The robot’s battery has less than 2 kWh of capacity, yet the few grams of adhesive inside are indeed crucial

In 2026, humanoid robots surpassed the threshold of mass production. TrendForce predicts that global shipments will exceed 50,000 units this year, marking a 700% year-on-year increase. Chinese robots can run 100 meters in under 9 seconds, marking the first time this industry has truly entered the “era of mass production.” While everyone focuses on the robots themselves, I focus on the battery they carry—more precisely, the few grams of gel inside the battery. A humanoid robot’s battery typically has a capacity of less than 2 kWh. The UR Space H1 has a capacity of 0.864 kWh, with a static runtime of under 4 hours; the Tesla Optimus Gen2, equipped with a 2.3 kWh high-nickel ternary battery, has a dynamic runtime of only about two hours. Compared to the battery packs in electric vehicles, which often range from 60 to 100 kWh, this is barely a fraction of that. Yet it is precisely this “small” capacity that has brought out the full temperament of the entire battery supply chain.

First, let’s see how difficult the battery is.

The challenges lie in four areas. First is space: the humanoid robot’s torso must house batteries, motors, reducers, control boards, and sensors, leaving only a narrow space with a highly constrained geometry for the batteries. Second is weight: robots are highly sensitive to every gram; if the battery adds one gram, the joints and motors must also bear an additional gram, which reduces both battery life and motion quality. Third is power: robots require very high instantaneous power peaks during acceleration, climbing, and rapid stopping, necessitating batteries capable of withstanding high-rate discharge. Fourth is heat: as energy density continues to increase, heat generation becomes more concentrated; dissipating this heat within a palm-sized space is far more challenging than for electric vehicles.

These four challenges are precisely why battery technology is undergoing a generational shift, and why solid-state batteries have gained prominence.

Solid-state batteries are the key. By replacing liquid electrolytes with solid-state electrolytes, they offer higher energy density, improved safety, and a more compact form factor. XPeng IRON utilizes a fully solid-state battery; according to official statements, this reduces weight by 30% while increasing capacity by 30%. GAC GoMate employs a fully solid-state design, achieving a battery life of 6 hours. Farasis’s first-generation sulfide fully solid-state cell delivers an energy density of 400 Wh/kg, which is said to enable robots to operate continuously for 8 to 12 hours. EVE Energy’s “Longquan II” achieves a volumetric energy density of 700 Wh/L, specifically targeting humanoid robots and low-altitude aircraft. Zhongchuang Innovation’s self-developed fully solid-state battery has reached over 430 Wh/kg. CATL’s fully solid-state battery is scheduled for small-batch production in 2027. Foreign companies are also active: Samsung SDI showcased a pouch-type fully solid-state battery at InterBattery 2026, emphasizing lightweight design; LG Energy Solution has directly planned a “anode-free” solid-state battery for robots, aiming to maximize energy density per unit volume.

There has also been no standardization of the technical pathways. The three approaches—oxides, sulfides, and polymers—are each pursuing their own direction, with soft-pack, prismatic, and anode-free configurations all under development. However, the overall direction remains the same: smaller, lighter, and denser.

Every step forward for solid-state batteries necessitates a corresponding step forward in the formulation of the battery encapsulant. This is where the term “miniaturization adaptation” originates. It is not merely a matter of reducing the amount of encapsulant; every single performance metric of the encapsulant must undergo a rigorous re-evaluation under more constrained space, more concentrated heat generation, and stricter weight requirements.The first challenge is space. The gaps within electric vehicle battery packs are on the millimeter scale, whereas in humanoid robots, the gaps between cells and between cells and the housing have reached the sub-millimeter scale. The encapsulant must have low viscosity—typically in the range of 1,000 to 2,000 cP—to allow it to seep into these narrow gaps via capillary action without trapping air bubbles; it must also enable precise dispensing, with both the amount and trajectory of the encapsulant tightly controlled to prevent it from flowing into unintended areas. The traditional high-flow potting approach completely fails in this context.。

Battery encapsulation and heat dissipation Stage Two: Heat dissipation. When the battery size is reduced, the energy density actually increases, causing heat generated during high-rate discharge to become more concentrated. The thermal conductivity of the thermal conductive encapsulant must be pushed higher—moving from the conventional level of around 1.0 toward the range of 1.5 to 3.0, or even higher. Furthermore, the encapsulant layer must be thin; the thinner it is, the lower the thermal resistance. However, adding more thermal conductive fillers increases viscosity, creating an inherent contradiction that requires careful formulation and balance. Simultaneously, the encapsulation process must have low heat release and low shrinkage; the delicate battery cells must not be damaged by heat generated during curing. Stage Three: Lightweight design and safety. Robots are highly sensitive to every gram; therefore, the encapsulant layer must be thin and the material usage must be calculated by grams, yet the bonding strength must not be compromised—the thin layer must withstand vibration and impact. Requirements for insulation, flame retardancy, and sealing have also become more stringent. While solid-state batteries are safer, the encapsulation, insulation protection, and moisture-proof sealing of pouch cells are equally indispensable. Thermal cycling resistance and vibration resistance are both critical thresholds.。

