IEC TR 62632: Nanoscale Electrical Contacts and Interconnects

IEC TR 62632:2013 provides a comprehensive survey of emerging nanoscale electrical contact and interconnect technologies. As conventional copper interconnects face fundamental physical limits at advanced process nodes, this technical report examines alternative technologies including carbon nanotubes (CNTs), graphene, nanowires, and organic conductors that may replace or complement existing metallization in future integrated circuits.

Motivation: At technology nodes below 10 nm, copper interconnects suffer from increased resistivity due to electron scattering at grain boundaries and surfaces, and current-carrying capacity is limited by electromigration — driving the search for alternative nanoscale conductors.

1. Nanotube and Nanowire Technologies

1.1 Carbon Nanotube Contacts and Interconnects

CNTs offer exceptional electrical properties: ballistic transport with mean free paths exceeding 1 micron, current densities up to 109 A/cm2 (100x higher than copper), and excellent thermal conductivity. However, practical implementation faces challenges:

  • Contact resistance between CNTs and metal electrodes remains high (typically >10 kohm)
  • Chirality control during growth is imperfect — metallic and semiconducting nanotubes grow together
  • Dense CNT bundle integration requires growth temperatures above 500 C, incompatible with back-end-of-line (BEOL) thermal budgets
Technology Resistivity Current Density Contact Resistance Maturity
Cu interconnects (7 nm node) ~8 micro-ohm-cm ~1e7 A/cm2 <1 ohm Production
CNT bundles (vias) ~10 micro-ohm-cm ~1e9 A/cm2 ~1-10 kohm Prototyping
Graphene interconnects ~5 micro-ohm-cm (SLG) ~1e8 A/cm2 ~100 ohm-micron Research
Ag nanowires ~3 micro-ohm-cm ~1e7 A/cm2 ~10 ohm Prototyping

2. Graphene and Emerging 2D Materials

Graphene — a single atomic layer of carbon atoms in a hexagonal lattice — offers the thinnest possible conductor with exceptional carrier mobility (>200,000 cm2/Vs). For interconnect applications, several forms are being investigated:

  • Single-layer graphene (SLG): Highest mobility but high sheet resistance (~100 ohm/sq)
  • Multi-layer graphene (MLG): Lower resistance but increased fabrication complexity
  • Graphene nanoribbons (GNRs): Confined geometry with bandgap opening for transistor channels
Breakthrough Potential: Graphene’s combination of high thermal conductivity (>3000 W/mK) and high current-carrying capacity makes it uniquely suited for on-chip interconnects that must simultaneously carry electrical signals and dissipate heat from densely packed transistors.

3. Standardization Needs for Commercialisation

The report identifies critical gaps that must be addressed before nanoscale contacts achieve widespread commercial adoption:

  • Reliability testing: No standardised accelerated life test exists for CNT or graphene interconnects
  • Metrology: Measurement methods for contact resistance at the nanoscale are not standardised
  • Defect characterisation: Techniques for identifying and quantifying defects in nanomaterials lack uniformity
  • Environmental stability: Standard protocols for evaluating oxidation, humidity, and temperature effects are needed
Engineering Reality: As of this report’s publication, no nanoscale contact or interconnect technology had been commercialized in mainstream semiconductor manufacturing. CNT vias were the closest, with prototype integration in 3D memory devices under evaluation.

Engineering Design Insights

  1. Contact resistance dominates — in nanoscale devices, the contact resistance between the nanomaterial and conventional metal electrodes often exceeds the intrinsic channel resistance, making contact engineering the primary challenge
  2. Integration temperature constraint — back-end-of-line (BEOL) processing limits temperatures to below 400 C; most high-quality CNT and graphene growth requires higher temperatures, necessitating transfer or low-temperature growth methods
  3. Density matters — sparse CNT forests or graphene wrinkles reduce effective cross-section; achieving >10% packing density in CNT vias is the minimum threshold for competitive performance with copper
  4. Reliability unknown — while individual CNTs show remarkable robustness, bundle reliability under thermomechanical cycling is poorly understood compared to the decades of data available for copper
  5. System-level co-design — the advantages of nanoscale interconnects extend beyond resistivity; their unique thermal, mechanical, and high-frequency properties require co-optimization with circuit design for maximum benefit

FAQs

Q: Why can’t copper interconnects continue to scale?

A: At nanoscale dimensions, copper’s resistivity increases dramatically due to electron scattering at surfaces and grain boundaries (size effect). Additionally, electromigration resistance degrades as dimensions shrink, and the required diffusion barrier layers consume a growing proportion of the cross-sectional area.

Q: Are CNT interconnects in commercial use?

A: Limited commercial adoption has occurred primarily in 3D memory devices where CNT bundles are used as through-silicon vias (TSVs), but mainstream logic and memory chip manufacturers continue to use copper with advanced barrier technologies.

Q: What role does IEC 62632 play in nanotechnology standardization?

A: This technical report surveys the state of the art and identifies standardization needs. It serves as a roadmap for future IEC work on nanoscale contact metrology, reliability testing, and material specifications in the broader context of IEC TC 113 (Nanotechnology standardization).

Leave a Reply

Your email address will not be published. Required fields are marked *