Using a 4-megapixel hybrid photon counting detector for fast, lab-based nanoscale x-ray tomography

Hybrid photon counting detectors (HPCDs) have unlocked new capabilities for x-ray-based measurements at synchrotrons around the world in the last 30 years. By leveraging independently optimized sensor and readout layers, they offer high quantum efficiency ($> 80 %$), ultra-low dark counts, sub-pixel point-spread function, and high count rates ($> 10^{6}$ counts per pixel per second). Furthermore, their small pixel size and large active area endow them with excellent coverage and resolution for both real-space and reciprocal space imaging. Here, we demonstrate that HPCDs are also well-suited for laboratory-based nanoscale x-ray tomography (nano-xCT). We perform nano-xCT on an integrated circuit fabricated at the 130-nm node and produce a 3D reconstruction with 40 times more photons collected 20 times faster than in this group’s previous work, for an overall speedup of 800$\times$. We review the technical considerations of using an HPCD for tabletop tomography. We quantify our reconstruction image quality using well-established metrics, including the modulation transfer function (MTF), Fourier shell correlation (FSC), and contrast-to-noise (CNR), to validate our choice of experimental parameters that provide sufficient resolution and imaging speed. Using these metrics, we determine that even under current experimental conditions, 160 nm wiring features are reconstructed at 75-80 nm spatial resolution.

💡 Research Summary

This paper demonstrates that a state‑of‑the‑art hybrid photon counting detector (HPCD), the DECTRIS Eiger2 R 4M, can dramatically accelerate laboratory‑based nanoscale X‑ray computed tomography (nano‑XCT) while delivering sub‑100 nm spatial resolution. The authors integrate the 4‑megapixel detector (pixel size 75 µm, active area 155 mm × 162 mm) into a custom SEM‑based cone‑beam CT system developed at NIST. An electron beam (<100 nm spot) strikes a Pt target, generating a broadband X‑ray source that illuminates a 130‑nm‑node integrated circuit (IC) sample thinned to 3.32 µm of circuitry and 8.6 µm of Si backing. The X‑rays exit the SEM vacuum chamber through a Be window and are captured by the HPCD positioned 256 mm from the source.

Because of the very short source‑sample distance (≈10 µm) the geometric magnification reaches ~25 000, theoretically projecting a 3 nm pixel on the sample. In practice the authors bin 16 × 16 detector pixels to reduce computational load, yielding an effective voxel size of ~1.2 µm for reconstruction. The key technical challenge addressed is the non‑uniform collection efficiency across the large flat‑panel detector. Two geometric effects dominate: (1) the solid‑angle reduction that scales as (z/r)³, where z is the distance to the nearest pixel and r the distance to an arbitrary pixel, causing up to 20 % intensity variation from centre to corner; (2) the obliquity factor (cos ϕ) that further reduces effective area for off‑axis pixels. The authors derive and apply a pixel‑wise correction factor that compensates both effects, eliminating the intensity gradients that would otherwise produce severe reconstruction artefacts.

A secondary effect, the angle‑dependent absorption probability in the silicon sensor, is shown to be negligible (<1 %) for the dominant photon energies (4–10 keV) in this experiment, but the paper notes that at higher energies or with larger detectors this term must be included.

Data acquisition proceeds over 10 hours, collecting 1800 projection images of a ~1200 µm³ volume. Compared with their previous setup that used a 240‑pixel X‑ray spectrometer (which required ~240 hours for the same volume), the HPCD provides 40× more photons and a 20× faster acquisition, amounting to an overall 800× speedup.

Image quality is quantified using three established metrics: Modulation Transfer Function (MTF), Fourier Shell Correlation (FSC), and Contrast‑to‑Noise Ratio (CNR). The MTF indicates a 10 % contrast cutoff at ~80 nm, the FSC 0.5 criterion yields a resolution of 75 nm, and the CNR analysis confirms that 160 nm wiring features are clearly distinguishable. These results demonstrate that laboratory‑scale nano‑XCT can now resolve features well below 200 nm, a regime previously accessible only at synchrotron facilities.

Beyond the experimental demonstration, the authors provide a generalized workflow for handling large‑area flat‑panel detectors in cone‑beam CT: (i) geometric correction of each pixel’s solid angle and obliquity, (ii) optional energy‑dependent absorption correction, (iii) binning strategy to balance resolution and computational demand, and (iv) integration with the TomoScatt reconstruction algorithm. This workflow is presented as a template for other groups seeking to retrofit HPCDs into existing CT rigs.

In summary, the study proves that hybrid photon counting detectors, originally designed for synchrotron applications, can be repurposed for fast, high‑resolution laboratory nano‑CT. By addressing the unique geometric artefacts of large flat‑panel detectors, the authors achieve a practical, non‑destructive metrology tool capable of imaging sub‑200 nm semiconductor features within a workday, opening new possibilities for in‑house failure analysis and rapid prototyping in the semiconductor industry.