The Li-Ion Rechargeable Battery: A Perspective

The Li-Ion Rechargeable Battery: A Perspective

John B. Goodenough * and Kyu-Sung Park

Texas Materials Institute and Materials Science and Engineering Program, The University of Texas at Austin, Austin, Texas 78712, United States

J. Am. Chem. Soc., 2013, 135 (4), pp 1167–1176

DOI: 10.1021/ja3091438

Publication Date (Web): January 7, 2013

Copyright © 2013 American Chemical Society

ABSTRACT: http://pubs.acs.org/doi/abs/10.1021/ja3091438

My motivation for choosing this article is largely due to the ubiquity of lithium-based batteries used in almost all laptops, smartphones, and electric cars.  I was also motivated to read this article (and another on new developments in Li ion batteries) because the author was prominently featured as a source of information in one of the books on my sabbatical reading list.  While a good portion of the paper consisted of detailed highly technical discussions that are not easy to follow, it does give a good overview of the development and chemical evolution of lithium ion batteries motivated by the challenges posed by earlier designs.  The paper shows that most of the chemistry and energetics involved are well-understood enough for these batteries to sustain progress in the search for materials optimizing energy density, cost, and safety.  No explicit description of the redox mechanism for the anode and cathode reactions was given.  The high-level technical discussion and the involvement of the Li+ as a working ion implies that the oxidation and reduction occur as a result of the spontaneous movement of Li+ from one electrode to another during discharge.  Other sources indicate the participation of the interconversion of CoIII and CoIV as part of the redox reaction.  [See Chemical Review article (2004) for a simple schematic diagram of the charge transfer process involved (below)]

In the introduction, the authors pose the following statements about the potential of rechargeable batteries and capacitors in providing an energy source, storage, and distribution system automobiles and household electricity needs.  Some of the important properties to consider in the design and performance of these battery systems are:

èSafety, cost, and driving range for electric vehicles

èShelf and cycle life, response rate to power outages and electric grid fluctuations, and capacity and cost for distributed or centralized energy storage in households using solar or wind power.

A battery is briefly described as having the following parameters:

“A battery is made of one or more interconnected electrochemical cells each giving a current at a voltage for a time Δt. The output current I and/or time Δt to depletion of the stored energy in a battery can be increased by enlarging the area of the electrodes or connecting cells in parallel; the voltage V for a desired power P = IV by connecting cells in series.”

ELECTROCHEMICAL CELLS

In this section, the author describes the general make-up of electrochemical cells: cathode, anode, and electrolyte.  The electrolyte may be a liquid or a solid.  Solid electrolytes generally work better with gaseous or liquid electrodes due to problematic solid-solid interfaces.  A schematic diagram of the first Li ion battery shows large surface area electrodes: C (graphite) at the anode and LiCoO₂ at the cathode.  The electrolyte conducts the ionic current and the electrons are forced to move through the external wire.  Electron conductivity is greater than ion mobility thus the large surface area of metal electrodes versus the thin electrolyte separator through which Li and other ions diffuse due to the small mobility of ions.  The Cu and Al metallic current collectors conduct the electron flow to the external wires.

The authors also define the following electrochemical cell properties:

Discharge voltage = Vdis = open circuit voltage – (discharge current x internal battery resistance)

Charging voltage = Vch = open circuit voltage + (discharge current x internal battery resistance)

The amount of polarization (Idis x Rb) is dependent on the state of charge q and the Idis.

% efficiency of a cell to store energy at a fixed current I =

Q = total charge per unit weight or unit volume (Ah/kg or Ah/L) at a given I charge and discharge current.

Q(I) = cell capacity for a given current I

èQ is dependent on I because the rate of diffusion through electrode/electrolyte interfaces is diffusion-limited at high currents.

Reversible loss of capacity can occur when Li particles are inserted into electrode particles at high charge/discharge rates.

Irreversible loss of capacity occurs due to charge/discharge cycling, loss of electrode volume, and/or electrode decomposition.

Passivating solid-electrolyte interphase (SEI) layer formation can also occur as a result of electrolyte-electrode reaction during the initial charge of a cell manufactured at a discharged state. This is separate from the irreversible capacity fade (loss?) that occurs upon cycling.

