State of Lithium-ion

Availability of information is a key requirement Most of the technology of commercialized electrochemistry is hidden behind the cadre of corporate lawyers to minimize loss of proprietary information and to limit the basis of litigation. Conversly, the open information from the Power Sources Conference is largely based on work funded by the U.S. taxpayer, providing a noncompetitive platform for free interchange of concepts which may form the basis of progressive usefulness in the growth of electrochemistry technology.

Depth is the second major trait of the contents of the Power Sources Conference. With a combination of multiple tracks, regimented schedules, and peer connectivity, the Conference draws presenters deep within and across the spectrum of batteries and fuel cells. There is a drive to provide information which has content worthy of extending the published knowledge of fundamental chemistry and system implementation. This depth is obtained from a group of 60 presentations chosen because each covers some aspect(s) of Lithium-ion batteries.

The proceedings of the 41st Power Sources Conference are recorded in the 547 page hardbound book shown here. The thirty-three sessions containing 157 presentations were often running in three simultaneous tracks, making the well-organized Proceedings necessary to complete one’s understanding of the information presented. Whether included in Lithium-ion sessions or Safety sessions or Testing sessions, there were a total of 60 Lithium-ion presentations, allowing one to infer that this chemistry is continuing to make major inroads into broad aspects of battery power.

This overview groups presentations by topics, and in future parts will highlight the important concepts within groups. In some cases, individual presentations are relevant to multiple groups such as Electrodes, Testing and Transportation and are therefore included within each appropriate category. +

Grouping considerations are a critical facet of this overview. Despite the excellent grouping by the organizers of the Power Sources Conference, there were many factors which did not allow optimal grouping of presentations, so the luxury of hindsight is used to recombine the presentations. Some presentations were included in multiple groups such as electrodes and nanomaterials.

Part one presents only the groups. Future parts will overview the flavor of each group so that the interested reader can be made aware of presentations which can be pursued in greater depth or just update understanding of the dynamic happenings within the field of Lithium-ion batteries.

One of the points in the scope of this summary, which may not have been sufficiently emphasized previously, focuses on the technology and applications rather than the business of Lithium-ion chemistry. We continue to overview presentations in groups as selected by BD in part one. This combination of part two continues alphabetically with Aerospace, State of Charge, COTS, Configuration and Construction. Future “Thinking” overviews will progress through the remaining groups. Because of content, some presentations could or should be included in multiple groups.

This presentation is also in the ‘COTS’ and ‘Safety’ segments because it focuses on safety testing of off-the-shelf batteries used in a PDA and satellite telephone. It is a good example of utilizing available high-technology devices for shuttle and space station applications. The potential danger is that the device, which works reasonably well on earth, may have different attributes in space. To provide the confidence for carrying the PDA and satellite phone into space with humans, these batteries were qualified with a well planned and documented program. While the small sample sizes limit confidence, thorough testing does show that the integral safety designs such as over Voltage and over current function properly.

From Presentation 1.2. Batteries powering commercial electronic devices used by astronauts are subjected to a wide spectrum of safety tests such as the crush test with performance shown above. It is assumed that a manufacturing defect could result in a foreign particle inside a cell which would loosen during launch vibration, resulting in an internal short. The NASA short test is performed by compressing a rod along the center of the flat portion of the cell just enough to cause deformation with subsequent shorting. ( Reproduction permission is by NASA, with special assistance from the author, Judith A. Jeevarajan.) +

This presentation is also included in the ‘COTS’ and ‘Safety’ segments because it focuses on using commercial batteries in a large array to power hydraulic systems for the Shuttle and Space Station. Specifically, in one application the module is part of a battery which is intended to replace the highly volatile hydrazine powered Auxiliary Power Unit on the Shuttle.

The module is proposed in both a large cell and a small cell configuration. This presentation focuses on the use of 230 individual 18650 cells in a single module. Testing followed a mission profile load for Aircraft and Directed Energy Applications: Modern military weapons have requirements which include high power and rapid turn-on. New Directed Energy Weapons include active armor and electric guns. These devices have very rapid implementation times which may last for very short time segments. Testing included cycle life, environmental temperature extremes, shorts, overcharge and collateral damage. The successful implementation of large groupings of small cells suggests that there may be terrestrial applications that could take advantage of their performance and cost.

Combining the requirements of aviation and space in a single battery chemistry such as Lithium-ion is extremely difficult. The space batteries must operate at very low temperatures, while aviation batteries must operate at both very low and very high temperatures. Space satellite applications require extremely long cycles lives. Aviation batteries may have to provide high power for control surfaces or landing gear. The benefit of employing Lithium-ion is to take advantage of the high energy density, resulting in lower package weight at lift off and flight time, providing greater economies of cost and performance. This presentation summarizes the results obtained in the pursuit of Lithium-ion for this wide range of applications. Battery management has been developed with specialized cells which have performance to 10C and cycling to 18,000 low earth orbit operations.

From Presentation 17.2. In a low Earth Orbit application, the battery is called upon at 90 minute intervals to power the satellite while the earth shades the photovoltaic array. Satellites should remain in orbit for many years to amortize development and launch costs. This means that the battery must provide many thousands of cycles with great calendar life. Neither cyclic nor calendar life are hallmarks of terrestrial Lithium-ion performance, so Lithion is building and testing cells with better life. One reason for greater cycle life is the limit of charge Voltage below the normal 4.2 Volts. In the above data, two cells are taper charged to 3.9 Volts and two to 3.7 Volts. After being subjected to a 40% depth of discharge, the end discharge Voltage is shown above. After 18,000 cycles, the cells still have an energy efficiency of greater than 92%. To get these data, cells have been operational for 5 years, giving some indication of calendar life. (Reproduction permission is by Lithion, with special assistance from the author, Chad Deroy.) +

Many applications have a need for great pulse power. Some of the concepts of a very high power battery were tested in an earlier program implementing 18650 cells, leading to the reported program using DD size cells which are capable of delivering current in excess of 2,000 Amperes. The pulse profile delivers up to 3,800 Watts in milliseconds with tens of milliseconds rest between pulses. Test results show these electrochemical devices to adequately perform, where previously only supercapacitors were considered fast enough in turn-on. Because of the military requirements, performance has been shown to -60 0C with the possibility of discharge rates to 120C. Work continues to expand the understanding of these very high power systems with rapid response to optimize designs.

