Energy Storage (ES) at the system level
Jun 16, 2023
This whitepaper will provide a solid overview of energy storage (ES) fundamentals, with a focus on ES solutions at the system level. Gain insights into capacitors vs. batteries, primary vs. secondary storage, and more.
1. How complicated can a two-terminal DC source be?
Have you ever stared at a battery that is located in your car or some handheld electronic device and wondered what the big deal is? As electrified transportation quickly envelops and drives the entire electronics industry, it seems more and more common to hear of not only economics but also the bleeding-edge designs and even the full supply chains (from raw material sourcing to end-of-life recycling) completely revolving around these devices. As we deploy many billions of tiny systems, commonly referred to as the Internet of Things (IoT)/ Industrial IoT (IIoT), the ability to run entire wireless sensor networks (WSN) off a tiny capacitor is an exciting prospect intertwined with the apocalyptic scenario of tossing ~100M batteries A DAY into the landfills of the world.
But even as one starts to see batteries take the center stage in many critical applications and market-driving use cases, they may still look at this basic-looking DC-source with only two terminals, and have a hard time resolving what all the fuss around batteries is about. As we shall come to see over the course of this article, there are plenty of reasons why batteries (and energy storage or ES, overall) should never be taken for granted. The first, obvious reason for this is that nothing in the electronic world runs without power, so being able to source and store energy is a necessity. Perhaps, not top of the mind until the last decade or so, but safety is a mission-critical requirement in just about any application, and therefore, it can be closely tied to the densest source of energy in the system. Even further from the purview of the many design considerations is the strategy of utilizing energy storage at multiple levels to reduce system sizes, maximize uptimes, and reduce the overall infrastructure (capital and operational expenditure or CAPEX/OPEX) savings.
ES is a huge topic that cannot be done justice to in a single, brief whitepaper, even though this one will serve to provide a solid overview of the fundamentals. But even fundamentally, focusing on basic physics alone still casts a very wide net to comprehensively cover all the major categories from MEMS-scale to utility-scale storage. Energy can be stored and converted from every type afforded to us by the physical world (even if not always pragmatic), whether be it in the form of electrochemical reaction (or other means of electrical potential), heat, light, radio frequency (RF), or potential energy/ motion/ electrodynamics. Wikipedia defines “energy storage” [1] as “…the capture of energy produced at one time for use at a later time to reduce imbalances between energy demand and energy production.”
Though this whitepaper will focus on energy storage solutions at the system level, it should be noted that thinking of all forms of energy as conservation that merely transfers/changes state, with some typically spent in the conversion/transfer process, is a worthwhile approach. This is not meant to sound like a fancy paraphrasing of the first law of thermodynamics but is rather intended to serve as an inspirational mantra that drives one to seek the perpetual optimization of energy storage and consumption as an overarching and constant objective.
Another area in which energy storage tends to be an outlier, compared to most of the other aspects of system advancement, is the rate at which significant, generational improvements in energy density occur. Moore’s Law keeps logic shrinking at an exponential pace and microelectromechanical systems (MEMS) shrink and integrate sensors to the point of being nearly invisible to the naked eye. Unfortunately, this kind of exponential improvement from one generation to the next does not bode well for energy storage solutions. Energy storage is directly limited by chemistry and physics, which tend to double on a timeline closer to a decade than a year. In general, the main goal in implementing and optimizing an energy storage solution is to keep the chemistry happy, but we shall discuss more what this means later.
As alluded to earlier, energy storage has an ever-increasing role in driving sustainable electronics and in the fight against climate change. Even rechargeable batteries tend to contain hazardous materials and may require rare-earth elements, whose global, lifetime supply exceeds demand (e.g., recycling is critical here), and may consume a whole lot of landfill space. As the number of non-tethered applications/ devices increases exponentially, so will all the associated energy storage, which multiplies the sustainability challenges summarized here. One apparent strategy to address this is by trying and converting the non-rechargeable to rechargeable energy storage solutions to mitigate much of the replacement efforts. Given the increased focus and attention on sustainability topics these days, even the very conservative industries dictating massive amounts of ES (with redundancy) are looking for ways to utilize energy storage more intelligently to meet uptime targets, while still reducing overall infrastructure.
