Dear Xiaoman,
The BEAM card is probably sufficient to achieve your desired settings: you
can specify the momentum spread, the x and y beam size, and the divergence
(equal for x and y, which is luckily your situation). Note:

The beam spread is expected as a momentum (p ) spread and not
as a kinetic energy (T) spread. The relationship is
dp = (T+m)/sqrt(T*(T+2m)) dT
Flair offers you the function dT2dp(T,dT,m) to convert from dT to dp
(see section F3.6 of the Flair manual).

Select Gaussian profiles for momentum, size, and divergence spreads.
All spreads are then interpreted as the full width at half maximum of
the Gaussian [note that FWHM=2sqrt(2ln(2))*sigma]. The conversion
can be done via the “fwhm” variable in Flair.
Regarding your emittance wishes: Let x’ be the angle of the initial
particle direction with respect to the x axis. In the xx’ plane, a setup
using a (FWHM) divergence fwhm * θx = 1.17741 mrad, and a (FWHM)
beam size fwhm * σx=1.05967 cm (beam size input is in cm in FLUKA
like all lengths) would populate a 2D Gaussian distribution with your
desired beam size σx and beam divergence θx. A measure of the
(normalized) emittance would be
εx / pi = σx * θx = 2.25 mm mrad (essentially your desired 2.22 mm mrad)
Likewise in yy’ space:
εy / pi = σy * θy = 2.30 mm mrad (essentially your desired 2.37 mm mrad)
If this fits your needs, make sure you are satisfied with the sampled
primaries before proceeding.
Cesc