Pioneering 2D Materials for semiconductor industry
2D material transfer challenges on a wafer scale
We welcomed over 300 attendees to this workshop, hosted by the Graphene Flagship's 2D-EPL, which focused on the different steps that are encountered during the transfer of 2D materials on Si wafers, which secure access to the most advanced nanofabrication technology. The challenges linked to the delamination of high quality 2D materials grown on epitaxial templates as well as the subsequent steps for bonding the 2D materials to the target wafer were subject to an in depth dive.
Agenda
14:00 - 14:15 | Cedric Huyghebaert, imec
2D Experimental Pilot Line update. Announcing the call for applications to participate in the 2D-Experimental Pilot Line's first Multi-Project Wafer (MPW) run.
14:15 - 14:55 | Jeewhan Kim, MIT
Deterministic wafer-scale growth and transfer of single-domain 2D materials
2D material-based devices have received great deal of attention as they can be easily stacked to obtain multifunctionality. With their ultrathin thicknesses, such multifunctioning devices become so flexible and conformal that they can be placed onto any 3D featured surfaces. However, 2D heterostructures are typically demonstrated as stacked flakes where single or few devices can be fabricated due to lack of strategies for layer-by-layer stacking of 2D materials at the wafer scale. While the growth technique could form wafer-scale 2D materials, it has been difficult to control monolayer-by-monolayer growths. In this talk, I will discuss about our unique strategy to isolate wafer-scale 2D materials into monolayers and stack them into a heterostructures by using a layer-resolved splitting (LRS) technique [1-2]. In addition, I will introduce our new technique to directly grow monolayer single-domain 2D materials at the wafer-scale and monolayer-by-monolayer growth of 2D heterostructures [3].
While 2D heterostructures promise interesting futuristic devices, the performance of 2D material-based devices is substantially inferior to that of conventional 3D semiconductor materials. However, 3D materials exist as their bulk form, thus it is challenging to stack them together for heterostructures. My group at MIT has recently invented a 2D materials-based layer transfer (2DLT) technique that can produce single-crystalline freestanding membranes from any compound materials with their excellent semiconducting performance [4-5]. I will also introduce my group’s 3D heterostructures by stacking various freestanding single-crystalline membranes to obtain new functionality [6].
References: [1] Science, 342, 833 (2013), [2] Science 362, 665 (2018), [3] unpublished, [4] Nature 544, 340 (2017), [5] Nature Materials 17, 999 (2018), [6] Nature, 578, 75 (2020)
14:55 - 15:25 | Steven Brems, imec
2D material transfer: The challenging route from LAB to FAB
Despite the sizeable effort to synthesise two-dimensional (2D) materials and demonstrate devices with these layers, the commercial adoption of 2D layers in the semiconductor industry did not yet fully materialise. Several challenges need to be solved (e.g., high-quality 2D growth, dielectric growth on a passivated 2D surface, controlled interfaces...) to achieve this goal. Chemical Vapour Deposition (CVD) is the most mature and industry relevant synthesis technique to produce oriented growth of graphene, h-BN and MX2 (e.g. WS2, MoS2). Oriented 2D layers are typically grown at high temperatures (often well above 700°C), and regularly even using a catalytic layer to enable high-quality 2D growth (i.e., graphene and h-BN). A possible 2D material integration route relies on the transfer of these 2D materials from the growth substrate to the device wafer. This presentation focusses on the 2D material transfer challenges to achieve a CMOS compatible 2D transfer.
An industrial relevant transfer process is being developed for graphene and MX2 layers at Imec, relying on a rigid carrier combined with a laser release layer. A rigid carrier is likely one of the necessary requirements to enable the use of today’s bonding infrastructures, while maintaining strain control in these 2D layer during the different transfer steps. An overview of the challenges related to 2D material debonding from the growth substrate for both graphene and MX2 will be given. An understanding of the debonding mechanism will be shown, which often relies on water intercalation when the 2D material is directly de-bonded from the growth surface. To end, an alternative transfer process for graphene will be discussed, avoiding the need for a graphene peeling step directly from the growth surface. Alternative debonding methods might become more important to enable a 2D material adaptation by the industry.
15:25 - 15:55 | Panel Discussion
Our speakers will continue the discussion on transfer challenges. The panel will be open to your questions relating to the previous presentations or other related topics.
Moderator: Cedric Huyghebaert
15:55 - 16:00 | Closing thoughts