Thermal Hazard Evaluation of Lauroyl Peroxide Mixed with Nitric Acid

Many thermal runaway incidents have been caused by organic peroxides due to the peroxy group, –O–O–, which is essentially unstable and active. Lauroyl peroxide (LPO) is also sensitive to thermal sources and is incompatible with many materials, such as acids, bases, metals, and ions. From the thermal decomposition reaction of various concentrations of nitric acid (HNO3) (from lower to higher concentrations) with LPO, experimental data were obtained as to its exothermic onset temperature (T0), heat of decomposition (ΔHd), isothermal time to maximum rate (TMRiso), and other safety parameters exclusively for loss prevention of runaway reactions and thermal explosions. As a novel finding, LPO mixed with HNO3 can produce the detonation product of 1-nitrododecane. We used differential scanning calorimetry (DSC), thermal activity monitor III (TAM III), and gas chromatography/mass spectrometer (GC/MS) analyses of the reactivity for LPO and itself mixed with HNO3 to corroborate the decomposition reactions and reaction mechanisms in these investigations.


Introduction
In many organic reactions organic peroxides are used as an initiator. Lauroyl peroxide (LPO) is also employed as a radical source as an initiator or catalyst [1][2][3]. LPO is an oxidizing agent that can ignite organic materials; hence a dangerous fire and explosion risk may inevitably occur if it is not handled properly. Strongly reduced materials, such as sulfides, nitrides, and hydrides, may react explosively. Typical examples include vigorous reactions with other reducing agents, such as with charcoal which sometimes ignites.
Information about organic peroxide mixed with acid or base has recently been reported [4][5][6][7][8]. This study was planned with a view to evaluate the hazard during the period of mixing of LPO contacted with nitric acid (HNO 3 ). With calorimetric tests, the hazardous characteristics with respect to specific inorganic acids could be established precisely. The DSC experimental data were obtained of exothermic onset temperature (T 0 ), heat of decomposition (ΔH d ), and other safety parameters for loss prevention of runaway reactions and thermal explosions [7][8][9][10]. The reactivity for LPO and itself mixed with HNO 3 was evaluated to corroborate the decomposition reactions and reaction mechanisms in these investigations.
Thermal activity monitor III (TAM III) is a micro-isothermal calorimeter that can test thermal behavior under various storage temperatures. We can apply the TAM III to calculate the self-accelerating decomposition temperature (SADT) and isothermal time to maximum rate (TMR iso ) under related parameters. All sample set-up and data acquisition steps are performed by a devoted software package, TAM III Assistant™, which can also perform most common types of data analyses. In the isothermal mode, the high level of control enables both long and short-term experiments to be performed with excellent baseline stability [11]. The scanning mode operates a linear ramp of temperature with time. Because the instrument records data continuously, also during the temperature change, phase transitions or other temperature-dependent effects will be detected. Since measurement is continuous, there are no breaks in the data collected.
In gas chromatography (GC), a gas or a vaporized sample is introduced and carried along by an inert carrier gas through a long, thin column where the sample components are separated. The components are flushed sequentially from the column and through a detector, and are identified by measuring the time from introduction to detection. The end of the GC column can be coupled directly to the mass spectrometer (MS). The mass spectrometer breaks up constituents into molecular ions and other fragments, which then pass through an electric and/or magnetic field that separates them according to their mass-to-charge ratio. Thus, the GC separates the components within a compound while the MS identifies these components [12,13].
GC/MS is a quantitative method [14][15][16][17][18][19] for determination of the components in a complex hydrocarbon mixture. We used GC/MS for the identification and structural analysis of compounds in the reaction of LPO with acids. We compared and analyzed the various compositions of incompatible decomposition by chromatography. We can propose the possible decomposition mechanism and major products under acid-catalyzed situations and acquire information on thermal hazards of LPO with acids of interest.

