Aluminum Clad Coin Cell Case: When It Helps (and Fails)

An aluminum clad coin cell case is an ordinary SS304 or SS316 can with a thin aluminum layer bonded to the inside of the positive shell, so the cathode side of your cell sits behind the same AlF₃ passivation film that protects your aluminum current collector. It exists to stop iron, chromium and nickel dissolving out of stainless steel above roughly 4.2 V and contaminating your data. But it is a conditional upgrade, not a universal one — with LiFSI or LiTFSI electrolytes, a thin or porous aluminum layer can corrode faster than bare steel. Here is how to decide, plus a three-cell test that tells you before you commit a full matrix.

What an aluminum clad coin cell case actually is

Al-clad means a layer of aluminum bonded to the stainless steel positive can, usually on the inner face and sometimes on both. The steel does the mechanical work of the crimp seal; the aluminum only handles electrochemistry.

A CR2032 set has five parts, and only one of them is clad:

• Positive can — 20.0 mm outside diameter, the deeper shell, in contact with the cathode. This is the clad part.
• Negative cap — shallower, carries the polypropylene gasket, stays plain stainless.
• Gasket — PP as standard; PFA or PTFE above about 60 °C.
• Spacer disc — SS304, 15.5–16.2 mm diameter, 0.2 / 0.5 / 1.0 mm thick.
• Wave spring — roughly 15.4 mm diameter, 1.1–1.4 mm free height.

Shell walls run 0.20–0.25 mm; the aluminum adds microns on top of that. [VERIFY cladding thickness against your supplier's certificate of analysis — most product listings omit it.]

The designation follows IEC 60086: C for lithium chemistry, R for round, 20 for 20 mm diameter, 32 for 3.2 mm height. Same family: CR2016, CR2025, CR2450 (24.5 × 5.0 mm). 

Clad, coated or plated — why the wording matters

Roll-bonded cladding is a metallurgical bond — continuous, and ductile enough to survive deep drawing and crimping. A deposited or electroplated coating is thinner and more likely to carry pinholes.

Suppliers use both terms interchangeably, so two vendors' parts with the same label can be different products. Ask for layer thickness and deposition method; if nobody can answer, assume it is a coating. The failure mode below is a direct consequence of a layer that is not dense.

The problem the aluminum layer solves

Stainless steel dissolves anodically above roughly 4.2–4.5 V vs Li/Li⁺, and the positive can sits in direct electrical contact with your cathode. It is a large, un-instrumented electrode you never accounted for.

Scale matters here. The floor of a CR2032 positive can is about 3.1 cm². A 14 mm cathode disc is 1.54 cm². The can is twice the area of the electrode you are studying, before you count the wall.

Dissolved Fe, Cr and Ni cross the separator, plate on the negative electrode, thicken the SEI and consume lithium inventory. Aluminum behaves differently in LiPF₆: fluoride released as PF₆⁻ decomposes builds an AlF₃ film only nanometres thick — thin enough for electron tunnelling, dense enough to block Al³⁺ dissolution. That is precisely why commercial cells use aluminum foil rather than steel as the positive current collector. The clad case extends the same chemistry to the can.

Four signatures of case corrosion in cycling data

• Coulombic efficiency is stuck 0.2–0.5% below your pouch or Swagelok baseline for the same electrode.
• Fade that accelerates after 30–50 cycles instead of decaying smoothly.
• Rising cell resistance and a hold current that never settles.
• On teardown: grey or brown discolouration inside the positive can, pits under a 10× loupe, faintly yellow electrolyte.

When an aluminum clad coin cell case makes corrosion worse

Three situations turn the clad case from an upgrade into a liability.

Imide salts: LiFSI and LiTFSI

AlF₃ passivation is a LiPF₆ phenomenon. With imide salts the protective film does not form the same way and aluminum dissolves continuously. Published work places a threshold near 4.3 V vs Li/Li⁺ for LiFSI at conventional 0.8–1.8 molal concentrations, above which corrosion runs away.

The nuance most guides skip: concentration changes the answer. At roughly 4–5 M LiFSI, a passivating AlF₃ layer does form, partly from fluoride present as a salt impurity. "LiFSI means no aluminum" is too blunt a rule.

The cladding may not be dense

In 2023 the Journal of The Electrochemical Society published a study in which graphite/NCM811 coin cells built in Al-clad cases performed worse than identical cells in plain stainless steel. On teardown the clad cases were severely corroded while the steel ones looked intact.

The aluminum layer was porous. Electrolyte permeated to the steel beneath and formed a galvanic couple with the aluminum, lowering the overpotential for aluminum oxidation and accelerating attack on the aluminum itself.

A poor aluminum layer is worse than no aluminum layer. Treat every new case lot as a new experimental variable, and do not change supplier partway through a study.

