(Invited) Performance of Graphene and Beyond Graphene 2D Semiconductor Devices

During the past decade, 2D (two-dimensional) materials have gained substantial interest in the electronic device community. The first 2D material studied in detail was graphene and since 2007 it has intensively been explored as channel material for transistors. It soon turned out, however, that the gapless nature of graphene causes problems for both logic (no switch-off) and radio frequency (poor power gain) transistors. While graphene transistors are still on the agenda, many researchers have extended their work to 2D materials beyond graphene. Meanwhile several hundred different 2D materials are known, such as the X-enes (graphene, silicene, germanene, phosphorene, etc.), the X-anes (graphane, silicane, germanane, etc.), the TMDs (transition metal dichalcogenide, e.g., MoS₂, MoSe₂, WS₂, WSe₂, etc.), and many more. A substantial portion of these 2D materials are semiconductors and therefore considered useful for transistors. Recently, experimental transistors with channels of 2D materials other than graphene have been demonstrated. In spite of the rapid progress in the field, the prospects of 2D transistors still remain vague. The intention of the present paper is to shed more light on the merits and drawbacks of 2D materials for transistor electronics. To this end, we compose a wish list of properties for a good transistor channel material and examine to what extent the different 2D materials fulfill the criteria of the list. Next, the state-of-the-art performance of 2D transistors is reviewed and both the pros and cons of these devices are elaborated. Here we focus on the switching behavior (on-off ratio) and the radio frequency performance in terms of the cutoff frequency f_T and the maximum frequency of oscillation f_max.

Based on the above considerations, the prospects of 2D materials for future electronics are discussed and the following conclusions are drawn. We see only limited potential of 2D transistors for mainstream electronics in the near to medium term while their prospects for flexible and transparent electronics look much better. In the field of flexible electronics, graphene MOSFETs show potential for radio frequency applications and TMD transistors, such as MoS₂ MOSFETs, are promising for both logic and radio frequency flexible circuits. In the long term, i.e., beyond the ITRS horizon at 5 and sub-5 nm gate length levels (provided the MOSFET can be scaled that far), certain 2D materials with a wide enough gap and heavy carrier effective mass m_eff (and thus lower mobility µ) become interesting for logic MOSFETs since they suffer much less from direct source-drain tunneling in the off-state compared to their light-m_eff, high-µ rivals. We show that for a gate length of 5 nm, source-drain tunneling in 2D TMD MOSFETs, particularly those with Mo-based channels, is orders of magnitude lower than in nanowire MOSFETs with III-V, Ge, and <100> Si channels (having lighter carrier effective masses than the TMD channels). We note, however, that Si nanowires with extremely small cross-section area and properly chosen orientation and strain may become another strong contender for MOSFETs beyond the ITRS horizon.