🔬 applied physics

From cathode to anode: Understanding lithium loss in 21700-type Ni-rich NCM||Graphite-SiOx cells

This study utilizes a combination of in operando neutron diffraction, neutron depth profiling, and X-ray computed tomography to demonstrate that capacity fade in large-format 21700-type Ni-rich NCM||Graphite-SiOx cells is primarily driven by microscopic lithium inventory loss and active anode material degradation, rather than macroscopic structural inhomogeneities.

Original authors: Thien An Pham, Hannah Bosch, Giovanni Ceccio, Lukas Keller, Hannes Wolf, Nicolas Bucher, Peter Müller-Buschbaum, Ralph Gilles

Published 2026-02-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Thien An Pham, Hannah Bosch, Giovanni Ceccio, Lukas Keller, Hannes Wolf, Nicolas Bucher, Peter Müller-Buschbaum, Ralph Gilles

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: The "Growing Pains" of Big Batteries

Imagine you are building a house. A small shed (like an old 18650 battery cell) is easy to manage; the temperature is even, and everything works smoothly. But now, imagine building a massive skyscraper (a new, larger 21700 battery cell). While the skyscraper holds more energy, it's harder to keep the temperature consistent from the top floor to the basement. Some parts get hotter, some get colder, and the stress isn't shared equally.

This study is about understanding why these "skyscraper" batteries wear out faster and where the energy goes when it disappears. The researchers looked inside a specific type of high-tech battery used in electric vehicles to see what happens after it has been charged and discharged many times.

The Cast of Characters

To understand the battery, think of it as a busy train station:

The Cathode (The Departure Station): Made of a special metal mix (NCM). It holds the "passengers" (Lithium ions) when the battery is empty.
The Anode (The Arrival Station): Made of Graphite mixed with a bit of Silicon Oxide (SiOx). This is where the passengers go when the battery is full.
The Lithium Ions (The Passengers): Tiny charged particles that zip back and forth between the stations to create electricity.
The SEI (The Security Guard): A thin layer that forms on the arrival station. It's necessary for safety, but if it gets too thick, it blocks the doors.

The Mystery: Where Did the Passengers Go?

When a battery ages, it loses its ability to hold a charge. The researchers wanted to know: Did the passengers get lost, or did the stations get too small to hold them?

They used two special "super-eyes" to look inside the battery while it was working:

Neutron Diffraction: Like an X-ray that can see through the metal casing to watch the atoms move and change shape.
Neutron Depth Profiling: Like a deep-sea scanner that measures exactly how many passengers are hiding at the surface versus deep inside the stations.

What They Found: The Two Main Culprits

The study revealed that the battery loses power for two main reasons, which they call LLI and LAAM.

1. LLI: The "Lost Passengers" (Loss of Lithium Inventory)

Imagine the Lithium ions are passengers. Over time, some passengers get stuck at the Arrival Station (the Anode). They get trapped in the "Security Guard" layer (the SEI) that keeps growing thicker and thicker.

The Analogy: It's like a bouncer at a club who keeps building a bigger and bigger wall. Eventually, the wall is so thick that new passengers can't get in, and the ones already inside can't get out.
The Finding: The researchers found that the Cathode (Departure Station) was running low on passengers, but the Anode (Arrival Station) was actually full of trapped passengers. The Lithium wasn't disappearing; it was just stuck on the wrong side of the station.

2. LAAM: The "Shrinking Stations" (Loss of Active Anode Material)

The Anode is made of Graphite and Silicon. Silicon is great because it can hold a lot of passengers, but it has a problem: it swells up like a sponge when wet.

The Analogy: Imagine the arrival platform is made of soft clay. Every time a passenger steps on it, the clay expands. After thousands of trips, the clay cracks and crumbles. Now, parts of the platform are broken and unusable.
The Finding: The study showed that the "platform" (the active material) was physically breaking apart. Because parts of the station were destroyed, fewer passengers could board, even if there were plenty of Lithium ions available.

The "Skyscraper" Problem: Uneven Aging

One of the most interesting discoveries was that the battery didn't age evenly.