Now that the bar has been set somewhat, let us shift our focus to the most ubiquitous forms of system-level energy storage - and consider what they are, how to deal with them, and the key items to look for while optimizing a system design.
But even as one starts to see batteries take the center stage in many critical applications and market-driving use cases, they may still look at this basic-looking DC-source with only two terminals, and have a hard time resolving what all the fuss around batteries is about. As we shall come to see over the course of this article, there are plenty of reasons why batteries (and energy storage or ES, overall) should never be taken for granted. The first, obvious reason for this is that nothing in the electronic world runs without power, so being able to source and store energy is a necessity. Perhaps, not top of the mind until the last decade or so, but safety is a mission-critical requirement in just about any application, and therefore, it can be closely tied to the densest source of energy in the system. Even further from the purview of the many design considerations is the strategy of utilizing energy storage at multiple levels to reduce system sizes, maximize uptimes, and reduce the overall infrastructure (capital and operational expenditure or CAPEX/OPEX) savings.
ES is a huge topic that cannot be done justice to in a single, brief whitepaper, even though this one will serve to provide a solid overview of the fundamentals. But even fundamentally, focusing on basic physics alone still casts a very wide net to comprehensively cover all the major categories from MEMS-scale to utility-scale storage. Energy can be stored and converted from every type afforded to us by the physical world (even if not always pragmatic), whether be it in the form of electrochemical reaction (or other means of electrical potential), heat, light, radio frequency (RF), or potential energy/ motion/ electrodynamics. Wikipedia defines “energy storage” [1] as “…the capture of energy produced at one time for use at a later time to reduce imbalances between energy demand and energy production.”
Though this whitepaper will focus on energy storage solutions at the system level, it should be noted that thinking of all forms of energy as conservation that merely transfers/changes state, with some typically spent in the conversion/transfer process, is a worthwhile approach. This is not meant to sound like a fancy paraphrasing of the first law of thermodynamics but is rather intended to serve as an inspirational mantra that drives one to seek the perpetual optimization of energy storage and consumption as an overarching and constant objective.
Another area in which energy storage tends to be an outlier, compared to most of the other aspects of system advancement, is the rate at which significant, generational improvements in energy density occur. Moore’s Law keeps logic shrinking at an exponential pace and microelectromechanical systems (MEMS) shrink and integrate sensors to the point of being nearly invisible to the naked eye. Unfortunately, this kind of exponential improvement from one generation to the next does not bode well for energy storage solutions. Energy storage is directly limited by chemistry and physics, which tend to double on a timeline closer to a decade than a year. In general, the main goal in implementing and optimizing an energy storage solution is to keep the chemistry happy, but we shall discuss more what this means later.
As alluded to earlier, energy storage has an ever-increasing role in driving sustainable electronics and in the fight against climate change. Even rechargeable batteries tend to contain hazardous materials and may require rare-earth elements, whose global, lifetime supply exceeds demand (e.g., recycling is critical here), and may consume a whole lot of landfill space. As the number of non-tethered applications/ devices increases exponentially, so will all the associated energy storage, which multiplies the sustainability challenges summarized here. One apparent strategy to address this is by trying and converting the non-rechargeable to rechargeable energy storage solutions to mitigate much of the replacement efforts. Given the increased focus and attention on sustainability topics these days, even the very conservative industries dictating massive amounts of ES (with redundancy) are looking for ways to utilize energy storage more intelligently to meet uptime targets, while still reducing overall infrastructure.
Now that the bar has been set somewhat, let us shift our focus to the most ubiquitous forms of system-level energy storage - and consider what they are, how to deal with them, and the key items to look for while optimizing a system design.
