Reaction calorimetry heat transfer

If a calorimetry experiment is carried out under a constant pressure, the heat transferred provides a direct measure of the enthalpy change of the reaction. Constant-volume calorimetry is carried out in a vessel of fixed volume called a bomb calorimeter. Bomb calorimeters are used to measure the heat evolved in combustion reactions. The heat transferred under constant-volume conditions is equal to A Corrections can be applied to A values to yield enthalpies of combustion. 

In those cases where concentrations are not measured directly, the problem of calibration of the in-situ technique becomes apparent. An assurance must be made that no additional effects are registered as systematic errors. Thus, for an isothermal reaction, calorimetry as a tool for kinetic analysis, heat of mixing and/or heat of phase transfer can systematically falsify the measurement. A detailed discussion of the method and possible error sources can be found in [34]. 

Develop Desired Reaction Reaction Calorimetry abdesired Power output, QKXN Heat transfer rate Accumulation, XAC Normally 0.1 to 2 liter scale Mimics normal operation Essential information for safe scale-up Very useful for process development... 

Use of medium-scale heat flow calorimeter for separate measurement of reaction heat removed via reaction vessel walls and via reflux condenser system, under fully realistic processing conditions, with data processing of the results is reported [2], More details are given elsewhere [3], A new computer controlled reaction calorimeter is described which has been developed for the laboratory study of all process aspects on 0.5-2 1 scale. It provides precise data on reaction kinetics, thermochemistry, and heat transfer. Its features are exemplified by a study of the (exothermic) nitration of benzaldehyde [4], A more recent review of reaction safety calorimetry gives some comment on possibly deceptive results. [5],... 

More recently, Sole391 studied the system to determine the extent, if any, of heterogeneous reaction. By means of differential calorimetry, he compared the heat transferred to the vessel walls with that transferred to a probe in the center of the vessel. He found that heterogeneous effects could be completely disregarded. 

In this chapter we discuss important issues as we move from laboratory to pilot plant and manufacturing. A review of batch process operation and pharmaceutical research is covered in Section 3.1.2, followed by laboratory vessels and reaction calorimetry in Section 3.1.3. In Section 3.1.4 heat transfer in process vessels is presented, including the effect of reactor type and heat transfer fluid on the vessel heat transfer capability. In Section 3.1.5 dynamic behavior based on simulation studies is discussed. 

The microanalytical methods of differential thermal analysis, differential scanning calorimetry, accelerating rate calorimetry, and thermomechanical analysis provide important information about chemical kinetics and thermodynamics but do not provide information about large-scale effects. Although a number of techniques are available for kinetics and heat-of-reaction analysis, a major advantage to heat flow calorimetry is that it better simulates the effects of real process conditions, such as degree of mixing or heat transfer coefficients. 

Regenass [10] reviews a number of uses for heat flow calorimetry, particularly process development. The hydrolysis of acetic anhydride and the isomerization of trimethyl phosphite are used to illustrate how the technique can be used for process development. Kaarlsen and Villadsen [11,12] provide reviews of isothermal reaction calorimeters that have a sample volume of at least 0.1 L and are used to measure the rate of evolution of heat at a constant reaction temperature. Bourne et al. [13] show that the plant-scale heat transfer coefficient can be estimated rapidly and accurately from a few runs in a heat flow calorimeter. 

FIGURE 3 Example reaction calorimetry study without reaction. The overall heat transfer coefficient area can be found during the steady-state temperature difference and known calibration probe heat flow, between 35 and 42 min. The heat capacity can then be found from the temperature ramp between 5 and 20 min. 

Various levels of models can be used to describe the behavior of pilot-scale jacketed batch reactors. For online reaction calorimetry and for rapid scale-up, a simple model characterizing the heat transfer from the reactor to the jacket can be used. Another level of modeling detail includes both the jacket and reactor dynamics. Finally, the complete set of equations simultaneously describing the integrated reactor/jacket and recirculating system dynamics can be used for feedback control system design and simulation. The complete model can more accurately assess the operability and safety of the pilot-scale system and can be used for more accurate process scale-up. 

In this chapter we have presented an overview of scale-up considerations involved as one moves from bench-scale reaction calorimetry to larger scale pilot plant and production reactors. Our focus has been on heat transfer and single-phase processes, addressing primarily the problem that the heat transfer area per unit reactor volume decreases with scale. Clearly, there are many challenging problems associated with multiphase vessels, with evaporation/distillation and crystallization as obvious examples, but these topics are beyond the scope of this chapter. 

The enthalpy change, dH = T dS + V dp, can be described as dH = dq - -V dp, and for a constant-pressure process, c/p = 0, we have dH = dqp. For a finite state change at constant pressure, qp = AH, that is, the heat transferred is equal to the enthalpy change of the system. This relation is the basis of constant pressure calorimetry, the constant-pressure heat capacity being Cp = dqldT)p. The relationship qp = AH is valid only in the absence of external work, w. When the system does external work, the first law must include dw. Then, the heat transferred to the system under constant-pressure conditions is qp = AH -f w. Thus, if a given chemical reaction has an enthalpy change of -50 kJ mol and does 100 kJ mol" of electrical work, the heat transferred to the system is —50 + 100 = 50 kJ mol". ... 

