Volkswagen’s PowerCo and Umicore obtain full regulatory approval for EU battery materials production
Commercially operated Wärtsilä engine runs on 25 vol% hydrogen blend

Amprius verifies 500 Wh/kg, 1300 Wh/L battery platform

Amprius Technologies, developer of the Silicon Anode Platform for batteries, has verified a lithium-ion cell delivering energy density of 500 Wh/kg, 1300 Wh/L, resulting in unparalleled run time. At approximately half the weight and volume of state-of-the-art, commercially available lithium-ion cells, the new battery cell delivers potential industry-disrupting performance with barrier-breaking discharge times.

The record 500 Wh/kg energy density performance was verified by Mobile Power Solutions, a leading testing house offering comprehensive battery regulatory compliance, safety, and performance testing. The results indicate that this cell model provides >504 Wh/kg and >1321 Wh/l at 25°C.


Amprius says that its next-generation cells are positioned to power products in the fast-growing aviation and, eventually, electric vehicles markets, estimated to be collectively more than $100 billion in battery demand by 2025.

The 500 Wh/kg battery platform significantly expands boundaries for customers and is a tailored solution for applications that require maximum discharge times without compromising key features such as aircraft payload and without having to increase vehicle weight.

The new batteries demonstrate both high gravimetric energy density (Wh/kg) and volumetric energy density (Wh/L) with exceptional adaptability. The customizable platform allows customers to select the option to either increase energy content in a battery pack without increasing weight, reduce weight in applications that target a fixed energy content, or combinations of both.

Higher energy is important for longer run times, range and endurance, while lighter packs increase energy efficiency—even for the same battery energy content.

We look forward to taking advantage of Amprius’ 500 Wh/kg cell to further develop Zephyr’s unrivalled stratospheric endurance capabilities. Amprius is a valued current supplier with a great track-record, and we are confident that Amprius’ battery will deliver the capability we need.

—Pierre-Antoine Aubourg, Chief Technical Officer at AALTO HAPS

AALTO HAPS is the Airbus subsidiary developing a 100% solar-electric High Altitude Platform Station for connectivity and earth observation applications.



Hmm, a post of mine seems to have disappeared, or maybe I pressed preview instead of post.

Anyway, I always look for the bits press releases are not talking about, in this case cycle life.

Here are NASA tests:

Something like 200 cycles.
Maybe enough for drones, but we are a long, long way from this lab top cell making its way into EVs.


I don't know what information source you contacted relating to cycle life of stated cell.
The cycle quoted at the above web address states 200 - 1,200 cycles.


I suspect that the 200 cycles you mention are worst case and 1,200 cycles perhaps under favorable ambient conditions.

Keep in mind that for a given application, double energy density delivers the equivalent of twice the cycles as its half-capacity competition. But at half the weight.

Even if this battery delivered only half the cycles of the Tesla / Panasonic battery, (let’s round that to 250Wh/kg to simplify the comparison), the total mileage would be the same 300-400,000 miles.

But the single charge range of a Model S could be 800 miles.

I’m not arguing for an 800 mile car. But 500Wh/kg smashes the idea that BEVs can not compete and win against gas or liquid fueled vehicles for consumer viability.


The NASA tests I cited give the specific test conditions, see page 8 for C/10 and C/2 discharge to 80%

At the very slow rate of C/10, a 70KWH battery pack would only be turning out 7KW.
At 3.5Kwh per mile, you had better keep your speed down to 2 mph or so, and even then the pack is only good for 350 cycles or so.

At a more realistic C/2, then the pack is good for under 300 cycles.

For the entirely unspecified output on the Amprius site they give 'up to 1200' cycles.
Sure, if you keep the draw low enough, and God knows what tiny rate of discharge they need for that, you get a lot of cycles.

You are not running a car, or even flying a drone on that though


eci said:

' Keep in mind that for a given application, double energy density delivers the equivalent of twice the cycles as its half-capacity competition. But at half the weight. '

I think that you have double dipped.
One or the other, but not both.

And the batteries are not free, if you want more KWh, you are pretty much going to double the cost, although to be sure you can realistically lug it around, which you couldn't with lower energy density batteries.


I think folk imagine that I am negative about Ampius.

I am not, but the results need taking for what they are, not imagined as though they are ready to stuff in a car to get double the mileage.

This is lab bench stuff, which hopefully can now start making its way into premium applications where weight is very important, and cost isn't, so the low cycle life at any realistic draw can be covered.

Presumably drones are high on the list.

None of that means that the tech can't be improved in cycle life etc, but we have not even got that working in the very controlled conditions of early samples in a lab.

Years and years away from something ready for commercial production for a car.


