Keywords: Battery, Coin Cell, Coin Cell DSC, DSC, Potentiostat, Mass Spectrometer
TA504
Abstract
Advanced battery development requires efficient methods to evaluate thermal safety early in the design process. Calorimetry, in particular differential scanning calorimetry (DSC), is widely used in battery research for detecting hazardous exothermic reactions and other thermal events. However, traditional DSC workflows require disassembling cells to analyze individual components — a destructive process that is time consuming, reduces reproducibility, and ignores critical cell-level interactions present in complete batteries.
This application note introduces the TA Instruments™ Coin Cell DSC, a DSC instrument specifically engineered to analyze intact 2032-format coin cells — a standard industry test bed. By eliminating the need for cell teardown, this system reduces sample preparation time by greater than 90%, preserves the structural integrity of the sample, and provides more representative thermal data. Critically, the Coin Cell DSC supports simultaneous collection of heat flow, evolved gas (via MS or FTIR), and electrochemical data (via potentiostat, battery cycler, or voltage monitoring device), enabling a comprehensive, multi-dimensional view of battery behavior in a single experiment. Direct coin cell calorimetry offers a practical, reliable solution for the early-stage safety screening of emerging battery materials and designs, ensuring higher data fidelity and more efficient development cycles.
As battery energy densities increase through new technologies and chemistries, it becomes more important to determine formulation safety and stability earlier in the development cycle. Critical safety defects related to potential material incompatibility or latent thermal instabilities will become much more costly, time-consuming, and potentially dangerous if only detected after scaling to larger cell formats. Evaluating battery safety begins at the materials level, but testing the components separately may not reveal the extent of their interactions [1].
DSC provides insight into thermal transitions and reaction heat flow, but its sample configuration imposes practical limitations. Most studies are confined to discrete battery components, mixtures of components, or materials harvested from aged or tested batteries [1–4]. This traditional workflow requires disassembling cells in a glovebox, harvesting electrode materials, and preparing hermetic or high-pressure DSC pans — a process that can take upwards of 30 minutes per sample. Beyond the time cost, this approach introduces variability through electrode handling, electrolyte loss, and atmospheric exposure, all which compromise reproducibility. Perhaps most importantly, it destroys the very celllevel interactions that are most relevant to safety: cathode–anode crosstalk, electrolyte decomposition cascades, and the internal short circuit pathways that drive thermal runaway in real batteries.
To address these limitations, the Coin Cell DSC enables battery researchers and developers to quickly and confidently assess the thermal properties of their batteries in a convenient coin cell format. This eliminates the need to teardown cells in a glovebox to test battery materials with different states-of-charge (SOC), formation protocols, or cycle-aging. The time savings are substantial: sample preparation for the Coin Cell DSC requires approximately 2 minutes — compared to over 30 minutes for a traditional teardown-and-pan workflow — representing a greater than 90% reduction in preparation time. This efficiency gain transforms thermal safety screening from a bottleneck into a highthroughput capability.
The Coin Cell DSC cell is based on the same proven design as the Discovery™ DSC but scaled larger to accommodate 20 mm diameter coin cells placed in specially constructed coin cell capsules. This allows complete coin cell batteries to be thermally tested without any additional preparation beyond mounting the cell into the chosen capsule.
The design of the DSC and capsules support multiple streams of data collection. Real-time gas analysis via mass spectrometry (MS) or Fourier-transform infrared spectroscopy (FTIR) is enabled through connections to the DSC lid with the heated transfer lines. Electrochemical measurements, including voltage monitoring and charge cycling, are supported through separate electrical connections on the lid.
With a temperature range spanning either -80 °C to 400 °C or room temperature to 600 °C, depending upon cooler selection, the Coin Cell DSC provides battery researchers with the capability to perform temperature-controlled charge cycling under typical battery operating conditions or heat-to-failure experiments. Figure 1 below shows the unit.
Single-layer Li-ion batteries were constructed at Waters Corporation in standard 2032 stainless steel casings with a polypropylene gasket to form a hermetic seal. The batteries were assembled in reverse order, with the graphite anode on the positive end. This was done to give a slight increase to the signal from the graphite thermal decomposition, as the flatter positive casing is placed against the sensor. The capacity is approximately 3 mAh. The casing, cathode, anode, and separator materials were purchased from MTI Corporation, while the electrolyte was purchased from Sigma-Aldrich. Material specifics are shown in Table 1 below.