Thermal Analysis by DSC
HNO 3 of different concentrations (0.1, 1, 2, 6, and 12 N) was used as contaminant. LPO was mixed with HNO 3 for DSC tests. The experimental results are displayed in Table 1 and Figure 1. In the investigation, most of the mixing conditions displayed a strong phenomenon: HNO 3 caused a high degree of thermal hazard during the experiments because the heat of decomposition increased. Especially for HNO 3 from a concentration of 1 N, in the thermal decomposition process we found the new detonation product 1-nitrododecane. In the 12 N nitric acid contamination even more thermal energy was emitted. Early studies of the decomposition peroxides indicated a relatively straightforward mechanism by which the diacyl peroxide broke down by scission of the oxygen-oxygen bond (-O-O-) to render lauroyloxy radicals. When free radicals are produced in solution they are invariably created in pairs. Since small free radicals react with very low activation energies it might be thought that they recombine immediately and no radicals would escape this early "gemination", period [20]. The homolysis of the major reaction mechanism [21,22] can be explained by the following equations: [2R‧]   R-R + R'CH 3 + R"CH=CH 2 (4) Accordingly, the initial step (1)(2)(3)(4) in the reaction is to generate the lauroyloxy radical; decarboxylation [23] of the lauroyloxy radical is a very rapid process. The decarboxylation of the lauroyloxy radical generates an undecyl radical singlet pair. From this mechanism and GC/MS analysis, we obtained hendecane, hendecene, and a cage product that undergoes a recombination into docosane.
In the second step (5-7), there is an intersystem crossing to the radical pair, then the radicals are combined together to yield undecyl ester. The presence of acid furnishes another mechanism pathway. Here, [RCOO‧], free radicals integrate with hydrogen producing dodecanoic acid; the secondary reaction that occurred formed the olefin hendecene.
According to the DSC alignment, when LPO was added to 1 N HNO 3 in the thermal decomposition process from GC/MS detector analysis, we found 1-nitrododecane. It means that its mechanism (8-10) was complex and showed two consecutive exothermic peaks in the overall reactions. Nitro compounds are organic compounds that contains one or more nitro functional groups (-NO 2 ). They are often highly explosive, especially when the compound contains more than one such nitro group. The presence of impurities or improper handling can trigger a violent exothermic decomposition.
The onset temperature of LPO is 50 °C [7]; when LPO is decomposed by thermal decomposition under the high concentration of 12 N HNO 3 , the [R‧] will be generated. At the same time, [R‧] will react with 12 N HNO 3 , and produce a great deal of nitro compound, that emits a good deal of thermal energy. The results are illustrated in Figure 1

Thermal Decomposition Analysis of LPO Mixed with Inorganic Acids by TAM III
A few calorimetric methods can be utilized to test runaway reactions and calculate the kinetic parameters, such as frequency factor (A), apparent activation energy (E a ), reaction order (n). Designing a suitable vessel and realizing the basic characteristics of a runaway reaction, such as kinetic and safety parameters (TMR iso , SADT) for LPO [7], are a prevailing way of reducing the consequences associated with the above risks in an emergency. TAM III was employed to determine isothermal reaction behaviors. Isothermal modeling is extensively used in evaluating the runaway reactions, requiring detailed consideration of several competing and often interacting heat transfer, mass-transfer, and thermodynamic processes. In this study, we elucidated a variety of basic thermal hazard characteristics for LPO with HNO 3 by using the TAM III tests.
Tables 2 to 5 present the thermal runaway decomposition of LPO mixed with the HNO 3 under four isothermal temperatures. Figure 2 presents the heat flow vs. time for the thermal decomposition of 95 mass% LPO with 1, 6, and 12 N HNO 3 , at 60 °C by TAM III tests. From the investigations, as the isothermal temperature increased, the time for emergency response became significantly shorter. As for the effects of 12 N HNO 3 , the heat flow at 50 °C was about 0.0048 W/g and heat of isothermal (∆H iso ) was about 2,060 J/g for the effects of 6 N HNO 3 , the heat flow was about 0.0022 W/g and ∆H iso was about 1,013 J/g, for the effects of 1 N HNO 3 , the heat flow was about 0.0027 W/g, and ∆H iso was about 793 J/g, when the concentration of HNO 3 (12 N) was decreased to HNO 3 (1 N), the degree of hazard was lessened significantly. Figure 2 presents the heat flow vs. time for the thermal decomposition of 95 mass% LPO with 1, 6, and 12 N HNO 3 , at 60 °C by TAM III tests. TMR iso was 11.6, 9.51, and 4.01 h under isothermal conditions of 50 °C for 1, 6, and 12 N HNO 3 . LPO mixed with HNO 3 [4,6] can produce the detonation product of 1-nitrododecane, which is a novel finding. In TAM III tests, we investigated LPO mixed with 12 N HNO 3 and found it can emit a large amount of heat at low temperature, but under higher isothermal condition the thermal reaction started before the samples reached a condition of the thermal equilibrium. In our study of LPO from VSP2 [7], LPO did melt at about 50 °C, and the autocatalysis behavior appeared, correspondingly. When LPO is dissolved completely, it would result in an instantaneous runaway, and generate huge pressure and huge amounts of heat.   In the meantime, LPO was attacked by high concentration of HNO 3 , and then the unstable structure was cleaved and released much more energy. Compared with the heat flow versus time for the thermal decomposition of 1, 6, and 12 N HNO 3 mixed with LPO, the heat flow was quite obviously different between acids. TMR iso under isothermal conditions of 50, 60, 70, and 80 °C for acids are demonstrated in Tables 2-5. Under isothermal conditions of 70 and 80 °C for LPO mixed acids, there is only 10 to 15 min to respond to the accident.  (11) where A, E a , R, and T represent frequency factor, activation energy, gas constant (R = 8.314 J/(mol K), and absolute temperature, respectively. The plot of lnk vs. 1/T is expected to be a straight line; the Arrhenius kinetic parameters, E a can be calculated from the plot accordingly to which show the determination of activation energy from the slope of lnQ vs. 1/(RT) for the thermal decomposition of 95 mass% LPO with 1, 6, and 12 N HNO 3 , at 50, 60, 70, and 80 °C by TAM III tests. As calculated from the Arrhenius equation, the value of E a for the thermal decomposition of 95 mass% LPO with 1, 6, and 12 N HNO 3 at 50, 60, 70, and 80 °C was 124.47, 126.14, and 107.78 J/mol, respectively.

