen bonding and delocalization of electron cloud compared to other dinitramide salts such as ADN and other proton bearing cations with dinitramide anions. Extensive hydrogen bonding in various dinitramide salts involving the hydrogen atoms and the dinitramide anions are reported from X-ray crystallographic studies [23, 24]. For e.g. all the four ammonium protons in ADN forms strong hydrogen bonds to the surrounding oxygen atoms of the dinitramide anion to set up a three dimensional arrangement [23]. Similarly in the case of guanidinium dinitramide (GDN), both the nitro groups forms a network of hydrogen bonds with crystallographically different independent guanidine cations [24]. In the case of GUDN, the protons from the guanylurea cation can form extensive hydrogen bonding framework with the dinitramide anion which is shown to be an important reason for the observed high thermal stability.
Simple models such as the Kissinger method, allow studying the decomposition process from a broad perspective. It provides with very little specific information on the various thermal
events happening during the decomposition reactions such as variation of Ea with α. It also fails to establish whether the process undergoes a single or multi-step kinetics. In this context, isoconversional methods provide more details as they determine the Ea dependence with respect to α. 3.3
Isoconversional analysis of decomposition of GUDN The isoconversional methods have the advantage of providing a comprehensive
description of the decomposition process of a heterogeneous solid-state reaction [25].The weight loss data from TGA can be converted to extent of conversion (α) using equation 3.
α=
Wi −Wa Wi −W f
(3)
Where, Wa, Wi & Wf are the sample mass at a given instant, initial and final respectively. The variation of conversion (α) with temperature (T) during the thermal decomposition of GUDN at various heating rates is shown in Figure 4. It is seen that the run at 2oC min-1 covers a temperature range of 200 to 217oC, whereas the run at 5oC min-1 covers about 205 to 223oC.
3.3.1
Vyazovkin’s method
Integral isoconversional method of Vyazovkin [26] was used to determine the apparent activation energy over conversion. The general assumption used in Vyazovkin’s method is that the pathway is independent of heating rate. For a set of experiments carried out at different heating rates, the activation energy Eaα can be determined at any particular desired value of α by finding the value of E aα for which the function Ω is minimum as shown in Eq. (4).
n
n
Ω = ∑∑ i =1 j≠i
β j I ( E aα , Tα i ) β i I ( E aα , Tα j )
(4)
The indexes i and j in Eq. (4) denote different heating rates, n is the total number of heating rates, and I is the temperature integral. The temperature integral I is given by Eq. (5).
I ( Eaα , Tαi ) = p ( x) =
Tαi
− E aα dT RT (5)
∫ exp 0
The temperature integral p(x), where x = Ea/RT in Eq. (5) is solved numerically using the in-built function ‘quadv’ in OCTAVE. The activation energy for any particular α is evaluated by minimizing Ω in OCTAVE using the medium scale quasi-Newton method with a mixed quadratic and cubic line search procedure using in-built unconstrained multivariable function ‘fminunc’. The minimization was done for each small intervals of conversion using the software and the Ea values were obtained with a relative error of ~8-15 kJ/mol. However a detailed determination of Ea error was not attempted. The plot of the dependence of Ea on α is shown in Fig. 5.
It can be seen from Figure 5 that the Ea steadily increases to a value of 419.5 kJ/mol at α=0.6 and then marginally increases over the rest of the conversion and reaches a maximum of 435 kJ/mol which indicates a multistep reaction mechanism. At low conversions, the Ea value is much closer to the reported Ea values of 149 kJ/mol (wood’s bath) and 277 kJ/mol (DSC) [9]. However it should be noted that the temperature range considered for the wood’s method is large i.e. from 190 to 240oC, while for the DSC it is from 200-225oC. The initial increase of Ea up to α=0.6 could be attributed predominantly to the rapid decomposition of the dinitramide anion of GUDN, whereas the marginal increase of Ea for the rest of the reaction is attributed to the formation of the guanylurea cation and its decomposition fragments.
3.3.2 Prediction of isothermal kinetics The model free isoconversional method proposed by Vyazovkin et al., can also be used to predict the isothermal conversion as a function of time from arbitrary temperature,T0 [26]. The time at which the given conversion will be reached at an arbitary temperature, can be computed using equation 6.
