Silicon (Si) has been regarded as a promising anode material for the lithium-ion batteries due to its high theoretical specific capacity about 4200 mAh/g and low costs. However, during the lithiation/delithiation processes, the Si particle experiences repeated expansion in size up to 300 %, which leads to associated particle pulverization, electrode delamination, and destabilization of solid-electrode interface (SEI) layer. The conventional polyvinylidene fluoride (PVDF) binder has poor cycling performance on the Si anode and a lot of efforts are currently focused on flexible and polar polymeric binders such as carboxymethyl cellulose (CMC), alginate, and poly(acrylic acid) (PAA), etc. While there have been many promising progress reported regarding those binders, very little work has been conducted to explore the relationship of the molecular properties and cell performance. In this report, we have attempted to probe one very simple polymer property, molecular weight, using PAA binders and try to correlate this particular variation to the performance of the silicon lithium-ion cells. Specifically, five commercially available PAA samples with various molecular weights ranging from 5k to 450k Dalton were procured. Their molecular weights were further confirmed by gel permeation chromatography (GPC) after pre-treatment of methylation. The viscosity of the lithiated PAA aqueous solutions with a constant concentration of 15% was measured by rheometer. It is overserved that higher molecular PAA sample affords higher viscosity, which is one desirable property for binders. Those binders were then evaluated in silicon half cells, where the anodes were consisted of 10 % lithiated PAA, 73 % Hitachi MagE, 15 % Si nanoparticles (70-130 nm) and 2 % Timcal C45. According to the results, cells with different PAA binders indeed showed different cycling performance. It is noticed that the cells with higher molecular weight PAA binders could afford higher initial specific capacities and better capacity retentions. While the fundamental reason is still not fully understood, we are conducting various characterizations to pursue the insights. DFT calculations are conducted to estimate the conformations, which can be used to estimate the chain length of PAA polymers with different molecular weights. We believe the chain lengths could be an important indicator to quickly evaluate the compatibility to silicon particles with various sizes.