2. Capacitors vs. Batteries
Batteries have longer discharge/charge cycles, higher energy densities, and improved self-discharge (i.e., internal resistance) characteristics compared to capacitors, but capacitors have a longer cycle lifetime than batteries.
The basic equation for a capacitor is shown below, demonstrating practically any two parallel, charged plates constituting a capacitive ES device. Any device storing energy in an electrical field in this manner is considered a capacitor. For comparison, the analogous energy storage in a magnetic field constitutes an inductor.
The basic equation for a capacitor is shown below, demonstrating practically any two parallel, charged plates constituting a capacitive ES device. Any device storing energy in an electrical field in this manner is considered a capacitor. For comparison, the analogous energy storage in a magnetic field constitutes an inductor.
Fig. 1: Capacitive Equation with Diagram, image courtesy of Wikimedia Commons [2]
A battery is an electrochemical capacitor that stores charge on the plates of dissimilar materials by providing a source of electrons that migrate from cathode (+) to anode (–), thus generating potential current across the terminals. The dissimilar materials form half-cell potentials that, when combined, form the battery potential (a.k.a., open-circuit terminal voltage). The combination of so many different materials is what lends to such a wide spectrum of batteries and other ES modalities.
The battery’s characteristics are completely tied to the cathode, anode, and dielectric material chemistry. Usable energy is generated via what is known as a “redox reaction” (short for reduction-oxidation, referring to the gain and loss, respectively, of electrons), which releases energy as a result of the free energy derived from the electron transfer. Every aspect of design from the chemistry to the geometry of the device impacts its energy performance and usable lifetime.
One may also refer to batteries as having wet or dry cells (or “wet/dry chemistry”). This refers to the physical state of the dielectric material. If it is a wet cell, then the dielectric is in liquid form under operating application temperatures. Wet cells typically require more careful handling to ensure that the electrolyte is not spilled/leaked, and may require regular maintenance, such as is the case with lead-acid batteries you might find in a car, boat, or home photovoltaic (PV) system. A dry cell uses a solid-state electrolyte that does not have any liquid moving around, which tends to make them more amenable to unique geometries and safer than its wet-cell counterparts. In general, this is because a dry cell is less likely to perpetuate a short between a cathode/anode or send hot, toxic electrolyte flying into someone’s face in the case of a catastrophic failure. There are some hybrid options, such as the “gel cell” or semi-solid electrolyte (i.e., polymer gel used in Li-polymer or Li-po batteries), but digging into those is beyond the scope of this paper.
The chemistry and the type of capacitor or battery at hand are not only important for the electrical design and management of a system, but they can also be the deciding factor in the overall system geometry. The characteristics described above can determine any number of form-factor options for your ES solution. This is probably one of the most perplexing aspects of ES devices, as solutions with seemingly very similar figures of merit (FOM) can vary widely in terms of size, cost, and performance. Some chemistries work much better for high-temperature environments of use in cases that require many charge cycles, whereas some are better for safety or for enabling flexible, conforming structures. Even geometries can change over time, such as in the case of Li-ion packs that can swell as a result of outgassing over time, with the accumulation of gas resulting from the redox reaction.
It can be quite surprising to note how slight changes in some FOM can have drastic impacts on life, especially the operating temperature and discharge rate (among many others to be reviewed a little later). For instance, the DC voltage rating on a multilayer ceramic capacitor (MLCC) takes on very different meanings based on the class and Electronic Industries Alliance (EIA) code. While two different MLCCs can be rated for 0.47μF at 16V with +/-5% tolerance operating from -55 to 125°C, the one with the X7R rating will have ~20% of its capacitance at 12VDC, whereas the one with the C0G rating will have ~90% of its capacitance at 12VDC. This difference can be just as stark over the operating temperature range.
When evaluating ES options, a critical area to consider is the use of the ES device over time. What is the operating environment (particularly temperature)? How many charge cycles can it experience? Is this a chemistry that has special requirements for charging and/or balancing multiple cells in a battery pack? There are so many factors that impact the USABLE (i.e., not rated) capacitance and the life-cycle performance of ES devices as well as tying directly to safety and fault-handling.
