On the afternoon of the 8th, halls 3 and 4 of Exhibition Center 1 at Kintex in Goyang, Gyeonggi Province. Even during weekday daytime hours, the "Nano Korea 2026" exhibition hall bustled with visitors. Between booths set up by Korea's leading materials, parts, and equipment corporations and startups, universities, and research institutions, businesspeople and students kept streaming by. Technology displays from a range of overseas equipment makers also caught the eye.
Samsung Electronics and LG Group put "invisible manufacturing base technologies" front and center at this year's Nano Korea rather than finished goods. Samsung Electronics placed sixth- and seventh-generation High Bandwidth Memory (HBM), HBM4 and HBM4E, up front and highlighted comprehensive semiconductor solutions for the artificial intelligence (AI) chip era through a 2-nanometer Gate-All-Around (GAA) wafer and next-generation packaging mockups.
LG Electronics, LG Innotek, and LG Chem shared a booth, placing semiconductor packaging equipment, glass substrate processing, wafer inspection, AI server cooling, automotive electronics parts, and carbon materials in one space. It was a scene showing that the front lines of AI competition are expanding beyond chip performance and finished goods to process, equipment, materials, metrology, and thermal management.
Nano Korea 2026 is Korea's largest nanotechnology event, co-hosted by the Ministry of Trade, Industry and Resources and the Ministry of Science and ICT and co-organized by the Korea Nano Convergence Industry Association and the Nano Technology Research Association. This is its 24th year. The event, themed "Innovation for future: as nano meets AI," runs for three days from today through the 10th. This year's event features 401 corporations and institutions from eight countries, filling 674 booths. Exhibitions across eight fields—nano convergence, advanced ceramics, smart sensors, adhesives, coatings and films, lasers, additive manufacturing, bio, and metrology instruments—are being held simultaneously.
◇ Samsung Electronics, from HBM4 to 2-nm GAA… "Comprehensive semiconductor solutions"
Inside the Samsung Electronics booth, under the phrase "Comprehensive semiconductor solutions spanning memory, logic, foundry, and packaging," were HBM4 and HBM4E, DDR5 server-registered dual in-line memory modules (RDIMMs), graphics DRAM (GDDR7), low-power DRAM (LPDDR5X), server solid-state drives (SSDs), image sensors, mobile processors, a 2-nm GAA wafer, and next-generation packaging structures.
The spot where the most visitors stopped was the HBM4 and HBM4E display. The exhibit panel described HBM4 and HBM4E as ultra-high-bandwidth memory that supplies data needed for graphics processing unit (GPU) computation at the fastest speeds. Samsung Electronics showcased that it applied the industry's first 1c DRAM and logic base die built on a 4-nm foundry to HBM4 and HBM4E. It put forward the message that it would secure performance and supply competitiveness through design and process optimization and advanced packaging technologies.
Samsung Electronics also exhibited RDIMMs with enhanced capacity scalability for operating large-scale AI models, GDDR7 that supports fast responsiveness and efficiency in AI inference and acceleration environments, and the LPDDR5X SOCAMM2 that improves power efficiency and space utilization in high-density AI server environments. Server SSDs such as the PM1763 (E1.S), which improves real-time data processing efficiency in AI servers, and the BM1773 (E3.S), which stores large-scale AI data stably over long periods, were placed side by side. In effect, it explained the entire process inside AI servers—how data is stored, moved, and processed—through its memory and storage product lineup.
Dedicated zones were also set up for on-device AI and chips for vehicles and robotics. ▲ The Isocell HP5, an ultra-compact module based on a 0.5-micrometer (μm) 200-megapixel image sensor ▲ the Exynos 2600, a flagship mobile processor built on a 2-nm process ▲ next-generation low-power memory LPDDR6 ▲ and the compact high-performance SSD PM9E1 conveyed that it supports the realization of "AI operating inside devices." In the semiconductor zone for cars and robots were ▲ a three-dimensional (3D) image sensor ▲ automotive UFS 4.1 and LPDDR5X ▲ and a detachable AutoSSD.
