The International Technology Roadmap for Semiconductors challenges the device research community to reduce the transistor footprint containing all components to 40 nanometers within the next decade. We report on a p-channel transistor scaled to such an extremely small dimension. Built on one semiconducting carbon nanotube, it occupies less than half the space of leading silicon technologies, while delivering a significantly higher pitch-normalized current density—above 0.9 milliampere per micrometer at a low supply voltage of 0.5 volts with a subthreshold swing of 85 millivolts per decade. Furthermore, we show transistors with the same small footprint built on actual high-density arrays of such nanotubes that deliver higher current than that of the best-competing silicon devices under the same overdrive, without any normalization. We achieve this using low-resistance end-bonded contacts, a high-purity semiconducting carbon nanotube source, and self-assembly to pack nanotubes into full surface-coverage aligned arrays.

https://science.sciencemag.org/content/sci/356/6345/1369.full.pdf

Carbon nanotube transistors scaled to a 40-nanometer footprint

Scaling trends of FinFET architecture, with focus on Front-End-of-Line (FEOL), and Middle-of-Line (MOL) device parameters, is systematically investigated. It is concluded that the combined requirements of device electrostatics together with the demands on contact resistance, presents a Contacted-Gate-Pitch (CGP) scaling limit for horizontal-transport FETs. FET drive is expected to significantly degrade below this CGP ~ 40 nm as a result. Good agreement between hardware data and TCAD simulations is achieved and employed to estimate the contact resistance values for aggressively scaled FinFETs. These observations show that FinFETs scaled below CGP of 40 nm will require the contact resistivity (ρ
 C
) of ~8 × 10
 -10
 Ω-cm
 2
, while fully ohmic contacts i.e., ρ
 C
 of ~1 × 10
 -10
 Ω-cm
 2
 will be required if FinFETs are to extend performance below CGP of 30 nm. Ultimately, transition to new device architectures in which contact area is independent of CGP and/or Fin-Pitch will be necessary.

Challenges and Limitations of CMOS Scaling for FinFET and Beyond Architectures

The world’s appetite for analyzing massive amounts of structured and unstructured data has grown dramatically. The computational demands of these abundant-data applications, such as deep learning, far exceed the capabilities of today’s computing systems and are unlikely to be met with isolated improvements in transistor or memory technologies, or integrated circuit architectures alone. To achieve unprecedented functionality, speed, and energy efficiency, one must create transformative nanosystems whose architectures are based on the salient properties of the underlying nanotechnologies. Our Nano-Engineered Computing Systems Technology (N3XT) approach makes such nanosystems possible through new computing system architectures leveraging emerging device (logic and memory) nanotechnologies and their dense 3-D integration with fine-grained connectivity to immerse computing in memory and new logic devices (such as carbon nanotube field-effect transistors for implementing high-speed and low-energy logic circuits) as well as high-density nonvolatile memory (such as resistive memory), and amenable to ultradense (monolithic) 3-D integration of thin layers of logic and memory devices that are fabricated at low temperature. In addition, we explore the use of several device and integration technologies in the N3XT beyond the specific ones mentioned earlier that are also used in our main nanosystem prototypes. We also present an efficient resiliency technique to overcome endurance challenges in certain resistive memory technologies. N3XT hardware prototypes demonstrate the practicality of our architectures. We evaluate the benefits of the N3XT using a simulation framework calibrated using experimental measurements. System-level energy-delay product of common implementations of abundant-data workloads improves by three orders of magnitude in the N3XT compared with conventional architectures. These improvements impact a broad range of application workloads and architecture configurations, from embedded systems to the cloud.

The N3XT Approach to Energy-Efficient Abundant-Data Computing

This paper describes a full-entropy 128-b key generation platform based on a 1024-b hybrid physically unclonable function (PUF) array, fabricated in 14-nm trigate high-k/metal-gate CMOS. Delay-hardened hybrid PUF cells use differential clock delay insertion to favor circuit evaluation in the desired direction while leveraging burn-in-induced aging for selective bit destabilization enabling quick identification and masking of unstable cells, and subsequent temporal-majority-voting with soft dark-bit masking to reduce PUF bit error by 3.9 times to 1.45% resulting in ~5 ppb failure probability. A stable full-entropy 128-b key is finally generated from the 1024 raw PUF bits using BCH error correction and AES-CBC-based entropy extraction. An all-digital design with compact PUF cell layout occupying $1.84~\mu \text{m}^{2}$ achieves: 1) 4-fJ/b energy-efficiency with 3-μW leakage at 0.65 V, 70 °C; 2) peak operating frequency of 1 GHz resulting in 1.2-μs key generation latency; 3) robust operation with stable key generation across 0.55–0.75 V, and 25 °C–110 °C; 4) 14 times separation between intra/inter-PUF hamming distances with 0.99993 entropy ensuring cryptographic quality randomness and uniqueness; 5) 48% higher PUF stability with long-term aging by leveraging transistor degradation to reinforce favorable cell bias; and 6) resiliency to power cycling attacks with common centroid clock routing measured from 49.5% hamming distance between array’s evaluation and wake-up states.

