Use of Statistical Experiment Design to Study Newspaper Deinking, 1999 Recycling Symposium Proceedings

John K. Borchardt, Kirk H. Raney, Ruth J. Blanco, Anne T. Coleman, Paul L.

Tortorici, Drew W. Matalamaki, P. Gregory Shpakoff, and Vevolyn G. Lott

Shell Chemical Company

The use of statistical design to optimize deinking process design has been studied for the flotation deinking of newspapers.

The interactions of two variables are often important in determining flotation

deinking efficiency. The interactions can be synergistic or antagonistic. These interactions would not be discovered using the traditional method of varying one parameter and holding the other variables constant. An experiment design in which two or more process parameters are varied simultaneously is required to identify these interactions.

A considerable number of conclusions could be drawn about the relative

importance of process variables on ink particle size distribution, ink removal

efficiency, and flotation yield loss in a wash – flotation deinking process.

However, these findings could be influenced by the deinking surfactant and the ink chemistry.