Table 1. Material specifics for the constructed batteries.
| Component | Material | Specifics |
|---|---|---|
| Cathode | NMC 5:2:3 on Aluminum Foil |
Areal Capacity: 1.774 mAh/cm2 Diameter: 15 mm |
| Anode | Graphite on Copper Foil |
Areal Capacity: 2.725 mAh/cm2 Diameter: 19 mm |
| Separator | Polypropylene | 2 sheets Diameter: 19 mm |
| Electrolyte | LiPF6 in EC/DEC/DMC |
1 M LiPF6 in EC/DEC/DMC 1:1:1 (v/v/v) |
After assembly, the cells were allowed to rest on the benchtop for 12 hours at room temperature to wet the electrodes. The formation method consisted of one cycle at C/10 (300 μA), followed by 10 cycles at C/5 (600 μA). Batteries were stored at approximately 50% SOC until ready for testing.
The Coin Cell DSC has two specially designed capsules for testing 2032-coin cells. For heat-to-failure tests discussed in this application note, the batteries were placed in Coin Cell Standard Capsules. The pan’s bottom and lid are joined with a threaded connection and there is a hole in the center of the lid to allow gases to escape. The hole in the lid will expose the coin cell casing, where an electrical connection to the cathode can be made. The other electrical lead contacts the anode via the top of the stainless-steel capsule lid. Figure 2 shows the electrical setup of leads and capsules. To prevent shorting of the battery through the capsule lid and bottom, a mica disc is placed between the lid and the battery.
Finally, a capsule spacer was placed on top of the mica disk to firmly hold the battery in position during heating. This helps prevent the battery from moving once the casing gasket melts and gases are vented from the capsule. Batteries were checked for proper voltage (~3.4 V) before being placed in the DSC. Figure 3 shows the Standard Capsule and the Cycling Capsule. The latter is a lighter-weight capsule used for improved sensitivity in nondestructive charge cycling experiments.
A potentiostat was used to perform a constant current (CC) charge of the batteries to 4.2 V under an isothermal set point of 25 °C. Then the open circuit voltage (OCV) was recorded during the 2 °C/ min heating ramp. The potentiostat is connected to the Discovery Coin Cell DSC through electrical connections at the top of the lid shown in Figure 4 below.
A Discovery mass spectrometer was used to monitor off-gas evolution. This unit is a benchtop quadrupole spectrometer with a range of 1 to 300 amu. The MS is connected to the lid of the Discovery Coin Cell DSC through a heated transfer line. Figure 4 above shows the connection point on the left-hand side.
With the experimental framework described above, the following sections trace the complete failure sequence of a fully charged NMC 523 coin cell during a heat-to-failure experiment. Rather than presenting each measurement modality in isolation, the results are organized chronologically — following the battery from initial stability through catastrophic failure. The mass spectrometry and voltage data are introduced at the points in the thermal profile where they become most informative, mirroring the way the Coin Cell DSC is designed to be used: as a single platform capturing multiple dimensions of battery behavior simultaneously.
Figure 5 presents an example of a complete heat flow profile for the coin cell, heated from room temperature to 600 °C at 2 °C/min. For all heat flow data presented, the temperature axis was calibrated using indium and lead standards. For a liquidelectrolyte Li-ion battery, there are typically five distinct stages of thermal, chemical, and electrical activity identifiable across this temperature range. Each stage reveals a different aspect of the cell’s progression toward failure.
Stage 1: Baseline Stability (Room Temperature to ~100°C)
During the initial phase of heating, a coin cell exhibits no significant thermal activity. The heat flow baseline remains flat, and the open-circuit voltage (OCV), recorded by a potentiostat connected through the DSC lid, holds steady near the expected value of approximately 4.15 V. This quiescent region confirms that the cell is electrochemically and thermally stable under mild heating for short periods of time, establishing a clean baseline against which subsequent events can be compared.
Importantly, both the thermal and electrochemical channels are active from the start of the experiment — a key advantage of the Coin Cell DSC platform. In a traditional DSC workflow, voltage monitoring during heating would likely not be possible because the cell would likely have already been disassembled.
Stage 2: Early Degradation — SEI Decomposition (~100–165°C)
The first indication of thermal degradation appears as a multi-peak, low-intensity exotherm with an onset near 117°C in Figure 5 and near 119°C in Figure 6, consistent with the decomposition of the solid electrolyte interphase (SEI) layer on the graphite anode. The SEI — a passivation film formed during initial battery cycling — is known to become unstable at elevated temperatures, releasing small amounts of heat as it breaks down.
Shortly after, at approximately 130°C, the open-circuit voltage begins to deviate significantly from its steady-state value. While the voltage data does not yet indicate an internal short circuit, it provides an early electrochemical signature that the cell’s internal chemistry is changing. The temperature correlation between the thermal signature of the SEI exotherm and the first voltage perturbation illustrates the value of simultaneous thermal and electrochemical measurement: neither signal alone would tell the full story, but together they reveal that thermal decomposition of the SEI is accompanied by a measurable disruption to the cell’s electrochemical equilibrium.