A 4-fJ/b Delay-Hardened Physically Unclonable Function Circuit With Selective Bit Destabilization in 14-nm Trigate CMOS

This paper describes architectural enhancements in the Altera Stratix? 10 HyperFlex? FPGA architecture, fabricated in the Intel 14nm FinFET process. Stratix 10 includes ubiquitous flip-flops in the routing to enable a high degree of pipelining. In contrast to the earlier architectural exploration of pipelining in pass-transistor based architectures, the direct drive routing fabric in Stratix-style FPGAs enables an extremely low-cost pipeline register. The presence of ubiquitous flip-flops simplifies circuit retiming and improves performance. The availability of predictable retiming affects all stages of the cluster, place and route flow. Ubiquitous flip-flops require a low-cost clock network with sufficient flexibility to enable pipelining of dozens of clock domains. Different cost/performance tradeoffs in a pipelined fabric and use of a 14nm process, lead to other modifications to the routing fabric and the logic element. User modification of the design enables even higher performance, averaging 2.3X faster in a small set of designs.

The Stratix™ 10 Highly Pipelined FPGA Architecture

A leading edge 14 nm SoC platform technology based upon the 2nd generation Tri-Gate transistor technology [5] has been optimized for density, low power and wide dynamic range. 70 nm gate pitch, 52 nm metal pitch and 0.0499 um2 HDC SRAM cells are the most aggressive design rules reported for 14/16 nm node SoC process to achieve Moore's Law 2x density scaling over 22 nm node. High performance NMOS/PMOS drive currents of 1.3/1.2 mA/um, respectively, have been achieved at 0.7 V and 100 nA/um off-state leakage, 37%/50% improvement over 22 nm node. Ultra-low power NMOS/PMOS drives are 0.50/0.32 mA/um at 0.7 V and 15pA/um Ioff. This technology also deploys high voltage I/O transistors to support up to 3.3 V I/O. A full suite of analog, mixed-signal and RF features are also supported.

A 14 nm SoC platform technology featuring 2 nd generation Tri-Gate transistors, 70 nm gate pitch, 52 nm metal pitch, and 0.0499 um 2 SRAM cells, optimized for low power, high performance and high density SoC products

Semiconductor device(s) including a transistor with a gate electrode having a work function monotonically graduating across the gate electrode length, and method(s) to fabricate such a device. In embodiments, a gate metal work function is graduated between source and drain edges of the gate electrode for improved high voltage performance. In embodiments, thickness of a gate metal graduates from a non-zero value at the source edge to a greater thickness at the drain edge. In further embodiments, a high voltage transistor with graduated gate metal thickness is integrated with another transistor employing a gate electrode metal of nominal thickness. In embodiments, a method of fabricating a semiconductor device includes graduating a gate metal thickness between a source end and drain end by non-uniformly recessing the first gate metal within the first opening relative to the surrounding dielectric.

Transistor gate metal with laterally graduated work function

A MOS antifuse with an accelerated dielectric breakdown induced by a void or seam formed in the electrode. In some embodiments, the programming voltage at which a MOS antifuse undergoes dielectric breakdown is reduced through intentional damage to at least part of the MOS antifuse dielectric. In some embodiments, damage may be introduced during an etchback of an electrode material which has a seam formed during backfilling of the electrode material into an opening having a threshold aspect ratio. In further embodiments, a MOS antifuse bit-cell includes a MOS transistor and a MOS antifuse. The MOS transistor has a gate electrode that maintains a predetermined voltage threshold swing, while the MOS antifuse has a gate electrode with a void accelerated dielectric breakdown.

Mos antifuse with void-accelerated breakdown

An embodiment includes an apparatus comprising: a non-planar fin having first, second, and third portions each having major and minor axes and each being monolithic with each other; wherein (a) the major axes of the first, second, and third portions are parallel with each other, (b) the major axes of the first and second portions are non-collinear with each other, (c) each of the first, second, and third portions include a node of a transistor selected from the group comprising source, drain, and channel, (e) the first, second, and third portions comprise at least one finFET. Other embodiments are described herein.

Non-linear fin-based devices

IC device structures including a lateral compound resistor disposed over a surface of a substrate, and fabrication techniques to form such a resistor in conjunction with fabrication of a transistor. Rather than being stacked vertically, a compound resistive trace may include a plurality of resistive materials arranged laterally over a substrate. Along a resistive trace length, a first resistive material is in contact with a sidewall of a second resistive material. A portion of a first resistive material along a centerline of the resistive trace may be replaced with a second resistive material so that the second resistive material is embedded within the first resistive material.

Pei-Chi Liu

Papers

A 14 nm SoC platform technology featuring 2 nd generation Tri-Gate transistors, 70 nm gate pitch, 52 nm metal pitch, and 0.0499 um 2 SRAM cells, optimized for low power, high performance and high density SoC products

Transistor gate metal with laterally graduated work function

Mos antifuse with void-accelerated breakdown

Non-linear fin-based devices

Compound lateral resistor structures for integrated circuitry