The project |
Meeting reports |
Progress reports |
News |
List of participants
M. Djafari Rouhani (LAAS-CNRS), A. Esteve (LAAS-CNRS),
D. Esteve (LAAS-CNRS), J. Greer (NMRC-Cork),
P. Hylgaard (Chalmers Univ. Tech.), G. Landa (LPS-Univ. Paul Sabatier),
J.L. Leray (CEA-DAM), M. Lhermine (Sigmaplus),
B. Lundqvist (Chalmers Univ. Tech.), E. Morvan (CNM-Barcelona),
P. Paillet (CEA-DAM), J. Rebollo (CNM-Barcelona),
F. Thievenaz (LAAS-CNRS)
This two-day kick off meeting was devoted to :
The following points were addressed during the meeting.
1 - The participants to the meeting introduce themselves and give briefly their main fields of interest.
2 - Adminitrative presentation
An overview of European networks in general and ATOMCAD in particular was given by F. Thievenaz, administrative coordinator of the network. The network has started february 1st. 30% of the total allowances will be available within 60 days, provided the general agreement has been signed by the participants.
3 - Participants activities
Participating group present in turn their specific activities.
3.1 - LAAS - CNRS
M. Djafari Rouhani presents the activities of LAAS-CNRS involving different levels of hierarchical models and applied to the following fields :
3.2 - NMRC - Cork
J. Greer presents the NMRC activities based on quantum chemistry calculations and molecular dynamics in the following fields :
3.3 - CNM - Barcelona
J. Rebollo presents the activities of CNM which are essentially using conventional tools and address :
3.4 - Sigmaplus
M. Lhermine presents the activities of Sigmaplus company devoted to the development and exploitation of softwares for external users. The Monte Carlo technique is widely used.
3.5 - Chalmers University of Technology
B. Lundvist presents the activities of CUT essentially based on ab initio calculations using Density Functional Theory and applied to the following areas :
3.6 - CEA - DAM
P. Paillet presents the CEA-DAM activities based on ab initio calculations and mesoscopic models applied to the following areas :
4 - ATOMCAD Network
M. Djafari Rouhani presents the ATOMCAD project to the participants with particular emphasis on :
5 - Global Strategy
The global strategy to be followed in each task and action has been presented by the different coordinators.
5.1 - Models and Tools Inventory
The detailed inventory of packages available within the participating groups, together with their origin and basic characteristics is summarized below.
5.2 - Software Development
As a conclusion of the general discussion, it has been decided to start the ATOMCAD network by focusing its activities within two particular themes :
This choice allows :
P. Paillet and J.L. Leray initiate the first point by suggesting a connexion between OXCAD and TRAPOX packages. It is postponed that capture cross sections of carriers by various defects in SiO2 be calculated using ab initio methods. These cross sections can then be injected into TRAPOX which is a continuum package relating the capture cross sections to the electrical characteristics of the devices. If the result is not completely satisfactory, then OXCAD can be used to take the defect distribution inside the layer into account, before parameters are injected into TRAPOX. This would constitute a first example for the use of multi-scale modelling.
J. Rebollo points out the SiC technological and industrial potentialities and introduces the various following lines along which the material could be investigated.
1 - Layer deposition. The experimental procedure concerns the growth of SiC/SiC with 3.5° angle tilt using CVD method with a SiH4-CH4 gas mixture at 1500 °C, or PVD method by sublimation of SiC at 2700 °C.
2 - Oxidation. The role of carbon is crucial and leads to unexpected mechanisms and experimental results. The OXCAD package could be improved and extended to the case of SiC.
3 - Doping by ion implantation. The boron diffusion during post implant annealing can be studied using ab initio models.
4 - Study of defect structures, using ab initio methods.
5 - Dopant-Defect interactions. The aluminum diffusion is known to involve interaction with vacancies, interstitials and extended defects.
6 - Metallization.
7 - Vibrational spectroscopy using VIBRATOM package.
All participants will try to find their own expertise and possible contribution within the proposed fields. Doctoral and postdoctoral students will be affected on these collaborative examples.
5.3 - Software Interface Development
It has been concluded that we should first focus on teleworking. This means that collaborative work should be done without transfer of the packages. Each user may have direct access, via internet, to competences and tools available at all sites.
LAAS and CNM have been candidates to operate Monte Carlo OXCAD and conventional ISE packages at their respective separate sites. Other experiments should follow.
In a second stage, as a result of the experience gained by teleworking, the necessary tools for an easy use of the distributed packages will be developed.
5.4 - Dissemination of the tools
D. Esteve points out that legal procedures for teleworking are available and should be respected between all the partners. He suggests that an agreement be written between the partners.
For the external dissemination, all the partners have been invited to think about the organization of an international conference around may 2002.
5.5 - Usability demonstration
The first demonstration prototypes will be based on the two examples described in section 5.2, namely silicon oxidation and silicon carbide devices. Simulations will be performed on different packages at different sites. An experimental validation may be undertaken via short runs performed in the CNM clean rooms.
5.6 - Training programme
Chalmers Univ. of Technology will prepare a proposal for the organization of a summer school during july 2002 or september 2002.
6 - Student recruitment
All the participants will advertise the available positions through internet and via specialized web sites (psi-k network, ccpt). In case of problems encountered in recruiting suitable students, it has been decided to :
7 - Next meeting
It has been decided to organize the next ATOMCAD meeting at CNM-Barcelona at mid september 2000 (probably sept.14 and 15th).
The schedule of the two day meeting will be :