NIRT: Architectures and Devices for Quantum-dot Cellular Automata

CCR-0210153 Kogge，Peer本提案是对NSF 01-157 NIRT类纳米科学与工程倡议的响应。量子点细胞自动机(QCA)是一种革命性的计算范式，非常适合于纳米电子的实现和分子维度的缩放。QCA的中心特征是二进制信息被编码在构成细胞的一组点中的单电子的位置。这代表着与基于晶体管的范例的重大突破，在该范例中，信息由晶体管电流开关的状态来编码。在QCA中，电子在细胞内的量子点之间交换，但细胞之间没有电流流动。这导致极低的功耗，避免了最终限制晶体管电路集成密度的发热问题。从体系结构和器件的角度来看，QCA电路的时钟已被证明是极其重要的。它允许QCA单元阵列被分解为子阵列进行流水线处理，并使单元能够产生信号功率增益以取代环境中损失的信号能量。已在铝/氧化物隧道结方案中演示了起作用的QCA器件，确认了QCA单元、移位寄存器、逻辑门和存储元件的操作。还实现了QCA移位寄存器的功率增益。该项目将推动QCA体系结构的发展，研究纳米电子器件中的开关速度问题，并开发先进的制造技术来实现体系结构和电路理论概念。由于QCA代表着与传统设备的戏剧性突破，因此需要在体系结构上进行重大改变，以充分利用QCA的功能。在QCA布局中，时间和架构密切相关，需要统一的设计方法。这类似于Mead和Conway开创的方法，该方法通过在架构和布局之间建立连接并在此连接上进行构建，使设计师能够快速综合大型和复杂的功能块，从而彻底改变了VLSI设计。同样，QCA系统设计者将能够利用布局之外的时序来产生高密度的功能设计。特别是，我们将研究基于简单但完整的QCA的“现场可编程门阵列”的开发，其中相同单元的2D阵列被平铺在一起，具有可编程的互连和功能。计时在QCA设计中起着至关重要的作用，因此对耦合电子阵列中的开关和开关动力学有一个完整的了解是至关重要的。最近的一些理论结果表明，电子开关速度将比从点和隧道结的电容和电阻预期的速度低几个数量级，这与巴黎圣母院的理论工作相反。为了解决这个问题，我们将高频测量技术应用到QCA单元和单元阵列中的开关研究中。目前，由于铝隧道结产生的大电容，QCA器件的实验演示仅限于少数单元。为了支持和实验证实在结构和电路理论方面取得的进展，我们将采用基于AFM光刻的先进制造技术来生产具有极大增强工作特性的QCA。这将使我们能够制造和测量具有显著范围和复杂性的单元阵列。QCA提供了一个独特的机会，可以对学生的教育体验和电子设备领域的研究产生广泛影响。我们将开发基于QCA模拟工具的教学模块，向本科生和研究生讲授QCA体系结构的概念。这些单元将向学生介绍其他架构概念，从而使他们受益。此外，通过开阔他们的视野，它将通过强调建筑概念的基本概念来加强他们对传统建筑的理解。

CCR-0210153Kogge, PeterThis proposal was received in response to the Nanoscale Science and Engineering initiative, NSF 01-157, category NIRT. Quantum-dot cellular automata (QCA) is a revolutionary computing paradigm that is well suited to nanoelectronic implementation and scaling to molecular dimensions. The central feature of QCA is that binary information is encoded in the position of single electrons among a group of dots forming a cell. This represents a significant break with the transistor-based paradigm in which information is encoded by the state of the transistor current switch. In QCA, electrons switch between quantum dots within a cell, but no current flows between cells. This leads to extremely low power dissipation, avoiding the problem of heat generation that will ultimately limit the integration density of transistor circuits. Clocking of QCA circuits has proven to be extremely important from the standpoint of both architectures and devices. It allows arrays of QCA cells to be broken into sub-arrays for pipelined processing, and it enables cells to produce signal power gain to replace signal energy lost to the environment. Functioning QCA devices have already been demonstrated in an aluminum/oxide tunnel junction scheme, confirming the operation of QCA cells, shift registers, logic gates, and memory elements. Power gain in a QCA shift register has also been achieved. This project will advance the architectural development of QCA, investigate questions of switching speed in nanoelectronic devices, and develop advanced fabrication techniques to implement the architectural and circuit theory concepts. Since QCA represents a dramatic break from conventional devices, significant changes in architecture are needed to fully exploit the capabilities of QCA. In QCA layout, timing, and architecture are intimately related, requiring a unified design approach. This is analogous to the approach begun by Mead and Conway which revolutionized VLSI design by making a connection between architecture and layout and building on that connection to enable designers to quickly synthesize large and complex functional blocks. Likewise, QCA system designers will be able to exploit timing in addition to layout to produce highdensity functional designs. In particular we will investigate the development of simple, yet complete, QCA based "Field Programmable Gate Arrays", where 2D arrays of identical cells are tiled together, with programmable interconnect and function. Timing plays a pivotal role in QCA designs, so it is vital to achieve a complete understanding of switching and switching dynamics in arrays of coupled electrons. Some recent theoretical results indicate that electron switching speeds would be orders of magnitude lower than that expected from the capacitances and resistances of the dots and tunnel junctions, contrary to theoretical work done at Notre Dame. To resolve this issue we will apply high frequency measurement techniques to the study of switching in QCA cells and in arrays of cells.At present, experimental demonstrations of QCA devices are limited to a small number of cells by the large capacitances produced by the aluminum tunnel junctions. To support and experimentally confirm the advances made in architecture and circuit theory, we will employ advanced fabrication techniques based on AFM lithography to produce QCA with greatly enhanced operating characteristics. This will allow us to fabricate and measure arrays of cells with significant extent and complexity. QCA presents a unique opportunity for a broad impact on the educational experience of students, and on research in the field of electronic devices. We will develop instructional modules based on QCA simulation tools to teach the concepts of QCA architecture to undergraduate and graduate students. These modules will benefit students by introducing them to alternative architectural concepts. In addition, by broadening their horizons, it will strengthen their understanding of conventional architectures by emphasizing the foundational concepts of architectural concepts.