Graphite is the layered crystalline form of carbon that already anchors lithium-ion anodes and laboratory surface science, and the current research frontier is not the material itself but control of its stacking order, intercalation chemistry and interfaces.
Summary
The parameters that decide graphite’s value in each application are different. For anodes, it is formulation and process feasibility rather than the carbon itself: an AI-guided iterative workflow on an industrial dataset raised the fraction of cells delivering at least 350 mAh/g from 28.4% to 84.8% and capacity retention from 42.1% to 97.3%, while moving from frequent process failures to 100% successful cell production. For quantum electronics, it is the abundance and area of rhombohedral stacking, which is scarce in natural graphite; a cryogenic shock exfoliation route plus low-pressure van der Waals assembly produced devices exceeding 1300 square micrometres at 90% fabrication yield, with a disorder mean free path above 200 micrometres. For intercalation compounds, it is the kinetics: sodium catalysis was shown to form superconducting CaC6 in Ca/Na/graphite mixtures at room temperature, a reaction previously thought to need elevated temperatures, with the CaC6 fraction scaling as the square root of storage time.
Graphite also appears constantly as a reference or an enabling component rather than the object of study. It is the benchmark surface against which graphene coatings are compared for rarefied-gas drag, the spectroscopic yardstick for judging whether reduced graphene oxide has recovered its sp2 network, with a pi*-sigma* separation of 7.0 eV against graphite’s 7.1 eV, the felt heater in ultrafast high-temperature sintering rigs, the standard sample for validating thermal metrology workflows, a proximity-screening layer that lengthened hot-carrier cooling by up to a factor of four and raised internal photoresponsivity by about 50% in a graphene photodetector, and a diamagnetically levitated proof mass in a room-temperature magnetometer reaching 4.5 pT per root hertz. Assessing graphite therefore means assessing a mature material whose upside sits in processing and stacking control, not in discovery.
Evidence base
- AI-guided iterative workflow on industrial graphite anode data raised the fraction of cells at or above 350 mAh/g from 28.4% to 84.8%, capacity retention from 42.1% to 97.3%, and fabrication success to 100% (2 Jun 2026).
- Cryogenic shock exfoliation plus low-pressure van der Waals assembly produced rhombohedral graphene devices above 1300 square micrometres at 90% yield, with uniform spin magnetism over 10 by 10 micrometres and disorder mean free path above 200 micrometres (7 Jul 2026).
- Sodium catalysis produced superconducting CaC6 in Ca/Na/graphite mixtures at room temperature, previously considered highly unlikely, with CaC6 amount scaling as the square root of storage time (18 Jun 2026).
- Density functional theory finds bulk rhombohedral graphite is a weak topological insulator with spin-orbit coupling, with drumhead surface states reaching approximately |Omega|/TrG = 1 at the K point (31 Jul 2026).
- Proximity screening by a nearby graphite layer lengthened hot-carrier cooling time by up to a factor of four and improved internal photoresponsivity by about 50% in a waveguide-integrated WSe2-graphene-WSe2 photodetector (7 Jun 2026).
- A diamagnetically levitated graphite plate coupled to a nano-trampoline resonator (Q = 6 million at 443 kHz) reached 4.5 pT per root hertz at room temperature without magnetic shielding (27 Jul 2026).
- The US Department of Energy is backing Solidion’s development of domestically made battery graphite, with no scale or timeline disclosed in the source (8 Dec 2025) ref.
Open questions
- Can rhombohedral stacking be grown or induced directly at wafer scale, rather than exfoliated and sorted, and at what defect density relative to the >200 micrometre mean free path already demonstrated?
- Does room-temperature, sodium-catalysed CaC6 formation survive electrochemical cycling in a working Ca-ion cell, and can the square-root-of-time kinetics be accelerated to useful rates?
- Does the AI-guided anode optimisation result generalise beyond the single industrial dataset and platform used, and do the 84.8% and 97.3% figures hold at pilot and production volume?
- What capacity, cost and timeline sit behind the DOE-backed push for US-made battery graphite, none of which the available source discloses?
Assessment drafted 2026-08-31 from up to 18 KB sources using the technology-scorecard framework; scores are a draft read pending review.