Higgins, Daniel Ronan
Advanced optical calibration of the Herschel HIFI heterodyne spectrometer.
PhD thesis, National University of Ireland Maynooth.
The Heterodyne Instrument for the Far-Infrared (HIFI) was launched aboard the Herschel
Space Observatory on the 14th of May 2009. HIFI's frequency range is spread over 7 mixer
bands. Bands 1-5 (480-1270 GHz) use Superconducting-Insulator-Superconducting (SIS) mixer
technology while bands 6 & 7 (1410-1910 GHz) use Hot Electron Bolometer (HEB) mixer
technology. HIFI is a double side band instrument and hence contains both the upper and
lower side band of the down converted sky signal. The gain in the upper and lower side band
is not always equal. This effect introduces a calibration uncertainty that must be understood
in order to achieve the HIFI calibration goal of 3%.
To determine the frequency dependent side band ratio for each mixer band, a gas cell test
set up was developed. During the instrument level testing a number of simple (12eo, 13eo and
OeS) and complex (CHgCN and CH30H) molecules were observed using the HIFI instrument.
Using a radiative transfer model with the measured pressure and optical path length of the gas
cell and molecular line parameters taken from the JPL and HITRAN catalogs, model spectra
can be generated. By comparing the generated spectra with the observed spectra the side band
gain can be determined.
In order to extract the side band ratio a number of additional instrumentation effects must
first be understood. Bands 1 and 2 shows good performance and the side band ratio was
extracted for these bands. The data showed good agreement with predicted antenna response.
Bands 3 and 4 had significant IF gain effects due to a diplexer mistuning problems. Band 5
had spurious LO signals at certain frequencies making the data observed unusable. Bands 6 &
7 had significant standing wave issues due to a non-optimal IF chain design. These instrument
effects are discussed in detail in this thesis.
The final part of the thesis details the first step in the analysis of the methanol (CHgOH)
data. This dataset makes up 80% of all data taken during the gas cell test campaign. Unfortunately
the pressure broadening (variation of line width with pressure) was not known before the
analysis. This was extracted from the data using a 2 parameter fit where the side band ratio
was the other free variable. This approach was shown to be problematic with multiple viable
solutions possible. It was proposed to use the analysis of the simpler molecules to constrain the
pressure broadening extraction process. The methanol pressure broadening parameters were
extracted from the data and showed some frequency dependence. The values extracted were in
reasonable agreement with the values predicted in other publications. A first pass extraction
of the side band ratio was undertaken at the lower edge of band 2, there was a large degree of
scatter in the measured data but an encouraging agreement was seen with the side band ratio
values extracted from the other simpler molecules.
Repository Staff Only(login required)
||Item control page