This is lab bench stuff
Unless you think the lab is at 70,000 ft.
“ Zephyr-S has now flown for 2,435 flight hours and demonstrated precise stratospheric manoeuvrability and station-keeping over points on the ground.”
Also :
These 2021 records were done using 435 Wh/kg batteries.
They also have plans for a 775,000 square foot facility in Brighton, Colorado. The factory, targeted to be operational in 2025, the initial phase will be 500 megawatt-hours (MWh) with the potential of up to 5 gigawatt-hours (GWh) within the initial footprint.



Don't get me wrong, I think this has got a lot of potential.
I am just trying to get a handle on where we are at the moment.

Have I got the figures right on cycle life, or have I dropped a decimal point somewhere?

Assuming I have it about right, as I said, initially this will be for premium applications where weight is more important than any costs arising from relatively short cycle life, and things like the Zephyr fit that exactly.


If I have got the figures right on the relationship between charge rates and cycle life for this type of battery at the moment, I sure hope that the batteries are used where they can be swapped rather than any attempt to fast charge.

I would hate to think what the cycle life would be at 10C.


Your NASA reference is 2016.
This is a slightly newer 2020 NASA reference:
This is an 12/2021 Extreme Fast Charge reference:

This battery tech could also be used on a Tesla Semi which has a 500 mile range, which would give a 600k mile life. This battery would definitely help on the weight of the 900 kWh battery (less than 2000 kg).


Thanks, Gryf.

Since I am just a tad short of infallible, I was hoping that someone would check to see if I have gone totally off the rails on this occasion! ;-)

However, looking at your references, I can't spot any substantial differences compared to the figures I based my comments on.

For the first reference, from Ampius, on page 8, the graph for degradation looks pretty well identical to the NASA figures I quoted, with something of the order of 350 cycles to 80%, given that they are using there a C/5 discharge rate.

In the second reference, sure, the battery can be fast charged, but there is no mention there of what cycle life they get if you do stuff in energy that fast, indeed, as your reference states:

' According to the press release, the energy density of the cells is amazing, at 370 Wh/kg, however, there is no word about the longevity and cycle life.'

I'm afraid that over the last 10 years or so I have seen all sorts of dodgy elisons, so that if you read the obvious into statements from the battery company, if they have not explicitly stated it, they later wriggle out of it!

Even Elon Musk was astounded by the misdirections and fakes in the battery industry, and he knows a thing or several about both.

I don't know if that is the case for Ampion, but until I see figures explicitly stating the cycle life under fast charge assessed by reputable third parties, I will assume that it hammers it.

I can't see anything to indicate authoritively that the cycle life and charging rates have increased at all since the NASA assessment I quoted.


Chemistries featuring a very good energy density and poor lifespan can still be useful on EVs, used in dual chemistry configuration (like the Gemini battery).

However, this type of cell would also need to be cheap and have some tolerance to fast charging.
We don't know about the production tech of this anode. Anything "nano" is always suspect of being the result of some slow, expensive process.


Amprius is using Plasma-enhanced chemical vapor deposition (PECVD), used in solar cell production. They have acquired Centrotherm PECVD for silicon-nanowire anode production.
They also have a DOE grant to study “High Throughput Source-less Plasma Deposition of Structured Silicon Anodes “,

Earlier (2016) Amprius used CVD equipment from the Dutch firm Meyer Burger,


Half as many cells could bring cost reductions


Amprius is using a “Nano” process which may produce the highest energy densities, however is a more expensive process compared to other silicon anodes.
So they will probably focus on Aerospace/Defense rather than EV.
Silicon Anodes are being developed by many other companies. UCSD and LG ES have shown that Silicon Anodes work well with Sulfide Solid State Electrolytes.
LG ES is working with Silicon “Micro” particles which are less expensive.
Maybe a 4680 Cylindrical cell using a Silicon Micro Particle Anode, a Hybrid Sulfide–Polymer Solid State Electrolyte, and NCM811 cathode would work on that Tesla Semi.
“Non-Conductive Polymers Enable Higher Ionic Conductivities and Suppress Reactivity in Hybrid Sulfide-Polymer Solid State Electrolytes”


Thanks Gryf.

Premium applications where short cycle life under high power draw may be an acceptable trade off.

I can't see any evidence from the information that they have provided that they currently have any way of combining charge and discharge rates with decent cycle life, although to be sure they do talk about high C rates without specifying what that does to cycle life.

Unfortunately over the last decade or so I have come across loads of instances of that sort of 'omission'

That is not to say that it can't be done by some future advance in the technology, but there is nothing at all to indicate that they have managed high charge rate combined with long cycle life for anything they have at the moment, even as a cell prototype.

The comments to this entry are closed.