A sizing tool for the vanadium redox flow battery (VRFB), a leading candidate for grid-scale energy storage. Adjust cell active area, stack configuration, electrolyte volume, vanadium concentration and current density, and see output power, stored energy and discharge duration update in real time. Experience the unique VRFB feature of decoupled power and energy scaling.
Parameters
Cell active area A_cell
cm²
Electrode area of the membrane-electrode assembly (MEA)
Cells per stack N_cell
Number of cells stacked in series within one stack
Number of stacks N_stack
Stacks connected in parallel
Total electrolyte volume V_elec
L
Sum of positive and negative tanks
Vanadium concentration c
mol/L
V concentration in sulfuric acid (1.6-1.8 mol/L is standard)
Operating current density j
mA/cm²
Higher j increases output but drops efficiency
Usable SOC range ΔSOC
%
Usable state-of-charge window (typically 80%, e.g. SOC 10-90%)
Results
—
Total cells
—
Operating current (A)
—
Stack voltage (V)
—
Output power (kW)
—
Stored energy (kWh)
—
Discharge duration (h)
—
VRFB system schematic — stack, electrolyte tanks & circulation pumps
Positive (yellow) and negative (purple) electrolytes are pumped through the central stack. Electrons (red → black) flow through the external circuit to deliver output. Hue shifts indicate SOC level.
Stored energy E and output power P. c = vanadium concentration (mol/L), V_electrolyte^half = half-cell electrolyte volume (L), F = 96485 C/mol, ΔSOC = usable SOC window, V_cell = mean cell voltage, N_cell = cells per stack.
Round-trip efficiency. ηC = coulombic (≈0.95), ηV = voltage = V_op/V_OCV, ηshunt = shunt-current loss (≈0.98), ηpump = pump/auxiliary loss (≈0.97).
Vanadium Redox Flow Battery (VRFB) — Power and Energy Sizing
🙋
"Vanadium redox flow battery" — that doesn't sound like a name I've heard before. Is it different from a lithium-ion battery? It has tanks and looks kind of like a chemical plant.
🎓
Yes, it's a completely different animal — usually called VRFB for short. While a lithium-ion battery packs positive material, negative material and electrolyte into one box, a VRFB separates the battery stack from the electrolyte tanks. Pumps push the electrolyte through the stack to react, then return it to the tanks. That's why it looks like a chemical plant. As a result it has become the go-to for grid-scale storage in the 100 kW to 100 MW range.
🙋
If the tanks are separate, what's the advantage? When I increase "electrolyte volume" on the left, stored capacity climbs quickly, but the output power (kW) stays the same.
🎓
Great catch — that is exactly the VRFB's killer feature. Power (kW) is set by the stack size, and capacity (kWh) is set by the tank size. You design the two independently. "Store the daytime solar surplus until night" needs 10 hours of discharge, so a 1 MW stack with a 10 MWh tank. "Inject big power for frequency regulation" can be 1 MW × 30 min = 0.5 MWh — a small tank. Li-ion has to add cells for either change, so it can't deliver this flexibility.
🙋
So you really can decouple power and energy. By the way, when I raise "current density j", the output goes up but the stack voltage drops slightly. What's happening?
🎓
That's the voltage-efficiency story. Open-circuit voltage (OCV) is fixed at about 1.25 V per cell, but passing current introduces ohmic loss and overpotential, dropping the working voltage: V_op = V_OCV − j·R_Ω. Higher j lowers V_op, so voltage efficiency ηV = V_op/V_OCV falls. Push j too low, however, and pump auxiliary power becomes relatively large. So VRFBs are typically operated around 60-100 mA/cm² for best efficiency. Move the slider: at 20 mA/cm² the voltage stays near OCV, while at 300 mA/cm² you can see a sizeable drop.
🙋
VRFB sounds great for big stationary storage. But why can't we use it like a Li-ion in a phone or EV? Is there something technically inferior?
