CEA has a long history of customizing optoelectronic components for space and astronomy applications. Based on this expertise, we are undertaking the development of cooled silicon bolometers for millimetre-wave (mm-wave) polarization detection in the next generation of space astronomy missions such as SPICA. Silicon bolometer technology has been demonstrated successfully in space conditions through the Herschel mission. There are many benefits of this technology such as the use of a simple and low-power read-out circuit that can be integrated below the detector array in an above-IC (Integrated Circuit) integration scheme. The advanced integration in a large array and the fabrication process based on microelectronics techniques are key challenges for these developments. This work presents the early results on the design, the fabrication and the first characterization of an innovative pixel for mm-wave polarization detection. The aim is to have an adapted absorption around λ=1.5 mm. These bolometers are composed of an absorbing layer and a thermometer, which are thermally insulated from the substrate. To increase the sensitivity, these detectors are working at very low temperature typically between 50 and 100 mK. The suspended thermometer is made of silicon implanted with Phosphorus and Boron species, and we optimized the design to have a high sensitivity with a 3D Variable Range Hopping conduction (Efros law) and a low 1/f noise at low temperature. The heat capacity of the bolometer is optimized by using a meander shape of the thermometer together with superconducting Ti/TiN thin films for the electromagnetic wave absorption. This sensor is implemented on a standard SOI substrate. Measurements of test structures at room temperature, and first results at very low temperature have been performed to evaluate the electrical performances of the fabricated detectors. The mechanical behaviour of released structures, including pixels with a pitch of 1200μm and 600μm, is presented and discussed.
Silicon-based vacuum packaging is a key enabling technology for achieving affordable uncooled Infrared Focal Plane Arrays (IRFPA) required by a promising mass market that shows momentum for some extensive consumer applications, such as automotive driving assistance, smart presence localization and building management. Among the various approaches studied worldwide, CEA, LETI in partnership with ULIS is committed to the development of a unique technology referred to as PLP (Pixel Level Packaging). In this PLP technology, each bolometer pixel is sealed under vacuum using a transparent thin film deposition on wafer. PLP operates as an array of hermetic micro caps above the focal plane, each enclosing a single microbolometer. In continuation of our on-going studies on PLP for regular QVGA IRFPAs, this paper emphasizes on the innate scalability of the technology which was successfully demonstrated through the development of an 80 × 80 pixel IRFPA. The relevance of the technology with regard to the two formats is discussed, considering both performance and cost issues. We show that the suboptimal fill factor inherent to the PLP arrangement is not so critical when considering smaller arrays preferably fitted for consumer applications. The discussion is supported with the electro-optical performance measurements of the PLP-based 80×80 demonstrator.
Silicon based vacuum packaging is a key enabling technology for achieving affordable uncooled Infrared Focal Plane Arrays (IRFPA) as required by the promising mass market for very low cost IR applications, such as automotive driving assistance, energy loss monitoring in buildings, motion sensors… Among the various approaches studied worldwide, the CEA, LETI is developing a unique technology where each bolometer pixel is sealed under vacuum at the wafer level, using an IR transparent thin film deposition. This technology referred to as PLP (Pixel Level Packaging), leads to an array of hermetic micro-caps each containing a single microbolometer. Since the successful demonstration that the PLP technology, when applied on a single microbolometer pixel, can provide the required vacuum < 10-3 mbar, the authors have pushed forward the development of the technology on fully operational QVGA readout circuits CMOS base wafers (320 x 240 pixels). In this outlook, the article reports on the electro optical performance obtained from this preliminary PLP based QVGA demonstrator. Apart from the response, noise and NETD distributions, the paper also puts emphasis on additional key features such as thermal time constant, image quality, and ageing properties.
Recent developments at the Infrared Lab (LIR) of CEA, LETI have been concentrated on the pixel size reduction of uncooled infrared detectors. With the support from French company ULIS, we have successfully demonstrated the technological integration of 12μm pixels on a commercial TV-format read-out circuit (VGA-ROIC) supplied by ULIS. The 12μm pixel has been designed, processed and characterized in CEA, LETI and first results showed exceptional performances. This paper presents the characterization and associated imagery results.
