Materials engineering for electronics

Electronics is nowadays confronted with miniaturization in order to increase storage densities, which leads to research involving the dimensionality of materials. In this context, the use of 2D (monolayer), quasi-2D (nanometric thin films) or 1D (nanometric-sized nanoparticles) materials with superior electronic, opto-electronic or optical properties (quantum confinement, 2D conductivity, etc.) is essential. Our team studies more particularly the physical processes of elaboration of this type of materials by advanced techniques (laser ablation, ion implantation, ...) and uses adequate post-deposition treatments (thermal or thermo-catalytic treatments under controlled atmosphere, laser, ...) for the ad hoc synthesis of low dimensional materials on/in substrates directly usable or easily integrable for the specifically targeted applications.

Developed topics

Quasi-D2D graphitic carbon films on insulating substrate for transparent electrode

Persons involved: F. Le Normand, C. Speisser, N. Javahiraly, D. Muller, N. Boubiche Collaborations: Prof. M. Abdesselam (University of Alger), F. Djeffal (University of Batna II)

For the production of transparent conductive films, a very thin graphite (graphenic) layer can be obtained directly on the surface by partial or total transformation of a Diamond Like Carbon (DLC) film deposited on quartz or glass followed by heat or thermocatalytic treatments. By pulsed laser ablation of graphite at room temperature, a whole class of DLC materials with variable properties (density, sp2/sp3 hybridization ratio, ...) can be obtained depending on the deposition conditions (fluence, thickness, ...). Heat or thermocatalytic treatments change the very thin precursor graphitic layer into a surface graphitic layer by aggregation of the graphitic domains and/or phase transformation of the sp3 carbon (diamond type) (Figure 1). From this point of view, the addition of a very small quantity of transition metal catalyst (less than one monolayer of Fe, Co, Ni), obtained by MBE (at the IPCMS, Strasbourg), proves very beneficial in lowering the temperature and kinetics of these transformations. In addition to its rigidity/hardness, chemical inertness, very low roughness (< 1 nm), the film is transparent in the visible and becomes conductive on the surface.

Figure 1: (left) Experimental film preparation scheme: metal@graphic film/DLC/quartz. (right) Figures of merit (conductivity of transparency) of Ni@graphic film/DLC/quartz and graphic film/DLC/quartz.

The figures of merit for these multilayer systems (graphic film/DLC/quartz or catalyst/graphic film/DLC/quartz) are of the same order of magnitude as for ITO/quartz-based electrodes. In the case of the nickel catalyst, these figures of merit (conductivity of transparency) are even much better than ITO (> 105 Siemens/cm from 400°C), whereas without the catalyst it is necessary to go up to 800-1100°C (Figure 2). The thickness of the initial DLC film, the laser fluence, the gaseous environment of the treatment, the nature rather than the metal concentration, the temperature and the kinetics are the parameters currently studied. They should allow us to develop an efficient device as transparent electrodes but also for other applications such as sensors.

Back to contents

Synthesis of transparent carbon-based electrodes synthesized by laser processes

Persons involved: F. Antoni, D. Muller, F. Stock

One of the major challenges that display technologies (LCD, OLeds...), optoelectronic and photovoltaic devices will have to face in the near future is to find an alternative to the use of transparent conductive oxides such as indium-tin oxide (ITO). Graphene, a 2D conductive and transparent carbon-based material appears as an attractive alternative to ITO. However, its transfer over large surfaces is complex and delicate. The elaboration process consists of two steps, described in Figure 1. First, a thin layer (about 20 nm) of Diamond-Like Carbon (DLC) is synthesized under vacuum and at room temperature by pulsed laser ablation (PLD) of a pure graphite target and deposited on an insulating and transparent substrate (glass, quartz...) in case a transparent electrode must be obtained. The deposit can however be obtained in an identical way on all types of substrates (conductive, flexible, polymers, etc.). The DLC has a good optical transmission in the visible range and is a perfect electrical insulator. However, it is partially opaque in the ultraviolet range. This property is exploited in a second step in which a uniform UV laser treatment at very low energy of the deposited layer allows the graphitization of its surface by scanning, making it conductive while maintaining its transparency.

Figure 2: Schematic representation of the electrode fabrication process.

This innovative and original process is based solely on laser technologies and offers the advantage of full compatibility with conventional microelectronic processes and also makes it possible to obtain a biocompatible material. No equivalent work has yet been published. We have been able to demonstrate that we are able to produce electrodes with performance equivalent to that of commercial transparent conductors (such as ITO), which is a remarkable result because this process uses only pure carbon deposited and processed at room temperature. Integration into an electrical circuit demonstrated the stability of the layer (Figure 2).

Figure 3: Use of the layer obtained as an electrical resistor supplying a LED to be used.