From Presentation 17.3 Directed energy weapons using lasers, or microwaves for active armor and electric guns (which neutralize projectiles such as mortars, artillery, and cruise missiles) have tremendous pulse power needs. Under a DARPA program, cell performance has progressed to levels shown in this graph. Despite discharge currents of 120C, 80 to 85 % of the total capacity can be delivered in 25 to 30 seconds. Because of the military realm of usage, operational temperature ranges are from - 400 C to + 700 C. Presentation data is shown to -600 C. (Reproduction permission is by SAFT America, with special assistance from the author, Kamen Nechev.) +

Pilotless aircraft have been more than a novelty since WWII when German V-1 flying buzz bombs harassed England. After the war, controlled pilotless aircraft were used to investigate the atmosphere created from nuclear weapons testing. Today, combat experiences have heightened the desire for reconnaissance and battlefield information with reduced risk to personnel using unmanned air-vehicles. Model airplanes built by hobbyists and powered by Lithium-ion batteries in the 1990’s showed that aerial observation platforms were possible.

From Presentation 1.1. Lithium-ion pouch cells were constructed with carbon-carbon composite anodes. The above test data shows the ability of the cell to withstand overcharge to 12 Volts at a 1C rate. The temperature spiked to 38 0C, but there was no fire smoke or explosion. The pouch enclosure remained pressurized as evident by swelling due to internal gas generation. It is believed that the carbon-carbon composite physical properties minimize side-reactions which generate heat during overcharge. (Reproduction permission is by LiTech, LLC, with special assistance from the author, Sohrab Hossain.) +

This paper describes a first generation military micro-air vehicle with 32 cm. wingspan powered by Lithium-ion polymer batteries. The radio controlled flying wing has a wingspan of only 32 cm. Lithium-ion polymer cells have been constructed so that the battery structure is a part of the aircraft structure, reducing total weight. This battery has flown for 1 hour and 47 minutes on a single charge, producing an average of 7.6 Watts. Cells are designed with rugged features both for electrochemical performance in overcharge and overdischarge plus mechanical robustness. Testing details show charge profiles, discharge capabilities to 2C, cycle life over 300 cycles and successful nail penetration testing. (Ed. note: An armada of the micro-planes might be more devastating to enemies than storms of killer bees.)

One of the most important safety issues is overcharge. Present Lithium-ion battery circuits rely on over protection circuits, PTC resistors, pressure sensitive rupture disks and temperature sensitive separator materials. Each safety feature increases the cost of battery while lowering total specific energy.

Intrinsic safety gained by increasing the exposed surface area of the cell will result in greater heat dissipation, but that will also compromise the compact design of the cell. It is more desirable to minimize the side reactions that generate excess heat and eliminate the need for extra expensive safety devices. This study looks at carbon-carbon composites to minimize side reactions which generate heat during overcharge. Pouch cells were constructed and tested at 700 and 2000 mA rates. Specific energy was about 140 Whrs/kg.

During overcharge, the package puffed up, indicating gas generation inside the cell from electrolytes decomposition but temperatures were less than 40 degrees C, with the 700 mA charge and stayed below 52 degrees with the larger overcharge current. It was determined that this combination of electrode - electrolytes and separator caused the separator shutdown mechanism to limit the current flow, keeping the cell safe.

This presentation was unusual in that it provided a combination of outcomes which include the time to battery cutoff, the initial battery state of charge prediction, the operational state of charge, a state of health and the remaining number of battery recharges, or as coined by the authors, ‘state of life’ (S0L).

Impedance measurement data is software processed in an electrochemical model which provides vector files to three different, but simultaneous, analyzers. One analyzer is a transfer function model estimator; the second is a fuzzy logic estimator and the third is a neural network estimator. The three estimation file outputs are delivered to a decision processor which compares the information to a knowledge history to throw out gross system errors. From this comparison, each estimate is ranked by measures of confidence. Lower confidence information is “deweighted” from the final predictions. Validation data is not presented.

From Presentation 27.2. Beginning with wide band impedance data, feature information is generated from an electrochemical model. That information is processed by three independent estimators. The results are evaluated based on a knowledge of history to eliminate estimations derived from non-processing factors such as loose connections. The goal is to obtain highly accurate SOC, SOH and SOL information about the battery. (Reproduction permission is by The Pennsylvania State University, with special assistance from the author, James Koslowski.) +

This project is targeted to 18650 cells which power a portable defibrillator. The typical profile of the application is to deliver 10 Amp pulses for five seconds, with quiescent 1.4 Amp. discharge at five minute intervals. After the pulse, a Voltage recovery profile is obtained and plotted with the number of pulses. From this data, a curve fitting equation is developed and coefficients are determined. The solution then identifies the cycle number for state of health obtained from real cell testing. Training and testing errors of 12-15% are being reported. Work continues to improve the model and determine the pulse number prediction (SOC).

Make or buy decisions involve availability, suitability, stubbornness and sometimes pride. NASA often has clear-cut guidelines for the decision. Since Shuttles are not readily available at Wal-Mart, they must be specially designed and built for space. What is carried on-board often falls into the clear-cut guidelines for purchasing commercial- off-the-shelf items. This was the case for the use of a PDA and an Iridium satellite phone. Unfortunately, these devices are not merely unpacked from their shipping containers and delivered directly to the launch pad. Qualification for flight involves detailed test and analysis of safety conditions for these devices, including their batteries. The battery part of the qualification was described in this presentation.