Stage 3: Cell Venting and Electrolyte Release (~165°C)
At 166.8 °C, the polypropylene gasket of the 2032 casing melts, and the coin cell vents. This event manifests in the heat flow signal as a sharp endothermic response — a distinctive signature that is clearly resolved by the Coin Cell DSC. Once the hermetic seal is broken, the volatile organic carbonate solvents in the electrolyte rapidly escape.
The Discovery mass spectrometer, connected to the DSC lid via a heated transfer line, immediately detects the released solvents (Figure 7). Mass fragments associated with all three electrolyte components — diethyl carbonate (DEC), dimethyl carbonate (DMC), and ethylene carbonate (EC) — are observed with no measurable delay between the venting endotherm and the MS response. This confirms both the sensitivity of the detection system and the efficiency of the gas transfer path. The rapid detection is expected at the venting temperature, as both DEC (boiling point: 125.9 °C) and DMC (boiling point: 90 °C) are well above their boiling points, while EC, though below its boiling point of 243 °C, possesses a sufficiently high vapor pressure (>10 kPa) [5] at this temperature to escape readily from the open capsule.
The simultaneous capture of both the thermal venting signature and the chemical identity of the released vapors provides information that neither DSC nor MS alone could deliver. The heat flow pinpoints when the cell loses containment; the mass spectrometer identifies what escapes. For battery developers evaluating electrolyte formulations, this combination offers direct, cell-level evidence of solvent release behavior under thermal abuse, and confirms that all three solvents are released, despite EC being below its boiling point.
Stage 4: Catastrophic Exothermic Reactions and Electrical Failure (~200–300 °C)
Above 200 °C, the heat flow signal is dominated by two intense exothermic peaks, with maxima at approximately 231 °C and 257 °C. These represent the most energetic events in the entire heating scan and correspond to the temperature region most closely associated with the onset of thermal runaway. Similar exothermic activity has been reported by Zhou et al. [1] in charged NMC 811 full cells, where it was attributed to cathode decomposition and cathode–anode crosstalk — reactions that generate substantial heat and can become self-sustaining.
The mass spectrometer provides critical chemical context for these thermal events. During the exothermic window between 200 °C and 300 °C, CO2 evolution is clearly detected — a signature consistent with the release of lattice oxygen from the cathode and its subsequent reaction with conductive carbon and residual organic material [1,6]. The detection of CO2 concurrent with the exothermic peaks directly links the observed heat generation to specific chemical processes within the cell.
The voltage data completes the picture. During this same temperature window, the open-circuit voltage undergoes a sharp drop near 200 °C before collapsing to zero at approximately 250 °C, marking the point of full internal short circuit (ISC) (Figure 8). Notably, the ISC occurs after the initiation of strong exothermic activity, suggesting that in this chemistry and cell configuration, thermal decomposition precedes — and likely precipitates — electrical failure, rather than the reverse [7].
The convergence of all three signals — thermal energy release in the heat flow, CO2 evolution in the mass spectrum, and voltage collapse in the electrochemical data — within the same narrow temperature window provides a comprehensive fingerprint of the thermal failure sequence. This level of integrated, time-resolved detail is not achievable through traditional workflows that test components separately or rely on a single measurement modality. It is precisely the type of multi-dimensional insight that the Coin Cell DSC was designed to deliver.
Stage 5: Post-Failure Decomposition (300–600°C)
Beyond the primary failure events, the heat flow signal reveals a final thermal transition at 410.9 °C, corresponding to the decomposition of the polypropylene separator and gasket materials. While this event occurs well above the temperatures of greatest safety concern, it completes the thermal inventory of the cell and may be of interest for studies of fire propagation behavior or post-mortem materials analysis.
Together, the five stages outlined above — from baseline stability through post-failure decomposition — constitute a complete thermal, chemical, and electrical analysis of a lithium-ion coin cell under thermal abuse. Each stage provides distinct safetyrelevant information, and the ability to capture all of it in a single experiment represents a fundamental advantage of the Coin Cell DSC approach.
Measurement Confidence: Reproducibility of Intact Cell Testing
A common concern with any calorimetric measurement on batteries is reproducibility. Traditional DSC workflows require disassembling the cell, harvesting individual components, and repackaging them into hermetic DSC pans — a process that introduces variability at every step. Electrode handling, electrolyte loss, atmospheric exposure, and inconsistent sample mass all contribute towards run-to-run scatter that can obscure meaningful differences between formulations.
