The Internet-of-Things (IoT), a proliferation of multitudes of interconnected sensors and processors, is arguably the most disruptive shift in technology since the origination of the Internet itself. It's a complex universe spanning communications, identification, location tracking, and security, enabled by multitudes of electronic equipment & devices and sensors.
Several technology advancements have been driving the IoT. The brains of these devices (embedded chips) are becoming more sophisticated and cheaper, as have reliable communications capabilities. Another reason: cloud storage for data and applications. And as the IoT expands, there are huge new opportunities for manufacturers of power source devices to make it all run.
Why Conventional Batteries Won't Work in the IoT
Let's look at the key criteria that are pretty common to the various types of IoT devices and functionalities:
- Various shapes and sizes
- Low power requirements
- High range and frequency
- Increasingly interconnected
To align with those criteria, power sources for IoT devices will need several key features of their own:
- Small/thin size and flexible shape
- Wireless connectivity -- able to charge a device on the go, easy to use anywhere
- Able to detect and select available energy resources (for wireless charging)
- Self-recharging, never needing to be replaced, with lifetime comparable to the device they're in
- Environmentally friendly, since IoT devices are used everywhere in all environments.
Consider this: the energy density of the lithium-ion battery in Apple's iPhone 5 is 142 mAh/cm3. That's 63% higher than conventional Li-ion batteries from nine years ago, and is far outpacing the ~5%/year energy density rate for Li-ion batteries. For all IoT devices including and beyond smartphones, future power source requirements -- rigid use (cycle life, specific energy and power), high tolerance capacity, flexibility (including wearability) -- don't bode well for bulky conventional batteries.
IoT Power Source Alternatives
Conventional batteries can't meet those IoT power-source requirements in most cases -- but other power sources can.
Inductive power supplies: Basically this involves wirelessly transferring energy from one device (transmitter or charging station) to another (receiver or portable device). Wireless charging was initially introduced for smartphones and still revolves around this utility; by the end of this year nearly all smartphones and tablets are expected to support it. Other inductive power applications include transcutaneous energy transfer systems in surgically implanted devices (e.g. artificial hearts) and environment-monitoring robots.
Inductive coupling is typically used to power RFID tags (IoT and RFID have common origins and extensive overlap). Such "passive" battery-free RFID sensor tags have an "on-demand" reliable source of energy, with no dependence on environmental conditions for the sensor to transmit data. They can be embedded in concrete (e.g. walls and pillars), inside piping systems, sealed within enclosures, and at many relatively inaccessible locations -- and they will never require battery-change maintenance.
With increasingly vast numbers of wirelessly connected IoT devices, the technology to power them is likely will evolve from inductive to resonant -- and from charging one device at a time to charging multiple devices concurrently. This could open up a myriad of IoT applications from smartphones to smart cars.
Thin-film and printed batteries: Thin-film batteries are being considered for, and even replacing conventional batteries in, several IoT applications such as smartphones, sensors, RFID tags, and smart cards and labels (including "smart packaging for food and medicine). Printed batteries are somewhat further away from commercial deployment for IoT applications such as electronic shelf labels and "smart shelves," though there are some efforts to bring products to the market. 3D printing of microbatteries is one intriguing area of research, with batteries as small as a grain of sand; another is in combining thin-film material layers and coatings with textiles.
Energy harvesting: This is arguably the most important sector for powering IoT devices, encompassing several technologies to facilitate ambient energy conversion and storage: PV solar cells, piezoelectric, thermoelectric, pyroelectric, geo-magnetic, electrostatic, and microwave conversion. The varying nature of ambient energy sources also means IoT wireless sensor nodes require microbatteries as backup energy sources, rechargeable once the nodes have harvested enough ambient energy.
Where IoT Power Sources Must Improve
Power source options for IoT devices currently have a number of hurdles to overcome:
Power efficiency: Technology for inductive and energy harvesting falls short of conventional wired charging in terms of power transfer efficiency, roughly 70% vs. 85%. That means it's significantly slower to charge a device, and/or requires tradeoffs in size and cost to close that gap.
Range: For wireless power (inductive/resonant coupling or energy harvesting), the range to devices must cover at least an entire room, and eventually a house.
Frequency choice: Selecting frequency to be compatible with charging standards around electromagnetic interference (EMI) and electromagnetic compatibility has been an issue for wireless charging technology.
Standards compliance and security: There are more than a dozen types of wireless communication technologies woven together to support the Internet of Things. Broad adoption of wireless charging will require standards compliance across several areas: frequency, induced electric field, induced current density, and specific absorption rate. This is a major issue, solvable by integrating them all into a single standard -- which also will solve security issues.
Between the most ubiquitous of those wireless technologies -- Bluetooth, WiFi, and ZigBee -- we see ZigBee expanding as the preferred protocol for IoT devices in terms of cost and efficiency.
Cost: Nearly all the devices in the IoT are small and low-cost, therefore so must be their power sources. Conventional chargers are generally provided with devices at no extra cost (or at least a transparent one), but wireless chargers generally come extra at cost, and often it's substantial (think $50 or more). Even thin-film and energy-harvesting batteries come at a higher cost than conventional power supplies.
Basically, two things need to happen for these power source technologies to usurp conventional batteries and take over the IoT: deliver better efficiency at lower cost, and create/adopt a wireless charging standard common for all devices and charging ports. If those happen, and limitations of these alternate power sources are overcome, then we see these technologies becoming dominant to power IoT devices.
By the Numbers: The Evolution of IoT Power Source Markets
IoT power-source technologies such as thin-film and printed batteries, energy harvesting modules, small flexible solar photovoltaic panels, and thermoelectric sources have enjoyed niche success and marginal revenues up to now. With the IoT’s emergence, however, NanoMarkets sees these products potentially generating hundreds of millions of dollars in annual revenues.
Thin-film and printed batteries make up the vast majority of today's total $57.1 million market for IoT power sources. Most of that is for mobile phones, considered the "eyes and ears" of applications connecting all the other IoT devices and networks, and mobile phones will continue to represent the main IoT application for these batteries. However, we anticipate several other applications -- notably smart cards, semiconductor/computing, and wearable electronics -- each will blossom into hundred-million-dollar battery markets by the end of this decade.
Beyond batteries, NanoMarkets sees the biggest growth opportunities over the next several years in IoT power sources addressed by inductive and energy harvesting technologies. Inductive power sources, almost exclusively used in wireless chargers, are barely a $5 million market today, but we look for this segment to crack the $100 million mark by 2018 and accelerate to $760 million by 2021. This is largely due to adoption in RFID tags, a segment that will surge to a $100 million market by 2019 and $583 million by 2021.
We see energy harvesting power sources remaining a small ($7 million) market through 2015, but then spiking to $41.5 million in 2016 thanks to rapid uptake for sensors/sensor networks. From there we see energy harvesting devices really establishing their IoT stride, to $161 million by 2018 and ultimately $557 million by 2021. Within that period we also anticipate the long-awaited arrival of wearable devices, ramping from next to nothing today to $82 million 2018 and a $200 million market by 2021.