ABSTRACT 

 

 

Title of Thesis: 
QUANTITATIVE UNDERSTANDING OF 

TEMPERATURE RISE AND SAFETY IN HIGH–

ENERGY SOLID–STATE BATTERIES. 

  

 TAIWO OLADAPO OGUNDIPE, MASTER OF 

SCIENCE, 2024. 

  

Thesis Directed By: DR. PAUL ALBERTUS, DEPARTMENT OF 

CHEMICAL AND BIOMOLECULAR 

ENGINEERING. 

 

The rising demand for renewable energy and electric transportation has increased the need for 

advanced and safer battery technologies. Conventional lithium – ion batteries face limitations in 

energy density and safety risks due to the reaction of oxygen from the decomposed cathodes with 

other battery components, which can cause thermal runaway, leading to fires or explosions. Solid 

– state batteries, which use a solid electrolyte, offer a promising solution by potentially 

improving both energy density and safety. This study focuses on the thermal behavior and heat 

generation of anode – cathode – electrolyte (ACE) (Li / LPSCl / NMC811) solid-state batteries 

using differential scanning calorimetry (DSC). The results show significant heat generation, 

ranging from 4000 to 5400 J/g NMC811, with a corresponding adiabatic temperature rise of 1300 – 

1750 ℃. When small amounts of liquid electrolyte are added, the onset temperature is lowered, 

and the heat release shifts to higher temperatures. However, the total heat generation remains 

within a similar range. These findings provide insights into the thermal stability of all – solid – 



 

 

state batteries , and solid – state batteries with small amount of liquid electrolyte, contributing to 

the development of safer and higher energy density energy storage systems. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 

 

 

 

 

 

QUANTITATIVE UNDERSTANDING OF TEMPERATURE RISE AND SAFETY 

IN HIGH – ENERGY SOLID – STATE BATTERIES. 

 

 

 

by 

 

 

Taiwo Oladapo Ogundipe 

 

 

 

 

 

Thesis submitted to the Faculty of the Graduate School of the  

University of Maryland, College Park, in partial fulfillment 

of the requirements for the degree of 

Master of Science  

[2024] 

 

 

 

 

 

 

 

 

 

Advisory Committee: 

Professor Paul Albertus, Chair 

Professor Peter Kofinas 

Professor Chunsheng Wang 

 

 

 

 

 

 

 



 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

© Copyright by 

Taiwo Oladapo Ogundipe 

2024 

 

 

 

 

 

 

 

 

 

 



 

ii 

 

 

Acknowledgements 

I would like to express my deepest gratitude to my advisor, Dr. Paul Albertus, for his 

invaluable guidance, (financial) support, and mentorship throughout this research 

journey. His deep knowledge and insightful feedback have been instrumental in 

shaping my work, and I am grateful for the opportunities he has provided me to grow 

both academically and professionally. 

A special thanks goes to Bhuvsmita Bhargava, Zixuan Wang, and Karl Larson, whose 

contributions to various aspects of this research were immensely helpful. Your 

collaboration, technical support, and insights made a significant difference, and I truly 

appreciate your efforts. I am also deeply thankful to all my colleagues and lab members 

for their support and assistance during this research. Working alongside such talented 

and supportive individuals was a privilege, and your willingness to offer advice, 

troubleshoot challenges, and share knowledge has been a constant source of 

encouragement. To my family and friends, your unwavering support and understanding 

have been a source of strength and motivation. Thank you for believing in me and 

encouraging me throughout this journey. 

Lastly, I am grateful to Ford, the Center for Research in Extreme Batteries (CREB), 

and the University of Maryland College Park for the resources, funding, and state – of 

– the – art laboratory facilities that made this research possible. This work would not 

have been accomplished without the technical infrastructure and support provided. This 



 

iii 

 

 

thesis is the result of a collective effort, and I am deeply thankful to everyone who 

played a role in its completion. 

 



 

iv 

 

 

Table of Contents 
 

Acknowledgements  ...................................................................................................... ii 
Table of Contents  ........................................................................................................ iv 
List of Tables ............................................................................................................... vi 
List of Figures ............................................................................................................. vii 
List of Abbreviations ................................................................................................. viii 
Chapter 1: Introduction ................................................................................................. 1 

1.1. Background on Solid – state Battery Safety ...................................................... 1 
   1.1.1. NMC Cathode Materials .............................................................................. 3 
   1.1.2. Argyrodite solid electrolyte .......................................................................... 5 
   1.1.3. High – capacity Anodes ............................................................................... 6 
1.2. Introduction to Differential Scanning Calorimetry (DSC) Measurements ........ 7 
1.3. Literature Review . ............................................................................................. 8 
1.4. Thesis Focus .................................................................................................... 10 

Chapter 2: Methodology ............................................................................................. 12 
2.1. Materials Preparation ....................................................................................... 12 
2.2. Split Cell Preparation ....................................................................................... 13 
2.3. Anode – Cathode – Electrolyte (ACE) Sample Preparation ............................ 13 
2.4. Differential Scanning Calorimetry Measurements .......................................... 14 
2.5. Method of Integration ...................................................................................... 15 

Chapter 3: Results and Discussion  ............................................................................. 17 
3.1. Component DSC Results ................................................................................. 17 

3.1.1. Li0.2NMC811 cathode sheet ...................................................................... 17 
3.1.2. LPSCl Shard ............................................................................................. 17 
3.1.3. Li0.2NMC811 cathode sheet + LPSCl Shard  ............................................ 19 
3.1.4. Liquid electrolyte (LiPF6  in EC: DEC: 1:1 v/v) . .................................... 19 
3.1.5. Li0.2NMC811  cathode sheet + Liquid Electrolyte  ................................... 22 
3.1.6. Li metal ..................................................................................................... 23 
3.1.7. Li metal + LPSCl Shard  ........................................................................... 23 

    3.2. ACE Sample DSC Results ............................................................................... 24 
3.2.1. ACE Sample (NMC811 + Hot – pressed LPSCl + Li metal)  ................... 24 
3.2.2. ACE Sample (NMC811 + Cold – pressed LPSCl + Li metal) .................. 26 

    3.3. Effect of Liquid Electrolyte on the Heat Generation ....................................... 27 
Chapter 4: Conclusion ................................................................................................ 31 
    4.1. Results Summary ............................................................................................. 31 
    4.2. Limitations to Current Work  ........................................................................... 32 
    4.3. Recommendations and Future Work  .............................................................. 33 
Appendices  ................................................................................................................. 34 
   A.1: Sample MATLAB Algorithms for fitting and integrating raw DSC heat flow 

data .............................................................................................................................. 34 
   A.2: DSC heat flow of LPSCl shard. Reproduced from 24 ...................................... 38 
   A.3: DSC heat flow of 1M LiPF6 in EC:DEC (1:1 v/v). Reproduced from Roth et 

al.25 .............................................................................................................................. 39 



 

v 

 

 

   A.4: DSC heat flow of 1M LiPF6 in EC:DEC (1:1 v/v). Reproduced from Wang et 

al.26 .............................................................................................................................. 39 
   A.5: Calculations to determine the maximum volume of liquid electrolyte used in 

ACE samples . ............................................................................................................. 40 
Bibliography  .............................................................................................................. 42 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

vi 

 

 

 

 

List of Tables 

 

Table 1: Comparison of the onset temperature, heat release, and adiabatic temperature 

rise of different battery chemistries …………………………………………………..30 



 

vii 

 

 

List of Figures 
 

Figure 1.1: Conventional Lithium – ion Battery …………………………………….. 2 

Figure 1.2: Mass spectrometry profiles of different NMC cathodes with varying nickel 

contents ………………………………………………………………………………. 4 

Figure 1.3: Challenges of nickel rich – NMC cathode materials ……………………. 4 

Figure 1.4: Challenges and improvement strategies of lithium metal anode 

material ………………………………………………………………………………. 7 

Figure 1.5: Schematic diagram of a differential scanning calorimetry ……………… 8 

Figure 1.6: Thermal stability and optical images of the solid-state pouch cell at 0% and 

100% SOC …………………………………………………………………………… 9 

Figure 1.7: An overview of my research …………………………………………… 11 

Figure 2.1: Li6PS5Cl sulfide solid electrolyte pellet (0.870 mm thick and 0.5 – inch 

diameter) ……………………………………………………………………………. 12 

Figure 2.2: ACE Sample Setup …………………...………………………………... 15 

Figure 2.3: (a). Variations in the raw empty pan baselines. (b). Raw heat flow and the 

Baseline – corrected heat flow ……………………………………………………… 16 

Figure 3.1: Raw DSC result of Li0.2NMC811 cathode sheet ………………………. 18 

Figure 3.2: Raw DSC result of LPSCl shard ……………………………………….. 18 

Figure 3.3: (a) Baseline – corrected DSC result of Li0.2NMC811 cathode sheet + 

LPSCl Shard. (b) Comparison with Li0.2NMC811 cathode sheet only. (c) Integrated 

area of Li0.2NMC811 cathode sheet + LPSCl Shard ………………………………... 20 

Figure 3.4: (a) Baseline – corrected DSC result of the liquid electrolyte. (b) Integrated 

