What’s happening in green technology this week? Siemens announced that its new 14 MW wind turbine, perovskite could bring us closer to hydrogen-based energy storage, Australia is powering a gold mine with a self-contained microgrid, and NREL released a free machine learning tool to help researchers develop cleaner, more efficient fuels.
A Mighty Wind Turbine
As wind turbine power records get blown away every few years, going from 8 MW in 2014 to 12 MW in 2018, one has to wonder just how powerful they can become without being unwieldy. Siemens Gamesa announced plans for the company’s latest venture into the big wind arena: the 14 MW SG 14-222 DD, a towering behemoth with a 222-meter (728-foot) diameter rotor. That’s a “wingspan” of nearly two and a half football fields.
Just one of those monstrous machines can power up to 18,000 homes and, over the course of its 25-year projected life, keep 1.4 million tons of CO2 out of the atmosphere. The company expects to test its first prototype next year, with commercial availability projected sometime in 2024. Will it be the world’s most powerful wind turbine by then? We’ll see.
Perovskite Converts Sunlight and Water into Hydrogen
Hydrogen is the most abundant element in the universe, and when combined with oxygen, it’s capable of generating clean energy. Unfortunately, most of the Earth’s hydrogen is locked up in molecules like water and hydrocarbons, and it takes a lot of energy to break the covalent bonds holding those molecules together. With current technology, which happens to be expensive, splitting water by electrolysis can be done with efficiencies in the low 70s. Fuel cells, which combine hydrogen and oxygen to produce electricity, are only about 60 percent efficient, so using hydrogen as an energy storage medium provides, at best, around 40 percent round-trip efficiency. Given this relatively poor level of efficiency, in order for hydrogen to be economically competitive and ecologically responsible, the energy used for electrolysis must be very inexpensive and carbon free.
Scientists at Rice University have developed a perovskite-based solar cell that, when immersed in water and exposed to light, splits water molecules into hydrogen and oxygen. Perovskite is a form of carbon with photovoltaic properties, and it’s much less expensive than the silicon used in most solar cells. Because perovskite tends to break down in sunlight, researchers are trying to find ways to stabilize it. The researchers at Rice developed a polymer coating to protect the material. Since the device uses sunlight as its only energy input and it’s made from inexpensive materials, it meets the criteria for being green and low cost. The scientists are working on ways to improve both the coating and the module’s efficiency in order to make it commercially viable.
Microgrid Powers Australian Gold Mine
Mines are often located in remote locations—places where it’s difficult to bring grid power and the associated infrastructure—so mining company Gold Fields decided to power its Agnew Gold Mine with a dedicated microgrid. Located in Western Australia, the 56 MW Agnew Hybrid Renewable Power Station, operated by EDL Energy, went live in May 2020. The microgrid includes five 3.6 MW wind turbines, a 4 MW solar array, a 13 MW/4 MWh battery bank, and a 21 MW gas/diesel generator. Under good weather conditions, the gold mine can derive 70 percent of its power from the microgrid’s renewable sources, with the gas generator handling the shortfall. Under typical conditions, it’s about a 50/50 split between renewables and gas. EDL says that the microgrid will prevent almost 50,000 tons of CO2 from entering the atmosphere every year—the equivalent of taking more than 12,000 cars off the road.
Machine Learning Predicts Chemical Reactions
In the quest for cleaner, more efficient fuels, scientists often use computer models to predict the thermodynamics of chemical reactions. While that’s faster than experimentation, the process still uses an enormous amount of processing power to calculate a molecule’s bond disassociation enthalpy (BDE)—the energy required to break a chemical bond. Researchers at the National Renewable Energy Laboratory (NREL) developed a web-based machine learning algorithm that speeds up the process: ALFABET (A machine-Learning derived, Fast, Accurate Bond dissociation Enthalpy Tool). Starting with a database of over 40,000 hydrocarbons whose chemical reactions were known, the researchers trained a machine learning model to predict the results. Then they tested it on other chemicals, comparing the results with those of more traditional computational methods. ALFABET was able to produce results with the same, if not better, accuracy compared to other methods, but in just a few seconds, as opposed to several days. Their research was published in the open-access journal, Nature Communications. ALFABET is free to use—simply describe the chemical structure in simplified molecular-input line-entry system (SMILES) format and the tool calculates the various BDEs for that chemical.
Speaking of NREL, as a public agency, part of its mission is to conduct R&D and provide results to the general public free of charge. Engineers and scientists at NREL have created a slew of resources, including design and optimization software, analysis tools, energy estimators, and research reports. Many of the resources have become industry standards, enabling engineers to design with state-of-the-art tools and helping renewable energy businesses grow and thrive.
Another federal agency, the U.S. Energy Information Association (EIA) provides educational resources, such as complete lesson plans for various grade levels. The lesson plans include teacher guides, student resources, and suggested activities. This is a great way for parents and teachers to introduce kids to the world of energy and promote scientific literacy.
This Week in Green Tech brings you the latest news about renewable energy, electric vehicles, net-zero buildings, energy harvesting, IoT, and more. Subscribe to engineering.com’s “Electronics Design News” section to stay in the loop!