On one side of the booth, a 2-nm GAA wafer and a model showing transistor structure changes were on display. It used models to show the evolution from planar transistors to FinFETs and then to Samsung Electronics' next-generation structure, the multi-bridge channel field-effect transistor (Samsung MBCFET). A heterogeneous-joining packaging mockup that spatially arranged HBM, upper and lower dies, a silicon bridge, and a substrate also drew attention.
◇ LG Electronics, semiconductor equipment instead of home appliances… "Starting external supply of internalized technologies"
The LG Group booth focused on equipment and materials for manufacturing sites. LG Electronics placed semiconductor packaging and back-end process equipment developed by its Manufacturing Technology Center at the front instead of TVs or home appliances. An LG Electronics official said, "Perhaps because it uses technologies verified internally, it's impressive how the equipment is designed to minimize barriers to adoption," and added, "Some are already being supplied externally."
The centerpiece of the LG Electronics booth was advanced packaging process equipment. The exhibit panel organized, step by step, substrate processes based on glass cores and printed circuit boards (PCBs) and wafer-level and panel-level packaging processes. LG Electronics introduced various equipment, including ▲ glass through-via (TGV) processing ▲ laser direct imaging (LDI) ▲ ceramic and PCB laser drilling ▲ in-wafer defect inspection ▲ die bonding ▲ redistribution layer (RDL) formation ▲ and six-side inspection. It presented, all at once, the equipment needed from substrate processing and patterning to via-hole formation, dicing, and inspection in the semiconductor packaging process.
The most eye-catching equipment was the high-resolution LDI system. Unlike conventional lithography processes, LDI forms circuit patterns directly on substrates by firing lasers based on design data without using photomasks. LG Electronics said LDI is used to create fine circuits on glass substrates and on substrates for semiconductor packaging. Its mass-production models use a 405-nanometer (nm) laser-diode module as the light source, and model resolutions range from 1.5 to 3.0 μm. It can also process large 600 mm × 600 mm substrates.
LG Electronics explained that its LDI equipment can help improve production yield in advanced packaging processes. It can implement line-and-space (L/S) patterns below 2 μm and compensate for input/output (I/O) positions and shapes. It highlighted as an advantage the ability to correct pattern distortion even when substrates warp or deform into a wavy shape.
A TGV laser system aimed at the era of glass substrates was also a key exhibit. TGV is a technology that drills microscopic holes inside glass substrates and fills them with conductive materials to connect the top and bottom circuit layers. As AI semiconductor packages become larger and more densely integrated, glass substrates are drawing attention alongside this technology.
LG Electronics also brought AI server heat issues into the realm of equipment technology. The booth featured a high-precision thermal verification system that simulated an Nvidia Blackwell GB200-class GPU environment and a fanless demo server using direct-to-chip (D2C) liquid cooling. LG Electronics said it can verify the reliability of cooling infrastructure by simulating GPU and HBM placement, hotspots, and real-time temperature feedback.
An executive at a domestic manufacturer met on site said, "It's impressive that the equipment was built to minimize barriers to adoption, perhaps because it uses technologies verified internally," adding, "It seems experience from actual mass-production sites was reflected in the equipment design."
LG Innotek exhibited electric-vehicle charging and power-conversion components. On display were a direct-current electric-vehicle supply equipment (DC EVSE) system that enables bidirectional energy flows among EVs, the power grid, and homes; an electric-vehicle communication controller (EVCC) that manages communications between in-vehicle controllers and charging infrastructure; and a direct-current-to-direct-current converter (DC-DC converter) that steps high voltage down to low voltage.
LG Chem put materials out front. Few-walled carbon nanotubes (FWCNTs) were introduced as a conductive additive that forms efficient conduction paths in battery electrodes to increase electrical conductivity. The exhibit included a dispersion and film samples. Proposed applications included thermal management systems, electromagnetic shielding films, transparent conductive films, heating fibers, and high-conductivity wires. LG Chem also introduced cases of applying AI to materials development and quality analysis, including an AI-based alloy design platform, real-time defect analysis, and AI vision automation to improve micro light-emitting diode (micro LED) assembly quality.