🎓
The fatal weakness is low energy density. VRFBs deliver about 20-40 Wh/L, while Li-ion delivers 200-500 Wh/L — an order of magnitude lower. It comes down to the vanadium sulfate solubility limit of about 1.6-2.0 mol/L, which is fundamentally hard to break. So mobility is out and VRFB is restricted to stationary use. In exchange, you get (1) a 20+ year lifetime with over 20,000 cycles, (2) non-flammable aqueous chemistry, (3) 100% depth of discharge, and (4) easy SOC monitoring through electrolyte colour change. In Japan Sumitomo Electric leads the world, with a 15 MW/60 MWh system running at Hokkaido Electric's Minami-Hayakita substation.
🙋
That's fascinating. Where do you start when you have to size one? This tool has lots of parameters and I don't know what to fix first.
🎓
Start from the spec: required power P and discharge duration t, say "1 MW × 4 h = 4 MWh". Then (1) stack design: P = I·V_stack = j·A_cell·V_op·N_cell·N_stack, typically with j = 80 mA/cm², V_op = 1.21 V, A_cell = 1000 cm², so you pick the product N_cell·N_stack. Next (2) tank design: E = c·V_half·F·ΔSOC·V_stack / 3600 gives V_half. With the defaults here you get 15.5 kW output, 1715 kWh capacity, ~111 h discharge — a very capacity-heavy "large-scale daily-cycle" sizing. The final check is whether energy ÷ power matches the required discharge time.
Frequently Asked Questions
The biggest difference is that a VRFB lets you design power (kW) and energy capacity (kWh) independently. Power is set by the stack (the cell assembly), and capacity is set by the size of the electrolyte tanks. In a lithium-ion battery the cell itself holds both capacity and output, so increasing the discharge duration requires more cells. Energy density is ~20-40 Wh/L for VRFB versus 200-500 Wh/L for Li-ion, an order of magnitude lower, so VRFB is limited to stationary, grid-scale storage.
Using half-cell electrolyte volume V_half [L], vanadium concentration c [mol/L], the Faraday constant F = 96485 C/mol and usable SOC range ΔSOC (typically 0.8), the available charge is Q = c·V_half·F·ΔSOC [C] and capacity [Ah] is Q/3600. The stored energy is E = Q·V_stack_nominal [Wh], where V_stack_nominal is N_cell × 1.25 V (the standard VRFB OCV). This tool assumes equal split between positive and negative electrolyte and computes from half volume.
Round-trip efficiency is the product of (1) coulombic efficiency ηC ≈ 95% (self-discharge and vanadium crossover through the membrane), (2) voltage efficiency ηV = V_op/V_oc (ohmic loss and overpotential), and (3) balance-of-plant losses ηBOP ≈ 95% (pump power, shunt currents). Higher current density drops ηV; very low j makes BOP dominant. Typical operating range is 60-100 mA/cm². This tool assumes ηC=0.95, ηshunt=0.98, ηpump=0.97 and computes ηV from the cell-voltage ratio.
VRFBs are used mainly in 100 kW to 100 MW grid-scale storage. Notable examples: Hokkaido Electric's Minami-Hayakita substation (Sumitomo Electric, 15 MW/60 MWh, renewable smoothing), the 100 MW/400 MWh Dalian project in China, and the Pellworm Island hybrid (wind+PV+VRFB) in Germany. Lifetime over 20 years (>20,000 cycles), full depth of discharge, and non-flammable aqueous chemistry suit daily cycling (storing solar surplus for nighttime release) and frequency regulation / capacity markets.
Real-World Applications
Grid-scale storage for renewable integration: Absorbing the variability of solar and wind and discharging at night or during calm periods is the headline use case for VRFBs. Examples include the 15 MW/60 MWh system at Hokkaido Electric's Minami-Hayakita substation (Sumitomo Electric), the 100 MW/400 MWh project in Dalian, China, and a 2 MW/8 MWh installation in California. With cycle life an order of magnitude longer than Li-ion, the economics work for daily 1-2 cycle operation over a 20+ year horizon.
Microgrids and island power: VRFBs are well suited to microgrids that combine diesel and renewables, such as Pellworm Island in Germany (wind + PV + VRFB) and King Island in Australia. The aqueous electrolyte is non-flammable with no ignition or explosion risk, easing siting standards. As fire-risk concerns tighten, VRFB adoption is increasing for island and residential-adjacent sites where Li-ion is harder to permit.