From Presentation 27.4. Difference Voltages at the beginning and end of a pulse are measured and plotted for each cycle within a battery discharge cycle. Curve fitting of the coefficients of a polynomial/exponential equation are determined from a fuzzy logic model. The outcome is a prediction of the cycle number (SOH). (Reproduction permission is by Villanova University, with special assistance from the author, Pritpal Singh.) +

The Lithium-ion batteries, both prismatic types, were subjected to charge/discharge cycles at a variety of temperatures, overcharge, overdischarge, short circuits, vacuum, vibration and crush testing. The presentation covers the results of a variety of these tests which led to a qualification for space flight with humans. Crush testing of both batteries resulted in explosions. These results were deemed extreme situations as vibration testing results removed the likelihood of internal shorts due to foreign particles. The batteries were thereafter approved for manned space flight.

The status of the battery within the system was also scrutinized. It was found that the satellite phone was independently two-failure tolerant to catastrophic hazards but the PDA had only one level of protection for overcharge. In this case, the system depends on the PDA charging circuitry for additional levels of protection.

Despite the fact that the devices required qualification safety testing, the use of commercial products provided a cost-effective implementation. The testing does not give carte blanche acceptability to any replacement battery in this application because commercial manufacturing factors could change the status, whether made by the original manufacturer or a replacement manufacturer.

Because of successes which provided rapid implementation of commercial batteries for space flight by the European Space Agency at the turn of the century, the Sony 18650 (hard carbon) cells have been considered suitable alternatives for space battery applications. In this presentation, the auxiliary power unit of the space shuttle, now powered by hydrazine, was considered for replacement with either large or small cell rechargeable Lithium-ion commercial-off-the-shelf (COTS) batteries. This presentation focused on the small battery (18650) approach. The small cells are placed in series and parallel combinations to provide a total of 28 kWh.

From Presentation 20.1. When assembling a large number of small cells into a module, the module can be characterized as a single ‘cell’ or ‘battery.’ The above data shows the ongoing cycling which will lead to end of life data for the module. The capacity fade for the module closely follows the fade of individual cells. (Reproduction permission is by Modular Energy Devices, Inc., with special assistance from the author, Stephen S. Eaves.) +

The mission profile for the battery was developed based on prior flight experience. Testing included a variety of external and internal shorts, collateral damage to external diodes and overcharge at hot and cold mission environmental temperatures.

Hot and cold mission profile testing showed that the cells’ PTCs tripped after being exposed to the hot temperature extremes during the mission phase. This led to the conclusion that for application, the cells will require a more robust PTC. While the testing showed that the current interrupt device (CID) in single cells offers protection for Voltage above 5.0 Volts, the CID cannot be relied on for overcharge protection in the module with many series/parallel combinations which make up a total battery. The reader would suspect that the test results would suggest a possibility for use of small COTS batteries for the application but that further work needs to be done to make them totally usable in this manned space flight application.

This presentation could have also been included in the commercial off-the-shelf (COTS) batteries section because it uses commercial cells. The central idea discussed was how large modules, made up of 18650 cells, are assembled and then used individually or in groups to provide very high energy capabilities. The typical module provides between 57 to 76 Amp hours with capabilities to 88 Amp hours in a 300+ cubic inch package. Data was shown for the discharge profile, cycle life and even life related to time on float, up to 10 years. The float date is especially interesting because it is difficult to find it for Lithium-ion applications. The projected usage is for Telecom and cable TV power backup, electric bicycles, and the Army’s Silent Watch battery program.

Coextrusion Spiral Die

From Presentation 11.2. To produce a rugged cell pouch material, which can withstand the rigors of testing to the Desert Cycle Profile, experience was borrowed from food packaging using an exterior polyester coat, an aluminum foil and then a 7-layer film substrate. The substrate is formed using an intricate spiral die as shown above. Direct sealing of the electrical leads to the laminate was succesfully accomplished. (Reproduction permission is by Pliant Corp. with special assistance from author, Gary Reich.) +

A 16 Volt, 9 Amp hour prototype Lithium-ion battery was built and tested. The four cell battery was constructed with packaging material suitable for Land Warrior applications, where light weight, long operating missions and environmental protection were important. The material selected was based on the flexible pouch material used in the military’s meals ready to eat (MRE) package because of its success in meeting storage conditions for over five years. The flexible packaging includes aluminum oxide coated polyester film, aluminum foil and a seven layer extruded film which then offers significant oxygen and moisture barrier performance. The cell was constructed with a low temperature electrolyte solution and polyethylene separator. Test data for temperature performance, discharge, and cycling data were presented.

Possibly the most difficult group of presentations in this overview of the 41st Power Sources conference is that of the electrodes, electrolytes and separators. The work reported is very complex because investigators are looking into the innermost interactions of the chemistry often to produce minuscule improvements in energy density, cycle life, calendar life and safety. Improvements sought are not earth shaking revelations of performance, but rather incremental improvements. If one envisions the millions of Lithium–ion batteries produced each month and then analyzes what a one percent improvement would do to that group of millions, and the ensuing billions thereafter, the small improvement becomes a gigantic contribution to the total energy produced. Because of this large base of Lithium-ion cells produced, the quest for continual incremental improvement is fueled with money to fund research such as is described below.

Research first has to point to fundamental paths which may lead to improved batteries, but it must also identify and explain those paths which do not lead to possible improvements. Because many of these investigations are in very fundamental electrochemistry rather than involving complete battery construction, the interest may be limited to only a few readers. Those with more global interests may only want to scan the work described, while some will find it an alert to obtain greater detail about the full presentation, and possibly connect with authors to share ideas which might lead to further growth of the technology.

5.1 High Performance Ni-Based Lithium-ion Cathode Material Designed for Potential Use in Hybrid-Electric Vehicles: Many researchers have worked with LiNiO2-based materials with the goal that these materials could be the next generation cathode materials for Lithium-ion batteries. Interest in nickel-based cathodes has been lead by two drivers, increased capacity and lower cost. However, maintaining adequate safety while keeping high capacity and power has been a challenge.

