3D high resolution imaging for microelectronics: A multi-technique survey on copper pillars

Area of Science:

• Microelectronics
• Materials Science
• Advanced Imaging Techniques

Background:

• 3D integration in microelectronics enables vertical stacking of dice or wafers for space-saving.
• The increasing adoption of 3D integration necessitates the development and assessment of advanced 3D imaging techniques.
• Copper pillars are critical components for interconnecting stacked components in 3D integrated circuits.

Purpose of the Study:

• To evaluate and compare various 3D characterization techniques for analyzing copper pillars used in microelectronic 3D integration.
• To assess both destructive and non-destructive imaging methods for their suitability in microelectronics applications.
• To provide insights into the resolution and sample throughput trade-offs of different 3D imaging modalities.

Main Methods:

• Destructive techniques: Focused Ion Beam/Scanning Electron Microscopy (FIB/SEM), FIB/FIB, and Plasma Focused Ion Beam/Plasma Focused Ion Beam (PFIB/PFIB) slice-and-view protocols.
• Non-destructive techniques: Laboratory-based computed tomography (CT) and synchrotron-based computed tomography.
• Comparative analysis focusing on resolution, throughput, and applicability to copper pillar characterization.

Main Results:

• Destructive techniques like FIB/SEM and PFIB/PFIB offer high-resolution cross-sectional views but are time-consuming and sample-destructive.
• Laboratory-based CT is suitable for rapid, punctual analysis of individual copper pillars.
• Synchrotron-based CT provides highly resolved 3D volumes, making it ideal for analyzing larger batches of samples efficiently.

Conclusions:

• The choice of 3D characterization technique for microelectronic copper pillars depends on specific industry needs, such as analysis scope and sample volume.
• Laboratory-based imaging is effective for targeted, single-sample investigations.
• Synchrotron-based computed tomography is the preferred method for high-throughput, detailed volumetric analysis of multiple samples in 3D integration research and development.