Energy evolution is the story of how living systems and human civilizations have found increasingly powerful ways to capture, store, and spend energy over roughly four billion years. It begins with simple chemical gradients on the early Earth, passes through the invention of photosynthesis and the rise of complex cells, and arrives at the fossil-fuel economy and the technologies poised to replace it. Each transition followed a similar pattern: a new energy source or conversion method unlocked possibilities that the previous one could not support, but also introduced new constraints and trade-offs.
Proton Gradients and the First Energy Currencies
Long before any cell existed, the early Earth already had energy gradients that could drive chemistry. The leading model for how life tapped into those gradients centers on alkaline hydrothermal vents, fissures on the ocean floor where warm, alkaline fluid seeps into acidic seawater. The pH difference across the thin mineral walls of these vents is strikingly similar in direction and strength to the proton gradients that modern cells use to make energy.1PubMed. Proton gradients at the origin of life Those mineral barriers contained iron-nickel-sulfur compounds whose structure resembles the catalytic centers of enzymes in living cells today, hinting that the earliest metabolism may have been a geochemical process before it was a biological one.2PubMed Central. An origin-of-life reactor to simulate alkaline hydrothermal vents
Once primitive biochemistry got going, it needed a portable way to carry energy from one reaction to another. That role fell to ATP, the molecule every known organism still uses as its primary energy currency. Research suggests ATP was not just an evolutionary accident that stuck around: its formation appears to be chemically favored in water under mild conditions resembling the prebiotic Earth, which may explain why it became universal rather than being replaced by something else.3PubMed Central. A prebiotic basis for ATP as the universal energy currency Before ATP took center stage, simpler molecules like acetyl phosphate and even bare electrons and hydrogen species served as energy carriers, a progression from raw electrochemistry to the sophisticated phosphate-based system life relies on now.4PubMed. The evolution and biophysical economics of energy currencies: from prebiotic electrochemistry to modern digital money
Energy Without Sunlight
Photosynthesis dominates most discussions of biological energy, but some of the deepest ecosystems on Earth run on a power source that has nothing to do with the sun. In the deep continental crust, the radioactive decay of uranium, thorium, and potassium splits water molecules and produces hydrogen gas. Fracture water collected from deep rock units in South Africa’s Witwatersrand Basin contains hydrogen concentrations up to two molar, far beyond what is found in shallow groundwater or marine sediments.5Geochemistry, Geophysics, Geosystems. Radiolytic H2 in continental crust: Nuclear power for deep subsurface microbial communities This radiolytic hydrogen feeds methane-producing microbes and even drives abiotic hydrocarbon synthesis, sustaining communities that are genuinely independent of surface photosynthesis.
The process can also generate oxygen. In ancient, long-isolated brines with high concentrations of radioactive elements, radiolysis produces enough oxidants to support aerobic metabolism deep underground over timescales of thousands to billions of years.6ISME Communications. Radiolytic support for oxidative metabolism in an ancient subsurface brine system These ecosystems matter for the broader story of energy evolution because they show that life’s energy options were never limited to a single source. Even today, nuclear-powered microbial communities persist kilometers below the surface, a reminder that biological energy strategies are more diverse than the textbook sun-to-plant-to-animal chain suggests.
The Photosynthesis Revolution
The single most consequential energy innovation in Earth’s history was the evolution of oxygenic photosynthesis, the ability to use sunlight to split water and release oxygen. Evidence suggests this capability appeared surprisingly early. Biomarker and geochemical data from rocks around 2.7 billion years old strongly imply that oxygen-producing cyanobacteria were already present, and some lines of evidence push the date back even further, possibly to 3.2 billion years ago or beyond.7PubMed Central. When did oxygenic photosynthesis evolve?
