Von Neumann Unitaries & Their Properties

• Von Neumann unitaries are the unitary elements in a von Neumann algebra, forming a topological group where conjugation coefficient functions reveal key dynamical properties.
• They exhibit weak and almost periodic behaviors that characterize finiteness and direct-sum decompositions through invariant means and norm compactness.
• Products of symmetries in von Neumann algebras illustrate factorization phenomena, with type II₁ cases requiring six or four factors depending on the spectral properties.

Von Neumann unitaries are the unitary elements of a von Neumann algebra $M\subset B(H)$, organized either as the topological group

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$

or through special subclasses such as symmetries, namely self-adjoint unitaries. Two complementary lines of analysis are particularly prominent: periodicity properties of conjugation coefficient functions $u\mapsto \langle uxu^*\xi\mid\eta\rangle$, which characterize finiteness and almost-periodicity phenomena of $M$, and factorization properties expressing unitaries as products of symmetries, especially in type $II_1$ von Neumann algebras (Jolissaint, 2020, Bhat et al., 2022).

1. The unitary group as a topological object

Let $M\subset B(H)$ be a von Neumann algebra acting on a Hilbert space $H$. Its unitary group $U_M$ is considered as a topological group under either the weak operator topology or the strong operator topology, and these topologies agree on $U_M$. A net $u_i\to u$ in the weak operator topology satisfies

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$0

whereas a net $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$1 in the strong operator topology satisfies

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$2

These are Polish group topologies on $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$3 (Jolissaint, 2020).

A basic family of functions on $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$4 is given by the conjugation coefficient functions

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$5

defined for $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$6 and $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$7. When $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$8, these functions encode the orbit of $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$9 under the inner automorphism action $u\mapsto \langle uxu^*\xi\mid\eta\rangle$0. In the setting of von Neumann algebras, this orbit structure becomes a bridge between operator-algebraic properties of $u\mapsto \langle uxu^*\xi\mid\eta\rangle$1 and dynamical properties of $u\mapsto \langle uxu^*\xi\mid\eta\rangle$2 as a topological group.

The same unitary group also supports a second, more algebraic viewpoint. In any von Neumann algebra $u\mapsto \langle uxu^*\xi\mid\eta\rangle$3, a symmetry is a self-adjoint unitary $u\mapsto \langle uxu^*\xi\mid\eta\rangle$4 with $u\mapsto \langle uxu^*\xi\mid\eta\rangle$5 and $u\mapsto \langle uxu^*\xi\mid\eta\rangle$6. Products of such symmetries generate large parts of the unitary group, and in type $u\mapsto \langle uxu^*\xi\mid\eta\rangle$7 algebras they admit explicit finite-length factorization results (Bhat et al., 2022).

2. Weakly almost periodic and almost periodic coefficient functions

For a topological group $u\mapsto \langle uxu^*\xi\mid\eta\rangle$8, let

$u\mapsto \langle uxu^*\xi\mid\eta\rangle$9

and let $M$0 denote the subalgebra of right-uniformly-continuous functions. A function $M$1 is weakly almost periodic if its left orbit $M$2 is relatively weakly compact in the Banach space $M$3, and it is almost periodic if that orbit is norm-relatively compact. Equivalently, weak almost periodicity may be tested by the right orbit or by Grothendieck’s double-limit criterion: $M$4 The space $M$5 is a unital $M$6-subalgebra of $M$7, invariant under left and right translations (Jolissaint, 2020).

A crucial structural fact is the existence of a unique invariant mean on $M$8. There is a unique state

$M$9

such that $II_1$0 if $II_1$1, $II_1$2, and

$II_1$3

Moreover, $II_1$4 lies in the closed convex hull of the left orbit of $II_1$5, and likewise of the right orbit. For $II_1$6, this invariant mean on $II_1$7 is the main device in the analysis of conjugation coefficients and the operator systems they generate (Jolissaint, 2020).

