An official website of the United States government
A precise and non-destructive metrology method for through-glass vias (TGVs) is critical for next-generation semiconductor packaging. Geometric features, including waist diameter and sidewall roughness, significantly influence the performance and reliability of the final product. Traditional inspection methods, such as scanning electron microscopy (SEM), are destructive, costly, and low-efficiency, making them unsuitable for in-line quality control. This paper presents a novel diffraction-based optical metrology method for non-destructive TGV inspection. In the proposed methods, a tailored incident laser is utilized to enhance the sensitivity to specific geometric features. A standard Gaussian beam is employed to characterize edge roughness, while a donut-shaped beam is specifically applied for the precise inspection of TGV waist diameter. A robust numerical simulation model, developed from a split-step procedure and the Rayleigh-Sommerfeld propagator, has been proposed to generate reference diffraction patterns for TGVs with an ideal geometry. The quality of a manufactured TGV was then quantified by calculating the 2D correlation coefficient between the experimentally captured diffraction pattern and the ideal simulated reference. Simulation results demonstrated high sensitivity to feature deviations, showing a clear monotonic decrease in correlation as edge roughness (sigma) and waist diameter error increased, respectively. The computational model was further experimentally validated using a custom optical setup to inspect an array of TGV samples (D1=150 μm, D2=84 μm). The experimental results showed a strong average 2D similarity (close to 0.80) with the simulation, confirming the model's accuracy despite fabrication imperfections. In summary, the proposed diffraction-based metrology provides a low-cost, high-throughput, and easy-to-implement solution for in-line and on-machine TGV inspection, offering strong potential for advanced semiconductor manufacturing.
more »
« less
This content will become publicly available on August 4, 2027
Silicon/Glass Interposer Metrology and Inspection by Shadowgraphy
Through-glass vias (TGVs) and through-silicon vias (TSVs) are critical vertical interconnects in advanced 3D packaging, where geometric defects and sidewall roughness directly degrade electrical reliability and yield. Conventional methods such as SEM and AFM require destructive preparation and offer limited throughput. This work presents a shadowgraphy approach for non-destructive TGV/TSV array metrology with sub-micron accuracy in diameter, depth, and sidewall roughness.
more »
« less
- Award ID(s):
- 2426512
- PAR ID:
- 10708039
- Publisher / Repository:
- US-Korea Conference
- Date Published:
- Format(s):
- Medium: X
- Location:
- Orlando FL
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Photonic integrated circuits (PICs) are vital for developing affordable, high-performance optoelectronic devices that can be manufactured at an industrial scale, driving innovation and efficiency in various applications. Optical loss of modes in thin film waveguides and devices is a critical measure of their performance. Thin film growth, lithography, masking, and etching processes are imperfect processes that introduce significant sidewall and top-surface roughness and cause dominating optical losses in waveguides and photonic structures. This roughness, as perturbations couple light from guided to far-field radiation modes, leads to scattering losses that can be estimated from theoretical models. Typically, with UV-based lithography, sidewall roughness is significantly larger than wafer-top surface roughness. Atomic force microscopy (AFM) imaging measurement gives a 3D and high-resolution roughness profile, but the measurement is inconvenient, costly, and unscalable for large-scale PICs and at wafer-scale. Here, we evaluate the sidewall roughness profile based on 2D high-resolution scanning electron microscope (SEM) imaging. We characterized the loss on two homemade nitride and oxide films on 3-inch silicon wafers with 12 waveguide devices on each and correlated the scattering loss estimated from a 2D image-based sidewall profile and theoretical Payne model. The lowest loss of guided fundamental transverse electric (TE0) mode is found at 0.075 dB/cm at 633 nm across 24 devices, a record at visible wavelength. Our work shows 100% success (edge continuity span exceeding 95% of image width/height) in edge detection in image processing of all images to estimate autocorrelation function and optical mode loss. These demonstrations offer valuable insights into waveguide sidewall roughness and a comparison of experimental and 2D SEM image processing based loss estimations with applications in loss characterization at wafer-scale PICs.more » « less
-
Abstract Achieving low optical loss is critical for scaling complex photonic systems. Thin‐film lithium niobate (TFLN) offers strong electro‐optic and nonlinear properties in a compact platform, making it ideal for quantum and nonlinear optics. WhileQfactors above 107are achieved, they remain below the intrinsic material limit. A systematic study of scattering losses due to roughness in TFLN racetrack cavities is presented, with isolating contributions from sidewall and interface roughness. Quality factors up to 2.7 × 107are demonstrated in waveguides with widths of 2.2λ (≈3.5 µm), where interface roughness dominates, and up to 1.2 × 107in narrower waveguides 0.8λ wide (≈1.2 µm), where sidewall roughness is the primary limitation. The modelling framework, based on 3D wave simulations informed by roughness measured by AFM, is applicable to any index contrast between waveguide core and cladding and widely applicable on integrated photonic platforms.more » « less
-
The atomic force microscopy (AFM) technology is a promising method for nanofabrication due to the high tunability of this affordable platform. The quality inspection and control significantly impact the manufacturing effectiveness for realizing the functionality of the achieved nanochannel. Particularly, the surface characteristics of nanochannel sidewalls, which play a significant role in determining the quality of the nanomachined products, can not be accurately captured using conventional surface integrity metrics (e.g., surface roughness). Therefore, it is necessary to propose a method to quantitatively characterize the surface morphology and detect the abnormal parts/regions of the nanochannel sidewall. This paper presents a statistical process control approach derived from the self-affine fractal model to detect the sidewall surface anomalies. It evaluates changes in the self-affine fractal model parameters (standard deviation, correlation length, and roughness exponent), which can be used to signify the changes on the sidewall surface; the statistical distributions of these parameters are derived and used to develop control charts to allow inspection of the sidewall morphology. The approach was tested on the AFM-based nanomachined samples. The results suggest that the presented approach can effectively reflect the abnormal regions on the machined parts, which opens up a new avenue toward guiding the quality control and rework for process improvement for AFM-based nanomachining.more » « less
-
Planar dielectric waveguides are key to silicon photonics but suffer from scattering losses due to sidewall roughness. This work revisits the classical analysis of Lacey and Payne for TE modes, providing a complete derivation of radiation loss using Green’s functions, Fourier analysis, and the Wiener–Khinchin theorem. Exponential and Gaussian autocorrelation models are examined, yielding closed-form expressions that link roughness statistics to attenuation. Results show Gaussian models predict higher peak loss and faster decay, while exponential models offer simpler, experimentally validated approximations. These findings provide scaling laws and practical design guidance for minimizing scattering and enabling ultralow-loss integrated waveguides.more » « less