Back to contents

Nanocrystals of GaN obtained by implantation in dielectric matrices and heat or thermocatalytic treatments in a nitrogenous reducing atmosphere

Persons involved: F. Le Normand, C. Speisser, N. Javahiraly, D. Muller Collaborations: Lilia Aggar (PhD student), Prof D. Bradai, M. Abdesselam (University of Alger) / C. Bouillet, M. Gallart (IPCMS/Strasbourg)

The formation of GaN nanocrystals in SiO2/Si was obtained by implantation of Ga+ and N+ ions followed by heat or thermocatalytic treatment at 450 to 950°C in nitrogen reducing atmospheres. Analyses by high-resolution electron microscopy (HRTEM), scanning electron microscopy (SEM), high-resolution X-ray diffraction (XRD), absorption fine structure spectroscopy (XAFS) at the K threshold of Ga, and photoelectron (XPS) and Raman spectroscopy show the formation of very small nanocrystals of GaN with wurtzite structure (2-5 nm). The addition of a surface catalyst controls and activates the decomposition of the dinitrogen to mono nitrogen, preventing the diffusion of ions at the surface, selectively activating the formation of GaN (at the expense of Ga or Ga2O3 formation) and increasing the crystallographic quality of the buried GaN particles (Figure 4).

Figure 4: (left) MET image of an implanted layer of GaN on thermal SiO2 in the presence of catalyst; (right) HRTEM of wurtzite GaN particles of size 3-6 nm.

In the presence of the catalyst, photoluminescence spectra characteristic of intense excitonic emission around 3.45 eV and little "yellow light" characteristic of a material where light is emitted by radiative recombination on defects are obtained. The effect of the dielectric matrix, the nature of the gaseous environment being processed and the temperature is currently being actively studied.

The aim is to control the average size (in a range below 3 nm where quantum confinement effects and a shift towards high energies are expected), density and depth distribution of these nanoparticles. In particular, it is envisaged to use this method to build photovoltaic tandem structures by implantation of GaN in transparent DLC.

Back to contents

Synthesis of graphene by implantation/diffusion of carbon in a metal matrix

Persons involved: F. Le Normand, C. Speisser, D. Muller

We approached the growth of graphene films by a specific process of carbon ion implantation in a thick metal matrix (Ni, Cu), followed by annealing diffusing the carbon either on the surface or at the interface, depending on the depth of implantation, with eventually a specific application concerning transparent electrodes for photovoltaics. The segregation of carbon films at the surface at low energy (20 keV) and at the interface at high energy (180 keV) was demonstrated. While low-pressure nucleation by such a method strongly limits the obtaining of a graphene film, on the other hand, at the interface with a substrate (MgO(111) or SiO2) it has been possible to obtain homogeneous films with both Cu and Ni. This process has also been tested to produce very thin hexagonal boron nitride films by N and B ion implantation and annealing. We will use for this purpose the new potential of very low energy ion implantation being installed in the C3-Fab platform (1 to 10 keV).

Back to contents

Nanoparticle engineering by ion implantation

Persons involved: D. Muller Collaborations: GPM (Rouen)

For many years we have been developing processes based on ion implantation and/or irradiation to fabricate, deform, dope and functionalize nano-objects buried in a matrix to make sensors, memories or optical devices.
The ability to dope quantum dots, and in particular silicon nanocrystals (Si-NC), is a key technological issue for their use in optoelectronic applications. For example, low doping levels are needed to enable applications such as Si-NC-based tandem cells and high doping levels are needed to achieve promising tunable optical properties (via Local Surface Plasmon Resonance - LSPR). (see topic "Functional materials and sensors").
In particular, we have demonstrated the possibility of synthesizing by ion implantation silicon nanocrystals doped with phosphorus and arsenic buried in a matrix of SiO₂. This method of synthesis offers good control of the size distribution of the synthesized nanocrystals and of the quantity of dopants incorporated in these nanocrystals. Ion beam synthesis allows the elaboration of silicon nanocrystals with an average size of a few nm. The volume density of dopant-containing nanocrystals ranges from 10¹⁸ to a few 10¹⁹ particles.cm^-3 with an average atomic concentration of dopant in the nanocrystals of ~ 8%. Similar studies are underway to characterize the effect of boron doping.

Figure 6: Spatial distribution of dopants in a Si nanoparticle (left) and concentration profile around a doped nanoparticle determined by Atomic Probe Tomography (APT).

Back to contents

Developing topics

Control of qubits in wide gap semiconductor

Persons involved:  : D. Muller Collaborations: J. Tribolet (Institut Chimie Strasbourg), C. Couteau, M. Lazar (L2n Troyes)

Quantum technologies are based on the exploitation of the properties of quantum physics for tomorrow's applications: quantum communications, computers and quantum simulators, but also high-resolution sensors (see "Functional materials and sensors" theme). Several wide-gap semiconductors such as diamond, ZnO and SiC will be studied. One part of the studies will consist in positioning by ion implantation paramagnetic defects or coloured centres in diamond or SiC for quantum sensor devices or for the realisation of a quantum register of a few qubits of spins but also for diamond-based photonic devices. Another approach will consist in incorporating by ion implantation magnetic atoms into epitaxial quantum boxes which have shown in the past that they can give rise to couplings between the electron spin of the box exciton and the magnetic spin of the defect in the vicinity of the box. The challenge is to duplicate the phenomenon in diamond and to demonstrate a spin-spin coupling between the spin of a coloured diamond center and the magnetic spin of an atom that would have been implanted not far from the center.

Back to contents