Review of Ultracapacitor Technologies for Vehicle Applications

Transcription

1 1 Review of Ultracapacitor Technologies for Vehicle Applications Andrew Burke Institute of Transportation Studies University of California, Davis Davis, California Abstract Ultracapacitor technologies are reviewed with the emphasis on carbon-based devices using organic electrolytes. Such devices have cell voltages of 2.5-3V. The energy density of commercially available large device ( F) have usable energy densities of 3-4 Wh/kg with pulse power capability of 3-4 kw/kg. A small (3V, 47F) carbon-based laboratory prototype device has an energy density of 8.5 Wh/kg with a pulse power density of greater than 10 kw/kg. The conceptual design of ultracapacitor modules and energy storage units for vehicle applications indicated that the commercially available devices could be used effectively for engine starting to assist batteries in passenger cars and large class 8 trucks and as the power assist energy storage unit for hybrid-electric drivelines for passenger cars and transit buses. Introduction Electrical energy storage is needed in both vehicles using conventional engines and those using electric and electric-hybrid powertrains. Presently in most of these applications, batteries are used as the energy storage device. The batteries must be sized to provide both the energy (kwh) and power (kw) required in the application. In some cases this results in the battery far exceeding one of the two requirements, as in the case where the power required is high and the energy storage required is relatively low. In such cases, the battery is said to be sized by power and the energy stored follows from the mass of the battery and the energy density (Wh/kg) of the battery chemistry. In these cases, an energy storage device having a very high power density (W/kg) and relatively low energy density (Wh/kg) could be a more effective solution to the energy storage problem in terms of weight, volume, and cost than a battery. The ultracapacitor is such an electrical energy storage device. The development of ultracapacitors for automotive applications has been in progress since about In this paper, (1) the technology of ultracapacitors is reviewed, (2) performance data presented for several advanced carbonbased devices, and (3) illustrations of how advanced ultracapacitors can be used in several automotive applications discussed. Why Ultracapacitors? Ultracapacitors are being developed as an alternative to pulse batteries. To be an attractive alternative, ultracapacitors must have much higher power and much longer shelf and cycle life than batteries. By much is meant at least one order of magnitude

2 higher. Ultracapacitors have much lower energy density than batteries and their low energy density is in most cases the factor that determines the feasibility of their use in a particular high power application. The characteristics of a number of ultracapacitors and pulse batteries are given in Table 1. Two approaches to the calculation of the peak power density of the batteries are indicated in the table. The first and more standard approach is to determine the power at the so-called matched impedance condition at which one-half the energy of the discharge is in the form of electricity and one-half is in heat. The maximum pulse power at this point is given by Pmi = Voc 2 /4 R b where V oc is the open-circuit voltage of the battery and R b is its resistance. The discharge efficiency at this point is 50%. For many applications in which a significant fraction of the energy is stored in the energy storage device before it is used by the system, the efficiency of the charge/discharge cycle is important to the system efficiency. In those cases, the use of the energy storage device should be limited to conditions that result in high efficiency for both charge and discharge. The discharge/charge pulse power for a battery as function of efficiency is given by P ef = EF*(1-EF)*V oc 2 /Rb where EF is the efficiency of the high power pulse. For EF=.95, P ef /Pmi =.19. Hence in applications in which efficiency is a primary concern, the useable pulse power of the battery is much less than the peak power (Pmi) often quoted by the manufacturer for the battery. In the case of ultracapacitors, the peak pulse power during a discharge between Vo and Vo/2, where Vo is the rated voltage of the device, is given by Puc = 9/16*(1-EF)*Vo 2 /Ruc where Ruc is the resistance of the ultracapacitor. The expression shown above accounts for the reduction of voltage during the discharge of the device. Peak pulse power values are shown in Table 1 for both matched impedance and high efficiency discharges of the batteries and ultracapacitors. It is apparent that in nearly all cases the power from the ultracapacitors is much higher than that from the batteries. Note that it is not correct to compare the high efficiency power density for the ultracapacitors with the matched impedance power density for the batteries as is often done. The power capability of both types of devices is primarily dependent on their resistance and knowledge of the resistance is key to determining the peak useable power capability. Hence measurement of the resistance of a device in the pulsed mode of operation is critical to an evaluation of its high power capability. In addition to high power capability, the other reason for considering ultracapacitors for a particular application is their long shelf and cycle life. This is especially true of ultracapacitors using carbon electrodes. Most secondary (rechargeable) batteries if left on the shelf unused for many months will degrade markedly and be essentially useless after this time due to self-discharge and corrosion effects. Ultracapacitors will self-discharge over a period of time to low voltage, but they will retain their capacitance and thus be capable of recharge to their original condition. Experience has shown that ultracapacitors can be unused for several years and remain in nearly their original condition. Ultracapacitors can be deep cycled at high rates (discharge times of seconds) for 500,000 to 1,000,000 cycles with a relatively small 2

