ABSTRACT
Among the disparate emerging technologies that have been proposed to overcome the limitations of “end-of-the-roadmap” CMOS (complementary metal–oxide–semiconductor), quantum-dot cellular automata (QCA) shows promising features to achieve both high computational throughput and low-power dissipation. The QCA computational paradigm [1,2] introduces highly pipelined architectures with extremely high speed (of the order of THz), while radically departing from the switch-based operation of CMOS. QCA manufacturability has been demonstrated both for metal-dot QCA [3] and molecular scale allowing room temperature operation. Recently, magnetic QCA (MQCA) based on Co nanomagnets has been analyzed [4–8]. The use of nanomagnets is very attractive because MQCA can operate at room temperature, and has been shown to be easier than the molecular implementation of an electrostatic QCA. Moreover, MQCA can also be integrated with other emerging technologies such as magnetic RAM for memory design. The clocking mechanism of MQCA is similar to electrostatic QCA; the use of abrupt switching in electrostatic QCA is unreliable [2] due to the possible generation of metastable states, so a quasi-adiabatic clocking scheme has been proposed to overcome the kink probability in QCA circuits [2]. For MQCA, a three-phase snake clock has also been proposed [9]. Finally, a technology-based solution has been proposed in Ref. [7] to stabilize the magnetization state of nanomagnets by adding biaxial anisotropy. This arrangement 226modifies the framework in which MQCA circuits can be designed, thus requiring further investigation into mechanisms (also at circuit level) to leverage the newly introduced functionalities.
