Principle of operation
A rechargeable battery cell is composed by two electrodes; each contains an active material and a current collector enabling the connection to the external electrical circuit. These active materials are those which are involved in the oxidation-reduction (redox) reactions which is the origin of the operation of any battery. These batteries therefore differ according to the technology used.
The electrodes are immersed in an electrolyte which allows the transport of charges in ionic form, necessary for the redox reactions to take place. The separator acts as an electrical insulation for both electrodes. It must prevent electron transfer internally but at the same time allows ionic transfer.
This basic assembly can be package in different ways depending on the shape of the battery. In a very general way, the electrochemical chain can be written as: (−)[ ]/[ ] // // [ ]/[ ](+) with: : active positive material, operating at high potential from +3 to +5V vs. Li+/Li : active negative material, operating at low potential below + 2 vs. Li+/Li
The operation of a battery cell consists of two redox half-reactions. The main mobile species in a Li-ion battery is the + cation and when the half-reaction at either electrode takes place, this cation is inserted directly into or extracted therefrom the crystal structure of the active material of the electrodes. The chemical half-reactions which take place at the active material of the negative electrode, and at the active material of the positive electrode, can be written as follows : ℎ With x, the number of + cations transported during charging or discharging.
The overall reaction of the electrochemical system is given by the equation: Overall chemical reaction of the system + →← + (1.3) ℎ
In a Li-ion battery, there are two charge carriers: the + cations in the electrolyte which was presented earlier and also the electrons in the active materials of both electrodes. These two types of charge carriers are called mobile species . As shown in Fig. 1.2, these two mobile species move at an opposite direction from each other in both cases; charging and discharging.
There are a few types of negative electrode: those with lithium metal oxide and those based on carbon or other compound as shown in Table 1.2. They have their own advantages and inconvenient.
However, the most common negative electrode is graphite because of its cost and abundancy. The lithium titanium dioxide (LTO) electrode for example is less common but nevertheless gains more interest these past years. The most important problem that the negative electrode faces is the problem of dendrite. This problem occurs during charging process when the deposition of lithium ions form a non-uniform layers. The dendrites could grow by time and puncture through the separator which then would create a short circuit.
However, carbon-based electrode is much less sensitive to the problem of dendrites to be used as lithium intercalation materials. It has a low and flat working voltage profile at about -2.7V/ESH which makes it an excellent negative electrode for lithium battery . Moreover, since carbon is used under its crystalline form (graphite), it has a high capacity and stability to receive the Li+ ions reversibly without altering its properties . Graphite is an insertion material in which Li-ions come to lodge into. The structure of graphite allows a high reversibility because it is constituted of multi-layered plane of graphene where the lithium ions penetrate right through and form a lithium/carbon intercalation compound LixC6 (Fig. 1.3).
For lithium titanium dioxide (LTO) electrode, even though it is more prone to have the problem of dendrite, its high chemical and thermal stability makes it a really good alternative towards graphite materials. Moreover, this compound is also really promising for the EV and HEV due to its capability for fast charging. Indeed, for this material, the insertion of lithium is made at an electrochemical potential superior to the one of metallic lithium. (~1.55V vs. Li+/Li). Thus, it can accepts high currents of charge without risk of lithium metal forming at the interface, contrary in the case of graphite.
There are many types of technologies for the active materials of the positive electrode. Table 1.3 shows some major characteristics of the materials used. As usual, they have their advantages and also their inconvenient.
However, only two materials for the positive electrode are considered for this study. First, it is the lithium iron phosphate (LFP) and secondly the nickel-manganese-cobalt (NMC).
For so long, lithium cobalt oxide, LiCoO2 (LCO) which is the first used and still widely used materials as the positive electrode. Unfortunately, the high costs of the cobalt material due to its limited worldwide reserves, some other alternatives materials for positive electrode are developed. Moreover, LCO becomes unstable at high temperature and releasing oxygen which could reacts with the organic electrolyte solvent with high risk of inflammation or explosion. This material is also toxic. Thus, for that reasons, the NMC material is one of the materials that have been developed as a substitution to the lithium cobalt oxide. This material harbors with a more moderate cost, higher specific capacity and better thermal tolerance.
