After a brief overview of possible failure mechanisms and failure defects that can occur in radio frequency microelectromechanical systems (RF MEMS), this chapter focuses on three specific reliability issues: charging, because it remains the most important problem for capacitive RF MEMS; electrostatic discharge, because it is less known as a possible failure cause for MEMS; and package hermeticity, because this is an often underestimated problem for RF MEMS.

This work demonstrates, for the first time, the use of a post deposition laser annealing technique to realize operational SiGe MEMS devices at deposition temperatures as low as 210°C. The patterned amorphous SiGe layers are treated by an excimer laser to induce crystallization. After the laser treatment, SiGe devices with good electrical and mechanical properties, such as contact resistivity values to a TiN electrode as low as 4.9×10−7Ωcm2 and a strain gradient of −1.6×10−6μm−1, are obtained. Devices such as an array of functional capacitive test structures and capacitive switches are realized.

We present an integrated reliability test methodology for electrostatic discharge (ESD) testing of micro-electromechanical systems (MEMS). It is shown that conventional ESD test methods for failure detection, like current and voltage waveforms during ESD stress and direct current leakage are insufficient for MEMS. A functionality-based approach using the mechanical response of the MEMS during ESD is needed to accurately and conclusively detect ESD failure in MEMS. A novel test setup with a probe-mountable human body model (HBM) tester is presented for this purpose. This setup can perform simultaneous measurements of MEMS out-of-plane displacement, HBM current and HBM voltage in the MEMS in situ during ESD stress. Using this setup, a few examples are demonstrated that show that traditional electrical characterization is overestimating the ESD robustness of MEMS devices. ESD testing of MEMS is performed at different pressures on RF MEMS actuators and show that more than one type of failure mechanism can occur due to ESD stress. ESD-induced charging and functionality degradation in RF MEMS actuators are also briefly discussed.

The mechanical response of electrostatically-actuated MEMS to ESD stress leads to contact breakdown or to discharges across micro-gaps. This is the root cause of most MEMS failures under ESD stress. This paper discusses improvement of the intrinsic ESD robustness of SiGe MEMS from Class0 (<;250V) to more than Class1 (>;500V), through smart design variations and higher mechanical stiffness. A MEMS-based one-time ESD-protection fuse with pre-defined trigger voltage is shown as an application.

With the ever-growing demand for faster and greater computational power and increased functionality and intelligence in electronic products, next generation devices and systems are undergoing a vast transformation in a fundamental sense. Gone are the days of traditional devices based on integrated circuit technology alone and operating solely in the electric domain. New concepts are emerging in the horizon which indicates the merging of multiple physical domains and technologies on a single chip. With sustainable energy and efficient health care growing into two of the most important challenges of this century, established semiconductor techniques are being reused in newer fields like solid state lighting and medical devices to extend the range and scope of current technology. These developments are driving system level heterogeneous integration and smart microsystems which in turn impose new challenges in packaging and assembly. This presentation aims to discuss some of the relevant directions in this scenario with the help of some implemented examples.

A silicon-controlled rectifier (SCR) is presented as an ESD protection device for microelectromechanical systems (MEMS) in a MEMS-on-CMOS process flow. Measurements on SiGe MEMS devices have been performed and the SCR is shown to provide the Class0 MEMS with protection levels up to 5.5kV HBM. The effect of the MEMS capacitance on the SCR robustness during ESD is investigated. Through simulations and measurements, the scope of the proposed ESD protection scheme has been evaluated. Current overshoots caused by large MEMS capacitances (>;100pF) are shown to be a potential issue in the SCR operation.

Abrasive blade dicing is the most common technique for die separation. In this work an alternative dry and non-abrasive die separation method, which is known as “Stealth dicing”, is assessed for surface-sensitive MEMS (Micro Electro Mechanical Systems) wafers. The dicing performance and capability of the system is investigated on 200mm full thickness wafers with and without MEMS structures. The diced wafers are analyzed with respect to the silicon cutting quality, possible particle contamination, the condition of functional structures and their mechanical and electrical functionality. In addition the performance and limitations of two different Stealth Dicing Engine (SDE) types, SDE01 and SDE03, are compared to each other with respect to their performance on MEMS wafer dicing.From this work design rules and proper dimensions of the scribe line can be determined. Process integration solutions, describing steps before and after the Stealth dicing process, including the contact-less dicing tape application to the wafer back side and the final die separation method by tape stretching, are presented. It was also found that the SDE03 laser with its outstanding performance in terms of process speed and separation quality can bring a breakthrough for applying this technology for MEMS wafers.

In this paper, electrostatic discharge (ESD) tests on electrostatically actuated torsional micromirrors are performed in various pressure conditions and the ESD failure levels are reported for the first time in literature. The influence of device design factors of the micromirror, such as the torsional hinge width and the actuation bottom electrode area, on the ESD sensitivity is investigated. HBM ESD stress tests are performed in standard atmospheric pressure and in vacuum. These design parameters and the ambient pressure conditions are shown to affect the ESD failure levels of the MEMS micromirrors. Two distinct failure signatures are observed for micromirrors, viz., pull-in and sticking to the bottom electrode or getting destroyed and broken under the impact of the ESD stress.

