Date of Graduation


Document Type


Degree Name

Bachelor of Science in Chemical Engineering

Degree Level



Chemical Engineering


Spicer, Tom

Committee Member/Reader

Walker, Heather

Committee Member/Second Reader

Smith, Chad

Committee Member/Third Reader

Lopes, Paolo


When gaseous contaminants are released into the atmosphere said contaminants form clouds that travel in the prevailing wind direction. An individual caught in the downwind path of one such cloud will accumulate a dosage, or toxic load, of said contaminant that is dependent on a multitude of environmental factors. This thesis quantifies unpredictability – or variation – in toxic load measurements using downwind concentration data captured from finite-duration gaseous contaminant releases conducted at set conditions in the Chemical Hazards Research Center ultra-low speed wind tunnel.

Finite-duration releases can be divided into two categories – time-limited steady releases and puffs – depending on release duration. Time-limited steady releases are found to consist of three distinct exposure phases with an ensemble average concentration in the middle, “steady-state” phase. This ensemble average concentration has been found experimentally to be the same concentration as a release with a constant release rate. Puffs consist of only two phases and do not have a middle phase. Toxic load is the measure typically used by toxicologists to assess the consequences of a release of a toxic chemical. Toxic load values can be calculated for each phase of a finite-duration release with the assumption that toxic load is equal to the integral of concentration. For any repeated releases under otherwise identical conditions, the evaluation of toxic load as an integral of concentration will be different from release to release. The variability of the toxic load for individual releases compared to the ensemble average is important to understand the potential variation of release consequences compared to predicted average release behavior. Such comparisons are often prepared in consequence assessments and HAZOPs. Variability itself is quantified with coefficients of variation calculated from mean toxic loads and their corresponding standard deviations.

Toxic load is found to vary most in the departure phase of finite-duration releases, or the time over which the trailing edge of the cloud passes through a downwind observation point. Steady-state toxic load variability is also shown to be constant. Finally, an increase in the magnitude of the toxic load factor is shown to increase toxic load variability across all eighteen tests sets.


toxic load, atmospheric dispersion, wind tunnel, Chemical Hazards Research Center, MATLAB