Using ab initio density functional theory (DFT) calculations, we demonstrate two molecular OR gates that are able to process binary signals encoded as molecular potentials. Thus, the possibility to implement logic gates of _1 nm is demonstrated. The advantage of this approach to post-microelectronics technologies is the tremendous low-power dissipation, the small feature size of molecular devices, and the compatible nature of input and output signals that would allow the implementation of complex logic.
LIUMING YAN,1 JORGE M. SEMINARIO1,2
1Department of Chemical Engineering, Texas A & M University, College Station, Texas 77843-3122
2Department of Electrical Engineering, Texas A & M University, College Station, Texas 77843-3122
Despite the tremendous progress of the complementary metal-oxide-semiconductor (CMOS) and related technologies to perform fast computing, we have to plan what to do when the physical limits of this technology are approached; most likely this will take place within the 5-year horizon. Faster computers are always needed, and societies are willing to pay the extra price for better computer performance because of its direct impact on prediction, design, control, and information on any aspect of human activities. The rate of improvement of computing technology follows an increasing exponential behavior known as Mooreâ„¢s law, which was empirically predicted more than 40 years ago, stating that every 18 months, the number of devices per unit of area is doubled [1, 2]. Thus, after more than 40 years, the prediction still holds; this is an extraordinary success of science and technology together not seen in any other activity of human endeavor. However, as devices scale down, we approach sizes between their electrodes and interfaces that are in the quantum tunneling region, deteriorating the switching behavior of the devices, in contrast to increasing tremendously heat removal problems from integrated circuits. Searching for new materials only provides marginal improvements thus a totally new scenario for data processing at atomistic scales needs to be developed.