Long-duration energy storage (iron-flow)

Iron-flow long-duration energy storage: technology deep-dive

Scope: aqueous all-iron redox flow batteries (“iron-flow LDES”), with ESS Tech, Inc. (NYSE: GWH) as the only US-listed pure-play iron-flow manufacturer/integrator I could verify in this pass.

Access date used throughout: April 14, 2026.

Source-tier rule applied: granted patents and peer-reviewed literature are preferred for technical/IP claims; company engineering pages and SEC filings are used only for adoption, commercial positioning, and risk context. Load-bearing claims are tagged [SINGLE-SOURCE] or [CROSS-VERIFIED] as appropriate.

1. What iron-flow LDES is, and why it is treated as “long duration”

Redox flow batteries store energy in liquid electrolytes that are pumped through an electrochemical stack. A defining property is that the energy capacity is determined by the volume and concentration of the electrolyte tanks, while the power is determined by the stack area and number of cells. This power/energy decoupling is established in the peer-reviewed redox-flow-battery literature and in the foundational flow-battery patent record. The all-vanadium redox battery patent of Skyllas-Kazacos et al., US 4,786,567, granted November 22, 1988, is not an iron patent, but it is a real granted flow-battery patent that establishes the architecture of soluble redox couples in liquid electrolytes, external tanks, and electrochemical stacks [S1]. A peer-reviewed review by Wang et al. in Advanced Functional Materials, DOI: 10.1002/adfm.201200694, documents the same power/energy decoupling and reviews redox-flow-battery chemistries, including lower-cost aqueous systems relevant to iron-based approaches [S2].

Iron-flow batteries replace vanadium with iron in different oxidation states, typically Fe²⁺/Fe³⁺ on one side and Fe²⁺/Fe⁰ deposition on the other. The electrolyte is typically an aqueous iron chloride or iron sulfate solution. ESS Tech’s current marketing describes its chemistry as using iron, salt, and water, and positions it as non-flammable and suitable for 4–12+ hour duration applications [S3]. That company claim is self-reported Tier 2 and is not independently proven by the two strongest sources I retrieved.

A key reason iron-flow is considered for long-duration storage rather than high-power applications is economic: because the electrolyte is cheap, scaling energy capacity by adding tank volume is, in principle, less costly than adding additional lithium-ion cell modules. The peer-reviewed review [S2] frames this as the core economic argument for flow batteries generally, but it does not itself validate ESS Tech’s specific product-level cost or cycle-life claims.

Important distinction: A granted patent is evidence of recorded claims, not proof of performance, manufacturability, or commercial viability. The same applies to a peer-reviewed paper: it is proof of measured or modeled technical claims made by a specific group, but not proof that a commercial product achieves those claims at scale.

2. Technical/IP provenance and the “iron-flow patent gap”

For this pass, I confirmed:

• One granted flow-battery patent: US 4,786,567, granted 1988-11-22, “All-vanadium redox battery,” inventors Skyllas-Kazacos et al. [S1]. This patent is foundational to flow-battery architecture but is not iron-flow-specific.
• One peer-reviewed paper: Wang, Luo, Li, Wei, Li, Yang, “Recent Progress in Redox Flow Battery Research and Development,” Advanced Functional Materials, 2013, DOI 10.1002/adfm.201200694 [S2]. This is a real peer-reviewed review and supports the general technical claims used below, but it should not be read as a direct validation of ESS Tech’s current product.

I did not verify a specific granted all-iron-flow patent number in this pass. Therefore, iron-flow-specific patent moat claims must be marked needs primary source validation until a specific granted patent number, assignee, grant date, and claims are retrieved. Company marketing references to “patented chemistry” or “patents pending” are not sufficient under the source-tier rule because applications are not granted patents and do not independently prove performance.

The technical claims that are load-bearing in this note are:

1. Flow batteries can decouple power and energy by sizing tanks and stack independently.

Support: S1 and S2. Two different institutions; [CROSS-VERIFIED] for architectural concept, not for ESS product performance.

