Safer Scale-Up of Batch and Semi-Batch Reactions Part 3: Applications of Adiabatic Calorimetry

Richard Kwasny, Ph.D., Senior Consulting Engineer and Gabe Wood, Manager Thermal Hazards Testing & Consulting, Fauske & Associates, LLC

Background

From a process safety perspective, we can use stirred-reaction calorimetry to measure the heat of a reaction and calculate the adiabatic temperature rise for an exothermic reaction, assuming there are no heat losses to the environment. The rise in batch temperature can be derived from the experimental data using ΔT_ad = ΔH_r / Cp,_s where the adiabatic temperature rise is ΔT_ad , the heat of the reaction is ΔH_r , and the specific heat of the reaction mass Cp,_s .

The adiabatic temperature rise provides for a thermodynamic estimate that if cooling were lost, we could predict the maximum temperature of the synthetic reaction (MTSR) using the relationship MTSR = T_OP + ΔT_ad , where T_OP is the temperature of the process.

However, this approach needs to be refined using the tempering effect of the batch solvent(s), incorporating the real change in heat capacity with increasing temperature, and accounting for any thermal instability from the reaction mixture at elevated temperature under adiabatic conditions. This type of worst-case-scenario, where the batch heats exponentially, and cooling is a linear function, is commonly referred to as a thermal runaway reaction. In many situations, we know the thermal properties of the desired reaction, but we are not aware of process safety hazards associated with the adverse reaction in terms of reaction kinetics, temperature/pressure rates, and the production of fixed gases and the corresponding maximum temperature/pressure.

Adiabatic Calorimeters

The first lab-scale attempt to study the adverse reaction using adiabatic calorimetry was the Accelerating Rate Calorimeter (ARC™). It was successful in identifying decomposition reactions that were driven by the kinetics of the reaction during a long-hold time under isothermal conditions or at elevated temperatures. However, the test data were obtained with a Φ factor much greater than one; and resulted in an estimated adiabatic temperature rise rather than actual scale-up conditions due to the significant amount of heat which was absorbed by the test cell.

The Φ-factor, or thermal inertia, can be calculated using Φ = 1 + (M_v *Cp,_v )/(M_s *Cp,_s ), where M_v and M_s are the masses of the test cell and sample, respectively; Cp,_v and Cp,_s are the corresponding specific heat of the test cell and the sample.

Fauske & Associates, LLC developed low Φ-factor agitated adiabatic calorimeters, which did not require any assumptions, and the adverse reaction was allowed to proceed adiabatically in an unhindered manner. The Advanced Reactive System Screening Tool (ARSST™) and Vent Sizing Package 2 (VSP2™) adiabatic calorimeters allowed for direct scale-up of the calorimetric temperature/pressure data allowing for the measurement of the actual adiabatic temperature rise as opposed to a theoretical calculation requiring several assumptions. These instruments can also be used to simulate various runaway scenarios, and the temperature and pressure rise rate data can easily be used to perform emergency relief sizing using the Design Institute for Emergency Relief Systems (DIERS) methodology.

Practical Applications of ARSST and VSP2 Test Data

These devices can be used to characterize the hazardous properties, tempering, and flow regimes of adverse chemical reactions, fire scenarios, and worst-case scenarios in terms of thermal runaways and thermal decompositions under adiabatic conditions.

Once the adverse reaction is characterized, it is possible to apply DIERS methodology to design engineering controls in terms of adequately sized emergency venting, thereby making the process safer in the event of an unwanted process deviation, e.g., loss of cooling.

Solid powders containing carbonate functional groups are often dried as part of the normal work-up process. Drying these types of organic substrates can suddenly result in decarboxylation resulting in significant pressure and pressure generation rates due to large volumes of evolved carbon dioxide and water during the endothermic decomposition.

The resulting off-gas and vapor can be initiated by exceeding the recommended isothermal drying time or raising the internal temperature of the dryer without knowledge of the consequences; both of which depend upon the kinetics of the decomposition reaction. Understanding the decomposition kinetics and time limits can better allow for proper dryer engineering controls, e.g., vent relief or use of a dryer with an adequate maximum allowable working pressure, etc.

A key advantage of using the ARSST or VSP2 is that the testing can be conducted in a batch or semi-batch mode with agitation as needed. Therefore, a test can be designed to study the adverse reaction(s) under real process conditions. Then the data can be used to determine and design the proper pressure relief needed using DIERS technology.

In summary, low Φ-factor adiabatic testing data can be used to determine:

• Onset temperature, adiabatic temperature rise, and heat of reaction for exothermic events,

• Moles of non-condensable gas generated by the reaction,

• Thermal runaway data, used for DIERS relief sizing,

• Identification of venting behavior (gassy, tempered, and hybrid),

• Determination of flow regime (two-phase or single-phase), and

• Kinetic data, e.g., Time to Maximum Rate (TMR) or Temperature of No Return (TNR).

If you are interested in learning more about how low Φ-factor adiabatic testing data can be used to improve your organization, contact us.

References

1. Stoessel, Francis, Thermal Safety of Chemical Processes, Wiley-VCH, (2008).
2. Grewer, Theodor, Thermal Hazards of Chemical Reaction, Volume 4, Elsevier, (1994).
3. Barton, John, and Rogers, Richard, Chemical Reaction Hazards, Second edition, Gulf Publishing Company, (1997).
4. Guidelines for Chemical Reactivity Evaluation and Application to Process Design, Center for Chemical Process Safety of the American Institute of Chemical Engineers,(1995).
5. Burelbach, J. P., “Advanced Reactive Systems Screening Tool (ARSST),” Presented at the Mary Kay O’Connor Process Safety Center Symposium, College Station, Texas, (October 26-27, 1999).
6. Fauske, H. K., “Managing Chemical Reactivity-Minimum Best Practice,” Process Safety Progress, Vol. 25, No. 2, (2006).
7. Fauske, H. K., “Properly Size Vents for Nonreactive and Reactive Chemicals,” Chem. Eng. Progress (February 2000).
8. Fauske, H. K., “Revisiting DIERS’ Two-Phase Methodology for Reactive Systems Twenty Years Later,” Process Safety Progress, Vol. 25, No. 3, pp. 180-188 (September 2006).
9. Leung, J. C., Fauske, H. K., and Fisher, H. G., “Thermal Runaway Reactions in a Low Thermal Inertia Apparatus,” Thermochimica Acta, 104, pp. 13-29 (1986).