Characterization of the Charging and Long-Term Performance of Cytop Electret Layers for MEMS Applications
- PDF / 8,127,666 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 43 Downloads / 172 Views
1134-BB08-17
Characterization of the Charging and Long-Term Performance of Cytop Electret Layers for MEMS Applications U. Bartsch, J. Gaspar, and O. Paul Microsystem Materials Laboratory, Department of Microsystems Engineering (IMTEK), University of Freiburg, Germany ABSTRACT This paper reports on the characterization of the charge stability of an amorphous fluoropolymer electret called Cytop. Cytop is a dissolved polymer material, compatible with standard micromachining fabrication technologies. In this study, Cytop layers are deposited and patterned on Pyrex and silicon substrates, followed by the electrical poling of the material by corona discharge using a customized charging station. The long-term performance of Cytop as an electret material is evaluated as a function of several relevant charging parameters. The results reveal highly stable layers, able to keep at least 92% of the initial charge 143 days after the corona charging stored at 23°C. INTRODUCTION Research in the field of polymer electrets compatible with microelectromechanical systems (MEMS) is of great interest because of the potential application of these materials in electrostatic micro energy harvesting devices [1], [2]. The characterization of the performance of these materials as electrets as well as their compatibility with standard MEMS fabrication processes are therefore necessary for their implementation in actual micro devices. The amorphous fluoropolymer Cytop CTL-809M (Asahi Glass, Tokyo, Japan) is available in solution, making it possible to spin-coat Cytop on standard substrates such as Pyrex and silicon (Si) wafers. Cytop is a multifunctional material, not only used as an electret material, but also for example for optical applications such as wave-guides [3]. In addition, both wafer and flip-chip bond processes using Cytop show great potential for the packaging of MEMS devices [4]. EXPERIMENTAL PART Fabrication of samples The process used to process the Cytop samples is schematically shown in Fig. 1. A Cr/Au/Cr multilayer with thickness values of 10 nm, 300 nm and 40 nm, respectively, is first evaporated on the front of a Si wafer. It is followed by the spin-coating of a multilayer of Cytop onto the metallization, using a rotation rate of 450 rpm during 20 sec, as shown in Fig. 1 (a). This step may be repeated in order to obtain the desired layer thickness. Layers thicker than 20 µm can be obtained in this way, if required. A softbake of 15 min at 100°C is performed after each spin-coating step. After deposition of the layers, the Cytop is hardbaked for 1h at 185°C. The metal layer is not patterned before the Cytop spin-coating. This has one main advantage: the single Cytop layers are more homogeneous and therefore lower rotational speeds may be used for producing thicker layers. As a result, the total number of spin-coating and process steps is reduced. After the photolithography illustrated in Fig. 1 (b), which uses the photoresist AZ9260
Figure 1. Fabrication of the Cytop samples: (a) Cytop and Photoresist are spun onto the me
Data Loading...