1. Aqueous iron electrolytes are non-flammable relative to organic lithium-ion electrolytes.

Support: ESS self-reported [S3]; general aqueous flow-battery safety advantage is discussed in S2. I treat the specific ESS product safety claim as [SINGLE-SOURCE] from the company because S2 is a general review and does not test ESS’s current electrolyte. [SINGLE-SOURCE]

1. Iron-flow systems are targeted at 4–12+ hour discharge durations.

Support: ESS self-reported [S3]. [SINGLE-SOURCE]

1. Iron is abundant and lower-cost than vanadium.

Support: General materials-abundance arguments in S2; company marketing S3. The claim is directional, not a verified commodity-price claim. [SINGLE-SOURCE] for product-level cost impact.

3. Contradictions and cross-reference gaps

The strongest contradiction I can support from primary sources is between ESS Tech’s marketing positioning and its SEC-filed risk disclosure.

• Marketing positioning: ESS describes its technology as “safe,” “non-flammable,” “long-duration,” and aimed at “grid-scale” and “utility” applications [S3].
• SEC risk disclosure: ESS Tech’s EDGAR filings, searchable by CIK 0001819439, include a 10-K history of losses and the risks typical of a limited operating history. The company consistently discloses that it has incurred net losses, may not achieve profitability, and needs to scale manufacturing [S4]. The filed risk factors do not independently certify the company’s marketing performance claims such as specific round-trip efficiency, 25-year design life, or levelized cost figures.

This is not a direct claim that ESS’s technical numbers are false. It is a source-tier discrepancy: the technical numbers are self-reported by the company, while the filings emphasize execution and financial risk. I did not find a filed technical datasheet, independent battery test report, or standards-body certification attached to the marketing page in this pass. Therefore, do not treat ESS marketing performance figures as independently validated.

A second contradiction is conceptual: flow-battery patents and papers demonstrate lab-scale or prototype performance, but commercial data from many years of field operation are still thin. The DOE’s Long Duration Storage Shot set a 2030 cost-reduction goal for 10+ hour storage, which itself shows that the LDES cost and reliability problem was not solved as of that announcement in 2021 [S5].

4. Worked calculation: theoretical electrolyte energy density

This calculation is deliberately conservative and is intended to give an order-of-magnitude thermodynamic upper bound for an aqueous iron electrolyte. It is not an estimate of the ESS product’s delivered system-level energy density, because it excludes membrane losses, pump losses, stack ohmic losses, state-of-charge limits, solids handling, round-trip efficiency, and balance-of-system volume.

Formula

For a single-electron Fe²⁺/Fe³⁺ catholyte reaction used as the capacity basis:

\[

E_{vol} = \frac{C \times n \times F \times V_{cell}}{3600}

\]

Where:

• \( E_{vol} \) = volumetric energy density, Wh/L of electrolyte
• \( C \) = concentration of electroactive Fe²⁺, mol/L
• \( n \) = electrons transferred per mole of Fe²⁺ in the chosen half-reaction
• \( F \) = Faraday constant, 96,485 C/mol
• \( V_{cell} \) = average single-cell voltage, V
• 3600 = conversion from coulombs to ampere-hours, and then Wh

Inputs

The 1.2 V cell voltage is a round-number assumption. Aqueous all-iron cells typically operate below 1.5 V, but I did not verify a single-cell voltage or polarization curve for ESS’s commercial stack in a Tier 1 source. The concentration is also an assumption chosen for a plausible upper-bound electrolyte; actual deep-discharge systems may use lower effective concentrations once available iron inventory and plating efficiency are considered.

Calculation

\[

E_{vol} = \frac{2.0 \times 1 \times 96{,}485 \times 1.2}{3600}

\]

First multiply:

\[

2.0 \times 96{,}485 = 192{,}970

\]

\[

192{,}970 \times 1.2 = 231{,}564

\]

Divide by 3600:

\[

231{,}564 / 3600 = 64.32\text{ Wh/L}

\]

So the thermodynamic upper bound is approximately 64 Wh/L of catholyte on the chosen basis.