3 3 change in characteristics (10-20% degradation in capacitance and resistance). This is not possible with batteries even if the depth-of-discharge is kept small (10-20%). Hence relative to batteries, the advantages of ultracapacitors as pulse power devices are high power density, high efficiency, and long shelf and cycle life. The primary disadvantage of ultracapacitors is their relatively low energy density (Wh/kg and Wh/l) compared to batteries limiting their use to applications in which relatively small quantities of energy are required before the ultracapacitor can be recharged. Ultracapacitors can, however, be recharged in very short times (seconds or fraction of seconds) compared to batteries if a source of energy is available at the high power levels required. How do Ultracapacitors store energy? The simplest capacitors store the energy in a thin layer of dielectric material that is supported by metal plates that act as the terminals for the device. The energy stored in a capacitor is given by ½ CV 2, where C is its capacitance (Farads) and V is the voltage between the terminal plates. The maximum voltage of the capacitor is dependent on the breakdown characteristics of the dielectric material. The charge Q (coulombs) stored in the capacitor is given by CV. The capacitance of the dielectric capacitor depends on the dielectric constant (K) and the thickness (th) of the dielectric material and its geometric area (A). C = KA/ th In a battery, energy is stored in chemical form as active material in its electrodes. Energy is released in electrical form by connecting a load across the terminal of the battery permitting the electrode materials to react electrochemically with the ions required in the reactions to be transferred through the electrolyte in which the electrodes are immersed. The useable energy stored in the battery is given as VQ, where V is the voltage of the cell and Q is the electrical charge (It) transferred to the load during the chemical reaction. The voltage is dependent on the active materials (chemical couple) of the battery and is close to the open-circuit voltage (Voc) for those materials. An ultracapacitor, sometimes referred to as an electrochemical capacitor, is an electrical energy storage device that is constructed much like a battery (see Figure 1) in that it has two electrodes immersed in an electrolyte with a separator between the electrodes. The electrodes are fabricated from high surface area, porous material having pores of diameter in the nanometer (nm) range. The surface area of the electrode materials used in ultracapacitors is much greater than that used in battery electrodes being m 2 /gm. Charge is stored in the micropores at or near the interface between the solid electrode material and the electrolyte. The charge and energy stored are given by the same expressions as cited previously for the simple dielectric capacitor. However, calculation of the capacitance of the ultracapacitor is much more difficult as it depends on complex phenomena occurring in the micropores of the electrode. It is convenient to discuss the mechanisms for energy storage in ultracapacitors in terms of double-layer and psuedo-capacitance separately. The physics and chemistry of these processes as they apply to electrochemical capacitors are explained in great detail in Reference 1. In the following sections, the mechanisms are discussed briefly in terms of how they relate to the properties of the electrode materials and electrolyte.

4 4 Double-layer capacitors Energy is stored in the double-layer capacitor as charge separation in the doublelayer formed at the interface between the solid electrode material surface and the liquid electrolyte in the micropores of the electrodes. A schematic of an ultracapacitor is shown in Figure 1. The ions displaced in forming the double-layers in the pores are transferred between the electrodes by diffusion through the electrolyte. The energy and charge stored in the electrochemical capacitor are ½ CV 2 and CV, respectively. The capacitance is dependent primarily on the characteristics of the electrode material (surface area and pore size distribution). The specific capacitance of an electrode material can be written as C/gm = (F/cm 2 ) act * (cm 2 /gm) act where the surface area referred to is the active area in the pores on which the double-layer is formed. In simplest terms, the capacitance per unit of active area is given by (F/cm 2 ) act = (K /thickness of the double-layer) eff As discussed in Reference 1, determination of the effective (eff) dielectric constant K eff of the electrolyte and the thickness of the double-layer formed at the interface is complex and not well understood. The thickness of the double-layer is very small (a fraction of a nm in liquid electrolytes) resulting in a high value for the specific capacitance of uf/cm 2. For a surface area of 1000 m 2 /gm, this results in a potential capacitance of F/gm of electrode material. As indicated in Table 2, the measured specific capacitances of carbon materials being used in ultracapacitors are in most cases less than these high values being in the range of F/gm for aqueous electrolytes and F/gm using organic electrolytes, because for most carbon materials a relatively large fraction of the surface area is in pores that can not be accessed by the ions in the electrolyte. This is especially true for the organic electrolytes for which the size of ions is much larger than in an aqueous electrolyte. Porous carbons for use in ultracapacitors should have a large fraction of their pore volume in pores of diameter 1-5 nm. Materials with small pores (< 1nm) show a large fall-off in capacitance at discharge currents greater than 100 ma/cm 2 especially using organic electrolytes. Materials with the larger pore diameters can be discharged at current densities of greater than 500 ma/cm 2 with a minimal decrease in capacitance. The cell voltage of the ultracapacitor is dependent on the electrolyte used. For aqueous electrolytes, the cell voltage is about 1V and for organic electrolytes, the cell voltage is V. Electrochemical capacitors utilizing pseudo-capacitance For an ideal double-layer capacitor, the charge is transferred into the double-layer and there are no Faradaic reactions between the solid material and the electrolyte. In this case, the capacitance (dq/dv) is a constant and independent of voltage. For devices that ultilize pseudo-capacitance, most of the charge is transferred at the surface or in the bulk near the surface of the solid electrode material. Hence in this case, the interaction between the solid material and the electrolyte involves Faradaic reactions which in most instances can be described as charge transfer reactions. The charge transferred in these reactions is voltage dependent resulting in the pseudo-capacitance (C = dq/dv) also being voltage dependent. Three types of electrochemical processes have been utilized in the development of ultracapacitors using pseudo-capacitance. These are surface

5 5 adsorption of ions from the electrolyte, redox reactions involving ions from the electrolyte, and the doping and undoping of active conducting polymer material in the electrode. The first two processes are primarily surface mechanisms and are hence highly dependent on the surface area of the electrode material. The third process involving the conducting polymer material is more of a bulk process and the specific capacitance of the material is much less dependent on its surface area although relatively high surface area with micropores is required to distribute the ions to and from the electrodes in a cell. In all cases, the electrodes must have high electronic conductivity to distribute and collect the electron current. An understanding of the charge transfer mechanism can be inferred from C(V), which is often determined using cyclic voltammetry. For assessing the characteristics of devices, it is convenient to use the average capacitance (C av ) calculated from C av = Qtot/Vtot where the Qtot and Vtot are the total charge and voltage change for a charge or discharge of the electrode. This permits a determination of the specific capacitance (C av /gm) of the material for the electrolyte of interest. As shown in Table 2, the specific capacitance of pseudo-capacitance materials are much higher than that of carbon materials. It is thus expected that the energy density of devices developed using the pseudo-capacitance materials would be higher. Status of the Development of Carbon-based Ultracapacitors Development has continued around the world on advancing the state-of-the-art of ultracapacitors (Reference 2). This work has included development of both carbon-based (double-layer) devices and devices using mixed metal oxide (psuedocapacitance). This paper will focus on the recent developments of carbon-based devices. Other papers at this conference will consider psuedocapacitance devices. The physical/chemical mechanisms responsible for energy storage in carbon-based devices are relatively simple. In addition, calculation of their performance and scaling to various sizes are relatively straight-forward once the properties of the active carbon and electrolyte are known. For carbon-based devices, there is minimum concern about either cycle or calendar life, because the energy storage mechanisms do not involve Faradaic processes. Charging of the carbon-based devices is also straight-forward with the rate of charging or time of charging having a small effect on their capacitance. Finally, technologies have been developed that have achieved or show promise of achieving both high energy density (5 Wh/kg) and high power density (2 kw/kg) in the same device. Even higher energy and power density appear to be likely using carbon as the active material. For these reasons, there is reason to be optimistic that carbon-based ultracapacitors can be marketed successful for the applications discussed later in the paper. The primary obstacle in the path of such commercialization is the high cost of the microporous carbon required for the devices. Except for very small devices, greatly reducing the cost of the carbon and to a lesser extent, the electrolyte for devices that use an organic electrolyte, are key to successfully marketing ultracapacitors. During the past year, I have had the opportunity to test in the EV Power Systems Laboratory at UC Davis several new ultracapacitor devices that are significantly advanced over devices I had previously available for testing. These devices were developed by Ness in Korea, Panasonic in Japan, and Superfarad/Skeleton Technologies

6 6 in Sweden and the Ukraine. The Ness and Panasonic devices were provided to UC Davis by the companies for evaluation. Extensive testing of the devices has been done over the last several months. The results of the testing are given in detail in Reference (3). The physical and performance characteristics of the devices are summarized in Table 3. All the devices use particulate carbon as the active material and an organic electrolyte. The Ness and Panasonic devices, which are fully packaged and suitable for use in application testing, are large with capacitances of 2500F and 1200F, respectively. The Skeleton Technologies device is a relatively small, lab-fabricated device (45F) and is not packaged for application testing. Photographs of the devices are shown in Figures 2-4. The testing of these devices indicated clearly that progress on ultracapacitor development is continuing to be made. It will be shown in the next section that in terms of performance these devices can be utilized to advantage in several automotive applications. A comparison of the performance of the three new ultracapacitor devices is shown in Figure 5 in which the higher performance of the prototype device from Skeleton Technologies is evident. The test data are for the constant power discharge of the devices for voltages between 3V and 1.5V and thus represents the useful energy that can be taken from the devices in a vehicle application. The ideal energy density (E=1/2CV2) for the devices would be at least 25% higher than that shown in Figure 5. The test data for the prototype device show the large advantage of a short RC time constant and low resistance on the power capability of an ultracapacitor. The prototype device has a power capability (W/kg) nearly a factor of ten greater than the other devices. The test data for the new devices indicate that useable energy densities of close to 10 Wh/kg and constant-power power densities of 3-4 kw/kg are achievable in carbon-based ultracapacitors. Automotive Applications of Ultracapacitors The following automotive-transportation applications will be discussed in this section: Engine starting 12V and 36-42V systems Hybrid-electric vehicles cars and buses Engine Starting Ultracapacitors can be used to augment the power from batteries for starting engines for both passenger cars and trucks. The two cases considered will be a mid-size passenger car and a large Class 8 truck. In the case of the passenger car, ultracapacitor modules that can be used with both the standard 12V and the future 36-42V systems will be discussed. In the case of the trucks, multiple capacitor modules are needed as multiple batteries are used to get adequate power at low, sub-zero ambient temperatures. The ultracapacitor characteristics used in defining the modules to be used with the batteries are those of the Ness capacitors that are presently commercially available. The other capacitors cited in the previous section could also be used, but the low resistance and high power capability of the Ness devices are particularly well-suited for this application. In sizing the ultracapacitors for the engine starting applications, it is necessary to know the usable energy storage (Wh) and the maximum power requirements. In the case of the passenger car, for normal starting of the engine, the maximum power is sustained for one second or less. For cold starting or repeated attempts to start the engine, more energy is required. In this study, the energy requirement was calculated based on a

7 maximum power of 6.5 kw for 5 seconds or 9 Wh. Assuming that 75% of the energy stored in the capacitor was useable, this resulted in a total energy storage requirement of 12 Wh for the passenger car. After discussions with a manufacturer of large trucks, the corresponding requirements for the truck were taken to be 10kW for 10 seconds, which results in a useable energy of 28 Wh and a total stored energy of 37 Wh. For the 12V systems, the passenger car requirements can be meet with a single module and for the large truck, it seemed appropriate to use three modules of the same design as used in the passenger car case. The ultracapacitor module would consist of five (5) cells connected in series. Each cell would have a capacitance of 2600F and be rated for use at 2.6V. The weight and volume of the cells in the module would be 3.25 kg and 2.67 liters, respectively. The corresponding energy density and pulse power for the modules are 3.7 Wh/kg and 2 kw/kg, respectively. These characteristics meet the power and energy requirements for both passenger car and truck cases. Using the equation cited previously for the calculation of pulse power from an ultracapacitor, the maximum power capability of the capacitor module at 25 deg C is very high being about 19 kw for a 20% IR drop (EF=.8) and 9.5 kw for a 10% IR drop (EF=.9). It is expected that the power capability will be significantly less at sub-zero temperatures where the ionic conductivity of the organic electrolyte will be reduced. At 20 deg C, the conductivity is reduced by 46% and at 40 deg C, it is reduced by 63%. Assuming the resistance of the cells is increased proportional to the decrease in conductivity, the pulsed power capability of the cells/modules will be reduced to 54% and 37% of the values at 25 deg C for 20 deg C and 40 deg C, respectively. This would result in a peak pulse power of about 7kW per module at 40 deg C. There is a considerable body of data (References 4-5) for capacitors tested at low temperatures between 20 and 40 deg C that shows the effect of temperature on device performance. The data indicate that the effect of temperature is design specific with some devices showing a large reduction in performance and others showing a relatively small reduction in performance. These differences between devices are particularly true of devices using an organic electrolyte. Tests of carbon/organic electrolyte devices (Reference 4,5) available from Panasonic and Ness at temperatures down to 30 deg C showed that the capacitance of the devices was essentially unchanged as the temperature was decreased and the increase in resistance was relatively small. This means that the energy storage of the five-cell capacitor module will not be significantly reduced at low temperatures. The characteristics of the capacitor units to be used in the engine starting applications are summarized in Table 4. Schematics of the 12V battery-capacitor configurations for the passenger car and large truck applications are shown in Figure 6. In both cases, the weights and volume of the capacitors are small compared to that of the standard batteries. It is likely that the size of the batteries can be reduced when they are combined with the capacitors, but the magnitude of the reduction depends on what fraction of the energy stored in the battery is needed to supply the accessory loads. That is a systems decision to be made by the vehicle designers. It is of interest to compare the results of the present analysis of the use of capacitors with batteries to start the engine in a large truck with those given in References (6-8). In Reference (8), the engine starting cycle was simulated for 20 deg C in detail using computer software and an assumed current vs. time profile. The capacitor performance was specified in terms of the total energy stored (kj) and the RC time constant (1 sec) of the capacitor. The results of that 7

8 8 study are completely consistent with those of the present study in that it was concluded that a capacitor having an energy storage capacity of 117 kj and a resistance of.65 mohm at 20 deg C could start the engine without the assistance of batteries. As shown in Figure 6, the capacitor modules are connected in parallel with the batteries without any interface electronics. In that case the battery and capacitors will share the starting current for the engine with the capacitor taking the larger fraction depending on its resistance and state-of-charge (voltage). The battery fraction of the current will increase as the capacitor discharges in order to match the capacitor voltage. The capacitors will be recharged by the battery at a rate (current) dependent on the combined resistance of the battery and capacitors. Further study is needed to determine the cost and performance trade-offs provided by using simple electronics to control the current from the battery during both engine starting and recharging the capacitors. Hopefully, the further work will indicate that no electronics is needed for long and reliable system operation. Consideration is presently being given to the use of volt systems in passenger car applications. Hence it is of interest to conceptual design a capacitor module for use with the volt electrical systems. The building block for such a module could be a 1500F cell similar to the 2600F cell used in the modules for the 12V module. Each of the 1500F cells would store 1.4 Wh of energy, weigh.375 kg, and have a resistance of.35 mohm at 25 deg C. The 21V module would consist of 8 cells, weigh 3 kg, and store 11 Wh of energy (8.4 Wh useable). The resistance of the module would be 2.8 mohm with a resultant calculated peak pulse power of about 17 kw. Two of the 21V modules would be used with the V battery system. The weight of the capacitor cells would be 6 kg, which would be much less than that of the batteries. The capacitors could provide most of the peak power required from the system and the batteries would recharge the capacitors during periods of relatively low system power demand. The capacitors could also be used to recover energy during regenerative braking if the vehicle was so equipped. The characteristics of the modules for this system are also shown in Table 4. Hybrid Vehicle Applications Ultracapacitors can also have applications in hybrid-electric drivelines using either an engine-generator or a fuel cell to generate on-board electricity. It is most likely that ultracapacitors would be considered for hybrid vehicles that utilize a power assist control strategy. In those cases, the engine, engine-generator, or fuel cell would load follow the power demand except for very rapid approaches to near peak demand at which times the capacitors would provide assistance. In addition, the capacitors would be utilized to recover energy during regenerative braking. Capacitor units for use in hybrid vehicles must be high voltage. For cars with electric drive systems, the voltage is likely to be about 300V and for buses about 600V. This would require about 125 cells in series for a car and 250 cells in series for a bus. If one used the same cells and modules as discussed previously for the 12V passenger car application, the hybrid-electric car would require 25 12Vmodules and the bus would require 50 modules. Modules having more than five (5) 2600F cells could be assembled to reduce the number of modules, but the performance of the high voltage units would be essentially the same. The energy stored per cell is 2.44 Wh at a rated voltage of 2.6V resulting in a total energy storage per string

9 9 in the high voltage unit of 305 Wh for the hybrid car and 610 Wh for the hybrid bus. The useable energy in each case is only 75% of the total. In the case of the car, the useable energy stored per string is about 30% less than the 300Wh specified in the PNGV design targets for a power assisted hybrid (Reference 9). This shortfall in energy stored could be made up by either increasing the rated cell voltage to 3V or the capacitance of the cells to 3500F. The later approach would increase the weight of the ultracapacitor unit for the car from 81 kg to 114 kg. Increasing the cell voltage from 2.6 to 3V seems like a better option because capacitor developers are already working to make this possible and the present designs are routinely tested at 3V in pulse testing of single cells. The power required by PNGV of the energy storage unit is 30-40kW. The peak power density required of the unit to meet this power would be less than 500 W/kg which is less than the constant-power peak power density capability of the Ness capacitors (see Figure 5). The weight (81 kg) and volume (67 liters) of the capacitor unit cells are about twice the weight (40 kg) and volume (32 liters) given in the PNGV specifications. Ultracapacitors with an energy density of about 10 Wh/kg (twice that of the present cells) would be required to meet the PNGV targets for weight and volume. The Ragone curves given in Figure 5 indicate that increases in energy density of large carbon-based capacitors are likely in the near-future. Each high voltage string of capacitors for the hybrid-electric bus application would store Wh of useable energy. Simulation results for buses using ultracapacitors given in Reference (10). The results indicate that for a bus utilizing a series hybrid driveline, the energy storage requirement is Wh. Hence least two 600V strings of the capacitors would be necessary to store sufficient energy for the hybrid bus application. Using two strings of 3V, 3500F cells, the unit would have a rated voltage of 750V and store 2200 Wh of energy. The weight of the unit would be 438 kg and it could provide 200 kw of power for both motoring and regenerative braking. Such a unit would be much lighter than the lead-acid battery packs currently used in hybridelectric buses (References 11, 12) and have a power capability much greater especially for regenerative braking. The low resistance of the capacitor units would also result in lower losses in storing the energy and as a result higher fuel economy than would be the case using batteries (References 10,13). The analyses of passenger car and bus hybrid-electric vehicle applications in this section of the paper indicate that ultracapacitors can be used effectively. The results of the analyses are summarized in Table 5. The key issues concerning whether ultracapacitors can be used in hybrid vehicles are no longer concerned with device performance, but rather the cost of the devices and more specially the unit cost ($/kg) of the carbon used to manufacturer the devices. There is, of course, a trade-off between device performance and material costs, which means that reasonable priced carbon must be developed for the high performance devices used in the present analyses. References 1. Conway,B.E., Electrochemical Capacitors: Scientific Fundamentals and Technological Applications, Kluwer Academic/Plenum, Burke, A.F., Ultracapacitors: why, how, and where is the technology, Journal of Power Sources, 91 (2000) 37-50

10 3. Burke, A.F. and Miller, M., Characteristics of Advanced Carbon-based Ultracapacitors, Proceedings of the 10 th International Seminar on Double-layer Capacitors and Similar Energy Storage Devices, Deerfield Beach, Florida, December Burke, A.F., Electrochemical Capacitors for Electric Vehicles: A Technology Update and Recent Test Results from INEL, Proceedings of the 36 th Power Sources Conference, Cherry Hill, N.J., June Miller, J.R., Property and Performance Measurements of Ness Electrochemical Capacitors, report to Ness, Soowon, Korea, November Miller, J.R. and etals, Truck Starting using Electrochemical Capacitors, SAE paper , Indianapolis, Ind., November Ong, W. and Johnston, R.H., Electrochemical Capacitors and Their Potential Application to Heavy Duty Vehicles, SAE paper xxxxxx, Portland, Oregon, December Miller, J.R., Engineering Battery-Capacitor combinations in High Power Applications: Diesel Engine Starting, Proceedings of the 9 th International Seminar on Double-layer Capacitors and Similar Energy Storage Devices, Deerfield Beach, Florida, December Sutula, R.A. and etals, Recent Accomplishments of the Electric and Hybrid Vehicle Energy Storage R&D Programs at the United States Department of Energy: Status Report, paper presented at the 17 th International Electric Vehicle Symposium, Montreal, Canada, October Burke, A.F. and Blank, E., Electric/Hybrid Transit Buses using Ultracapacitors, Proceedings of Electric Transportation, Power Systems 95, Long Beach, California, September Hybrid-Electric Transit Buses: Status, Issues, and Benefits, TCRP Report 59, Transportation Research Board, King, R.D. and etals, Heavy Duty (225 kw) Hybrid-Electric Propulsion System for Low-Emission Transit Buses Performance, Emissions, and Fuel Economy Tests, Proceedings of the 14 th International Electric Vehicle Symposium, Orlando, Florida, December Burke, A.F. and Miller, M., Assessment of the Greenhouse Gas Emissions Reduction Potential of Ultra-clean Hybrid-Electric Vehicles, Report No. UCD- ITS-RR 97-21, December

11 11 (1) Table 1: Comparisons of the Performance Characteristics of Various Ultracapacitors and High Power Batteries Ultracapacitor Devices SKT 47F Ness 2600F Panasonic 1200F 800F Maxwell 2700F Montena 1800F V Ah Weight (kg) Resist. (mohm) Wh/kg W/kg 95% Effic. W/kg Match. Imped >80K (unpack.) Batteries Devices Panasonic Nmthd (spirwd) Nmthd (prismt.) Ovonic Nimthd Hawker Pb-acid Optima Pb-acid Bolder Technol. Pb-acid Shin Kobe Lithium ion

12 12 Table 2: Specific Capacitance of Selected Electrode Materials Material Carbon cloth Carbon black Aerogel carbon SKT (1) Particulate carbon from SiC Density gm/cm3 Electrolyte F/gm F/cm3.35 KOH Organic KOH KOH Organic.7 KOH Organic SKT Particulate carbon from TiC Anhydrous RuO2 Hydrous RuO2 Doped conducting polymers.7 KOH Organic Sulfuric acid Sulfuric acid.7 Organic (1) Nano-porous carbons produced from carbides by Skeleton Technologies

13 13 Table 3: Summary of the characteristics of advanced prototype and commercially available carbon-based ultracapacitors Device V C R RC W/kg W/kg Wgt. Vol. (F) (mohm) (sec) (95%) Match. (kg) lit. (1) (2) Imped. Skeleton Techn.R4* >80K Maxwell Ness Panasonic Montena *unpackaged (1) Energy density based on E=1/2CV2, Vrated=3V (2) Power based on P=9/16*(1-EF)*V2/R, EF=efficiency of discharge, V=3V Table 4: Capacitor Module/System Characteristics for Engine Starting Applications Application System Voltage Batteries No. Ah Wgt. (kg) Capacitor Modules No. V Wh Pulse Total Power (2) (kw) (3) Wgt. (kg) Passenger car 12V Passenger car 36-42V (4) Class 8 Truck 12V (2) Total energy stored in the capacitor unit (useable energy 75% of total) (3) Pulse power at EF=.8, 25 deg C (4) Battery for the36-42v system assumed to provide twice the energy storage and pulse power of the standard 12V battery and have an energy density of 40 Wh/kg

14 14 Table 5: Capacitor Cell/Unit Characteristics for Hybrid-electric Passenger Car and Transit Bus Applications Application System Voltage Capacitor cell Capacitor unit V F Wh Cells No. Wh Wgt. Per Of Total (kg) String Strings (1) (2) Passenger Car Transit Bus (1) Total energy stored in the unit (useable energy is 75% of total) (2) Maximum steady power at 750W/kg Max. Power

15 15

16 16 Figure 2: Photograph of the Skeleton Technologies 2.3v, 46F Capacitor Figure 3: Photograph of the Ness 2.3V, 2500F Capacitor

17 Figure 4: Photograph of the Panasonic 2.3V, 1200F Capacitor 17

18 18

19 19