Effects of Sputtering Metal Deposition Parameters in the Photoresist used as a Sacrificial Layer in Micro-electromechani
- PDF / 2,022,187 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 3 Downloads / 188 Views
1075-J07-01
Effects of Sputtering Metal Deposition Parameters in the Photoresist used as a Sacrificial Layer in Micro-electromechanical System Devices David Molinero electronics department, Universitat Politecnica de Catalunya, jordi girona 1-3, campus nord modul c4, barcelona, E-08034, Spain ABSTRACT The present letter describes a reliability study of the micro-electromechanical system fabrication with a photoresist layer used as sacrificial layer with an aluminum beam deposited by means of RF sputtering method. This work reports changes of the roughness and planarity of the sacrificial layer beneath the aluminum film following the sputtering deposition. Such changes may be attributed to the alteration of the photoresist properties due principally to the outgassing of hydrogen by decomposition of C-H bonds under argon plasma. A safe deposition parameters area was identified where the photoresist layer keeps its properties and may be used as sacrificial layer. INTRODUCTION The fabrication of Micro-electromechanical system (MEMS) devices, such as metal bridges that can be used as switches or capacitors [1], uses different materials as sacrificial layer, such as silicon oxide, silicon nitride or polyimide [2][3]. Complex processes are needed to deposit, etch and release these films. Some MEMS devices fabrication processes are much simplified if a photoresist is used as sacrificial layer. This technological option implies that metal deposition takes place after the photoresist has been spun-on and cured. It becomes relevant for the reproducibility of the devices that the sputtering deposition parameters do not significantly modify the thickness and planarity of the photoresist. We have observed, in our fabrication experience, that these reliability problems do arise and have launched a limited but systematic study to identify suitable conditions for the sputtering step and this is the object of this work. Although the study is restricted to the sputtering system available in our Lab, the conclusions may help researchers using different equipment to find suitable experimental conditions for their own fabrication process. EXPERIMENT A simple structure and procedure to test the photoresist as a sacrificial layer has been designed (Fig. 1), where a silicon wafer was used as starting material. Photoresist was spun-on and patterned to define the sacrificial layer of MEMS device. In our case a Shipley SPR220-1.2 photoresist was used and a thickness of 1.1µm was spun using a spinner velocity of 3000 rpm for 30 seconds. Two minutes in a hotplate at 115ºC was applied as softbake and the same time was used as hardbake after exposure of the photoresist. Several measures with a profilometer Dektak 3030 have been done to test the planarity and thickness of the photoresist layer and, in all cases a good planarity and thickness reproducibility was obtained. A metal beam structure was formed on top of the photoresist with a metal layer deposited with an RF sputtering system Edwards Sputters coater ESM100 NS1719.
Aluminum metal has b
Data Loading...
