Feasibility of the Alcock Paczynski (AP) test by stacking voids in the 21cm
line intensity field is presented. We analyze the Illstris-TNG simulation to
obtain the 21cm signal map. We then randomly distribute particles depending on
the 21cm intensity field to find voids by using publicly available code, VIDE.
As in the galaxy clustering, the shape of the stacked void in the 21cm field is
squashed along the line of sight due to the peculiar velocities in
redshift-space, although it becomes spherical in real-space. The redshift-space
distortion for the stacked void weakly depends on redshift and we show that the
dependency can be well described by the linear prediction, with the amplitude
of the offset being free parameters. We find that the AP test using the stacked
voids in a 21cm intensity map is feasible and the parameter estimation on
$Ømega_m$ and $w$ is unbiased.
The Alcock Paczynski test with 21cm intensity field