Electrical and optical interconnects for high-performance computing

Technology scaling has enabled drastic growth in the computational and storage capacity of integrated circuits (ICs). This constant growth drives an increasing demand for high-bandwidth communication between and within ICs. In this dissertation we focus on low-power solutions that address this demand. We divide communication links into three subcategories depending on the communication distance. Each category has a different set of challenges and requirements and is affected by CMOS technology scaling in a different manner. We start with short-range chip-to-chip links for board-level communication. Next we will discuss board-to-board links, which demand a longer communication range. Finally on-chip links with communication ranges of a few millimeters are discussed.

Electrical signaling is a natural choice for chip-to-chip communication due to efficient integration and low cost. IO data rates have increased to the point where electrical signaling is now limited by the channel bandwidth. In order to achieve multi-Gb/s data rates, complex designs that equalize the channel are necessary. In addition, a high level of parallelism is central to sustaining bandwidth growth. Decision feedback equalization (DFE) is one of the most commonly employed techniques to overcome the limited bandwidth problem of the electrical channels. A linear and low-power summer is the central block of a DFE. Conventional approaches employ current-mode techniques to implement the summer, which require high power consumption. In order to achieve low-power operation we propose performing the summation in the charge domain. This approach enables a low-power and compact realization of the DFE as well as crosstalk cancellation. A prototype receiver was fabricated in 45nm SOI CMOS to validate the functionality of the proposed technique and was tested over channels with different levels of loss and coupling. Measurement results show that the receiver can equalize channels with maximum 21dB loss while consuming about 7.5mW from a 1.2V supply. We also introduce a compact, low-power transmitter employing passive equalization. The efficacy of the proposed technique is demonstrated through implementation of a prototype in 65nm CMOS. The design achieves up to 20Gb/s data rate while consuming less than 10mW.

An alternative to electrical signaling is to employ optical signaling for chip-to-chip interconnections, which offers low channel loss and cross-talk while providing high communication bandwidth. In this work we demonstrate the possibility of building compact and low-power optical receivers. A novel RC front-end is proposed that combines dynamic offset modulation and double-sampling techniques to eliminate the need for a short time constant at the input of the receiver. Unlike conventional designs, this receiver does not require a high-gain stage that runs at the data rate, making it suitable for low-power implementations. In addition, it allows time-division multiplexing to support very high data rates. A prototype was implemented in 65nm CMOS and achieved up to 24Gb/s with less than 0.4pJ/b power efficiency per channel. As the proposed design mainly employs digital blocks, it benefits greatly from technology scaling in terms of power and area saving.

As the technology scales, the number of transistors on the chip grows. This necessitates a corresponding increase in the bandwidth of the on-chip wires. In this dissertation, we take a close look at wire scaling and investigate its effect on wire performance metrics. We explore a novel on-chip communication link based on a double-sampling architecture and dynamic offset modulation technique that enables low power consumption and high data rates while achieving high bandwidth density in 28nm CMOS technology. The functionality of the link is demonstrated using different length minimum-pitch on-chip wires. Measurement results show that the link achieves up to 20Gb/s of data rate (12.5Gb/s/$\mu$m) with better than 136fJ/b of power efficiency.

共焦扫描系统扫描深度与厚度失真分析

采用几何光学方法对共焦系统扫描时产生的扫描深度与厚度失真进行了理论分析，对名义扫描深度与实际扫描深度之间的关系进行了研究．以若丹明6G薄膜与玻片组成的多层样品为模型，对其进行了模拟计算，得到了扫描深度与厚度失真与系统数值孔径、折射率和样品厚度之间的关系．在实验上分别采用单光子荧光和双光子荧光作为检测信号，在反射式共焦扫描系统上进行了纵向扫描实验，并与模拟计算的结果进行了比较和分析．

Dynamic storage performance of two cyanine dye films and optimization design for a recordable compact disk

The optical constants of two cyanine dye films that we prepared were measured with a RAP-1-type (RAP is rotating analyzer and polarizer) spectroscopic ellipsometer. Toward making a simplified model for the wafers of a recordable compact disk (CD-R), we give their optimization designs developed with the cyanine dye films. in addition, the dynamic storage performances of two sample disks were tested by our dynamic storage testing system. Measurement results of the sample disks were obtained to test and verify our film designs. (C) 2000 Optical Society of America. OCIS codes: 160.4890, 160.4760, 210.4810.