=
(6)
Where Tα is the temperature of the nonisothermal process at a given conversion α, with constant heating rate β, and at a fixed temperature, T0. Equation 6 was applied on the decomposition of GUDN to predict the isothermal conversion as a function of reaction time at two different temperatures of 205 and 210oC and the results are shown in figure 6.
It can be observed from fig.6, that the predicted time(t) and conversion(α) data for the decomposition of GUDN shows that the decomposition proceeds very fast at 210oC than at 205oC. As the decomposition of GUDN proceeds in a narrow temperature range of 200-220oC, the time required for 90% conversion at isothermal temperatures of 205oC and 210oC were predicted to be ~150 and ~50mins respectively. The results show that the thermal stability of GUDN is less at temperatures above 200oC and is thermally stable below 200oC.
3.4
Identification and analysis of evolved gases by mass spectrometry Analysis of the evolved gases by TG-MS during the decomposition of GUDN revealed
the formation of ions corresponding to m/z 14, 16, 18, 28, 30, 42, 43, 44, 46 and 60amu. The complete mass spectra showing the reaction products are given in figure 7.
The ions are identified as 14 (N2), 16 (O2), 17 (NH3), 18 (H2O), N2 (28), 30 (NO), 42 (HN=C=NH, methane diimine), 43 (HNCO, isocyanic acid), 44 (N2O), 46 (NO2) and 60 (NH2CO-NH2, urea). The temperature range of the evolution of products along with the corresponding TG data is shown in figure 8.
It is seen from figure 8, that the gaseous products start to appear from temperature as early as 195oC until 220oC. Though all the products were identified in this narrow decomposition range, the products detected are N2O and NO2 which appeared in the mass spectrum initially
when the time of appearance of mass fragments was considered, and are followed by HNCO, HN=C=NH. Whereas, the presence of urea (m/z 60) was detected in trace levels.
3.5 Mechanistic aspects and comparison of results Among the many dinitramide salts, the decomposition of ADN has only been studied by TG-MS so far. The mass spectrometric study on the decomposition of ADN revealed the presence of NH3, N2O and NO in the earlier stages (<100oC), while at temperatures above 150oC, the concentration of NO, H2O and NO2 increases [27]. In the case of GUDN decomposition, most of the products appeared in the mass spectra are evolved during the narrow temperature regime between 200-220oC predominantly forming N2O, NO2, isocyanic acid and methane diimine. The mechanism of decomposition and the fractions from GUDN are represented in scheme 1.
In the isoconversional methods, the effective activation energy computed for a multi-step process varies with the extent of conversion. Revealing the dependence of Ea with α helps to understand the complexity of the process and also to shed some light to identify the kinetic scheme. Vyazovkin et al. have given a detailed account on the shapes of the dependence of Ea with α to identify the kinetic scheme of the process [28]. An increasing dependence of Ea on α is attributed to the competing reactions. These dependencies were observed for the thermal decomposition of many linear polymers [28]. In the present study, the observed dependencies (Ea increases with α) by VZ method for the thermal decomposition of GUDN could be attributed to the competing reaction mechanism. The mean Ea and Ea(max) values could be used for comparing the activation energy from other methods which a gives single activation energy. However this is only a qualitative exercise in knowing the general trend as there is no direct method of correlation. The computed mean Ea for the decomposition of GUDN is 358 kJ/mol by VZ method. The Ea(max) by VZ method ~ 435 kJ/mol which is also closer to the Ea value obtained by Kissinger method. However, a higher value of ~80 kJ/mol by the Kissinger method to the mean Ea value obtained by VZ method could be because Kissinger method relies on the relative positions of Tp rather than considering the whole decomposition profile of GUDN.
4 Conclusions Non-isothermal kinetic investigations of the decomposition of GUDN were carried out using nonlinear isoconversional method of Vyazovkin. The variation of Ea with α at various conversions was evaluated. The Ea values increases steadily in the first stage (0 ≤ α ≤ 0.6) and in the later stage (0.6 ≤ α ≤ 1) it increases only marginally. The observed Ea(max) of ~435 kJ/mol by VZ method is close to the Ea value of 441 kJ/mol obtained from Kissinger method. Isothermal kinetic prediction at two different temperatures was also made by a model free isoconversional method and the results show that GUDN is thermally stable below 200oC. The major decomposition products from GUDN were identified from TG-MS as N2O, NO2, N2, isocyanic acid and methane diimine and a decomposition mechanism was proposed.
