If you assemble battery packs, integrate containerised BESS or run a storage EPC project and you buy prismatic LiFePO4 cells by the container (from a few hundred to several thousand units), the biggest risk in this purchase is not price — it is cell grade. On the grey market, Grade B is routinely sold as Grade A: relabelled, factory markings lasered off, "like new". A pack built from those cells looks fine at intake and falls apart within a year. Below is what LiFePO4 is, which specs to lock into the contract, and step by step how to verify that Grade A — not B — is what actually ships in your container.
Who buys cells direct, and why
Buying cells rather than finished packs makes sense for those who add their own value: BESS integrators, pack assemblers, storage EPC contractors. Buying the cell from the factory instead of a pack from a distributor gives you control over the BMS, pack topology, warranty and cost per kWh. But with that margin you take on what the pack maker used to cover: grade verification, IR matching, certification and dangerous-goods logistics.
Storage demand is climbing across every serious energy market — from C&I peak-shaving and backup in Europe, to utility-scale renewables build-out in the MENA region and Africa. Incentive regimes differ by country (for example, Ukraine exempts lithium storage above 100 Ah from VAT and import duty through 1 January 2029, verified at clearance), so treat local tax rules as one input, not the frame. We run cell and system sourcing end to end on our BESS energy-storage page, and cover the full import picture in how to import BESS from China.
What LiFePO4 is, and why it dominates stationary storage
LiFePO4 (lithium iron phosphate, LFP) is the cathode chemistry that displaced NMC in stationary storage for four reasons:
- Safety. LFP has the most thermally stable cathode: a high thermal-runaway onset temperature and far lower fire risk under puncture or overcharge. For an indoor containerised installation, that is decisive.
- Cycle life. A modern LFP cell delivers 6,000–10,000+ cycles at 80% DoD — a multiple of NMC. That figure directly sets your lifetime cost per kWh.
- Cost and raw materials. No cobalt or nickel means cheaper, more stable cost and less exposure to scarce metals.
- Regime tolerance. LFP handles full charge and deep discharge more gracefully, simplifying pack management.
The trade-off is lower gravimetric energy density than NMC. For stationary storage, where weight is not critical, that is an acceptable compromise — which is why the BESS market has effectively become an LFP market.
Key specifications you must pin
"How much is a cell" without these parameters is an unanswerable question. Put these in the contract and datasheet:
- Capacity (Ah). The 2026 mainstream is 314Ah / 320Ah (successors to the legendary 280Ah). The new utility-scale generation is 500+Ah: CATL, EVE (628Ah MB56), REPT (587/625Ah) and Hithium are already deploying these in large projects. Choose the format to fit your pack architecture, not "whatever is newest".
- Nominal voltage ~3.2 V, window 2.5–3.65 V. The pack is built from this up.
- Cycle life at DoD. Demand it as "N cycles at 80% DoD to 80% SOH" — a bare "10,000 cycles" with no conditions means nothing.
- Internal resistance (IR). Low and consistent across the batch — critical for pack matching.
- C-rate. Stationary BESS is typically 0.5C charge/discharge; peak-duty requirements are specified separately.
- Format and dimensions (prismatic, size, terminal type) — to fit your assembly.
- Manufacturer cycle warranty, not a trader's warranty.
Grade A / B / C: the risk nobody talks about
This is the layer sellers rarely spell out, and it decides the fate of the pack. Factories (EVE, CATL and others) sort cells by capacity, internal resistance and appearance:
- Grade A — full spec compliance: capacity in tolerance, low and even IR, no cosmetic defects. Goes into EVs, grid storage and premium BESS.
- Grade B — capacity shortfall, higher or scattered IR, cosmetic flaws; cells sorted out specifically to keep them out of automotive. Not "scrap" — honestly labelled and correctly matched, B-grade has a place in budget storage, but it must be priced accordingly.
- Grade C — clear deviations, out-of-spec capacity/IR, sometimes rebuilt or reused cells; unfit for serious BESS.
How B-grade ends up in your container labelled as A:
- Relabelling. The factory QR and grade are lasered off and an "A" is applied — you cannot tell the cell apart physically. This is why leading resellers warn about "conglomerates lasering off QR codes and passing Grade B off as Grade A".
- Batch mixing. Under-spec cells with scattered IR are blended into a pallet — the pack will never balance.
- CATL grey market. CATL does not sell cells to the public directly: anything listed on Alibaba or Made-in-China as "CATL Grade A" should be treated as suspect — often it is B-grade at best.
A Grade A cell and a relabelled Grade B are indistinguishable by eye. Only manufacturer traceability, matching data and a cycle test prove it — everything else is the seller's word.
The verification stack: how to prove Grade A
Protection is built from four layers that only work together:
- Traceability to the maker. Every cell carries a QR/serial that validates in the CATL / EVE / REPT database. A lasered, unreadable or "not-found" code is a red flag. First confirm the seller is a real factory or authorised channel, not a trader with an unknown BOM: how to find and verify a factory in China.
- Capacity and IR matching. Demand the grading report: capacity and internal-resistance spread across the batch must sit within tolerance (single-digit percent for a quality pack). This is a direct proxy for grade.
- Independent cycle / EL test. Sample-test real capacity under load and hidden internal defects. The classic "perfect sample, cheapened batch" scam is only closed by pre-shipment quality inspection.
- Full datasheet plus certificates for your batch. Not "available in the range", but for the specific model and lot, cross-checked against the QR.
A datasheet without QR verification is paper that can be drawn up; a QR without a cycle test is a number with no proof of capacity. Together they make grade substitution nearly impossible — this is what separates full-cycle sourcing from "just ship it".
Cell certification: what to require
For the cell and the pack, anchor on three key standards (always the current edition for your model — verify status):
- IEC 62619 — safety of industrial/stationary lithium cells and batteries (short circuit, overcharge, thermal and mechanical abuse).
- IEC 62620 — performance requirements (capacity, endurance), the companion to 62619.
- UN 38.3 — mandatory transport tests (altitude, vibration, shock, thermal cycling); without it a cell cannot legally ship by sea or air.
We break down certification and its traps separately in BESS certification — UN 38.3, IEC and more. Primary sources: IEC 62619 in the IEC catalogue and the UN Manual of Tests, Part 38.3. Note: lithium cells are dangerous goods (UN 3480/3481), and mishandled DG paperwork means containers held or bumped off the vessel.
Buying cells vs finished packs
A cell gives you full control over BMS, warranty and cost, but demands competence in assembly, matching and DG. A finished pack or rack is simpler to ship and to certify at system level, but you pay for someone else's build and trust their choice of cell grade. The rule is simple: integrators and assemblers buy cells; EPCs without an assembly line more often buy packs. Either way, cell grade must be proven — in a pack, that means the factory cell certificate for the cells inside.
Request Grade A cell pricing
Silk Way Sourcing runs LiFePO4 cells end to end: factory/channel vetting, QR grade verification back to the maker, capacity and IR matching reports, independent cycle testing and dangerous-goods logistics to your port.
Send us the format, capacity, volume and pack application, and we will request Grade A cell pricing with verification: contact@silkwaysourcing.com or WhatsApp +380 97 883 4765.