靶镜光学质量检测技术的研究

提出了一种利用扫描型哈特曼检测装置检验靶镜光学质量的技术.该装置对传统哈特曼检验装置的光阑进行了改进,通过扫描型哈特曼光阑的旋转扫描,可对被检靶镜全口径范围内连续采样.利用该扫描型哈特曼检测装置对一块口径为φ270 mm的非球面靶镜的能量集中度和波像差进行了检验,其结果与激光数字波面干涉仪的测量结果相吻合,其中能量集中度的相对测量误差为7.7%,波像差的相对测量误差为10.2%,验证了该检测技术的有效性.

双频激光干涉仪的最大可测量速度研究

现代制造业对双频激光干涉仪的最大可测量速度提出了越来越高的要求。最大可测量速度是双频激光干涉仪的一项重要指标，它主要受双频激光光源所输出的频差、干涉仪的光学结构以及电子带宽等因素的限制。本文从理论和实验两方面对干涉仪的最大可测量速度进行了研究，搭建了基于自由落体运动的实验装置。实验结果表明，实际最大可测量速度略低于其理论值。另外，文中还分析了上述三种因素对最大可测量速度的影响。实验装置和结果可供工业应用提供参考。

Yb：YAG晶体的红外闪烁特性

用脉冲电子束激发测量了不同Yb^3+掺杂浓度的Yb：YAG晶体的红外(IR)闪烁发光性能。Yb：YAG晶体的IR闪烁发光具有高的光产额和长的衰减时间，但存在浓度猝灭效应和温度依赖关系。Yb：YAG晶体的IR闪烁性能还与晶体品质有关，相同掺杂浓度的Yb：YAG晶体，品质优异的会获得更高的光产额。这一初步的研究成果表明，部分掺Yb^3+晶体有可能用于医学成像装置。

Diffraction at 8-keV x-rays of Bragg-Fresnel optics

The basic elements of Bragg-Fresnel multilayer optics are multilayer gratings. In this paper, the fabrication process for the simplest Bragg-Fresnel multilayer optics is given; the diffraction measurements at 8-keV x rays of multilayer gratings are presented; and the measurement results in the diffraction spectrum are analyzed and discussed in detail. (C) 1996 Society of Photo-Optical Instrumentation Engineers.

Enhancement of near-band edge photoluminescence of ZnO thin films by employing MgF2 buffer layer

ZnO/MgF2/ZnO sandwich structure films were fabricated. The effects of a buffer layer on structure and optical properties of ZnO films were investigated by X-ray diffraction, photoluminescence, optical transmittance and absorption measurements. Measurement results showed that the buffer layer had the effects of improving the quality of ZnO films and releasing the residual stresses in the films. The near-band edge emissions of ZnO films deposited on the MgF2 buffer layer were significantly enhanced compared with those deposited on bare substrate due to the smaller lattice mismatch between MgF2 and ZnO than that between fused silica and ZnO. (c) 2006 Elsevier B.V. All rights reserved.

ZnO : Zn phosphor thin films prepared by face-to-face annealing

ZnO:Zn phosphor thin films were prepared by face-to-face annealing at 450 degrees C in air. The effects of the face-to-face annealing on the structural and optical properties of the ZnO films were investigated by X-ray diffraction (XRD), photoluminescence (PL), optical transmittance and absorption measurements. Measurement results showed that the crystal quality of ZnO films was improved by face-to-face annealing. Both UV light emission and visible light emission were enhanced compared to those of open annealing films. The UV emission peak was observed to have a blueshift towards higher energy. The optical band-gap edge of as-annealed films shifted towards longer wavelength. (c) 2005 Elsevier B.V.. All rights reserved.

