What Happens to New Zealand's Grid-Scale Batteries When They Wear Out?

 New Zealand commissioned its first utility-scale battery in 2024. Several more are coming through 2026 and 2027 — Glenbrook, Huntly, Ruakākā, the Eku Energy Waikato project. None of them is anywhere near end of life. Which is precisely why this is the moment to ask what happens when they reach it, because the answer will shape national infrastructure decisions a decade before the first system needs replacing.

Energy storage is a long-cycle asset class. The decisions being made in 2026 will, in the case of lithium-ion installations, produce a wave of end-of-life material around 2036 to 2040. That sounds far off, until you remember that the recycling infrastructure to handle it would need to be built years in advance, and that countries further down the deployment curve than New Zealand are already discovering how poorly that timing tends to work.

The Li-Cycle warning

In January 2025, one of North America's most prominent lithium-ion battery recyclers, Li-Cycle, filed for bankruptcy protection. Its half-built flagship Rochester plant — intended to process the first major wave of retired lithium batteries from electric vehicles and grid storage — became a case study in how hard the economics of battery recycling actually are. Cost overruns. Engineering delays. The kind of difficulties that look manageable on a spreadsheet and become existential when scaled to industrial throughput.

The Li-Cycle story did not get much coverage outside trade press, but it told the industry something important. The headline recycling rate frequently quoted for lithium-ion batteries — sometimes as high as 95% — describes what is theoretically achievable in a fully-built recycling ecosystem. The actual recycling rate of installed lithium batteries globally sits much closer to 5 to 15 per cent, with most retired cells still routed to landfill or specialised hazardous waste streams. The gap between the marketing figure and the operating figure is the gap that bankrupted Li-Cycle.

New Zealand currently has no lithium-ion battery recycling infrastructure at the scale that will be needed when Rotohiko, Ruakākā, and their successors reach end of life. That is not a criticism of any specific project — it is simply a fact about a country whose first utility-scale battery commissioned 18 months ago. But it does mean that the disposal pathway for those batteries, in the late 2030s, is currently unspecified.

The chemistry that does not have this problem

The end-of-life conversation for vanadium flow batteries is meaningfully different, and worth contrasting honestly. Sumitomo Electric's Yokohama Works installation has now demonstrated 24 years of continuous service on the same vanadium electrolyte — first in a customer system from 2001 to 2011, then transferred to a different battery at the Yokohama facility from 2012 to present. The electrolyte has not been "used up" in the chemical sense, because vanadium ions cycle between oxidation states without being consumed. At end of system life, the electrolyte is functionally the same material that entered the system on day one.

Sumitomo's published total recyclability rate for a complete vanadium flow battery system is 99.2 per cent — 70 per cent of which is electrolyte reused, 29.2 per cent components recycled, and 0.8 per cent disposed. U.S. Vanadium demonstrated a 97 per cent recovery rate from decommissioned VRFB electrolyte in 2021, using the same plant infrastructure originally built to manufacture electrolyte. Peer-reviewed life cycle assessment in Energy Technology (Wiley, 2024) has validated the process design across multiple contamination scenarios.

None of this is theoretical. It is operational fact, backed by independent academic publication.

What the next decade of NZ procurement should be modelling

The implication is not that lithium installations are bad investments. They are mature, well-understood, and appropriate for the duty cycles they currently serve. The implication is narrower and more specific: end-of-life cost should be a line item in the project finance model for every grid-scale battery being scoped in New Zealand right now, and it is currently not.

For lithium installations, that line item should reflect the actual recycling economics of the late 2030s, not the optimistic recycling economics of the marketing materials. For flow installations, it should reflect the documented residual value of recoverable electrolyte at end of system life — typically 30 to 50 per cent of original system cost, depending on vanadium commodity pricing at the time. Comparing the two chemistries on full lifecycle terms looks substantially different from comparing them on capex alone.

For council buyers, ESG-conscious corporate procurement teams, and any utility commissioning long-life infrastructure, the question of what happens to a battery in 2038 deserves a written answer in 2026. The recycling and disposal landscape is going to look very different by then. The procurement decisions made today are betting, implicitly, on which version of that landscape arrives.

The unfashionable timing question

End-of-life is the least exciting part of the energy transition conversation. It is also the part that determines whether the transition is genuinely sustainable or merely fashionable. The Li-Cycle bankruptcy was a small story in early 2025. By the late 2030s, when the first major wave of retired grid-scale lithium reaches end of useful life, it will be a much larger story — and the countries that quietly built the right disposal and replacement infrastructure in the years between will look noticeably better than the ones that did not. New Zealand still has time to be in the first group. The decisions that get it there are being made now.

The Zion Technologies team are New Zealand's exclusive distributor for Rongke Power vanadium flow battery systems. Based in Pokeno, Waikato, the company supplies utility, commercial, and community-scale energy storage projects across New Zealand, Fiji, Tonga, Samoa, and the Cook Islands.

The views expressed in this article are those of the authors.