
A Weyl semimetal (WSM) is a topological material that hosts Weyl fermions as
quasiparticles in the bulk. We study the effects of diagonal and offdiagonal
disorders on WSMs by adopting a tightbinding model that supports the WSM, the
threedimensional quantum anomalous Hall (3D QAH) and the normal insulator (NI)
phases in the clean limit. Basing on the calculation of the localization length
and the Hall conductivity, we present rich phase diagrams when WSMs are subject
to the two types of disorders. We find that the WSM with wellseparated Weyl
nodes is stable against weak disorder. Weak diagonal disorder may annihilate
the Weyl nodes in the WSM phase close to the 3D QAH phase, resulting in a 3D
QAH state, and it could also create Weyl nodes in the NI phase near the WSM
phase, leading to a WSM. Reversely, weak offdiagonal disorder may nucleate
Weyl nodes in the 3D QAH phase in proximity to the WSM phase, causing a 3D
QAHWSM transition, or it could destroy Weyl nodes in the WSM phase near the NI
phase, bringing about a WSMNI transition. We observe a diffusive anomalous
Hall metal (DAHM) phase at moderate disorder strength. The DAHM appears in a
wide range of the phase diagram when the diagonal disorder dominates over the
offdiagonal disorder, while the DAHM is found to exist in a narrow region or
be missing in the phase diagram when the offdiagonal disorder dominates.

We utilize variational method to investigate the Kondo screening of a
spin1/2 magnetic impurity in tilted Dirac surface states with the Dirac cone
tilted along the $k_y$axis. We mainly study about the effect of the tilting
term on the binding energy and the spinspin correlation between magnetic
impurity and conduction electrons, and compare the results with the
counterparts in a two dimensional helical metal. The binding energy has a
critical value while the Dirac cone is slightly tilted. However, as the tilting
term increases, the density of states around the Fermi surface becomes
significant, such that the impurity and the host material always favor a bound
state. The diagonal and the offdiagonal terms of the spinspin correlation
between the magnetic impurity and conduction electrons are also studied. Due to
the spinorbit coupling and the tilting of the spectra, various components of
spinspin correlation show very strong anisotropy in coordinate space, and are
of powerlaw decay with respect to the spatial displacements.

We utilize the variational method to study the Kondo screening of a
spin$1/2$ magnetic impurity in a threedimensional (3D) Weyl semimetal with
two Weyl nodes along the $k_z$axis. The model reduces to a 3D Dirac semimetal
when the separation of the two Weyl nodes vanishes. When the chemical potential
lies at the nodal point, $\mu=0$, the impurity spin is screened only if the
coupling between the impurity and the conduction electron exceeds a critical
value. For finite but small $\mu$, the impurity spin is weakly bound due to the
low density of state, which is proportional to $\mu^2$, contrary to that in a
2D Dirac metal such as graphene and 2D helical metal where the density of
states is proportional to $\mu$. The spinspin correlation function
$J_{uv}(\mathbf{r})$ between the spin $v$component of the magnetic impurity at
the origin and the spin $u$component of a conduction electron at spatial point
$\mathbf{r}$, is found to be strongly anisotropic due to the spinorbit
coupling, and it decays in the powerlaw. The main difference of the Kondo
screening in 3D Weyl semimetals and in Dirac semimetals is in the spin $x$
($y$) component of the correlation function in the spatial direction of the
$z$axis.

We study localized ingap states and quantum spin Hall effect in Sidoped
InAs/GaSb quantum wells. We propose a model describing donor and/or acceptor
impurities to describe Si dopants. This model shows ingap bound states and
wide conductance plateau with the quantized value $2e^2/h$ in light dopant
concentration, consistent with recent experiments by Du et al. We predict a
conductance dip structure due to backward scattering in the region where the
localization length $\xi$ is comparable with the sample width $L_y$ but much
smaller than the sample length $L_x$.