ES solutions and DC/DC converters go hand-in-hand. One cannot take advantage of the full capacity, optimal energy efficiency conversion, and maximum reliability without the use of such converters. While a converter should be used to stabilize the ES solution’s output voltage, it should also be lightweight (particularly for portable equipment) and support a wide-input voltage range to utilize the maximum energy from the battery or capacitor. High converter efficiency will support all these objectives, while also ensuring that the converter itself is not a primary contributor to the overall system power budget (via its conversion losses). Great examples of such a solution are R-78B-2.0 and RPMB-2.0.
The battery’s characteristics are completely tied to the cathode, anode, and dielectric material chemistry. Usable energy is generated via what is known as a “redox reaction” (short for reduction-oxidation, referring to the gain and loss, respectively, of electrons), which releases energy as a result of the free energy derived from the electron transfer. Every aspect of design from the chemistry to the geometry of the device impacts its energy performance and usable lifetime.
One may also refer to batteries as having wet or dry cells (or “wet/dry chemistry”). This refers to the physical state of the dielectric material. If it is a wet cell, then the dielectric is in liquid form under operating application temperatures. Wet cells typically require more careful handling to ensure that the electrolyte is not spilled/leaked, and may require regular maintenance, such as is the case with lead-acid batteries you might find in a car, boat, or home photovoltaic (PV) system. A dry cell uses a solid-state electrolyte that does not have any liquid moving around, which tends to make them more amenable to unique geometries and safer than its wet-cell counterparts. In general, this is because a dry cell is less likely to perpetuate a short between a cathode/anode or send hot, toxic electrolyte flying into someone’s face in the case of a catastrophic failure. There are some hybrid options, such as the “gel cell” or semi-solid electrolyte (i.e., polymer gel used in Li-polymer or Li-po batteries), but digging into those is beyond the scope of this paper.
The chemistry and the type of capacitor or battery at hand are not only important for the electrical design and management of a system, but they can also be the deciding factor in the overall system geometry. The characteristics described above can determine any number of form-factor options for your ES solution. This is probably one of the most perplexing aspects of ES devices, as solutions with seemingly very similar figures of merit (FOM) can vary widely in terms of size, cost, and performance. Some chemistries work much better for high-temperature environments of use in cases that require many charge cycles, whereas some are better for safety or for enabling flexible, conforming structures. Even geometries can change over time, such as in the case of Li-ion packs that can swell as a result of outgassing over time, with the accumulation of gas resulting from the redox reaction.
It can be quite surprising to note how slight changes in some FOM can have drastic impacts on life, especially the operating temperature and discharge rate (among many others to be reviewed a little later). For instance, the DC voltage rating on a multilayer ceramic capacitor (MLCC) takes on very different meanings based on the class and Electronic Industries Alliance (EIA) code. While two different MLCCs can be rated for 0.47μF at 16V with +/-5% tolerance operating from -55 to 125°C, the one with the X7R rating will have ~20% of its capacitance at 12VDC, whereas the one with the C0G rating will have ~90% of its capacitance at 12VDC. This difference can be just as stark over the operating temperature range.
When evaluating ES options, a critical area to consider is the use of the ES device over time. What is the operating environment (particularly temperature)? How many charge cycles can it experience? Is this a chemistry that has special requirements for charging and/or balancing multiple cells in a battery pack? There are so many factors that impact the USABLE (i.e., not rated) capacitance and the life-cycle performance of ES devices as well as tying directly to safety and fault-handling.
ES solutions and DC/DC converters go hand-in-hand. One cannot take advantage of the full capacity, optimal energy efficiency conversion, and maximum reliability without the use of such converters. While a converter should be used to stabilize the ES solution’s output voltage, it should also be lightweight (particularly for portable equipment) and support a wide-input voltage range to utilize the maximum energy from the battery or capacitor. High converter efficiency will support all these objectives, while also ensuring that the converter itself is not a primary contributor to the overall system power budget (via its conversion losses). Great examples of such a solution are R-78B-2.0 and RPMB-2.0.
