studies have reported deceleration in the growth rate of polymer chains as polymerization progressed. [ 5,6 ] There are two schools of theories that could explain this experimental phenomenon, propagation and termination theories. The propagation theory explains this trend through reduction of available monomers in the grafted layer due to crowding. On the other hand, the termination theory justifi es the deceleration through decreased active chain ends due to surface radical termination, which is assisted by radical migration through the reversible activationdeactivation reactions involved in ATRP. These two theories have also been used to offer explanation under which conditions the solution and surface chains possess similar or different properties in a simultaneous solution and surface polymerization. [ 7,8 ] This helps to evaluate whether the properties of free polymer chains formed in solution can be used as an estimate for those of the grafted poly mer brush on surface for calculation of grafting density, as often assumed in experimental studies.The dry thickness of polymer brush ( δ ) and its grafting density ( τ ) are two important parameters of grafted In surface-initiated atom transfer radical polymerization, knowledge of grafting density is of signifi cant interest because it is one of the determining properties of grafted polymer. It is well known that not all of the immobilized initiators can grow into polymer chains. However, little is known about why this happens and what affects the grafting effi ciency. The lack of information is partly due to the diffi culty in experimental determination of grafting density on fl at substrates. To circumvent the problem, Monte Carlo simulation with bond fl uctuation model is used in this study to investigate the effects of various reaction conditions on the grafting density. The simulation results show lower grafting density when less deactivator is present. In systems with lower deactivator concentration, the number of monomer added per activation cycle is higher. Coupling this with close proximity of immobilized initiators results in chains initiated at earlier time to shield their neighboring initiator moieties from adding mono mers, thus lowering the grafting density in such a system. These simulation results also provide an explanation to the seemingly confl icting trend reported in the literatures.