Advanced packaging now drives much of the progress in semiconductor performance. It needs interconnects and dielectrics patterned at scales and speeds that conventional lithography and plating struggle to reach. We develop additive processes that build these structures directly. Holographic metasurface nanolithography (HMNL) uses sub-wavelength metasurface masks to project multi-color holograms into a hybrid metal–polymer resin. This patterns entire 3D conductor–insulator structures with roughly 500 nm resolution in a single exposure. Microscale selective laser sintering (µ-SLS) uses a digital micromirror array to sinter copper nanoparticle films, writing micron-scale metal interconnects at high throughput. We also use two-photon photoreduction to write embedded 3D silver networks inside polymer dielectrics. Together these processes target redistribution layers, interconnects and heterogeneous integration for next-generation electronics packages.
You are here
Research
Research in the NDML focuses on a simple question: how do we make things at the micro- and nanoscale reliably, at the scale and cost that matter outside the lab? A new process alone is rarely the answer. We develop additive manufacturing methods, such as holographic metasurface nanolithography, microscale selective laser sintering, and two-photon printing, that can build three-dimensional structures no conventional process can. We pair these processes with the metrology, modeling, and machine learning needed to see what we are making and to predict what we will make, from in-line atomic force microscopy to digital twins of hybrid bonding and in-situ monitoring of metal printing. We also design the precision machines and instruments that make this work possible. Much of our research is driven by semiconductor advanced packaging, where the industry now needs new ways to build and inspect interconnects, but the same tools also let us create new materials, from nanoporous structures for critical element recovery to high-performance copper conductors. We believe that this type of progress in nanomanufacturing could help lead to a more secure domestic semiconductor supply chain and a more productive, sustainable society.
Hybrid bonding joins chips face to face through direct copper–copper and oxide–oxide bonds and is central to 3D heterogeneous integration. Bond quality, however, depends on nanometer-scale surface topography, material properties and process conditions that are hard to measure directly. We build multiscale simulations and digital twins of the Cu–SiO2 hybrid bonding process that link these inputs to bond formation, stress and yield. By combining physics-based models with machine-learning surrogates and in-line metrology data, these virtual metrology tools aim to predict bond quality before parts are bonded and to guide process decisions in advanced packaging manufacturing.
Scaling up nanomanufacturing requires measuring nanoscale features fast enough to control the process. We build in-line metrology systems and the data tools that make them practical. Our roll-to-roll inspection platform integrates MEMS-based atomic force microscopes into a moving web to measure nanopatterned films during production. Fast scans are noisy, so we develop machine-learning methods that reconstruct high-resolution topography from fast, sparse measurements and remove scan artifacts. We also develop alignment metrology for holographic lithography that detects voxel-scale misalignment between metasurface-generated holograms.
Metal additive manufacturing is only as reliable as our ability to see what happens in the melt pool and to quantify our confidence in each part. We use infrared thermography and synchronized in-situ measurements to monitor laser powder bed fusion of alloys such as 316L stainless steel and IN718. We have also developed methods to measure the emissivity of metal powders, so thermal images can be converted into accurate temperatures. Building on metrology principles, we apply statistical quality assessment and measurement-uncertainty analysis to judge whether a process is in control. We also study how factors such as powder particle size affect build quality. For directed energy deposition, we model the process dynamics to understand how disturbances affect part geometry.
Many of our manufacturing and metrology advances depend on custom precision hardware. We design long-travel, flexure-based nanopositioning stages that combine centimeter-scale range with nanometer-level precision, and we add features such as eddy-current damping to improve their dynamics. At smaller scales, we build MEMS-based instruments for measuring forces and topography at the nanoscale, including a MEMS interfacial force microscope and single-chip atomic force microscopes. With collaborators, we develop programmable mechanical metamaterials and mechanical neural networks driven by meso-scale thermal actuators. We also design precision machines for manufacturing, such as an r–θ projection micro-stereolithography system for large-area optics and ultraprecision spindles.
Additive manufacturing lets us design materials from the microstructure up. With collaborators in chemistry, we 3D print dual-scale gyroid structures whose nanoporous walls carry custom molecular receptors. These form modular cartridges that selectively capture critical elements from mixed streams. We also develop graphene–copper composite conductors that pair copper’s conductivity with improved thermal stability for aerospace power applications, and we study laser powder bed fusion of pure copper. Using vat photopolymerization with in-situ phase separation, we print highly porous polymer structures whose pore architecture can be tuned to control how drugs are released from implants.