The Coin Cell DSC eliminates these sources of variability by testing the cell as-built. Figure 9 presents an overlay of three independent heating scans performed on three separate coin cells of identical construction and state-of-charge. These data show excellent reproducibility in both the type and temperature of every major thermal feature — including the SEI decomposition onset, the venting endotherm, the two primary exothermic peaks, and the high-temperature polymer decomposition. This consistency is particularly notable given the complexity of the sample and the number of overlapping reactions involved.
This level of reproducibility gives researchers confidence that observed differences between formulations, states-of-charge, or aging protocols reflect genuine material behavior rather than sample preparation artifacts. It also enables meaningful statistical comparison across experimental conditions — a prerequisite for systematic safety screening tests.
The TA Instruments Coin Cell DSC is a specialized DSC designed for the direct analysis of Li-ion batteries in standard 2032 coin cells. By circumventing the traditional, labor-intensive process of cell teardown and reconstruction — and reducing sample preparation time by greater than 90% — this platform removes a critical bottleneck in battery safety evaluation while preserving the cell-level interactions that are most relevant to thermal runaway.
The heat-to-failure experiments presented here on single-layer NMC 523 secondary Li-ion batteries demonstrate the power of simultaneous, multi-modal data collection. By tracking heat flow, evolved gas composition, and open-circuit voltage through a single controlled heating experiment, the Coin Cell DSC revealed a complete five-stage failure sequence. The five stages examined were: baseline stability, SEI decomposition with concurrent voltage deviation, cell venting with immediate solvent detection, catastrophic exothermic reactions coinciding with CO2 evolution and internal short circuit, and post-failure polymer decomposition.
The convergence of thermal, chemical, and electrical signals within the critical 200–300°C window provided an integrated fingerprint of the thermal failure sequence that single measurement techniques do not capture alone.
Triplicate measurements confirmed excellent reproducibility of all major thermal features, underscoring the advantage of testing intact cells over reconstructed samples. This consistency enables meaningful comparison across formulations, states-of-charge, and aging conditions — the foundation of systematic safety screening.
By bridging the divide between material-level and cell-level testing, the Coin Cell DSC provides researchers and developers a new tool to evaluate thermal safety earlier in the development cycle. Looking ahead, the platform’s rapid throughput and minimal sample preparation make it well-suited for integration into highthroughput screening workflows, automated testing pipelines, and studies of emerging battery chemistries — including solid-state electrolytes, sodium-ion systems, and next-generation high-nickel cathodes — where early-stage thermal safety data is essential to guide material selection and accelerate safe commercialization.
References
- Hanwei Zhou, Mukul Parmananda, Kyle R. Crompton, Michael P. Hladky, Martin A. Dann, Jason K. Ostanek, and Parthat P. Mukherjee, Energy Storage Materials, 44, pp. 326–341 (2022)
- Nathan B. Johnson, Bhuvsmita Bhargava, Jonathan Chang, Samaa Zama, William Schubert, and Paul Albertus, Applied Materials and Interfaces, 15, pp. 57134–57143 (2023)
- Zixuan Wang, Bhuvsmita Bhargava, Nathan Brenner Johnson, Alex Martin Bates, Loraine Torres-Castro, M. Florencia Petracci, Michael Grady McKee, William Shubert, Samaa Zaman, and Paul Albertus, App. Matls. & Interfaces, 17, pp. 59289–59300 (2025)
- Safety Evaluation of Lithium-ion Battery Cathode and Anode Materials Using Differential Scanning Calorimetry, TA 467
- Vaclav Pokorny, Vojtech Stejfa, Michael Fulem, Ctirad Cervinka, and Kvetoslav Ruzicka, J. Chem. & Eng. Data, 62, pp. 4174–4186 (2017)
- Zhenxiang Tao, Haozhi Zhou, Yuyao Cao, Xuob Huang, Wei Wang, and Hui Zhang, Thermal Science and Engineering Progress, 69, pp. 104472–104493 (2026)
- Donal P. Finegan, Eric Darcy, Matthew Keyser, Bernhard Tjaden, Thomas M. M. Heenan, Rhodri Jervis, Josh J. Bailey, Romeo Malik, Nghia T. Vo, Oxana V. Magdysyuk, Robert Atwood, Michael Drakopoulos, Marco DiMichiel, Alexander Rack, Gareth Hinds, Dan J. L. Brett, and Paul R. Shearing, Energy Environ. Sci., 2017, 10, pp. 1377–1388
Acknowledgement
For more information or to request a product quote, please visit www.tainstruments.com to locate your local sales office information.
This paper was written by Gray Slough, Matt Lewis, Jeremy May, and Hang Lau, application scientists at Waters Corporation.
TA Instruments and Discovery are trademarks of Waters Corporation or its affiliates.
Click here to download the printable version of this application note.