A groundbreaking study published in the journal Nature Sustainability has quantified what many in the cryptocurrency space have long suspected: mining Bitcoin requires significantly more energy than extracting equivalent value from traditional precious metals. The research, conducted by scientists at the Oak Ridge Institute for Science and Education in Cincinnati, Ohio, found that producing one dollar’s worth of Bitcoin consumes approximately 17 megajoules of energy — more than three times the 5 megajoules needed to mine a dollar’s worth of gold.
TL;DR
- Oak Ridge Institute study finds Bitcoin mining uses 17 MJ per $1 mined — 3x more than gold
- Ethereum and Litecoin each consume 7 MJ per dollar, Monero uses 14 MJ per dollar
- Copper requires just 4 MJ, gold 5 MJ, and platinum 7 MJ to mine $1 worth
- Four tracked cryptocurrencies responsible for 3-16 million tons of CO2 emissions annually
- Bitcoin network consumed approximately 30.1 billion kWh as of mid-2018
The Numbers Behind the Energy Gap
The researchers tracked the daily energy demand and hashrate of four major cryptocurrencies — Bitcoin, Ethereum, Litecoin, and Monero — over an 18-month period spanning January 1, 2016 through June 30, 2018. Using average daily market prices and block rewards, they calculated the energy cost per dollar of cryptocurrency mined.
Bitcoin’s 17 megajoules per dollar dwarfed not only gold’s 5 MJ but also copper’s 4 MJ and platinum’s 7 MJ. Ethereum and Litecoin each required 7 MJ to produce one dollar of value, while Monero — a privacy-focused cryptocurrency — consumed 14 MJ per dollar. For context, aluminum mining remains the most energy-intensive traditional commodity at 122 MJ per dollar.
At current prices near $6,461, the Bitcoin network’s aggregate energy consumption had reached approximately 30.1 billion kilowatt-hours as of July 2018, according to the study. With Bitcoin’s market capitalization hovering around $112 billion, the scale of energy expenditure has drawn increasing scrutiny from environmental researchers and policymakers.
CO2 Emissions and Geographic Disparities
The study estimated that mining the four tracked cryptocurrencies is responsible for between 3 and 16 million metric tons of CO2 emissions. However, the researchers emphasized that the environmental impact varies dramatically depending on where mining operations are located.
“Any cryptocurrency mined in China would generate four times the amount of CO2 compared to the amount generated in Canada,” the authors wrote, underscoring how the carbon intensity of local power grids directly influences the ecological footprint of mining operations. China’s reliance on coal-fired power plants versus Canada’s significant hydroelectric capacity accounts for much of this disparity.
The researchers used median values across the 18-month study period to account for the extreme price volatility that characterized cryptocurrency markets during that timeframe. Even through dramatic market swings, the hashrate of all four tracked cryptocurrencies consistently increased, suggesting that energy consumption would continue rising regardless of price direction.
A Growing Debate on Sustainability
The study represents the first effort to evaluate cryptocurrency mining from an energy-cost-per-dollar perspective, providing a direct comparison with traditional mineral extraction. The authors stated their goal was to “encourage debate on whether these energy demands are both sustainable and appropriate given the product that results from relatively similar energy consumption.”
The findings add fuel to an already heated debate about Bitcoin’s environmental impact. Previous estimates have drawn striking comparisons — one analysis placed the Bitcoin network’s power consumption on par with the entire nation of Ireland, while another calculated that it produced carbon emissions equivalent to one million transatlantic flights per year.
Notably, the Ethereum project has publicly discussed transitioning from energy-intensive proof-of-work consensus to a proof-of-stake model, which would dramatically reduce its energy requirements. However, as of November 2018, such a transition remained largely theoretical.
With Ethereum trading at approximately $218 and the broader crypto market capitalization above $209 billion, the energy question is no longer a niche concern. As institutional interest grows and mining operations scale, the tension between cryptocurrency’s decentralized promise and its environmental cost will only intensify.