The Center vs. The Edge: The researchers took samples from the center of the battery and the bottom edge.
The Result: The center was much more damaged than the edges. It was like the middle of the skyscraper was overheating and cracking, while the outer walls were still fine.
Why? This is likely due to how the liquid electrolyte (the "fuel" that helps the passengers move) moves around. In a tall, round battery, the liquid doesn't mix perfectly, creating "traffic jams" in the middle that cause more wear and tear.

The Silicon Factor

The battery used a mix of Graphite and Silicon. Silicon is the "star player" with high energy, but it's also the "troublemaker."

The study found that the Silicon part of the anode was degrading faster than the Graphite. As the Silicon broke down, it contributed heavily to the "shrinking station" problem (LAAM).

The Takeaway: What Does This Mean for Us?

This research is like a mechanic diagnosing a car engine that is making a weird noise. They found out:

The passengers are getting stuck in a thick security wall on the arrival side.
The arrival platform is physically breaking because of the swelling Silicon.
The middle of the battery is suffering the most because of uneven conditions.

Why does this matter?
By understanding exactly where and how the battery fails, engineers can design better batteries. They might:

Create better "security guards" (SEI) that don't get too thick.
Reinforce the "platforms" so the Silicon doesn't crack as easily.
Design the battery shape or cooling systems to ensure the middle doesn't get hotter than the edges.

In short, this paper helps us build electric vehicles that last longer, charge faster, and don't lose their power as quickly as they age.

1. Problem Statement

As electric vehicle (EV) batteries transition to larger cylindrical formats (e.g., 21700 and 4680 cells) to increase energy density, new challenges arise regarding spatial inhomogeneities. Scaling up cell size introduces thermal gradients and mechanical stresses that can lead to uneven degradation. Specifically, there is a lack of detailed understanding regarding:

How Ni-rich NCM cathodes and Graphite/SiOx composite anodes degrade together during cyclic aging in large-format cells.
The specific mechanisms of Lithium Inventory Loss (LLI) and Loss of Active Anode Material (LAAM).
The spatial distribution of degradation (e.g., differences between the cell center and edges) and how lithium is redistributed between electrodes.

2. Methodology

The study employed a multi-modal approach combining electrochemical testing with advanced neutron-based characterization techniques on prototype 21700-type cells.

Cell Chemistry:
- Cathode: LiNi $_{0.83}$ Co $_{0.12}$ Mn $_{0.05}$ O $_2$ (NCM-831205).
- Anode: Graphite blend with ~14 wt.% Silicon Oxide (SiO $_x$ ).
- Electrolyte: 1.15 M LiPF $_6$ in DMC:FEC:EMC (70:20:10 vol.%).
Aging Protocol:
- Cells were cycled at 25°C between 2.5 V and 4.2 V.
- Three states of health (SOH) were analyzed: SOH100 (uncycled reference), SOH80, and SOH70.
- Aging involved a mix of C/2 cycling and periodic "Check-Up" cycles at lower C-rates (C/3, C/10) for electrochemical analysis.
Characterization Techniques:
1. Electrochemical Analysis: Differential Voltage Analysis (DVA) and Incremental Capacity Analysis (ICA) were used to identify degradation modes (LLI vs. LAAM) based on peak shifts in voltage-capacity curves.
2. In Operando Neutron Diffraction (ND): Performed at the Paul Scherrer Institute (SINQ). This allowed for real-time tracking of:
  - NCM Cathode: Lattice parameter changes ( $a$ and $c$ ) to monitor delithiation depth and unit cell volume changes.
  - Graphite Anode: Phase transitions from C $\to$ LiC $_{12}$ (Stage 2) $\to$ LiC $_6$ (Stage 1) to quantify active material availability and lithium storage capacity.
3. Neutron Depth Profiling (NDP): Performed on extracted electrodes at the Nuclear Physics Institute (Czech Republic). This non-destructive technique measured lithium concentration profiles as a function of depth (from surface to bulk) at different positions (Center vs. Edge) to validate lithium redistribution and SEI growth.