The superiority of this technique, especially in comparison to the so-called heat flux calorimetry, ch still remains to be described, lies in the fact that the measured signal is completely independent of the size of the heat transfer area, which may change due to a feed process, or of any other substance properties of the mixture, such as density or viscosity. These properties determine tiie heat transfer on the side of the reaction mixture or, in other words, the film heat transfer coefficient, as is well known from process engineermg. 

When discussing the sensitivity of heat flow calorimetry a representative example was chosen of a reaction releasing 27 W/batch on average when performed in a 2 liters glass vessel. A frequently found value for the coolant mass flow rate of a heat balance calorimeter amounts to 70 1/hour. Assuming a specific heat capacity of the coolant of 2600 J/kg K, this reaction power is transferred into a temperature difference between coolant inlet and outlet of 0,54 K. If the heat balance calorimeter and the heat flow calorimeter are to be of equal sensitivity, it follows that a resolution down to 1/100 K is required for the temperature difference. 

Enthalpies of reaction in solution are generally measured in an isothermal jacketed calorimeter. This consists of a calorimetric vessel that contains a certmn amount of one of the reactants that is either a liquid or, if a solid is involved, it has been dissolved in a suitable solvent. The other reactant is sealed in a glass ampoule that is placed in a holder. The vessel is enclosed in a container, which is placed in a thermostatted bath with the temperature controlled to 0.001 °C. When the system has reached thermal equilibrium, the ampoule is broken and the reaction is initiated. Throughout the experiments the temperature is measured as a function of the time and a temperature-time curve with approximately the same shape as the ones obtmned in combustion calorimetry, vdth fore-period, reaction-period and after-period is obtained. The observed temperature rise is due to several sources die heat transferred from the thermostatted bath, the energy of the reaction and the stirring energy. To correct... 

Generally, in classical reaction calorimetry only the liquid phase is taken into account in the heat balance. This means that the gas phase in equihbrium with it is neglected because of its small contribution in terms of heat transfer and heat capacity. The situation with supercritical fluids becomes complicated as soon as they occupy all the available volume. This implies that the whole inner reactor surface has to be thermally perfectly controlled when working with supercritical fluids. In this case, the cover and the flange temperature are adjusted on-line to the reaction temperature in order to neglect the heat accumulation term. 

Heat transfer is even more serious for an exothermic reaction. Minor exotherms in the laboratory may not even be apparent or if they are noticed can be easily controlled with cooling. However, the same exotherm on a large scale can be difficult to control and even present a safety hazard. Therefore it is important to thoroughly understand the thermodynamics before increasing reaction scale. This is often done by calorimetry experiments. 

Different alternatives have been presented to circumvent this issue in heat flow calorimetry. A priori off-line determination of the dependence of UA [9], adaptive calorimetry using an additional off-line measurement [12] and cascade state estimation observers [14] proven to work, will be discussed in the following section. Obviously, another alternative is to use heat balance calorimetry and to solve the energy balances given by Equations 7.1 and 7.2 simultaneously to compute the evolution of the heat of reaction, Qp and the overall heat transfer coefficient, UA. This approach will be addressed in Section 7.2.3. 

The overall heat transfer coefficient can be estimated online by using an additional process measurement (e.g., gravimetric conversion or solids content) together with state (parameter) estimation techniques to update the value of the overall heat transfer coefficient. This approach referred as adaptive calorimetry has been mainly exploited by Fevotte and coworkers [12] to monitor emulsion (co) polymerization reactions. They used a dependence of U with conversion... 

Temperature Oscillation Calorimetry A more elegant way to estimate online the overall heat transfer coefficient without any additional measurement was developed by Carloff [ 11] by the technique known as temperature oscillation calorimetry, TOC. In this approach, the unknown product UA is computed from the analysis of the sine-shaped oscillations, which are superposed on either the reactor temperature or jacket temperature. The objective is to decouple the slow dynamic of the chemical heat production from the fast dynamic variable heat transfer during the reaction. The oscillations can be achieved either by adding a calibration heater to the system or by adding a sine signal to the set point of either T or Ty Figure 7.2 shows the evolution of the reactor and jacket temperatures in a reaction calorimeter where a sine wave temperature modulation was superimposed on the reactor jacket temperature. 

Reaction calorimetry can also provide useful information for process design such as the necessary cooling power, the adiabatic temperature rise, and the heat transfer for scale-up. 

Reaction calorimetry, as discussed in Sections 7.2.2 and 7.2.3, allows the heat transfer coefficient of the reactor to be estimated. The information of the heat transfer coefficient calculated on a daily basis can be very useful to detect the level of fouling in the reactor walls and to anticipate and therefore carry out actions that minimize its deleterious effects on the performance of the reactors [31]. 

---