Acknowledgements Deepthi Thomas, Analytical and Spectroscopy Division, Vikram Sarabhai Space Centre (VSSC), Trivandrum is thanked for her help in TG-MS analysis and also for the inputs on decomposition mechanism. Dr. C.P. Reghunadhan Nair, Deputy Director, PCM Entity, VSSC and Dr. Dona Mathew, Polymers and Special Chemicals Division, VSSC is thanked for critically going through the manuscript and for the comments.
5 References [1]
Ang How Ghee, G. Santhosh, Advances in Energetic Dinitramides – An Emerging Class of Inorganic Oxidizers, World Scientific, Singapore, 2008.
[2]
S. Venkatachalam, G. Santhosh, K.N. Ninan, An overview on the synthetic routes and properties of ammonium dinitramide (ADN) and other dinitramide salts, Prop. Expl.Pyro. 29(3) (2004) 178-187.
[3]
H. Östmark, A. Helte, T. Carlsson, R. Adolfsson, L. Bodin, C. Eldsäter, H. Edvinsson, J. Lundgren, H. Örnhed, N-guanylurea dinitramide (FOX-12): Properties, FOI-R--2312—SE, 2007.
[4]
J. Dahlberg, New low-sensitivity modular charge propellant based on GUDN, Int. Annu. Conf. ICT, 46/1-46/5, 2006.
[5]
P. Sjoberg, Gas-generating material for gas-actuated car safety devices, PCT Int. Patent Appl. No. WO 00/40523, 2000.
[6]
Menke Klaus, Eisele
Siegfried, Gas generators and propellants based on guanylurea
dinitramide with additional energetic components, DE102011100113A1, 2012. [7]
Michienzi, Christine Marie, Knott, Christine Dianne, Insensitive primers and munitions containing low-sensitivity, low-vulnerability compositions, US Patent No. 8425703B1, 2013.
[8]
Siegfried Eisele, Klaus Menke, Gas generator propellants based on GUDN, 41st Int. Annu. Conf. ICT, 43/1 to 43/10, 2010.
[9]
H.
Ostmark, U. Bemm, H. Bergman, A. Langlet, N-guanylurea-dinitramide: a new
energetic material with low sensitivity for propellants and explosives applications, Thermochim. Acta 384 (2002) 253-259. [10] Vorde Carin, Skifs Henrik, Method of producing salts of dinitramidic acid, US Patent No. 7981393 B2, 2011. [11] D.M. Badgujar, Rashmi M. Wagh, S.J. Pawar, A.K. Sikder, Process optimization for synthesis of guanylurea dinitramide (GUDN), Prop. Expl. Pyro. 39 (2014) 658-61. [12] G. Santhosh, Robbin Poh Cheng Tien, Ang How Ghee, Thermal decomposition kinetics of ammonium dinitramide-guanylurea dinitramide mixture analyzed by isoconversional methods, Thermochim. Acta 480(2008) 43-48. [13] F.Q. Zhao, P. Chen, H.A. Yuan, S.L. Gao, R.Z. Hu, Q.Z. Shi, Thermochemical properties and non-isothermal decomposition reaction kinetics of N-guanylurea dinitramide (GUDN),Chin. J. Chem. 22 (2004) 136-141. [14] H.X Gao, H. Zhang, F.Q. Zhao, R.Z. Hu, H.X. Ma, K.Z. Xu, J.H.Yi, S.Y. Xu, Y. Gao, Kinetic behaviour of the exothermic decomposition reaction of N-guanylurea dinitramide, Acta Phys. Chim. Sin.24(3) (2008) 453–8. [15] Jiao-qiang Zhang, Wei-wei Zhao, Tie-zheng Ji,Hong-xu Gao, Rong-zu Hu, The dissolution properties of N-guanylurea-dinitramide (FOX-12) in dimethyl sulfoxide (DMSO), J. Therm. Anal. Calorim. 115 (2014) 641-646.