Never on the negative side

Aluminum alloys with lithium below about 0.4 V vs Li/Li⁺. Graphite operates from 0.2 V down toward 0.05 V and lithium metal sits at 0 V, so aluminum there would consume lithium inventory, swell, and can distort the crimp seal. That is why commercial kits clad the positive can only. For the negative side the specialty option is copper — Cu-clad, or the Ni/SS/Cu tri-clad long used in alkaline button cells.

Aqueous and alkaline systems

Aluminum corrodes in both strongly acidic and strongly alkaline aqueous electrolytes, with no passivation advantage. For Zn-ion, Zn-air or alkaline work, specify SS316L, Ti-clad or Ni/Cu clad sets.

Match the case to your salt and voltage window

Choose from two inputs — the salt and the upper cutoff — rather than from the chemistry name.

Your system	Upper cutoff	Salt	Case to specify
LFP, LTO, most half cells	≤ 4.0 V	LiPF₆	SS304 — cladding adds nothing
NMC532/622, LCO	4.2–4.3 V	LiPF₆	SS316L; Al-clad optional
NMC811, LMR, LNMO	4.4–5.0 V	LiPF₆	Al-clad positive can — the design case
Any cathode, 1 M LiFSI or LiTFSI	> 4.2 V	imide	Screen first; consider SS316L plus ~4 M HCE
Li–S	~2.8 V	LiTFSI + LiNO₃	SS316L — polysulfides attack 304 more readily
Sodium-ion	4.0–4.5 V vs Na	NaPF₆	Al-clad works well
Aqueous Zn-ion, Zn-air	—	ZnSO₄, KOH	SS316L, Ti-clad or mesh can — not aluminum
Solid-state pellet cells	varies	—	SS316L with thick spacer, or a split cell

 

Sodium is the useful exception. It does not alloy with aluminum at battery potentials, which is why sodium-ion cells run aluminum current collectors on both electrodes — and why clad hardware is more forgiving in Na cells than Li cells.

A three-cell, 24-hour screen that qualifies a case lot

Before building a 40-cell matrix on a new lot, spend three cells and one day proving the case is stable in your electrolyte.

1. Build blank cells. Lithium metal on the negative side, no cathode at all, so the bare inside of the positive can faces the separator. Production separator, electrolyte and volume — 60–120 µL for CR2032.
2. Rest 6–12 hours at open circuit until OCV is stable.
3. Hold the can at your upper cutoff — 4.5 V for NMC811, 4.9 V for LNMO. A potentiostat channel is cleanest, but most cyclers run a constant-voltage hold.
4. Log current for 24 hours, normalised to wetted area: about 3.1 cm² for the can floor.
5. Read the curve. Decaying below roughly 1 µA/cm² within a few hours means the surface is passivating. Settling at 1–10 µA/cm² is marginal — fine at 4.3 V, not at 4.9 V. A plateau at tens of µA/cm², or a rising current, means reject the lot.
6. Open one cell. Discolouration, dark spotting or tinted electrolyte confirms what the current told you.

Always run a plain SS304 case in the same batch as your control. This is the measurement the current-collector literature runs on aluminum foil coupons — you are simply using the can as the working electrode.

Worked example: an LNMO half cell that was not the cathode's fault

Symptoms. LNMO half cells cycled 3.5–4.9 V in 1 M LiPF₆ EC/DMC. Coulombic efficiency sits at 96–97% and does not improve after formation, capacity drops a few percent every ten cycles, and one or two cells per batch stop holding voltage.

Work through it cheapest test first:

1. Rule out leakage. Weigh cells before and after 48 hours at 45 °C. Consistent mass loss means a bad crimp, not corrosion. [VERIFY the threshold against your balance — you need 0.1 mg resolution.]
2. Drop the cutoff to 4.3 V on five cells. If CE jumps above 99%, the problem is potential-dependent: corrosion or electrolyte oxidation, not the active material.
3. Run the blank-cell hold at 4.9 V, three SS304 against three Al-clad. Similar currents on both means the electrolyte is oxidising; a large gap means the case.
4. If steel is the culprit, switch to Al-clad — then re-run step 3 on the new lot rather than trusting the label.
5. If both run high, no case will fix it. You need an additive or a different solvent system.

The case is the cheapest of the four candidates to test, which is why it belongs early in the sequence rather than last.

Assembly variables that matter more than the cladding

Crimp quality and stack height produce more cell-to-cell scatter than case material does.

