We construct the physics concepts network in which the physics concepts is considered as nodes. We find that this network is clearly not a random network but a scale-free network in which the distribution P(k) of the degree k follows a power law, P(k) similar to k(-gamma) with exponent gamma approximate to 2.41. The relevance between the physics concepts and the important concepts are studied by measuring the degree, the betweenness centrality, and the relevance strength and we find that the energy concept is most important. Also we find that the relevance strength s(k) as a function of the degree k exhibits a power-law behavior, s(k) similar to k(alpha) with the exponent alpha approximate to 0.15 and s(k(nn)) as a function of the average neighbor's degree k(nn) follows a power-law behavior, s(k(nn)) similar to k(nn)(delta) with the exponent delta approximate to 0.17 for small k(nn). We also measured the other quantities which describe the topological properties of the network and find that it has the hierarchical property and tree-like patterns.