Plasma Etching Technology for Advanced Packaging to Lead Moore’s Law

2022 ◽  
Vol 25 (1) ◽  
pp. 47-53
Author(s):  
2016 ◽  
Vol 2016 (DPC) ◽  
pp. 001302-001327 ◽  
Author(s):  
Tom Swarbrick ◽  
Keith Best ◽  
Casey Donaher ◽  
Steve Gardner

Advanced packaging technologies continue to enable the semiconductor industry to meet the needs for ever thinner, smaller and faster components required in mobile devices and other high performance applications. However, the increase in chip I/O count, driven by Moore's law, and the ability to produce FinFETs below 10nm has presented numerous additional challenges to the existing advanced packaging processes. Furthermore, unlike Moore's law, which predicted the number of transistors in a dense integrated circuit to double approximately every two years, advanced packaging is experiencing an alternate “law”; where instead of the number of transistors increasing, it is the number of functions increasing, within the ever decreasing volume constraints of the final product that drives the technology roadmap. Inevitably, as functionality increases, so does the process complexity and cost. And in the very cost sensitive advanced packaging arena, Outsourced Semiconductor Assembly and Test suppliers (OSATs) need to compensate by reducing their manufacturing costs. This requires the OSAT to reduce material costs, increase throughput, yield, and look for new ways to reduce the number of process steps. One of the ways in which the OSATs have reduced the cost of materials is by removing the silicon wafer from the backend processing altogether; using epoxy mold compound (EMC) to create reconstituted wafers, or by using glass carriers. In the case of glass carriers, it is often the case, where the dice are attached face down on the carrier and subsequent processing prevents the front side patterns from being visible from the top side of the composite stack, even with Infrared (IR) imaging. In this particular case, an additional lithography “clear out” window is defined in photoresist over the alignment mark so the opaque film can be etched away from the alignment mark, the resist is then stripped and cleaned. This additional processing is obviously costly and time consuming. This paper specifically focuses on the concepts, methodology, and performance of a stepper based photolithography solution that utilizes a photoresist latent image to provide temporary alignment marks for the lithography process, removing the need for the additional patterning and etching steps. This revolutionary system employs a backside camera, to align to die through the carrier. A separate exposure unit, calibrated to the alignment camera center, exposes temporary latent image targets which are then detected by the system's regular alignment system during the normal stepper lithography operation. The performance data for the alignment, overlay, and latent image depth control are discussed in detail. The final analysis proves that overlay of < 2um is readily achievable, with no impact on system throughput.


2016 ◽  
Vol 2016 (1) ◽  
pp. 000315-000320 ◽  
Author(s):  
Keith Best ◽  
Steve Gardner ◽  
Casey Donaher

Abstract Advanced packaging technologies continue to enable the semiconductor industry to meet the needs for ever thinner, smaller and faster components required in mobile devices and other high performance applications. However, the increase in chip I/O count, driven by Moore's law, and the ability to produce FinFETs below 10nm have presented numerous additional challenges to the existing advanced packaging processes. Furthermore, unlike Moore's law, which predicted the number of transistors in a dense integrated circuit to double approximately every two years, advanced packaging is experiencing an alternate “law”; where instead of the number of transistors increasing, it is the number of functions increasing, within the ever decreasing volume constraints of the final product that drives the technology roadmap. Inevitably, as functionality increases, so does the process complexity and cost. In this very cost sensitive advanced packaging arena, outsourced semiconductor assembly and test suppliers (OSATs) need to compensate by reducing their manufacturing costs. This requires the OSAT to reduce material costs, increase throughput, yield, and look for new ways to reduce the number of process steps. One of the ways in which the OSATs have reduced the cost of materials is by removing the silicon wafer from back-end processing altogether; using epoxy mold compound (EMC) to create reconstituted wafers, or by using glass carriers. In the case of glass carriers, it is often the case, that the dice are attached face down on the carrier and subsequent processing prevents the front side patterns from being visible from the top side of the composite stack, even with infrared (IR) imaging. In this particular case, an additional lithography “clear out” window is defined in photoresist over the alignment mark so the opaque film can be etched away from the alignment mark, the resist stripped, and the lithography layer reworked. This additional processing is obviously costly and time consuming. This paper specifically focuses on the concepts, methodology, and performance of a stepper-based photolithography solution that utilizes a photoresist latent image to provide temporary alignment marks for the lithography process, removing the need for the additional patterning and etching steps. This revolutionary system employs a backside camera, to align to die through the carrier. A separate exposure unit, calibrated to the alignment camera center, exposes temporary latent image targets which are then detected by the system's regular alignment system during the normal stepper lithography operation. The performance data for the alignment, overlay, and latent image depth control are discussed in detail. The final analysis proves that overlay of < 2μm is readily achievable, with no impact on system throughput.


Author(s):  
David Segal

Chapter 3 highlights the critical role materials have in the development of digital computers. It traces developments from the cat’s whisker to valves through to relays and transistors. Accounts are given for transistors and the manufacture of integrated circuits (silicon chips) by use of photolithography. Future potential computing techniques, namely quantum computing and the DNA computer, are covered. The history of computability and Moore’s Law are discussed.


Author(s):  
Daniel Pargman ◽  
Aksel Biørn-Hansen ◽  
Elina Eriksson ◽  
Jarmo Laaksolahti ◽  
Markus Robèrt
Keyword(s):  

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Prasanna Date ◽  
Davis Arthur ◽  
Lauren Pusey-Nazzaro

AbstractTraining machine learning models on classical computers is usually a time and compute intensive process. With Moore’s law nearing its inevitable end and an ever-increasing demand for large-scale data analysis using machine learning, we must leverage non-conventional computing paradigms like quantum computing to train machine learning models efficiently. Adiabatic quantum computers can approximately solve NP-hard problems, such as the quadratic unconstrained binary optimization (QUBO), faster than classical computers. Since many machine learning problems are also NP-hard, we believe adiabatic quantum computers might be instrumental in training machine learning models efficiently in the post Moore’s law era. In order to solve problems on adiabatic quantum computers, they must be formulated as QUBO problems, which is very challenging. In this paper, we formulate the training problems of three machine learning models—linear regression, support vector machine (SVM) and balanced k-means clustering—as QUBO problems, making them conducive to be trained on adiabatic quantum computers. We also analyze the computational complexities of our formulations and compare them to corresponding state-of-the-art classical approaches. We show that the time and space complexities of our formulations are better (in case of SVM and balanced k-means clustering) or equivalent (in case of linear regression) to their classical counterparts.


2015 ◽  
Vol 59 (1) ◽  
pp. 33-35 ◽  
Author(s):  
Michael A. Cusumano ◽  
David B. Yoffie
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document