All posts

2026

Battery Chemistry Mini-Series: A Field Guide to What's Powering (and What Might Power) the Future

A tour through seven battery (and battery-adjacent) chemistries — the pros, the cons, and my personal excitement, readiness, and cost/supply-chain scores for each.

Part 1

LFP

LiFePO₄ — Lithium Iron Phosphate

Pros

  • No cobalt or nickel means real supply chain and cost benefits.
  • The lower voltage also means it won't decompose the organic electrolyte as quickly, giving it a longer cycle life.
  • The lower voltage makes it safer too; fast charging carries less risk of the voltage climbing so high that O₂ starts to evolve out of the crystal structure.

Cons

  • As a cathode (positive electrode), LFP has a lower voltage (~3.2V) than its nickel/cobalt-containing counterparts (~3.7V), so it provides lower energy, which isn't great.

A change of mind

A younger version of me thought the low energy would eliminate LFP from EV use and make it only good for grid-scale storage. The Chinese EVs — and now some Teslas — using LFP have changed my mind. Here's why: the higher safety means less stringent thermal management (and thus less weight) is needed at the pack level. And if you can safely fast-charge LFP for your EV, you can eliminate range anxiety without a massive battery pack — just an extensive fast-charging network instead.

Excitement

6/10

World's readiness

10/10 (it's already being used)

Cost & supply chain

8.5/10 (still requires lithium; not as abundant as other resources)

Part 2

NMC

Nickel Manganese Cobalt — LiNiₓMnᵧCo_zO₂, typically written as NMCxyz, like NMC811

Pros

  • NMC has a much higher voltage than LFP (3.7V vs 3.2V), so it stores more energy in less weight.
  • As EVs became super popular, so did the demand for more batteries and the materials inside them.

Cons

  • Cobalt is expensive and mostly mined in the Congo using "artisanal" miners — neither of which is great. So the name of the game has been decreasing cobalt content while maintaining high energy and cycle life, which means using more nickel instead.
  • Manganese is cheap, but too much of it can dissolve into the liquid electrolyte, which lowers stored charge.
  • Nickel is great, but at high concentrations it can block lithium-ion diffusion pathways in the crystal structure, lowering charge/discharge rates. AKA safety concerns.

Exciting players in the market

CATL, BYD, multiple Asian battery companies

Excitement

6.5/10 (performance OK but safety risks and not good enough for eVTOLs)

World's readiness

10/10 (already being used)

Cost & supply chain

5/10

Part 3

Sodium-Ion

Pros

  • Lower costs, since you skip the lithium, cobalt, and nickel supply chains entirely.

Cons

  • Lower performance. Fundamentally, sodium is heavier than lithium, with a higher voltage.
  • You also have to switch the graphite anode used in lithium-ion batteries for a glassy carbon anode instead.

Outlook

Overall, energy density compared to lithium-ion isn't great — but it could improve if using a sodium metal anode. That'll get us into a discussion of solid-state batteries in a future post.

Exciting players in the market

CATL, as they scale up mass production for EV batteries.

Excitement

3/10 (lower performance)

World's readiness

9/10 (can mostly reuse lithium ion manufacturing lines)

Cost & supply chain

10/10 (no lithium, no cobalt)

Part 4

Solid State

Pros

  • The promise of nonflammability, high safety, and high energy density by enabling the use of a lithium metal anode — the highest charge density element available.

Cons

  • It's a pain in the butt to make, and there's been more hype than substance over solid state batteries. The main reason for this is that the solid electrolytes tend to be much thicker than the anode or cathode layers, so you end up sacrificing most of the energy density benefits of the lithium metal anode due to the thick, heavy solid electrolyte.
  • Manufacturing is also a pain in the butt because you've got all of these solid-solid particle contact interfaces. And if you're sintering two different solids together that have different coefficients of thermal expansion, having them maintain intimate contact at all temperatures is incredibly difficult, if not impossible.

The pragmatic path

I think the real approach here is probably something like what QuantumScape is doing: using a thin solid electrolyte to enable an anode-less lithium metal anode, but then using a tiny amount of liquid electrolyte to solve the interface contact issues between the solid electrolyte and the cathode. Sure, it might not be 100% nonflammable, but it should be much less of a hazard than a standard NMC lithium-ion cell, and could potentially allow for much higher energy densities — assuming the solid electrolyte fraction is kept low.

Rivalry

This chemistry's main rival is silicon anodes.