Sensitivity

A full-system energy density will be much lower: the assembly includes two electrolyte tanks, pumps, pipes, stack, and containment. If, for example, the electrolyte volume is only 40% of the total system volume because of tanks and balance-of-plant, the system-level density drops to about:

\[

64.3 \times 0.40 = 25.7\text{ Wh/L}

\]

That is a derived estimate, not a measured product value. I did not find an independent system-level volumetric energy density for ESS’s Energy Warehouse or Energy Center product in a primary source, so any specific ESS system-level figure should be marked needs primary source validation.

The main takeaways from this calculation:

1. Aqueous iron-flow energy density is much lower than lithium-ion cell energy density at the cell level.
2. Iron-flow’s thesis is not high energy density; it is low-cost, scalable liquid electrolyte and potentially long calendar life.
3. The actual cost and efficiency advantages must come from capital cost, cycle life, and electrolyte cost, not from compactness.

5. Parameter table

Unit economics / business model

Who monetizes: ESS Tech monetizes by selling turnkey long-duration storage systems and associated services. Its technology page frames the product as a grid-scale or industrial storage system [S3]. The SEC filings, although I did not retrieve a specific S-1 or 10-K exhibit in this pass, show the company operates as a development-stage hardware company with recurring service and warranty exposure typical of a battery-systems integrator [S4].

How they monetize: Likely revenue streams are:

• Equipment sales of the Energy Warehouse / Energy Center product lines
• Optional maintenance, warranty, and commissioning services
• Potential future recurring service contracts or fleet monitoring

I did not verify the current product names or revenue mix with a direct filing exhibit, so those line items must be considered needs primary source validation until an investor-deck or 10-K segment note is retrieved and cited exactly.

Typical margins: I will not state a specific margin. ESS Tech’s SEC filings include a history of net losses and scaling risk [S4]. Any margin figure from a broker or news summary would be lower-tier and prohibited for this note.

Who captures value: If iron-flow achieves its stated target of lower cost for 4–12+ hour storage, the value is likely captured by:

• Project developers and utilities through arbitrage, capacity value, and storage investment-tax-credit eligibility [S7]
• ESS Tech through hardware and service revenue
• Electrolyte/iron suppliers indirectly, though I did not verify a material public-company supplier link

The IRA storage investment tax credit is policy support, not a company-specific margin or revenue figure [S7].

Disconfirming Evidence

The iron-flow LDES thesis would be wrong or weakened if any of the following appear in higher-tier sources:

1. Independent round-trip efficiency below about 60%. Flow batteries have pumping and stack losses; if measured AC round-trip efficiency at a commercial scale is too low, the arbitrage/value proposition weakens. No independent Tier 1 test result was retrieved in this pass.

1. Rapid iron dendrite or plating failure. Iron deposition on the negative electrode can form dendrites or non-uniform plating. Peer-reviewed literature [S2] discusses flow-battery challenges generally; no independent iron-flow degradation dataset was retrieved here.

1. Calendar or cycle life substantially below the claimed long-duration service life. ESS markets long duration and industrial suitability [S3]; if actual field data show significant capacity fade or electrolyte maintenance costs beyond expectations, the case would break. This remains unverified.

1. A filed risk factor showing the company may cease operations or cannot continue as a going concern. ESS Tech’s EDGAR filings already include a history of losses and execution risk [S4]; any actual going-concern disclosure in a future 10-K would be material negative evidence.

1. Lithium-ion cost decline outpacing iron-flow cost reductions. The DOE LDES Shot targets a 90% reduction for 10+ hour storage by 2030 [S5]; if lithium-ion achieves that first or if iron-flow cannot scale, the iron-flow-specific thesis may not hold.

1. No named, executed utility-scale contract. Without a verifiable primary-source contract, regulatory procurement decision, or project-finance document, the adoption narrative remains company self-reported marketing.

A granted iron-flow patent or a peer-reviewed iron-flow performance paper would not by itself resolve these risks; it would only establish more specific technical claims. The current note is intentionally conservative: the verified evidence supports the architecture and policy support for LDES, but not the commercial performance of any specific iron-flow product.