Grid-scale battery storage — projects at or above 1 MW nameplate — absorbed the bulk of new utility storage in 2021, when US capacity tripled year-on-year [S4]. The market continues to be anchored to lithium-ion cells manufactured overwhelmingly in China, with downstream Bankability now dictated as much by cell traceability and tariff exposure as by round-trip efficiency [S5][S6].
Process engineers spec'ing a BESS today are no longer asking only kWh vs. cost. They are weighing LFP vs. NMC cathodes, closed-loop frequency response duty cycles, and whether their cell supplier sits inside or outside the 90-day US-China tariff window that opened on 2025-05-12 [S5]. Industrial Info Resources now tracks mining, refining, cell, pack, and integration assets on a single platform precisely because the chain has become the project risk [S6].
Cell chemistry and pack architecture: what the 2026 spec actually says
Lithium-ion remains the workhorse chemistry for grid BESS, favored for its high efficiency, long cycle life, and modular packaging versus pumped hydro, CAES, or flywheel alternatives [S2]. The Nature Reviews Clean Technology assessment published 2025-06-20 confirms LIB dominance at grid scale but flags a parallel push toward aqueous zinc, redox-flow, and high-temperature sodium chemistries for long-duration service where cycle cost per kWh matters more than footprint [S3].
For a 4-hour utility system, the practical trade is between LFP (lower energy density, longer cycle life, cobalt-free) and NMC (higher energy density, higher specific energy per rack). Inside a storage rack, this maps to more or fewer parallel strings for a given MWh. Choice of cathode then cascades into the dc power supply sizing for the DC bus and the thermal envelope required to keep cells inside vendor warranty limits. Closed-loop frequency response simulations on a variable-inertia grid — University of Sheffield work indexed 2026-05-16 — show battery round-trip efficiency of roughly 85-90% as the design point below which reserve service revenue collapses [S1].
Supply chain geography and the 2025-2026 tariff shock
China still controls the majority of the lithium battery value chain, from upstream refining through cell, module, and pack assembly [S5][S6]. On 2025-05-12, the US and China agreed to a 90-day reciprocal tariff reduction, described by US Treasury Secretary Scott Bessent as direct relief for a battery industry whose headline tariff scenarios had rendered many project pipelines unbankable [S5].
The practical sourcing consequence is a shift toward qualifying non-China cell suppliers — Korean (LG, Samsung SDI), Japanese (Panasonic), and emerging US gigafactories — for IRA-eligible projects, while keeping China-sourced packs for merchant and international builds where tariff pass-through is acceptable. The Industrial Info Resources database (updated 2026-06-05) tracks mining, refining, cell, pack, and storage integration assets globally, with the operator noting that the 2026/27 ecosystem is now read as a single risk surface rather than a series of independent purchase orders [S6]. A useful parallel is the PV-bundled approach covered in the mid-2026 energy storage sourcing note, which documents how integrators are co-sourcing cathode precursor and module to lock in cell allocation.
Market sizing: $24.5 B by 2030, but bankability is the gate
IndustryArc's December 2024 update sizes the global grid-scale battery market at a projected USD 24.5 billion by 2030, expanding at a 22.6% CAGR over 2024-2030, driven by renewable integration mandates and falling cell prices [S7]. That headline, however, is filtered by what financing parties will underwrite, and that is where chemistry traceability, cycle-life warranties, and tariff-pass-through clauses become the binding constraints.
The 2021 EIA data — US utility-scale battery capacity tripled in a single year, the largest single-year capacity addition on record at the time — established the run-rate that the 2024-2030 forecast extrapolates [S4]. From a process-engineering standpoint, the question is no longer whether to deploy batteries, but how to spec the BESS to survive 6,000+ cycles at 85-90% round-trip efficiency while remaining compliant with shifting US, EU, and UK cell-content rules [S1][S2].
BOS, PCS, and the 4-20 mA/HART layer that ties it together
Outside the cells, the balance of system is dominated by the power conversion system (PCS), the industrial UPS architecture for ride-through, and the switching power supply rails that feed protection relays and the storage cage monitoring layer. For new builds, process engineers should verify that the PCS is rated for the project's continuous C-rate (typically 0.25C to 0.5C for 4-hour systems) and that the BMS can publish state-of-health, state-of-charge, and cell-level voltage to a SCADA over a documented protocol. [S2]
The 4-20 mA analog loop with HART (FSK superimposed on the analog current) remains the dominant field-instrument layer for temperature, smoke, and gas sensing inside a storage rack enclosure. This is a different protocol family from Foundation Fieldbus or PROFIBUS PA, which are fully digital; engineers retrofitting older sites must keep HART devices on their 4-20 mA loops rather than attempting to bridge them onto the digital segments, and new builds should preserve that split for spare-parts commonality across fleets.
Selection criteria: who BESS is for, and where it loses
Battery storage is the right spec for 0.25-4 hour shifting, fast frequency response, and inertia replacement on low-rotational-mass grids; it loses to pumped hydro for 8-24 hour discharge at fixed sites with suitable topography, and to thermal or hydrogen storage for multi-day seasonal shifting [S3]. Plants that already host SCADA, have an established industrial UPS maintenance contract, and operate in a market with a frequency-response ancillary service should see the strongest project IRR.
Operators in remote microgrids, behind-the-meter C&I sites, and merchant developers co-located with solar or wind should evaluate BESS on three criteria: (1) round-trip efficiency at the operating C-rate, (2) warranted cycle count at 80% depth-of-discharge, and (3) tariff scenario sensitivity on the cell supply contract. The third criterion is the one that has changed most sharply since the 2025-05-12 tariff relief and the subsequent 90-day renewal cycles [S5].
Standards, sourcing signals, and what to track next
Engineers should anchor BESS procurement to the relevant UL 9540 / UL 9540A, IEEE 1547, and IEC 62933 series, with NFPA 855 governing outdoor enclosure spacing. These are the documents that financiers and AHJs read, and they are the cleanest place to align spec text with warranty language. Sourcing decisions should be cross-checked against the IIR supply-chain database, which now exposes cell-factory commissioning dates, cathode-active-material precursor flows, and project-level capex [S6].
Two signals worth tracking into the second half of 2026: whether the US-China 90-day tariff window is renewed, extended, or made permanent, and whether non-China LFP capacity (US, Korean, Indonesian) reaches the gigawatt-hour scale needed to underwrite IRA-compliant merchant storage. A third, longer-cycle signal is the 2025-06-20 Nature Reviews note that redox-flow and high-temperature chemistries are moving from pilot to multi-MWh demonstration, which will eventually re-open long-duration service for batteries that today's LFP-and-PCS architecture cannot serve economically [S3].