2.5D and 3D Heterogeneous Integration: Emerging applications

Sheikh, Farhana; Nagisetty, R.; Karnik, Tanay; Kehlet, David

doi:10.1109/mssc.2021.3111386

Cited by 58 publications

(15 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RFID tag [99]. Third, the possible 3D-stacked integration can provide performance improvement by orders of magnitude with high-bandwidth sensor data readout and weight reload, which has also attracted much attention in the field of advanced CMOS-based sensors and heterogeneous-integrated computing systems [113]. Moreover, exploration space still remains for the TFT-based edge system design with data exhibition modules integrated such as displays and haptic feedbacks.…”

Section: Future Of Tft-based Cim and Edge Computing Systemmentioning

confidence: 99%

Computing-in-memory with thin-filmtransistors: challenges and opportunities

Tang¹,

Liu²,

Li³

et al. 2022

Flex. Print. Electron.

View full text Add to dashboard Cite

Thin-film transistor (TFT) has attracted enormous interests recently for its great potential in a wide range of edge computing applications, benefitting from its large-area low-cost flexible fabrications, and well integration with sensors and displays. With the support of in-situ processing of sensor data, TFT-based edge systems show their advantages in large-scale dense sensing with real-time energy-efficient processing and interaction, and more excitingly, it provides the opportunity of eliminating the massive data transfer to the cloud servers. However, the design of high-performance processing modules based on TFT is difficult, due to large device variation, poor stability, and low mobility. Computing-in-memory (CiM), which has been proposed recently as a high-efficiency high-parallelism computing approach, is expected to improve the capacity of TFT-based edge computing systems. In this survey, various recent works on TFT-based CiM have been summarized, showing the superiority to conventional processing flow by efficient in-memory analog computation with mitigation of data transfer, and reduced ADC usage for sensor data. With both opportunities and challenges, the design space and trend of TFT-based CiM to be explored are then described. Finally, further development and co-optimization from device to system are discussed for flourishing of the next-generation intelligent TFT-based edge system.

show abstract

Section: Future Of Tft-based Cim and Edge Computing Systemmentioning

confidence: 99%

Computing-in-memory with thin-filmtransistors: challenges and opportunities

Tang¹,

Liu²,

Li³

et al. 2022

Flex. Print. Electron.

View full text Add to dashboard Cite

show abstract

“…In the semiconductor industry, three-dimensional (3-D) integration technology in the 'More than Moore' era is a promising solution for meeting the increasing requirements of high transistor density and fast data transmission in the sixth generations mobile communications, Internet of Things, and artificial intelligence fields, playing a key role in advanced package technologies such as system in package, package on package, waferlevel packaging, and system on chip [1][2][3]. Bonding is a key technology for 3-D vertical interconnection.…”

Section: Introductionmentioning

confidence: 99%

Low-temperature, short-time, wafer-level bonding for Cu/Sn/Cu solid-state-diffusion interconnect in 3-D integration

Li¹,

Wang²,

Li³

et al. 2023

Phys. Scr.

View full text Add to dashboard Cite

In this paper, Cu/Sn/Cu solid-state diffusion (SSD) under low temperature is proposed and investigated for three-dimensional (3-D) integration. Cu and Sn films were deposited by high-efficiency and low-cost physical vapor deposition to fabricate 40-µm-pitch daisy-chain structures. Subsequently, the Cu bump surface was treated with Ar (5% H2) plasma. The Cu/Sn/Cu structure was bonded face to face at 200 ℃ for 15 min. The interfacial composition of the as-bonded dies comprised five layers, Cu/Cu3Sn/Cu6Sn5/Cu3Sn/Cu, with no Sn remaining and no overflow. After annealing at 200 ℃ for 15 min under N2 atmosphere, as the Cu6Sn5 completely transformed into Cu3Sn, the microstructure changed to stable three layers: Cu/Cu3Sn/Cu. Additionally, the average bonding shear strength reached 27.0 MPa, which is higher than that for conventional Cu/Sn SSD bonding. The measured bonding resistance value was maintained at the theoretical value. Moreover, the parabolic growth constant of Cu3Sn reached 1.86×10−15 m2/s. Our study demonstrates the feasibility of using Cu/Sn/Cu SSD for low-temperature, short-time, wafer-level bonding.

show abstract

“…Slow-down of Moore's law and economical concerns of yield are pushing the semiconductor industry towards heterogeneous integration, where multiple dies are interconnected through an advanced chip package, resulting in a System-in-Package (SiP). Heterogeneous integration allows a "more-than-Moore" approach that enables cutting-edge computing 6,7,8,9,10 . In such advanced packaging, the interconnect parasitics can no longer be modeled using isolated circuit models available in closed-form 11,12 requiring blackbox models for time-domain circuit simulation of complex electromagnetic systems as shown in Fig.…”

Section: Introductionmentioning

confidence: 99%

Orthogonal Rational Approximation of Transfer Functions for High-Frequency Circuits

Ma,

Engin

2022

Preprint

View full text Add to dashboard Cite

Rational function approximations find applications in many areas including macromodeling of high-frequency circuits, model order reduction for controller design, interpolation and extrapolation of system responses, surrogate models for highenergy physics, and approximation of elementary mathematical functions. The unknown denominator polynomial of the model results in a non-linear problem, which can be replaced with successive solutions of linearized problems following the Sanathanan-Koerner (SK) iteration. An orthogonal basis can be obtained based on Arnoldi resulting in a stabilized SK iteration. We present an extension of the stabilized SK, called Orthogonal Rational Approximation (ORA), which ensures real polynomial coefficients and stable poles for realizability of electrical networks. We also introduce an efficient implementation of ORA for multi-port networks based on a block QR decomposition.

show abstract

2.5D and 3D Heterogeneous Integration: Emerging applications

Cited by 58 publications

References 28 publications

Computing-in-memory with thin-filmtransistors: challenges and opportunities

Computing-in-memory with thin-filmtransistors: challenges and opportunities

Low-temperature, short-time, wafer-level bonding for Cu/Sn/Cu solid-state-diffusion interconnect in 3-D integration

Orthogonal Rational Approximation of Transfer Functions for High-Frequency Circuits

Contact Info

Product

Resources

About