This paper, discussed TIAX, LLC’s results of calorimetric studies on a new Ni-based (doped) cathode material formulated into electrodes for use in HEV (Hybrid Electric Vehicle) electrodes. Results of their on-going study show that new LiNiO2-type cathodes can be made safer than previous results reported due to lower chemical reactivity. Optimizing the formula for coating the cathodes has made an improvement in safety. With this optimization, gains in power are also being maximized. Long term cycle life evaluation is in progress for HEV type applications. As of the date of the presentation, the new LiNiO2-based cathode material shows extended cycling at 10% DOD (80%SOC), with greater than 15,000 cycles. Testing is on-going. Results to date look promising.

5.4 Thermal Behavior of Vanadium Pentoxide Aerogel and Ambigel Cathode Materials: The U.S. Naval Surface Naval Warfare Center is studying the electrochemical behavior of amorphous vanadium pentoxides hydrogels which have shown potential as high energy density electrodes in rechargeable lithium batteries.

Figure 8: Isothermal plots of % crystallization vs. time.

From presentation 5.4 These isothermal plots are very useful for predicting the behavior of thermal unstable materials. The plots show that the fraction of material that would crystallize at these temperatures is very small. In fact, at temperatures of 120-1800C, the fraction is insignificant. (Reproduction permission is by Naval Surface Warfare Center’s Carderock Division, with special assistance from the author, S. Dallek.) +

The content in this research included thermoanalytical methods for determining water content, vanadium oxidation state and crystallization kinetics of vanadium pentoxide aerogel and ambigel cathode materials. Water content and vanadium oxidation state were determined by TGA (thermogravimetric analysis) and crystallization behavior was determined by DSC (differential scanning calorimetry). Results of these tests allow for correlation of thermal analysis results with electrochemical discharge behavior. The isothermal plots shown in Figure 8 are very useful for predicting the behavior of thermal unstable materials. The plots show that the fraction of material that would crystallize at temperatures of 120-1800 C is very small. This valuable information can provide the basis for further development of these rechargeable cathode materials.

20.6 Degradation of Lithium Rechargeable Batteries: This presentation used one of the classical techniques for technology progression - the forensic investigation of the internal elements of the batteries from the normal, worn out or otherwise failed condition. The target chemistry in this case is Lithium-vanadium pentoxide, which is a rechargeable lithium battery. The degradation mechanisms observed are similar to those reported for other lithium rechargeable chemistries. Analysis consisted of a battery test system, electron microscopy, and NMR investigation. Depositions of lithium and aluminum on the surface of the cathodes in depleted cells was observed, with reduced porosity of the electrode. Lithium deposition leads to dendritic growth shorting of the battery and such shortings have also been reported in Lithium- polymer cell investigations. The combined observations suggested that the capacity degradation was due to the decomposition of the organic electrolyte during charge and discharge cycling. Depletion of the solvent plus thin film deposition on the cathode increased cell impedance leading to battery failure.

23.1 LiNixCo1-xPO4 (0#X#1) Cathodes: Investigations centered on the use of lithium transition metal phosphates with an ordered olivine structure LiMPO4 (specifically, LiNiPO4 -LiCoPO4 solid solutions) as a potential cathode improvement of capacity and Voltage. The goal was to study the discharge capacity of a series of materials based on these solutions. The findings noted that LiNiPO4 and LiNi0.8Co0.2PO4 yielded no capacity on discharge. However, LiNi0.5Co0.5PO4, LiCo0.8Ni0.2PO4 and LiCoPO4 exhibited Voltage plateaus around 4.65 - 4.8 Volts versus Lithium on discharge with capacities between 35-72 mAh/g. The discharge capacity was enhanced with increasing amount of Co in the materials.

29.3 Nanostructured Electrodes for Next Generation Rechargeable Electrochemical Devices: Based on the promised improvements from nano structured materials, the concept has been applied to cathodes of Lithium – ion batteries. The target is to improve the performance of manganese dioxide based cathodes because of the lower-cost and safety over cobalt and nickel cathodes. A major drawback of ordinary manganese based cathodes is that they exhibit significant capacity fade and do not come close to the theoretical energy density of 950 Wh/kg. It is thought that the material suffers from a poor rate of diffusion of lithium ions into the cathode particles. Producing cathodes with much smaller particles, as is done in nanostructures, can provide the needed shorter diffusion distance for the intercalating lithium ions.

In this reported work, nanostructure powders were prepared with nominal 50 nm crystal size. From these materials, test cells were produced to obtain the charge and discharge Voltage curves, the retention capacity with cycling and the specific capacity of the cathodes. The energy density measured was 550 Wh/kg for the cathodes, and discharge Voltages were comparable to current cells. Capacity fade for the nano structured cells was not significant in limited cycling. As this technology is refined, the energy density of cells will be improved and will allow less expensive and safer materials to provide performance better than that seen in today’s cells.

5.5 High Capacity Anode Materials for Lithium-ion Batteries: Carbon anodes have been popular in Lithium-ion batteries because of their characteristics of exceptional reversibility and safety. Alternative, silicon-based anodes, are interesting to explore because their high theoretical discharge capacities of 4000 mAh/g in comparison to typical carbon-based anodes which have 290 mAh/g. But because there is poor reversibility of the alloying reaction between lithium and silicon at room temperature, work has been done to successfully coat carbon onto silicon powder using a CVD (chemical vapor deposition) method to improve electrochemical performance.

From presentation 5.5 The data shows the first ten cycles for all of the 7 Ah cells. Significant differences exist between the carbon-coated silicon (Si) material and the Si material without carbon coating. The CVD procedure for carbon deposition improved specific capacity and fade rate. PNG-Si cells peaked at 7.1 Ah and faded to 6.3 Ah by the tenth cycle. The PNG-Si-C cells had a higher capacity of 8.0 Ah and a smaller fade rate that resulted in 7.3 Ah after ten cycles. Coating of the silicon particles that are on the graphite surface have greater reversibility than noted with the silicon particles on PNG-Si. ( Reproduction is by permission of Lithion, Inc., with special thanks to the author, Santo Iaconetti.) +

The study found that silicon-based anode materials showed improved electrochemical performance when particles were carbon-coated via CVD. Milled silicon powder displayed much less irreversibility than the CVD coated version of silicon, which remained coulombicly efficient for about 40 cycles. It has been shown that optimizing the CVD system for more uniform and repeatable coatings can extend cycle life.