[16] Na Li , Feng Qi Zhao, Yang Luo, Hong Xu Gao, Li Bai Xiao, Rong Zu Hu, Rong Hui Ju, Dissolution properties of N-guanylurea dinitramide (GUDN) in dimethyl sulfoxide and Nmethyl pyrrolidone, J. Therm. Anal. Calorim. 115 (2014) 869-873. [17] Nancy N. Perreault, Annamaria Halasz, Sonia Thiboutot, Guy Ampleman, Jalal Hawari, Joint Photomicrobial Process for the Degradation of the Insensitive Munition Nguanylurea-dinitramide (FOX-12), Environ. Sci. Technol. 47 (2013) 5193-5198. [18] Helen Stenmark, Sune Sjökvist, Toxicological Testing of ADN, GUDN and FOX-7, 41st Int. Annu. Conf. ICT, 56/1 to 56/3, 2010. [19] John W. Eaton, David Bateman, Soren Hauberg, GNU Octave version 3.6.4, A high level interactive language for numerical computations, http://www.octave.org. [20] S. Vyazovkin, A.K. Burnham, Hose M. Criado, Luis A. Perez-Maqueda, Crisan Popescu, Nicolas Sbirrazzuoli, ICTAC kinetics committee recommendations for performing kinetic computations on thermal analysis data, Thermochim. Acta 520 (2011) 1-19. [21] S. Vyazovkin, A unified approach to kinetic processing of nonisothermal data, Int. J. Chem. Kinetics 28 (1996) 95-101. [22] H.E. Kissinger, Reaction kinetics in differential thermal analysis, Anal. Chem. 29 (1957) 1702-1706. [23] R. Gilardi, J. Flippen-Anderson, Clifford George, Ray J. Butcher, A new class of flexible energetic salts: The crystal structures of the ammonium, lithium, potassium and cesium salts of dinitramide, J. Am. Chem. Soc 119 (1997) 9411-9416. [24] R. Gilardi, Ray J. Butcher, A new class of flexible energetic salts: Part 7 – The structures of the guanidinium and hydroxyguanidinium salts of dinitramide, J. Chem. Cryst 32 (2002) 477-484. [25] S. Vyazovkin, C.A. Wight, Model-free and model-fitting approaches to kinetic analysis of isothermal and nonisothermal data, Thermochim.Acta340-341 (1999) 53-68. [26] S. Vyazovkin, C.A. Wight, Isothermal and non-isothermal kinetics of thermally stimulated reactions of solids, Int. Rev.Phys. Chem. 17 (3) (1998) 407-433. [27] S. Vyazovkin, C.A. Wight, Ammonium Dinitramide: Kinetics and Mechanism of Thermal Decomposition, J. Phys. Chem. A.101(31) (1997) 5653-5658. [28] S. Vyazovkin, C.A. Wight, Kinetics in solids, Annu. Rev. Phys. Chem. 48 (1997) 125-149.
Figure legends
Figure 1: Chemical structure of GUDN Figure 2: TG-DTG curve of GUDN at a heating rate of 2oC/min Figure 3: Kissinger plot for GUDN Figure 4: Conversion (α) as a function of temperature (T) at various heating rates for GUDN Figure 5: Dependence of Ea with α by Vyazovkin method for the decomposition of GUDN. Figure 6:Isothermal kinetic prediction at 205 and 210oC using eqn.6 for the decomposition of GUDN Figure 7: TG-MS spectrum of decomposition products from GUDN Figure 8: Mass-to-charge ratio of major ions detected during the decomposition of GUDN from TG-MS experiment at 5oC/min Scheme 1: Decomposition mechanism and products from GUDN
Table 1: Characteristic temperature of decomposition for GUDN from TG H.R
To (oC)
Tp (oC )
Tf (oC)
2
179.8
216.5
227.1
3
182.4
218.0
230.0
4
184.8
219.5
233.2
5
187.3
220.5
235.5
(oC min-1)
Figure2 .
Figure3 .
Figure4 .
Figure5 .
Figure6 .
Figure7 .
Figure8 .