Definition

Let AA be a . Its state space, denoted S(A)S(A), is the set of all on AA:

S(A)={φA:φ0, φ=1}.S(A)=\{\varphi\in A^*:\varphi\geq 0,\ \|\varphi\|=1\}.

It is a in the Banach dual AA^*, equipped most often with the . If AA is unital, the norm condition is equivalent to φ(1A)=1\varphi(1_A)=1, and S(A)S(A) is weak-star compact. For nonunital AA, S(A)S(A) need not be weak-star compact, although the of norm at most one form a weak-star .

Convex structure

A state is exactly when it is an extreme point of S(A)S(A). Thus convex decompositions express mixed states as combinations of other states. The Krein–Milman theorem implies that, for unital AA, the compact convex set S(A)S(A) is the closed of its pure states. This convex viewpoint is developed systematically in Alfsen and Shultz, Part I.

Standard examples

For A=C(X)A=C(X) with XX compact Hausdorff, S(A)S(A) is the space of Radon on XX; its pure states are the point masses. For A=Mn(C)A=M_n(\mathbb C), states correspond to positive matrices ρ\rho with Tr(ρ)=1\operatorname{Tr}(\rho)=1 through φ(a)=Tr(ρa)\varphi(a)=\operatorname{Tr}(\rho a). The pure states correspond to rank-one ρ\rho.

Topological caveat
References
  1. Gerard J. Murphy, C-Algebras and Operator Theory*, Academic Press, 1990. DOI record. Relevant: §3.2 on states, convexity, and weak-star compactness.
  2. Erik M. Alfsen and Frederic W. Shultz, State Spaces of Operator Algebras: Basic Theory, Orientations, and C-Products*, Birkhäuser, 2001. DOI record. Relevant: Part I on compact convex state spaces.