ETerostrutture iNnovative AlGaN/GaN per dispositivi ad Alta efficienza energetica (ETNA)

Responsabili di progetto

Fabrizio Roccaforte, Michal Leszczynski

Accordo

POLONIA - PAS (NUOVO ACCORDO) - Polish Academy of Sciences/Polska Akademia Nauk

Bando

CNR-PAN 2017-2019

Dipartimento

Scienze fisiche e tecnologie della materia

Area tematica

Scienze fisiche e tecnologie della materia

Stato del progetto

Rinnovo

Relazione per il rinnovo

joint-report-etna-2014-2016-finale.PDF

Proposta di ricerca

Nowadays, it is recognized that the steadily increasing demand of electricity in the world must be faced through a more efficient use of the energy in the conversion and distribution processes. In particular, it has been estimated that the introduction of new materials and technologies in solid state power electronics devices can lead to an overall reduction of the energy consumption in the conversion processes up to about 20%.
Gallium Nitride (GaN) and its related AlxGa1-xN alloys are excellent materials to meet the demands of future RF power devices, in terms of power density, operation speed and reduction of energy losses. In fact, their high critical electric field, combined with the presence of a two dimensional electron gas (2DEG) at AlGaN/GaN interfaces, permits the fabrication of high electron mobility transistors (HEMTs) with a low specific on-resistance, high breakdown voltage, and increased efficiency at high frequencies operation. Moreover, the direct band gap of GaN, tunable with the Al-concentration in AlxGa1-xN alloys, allows the fabrications of a variety of opto-electronics devices (Light Emitting Diodes, Laser Diodes, UV-detectors, etc.).
P-type doped layers of GaN are important building blocks in many of these devices. Indeed, since decades the physics related to p-type GaN in electronics has been a relevant scientific topic, as recognized in the motivation of the assignation of the Nobel Prize for Physics in 2014. However, the growth and integration of p-type GaN into power devices present still several issues. As an example, due to the high ionization energy of Mg p-type dopant and the difficulty in the electrical activation (strongly affected by the presence of hydrogen in the material), the control of the electrical properties of these layers is extremely challenging. Moreover, in some specific applications (such as in normally-off HEMTs with a p-GaN gate, or in vertical diodes), the control of the selective dry etch of p-GaN or selective p-type doping is extremely difficult, and represent a bottleneck for the development of such technologies.
Clearly, a further development of GaN electronics requires additional efforts to better understand some fundamental physical aspects associated to the p-type regions in GaN-based materials.
During the CNR-PAS Cooperation Agreement 2014-2016, in the framework of the project ETNA between CNR-IMM of Catania and the Institute of High Pressure Physics (Unipress - PAS) of Warsaw, first important results on the selective growth of p-GaN regions onto AlGaN/GaN heterostructures have been achieved. However, it has been demonstrated that the electrical properties and the uniformity of the re-grown GaN layer depend on the devices geometry, as well as on the hard-mask used for the developed process. To solve the issues encountered in the previous activity, one of the proposed research topic for the prosecution of the project is to study the impact of "wettable" hard masks (e.g., AlN-based) on the morphology and electrical properties of p-GaN layers, and to finally demonstrate the integration feasibility of such new layers into specific devices (e.g., normally-off HEMTs, Schottky diodes or p-n junction on bulk GaN substrates). Another important issue in MOVPE (metalorganic chemical vapour phase epitaxy) growth of p-type (Mg-doped) GaN is related to the hydrogen incorporation with Mg. In fact, thermal treatments must be used to activate Mg through hydrogen removal. However, if a n-type layer is grown onto the p-type GaN (e.g., as it may occur in many parts of the proposed devices), this activation becomes extremely difficult and can be affected by the materials defects. Moreover, hydrogen is suspected to lower the device life-time, as it can diffuse easily upon electric current flow. Therefore, these issues will be raised in the proposed project, combining the competences of Unipress in the growth of p-GaN and of CNR-IMM in the devices processing and characterization, as well as dopant profiling by SPM techniques.
The collaboration between CNR-IMM and Unipress-PAS started already in 2010 within an ENIAC-JU project (Last Power). After that, the fruitful interaction of the two partners continued during the previous CNR-PAS Agreement (2014-2016) and resulted not only in the publication of joint papers on international journals, but also in a series of initiatives (one of them, the project Flag-Era GraNiTE, involving the Unipress spin-off TopGaN, has been approved for funding from the EU commission) to attract new funds in the framework of European calls (see final Joint Report 2014-2016).
The present exchange project will enable to improve the "know-how" developed in the last three years, to add new aspects of the research and to strengthen the collaboration between the two groups through visits, seminars, exchange of materials and mutual transfer of knowledge. The renewal of the ETNA project for the period 2017-2019 will give the possibility to the partners to integrate their ongoing funded programs, favouring also the training of newly enrolled PhDs and Post-Docs researchers of CNR-IMM with a research activity carried out in an international context. For that purpose, parallel to the experimental activities, a detailed program of mutual visit is proposed, including seminars (in the form of small lectures) and training in the labs for the youngest researchers. Within this new project, CNR-IMM and Unipress-PAS intend to submit in 2017 the proposal for the organization of a symposium at the EMRS-Fall meeting to be held in Warsaw in September 2018.

Obiettivi della ricerca

The focus of the project is the study of novel processes for p-GaN layers, with the following objectives:

1) p-type GaN layers are important for many devices. However, in specific applications, the control of selective plasma etch of p_GaN can be a serious limitation. Hence, the first objective will be to optimize the technology of p-GaN re-growth, going beyond the "know-how" developed during the previous agreement 2014-2016, by employing new "wettable" hard marks and to characterize the re-grown layers under the morphological, electrical and structural point of view.

2) The electrical activation of p-type Mg dopant is strongly related to the out-diffusion of hydrogen during annealing. Hence, to employ p-type layers into practical devices it is important to study the activation processes, in relation to the presence of material defects (such as holes in n-p-n structures). In this context, the second objective will be the nanoscale SPM study of the surface electrical properties as a function of a distance from these defects.

3) Due to the importance of p-GaN technology, the third objective will be to demonstrate the integration feasibility of the developed processing steps into lateral and vertical devices. For this purpose, the following test vehicles will be used: (i) normally-off AlGaN/GaN HEMTs with a re-grown p-GaN gate; (ii) Schottky diodes on bulk GaN substrates with a re-grown edge termination; (iii) P-N junctions on bulk GaN substrates with a re-grown p-GaN anode.

Ultimo aggiornamento: 04/04/2025