DSC result of the liquid electrolyte …………………………………………………. 21 

Figure 3.5: (a). Raw DSC result of the Li0.2NMC811 cathode sheet with liquid 

electrolyte. (b) Raw DSC result of the comparison of Li0.2NMC811 sheets with 

(Li0.2NMC811 sheets + liquid electrolyte) ………………………………………….. 22 

Figure 3.6. Raw DSC result of the Li metal ………………………………………... 23 

Figure 3.7: Raw DSC result of the reaction of lithium with LPSCl shard …………. 24 

Figure 3.8: (a) Baseline – corrected DSC result of the ACE sample with hot – pressed 

LPSCl shard. (b) Cumulative heat flow of the ACE sample ………………………... 25 

Figure 3.9: (a). Baseline – corrected DSC result of the ACE sample with hot – pressed 

LPSCl shard. (b). Cumulative heat flow of the ACE sample ……………………….. 26 

Figure 3.10: Baseline – corrected DSC result of the ACE sample with varying amounts 

of liquid electrolyte ………………………………………………………………..... 27 

Figure 3.11: Raw DSC result of the ACE sample (a) without and (b) with 0.5 µL liquid 

electrolyte …………………………………………………………..………………. 28 

 

 

 



 

viii 

 

 

List of Abbreviations 

ACE – Anode – Cathode – Electrolyte  

ARC – Accelerated Rate Calorimetry 

ASSB – All – Solid – State – Battery 

CC – Constant Current 

CV – Constant Voltage 

DEC – Diethyl Carbonate 

DSC – Differential Scanning Calorimetry 

EC – Ethyl Carbonate 

EV – Electric Vehicle 

LE – Liquid Electrolyte 

LiBs – Lithium-ion Batteries 

LPSCl – Lithium Thiophosphate (Li6PS5Cl) 

MS – Mass Spectrometry 

NMC – Nickel Manganese Cobalt Oxide (LiNixMnyCo1-x-yO2) 

PVDF – Polyvinylidene Fluoride 

SE – Solid Electrolyte 

SSE – Solid – State Electrolyte 

TGA – Thermal Gravimetric Analysis 

XRD – X – ray Diffraction 

XPS – X – ray Spectrometry  

 

 

 

 

 

 

 

 

 

 

 



 

1 

 

 

Chapter 1: Introduction 

1.1. Background on Solid – state Battery Safety.  

Lithium-ion batteries (LIBs) have emerged as one of the most extensively used energy 

storage systems, playing a crucial role in powering various technologies ranging from 

personal electronics like cellphones and laptops to large – scale applications, such as 

electric vehicles (EVs) and grid storage.1 (Note: all acronyms are defined in the List of 

Abbreviations above). The basic structure of a conventional lithium – ion battery 

consists of an anode typically made of graphite, a cathode composed of lithium metal 

oxides, a separator, and a liquid organic electrolyte containing lithium salt that 

facilitates the movement of lithium ions between the electrodes during charge and 

discharge cycles,2 as shown in Figure 1.1. As the world increasingly shifts towards 

renewable energy sources and electric transportation to mitigate the effects of climate 

change, the demand for advanced battery technologies has surged dramatically. It is 

projected that global battery demand will continue to rise exponentially in the coming 

decades, primarily due to the increased adoption of EVs and the need for large – scale 

energy storage solutions.3 However, despite their widespread use, conventional lithium 

– ion batteries have limited energy density (the amount of energy stored per unit 

volume or mass). Moreover, they pose safety concerns as the liquid electrolytes are 

flammable and prone to leakage.4 Battery safety is important to prevent the release of 

gases within the cell as a result of overcharging, damage, or overheating, which can 

lead to combustion, rupture, or thermal runaway where the temperatures exceed 500°C. 

To address these challenges, current research is focused on developing new battery 



 

2 

 

 

materials that offer higher energy densities to deliver longer ranges and improved 

safety. 

                      
 

Figure 1.1: Conventional Lithium-ion Battery. Reproduced from 2. 

 

More research is also focused on developing high energy density cathode materials and 

high – capacity anode materials which can store more energy. Solid – state batteries, 

which replace the liquid electrolyte in conventional Li – ion batteries with a solid 

electrolyte, are seen as a good alternative as the solid electrolyte is assumed to address 

many of the safety concerns associated with liquid electrolytes, reducing the risk of 

thermal runaway and improving the safety of the battery. However, solid – state 

batteries still show substantial safety challenges. While solid state batteries offer 

increased energy density, the high – capacity cathodes are more prone to thermal 

instability, posing a higher risk of heat generation during thermal runaway.5  



 

3 

 

 

1.1.1. NMC Cathode Materials.  

Lithium nickel manganese cobalt oxide, LiNixMnyCo1-x-yO2 (NMC), is recognized as a 

highly promising cathode material for next – generation batteries, offering a high 

specific capacity of approximately 200 mAh g–1 and a moderately high average 

discharge potential of around 3.8 V vs Li+ / Li. Among these, Ni – rich NMC cathodes 

(x > 0.5), especially LiNi0.8Mn0.1Co0.1O2 (NMC811), are attractive for their increased 

energy density due to higher nickel content. However, the increased nickel content also 

increased the reactivity of the cathodes due to the instability of nickel ions at higher 

temperatures, triggering oxygen release that causes exothermic reactions and making 

it thermally unstable.6  Bak et al investigated the thermal stability of NMC cathode 

materials with varying nickel contents using Time – Resolved X-ray diffraction (TR – 

XRD) and mass spectrometry (MS) up to 600 °C. The results showed that thermal 

stability of the NMC sheets decreases with increasing nickel content. Specifically, 

NMC811, which has the highest nickel content among the tested NMC samples, shows 

the lowest onset temperature for phase transitions and a sharper peak in oxygen release 

compared to NMC materials with lower nickel content due to the instability of nickel 

as it undergoes a rapid reduction from Ni4+ to Ni2+ during thermal decomposition,7 as 

shown in Figure 1.2. Furthermore, the effect of varying Ni contents on the 

electrochemical properties and thermal stability of NMC cathode sheets was studied. It 

was found that increasing nickel content led to a reduction in the thermal stability and 

capacity retention of the NMC materials.8 Also, Ni – rich NMCs exhibit poorer 

chemical stability under ambient conditions, and experience side reactions with the 

electrolyte at high voltages.9 At about 170 °C, the NMC cathode materials start to 



 

4 

 

 

decompose and release oxygen, which reacts with electrolyte, resulting in exothermic 

reactions and potential thermal runaway of LIBs10 (Figure 1.3). 

                             

Figure 1.2: Mass spectrometry profiles of different NMC cathodes with varying nickel 

contents. Reproduced from 7. 

 

 

                

Figure 1.3: Challenges of nickel rich – NMC cathode materials. Reproduced from 9. 



 

5 

 

 

1.1.2. Argyrodite solid electrolyte.  

Liquid electrolytes (LEs) used in traditional LIBs are flammable, which pose safety 

issues and limit the operation temperature range of LIBs. These limitations necessitate 

the development of safer alternatives, leading to the adoption of solid electrolytes, 

which offer potentially enhanced safety and performance. Recent research has 

increasingly focused on inorganic solid electrolytes, including sulfides, hydrides, 

oxides, nitrides, and phosphates. Among these, sulfide – based solid electrolytes have 

attracted significant attention due to their exceptional ionic conductivity, often 

surpassing 10⁻3 S / cm at room temperature, distinguishing them from other types of 

solid electrolyte.11 One of the key advantages of sulfide solid-state electrolytes (SSEs) 

is their lower Young’s modulus, which results in reduced stress formation and 

minimized contact loss during volume changes in cell operation. However, sulfide 

SSEs generally have lower chemical stability compared to oxides and are highly 

sensitive to atmospheric conditions, leading to a rapid decline in ionic conductivity 

when exposed to air. As a result, sulfide electrolytes are typically handled in inert – gas 

– filled gloveboxes to prevent degradation.12 Among the types of solid electrolytes, 

argyrodites have emerged as a promising class due to their high ionic conductivity at 

room temperature, low cost and good compatibility towards lithium metal.13 Argyrodite 

– structured sulfide solid electrolytes, such as Li6PS5X (X = Cl, Br), have become the 

dominant choice for all – solid – state battery fabrication due to their exceptional 

lithium – ion conductivities, reaching up to 2.4 mS·cm−1 at room temperature,13 making 

them among the most promising solid-state electrolytes for next-generation solid-state 

batteries.14  



 

6 

 

 

1.1.3. High – capacity Anodes.  

High – capacity anode materials are crucial for advancing lithium – ion batteries, 

especially as demand for high – performance batteries grow for electric vehicles (EVs) 

and portable electronics. Traditional Li – ion batteries use graphite as the anode 

material due to its stable electrochemical performance, but the theoretical specific 

capacity of the graphite anode material (372 mAh/g) limits its potential for Li – ion 

batteries.15 Unlike graphite which store lithium ions within its structure, lithium metal 

anode uses lithium in its metallic form with a theoretical specific capacity of 3860 

mAh/g and an ultra – low redox potential of – 3.04 V versus the standard hydrogen 

electrode, making it a highly promising anode material for achieving extremely high 

energy density batteries.16  However, one of the major problems with Li anode is that 

it forms dendrites (needle – like lithium deposits that can grow during charging cycles), 

and penetrates the separator between the anode and cathode, causing short circuits and 

potentially leading to thermal runaway17 (Figure 1.4).  

