At the center of every galaxy there is a super-massive black hole of a million or more solar masses. In most galaxies the presence of this black hole can only be detected through its gravitational attraction, which affects the motion of nearby stars. However, in about 10% of the galaxies the super-massive black hole is the engine of one of the most luminous phenomena in the universe: an active galactic nucleus (AGN). In the local universe there are two types of AGN: ‘Radiative-mode’ and ‘Jet-mode’ AGN. In this thesis I show that these two AGN types are hosted by different galaxies and have different infrared properties. ‘Radiative-mode’ AGN are the ‘classical’ AGN which are bright emitters across the entire electromagnetic spectrum. They are thought to be powered by a super-massive black hole accreting matter at a high rate. I show that ‘radiative-mode’ AGN are predominantly found in intermediate mass galaxies with blue and green optical colors. These colors are indicative of active or recently terminated star formation and a young stellar population. Due to the presence of torus of hot dust near the black hole, galaxies with a ‘radiative-mode’ AGN typically show an excess of mid-infrared emission. ‘Jet-mode’ AGN lack the bright optical emission and excess infrared emission of ‘jet-mode’
AGN and can only be identified by means of their radio jet – a stream of relativistic particles that can reach far outside the AGN’s host galaxy. This absence of electromagnetic radiation and prominence of the radio jet is thought to be the result of the low accretion rate of the super-massive black hole driving this AGN type. I show that ‘jet-mode’ AGN have a strong preference for the most massive galaxies, which typically have little star formation. The presence or absence of a dusty torus and the resulting difference in broadband mid-infrared emission could be a powerful tool to separate ‘radiative-mode’ and ‘jet-mode’ AGN without using spectroscopy. Unfortunately, the inherent scatter in the mid-infrared emission of galaxies due to dust heated by stars is too large to separate the two populations reliably.
Far-infrared observations could help resolve this, by constraining the mid-infrared contribution of dust heated by stars. However, current far-infrared surveys do not have the depth or the area to give the number statistics required to calibrate this procedure. In this thesis I have investigated the properties of microwave kinetic inductance detectors. These detectors will enable the far-infrared instruments with 10.000 pixels as a result of their inherent potential for frequency domain multiplexing. This is a huge leap from the 100 pixel far-infrared instruments currently on telescopes. I have shown that microwave kinetic inductance detectors made from NbTiN and Al can satisfy all the requirements to enable a new generation of large format far-infrared cameras, which are required to constrain the far-infrared emission of many galaxies.

This thesis reports on progress in various Nikhef solutions for the alignment of the proposed Compact Linear Collider. Optical three-point-alignment systems are proposed for the small (60 mm), intermediate (2 m) and large (200 m) alignment distances. A projective alignment system, called Rasnik and consisting of a back-illuminated chess _eld mask, a lens and an image sensor, is used for the small and intermediate distances. An alignment system that uses diffraction for image generation, called Rasdif and consisting of a diode laser, a diffraction plate and an image sensor, is used on the intermediate and large distances. A 4 m set-up at Nikhef and a 140 m set-up at CERN were used for validation of the performance of the optical alignment systems for the different alignment distances. A 0.5-1 _m resolution is reported for thermally shielded 4 m Rasniks and Rasdifs in open air and a 1.5-2 _m resolution is reported for a 140 m Rasdif in a 1.5 _ 10􀀀4 mbar vacuum. A Rasdif anomaly was discovered during preliminary measurements on the 4 m set-up: the virtual departure of the spherical waves of light, the point where the waves seem to be coming from, did not coincide with the physical location of the laser. The translation of the image is governed by the translation of this virtual point, whereas governance by the physical point is desired. A Diode Pumped Solid State (DPSS) laser came out as best of measurements that searched for laser with a near virtual point. Translational linearity tests on the 4 m set-up pin the virtual point of departure about 4 cm behind the these lasers. Translational measurements of the 140 m Rasdif show great discrepancies in two separate measurements and this is subject to future investigation. Cramér-Rao Lower Bound (CRLB) analysis on simulated Rasdif images show the current diffraction plates, with an outer radius of 1.44 mm, have to be scaled to a design with an outer radius of 2.55 mm to obtain images with more gradient energy. For the alignment of the _nal focusing magnets of the collider, position information about the location of these magnets is radially transferred to non-near-vertex locations by Zerodur spokes. The ends of these spokes are then aligned by Rasniks and/ or Rasdifs. Preliminary results of a 1 m Zerodur spoke set-up confirm the zero coffcient of expansion and Young's modulus of the material. Contact repeatability for one end of Zerodur, with a steel semi-sphere glued to it, and a reference block is measured to be within 150-200 nm by a 60 mm Rasnik attached to the other side of the spoke.