The Chromospheric Lyman-Alpha SpectroPolarimeter (CLASP) is a sounding rocket instrument aiming to measure for the first time the polarization signal of the hydrogen Lyman-Alpha line (121.567nm) emitted in the solar upper-chromosphere and lower-transition region. Anisotropic radiation in the solar atmosphere produces atomic level population imbalance, which creates sizeable linear polarization signal. Due to the Hanle effect, the linear polarization from such scattering processes is sensitive to magnetic field between 10G to 250G for the Lyman-Alpha line. CLASP will use the measurement of the linear polarization to retrieve the magnetic field strength and orientation of the solar chromosphere. The polarization accuracy for measuring these linear polarization amplitude, expected from simulations to be below 1%, is a crucial part of CLASP requirement, with the spurious polarization being especially important since the VUV continuum around the Lyman-Alpha line cannot be used to calibrate it.
Due to these scientific requirements, CLASP needs a 0.1% polarization accuracy in order to resolve the Hanle effect. Reaching such high polarization accuracy requires an unprecedented 0.01% sensitivity on the spurious polarization measurement which was achieved with an extensive polarization calibration of the spectro-polarimeter part. A specially designed Lyman-Alpha light source composed of a Deuterium lamp, an MgF2 converging lens and two polarizers with highly reflective multilayer coating was used to illuminate the slit with an almost perfectly linearly polarized Lyman-Alpha light and with a F-number similar to the telescope. Two independent methods were used to change the linear polarization input to the four Stokes parameters (+Q, +U, -Q and -U) needed to determine the instrument response matrix: by rotating the entire light-source and its vacuum chamber, or by using a half-waveplate located after the light-source's analysers instead.
As a result, all the parameters of the response matrix were measured with the required accuracy for both channel independently. It was also observed that the instrument shows a very small inherent spurious polarization at the 0.01% order when using both channel combined. These parameters were measured at three locations along the slit (top, center and bottom) to confirm the spatial dependency of the response matrix. In addition, the polarization calibration will be confirmed in-flight by measuring the Lyman-Alpha polarization at the disk center during 10s at the beginning of the flight. The atomic polarization at disk center is expected to be close to zero because of symmetry by averaging spatially pixels along the slit.