Development of Instrumental and Computational tools for investigation of polymer flammability

Abstract

This thesis describes published work undertaken over the last 17 years. The main focus is the development and utilization of the microscale combustion calorimeter (MCC) and how it helps us understand the flammability of materials.

A reproducible way to quantitatively assess material flammability was needed. The simplest approach is based on the molecular structure of a material to determine which moieties influence the flammability. This approach is based on material properties that can be measured using small-scale thermal analysis methods such as Thermogravimetric Analysis (TGA), Differential Scanning Calorimetry (DSC), and the MCC. Properties such as thermal stability, heat of gasification, and heat of combustion provide key information about materials' flammability.

Tests such as LOI, UL-94 and other Bunsen burner type tests provide pass/fail type results. These types of tests are not quantitative and are dependent on physical properties and extrinsic parameters such as sample geometry and orientation. They do not provide a true measure of flammability, and formulations can be optimized to pass the test even though they are highly flammable.

A more quantitative approach is to use larger bench scale flammability tests such as cone calorimeter and Ohio State University (OSU) fire calorimeter. Although these tests are also dependent on the physical properties and test geometry, they measure properties such as burning rates, mass loss rates, and combustion efficiency under different imposed fire scenarios from which material properties can be derived.

The ultimate flammability test for a material is to subject it to a fire in a real-life scenario where other materials are present, a full-scale fire test. These tests can involve a single item or combinations of items. These tests are highly dependent on physical and material properties and are useful for making direct comparisons of one scenario to the next, for properties like flame spread and time to flashover.

Measurements have been made using all of the thermal analysis and fire test methodologies listed above. Correlations between the test methods have been drawn and the theory relating them derived. Predictive methods for estimating polymer flammability from molecular structure have been formulated using a molar group contribution approach. Methods for predicting fire performance in the bench scale tests from the small scale test measurements have also been derived. Modelling the bench scale fire performance in the quantitative tests as well as determining a probability for the rating in the pass/fail type bench scale tests for a range of polymeric materials has been undertaken. This type of work in the small- and bench-scale has helped identify materials that perform well when subjected to the harshest fire conditions in the full-scale. The ultimate goal being to save lives by preventing deaths due to fire through the development of more fire-safe materials.