Semiconducting lithium indium diselenide, 6 LiInSe 2 or LISe, has promising characteristics for neutron detection applications. The 95% isotopic enrichment of 6 Li results in a highly efficient thermal neutronsensitive material. In this study, we report on a proof-of-principle investigation of a semiconducting LISe pixel detector to demonstrate its potential as an efficient neutron imager. The LISe pixel detector had a 4x4 of pixels with a 550 µm pitch on a 5x5x0.56 mm 3 LISe substrate. An experimentally verified spatial resolution of 300 µm was observed utilizing a super-sampling technique. 1 INTRODUCTION Cold neutron radiography and tomography comprise powerful detection tools for investigating both natural and engineered materials. Its sensitivity for real-space imaging relies on the total neutron interaction cross section with the material under investigation, which includes both scattering (neutronnuclei and neutron-lattice) and absorption [1]. Neutron imaging shares many similarities with X-ray imaging systems, albeit the mode of interaction of the interrogating beam is different. The comparison between these two imaging techniques is outside the scope of this study, but many references are available [1-3]. In the advancement of the capabilities of neutron imaging facilities, it has been indicated that there is a need for advances in suitable detection systems to reach the full potential of current and future neutron facilities, such as the Spallation Neutron Source (SNS) at Oak Ridge National Laboratory (ORNL). Specifically, a simultaneous enhancement of the detection efficiency (ideally 100%), spatial resolution (less than 10 µm), large signal-to-noise ratio, (good gamma-ray rejection capability), and high temporal resolution (less than 1 µs) is desired [4]. Achieving the desired performance metrics outlined is a challenging prospect, but the use of a semiconductor-based imaging plane may provide the necessary performance.

This paper describes various package related failure mechanisms observed in the plastic surface mount Ball Grid Array (BGA) package. Two types of plastic BGA packages commonly known as 225 OMPAC™ (Over Molded Pad Array Carrier) and 225 GTPAC (Glob Top Pad Array Carrier) are covered in this paper. The GTPAC is not offered as a production package, but it is used for commercial prototypes and evaluations. The failure analysis results discussed in this paper are primarily of the devices which failed at different times during various reliability and qualification testing over a period of two years. The failure analysis results of field returns (about 10% of al the devices analyzed) from customers for the same period are also included in this study. Of all the devices in the BGA packages which were failure analyzed, about 50% lailed due to package related problems. All the package related failures fall into two major categories of failure mechanisms, package delamination and cracked open copper traces on the printed circuit board (PCB). The delamination resulted in a variety of physical damage such as lifted ball bonds at the die pads, fractured bond wires in the span as well as at the heel of the crescent bonds on the PCB substrate, and cracking of the encapsulant. The copper traces cracked from two types of stresses, mechanical and thermal. In addition, some of the techniques used for the failure analysis are briefly discussed in this paper.

In the last several years emission microscopy has become an essential tool for failure analysis, specifically for VLSI devices. This paper describes various die related failure mechanisms in CMOS ASIC devices which were detected by emission microscopy. The failure analysis results discussed in this paper are primarily of the devices which were analyzed over the period of the last three years, 1994 - 1996. These devices were from a broad spectrum of final test failures, qualification and reliability test failures, special evaluation failures, testing and assembly failures at customer sites, and end user field failures. In addition to the failure mechanism statistic scanning electron micrographic illustrations of some of the failure mechanisms and associated damage are presented in this paper. The data presented in this paper clearly show the effectiveness of photon emission microscopy. The value of emission microscopy really lies in quick detection of failure locations on the die which failed functionally or due to excessive static IOD, functional IOD, or input/output leakage currents. It has certainly impacted tum around time of the analysis as significant reduction in analysis time has been achieved. In some cases same day turn around was possible.