To main content

Cost-Efficient Wafer-Level Capping for MEMS and Imaging Sensors by Adhesive Wafer Bonding

Abstract

Device encapsulation and packaging often constitutes a substantial part of the fabrication cost of micro electro-mechanical systems (MEMS) transducers and imaging sensor devices. In this paper, we propose a simple and cost-effective wafer-level capping method that utilizes a limited number of highly standardized process steps as well as low-cost materials. The proposed capping process is based on low-temperature adhesive wafer bonding, which ensures full complementary metal-oxide-semiconductor (CMOS) compatibility. All necessary fabrication steps for the wafer bonding, such as cavity formation and deposition of the adhesive, are performed on the capping substrate. The polymer adhesive is deposited by spray-coating on the capping wafer containing the cavities. Thus, no lithographic patterning of the polymer adhesive is needed, and material waste is minimized. Furthermore, this process does not require any additional fabrication steps on the device wafer, which lowers the process complexity and fabrication costs. We demonstrate the proposed capping method by packaging two different MEMS devices. The two MEMS devices include a vibration sensor and an acceleration switch, which employ two different electrical interconnection schemes. The experimental results show wafer-level capping with excellent bond quality due to the re-flow behavior of the polymer adhesive. No impediment to the functionality of the MEMS devices was observed, which indicates that the encapsulation does not introduce significant tensile nor compressive stresses. Thus, we present a highly versatile, robust, and cost-efficient capping method for components such as MEMS and imaging sensors.

Category

Academic article

Language

English

Author(s)

Affiliation

  • Royal Institute of Technology
  • SINTEF Digital / Smart Sensors and Microsystems

Year

2016

Published in

Micromachines

ISSN

2072-666X

Publisher

MDPI

Volume

7

Issue

10

View this publication at Cristin