Silicon is another promising high – capacity anode material used for batteries due to its 

extraordinary theoretical capacity of about 4200 mAh/g, which is more than ten times 

that of conventional graphite anodes.18 This significant capacity could dramatically 

increase the energy density of Li – ion batteries, enabling longer driving ranges for 

EVs, extended battery life for electronic devices, and more efficient energy storage 

solutions. However, silicon causes large volume expansion during lithiation, causing 

the silicon particles to fracture, and leading to rapid capacity degradation over repeated 

charge – discharge cycles.19  



 

7 

 

 

                 

Figure 1.4: Challenges and improvement strategies of lithium metal anode material. 

Reproduced from 20. 

 

 

 

1.2. Introduction to Differential Scanning Calorimetry (DSC) 

Measurements.  

Differential Scanning Calorimetry (DSC) is a widely utilized thermal analysis 

technique that provides valuable insights into thermal transitions (endothermic or 

exothermic), including solid – solid and solid – liquid transitions of the components of 

a cell. By measuring the heat absorbed or released by a material as a function of 

temperature or time, DSC allows for the precise characterization of thermal properties 

and behaviors in different materials.21 In our heat flux DSC setup, the analysis takes 

place within a chamber containing two pans isolated from the surrounding 

environment. One pan holds the sample to be analyzed, while the other pan remains 



 

8 

 

 

empty and serves as a reference. Both pans are placed on top of two areal 

thermocouples to measure the temperature changes on both pans, and heated in the 

chamber through a gas flow stream. As the temperature changes at a constant rate, heat 

is transferred to the sample and reference, as shown in Figure 1.5, and the heat flow is 

calculated from the temperature difference between the two pans.22 The DSC analysis 

can give valuable information such as the total heat released, onset temperature, and 

the thermal transitions of the sample in the DSC pans. 

 

Figure 1.5: Schematic diagram of a differential scanning calorimetry. Reproduced 

from 22.  

 

1.3. Literature Review.  

Few studies have been conducted to investigate the thermal runaway of solid – state 

batteries. Yang et al investigated the thermal stability and oxygen – induced thermal 

runaway mechanisms in a 3.8 Ah pouch cell comprising lithium metal anode, a 



 

9 

 

 

Li6PS5Cl solid-state electrolyte, and LiNi0.5Mn0.3Co0.2O2 (NMC532) cathode, using 

extended-volume accelerating rate calorimetry at 0% and 100% state – of – charge 

(SOC). At 0% SOC, the cells exhibit high thermal stability, showing only slight self-

heating without thermal runaway, due to the formation of a passivating LiCl layer, 

which results from the mild reaction between lithium metal and the Li6PS5Cl 

electrolyte, limiting further reactions. However, at 100% SOC, thermal runaway occurs 

at temperatures above 275°C due to the reaction between oxygen released by the 

decomposing cathode and the solid – state electrolyte (Figure 1.6), while the lithium 

metal anode itself plays a secondary role, reacting only after the initial runaway is 

triggered.23 

                 

Figure 1.6: Thermal stability and optical images of the solid-state pouch cell at 0% 

and 100% SOC. Reproduced from 23. 



 

10 

 

 

Also, Rui et al,24 explored the thermal runaway mechanisms of sulfide – based all – 

solid – state – batteries (ASSBs) by investigating two distinct thermal runaway 

pathways (gas – solid reactions and solid – solid reactions) by using different sulfide 

solid electrolytes: glassy – ceramic SEs (such as Li3PS4 and Li7P3S11) and crystalline 

SEs (such as Li6PS5Cl and Li10GeP2S12). Results showed that the glassy – ceramic SEs 

undergo gas – solid reactions at around 200°C, triggered by the release of oxygen from 

the NMC cathode, resulting in substantial heat and toxic sulfur dioxide (SO2) gas 

generation. On the other hand, crystalline SEs remain stable against oxygen up to 

around 200°C but react with the decomposed products of the NMC cathode, such as 

transition – metal oxides and very little SO2 gas, to give a solid – solid reaction pathway. 

Differential scanning calorimetry – mass spectrometry (DSC – MS) and other 

characterizations were used to validate the thermal runaway mechanisms. The gas – 

solid reaction pathway (observed in glassy – ceramic SEs) is characterized by the direct 

oxidation of the SE by oxygen released from the NMC cathode, producing SO2 and 

substantial heat. In contrast, the solid – solid reaction pathway (observed in crystalline 

SEs) involves reactions with the solid decomposition products of the NMC cathode, 

resulting in different thermal failure behaviors. 

1.4. Thesis Focus.  

This thesis investigates the thermal runaway mechanism, reaction pathways, and the 

temperature rise in solid-state batteries, especially Li / Li6PS5Cl / NMC811 anode – 

cathode – electrolyte (ACE) samples using the differential scanning calorimetry (DSC). 

By examining the thermochemistry of these samples, the research aims to elucidate the 



 

11 

 

 

reaction pathways, and the temperature rise that leads to thermal runaway, a critical 

safety concern in solid – state battery technology. Furthermore, this thesis also explores 

the effect of small amounts of liquid electrolyte on the thermal behavior of these solid-

state batteries, as shown in Figure 1.7. The findings from this research contribute to the 

broader knowledge base necessary for the development of safer and more efficient 

energy storage systems, particularly in applications where high energy density and 

thermal stability are paramount. 

 

Figure 1.7: An overview of my research. 



 

12 

 

 

Chapter 2: Methodology 

2.1. Materials Preparation.  

Lithium Nickel Manganese Cobalt Oxide (LiNi0.8Mn0.1Co0.1O2) cathode sheets, 

commonly referred to NMC811 cathode, with a composition of 90% active material, 

5% PVDF binder, and 5% conductive carbon, were purchased from NEI corporation 

and stored in the glovebox. Li metal sheets and ingots were purchased from MSE 

Supplies and Sigma Aldrich, respectively. The lithium ingots were stored in mineral 

oil to avoid being oxidized. Argyrodite Li6PS5Cl sulfide solid electrolyte powders, 

commonly referred to as LPSCl were purchased from Ampcera with particle size of ~1 

µm, as shown in Figure 2.1. The powders were hot pressed (at 150 °C and 300 MPa for 

one hour) and cold pressed (at 20 °C and 300MPa for about ten minutes) in a 18mm 

custom made die, with a density of ~1.63 g/cm3. The LPSCl pellets were fractured to 

obtain shards for DSC pans. Celgard separators were purchased from Celgard, and used 

in their pristine form. 

 

       
 

Figure 2.1: Li6PS5Cl sulfide solid electrolyte pellet (0.870 mm thick and 0.5 – inch 

diameter).  



 

13 

 

 

2.2. Split Cell Preparation.  

The split cell preparation was carried out in the glovebox. The purchased NMC811 

cathode sheets, aluminum foils, and Lithium metal foils were punched into 18 mm 

diameter sheets. The Celgard separator was punched into 20 mm diameter sheet to 

adequately prevent contact of the anode and cathode materials. Both sides of the 

Lithium metal foil were scrapped to remove any oxidized layer deposited on the metal 

foil. The materials were layered into the split cell device in the following order from 

top to bottom: one sheet of the LiNi0.8Mn0.1Co0.1O2 sheets (with the current collector 

facing up), the Celgard separator, the Lithium metal disk, and the aluminum current 

collector. 4 – 5 drops of 1M LiPF6 in EC/DEC (1:1 v/v) electrolyte was added on the 

lithium metal disk and on the Celgard separator. The NMC811 cathode sheet was 

charged CC / CV to 4.3 V vs Li+ / Li in the split cell device to a lithium content of 0.2. 

The cathode sheet was removed from the split cell, rinsed with diethyl carbonate (DEC) 

about 10 times and dried at 70 °C in an oven under vacuum overnight to remove the 

liquid electrolyte salts. 

2.3. Anode – Cathode – Electrolyte (ACE) Sample Preparation.  

For the ACE sample preparation, two 4 mm diameter cathode sheets were used for the 

DSC tests. The ACE samples were assembled in a hermetically sealed stainless – steel 

pan (rated up to 200 bar), with around 0.15 mg lithium (cut from  bulk lithium ingot) 

put at the base of the pan, followed by ~11 mg of LPSCl shard, and two sheets of the 

Li0.2NMC811 (~ 4 mg with the current collectors facing down), as shown in Figure 2.2. 

