By David M. Fried
Today we officially released SEMulator3D 2014.100. Typically, I wouldn’t be so excited about a “point release”, but this is clearly the biggest interim software release in recent SEMulator3D memory. We’ve added significant capability to an already industry-leading virtual fabrication platform. Many of the features of recent SEMulator3D releases have been focused on Etch enhancements. To complement these enhancements, we’ve stepped up the predictive accuracy of several other process models in SEMulator3D 2014.100, including Deposition and CMP.
The highlight of this release is a new Visibility-Limited Deposition model. This model dramatically improves the predictive accuracy for directional depositions, like Physical Vapor Deposition (PVD) and other plasma enhanced deposition processes. As with other process models in SEMulator3D, we’ve made this process simple to implement and calibrate using a reduced set of process parameters. The key features of this Visibility-Limited Deposition model are the “Source Sigma”, reflecting the directional distribution of the process, and the “Isotropic Ratio”, reflecting the non-visibility-limited component of the deposition process. This model enables a large variety of processes, with a wide range of results.
Figure 1. Cross-sectional images of 3D models using Visibility-Limited Deposition model.
The modeling results above demonstrate the process capability range in 2D cross-sections, from very directional depositions in the top row to more conformal processes in the bottom row. Varying source distributions also exhibit different voiding and void shape during the deposition process. This capability will be used extensively in the modeling of state-of-the-art high-aspect ratio device geometries, including DRAM, 3D NAND Flash, MEMS and scaled FinFET CMOS.
SEMulator3D 2014.100 also includes a new Planarizing Deposition model, to more accurately reflect spin-on processes and flowable depositions. While these processes are meant to deliver a planar result, they inevitably retain some non-planarity due to underlying topography. The new Planarizing Deposition model delivers predictive accuracy for many different materials and processes, with various different planarizing behaviors.
Figure 2. Cross-sectional images of 3D models using Planarizing Deposition model.
This model also accounts for deposition processes thinner than the underlying topography, resulting in exposed structures. This capability will be utilized to model several of the novel high-aspect-ratio fill processes as well as many of the innovative patterning film-stacks emerging in a semiconductor process environment still waiting for EUV.
There are several other features in SEMulator3D 2014.100, including a new CMP model that more accurately predicts dishing and over-polish behavior in the presence of complex underlying topography, and modeling performance/accuracy enhancements. We’ve also added a helpful process comparison tool that helps developers keep their flows accurately reflecting the fabrication process.
With this long list of features and enhancements, it’s pretty easy to see why I’m so excited for this interim release. These features answer the needs of our growing customer set, and address new complexities in advanced processes. I can’t wait to spend some time with our customers and exploit these capabilities to accelerate their technology development projects!