Potential applications of electromagnetic absorbers strongly increased over the past few years. Radar absorbing materials were mainly used for stealth applications in the past but are now also integrated in industrial processes (electromagnetic compatibility in RF systems, antennas…). Moreover, the strong development of wireless technologies has led to an increase in the human exposure to electromagnetic waves. This fact gives rise to new public health issues and house protection against electromagnetic radiations is thus a pretty hot topic. Potential applications of radar absorbers are nowadays numerous and new technologies have thus to be developed to answer to these growing needs. This project has two main objectives: i) ultra-thin absorbers for low frequency applications (<4 GHz) and ii) 3D absorbers or Frequency Selective Surfaces (FSS). The need in ultra-thin low-frequency absorbers concerns both military and civil engineering. Indeed, at these frequencies, the most efficient solutions consist in using ferrite ceramics (heavy and expensive) or loaded polymer foams (thick). Flexible magnetic composites can also be used but their absorption capacities are lower. This project proposes to design and fabricate ultra-thin absorbers thanks to the coupling of metasurfaces and composite materials. Considering the frequency band of interest (1-4 GHz), potential applications will concern not only stealthiness of military systems but also the house protection against radiations (GSM, Wifi, 3G, 4G) and a decrease of the electromagnetic interactions between civil radars and wind plants. The second objective of the project is to develop technological means for the realization of 3D absorbers and FSS. These 3D objects will be applied to electronic war (protection against electromagnetic attacks) or to electromagnetic compatibility issues (absorbent packaging for microwave devices). 3D printing of composite materials and 3D selective metallization processes will be used to realize the demonstrators.

New microelectronic applications such as wireless communications or radar detections require increasingly high data rates or resolutions. That implies to work at very high frequencies, in the millimetre waves domain. More specifically, in the frequency range 140-220 GHz (G-band), microelectronic circuits are emerging but suffer from a lack of complete characterization tools. There is a strong need for in wafer integrated measurement set-ups. Hence the BISCIG project aims to integrate, for the first time, a measurement system that would directly and completely measure incoming and outgoing powers, at all ports, and very close to the Device Under Test (DUT). The set-up is proposed in G-band. This project includes two versions. The first version (called "load-pull") concerns large signal power measurements to characterize power amplifiers in millimeter and sub-millimeter-wave bands. External current measurement devices, such as commercial impedance tuners, cannot do that efficiently. Because of their intrinsic losses in G-band, they cannot cover all the impedances of the complex plane to be presented at the output of the power amplifier. The second version will enable to characterize 4-ports DUT with small signal analysis (called “S-parameters”) and to perform differential measurements. Indeed, such instrument does not exist beyond 110 GHz. The BISCIG project therefore meets an industrial need for characterization of devices for new applications and expanding in G-band (high-speed communication systems, radar detection, imagers). Our solution consists in addressing the integrated measurement set-up with a well-known signal covering the 35-55 GHz spectrum. This microwave signal is then amplified and frequency quadrupled in order to address the G-band in the same technology as the DUT. Finally, we measure DC output signals as images of the detected powers, to characterize the behaviour of the DUT. The technology, provided by STMicroelectronics, is the SiGe BiCMOS 55 nm which is very powerful in the millimeter-wave band. The Back End of Line is very well suited to realize passive devices (thick metals in the upper layers). The Front End of Line is completely suitable as well for active devices (fT/fmax = 300/400 GHz). The academic partners will work closely, hand in hand, with the manufacturer STMicroelectronics providing the technology. IMEP-LAHC will design the measurement systems while IEMN will handle the characterization of the component blocks as well as the various sub- systems.

The control of wave propagation, which relies on the artificial media design, is an important topic in fundamental and applied research as well. Unusual acoustic properties have been observed in both phononic crystals, periodic structures allowing the control of the wave at the scale of the wavelength, and metamaterials in which the effect is expected at large wavelength compared to structuration size. Numerous functionalities have been demonstrated such as frequency filtering and demultiplexing, acoustic insulation, wave guiding, acoustic cloaking, negative refraction and super-resolution, pulse delaying and compression with different application fields such as telecommunication components, imaging and acoustic stealth. However, few phononic crystals and metamaterials have ended up as actual devices because of their lack of tunability: the control of wave propagation, often obtained for a limited frequency range, is completely defined by the geometry and physical properties of the constitutive materials at the fabrication stage. To bring to these devices tunability and re-configurability, which are nowadays essential to satisfy most professional system requirements, MIRAGES project aims at developing tunable and reconfigurable active phononic crystals and metamaterials which will include piezoelectric or magnetostrictive materials and will be controlled by electric or magnetic fields. Tuning of active materials elasticity will be insured by external electrical impedance variations for piezoelectric constituents or by DC magnetic bias for magnetostrictive materials. The establishment of generic models and technologies for the design, elaboration and optimization of active phononic crystals and metamaterials constitutes the first goal of the project. Theoretical and numerical models, elaboration processes and specific characterization set-ups will be developed to evaluate quantitatively the magnitude of properties variations provided by the different materials, structures and control methods and to identify the optimal solution in terms of tunability range, frequency and fabrication technology. In connection with this general concept, two goals related to specific applications are forecast in MIRAGES project: • the demonstration that the integration of a controllable active material within a phononic crystal bring a clear added value to an existing acoustic MEMS component used in telecommunications. Electric or magnetic control of one-dimensional magnetostrictive phononic crystal will be studied in the MHz range to realize either a switchable Bragg mirror, a post-fabrication finely tunable (less than 1%) Coupled Resonator Filter or a Fabry-Perot acoustic resonator with broad tuning (more than 10%). • the first realization of an electrically controlled gradient-index metamaterial or phononic crystal for stealth applications. To mimic sonar, the demonstrator will take the form of an active wall located between an acoustic source and a target. Constituted by a two-dimensional phononic crystal or metamaterial with piezoelectric inclusions, the wall will act as a countermeasure on the position, the speed and the orientation of a target. This demonstrator will fully exploit the possibilities given by the static, dynamic and real-time electric control of the artificial medium to shape the delay, the frequency content and the reflection angle of the reflected beam.