Exciting players in the market

QuantumScape, ION Storage Systems, and Solid Power (although I'm curious how sulfide solid electrolytes would scale due to reactions with water vapor).

Excitement

8/10 (QS showed some really high (>100 mA/cm²) current densities)

World's readiness

5/10 (QS has invested a lot of treasure into making their pilot lines. Science is tough!)

Cost & supply chain

6.5/10 (still uses Li/Ni/Co)

Part 5

Silicon Anode

Pros

  • Much higher charge density on the anode side, so overall you're storing more energy in a smaller form factor and lighter weight.
  • These are already being sold on the market today, unlike solid-state batteries. So currently, they're the leading new high-energy battery chemistry on the market.

Cons

  • The volume expansion/contraction of the silicon anode reduces cycle life a whole lot. Clever scientists and engineers have mostly addressed this by nano-sizing the silicon anode into nanorods, which helps accommodate the volume expansion/contraction better.
  • Even if you've solved the cycle life issue, calendar life is still a problem, because the silicon anode will have chemical reactions with the liquid electrolyte that ultimately shorten battery lifetime. But for shorter-term applications — think one-way drones or consumer applications that aren't EV or grid-scale storage — this could help a whole lot.

Exciting players in the market

Amprius, since they're already selling these batteries for drones — which I think is a great early market for aspiring battery companies.

Excitement

8/10 (higher energies ~330–380 Wh/kg; selling on the market, but pretty expensive)

World's readiness

8/10 (pretty expensive and calendar life questions remain)

Cost & supply chain

3/10 (still use Li/Ni/Co and are expensive. Costs should come down over time)

Part 6

Hydrogen Fuel Cells

Technically not a battery — it's a fuel cell — but it's energy storage.

Pros

  • Very high power, fast charging.
  • The byproduct is water, and it uses oxygen from the air. Sounds amazing.

Cons

  • It's incredibly energy inefficient to be producing hydrogen gas from water electrolysis only to use it for powering a hydrogen fuel cell. After accounting for all the thermodynamic and practical inefficiencies from doing water electrolysis into fuel cells, you might have a round-trip efficiency anywhere between 25% to, at best, 50%. So you'd only use it for applications where energy costs aren't very important — maybe electric aircraft, maybe shipping?
  • One of the big glaring problems for powering cars is the lack of available refueling infrastructure. It's very expensive to pump hydrogen gas, or any gas, over long distances in a pipe.
  • Hydrogen gas likes to diffuse out of the tank it's in, and a car would need to carry very heavy tanks with pressurized hydrogen gas at 700 atmospheres. So you're essentially driving around a bomb.

The economics

I worked on a DOE grant where the DOE assumed $4–5 total/gallon equivalent for producing and transporting hydrogen gas. So essentially, you're asking consumers to pay for a more expensive car with no refueling stations (outside of the 40–50 in California), for gas that costs $5 per gallon equivalent. That's a very hard sell over battery-powered electric vehicles that cost 5X less to drive than gas-powered cars, with readily available charging infrastructure.

Excitement

2/10 (cool promise/high power, but the remaining practical challenges are tough)

World's readiness

1/10

Cost & supply chain

2/10 (H₂ = expensive, O₂ = cheap)

Part 7

Metal-Air Batteries

Lithium-air, aluminum-air, zinc-air, sodium-air, magnesium-air, and more.

Pros

  • You'd unlock ultra-high energy densities — think north of 700 to 1,000 Wh/kg.
  • Imagine flying cars like in the Jetsons, Teslas with a thousand-mile range easily. Imagine electric cars costing less than a Honda Civic.

Cons

  • It depends on the chemistry, but if you're using a non-aqueous liquid electrolyte, your air electrode will almost always clog up with insoluble discharge products — typically some kind of alkali peroxide — suffocating your oxygen electrode. So in practice, even though these chemistries have very high theoretical energy density, their practical energy density is very low, except at very low currents/power densities, to the point where they're not practically useful.
  • The way around this is to use a water-based liquid electrolyte, which solves many of the insoluble discharge product problems. But that also requires a solid electrolyte, which introduces some of the same challenges from solid-state batteries mentioned earlier.

Note

I'm not going to get into the nitty-gritty details of DayLyte's battery, but metal-air batteries have a whole lot of potential. Overall, I'm very excited about this space and what DayLyte has been working on.

Excitement

12/10

World's readiness

4/10

Cost & supply chain

10/10