Texas is proud of its oil and gas industry, but the state is blessed with abundant solar and wind power potential. Tapping that potential requires more than simply building out more wind turbines and solar panels–Texas will need a large, but achievable energy storage system. This study analyzes if Texas coulld become a green state in the future.
Since the 1980s, the state tourism office has described Texas as “like a whole other country.” Texas definitely has one of the more distinctive characters of any of the United States, known for its swagger, independent spirit, and economic vigor. If Texas actually were an independent country it would have the 10th largest economy in the world, just behind Italy.
The state is also intensely identified with the energy industry. The oilman is a recognized Texas character, from fictional J.R. Ewing to real-life H.L. Hunt, Clint Murchison, and Sid Richardson. For decades, one of the state’s NFL teams was known as the Oilers.
While Texas is best known for its petroleum and gas, it is also rich in potential solar and wind power resources. According to the American Wind Energy Association, the state is first in the nation in installed wind power capacity, with 23,421 MW from 12,793 wind turbines that are now ubiquitous in the Texas landscape. This generates 62.2 TWh of electricity per year, or 17.4 percent of the power on the state’s electric grid. The technically achievable wind power potential stands at a staggering 1,347,992 MW.
The potential for solar power in Texas is also enormous. In 2017, the solar farms in the state generated 2,119 GWh of electricity; according to a 2016 report from the National Renewable Energy Laboratory, Texas has more than 20,400,000 MW of potential utility-scale solar power, which could generate as much as 41,300 TWh per year.
With all that potential, it is worth looking at whether it is technically possible to run the state’s electrical grid without emitting any carbon dioxide, using wind, solar, hydroelectric, and the existing nuclear power. Political writers like to classify Texas as a red state trending toward purple, but could it become in the future a green state?
My colleagues and I have modeled the potential electricity supply and demand, and we have found that, yes, Texas could become a green state. It is not as simple as building out more wind turbines and solar panels. But because Texas is so large, the solutions for the reduction of CO2 emissions that may work here will also work without scaling in most of the developed economies.
All Its Own
While Texas’s individual streak is famous, the state also goes its own way with electricity. Alone among the 48 contiguous states, most of Texas is on its own electric grid, separated from the large Eastern and Western Interconnections. The Electric Reliability Council of Texas (ERCOT) manages the production, distribution, and supply of electric power—357.3 TWh in 2017, more than Italy or the United Kingdom—to more than 92 percent of the state's inhabitants. In addition to wind and solar power contributions, nuclear power provides 11 percent of the annual electricity production, natural gas fuels about 38 percent, and coal another 31 percent.
That independent grid creates some problems for replacing the carbon-generated electricity. Texas is big, but it isn’t so big that weather couldn’t affect most of the state at the same time, dimming solar power production with heavy overcast skies or stilling wind turbines with calm air. Other states faced with those conditions could draw wind power from the northern Great Plains or solar power from Nevada and Arizona. Texas will have to do without those out-ofstate resources.
“The only way to achieve higher penetration of renewable energy is to mitigate the effects of the variability by storing energy.”
Being on its own grid presents another, almost opposite problem, too: oversupply. Increase wind and solar power production enough, and on sunny or windy low-demand days, the amount of power produced by nuclear and renewables could easily outpace the net demand.
The swing from oversupply to undersupply of renewable electricity is captured in the so-called duck curve, which shows how the power production goes out of sync with the consumers’ demand as the sunshine increases and decreases on a daily basis. Once the market penetration of renewable wind or solar energy over the course of a year reaches levels higher than 30 percent, the supply of solar-powered electricity can encroach on the nuclear power output or even overwhelm the total demand during mid-day. Once the sun starts to go down and demand increases due to air conditioners being turned on in the afternoon heat, the need for non-solar power ramps up extraordinarily fast.
The situation is even worse with the intermittent wind power generation, which produces the rattlesnake curve for the rest of the power-generating units in the area. It would be impossible to operate nuclear units with this type of electric power demand and without utility-level storage. In addition, the reliability of the entire electric power grid would be severely compromised. Similar problems with grid reliability arise when the penetration of renewables exceeds the range 20 to 25 percent of the annually produced electricity, not only in this region, but in most electric energy markets.
It quickly becomes apparent that the only way to achieve higher penetration of renewable energy in the ERCOT region is to mitigate the effects of the variability of the renewable energy sources by storing energy, when excess electric energy is produced, and using the stored energy at a later time, when the demand is high.
Energy storage is a bit tricky in Texas. Pumped hydro storage requires a sort of terrain that is absent in Texas. And while battery storage may become an option in the future, the current generation of batteries is not suitable for utility-level storage, especially for storing energy for several months to better match the seasonal peaks of solar and wind power with the seasonal electrical demand peaks.
Hydrogen storage is a more attractive option. The processes of production, storage, and conversion of hydrogen are sufficiently advanced for hydrogen storage to be implemented on a large scale: Production of hydrogen by electrolysis has been used for almost two centuries. Hydrogen storage tanks at pressures as high as 700 bar are currently used by several commercial automotive companies that produce hydrogen-driven cars and buses. Fuel cells have also been used for almost two centuries to directly convert the chemical energy of hydrogen to electricity.
To be sure, significant thermodynamic irreversibilities occur during the processes of hydrogen production and hydrogen-to-electricity conversion in fuel cells. With current technology, the round-trip efficiency of the electricity-to-hydrogen-to-electricity processes is in the range of 40 to 50 percent. Such losses would have to be accounted for.