Gallium Nitride (GaN) devices are foreseen as the next generation of RF power transistor technology in the millimeter wave range. A number of groups (HRL, Triquint, UCSB, and Fujitsu) have demonstrated the unique combination of higher power, higher efficiency and wider bandwidth available with GaN as compared to competing GaAs and Si based technologies. In this frame, a new heterostructure based on high quality AlN/GaN has been developed and enables to boost transistor carrier density 2-3 times higher than AlGaN/GaN HEMTs, while offering high aspect ratio (Lg/a) needed for millimeter wave operation due to the possibility to use ultra thin barrier. Using this novel double heterostructure AlN/GaN/AlGaN on Si, developed in collaboration with EpiGaN, IEMN has already demonstrated achieved state of the art results which have shown: *Cut-off frequencies Fmax about 200 GHz together with a voltage handling of more than 100V which allow to reach very high power density in Ka band (2.5W/mm @40 GHz). *Very low leakage currents, high gain, reduced trapping effects which have led to the lowest Noise Figure ever measured in Ka band with GaN HEMTs (NFmin=1.2dB @40GHz). In addition a preliminary reliability study developed with the University of PADOVA has shown quite promising results in terms of stability and robustness with several hours of measurements on 100nm gate length devices. These first reliability results demonstrate that the reproducibility is good enough to foresee the realization of complex circuits in Ka band. The CROCUS project aims at the design, realization and test of robust circuits in Ka band based on the AlN/GaN on Si technology with outstanding performance for civilian and military applications. This paves the way to a European source of reliable millimeter wave MMIC GaN on Si circuits telecommunication and radar systems. This project proposes to design, realize in hybrid integration and test some key Ka band circuits such as power amplifier (HPA), Low noise amplifier (LNA) and mixer, using EpiGaN epitaxies. These circuits will be developed with respect to Thales communication (military communication systems) and BLUWAN specifications (Civilian LMDS & Satellite applications). IEMN, XLIM, THALES Com and BLUWAN will join their efforts and combine their competences to reach the ambitious goals of this project which should lead to a real technological breakthrough in terms of power, efficiency, linearity and self protection. Even though Thales is not an official partner of this project the society is interested in following it and is willing to provide operational specifications according to their military applications and to promote an industrial development of this technology (Cf joint letter of Thales)

A large number of chemical species have absorption lines lying in the Terahertz frequency range, which is thus a very interesting spectral range for molecular spectroscopy applied to earth, planetary, environmental and space science. In spite of this great scientific interest, the THz range still remains one of the less used spectral band. It is indeed called the THz gap because of the lack of low cost, compact and reliable sources and detectors as compared to its two adjacent frequency ranges, i.e. the microwave and the optical waves, which rely respectively on the efficiency of the photonics and electronics technology. A dramatic improvement of the detectors performances have been done in the last years, motivated by the emergence of new applications such as security or medical imaging, DNA study or even painting analysis. Background limited THz detectors are now available. On the other hand, the generation of THz wave remains a challenge and a room-temperature powerful, narrow linewidth, and compact THz source is still needed. Spectral purity is indeed as important as power level. It is obvious when a heterodyne mixer is used as receiver, whose minimum detectable power is related to the frequency width of the intermediate frequency filter. One of the most promising THz continuous-wave (CW) sources working at room temperature is based on the photodetection of the beating frequency generated by two spatially overlapped infrared lasers. It is the so called photomixing THz source. This down converting between very high frequency (~300 THz) infrared lasers is intrinsically wideband. Nevertheless, THz sources based on photomixing suffer, up to now, from a lack of power, around 10 µW at 1 THz. The output power is indeed limited by the trade-off between the small size of the photodetector used in order to minimize its electrical capacitance and the photocurrent needed in order to generate a high output power. A new type of photomixer is required in order to break this trade-off and reach the milliwatt level at 1 THz. In PHENIX, we propose to study a new architecture of photomixers in order to develop a wideband and powerful photomixing source. It is based on a Highly Distributed Photoconductor (HDP). The HDP, through its highly distributed nature (i.e a length reaching a few THz wavelengths), will offer both high-frequency and high-power operation unreachable with vertically illuminated photodetectors or standard distributed photodetectors. A photodetector of length reaching 1 mm will absorb up to 1 W of optical power, a value at least ten times higher than the one in the current high frequency photodetectors. An output power of around 10 mW at 300 GHz and 1 mW at 1 THz is expected with such a wide band CW THz source. Besides, the achievement of an efficient THz CW source through photomixing technique relies on the availability of a narrow linewidth optical beatnote at the frequency of interest and at the absorption wavelength of the photomixer (~800 nm). The straightforward solution consists of using two laser diodes since such diodes are commercially available and can be easily implemented in the system. This solution however suffers from a limited frequency stability of the beat note and a poor spectral purity. In PHENIX we propose to develop a compact beatnote source working at lambda=800 nm, which will offer a high beat frequency stability and spectral purity. It will be based on a Solid state Dual Frequency Lasers (DFL) using a titanium doped sapphire as active medium and exhibiting an intrinsically narrow beatline (<30 kHz in free regime). We have recently performed a preliminary demonstration. Nevertheless, several improvements are still required to end up with a compact continuously tunable and robust laser.