The general electrochemical reaction for NMC positive electrode is: ++ −+ 1⁄3 1⁄3 ⇋ 1⁄3 1⁄3 (1.6) 1− 1⁄3 5 1⁄3 5
For the past few years now, a new alternative of the lithiated metal oxide is introduced. The new positive electrode is constituted by phosphates of transition metals. In this study, the lithium iron phosphate LiFePO4 is the main battery technology that is used as the reference. This materials crystalline structure is isomorphic and allows the reversible insertion of lithium. They are very attractive materials because in spite of a nominal voltage lower than that of other materials, they are less dangerous in case of abuse conditions [14,15].
The general electrochemical reaction for LFP positive electrode is: ++ −+ ⇋ (1.7)
There are many types of electrolyte such as liquid electrolyte, solid electrolyte and gel electrolyte. Liquid electrolytes are the most common electrolytes used in the rechargeable battery domain. Practically, a Li-ion does not pass directly from one electrode to the other but is transferred from one solvent molecule to the next and reaches the other electrode . So, these electrolytes can only be used in a Li-ion battery if they contain lithium salt which is dissolved in an organic solvent. An example of the latter is the lithium salt LiPF6 which is dissolved in an organic electrolyte called ethylene-dimethyl-carbonate. However, this type of electrolyte presents two big inconvenient which are low conductivity and also strong inflammability. All of the battery that has been used in this study possesses this type of electrolyte.
For solid electrolyte, the lithium salts are incorporated into a polymer material and, after various manufacturing steps, a film is obtained. This film can also be used directly as a separator. Solid electrolytes have low conductivity at room temperature and often required heating (60°C to 80°C) for the battery to be operational. The first solid electrolytes have been developed for negative electrode cells in lithium metal, in order to mechanically restrain the growth of dendrites. They now have also been used for lithium-ion cells. However, a difficulty exists at the solid graphite/electrolyte interface. In fact, the contact is bad compared to the liquid electrolyte which can infiltrate inside the pores of the graphite.
The gel electrolytes are obtained by dissolving a solid membrane in a solvent. Compared to solid electrolytes, they provide better contact with the electrodes and a sufficient conductivity at room temperature. They have the advantage over liquid electrolytes of allowing cells to be made in a flexible package (pouch cells), but the contact resistance of the batteries remains higher than that of batteries with liquid electrolyte .
Separator and collector
The separator is an important part of the battery. It is a mono or multilayer microporous polyethylene and polypropylene sheet isolating both electrodes. It must therefore have these following properties :
• Good mechanical strength in order to avoid electrical contact between the electrodes
• low electronic conductivity
• A good wettability which allows the separator to soak with electrolyte and thus obtain a good ionic conductivity
• Chemical neutrality with respect to other elements of the battery, i.e. it must not react with the different chemical components of the battery.
Some separators play an active role in the safety of the cell by becoming impermeable to ions from a certain temperature, which blocks the chemical reaction in case of thermal runaway. These separators are called « shutdown separator »  and it should be noted that this phenomenon is irreversible.
The current collector is responsible for transferring the electrons from the active material of the electrode to the external circuit. Aluminum is a good candidate for this, very light and good electrical conductor. However, it reacts with Li-ions below 0.7V vs Li+/ Li. Copper, also a very good electrical conductor, is interesting but oxidizes above 3.2V vs Li+/Li.
The graphite has a potential of 0.3 V vs Li+/Li and most of the positive electrode materials presented above have, at the end of the charge, a voltage greater than 3.2 V vs Li + / Li. Hence the choice frequently encountered: aluminum collector for the positive electrode and copper for the negative electrode.
State of the art of fast charge
Fast charging for Li-ion battery is crucial for our everyday use of electronic devices and transportation. However, Li-ion batteries still have some serious drawbacks especially for fast charging protocols. Many of the problems can be related for one part to surface phenomena occurring on the negative and positive electrodes and for another part to structural modifications (expansion-contraction, crystal disorder), in simpler words, internal degradations of active materials. Moreover, safety protocol is also the main concern during the manipulation of this type of battery especially during high temperature elevation because of its high specific energies. In regards of fast charging process, the high temperature elevation is something that is unavoidable.
