The non-parametric kinetics (NPK) method has been recently developed for the kinetic treatment of thermoanalytical data. The
most significant feature of this method is its ability to provide information about the reaction kinetics without any assumptions
either about the functionality of the reaction rate with the degree of conversion or the temperature. This paper presents
the results of the application of the method to adiabatic calorimetry. Some data have been obtained by numerical simulation,
but also the thermal decomposition of DTBP, a well known first order reaction, has been studied, being the obtained results
in good agreement with literature.
For a great number of European safety groups, reaction calorimetry is the key technique for analysis of the main reaction in the risk assessment of chemical processes. A comparison of calorimetric studies of model reactions, the N-oxidation of two substituted pyridines with hydrogen peroxide, made by several European groups, can open the door to standardization of the methodologies used. However, the intrinsic experimental complexity of the model reactions, which included dosing at high temperature, a multiphase system and evaporation, and the different evaluation criteria, produced a considerable dispersion between the results obtained by the various groups.
Traditionally, the kinetic treatment of adiabatic calorimetry data has been based on the results of one or more experiments, but always with the assumption of the kinetic model that the reaction follows to calculate the kinetic parameters. In this paper a method for the determination of the activation energy that uses a set of adiabatic calorimetry data is developed. To check the method, the thermal decompositions of two peroxides were studied.
A small scale (100 mL) calorimeter is developed. It includes a glass vessel submerged in a thermostatic bath, a compensation
electrical heater, and a control system. The typical operation mode consists on introducing the solvents and part of the reactants
into the vessel, to stabilise a temperature of the bath (Tj) some degrees below the desired process temperature (Tp) and to adjust the reaction mass temperature (Tr) to Tp using the electrical heater. An oscillating set point is established for Tr, which produces an oscillating response of the
applied compensation power (Qc). Finally, the rest of reactants are dosed to the vessel. A small deviation of Tr and Tp is observed. Even though it can be avoided improving the tuning of the controller, it can be useful for enhancing the calculation
of the heat capacity of the reaction mixture (CP). The signals of Tr, Qc and Tj are processed on-line using the FFT (Fast Fourier Transform) method as the mathematical tool used to analyse the data obtained,
producing accurate values of the heat evolved (Qc) by the process, the heat transfer coefficient (UA), and the heat capacity of the reaction mixture (CP).
Titanium nitride and carbide oxidation have been studied using TG and DSC. Titanium nitride shows a oxidation behavior were both techniques detect a unique phenomenon. Titanium carbide shows a variable behavior depending on the heating rate and sample size. Low masses and heating rates provide similar results to titanium nitride. However, using moderate sample sizes and scanning rates a two-stage oxidation is observed. The first step is extremely fast and exothermic, consuming the oxygen trapped inside the nanoparticle bed. The second is controlled by the diffusion of the oxygen and CO2 through the sample. Thermal safety conclusions are extracted from this observation. Energies of activation calculated using traditional kinetic models are lower than those found in the literature, being an indication of the influence of the specific surface of the material.
Estimation methods developed over years by S. W. Benson and co-workers for calculation the thermodynamic properties of organic
compounds in the gas phase are applied to a pharmaceutical real process with all type of non-idealities. The different strategies
used to calculate the reaction enthalpy of a chemical process, in the absence of data for complex molecules, using the Benson
group additivity method are presented and also compared with the experimental value of reaction enthalpy obtained using reaction
calorimetry (Mettler-Toledo, RC1). We demonstrate that there are some strategies that can be followed to obtain a good estimation
of the reaction enthalpy in order to begin the safety assessment of a chemical reaction. This work is part of an industrial
project  in which the main objective was the risk assessment of chemical real and complex processes using the commonly
available tools for the SMEs (with limited resources).