Here is the puzzle: if photosynthesis showed up that early, why did oxygen not accumulate in the atmosphere until the Great Oxidation Event roughly 2.4 billion years ago? Modeling work suggests a counter-intuitive answer. An earlier origin of oxygenic photosynthesis actually leads to a longer delay before oxygen builds up in the atmosphere, because early photosynthesizers faced a severe bottleneck. Without enough phosphorus dissolved in the oceans to fuel large-scale biological productivity, oxygen production stayed low, and what oxygen did appear was quickly consumed by reacting with reduced minerals and volcanic gases.8PubMed Central. An early origin of oxygenic photosynthesis delays the Great Oxidation Only once surface phosphorus reservoirs grew large enough did photosynthetic productivity cross the threshold needed to permanently oxygenate the atmosphere.
When it did, everything changed. Oxygen made aerobic respiration possible, which extracts roughly an order of magnitude more energy from the same food molecules than anaerobic pathways. That surplus underwrote every subsequent leap in biological complexity.
Mitochondria and the Rise of Complex Cells
Prokaryotes, the bacteria and archaea that dominated the first two billion years of life, are metabolically impressive but structurally limited. The energy transition that unlocked complex, large-celled life was the acquisition of mitochondria. When an ancient archaeal cell engulfed a bacterium capable of oxidative phosphorylation, it gained an internal power plant. Over evolutionary time, the engulfed bacterium lost most of its own genome to the host’s nucleus but kept the genes needed for local control of energy production on its own membranes.9Current Biology. How energy flow shapes cell evolution
This arrangement created an extreme asymmetry: a small mitochondrial genome devoted almost entirely to energy production, and a large nuclear genome free to expand and diversify. The result was thousands of times more energy available per gene than any prokaryote could manage. That surplus did not just allow bigger cells. It supported more genes, more regulatory DNA, more protein synthesis, and ultimately the morphological complexity that distinguishes animals, plants, and fungi from bacteria.
Whether mitochondria were strictly necessary for the earliest eukaryotes remains debated. Recent analyses suggest that at the cell sizes and genome sizes plausible for the last common ancestor of eukaryotes, prokaryotic-style energy budgets might have been sufficient, and mitochondria became essential only as cells grew larger and genomes ballooned in size.10PubMed Central. The role of mitochondrial energetics in the origin and diversification of eukaryotes Either way, once mitochondria were in place, the ceiling on biological complexity lifted dramatically.
Metabolic Scaling and the Energy Cost of Getting Bigger
As organisms grew larger across evolutionary time, the relationship between body size and energy use changed in telling ways. The classic observation, often called Kleiber’s law, holds that an animal’s metabolic rate increases with body mass but not in lockstep. A mouse burns far more energy per gram than an elephant. The scaling exponent, however, is not a universal constant. It shifts across the major domains of life: metabolic rate scales faster than body size in prokaryotes, roughly proportionally in single-celled eukaryotes, and slower than body size in multicellular animals.11PubMed Central. Shifts in metabolic scaling, production, and efficiency across major evolutionary transitions of life
These shifts track the major evolutionary transitions in how cells organize themselves. A free-living bacterium that doubles its size roughly doubles its metabolic rate. A multicellular animal that doubles its mass adds less than double the metabolic rate, in part because larger bodies have proportionally less active surface area and more tissue devoted to structural support. Contemporary work explores how cell size, mitochondrial dynamics, and energy storage strategies within individual cells shape the metabolic rate of the whole organism.12PubMed. Beyond Kleiber’s Law: Variation and Mechanisms of Metabolic Scaling For the energy evolution story, the key point is that each new level of biological organization came with a different energy economy, not just more of the same.
One particularly dramatic example involves the human brain. The brain is metabolically ravenous, consuming a disproportionate share of the body’s energy budget. The expensive-tissue hypothesis proposes that humans could afford such a costly organ because other metabolically expensive tissues, particularly the gut, shrank in compensation.13PubMed Central. The Expensive-Tissue Hypothesis in Vertebrates: Gut Microbiota Effect, a Review A shorter gut, in turn, demanded higher-quality food that could be digested quickly, which likely drove early humans toward cooking and meat consumption. In this way, the brain’s energy demands shaped not just anatomy but behavior and, eventually, culture.
