We study bipartite entanglement in a general one-particle state, and find that the linear entropy, quantifying the bipartite entanglement, is directly connected to the paricitpation ratio, charaterizing the state localization. The more extended the state is, the more entangled the state. We apply the general formalism to investigate ground-state and dynamical properties of entanglement in the onedimensional Harper model. [5]. On the other hand, entanglement has been proved to be playing an important role in condensed matter physics. There are many studies on entanglement in the Heisenberg spin models [6,7,8], Ising models in a transverse magnetic field [9,10], and related itinerant fermionic systems [11]. In the context of quantum phase transition, entanglment is also an indicator of the transition which can not be captured by common statistical physics [12,13].Recently, pairwise entanglement sharing in one-particle states was studied [14] using the concurrence [15] in the Harper model [16]. Here, we study another type of entanglement of one-particle states, the bipartite entanglement, which refers to entanglement between two subsystems when a whole system is divided into two parts. We reveal that the average bipartite entanglement directly connects to state localization.The one-particle states permeate many physics systems. For examples, for one electron moving on a substrate potential, the eigenfunctins are one-particle states. In quantum spin chain models with only one spin up (down) and all other spins down (up), the eigenfuntions of the model are one-magnon states.We consider a system containing N two-level systems (qubits) with |0 being the excited state and |1 the ground state. A general one-particle state is then written asHere, {|ψ n | 2 } is a probability distribution, satisfying the normalization condition N n=1 |ψ n | 2 = 1. When |ψ n | = 1/ √ N , state |Ψ reduces to the W state [8,17], one representative state in quantum information theory. We now consider bipartite entanglement between a block of L qubits and the rest N − L qubits. Bipartite entanglement of a pure state can be measured by the linear entropy of reduced density matrices [18].where ρ i is the reduced density matrix for subsystem i.To calcualte bipartite entanglement, we first consider a simple situation, namely, the entanglement between the first qubit and the rest N − 1 qubits. The one-particle state can be written in the following form: