The one-dimensional linear eddy model (LEM), which captures both molecular and turbulence mixing mechanisms, has great potential for providing useful insights into the effects of inhomogeneity and turbulence on HCCI combustion. In this work, the one dimensional LEM code has been enhanced so that detailed/reduced chemical kinetics can be incorporated in a straightforward manner. Additionally, a chemistry agglomeration model is employed to speed up the chemistry calculations, while maintaining accurate predictions of combustion quantities such as pressure, temperature and chemical species concentrations. The chemistry agglomeration model clusters computational cells with "similar" thermo-chemical compositions and performs chemistry calculations on the clusters of cells instead of individual cells. By doing this, significant savings in computational time are achieved. This model communicates with the LEM code at each time step and the thermo-chemical composition of the computational cells is mapped back and forth between them. Furthermore, a series of LEM calculations with reduced chemical kinetics are performed to explore the effects of fuel distribution on HCCI combustion for both n-heptane and iso-octane fuels. Results show that the effect of fuel stratification on combustion phasing is fuel dependent. Coarsening the fuel distribution monotonically advances combustion phasing significantly for n-heptane, but lesser so for iso-octane. With fuel stratification, increases in NOx and CO emissions are observed. Results also show that turbulence retards combustion phasing and its relative effect is dependent on the fuel. The relative importance of c (i.e., the ratio of specific heats) due to the different thermal properties of each species and effect of chemistry in generating thermal gradients is also studied. (C) 2012 Elsevier Ltd. All rights reserved.