Vacuum packaging is definitely a major cost driver for uncooled IRFPA and a technological breakthrough is still
expected to comply with the very low cost infrared camera market. To address this key issue, CEA-LETI is developing a
Pixel Level Packaging (PLP) technology which basically consists in capping each pixel under vacuum in the direct
continuation of the wafer level bolometer process. Previous CEA-LETI works have yet shown the feasibility of PLP
based microbolometers that exhibit the required thermal insulation and vacuum achievement.
CEA-LETI is still pushing the technology which has been now applied for the first time on a CMOS readout circuit. The
paper will report on the recent progress obtained on PLP technology with particular emphasis on the optical efficiency of
the PLP arrangement compared to the traditional microbolometer packaging. Results including optical performances,
aging studies and compatibility with CMOS readout circuit are extensively presented.
The lifetime of optical components submitted to high laser fluences is degraded under organic contaminant environment.
The molecular background of the Ligne d'Integration Laser (LIL), prototype of the future Laser Megajoule, might reduce
the laser damage threshold of exposed fused silica surfaces. This paper reports the interaction effects between pure
model contaminant deposits and a pulsed 1064 nm laser radiation on the coming out of mirror damage. The experimental
setup allowed us to condense nanolayers of model contaminants on optics, the deposit impacts were then investigated by
Laser Induced Damage Threshold (LIDT) tests in Rasterscan mode. In order to highlight physical processes emphasizing
the emergence of optics damage, we characterized the irradiated deposit using interferometric microscopy analysis and
spectrophotometric analysis. The challenge was to determine physical and phenomenological processes occurring during
the irradiation of a pure contaminant deposit with a 1064 nm pulsed laser and to study the impact of this model
contaminant on the LIDT of dielectric SiO2/HfO2 mirrors.
Under vacuum conditions, the accumulation of low fluence laser pulses generally leads to an organic
contamination of the surface irradiated. This phenomenon reduces the optical component lifetime. Experimental
conditions such as laser characteristics, environment composition and structure of the coating strongly influence the
contamination mechanisms. Silica being the most employed material for optical coatings, this study aims at describing
the laser-induced contamination influence of silica coatings deposition techniques. E-Beam evaporated and Ion Beam
Sputtered silica thin films have been exposed to several billions 600 mJ/cm2 - 532 nm laser pulses under vacuum. This
paper presents the observations made on laser-induced contamination and discusses the physical mechanisms involved.
It is still assumed that optical components submitted to laser fluences orders of magnitude below their laser induced damage threshold (LIDT) will last for ever. However, depending upon environmental conditions, the accumulation of low fluence laser pulses leads to a progressive contamination and eventually to a damage of the optical components. In order to study the physics of the laser induceded contamination, a laser test bench has been developed. The experimental cell is dry-pumped and a mass spectrometer controls the environment around the optical component. An infrared camera diagnosis follows the sample surface temperature. This paper contains preliminary results obtained on anti-reflective coatings on fused silica tested at 532 nm with a pulse repetition rate of 10 kHz and a pulse width of 100 ns.
The pulsed Laser Induced Damage Threshold (LIDT) of optical components usually reaches several hundreds of MW/cm2. When exposed to laser power several order of magnitude below their LIDT, the optical component lifetime is, by default, considered infinite. Under specific conditions, the accumulation of laser pulses may lead to a contamination of the surface and a degradation of its optical properties and LIDT. In the first order, these phenomena depend on the experimental conditions such as the irradiation time, the laser power, and the environment. In order to better understand the physics emphasizing this degradation, we developed an experimental cell with an in-situ spectroscopic ellipsometry diagnostic. The dry-pumped cell sheltering the sample is associated with a mass spectrometer that enables us to follow the environmental conditions in which we experiment the ageing. Anti-reflection coatings on fused silica were tested under 10 kHz-532 nm laser ageing. We present first results of degradation obtained in these conditions.