In one test, 1) Natural graphite (PNG) materials and 2) PNG materials with carbon coating (PNG-C) and 3) silicon embedded onto the graphite surface (PNG-Si) and 4) a carbon-coated version of PNG-Si (PNG-Si-C) were analyzed in 7 Ah prismatic cells. Discharge capacity was compared to the number of cycles as well as the specific capacity compared to the number of cycles. Some findings were: The fade rate was less with the carbon coating. Specific capacity increased with the coated material. There are similarities in the specific capacity and reversibility in the initial cycling of the PNG and PNG-C materials. Cycle life improvement can be enhanced with carbon coating on the graphite. More cycles need to be completed for additional performance data on the carbon coated material. See above data chart, “Discharge Capacity vs. Cycle Number.”

8.1 Carbon-Carbon Composite - A High Capacity Anode for Lithium-ion Battery Systems: LiTech , LLC’s researchers have proposed carbon-carbon (C-C) composite as an anode for Lithium-ion batteries where the practical capacity is in the range of 300-320 mAh/g. Because carbon-carbon does not contain any inactive materials , the entire composite electrode is active for Lithium-ion intercalation. The paper described preparation of the C-C composite as well as its physical mechanical and thermal properties in the range of 300-320 mAh/g.

• low irreversible capacity loss/low surface area and no foreign materials and surface functional groups. The irreversible capacity loss of this C-C composite anode is only 3% compared to 8% for most carbon/graphite materials.

• high cycle life/high interlayer spacing with low surface area and strong mechanical integrity. In a cycling test at 200C, the cell was discharged at a constant current of 1C and charged at the same constant current rate to 4.2 V and then held at a constant Voltage for 3 hours. The behavior of a C-C composite-based Lithium-cell delivered over 500 cycles with less than 10% capacity loss.

• low heat generation during overcharge along with low surface area and low coefficient of thermal expansion and absence of thermal gradient. Carbon-carbon composite’s physical properties are very favorable to minimize the side reactions that generate heat during overcharge.

Voltage, Temperature, Current Response during Charge-Overcharge of a Lithium-ion Cell at 1C rate: Anode: C-C Composite; Cathode: LiCoO2.

From Presentation 8.1 Voltage and temperature response are shown during charge/overcharge at a constant current of 700 mA (1C) until the cell Voltage reached the upper set-up limit of the channel (12V) when it was allowed to go to a rest period. Although there was an increase in cell temperature during overcharge to 380C from 21.40C, there was no fire, smoke or explosion. (Reproduction is by permission of LiTech, LLC, with special thanks to the author, Sohrab Hossain.)

10.4 A Study on the Characteristics of Passive Films on the Surface of Graphite Anodes in Polysiloxane based Electrolyte: The mechanism of passive film formation on the surface of highly oriented pyrolytic graphite (HOPG) charged in polysiloxane-based electrolyte was studied. Polysiloxane has been studied as a suitable electrolyte in a lithium battery because of its high conductivity which exceeds that of polyethylene oxide (PEO) , a well-known polymer currently being used.

In the study, the researchers at the Material Science department of the University of Southern California and Quallion, LLC centered their work on understanding why polysiloxane results in poor cycling capability. The researchers correlated this characteristic to the nature of the SEI (solid electrolyte interface) film. In the investigation, the graphite surface was charged with polysiloxane-based electrolyte with lithium bis(oxalo)borate(LiBOB). To better understand the formation mechanism and structure of the SEI layer on the graphite, scanning microscope (SEM) examination and Auger electron spectroscope (AES) measurements were completed.

Figure 10. Discharge curves of coin cells with polysiloxane based electrolyte added with different amounts of VEC

From Presentation 10.4 Vinyl ethylene carbonate (VEC) was used as an additive in the experiment. The discharge curve of coin cells with polysiloxane-based electrolyte added with different amounts of VEC are shown. The polysiloxane-based electrolyte cell without VEC does not deliver any discharge capacity. (Reproduction permission is from Quallion LLC, with special thanks to the author, Hiroshi Nakahara.) +

Vinyl ethylene carbonate (VEC) was added to the polysiloxane-based electrolyte to reform graphite surface (suppress film formation) which led to a positive gain in the discharge property of the LiCoO2 VEC graphite lithium battery. VEC additives were shown to increase conductivity of the system. These results could lead to a pathway of improvement of intercalation/deintercalation in lithium batteries having a graphite anode and polysiloxane-based electrolyte with LiBOB.

23.2 Electrochemical Behavior of Tin Oxide Nanoparticles as a Material for Negative Electrodes of Li-ion Batteries: Lithium alloys have been considered as alternative electrode materials for anodes of rechargeable Lithium-ion batteries. The Li-Sn compounds are of special interest because lithium can be inserted electrochemically, reversibly, into tin to form alloys of high Lithium content up to Li17Sn4, having 790 mAh.g. However, a problem exists in that in the process of insertion, pronounced volume changes occur which lead to an intrinsic instability of the lithiated alloys in solutions. This problem can be avoided if the size of the metallic host is kept small. The purpose of this paper centered on an investigation of the electrochemical lithiation/delithiation process of negative electrodes based on nanosized particles of tin oxide (SnO) for rechargeable lithium batteries.

Based on initial results of the experiment, nanometer size SnO power was found to be more effective as an active electrode material than micrometer size SnO. SnO nanoparticles for Lithium battery anodes is a promising material but extensive practical, engineering-related work must be done to make the SnO nanomaterial useful for this battery application.

23.5 Performance Characteristics of Lithium-ion Cells Possessing Carbon-Carbon Composite-Based Anodes Capable of Operating over a Wide Temperature Range: Solving the challenging temperature requirements, especially low temperatures to -400 C, for batteries has been a strong goal for NASA. In this presentation, researchers at JPL (Jet Propulsion Laboratories) shared results of the performance characterization of prototype Lithium-ion cells with carbon-carbon composite anodes developed by LiTech, LLC (MER Corporation). (See Presentation 8.1 in this summary for more information on the carbon-carbon composite electrode developed by LiTech, LLC.)