Why This Matters
This study marked a pivotal moment in the crypto energy debate by providing the first rigorous, peer-reviewed comparison between digital and physical mining. The 17 MJ per dollar figure for Bitcoin became a widely cited benchmark in subsequent discussions about cryptocurrency sustainability. As the industry matured, this research laid the groundwork for the eventual shift toward proof-of-stake consensus mechanisms and the growing emphasis on renewable energy sources in mining operations — trends that would reshape the crypto landscape in the years that followed.
Disclaimer: This article is for informational purposes only and does not constitute financial advice. Cryptocurrency markets are highly volatile. Always conduct your own research before making investment decisions.
17 megajoules per dollar vs 5 for gold. the environmental crowd never lets you forget this one
but gold mining destroys actual land and ecosystems. bitcoin just uses electricity. false equivalence
this. gold mining literally removes mountaintops and poisons water tables. btc mining uses electricity that can come from anywhere including stranded hydro and flared gas. the comparison is dishonest
17 mj per dollar was true in 2018 when the study came out. network efficiency has improved massively since then with asic upgrades. wonder if anyone reran the numbers post-china ban
Tanja W. someone did rerun it in 2023 using Cambridge data and BTC was around 9-11 MJ per dollar at higher prices. still higher than gold but way closer
ferro_hash_ glad someone mentioned the Cambridge redo. 9 to 11 MJ per dollar at higher prices completely changes the narrative but nobody cites the updated numbers
thermodynamic_ the Cambridge redo getting 9-11 MJ per dollar completely changes the conversation. but activists still cite the 2018 Oak Ridge numbers because the headline is scarier
the 2018 numbers are hopelessly outdated. post-China ban the hashrate shifted massively to North America where renewable penetration is way higher. the 17 MJ figure needs a 2026 update
renewable_btc the 17 MJ number is from 2018 hardware efficiency. S21 hydrominers do like 0.02 J/GH. the study needs a complete redo with post-2021 fleet data
ines_m S21 hydrominers at 0.02 J/GH make the 17 MJ figure ancient history. the fleet turned over twice since 2018. anyone still citing that number is being dishonest
Tanja W. the Cambridge redo found 9-11 MJ but nobody updated the headlines. activists prefer the scary number
nature sustainability publishing a btc hit piece. surprise surprise. wonder who funded it
comparing energy per dollar mined treats gold and BTC as equivalent goods when they serve completely different functions. gold sits in vaults, BTC settles transactions globally
comparing joules per dollar for gold extraction vs bitcoin mining is comparing a physical process to a digital settlement network. apples to oranges dressed up as science
comparing joules per dollar is meaningless when gold sits in a vault for decades doing nothing and btc settles billions in transfers daily
17 MJ per dollar in 2018 when the study dropped. Cambridge revised to 9-11 in 2023. still gets cited because the headline number is scarier
comparing joules per dollar for gold extraction vs a global settlement network is apples to oranges. gold sits in a vault doing nothing after you dig it out
17 MJ per dollar was measured on S9 hardware in 2018. S21 hydrominers do 0.02 J/GH. the study is archaeology at this point not current data
s21_meta the cambridge centre redid the numbers in 2023 and got 9-11 MJ per dollar. still higher than gold but the gap closed massively. oak ridge used S9 data which is 3 hardware generations old now
s21_meta Cambridge redid the numbers in 2023 and got 9-11 MJ. still higher than gold but the gap closed significantly. nobody cites the updated figures though because scary headlines get more traction
comparing joules per dollar for gold extraction vs BTC settlement is comparing a mining pick to a global payment rail. the energy serves completely different purposes
oak ridge is a legit research institute but comparing energy per dollar mined ignores that bitcoin provides a global settlement network not just a commodity you dig out of the ground
dead_stoat_ exactly. gold sits in a vault doing nothing for centuries. BTC settles billions in value transfer daily. comparing joules per dollar without accounting for utility is like comparing a truck to a bicycle by fuel consumption
17 MJ per dollar was the 2018 number. Cambridge revised it to 9-11 MJ in 2024 using post-China-ban data. the study is still cited because the headline is scarier than the update
comparing joules per dollar for gold extraction vs a global settlement network is comparing a shovel to a banking system. the framing is fundamentally broken