It is easy to envision a Texas-based grid that largely avoids the emission of carbon dioxide, relying on nuclear, solar, wind, and hydropower for the generation of electricity, with mismatches between supply and demand mediated via the production and consumption of hydrogen. In many respects, electrolyzed hydrogen would replace natural gas. It could be produced throughout the state, stored in local facilities, transported via pipeline as needed, with market prices used to ensure the adequate supply of the gas in all the regions of the state during every season of the year.
The question remains, though, whether such a system would be practical. That’s where modeling comes in.
Fossil Fuel Replacement
Together with Matt Leonard and Dimitri Michaelides, I calculated many of the main parameters needed for a “green Texas” model. We started with the hourly electricity demand data of the year 2017 in the entire ERCOT system, then added the hourly irradiance in the state (averaged over a period of five years) and the hourly available wind energy.
The first, obvious question is, how can Texas forego the contribution of coal power plants that currently provide 31 percent of the annual electric energy in the ERCOT region? Removing coal still leaves four nuclear reactors and a number of gas-fired plants in the mix in addition to wind, solar, and a small amount of hydropower. Allowing the existing nuclear and gas plants to operate as they do today, we looked at how much hydrogen storage would be needed to account for the inevitable lulls in wind and overcast days.
“Modeling shows that storing enough hydrogen is possible and not even technically challenging.”
Modeling the minimum storage requirements is not straightforward. We had to account for thermodynamic irreversibilities and the large fraction of lost electric energy associated with the storage process, mentioned above. A second constraint is that at the annual minimum storage level, the storage systems should contain sufficient hydrogen to provide the grid during the next 10 days with the surplus power that was taken from storage in the previous ten days. That constraint allows the grid operators to produce additional hydrogen or to purchase hydrogen in the market in cases of spurious demand or system malfunction. The constraint also implies that the stored energy in the entire system does not reach zero at any time in the year, but maintains a minimum value.
Our hourly simulations of the demand and supply showed that, for the complete elimination of coal from the electric power production mix, the electric utilities will need to add 22,130 MW (rated) wind power and 9,320 MW (rated) PV capacity. In addition, approximately 45,800 m3 of hydrogen storage capacity is needed to be developed in order to store an annual maximum of approximately 700 Mmol H2 (equivalent to 46 GWh of electricity) at the maximum pressure 500 bar. That storage volume, 45,800 m3, is equal to about 1.6 million cubic feet, which is a rounding error compared to the 853,043 million cubic feet of natural gas storage in Texas, as reported by the Energy Information Administration.
The intersection of weather and demand means that storage and depletion follows a regular cycle. Energy is stored primarily during the spring season and the energy is used in the late summer, when the sun and wind supply less electric power, but the demand is still high because of air-conditioning. (High temperatures in Fort Worth peak in early August, but that month has about as much daily solar insolation as the much more temperate May.)
We also calculated what it would take to substitute all fossil fuels—both coal and gas—with solar and wind to power the ERCOT system, leaving only nuclear as the nonrenewable source of electricity. Obtaining agreement on such a complete switch to renewables would be a formidable challenge, as natural gas is produced in Texas and its consumption has become a patriotic task for many Texans. But modeling such a system is useful to determine the scale of energy storage needed to accomplish it.
To replace the 69 percent of electricity generation now produced by coal and gas power, utilities would need to add approximately 30,800 MW (rated) of wind power and 52,200 MW (rated) PV capacity, in addition to the existing capacity. The needed hydrogen storage capacity is approximately 15.9 million m3, or 562 million cubic feet, that would store a maximum of 243,000 Mmol H2 (equivalent to 16,070 GWh) at the maximum pressure of 500 bar.
As with the more limited scenario, in this complete replacement scenario would see stored hydrogen accumulate in the spring—or more precisely, from February to mid-June—with another small bump up in autumn. A small fraction of the stored energy is used during the winter months when solar input is minimal, but most of the stored energy is spent in the summer, when the air-conditioning demand is high. From mid-June to late September, the storage system would discharge the full 16,070 GWh, or a bit less than 12 percent of the electricity demand during that season in 2017.
More Like Texas
An electricity system dominated by wind, solar, and storage would look quite a bit different from the current one. One major change would be the size and distribution of the facilities. Currently, coal and nuclear power plants are scaled in many hundreds of megawatts, so only a handful are needed. There are less than two dozen coal and nuclear power stations scattered across Texas.
Solar and wind power are diffuse, and to effectively capture that power, facilities need to be widely distributed. Some places are windier and some are sunnier, so the largest wind and solar farms will be located there. But rooftop solar, for instance, will bring electricity production to nearly every neighborhood, and will likely feed power into small districts served by microgrids.
To match this distributed generation, hydrogen production and storage may also be distributed throughout the ERCOT system area. The estimated 15.9 million m3 of hydrogen storage capacity needed to support wind and solar power corresponds to approximately 1.5 m3 or 53 gallons per household. One can imagine the electricity system shored up with relatively small hydrogen tanks in each household.
Texas is just one state of 50, but it plays a large and out-sized role in the energy industry. And it consumes a lot of electricity—so much so that converting the state to “green” power production would reduce global CO2 emissions by 0.71 percent.
Our modeling shows that storing enough hydrogen to buffer a wind- and solar-dominated electrical system in Texas is possible and not even technically challenging. We did not calculate the cost of adding the green generating and storage capacity because prices drop precipitously when such systems are widely used as household items—as the history of the dramatic price reductions of refrigerators, microwaves, and personal computers has shown. We also did not consider any political resistance that might be raised by owners of coal and gas power assets.
But if converting the electrical system to non-carbon sources can work in Texas—and I believe it can—then it can be a model for most places on Earth. Texas may be a whole other country, but in terms of energy, most countries can be like Texas.