Two sheets were used to match the capacity ratio of the lithium anode (N : P ratio of 



 

14 

 

 

1 : 1).  The mass of the LPSCl shard used is very important because it reacts with other 

components of the cell up to 550 °C. Furthermore, the effect of liquid electrolyte on 

the ACE samples were also determined. In this case, varying amount of liquid 

electrolyte (0.1 µL, 0.5 µL, and 1 µL) were added to the top of the Li0.2NMC811 sheet 

in the ACE sample assembly. The maximum volume of liquid electrolyte added was 

calculated based on the total volume of ACE samples within the DSC pan. To avoid 

exceeding the pan’s maximum operating pressure, 20% of the remaining volume was 

allocated for the liquid electrolyte, accounting for the evolved gases from the thermal 

decomposition of ACE samples, which also contribute to pressure buildup. This limits 

the maximum liquid electrolyte volume to ~ 2 µL. The calculations are shown in 

Appendix A.5. In addition, a single cathode sheet has a capacity of about 0.25 mAh. 

As a commercial Li – ion cell (with a porous anode and separator, in addition to 

cathode) has an electrolyte : capacity ratio of about 1.3 g / Ah, that would translate to 

about 0.65 mg of electrolyte within our DSC pans (for two cathode sheets), or about 

0.5 µL. Hence, the tests with 0.5 µL roughly match the electrolyte : capacity ratio 

expected in a cathode sheet alone.  

2.4. Differential Scanning Calorimetry Measurements.  

The sample and reference DSC pans were placed in the DSC equipment. The stainless 

– steel DSC pans were initially equilibrated at 40 °C for 15 minutes to ensure thermal 

uniformity. 



 

15 

 

 

             
 

Figure 2.2: ACE Sample setup. 

 

After equilibration, the pans were gradually heated to 550 °C at a ramp rate of 2°C / 

min to allow for the accurate measurement of the thermal transitions across the 

specified temperature range and minimize the effect of thermal lags from the pan on 

the measured heat, as the sample mass is very small compared to the pan mass. 

2.5. Method of Integration.  

To estimate the total heat released based on the variation in the different empty pan 

baselines (Figure 2.3a), the raw DSC heat flow is not subtracted from empty pan 

baseline. Instead, multiple points from the no – heat region of the raw heat flow curves 

are selected and fitted using the MATLAB algorithms I developed to form an 

integration baseline (Appendix A.1). This baseline is then used to correct the heat flow 

data, as shown in Figure 2.3b. Subsequently, the baseline – corrected heat flow is 

integrated across the exothermic regions to calculate the total heat released. 



 

16 

 

 

 
 

Figure 2.3: (a). Variations in the raw empty pan baselines. (b). Raw heat flow and the 

baseline – corrected heat flow.   



 

17 

 

 

Chapter 3: Results and Discussion 
 

3.1. Component DSC Results  

Conducting differential scanning calorimetry (DSC) tests for all components of the cell 

is important before performing the ACE sample DSC analyses to provide a 

comprehensive understanding of the thermal properties of each component, such as 

melting points, glass transition temperatures, and decomposition behavior, and predict 

how the materials behave when combined in an ACE sample. The DSC results (with 

the normalized figures) of the components are explained below. 

  3.1.1. Li0.2NMC811 cathode sheet 

Two 18 mm diameter sheets of Li0.2NMC811 (with current collector facing down) were 

placed in a sealed DSC pan. Two DSC replicate tests were performed at a heating rate 

of 2 ℃ / min from 40 ℃ to 550 ℃. From Figure 3.1, there are exothermic peaks around 

200 ℃, which are likely the decomposition of the cathode sheets. The exothermic peaks 

at 300 ℃ to 400 ℃ are likely from reactions of the oxygen from the decomposed 

Li0.2NMC811 with carbon. Although there is no exothermic peak around 400 ℃, it is 

expected that PVDF breakdown and following exothermic reactions might be present 

at this temperature and above. 

  3.1.2. LPSCl Shard 

Figure 3.2 shows three replicate DSC test results of LPSCl shard alone from 40 ℃ up 

to 550 ℃ at heating rates of 1 ℃ / min, 2 ℃ / min, and 4 ℃ / min, respectively. The 

LPSCl pellet was broken into shards and was sealed in a DSC pan. The result shows 

that LPSCl shows no obvious exothermic or endothermic peak up to 500 ℃, 



 

18 

 

 

highlighting its high thermal stability. Our result conforms with the results found in 

literature24 (Appendix A.2).  

 

Figure 3.1: Raw DSC result of Li0.2NMC811 cathode sheet.  

   

Figure 3.2: Raw DSC result of LPSCl shards.  

 



 

19 

 

 

  3.1.3. Li0.2NMC811 cathode sheet + LPSCl Shard 

The DSC result of the three replicates of the component NMC811 + LPSCl shard is 

shown in Figure 3.3a. It is observed that, compared to the DSC result of Li0.2NMC811 

only (Figure 3.3b), there is a large exotherm around 200 ℃ to 300 ℃, which is likely 

due to the reaction of the LPSCl with oxygen released from decomposed Li0.2NMC811, 

as seen in equation 1 (this gives reactants and products, but not exact stoichiometry): 

            O2 + Li6PS5Cl → S, SO2, Li3PO4, LiCl      (1) 

Also, at around 400℃, the PVDF in the Li0.2NMC811 cathode sheet breaks down, 

releasing hydrogen fluoride gas (HF), which can react with the products in equation (1) 

to form the possible products shown in equation (2). The total heat generation from 

50 ℃ to 550 ℃ is around 3100 – 3400 J/g NMC811.  

  HF + S + SO2 + Li3PO4 + LiCl → H2S, LiF, H3PO4, …      (2) 

  3.1.4. Liquid electrolyte (LiPF6  in EC: DEC: 1:1 v/v). 

Figure 3.4 shows two replicates of the DSC result of 0.5 µL liquid electrolyte, LiPF6 in 

EC: DEC (1:1 v/v) only. A small endotherm was observed at around 250 ℃, followed 

by two successive exotherms at around 275 ℃ and 280 ℃, respectively. The endotherm 

around 250 ℃ is likely due to electrolyte salt decomposition (equation 3). At 275 ℃, 

DEC may react with PF5 from the decomposed LiPF6  (equations 4). The heat 

generation from 50 ℃ to 550 ℃ is around 65 J/g LE, as seen in Figure 3.4b. 



 

20 

 

 

                   

Figure 3.3: (a) Baseline – corrected DSC result of Li0.2NMC811 cathode sheet + 

LPSCl Shard. (b) Comparison with Li0.2NMC811 cathode sheet only. (c) Integrated 

area of Li0.2NMC811 cathode sheet + LPSCl Shard. 



 

21 

 

 

LiPF6 → LiF + PF5                                           (3) 

C2H5OCOOC2H5 + PF5 → C2H5OCOOPF4 + HF + C2H4                                  (4) 

At 280 ℃, there may be a polymerization of EC, according to equation 5: 

       (CH2O)2CO → [(CH2CH2O)n COO]m + CO2        (5) 

As a result, our result conforms to the results of Roth et al25 and Wang et al26, as shown 

in Appendix A.3 and A.4, respectively. 

          
 

Figure 3.4: (a) Baseline - corrected DSC result of the liquid electrolyte. (b) Integrated 

DSC result of the liquid electrolyte.  



 

22 

 

 

  3.1.5. Li0.2NMC811  cathode sheet + Liquid Electrolyte 

Two replicates of the DSC test of Li0.2NMC811 cathode sheet + the liquid electrolyte 

were performed.  The 0.5 µL LE was added on top of the two sheets of Li0.2NMC811, 

with the cathode current collectors facing down, in a closed stainless – steel DSC pan. 

The DSC test was performed at 2 ℃ / min, and compared with Li0.2NMC811 only with 

the result shown in Figure 3.5a. It is observed that the addition of the liquid electrolyte 

to the Li0.2NMC811 sheets led to bigger exotherms around 180 ℃ to 250 ℃ compared 

to Li0.2NMC81 only (Figure 3.5b), showing an increased heat production from the 

reaction of Li0.2NMC811 and LE. 

               



 

23 

 

 

Figure 3.5: (a). Raw DSC result of the Li0.2NMC811 cathode sheet with liquid 

electrolyte. (b) Raw DSC result of the comparison of Li0.2NMC811 sheets with 

(Li0.2NMC811 sheets + liquid electrolyte). 

 

  

3.1.6. Li metal 

Figure 3.6 shows the DSC result of lithium metal at a heating rate of 2 ℃ / min, from 

40 ℃ to 500 ℃. As seen in the figure, there is an endotherm at around 183 ℃, which 

shows the melting point of lithium. After the lithium melting, there was no endothermic 

or exothermic peak. 

    

Figure 3.6: Raw DSC result of the Li metal. 