% Coulombic efficiency associated with capacity fade = 100 x (Qdis/Qch)

Qdis = total capacity at discharge

Qch = total capacity at charging

Cycle life of a battery = number of cycles until the capacity fades to 80% of its initial reversible value

Gravimetric energy density is measured in Wh / kg or mWh / g and is dependent on Idis because Q is dependent on Idis:

Volumetric energy density measured in Wh / L is an important factor for batteries used in handheld devices, laptops, and other small electronic devices.

THE CHALLENGE

Improvements on volumetric energy density depend on controlling the size and morphology of the active particles in the electrode and the architecture of the current collectors and the engineering and synthetic methods for these have reached their limits, according to the authors. This poses a problem for the battery needs of electric vehicles and energy storage units for solar and wind power.

Challenges with capacity stem from the Li-based cathode.  The extraction of Li from the oxide host “is limited to the reversible solid−solution range of Li in the cathode host structure operating on the redox energy of a single transition metal cation; and where a passivating layer forms on the anode during the first charge, the capacity is further reduced by an irreversible loss of Li from the cathode in the Li+-permeable SEI layer formed on the anode.”  Despite this, rechargeable batteries capable of 30,000 safe charge/discharge cycles at an acceptable rate and a 10 year operational life have been successfully built.  As the authors noted, the challenge here is to build safe, affordable, higher energy density batteries with the same capabilities.

Limitations on the Voc:  The next discussion detailed the complex electrode-electrolyte chemistry and energetics considerations required to control the open circuit voltage which is the difference between the anode and the cathode potential.  This open circuit voltage is limited by the “window” of the electrolyte (energy gap between the HOMO and the LUMO of a liquid electrolyte or between the conduction and valence band of a solid electrolyte) or the top of the anion-p bands of the cathode.  The choice of anode and cathode material needs to consider the energetics of redox chemistry associated with electrolyte substance.  Cells using an aqueous electrolyte are limited to a maximum stable voltage of 1.5 V. The detailed discussion ended with a limiting cathode voltage of 5 V versus Li, requiring a larger cathode capacity than what a Li-insertion host is capable of for higher stored energy density (and an organic solvent).

RECHARGEABLE LI-ION BATTERIES: THEIR EVOLUTION

For a battery to be rechargeable, the chemical reactions at the electrodes have to be reversible.  The two types of reversible reactions at solid electrodes are insertion and displacement reactions. Insertion reactions normally occur at the cathode while anode mostly have displacement reactions (although can be insertion reactions as well).

In an insertion reaction, the working cation, Li+ or H+, is inserted into or extracted from an electronically conducting host reversibly over a finite solid-solution range.

An example is the following reactions occurring in an aqueous alkaline electrolyte:

Cathode reaction (an insertion):                         0 <x < 1

Anode reaction (a displacement):

This battery assembled in a charged state gives a V of ~=1.5 V but the window of the aqueous electrolyte restricts the voltage of a battery with a stable shelf life to < 1.5 V.

The motivation for the Li ion battery is the need for a rechargeable battery with a higher energy density, a larger voltage which requires a non-aqueous electrolyte (to exceed the 1.5 V limit of aqueous electrolytes).  Li+ was chosen as the working ion because it is mobile in a non-aqueous electrolyte whereas H+ is not.  Lithium salts are known to be soluble and dissociate in ether or in organic dimethyl or diethyl liquid carbonates, electrolytes that offer a window of ~ 3 V. 

…Some detailed technical history on the emergence of the materials used in the first lithium ion battery points to the problem of dendrite formation on the lithium surface upon charging which led to some explosive reactions.  The first lithium ion battery avoided this by using a carbon graphite anode that restricts the rate but avoided the dendrite problem.  The two half reactions at DISCHARGED STATES in Figure 1 are:

Insertion of a Li+ ion into a layered host like CoO2 or C was originally referred to as intercalation. An early challenge is the fabrication of the lithium metal oxide which requires that the M(III) ion have a smaller ionic radius than the Li+ ion to form a stable octahedral site planes on a CCP oxide.  Because of the energetics involved, O2 evolution was an issue that needed to be addressed under the parameters for x imposed by using Ni or Co.  Addition of ~10% Al3+ stabilized the layered oxides against electrode-electrolyte interface reactions during the Li extraction but lowered the capacity.  Other challenges addressed is the diffusion of Li+ through the separator which caused short circuits.

Si or alloys of Sn or Sb are replacing carbon because of its lower capacity and the plating of Li on its surface during a fast charge.  These, however, have problems of lattice expansion (up to 3x for Si) upon Li+ insertion and therefore have to be fabricated to accommodate the volume change.

