The Russo–Dye theorem is a result in functional analysis concerning the structure of the unit ball in a unital C*-algebra. It asserts that the closed unit ball of a unital C*-algebra is the closed convex hull of its unitary elements.
Statement
Let $A$ be a unital C*-algebra with unit $1$. Denote by
$$
\mathcal{U}(A)={u\in A : u\text{ is unitary, } uu^{}=u^{}u=1}
$$
the set of unitary elements of $A$. Then
$$
{a\in A : |a|\le 1}
= \overline{\operatorname{co}}\bigl(\mathcal{U}(A)\bigr),
$$
where $\operatorname{co}$ denotes the convex hull and the overline indicates norm‑closure.
Historical background
The theorem is named after B. Russo and H. A. Dye, who proved it in 1966. Their original paper, “A note on unitary operators in C*-algebras,” appeared in Duke Mathematical Journal and established the result for general unital C*-algebras, extending earlier observations for specific algebras such as $B(H)$, the bounded operators on a Hilbert space.
Proof outline
A standard proof proceeds by showing that any element $a$ with $|a|\le 1$ can be approximated in norm by finite convex combinations of unitaries. The key steps are:
- Polar decomposition: For any $a$ with $|a|\le 1$, write $a = v|a|$ where $v$ is a partial isometry.
- Functional calculus: Approximate the positive contraction $|a|$ by convex combinations of projections, and then lift these projections to unitaries using the functional calculus on the C*-algebra.
- Approximation: Combine the approximations to obtain a convex combination of unitaries that converges to $a$ in norm.
The argument relies on the spectral properties of elements in a C*-algebra and on the fact that the unitary group is norm‑dense in the set of invertible elements of norm one.
Consequences and applications
- Numerical radius: The theorem implies that, for any element $a$ in a unital C*-algebra, the numerical radius satisfies $w(a)=\sup{|\phi(a)| : \phi\text{ is a state on }A}$. This relationship is often used in norm estimates.
- Structure of the unit ball: It provides a geometric description of the unit ball, facilitating the study of extreme points, faces, and the convex geometry of operator algebras.
- Operator theory: In $B(H)$, the theorem yields that every contraction can be approximated by convex combinations of unitary operators, a fact employed in dilation theory and the study of completely positive maps.
- Banach space theory: The result is applied to characterize norm‑preserving linear maps between C*-algebras and to investigate isometries.
Related results
- Kadison’s transitivity theorem – deals with the density of unitary orbits in certain contexts.
- Wigner’s theorem – describes symmetries of the set of unitary operators in quantum mechanics, though in a different setting.
- Bishop–Phelps theorem – concerning the density of norm‑attaining functionals, which shares a convex‑geometric perspective.
References
- B. Russo and H. A. Dye, “A note on unitary operators in C*-algebras,” Duke Mathematical Journal, vol. 33, no. 2, pp. 413–421, 1966.
- G. J. Murphy, C*-Algebras and Operator Theory, Academic Press, 1990.
- J. B. Conway, A Course in Operator Theory, Graduate Studies in Mathematics, Vol. 79, American Mathematical Society, 2000.
- M. Takesaki, Theory of Operator Algebras I, Springer, 2002.
