PhD - Leuven | More than two weeks ago
With the end of Moore’s law in sight, chip manufacturers are turning to 3D integration to continue the scaling trend and add more functionality per unit area. This is done by integrating circuits on top of each other, using 3D interconnections. The technology also enables the integration of heterogeneous wafers with different functionalities and materials, such as logic, memory, sensors and photonics. As such it provides shorter paths between the processor and its memory, helping to break down the memory wall, a bottleneck for artificial intelligence performance.
One of the topics in imec’s 3D systems integration program is the scaling of wafer-to-wafer hybrid bonding to ultra-low 3D interconnect pitches. In this technology, full wafers are bonded using dielectric bond surfaces containing Cu contact pads. Currently imec has demonstrated electrical yield for pad pitches down to 400 nm. At these small dimensions, contacting and bonding of the Cu pads is enabled by surface diffusion driven topography changes. For further scaling, this phenomenon must be better understood and controlled. This requires modeling and experimental measurement of the underlying self-diffusion driven by gradients of stress & strain and curvature, which cannot be achieved using commercial Finite Element software packages.
Your role will be to develop a physics-based model allowing to simulate and understand in detail the observed surface self-diffusion phenomena and their dependence on geometric, material and processing conditions. Based on this, you will develop a predictive model and use it to forecast the Cu surface topography changes.
Data for model verification and calibration will be collected in experiments both at coupon and at 300 mm wafer level. Additional supporting experiments will include contact angle and surface diffusion coefficient measurements.
You have a Materials Science or Physics degree with good grades and a keen interest in modeling, surpassing common use of commercial simulators.
We offer a chance to be part of an enthusiastic international team, working ahead of the industry on scaling of 3D technologies. Collaboration with scientists is expected in the expertise fields of materials science, thermo-mechanical modeling, metrology, and processing. You will work in daily contact with representatives of industry leading companies and have the opportunity to present your work at high-ranking conferences.
Expertise Center - AR2T
Required background: Materials Science or Physics
Type of work: 60% modeling/simulation, 20% experimental, 10% data analysis, 10% literature/reporting
Supervisor: Clement Merckling
Co-supervisor: Joke De Messemaeker
Daily advisor: Joke De Messemaeker
The reference code for this position is 2024-054. Mention this reference code on your application form.