3.1.7. Li metal + LPSCl Shard 

Furthermore, the DSC tests for lithium ingot and LPSCl shard at 2 ℃ / min from 40 ℃  

to 500 ℃ were performed. It was observed that there was no visible exothermic or 

endothermic peak throughout the test as shown in Figure 3.7, which indicates that there 



 

24 

 

 

is limited reactivity between the lithium metal and the LPSCl. This was also confirmed 

by Yang et al.23. 

 
 

Figure 3.7: Raw DSC result of the reaction of lithium with LPSCl shard. 

 

 

 

3.2. ACE Sample DSC Results 

ACE sample DSC tests are essential in evaluating the combined thermal behavior of 

all the components. While component DSC tests provide insight into the thermal 

properties of each component, the ACE sample DSC analysis reveals how these 

materials interact when combined together. The DSC results of the ACE samples with 

hot – pressed and cold – pressed LPSCl shards are discussed below.  

        3.2.1. ACE Sample (NMC811 + Hot – pressed LPSCl + Li metal) 

4 repeated runs of the DSC tests were performed for the ACE sample (Li / hot – pressed 

LPSCl / Li0.2NMC811) from 40 ℃ to 550 ℃ with a heating rate of 2 ℃ / min (Figure 

3.8a) with their respective cumulative heat flows (Figure 3.8b). It is observed that after 



 

25 

 

 

the lithium melting around 180 ℃, there is a broad exothermic peak from around 

200 ℃ to 350 ℃,  which is likely due to the reactions between oxygen from 

decomposed Li0.2NMC811 and the hot – pressed LPSCl. However, there is no 

exothermic peak around 400 ℃ to 500 ℃ in the PVDF breakdown region, which may 

indicate there are no species still present that are able to undergo exothermic reactions 

with the PVDF breakdown products.  

        
 

Figure 3.8: (a) Baseline – corrected DSC result of the ACE sample with hot – pressed 

LPSCl shard. (b) Cumulative heat flow of the ACE sample. 



 

26 

 

 

         3.2.2. ACE Sample (NMC811 + Cold – pressed LPSCl + Li metal) 

The hot – pressed LPSCl shard was replaced with cold – pressed LPSCl shards in the 

ACE sample to understand the effect of the heat treatment of the LPSCl pellet on the 

heat generation of the ACE sample. Two repeated DSC runs were performed. From 

Figure 3.9a, there is an exothermic peak immediately after lithium melting to around 

400 ℃, compared to the hot – pressed LPSCl ACE sample, leading to similar heat 

generation.  Figure 3.9b shows the cumulative heat flows of the ACE samples.  

                 
 

Figure 3.9: (a). Baseline – corrected DSC result of the ACE samples with cold – 

pressed LPSCl shard. (b). Cumulative heat flow of the ACE samples.       



 

27 

 

 

3.3. Effect of Liquid Electrolyte on the Heat Generation 

Liquid electrolyte (1M LiPF6 (EC : DEC 1:1 v/v)) was added to the Li / hot – pressed 

LPSCl / NMC811 ACE sample to further understand its effect on the thermal runaway 

mechanism. Here, various amounts of the liquid electrolyte were added to the cathode 

active material in the closed DSC pan. Figure 3.10 shows the DSC result of the ACE 

samples with 0.1 µL, 0.5 µL, and 1 µL of liquid electrolyte, respectively. It is observed 

that there is an increase in the exothermic peak of the DSC result around 100 ℃ as the 

liquid electrolyte introduced increased. Also, there are exothermic peaks from around 

180 ℃ to 350 ℃, which may be attributed to the reaction of the decomposed electrolyte 

with the decomposed product of the Li0.2NMC811 cathode sheets and LPSCl shard. At 

around 450 ℃, there is an exothermic peak, which is likely due to the PVDF breakdown 

and reaction with other decomposed products, which is not evident in the ACE samples 

without liquid electrolyte.  

     

Figure 3.10: Baseline – corrected DSC result of the ACE sample with varying amounts 

of liquid electrolyte.  



 

28 

 

 

To further understand the reaction mechanism, the DSC tests of the ACE samples 

(without and with 0.5 µL liquid electrolyte) were performed up to 350 ℃ with a lithium 

remelt test from 170 ℃  to 190 ℃. After the lithium remelt test, it is observed that the 

lithium has completely reacted with the other components at 350 ℃, for the ACE 

sample without liquid electrolyte (Figure 3.11a). However, for the ACE sample with 

0.5 µL liquid electrolyte, only about 70% of the lithium metal has been consumed 

(Figure 3.11b).  

              
 

Figure 3.11: Raw DSC result of the ACE sample (a) without and (b) with 0.5 µL liquid 

electrolyte. 



 

29 

 

 

Based on the above results, it is observed that the addition of the liquid electrolyte 

lowers the onset temperature and shifts the heat release to higher temperatures, while 

the total heat released remains in a similar range. Compared to a Li – ion material set, 

the onset temperature is higher for the solid-state material sets, but the total heat release 

is similar, as shown in Table 1.   

For a 350 
𝑊ℎ

𝑘𝑔
 cell with 3.8 V,  we have 92 Ah per 1000 gcell 

 

Assume 0.2 
𝐴ℎ

𝑔𝑁𝑀𝐶811
 = 330 𝑔𝑁𝑀𝐶811 per 1000 𝑔𝑐𝑒𝑙𝑙. 

 

For a heat generation of 4000 
𝐽

𝑔𝑁𝑀𝐶811
 ,  

 

Therefore 4 
𝑀𝐽

𝑔𝑁𝑀𝐶811
 × 

330 𝑔𝑁𝑀𝐶811

1000 𝑔𝑐𝑒𝑙𝑙
  = 1320 

𝐽

𝑔𝑐𝑒𝑙𝑙
  

 

Assuming the heat capacity is 1 
𝐽

𝑔𝑐𝑒𝑙𝑙𝐾
,  

 

4000 
𝐽

𝑔𝑁𝑀𝐶811
 = 1320 

𝐽

𝑔𝑐𝑒𝑙𝑙
 / 1 

𝐽

𝑔𝑐𝑒𝑙𝑙𝐾
 = 1320 ℃ 

 

The table below shows the comparison of the onset temperature, heat release, and 

adiabatic temperature rise of the tested ACE samples with and without liquid 

electrolyte. For the ACE samples with varying amount of liquid electrolyte, it is 

observed that the onset temperature of the ACE samples is lowered from ~ 190 ℃ to ~ 

80 ℃ as the amount of added liquid electrolyte increases from 0.1 µL to 1 µL. 

However, the total heat released, and the adiabatic temperature rise are in the same 

range. For the ACE samples with hot – pressed and cold – pressed LPSCl, their onset 

temperatures remained the same. Moreover, their heat generation and adiabatic 

temperature rise are within the same range. Based on the results, it can be deduced that 

there is no obvious trend in how the amount of liquid electrolyte affects the total heat 



 

30 

 

 

release. The only effect of the liquid electrolyte appears to be a shift in heat release to 

higher temperatures. The significant variability in the heat flow results may indicate 

that the complexity of the ACE samples, including a large number of components and 

the importance of the lithium metal passivation, results in inherently stochastic reaction 

paths.  

 Table 1: Comparison of the onset temperature, heat release, and adiabatic temperature 

rise of different battery chemistries. 

 

 

 

 

 

 

 

 

 
 

 

Battery Chemistries Onset 

Temperature 

of Exotherm 

(°C) 

Total Heat 

(J/gNMC811) 

~50 °C to ~ 550 °C 

Adiabatic 

temperature 

rise (°C) 

Li / LPSCl / NMC811 

(hot pressed LPSC) 
~ 190 ℃ 4000 – 5300 1300 – 1750 

Li / LPSCl / NMC811 

(cold pressed LPSC) 
~ 190 ℃ 4600 – 5300 1500 – 1750 

Li / LPSCl / NMC811 +  

1 µL LE 
~ 80 ℃ 3500 – 5000 1150 – 1650 

Li / LPSCl / NMC811 + 

0.5 µL LE 
~ 165 ℃ 2600 – 5400 860 – 1800 

Li / LPSCl / NMC811 + 

0.1 µL LE 
~ 190 ℃ 3400 – 4300 1100 – 1400 



 

31 

 

 

Chapter 4: Conclusion 

4.1. Results Summary 

The thermal runaway and quantitative heat generation in solid – state battery, especially 

the ACE (Li / hot – pressed LPSCl / NMC811) samples were investigated using the 

differential scanning calorimetry (DSC) tests. The DSC result of LPSCl alone showed 

no obvious exothermic or endothermic peak up to 500 ℃ highlighting its excellent 

thermal stability. However, when incorporated in an ACE sample, it reacts with oxygen 

from the decomposition Li0.2NMC811, resulting in a large exotherm around 200 ℃ to 

350 ℃, giving a heat generation of around 4000 - 5300 J/g NMC811. When compared to 

an ACE sample with cold – pressed LPSCl (4600 – 5300 J/g NMC811), they are within 

the same range.  