Frequency regulation and ancillary services: VRFBs also serve frequency-regulation markets that demand second-to-minute scale charge/discharge cycling. Here the priority is cycle count rather than response speed, and the VRFB's tens-of-millisecond response combined with essentially unlimited cycle life is a great match. Pilot operations on Germany's and PJM's markets have run, and demonstrations are ongoing in TEPCO's service area in Japan. Sizing uses a high-P/E ratio (small energy relative to power).
UPS and data-center backup: Traditionally dominated by lead-acid and Li-ion, data centers are starting to adopt VRFB for fire-risk reduction and long life. The decoupled power and energy design lets a single installation cover both "short backup + ride-through" and "long autonomous run", offering useful flexibility. A 20-year lifetime aligns with facility refresh cycles, giving competitive life-cycle cost.
Common Misconceptions and Pitfalls
The first misconception is that you can grow capacity endlessly by growing the electrolyte tanks. Capacity E does scale with V_electrolyte, so doubling tanks doubles capacity. However, vanadium concentration is bounded by solubility in sulfuric acid (about 1.7-2.0 mol/L, temperature-dependent). Pushing it higher causes V(V) (VO₂⁺) to precipitate at low temperature, clogging plumbing and degrading capacity. Real systems run at 10-40 °C controlled, with typical concentration around 1.6 mol/L. Before enlarging the tank, optimise concentration and operating temperature.
Next, "raising current density increases output proportionally" is wrong. Going from 80 to 200 mA/cm² gives 2.5× current, but cell voltage drops to V_op = 1.25 − 0.2·0.5 = 1.15 V, so output only rises to 2.5 × (1.15/1.21) = 2.38×. Voltage efficiency drops from 92% to 88%, and waste heat (I²R) grows. The rule of thumb is to reserve high-current operation for short peak events and run steady-state at the economic optimum (typically 60-80 mA/cm²). Try setting j to 300 mA/cm² in this tool to see how voltage efficiency collapses.
Finally, "VRFB lasts longer so its high capex always pays back" is too simplistic. Stack and electrolyte can run 20+ years, but BOP (Balance of Plant) — pumps, piping, heat exchangers, controls — typically needs renewal every 5-10 years. Slow capacity fade from V³⁺/VO₂⁺ crossover also forces periodic rebalancing (catholyte/anolyte mixing or electrochemical regeneration) every 3-5 years. When you do LCOE (levelised cost of energy) comparisons, include both maintenance and BOP renewal costs. The true win against Li-ion shows in long-term cycling well beyond 5,000 cycles.
How to Use
Set the electrode area in cm² (typical range 100–500 cm² for commercial stacks) using the areaNum input or areaRange slider.
Specify the number of cells per stack (20–60 cells standard) and total number of stacks in your battery system.
Input vanadium electrolyte volume in liters (500–5000 L for grid applications); the simulator calculates operating current, stack voltage, output power (kW), total stored energy (kWh), and discharge duration (hours) based on V⁵⁺/V⁴⁺ and V³⁺/V²⁺ redox reactions at nominal cell potential of 1.25 V per cell.
Worked Example
A 2 MW VRFB installation uses 4 stacks with 40 cells each, electrode area 300 cm², and 2000 L vanadium electrolyte (50 mol/m³ concentration). Operating current = 2500 A; stack voltage = 50 V (40 cells × 1.25 V nominal); output power = 125 kW per stack (500 kW total at 200 A discharge); stored energy ≈ 250 kWh; discharge duration ≈ 30 minutes at rated output, or 4 hours at 50 kW continuous.
Practical Notes
Increase electrode area and electrolyte volume together to boost both power density (A/cm²) and energy capacity without exceeding stack thermal limits (typically 60°C max in vanadium systems).
Stacking multiple VRFB units in series increases voltage; parallel stacking increases current capacity—choose based on grid interconnection requirements (480 V AC or 600 V DC common in North America).
Vanadium crossover through membranes increases with current density; limit to <500 mA/cm² for Nafion membranes to preserve round-trip efficiency above 75%.