From Muscle to Machine on the Farm
For most of human history, the primary energy converters in food production were muscles, human and animal. As recently as 1950, about 80% of on-farm physical work was still powered by people and draft animals, and global land productivity had been roughly stagnant at about 1.7 million kilocalories per hectare of cropland since 1800.14SpringerLink / CrossRef. The Contributions of Muscle and Machine Work to Land and Labor Productivity in World Agriculture Since 1800 Muscle power is low in both raw output and efficiency, so even with widespread use of oxen and horses, there was a hard ceiling on how much food a given area of land could produce.
From 1950 to 2012, the introduction of tractors, combine harvesters, irrigation pumps, and the fossil-fuel-derived chemicals that feed modern agriculture nearly tripled land productivity. But this gain came at a cost: the roughly threefold increase in food output per hectare required a nearly 4.5-fold increase in physical work per hectare, pointing to diminishing returns. Each additional calorie of food now demands more mechanical energy input than the last. The agricultural energy transition mirrors a broader pattern in energy evolution: new energy sources unlock spectacular gains early on, then run into efficiency walls as the easy improvements are exhausted.
Fossil Fuels and the Decline of Energy Return
The industrial revolution was fundamentally an energy revolution. Coal, oil, and natural gas offered energy densities that no biological system could match, and their exploitation transformed every aspect of civilization. But the energy return on these fuels has not been constant. Using price-based analysis of global production data stretching back to 1800, researchers have estimated that the energy return on investment (EROI) of oil peaked in the 1930s and 1940s at around 50 to 1, meaning that for every unit of energy spent extracting oil, about 50 units came back. Natural gas peaked even higher, around 150 to 1, during roughly the same period. Both have declined since, as the easiest-to-reach deposits were tapped first and extraction has moved to more challenging environments.15Ecological Economics. Long-Term Estimates of the Energy-Return-on-Investment (EROI) of Coal, Oil, and Gas Global Productions
Coal’s EROI has followed a different trajectory and may not yet have peaked globally. This partly reflects the vast remaining reserves and the fact that surface mining in some regions is still relatively cheap in energy terms. But declining EROI across fossil fuels as a class means that society spends a growing fraction of its energy budget just obtaining energy, leaving less net energy for everything else. That dynamic is one of the structural pressures behind the push toward renewables.
The Planet Feels It
The fossil-fuel era has also altered the planet’s energy balance in a more literal sense. Earth absorbs slightly more energy from the sun than it radiates back to space, and that imbalance is growing. Measurements from 2005 to 2010 put the planetary energy imbalance at about 0.58 watts per square meter, driven primarily by human-produced greenhouse gases.16Atmospheric Chemistry and Physics. Earth’s energy imbalance and implications Aerosols from burning fossil fuels partially mask the warming by reflecting sunlight, exerting an estimated cooling effect of about −1.6 watts per square meter. If aerosol emissions were cleaned up without reducing greenhouse gases, the warming effect would accelerate. This is the planetary-scale consequence of humanity’s energy choices, and it feeds directly into the urgency around developing cleaner energy systems.
Solar Cells, Batteries, and the Efficiency Frontier
The transition to renewable energy is, at its core, a return to solar power, the same energy source photosynthesis tapped billions of years ago, but with engineered materials instead of chlorophyll. Single-junction silicon solar cells face a theoretical efficiency ceiling of about 33%, known as the Shockley-Queisser limit, set by the physics of how a single semiconductor material absorbs light. Recent experimental work has pushed well past this barrier: by suppressing the conversion of light into waste heat at low temperatures, researchers have achieved power conversion efficiencies of 50 to 60% in single-junction silicon cells.17PubMed Central. Surpassing Shockley-Queisser Efficiency Limit in Photovoltaic Cells Other strategies being explored include using the crystal lattice itself as an energy reservoir to recycle heat losses back into useful electricity.18EES Solar. Moving beyond the Shockley–Queisser limit: current bottlenecks and a new direction in solar energy conversion
Capturing sunlight is only half the problem. Storing that energy for use at night and during cloudy periods requires batteries, and lithium-ion technology has been the dominant solution. Early lithium-ion cells stored around 100 watt-hours per kilogram. Contemporary cells exceed 200 watt-hours per kilogram, and experimental silicon-graphene composite anodes have pushed energy density to roughly 300 watt-hours per kilogram.19Journal of Energy Storage. Advancing energy storage: The future trajectory of lithium-ion battery technologies That threefold improvement in a few decades is impressive, though batteries still store orders of magnitude less energy per kilogram than fossil fuels, which is one reason liquid fuels remain hard to replace in aviation and shipping.