To further understand the effect of liquid electrolyte on the thermal runaway of all – 

solid – state cells, small amounts of liquid electrolyte were added to the ACE samples. 

It is observed that the addition of the liquid electrolyte lowered the onset temperature 

and shifted the heat release to higher temperatures. However, the total heat generation 

remained within a similar range. Compared to a Li – ion material set, the onset 

temperature is higher for the solid-state material sets, but the total heat release is 

similar. This analysis highlights that while solid-state batteries offer enhanced 

performance due to their higher energy densities, they present significant safety 

concerns. As a result, they may not be as safe as traditional lithium – ion batteries, as 

their heat generation is higher. These findings are crucial for understanding the 

temperature rise, heat generation, and safety of solid – state batteries.  



 

32 

 

 

4.2. Limitations to Current Work 

This research presents some limitations that impact the comprehensiveness of the  

result findings. First, the reliance on Differential Scanning Calorimetry (DSC) only as 

the primary analytical tool limits the scope of data collection. While DSC is highly 

effective at measuring thermal transitions, it may not capture other critical factors, such 

as electrochemical reactions that also contribute to the thermal runaway of solid – state 

batteries. The results from the DSC measurements may not fully translate to large – 

format cells where external factors also influence the thermal behavior and safety. In a 

commercial cell, the cell layers are designed with optimized electrode and electrolyte 

configurations to facilitate efficient ion transport, enhancing electrochemical 

performance. Meanwhile, in a DSC pan, the layered configuration is not engineered for 

electrochemical processes, resulting in less efficient transport between layers. Also, the 

sensitivity of LPSCl to atmospheric exposure makes accurate post-calorimetry analysis 

(such as X-ray Photoelectron Spectroscopy (XPS) and X-ray Diffraction (XRD)) 

challenging.  

Another significant constraint in this study is the inability of our Differential Scanning 

Calorimetry (DSC) setup to perform gas chromatography-mass spectrometry (GC – 

MS) analysis of gases evolved during thermal decomposition. The DSC pans used in 

this research are hermetically sealed to prevent material loss during heating. While this 

ensures accurate thermal measurements, it limits the analysis of volatile gases released 

during thermal runaway, which are crucial for understanding the full decomposition 

reactions. Lastly, the laboratory environment used for this research differs from real – 



 

33 

 

 

world operating conditions. The controlled laboratory setting does not account for 

external factors which could significantly affect battery safety during operation in 

electric vehicles or grid – scale energy storage systems.  

 

4.3. Recommendations and Future Work 

To address the limitations identified in this work, the following recommendations and 

future work are highlighted. Further post – calorimetry characterization techniques, 

such as in-situ XPS or XRD, should be employed to monitor real-time reactions within 

the battery, thus providing a clearer understanding of the degradation processes. 

To enhance the understanding of thermal runaway and reaction mechanisms, future 

research should incorporate Gas Chromatography-Mass Spectrometry (GC – MS) 

analysis with Differential Scanning Calorimetry (DSC). This would require a 

modification of the DSC setup, allowing for the collection and analysis of evolved 

gases during thermal decomposition. By coupling GC – MS with DSC, the study could 

identify critical gas – phase reactions that contribute to thermal runaway.  

Future research could also investigate the use of silicon anodes as a potential 

replacement for lithium metal in studying the thermal runaway behavior of ACE 

samples. Given the growing interest in silicon anodes for next – generation energy 

SSBs, incorporating them into future thermal runaway studies would be a logical step 

toward developing safer, high – performance solid – state batteries.  

 



 

34 

 

 

Appendices 
 

 

A.1: Sample MATLAB Algorithms for fitting and integrating raw DSC heat flow 

data. 

 

close all 
clear  
% ACE sample with 1uL LE 
 
MC_LE1 = readmatrix ('C:\Users\dapot\OneDrive\Documents\UMD PhD\Paul 
Albertus Group\joined\LPSC\2024\June 2024\MC with LE 1uL\2024-04-12 Full 
Microcell with 1uL LE.csv'); 
MC_LE2 = readmatrix ('C:\Users\dapot\OneDrive\Documents\UMD PhD\Paul 
Albertus Group\joined\LPSC\2024\June 2024\MC with LE 1uL\2024-05-22 Full 
Microcell 1 w LE 1uL DSC2A-02873.csv'); 
MC_LE3 = readmatrix ('C:\Users\dapot\OneDrive\Documents\UMD PhD\Paul 
Albertus Group\joined\LPSC\2024\June 2024\MC with LE 1uL\2024-05-22 Full 
Microcell 2 w LE 1uL DSC2A-02873.csv'); 
 
total_mass = [14.972; 16.792; 16.229]; 
MC_LE1 (:,3) = MC_LE1 (:, 3).*(total_mass (1)); %mW 
MC_LE2 (:,3) = MC_LE2 (:, 3).*(total_mass (2)); %mW 
MC_LE3 (:,3) = MC_LE3 (:, 3).*(total_mass(3)); %mW 
 
figure (1) 
plot (MC_LE1(17:153016,2), MC_LE1(17:153016,3)); 
a(1) = 3402; a(2) = 85631; a(3) = 152996; 
nmc_f = polyfit ([MC_LE1(a(1),2); MC_LE1(a(2),2); 
MC_LE1(a(3),2)],[MC_LE1(a(1),3); MC_LE1(a(2),3); MC_LE1(a(3),3)],2); 
nmc_f1 = polyval(nmc_f,MC_LE1(17:153016,2)); 
nmc_f2 = MC_LE1(17:153016,3) - nmc_f1; 
 
figure (2) 
plot (MC_LE2(18:153016,2), MC_LE2(18:153016,3)); 
b(1) = 4467; b(2) = 33241; b(3) = 129799; b(4) = 152920;  
nmc_fb = polyfit ([MC_LE2(b(1),2); MC_LE2(b(2),2); MC_LE2(b(3),2); 
MC_LE2(b(4),2)],[MC_LE2(b(1),3); MC_LE2(b(2),3); MC_LE2(b(3),3); 
MC_LE2(b(4),3)],2); 
nmc_fb1 = polyval(nmc_fb,MC_LE2(18:153016,2)); 
nmc_fb2 = MC_LE2(18:153016,3) - nmc_fb1; 
 
figure (3) 
plot (MC_LE3(18:153017,2),MC_LE3(18:153017,3)); 
c(1) = 3487; c(2) = 26264; c(3) = 51229; c(4) = 86712; c(5) = 152298;  
nmc_fc = polyfit ([MC_LE3(c(1),2); MC_LE3(c(2),2); MC_LE3(c(3),2); 
MC_LE3(c(4),2); MC_LE3(c(5),2)],[MC_LE3(c(1),3); MC_LE3(c(2),3); 
MC_LE3(c(3),3); MC_LE3(c(4),3); MC_LE3(c(5),3)],4); 
nmc_fc1 = polyval(nmc_fc,MC_LE3(18:153017,2)); 
nmc_fc2 = MC_LE3 (18:153017,3) - nmc_fc1; 
 



 

35 

 

 

figure (1) 
plot (MC_LE1(17:153016,2),nmc_f2); 
a(6) = 3565; a(7) = 43066; a(8) = 85631; a(9)= 152996;  
Area_A = trapz(MC_LE1 (a(6): a(9),2), nmc_f2 (a(6): a(9),1) - nmc_f2 
(a(6),1)).*(30/(4.011*0.7)); 
Area1a = trapz(MC_LE1 (a(6): a(7),2), nmc_f2 (a(6): a(7),1) - nmc_f2 
(a(6),1)).*(30/(4.011*0.7)); 
Area1b = trapz(MC_LE1 (a(7): a(8),2), nmc_f2 (a(7): a(8),1) - nmc_f2 
(a(6),1)).*(30/(4.011*0.7)); 
Area1c = trapz(MC_LE1 (a(8): a(9),2), nmc_f2 (a(8): a(9),1) - nmc_f2 
(a(6),1)).*(30/(4.011*0.7)); 
 
figure (2) 
plot (MC_LE2(18:153016,2),nmc_fb2); 
b(6) = 7758; b(7) = 42815; b(8)= 129799; b(9) = 152995;  
Area_B = trapz(MC_LE2 (b(6): b(9),2), nmc_fb2 (b(6): b(9),1) - nmc_fb2 
(b(6),1)).*(30/(3.912*0.7)); 
Area2a = trapz(MC_LE2 (b(6): b(7),2), nmc_fb2 (b(6): b(7),1) - nmc_fb2 
(b(6),1)).*(30/(3.912*0.7)); 
Area2b = trapz(MC_LE2 (b(7): b(8),2), nmc_fb2 (b(7): b(8),1) - nmc_fb2 
(b(6),1)).*(30/(3.912*0.7)); 
Area2c = trapz(MC_LE2 (b(8): b(9),2), nmc_fb2 (b(8): b(9),1) - nmc_fb2 
(b(6),1)).*(30/(3.912*0.7)); 
 
