Angular resolution improvement of slumped thin glass optics for x-ray telescopes.

Current X-ray telescopes such as Chandra and XMM were designed and built to privilege the angular resolution or the collecting area respectively, but using two different manufacturing techniques, and these two key aspects could not be optimised at the same time. The result is that the high-redshift X-ray Universe is still too unresolved to provide us with the understanding of the time evolution of the universe as we see it nowadays. Now the ATHENA mission, approved for the L2 slot in the Cosmic Vision program, is required to have an effective area of 2 m2 at 1 keV and an angular resolution better than 5 arcsec Half-Energy Width (HEW). At this regard, in addition to the baseline solution based on Silicon Pore Optics (SPO), thin glass foils are considered a viable alternative. Slumped glass foils are also considered as substrates for X-ray optics being developed in other projects; for example, the X-ray Surveyor mission being developed at SAO/CfA in USA, and the AXYOM project being developed in Italy. They both foresee the use of piezo-electric actuators to correct the shape errors of the mirrors. The Brera Astronomical Observatory (INAF-OAB, Merate - Italy) has been working, from 2009 till 2013, under ESA contract aimed to develop in Europe a Slumped Glass Optics (SGO) technology, alternative to the one based on silicon pores, and based on the slumping of thin glass foils, in parallel to the work being carried out at NASA/GSFC and other institutes. The INAFOAB innovation is the use of pressure in the hot slumping process to replicate the mould figure. This technology is coupled with an integration process able to damp low frequency errors. My Ph.D. activity, carried out at the INAF-OAB, is devoted to the advancement in the hot slumping technology assisted by pressure, for the production of glass mirrors for future X-ray telescopes. As a preliminary work, I have developed a new method to precisely characterise the mid frequency errors of the glass foils. An existing model, developed by Jimenez-Garate in 2003 to account for the relaxation of ripples in the slumped glass foil when in contact with the slumping mould, was modified to include the application of pressure, and the model was compared to the experimental results. The pressure was found to be essential, also from the the theory viewpoint, to reduce mid-frequency errors in the profile of slumped glass foils, which crucially degrade the performances of the optics. As for the slumping process, I have introduced a new glass material (Corning Eagle XG) combined with the Schott Zerodur K20, already chosen at earlier times as slumping mould material for its anti-sticking properties. I have developed cleaning protocols, used different thermal cycles and different pressures conditions, and reconditioned the pressure control system. I have also optimised the mould height with experiments based on Finite Element Analysis, and reduced the thermal gradients inside the mould and the glass foils. The final result obtained in this research are slumped glass foils with angular resolution improved from the initial 7 arcsec to 2.2±0.3 arcsec in single reflection, as expected from metrology at 1 keV X-ray energy and 0.7 incidence angle, and computed by simulating a perfect integration. This result has to be compared with 2 arcsec defined as the error budget allocated for the slumped glass foils. About 1 arcsec is due to replication of the mid-frequency errors in the slumping mould which, owing to the improved result in the slumping, are now a limiting factor in the quality of the slumped glass foils. From the roughness point of view, the contribution to the HEW is guaranteed to be less then 1 arcsec at 1 keV, with an rms improvement from 21 to 11 Å throughout my Ph.D. Moreover, the last tests carried out proved that the roughness of the slumped glass foils can be further reduced to _ 7 Å, making the technology attractive also for higher X-ray energies and higher incidence angles. The glass foils slumped during this Ph.D. have been used for the production of two SGO Proof Of Concept prototypes (POC) for the ATHENA mission: the POC#3, (with 2 glass foils assembled) is characterised by HEW value of 5.5 arcsec at 0.27 keV in the best portion of the module, as tested at the PANTER facility; the POC#4 (with 8 glass foils) has expected HEW of about 15 arcsec from metrological data, but the result in X-ray was much higher because of accidental problems during the integration. The next prototype, the POC#5 (with the best eight glass foils produced during my Ph.D. activity), will soon be integrated and tested in X-rays, to assess the improvement already proven by metrology and simulations. Some of the produced slumped glass foils have also been used for studies of active profile corrections with piezoelectric elements, both for the AXYOM project and X-ray Surveyor mission. Alternative materials and techniques have been used in this research to further improve the result. To reduce the surface micro-roughness of the slumped glass foils, I have proven that the slumping technique with pressure application can also be suited in the indirect slumping, in which the optical surface of the glass is not in contact with the mould, thus preserving its initial surface smoothness. Moreover, the dip coating technique was studied to fill the micro-pores present on the surface of the glass foils slumped with the direct technique. Alternative materials for the slumping mould were tested, defining Si3N4 as a very promising candidate, owing to its higher rigidity with respect to Zerodur K20 and therefore preferable to avoid mould deformations with the slumping, as experienced with K20. The Gorilla glass (normally used for smartphones and tablets), was proven to preserve the surface quality of the glass foils, once slumped and chemically tempered to increase the mirror endurance against the vibrations experienced at launch. Finally, I have also studied the scattering and the reflectivity properties of multilayer coatings for X-ray optics. I have upgraded an IDL code to simulate the roughness growth in the multilayer deposition process, extending the computations from periodic to the more general case of graded multilayers. I have also contributed to the X-ray measurements at the BEAR beamline of the Elettra synchrotron in Trieste, on multilayers deposited on glass, silicon and electroformed nickel, for the polarimetric LAMP project