Fusion and the Pursuit of Stellar Energy
If solar panels represent a return to the sun’s energy, fusion aims to recreate the sun’s own process on Earth. Fusing light atomic nuclei, typically deuterium and tritium, releases enormous amounts of energy with no carbon emissions and negligible long-lived radioactive waste compared to fission. The central challenge is expressed by the Lawson criterion, which relates the density, temperature, and confinement time of a plasma to whether it produces net energy gain.20Physics of Plasmas. Progress toward fusion energy breakeven and gain as measured against the Lawson criterion
Meeting those conditions has been the work of decades. Recent theoretical and simulation work has explored ways to improve the odds, including exploiting non-Maxwellian ion distributions (where the particles in the plasma do not follow the standard statistical spread of speeds) to increase fusion reaction rates under certain conditions.21Physics of Plasmas. Enhancement of the fusion reactivity due to the D-T non-Maxwellian ion distribution and its impact on Lawson criterion Researchers are also evaluating aneutronic fuels like proton-boron-11, which would produce fewer damaging neutrons, though the temperature and confinement requirements are far more demanding than for standard deuterium-tritium fuel.22Frontiers in Nuclear Engineering. Evaluation of the Lawson criterion for aneutronic proton-boron-11 fusion: effects of ion temperature and bremsstrahlung losses Fusion remains a technology that is not yet commercially viable, but the pace of investment and experimental progress has accelerated in recent years.
Measuring Civilization on the Energy Scale
One way to step back and view the entire arc of energy evolution, from hydrothermal vents to fusion reactors, is through the Kardashev Scale, which ranks civilizations by their total energy consumption. A Type I civilization uses all the energy available on its home planet, a Type II harnesses the full output of its star, and a Type III commands the energy of an entire galaxy. By this measure, humanity is not yet a Type I civilization. Machine-learning projections based on current consumption trends place us at roughly 0.73 on the scale today, with global energy consumption expected to reach about 887 exajoules by 2060, bringing us to about 0.74.23PubMed Central. Forecasting the progression of human civilization on the Kardashev Scale through 2060 with a machine learning approach
Those numbers are humbling. After four billion years of biological energy innovation and a few centuries of industrial acceleration, our civilization still captures only a tiny fraction of the energy our planet receives from the sun. The gap between where we are and what physics allows is enormous. Whether closing it looks like planet-spanning solar arrays, working fusion reactors, or something not yet imagined, the direction of travel is the same one it has always been: each era figures out how to unlock an energy source the previous era could not reach, and that surplus reshapes everything that follows.
The Energetic Cost of Information
One frontier that links biological and technological energy evolution is the cost of processing information. Cells are, among other things, information-processing systems: they sense their environment, regulate gene expression, and make decisions about growth and division. Research estimating how much energy biological systems spend on processing a single bit of information has found that real cells pay between ten thousand and ten million times the theoretical minimum, known as the Landauer bound, which represents the absolute floor set by thermodynamics.24Physics of Plasmas. Progress toward fusion energy breakeven and gain as measured against the Lawson criterion Modern computers are similarly inefficient relative to physical limits, though for different reasons.
This inefficiency is not a flaw so much as a signature of how energy evolution works in practice. Theoretical limits set the floor, but real systems operate far above it because reliability, speed, and robustness all cost extra energy. Biology has had billions of years to optimize, and it still runs at a steep premium over the thermodynamic minimum. As computing and artificial intelligence consume a growing share of global electricity, understanding and reducing the energy cost per bit of computation is becoming its own chapter in the story of energy evolution, one where lessons from biological systems and engineered ones may eventually converge.