| Traditionally, single constant Kubelka-Munk type colorant formulation algorithms have been used for color control in the paper industry. Tuning data is derived from colored handsheets representing dyeing of a particular color grade, applicable to a substrate of similar properties. Due to furnish variation and changes in the chemical environment, such tuning data is of limited accuracy in practice. Kubelka-Munk approaches have numerous other limitations, in part due to their physically unrealistic assumptions. In particular, they neglect fluorescence phenomena, the interdependence of absorption and scattering, and nonlinearities due to colorant interactions. This thesis addresses those problems.; A set of colored handsheets was made, employing several anionic direct dyes and fluorescent colorants, individually and in various combinations. Both a spectrophotometer and a spectrofluorimeter were used for measuring color properties. An extended Langmuir adsorption isotherm was used in modelling the dye-on-fiber in each dyeing. Kubelka-Munk absorption and scattering coefficients were then modelled based on dye-on-fiber, and a number of the limitations of the Kubelka-Munk approach were clearly demonstrated.; An extended phenomenological model was derived, incorporating fluorescence and interdependence of absorption and scattering. This model predicts illuminator-independent radiance transfer factors based on dye-on-fiber, from which total radiance factor responses under arbitrary illumination can be computed. It requires spectrofluorometric measurements to characterize the coloring process.; A new reflectance factor model, based on the same adsorption isotherm approach, was derived for non-fluorescent colorants. A corresponding total radiance factor model, which is illuminator-dependent, was derived for fluorescent colorants. These models have provision for phenomena such as broadening of absorption and scattering bands, which are encountered in practice. Being based on spectrophotometric measurements, they are directly applicable in industrial settings, and predict colorant responses reliably under wider ranges of conditions than the Kubelka-Munk approach. |