• Crimp pressure — hydraulic crimpers typically run 750–1,000 psi for CR2032 [VERIFY against your die and press specification]. Log it every batch; it is the most common uncontrolled variable in lab coin cell data.
• Stack height — a CR2032 has about 3.2 mm of internal room. If the wave spring bottoms out, contact pressure is uncontrolled. Swap a 1.0 mm spacer for 0.5 mm with thick electrodes or lithium foil.
• Electrolyte volume — pipette it, 60–120 µL, identical across the matrix. Over-flooding hides corrosion behind excess salt.
• Gasket rating — PP creeps above roughly 60 °C and the seal opens; specify PFA or PTFE for aging studies. [VERIFY ratings with your supplier.]
• Handling — a burr on the crimp lip is a leak path, and the soft aluminum layer scratches easily. Never reuse cases.

The wider specialty case family

Case type	Construction	Typical use	Watch out for
SS304 standard	304	LFP, LTO, graphite, ≤ 4.2 V	Not for high voltage or aqueous
SS316 / 316L	316L, low carbon	High-V, Li–S, chloride or sulfur	316L (~0.03% C) pits less than 316 (~0.08% C)
Al-clad positive can	Al on 304/316	> 4.3 V cathodes in LiPF₆	Layer density; not with 1 M imide salts
Ti-clad / Ti-coated	Ti on SS	High voltage and corrosive aqueous	Cost and limited availability
Cu-clad, Ni/SS/Cu tri-clad	Cu inner face	Anode-side contact, alkaline and Zn button cells	Copper is not for the positive side
Mesh or vented can	304 with mesh window	Li-air, Zn-air, gas-consuming cells	Needs a controlled-atmosphere rig
Window can (Kapton, Be, glass)	SS plus bonded window	Operando XRD and Raman	Seal reliability; short experiments only

 

Most labs need two SKUs on the shelf: bulk SS304 for routine work, and a small stock of Al-clad or 316L for the high-voltage matrix. Five specialty variants used twice a year mostly age in a drawer.

Cost, lead time and when to skip the upgrade

Al-clad sets cost a multiple of plain SS304, not a small premium, and usually ship in packs of 10–60 rather than 100. [VERIFY current pricing — coin cell case prices move with steel cost and pack size.] Cladding is a small-batch process, so lead times run longer. Buy one lot large enough to cover a whole study, because lot-to-lot variation surfaces later as scatter you cannot explain.

Skip it below 4.2 V, with LFP or LTO, or in any aqueous system. Specify it for any cathode charged above 4.4 V in a LiPF₆ electrolyte, especially long cycling runs and elevated-temperature aging, where corrosion has time to accumulate.

FAQ 

Is an aluminum clad coin cell case clad on both shells?

No — in almost all commercial kits only the positive can carries the aluminum layer, and the negative cap stays plain stainless steel. Aluminum alloys with lithium below roughly 0.4 V vs Li/Li⁺, so aluminum on the negative side would consume lithium inventory, swell, and can distort the crimp seal.

Can I use an aluminum clad coin cell case with LiFSI electrolyte?

Not without testing it first. LiFSI and LiTFSI do not build the protective AlF₃ film that LiPF₆ does, and published work reports aluminum corroding above roughly 4.3 V vs Li/Li⁺ at conventional concentrations. Highly concentrated LiFSI, around 4–5 M, behaves differently and can passivate. Screen your specific electrolyte.

What is the difference between Al-clad and Al-coated coin cell cases?

Clad normally means a roll-bonded metallurgical layer; coated means a thinner deposited film. Suppliers use both terms loosely, so the labels are unreliable. What matters is whether the layer is dense — a porous layer lets electrolyte reach the steel underneath and can accelerate corrosion. Ask for thickness and method.

Do I need Al-clad cases for LFP or graphite half cells?

No. LFP cuts off near 3.8 V and graphite half cells stay well below the stainless steel corrosion window, so plain SS304 performs identically. Reserve Al-clad stock for cathodes charged above about 4.4 V, where steel dissolution genuinely distorts capacity and coulombic efficiency measurements.

How do I tell whether a coin cell failed from case corrosion?

Look for coulombic efficiency stuck below your pouch-cell baseline, fade that accelerates after 30–50 cycles, and hold current that never settles. On teardown, check the inside of the positive can for grey or brown discolouration and pitting, and the electrolyte for a yellow tint.

Are Al-clad coin cell cases suitable for sodium-ion research?

Generally yes. Sodium does not alloy with aluminum at battery potentials, which is why sodium-ion cells use aluminum current collectors on both electrodes. That makes Al-clad hardware more forgiving in sodium cells than in lithium cells. Still confirm compatibility with your specific salt and upper cutoff.

Conclusion

The aluminum clad coin cell case solves one specific problem — anodic dissolution of stainless steel at the positive electrode above about 4.2 V in LiPF₆ electrolytes — and it solves it well when the aluminum layer is dense. It is not a general-purpose upgrade. With imide salts, on the negative side, or in aqueous systems it ranges from useless to actively harmful.