• At temperatures lower than 200 C, the quaternary low temperature electrolyte outperformed the binary mixtures. Discharge capacity as a function of discharge current and temperature was measured for a 700 mAH LiTech Lithium-ion cell containing a quaternary JPL low ethylene carbonate -content electrolyte. With discharge rate (C/5 to C/20) and temperature (-200 to -500 C) reasonable performance was obtained to -500C at a C/20 rate.

26.4 Safe Design Composite Material in Negative Electrode for Lithium-ion Cell: The Army is investigating polymer pouch cells to gain higher energy for a given volume. This paper specifically discusses the possibilities of the MER Corp. (LliTech, LLC) cell which has the carbon fiber and has been reported to handle overcharge and undercharge without incident.

To conduct the study, the U.S. Army Research, Development, and Engineering Command’s Communications, Electronics, Research, Development and Engineering Center (CERDEC) utilized Sony’s UP383562A cells as a comparison and control with the MER cells. The MER cells performed well in tested conditions such as storage, low temperature and high rate discharge and cycle life. These cells also tolerated at least 50% overcharging capacity and overVoltage to 4.6 Volt without any physical changes. The cell cycling test results indicated that the cycling between a 4.2 Volt charge and 0 capacity for discharge appeared to decrease the useable capacity rapidly. No dimensional changes (bulges) were noted in the MER cell during low Voltage discharge cycling.

26.5 Evaluation of Novel Carbon Based Additives in Li ion Anodes: ITT Industries -AES, Inc. is researching fullerene soots as an additive to the graphite-based anode for the Lithium-ion battery in order to improve graphite’s rapid capacity fade when repeatedly cycled. Various ratios of fullerene soots to graphite were used in electrode preparation as well as types of soot. Arc generated soots using 10% fullerene soot to 90% graphite have displayed the best results. An initial specific capacity of 334 mAh/g was observed and showed less than 1.5 % decrease over 200 cycles.

7.4 Non-Flammable Polyphosphonate Electrolytes: This work involves the use of phosphorus chemistry to develop electrolytes for better flammability performance. Using differential scanning calorimetry and thermogravimetric analysis, comparisons of the experimental polyphosphonates to control electrolytes showed polyphosphonates exhibited lesser reactions prior to lithium melting. This suggests a smaller contribution of that electrolyte to thermal runaway. Identification of such chemistry improvements is the basis for choice of materials for future investigations to increase safety.

10.1 Conductivity of POSS-PEO(n)8 Based Solid-State Electrolytes: In search of solid-state electrolytes which have liquid – like conductivity for Lithium-ion batteries, polymers with low glass transition temperatures are candidates to provide high conductivity with the mechanical properties of a thermoplastic film. Materials with chemical designations, far more complex than are usable in this overview, were investigated for conductivities. Conductivity variation with the amount of inorganic materials were related to variations in the glass transition temperatures which ideally will be low for battery applications. This is fundamental work can be applied to a next generation of investigation which would include such materials for polymer electrolyte applications.

10.2 Ionic Transport Properties of Polyimide Based Electrolyte Films: This work is fundamental research to investigate the polyimide Matrimid from the viewpoint of using it as a high-energy lithium battery electrolyte. Films were cast from solutions out of polymer and salt, then dried and experimentally measured for conductivity and transference. Data of the impedance and melt temperatures are included. The materials were found difficult to dry, and all materials formed contained some solvent. Conductivity which is low at room temperature can be improved with plasticized material to increase conductivity to 10-4 S/cm. Future investigation will center on an improvement in conductivity. Low transference is also suggested to be a critical factor in limiting performance in high-power or in low-temperature applications.

14.1 18650 Li-Ion Cell Building for Electrochemical and Thermal Abuse Testing at Sandia National Laboratories: This presentation describes a program in which 18650 size cells were constructed using a special fire retardant additive in the electrolyte to determine ways to improve the thermal abuse tolerance of cells. Three salts and two additives were investigated. Testing results for the cells included cycling, impedance measurements and temperature variations to characterize performance. Baseline cells vented and went into thermal runaway at 192 0 C. Cells with supposed improved thermal abuse tolerance provided more gentle venting response in some cases but reduced thermal runaway at temperatures as low as 140 0C. A major milestone for this project was to prove an in-house capability for fabricating Lithium – ion cells with new materials which could provide the high reproducibility needed for quantities of experimental cells.

Figure 4: Impedance comparisons for SNL cell with that of two commercial cells

From Presentation 14.1 The data shows the lower impedance of the SNL built 18650 cells which are optomized for power rather than capacity. The commercial cells have capacity of 2,200 mAh whereas the SNL cell has only 920 mAH capacity which is the result of thinner electrodes to maximize power. Other data showed the effects of two flame retardant additives and a comparison of three salts in electolytes.(Reproduction permission is from Sandia National Laboratories, with special thanks to the author, G. Nagasubramanian.) +

26.3 Suppression of Decomposition Reactions of Lithium-Ion Battery Electrolytes: Working with a concept that Lithium-ion batteries have power loss and capacity loss, especially at high temperatures, this presentation pursued thermal decomposition of electrolytes as a major contributor. Lewis bases such as pyridine can be added into the electrolyte to prevent the decomposition of the electrolytes. It is thought that the additives prevent the generation of catalytic fluorophosphates. To test the concept, cathode materials were prepared and subjected to various electrolytes to determine the formation of surface films. Spectroscopy was used to determine the results. It was concluded that the addition of these additives (i.e. Pyridyne) to the electrolyte retards the growth of surface films.

Co and Ni content on the surface of LiCoxNi1-xO2 particles stored at 100 oC for 336 h.