Thus, reliable indicators and reliable fast-charging protocols have to be developed and improved to properly charge the battery until a certain state of charge and monitor the state of health of the battery.
Standard charging method
Generally, the charging process of a Li-ion battery is divided into 2 charging stages; these are the constant current stage (CC) and constant voltage stage (CV). During the CC stage, the battery is charged at a chosen constant current (i.e. charging rate) until a certain upper voltage limit is reached before switching to CV stage. The upper voltage limit is predetermined by the manufacturer; it is designed to ensure longer battery life-span by avoiding side reactions. During the CV stage, the battery is normally charged more slowly with a degrading current to maintain the battery at a constant voltage until the current limit called cut-off, is attained. Even though the CV stage is slow, it allows the relaxation of the species concentrations inside the electrolyte and electrode materials. The duration of the total charging period depends on the charging rate applied during the CC stage.
Fast charging methods
In the context of EV applications, Li-ion batteries are faced with reliability and durability issues. Nevertheless, it is a mandatory requirement in EV applications to minimize the battery charging time; so, a fast-charging method must be developed properly. If fast-charging is applied, theoretically the battery can be recharged in a shorter period, which is not the case for current EVs.
Different methods are proposed in the literature for fast-charging. The fast-charging process for the Li-ion battery can be performed by increasing the C-rate of the CC stage [20, 21, 22].
Lopez and al.  observed that the total charging time is reduced quite significantly when the charging C-rate increases as shown in Fig. 1.4. For example, they save about 42% of charging time for a fast charge at 1.5C-rate compared to the nominal C-rate of C/2 to reach 90% of C-rate. Nevertheless, their results show that the energy efficiency of the charging process decreases greatly when the charging C-rate rises and their result mainly focus on the first stage of charging process which is the constant current stage.
On the other hand, Huang and al. improve their fast charging strategies by evaluating and characterizing Li-ion battery . Brief current of charging and discharging interrupts the charging process during constant current period to restrict the hysteresis effect. There’s also a fast charging strategy of implying a multistage charging process during the constant current (CC) period  and constant voltage (CV) period  to reduce the charging time as shown in Fig 1.5.
Figure 1.6: Schematic of fast charging process of different pulse sequences. (A) Constant amplitude pulsed current interspersed with identical rest periods; (B) constant amplitude pulsed current interspersed with alternating constant amplitude discharge pulses; and (C) pulsed current consisting of a sequence of different amplitudes charge current density pulses .
Other methods for fast-charging are also proposed in the literature based on pulsing current [24, 25, 26]. This latter method has shown good global performance, but an optimal configuration remains a challenge. For its part, the pulse charging is based on constant current steps of short duration followed by relaxation periods. The sequences can vary depending on the pulse amplitude and duration, as well as on the relaxation time as shown in Fig. 1.6. This method is highly recommended for fast-charging purpose as the current imposed is really high. Pauses are required to reduce or prevent the metallic lithium formation. The metallic lithium formation can greatly interfere with the charging process according to Purushothaman and al. . However, their studies only focus on the empirical calculation and simulation without any real experiment implementation. Finally, Ohmic Drop Compensation (ODC) method might also be used for fast-charging. As far as we know, this latter electrochemical method [27, 28], has never been used for Li-ion battery fast-charging. Nonetheless, there are some studies that are based on this method for developing fast-charging chargers [29, 30, 31, 32, 33]. For instance, Saint-Pierre  conducted his studies, which focused on the charger’s electronic circuitry rather than the impact of this method on the battery itself, which is in contrary, the main purpose of this paper. His study shows that with this method, the charging time is reduced by more than 30 minutes. Lin and al. [30, 31] also proposed a fast-charging charger with built-in resistance compensator (BRC) to achieve a fast and stable charging process on Li-ion battery packs.
Figure 1.7: Waveforms of the voltage with and without the BRC technique. The CC stage of the original design is extended to the CC’ stage with the BRC design .
