Experimental evidence has shown that composites comprised Si and Sn nanoparticles embedded inside a matrix are the most promising next generation anodes for Li-ion batteries. This is due to the ability of the matrix material to constrain/buffer the up to 300% volume expansion that Sn and Si undergo upon the formation of lithium rich alloys. Damage still occurs at the nanoparticle/matrix interface, and hence further materials design is required in order to commercialize such anodes. Initial theoretical works have predicted that low volume fractions and high aspect ratios of the nanoparticles result in a greater mechanical stability and hence better capacity retention. The most important design parameters, however, such as particle size and spacing have not been considered theoretically. In the present study, therefore, a gradient enhanced damage model will be employed to predict that damage during Li-insertion, is negligible when the particle size is 20 nm, and the interparticle half-spacing greater then 1.5 times the particle diameter. Furthermore, from the matrix materials considered herein graphene is predicted to be the most promising matrix, which is consistent with recent experimental data. (C) 2012 Elsevier B.V. All rights reserved.