1. This follows immediately from the Great Orthogonality Theorem. 2. Now $E^{jk}e_l=\d_{kl}e_j$, i.e. the matrix of $E^{jk}$ with respect to the basis $e_1,\ldots,e_n$ has just one non-zero entry: the $(j,k)$ entry, which is $1$. Hence we get $$ P_{ex}(e_l\otimes e_m) =\sum_{j,k}E^{kj}e_l\otimes E^{jk}e_m =\sum_{j,k}\d_{jl}\d_{km}e_k\otimes e_j =e_m\otimes e_l $$ and therefore $P_{ex}(x\otimes y)=y\otimes x$. Obviously $P_{ex}^2=1$ and $P_{ex}^*=P_{ex}$, for $$ \la P_{ex}(x\otimes y),u\otimes v\ra =\la y,u\ra\la x,v\ra =\la x\otimes y,P_{ex}(u\otimes v)\ra, $$ i.e. $P_{ex}$ is unitary and the only possible eigen-values are $\pm1$. For $A\in\Hom(E)$ and $x,y\in E$ we have: $$ (A\otimes A)P_{ex}(x\otimes y) =Ay\otimes Ax =P_{ex}(A\otimes A)(x,y) $$ 3. Finally for any $A\in\Hom(E)$ we have: \begin{eqnarray*} \tr_1((A\otimes1)P_{ex}) &=&\sum_{j,k}\tr(AE^{kj})\otimes E^{jk}\\ &=&\sum_{j,k,l,m}(a_{lm}\d_{mk}\d_{jl})E^{jk} =\sum_{j,k}a_{jk}E^{jk} =A~. \end{eqnarray*}