Silicon Carbide (SiC): The Secret Weapon for Improving Distributed Solar Power Efficiency
Click on the blue words to follow us
Some of SiC's advantages come into play in power boost circuits, which make solar energy conversion more efficient. This article focuses on a circuit design for matching the output impedance of a solar array (which varies with the level of incident light) to the input impedance required by the inverter for the most efficient conversion.
Consumers, industries and governments are taking steps to increase the use of renewable energy. This is driving a reshaping of the power generation and distribution system from a centralized hub-and-spoke architecture to a more grid-based, localized generation and consumption, with smart grid interconnections to smooth supply and demand.
According to the International Energy Agency’s (IEA) October 2019 Fuels Report, electricity generation from renewable sources will grow by 50% by 2024.
This means that global renewable energy generation will increase by 1,200GW, equivalent to the current installed capacity of the United States. The report predicts that 60% of the increase in renewable energy generation will be in the form of solar photovoltaic (PV) equipment.
Figure 1: Renewable energy generation growth by technology, 2019-2024
The report also highlights the importance of distributed photovoltaic systems as consumers, commercial buildings and industrial facilities begin to generate their own electricity. It predicts that distributed photovoltaic capacity will more than double to more than 500 GW by 2024. This means that distributed photovoltaic power generation will account for nearly half of the total growth in solar photovoltaic power generation.
Figure 2: Growth in distributed photovoltaic (PV) power generation by market segment, 2007-2024
Why does solar photovoltaic power generation take such a leading position in the growth of renewable energy power generation?
One obvious reason is that the sun shines on us all, so its energy is widely available. This allows generation to be closer to consumption, delivering power to off-grid locations, which is particularly useful for reducing distribution losses.
Another obvious reason is that there is a lot of solar energy. There are many nuances to calculating how much energy the Earth receives from the Sun, a rule of thumb is 1kW per square meter on a sunny day at sea level, or 6kWh per square meter per day when factoring in day/night cycles, angle of incidence, seasonality, etc.
Solar cells use the photoelectric effect to convert incident light into electrical energy in the form of a stream of photons. Photons are absorbed by semiconductor materials such as doped silicon, and their energy excites electrons from their molecular or atomic orbits. These electrons are then free to dissipate the excess energy as heat and return to their orbits, or to travel to the electrodes and become part of the current to offset the potential difference they create across the electrodes.
As with all energy conversion processes, not all of the energy input to a solar cell is output in the preferred form of electrical energy. In fact, for many years, the efficiency of single-crystal silicon solar cells has hovered between 20% and 25%. However, the opportunity for solar photovoltaics is so great that research teams have been working for decades to improve cell conversion efficiency using increasingly complex structures and materials, as shown in this graphic from NREL.
Figure 3: Global best solar cell conversion efficiency – 1976 to 2020 (NREL) (Source: National Renewable Energy Laboratory, Golden, Colorado, USA)
The higher energy efficiencies shown are usually achieved at the expense of using multiple different materials and more complex and expensive manufacturing techniques.
Many solar photovoltaic devices are based on various forms of crystalline silicon or thin films of silicon, cadmium telluride or copper indium gallium selenide, with conversion efficiencies ranging from 20% to 30%. The cells are built into modules, which installers use as the basic units to build solar photovoltaic power generation systems.
Photovoltaic conversion converts 1 kilowatt of solar energy incident on the Earth into 200 to 300 watts of electrical energy per square meter of surface. Of course, this is under ideal conditions. But the conversion efficiency may be reduced due to the deposition of rain, snow and dust on the cell surface, the effects of aging of semiconductor materials, and increased shading due to environmental changes such as the growth of vegetation or the construction of new buildings.
So the reality is that, while solar energy is free, generating useful electricity from it requires careful optimization of each stage of collection, storage, and ultimate conversion to electricity. One of the biggest opportunities for improving energy efficiency is in the design of the inverter, which converts the DC output of a solar array (or its battery storage) into AC current for direct use or transmission over the grid.
