Eur. Phys. J. Appl. Phys.
Volume 73, Number 2, February 2016
|Number of page(s)||11|
|Section||Plasma, Discharges and Processes|
|Published online||29 February 2016|
DBD reactor design and optimization in continuous AP-PECVD from HMDSO/N2/N2O mixture
Université de Toulouse; UPS, INPT; LAPLACE (Laboratoire Plasma et Conversion d’Energie); 118 route de Narbonne, 31062 Toulouse cedex 9, France
2 CNRS; LAPLACE; 31062 Toulouse, France
a e-mail: firstname.lastname@example.org
Revised: 19 January 2016
Accepted: 21 January 2016
Published online: 29 February 2016
Dielectric barrier discharge (DBD) deposition of thin films is increasingly studied as a promising alternative to other non-thermal processes such as low-pressure plasma-enhanced chemical vapor deposition (PECVD) or wet-coating. In this paper we demonstrate how optimizing gas injection in the DBD results in an improvement in the reactor performance. We propose to confine the precursor gas close to the deposition substrate by an additional gas flow. The performance of this design is studied though simulation of mass transport. To optimize the deposited thickness, gas cost and reactor clogging, we assess the influence of the confinement, total gas flow rate and DBD length. The confinement is found to reduce reactor clogging, even for long DBD, and increase the deposit thickness. This increase in thickness requires a proportionate increase in the gas flow-rate, making the gas-cost the main limitation of the proposed design. We show, however, that by fine-tuning the operating conditions a beneficial compromise can be obtained between the three optimization objectives.
© EDP Sciences, 2016
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.