1
$\begingroup$

I conjecture that in the complex vector space $\mathbb{C}^{2^N \times 2^N}$, where $N$ is a positive integer, there is a basis $\mathcal{B}$ whose elements are Hermitian and unitary. However, I don't know how to prove (or counterprove) it.

Given a number $N$, it is always possible to find $2N$ distinct Hermitian and unitary matrices other than the identity $\{\gamma_1. \cdots. \gamma_{2N}\}$ such that each pair of them anti-commutes: $$\gamma_i \gamma_j + \gamma_j \gamma_i = 0 \ \text{if} \ i \neq j.$$ Physicists usually call those $\gamma$ Majorana operators, and the reason why it is always possible to find $2N$ such matrices is based on the relation between fermionic operators and Majorana operators. For example, if $N = 1$, we define a so-called fermionic annihilation operator as $$d = \begin{bmatrix} 0 & 1\\ 0 & 0 \end{bmatrix}. $$ Then the matrix $$\gamma_1 = d + d^\dagger = \begin{bmatrix} 0 & 1\\ 1 & 0 \end{bmatrix} $$ and the matrix $$\gamma_2 = i(d^\dagger - d) = \begin{bmatrix} 0 & -i\\ i & 0 \end{bmatrix} $$ are Majorana operators.

The set of $2N$ Majorana operators can be extended to a set of $4^N$ Hermitian and unitary matrices: $$\mathcal{B}=\left\{i^{\frac{n(n-1)}{2}} (\gamma_{a_1} \cdots \gamma_{a_n}): \ n, a_i \in \{1,\cdots,2N\}, \, a_i < a_{i+1}\right\} \bigcup \{1\}.$$ For instance, when $N = 1$, $$\mathcal{B}=\left\{1, \gamma_1,\gamma_2, i \gamma_1 \gamma_2\right\}.$$

A direct calculation shows $$|\mathcal{B}| = 1 + \sum_{n=1}^{2N} \begin{pmatrix} 2N\\n \end{pmatrix} = \sum_{n=0}^{2N} \begin{pmatrix} 2N\\n \end{pmatrix} = 2^{2N} = 4^N = \dim\left(\mathbb{C}^{2^N \times 2^N}\right).$$ I feel this set $\mathcal{B}$ is a basis for the complex vector space. Indeed, it is true at least when $N = 1$: $\mathcal{B}$ is the set of Pauli matrices. Nevertheless, I don't know how to prove it in higher dimensional spaces.

Thanks in advance for your help. :)

$\endgroup$
0
$\begingroup$

I worked it out myself. This is the link to my proof.

$\endgroup$

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.