…The next phase in the evolution of lithium ion batteries considered layered lithium metal oxides with exotic formulas involving Mn, Co, Ni in an effort to control site occupations and resulting energetics at varying x values for each atom.

The authors bring back the discussion to the primary challenge posed by battery capacity if it were to replace the internal combustion engine: “To increase the energy density of a rechargeable battery with

solid electrodes to where it can compete with the internal combustion engine, it will be necessary to find a way to raise <V(q)> while retaining a large cathode Q at high currents I.”  Further research on materials to optimize energy densities led to the exploration of olivines of the type lithium metal phosphates, e.g. LiFePO4.  Studies of the energetics of this cathode host pointing to poor electronic conductivity have led to considerations of optimizing electron conductivity of the cathode host; low electronic conductivity limits the charge/discharge rates.  One solution was the use of smaller nanoparticles to increase the surface area to accommodate a higher number of single-phase reactions and/or apply an electronically conductive surface layer.  For example, nanoparticles of LiFePO4 have been prepared with a coating of a thin layer of a conductive polymer. The redox energy of the coated nano-sized particle has to be kept within the electrolyte window.

A summary given by the authors brings back the issue to the more practical challenge of energy density suitable to compete with internal combustion engines:

“In summary, a LiB using solid rechargeable electrodes is capable of a long cycle life at acceptable rates of charge/discharge, but the energy density of individual cells, even with a 4 V cell, makes difficult the manufacture of a cost-competitive battery of sufficient energy density to displace the internal combustion engine of an automobile with long driving range between rapid and convenient liquid-fuel refills.”

…and the need for stationary energy storage for wind and solar-generated energy:

“Stationary storage of electrical energy from alternative energy sources (wind, solar, nuclear) calls for larger capacities than can be realized with an oxide-host cathode, but the energy density requirement of a mobile battery is relaxed.”

STRATEGIES WITH SOLID ELECTROLYTES:

Sodium-sulfur battery: A solid electrolyte separates the molten Na anode and the molten S cathode impregnated with C felt.  (Commercialized in Japan)

Zebra cell: the S cathode is replaced by a discharged NaCl and Fe particles (developed in South Africa and being developed in the US by GE Corporation)

Advantages of sodium cells: lower cost than lithium and larger capacity for large-scale energy stationary energy storage.

Disadvantage: corrosion due to high temperatures required for operations (sodium has to be kept in the molten state).

Solid electrolytes: allow the use of gaseous or liquid reactants like the solid oxide fuel cell which uses a solid oxide electrolyte and gaseous reactants.

Solid-state Li battery:  in principle, would require an inorganic solid or a Li+ polymer electrolyte.  Dendrite formation and maintenance of a good electrolyte-electrode contact are important issues in choosing the electrode and electrolyte material along with other parameters for optimization.  See article for optimum characteristics of a solid-electrolyte membrane separator.

Air cathode: High capacity can be achieved with an air cathode but the reversible reaction requires inexpensive (Pt and Au, expensive catalysts, have been tested with poor results) catalysts for the oxidation reduction reaction and the oxygen evolution reaction:

Zn-air cell:  Voltage is limited to ~1.5 V but modifications of the separator and anode using lithium can raise this to 3.5 V.

Some helpful statements from the summary to wrap up the most salient points in this mostly technical paper:

“Electrical energy can be stored efficiently in a rechargeable battery, and the shift from an aqueous to the organic liquid-carbonate electrolyte in a LIB increased the energy density of a rechargeable battery sufficiently to enable battery-powered handheld electronic devices and power tools; however these applications do not compete with devices powered by fossil fuels.”

“…Another near-term target is storage of electrical energy generated by solar or wind power or for stabilizing the grid against variable demand for power. These latter applications appear to require increasing significantly the energy and power density over what is possible with a strategy that relies on a cathode composed of a solid host into which a singly charged cation is inserted reversibly over a finite solid−solution range.”

“Realization of this situation has led to consideration of either multiple-electron redox couples and/or multivalent working ions such as Mg2+ in place of Li+. This shift of emphasis leads inevitably to the electrolyte, catalysts, and organic multiple-electron redox centers.  We have emphasized here the potential of a Li+ electrolyte membrane separating two different liquid electrolytes, a material that will require a composite of a polymer and an inorganic Li+ electrolyte.”