Huang and al. and Peng and al. also conducted their study on a similar area to improve their fast-charging charger [32, 33]. Lin and al.’s system is able to charge the battery pack with adequate current and speed up the charging process time. They managed to shorten the charging time by using an ohmic-drop compensation method. This technique is applied to a battery pack and the compensated resistance consists of the external resistance of the battery pack as well as the connections. The compensation technique is performed for a very short delay as shown in Fig. 1.7. The results establish the improvement of the charging time by means of external resistance compensation technique of the battery pack.
Ohmic-drop compensation fast charging method
Our fast charging method, called ohmic-drop compensation (ODC) method, which consists in compensating the ohmic-drop of the battery voltage caused by the internal resistance of the battery cell by changing the upper-bound voltage limit at the end of CC stage. This section will explain on how this method functions.
Table of contents :
CHAPTER 1: LI-ION BATTERY; PRESENTATION, OPERATION, FAST-CHARGING AND AGEING
1. Battery technologies and application
2. Li-ion Batteries
2.1. Principle of operation
2.2. Negative Electrode
2.3. Positive Electrode
2.5. Separator and collector
3. State of the art of fast charge
3.1. Standard charging method
3.2. Fast charging methods
3.3. Ohmic-drop compensation fast charging method
4. Lithium-ion battery ageing
CHAPTER 2: EXPERIMENTAL SETUP, MATERIALS, EQUIPMENT AND BATTERY CHARACTERIZATIONS
2. Lithium battery
2.1. Lithium iron phosphate battery, C/LFP
2.2. Nickel manganese cobalt oxide battery, C/NMC
2.3. Lithium titanate oxide – lithium iron phosphate battery, LTO/LFP
3. Battery test bench, characterization and charging methods
3.1. Test bench
3.2. Charging and discharging protocols
3.3. Battery characterizations
3.3.1. Internal resistance
3.3.2. State Of Charge, SOC
3.3.3. State Of Health, SOH
3.3.4. Energy and coulombic efficiencies
4. The ventilation test bench: Conception of the air canal
4.1. Air canal characteristics
4.2. Air flow measurements
4.3. Determination of heat transfer coefficient by forced convection
5. Determination of heat capacity of C/LFP and the heat transfer coefficient inside climatic chamber
6. Post-mortem analysis methodology
6.1. X-ray tomography
6.2. SEM Analysis
6.3. Gaseous Chromatography analysis
6.4. Electrochemical characterization with coin cell
CHAPTER 3: FAST-CHARGING OF LI-ION BATTERIES
2. Fast charge process for C/LFP battery
2.1. Fast charge of the battery with normal method
2.2. Fast charge process of the C/LFP battery with ODC
2.2.1. High level ODC
2.2.2 Influence of the compensation rate
2.3. Temperature and thermal effect during fast charge of C/LFP Battery
2.3.1. Fast charge of C/LFP battery with normal method
2.3.2. Fast charge of C/LFP battery with ODC method
2.3.3. Ventilation system effect
3. Fast charge of C/NMC battery with ODC method
4. Fast charge of LTO/LFP battery with ODC method
5. Comparison of the three batteries
CHAPTER 4: C/LFP LI-ION BATTERY THERMAL MODELLING
2. C/LFP Li-ion battery thermal model description
2.1 Battery Description
2.2 Development of the thermal model
2.2.1 Heat sources in Li-ion battery during charge
2.3 Determination of the model parameters
2.3.1. Reversible heat source
2.3.2. Irreversible heat source
3. 0D thermal model of Li-ion battery
3.1. Thermal model description
3.2. Simulation results and validation
4. “3D” thermal model of Li-ion battery
4.1. 2D axi-symmetric battery model
4.2. Sensibility analysis of the thermal conductivity
4.3. Simulation of internal gradient of temperature
4.4. Influence of the ambient temperature
4.5. Influence of the battery cooling
CHAPTER 5: AGEING STUDY OF C/LFP BATTERY
2. Cycling ageing
3. Characterization of the aged LFP batteries
3.1. X-Ray tomography photographs of batteries
3.2. Internal visual inspection of the batteries
4. Post-mortem electrochemical characterisation
5. Post-mortem analysis of the electrolyte by GC-MS