Available online at www.sciencedirect.com J. of Supercritical Fluids 45 (2008) 121–131 Optimization of reaction calorimetry with supercritical fluids: A complete term-by-term analysis of the heat flow equation Charalampos A. Mantelis, Thierry Meyer ∗ Ecole Polytechnique F´ ed´ erale de Lausanne, Institute of Chemical Sciences and Engineering, Group of Macromolecular Processes, Station 6, CH-1015, Lausanne, Switzerland Received 19 March 2007; received in revised form 16 November 2007; accepted 7 December 2007 Abstract A complete analysis of the heat flow equation in reaction calorimetry with supercritical fluids is presented. An especially developed reaction calorimeter is employed to study the challenges introduced by the supercritical state of the reaction mixture and the particularities that they generate in the calorimetric calculations. As a model reaction the free radical dispersion polymerization of methyl methacrylate is chosen. Each term of the heat flow equation is optimized with special attention to the accumulation term and the injection term. The overall heat transfer coefficient, being the most important variable of the heat flow equation, is estimated using four different approaches, both theoretical and experimental. Moreover, the injection phase of additional reactants is shown to generate undesirable temperature oscillations and an optimized injection strategy is found to eliminate the erroneous calorimetric calculations. As a result the enthalpy of reaction and the heat released by the reaction are very accurately estimated. © 2007 Elsevier B.V. All rights reserved. Keywords: Calorimetry; Heat transfer; Monitoring; Optimization; Polymerization; Supercritical fluids 1. Introduction Supercritical fluids (SCFs) have been continuously attracting scientific interest during the past decades because they represent an environmentally benign alternative for volatile organic com- pounds (VOCs) in the chemical industry [1]. The production and consumption of such organic compounds has been progressively phased out starting from the Montreal Protocol on Substances that Deplete the Ozone Layer, signed in 1987. The particularity that renders SCFs attractive is the possibility to tune their physi- cal and chemical properties by slight changes of the temperature and/or the pressure in order to resemble those of a liquid or a gas [2]. Although research on SCFs initially started with water, the most widely used nowadays is carbon dioxide (CO 2 ). From a process development point of view CO 2 has a relatively easily attainable critical point (T c = 31.1 ◦ C and P c = 7.38 MPa) and it is non-toxic and non-flammable. Furthermore, it is very cheap and it can be provided in high purity. Finally it is advantageous for ∗ Corresponding author. Tel.: +41 21 6933614; fax: +41 21 6933190. E-mail address: thierry.meyer@epfl.ch (T. Meyer). biomedical and pharmaceutical applications due to its generally regarded as safe (GRAS) status. Beginning from the first reactions reported with SCFs in the second half of the 19th century their industrial applica- tions increased and the two most important applications came later with the commercialization of the BASF process for the production of ammonia in 1913 and the polymerization of ethy- lene under supercritical conditions in 1939 [3,4]. More recently, CO 2 has been widely employed as an extracting agent for the extraction of natural products, mainly for food ingredients and phytopharmaceuticals, and as a solvent for polymerization reac- tions [5,6]. Overall, supercritical reaction chemistry is advancing at a high pace and the need for kinetics information and safety analysis for process development and scale-up studies is increas- ing accordingly. Heat flow reaction calorimetry is one of the analytical tools that can provide such information [7]. By definition it is a ther- mal analysis technique that monitors the evolution of a chemical reaction by measuring the heat that flows in or out of a reactor [8]. It provides an extensive engineering insight of the chemical process allowing for the development of model-based predic- tive control schemes for improved process control. Overall heat 0896-8446/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.supflu.2007.12.001