3. Primary vs. secondary storage
When it comes to ES systems, one may hear of them being classified as either primary or secondary storage. This classification refers to their recharge capabilities. A primary device is considered non-rechargeable, whereas a secondary device is considered rechargeable. It is common to see these terms used interchangeably (i.e., primary/non-rechargeable or secondary/rechargeable).
Aside from the obvious characteristic of rechargeability, the distinction between primary and secondary storage devices is important for many of the operational and handling parameters previewed above and detailed much more in the following section. Therefore, they must be treated differently in terms of technical operation and economics. For instance, the maintenance costs of secondary cells can be significant, but they must be weighed against the cost of primary replacements and the possible disruption of service.
Primary cells tend to have a greater energy density than their secondary counterparts, but they must also be replaced if a single solution will not meet the operational lifetime of the system/application. A primary battery may be considered more reliable in a given application, but a secondary battery can be used to reduce peak power requirements, thereby reducing stress on the other components and improving the overall system reliability.
Secondary batteries and even capacitor banks can have special requirements for charge cycles to maximize life and performance. Charging may require a very complicated series of controlling different voltage/current combinations. If multiple cells are involved, then there may be additional requirements to balance cell charge for stable operation and reliability.
A battery management system (BMS) can be very advantageous for system designers interested in getting the best out of their ES. A BMS can allow for the successful operation of the battery without it being burdened by the circuitry and specialized control described throughout this white paper. If all of these factors seem overwhelming, then the good news is that there are many solutions available to help make the BMS more of a turnkey solution. Numerous semiconductor manufacturers are offering BMS ICs or even integrating the functionality, along with other energy management/conversion features. Given the risks associated with poor ES management and the benefits that come with an optimally-managed solution, utilizing a known, verified BMS or charging solution can make a real difference. The reduction in the maintenance costs alone can more than justify the added cost of an external solution. For small systems in the IoT realm, there are complete power modules to serve as BMS to common ES sources, such as Li-ion or supercapacitors, and support energy harvesting sources.
RECOM has a customer using our products for cell balancing in an electric bus. The cells are charged …
Aside from the obvious characteristic of rechargeability, the distinction between primary and secondary storage devices is important for many of the operational and handling parameters previewed above and detailed much more in the following section. Therefore, they must be treated differently in terms of technical operation and economics. For instance, the maintenance costs of secondary cells can be significant, but they must be weighed against the cost of primary replacements and the possible disruption of service.
Primary cells tend to have a greater energy density than their secondary counterparts, but they must also be replaced if a single solution will not meet the operational lifetime of the system/application. A primary battery may be considered more reliable in a given application, but a secondary battery can be used to reduce peak power requirements, thereby reducing stress on the other components and improving the overall system reliability.
Secondary batteries and even capacitor banks can have special requirements for charge cycles to maximize life and performance. Charging may require a very complicated series of controlling different voltage/current combinations. If multiple cells are involved, then there may be additional requirements to balance cell charge for stable operation and reliability.
A battery management system (BMS) can be very advantageous for system designers interested in getting the best out of their ES. A BMS can allow for the successful operation of the battery without it being burdened by the circuitry and specialized control described throughout this white paper. If all of these factors seem overwhelming, then the good news is that there are many solutions available to help make the BMS more of a turnkey solution. Numerous semiconductor manufacturers are offering BMS ICs or even integrating the functionality, along with other energy management/conversion features. Given the risks associated with poor ES management and the benefits that come with an optimally-managed solution, utilizing a known, verified BMS or charging solution can make a real difference. The reduction in the maintenance costs alone can more than justify the added cost of an external solution. For small systems in the IoT realm, there are complete power modules to serve as BMS to common ES sources, such as Li-ion or supercapacitors, and support energy harvesting sources.
RECOM has a customer using our products for cell balancing in an electric bus. The cells are charged …
Want to read the whole Whitepaper?
Applications