From Presentation 26.3 This data provides a measure of the Co and Ni content of the cathode materials. While the data for Ni and Co is not as clear as that previously observed for F and P, the concentrations of Ni and Co on the surface of the particles is higher in samples that contain thermal stabilizing additives. HMPA appears to inhibit the formation of surface films on LiCoxNi1-xO2. (Reproduction is by permission of University of Rhode island, with special thanks to the author, Brett Lucht.) +

32.2 Characterization and Performance of LiBOB as Electrolyte Solute for Li-ion Devices: The lithium salt LiBOB [Lithium bis(oxalato)borate] is considered a possible alternative to the standard LiPF6 because it exhibits stable performance at elevated temperatures. This investigation extended the understanding of the good and bad features of LiBOB. Electrochemical, spectroscopic and assembled cell performance analysis were made.

Photoelectron spectroscopy was used to characterize the anodes surface chemistry when contacted by LiBOB. The stability of the electrolytes with the cathodes was investigated in a kinetic manner. Interactions with the cobalt-based cathode materials are different from that with the nickel based cathode. A 12 percent post storage capacity loss was observed with LiBOB and a cobalt cathode. Impedance spectroscopy showed a tenfold increase in cell resistance, indicating a continuous reaction between the LiBOB and the cathode surface.

The OCV Decay of th e Li-ion cells based on

LiNO2-cathode at 70 0C

From Presentation 32.2 There is an apparent gap between the thermal stability of the two electrolyes on the nickel based cathode surface at higher temperature. The chemical inertness of the BOB- anion resulted in better OCV retention, supporting the idea that LiBOB contributes to improved cell performance at high temperature. (Reproduction is by permission of Army Researech Laboratory, with special thanks to the author, Kang Xu.) +

32.3 Effect of LiBF4 on the Cycling Performance of Li-ion Batteries: There is an alternate lithium salt, lithium bis(oxalato)borate more commonly referred to as LiBOB which may have as advantages of cycling stability at elevated temperatures and a unique ability to form protective films on graphitic anode surfaces. This presentation is a continuing analytical effort to determine the possible uses of LiBOB in place of the current LiPF6. The effects of standard and LiBOB electrolytes were analyzed both for cathode and anode materials used in Lithium – ion batteries. The results showed that there is promise for this approach, but that additional changes to the salt must be made before use in a practical battery.

Dischare Voltage profile for selected cells in Table 1 at -200 C.

From Presentation 32.3 Every cell that contained some amount of LiBF4 showed evidence of lithium plating in the discharge curve at - 200 Cand irreversible capacity loss during subsequent room temperature cycling. These results lead to the conclusion that LiBF4 is involved in the formation of a more resistive SEI layer on the anode that lowers the rate of lithium action into the anode more than other common salts. (Reproduction is by permission of Yardney Techncal Products, Inc., with special thanks to the author, Thomas Bararich.) +

32.4 Change of Conductivity with Salt Molality, Solvent Composition and Temperature and Its Mechanisms for PC-DEC and PC-EC Solutions of LiBF4, LiPF6, LiBOB, Et4NBF4, and Et4NPF6: This work brings together a comparison of conductivity and viscosity for a variety of common, new and proposed electrolytes from data of previous studies. With such a comprehensive overview, choosing salts based on the application requirements will be more effective.

From Presentation 32.4. Comparison of Tg-surface of glass transiton temperature in the coordinates of salt molality m and solvent mass fraction w for the PC-EC solutions of LiBOB. LiPF6, LiBF4, Et4NBF4 as indication for the changes in their viscosity. (Reproduction is by permission of Army Research Laboratory, with special thanks to the author, Michael Ding.) +

The data showed that the behavior of the LiBOB electrolyte is governed mostly by the high viscosity, whereas the LiBF4 is governed by its strong ion association and the LiPF6 by both. LiPF6 is more conductive than LiBOB. LiBF4is favored for low temperature applications, but LiBOB is not suitable for low temperature applications. The lithium salts seem to have a higher solubility in carbonate solvents than the ammonium salts. The ammonium salts have more conductive electrolyes when dissolved in carbonate solvents than have the lithium salts.

8.3 Performance Evaluation of Battery Separator Materials for use in Organic and Alkaline Electrolytes: The target application for this presentation is lithium primary and a Silver-zinc battery chemistries. Because Lithium-ion also uses organic electrolytes, the information is relevant to this chemistry. Data is presented on absorption, porosity, dry and wet thickness, and resistance and resistivity. Performance with alkaline electrolytes in various concentrations from 25 percent to 45 percent are included. At the conclusion of the paper, the suggestion is made,

Today, making a battery is one thing, but to make a high energy or power battery is a special case, especially with Lithium-ion, because its high energy and power demand unique safety features. Even a Lithium-ion battery in a PDA could explode and cause burns to the user’s hand. As the power or capability increases, the potential for a higher rate exothermic reaction opens the door to an even greater hazard. With high-energy batteries, there is an even greater opportunity for a large explosion or fire.

Safety is not the only concern for high energy and power batteries, but there is also, oftentimes, a requirement for performance in difficult environmental conditions, usually high and low temperatures. A battery powering a Rover on Mars may get very cold. A battery living in a vehicle engine compartment may get very hot. While there are engineering possibilities for dealing with the environment, the most straightforward approach requires a battery which could exist in those environmental conditions. Even the battery itself has requirements for getting rid of heat generated as it is providing energy.

The overviews of the following presentations, separated into ‘High Energy’ and ‘High Power’ sections, bring out many of the challenges and stepping stones of progress which may provide improvements or solutions which will extend Lithium-ion applications.

14.3 A New Strategy for Li-ion Microbattery Development as an Autonomous Micropower Source: High Capacity LiCo02 Li2RuO3 Electrodes: There is a need for microbatteries to be used in conjunction with ever expanding microsensor and MEMS applications. The very size of a microbe battery implies a very small amount of energy available, so one of the major goals is to increase the energy density of a microbattery. In this paper, the work shows that the addition of ruthenium oxide to cobalt based cathodes provides a 34 percent increase in discharge capacity.