Sample Preparations
95 mass% LPO was directly purchased from the Fluka Co., and both density and concentration were measured. Then, LPO was stored in a refrigerator at 4 °C. Deionized water (H 2 O) was used as the diluent in preparing different concentrations of commonly used inorganic acids, which were selected to combine with LPO in this study for studying its incompatible reactions. The typical reactions of LPO with 0.1, 1, 2, 6, and 12 N HNO 3 were selected for comparative study.

Differential Scanning Calorimetry (DSC)
Dynamic scanning experiments were performed on a Mettler TA8000 system coupled with a DSC 821 e measuring test crucible (Mettler ME-26732) that could withstand relatively high pressure of about 100 bars. STAR e software was employed for acquiring thermal curves. To have a better thermal equilibrium, the heating rate (β) was chosen at 4 °C/min. The test cell was sealed manually by a special tool equipped with Mettler's DSC. We conducted dynamic scanning by starting the programmed setting. The range of temperature rise was chosen from 30 to 300 °C for each condition of experiments. This apparatus could adequately measure the heat flow, and then via experimental data to calculate the T 0 , TMR iso , and ΔH d by software imbedded in the interior function of DSC.

Gas Chromatography/Mass Spectrometer (GC/MS)
Thermal decomposition of LPO with acids and its decomposition products was determined by GC/MS [14]. It was an Agilent Technologies 6890N model equipped with a split/splitless injector. The details of the column are: DB-WAX; Length 30; I. D. 0.25 mm; Film (μm) 0.25. The injector temperature was 280 °C and the detector temperature was 250 °C. The GC/MS oven temperature was programmed from 60 °C kept at 1 min then 20 °C/min to 250 °C, which was held for 15 min. Here, helium was used as carrier gas at a flow rate of 1.2 mL/min.

Thermal Activity Monitor III (TAM III)
Isothermal microcalorimetry (TAM III) is generally applied on a range of products for thermal measurements originally manufactured by Thermometric in Sweden. TAM III was used to investigate the runaway reaction of LPO mixed with HNO 3 at 50, 60, 70, and 80 °C. The absolute temperature could be adjusted to within 0.02 K; while operating in isothermal mode, the bath mean temperature fluctuations were within 10 -5 K. The maximum scanning rate is ±2 K/h for nearly chemical and physical equilibrium. We used the software of TAM III Assistant™ to govern the thermostat. The thermostat liquid is mineral oil with a total volume of 22 L, and the temperature range of the thermostat is 15-150 °C when mineral oil is employed. This study investigated LPO mixed with HNO 3 solution under a specified isothermal condition. Under the isothermal conditions, a series of LPO homolytic scissions were carried out, and the reaction and thermal decomposition of LPO individually mixed with HNO 3 solution were confirmed completely.

Conclusions
Many organic peroxides have been defined as hazardous materials, and LPO is among them. From our study of the thermal decomposition of LPO and acids, these have thermal hazard potentials due to their unstable and sensitive characteristics. According to the experimental results, LPO is very sensitive to inorganic acids (here HNO 3 ), especially at high concentrations (12 N).
We believe that many peroxides in storage and in transport may potentially be contaminated by acids or other pollutants. Instantaneous decomposition or combustion of such unstable substances may engender explosions and create thermal hazards. Through the results of this study, we could provide information on the hazards and instruct the relevant staff on how to forestall and mitigate an accident before or during the period of runaway triggered by LPO along with the above-mentioned contaminants.