figure (3) 
plot (MC_LE3(18:153017,2), nmc_fc2); 
c(6) = 6088; c(7) = 42839; c(8) = 85371; c(9) = 152146; 
Area_C = trapz (MC_LE3 (c(6): c(9),2), nmc_fc2 (c(6): c(9),1) - nmc_fc2 
(c(6),1)).*(30/(3.858*0.7)); 
Area3a = trapz (MC_LE3 (c(6): c(7),2), nmc_fc2 (c(6): c(7),1) - nmc_fc2 
(c(6),1)).*(30/(3.858*0.7)); 
Area3b = trapz (MC_LE3 (c(7): c(8),2), nmc_fc2 (c(7): c(8),1) - nmc_fc2 
(c(6),1)).*(30/(3.858*0.7)); 
Area3c = trapz (MC_LE3 (c(8): c(9),2), nmc_fc2 (c(8): c(9),1) - nmc_fc2 
(c(6),1)).*(30/(3.858*0.7)); 
 
figure (1) 
plot (MC_LE1(17:153000,2), (nmc_f2 (17:153000,1) - nmc_f2 
(a(2),1)) ./(4.011*0.7), 'b', 'linewidth',1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Heat flow W/g_{(NMC811)}', 'fontsize', 20); 
hold on 
area (MC_LE1 (a(6): a(7),2), (nmc_f2 (a(6): a(7),1) - nmc_f2 
(a(6),1))./(4.011*0.7));  
area (MC_LE1 (a(7): a(8),2), (nmc_f2 (a(7): a(8),1) - nmc_f2 
(a(6),1))./(4.011*0.7));  
area (MC_LE1 (a(8): a(9),2), (nmc_f2 (a(8): a(9),1) - nmc_f2 
(a(6),1))./(4.011*0.7));  
ax=gca; 
ax.FontSize=30; 
ylim ([-0.5, 4]); 
xticks(0:50:550); 
legend ('Full Microcell 1 + 1 µL LE (2℃/min)'); 



 

36 

 

 

% Convert heat flow (mW) to J/g by integrating and dividing by heating rate 
(assuming 2°C/min) 
heating_rate = 2 / 60; % °C/s 
cum_energyA = cumtrapz(MC_LE1(17:153016, 2), nmc_f2) / (4.011 * 0.7 * 
heating_rate); 
% Plot the cumulative energy vs temperature 
figure (2) 
plot(MC_LE1(17:153016, 2), cum_energyA, 'b', 'linewidth', 1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Cummulative Heat flow J/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
yticks(0:500:6000); 
xticks(0:50:550); 
legend ('Cumulative Heat Flow J/g for MC1 + 1 µL LE'); 
 
 
figure (3) 
plot (MC_LE2(18:152999,2), (nmc_fb2 (18:152999,1) - nmc_fb2 
(b(2),1)) ./(3.912*0.7), 'r', 'linewidth',1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Heat flow W/g_{(NMC811)}', 'fontsize', 20); 
hold on 
area (MC_LE2 (b(6): b(7),2), (nmc_fb2 (b(6): b(7),1) - nmc_fb2 
(b(6),1))./(3.912*0.7));  
area (MC_LE2 (b(7): b(8),2), (nmc_fb2 (b(7): b(8),1) - nmc_fb2 
(b(6),1))./(3.912*0.7));  
area (MC_LE2 (b(8): b(9),2), (nmc_fb2 (b(8): b(9),1) - nmc_fb2 
(b(6),1))./(3.912*0.7));   
ax=gca; 
ax.FontSize=30; 
ylim ([-0.5, 4]); 
xticks(0:50:550); 
legend ('Full Microcell 2 + 1 µL LE (2℃/min)'); 
% Convert heat flow (mW) to J/g by integrating and dividing by heating rate 
(assuming 2°C/min) 
heating_rate = 2 / 60; % °C/s 
cum_energyB = cumtrapz(MC_LE2(18:153016, 2), nmc_fb2) / (3.912 * 0.7 * 
heating_rate); 
% Plot the cumulative energy vs temperature 
figure (4) 
plot(MC_LE2(18:153016, 2), cum_energyB, 'r', 'linewidth', 1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Cummulative Heat flow J/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
yticks(0:500:6000); 
xticks(0:50:550); 
legend ('Cumulative Heat Flow J/g for MC2 + 1 µL LE'); 
 
 
figure (5) 
plot (MC_LE3(18:152998,2), (nmc_fc2 (18:152998,1) - nmc_fc2 
(c(2),1)) ./(3.858*0.7),'k', 'linewidth',1); 



 

37 

 

 

hold on 
area (MC_LE3(c(6): c(7),2), (nmc_fc2 (c(6): c(7),1) - nmc_fc2 
(c(6),1))./(3.858*0.7));  
area (MC_LE3(c(7): c(8),2), (nmc_fc2 (c(7): c(8),1) - nmc_fc2 
(c(6),1))./(3.858*0.7));  
area (MC_LE3(c(8): c(9),2), (nmc_fc2 (c(8): c(9),1) - nmc_fc2 
(c(6),1))./(3.858*0.7));  
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Heat flow W/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
ylim ([-0.5, 4]); 
xticks(0:50:550); 
legend ('Full Microcell 3 + 1 µL LE (2℃/min)'); 
% Convert heat flow (mW) to J/g by integrating and dividing by heating rate 
(assuming 2°C/min) 
heating_rate = 2 / 60; % °C/s 
cumm_energyC = cumtrapz(MC_LE3(18:153017, 2), nmc_fc2) / (3.858 * 0.7 * 
heating_rate); 
% Plot the cumulative energy vs temperature 
figure (6) 
plot(MC_LE3(18:153017, 2), cumm_energyC, 'k', 'linewidth', 1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Heat flow J/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
yticks(0:500:6000); 
xticks(0:50:550); 
legend ('Cumulative Heat Flow J/g for MC3 + 1 µL LE'); 
 
figure (7) 
plot(MC_LE1(17:153016, 2), cum_energyA, 'b', 'linewidth', 1); 
hold on 
plot(MC_LE2(18:153016, 2), cum_energyB, 'r', 'linewidth', 1); 
hold on 
plot(MC_LE3(18:153017, 2), cumm_energyC, 'k', 'linewidth', 1); 
xlabel ('Temperature (℃)', 'fontsize', 20); 
ylabel ('Cummulative Heat flow J/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
yticks(0:1000:7000); 
xticks(0:50:550); 
legend ('MC1 + 1 µL LE', 'MC2 + 1 µL LE', 'MC3 + 1 µL LE'); 
 
figure (8) 
plot (MC_LE1(17:153000,2), (nmc_f2 (17:153000,1) - nmc_f2 
(a(2),1)) ./(4.011*0.7), 'b', 'linewidth',2); 
hold on 
plot (MC_LE2(18:152999,2), (nmc_fb2 (18:152999,1) - nmc_fb2 
(b(2),1)) ./(3.912*0.7), 'r', 'linewidth',2); 
hold on 
plot (MC_LE3(18:152998,2), (nmc_fc2 (18:152998,1) - nmc_fc2 
(c(2),1)) ./(3.858*0.7),'k', 'linewidth',2); 
xlabel ('Temperature (℃)', 'fontsize', 20); 



 

38 

 

 

ylabel ('Normalized Heat flow W/g_{(NMC811)}', 'fontsize', 20); 
ax=gca; 
ax.FontSize=30; 
ylim ([-1 5]); 
xticks(0:50:550); 
legend ('MC1 + 1 µL LE', 'MC2 + 1 µL LE', 'MC3 + 1 µL LE'); 
 
 
 
 
 
 
 
 
 

A.2: DSC heat flow of LPSCl shard. Reproduced from 24.  
 

 

  



 

39 

 

 

A.3: DSC heat flow of 1M LiPF6 in EC:DEC (1:1 v/v). Reproduced from Roth et 

al.25 
 

 

 

 

A.4: DSC heat flow of 1M LiPF6 in EC:DEC (1:1 v/v). Reproduced from Wang et 

al.26 

 

 

 

 

 

 



 

40 

 

 

A.5: Calculations to determine the maximum volume of liquid electrolyte used in 

ACE samples.  

From the DSC pan specifications, the internal volume is 20 µL (0.02 cm3) and the 

maximum operating pressure is 217 bar. 

The mass of lithium used is ~ 0.2 mg (0.0002 g), and the density of lithium is 0.534 g 

/ cm3
. Volume of lithium used = Mass of lithium / Density of lithium =  0.0002 g / 

0.534 g/ cm3 = 0.0003745 cm3. 

The mass of LPSCl used is ~ 11 mg (0.011 g) and its density is 1.64 g cm-3. Therefore, 

volume of LPSCl used = Mass of LPSCl / Density of LPSCl =  0.011 g / 1.64 g cm-3 = 

0.0067073 cm3.  

