Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/89873
Title: Pulsed laser scan methodology for single event effect (SEE) qualification
Authors: Chua, Chung Tah
Keywords: DRNTU::Engineering::Materials::Microelectronics and semiconductor materials
DRNTU::Science::Physics::Radiation physics
Issue Date: 2019
Source: Chua, C. T. (2019). Pulsed laser scan methodology for single event effect (SEE) qualification. Doctoral thesis, Nanyang Technological University, Singapore.
Abstract: This thesis addresses the need for alternative methods of single event effect (SEE) radiation testing in a period when microelectronic technology nodes are scaling to ever smaller dimensions. This results in increasingly complex devices and compels radiation test engineers to demand radiation test techniques which are capable of providing more information (both spatially and temporally) as compared to conventional broad beam heavy ion radiation test. Moreover, reliability and lifetime of advanced microelectronic devices are getting lower and thus, this thesis aims to study the influence of device ageing on its SEE radiation response. These issues are further complicated by demands for efficient radiation test methods to cope with the industry-wide shift towards the use of large amount of commercially-off-the-shelf (COTS) components which are not manufactured with radiation hardness and high reliability as primary priorities. This thesis initiates with an introduction of the space radiation environment and its associated radiation damage on semiconductor materials, with particular focus on a type of spontaneous radiation damage called SEE. Through the discussion on the overview of Singapore’s small satellite landscape and limited land area, this thesis first aims to establish an efficient radiation test methodology, based on the proposed pulsed laser scanning technique, in a laboratory setting. Details of the considerations made for the choice of scanning microscope system, laser source and optical delivery fiber are elaborated. Characterization of the laser spot size and measurement of pulsed laser energy stability are reported. Pulsed laser SEE tests were conducted on two separate devices fabricated with bipolar and CMOS process technologies, in order to benchmark this proposed scan system to other existing facilities. The second objective of this thesis is to study the effects of ageing on SEE sensitivity in old and advanced microelectronic devices. LM124 operational amplifier, which was fabricated with old bipolar process technology, was investigated via use of various load resistors in order to vary the device’s output current. Changes in electrical parameters were monitored but little change in the SEE sensitivity of this device was observed after 1002 hours of electrical and thermal stress. In contrast for advanced microelectronic devices, significant change (at least doubling) in the SEE sensitivity of 65 nm CMOS chain of flip flops was observed after 130 hours of voltage stress. The third and final objective of this thesis is to demonstrate the applications of the proposed pulsed laser SEE scan methodology and explore its advantages and possible areas of concern. This methodology was shown to be capable of assessing the various SEE sensitivities of LM124 in different bias conditions. In addition, the advantage of pulsed laser SEE scan methodology in localizing SEE-sensitive regions and applying mitigation schemes were demonstrated on PIC16 microcontroller. Results revealed SEE-sensitive structures and subsequent laser probing technique identified the source of sensitivity. Mitigation was carried out and SEE sensitivity was eventually resolved. For all of its benefits, the technique is not without vulnerabilities. A heavy ion test campaign was conducted on test structures containing chains of inverters. Pulsed laser test was conducted in an attempt to further understand the heavy ion results but no meaningful results were obtained. Further analysis via laser probing technique and frequency tests revealed issues with test structure design. Subsequent test board re-design improved bandwidth of the test setup but still fell short of observing meaningful results. In such a case, the laser scan methodology can be used as a screening test before heavy ion test campaign in order to identify possible bottlenecks in chip or test board design. Lastly, the proposed pulsed laser SEE scan methodology was demonstrated to induce damage in advanced microelectronic devices. A series of optical failure analysis and fault localization techniques were utilized to pinpoint the location of the laser-induced damage. Thereafter, focused ion beam cross-sectioning, transmission electron microscopy and energy dispersive x-ray spectroscopy found that the damage was related to nickel (in silicide layer) diffusion towards the gate terminal of the transistor. In overall, this proposed methodology has shown the potential to support radiation testing, especially in Singapore where limited land area renders particle accelerators as an impractical method of test. This methodology builds on existing laser system by employing the scan methodology to improve efficiency and repeatability. On the other hand, this thesis also highlights the potential areas of concern in applying this methodology on advanced microelectronic devices as possible laser-induced damage and importance of test board design has to be carefully considered.
URI: https://hdl.handle.net/10356/89873
http://hdl.handle.net/10220/47728
DOI: 10.32657/10220/47728
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:MSE Theses

Files in This Item:
File Description SizeFormat 
CT Thesis Full vFinal.pdfThesis vFinal9.22 MBAdobe PDFThumbnail
View/Open

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.