In this framework, the distinction between weak almost periodicity and almost periodicity is substantive rather than terminological. Weak compactness of orbits is sufficient to recover finite-type structure through invariant averaging, whereas norm-compactness forces a significantly stronger decomposition of the underlying algebra.

3. Finiteness characterized by weakly almost periodic conjugation

A central theorem identifies finiteness of a von Neumann algebra with weak almost periodicity of all conjugation coefficient functions. For $II_1$8, the following are equivalent:

1. $II_1$9 is finite, i.e. it admits a faithful normal center-valued trace.
2. For every $M\subset B(H)$0 and $M\subset B(H)$1, the coefficient function

$M\subset B(H)$2

belongs to $M\subset B(H)$3 (Jolissaint, 2020).

When these conditions hold, the invariant mean yields a completely positive, unital, $M\subset B(H)$4-bimodular conditional expectation

$M\subset B(H)$5

whose restriction to $M\subset B(H)$6 is precisely the center-valued trace of $M\subset B(H)$7. Thus the periodicity of scalar-valued coefficient functions on the unitary group recovers a canonical operator-algebraic averaging map.

The proof mechanism has two directions. If every $M\subset B(H)$8 has weakly almost periodic coefficients, then $M\subset B(H)$9 sits inside the operator system $H$0 of operators whose coefficients lie in $H$1, and the invariant mean defines

$H$2

Restricted to $H$3, this map is a unital $H$4-bimodule map satisfying $H$5, which forces $H$6 to agree with the unique center-valued trace. Conversely, if $H$7 is finite, then the group of inner automorphisms

$H$8

is relatively weak$H$9-compact in the Banach space $U_M$0, and Grothendieck’s criterion shows that each conjugation coefficient lies in $U_M$1 (Jolissaint, 2020).

This theorem places finiteness in a dynamical form: it is detected not by traces alone, but by weak compactness of the scalar orbit data generated by conjugation under unitaries.

4. Almost periodicity, direct-sum structure, and minimal almost periodicity

The almost-periodic analogue is more restrictive. For $U_M$2, the following are equivalent:

1. For every $U_M$3 and $U_M$4, the coefficient function

$U_M$5

lies in $U_M$6.

2. $U_M$7 decomposes as a direct sum

$U_M$8

where $U_M$9 is a diffuse abelian von Neumann algebra and each $U_M$0 is a finite-dimensional factor (Jolissaint, 2020).

Equivalently stated in the source, all conjugation-coefficient functions are almost periodic if and only if no infinite faithful atomic part of $U_M$1 remains. The proof again proceeds through an expectation on the operator system of almost-periodic operators $U_M$2. If $U_M$3, then all central summands of type $U_M$4, type $U_M$5, or type $U_M$6 must vanish, and on the atomic type $U_M$7 part one recovers only finite-dimensional blocks. Conversely, for

$U_M$8

with $U_M$9 abelian and diffuse, each coefficient function is a uniform limit of matrix coefficients of finite-dimensional unitary representations of $u_i\to u$0, hence is almost periodic (Jolissaint, 2020).

A related but distinct notion is minimal almost periodicity. A topological group $u_i\to u$1 is minimally almost periodic if its only continuous finite-dimensional irreducible unitary representation is the trivial one; equivalently, $u_i\to u$2. If $u_i\to u$3 is a diffuse von Neumann algebra, then $u_i\to u$4 is minimally almost periodic. The proof proceeds by showing first that a diffuse maximal abelian subalgebra $u_i\to u$5 has unitary group $u_i\to u$6 with no non-trivial characters, and then passing from the abelian case to general diffuse $u_i\to u$7 by restriction to maximal abelian subalgebras (Jolissaint, 2020).

These statements isolate two different levels of periodicity. The source material implies that the condition $u_i\to u$8 for diffuse $u_i\to u$9 concerns finite-dimensional unitary representations of the group itself, whereas the stronger requirement that all conjugation-coefficient functions be almost periodic forces a direct-sum decomposition with a diffuse abelian part and finite-dimensional factors.