Using the Naval Research Laboratories’ protocols for stenciling electrodes and creatingprecision laser deposition of electrode materials, microbattery cathodes were built with standard lithium cobalt oxide, ruthenium oxide and a mix of the two. While the ruthenium oxide performed similarly to cobalt oxide, the mixture-based cells provided significant improvement in discharge capacity when tested to 2.0 Volt levels at 1C rates. Despite the additional cost of the ruthenium oxide, the small amounts needed in microbatteries may be justifiable to produce this additional capacity.

14.4 Design, Thermal Analysis and testing of Very Large Lithium-ion Cells: Lithion, Inc. developed and tested a 200 Ah cell to power an unmanned aerial vehicle and a 400 Ah cell for a submersible crew transport in a U.S. Naval application. Because heat generation and heat transfer rates are most critical in such high energy/high power applications for maintaining safe operating temperature for the battery and its accompanying electronics, thermal modeling of the system (battery and electronics) under specified power cycles were characterized.

Prismatic cell designs were used because thermal modeling of the pack showed that the cylindrical cell construction temperature increase would be 3.5 times greater than that of the prismatic shape. The design for the 200 Ah cell needed to fit in a rectangular envelope so the design of the cell was a rectangle. However, the 400 Ah cell had to fit into a round tube, so the design construction had to be modified. A trapezoidal design with a rounded bottom was chosen for this 400 Ah cell with 1400 Wh of energy. Special design features included matching the curvature of the bottom of the cells to the radius of the tube to give a clean path, in a direction parallel to the electrode plates, directly to the metal tube. Thus, heat flows directly out of the bottom of the cells and battery pack.

In one test, to show how the large 400 Ah cell would move heat efficiently, a 120 Amp discharge and temperature profile was run. The cell chamber was set to 250 C and was well insulated with no leakage of heat. Several different temperature heat generation rates were noted. First, the temperature increased linearly until reaching a plateau at 65% state or charge. The plateau existed to about 30% state of charge. The majority of heat generated was in the remaining 30% discharge. It was determined that cycling the cells in the battery between 100% and 30% state of charge would be the preferred technique for thermal management. The presenter, Seth Cohen, noted that (to date) the cell demonstrated excellent cycle life and specific energy up to 216 Wh/kg.

Rather than focus on a single battery, this presentation overviewed Lithion’s progress over the recent years in developing aerospace Lithium-ion batteries. These batteries are used to provide power to parts of systems, or to provide the total power such as in the situation where a satellite encounters the shadowed part of its orbit, being deprived of its photovoltaic power. The years have included research, development, testing and now delivery of batteries for applications ranging from the B 2 bomber to Mars Landers and Rovers

From Presentation 17.2. Early Lithium-ion cells had limited cycling ability, restricting their possible use in satellite applications. Through electrochemical improvements plus limiting the peak charge Voltage and depth of discharge, cycle life has improved to make their use possible in orbital applications. The end discharge Voltages can be seen in the above figure. With more than 22,000 cycles, the cells still have an energy efficiency of approximately 92%. This test has been operational for over 5 years, adding confidence to the calendar life of Lithium-ion chemistry. (Reproduction is by permission of Yardney/Lithion, with special thanks to the author, Chad Deroy.) +

Critical to the batteries success is a battery management system (BMS), which is either integral to the battery or integrated within the craft control system. Cell management can be monitored at the individual cell level or can be used to track performance for a group of four cells. BMS must provide for overcharge protection, safety and the regulation of charge via battery dedicated automatic control or space craft central control.

Data from representative cells shows excellent performance with high rate and orbital cycling. In one set of data, cells under test for 5 years displayed excellent performance for 18,000 LEO cycles which, in addition to validating cycle life, begins to address the question of calendar life. This work shows that Lithium-ion is moving into a workhorse role in aerospace power. These results will extend Lithium-ion battery use in aerospace an terrestrial applications.

17.4 High Capacity Li-ion BB-2590: Performance and Safety Characteristics: Military electronic packages employ standard replaceable battery packs which are updated to use new chemistries and internal configurations so that the energy density of the pack is extended. This is the case with the BB 2590 battery, a pack which had eight D size Lithium-ion cells which only utilized part of the internal space. To utilize more of the space, a larger diameter cell with the same height was designed, tested and implemented. This gave the battery a greater energy density.

This presentation provided the details of the design with safety testing and performance data to show the improved capabilities with the new design. Each cell is constructed with a three layer shutdown separator, a pressure activated circuit breaker which opens at 7 to 9 bar, and a coined vent which opens at 18- 21 bars to prevent rupture of the can. Short circuit testing showed a temperature rise to 600 C maximum. An internal protection circuit monitors Voltage, current and temperatures of all cells. The circuit interrupts when abusive conditions are encountered. Testing at temperatures to - 400 C was presented along with discharge capacity to 1.5C. Cycling data showed 94 percent capacity remaining after 200 cycles of full depth of discharge at a C/2 rate. It is anticipated that further improvements will lead to greater cell capacity beyond the present 6 Ah.

20.1 Large, Low Cost, Rapidly Configurable Lithium-ion Battery Modules Constructed from Small Commercial Cells: This presentation was previously reviewed in Part 2 (p.108-7). Large battery assemblies are built up from many 18650 commercial-off-the-shelf cells.

20.4 Development of a 300 Wh/kg Solid-State Rechargeable Lithium Battery: Front Edge Technology (FET) has successfully produced and improved microbatteries based on Oak-Ridge National Labs’ rechargeable lithium system LiPON technology. The basic cell has a lithium-metal anode, lithium-phosphorous-oxynitride (LiPON) ceramic electrolyte, and solid lithium-cobalt-oxide cathode. These cell materials are deposited as thin films on a thin mica substrate by vacuum sputter deposition. These 4.2 V cells can be stacked and range in size from 0.01 mAh to 2.5 mAh (one-hour discharge capacity). Simon Nieh discussed the characteristics of 2 to 4 cell stacks of 0.25 to 1.0 mAh capacity. Some highlights of the cells (batteries) are as follows: