Hello,
A yearly allocation for the LCRC cluster has been requested with the
following updated information:
Submitter/PI: Paul Redfern
Project Name: Redoxshuttles
Division: MSD
Project title: Redox Shuttles and Additives for Lithium-Ion Batteries
Associated funding: EERE-DOE
Other Systems: Small Group Cluster
Science: There are several types of electrolytes that are used to improve the performance of the batteries.
Some electrolyte additives are used to improve lithium-ion battery performance by facilitating solid-electrolyte (SEI) formation, reducing irreversible capacity and gas formation, improving thermal stability, protecting cathode materials, improving electrolyte properties, decreasing flammability and providing overcharge protection. It is also desirable to develop additives that can stabilize electrolytes near 5 V in order to deliver energy at higher density and potential. Numerous additives have been developed that improve some aspect of battery performance, but many introduce negative side effects.
A second type is used to prevent dangerous overcharge of lithium-ion batteries occurs when charge from current forced through a cell exceeds its charge storing capability, leading to chemical and electrochemical reactions, rapid temperature rise, self accelerating reactions and explosion. Redox shuttle molecules are additives that are reversibly oxidized and reduced at potentials slightly higher than the end of charge cathode potential, locking the cathode potential at the oxidation potential of the redox shuttle and thereby preventing overcharging. The shuttling mechanism can be regarded as an internal short that can be repeated continuously due to reversible oxidation and reduction of the shuttle molecule. Redox shuttles with excellent performance and compatibility are critically important to the development of lithium-ion battery technology. Development of new redox shuttles is difficult and has been largely based on trial and error due to the multidisciplinary knowledge r
equired. Ideal redox shuttles should be electrochemically reversible, have a redox potential slightly higher than the cathode end of charge potential, be electrochemically stable, have good solubility and high diffusion coefficients, and be compatible with the cell, i.e., they should not adversely affect normal cell performance. The shuttle should not react with electrolytes, the solid-electrolyte interface, separators, current collectors, binders, etc. Interface resistance and thermal properties should also be considered.
Project description: Over the past year our DFT calculations were used to provide insight into the mechanism for formation of the core-shell Ir3Li particles. The Ir3Li core-shell nanoparticles discovered here provide a new direction in nanomaterials research for reducing charge overpotentials and finding a closed system in Li-O2 batteries. Knowledge of the solubilities of Li2O2 and LiO2 in aprotic solvents is important for insight into the discharge and charge processes of Li-O2 batteries, but these quantities are not well known. In this contribution, the solvation free energies of molecular LiO2 and Li2O2 in various organic solvents were calculated using various explicit and implicit solvent models, as well as ab initio molecular dynamics (AIMD) methods. The solvation energies from these calculations along with calculated lattice energies of Li2O2 and LiO2 were used to determine the solubility of bulk LiO2 and Li2O2. The calculated LiO2 and Li2O2 solubilities provide import
ant information for fundamental studies of discharge and charge chemistries in Li-O2 batteries. Oxidation and reduction potentials of the synthesized fluorinated sulfones were calculated using the density functional theory (DFT) method. The fluorination lowers the HOMO and LUMO energy levels and raises the oxidation potential of the sulfone molecules. A significant increase (+0.42 V) in the oxidation potential of FMES over EMS suggests its greater oxidation stability on the surface of the charged cathode. Trifluoromethyl substituted sulfones proved to be a promising next generation electrolyte for high voltage, high energy application as evidenced by the long term cycling performance in a NMC532/graphite cell cycled at 4.6 V. urprisingly, the hydrolysis of solid LiO2 is significantly different from that of NaO2 and KO2. Unlike KO2 and NaO2, the hydrolysis of LiO2 does not produce H2O2. Similarly, the reactivity of Li2O2 toward water differs from LiO2, in that Li2O2 results i
n H2O2 as a product. The difference in the LiO2 reactivity with water is due to the more exothermic nature of the formation of LiOH and O2 compared with the corresponding reactions of NaO2 and KO2. The effect of inclusion of water molecules on these reactions based on molecular states was calculated using high level quantum-chemical calculations. The results show a significant shift from being unfavorable in the gas phase to being favorable when solvation effects are included. This effect is consistent with both the observed hydrolysis of these solids and the use of these reactions for quantification of Li2O2, NaO2, and KO2 by titration.
Electric vehicles and energy storage require higher capacity and lower cost than can be achieved by lithium ion batteries. The lithium-O2 battery, with its high (theoretical) energy density, could transform energy storage. Several problems must be overcome before lithium-O2 batteries become practical. Li-O2 batteries typically contain a negative lithium anode, a positive porous cathode and a nonaqueous electrolyte. During discharge O2 gas bubbles into the porous cathode and is reduced to superoxide anion which then combines with Li+ ions in the electrolyte to form solid Li2O2. On charging the Li2O2 is oxidized back to O2 and Li+ which has proven to be difficult. Li2O2 in contact with the electrode surface easily oxidizes but most Li2O2 is farther from the electrode and very difficult to oxidize since Li2O2 is a wide band gap insulator. The voltage increases by as much as 1 V which is impractical. A high overpotential leads to electrolyte decomposition and corrosion of the cat
hode. During discharge Li2O2 builds up on the electrode surface, shutting it down and limiting the amount of it that can be produced (discharge capacity).
Two approaches have been tried to improve charging, making Li2O2 more soluble and using a redox shuttle (mediator). Anion receptors have been tried which complex with O2- to make Li2O2 more soluble but the process is not reversible. Redox mediators are solvated molecules which are oxidized at the electrode and then diffuse through the solution to oxidize solid Li2O2, at which point the redox mediator is reduced. The (0001) surface of Li2O2 is the most stable on charging and is where oxygen would evolve. The equilibrium potential of the redox mediator must be greater than the potential required to oxidize this surface.
On discharge a different redox mediator would be used which gets reduced at the electrode and then carries the electron through the solution to reduce O2 to O2-, which reacts with Li+ to form LiO2, and then another reduced redox mediator and Li+ combine with the LiO2 to produce Li2O2. If this is done in solution instead of on the surface of the electrode, the discharge capacity (amount of Li2O2) produced increases because the insulating Li2O2 no longer covers the electrode. The redox mediator must be able to reduce oxygen to Li2O2 and must not evolve oxygen from Li2O2. A different redox mediator is required on discharge than on charge because its redox potential must be lower than that for oxidizing Li2O2.
The ionization potential (HOMO) of the solvent must be greater than that of the redox mediator for charging and the electron affinity of the redox mediator (LUMO) should be lower than that of the solvent so that the solvent is not oxidized or reduced by the redox mediator. In addition, while a high donor number and acceptor number solvent is desirable for dissolving LiO2, such solvents are very reactive to nucleophilic attack or proton abstraction by O2- resulting in unwanted side reactions so the search is on for weakly solvating (low donor number) solvents. In addition, salt anions can also affect O2- solubility since a high donor number anion can complex with the solvent-Li+ complex which makes Li+ a softer acid, which binds more strongly to the O2- anion (which is a soft base) according to the theory of hard and soft acids and bases so more O2- dissolves and more Li2O2 is formed in solution. High donor number salts increase the discharge capacity and lower the discharge p
otential.
It is important to tune the redox mediator’s equilibrium potential so as to minimize the overpotentials on charge and discharge and to meet stability requirements. Substituent groups on the redox mediator can be used to tune the redox potential.
TDPA is a triphenyl amine that has been used successfully as a redox mediator in the Li-O2 battery. Oxidized TDPA has been shown to oxidize Li2O2 with O2 evolution and TDPA improves O2 solubility and diffusivity. The charging overpotential is lowered and the discharge capacity is increased with TDPA.
Tetrathiafulvalene (TTF) has also been used as a redox mediator. The cell was reversibly cycled for 100 cycles at rates much higher than without TTF.
TEMPO (2,2,6,6-tetramethylpiperidinyloxyl) was used as a redox mediator resulting in longer cycle life. Modification of substituent groups could decrease the charging potential even further. Reduction potentials have been calculated for many nitroxide antioxidants.
DBBQ (2,5-di-tert-butyl-1,4-benzoquinone) in a low donor number electrolyte stimulates O2 reduction, decreases the discharge overpotential, increases discharge capacity by decreasing Li2O2 film growth on the electrode and increases discharge rates. Viologens, phthalocyanines and quinones have been tried as redox mediators but do not show increased discharge capacity or provide direct evidence for Li2O2 formation. DBBQ reacts by a new mechanism which bypasses LiO2 formation which is a thermodynamic overpotential since typically LiO2 must be formed before Li2O2 and DBBQ is reduced more easily. In this case DBBQ is reduced with Li+ to form LiDBBQ in solution, which then complexes O2 in solution; two of these complexes then disproportionate in solution to form Li2O2 and O2 or LiDBBQ reacts with LiDBBQO2 to form Li2O2.
LiNO3 was also investigated as a redox mediator in Li-O2 cells. NO3- or NO2- can’t oxidize Li2O2 but even trace amounts of NO2- are oxidized to NO2 which can oxidize Li2O2 at low overpotential. This is catalytic since the resulting NO2- is then immediately reoxidized. In addition NO3- anion has a high donor number (22.2) and can complex with Li+ and solvent to soften Li+ and bind O2- strongly, increasing its solvation and leading to large crystals of Li2O2 remote from the electrode since charged particles diffuse in solution and transfer charge in all directions. In other studies high donor number salt anions like NO3- or CF3SO3- (donor number 16.9) have been shown to increase discharge capacity.
Heme can also act as a redox mediator with the additional ability of complexing with oxygen to increase oxygen solubility. A salt anion with a high donor number such as ClO4- (donor number 8.4) is required to stabilize binding of O2-. PF6- (donor number 2.5) does not work. Other biomolecules based on porphyrin can be examined to improve charging efficiency.
The proposed research would calculate reduction and oxidation potentials of many potential redox mediators using DFT with Gaussian 09/16. Various porphyrin based species, quinones, nitroxide antioxidants, aromatic amines as well as many others will be studied. These will then be examined see which ones have a low overpotential for charging and discharging and to make sure that the redox mediators will not oxidize or reduce a low donor number low acceptor number solvent. Binding with O2 and O2- will also be calculated to see if oxygen solubility is enhanced. Binding with salt anions will be calculated to see if this can be used to tune the redox mediator. Interactions between redox mediators and Li2O2 clusters will also be calculated to examine oxidation during charging.
Software
We intend to use Gaussian 09/16 for quantum chemical calculations. In Gaussian 09/16 the DFT calculations scale well up to 64 processors.
Industry partnership:
Project URL:
Current FY Hours Used: undetermined amount
New FY Requested allocation: 475000
Q1: 100000
Q2: 125000
Q3: 125000
Q4: 125000
Justification:
Storage requirements:
Thank You,
The LCRC Accounts System