5. Products of symmetries in von Neumann algebras

In a von Neumann algebra $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$00, the set of symmetries is

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$01

and for $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$02,

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$03

For type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$04 von Neumann algebras, the factorization theory is especially explicit (Bhat et al., 2022).

Setting	Conclusion	Citation
$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$05 type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$06, arbitrary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$07	Every unitary is a product of six symmetries	(Bhat et al., 2022)
$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$08 type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$09, $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$10 with finite spectrum	Every such unitary is a product of four symmetries	(Bhat et al., 2022)
$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$11 type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$12	$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$13	(Bhat et al., 2022)
Arbitrary von Neumann algebra $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$14	$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$15	(Bhat et al., 2022)

The failure at three factors is not merely a failure of exact generation. The source exhibits a non-empty open set of unitaries—such as those whose spectrum sits inside one of the four open arcs of the circle obtained by removing $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$16—none of which can be written as a product of three symmetries. Thus the three-symmetry obstruction is topologically stable in the norm topology (Bhat et al., 2022).

This produces a sharp asymmetry between four and three factors in the type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$17 setting: four symmetries are already norm-dense, while three are never norm-dense in any von Neumann algebra.

6. Decomposition mechanisms and classical placement

The six-symmetry theorem for type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$18 algebras is proved by an infinite block-diagonal cutting argument. In a maximal abelian subalgebra $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$19 containing a unitary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$20, one chooses mutually orthogonal projections

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$21

where $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$22 is the center-valued trace. On each block $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$23, one peels off four symmetries $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$24, together with a decomposition

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$25

satisfying

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$26

and a unitary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$27 on the remaining piece, so that

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$28

The remainders $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$29 are then reassembled into a single unitary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$30, and $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$31 is itself a product of two symmetries, yielding a total of six (Bhat et al., 2022).

The finite-spectrum four-symmetry theorem is obtained by diagonalizing

$U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$32

in a maximal abelian von Neumann algebra containing $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$33, then showing that each scalar unitary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$34 is a product of four symmetries in the corresponding corner. Norm-density of $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$35 follows by writing any $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$36 as $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$37 for some self-adjoint $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$38 with spectrum in $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$39, approximating $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$40 in norm by finite-spectrum self-adjoints $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$41, and setting $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$42 with $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$43 in norm. The obstruction at three factors is linked to the characterization of $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$44: a unitary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$45 is a product of two symmetries if and only if $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$46 is unitarily equivalent to its adjoint $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$47, so $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$48 must be symmetric under complex conjugation about the real axis (Bhat et al., 2022).

The type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$49 results are situated in an existing classical picture. Halmos–Kakutani showed that in $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$50, a type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$51 factor, every unitary is a product of four symmetries and that three symmetries cannot generate all unitaries. Fillmore later extended the four-symmetry result to all properly infinite and type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$52 von Neumann algebras. Broise, and later Dowerk–Thom, showed that in a $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$53-factor every unitary is a product of finitely many symmetries; the Dowerk–Thom bound was $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$54 factors. The six-symmetry theorem sharpens this to six in arbitrary $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$55 von Neumann algebras, while the finite-spectrum bound drops to four. The resulting summary given in the source is:

• Type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$56, $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$57, $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$58, $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$59: four symmetries suffice.
• Type $U_M=\{\,u\in M\mid u^*u=uu^*=1\,\}$60: six in general; four for finite-spectrum unitaries; four are already norm-dense; three never suffice (Bhat et al., 2022).

Taken together, these results present von Neumann unitaries as objects with simultaneously topological, representation-theoretic, and factorization-theoretic structure. Weak and norm compactness of conjugation orbits detect finiteness and direct-sum structure, while products of symmetries quantify how unitary groups in different von Neumann types are generated and approximated.