Physical etching/deposition simulation with collision-free boundary movement

Hsiau, Ze-Kai; Kan, Edwin C.; McVittie, J.P.; Dutton, R.W.

doi:10.1109/iedm.1995.497192

Cited by 5 publications

(4 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shimada et al [59] simulated the SiO 2 trench etching with both charging and polymer deposition. Hsiau et al [66] incorporated the level-set method into the SPEEDIE code to simulate etching and deposition. Kokkoris et al [65] simulated Si trench profile in the Bosch process with alternating SF 6 and fluorocarbon gases.…”

Section: Level-set Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Review of profile and roughening simulation in microelectronics plasma etching

Wei¹,

Sawin²

2009

J. Phys. D: Appl. Phys.

View full text Add to dashboard Cite

Plasma etching of thin films is essential for microelectronics manufacturing. With current feature sizes of 35 nm in production and processes for smaller devices being developed, the sidewall roughness is within the order of magnitude of the gate length of the device, and therefore significantly impacts the devices' performance. In this paper we review the modelling of the surface profile evolution in plasma etching. Both two-dimensional (2D) and three-dimensional (3D) models have been developed using a number of representations and solution algorithms. String algorithms and the method of characteristics use a segmented string which is incrementally advanced. Level-set representations describe the profile evolution as a moving interface in response to a velocity field. Cellular representations in which the area or volume domain is divided into discrete cells have been used with flux and surface kinetics based on Monte Carlo calculations. We discuss our work in the modelling of profile evolution with surface roughening using a 3D cellular Monte Carlo simulation. The formation of perpendicular and parallel ripple formation on planar surfaces as a function of ion bombardment incidence angle and the transformation from perpendicular to parallel as etching progresses has been modelled. The smoothing and/or roughening of resist masks has been demonstrated along with the pattern transfer of roughness into the underlying layers being etched.

show abstract

Section: Level-set Methodsmentioning

confidence: 99%

“…The thickness of sidewall passivation layers and trench profiles was assumed to be the difference of the initial sidewall and the advancing frontier. [53][54][55][56][57][58][59][60][61][62][63][64][65][66]. The level-set method for profile evolution in microelectronics was introduced by Osher and Sethian [53].…”

Section: 32mentioning

confidence: 99%

Review of profile and roughening simulation in microelectronics plasma etching

Wei¹,

Sawin²

2009

J. Phys. D: Appl. Phys.

View full text Add to dashboard Cite

show abstract

“…The level set method on adaptive quad/oct-tree mesh [26] is a more straightforward alternative, and is more compatible with the adaptive gridding schemes in process simulation environment [27]. In our implementation, adaptive quad-tree mesh, with grid density adapted to , is used for the spatial discretization of boundary movement [28]. The stopping criteria for adaptivity is cell area or maximum grid density has been achieved (12) where is the average distance of over the quad/oct-tree mesh element.…”

Section: B Discretization and Solution Schemesmentioning

confidence: 99%

Robust, stable, and accurate boundary movement for physical etching and deposition simulation

Hsiau¹,

Kan²,

McVittie³

et al. 1997

IEEE Trans. Electron Devices

Self Cite

View full text Add to dashboard Cite

The increasing complexity of VLSI device interconnect features and fabrication technologies encountered by semiconductor etching and deposition simulation necessitates improvements in the robustness, numerical stability, and physical accuracy of the boundary movement method. The volume-meshbased level set method, integrated with the physical models in SPEEDIE, demonstrates accuracy and robustness for simulations on a wide range of etching/deposition processes The surface profile is reconstructed from the well-behaved level set function without rule-based algorithms. Adaptive gridding is used to accelerate the computation. Our algorithm can be easily extended from two-dimensional (2-D) to three-dimensional (3-D), and applied to model microstructures consisting of multiple materials. Efficiency benchmarks show that this boundary movement method is practical in 2-D, and competitive for larger scale or 3-D modeling applications.

show abstract

“…Fundamental to MOS technology, the patterning of polysilicon structures to form the gates and self-aligned contacts presents challenges in the area of deposition and etching. This includes both algorithmic issues for moving boundaries and the surface chemistry necessary to achieve anisotropic etching [4]. The kinetics of forming ultrathin layers and their resulting electrical and interface properties continue to be a key process step where predictive models are needed.…”

Section: Intrinsic Devicesmentioning

confidence: 99%