The inverter changes the polarity of the DC input current to approximate the AC output. The higher the switching frequency, the more efficient the conversion. Simple switches produce a square wave output that can drive resistive loads, but with harmonics that can damage more complex electronic devices powered by pure sine wave AC. Therefore, inverter design becomes a key balance between increasing the switching frequency to improve energy efficiency, operating voltage, and power generation, while minimizing the cost of auxiliary components used to smooth the square wave.
Silicon carbide (SiC) has several material advantages over silicon in solar energy management due to its wide bandgap, with thermal conductivity almost three times that of silicon. This means that SiC devices withstand almost 10 times the breakdown electric field of silicon, allowing SiC devices to operate efficiently at much higher voltages than similarly structured silicon. SiC devices also have much lower on-resistance, gate charge, and reverse recovery charge characteristics than silicon, as well as higher thermal conductivity. These characteristics mean that SiC devices can be switched at higher voltages, frequencies, and currents than their silicon equivalents, while managing heat buildup more efficiently.
SiC is used to make devices that are not suitable for silicon. MOSFETs are favored in switching applications because they are unipolar devices, which means they do not use minority carriers. Silicon bipolar devices use both majority and minority carriers and can operate at higher voltages than silicon MOSFETs, but their switching speeds are slower because they need to wait for electrons and holes to recombine and dissipate recombination energy when switching.
Silicon MOSFETs are widely used in switching applications up to about 300 V, above which the on-resistance of the device rises to the point where designers have to turn to slower bipolar devices. SiC’s high breakdown voltage means it can be used to make higher voltage MOSFETs than silicon while retaining the fast switching advantages of low-voltage silicon devices. Switching performance is also relatively unaffected by temperature, resulting in stable performance as the system heats up.
Since power conversion efficiency is directly related to switching frequency, SiC can both handle higher voltages than silicon and ensure the high-speed switching required for high conversion efficiency.
SiC has three times the thermal conductivity of silicon, so it can operate at higher temperatures. While silicon stops acting as a semiconductor at around 175°C and becomes a conductor at around 200°C, SiC does not become a conductor until it reaches 1000°C. The advantages of SiC's thermal properties can be used in two ways. First, it can be used to create power converters that require less cooling than equivalent silicon systems. In addition, SiC's stable operation at higher temperatures can be used to create extremely high-density power conversion systems for applications where space is at a premium, such as vehicles and cellular base stations.
Figure 4: Introducing SiC devices to improve conversion efficiency of solar boost circuits (ON Semiconductor)
We can see these advantages of SiC come into play in power boost circuits, which make solar energy conversion more efficient.
The circuit is designed to match the output impedance of the solar array (which varies with the level of incident light) to the required input impedance of the inverter for the most efficient conversion.
The leftmost figure shows the lowest cost approach, using silicon diodes and MOSFETs. The first optimization scheme, shown in the middle figure, is to replace silicon diodes with SiC diodes, which will increase the power density and conversion efficiency of the circuit, thereby reducing system cost. It is also possible to replace silicon MOSFETs with SiC equivalents, as shown in the right figure, which provides designers with more switching frequency options, further improving the conversion efficiency and power density of the circuit.
In response to the need to utilize SiC in solar photovoltaic equipment, ON Semiconductor has also developed a series of two-channel or three-channel SiC boost modules for solar inverters.
SiC power devices have superior performance to silicon devices, including their ability to switch high voltages and currents at high speeds, low losses, and good thermal performance.
Although they may currently be more expensive than equivalent silicon products (if silicon alternatives are available), their system-level performance can save costs and optimize cooling complexity. There is also an estimate of conversion efficiency: if the deployment of SiC can improve the power conversion efficiency of all distributed solar photovoltaic systems, the IEA predicts that even if only 2% is installed by 2024, it will generate an astonishing 10GW of additional power generation.