ESD tests have been performed on RF MEMS capacitive switches in different ambient pressure and environmental conditions. This is done using an integrated measurement setup which can measure out-of-plane displacement as well as current and voltage in the MEMS during an HBM ESD stress event. The effect of ESD on the switch characteristics is investigated.

The sensitivity of electrostatically actuated SiGe microelectromechanical systems to electrostatic discharge events has been investigated in this paper. Torsional micromirrors and RF microelectromechanical systems (MEMS) actuators have been used as two case studies to perform this study. On-wafer electrostatic discharge (ESD) measurement methods, such as the human body model (HBM) and machine model (MM), are discussed. The impact of HBM ESD zap tests on the functionality and behavior of MEMS is explained and the ESD failure levels of MEMS have been verified by failure analysis. It is demonstrated that electrostatic MEMS devices have a high sensitivity to ESD and that it is essential to protect them.

The mechanical response of electrostatic actuators during ESD stress has been measured in-situ and is reported for the first time. Two different breakdown mechanisms during ESD - air-gap breakdown and dielectric breakdown - have been distinguished. This has been achieved by simultaneously measuring electrical and mechanical quantities on the MEMS device under test.

The response of electrostatic actuators to electrical overstress is studied in this paper. More specifically, the use of passive components (resistors and capacitors) to diminish the harmful effects of electrical overstress is demonstrated. The influence of these passives on the normal operation of the devices is also studied.

Since a laser Doppler vibrometer (LDV) allows one to directly measure mechanical quantities, this technique is particularly well suited for characterising micro-electromechanical systems (MEMS) and for performing long-term reliability tests on such devices. In the work presented here, a microscope-mounted system is used at different stages of the MEMS device qualification process. This is illustrated by a wide range of applications.Mounted on a semi-automatic probestation, the system is used for mapping the within-wafer uniformity of several parameters of MEMS devices, such as pull-in voltage, out-of-plane displacement or resonance frequency. In turn, these quantities can be translated into material properties such as Young's modulus, residual stress and stress gradient. The same LDV system was also used on a manual prober for reliability testing of MEMS devices. Since scanning LDV measurements are performed by successively pointing a laser beam at a predefined set of locations, it is of the utmost importance to have no movement of the MEMS device relative to the laser source. A compensation algorithm was developed in order to compensate for the inevitable drift that occurs over extended periods of time.Electrostatic discharge (ESD) testing is another application to which the LDV system was adapted For this purpose, a Human Body Model (HBM) ESD tester was homemade. The tester is small enough to fit inside a vacuum probe station, while the LDV system is mounted above a measurement window. This setup allows electrical as well as mechanical measurements to be performed during ESD events on MEMS devices. As LDV measurements can be performed through translucent media, measurements can even be performed on MEMS devices after packaging, provided that transparent capping materials are used Hermeticity of wafer-level packaged MEMS devices was evaluated in this manner.

In this paper various non-standard methods and instruments for the functional, yield and reliability analysis of MEMS are discussed. Most of these methods are based on existing instruments, involving electrical, optical or mechanical measurements. We present either alternative applications of existing techniques, new methodology for data extraction, or adaptation/automation of the techniques for automatic chip or wafer level measurements.

The electrical characterization of devices and circuits regarding their electrostatic discharge (ESD) robustness is done by using several measurement tools. Transmission line pulsing (TLP) and human body model (HBM) testing are the commonly used methods. In this paper, TLP and HBM on-wafer setups are presented regarding their electrical schematics, the type of data that is obtained, and the required calibration methodologies. By using three case studies, both test methods are compared by showing their advantages and disadvantages. It is demonstrated that pulsed measurement methods like TLP testing are not always a suitable tool to fully assess the ESD performance of devices or circuits.

In this work, the mechanical response of electrostatic MEMS actuators during ESD stress has been measured and reported for the first time. The failure mechanism of the actuators during ESD has been studied and compared with failure under low frequency (~DC) voltage overstress. Electrical and mechanical failure modes have been distinguished and correlated to enable better understanding of the failure physics. Measuring the mechanical response during ESD stress tests has been demonstrated to be very important to characterize the reliability of electrostatic MEMS actuators. An experimental set-up for the same has been demonstrated and compared with a conventional ESD tester.

For the first time a study of electrostatically actuated torsional micromirrors under ESD stress is presented. Together with on-wafer ESD measurements their electrical and mechanical behavior under ESD stress is explained. HBM and MM measurements have been performed and their relevance in calculating the ESD failure level of the micromirrors is compared and discussed in detail. It is demonstrated that these optical MEMS devices have a high sensitivity to ESD and that they need ESD protection.