For the NMC811 used, the diameter is 4 mm (0.4 cm). The area of the NMC811 sheet 

used is 0.12568 cm3. The thickness of the current collector is 16 µm and the coating is 

63 µm, total thickness is ~ 80 µm (0.008 cm). 

Volume = Area × Thickness = 0.12568 cm3 × 0.008 cm =  0.0010054 cm3.  Since two 

sheets were used, the total volume of NMC811 used is 0.0020108 cm3.  

The total volume of the ACE sample in the DSC pan = Volume of lithium + Volume 

of LPSCl + Volume of NMC811 = 0.009093 cm3. 

The remaining volume = 0.02 cm3 – 0.009093 cm3  = 0.010907 cm3. 

To avoid reaching the maximum operating pressure of the DSC pan considering the 

oxygen release from NMC811 which reacts with other components to give other 

gaseous products, a conservative approach is to use 20% of the remaining volume for 



 

41 

 

 

the liquid electrolyte. Therefore, the maximum volume of 1M LiPF6 (EC / DEC) that 

can be added to the DSC pan considering oxygen release and pressure constraint = 0.2 

× Vremaining = 0.2 × 0.010907 cm3 = 0.0021814 cm3  = ~ 2 µL.  

 

 



 

42 

 

 

Bibliography 
 

 

1. Kim, T. H.;  Park, J. S.;  Chang, S. K.;  Choi, S.;  Ryu, J. H.; Song, H. K., The 

Current Move of Lithium-Ion Batteries Towards the Next Phase. Advanced Energy 

Materials 2012, 2 (7), 860-872. 

2. Daniel, C.;  Mohanty, D.;  Li, J.; Wood, D. L. In Cathode materials review, 

2014; AIP Publishing LLC. 

3. Huang, Y.; Li, J., Key Challenges for Grid‐Scale Lithium‐Ion Battery Energy 

Storage. Advanced Energy Materials 2022, 12 (48), 2202197. 

4. Mauger, A.; Julien, C. M., Critical review on lithium-ion batteries: are they 

safe? Sustainable? Ionics 2017, 23 (8), 1933-1947. 

5. Bates, A. M.;  Preger, Y.;  Torres-Castro, L.;  Harrison, K. L.;  Harris, S. J.; 

Hewson, J., Are solid-state batteries safer than lithium-ion batteries? Joule 2022, 6 (4), 

742-755. 

6. Märker, K.;  Reeves, P. J.;  Xu, C.;  Griffith, K. J.; Grey, C. P., Evolution of 

Structure and Lithium Dynamics in LiNi0.8Mn0.1Co0.1O2 (NMC811) Cathodes during 

Electrochemical Cycling. Chemistry of Materials 2019, 31 (7), 2545-2554. 

7. Bak, S.-M.;  Hu, E.;  Zhou, Y.;  Yu, X.;  Senanayake, S. D.;  Cho, S.-J.;  Kim, 

K.-B.;  Chung, K. Y.;  Yang, X.-Q.; Nam, K.-W., Structural Changes and Thermal 

Stability of Charged LiNixMnyCozO2 Cathode Materials Studied by Combined In Situ 



 

43 

 

 

Time-Resolved XRD and Mass Spectroscopy. ACS Applied Materials & Interfaces 

2014, 6 (24), 22594-22601. 

8. Noh, H.-J.;  Youn, S.;  Yoon, C. S.; Sun, Y.-K., Comparison of the structural 

and electrochemical properties of layered Li[NixCoyMnz]O2 (x = 1/3, 0.5, 0.6, 0.7, 0.8 

and 0.85) cathode material for lithium-ion batteries. Journal of Power Sources 2013, 

233, 121-130. 

9. Shadike, Z.;  Chen, Y.;  Hu, E.;  Zhang, J.; Yang, X.-Q., Interphasial 

engineering for Ni-rich NMC cathode materials. Trends in Chemistry 2023, 5 (10), 

775-787. 

10. Akhilash, M.;  Salini, P. S.;  John, B.;  Supriya, N.;  Sujatha, S.; Mercy, T. D., 

Thermal stability as well as electrochemical performance of Li-rich and Ni-rich cathode 

materials—a comparative study. Ionics 2023, 29 (3), 983-992. 

11. Yu, C.;  Zhao, F.;  Luo, J.;  Zhang, L.; Sun, X., Recent development of lithium 

argyrodite solid-state electrolytes for solid-state batteries: Synthesis, structure, stability 

and dynamics. Nano Energy 2021, 83, 105858. 

12. Chen, Y.-T.;  Marple, M. A. T.;  Tan, D. H. S.;  Ham, S.-Y.;  Sayahpour, B.;  

Li, W.-K.;  Yang, H.;  Lee, J. B.;  Hah, H. J.;  Wu, E. A.;  Doux, J.-M.;  Jang, J.;  Ridley, 

P.;  Cronk, A.;  Deysher, G.;  Chen, Z.; Meng, Y. S., Investigating dry room 

compatibility of sulfide solid-state electrolytes for scalable manufacturing. Journal of 

Materials Chemistry A 2022, 10 (13), 7155-7164. 



 

44 

 

 

13. Zhou, L.;  Minafra, N.;  Zeier, W. G.; Nazar, L. F., Innovative Approaches to 

Li-Argyrodite Solid Electrolytes for All-Solid-State Lithium Batteries. Accounts of 

Chemical Research 2021, 54 (12), 2717-2728. 

14. Bai, X.;  Duan, Y.;  Zhuang, W.;  Yang, R.; Wang, J., Research progress in Li-

argyrodite-based solid-state electrolytes. Journal of Materials Chemistry A 2020, 8 

(48), 25663-25686. 

15. Roy, K.;  Banerjee, A.; Ogale, S., Search for New Anode Materials for High 

Performance Li-Ion Batteries. ACS Applied Materials & Interfaces 2022, 14 (18), 

20326-20348. 

16. Lin, D.;  Liu, Y.; Cui, Y., Reviving the lithium metal anode for high-energy 

batteries. Nature Nanotechnology 2017, 12 (3), 194-206. 

17. Cheng, X.-B.;  Zhao, C.-Z.;  Yao, Y.-X.;  Liu, H.; Zhang, Q., Recent Advances 

in Energy Chemistry between Solid-State Electrolyte and Safe Lithium-Metal Anodes. 

Chem 2019, 5 (1), 74-96. 

18. Liang, B.;  Liu, Y.; Xu, Y., Silicon-based materials as high-capacity anodes for 

next generation lithium ion batteries. Journal of Power Sources 2014, 267, 469-490. 

19. Ko, M.;  Chae, S.; Cho, J., Challenges in Accommodating Volume Change of 

Si Anodes for Li‐Ion Batteries. ChemElectroChem 2015, 2 (11), 1645-1651. 



 

45 

 

 

20. Shi, J.;  Jiang, K.;  Fan, Y.;  Zhao, L.;  Cheng, Z.;  Yu, P.;  Peng, J.; Wan, M., 

Advancing Metallic Lithium Anodes: A Review of Interface Design, Electrolyte 

Innovation, and Performance Enhancement Strategies. Molecules 2024, 29 (15), 3624. 

21. Cooper, A.; Johnson, C. M., Differential Scanning Calorimetry. Humana Press: 

pp 125-136. 

22. Steinmann, W.;  Walter, S.;  Beckers, M.;  Seide, G.; Gries, T., Thermal 

Analysis of Phase Transitions and Crystallization in Polymeric Fibers. In Applications 

of Calorimetry in a Wide Context, Amal Ali, E., Ed. IntechOpen: Rijeka, 2013; p Ch. 

12. 

23. Yang, S.-J.;  Hu, J.-K.;  Jiang, F.-N.;  Cheng, X.-B.;  Sun, S.;  Hsu, H.-J.;  Ren, 

D.;  Zhao, C.-Z.;  Yuan, H.;  Ouyang, M.;  Fan, L.-Z.;  Huang, J.-Q.; Zhang, Q., 

Oxygen-induced thermal runaway mechanisms of Ah-level solid-state lithium metal 

pouch cells. eTransportation 2023, 18, 100279. 

24. Rui, X.;  Ren, D.;  Liu, X.;  Wang, X.;  Wang, K.;  Lu, Y.;  Li, L.;  Wang, P.;  

Zhu, G.;  Mao, Y.;  Feng, X.;  Lu, L.;  Wang, H.; Ouyang, M., Distinct thermal runaway 

mechanisms of sulfide-based all-solid-state batteries. Energy & Environ. Sci. 2023,  

(16), 3552-3563. 

25. Roth, E. P.;  Doughty, D. H.; Franklin, J., DSC investigation of exothermic 

reactions occurring at elevated temperatures in lithium-ion anodes containing PVDF-

based binders. Journal of Power Sources 2004, 134 (2), 222-234. 



 

46 

 

 

26. Wang, Q.;  Sun, J.;  Yao, X.; Chen, C., Thermal stability of LiPF6/EC+DEC 

electrolyte with charged electrodes for lithium ion batteries. Thermochimica Acta 2005, 

437 (1), 12-16.