A bounded operator $A$ in a Hilbert space is called nilpotent if there exists $n$ such that $A^{n}=0$. An operator is called quasi-nilpotent iff $$ \limsup_{n\to\infty}{ \|A^{n}\|^{1/n}}=0. $$

Every nilpotent operator is clearly quasi-nilpotent. The family of quasi-nilpotent operators is very important for the hyperinvariant subspace problem. For instance, it was proved by Haagerup and Schultz, that if the Brown measure of an operator in a $\mathrm{II}_1$-factor is concentrated in more than one point then it has a non-trivial hyperinvariant subspace. A subspace is called $A$-hyperinvariant if it is invariant for all the operators that commute with $A$.

I seem to recall from Herrero's book that the norm closure of the nilpotent and quasi-nilpotent operators in $B(H)$ is pretty well understood. However, I don't have a copy with me at the moment. My question is: Is it known what is the norm closure of the nilpotent operators or/and quasi-nilpotent operators in the hyperfinite $\mathrm{II}_1$-factor? In any other $\mathrm{II}_1$-factor?



The definition of quasi-nilpotent is that the spectrum of $A$ is zero, which is the same as $\|A^n\|^{1/n} \to 0$.

See Apostol's paper On the norm-closure of nilpotents, Rev. Roumaine Math. Pures Appl. 19 (1974), 277-282 or the later paper Apostol, C.; Foiaş, C.; Pearcy, C. That quasinilpotent operators are norm-limits of nilpotent operators revisited. Proc. Amer. Math. Soc. 73 (1979), no. 1, 61–64 for info about the closure of the nilpotent operators. For recent papers giving info about hyperinvariant subspaces for quasi-nilpotent operators, see recent papers by Foias, Pearcy, et al.

  • $\begingroup$ Oops! That should have been a comment rather than a remark, but I'll leave it as is. $\endgroup$ Apr 13 '11 at 20:32
  • $\begingroup$ Thanks Bill! I corrected. These papers focus on the $B(H)$ case right? I'm more interested in the $\mathrm{II}_{1}$ factor situation, do you know any references there? $\endgroup$
    – ght
    Apr 13 '11 at 21:11
  • $\begingroup$ Certainly their focus was on $B(H)$, Gabriel. I did not study the papers to see what all they did. Have you tried to check on math sci-net? $\endgroup$ Apr 13 '11 at 22:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.