BASIC BUILDING BLOCKS FOR EFFECTIVE SINGLE ELECTRON TUNNELING TECHNOLOGY BASED COMPUTATION Cor Meenderinck and Sorin Cotofana Computer Engineering Laboratory Delft University of Technology, Delft, The Netherlands {Cor,Sorin}@ce.et.tudelft.nl Abstract Single Electron Tunneling (SET) technology appears to be a promising alternative for CMOS as it exhibits excellent power consumption and scalability features. Moreover, this new technology opens up avenues for new computational paradigms, which require building blocks with unconventional behavior. In this paper we discuss a number of basic building blocks that allow to effectively implement computational structures in those new paradigms, and analyze them in terms of area, delay, and energy consumption. Keywords: single electron tunneling, circuit design, building blocks. 1. INTRODUCTION It is generally expected that current semiconductor technologies, i.e., CMOS, cannot be pushed beyond a certain limit because of problems arising in the area of power consumption and scalability. A promising alternative is Single Electron Tunneling (SET) technology [1], which has the potential of performing computation with much lower power consumption than CMOS and it is scalable to the nanometer region and beyond [2]. SET technology is fundamentally different from CMOS as it is based on tunneling of electrons. This difference opens up avenues for new computational paradigms of which a number have been proposed [3,4,5], and which try to effectively use the basic SET properties. Theoretical results on the complexity of arithmetic operations using those new paradigms indicate great potential. However, the actual practical results depend on the capabilities of the utilized building blocks. In previous research we already identified a number of such basic building blocks. In this paper we analyze these building blocks with respect to limitations, area, delay, and energy consumption. This paper is organized as follows. In Section 2 we briefly present some background on SET technology. In Section 3 we present five basic building blocks for SET based computation, and analyze them. Section 4 concludes the paper. 1. BACKGROUND SET circuits are based on tunnel junctions, which consist of an ultra-thin insulating layer in a conducting material. In classical physics no charge transport is possible through an insulator. However, when the insulating layer is thin enough the transport or tunneling of charge can be controlled in a discrete and accurate manner, i.e., one electron at a time. Tunneling through a junction becomes possible when the junction's current voltage V j exceeds the junction's critical voltage ( ) j e e c C C q V + = 2 [6], where q e =1.602*10 - 19 C, C j is the capacitance of the junction, and C e is the capacitive value of the remainder of the circuit as seen from the junction. In other words, tunneling can occur if and only if |V j | ≥ V c . Electron tunneling is stochastic in nature and as such the delay cannot be analyzed in the traditional sense. Instead, for each transported electron one can describe the switching delay as ( ) c j t e error d V V R q P t − − = ln , where R t is the junction's resistance and P error is the chance that the desired charge transport has not occurred after t d seconds. In this paper we assume R t =10 5 Ω and P error =10 -8 . Each transported electron reduces the system energy by ( ) c j e V V q E − = Δ from which the consumed energy can be calculated. Note that SET technology can physically be implemented in various ways, e.g., classical semiconductor lithography and by carbon nanotubes. Therefore, for the blocks we discuss in this paper, the circuit area is evaluated in terms the total number of circuit elements (capacitors and junctions). 2. BASIC BUILDING BLOCKS SET technology enables accurate control of the transportation of discrete electrons. Moreover, SET allows the representation of values by number of electrons, i.e., Boolean values may be represented by the presence or absence of one electron, while integer values may be represented by the corresponding number of electrons. To effectively utilize this encoding in arithmetic and logic operations, building blocks are required that perform basic signal operations on this Boolean and multi-value signals. Previous investigations suggested that Boolean operations can be implemented using threshold logic gates and inverting buffers [4], while in order to perform arithmetic operations via direct charge