3. Key Contributions

Correlation of Structural and Electrochemical Data: The study successfully links macroscopic electrochemical capacity fade with microscopic structural changes (lattice parameters and phase transitions) in real-time.
Spatial Inhomogeneity Mapping: By sampling electrodes from the center and bottom edge of the 21700 cells, the research quantifies the extent of degradation gradients within a single large-format cell.
Mechanism Elucidation: It distinguishes between LLI (consumption of lithium) and LAAM (loss of active graphite/SiO $_x$ ) and determines their relative contributions to capacity fade in SiO $_x$ -containing systems.
Validation of Li Redistribution: It provides direct experimental evidence that lithium lost from the cathode is not merely "lost" from the system but is trapped in the anode (specifically in the SEI and bulk), confirming the "Li shuttle" mechanism of degradation.

4. Key Results

A. Electrochemical Findings

Capacity Fade: Both SOH80 and SOH70 cells showed consistent capacity loss.
DVA/ICA Analysis:
- LAAM: The DVA peak associated with SiO $_x$ lithiation diminished and shifted, indicating a loss of active SiO $_x$ material. The transition from LiC $_{12}$ to LiC $_6$ (MaxHi peak) shifted to lower capacities, confirming a reduction in active graphite.
- LLI: A shift in DVA curves indicated a loss of cyclable lithium inventory.

B. In Operando Neutron Diffraction (Structural Changes)

Cathode (NCM):
- Reduced Delithiation: In aged cells, the NCM unit cell expansion (specifically the $c$ -parameter) during charging was reduced, indicating less lithium could be extracted.
- Increased Over-delithiation: In the charged state, aged cells showed higher delithiation levels (lower $c$ -parameter), suggesting the cathode potential shifts higher to compensate for anode limitations.
- Volume Stress: The unit cell volume change became more pronounced with aging, increasing the risk of particle cracking due to anisotropic stress.
Anode (Graphite):
- Phase Transition Shifts: The transition to the LiC $_{12}$ phase occurred at lower capacities in aged cells, confirming LAAM.
- LiC $_6$ Formation: The onset of the fully lithiated LiC $_6$ phase shifted to lower capacities, and the final intensity of the LiC $_6$ peak decreased, indicating that less lithium is available to fully lithiate the graphite (LLI).
- Inhomogeneity: A "shoulder" in the diffraction peaks for the SOH70 cell suggested inhomogeneous lithiation within the electrode.

C. Neutron Depth Profiling (Li Distribution)

Cathode: The total lithium concentration in the NCM cathode decreased significantly (approx. 38% at the center) from SOH100 to SOH70.
Anode: The total lithium concentration in the anode increased significantly in aged cells.
- SEI Growth: A sharp surface peak in the Li profile (attributed to the Solid Electrolyte Interphase) was broader and higher in aged cells, confirming continuous SEI formation consumes lithium.
- Bulk Trapping: The bulk anode also showed higher Li retention, suggesting trapped lithium.
Spatial Gradients:
- Degradation was more severe at the cell center compared to the bottom edge.
- The center cathode showed a larger drop in Li concentration, and the center anode showed a larger increase in Li concentration (and SEI thickness) compared to the edge.

5. Significance

This study provides critical insights for the development of next-generation EV batteries:

Degradation Source Identification: It confirms that in NCM|Graphite/SiO $_x$ 21700 cells, capacity fade is driven by a combination of LLI (primarily due to SEI growth on the anode) and LAAM (mechanical breakdown of SiO $_x$ and graphite).
Design Implications: The finding that degradation is more severe at the cell center than the edges highlights the impact of thermal and electrolyte gradients in large cylindrical cells. This suggests that cell design and thermal management systems must be optimized to minimize these inhomogeneities to extend cycle life.
Methodological Benchmark: The combination of in operando neutron diffraction and NDP offers a powerful, non-destructive framework for diagnosing aging mechanisms that X-ray techniques might miss (due to the low scattering cross-section of light elements like Li).

In conclusion, the paper demonstrates that lithium is effectively "consumed" on the anode side (via SEI and trapping) while the cathode becomes increasingly depleted, a process exacerbated by spatial inhomogeneities in larger cell formats.