Dispensing on the production line: Once these three stages are completed, there is a fourth stage: the process. As battery sizes shrink, the level of production line automation increases, requiring the curing method for the adhesive to be adapted: either UV photopolymerization with seconds of positioning, or low-temperature rapid curing integrated into the press cycle. Batch consistency must be far more stringent than before; after all, the amount of adhesive in a single robot battery pack may be less than 10 grams—any slight batch fluctuation can render the entire pack unusable. Thus, you can see that the “miniaturization and adaptation” of battery adhesives represents a comprehensive reconfiguration spanning from formulation to process. Who is competing for this market share? Foreign companies have a first-mover advantage, with firms like 3M, Henkel, and Dow having deep expertise in electronic adhesives and thermal conductive materials. However, domestic manufacturers also have their opportunities: in scenarios such as battery thermal interface packaging and magnetic steel bonding, Shanghai Jome has been deeply engaged for years, offering performance on par with international standards, faster delivery times, and quicker response times. Adhesives for humanoid robots represent another new battleground for domestic substitution. According to forecasts from TrendForce, demand for solid-state batteries will surge from 0.05 GWh in 2025 to 74.2 GWh in 2035—a growth of over a thousandfold. Battery adhesives will follow this trend, making this incremental growth space a tangible reality. A volume of 50,000 humanoid robots is merely the starting point. As batteries become smaller, adhesives are becoming thinner; competition in this sector has only just begun.

Mapping JOME Adhesives to the Four Critical Zones of a Sub-2 kWh Robot Battery Pack

Translating torso-level energy-density targets into a manufacturable humanoid or quadruped battery module requires four distinct polymer functions working inside the same compact enclosure: structural cell-to-cold-plate bonding, BMS high-current MOSFET cooling, inter-cell dielectric cushioning, and serviceable IP67 lid sealing for field pack swaps. Using a rigid brittle epoxy across thin-walled 21700 cylindrical cans or high-silicon pouch cells risks pouch seam tearing during 15G torso drop events, whereas an unbonded thermal grease provides zero structural shear retention.

Robot Battery Pack Zone Electro-Thermal & Drop-Shock Requirement Verified JOME Grade Key Specification & Assembly Benefit
Cell-to-Cooling-Plate Structural Base ≥8–12 MPa lap shear + conductive heat extraction without heating cells JOMP 2650 A/B (1.3 W/m·K) & JOMP 2652 A/B (2.0 W/m·K) 2K room-temp cure, low VOC, tough elastomeric peel resistance under impact
BMS Power Board & Current Shunt TIM Low thermal resistance across uneven MOSFETs & Hall sensors JOMG 1980 (8.0 W/m·K 2K Filling Gel) Rapid 2K cure with zero slump under multi-axis humanoid torso tilt
Wire-Bond / Busbar & NTC Sensor Potting Vibration damping, arc suppression & wide -60/+200°C window JOMS 1850-20 A/B (2.0 W/m·K Silicone Gel) Soft stress-relieving addition cure protects delicate ultrasonic wire bonds
Magnesium / CFRP Pack Housing Lid Gasket IP67 dust/moisture seal + non-destructive opening for cell service JM 3302 UV-Silicone Dual-Cure CIPG Sealant Replaces hand-laid O-rings; cures in seconds under inline UV LED

Thin-Bead Weight Budgeting & Dispensing Control for Mass-Produced Humanoid Packs

In a 1.4 kWh torso pack containing 80 to 110 cylindrical cells, trimming average bond-line thickness from 1.2 mm down to 0.45 mm cuts parasitic polymer weight by more than 140 grams while simultaneously halving conductive thermal resistance (ΔT across the adhesive layer). Achieving that thin, void-free footprint without metal-can-to-plate electrical shorting requires controlled filler particle top-cut and dielectric breakdown margins—parameters you can model directly in our potting & thermal bond-line dosage calculator. For full-robot actuator, harmonic drive, and sensor bonding outside the battery bay, see our companion robot manufacturing & actuator assembly adhesive guide and TIM supplier OEM qualification framework.

Shanghai Jome Industrial Co., Ltd. manufactures all battery-grade 2K thermal polyurethanes, silicone gels, and CIPG sealants under ISO 9001:2015 (86724Q10331R0S) and ISO 14001:2015, verified by 11 CTI RoHS 2.0 (EU 2015/863) reports. Request evaluation cartridges for cell-pack drop and thermal-runaway testing via our technical engineering team.

Frequently Asked Engineering Questions

Why do humanoid robot battery packs (<2 kWh) require structural thermal adhesives instead of mechanical cell brackets?

Torso cavity volume in bipedal humanoid robots is severely constrained; bonding cylindrical 21700/4680 or pouch cells directly to thin cooling plates with JOMP 2650 A/B (1.3 W/m-K) or JOMP 2652 A/B (2.0 W/m-K) 2K thermal polyurethane structural adhesive eliminates bracket mass while withstanding dynamic walking and fall-impact loads.

How do battery pack engineers balance heat dissipation during fast charging against cell-to-cell thermal runaway propagation?

Engineers create anisotropic thermal paths: high-conductivity structural PU adhesive or TCMP 140# thermal gel couples cell bottoms into the cold plate for rapid heat extraction, while lightweight silicone encapsulating gels and dielectric barriers isolate adjacent cell sidewalls.

Can robotics OEMs request free trial cartridges of JOME battery structural and potting adhesives?

Yes. Shanghai Jome Industrial Co., Ltd. supplies 2K thermal PU structural adhesives, silicone potting gels, and UV-silicone CIPG housing gaskets backed by ISO 9001:2015 and 11 CTI RoHS 2.0 test reports, with free evaluation samples shipped via UPS freight collect.

For Immediate Assistance, Please Contact us Now!

InquiryWhatsApp