Techniques for fabrication of large-size graded reflectivity mirrors

According to the parameter requirements of a graded reflectivity mirror with a Gaussian profile, the layer structure and the mask pattern are designed using a graded-thickness middle layer. The mask and the automatic mask-switchover equipment are designed considering the actual requirement of the thin films and the specific deposit facility. The uniformity of the layer thickness is analyzed. The measurement results indicate that samples prepared with this technique are basically in accordance with the design parameter. The scattering effect between the material molecules and the mask, thickness errors, and the alignment error between the mask and the substrate are the main factors that influence the deposit result. (c) 2008 Society of Photo-Optical Instrumentation Engineers. [DOI: 10.1117/1.3027595]

Effects of nonuniform reactant stoichiometry on combustion instability

This paper analyzes the forced response of swirl-stabilized lean-premixed flames to acoustic forcing in a laboratory-scale stratified burner. The double-swirler, double-channel annular burner was specially designed to generate acoustic velocity oscillations and radial fuel stratification at the inlet of the combustion chamber. Temporal oscillations of equivalence ratio along the axial direction are dissipated over a long distance, and therefore the effects of time-varying fuel/air ratio on the flame response are not considered. Simultaneous measurements of inlet velocity and heat release rate oscillations were made using a hot wire anemometer and photomultiplier tubes with narrowband OH*/CH* interference filters. Time-averaged CH* chemiluminescence intensities were measured using an intensified CCD camera. Results show that flame stabilization mechanisms vary depending on stratification ratio for a constant global equivalence ratio. For a uniformly premixed condition, an enveloped M-shaped flame is observed. For stratified conditions, however, a dihedral V-flame and a detached flame are developed for outer stream and inner stream fuel enrichment cases, respectively. Flame transfer function (FTF) measurement results indicate that a V-shaped flame tends to damp incident flow oscillations, while a detached flame acts as a strong amplifier relative to the uniformly premixed condition. The phase difference of FTF increases in the presence of stratification. More importantly, the dynamic characteristics obtained from the forced stratified flame measurements are well correlated with unsteady flame behavior under limit-cycle pressure oscillations. The results presented in this paper provide insight into the impact of nonuniform reactant stoichiometry on combustion instabilities, which has not been well explored to date. Copyright © 2011 by ASME.

Silicon pyramids for plasmonic enhanced schottky photodetectors

We demonstrate the design, fabrication and characterization of plasmonic enhanced free space Schottky detector for telecom wavelength. Unique fabrication technique, simulation and measurement results will be presented and discussed. © OSA 2013.

A high temperature and low power SOI CMOS MEMS based thermal conductivity gas sensor

The design, 3D FEM modelling and measurement results of a novel high temperature, low power SOI CMOS MEMS thermal conductivity gas sensor are presented here. The sensor consists of a circular membrane with an embedded tungsten micro-heater. The high sensing capability is based on the temperature sensitivity of the resistive heating element. The sensor was fabricated at a commercial foundry using a 1 μm process and measures only 1×1 mm 2. The circular membrane has a 600 μm diameter while the heating element has a 320 μm diameter. Measurement results show that for a constant power consumption of 75 mW the heater temperature was 562.4°C in air, 565.9°C in N2, 592.5°C for 1 % H2 in Ar and 599.5°C in Ar. © 2013 IEEE.

High efficiency and broad bandwidth grating coupler between nanophotonic waveguide and fibre

A high efficiency and broad bandwidth grating coupler between a silicon-on-insulator (SOI) nanophotonic waveguide and fibre is designed and fabricated. Coupling efficiencies of 46% and 25% at a wavelength of 1.55 mu m are achieved by simulation and experiment, respectively. An optical 3 dB bandwidth of 45 nm from 1530 nm to 1575 nm is also obtained in experiment. Numerical calculation shows that a tolerance to fabrication error of 10 nm in etch depth is achievable. The measurement results indicate that the alignment error of +/-2 mu m results in less than 1 dB additional coupling loss.

A photonic wire-based directional coupler based on SOI

A photonic wire-based directional coupler based on SOI was fabricated by e-beam lithography (EBL) and the inductively coupled plasma (ICP) etching method. The size of the sub-micron waveguide is 0.34 mu m x 0.34 mu m, and the length in the coupling region and the separation between the two parallel waveguides are 410 and 0.8 mu m, respectively. The measurement results are in good agreement with the results simulated by 3D finite-difference time-domain method. The transmission power from two output ports changed reciprocally with about 23 nm wavelength spacing between the coupled and direct ports. The extinction ratio of the device was between 5 and 10 dB, and the insertion loss of the device in the wavelength range 1520-1610 nm was between 22 and 24 dB, which included an about 18.4 +/- 0.4 dB coupling loss between the taper fibers and the polished sides of the device. (c) 2008 Elsevier B.V. All rights reserved.